Treatment of the Postmenopausal Woman Basic and Clinical Aspects THIRD EDITION
This Page Intentionally Left Blank
Treatment of the Postmenopausal Woman Basic and Clinical Aspects THIRD EDITION EDITOR
Rogerio A. Lobo, M.D. Department of Obstetrics and Gynecology Columbia University College of Physicians and Surgeons New York, New York
Academic Press is an imprint of Elsevier 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, California 92101-4495, USA 84 Theobald's Road, London WCIX 8RR, UK
This book is printed on acid-free paper.
Copyright 6) 2007, Elsevier Inc. All rights reserved. Portions of this work have been adapted from Menopause:Biology and Pathobiology by Rogerio A. Lobo, Jennifer Kelsey, and Robert Marcus (Academic Press, 2000), and Treatment of the Postmenopausal Woman:Basic and Clinical/Ispects, SecondEdition, by Rogerio A. Lobo (Lippincott Williams & Wilkins, 1999).
Notice No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permissions may be sought directly from Elsevier's Science &Technology Rights Department in Oxford, UK: phone: (+44) 1865 843830, fax: (+44) 1865 853333, E-mail:
[email protected] may also complete your request online via the Elsevier homepage (http://elsevier.com),by selecting "Support & Contact" then "Copyright and Permissions" and then "Obtaining Permissions." The Publisher
Library of Congress Cataloging-in-Publication Data 2007926088
British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 978-0-12-369443-0
For information on all Elsevier Academic Press publications visit our Web site at www.books.elsevier.com Aquisitions Editor: Tari Broderick Developmental Editor: Renske van Dijk Publishing Services Manager: Melissa Lastarria Project Manager: Mary Pohlman Marketing Manager: Katherine Clevenger
Working together to grow libraries in developing countries www.elsevier.com I www.bookaid.org I www.sabre.org Printed in the United States of America 07 08 09
10 11 12 9 8 7 6 5 4 3 2
1
Contributors
Numbers in parentheses indicate the pages on which the authors' contributions begin.
David F. Archer (169, 847) Department of Obstetrics and Gynecology, CONRAD Clinical Research Center, Eastern Virginia Medical School, Norfolk, Virginia 23507 Cecilia Artacho (655) Department of Obstetrics/Gynecology, Columbia University, New York, New York 10032 Gloria Bachmann (263) Department of Obstetrics/Gynecology & Medicine, UMDNJ-Robert Wood Johnson Medical School, New Brunswick, New Jersey 08901 Randall B. Barnes (767) Department of Obstetrics and Gynecology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois 60633 Kurt T. Barnhart (1) Division of Reproductive Endocrinology and Infertility, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104 Yves Muscat Baron (227) Department of Obstetrics and Gynecology, St. Luke's Hospital Medical School, University of Malta, Msida MSD 06, Gwardamangia, Malta Narender N. Bhatia (693, 739) Division of Female Pelvic Medicine and Reconstructive Surgery, Department of Obstetrics and Gynecology, Harbor-UCLA Medical Center, David Geffen School of Medicine, University of California at Los Angeles, Torrance, California 90509 Stephanie V. Blank (593) Section of Gynecologic Oncology, Department of Obstetrics and Gynecology, New York University School of Medicine, New York, New York 10016 Mark Brincat (227) Department of Obstetrics and Gynecology, St. Luke's Hospital Medical School, University of Malta, Msida MSD 06, Gwardamangia, Malta
Henry G. Burger (67) Prince Henry's Institute, Clayton, Victoria 3168, Australia John E. Buster (821) Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology and Infertility, Baylor College of Medicine, Houston, Texas 77030 R.Jeffrey Chang (49) University of California, San Diego, School of Medicine, La Jolla, California 92093 Ru-fongJ. Cheng (263) Department of Obstetrics, Gynecology, & Reproductive Sciences, UMDNJ-Robert Wood Johnson Medical School, New Brunswick, New Jersey 08901 Judi L. Chervenak (157) Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology and Infertility, Montefiore Medical Center, Albert Einstein College of Medicine, New York, New York 10461 Claus Christiansen (377, 393) Center for Clinical and Basic Research, 2750 Ballerup, Denmark Thomas B. Clarkson (509) Wake Forest University School of Medicine, Comparative Medicine Clinical Research Center, WinstonSalem, North Carolina 27157 Felicia Cosman (323, 837) Helen Hayes Hospital, West Haverstraw, New York 10993; Columbia University College of Physicians and Surgeons, New York, New York 10032 John P. Curtin (585) Department of Obstetrics and Gynecology, New York University School of Medicine, New York, New York 10016 Ivaldo da Silva (199) Gynecology Department, Federal University of S~o Paulo, Brazil 04038-031
Vi
Susan R. Davis (799) Women's Health Program, Monash University (CECS), Alfred Hospital, Prahran, Victoria 3181, Australia Lorraine Dennerstein (271) Office for Gender & Health, Department of Psychiatry, The University of Melbourne, Parkville 3010, Australia Martina DSren (813) Charitd-Universitiitsmedizin Berlin, Campus Benjamin Franklin, Clinical Research Center of Women's Health, D-12200 Berlin, Germany Paul S. Dudley (821) Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology and Infertility, Baylor College of Medicine, Houston, Texas 77030 Richard Eastell (337) Academic Unit of Bone Metabolism, University of Sheffield, Metabolic Bone Centre, Sorby Wing, Northern General Hospital, Sheffield, United Kingdom $5 7AU Gregory F. Erickson (49) University of California, San Diego, School of Medicine, La Jolla, California 92093 David A. Fishman (593) Section of Gynecologic Oncology, New York University School of Medicine, New York, New York 10016 Ian S. Fraser (149) Department of Obstetrics and Gynaecology, University of Sydney, NSW 2006, Australia Robert R. Freedman (187) Departments of Obstetrics and Gynecology and Psychiatry and Behavioral Neurosciences, Wayne State University School of Medicine, Detroit, Michigan 48201 Adriane Fugh-Berman (683) Department of Physiology and Biophysics, Georgetown University Medical Center, Washington, DC 20057 Ray Galea (227) Department of Obstetrics and Gynecology, St. Luke's Hospital Medical School, University of Malta, Msida MSD 06, Gwardamangia, Malta J.C. OaUagher (847) Department of Medicine, Bone Metabolism Unit, Creighton University School of Medicine, Omaha, Nebraska 68178 Margery L.S. Gass (611) Clinical Obstetrics and Gynecology, University of Cincinnati College of Medicine, Cincinnati, Ohio 45267 Radhika Gogoi (593) Section of Gynecologic Oncology, Department of Obstetrics and Gynecology, New York University School of Medicine, New York, New York 10016 Ellen B. Gold (77) Division of Epidemiology, Department of Public Health Sciences, University of California, Davis, California 95616
CONTRIBUTORS
Juan Gonzalez (1) University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104 Eiran Zev Gorodeski (405) Department of Cardiovascular Medicine, Cleveland Clinic, Cleveland, Ohio 44195 George I. Gorodeski (405) Department of Reproductive Biology, CASE (Case Western Reserve) University, Cleveland, Ohio 44106 Gall A. Greendale (77) Divisions of Geriatrics/General Internal Medicine, Department of Internal Medicine, David Geffen School of Medicine, University of California at Los Angeles, Torrance, California 90095 Allison R. Hagey (655) Department of Obstetrics/Gynecology, Columbia University, New York, New York 10032 Georgina E. Hale (149) Department of Obstetrics and Gynaecology, University of Sydney, NSW 2006, Australia RosemaryA. Hannon (337) Academic Unit of Bone Metabolism, University of Sheffield, Metabolic Bone Centre, Sorby Wing, Northern General Hospital, Sheffield, United Kingdom $5 7AU Victor W. Henderson (295) Departments of Health Research and Policy, and of Neurology and Neurological Sciences, Stanford University, Stanford, California 94305 William H. Hindle (579) Department of Obstetrics and Gynecology, University of Southern California; The William Hindle, M.D. Breast Diagnostic Center, Los Angeles, California 90033 Mat H. Ho (693, 739) Division of Female Pelvic Medicine and Reconstructive Surgery, Department of Obstetrics and Gynecology, Harbor-UCLA Medical Center, David Geffen School of Medicine, University of California at Los Angeles, Torrance, California 90509 Howard N. Hodis (529) Atherosclerosis Research Unit, Division of Cardiovascular Medicine, Keck School of Medicine, University of Southern California, Los Angeles, California 90033 Christian F. Holinka (863) PharmConsult | New York, New York 10468 Diane M.Jacobs (287) Consultant, San Diego, California; Department of Psychiatry, Mount Sinai School of Medicine, New York, New York 10468 Jay R. Kaplan (509) Wake Forest University School of Medicine, Comparative Medicine Clinical Research Center, Winston-Salem, North Carolina 27157
CONTRIBUTORS
Richard H. Karas (453) Molecular Cardiology Research Center, Tufts-New England Medical Center, Tufts University School of Medicine, Boston, Massachusetts 02111 Morten A. Karsdal (393) Pharmos Bioscience, 2730 Herlev, Denmark Harry Irving Katz (237) Department of Dermatology, University of Minnesota, Minneapolis, Minnesota 55455 Fergus S.J. Keating (351) Giggs Hill Surgery, Thames Ditton, Surrey, United Kingdom KT7 0EB Michael Kleerekoper (331) St. Joseph Mercy Hospital, Ann Arbor, Michigan; Department of Internal Medicine and Obstetrics and Gynecology, Wayne State University School of Medicine, Detroit, Michigan; St. Joseph Mercy Reichert Health Center, Ypsilanti, Michigan 48197 Kwang Kon Koh (471) Vascular Medicine and Atherosclerosis Unit, Division of Cardiology, Gil Medical Center, Gachon Medical School, Namdong-Gu, Incheon, Korea 405-760 Ronald M. Krauss (461) Children's Hospital & Research Center Oakland, Oakland, California 94609 Carlo La Vecchia (599) Istituto di Ricerche Farmacologiche "Mario Negri," 20157 Milano, Italy; Istituto di Statistica Medica e Biometria, Universit~ degli Studi di Milano, 20133 Milano, Italy Sulggi Lee (569) Department of Preventive Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, California 90033 Seth G. Levrant (767) Tinley Park, Illinois 60477 Robert Lindsay (323, 837) Helen Hayes Hospital, West Haverstraw, New York 10993; Columbia University College of Physicians and Surgeons, New York, New York 10032 Rogerio A. Lobo (875) Department of Obstetrics & Gynecology, Columbia University College of Physicians and Surgeons, New York, New York 10032 WendyJ. Mack (529) Department of Preventive Medicine, Division of Biostatistics, Keck School of Medicine, University of Southern California, Los Angeles, California 90033 Sten Madsbad (501) Department of Endocrinology, Hvidovre Hospital, University of Copenhagen, Denmark
vii
Pauline M. Maki (279) Departments of Psychiatry and Psychology, Center for Cognitive Medicine, University of Illinois at Chicago, Chicago, Illinois 60612 JoAnn E. Manson (619) Division of Preventive Medicine, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts 02215 Donald P. McDonnell (17) Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, North Carolina 27710 C. Noel Bairey Merz (471) Division of Cardiology, Department of Medicine, Cedars-Sinai Research Institute, Cedars-Sinai Medical Center; Department of Medicine, University of California School of Medicine, Los Angeles, California 90048 Karin B. Michels (619) Obstetrics and Gynecology Epidemiology Center, Department of Obstetrics, Gynecology and Reproductive Biology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts 02115 Frederick Naftolin (199) Department of Obstetrics and Gynecology, New York University School of Medicine, New York, New York 10016 Morris Notelovitz (481,855) Adult Women's Health & Medicine, Boca Raton, Florida 33496 James M. Olcese (829) Department of Biomedical Sciences, Florida State University College of Medicine, Tallahassee, Florida 32306 Annlia Paganini-Hill (627) Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, California 90033 C. Leigh Pearce (569) Department of Preventive Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, California 90033 Alison C. Peck (157) The Fertility Institutes, Encino, California 91436 James H. Pickar (863) Wyeth Research, Collegeville, Pennsylvania 19426 Malcolm C. Pike (569) Department of Preventive Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, California 90033
viii
Janet Hill Prystowsky (237) St. Luke's-Roosevelt Hospital Center, New York, New York 10021 Nancy King Reame (127) School of Nursing, Columbia University, New York, New York 10032 Robert W. Rebar (99,471) American Society for Reproductive Medicine, Birmingham, Alabama 35216 Catherine A. Roca (307) Behavioral Endocrinology Branch, National Institute of Mental Health, Bethesda, Maryland 20892 Elisheva Rovner (263) Women's Health Institute, UMDNJ-Robert Wood Johnson Medical School, New Brunswick, New Jersey 08901 David R. Rubinow (307) Department of Psychiatry, University of North Carolina, Chapel Hill, North Carolina 27599 Ichiro Sakuma (471) Cardiovascular Medicine, Hokko Memorial Hospital, Sapporo, Japan G6ran Samsioe (251,813) Department of Obstetrics & Gynecology, Lund University Hospital, 221 85 Lund, Sweden Mary Sano (287) Alzheimer's Disease Research Center, Department of Psychiatry, Mount Sinai School of Medicine, James J. Peters VA Medical Center, Bronx, New York 10468 Nanette Santoro (157) Division of Reproductive Endocrinology, Department of Obstetrics, Gynecology & Women's Health, Albert Einstein College of Medicine, Bronx, New York 10461 MarkV. Sauer (111) Columbia University Medical Center, New York, New York 10032 PeterJ. Schmidt (307) Behavioral Endocrinology Branch, National Institute of Mental Health, Bethesda, Maryland 20892 Hermann P.G. Schneider (639) Department of Obstetrics & Gynecology, University of Muenster, 48149 Muenster, Germany Andrea B. Sherk (17) Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, North Carolina 27710 Barbara B. Sherwin (217) Department of Psychology, McGi11 University, Montreal, Quebec, Canada H3A 1B1 Joe Leigh Simpson (29) Department of Obstetrics and Gynecology, Human Molecular Genetics, Florida Internal University, Miami, Florida 33199
CONTRIBUTORS
Sven O. Skouby (501) Department of Obstetrics and Gynecology,Frederiksberg Hospital, University of Copenhagen, Denmark Leon Speroff (1) Department of Obstetrics and Gynecology, Oregon Health & Science University, Portland, Oregon 97201 DarcyV. Spicer (569) Department of Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, California 90033 Frank Z. Stanczyk (779) Departments of Obstetrics & Gynecology, and Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, California 90033 John C. Stevenson (351) National Heart & Lung Institute, Imperial College London, Royal Brompton Hospital, London, United Kingdom SW3 6NP Lfiszl6 B. Tank6 (377, 393) Center for Clinical and Basic Research, 2750 BaUerup, Denmark HelenaJ. Teede (67) Jean Hailes Research Group, Monash Institute for Health Services Research, Clayton, Victoria, 3168 Australia Prati Vardhana (111) Columbia University Medical Center, New York, New York 10032 MicheUe P. Warren (655) Department of Obstetrics/Gynecology, Columbia University, New York, New York 10032 Carolyn Westhoff (491) Columbia University Medical Center, New York, New York 10032 Anna H. Wu (569) Department of Preventive Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, California 90033 Byung-Koo Yoon (471) Obstetrics and Gynecology, Samsung Medical Center, Sungkyunkwan University, School of Medicine, Seoul, Korea RalfC. Zimmermann (829) Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology, Columbia University College of Physicians and Surgeons, New York, New York 10032
Preface
This is the third edition of Treatment of the Postmenopausal Woman."Basic and Clinical Aspects. In reality it is a merging of chapters from the second edition of this book and a book I edited together with Bob Marcus and Jennifer Kelsey, entitled Menopause, which dealt with more basic and research issues. Although these books were published almost 8 years ago, it was important for me to wait until now to update the book because of the various randomized clinical trials on hormonal therapy that were being conducted at the time of the last edition. Now that we have a great deal of new clinical information that has been synthesized and updated, it is appropriate to try to put these findings into perspective. It should be stated, however, that menopausal management is more than just hormonal therapy, and it is the aim of this edition to approach menopausal management more globally. There are now close to 50 million postmenopausal women in the United States and many more worldwide. In developed countries, once a woman reaches the age of natural menopause, her life expectancy is approximately 84 years. As life expectancy increases in women, the total population of postmenopausal women increases, and this increasing segment of the population requires comprehensive health care. Most countries around the world have established national menopause societies, lending credence to the importance and interest in this field of health care. For women in developed countries, the leading cause of death is still cardiovascular disease (coronary disease and stroke), followed by total cancer deaths; lung cancer mortality exceeds that of breast cancer. All of these concepts and findings will be discussed in this book. The book has been written with a clinical orientation and is appropriate reading for all providers of health care for women, as well as for medical professionals in training. Newer sections have been added since the last edition, and as in the previous edition, each section is preceded with a preface, in which I have attempted to orient the reader to the key issues contained therein. Once again, I hope the reader will enjoy this book. It is my desire that this book will help provide the basis and rationale for strategies that will result in better health care for the mature woman, a growing segment of the U.S. population and the world.
Rogerio A. Lobo, M.D.
ix
This Page Intentionally Left Blank
Contents
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
The Menopause" A Signal for the Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Leon Speroff, Kurt T. Barnhart, and Juan Gonzalez
Section I" Basics to Enhance Our Understanding 2.
Molecular Pharmacology of Estrogen and Progesterone Receptors . . . . . . . . . . . . . . . . . . . . . . Andrea B. Sherk and DonaM P McDonnell
17
3.
Genetic Programming in Ovarian Development and Oogenesis . . . . . . . . . . . . . . . . . . . . . . . Joe Leigh Simpson
29
4.
Basic Biology: Ovarian Anatomy and Physiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gregory E Erickson and R. Jeffrey Chang
49
5.
Endocrine Changes During the Perimenopause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Henry G. Burger and Helena J. Teede
67
6.
Epidemiology of Menopause: Demographics, Environmental Influences, and Ethnic and International Differences in the Menopausal Experience . . . . . . . . . . . . . . . . . . Ellen B. Gold and Gaild. Greendale
77
Section II: Ovarian Senescence and Options 0
0
Premature Ovarian Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert W. Rebar Reproductive Options for Perimenopausal and Menopausal Women . . . . . . . . . . . . . . . . . . . . . Mark V. Sauer and Prati Vardhana
99 111
Section III: The Perimenopause Neuroendocrine Regulation of the Perimenopause Transition . . . . . . . . . . . . . . . . . . . . . . . . Nancy King Reame
127
10.
Changes in the Menstrual Pattern During the Menopause Transition . . . . . . . . . . . . . . . . . . . . Georgina E. Hale and Ian S. Fraser
149
11.
Decisions Regarding Treatment During the Menopause Transition . . . . . . . . . . . . . . . . . . . . . Alison C. Peck,Judi L. Chervenak, and Nanette Santoro
157
12.
Use of Contraceptives for Older Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . David E Archer
169
0
xi
xii
CONTENTS
Section IV: Changes Occurring After Menopause 13.
Menopausal Hot Flashes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
187
Robert R. Freedman 14.
Clinical Effects of Sex Steroids on the Brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
199
Ivaldo da Silva and Frederick Naftolin
15.
Impact of the Changing Hormonal Milieu on Psychological Functioning . . . . . . . . . . . . . . . . . .
217
Barbara B. Sherwin 16.
Connective Tissue Changes in the Menopause and with Hormone Replacement Therapy . . . . . . . . .
227
Mark Brincat, Ray Galea, and Yves Muscat Baron 17.
Menopause and the Skin . . . . . . . . .
..................................
237
Harry Irving Katz and Janet Hill Prystowsky
18.
Urogenital Symptoms around the Menopause and Beyond . . . . . . . . . . . . . . . . . . . . . . . . . .
251
GO'ran Samsioe
19.
Vulvovaginal Complaints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
263
Gloria Bachmann, Ru-fongJ. Cheng, and Elisheva Rovner 20.
Sexuality: Clinical Implications from Epidemiologic Studies . . . . . . . . . . . . . . . . . . . . . . . . .
271
Lorraine Dennerstein
Section V: Brain Function 21.
Menopause and Cognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
279
Pauline M. Maki 22.
Cognitive Health in the Postmenopausal Woman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
287
Diane M. Jacobs and Mary Sano 23.
The Role of Sex Steroids in Alzheimer's Disease: Prevention and Treatment . . . . . . . . . . . . . . . .
295
Victor W. Henderson 24.
Estrogens and Depression in Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
307
David R. Rubinow, Catherine A. Roca, and PeterJ. Schmidt
Section VI: Bone Changes 25.
Pathogenesis of Osteoporosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
323
Robert Lindsay and Felicia Cosman 26.
Assessment of Bone Density and Bone Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
331
Michael Kleerekoper 27.
Biochemical Markers of Bone Turnover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
337
Richard Eastell and Rosemary A. Hannon 28.
Interventions for Osteoporosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
351
Fergus S.J. Keating and John C. Stevenson
29.
Treatment of Osteoporosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
377
Ldszl6 B. Tank6 and Claus Christiansen
30.
Potentials of Estrogens in the Prevention of Osteoarthritis: W h a t Do We Know and W h a t Questions are Still Pending? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ldszl6 B. Tank5, Claus Christiansen, and Morten A. Karsdal
393
xiii
CONTENTS
Section VII" Cardiovascular 31.
Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women . . . . . . . . . . . Eiran Zev Gorodeski and George I. Gorodeski
405
32.
Mechanisms of Action of Estrogen on the Cardiovascular System . . . . . . . . . . . . . . . . . . . . . . Richard H. Karas
453
33.
Lipids and Lipoproteins and Effects of Hormone Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . Ronald M. Krauss
461
34.
The Effects of Hormone Therapy on Inflammatory, Hemostatic, and Fibrinolytic Markers in Postmenopausal Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kwang Kon Koh, Byung-Koo Yoon, C. Noel Bairey Merz, Icloiro Sakuma, and Robert W. Rebar
471
35.
Hormone Therapy and Hemostasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Morris Notelovitz
481
36.
Risk of Pulmonary Embolism/Venous Thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carolyn Westhoff
491
37.
Glucose Metabolism After Menopause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sven O. Skouby and Sten Madsbad
501
38.
Stage of Reproductive Life, Atherosclerosis Progression and Estrogen Effects on Coronary Artery Atherosclerosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Thomas B. Clarkson and Jay R. Kaplan
509
Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy on Cardiovascular Disease: Facts, Hypotheses, and Clinical Perspective . . . . . . . . . . . . . . . . . . . . . Howard N. Hodis and WendydI. Mack
529
39.
Section VIII: Neoplasia 40.
Body Weight, Menopausal Hormone Therapy, and Risk of Breast Cancer . . . . . . . . . . . . . . . . . . Anna H. Wu, C. Leigh Pearce, Darcy V. Spicer, Sulggi Lee, and Malcolm C. Pike
569
41.
Postmenopausal Breast Surveillance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . William H. Hindle
579
42.
Endometrial Cancer and Hormonal Replacement Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . John P. Curtin
585
43.
Ovarian Cancer and Its Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Radhika Gogoi, Stephanie V. Blank, and Davidd. Fishman
593
Hormone Replacement Therapy in Menopause and Breast, Colorectal, and Lung Cancer: An U p d a t e . . . Carlo La Vecchia
599
4Q
Section IX: Clinical Trials and Observational Data 45.
The Women's Health Initiative~Data and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . Margery L.S. Gass
46.
Postmenopausal Hormone Therapy in the 21st Century: Reconciling Findings from Observational Studies and Randomized Clinical Trials . . . . . . . . . . . . . . . . . . . . . . . . . Karin B. Michds andJodnn E. Manson
47.
Morbidity and Mortality Changes with Hormone Therapy . . . . . . . . . . . . . . . . . . . . . . . . . dnnlia Paganini-Hill
611
619 627
CONTENTS
xiv
Section X: Life Cycle and Q OL 48.
Issues Relating to Quality of Life in Postmenopausal Women and Their Measurement . . . . . . . . . . . Hermann P.G. Schneider
639
49.
Role of Exercise and Nutrition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michelle P. Warren, CeciliaArtacho, and Allison R. Hagey
655
50.
Herbs, Phytoestrogens, and Other CAM Therapies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adriane Fugh-Berman
683
Section Xl: Urinary Symptoms and Pelvic Support 51.
Lower Urinary Tract Disorders in Postmenopausal Women . . . . . . . . . . . . . . . . . . . . . . . . . Mat H. Ho and Narender N. Bhatia
693
52.
Pelvic Organ Prolapse in Postmenopausal Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mat H. Ho and Narender N. Bhatia
739
Section Xll: Hormonal Therapy 53.
Pharmacology of Estrogens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Randall B. Barnes and Seth G. Levrant
54.
Structure-Function Relationships, Pharmacokinetics, and Potency of Orally and Parenterally Administered Progestogens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Frank Z. Stanczyk
767
779
55.
Intervention: Androgens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Susan R. Davis
799
56.
Future Strategies in Climacteric Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GOran Samsioe and Martina DO'ren
813
Section Xlll: Some Alternative Medical Therapies 57.
Alternative Therapy: Dehydroepiandrosterone for Menopausal Hormone Replacement . . . . . . . . . . . Paul S. Dudley and John E. Buster
821
58.
Melatonin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ralf C. Zimmermann and James M. Olcese
829
59.
Selective Estrogen Receptor Modulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Felicia Cosman and Robert Lindsay
837
60.
Tibolone: Selective Tissue Estrogen Activity Regulator Utilization in Postmenopausal Women . . . . . . . David E Archer andJ. C. Gallagher
847
Section XIV: Women's Centers, Information Needed for Clinical Trials, and the Future 61.
Integrated Adult Women's Medicine: A Model for Women's Health Care Centers . . . . . . . . . . . . . Morris Notelovitz
853
62.
Clinical Trials in Postmenopausal Hormone Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christian E Holinka and James H. Pickar
863
63.
The Furore of Therapy and the Role of Hormone Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . Rogerio d. Lobo
875
Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
881
~HAPTER
_
The Menopause: A Signal for the Future LEON SPEROFF
Department of Obstetrics and Gynecology, Oregon Health & Science University, Portland, OR 97201
K U R T T. B A R N H A R T
Division of Reproductive Endocrinology and Infertility, University of Pennsylvania Medical Center, Philadelphia, PA 19104
JUAN G O N Z A L E Z
University of Pennsylvania Medical Center, Philadelphia, PA 19104
Throughout recorded history, multiple physical and mental conditions have been attributed to the menopause. Although medical writers often wrote colorfully in the past, they were also less than accurate, unencumbered by scientific information and data. A good example of the stereotypical, inaccurate thinking promulgated over the years is the following passage, which was written in 1887 (1). The ovaries, after long years of service, have not the ability of retiring in graceful old age, but become irritated, transmit their irritation to the abdominal ganglia, which in turn transmit the irritation to the brain, producing disturbances in the cerebral tissue exhibiting themselves in extreme nervousness or in an outburst of actual insanity. The belief that behavioral disturbances are related to manifestations of the female reproductive system is an ancient one that has persisted in contemporary times. This belief regarding menopause is not totally illogical; there is reason to associate the middle years of life with negative experiences. The events that come to mind are impressive: onset of a major illness or disability (even death) in a spouse, relative, or friend; retirement from employment; financial insecurity; the need to provide care for very old parents and relatives; and separation
TREATMENT OF THE POSTMENOPAUSAL W O M A N
from children. It is therefore not surprising that a middle-age event, the menopause, shares in this negative outlook. The scientific study of all aspects of menstruation has been hampered by the overpowering influence of social and cultural beliefs and traditions. Problems arising from life events have often been erroneously attributed to the menopause. However, data, especially more reliable communitybased longitudinal data, have established that the increase of most symptoms and problems in middle-aged women reflects social and personal circumstances, not the endocrine events of the menopause (2-8). The Massachusetts Women's Health Study, a large and comprehensive prospective, longitudinal study of middleaged women, provides a powerful argument that menopause is not and should not be viewed as a negative experience by most women (3,9). The cessation of menses was perceived by these women (as by the women in other longitudinal studies) to have almost no impact on subsequent physical and mental health. Most women expressed positive or neutral feelings about menopause. An exception was the group of women who experienced surgical menopause, but the reasons for the surgical procedure usually were more important than the cessation of menses. Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
2
SPEROFF ET AL.
The Study of Women's Health across the Nation (SWAN), a large multiethnic cohort study, found that menopause-related symptoms exert a stronger influence on impaired functioning. Menopause in itself does not impact quality of life. Experiencing night sweats, urinary incontinence, and hot flashes is, however, associated with lower health-related quality of life. SWAN also showed that significant ethnic differences exist regarding health-related quality of life (ga). Among the ethnic groups studied, Bodily Pain and Social Functioning subscales remained significant even in adjusted analyses. Alterations in menstrual function are not symbols of some ominous "change." There are good physiologic reasons for changing menstrual function, and understanding the physiology can do much to reinforce a healthy, normal attitude. Attitude and expectations about menopause are important. Women who have been high users of health services and who expect to have difficulty do experience greater symptoms and higher levels of depression (4,9). The symptoms that women report are related to many variables within their lives, and the hormonal change at menopause cannot be held responsible for the common psychosocial and lifestyle problems we all experience. It is time to stress the normalcy of this physiologic event. Menopausal women do not suffer from a "hormone deficiency disease." Hormone therapy remains effective for treating vasomotor symptoms, vaginal atrophy, and retardation of osteoporosis in select patients (gb). The risks of hormone therapy exceed the benefits for the prevention of chronic diseases. It can be further argued that physicians have had a biased negative point of view regarding women's experience of menopause, because most women who are healthy and happy do not seek contact with physicians (10,11). Clinicians must be familiar with the facts about menopause and have an ap-
propriate attitude and philosophy regarding this period of life. Medical intervention at this point of life should be regarded as an opportunity to provide and reinforce a program of preventive health care. The issues of preventive health care for women are familiar ones. They include family planning, cessation of smoking, control of body weight and alcohol consumption, prevention of accidents and trauma, prevention of cardiovascular disease and osteoporosis, maintenance of mental well-being (including sexuality), cancer screening, and treatment of urologic problems.
I. AGE OF MENOPAUSE Menstrual irregularity is the only marker used to define and establish the perimenopausal transition. The menopause is permanent cessation of menstruation after the loss of ovarian activity. Menopause is derived from the Greek words men (month) and pausis (cessation). The years before menopause that encompass the change from abnormal ovulatory cycles to cessation of menses are known as the perimenopausal transitional years, marked by irregularity of menstrual cycles. Climacteric indicates the period when a woman passes from the reproductive stage of life through the perimenopausal transition and the menopause to the postmenopausal years. Climacteric is derived from the Greek word for ladder. In July 2001 the Stages of Reproductive Aging Workshop (STRAW) was convened to establish nomenclature and guidelines (Fig. 1.1) (lla). Designating the average age of menopause has been somewhat difficult. Based on cross-sectional studies, the median age was estimated to be between 50 and 52 (12). These studies relied on retrospective memories and the subjective vagaries of the individuals being interviewed.
FIGURE 1.1 The STRAWstaging system.SoulesMR, Sherman S, Parrott E, e t al.
Executive summary:Stagesof ReproductiveAgingWorkshop (STRAW).Fertilityand Sterility 2001;76:874-878.
3
CHAPTER 1 The Menopause: A Signal for the Future SWAN provided a cross-sectional examination of the relation of lifestyle and demographic factors to age at natural menopause in seven U.S. centers and five ethnic groups. Median age of menopause was found to be 51.4 years. Current smoking, lower educational attainment, being separated/ widow/divorced, nonemployment, and history of heart disease were all independently associated with earlier natural menopause. In the contrary parity, prior use of oral contraceptives and Japanese ethnicity were associated with later age of menopause (13). A median age of menopause means that only one-half the women have reached menopause at this age. In the classic longitudinal study by Treolar, the average age of menopause was 50.7 years, and the range that included 95% of the women was 44 to 56 years (14). In a survey in the Netherlands, the average age of menopause was 50.2 years (15). About 1% of women experience menopause before the age o f 40 (16). Clinical impression has suggested that mothers and daughters tend to experience menopause at the same age, and two studies indicate that daughters of mothers with an early menopause (before age 46) also have an early menopause (17-19). There is sufficient evidence to believe that undernourished women and vegetarians experience an earlier menopause (17,20). Because of the contribution of body fat to estrogen production, thinner women experience a slightly earlier menopause (21). Consumption of alcohol is associated with a later menopause (18). This is consistent with the reports that women who consume alcohol have higher blood and urinary levels of estrogen and greater bone density (22-26). There is no correlation between age of menarche and age of menopause (14,15,17). In most studies, race, parity, and height have no influence on the age of menopause; however, two cross-sectional studies found later menopause to be associated with increasing parity (15,17,21). Two studies have found that irregular menses among women in their forties predicts an earlier menopause (27,28). A French survey detected no influence of heavy physical work on early menopause (before age 45) (29). An earlier menopause is associated with living at high altitudes (30). Multiple studies have consistently documented that an earlier menopause (average of 1.5 years earlier) is a consequence of smoking. There is a dose response relationship with the number of cigarettes smoked and the duration of smoking (31,32). Even former smokers show evidence of an impact. There is reason to believe that premature ovarian failure can occur in women who have previously undergone abdominal hysterectomy, presumably because ovarian vasculature has been compromised (33). Unlike the decline in age of menarche that occurred with an improvement in health and living conditions, most historical investigation indicates that the age of menopause has changed little since the reports from ancient Greece (34,35). A few authorities have disagreed, concluding that the age of
menopause did undergo a change, starting with an average age of about 40 years in ancient times (36). If there has been a change, however, history indicates it has been minimal. Even in ancient writings, 50 is usually cited as the age of menopause.
II. SYMPTOMS OF MENOPAUSE During the menopausal years, some women experience multiple severe symptoms, but others have no reactions or minimal reactions that can go unnoticed. The differences in reactions to menopausal symptoms across different cultures is poorly documented, and it is difficult to do so. Individual reporting is so conditioned by sociocultural factors that it is hard to determine what is caused by biologic or cultural variability. Women often seek medical assistance for disturbances in menstrual pattern, hot flushes, atrophic conditions, and psychologic symptoms. Disturbances in the menstrual pattern include anovulation and reduced fertility, decreased or increased flow, and irregular frequency of menses. Vasomotor instability results in the hallmark symptom of menopause, the hot flush (Table 1.1). The vasomotor flush is viewed as the hallmark of the female climacteric, experienced to some degree by most postmenopausal women. The term hotflush is descriptive of a sudden onset of reddening of the skin over the head, neck, and chest, accompanied by a feeling of intense body heat and concluded by sometimes profuse perspiration. The duration varies from a few seconds to several minutes or, rarely, for an hour. They may occur rarely or recur every few minutes. Flushes are more frequent and severe at night (when a woman is often awakened from sleep) or during times of stress. In a cool environment, hot flushes are fewer, less intense, and shorter in duration compared with a warm environment (37). In a longitudinal follow-up study of a large number of women, fully 10% of the women experienced hot flushes
TABLE 1.1
Characteristics of Hot Flushes
Premenopausal Postmenopausal 9 Number of flushes: 9 Daily flushing: 9Average duration: 9 5 + years' duration: Other causes 9 Psychosomatic 9 Stress 9Thyroid disease ~ Pheochromocytoma 9 Carcinoid 9 Leukemia 9Cancer
15-25% of women 15-25% of women 15-20% of women 1-2 years 25% of women
4 before menopause, but in other studies, as many as 15% to 25% of premenopausal women reported hot flushes (38,39). In the Massachusetts Women's Health Study, the incidence of hot flushes increased from 10% during the premenopausal period to about 50% just after cessation of menses. By approximately 4 years after menopause, the rate of hot flushes declined to 20%. In a community-based Australian survey, 6% of premenopausal women, 26% of perimenopausal women, and 59% of postmenopausal women reported hot flushing (40). Although the flush can occur in the premenopause, it is a major feature of postmenopause, lasting in most women for 1 to 2 years but in as many as 25% for longer than 5 years. In cross-sectional surveys, up to 40% of premenopausal women and 85% of menopausal women report vasomotor complaints (39). In the United States, there is no difference in the prevalence of vasomotor complaints in surveys of black and white women (41,42). In a massive review of hot flushes, it was concluded that exact estimates on prevalence are hampered by inconsistencies and differences in methodologies, cultures, and definitions (43). The physiology of the hot flush is still not understood, but it apparently originates in the hypothalamus and is brought about by a decline in estrogen. However, not all hot flushes are caused by estrogen deficiency. Flushes and sweating can result from diseases, including pheochromocytoma, carcinoid, leukemias, pancreatic tumors, and thyroid abnormalities (44). Unfortunately, the hot flush is a relatively common psychosomatic symptom, and women often are unnecessarily treated with estrogen. When the clinical situation is not clear, estrogen deficiency as the cause of hot flushes should be documented by elevated levels of folliclestimulating hormone (FSH). The correlation between the onset of flushes and estrogen reduction is clinically supported by the effectiveness of estrogen therapy and the absence of flushes in hypoestrogen states, such as gonadal dysgenesis. Only after estrogen is administered and withdrawn do hypogonadal women experience the hot flush. Although the clinical impression that premenopausal surgical castrates suffer more severe vasomotor reactions is widely held, this is not borne out in objective study (45). Although the hot flush is the most common problem of the postmenopause, it presents no inherent health hazard. The flush is accompanied by a discrete and reliable pattern of physiologic changes (45a). The flush coincides with a surge of luteinizing hormone (LH), not FSH, and is preceded by a subjective prodromal awareness that a flush is beginning. This aura is followed by measurable increased heat over the entire body surface. The body surface experiences an increase in temperature, accompanied by changes in skin conductance and followed by a fall in core temperature--all of which can be objectively measured. In short, the flush is not a release of accumulated body heat but is a sudden inappropriate excitation of heat release mechanisms. Its relation to the LH surge
SPEROFF ET AL.
and temperature change within the brain is not understood. The observation that flushes occur after hypophysectomy indicates that the mechanism does not depend on LH release. The same hypothalamic event that causes flushes also stimulates gonadotropin-releasing hormone (GnRH) secretion and elevates LH. This probably results in changes in neurotransmitters that increase neuronal and autonomic activity. A strong interaction exists between estrogens and the serotonergic system. Serotonin levels fall with menopause. The 5-HT 2A receptor is postulated to underlie changes in thermoregulation. Stimulation of the receptor leads to changes in the set temperature, leading to autonomic changes that cool the body. Figure 1.2 illustrates the postulated mechanism of hot flashes (46). Premenopausal women experiencing hot flushes should be screened for thyroid disease and other illnesses. A comprehensive review of all possible causes is available (47). Clinicians should be sensitive to the possibility of an underlying emotional problem. Looking beyond the presenting symptoms into the patient's life can be an important service to the patient and her family that eventually will be appreciated. This is more difficult than prescribing estrogen, but confronting problems is the only hope of reaching some resolution. Prescribing estrogen inappropriately (i.e., in the presence of normal levels of gonadotropins) only temporarily postpones by a placebo response dealing with the underlying issues.
in, or external stimulus - i (coffee, anxiety, alcohol etc) I
eetrogen withdrawal ...........
,
_
,. ,,H,,
.....................
decreased 5-HT level in I
1_
reietee
of
5-HT
modullne
]
,
ii
.. "1
blockade of 5-HT1B receptors ........ (eut~,,, ,HH~
Increased 5.HT2A receptor sensitivity in hypothalamus (involved in thermoregulation)
, J~
.........
l
,,,
,~
Increased'"r'elease~ 5"HT
......, ,,
,.
,.
J
'""
I
,,,
l st'imUlationo;"5-HT2Areceptors--] in hypothalamus '" ~ ........ J
i'"
9
Cha'nge"o;"setPolnt,,', i'--] temperature ...... ,,,.
i,
, . ;, HH I autonomic reactions to cool........i down the b..ody
,,
1
HOT FLUSH FIGURE 1.2 Possible mechanism by which a hot flush is induced. Berendsen HH. The role of serotonin in hot flushes. Maturitas 2000;36: 155-164.
CHAPTER 1 The Menopause: A Signal for the Furore A striking and consistent finding in most studies dealing with menopause and hormonal therapy is a marked placebo response in a variety of symptoms, including flushing. In an English randomized, placebo-controlled study of women being treated with estrogen implants and requesting repeat implants, there was no difference in outcome in terms of psychologic and physical symptoms comparing the women who received an active implant with those receiving a placebo (48). Hormone therapy is indicated for the treatment of moderate-to-severe vasomotor symptoms associated with menopause. Therapy should be used at the lowest effective dose and for the appropriate duration (48a). Atrophic conditions include atrophy of the vaginal epithelium; formation of urethral caruncles; dyspareunia and pruritus from vulvar, introital, and vaginal atrophy; and urinary difficulties such as stress incontinence, urgency, and bacterial urethritis and cystitis. Psychologic symptoms include anxiety, mood depression, irritability, insomnia, and decreased libido. The view that menopause has a deleterious effect on mental health is not supported in the psychiatric literature or in surveys of the general population (38,39,49,50). The concept of a specific psychiatric disorder (i.e., involutional melancholia) has been abandoned. Depression is less common among middle-aged women than younger women, and the menopause cannot be linked to psychologic distress (2-8,51). The longitudinal study of premenopausal women indicates that hysterectomy with or without oophorectomy is not associated with a negative psychologic impact among middle-aged women. Longitudinal data from the Massachusetts Women's Health Study document that menopause is not associated with an increased risk of depression (52). Although women are more likely to experience depression compared with men, this sex difference begins in early adolescence, not at menopause (53). The U.S. National Health Examination Follow-up Study includes longitudinal and cross-sectional assessments of a nationally representative sample of women. This study found no evidence linking natural or surgical menopause to psychologic distress (54). The only longitudinal change was a slight decline in the prevalence of depression as women aged through the menopausal transition. Results in this study were the same for estrogen users and nonusers. A negative view of mental health at the time of menopause is not justified; many of the problems reported at menopause are caused by the vicissitudes of life (55,56). There are problems encountered in the early postmenopause that are seen frequently, but their causal relation with estrogen is unlikely. These problems include fatigue, nervousness, headaches, insomnia, depression, irritability, joint and muscle pain, dizziness, and palpitations. Men and women at this stage of life express a multitude of complaints that do not reveal a gender difference that could be explained by a hormonal cause (57).
5 Attempts to study the effects of estrogen on these problems have been hampered by the subjectivity of the complaints (i.e., high placebo responses) and the "domino effect" of what reduction of hot flushes does to the frequency of the symptoms. Using a double-blind, crossover, prospective study format, Campbell and Whitehead concluded many years ago that many symptomatic "improvements" ascribed to estrogen therapy result from relief of hot flushes~a domino effect (58). A study of 2001 women between the ages of 45 and 55 focused on the use of the health care system by women in the perimenopausal period of life (10). Health care users in this age group were frequent previous users of health care, less healthy, and had more psychosomatic symptoms and vasomotor reactions. These women were more likely to have had a significant previous adverse health history, including a history of premenstrual complaints. This study emphasized that perimenopausal women who seek health care help are different from those who do not seek help, and they often embrace hormone therapy in the hope it will solve their problems. This population is seen most often by clinicians, producing biased opinions regarding menopause among physicians. We must be careful not to generalize to the entire female population the behavior experienced by this relatively small group of women. Most importantly, perimenopausal women who present to clinicians often end up being treated with estrogen inappropriately and unnecessarily. Nevertheless, it is well established that a woman's quality of life is disrupted by vasomotor symptoms, and estrogen therapy provides impressive improvement (59,60). Once improvement has been reached, it is recommended that the decision to maintain estragon therapy be revisited periodically. If upon stopping estrogen therapy, vasomotor symptoms are not severe, therapy should be reinitiated based on other indications. Emotional stability during the perimenopausal period can be disrupted by poor sleep patterns. Hot flushing does have an adverse impact on the quality of sleep (61). Estrogen therapy improves the quality of sleep, decreasing the time to onset of sleep and increasing the rapid eye movement (REM) sleep time (59,62). Perhaps flushing may be insufficient to awaken a woman but sufficient to affect the quality of sleep, thereby diminishing the ability to handle the next day's problems and stresses. The overall quality of life reported by women can be improved by better sleep and alleviation of hot flushing. However, it is still uncertain whether estrogen treatment has an additional direct pharmacologic antidepressant effect or the mood response is an indirect benefit of relief from physical symptoms and, consequently, from improved sleep. Using various assessment tools for measuring depression, improvements with estrogen treatment have been recorded in oophorectomized women (63,64). In the large, prospective cohort study of the Rancho Bernardo retirement community, no
6 benefit could be detected in measures of depression in current users of postmenopausal estrogen compared with untreated women (65). Treated women had higher depressive symptom scores, presumably reflecting treatment selection bias; symptomatic and depressed women seek hormone therapy. Nevertheless, estrogen therapy is reported to have a more powerful impact on women's well-being beyond the relief of symptoms such as hot flushes (66). In elderly, depressed women, improvements in response to fluoxetine were enhanced by the addition of estrogen therapy (67). Sexuality is a lifelong behavior with evolving changes and development. It begins with birth (perhaps before) and ends with death. The notion that it ends with aging is inherently illogical. The need for closeness, caring, and companionship is lifelong. Old people today live longer, are healthier, and have more education and leisure time, and they have had their consciousness raised in regard to sexuality. Younger people, especially younger physicians, underrate the extent of sexual interest in older people. In a random sample of women between the ages of 50 and 82 in Madison, Wisconsin, nearly one-half reported an ongoing sexual relationship (68). In the Duke longitudinal study on aging, 70% of men in the 67 to 77 age group were sexually active, and 80% reported continuing sexual interest, whereas 50% of all older women were still interested in sex (69). In the Postmenopausal Estrogen/Progestin Interventions (PEPI) trial, 60% of women 55 to 64 years old were sexually active (70). The decline in sexual activity with aging is influenced more by culture and attitudes than by nature and physiology (or hormones). The two most important influences on older sexual interaction are the strength of a relationship and the physical condition of each partner (70). The single most significant determinant of sexual activity for older women is the unavailability of partners because of divorce and the fact that women are outliving men. Given the availability of a partner, the same general high or low rate of sexual activity can be maintained throughout life (4,71). Longitudinal studies indicate that the level of sexual activity is more stable over time than previously suggested (72-74). Individuals who are sexually active earlier in life continue to be sexually active into old age.
III. G R O W T H O F T H E OLDER POPULATION We are experiencing a relatively new phenomenon: We can expect to become old. We are on the verge of becoming a rectangular society, one of the greatest achievements of the 20th century. This is a society in which nearly all individuals survive to advanced age and then succumb rather abruptly over a narrow age range centering on 85 years. In 1000 sc, life expectancy was only 18 years. By 100 BC, the time of Caesar, it had reached 25 years. In 1900 in the United
SPEROFF ET AL.
States, life expectancy had reached only 49 years. In 2000, the average life expectancy is 79.7 years for women and 72.9 for men (75). Today, after a man reaches 65, he can expect to reach 80.5, and a woman who reaches 54 can expect to reach the age of 84.3 years. We can anticipate that eventually about two-thirds of the population will survive to 85 years or longer, and more than 90% will live past the age of 65, producing the nearly perfect rectangular society (76,77). Sweden and Switzerland are closest to this demographic composition. A good general definition of elderly is 65 and older, although it is not until age 75 that a significant proportion of older people show the characteristic decline and health problems. The elderly population is the largest contributor to illness and human need in the United States (78). There are more old people (with their greater needs) than ever before (79). In 1900, there were approximately 3 million Americans 65 years or older (about 4% of the total population). By 2030, the elderly population will reach about 57 million (17% of the total population). Population aging will soon replace population growth as the most important social problem. Two modern phenomena have influenced the rate of change. The first was the baby boom after World War II (1946 through 1964) that temporarily postponed the aging of the population but now is causing a faster aging of the general population. The second major influence has been the modern decrease in old-age mortality. Our success in postponing death has increased the upper segment of the demographic contour. By 2050, the current developed nations will be rectangular societies. By 2050, China will contain more people older than 65 years of age (270 million) than the number of people of all ages currently living in the United States. This is a worldwide development, not limited to affluent societies (80). The population of the earth will continue to grow until the year 2100 or 2150, when it is expected to stabilize at approximately 11 billion, and 95% of this growth will occur in developing countries. The poorest countries today (in Africa and Asia) account for about one-half of the global population, and in 2000, 87% of the world's population will be living in what are now called developing countries. In most developing countries, the complications associated with pregnancy, abortion, and childbirth are the first or second most common cause of death, and almost one-half of all deaths are children younger than 5 years of age. Limiting family size to two children would cut the annual number of maternal deaths by 50% and infant and child mortality by 50% (81). It is appropriate to focus attention on population control; however, even in developing countries, this will change. In 1950, only 40% of people 60 and older lived in developing countries. By 2025, more than 70% will live in those countries (Table 1.2). In 1900, older men in the United States outnumbered women by 102 to 100. In the 1980s, there were only 68 men
7
CHAPTER 1 The Menopause: A Signal for the Future TABLE 1.2
Projected Size of the Population 60 Years of Age and Older
Year
World (millions)
Developing countries (millions)
1950 1975 2000 2025
200 350 590 1100
80 (40%) a 178 (51%) 355 (60%) 792 (72%)
aThe percentageswithin parentheses compare the populations in developing countries with the world populations. From ref. 80, with permission.
for every 100 women older than 65 years. By age 85, only 45 men are alive for every 100 women. Nearly 90% of white American women can expect to live to age 70. Vital statistics data indicate that this gender difference is similar in the black and white populations in the United States (82). Approximately 55% of girls but only 35% of boys live long enough to celebrate their 85th birthday (83). Men and women reach old age with different prospects for older age, a sex differential that in part results from the sex hormone-induced differences in the cholesterol lipoprotein profile and other cardiovascular factors, producing a greater incidence of atherosclerosis and earlier death in men (84). From a public health point of view, the greatest impact on the sex differential in mortality would be gained by concentrating on lifestyle changes designed to diminish atherosclerosis in men: low-cholesterol diet, no smoking, optimal body weight, and active exercise. The death rate is higher for men at all ages. Coronary heart disease accounts for 40% of the mortality difference between men and women. Another one-third is from lung cancer, emphysema, cirrhosis, accidents, and suicides. In our society, the mortality difference between men and women is largely a difference in lifestyle. Smoking, drinking, coronaryprone behavior, and accidents account for most of the higher male mortality rate after age 65. It has been estimated that perhaps two-thirds of the difference has been because of cigarettes alone, but this results from a greater prevalence of smoking among men. Women whose smoking patterns are
TABLE 1.3 Age 55-64 65-74 >75 Total
similar to those of men have a similar increased risk of morbidity and mortality. Perhaps because more women are smoking, drinking, and working, the mortality sex difference has begun to lessen. The U.S. Census Bureau projects that the difference in life expectancy between men and women will increase until the year 2050 and then level off (79). In 2050, life expectancy will be 82 years for women and 76.7 years for men (75). There will be 33.4 million women 65 or older, compared with 22.1 million men (Table 1.3). Unmarried women will be an increasing proportion of the elderly. By 1983, 50% of American women between the ages of 65 and 74 were unmarried (some were divorced, but most were widowed), and after age 75, 77% were unmarried (85). One-half of men 85 or older live with their wives, but only 10% of elderly women live with their husbands. Because the unmarried tend to be more disadvantaged, there will be a need for more services for this segment of the elderly population. Older unmarried people are more vulnerable, demonstrating higher mortality rates and lower life satisfaction. In addition to the growing numbers of elderly people, the older population itself is getting older (86). For example, in 1984, the 65 to 74 age group was more than seven times larger than in 1900, but the 75 to 84 group was 11 times larger, and the 85 and older group was 21 times larger. The most rapid increase is expected between 2010 and 2030, when the baby boom generation hits 65. In the next century, the only age groups in the United States expected to experience significant growth will be those past the age of 55. In this elderly age group, women outnumber men by 2.6 to 1. By the year 2040, there will be 8 million to 13 million people 85 years of age or older; the estimate varies according to pessimistic to optimistic projections regarding disease prevention and treatment.
IV. T H E RECTANGULARIZATION OF LIFE The life span is the biologic limit to life, the maximal obtainable age by a member of a species. The general impression is that the human life span is increasing, but in fact
The Older U.S. Female Population
1990 (millions)
2000 (millions)
10.8 (8.6%)a 10.1 (8.1%) 7.8 (6.2%) 28.7
12.1 (9.0%) 9.8 (7.3%) 9.3 (7.0%) 31.2
2010 (millions) 17.1 (12.1%) 11.0 (7.8%) 9.8 (6.9%) 37.9
2020
(millions) 19.3 (12.9%) 15.6 (10.4%) 11.0 (7.3%) 45.9
aThe percentageswithin parentheses compare the group populations with world populations. From ref. 79, with permission.
8
the life span is fixed, and it is a biologic constant for each species (87). Differences in species' life spans argue in favor of a species-specific genetic basis for longevity. If life span were not fixed, it would mean an unlimited increase in the number of elderly. But a correct analysis of survival reveals that death converges at the same maximal age; what has changed is life expectancy, the number of years of life expected from birth. Life expectancy cannot exceed the life span, but it can closely approximate it. The number of old people will eventually hit a fixed limit, but the percentage of a typical life spent in the older years will increase. Our society has almost eliminated premature death. Diseases of the heart and the circulation and cancers are the leading causes of death. The reason for this is not an increase or an epidemic; it is a result of our success in virtually eliminating infectious diseases. The major determinant is chronic disease, affected by genetics, lifestyle, the environment, and aging itself. The major achievement left to be accomplished is in cardiovascular diseases, but even if cancer, diabetes, and all circulatory diseases were totally eliminated, life expectancy would not exceed 90 years (76). Fries describes three eras in health and disease (88). The first era existed until sometime in the early 1900s and was characterized by acute infectious diseases. The second era, highlighted by cardiovascular diseases and cancer, is beginning to fade into the third era, marked by problems of frailty (e.g., fading eyesight and hearing, impaired memory and cognitive function, decreased strength and reserve). Much of our medical approach is still based on the first era (i.e., find the disease and cure it), but we have conditions that require a combination of medical, psychologic, and social approaches. Our focus has been on age-dependent, fatal chronic diseases. The new challenge is with the nonfatal, age-dependent conditions such as Alzheimer's disease, osteoarthritis, osteoporosis, obesity, and incontinence. It can be argued that health programs in the future should be evaluated by their impact on years free of disability, rather than on mortality.
V. SUCCESSFUL AGING: THE ROLE OF PREVENTIVE HEALTH CARE Chronic illnesses are incremental in nature. The best health strategy is to change the slope, the rate at which illness develops, postponing the clinical illness, and if it is postponed long enough, effectively preventing it. There has been a profound change in public consciousness toward disease. Disease is increasingly seen as something not necessarily best treated by medication or surgery but by prevention or, more accurately, by postponement. Postponing illness is expressed by J.E Fries as the compression of morbidity (87,89). We would live relatively healthy lives and compress our illnesses into a short period just be-
SPEROFFE T
AL.
fore death. Is this change really possible? The mean national body weight has decreased by 5 pounds despite a slight increase in the national average height. There has been a decrease in atherosclerosis in the United States. Reasons include changes in the use of saturated fat, more effective detection and treatment of hypertension, increased exercise, and decreased smoking. Smoking initiation has decreased markedly in men but unfortunately has remained essentially unchanged in women. Female smokers begin smoking at a younger age. More young women (including teenagers) smoke than young men. Smoking appears to have a greater adverse effect on women compared with men (90). Physician smokers have declined from a high of 79% to a small minority (91). The greatest decrease has been among pulmonary surgeons, and the least decrease has been among proctologists. From the mid-1970s to the early 1990s, smoking among physicians declined from 18.8% to 3.3%. Unfortunately, that still amounts to approximately 18,000 physicians who smoke. Fifteen percent of Registered Nurses smoke. That is about 388,960 of the 2.6 million Registered Nurses in the United States. Twentyeight percent of Licensed Practical Nurses smoke. Smoking is declining among Registered Nurses but is higher than the Healthy People 2010 goal of 12% for the general population (91a). In the year 2005, approximately 20.9% (45.1 million) of U.S. adults were smokers. Current smoking was higher among men (23.9%) than women (18.1%). By education level, smoking prevalence was higher among adults who had earned a General Educational Development diploma (43.2%) and those with 9 to 11 years of education (32.6%) and prevalence decreased with increasing education (91b). Cigarette smoking therefore continues to be the single most preventable cause of premature death in the United States. The use of chewing tobacco, pipe smoking, and cigars contributes significantly to morbidity and mortality. Physicians and older patients may be skeptical that quitting after decades of smoking could be beneficial. In a longitudinal study of 2674 persons between the ages of 65 and 74, the mortality rates for ex-smokers were no higher than for nonsmokers (92). The effects are at least partly reversible within 1 to 5 years after quitting. Even older patients who already have coronary artery disease have improved survival after they quit smoking (93). No matter how old a person is, if he continues to smoke, he has an increased relative risk of death, but if he quits smoking, his risk of death decreases. Since 1970, the death rate from coronary heart disease has declined by approximately 50% in the United States (90). During the 20 year period of 1970 to 1990 age-adjusted coronary heart disease mortality rates decreased 3% per year. However, during the 7-year period between 1990 and 1997 coronary heart disease mortality declined at a rate 2.7% (93a).
CHAPTER I The Menopause: A Signal for the Future
Despite our progress, we must continue to exert preventive efforts on the risk factors associated with cardiovascular disease, especially obesity, hypertension, and lack of physical activity. The effort to improve the quality of life has an important value to society; it will decrease the average number of years that people are disabled and a liability. Frailty and disability have become the major health and social problems of society. Most significantly, this is a major financial challenge for health care systems and social programs. With evolution toward a rectangular society, the ratio of beneficiaries to taxpayers grows rapidly, jeopardizing the financial support for health and social programs. Compression of morbidity is at least one attractive solution to this problem.
VI. MENOPAUSE AS AN OPPORTUNITY Clinicians who interact with women at the time of menopause have a wonderful opportunity and therefore a significant obligation. Medical intervention at this point of life offers women years of benefit from preventive health care. This represents an opportunity that should be seized. It is logical to argue that health programs should be directed to the young. It makes sense to create good lifelong health behavior. Although not underrating the importance of good health habits among the young, it can be argued that the impact of teaching preventive care is more observable and more tangible at middle age. The prospects of limited mortality and the morbidity of chronic diseases are now viewed with belief, understanding, and appreciation during these older years. The chance of illness is higher, but the impact of changes in lifestyle is greater.
VII. THE MENOPAUSE AS A SIGNAL FOR THE FUTURE The menopause should remind patients and clinicians that this is a time for education. Preventive health care education is important throughout life, but at the time of menopause, a review of the major health issues can be especially rewarding. Diseases of the heart are the leading cause of death for women in the United States, followed by malignant neoplasms, cerebrovascular disease, and motor vehicle accidents. Of the 550,000 people in the United States who die each year of heart disease, 250,000 are women (94). Nearly onethird of heart disease mortality in women occurs before the age of 65. Most cardiovascular disease results from atherosclerosis in major vessels. The risk factors are the same for men and women: high blood pressure, smoking, diabetes mellitus, and
9 obesity. When controlling for these risk factors, men have a 3.5 times greater risk of developing coronary heart disease than women. Even taking into consideration the changing lifestyle of women (e.g., employment outside the home), women still maintain their advantage in terms of risk for coronary heart disease. However, with increasing age, this advantage is gradually lost, and cardiovascular disease becomes the leading cause of death for older women and men. In the past 30 years, stroke mortality has declined by 60% and mortality from coronary heart disease by 50% in the United States (90). Improvements in medical and surgical care can account for some of this decline, but 60% to 70% of the improvement results from preventive measures. Excellent data from epidemiologic studies and clinical trials demonstrate a decline in stroke and heart disease morbidity and mortality from smoking cessation, blood pressure reduction, and lowering of cholesterol (95,96). There is a strong and growing scientific basis for preventive medicine and health promotion efforts in clinical practice. Multiple observational studies in younger postmenopausal women support the conclusion that hormone therapy has cardiovascular benefit in postmenopausal women. Data from randomized control trials of older postmenopausal women have not supported the observational data. Risk factors for cardiovascular disease, such as diabetes and subclinical arthrosclerosis, increase with age. It is possible that potential cardiovascular benefits of hormone therapy may decrease the further from the onset of menopause that estrogen therapy is initiated. However, initiation or continuation of hormone therapy should not be based on primary or secondary prevention of cardiovascular disease (96a). Osteoporosis is a major global public health problem, and it is epidemic in the United States, affecting more than 20 million individuals (97). The increase in osteoporotic fractures in the developed world is partly caused by an increase in the elderly population. A comparison of bone densities in proximal femur bones in specimens from a period of over 200 years suggested that women lose more bone today, perhaps because of less physical activity and less parity (98). Other contributing factors include a dietary decrease in calcium and an earlier and greater loss of bone because of the impact of smoking. Our Stone Age predecessors consumed a diet high in calcium, mostly from vegetable sources (99). However, the impact of the tremendous increase in the elderly population throughout the world cannot be underrated. Because of this demographic change, the number of hip fractures occurring in the world each year will increase approximately sixfold from 1990 to 2050, and the proportion occurring in Europe and North America will fall from 50% to 25% as the numbers of old people in developing countries increase (100). The onset of spinal bone loss begins in the twenties, but the overall change is small until menopause. Bone density in the femur peaks in the middle to late twenties and begins to
10
SPEROFF ET AL.
decrease around age 30. In general, trabecular bone resorption and formation occur four to eight times as fast as cortical bone. Beyond age 30, trabecular resorption begins to exceed formation by about 0.7% per year. This adverse relationship accelerates after menopause, and up to 5% of trabecular bone and 1% to 1.5% of total bone mass loss per year occurs after menopause. This accelerated loss continues for 10 to 15 years, after which bone loss is considerably diminished but continues as the aging-related loss. For the first 20 years after cessation of menses, menopause-related bone loss results in a 50% reduction in trabecular bone and a 30% reduction in cortical bone (101,102). Estrogen therapy provides a 50% to 60% decrease in fractures of the arm and hip (103-105), and when estrogen is supplemented with calcium, an 80% reduction in vertebral compression fractures occurs (106). This reduction is seen primarily in patients who have taken estrogen for more than 5 years (107,108). Protection against fractures wanes with age, and long-term estrogen use is necessary to maximally reduce the risk of fracture after age 75. Because most osteoporotic fractures occur late in life, women and clinicians must understand that the short-term use of estrogen immediately after menopause cannot be expected to protect against fractures in the seventh and eighth decades of life. Some long-term protection is achieved with 7 to 10 years of estrogen therapy after menopause, but the impact is minimal after age 75 (109). In a prospective cohort study of women 65 years of age or older, in the women who had stopped using estrogen and in those who were older than 75 and had stopped using estrogen even if they had used estrogen for more than 10 years, there was no substantial effect on the risk for fractures (110). The effective impact of estrogen requires initiation within 5 years of menopause and use extending into the elderly years. The protective effect of estrogen rapidly dissipates after treatment is stopped, because estrogen withdrawal is followed by rapid bone loss. In the 3- to 5-year period after loss of estrogen, whether after menopause or after cessation of estrogen therapy, there is accelerated loss of bone (111-113). The benefits of estrogen-progestin treatment suggest that other treatment regimens should be considered as well. One indication for estrogen treatment for osteoporosis includes concomitant treatment of vasomotor symptom. Maximal protection against osteoporotic fractures therefore requires lifelong therapy, and even some long-term protection requires 10 or more years of treatment.
VIII. CONCLUSION The menopause has been overly laden with negative symbolism. Many of the behavioral complaints at the time of menopause, however, can be explained by psychologic and sociocultural influences. That is not to say that important interactions among biology, psychology, and culture do not
occur, but it is time to stress the normalcy of this life event. Menopausal women do not suffer from a hormone deficiency disease. Hormone replacement therapy should be viewed as specific treatment for symptoms in the short term. Therapy remains effective for treating women with vasomotor symptoms and vaginal atrophy and selected women with osteoporosis. The benefits and risks of hormone therapy should be balanced in each individual. Part of the reason for our negative stereotypical views of menopause is that the initial characterization of menopause was derived from women presenting with physical and psychologic difficulties. The variability in menopausal reactions makes the cross-sectional study design particularly unsuitable. More and larger longitudinal studies are needed to document what is normal and the variations around normal. It is important to educate women and clinicians about the normal events of this time. Changes in menstrual function are not symbols of some ominous change. There are good physiologic reasons for changing menstrual function, and understanding the physiology can do much to reinforce a healthy, normal attitude. The menopause serves a useful purpose. This physiologic event brings clinicians and patients together, providing the opportunity to enroll patients in a preventive health care program. Contrary to popular opinion, menopause is not a signal of impending decline, but rather a wonderful phenomenon that can signal the start of something positive, such as a good health program. Rather than being a lightning rod for social and personal problems, menopause can be a signal for the future.
References 1. Farnham AM. Uterine disease as a factor in the production of insanity. Alienst Neurologist 1887;8:532. 2. HaUstrom T, Samuelsson S. Mental health in the climacteric: the longitudinal study of women in Gothenburg. Acta Obstet Gynecol &and Supp11985;130:13-18. 3. McKinlay SM, McKinlay JB. The impact of menopause and social factors on health. In: Hammond CB, Haseltine FP, SchiffI, eds. Menopause: evaluation, treatment, and health concerns. New York: Alan R Liss, 1989:137-161. 4. Matthews KA, Wing RR, KuUer LH, et al. Influences of natural menopause on psychological characteristics and symptoms of middleaged healthy women. J Consult Clin Psycho11990;58:345-351. 5. Koster A. Change-of-fife anticipations, attitudes, and experiences among middle-aged Danish women. Health Care Women 1991;12:1-13. 6. Holte A. Influences of natural menopause on health complaints: a prospective study of healthy Norwegian women. Maturitas 1992;14:127. 7. Kaufert PA, Gilbert P, Tate R. The Manitoba Project: a re-examination of the link between menopause and depression. Maturitas 1992;14:143. 8. Dennerstein L, Smith AMA, Morse C, et al. Menopausal symptoms in Australian women. MedJAust 1993;159:232. 9. Avis NE, McKinlay SM. A longitudinal analysis of women's attitudes toward the menopause: results from the Massachusetts Women's Health Study. Maturitas 1991;13:65-79.
CHAPTER 1 The Menopause: A Signal for the Future 9a. Avis NE, Ory M, Matthews KA, et al. Health-related quality of life in a multiethnic sample of middle-aged women. Medical Care 2003;41:1262-1276. 9b. American College of Obstetricians and Gynecologists Women's Health Care Physicians. Executive summary, hormone therapy. Obstet Gyneco12004;104:1S-4S. 10. Morse CA, Smith A, Dennerstein L, et al. The treatment-seeking woman at menopause. Maturitas 1994;18:161-173. 11. Defey D, Storch E, Cardozo S, Diaz O, Fernandez G. The menopause: women's psychology and health care. Sac Sci Med 1996;42:1447-1456. 11a. Soules MR, Sherman S, Parrott E, et al. Executive summary: Stages of Reproductive Aging Workshop (STRAW). Fertil Steril 2001;76:874-878. 12. McKinlay SM, Bigano NL, McKinlay JB. Smoking and age at menopause. Ann Intern Med 1985;103:350. 13. Gold EB, Bromberger J, Crawford S, et al. Factors associated with age at natural menopause in a multiethnic sample of midlife women. Am J Epidemio12001;153:865-874. 14. Treolar AE. Menarche, menopause and intervening fecundability. Hum Bid 1974;46:89. 15. Van Noord PAR, Dubas JS, Dorland M, Boersma H, te Velde E. Age at natural menopause in a population-based screening cohort: the role of menarche, fecundity, and lifestyle factors. Fertil Steril 1997;68:95-102. 16. Coulam CB, Adamsen SC, Annegers JF. Incidence of premature ovarian failure. Obstet Gyneco11986;67:604. 17. Torgerson DJ, Avenell A, Russell IT, Reid DM. Factors associated with onset of menopause in women aged 45-49. Maturitas 1994;19:83. 18. Torgerson DJ, Thomas RE, Campbell MK, Reid DM. Alcohol consumption and age of maternal menopause are associated with menopause onset. Maturitas 1997;26:21-25. 19. Cramer DW, Xu H, Harlow BL. Family history as a predictor of early menopause. Fertil Steri11995;64:740-745. 20. Baird DD, Tylavsky FA, Anderson JIB. Do vegetarians have earlier menopause? Am J Epidemio11988;128:907. 21. MacMahon B, Worcester J. Age at menopause: U.S. 1960-62. Vital Health Stat 1966;19:1. 22. Katsouyanni K, Boyle P, Trichopoulos D. Diet and urine estrogens among postmenopausal women. Oncology 1991;48:490-494. 23. Gapstur SM, Potter JD, Sellers TA, Folsom AR. Increased risk of breast cancer with alcohol consumption in postmenopausal women. AmJEpidemio11992; 136:1221-1231. 24. Gavaler JS, Van Thiel DB. The association between moderate alcoholic beverage consumption and serum estradiol and testosterone levels in normal postmenopausal women: relationship to the literature. Alcohol Clin Exp Res 1992; 16:87-92. 25. Holbrook TC, Barrett-Connor E. A prospective study of alcohol consumption and bone mineral density. Br MedJ 1993; 306:1506-1509. 26. Ginsburg EL, Mello NK, Mendelson JH, et al. Effects of alcohol ingestion on estrogens in postmenopausal women. JAMA 1996; 276:1747-1751. 27. Brambila DJ, McKinlay SM, Johannes CB. Defining the perimenopause for application in epidemiologic investigations. Am J Epidemiol 1994;140:1091-1105. 28. Bromberger JT, Matthews KA, Kuller LH, et al. Prospective study of the determinants of age at menopause. Am JEpidemio11997;145:124-133. 29. Cassou B, Derriennic F, Monfort C, Dell'Accio P, Touranchet A. Risk factors of early menopause in two generations of gainfully employed French women. Maturitas 1997;26:165-174. 30. Gonzales OF, Villena A. Age at menopause in Central Andean Peruvian women. Menopause 1997;4:32-38. 31. Willett W, Stampfer MJ, Bain C, et al. Cigarette smoking, relative weight, and menopause. Am J Epidemio11983;117:651-658. 32. Midgette AS, Baron JA. Cigarette smoking and the risk of natural menopause. Epidemiology 1990;1:474.
11 33. Siddle N, Sarrel P, Whitehead M. The effect of hysterectomy on the age at ovarian failure: identification of a subgroup of women with premature loss of ovarian function and literature review. Fertil Steril 1987;47:94. 34. Amundsen DW, Diers CL. The age of menopause in classical Greece and Rome. Hum Bid 1970;42:79. 35. Amundsen DW, Diers CL. The age of menopause in medieval Europe. Hum Bid 1973;45:605. 36. Frommer DL. Changing age at menopause. Br MedJ 1964;2a:349. 37. Kronnenberg F, Barnard RM. Modulation of menopausal hot flashes by ambient temperature. J Therm Bid 1992;17:43-49. 38. Hunter M. The South-East England longitudinal study of the climacteric and postmenopause. Maturitas 1992; 14:17-26. 39. Oldenhave A, Jaszmann LJB, Haspels AA, Everaerd WTAM. Impact of climacteric on well-being. Am J Obstet Gyneco11993;168:772-780. 40. Guthrie JR, Dennerstein I, Hopper JL, Burger HG. Hot flushes, menstrual status, and hormone levels in a population-based sample of midlife women. Obstet Gyneco11996;88:437-442. 41. Schwing PJ, Hulka BS, Harlow SD. Risk factors for menopausal hot flashes. Obstet Gyneco11994;84:29-34. 42. Pham KT, Grisso JA, Freeman EW. Ovarian aging and hormone replacement therapy: hormonal levels,symptoms, and attitudes of AfricanAmerican and white women. J Gen Intern Med 1997;12:230-236. 43. Kronnenberg F. Hot flashes: epidemiology and physiology. Ann N Y Acad Sci 1990;592:52-86. 44. Wilkin JR. Flushing reactions: consequences and mechanisms. Ann Intern Med 1981;95:468-476. 45. Aksel S, Schomberg MY, Tyrey I, Hammond CB. Vasomotor symptoms, serum estrogens and gonadotropin levels in surgical menopause. d m J Obstet Gyneco11976;126:165. 45a. Swartzman LC, Edelberg R, Kemmann E. The menopausal hot flush: symptom reports and concomitant physiological changes. J Behav Med 1990;13:15. 46. Berendsen HH. The role of serotonin in hot flushes. Maturitas 2000;36:155-164. 47. Mohyi D, Tabassi K, Simon I. Differential diagnosis of hot flashes. Maturitas 1997;27:203-214. 48. Pearce J, Hawton K, Blake F, et al. Psychological effects of continuation versus discontinuation of hormone replacement therapy by estrogen implants: a placebo-controlled study.J PsychosomRes 1997;42:177-186. 48a. The North American Menopause Society. Estrogen and progestogen use in peri- and postmenopausal women: September 2003 position statement of The North American Menopause Society. Menopause 2003;10:497-506. 49. Ballinger CB. Psychiatric aspects of the menopause. Br J Psychiatry 1990;156:773-787. 50. Schmidt PJ, Rubinow DR. Menopause-related affective disorders: a justification for further study. Am J Psychiatry 1991;148:844-854. 51. Gath D, Osborn M, Bungay G, et al. Psychiatric disorder and gynaecological symptoms in middle aged women: a community survey. Br MedJ 1987;294:213-218. 52. Avis NE, Brambilla D, McKinlay SM, Vass K. A longitudinal analysis of the association between menopause and depression: results from the Massachusetts Women's Health Study. Ann EDidemio11994;4:214-220. 53. Kessler RC, McGonagle KA, Swartz M, Blazer DG, Nelson CB. Sex and depression in the National Comorbidity Survey 1: lifetime prevalence, chronicity and recurrence. JAffect Disord 1993;29:85. 54. Busch CM, Zonderman AB, Costa PT Jr. Menopausal transition and psychological distress in a nationally representative sample: is menopause associated with psychological distress?JAging Health 1994;6:209-228. 55. Dennerstein I, Smith AMA, Morse C. Psychological well-being, midlife and the menopause. Maturitas 1994;20:1. 56. Mitchell ES, Woods NE Symptom experiences of midlife women: observations from the Seattle midlife women's health study. Maturitas 1996;25:1-10.
12 57. Van Hall EV, Verdel M, Van Der Velden I. "Perimenopausar' complaints in women and men: a comparative study. J Womens Health 1994;3:45-49. 58. Campbell S, Whitehead M. Estrogen therapy and the menopausal syndrome. Clin Obstet Gyneco11977; 4:31-47. 59. Wiklund 1, KarlbergJ, Mattsson I.-A. Quality of life ofpostmenopausal women on a regimen of transdermal estradiol therapy: a double-blind, placebo-controlled study.Am J Obstet Gyneco11993;168:824-830. 60. Daly E, Gray A, Barlow D, et al. Measuring the impact of menopausal symptoms on quality of life. Br MedJ 1993;307:836. 61. Woodward S, Freedman RR. The thermoregulatory effects of menopausal hot flashes on sleep. Sleep 1994;17:497-501. 62. Schiff I, Regestein Q~ Tulchinsky D, Ryan KI. Effects of estrogens on sleep and psychological state of hypogonadal women. JAMA 1979;242:2405-2407. 63. Dennerstein I, Burrows OD, Hyman GJ, Wood C. Hormone therapy and affect. Maturitas 1979;1:247-254. 64. Sherwin BB. Affective changes with estrogen and androgen replacement therapy in surgically menopausal women. J Affect Disord 1988;14:177-187. 65. Palinkas LA, Barrett-Connor E. Estrogen use and depressive symptoms in postmenopausal women. Obstet Gyneco11992;80:30-36. 66. Limouzin-Lamothe M-A, Mairon N, Joyce CRB, Le Gal M. Quality of life after the menopause: influence of hormonal replacement therapy. Am J Obstet Gyneco11994;170:618-624. 67. Schneider LS, Small GW, Hamilton SH, et al. Estrogen replacement and response to fluoxetine in a multicenter geriatric depression trial. A m J Geriatr Psychiatry 1997;5:97-106. 68. Traupman J, Eckels E, Hatfield E. Intimacy in older women's lives. Gerontologist 1982;2:493. 69. Pfeiffer E, Verwoerdt A, Davis GC. Sexual behavior in middle life. Am J Psychiatry 1972;128:1262. 70. Greendale GA, Hogan P, Shumaker S, for the Postmenopausal Estrogen/ Progestin Interventions (PEPI) Trial Investigators. Sexual functioning in postmenopausal women: the Postmenopausal Estrogen/Progesfin Interventions (PEPI) trial.J Womens Health 1996;5:445-456. 71. Martin CE. Factors affecting sexual functioning in 60-79-year-old married males. Arch Sex Behav 1981;10:399. 72. George LK, Weiler SI. Sexuality in middle and late life. Arch Gen Psychiatry 1981;38:919. 73. White CB. Sexual interest, attitudes, knowledge, and sexual history in relation to sexual behavior in the institutionalized aged. Arch Sex Behav 1982;11:11. 74. Renshaw DC. Sex, intimacy, and the older woman. J Womens Health 1983;8:43-54. 75. Annual report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds. Washington, DC: U.S. Government Printing Office, 1995. 76. Olshansky SJ, Carnes BA, Cassel C. In search of Methuselah: estimating the upper limits to human longevity. Science 1990;250:634. 77. Olshansky SJ, Carnes BA, Cassel C. The aging of the human species. Sci Am April 1993;268:46-52. 78. Rowe JW, Grossman E, Bond E. Academic geriatrics for the year 2000: an Institute of Medicine report. N EnglJMed 1987;316:1425. 79. U.S. Bureau of the Census. Current population reports:projections of the population of the United States: 1977 to 2050. Washington, DC: U.S. Government Printing Office, 1993. 80. Diczfalusy E. Menopause, developing countries and the 21st century. Acta Obstet Gynecol Scand Supp11986;134:45. 81. Duke RC, Speidel JL. Women's reproductive health: a chronic crisis. JAMA 1991;266:1846. 82. Miles TP, Bernard MA. Morbidity, disability, and health status of black American elderly: a new look at the oldest-old. J Am Geriatr Soc 1992;40:1047.
SPEROFF ET AL.
83. Day JC, U.S. Bureau of the Census. Current population reports. Population projections of the United States, by age, sex, race, and Hispanic origin: 1993 to 2050. Washington, DC: U.S. Government Printing Office, 1993. 84. Hazzard WR. Biological basis of the sex differential in longevity.JAm Geriatr Sac 1986;34:455. 85. Keith PM. The social context and resources of the unmarried in old age. IntJAging Hum Dev 1986;23:81. 86. U.S. Bureau of the Census. Current population reports, special studies." sixty-five plus in America. Washington, DC: U.S. Government Printing Office, 1993. 87. Fries JF, Crapo LM. Vitality and aging. San Francisco: WH Freeman, 1981. 88. Fries JE The sunny side of aging. JAMA 1990;263:2354. 89. Fries JE Strategies for reduction of morbidity. Am J Clin Nutr 1992;55:1257S. 90. Davis DL, Dinse GE, Hoel DG. Decreasing cardiovascular disease and increasing cancer among whites in the United States from 1973 through 1987. JAMA 1994;271:431. 91. Nelson DE, Giovino GA, Emont SL, et al. Trends in cigarette smoking among U.S. physicians and nurses. JAMA 1994;271:1273. 91a. Tobacco Free Nurses Initiative. The facts on nurses and smoking. Los Angeles: School of Nursing, University of California, Los Angeles, 2006. 91b. Centers for Disease Control and Prevention. Tobacco use among adults--United States, 2005. M M W R Morb Mortal Wkly Rep 2006;55:1145-1148. 92. Jajich CL, Ostfeld AM, Freeman DH Jr. Smoking and coronary heart disease mortality in the elderly.JAMA 1984;252:2831. 93. Hermanson B, Omenn GS, Kronmal RA, Gersh BJ. Beneficial sixyear outcome of smoking cessation in older men and women with coronary artery disease: results from the CASS registry. NEnglJMed 1988;319:1365. 93a. Cooper R, Cutler J, Desvigne-Nickens P, et al. Trends and disparities in coronary heart disease, stroke, and other cardiovascular diseases in the United States: Findings of the National Conference on Cardiovascular Disease Prevention. Circulation 2000;102:3137-3147. 94. Wenger NK, Speroff I, Packard B. Cardiovascular health and disease in women. N E n g l J M e d 1993;329:247. 95. Vartiainen E, Puska P, Pekkanen J, Tuomilehto J, Jousilahti E Changes in risk factors explains changes in mortality from ischaemic heart disease in Finland. Br MedJ 1994; 309:23. 96. Vartiainen E, Sarti C, Tuomilehto J, Kuulasmaa K. Do changes in cardiovascular risk factors explain changes in mortality from stroke in Finland? Br MedJ 1995; 310:901. 96a. Mosca L, Collins P, Herrington DM, et al. Hormone replacement therapy and cardiovascular disease: a statement for healthcare Professionals from the American Heart Association. Circulation 2001;104:499-503. 97. Dempster DW, Lindsay R. Pathogenesis of osteoporosis. Lancet 1993;341:797. 98. Lees B, Molleson T, Arnett TR, Stevenson JC. Differences in proximal femur bone density over two centuries. Lancet 1993;341:673-675. 99. Eaton SB, Nelson DA. Calcium in evolutionary perspective. Am J Clin Nutr 1991;54:281S. 100. Cooper C, Campion G, Melton LJ III. Hip fractures in the elderly: a world-wide projection. OsteoporosInt 1992;2:285. 101. Ettinger B. Prevention of osteoporosis: treatment of estradiol deficiency. Obstet Gynecolo11988;72:125. 102. Lindsay R. Prevention and treatment of osteoporosis. Lancet 1993;341:801. 103. Weiss NC, Ure CL, Ballard JH, Williams AR, Dating JR. Decreased risk of fractures of the hip and lower forearm with postmenopausal use of estrogen. N EnglJ Med 1980;303:1195.
13
CHAPTER 1 The Menopause: A Signal for the Future 104. Ettinger B, Genant HK, Cann CE. Long-term estrogen replacement therapy prevents bone loss and fractures.Ann Intern Med 1985;102:319. 105. Kiel DP, Felson DT, Anderson JJ, Wilson PW, Moskpwitz MA. Hip fracture and the use of estrogen in postmenopausal women. The Framingham Study. N E n g l J M e d 1987; 317:1169. 106. Riggs BL, Seeman E, Hodgson SF, Taves DR, O'Fallon WM. Effect of the fluoride/calcium regimen on vertebral fracture occurrence in postmenopausal osteoporosis. N EnglJ IVied 1982; 306:446. 107. O.uigley MET, Martin PI, Burnier AM, Brooks E Estrogen therapy arrests bone loss in elderly women. Am J Obstet Gynecol 1987;156:1516. 108. Lafferty FW, Fiske ME. Postmenopausal estrogen replacement: a long-term cohort study. Am J Med 1994;97:66. 109. Felson PT, Zhang Y, Hannan MT, et al. The effect of postmenopausal estrogen therapy on bone density in elderly women. N EnglJ Med 1993;329:1141-1146. 110. Cauley JA, Seeley DG, Enbsrud K, et al. Estrogen replacement therapy and fractures in older women. Ann Intern Med 1995;122:9-16. 111. Lindsay R, MacLean A, Kraszewski A, Clark AC, Garwood J. Bone response to termination of estrogen treatment. Lancet 1978;1:1325. 112. Horsman A, Nordin BEC, Crilly RG. Effect on bone of withdrawal of estrogen therapy. Lancet 1979;2:33. 113. Christiansen C, Christiansen MS, Transbol IB. Bone mass in postmenopausal women after withdrawal of oestrogen/gestagen replacement therapy. Lancet 1981;1:459.
Additional References American College of Obstetricians and Gynecologists, Women's Health Care Physicians. Executive summary: hormone therapy. Obstet Gynecol 2004;104:1S-4S. Avis NE, Ory M, Matthews KA, et al. Health-related quality of life in a multiethnic sample of middle-aged women. Med Care 2003;41: 1262-1276. Berendsen HH. The role of serotonin in hot flushes. Maturitas 2000;36:155-164. Gold EB, Bromberger J, Crawford S, et al. Factors associated with age at natural menopause in a multiethnic sample of midlife women. Am J Epidemio12001;153:865-874. Mendelsohn ME, Karas RH. The time has come to stop letting the HERS tale wag dogma. Circulation 2001;104:2256-2259. Mosca L, Collins P, Herrington DM, et al. Hormone replacement therapy and cardiovascular disease: a statement for healthcare professionals from the American Heart Association. Circulation 2001;104:499-503. Soules MR, Sherman S, Parrott E, et al. Executive summary: Stages of Reproductive Aging Workshop (STRAW). Ferti! Steri! 2001;76:874-878. The North American Menopause Society. Estrogen and progestogen use in peri- and postmenopausal women: September 2003 position statement of The North American Menopause Society. Menopause 2003;10:497-506.
This Page Intentionally Left Blank
SECTION I
Basics to Enhance Our Understanding The aim of this book, and its prior editions, is to provide a comprehensive basis for the treatment and care of the postmenopausal woman. Accordingly, both basic science information and clinical data will be presented for all areas that are relevant for the health care of postmenopausal women. In this first section of the book, several chapters are devoted to basic information that will help the reader to understand the changes that occur around the time of menopause and thereafter, as the basis for choosing a treatment, if necessary. I am extremely gratified that in this section as well as in other areas of the book, the chapters are written by true experts who are highly acknowledged for their particular expertise in the areas on which they write. Chapter 2, by Sherk and McDonnell, provides an update on the mechanisms of action of estrogen and progesterone. This story has become extremely complex but has been explained in a very clear way. Particularly important concepts include nongenomic effects, receptor isoforms that affect agonist/antagonist actions, and the role of coactivators and corepressors in various tissues. The third chapter, by Simpson, describes the genetics of ovarian failure and the genes that are involved for normal function, as well as those genes that may be implicated in premature ovarian failure, which will be covered in the next section. Chapter 4, by Erickson and Chang, provides a description of the basic biology of ovarian failure. This offers us a deeper understanding of the fertility concerns of older women prior to menopause, as well as the endocrinologic changes that occur at this time. Burger and Teede next describe the endocrine changes around the perimenopause--that is, the first series of changes that occur leading up to menopause, which will set the stage for our understanding of hormonal and other changes to follow. In Chapter 6 Gold and Greendale provide a very scholarly and up-to-date review of the epidemiology of the menopause and point out areas that are still not clear, stressing the importance of the need for more longitudinal studies.
This Page Intentionally Left Blank
_~HAPTER z
M ole cular P h arm acology
of Estrogen and Progesterone Receptors ANDREA
B.
SHERK
Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, NC 27710
DONALD P. M c D O N N E L L
Departmentof Pharmacology and Cancer Biology,Duke UniversityMedical Center, Durham, NC 27710
II. E S T R O G E N A N D P R O G E S T E R O N E RECEPTORS
I. I N T R O D U C T I O N The steroid hormones estrogen and progesterone are low-molecular weight, lipophilic hormones that, through their action as modulators of distinct signal transduction pathways, are involved in the regulation of reproductive function (1,2). These hormones have also been shown to be important regulators in bone, the cardiovascular system, and the central nervous system (3-5). Despite their different roles in these systems, however, it has become apparent that estrogens and progestins are mechanistically similar (6). Insights gleaned from the study of each hormone, therefore, have advanced our understanding of this class of molecules as a whole. This review highlights some of the recent mechanistic discoveries that have occurred in the field and explores the subsequent changes in our understanding of the pharmacology of this class of steroid hormones. TREATMENT OF THE POSTMENOPAUSAL WOMAN
Estrogen and progesterone bind to high-affinity intracellular receptors, which act as ligand-activated transcription factors. The estrogen receptor (ER) and progesterone receptor (PR) cDNAs have been cloned and used to develop specific ligand-responsive transcription systems in heterologous cells, permitting the use of reverse genetic approaches to define the functional domains within each of the receptors (Fig. 2.1) (6). The largest domain (approximately 300 amino acids) that is responsible for ligand binding is located at the carboxyl terminus of each receptor. Crystallographic analysis of the agonist-bound forms of ERs and PRs has indicated that this domain consists of 12 short o~-helical structures that fold to provide a complex ligand-binding pocket (7,8). The ligand-binding domain also contains sequences that facilitate receptor homodimerization and permit the interaction of
17
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
18
SHERK AND McDONNELL
FIGURE 2.1 The domain structures of the estrogen and progesterone receptors are similar. AF-1, activation function 1; AF-2, activation function 2.
apo-receptors with inhibitory heat-shock proteins. An activation function (AF-2) required for receptor transcriptional activity is also contained within the ligand-binding domain (9,10). An additional activation function (AF-1) is located within the amino terminus of each receptor (11). The DNA-binding domain (DBD) is a short region (approximately 70 amino acids) located in the center of the receptor protein (12). This permits the receptor to bind as a dimer to target genes. Within the D B D there are nine cysteine residues, eight of which can chelate two zinc atoms, thereby forming two fingerlike structures that allow the receptor to interact with D N A (13). All the information required to permit target gene identification by ligandactivated ERs and PRs is contained within this region.
III. ESTABLISHED MODELS OF ESTROGEN AND PROGESTERONE ACTION The steroid hormones estrogen and progesterone are representative members of a larger family of steroid hormones, all of which appear to share a common mechanism of action. It is generally believed that steroid hormones enter cells from the bloodstream by simple passive diffusion, exhibiting activity only in cells in which they encounter a specific high-affinity receptor protein (14). These receptor proteins are transcriptionaUy inactive in the absence of ligand, sequestered in a large oligomeric heat-shock protein complex within target cells (15). On binding ligand, however, the receptors undergo an activating conformational change that promotes the dissociation of inhibitory proteins (16). This event permits the formation of receptor homodimers that are capable of interacting with specific high-affinity DNA-response elements located within the regulatory regions of target genes (17). The DNA-bound receptor can then exert a positive or negative influence on target gene transcription (Fig. 2.2).
FIGURE 2.2 Established models of estrogen and progesterone action. The classic models of estrogen and progesterone action suggested that, in the absence ofligand, the steroid receptor (SR) exists in target cells in an inactive form. On binding an agonist, the receptor would undergo an activating transformation event that displaces inhibitory heat-shock proteins (HSP) and facilitates the interaction of the receptor with specific DNA steroid response elements (SRE) within target gene promoters.The activated receptor dimer could then interact with a complex of proteins and positively or negatively regulate target gene transcription. In this model, the role of the agonist is that of a "switch"that merely converts the ER or PR from an inactive to an active form. Thus, when corrected for affinity, all agonists would be qualitatively the same and evoke the same phenotypic response. By inference, antagonists, compounds that oppose the actions of agonists, would competitivelybind to their cognate receptors and freeze them in an inactive form. As with agonists, this model predicted that all antagonists are qualitatively the same. Within the confines of this classic model it was difficult to explain the molecular pharmacology of the known ER and PR agonists and antagonists. TA, transcriptional apparatus.
CHAPTER 2 Molecular Pharmacology of Estrogen and Progesterone Receptors
In the classic models of steroid hormone action, it was proposed that progestins and estrogens function merely as switches that, on binding to their cognate receptor, permit conversion of the receptor, in an all or nothing manner, from an inactive to an active state (18). This implied that ER and PR pharmacology was very simple, and that when corrected for affinity, all progestins and estrogens were qualitatively the same. Furthermore, it suggested that antihormones (antagonists) function simply as competitive inhibitors of agonist binding, freezing the target receptor in an inactive state within the cell (see Fig. 2.2). Under most experimental conditions, this simple model was sufficient to explain the observed biology of known PR and ER agonists and antagonists. However, the results of several clinical studies of estrogens and antiestrogens suggested that the pharmacology of ER was far more complex than predicted from these classic models of hormone action. Studies probing the complex pharmacology of the antiestrogen tamoxifen, for instance, have been very informative with respect to understanding the inadequacies of the classic model. Tamoxifen is widely used as a breast cancer chemotherapeutic and as a breast cancer chemopreventive in high-risk patients (19,20). In ER-positive breast cancers, tamoxifen opposes the mitogenic action of estrogen(s) by binding to the receptor and competitively blocking agonist access. However, it has become clear that tamoxifen is not a pure antagonist, because in some target organs, it can exhibit estrogen-like actMty. This is most apparent in both the skeletal system, where tamoxifen, like estrogen, increases lumbar spine bone mineral density, and the cardiovascular system, where both tamoxifen and estradiol have been shown to decrease low-density lipoprotein (LDL) cholesterol (21,22). These in vivo properties of tamoxifen led to its being reclassified as a selective estrogen receptor modulator (SERM), rather than an antagonist. Similarly, selective progesterone receptor modulators (SPRMs), which function in a tissueselective manner, have been discovered. The most promising of these is asoprisnil, which is currently being developed for the treatment of women with uterine fibroids and dysfunctional uterine bleeding. Asoprisnil exhibits antiproliferative effects on the uterine endometrium, while maintaining ovarian estrogen production, similar to progestins, but lacks progestin-induced irregular endometrial bleeding and the mitogenic effects of progestins in the breast (23). Therefore, the ability of ligands to exert differential effects through their cognate receptors depending on the tissue environment is a common theme across the steroid receptor family. The observation that different ligand-receptor complexes were not recognized in the same manner in all cells was at odds with the established models of ER and PR action. The observed pharmacology of tamoxifen and similar compounds begged a reevaluation of the classic model of ER and PR action and initiated the search for the cellular systems that enable ER/PR-ligand complexes to manifest different biologies in different cells.
19
IV. NONCLASSICAL MODELS OF ESTROGEN AND PROGESTERONE RECEPTOR ACTION It has become increasingly clear that estrogens and progestins utilize alternative pathways, in addition to the classical mode of action, to exert their biological effects. It is possible that differential activation of these pathways in various tissues might contribute to the tissue-specific activity of selective receptor modulators. As described earlier, the classical model of action of steroid hormones consists of hormone binding to its receptor and subsequent association of the receptor dimer with DNA response elements in the promoter regions of target genes (Fig. 2.3, E). However, ligand-bound ER and PR are also able to regulate genes without directly binding to their promoter regions. For example, ligand-associated ER is able to interact with and modulate the activity of the NFKB and Spl transcription factors, as well as Fos and Jun heterodimers at AP-1 binding sites (see Fig. 2.3, F and G) (24-26). Consequently, genes that lack hormone response elements can be regulated by steroid hormones due to the indirect recruitment of ligand-activated receptors to promoter regions via interactions with other transcription factors. Interestingly, tamoxifen can activate AP-1 target genes in endometrial cells but not in breast tumor cells, reflecting its growth effects on these tissues (24). Therefore, the ability of SERMs to modulate ER action at AP-1 target genes in a cell-selective manner might be a key determinant in the tissue-specific growth effects of these ligands. However, the extent to which these in vitro observations reflect activities that occur in vivo remains to be determined. Estrogens and progestins can also elicit rapid responses in cells, which occur so quickly that they are clearly not mediated by the transcriptional activity of ER and PR and are thus described as "nongenomic." For example, estrogen and progesterone can increase intracellular levels of calcium and cyclic adenosine monophosphate (cAMP), second messengers for specific signal transduction pathways, and can initiate mitogen-activated protein (MAP) kinase phosphorylation cascades (27-29). Although evidence for these rapid effects has been mounting for decades, the mechanism by which they occur is just beginning to be elucidated. It is unclear what type of receptor mediates these hormonal effects, although several models have been proposed. There is evidence that these "nongenomic" effects occur via activation of a subpopulation of the classical ER and PR or a splice variant that has been localized to the plasma membrane, either directly (through palmitoylation of the receptor) or indirectly, through interactions with other membrane-associated proteins (such as caveolin) (see Fig. 2.3, A and B) (30-32). Indeed, the classical PR and ER have both been shown to associate with the Src
20
SHERK AND McDONNELL
FIGURE 2.3
Estrogen and progesterone exert their biological effects via both genomic and nongenomic mechanisms. Nongenomic effects of estrogen have been reported to occur through the classical nuclear receptor that is localized to the membrane directly through palmitoylation of the receptor (B), through interactions with other proteins, such as caveolin (A) or MNAR and c-Src (C). GPR30 is a G-protein-coupled receptor that has been reported to bind estrogen, resulting in intracellular calcium mobilization (D). There is evidence that GPR30 might be localized to the endoplasmic reticulum, rather than the plasma membrane. E: The classical mode of action of estrogen, in which the nuclear ER dimer binds to an estrogen response element (ERE) within the promoter regions of target genes. The nuclear ER can also modify transcription of target genes by being indirectly tethered to DNA through additional transcription factors, such as Fos and Jun on AP1 sites (F) or NFKB (G). Similar mechanisms are involved in progesterone action. MNAR, modulator of nongenomic activity of estrogen receptor; CoA, coactivator.
family of nonreceptor tyrosine kinases at the plasma membrane, either directly (PR) or through the interaction with the adaptor protein MNAR (ER) (see Fig. 2.3, C) (33,34). Ligand-bound ER and PR can activate Src, initiating a MAP kinase phosphorylation cascade, which ultimately influences ER and PR transcriptional activity (26,35). Therefore, although these responses are termed "nongenomic," they do eventually modulate the transcriptional activity or "genomic" response of ER and PR. Recently, a family of transmembrane G protein coupled receptors (GPCRs) that bind progesterone was identified, first in the sea trout, and then in the mouse and human
(36,37). Subsequently, another GPCR, GPR30, was shown to elicit intracellular calcium mobilization in response to estrogen binding (38,39). Surprisingly, GPR30 appears to be localized to the endoplasmic reticulum, rather than the plasma membrane (see Fig. 2.3, D) (39). It remains to be determined if these are the membrane receptors that mediate the nongenomic effects of estrogen and progesterone. The contribution of these nonclassical mechanisms of action to estrogen and progesterone physiology is yet to be determined. However, one of the most widely studied nongenomic effects of estrogen is stimulation of nitric
CHAPTER 2 Molecular Pharmacology of Estrogen and Progesterone Receptors
oxide production in endothelial cells, which may contribute to estrogen's antiatherogenic and antiatherosclerotic effects (26). It is possible that differential activation of membrane receptors and nuclear receptors is partially responsible for the tissue-specific effects of selective receptor modulators.
V. ESTROGEN AND PROGESTERONE RECEPTOR ISOFORMS AND SUBTYPES Another mechanism to explain the cell-selective action of steroid receptor ligands is the likelihood that they may activate different receptor isoforms (derived from the same gene) or subtypes (derived from similar genes). This concept has been well established for the ci- and [3-adrenergic systems, where it has been shown that different receptor subtypes have distinct ligand preferences and that selectivity can be explained by differences in the expression of these subtypes. Until recently, the parallel between this system and that of the nuclear receptors was not obvious. However, the identification and characterization of ER and PR isoforms and subtypes has shed new light on this issue (Fig. 2.4).
A. Progesterone Receptor Isoforms The progesterone receptor was the first receptor for which bona fide isoforms were shown to exist. Human PRs can exist within target cells in either of two distinct forms, hPR-A (94 kDa) or hPR-B (114 kDa) (40). These pro-
Estrogen Receptor SubtyDe~ 1
hER-a
DBo I
NH2 I
.....I............. 1
hER-I]
NH2
595
.........
.......
I
530
1
IDBD
!
I
Progesterone Receptor Isoform~ 933
1
I
769
FIGURE 2.4 At least two distinct forms of the estrogen and progesterone receptors exist in target cells. DBD, DNA-binding domain; LBD, ligand-binding domain.
21
teins, differing only in that the hPR-B isoform contains an additional 164-amino acid extension at its amino terminus, are produced from distinct mRNAs that are derived from different promoters within the same gene (9). In most progesterone-responsive tissues these two receptor isoforms are expressed in equimolar amounts. This apparent 1:1 relationship is so widespread that until about 10 years ago, the hPR-A isoform was thought to be merely an artifact derived from hPR-B by proteolysis during biochemical fractionation. It has now been established that these two proteins are produced in a deliberate manner by the cell, and that they are not functionally equivalent (40-42). The first evidence in support of this hypothesis came following the cloning and subsequent functional analysis of the chicken progesterone receptor (cPR) cDNA (43). Specifically, on expression in heterologous cells, it was found that although the A and B forms of cPRs display identical ligand-binding preferences, they activate different target genes (43). It was subsequently shown that the amino-terminal sequences, which distinguish cPR-B from cPR-A, are important in determining target gene selectivity. This concept was reaffirmed when the cloned hPR-B and hPR-A were analyzed in a similar manner (41). Studies in T47D breast cancer cells in which one or the other isoforms were overexpressed provided evidence that suggests hPR-A and hPR-B regulate a distinct subset of genes. Interestingly, the target genes identified in this manner showed very little overlap. The number of hPR-B regulated genes was far greater than those modulated by hPR-A (44). Further analysis has revealed that hPR-A functions as a ligand-dependent transdominant modulator of the transcriptional activity of hPR-B, the ability of hPR-B to activate target gene transcription being influenced by the cellular concentration of hPR-A (41,45). Surprisingly, it was also determined that ligand-activated hPR-A can inhibit the transcriptional activity of agonistactivated ERs, androgen receptors (ARs), and mineralocorticoid receptors (MRs) (41). Analysis of mice in which PR-A or PR-B was genetically disrupted has demonstrated that the two isoforms have distinct roles in progesterone biology. Mice lacking PR-A display infertility, severe endometrial hyperplasia, and anovulation but exhibit normal breast morphogenesis. On the other hand, PR-B knockout mice are fertile with a normal uterus but exhibit reduced pregnancy-induced mammary ductal morphogenesis. Collectively, these findings suggest that PR-A activation is responsible for progesterone's reproductive functions and, specifically, the "antiestrogenic" actions of progestins in the uterus, whereas PR-B is required for progesteroneinduced, pregnancy-related mammary morphogenesis (46). Thus, by virtue of having two functionally distinct receptor isoforms, a single hormone such as progesterone can have completely different functions in target cells.
22
SHERK AND McDONNELL
B. Estrogen Receptor Subtypes The identification of functionally distinct PR isoforms introduced a new dimension to progesterone action, although it was not until a second estrogen receptor was cloned in 1995 that the general significance of isoforms (or subtypes) in steroid receptor signaling was established (47). Unlike the case of PRs, ERe~ and ER[3 are encoded by different genes, and although they share significant amino acid homology in their ligand-binding domains, they are not pharmacologically equivalent. Both receptors bind the endogenous estrogen, 1713-estradiol, with equivalent affinity (48). However, when binding analysis was extended to additional compounds, significant differences in ligand preferences were noted. The biological and pharmacological consequences of these differences remain to be determined. Although the discovery of ER[3 has occurred relatively recently, significant progress has been made in elucidating its role in estrogen signaling. It has been determined that the expression pattern of ER[3 does not mirror that of ERa (49,50). Expression of both isoforms is found in some tissues, whereas ER[3 alone occurs in others, such as the lung, the urogenital tract, and the colon (48). Two major approaches have been undertaken to delineate the relative contribution of these receptor subtypes to estrogen biology. The first approach involved the generation of mice whose ERos, ER[3, or both of the receptors have been genetically disrupted. The phenotype of the ER[3 knockout mice is different from that of ERa knockout mice, reflecting the distinct roles of these two receptors in the endocrinology of estrogen (51). The other approach utilized ER-subtypeselective ligands, which has corroborated observations made in the knockout mice and revealed novel functions of ER[3. Data from both these sources suggest ERc~ is responsible for the bone-protective, uterotrophic, and mammotrophic effects of estrogens. Estrogen acting through ER[3, on the other hand, has an antiproliferative, prodifferentiative effect in the breast, as well as in the prostate (51,52). Recently, work utilizing the ER[3-selective ligand, ERB-041, has uncovered a role for ER[3 in the immune system. Studies with ERB-041 in rodent models indicate that an ER[3-selective ligand might be beneficial for the treatment of chronic inflammatory diseases, such as irritable bowel syndrome and arthritis (53). In fact, ERB-041 is currently being developed for the treatment of rheumatoid arthritis and endometriosis. Therefore, estrogen can have very different, even opposing, effects by acting through two different receptor isotypes. The identification of estrogen and progesterone receptor isoforms and subtypes and the definition of specific functions that they modulate have introduced a new dimension in steroid hormone action. Understanding the regulatory
mechanisms that control the expression levels of the individual forms of each receptor is likely to provide novel targets for pharmaceutical intervention.
VI. REGULATION OF ESTROGEN AND PROGESTERONE RECEPTOR FUNCTION BY LIGANDS The finding that ERs and PRs could exist in multiple forms within target cells suggests that some of the tissueselective actions of their cognate agonists and antagonists can be explained by their ability to regulate differentially the action of one specific receptor isoform or subtype. It has become apparent, however, from the study of antiestrogens that the identical ligand operating through the same receptor can manifest different biological activities in different target cells (54). In breast tissue, for instance, where ERo~ predominates, all the known antiestrogens oppose the mitogenic actions of estrogen (55). In the endometrium, however, where ERc~ also predominates, it has been found that tamoxifen functions as a partial estrogen mimetic (56,57), whereas compounds such as raloxifene, GW5638, and ICI182,780 function as pure antiestrogens. Thus, the same compounds, acting through ERoL, manifest different biological activities in the breast and the endometrium. This finding is not in agreement with the classic models of ER action that indicate that ligands basically fall into two classes, agonists and antagonists. This paradox has been the subject of much investigation, leading to the observation that different compounds can induce different alterations in ER structure and that not all structures are functionally identical. It is implied, therefore (discussed in more detail later), that the cell possesses the cellular machinery to distinguish between these dissimilar complexes and that the identification and characterization of the specific components of these systems are the keys to the development of the next generation of tissue-selective ER and PR modulators. Much of what we know about the effect of ligands on steroid receptor structure has come from studies of different ER-ligand complexes. Initially, using differential sensitivity to proteases, it was demonstrated that the hormone-binding domain within the ER adopts different shapes on binding estradiol and tamoxifen and that these structures are dissimilar to that of the apo-receptor (54,58). Thus, receptor conformation is affected by the nature of the bound ligand. This relationship between structure and function was later confirmed by the observation that agonists and antagonists induce different alterations in PR structure (59,60). Further analysis has revealed that the majority of the structural changes that occur in the PR are located at the extreme carboxyl tail of the receptor and that removal of the
CHAPTER 2 Molecular Pharmacology of Estrogen and Progesterone Receptors carboxyl-terminal 42 amino acids of hPR-B permits the antagonist RU486 to function as an agonist (59). Interestingly, a similarly positioned domain enables the ER to discriminate between different compounds, and, not surprisingly, removal of 35 amino acids from the C-terminal tail of the ER abolishes its ability to distinguish between agonists and antagonists (61). The recent determination of the crystal structures of the ER-estradiol and ER-tamoxifen complexes confirmed the important role of the carboxyl tail in determining the pharmacology of steroid receptors (7,62,63). This new structural information has also revealed that agonist activation of the ER permits the formation of a unique surface (or pocket) on the receptor that allows it to interact with coactivator proteins. In the presence of the SERM tamoxifen, however, the carboxyl tail of the ER is positioned in such a manner that it occludes this coactivator binding pocket, preventing a productive association with coactivators that utilize this surface to interact with the receptor. In addition to tamoxifen, several additional SERMs manifest distinct activities in vivo. One of these compounds, raloxifene, has been approved as a SERM for the treatment and prevention of osteoporosis (64). This compound distinguishes itself from tamoxifen in that it does not exhibit estrogenic action in the postmenopausal endometrium (65,66). Although clearly different biologically, the crystal structures of the ER-tamoxifen and ERraloxifene complexes were shown to be virtually indistinguishable. Although these results appear to be at odds with the hypothesis that links receptor structure to function, some data from our group have reconciled these potential discrepancies. We have used phage display technology to identify small peptides, the ability of which to bind ERs is affected differentially by the nature of the ligand bound to the receptor (Fig. 2.5) (67-69). The rationale behind this approach is that because of the vast complexity of the peptides available in these libraries, it may be possible to find peptides that have the ability to distinguish between two very similar receptor-ligand complexes. This approach has led to the identification of a series of high-affinity peptide probes that, in addition to being able to distinguish between ERestradiol and ER-tamoxifen complexes, are also able to distinguish among several different E R - S E R M complexes. This approach has been extended to the study of the PR, and it was similarly observed that various PR ligands manifest different biologies in different cells, allowing the identification of peptide probes whose interaction with the receptor is influenced by the nature of the ligand bound to the PR. All these findings establish a firm relationship between the structure of a receptor-ligand complex and biological activity and suggest that novel ER and PR ligands with unique pharmaceutical properties may be developed by exploiting this observation.
23
FIGURE 2.5 Fingerprintingthe surfaces of different ER-ligand complexes using conformation-sensitivepeptide probes.A: Random peptide libraries were constructed in an M13 bacteriophage; each of the resulting bacteriophages expressed a unique random peptide on its surface pilus. Screens were subsequentlyperformed to identifyspecificpeptides (bacteriophage) whose interaction with the ER was influenced by the nature of the bound ligand. The bacteriophage identified in this manner were used to develop an enzyme-linked immunoassayto monitor changes that occur in the ER on its interaction with different ligands. Specifically, a biotinylated estrogen responseelement (ERE) was used to immobilizerecombinant ERs on streptavidin-coated plates. After incubation of this complexwith the ligand to be tested, to each well was added an aliquot of a different class of ER-interacting bacteriophage. Binding of the bacteriophage was assessed enzymatically using an anti-M13 antibody coupled to horseradish peroxidase (HRP). B: Fingerprint analysisof ER conformationin the presence of saturating concentrations of the indicated ligands; the resulting complexes were incubated with aliquots of bacteriophage expressing eight different peptides. Tam, tamoxifen; DES, diethylstilbestrol;Prog, progesterone.This figure has been publishedpreviouslyin a similarform (68) and is reproduced and presented here with permission (Copyright 1999 National Academyof Sciences, U.S.A.).
24
SHERK AND McDONNELL
VII. ESTROGEN AND PROGESTERONE RECEPTOR ASSOCIATED PROTEINS The estrogen and progesterone receptors are liganddependent transcription factors that, on activation by ligands, associate with specific DNA response elements located within the regulatory regions of target genes (14). The DNA-bound receptor can then positively or negatively influence gene transcription by recruiting a complex of proteins that can modulate chromatin structure and either facilitate or prevent the interaction of RNA polymerase II with the promoter. In the past few years, it has become clear that at least two functional classes of proteins are involved in recognizing the activated receptor. One class includes components of the basic transcription machinery, the general transcription factors, whose expression levels are generally invariant from cell to cell. The second class of proteins, cofactors, is not a part of the general transcription machinery and can exert either a positive or negative influence on SR transcriptional activity (70). Those cofactors that interact with agonist-activated SRs have been called coactivators, whereas those that interact with apo-receptors or antagonistactivated receptors have been called corepressors. Interestingly, it has become apparent that differences in the relative expression levels of coactivators and corepressors can have a profound effect on the pharmacology of estrogen and progesterone receptor ligands (42,70,71).
A. Coactivator Proteins One of the most well-characterized coactivator proteins, steroid receptor coactivator 1 (SRC-1), was identified in a yeast two-hybrid screen as a protein that interacted with agonist-activated PR (73). Subsequently, this protein has been shown to also interact with estrogen, glucocorticoid, and androgen receptors. It appears that SRC-1 increases target gene transcription by linking the hormone-activated receptor with the general transcription machinery, stabilizing the transcription preinitiation complex, and nucleating a large complex of proteins that together have the ability to acetylate histones and facilitate chromatin decondensation (74,75). SRC-1 belongs to a family of p160 proteins that also includes SRC-2/GRIP1 and SRC-3/AIB1. The contribution of each of the p160 family members to the biology of estrogens and progestins has been dissected through the use of knockout mice. Although there is partial functional redundancy among the family members, the phenotypes of the three p160 single knockout strains are clearly different from one another. Mice lacking SRC-1 exhibit normal growth and fertility but demonstrate partial resistance to hormones. For example, these mice display reduced uterine
growth, uterine decidual response, mammary ductal branching and alveolar development and have an estrogen-resistant skeletal phenotype. Female SRC-3 knockout mice exhibit abnormal reproductive development and function. Estrogen levels are reduced in these mice, which might account for delayed pubertal development and reduced mammary gland growth. In addition, estrogen- and progesterone-stimulated mammary gland alveolar development is attenuated in SRC-3 knockout mice. Although both male and female SRC-2 knockout mice have impaired fertility, this is probably not due to an impairment of ER or PR signaling pathways (76). Collectively, these in vivo murine models suggest SRC-1 and SRC-3 are important for normal estrogen and progesterone biology. Interestingly, SRC-3 is overexpressed in many breast cancers, and overexpression in transgenic animals results in mammary hyperplasia and mammary tumor development, suggesting it is a potent oncogene, and implicates a role for SRC-3 overexpression in estrogenmediated carcinogenesis (77). In addition to the p160 family, more than 200 additional coactivators have been identified and characterized. These include proteins that possess enzymatic activities or recruit protein complexes that exhibit enzymatic activities (such as kinase, acetyl- or methyltransferase, ubiquitin- or SUMO-ligase activities) that modify the receptor, other associated proteins, or histones to modulate gene expression (78). Cumulatively, these studies have revealed that (1) the expression levels of these coactivators vary from cell to cell, (2) coactivators demonstrate specific receptor preferences, (3) a given receptor can interact with more than one type of coactivator, and (4) the conformation of the receptor adopted in the presence of a specific ligand can determine which coactivators are engaged. These findings strongly support the hypothesis that differential cofactor expression is the most important determinant of estrogen and progesterone receptor pharmacology. With the discovery of the nuclear receptor coactivators and the characterization of their biochemical properties has come a new understanding of the mechanism by which differently conformed receptor-ligand complexes are recognized in the cell. The studies that have been performed with ERs are the most informative. As described previously, it has been shown that ERs in the presence of estradiol undergo a conformational change that allows the presentation of surfaces on the receptors, permitting them to interact with coactivators. Because estradiol induces the same conformational change within ERs in all cells, the phenotypic consequence of the exposure of a cell to estradiol will depend on the properties of the coactivators expressed in that cell. The situation gets more complicated, however, when considering the role of coactivators in mediating the cellselective action of SERMs such as tamoxifen. It has been shown that the tamoxifen-induced conformational change within the ER does not allow the coactivator binding pocket
CHAPTER 2 Molecular Pharmacology of Estrogen and Progesterone Receptors
to form properly, preventing or hindering the interaction of those coactivators that require the coactivator binding pocket in order to interact with the ER (67). In cells in which this type of coactivator is important, therefore, tamoxifen can function as an antagonist. It is becoming clear, however, that not all coactivators rely on the coactivator binding pocket to the same degree. Thus, the relative agonist-antagonist activity of tamoxifen depends on the ability of the tamoxifen-ER complex to engage a compatible coactivator in target cells (Fig. 2.6) (54). As the repertoire of cofactors increases, we are likely to find that targeting specific cofactor-receptor complexes will yield pharmaceuticals that manifest their activities in a cell- or tissue-restricted manner. Many coactivators for ER and PR are RNA-binding proteins or proteins involved in RNA processing and maturation. The significance of this was not appreciated until recently. Previously, it was believed that transcription and RNA maturation were two separate processes. However, recent evidence suggests that transcription and RNA maturation are functionally coupled. Estrogen and progesterone can alter the alternative splicing decisions of transcripts synthesized from hormone-sensitive promoters. This coordinate control of transcription and splicing is dependent upon "coupling" coactivator proteins (79). Thus, not only can ER and PR influence the transcription rate of target genes, they can affect the exon content of the resultant
FIGURE 2.6 A molecular explanation for the tissue-selective agonist/antagonist activity of the SERM tamoxifen. The estrogen receptor undergoes different conformational changes on binding agonist, antagonist, or SERMs. The agonist-induced conformation allows the ER to interact with any coactivator protein expressed in target cells, and thus it can activate transcription. The tamoxifen-induced conformational change, on the other hand, is more restrictive and allows the interaction of the ER with only a subset of available coactivators. In those cells in which the tamoxifen-ER complex can engage a coactivator, this compound can manifest agonist activity. In other contexts, tamoxifen interacts with corepressors and functions as an antagonist. ER, estrogen receptor; ERE, estrogen response elements; CoA, coactivator; CoR, corepressor.
25
transcripts. This introduces another level of complexity into estrogen and progesterone action.
B. Corepressor Proteins Two nuclear corepressor proteins that appear to be important in ER and PR pharmacology were initially identified. These proteins, NCoR and SMRT, originally found as proteins that interact with DNA-bound thyroid hormone or retinoid X receptors, repress basal transcription in the absence of hormone (80,81). However, it has now been shown that these proteins can interact with either PR or ER, either in the absence of hormone or in the presence of antagonists (71,72). Under these conditions, the corepressors nucleate a large protein complex, which represses target gene transcription by deacetylating histones and facilitating chromatin condensation. There is little evidence suggesting that NCoR or SMRT are involved in normal estrogen or progesterone physiology, as mice that are genetically null for NCoR are embryonically lethal and there is no reported SMRT knockout mouse model (82). However, the physiological importance of corepressors in ER pharmacology was suggested by the studies of Lavinsky and coworkers, who found that passage of breast tumors in mice from a state of tamoxifen sensitivity to an insensitive state was accompanied by a decrease in the expression level of NCoR (83). A similar process, if occurring in humans, could explain how cells become resistant to the antiestrogenic actions of tamoxifen. Interestingly, tamoxifen functions as a strong agonist in cells lacking NCoR, suggesting NCoR is important for the antagonistic actions of tamoxifen and other SERMs (82). These data suggest that NCoR and perhaps SMRT are more important in the pharmacological actions of ER antagonists and SERMs, rather than normal ER physiology. Another recently described corepressor, REA (repressor of estrogen receptor activity), is able to associate with and repress the activity of both antagonist- and agonist-bound receptors, suggesting that it might act in a physiological manner to attenuate estrogen action (84). Notably, mice expressing only one copy of the REA gene exhibit an enhanced response to estrogen at the genomic level, resulting in increased uterine growth and epithelial hyperproliferation (85). This strongly suggests that REA functions as a negative regulator of estrogen-dependent signaling.
VIII. AN UPDATED MODEL OF ESTROGEN AND PROGESTERONE RECEPTOR ACTION On ligand binding, the activated receptor (ER or PR) can interact as a dimer with specific DNA response elements within target genes. It is now apparent that the conformation
SH~RK AND McDONNELL
26 of the resulting receptor is influenced by the nature of the b o u n d ligand and that the shape of the resulting receptorligand complex is a critical determinant of w h e t h e r it can activate transcription. In the presence of a full agonist, the conformation adopted by the receptor facilitates the displacement of corepressor proteins and recruitment of coactivator proteins with a concomitant increase in target gene transcription. Pure antagonists, on the other hand, drive their cognate receptor into a conformation that favors corepressor interaction. T h e activity o f mixed agonists/antagonists appears to relate to their ability to alter receptor conformation differentially and the ability of corepressor and coactivator proteins within a given target cell to recognize these complexes. Clearly, the classic models of estrogen and progesterone action need to be updated to accommodate the insights that have emerged from the study of the genetics and m o lecular pharmacology of these two receptors.
References 1. Clark JH, Markaverich BM. In: The physiology of reproduction, E Knobil, J Neill, eds. New York: Raven Press, 1988:675-724. 2. Clarke CL, Sutherland RL. Progestin regulation of cellular proliferation. Endocrine Rev 1990;11:266 - 301. 3. Colditz GA, Hankinson SE, Hunter DJ, et al. The use of estrogens and progestins and the risk of breast cancer in postmenopausal women. N E n g l J M e d 1995;332:1589-1593. 4. Barzel US. Estrogens in the prevention and treatment of postmenopausal osteoporosis. Am J Med 1988;85:847- 850. 5. Grodstein F, Stampfer MJ, Manson JE, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N EnglJ IVied 1996;335:453-461. 6. McDonnell DP, Vegeto E, Gleeson MAG. Nuclear hormone receptors as targets for new drug discovery. Biol Technology 1993;11:1256-1260. 7. Brzozowski AM, Pike AC, Dauter Z, et al. Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 1997;389:753-758. 8. Williams SP, Sigler PB. Atomic structure of progesterone complexed with its receptor. Nature 1998;393:392-396. 9. Giangrande PH, McDonnell DE The A and B isoforms of the human progesterone receptor: two functionally different transcription factors encoded by a single gene. Rec Prog Hormone Res 1999;54:291-314. 10. Tzukerman MT, Esty A, Santiso-Mere D, et al. Human estrogen receptor transactivational capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrino11994;8:21- 30. 11. Gronemeyer H. Control of transcriptional activation by steroid hormone. FASEB J 1992;6:2524- 2529. 12. Kumar V, Green S, Stack G, et al. Functional domains of the human estrogen receptor. Cell 1987;51:941-951. 13. Schwabe JWR, Chapman L, Finch JT, Rhodes D. The crystal structure of the estrogen receptor DNA-binding domain bound to DNA: how receptors discriminate between their response elements. Cell 1993; 75:567-578. 14. O'Malley BW, Tsai SY, Bagchi M, et al. Molecular mechanism of action of a steroid hormone receptor. Rec Prog Hormone Res 1991;47: 1-26. 15. Bagchi MK, Tsai SY, Tsai MJ, O'Malley BW. Identification of a functional intermediate in receptor activation in progesterone-dependent cell-free transcription. Nature 1990;345:547-550.
16. Allan GF, Tsai SY, Tsai M-J, O'Malley BW. Ligand-dependent conformational changes in the progesterone receptor are necessary for events that follow DNA binding. Proc Natl Acad Sci USA 1992; 89:11750-11754. 17. Bagchi MK, Elliston JF, Tsai SY, et al. Steroid hormone-dependent interaction of human progesterone receptor with its target enhancer element. Mol Endocrino11988;2:1221-1229. 18. Clark JH, Peck EJ. Female sex steroids: receptors andfunction. New York: Springer-Verlag, 1979. 19. Jordan VC. Tamoxifen: a guide for clinicians and patients. Huntington, NY: PRR, Inc, 1996. 20. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study.JNatl Cancer Inst 1998;90:1371-1388. 21. Love RR, Wiebe DA, Newcombe PA, et al. Effects of tamoxifen on cardiovascular risk factors in postmenopausal women. Ann Int Med 1991;115:860-864. 22. Love RR, Mazess RB, Barden HS, et al. Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. N Engl J M e d 1992;326:852-856. 23. Chwalisz K, Perez, MC, DeManno D, et al. Selective progesterone receptor modulator development and use in the treatment of leiomyomata and endometriosis. Endocrine Rev 2005;26:423-438. 24. Kushner PJ, Agard DA, Greene GL, et al. Estrogen receptor pathways to AP-1. J Steroid Biochem Mol Bid 2000;74:311-317. 25. Porter W, Saville B, Hoivik D, Safe S. Functional synergy between the transcription factor Spl and the estrogen receptor. ]Viol Endocrinol 1997;11:1569-1580. 26. Bj6rnstr6m L, Sj6berg M. Mechanisms of estrogen receptor signaling: convergence of genomic and nongenomic actions on target genes. Mol Endocrino12005;19:833- 842. 27. L6sel R, Wehling M. Nongenomic actions of steroid hormones. Nat Rev Mol Cell Bid 2003;4:46-56. 28. Improta-Brears T, Whorton AR, Codazzi F, et al. Estrogen-induced activation of mitogen-activated protein kinase requires mobilization of intracellular calcium. Proc NatlAcad Sci USA 1999;96:4686-4691. 29. Aronica SM, Kraus WL, Katzenellenbogen BS. Estrogen action via the cAMP signaling pathway: stimulation of adenylate cyclase and cAMP-regulated gene transcription. Proc Natl Acad Sci USA 1994; 91:8517-8521. 30. Li L, Haynes MP, Bender JR. Plasma membrane localization and function of the estrogen receptor a variant (ER46) in human endothelial cells. Proc NatlAcad Sci USA 2000;97:5930-5935. 31. Acconcia F, Ascenzi P, Fabozzi G, Visca P, Marino, M. S-palmityolation modulates human estrogen receptor-el functions. Biochem Biophys Res Comm 2004;316:878-883. 32. Razandi M, Oh P, Pedram A, Schnitzer J, Levin ER. ERs associate with and regulate the production of caveolin: implications for signaling and cellular actions. Mol Endocrino12002;16:100-115. 33. Boonyaratanakornldt V, Scott MP, Ribon V, et al. Progesterone receptor contains a proline-rich motif that directly interacts with SH3 domains and activates c-Src family tyrosine ldnases. Mol Cell 2001;8:269-280. 34. Wong C-W, McNally C, Nickbarg E, Komm BS, Cheslds BJ. Estrogen receptor-interacting protein that modulates its nongenomic activitycrosstalk with Src/Erk phosphorylation cascade. Proc NatlAcad Sci USA 2002;99:14783-14788. 35. Lange CA. Making sense of cross-talk between steroid hormone receptors and intracellular signaling pathways: who will have the last word? Mol Endocrino12004;18:269-278. 36. Zhu Y, Bond J, Thomas P. Identification, classification, and partial characterization of genes in humans and other vertebrates homologous to a fish membrane progestin receptor. Proc Natl Acad Sci USA 2003;100:2237- 2242.
CIJAVTv.~ 2 Molecular Pharmacology of Estrogen and Progesterone Receptors 37. Zhu Y, Rice CD, Pang Y, Pace M, Thomas R Cloning, expression, and characterization of a membrane progestin receptor and evidence it is an intermediary in meiotic maturation of fish oocytes. Proc Natl Acad Sci USA 2003;100:2231-2236. 38. Thomas P, Pang Y, Filardo EJ, Dong J. Identity of an estrogen membrane receptor coupled to a G protein in human breast cancer cells. Endocrinology 2005;146:624-632. 39. Revankar CM, Cimino DF, Sklar LA, Arterburn JB, Prossnitz ER. A transmembrane intracellular estrogen receptor mediates rapid cell signaling. Science 2005;307:1625-1630. 40. Lessey BA, Alexander PS, Horwitz KB. The subunit characterization of human breast cancer progesterone receptors: characterization by chromatography and photoaffinity labelling. Endocrinology 1983;112: 1267-1274. 41. Vegeto E, Shahbaz MM, Wen DX, et al. Human progesterone receptor A form is a cell- and promoter-specific repressor of human progesterone receptor B function. Mol Endocrino11993;7:1244-1255. 42. Jackson TA, Richer JK, Bain DL, et al. The partial agonist activity of antagonist-occupied steroid receptors is controlled by a novel hinge domain-binding coactivator L7/SPA and the corepressors NCoR or SMRT. ]ViolEndocrino11997;11:693- 705. 43. Conneely OM, Dobson ADW, Huckaby C, et al. Structure-function relationships of the chicken progesterone receptor. In: Steroid hormone action, G Ringold, ed. New York: Alan R. Liss, 1988:227-236. 44. Richer JK, Jacobsen BM, Manning NG, et al. Differential gene regulation by the two progesterone receptor isoforms in human breast cancer cells.J Biol Cbem 2002;277:5209-5218. 45. Wen DX, Xu Y-F, Mais DE, Goldman ME, McDonnell DR The A and B isoforms of the human progesterone receptor operate through distinct signaling pathways within target cells. Mol Cell Biol 1994;14: 8356-8364. 46. Li X, Lonard DM, O'Malley BW. A contemporary understanding of progesterone receptor function. Mech Ageing Dev 2004;125:669-678. 47. Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson J-A. Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc NatlAcad Sci USA 1996;93:5925-5930. 48. Kuiper GGJM, Carlsson B, Grandien K, et al. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors ci and [3. Endocrinology 1997;138:863-870. 49. Shughrue PJ, Komm B, Merchenthaler I. The distribution of estrogen receptor-[3 mRNA in the rat hypothalamus. Steroids 1996;61: 678-681. 50. Shughrue PJ, Lane MV, Merchenthaler I. Comparative distribution of estrogen receptor-cl and -[3 mRNA in the rat central nervous system.J Comp Neuro11997;388:507-525. 51. Hewitt SC, Harrell JC, Korach KS. Lessons in estrogen biology from knockout and transgenic animals. Ann Rev Pbysio12005;67:285-308. 52. Koehler KF, Helguero LA, Haldos6n L-A, Warner M, Gustafsson, J-A. Reflections on the discovery and significance of estrogen receptor [3. Endocrine Rev 2005;26:465-478. 53. Harris HA, Albert LM, Leathurby Y, et al. Evaluation of an estrogen receptor-beta agonist in animal models of human disease. Endocrinology 2003;144:4241-4249. 54. McDonnell DP, Clemm DL, Hermann T, Goldman ME, Pike JW. Analysis of estrogen receptor function in vitro reveals three distinct classes of antiestrogens. Mol Endocrino11995;9:659-668. 55. Jordan VC. The strategic use of antiestrogen to control the development and growth of breast cancer. Cancer 1992;70:977-982. 56. Kilackey MA, Hakes TB, Pierce VK. Endometrial adenocarcinoma in breast cancer patients receiving antiestrogens. Cancer Treat Rep 1985;69:237-238. 57. Gottardis MM, Robinson SP, Satyaswaroop PG, Jordan VC. Contrasting actions of tamoxifen on endometrial and breast tumor growth in the athymic mouse. Cancer Res 1988;48:812-815.
27
58. Allan GF, Leng X, Tsai SY, et al. Hormone and antihormone induce distinct conformational changes which are central to steroid receptor activation. J Biol Chem 1992;267:19513-19520. 59. Vegeto E, Allan GF, Schrader WT, et al. The mechanism of RU486 antagonism is dependent on the conformation of the carboxy-terminal tail of the human progesterone receptor. Cell 1992;69:703-713. 60. Wagner BL, Pollio G, Leonhardt S, et al. 16 alpha-substituted analogs of the antiprogestin RU486 induce a unique conformation in the human progesterone receptor resulting in mixed agonist activity. Proc Natl Acad Sci USA 1996;93:8739-8744. 61. Mahfoudi A, Roulet E, Dauvois S, Parker MG, Wahli W. Specific mutations in the estrogen receptor change the properties of antiestrogens to full agonists. Proc NatlAcad Sci USA 1995;92:4206-4210. 62. Tanenbaum DM, Wang Y, Williams SP, Sigler PB. Crystallographic comparison of the estrogen and progesterone receptor's ligand binding domains. Proc NatlAcad Sci USA 1998;95:5998-6003. 63. Shiau AK, Barstad D, Loria PM, et al. The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 1998;95:927-937. 64. Delmas PD, Bjarnason NH, Mitlak BH, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. N Engl J Med 1997; 337:1641-1647. 65. Sato M, Rippy MK, Bryant HU. Raloxifene, tamoxifen, nafoxidine, or estrogen effects on reproductive and nonreproductive tissues in ovariectomized rats. FASEBJ 1996;10:905-912. 66. Ashby J, Odum J, Foster JR. Activity of raloxifene in immature and ovariectomized rat uterotrophic assays. Reg Toxicol Pharmacol 1997;25:226-231. 67. Norris JD, Paige LA, Christensen DJ, et al. Peptide antagonists of the human estrogen receptor. Science 1999;285:744-746. 68. Paige LA, Christensen DJ, Gron H, et al. Estrogen receptor (ER) modulators each induce distinct conformational changes in ERcx and ER[3. Proc NatlAcad Sci USA 1999;96:3999-4004. 69. Wijayaratne AL, Nagel SC, Paige LA, et al. Comparative analyses of the mechanistic differences among antiestrogens. Endocrinology 1999; 140:5828 - 5840. 70. Horwitz KB, Jackson TA, Bain DL, et al. Nuclear receptor coactivators and corepressors. Mol Endocrino11996; 10:1167 - 1177. 71. Smith CL, Nawaz Z, O'Malley BW. Coactivator and corepressor regulation of the agonist/antagonist activity of the mixed antiestrogen, 4-hydroxytamoxifen. Mol Endocrino11997;11:65 7-666. 72. Wagner BL, Norris JD, Knotts TA, Weigel NL, McDonnell DR The nuclear corepressors NCoR and SMRT are key regulators of both ligandand 8-bromo-cyclic AMP-dependent transcriptional activity of the human progesterone receptor. Mol Cell Bio11998;18:1369-1378. 73. Onate SA, Tsai S, Tsai M-J, O'Malley BW. Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 1995;270:1354-1357. 74. Spencer TE, Jenster G, Burcin MM, et al. Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 1997;389:194-198. 75. Shibata H, Spencer TE, Onate SA, et al. Role of co-activators and corepressors in the mechanism of steroid/thyroid receptor action. Rec Prog Horm Res 1997;52:141-165. 76. Xu J, Li Q:. Review of the in vivo functions of the p160 steroid receptor coactivator family. Mol Endocrino12003;17:1681 - 1692. 77. Torres-Arzayus MI, Mora JFD, Yuan J, et al. High tumor incidence and activation of the PI3K/AKT pathway in transgenic mice define AIB1 as an oncogene. Cancer Cell 2004;6:263-274. 78. Lonard DM, O'Malley BW. Expanding functional diversity of the coactivators. Trends Biocbem Sci 2005;30:126-132. 79. AuboeufD, Dowhan DH, Dutertre M, et al. A subset of nuclear receptor coregulators act as coupling proteins during synthesis and maturation of RNA transcripts. Mol Cell Bio12005;25:5307-5316.
28 80. HSrlein AJ, N~i~irAM, Heinzel T, et al. Ligand-independent repression by the thyroid hormone receptor mediated by a nuclear receptor co-repressor. Nature 1995;377:397-403. 81. Chen JD, Evans RM. A transcriptional co-repressor that interacts with nuclear hormone receptors. Nature 1995;377:454-457. 82. Privalsky ML. The role of corepressors in transcriptional regulation by nuclear hormone receptors. Ann Rev Physio12004;66:315-360. 83. Lavinsky RM, Jepsen K, Heinzel T, et al. Diverse signaling pathways modulate nuclear receptor recruitment of N-CoR and SMRT complexes. Proc NatlAcad Sci USA 1998;95:2920-2925.
SHERK AND McDONNELL
84. Delage-Mourroux R, Martini PGV, Choi I, et al. Analysis of estrogen receptor interaction with a repressor of estrogen receptor activity (REA) and the regulation of estrogen receptor transcriptional activity by REA. J Biol Cbem 2000;275:35848-35856. 85. Park S-E, Xu J, Frolova A, et al. Genetic deletion of the repressor of estrogen receptor activity (REA) enhances the response to estrogen in target tissues in vivo. Mol Cell Bio12005;25:1989-1999.
~HAPTER
Genetic Programming in Ovarian Development and Oogenesis Joe L E I G H
SIMPSON
Department of Obstetrics and Gynecology, Human Molecular Genetics, Florida Internal University, Miami, FL 33199
exists even in 46,XY-phenotype females, such as in infants with XY gonadal dysgenesis (2) or the genitopalatocardiac syndrome (3). Oocyte development in the presence of a Y chromosome is also well documented in mice (4). Thus, the pathogenesis of germ cell failure in humans can be deduced to be increased germ cell attrition. If two intact X chromosomes are not present, ovarian follicles in 45,X individuals usually degenerate by birth. Genes on the second X chromosome are thus responsible for ovarian maintenance, rather than ovarian differentiation. One might expect existence of a specific gene product for primary ovarian differentiation, but this has proved elusive. Once, a popular candidate was the gene initially termed AHC (adrenal hypoplasia congenital), the gene that encodes DAX1. Following observation that duplication of X p21 resuited in 46,XY embryos differentiating into females (5), it was reasoned that this region could play a primary role in ovarian differentiation in 46,XX individuals. The region contained AHC, which includes or is identical to DAX1 (dosage-sensitive sex reversal/adrenal hypoplasia critical region X); the mouse homolog is Ahch. Ahch is upregulated in the XX mouse ovary, as predicted if Ahch (DAX1) were to play a pivotal role in primary ovarian differentiation. Transgenic XY mice overexpressing Ahch develop as females.
Failure of germ cell development is associated with complete ovarian failure, resulting in lack of secondary sexual pubertal development (primary amenorrhea). Decreased number but not total absence of germ cells is more likely associated with premature ovarian failure, presenting with infertility or secondary amenorrhea. Yet complete and premature ovarian failure may be different manifestations of the same underlying pathogenic and etiologic processes. Chromosomal abnormalities, mutations of autosomal or X-linked genes, and polygenic/multifactorial elements all play a role. In this contribution, we shall enumerate clinical disorders associated with germ cell abnormalities, deducing etiologic factors responsible for ovarian differentiation and oogenesis in normal females.
I. OVARIAN DIFFERENTIATION REQUIRES ONLY ONE X (CONSTITUTIVE) In the absence of the Y chromosome, the indifferent embryonic gonad always develops into an ovary. Germ cells exist in 45,3( human fetuses (1). Oocyte development initially TREATMENT OF THE POSTMENOPAUSAL W O M A N
29
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
JoE LEIGH SIMPSON
30
However, XX mice lacking Ahch (knockout) surprisingly showed normal ovarian differentiation, ovulation, and fertility (6), whereas XY mice mutant for Ahch show testicular germ cell defects. Thus, Ahch cannot be responsible for primary ovarian differentiation in mice, nor presumably could DAX1 (AHC) in humans.
II. POLYGENIC A N D STOCHASTIC C O N T R O L OVER OOCYTE NUMBER It is to be expected that oocyte number (reservoir) will be low in some women simply on statistical (stochastic) grounds. Normal distribution exists for all common anatomic traits (e.g., height), and this principle should apply to oocyte number and reservoir at birth. That a normal distribution of germ cell number exists in ostensibly normal females is well established in animals but difficult to prove in humans. Different rodent strains show characteristic breeding duration, implying genetic control over either the rate of oocyte depletion or the number of oocytes initially present. It follows that some ostensibly normal (menstruating) women may have decreased oocyte reservoir or increased oocyte attrition on a genetic basis, analogous to animal models. In humans a genetic basis for this can be presumed by analogy to the heritability of age at human menopause, a characteristic that clearly shows familiar tendencies. Assessing heritability of age of menopause is complicated because iatrogenic behavior (e.g., hysterectomy) and other confounding factors (e.g., leiomyomata or uterine cancer) must be taken into account. However, within the decade several studies have directly addressed the issue. Cramer et al. (7) performed a case control study on 10,606 U.S. women between 45 and 54 years of age. Women with an early menopause (40 to 45 years) were age-matched with control women who were either still menstruating or had experienced menopause after age 45 years. Of 129 early menopause cases (less than 46 years), 37.5% had a family history of a similarly affected mother, sister, aunt, or grandmother. Only 9% of control subjects had such relatives (odds ratio after adjustment 6.1, 95% confidence interval [CI] 3.9 to 9.4). As predicted on the basis of expectations for a polygenic condition, the odds ratio was greatest (9.1) for sisters and for when menopause occurred prior to 40 years. The frequency of galactose-l-phosphate uridylyltransferase (GALl-) variants (N314D or Q188R) did not deviate from what was expected in early menopause cases, in contrast to previous studies by some of the same authors (8). Torgerson and colleagues (9) reported that if women underwent menopause during the 5-year centile aged 45 to 49 years, the likelihood was increased that menopause would occur in a similar 5-year centile in their daughters. Twin studies have been used to estimate heritability of age at menopause. Two studies have shown similar results
(10). Snieder et al. (11) studied 27 M Z and 353 DZ twin pairs in the United States. For age at menopause, correlation (r) was 0.58 in M Z twins and 0.39 in DZ twins (heritability [h2] = 63%). Treloar et al. (10,12) performed a similar study in 1177 M Z and 711 DZ Australian twin pairs. For age at menopause, correlations (r) were 0.49 to 0.57 for M Z and 0.31 to 0.33 for DZ. Heritability was 31% to 53% (10). Differences between M Z and DZ held when iatrogenic menopause (hysterectomy for leiomyomata or endometriosis) was taken into account. More recent studies continue to show the existence of heritable factors. In the Framingham Heart Study, Murabito et al. (13) determined age of menopause in 1500 U.S. women and 932 of their offspring. Mean ages at natural menopause were 49.1 and 49.4 years, respectively. Heritability was estimated to be 0.49 for mother-daughter and 0.52 for sistersister pairs. Very similar results were reported by van Asselt et al. (14) studying 164 Dutch mother-daughter pairs. Their mean difference in age of menopause was I year; heritability was 44%. Overall, the approximately 50% heritability for age of menopause offers evidence that ovarian capacity and, hence, development is similar in relatives. Interestingly, age of menarche shows similar heritability (15), although responsible biologic mechanisms are probably different.
IIl. M O N O S O M Y X The complement most frequently associated with ovarian dysgenesis is 45,X. The proportion of 45,X individuals in a given sample will depend on the method of ascertainment. Fewer 45,X individuals will be detected if primary amenorrhea is the presenting complaint than if short stature or various somatic anomalies are the presenting complaints. Primary amenorrhea in women is more likely to be the presenting complaint ascertained by gynecologists, whereas short stature in children is likely ascertained by pediatricians. Overall, about 50% of all patients with gonadal dysgenesis have a 45,X complement; 25% have sex chromosomal mosaicism with a structural abnormality (e.g., 45,X/46,XX). Far fewer have a structurally abnormal X or Y chromosome or no detectable chromosomal abnormality (16,17). In 80% of cases the paternally derived X has been lost (18). With one possible exception to be noted, the phenotype does not differ between 45,X m and 45,XP cases (Xm,X of maternal origin; XP,X of paternal origin). That is, in general no evidence exists for imprinting (19,20). In structurally abnormal X chromosomes, it is also the paternal X that is lost (21,22).This indicates that X m and XP chromosomes are lost at random (23). Because 45,Y is lethal, the theoretical percentage of 45,X m cases would be 67%, not greatly different from the 80% actually observed.
CHAPTER 3 Genetic Programming in Ovarian Development and Oogenesis
A. Gonads In most 45,X adults with gonadal dysgenesis, the normal gonad is replaced by a white fibrous streak, 2 to 3 cm long and about 0.5 cm wide, located in the position ordinarily occupied by the ovary. A streak gonad is characterized histologically by interlacing waves of dense fibrous stroma, indistinguishable from normal ovarian stroma (Fig. 3.1). That germ cells are usually completely absent in adults but present in 45,X embryos is the basis for the belief that the pathogenesis of germ cell failure is increased atresia, not failure of germ cell formation. Speed (24,25) has shown that in monosomy X, oogenesis ceases in meiosis I at or before the pachytene meiotic stage. Degenerating pachytene oocytes are observed. A variety of pairing abnormalities and disruptions are seen at later stages of meiosis I, but oocytes in dictyotene are rare. Ogata and Matsuo (23) argue that the ovarian failure found in monosomy X is caused by generalized (nonspecific) meiotic pairing errors, the extent of ovarian failure correlating with extent of pairing failure. Ovarian rete tubules, which probably originate from either mesonephric tubules or medullary sex cords, are present in the median portion of most streak gonads. Hilar cells are usually detected in streak gonads of patients past the age of expected puberty.
FIGURE 3.1
31
That 45,X humans manifest streak gonads is not so obvious as one might expect. Relatively normal ovarian development occurs in many other monosomy X mammals (e.g., mice). These observations are, incidentally, at odds with the hypothesis I (24,25) that ovarian failure merely reflects meiotic pairing errors. If the hypothesis were true, the monosomy X mouse would not differ from the monosomy X human. The more likely explanation is that in humans not all loci on the normal heterochromatic (inactive) X are inactivated. In addition, X inactivation never exists in oocytes, because X reactivation of germ cells occurs before entry in meiotic oogenesis (26). X inactivation could also occur only after some crucial time of differentiation, beyond which only a single euchromatic (active) X is necessary for continued oogenesis.
B. S e c o n d a r y Sexual D e v e l o p m e n t Although streak gonads are usually present in 45,X individuals, about 3% of adult cases menstruate spontaneously and 5% show breast development (Table 3.1). Occasionally, the interval between menstrual periods appears normal in 45,X patients, and fertile patients have been reported. Although an undetected 46,XX cell line should always be
Streakgonad, showinglack of oocytes. From ref. 17.
JOELEIGH SIMPSON
32 suspected in menstruating 45,X patients, it is plausible that a few 45,X individuals could be fertile, inasmuch as germ cells are present in 45,X embryos. The rare offspring of 45,X women are probably not at greatly increased risk for chromosomal abnormalities (27,28), although theoretically they should be. Some authors disagree with this statement (29). Low-grade 45,X/46, XX mosaicism has been claimed to occur in women experiencing repeated abortion (30). Irrespective, menstruation and fertility occur so rarely that 45,X patients should be counseled to anticipate primary amenorrhea and sterility. After hormone therapy is initiated in such women, uterine size becomes normal. This permits 45,X women to carry pregnancies in their own uterus after receipt of donor embryos or donor oocytes mixed with their husband's sperm.
IV. X CHROMOSOMAL MOSAICISM:
45,X/46,XXAND 45,X/47,XXX If nondisjunction or anaphase lag occurs in the zygote and embryo, two or more cell lines may result (mosaicism) (Fig. 3.2). The final complement will depend on the stage at which abnormal cell division occurs and on the types of daughter ceils that survive following nondisjunction or anaphase lag. Detection of mosaicism depends on the number of cells analyzed per tissue and on the number of tissues analyzed (16,17). The most common form of mosaicism associated with gonadal dysgenesis is 45,X/46,XX. Individuals with a 45~X/ 46,XX complement predictably show fewer anomalies than do 45,X individuals. Simpson (16) tabulated that 12% of 45,X/46,XX individuals menstruate, compared with only 3% of 45,X individuals. Among 45,X/46,XX indMduals, 18% undergo breast development, compared with 5% of 45,X individuals. Mean adult height is greater with a 45,X/46,XX complement than with 45,X; more mosaic (25%) than nonmosaic (5%) patients reach adult heights greater than 152 cm (16). Somatic anomalies are less likely to occur in 45,X/46,XX than in 45,X. In Sybert's review and analysis (31), spontaneous menstruation occurred in 45% to 57% (depending on whether mode of ascertainment was in her clinic or from published reports, respectively); frequency of short stature (below the third percentile) was 45% (5/11) and 87% (7/8); fertility occurred in 14% (1/7) and 69% (9/13). Less common but phenotypically similar to 45,X/46,XX individuals is 45,X/47,XXX. 45,X/47,XXX occurs less often but is phenotypically similar to 45,X/46,XX. Individuals with 45,X/46,XY may also show bilateral streak gonads; however, more often they show a unilateral streak gonad and a contralateral dysgenetic testis (mixed gonadal dysgenesis).
TABLE 3.1 Somatic Features Associated with 45,X Chromosomal Complement Growth 9 Decreased birth weight 9 Decreased adult height (141-146 cm) 9 Intellectual function: Verbal IQ.greater than performance IQ_ Cognitive deficits (space-form-blindness) 9 Craniofacial: Premature fusion of spheno-occipital and other sutures, producing brachycephaly Abnormal pinnae Retruded mandible Epicanthal folds (25%) High-arched palate (36%) Abnormal dentition Visual anomalies, usually strabismus (22%) Auditory deficits; sensorineural or secondary to middle ear infections 9 Neck: Pterygium coli (46%) Short broad neck (74%) Low nuchal hair (71%) 9 Chest: Rectangular contour (shield chest) (53%) Apparent widely spaced nipples Tapered lateral ends of clavicle 9 Cardiovascular: Coarctation of aorta or ventricular septal defect (10-16%) 9 Renal (38%): Horseshoe kidneys Unilateral renal aplasia Duplication of ureters 9 Gastrointestinal: Telangiectasias 9 Skin and lymphatics: Pigmented nevi (63%) Lymphedema (38%) due to hypoplasia of superficial vessels 9 Nails: Hypoplasia and malformation (66%) 9 Skeletal: Cubitus valgus (54%) Radial tilt of articular surface of trochlear Clinodactyly V Short metacarpals, usually fourth (48%) Decreased carpal arch (mean angle 117 degrees) Deformities of medical tibial condyle 9 Dermatoglyphics: Increased total digital ridge count Increased distance between palmar triradii a and b Distal axial triradius in position t Modified from ref. 16.
V. PITFALLS IN LOCALIZING OVARIAN MAINTENANCE GENES TO SPECIFIC REGIONS OF THE X The first step in understanding normal ovarian differentiation and in producing gene products of therapeutic benefit is delineating the region (genes) on the X responsible
CHAPTER3 Genetic Programming in Ovarian Development and Oogenesis
NORMAL MITOSIS
33
MITOTIC NONDISJUNCTION
FIGURE 3.2 Diagrammaticrepresentation of the products of (A) normal mitosis and (B) mitosischaracterizedby nondisjunctionof a Y chromosome.If all daughtercells survived, the complementwouldbe 45,X/46,XY/47,XYY.From res 17.
for ovarian maintenance. Until the 1990s, phenotypickaryotypic correlations to deduce location of gonadal and somatic determinants relied solely on metaphase analysis. Prometaphase karyotypes allow 1200-band analysis, whereas traditional banding is 400 to 500; however, each band still contains considerable DNA. More refined analysis is possible using polymorphic DNA markers, which allow precise resolution far beyond the capacity of light microscopy. Progress has, nonetheless, been slow compared with that achieved in delineating the regions of the Y necessary for testicular differentiation (SRY) or spermatogenesis (DdZ). Several impediments are responsible for this relative lack of progress. The incidence of X deletions is low, so the ideal approach of analyzing cases ascertained by population-based methods is impractical. No individuals with X deletions were discovered among 50,000 consecutively born neonates (32). Most del(Xp) or del(Xq) individuals have been identified only because they manifested clinical abnormalities, exceptions being familial cases or cases detected in fetuses at the time of prenatal genetic diagnosis for their mother's advanced maternal age. Doubtless many less severely affected individuals escape detection. Mode of ascertainment ideally should be considered in phenotypic-karyotypic analysis, but in reality this is impractical because sample sizes are too small. Inevitably biases of selection arise. Another pitfall impeding molecular analysis of X-ovarian maintenance genes is that analysis is not always derived from individuals who are cytogenetically well studied. Mosaicism in nonhematogenous tissues has not always been excluded to the extent reasonably possible. Individuals with unstable aberrations (rings, dicentrics) should probably be excluded from phenotypic-karyotypic deductions because monosomy X and other cell lines may arise secondarily, sometimes in tissues (e.g., gonads) relatively inaccessible to
study. Utilizing X-autosome translocations for analysis may also be hazardous because of vicissitudes of X inactivation and because autosomal regions are not devoid of significance for gonadal differentiation.
VI. CYTOGENETICS OF X CHROMOSOMAL ABERRATIONS A. 46,X, del(Xp) or 45,X/46,X, del(Xp) Deletions Deletions of the short arm of the X chromosome show variable phenotypes, depending upon the amount of Xp persisting. The most common breakpoint for terminal deletions is Xp11.2611.4. In 46,X,del(X)(p11) only proximal Xp remains; the del(Xp) chromosome thus appears acrocentric or telocentric. Deletions characterized by progressively more distal breakpoints has been reported: Xp21, 22.1, 22.3. X autosomal translocations leading to Xp interstitial deletions have been reported and are analytically useful, albeit subject to caveats noted in the previous section. The standard for refined analysis is using polymorphic DNA markers to determine precise breakpoints, but relatively few cases have been subjected to refined molecular analysis. Approximately half of 46,X,del(Xp)(p11) individuals show primary amenorrhea and gonadal dysgenesis. Others menstruate and usually show breast development. In a 1989 tabulation by the author, 12 of 27 reported del(X)(p11.2) (11.4) individuals menstruated spontaneously; however, menstruation was rarely normal (33). More recent compilations have not altered these general conclusions (34-36). In another tabulation, Ogata and Matsuo (23) found that 50% of del(X)p11 cases showed primary amenorrhea, with 45% showing secondary amenorrhea. Ovarian function is thus
34 observed more often in individuals with a del(Xp11) chromosome than in 45,X individuals. Women with more distal deletions (del[X][p21.1 to p22.1.22]) menstruate more often, but many are still infertile or even have secondary amenorrhea (Fig. 3.3). Thus, Xp (X[Xpter--+p21]) retains a role in ovarian development (34-36). The distal region of importance must involve Xp21,22.1 or 22.2 because del(X) (p22.3) cases do not show primary amenorrhea. Most women with deletions of Xp are short. Thus, statural determinants--that is, regions with genes conferring stature--must exist on Xp. Because del(Xp) women may menstruate but still be short, regions on Xp responsible for ovarian and statural determinants must be distinct (34-38). Clinically it is important to realize that del(Xp) women may be short despite manifesting normal ovarian function. Both mother and daughter may show the same Xp deletions, not only in association with X autosome translocation but also in association with terminal deletions. In 1977, Fraccaro et al. (37) first called attention to familial distal Xp
FIGURE 3.3 Schematicdiagram of the X chromosomeshowing ovarian function as a function of nonmosaic terminal deletion. References initially provided by Simpson (33). Nonmosaic cases described since that report include Naguib et al. (155); Massa et al. (156); Veneman (157); Schwartz (158), which is a molecularupdate of Fitch et al. (42);Tharapel et al. (43); Zinn et al. (159); Zinn et al. (160); Ogata and Matsuo (23); Marozzi et al. (161); Davison et al. (162);James et al. (22); and Susca et al. (163). In familial aggregates, all affected cases are included because their phenotypes are not alwaysconcordant. In some case,patients are described as havingpremature ovarianfailure, but no information is provided on fertility; in the absence of explicit information it is assumed that no pregnancy has occurred. In some younger patients (e.g., older than 14 years but youngerthan 20-25 years),there has been little opportunity to demonstrate pregnancy, nor is there assurance regular menses will continue. Nonetheless, they are designated as having "regular menses/fertility."From ref. 164.
JOE LEIGH SIMPSON deletions. Among 10 del(Xp) cases studied by James et al. (22) were two mother-daughter pairs; only 6 of their 10 cases arose de n o v o . Familial cases involved deletion at X p l l as well as Xp22-12. Xp interstitial deletions involving Xp 11-22 and Xp11.422.3 (39,40) have been reported.
B. Isochromosomes for Xq (i[Xq]) Division of the centromere in the transverse rather than the longitudinal plane results in an isochromosome, a metacentric chromosome consisting of isologous arms. Both arms are structurally identical and contain the same genes. An isochromosome for the X long arm (i[Xq]) differs from a terminal deletion of Xp in that not just the terminal portion but all of the Xp is deleted. Many isochromosomes for Xq are actually isodicentrics, the clinical significance of which is that a minute portion of Xp is duplicated and retained in addition to duplication of the entire Xq. An isochromosome for the X long arm is the most common X structural abnormality, but coexisting 45,X cell lines (mosaicism) are typical. Nonmosaic cases are relatively uncommon. 46,X,i(Xq) individuals almost always have streak gonads and primary amenorrhea. Occasionally menstruation is observed, but surveys continue to agree with those published by the author (16) more than 25 years ago in showing rarity of menstruation (23). The near complete lack of gonadal development in 46,X,i(Xq) contrasts to that in 46,X, del(X)(p11) individuals, about half of whom menstruate or develop breasts. The contrast is even greater with more distal Xp deletions. Phenotypic differences could be explained if gonadal determinants were present at several different locations on Xp, one locus being deficient in i(Xq) yet retained in del(X)(p11). Alternatively, 46,XX cells may be associated with del(Xp) more often than generally appreciated. Irrespective, duplication of X q - - t h a t is, i(Xq) - - fails to compensate for deficiency of Xp. Thus, gonadal determinants on Xq and Xp have different functions. Another possibility is that all loci on i(Xq) chromosomes are completely inactivated, as indeed study of candidate genes indicates. It seems unlikely that duplication of Xq per se produces abnormalities, given that 47,XXX is often clinically normal. Almost all reported 46,X,i(Xq) patients are short. Their mean height seems to be less than in 45,X (Table 3.2). The mean height of nonmosaic 46,X,i(Xq) patients is 136 cm (16), and many somatic features of the Turner stigmata are observed (16). Somatic anomalies occur as frequently in 46,X,i(Xq) individuals as in 45,X individuals, and the spectrum of anomalies associated with the two complements is similar.
35
CHAPTER 3 Genetic Programming in Ovarian Development and Oogenesis TABLE 3.2
Ovarian Function as Tabulated on Basis of Cases Reviewed in 1995 by Ogato and Matsuo (23)* Ovarian failure (percentage) in X deletions (tabulation of Ogato and Matsuo [23]) Partial (secondary amenorrhea or Complete (primary amenorrhea or abnormal menses) streak gonads)
Monosomy X (45,X) Short arm deficiency del (X)(p11) del (X)(p21-22.2) del (X)(p22.3) i(Xq) idic(Xq) Long arm deficiency del(X)(q13-21) del(X)(q22-25) del(X)(q26-28) idic(Xp)
None (presumed normal)
88%
12%
50 13 0 91 80
45 25 0 9 20
5 65 100 0 0
69 31 8 73
31 56 67 27
0 13 25 0
0
*Ogato and Matsuo provided data in first two columns,with the assumptionbeing that remainder of cases have normal ovarian function--for example, 5% in del(X)(pl1). Publications surveyedoverlap in large part those used for analysisof Simpson (Fig. 3.4)
C. 4 6 , X . d e l ( X q ) a n d 4 5 , X / 4 6 . X , d e l ( X q ) Deletions Deletions of the X long arm are well known (33-36) and vary in composition. If the breakpoint leading to a terminal deletion originates at band Xql3, the derivative chromosome resembles number 17 or number 18; a breakpoint at band Xq21 produces a chromosome resembling number 16. Almost all deletions originating at Xq13 are associated with primary amenorrhea, lack of breast development, and complete ovarian failure (36).Xq13 thus seems to be an important region for ovarian maintenance. Key loci could lie in
proximal Xq21, but not more distal, given that del(X)(q21) to (q24) indMduals menstruate far more often (see Fig. 3.3). Menstruating del(X)(q21) women might have retained a region that contained an ovarian maintenance gene, whereas del(X)(q13 or 21) women with primary amenorrhea might have lost such a locus (36). X autosomal breakpoints associated with ovarian failure span the entire Xq21 region. This was the first indication that a single gene on Xq cannot be responsible for all cases of ovarian failure. The caveat is that balanced X autosome translocations could result not from disruption of a gene per se, but rather from generalized cytologic (meiotic) perturbation.
[BIOSYNTHETIC P.ATHWAYS]
Acetate Cholesterol Precjnenolone
Procjesterone ......
c
C
D ~.. 17 o< - OH precjnenolone ~ Dehydroepmndrosterone
17 c~ - OH progesterone ~
II - deoxycorticosterone
I I-deoxycortisol
Corticosterone
Cortlsol
Androstenedione ~ T e s t o s t e r o n e Estrone ~ E s t r a d l o l
Aldosterone
FIGURE3.4 Importantadrenal and gonadal biosyntheticpathways.Letters designate enzymesrequired for the appropriate conversions. (A) 20ot-hydroxylase,22c~-hydroxylase,and 20,22-desmolase; (B) 38fh-ol-dehydrogenase;(C) 17oL-hydroxylase; (D) 17,20-desmolase;(E) 17-ketosteroidreductase; (F) 21-hydroxylase;(G) 11-hydroxylase.Modified from ref. 17.
36
JoE LEIGH SIMPSON
In more distal Xq deletions, the more common phenotype is not primary amenorrhea but premature ovarian failure (34,35,41,42). Although distal Xq seems less important for ovarian maintenance than proximal Xq, the former must still have regions important for ovarian maintenance. Informative cases have included terminal deletions originating at various sites as well as interstitial deletions (41,43). These interstitial deletions point out hazards of interpretation without molecular studies. Although demarcation into discrete regions is not possible, it is heuristically useful to stratify terminal deletions into those occurring in these regions: Xq13--+21, Xq22--->25, Xq26-+28. Table 3.2 shows ovarian function as tabulated by Ogato and Matsuo (23) using such stratification. Figure 3.3 shows the author's tabulation using a different format. Both estimates are based on pooled cases, and both are generally consistent. Distal Xq deletions may be familial. Some familial Xq deletions are derivative of Xq autosome translocations, but familial terminal or interstitial deletions also exist (43). Familial Xp terminal or interstitial deletions have been characterized by various breakpoints between Xq25 to Xq28. Breakpoints near or in Xq27 seem most common. Some families have been ascertained for reasons other than premature ovarian failure, a case reported by our group having been ascertained following amniotic fluid analysis in a fetus (43). As will be discussed later, the locus for fragile X (FRAXA) is also in the region. About 15% of women with >55CGG repeats show premature ovarian failure. Distal Xq deletions seem to have a less deleterious effect on stature than proximal deletions. Somatic anomalies of the Turner stigmata are uncommon and perhaps no different than in the general population.
VII. OVARIAN GENES ON T H E X Although chromosomal regions on Xp (and Xq) are presumed to contain genes pivotal to ovarian germ cell function, the actual gene and gene products remain frustratingly unclear. A host of candidate genes are being studied.
A. Candidate Genes on Xp 9 USP9X (ubiquitin specific protease 9): This gene maps to Xpll.4 (44) and is expressed in multiple tissues. The Drosophila orthologue of USP9X is required for eye development and oogenesis. The role USP9X plays in human gonadal development is still unclear, but its location in the appropriate region is tantalizing. 9 Z F X (zinc finger X): Mapped to Xp22.1--+21.3, Z F X is a candidate gene for short stature and ovarian failure (45). It has attracted attention on the basis of being homologous
to ZF~, once the prime candidate gene for male sex determination. Mice null for Zfx are small, less viable, less fertile, and characterized by diminished germ cell number in ovaries and testes (46). Their external and internal genitalia are otherwise normal. 9 B M P 1 5 (bone morphogenetic protein 15): Bone morphogenetic protein 15 (BMP15) is a member of the transforming growth factor-J3 (TGF-[3) superfamily. These genes direct many developmental pathways through binding and activating transmembrane serine/threonine kinase receptors. B M P is involved in folliculogenesis and embryonic development, being expressed in gonads. The B M P 1 5 gene is located on Xpll.2 and has two exons. The gene is a member of the TGF family. Prior to human cases being reported, animal studies had suggested that perturbations of B M P 1 5 could be important in ovarian development. Heterozygous Inverdale sheep carrying a mutation in the B M P 1 5 gene show an increased ovulation rate, with twin and triplet births. Primary ovarian failure occurs in homozygotes (47). B M P 1 5 knockout mutant female mice are subfertile, showing decreased ovulation rates, reduced litter size, and decreased number of litters per lifetime (48). In humans (49), Di Pasquale et al. reported a heterozygous Y235C missense mutation in the second exon of the B M P 1 5 gene in each of two sisters having ovarian failure. The proband had streak gonads and elevated follicle-stimulating hormone (FSH 80 mlU/mL); the younger sibling had one episode of vaginal spotting but at age 18 years had an FSH level of 67 mlU/mL. The mother was homozygously normal (Y235) at this allele, and C235 was transmitted from the father. The authors presented in vitro evidence of a dominant negative mechanism. Another TGF family member, growth differentiation factor 9 (GDF9; see p. 39) has also been implicated in ovarian failure. In these cases the mutation was also heterozygous (50). Given that proteins of TGF family members (BMP15, GDF9) may form heterodimers, a single mutation could plausibly generate a dysfunctional gene product. ~ Region Localized by Q.gantitative Linkage Analysis: Genomewide linkage analysis has been applied in order to determine which genes influence age of menopause. One study has shown association between X p l l and the age of menopause. This suggests the existence of a locus of importance.
B. Candidate Genes on Xq 9 XIST" Xql3 contains the X inactivation center and XIST. Loss of germ cells may or may not be the direct result of perturbation of XIST, despite years of speculation that
CHAPTER 3 Genetic Programming in Ovarian Development and Oogenesis disturbances of X inactivation per se lead to ovarian failure. The concept of a well-defined "critical regioN' necessary for ovarian development receives less attention than in the past, but this does not exclude a region rich in pivotal genes. 9 D I d P H 2 : Human DIHPH2 is the homolog of Drosophila melanogaster diaphanous (dia). In Drosophila, dia is a member of a family of proteins that help establish cell polarity, govern cytokinesis, and reorganize the actin cytoskeleton. In both males and females, dia causes sterility in flies (51). A human Xq21 autosome translocation was found to have a disruption of the last intron o f DIHPH2 (52). Bione and Toniolo (53) and Prueitt et al. (54) found disruption of X P N P E P 2 in an Xq autosome. D I A P H 2 on X P N P E D 2 could be relevant m so-called POF2. 9 F M R 1 : On Xq27 lies the F M R 1 locus, the gene for fragile X syndrome. About 20% of females with an F M R 1 premutation (55 or more CGG repeats) develop premature ovarian failure, although paradoxically those with the full mutation do not. (Fragile X syndrome is discussed later.) This locus cannot logically correspond to that which when deleted causes ovarian failure in del(Xq) (2.7 or 2.8).
VIII. AUTOSOMAL CHROMOSOMAL ABNORMALITIES
A. Trisomy Autosomal trisomy has long been known to affect adversely ovarian development. The question remains whether this effect is mediated by nonspecific meiotic perturbation or by chromosome-specific genes, perhaps acting in double dose. Trisomies 13 and 18 are frequently associated with ovarian failure, as indicated by necropsy observations in stillborn fetuses or deceased neonates. Few longitudinal data are available in trisomy 13 or 18 because few females survive until menarche. In trisomy 21, however, ovarian function may be normal. There seems to be no objective information on age at menopause. Pregnancies occur in trisomy 21 females (55). About one third of offspring are aneuploid (fewer than the theoretically expected 50%). If nonspecific meiotic breakdown is merely secondary to an uneven number of chromosomes, the effect should be the same with chromosome 21 as with chromosomes 18 or 13. The ostensibly normal ovarian function in trisomy 21 argues for the existence of specific ovarian genes on chromosomes 13 and 18 but not on 21.
B. Translocations Chromosomal rearrangements, specifically balanced autosomal reciprocal translocations, are not infrequently observed in otherwise normal women with complete or partial
37
ovarian failure. As with autosomal trisomy, it is unclear whether this association reflects disruption of autosomal loci integral for ovarian preservation and oogenesis. That no chromosome is consistently involved suggests nonspecific meiotic perturbation. In fact, men who are azoospermic or oligospermic but otherwise normal clinically show balanced autosomal translocations far more often than expected: About 1% of men requiring intracytoplasmic sperm injection (ICSI) show a balanced autosomal rearrangement, typically a balanced translocation (56). A problem of comparable magnitude probably exists in women. The pathogenesis leading to meiotic breakdown presumably involves malalignment or failure of synapsis. Recognizing individuals with autosomal rearrangements is important because their offspring are at increased risk for gametes showing unbalanced segregation.
IX. AUTOSOMAL GENES In addition to ovarian failure resulting from monosomy X and X deletions, perturbation of a host of autosomal genes can be deduced to be pivotal because their disturbance causes ovarian dysgenesis in 46,XX indMduals. Sometimes the causative gene is known, whereas in other cases a specific phenotype merely allows us to deduce its presence. Table 3.3 shows the spectrum of genetic causes of ovarian failure in 46,XX women. The archetypal form of XX gonadal dysgenesis is that characterized by streak gonads nat associated with somatic anomalies. Inheritance is autosomal recessive (57). Affected individuals are normal in stature (58), and Turner stigmata are absent. Individuals with XX gonadal dysgenesis as defined are heterogenous and should be more precisely delineated. This may or may not be possible. Of clinical interest is that in XY gonadal dysgenesis variable expressivity occurs. In many families one sibling may show streak gonads, whereas another may show ovarian hypoplasia but not streak gonads per se. A mutant gene operative in this fashion may thus be responsible for isolated cases of premature ovarian failure (POF).
A. FSH-f3 (FSH) Mutations Coded by a gene on chromosome 11, FSH is composed of a unique [3 subunit and an (x subunit shared in common with thyroid-stimulating hormone (TSH), luteinizing hormone (LH), and human chorionic gonadotropin (HCG). Cellular action requires an FSH receptor (FSHR), the gene for which is located on chromosome 2. Mutations in FSH-f3 are rare. Two affected females showed neither thelarche nor menarche. Matthews et al. (59) described a homozygous 2 bp deletion (GT) in exon
JOELEIGH SIMPSON
38 TABLE 3.3 Genetic Causes of Complete Ovarian Failure and Premature Ovarian Failure (POF) in 46,XX Women FSH-13 (FSH) mutations FSHR mutations Inactivating LH receptor (LHR) mutations Inhibin a (INHA) mutations Growth differentiating factor 9 (GDFg) mutations Perrault syndrome (neurosensory deafness) Cerebellar ataxia with XX gonadal dysgenesis Malformation syndromes associated with XX gonadal dysgenesis Microcephaly and arachnodactyly (94) Epibulbar dermoids (95) Short stature and metabolic acidosis (96,136) Mendelian disorders characterized predominately by somatic features (see Table 3.4) Blepharophimosis-ptosis-epicanthus (BPE) (FOXL2) 17(z-hydroxylase/17,20 desmolase deficiency (CYP17) A_romatase mutations deficiencies (CYP19) Germ cell failure in both sexes (nonsyndrome) Germ cell failure in both sexes (syndrome) Hypertension and deafness (Hamet et al. [137]) Alopecia (A1Awadi et al.) (110) Microcephaly and short stature (Mikati et al. [111]) Galactosemia Carbohydrate-deficient glycoprotein (CGD) Agonadia (46,XX cases) Dynamic mutations (triplet repeat) Fragile X Myotonic dystrophy Ovarian-specific autoimmune syndrome Polyglandular autoimmune syndrome
3 at codon 61 that resulted in a premature stop codon (Va161X). Layman et al. (60) reported a compound heterozygote: one allele was Va161X, whereas the other was Cys51Gly.
B. FSHR Mutations FSHR mutations are not uncommon in Finland but seem rare elsewhere. Aittomaki searched Finnish hospitals and cytogenetic labs to identify 75 women with 46,XX primary or secondary amenorrhea and a serum FSH of 40 mIU/mL or greater. The gene was first localized to chromosome 2p. Then, a missense mutation in the FSH receptor gene (FSHR) in exon 7 (566C---~T or Ala566Val) was found in six families (61,62). Women heterozygous for the mutation showed no decrease in fertility. Outside Finland, Ala566Val seems uncommon. No mutations in FSHR were found in North American women having either 46,XX hypergonadotropic hypogonadism (63) or premature ovarian failure (64). Similar findings were reported in 46,XX premature ovarian failure or primary amenorrhea cases from Germany (65), Brazil (66), and Mexico (67). Awaited are studies examining all
exons of the FSHR gene, rather than just the Ala566Val missense mutation. Two reports describe compound heterozygosity for FSHR mutation (68,69). Genotypes were Ile160Thr/Arg573Cys and Asp224Val/Leu602Val, respectively.
C. Inactivating LH Receptor Mutations Analogous to the FSH receptor gene, the LH receptor gene (LHR) is relatively large, 75 kD in length and consisting of 17 exons. Located on 2p near the locus for FSHR, the first 10 exons in LHR are extracellular, the l lth transmembrane, and the remaining six intracellular. Mutations have been detected only in the transmembrane domain. Most reported LHR mutations are 46,XY and have been found in individuals with XY sex reversal (70). The latter have Leydig cell hypoplasia. LHR mutations in 46,XX women cause the phenotype of XX gonadal dysgenesis or premature ovarian failure. The phenotype is oligomenorrhea or more often primary amenorrhea. Ovulation does not occur, even though gametogenesis proceeds until the preovulatory stage. This is consistent with findings using mouse knockout models (71). 46,XX cases of LHR mutation have usually been ascertained in sibships in which they have been affected 46,XY siblings. Latronico et al. (72) reported a 22-year-old woman who presented with primary amenorrhea due to an LHR mutation. In that family, three 46,XY siblings with Leydig cell hypoplasia had the same homozygous C 5 5 4 ~ X mutation, which resulted in a truncated protein consisting of five rather than seven transmembrane domains. The 46,XX sibling had breast development but only a single episode of menstrual bleeding at age 20; LH was 37 mlU/mL, FSH 9 mlU/mL. The mutation reduced signal transduction activity of the LH receptor gene. In another 46,XX case, Latronico et al. (72) observed secondary amenorrhea; LH and FSH were 10 and 9 mlU/mL, respectively. The mutation was homozygous Ala593Pro.
D. Inhibin alpha (INHA)Mutations Inhibins (INH) are heterodimeric glycoproteins. They consist of an oL subunit and one of two 13 submits (~A and 13B), producing INHd and INHB, respectively. These gene products exert negative feedback inhibition on FSH. Inhibins are opposed by activins, which enhance FSH secretion. Inhibins are synthesized by granulosa cells. In premature ovarian failure, serum inhibin is low and FSH elevated. Elevated FSH and low inhibin thus indicate reproductive aging. Perturbation of inhibins could very plausibly cause ovarian failure.
CHAPTER 3 Genetic Programming in Ovarian Development and Oogenesis Three studies have shown an association between POF and a particular I N H A missense mutation or polymorphismmG769A (50,73,74). Studying cases from New Zealand, Shelling et al. (73) found G769A in 3 of 43 POF patients (7%) versus only 1 of 150 normal control subjects (0.7%). However, the clinically normal mother of one of the three G769A cases had the same perturbation. Marozzi et al. (74) found G769A in 7 of 157 Italian POF cases, 3 of 12 primary amenorrhea cases, and 0 of 36 women with early menopause (40 to 45 years). Familial POF cases were relatively more likely to have G769A than sporadic cases. Dixit et al. (50) found G796A in 9 of 80 Indian POF cases; no mutations were found in I N H B B or INHBA. On the other hand, normal individuals have also shown the G769A transition. Individuals with G769A may also be abnormal even if another G769A family member has POF; thus, G769A itself is not necessarily paramount, but could be in linkage disequilibrium or require another perturbation in cis.
E. GDF9 Mutations GDF9 is a member of the TGFB family. Located on chromosome 5, the gene consists of 2 exons encoding a 454-amnio-acid peptide. This arrangement is similar to that of other members of the TGFB family. GDF9 is expressed in oocytes and plays an essential role both in early and late folliculogenesis. GDF9 protein promotes cumulus expansion in cumulus cell-oocyte complexes (75), whereas its suppression prevents cumulus expansion in sheep (76). Immunization against GDF9 in sheep disrupts early folliculogenesis, leading to the absence of normal follicles beyond the primary stage of development. That GDF9 perturbations can cause ovarian failure in humans is plausible given its known role in oocyte development and demonstration of lack of ovarian development in GDF9 knockout (null) mice (77). Dixit et al. (78) sought perturbations in the GDF9 coding region in 127 women with POF (FSH greater than 40 mlU/ mL), 58 with primary amenorrhea and 10 with secondary amenorrhea (cessation of menses before age 40). A total of 220 control subjects were studied. Two missense mutations were found and considered causative. A199C was found in five women, four with POF and one with secondary amenorrhea; G646A was found in two additional POF subjects. No control women showed a mutation. A variety of other sequence variants (single nucleotide polymorphisms [SNPs]) and silent mutations (no amino acid alteration) were found, but these are unlikely to be causative. Reported GDF9 mutations have been heterozygous. If pathogenesis of ovarian failure were recessive, the G646A and A199C mutations might simply connote heterozygosity of no clinical significance. That is, heterozygous status for an enzyme deficiency is not abnormal because 50% enzyme
39
level more than suffices. That GDF9 forms heterodimers with other member of the TGFB superfamily is relevant. One such member is BMP15, mutation of which also is associated with POF (79). A dominant negative interfering with dimerization is an attractive hypothesis. The proportion of Indian women with GDF9 perturbation was 2.7% (6/220) in the Dixit et al. (78) study, suggesting a not insignificant role in this population. In Japan, 2 of 53 Japanese women showed a mutation for either GDF9 or BMP15 (80). Fifteen had POF, and 38 had polycystic ovarian syndrome. In the U.S., Kovanci et al. (81) found one GDF9 mutation in 61 women with POE Although obviously not common, detection of disturbances in GDF9 in a second population confirms that this is responsible for a proportion of POF cases.
F. Perrault Syndrome XX gonadal dysgenesis with neurosensory deafness is Perrault syndrome (82). Perrault syndrome is inherited in autosomal recessive fashion (83-86). Endocrinologic features seem identical to those of XX gonadal dysgenesis without deafness. Attractive candidate genes are those in the connexin family, such as connexin 37 (Gap Junction alpha 4 or GJA4). Mutations in the connexin gene family are responsible for many forms of congenital deafness in humans, a fact of obvious relevance to Perrault syndrome. In the murine knockout model for connexin 37 (87), null mice show gonadal failure due to arrest at the antral stage of oogenesis.
G. Cerebellar Ataxia with XX Gonadal Dysgenesis Ataxia and hypergonadotropic hypogonadism were first associated by Skre et al. (88), who described cases in two families. In one family a 16-year-old girl was affected, whereas in the other family three sisters were affected. In the sporadic case and in one of the three sisters, ataxia was first observed shortly after birth; in the two other sisters age of onset of ataxia began later during childhood. Cataracts were present in all the cases reported by Skre et al. (88). Hypergonadotropic hypogonadism and ataxia was later reported by De Michele et al. (89), Linssen et al. (90), Gottschalk et al. (91), Fryns et al. (92), Nishi et al. (86), and Amor et al. (93). Ataxia differed clinically among these cases. Ataxia was not progressive in the cases of Skre et al. (88), De Michele et al. (89), Nishi et al. (86), and Amor et al. (93). Mitochondrial enzymopathy was found by De Michele et al. (89). Only Skre et al. (88) observed cataracts, whereas amelogenesis was observed only by Linssen et al. (90). Neurosensory deafness reminiscent of Perraut syndrome was reported by Amor et al. (93). Mental retardation was also variable (93).
JoE LEIGH SIMPSON
40 Overall, genetic heterogeneity exists in the hypergonadotropic hypogonadism disorders showing cerebellar ataxia. However, not every family need be unique.
H. Malformation Syndromes Associated with XX Gonadal Dysgenesis XX gonadal dysgenesis is found in three rare malformation syndromes, all presumed autosomal recessive on the basis of multiple affected siblings. These include XX gonadal dysgenesis, microcephaly, and arachnodactyly (94); XX gonadal dysgenesis and epibulbar dermoid (95); and XX gonadal dysgenesis, short stature, and metabolic acidosis (96).
I. Mendelian Disorders Characterized Predominately by Somatic Features Ovarian failure is a feature, albeit not universal, in several well-established Mendelian disorders. These are enumerated in Table 3.4.
TABLE 3.4
j. Blepharophimosis-Ptosis-Epicanthus (FOXL2) Blepharophimosis-ptosis-epicanthus syndrome (BPE) is an autosomal dominant malformation syndrome. Type II BPF is characterized not only by ocular findings but also by premature ovarian failure (POF) (97,98). The BPE gene proved to be encoded on 3q21-24, and is forkhead box L2 (FOXL2). FOXL2 consists of only one exon. The gene is expressed predominately in eyelids and ovaries (99) and like other forkhead DNA-binding proteins is crucial in signal induction. Crisponi et al. (99) showed that in four families FOXL2 mutations cosegregated with BPE and POE Mutations included stop codons as well as a 17bp duplication that resulted in a frameshift and, hence, truncated protein. In the absence of somatic features, FOXL2 mutations are rare causes of P O E De Baere et al. (100) found no FOXL2 mutations in 30 POF patients having no eyelid abnormalities. Harris et al. (101) found two mutations in 70 POF cases.
Mendelian Disorders Associated with Ovarian Failure (Hypergonadotropic Hypogonadism)* Somatic features
Cockayne syndrome (Nance and Berry [139]) Martsolf syndrome (Martsolf et al. [141]) Nijmegen syndrome (Weemaes et al. [144]) Werner syndrome (Goto et al. [147]) Rothmund-Thompson syndrome (Hall et al. [148]) Ataxia-telangiectasia
Bloom syndrome (German [152]; German et al. [153]; German [154])
Dwarfism, microcephaly, mental retardation, pigmentary retinopathy and photosensitivity, premature senility; sensitivity to ultraviolet light Short stature, microbrachycephaly, cataracts, abnormal facies with relative prognathism due to maxillary hypoplasia Chromosomal instability, immunodeficiency, hypersensitivity to ionizing radiation, malignancy Short stature, premature senility, skin changes (scleroderma) Skin abnormalities (telangiectasia, erythema, irregular pigmentation), short stature, cataracts, sparse hair, small hands and feet, mental retardation, osteosarcoma Cerebella ataxia, multiple telangiectasias (eyes, ears, flexor surface of extremities), immunodeficiency, chromosomal breakage, malignancy, x-ray hypersensitivity Dolichocephaly, growth deficiency, sunsensitive facial erythema, chromosomal instability (increased sister chromatical exchange), increased malignancy
*From ref. 138 where referencesare provided.
Ovarian anomalies
Etiology
Ovarian atrophy and fibrosis (Sugarman et al. [140])
Autosomal recessive
"Primary hypogonadism" (Harbord et al. [142]; Hennekam et al. [143])
Autosomal recessive
Ovarian failure (primary) (Conley et al. [145]; Chrzanowska et al. [146]) Ovarian failure (Goto et al. [147]) Ovarian failure (primary hypogonadism or delayed puberty) (Starr et al. [149])
Autosomal recessive (7;14 rearrangement) Autosomal recessive
"Complete absence of ovaries," "absence of primary follicles" (Zadik et al. [150]; Waldmann et al. [151]) Ovarian failure (German [154])
Autosomal recessive
Autosomal recessive
Autosomal recessive
41
CHAPTER3 Genetic Programming in Ovarian Development and Oogenesis
K. 17o~-hydroxylase/17,20 Desmolase Deficiency ( c w m 7) Deficiency of 17ci-hydroxylase/17,20 desmolase should be considered an uncommon cause of 46,XX hypergonadotropic hypogonadism. Patients with 46,XX present with primary amenorrhea or premature ovarian failure. Hypertension often coexists. Ovaries are hypoplastic and sometimes streaklike in appearance. Oocytes appear incapable of reaching diameters greater than 2.5 mm (102). Stimulation with exogenous gonadotropins can produce oocytes capable of fertilization in vitro (103).
L. Aromatase Mutations in Females (46,XX)
(C-TP19) Conversion of androgens (A 4-androstenedione) to estrogens (estrone) requires cytochrome P-450 aromatase (GYP19), an enzyme that is the gene product of a 40-kb gene located on chromosome 15q21.1 (104). The gene consists of 10 exons. Phenotypic female patients with 46,XX aromatase deficiency may present with primary amenorrhea. Ito et al. (105) reported an aromatase mutation (CYP19) in a 46,XX 18-year-old Japanese woman having primary amenorrhea and cystic ovaries. The patient was a compound heterozygote who showed two different point mutations in exon 10. No enzyme activity was evident in vitro. Conte et al. (106) also reported aromatase deficiency in a 46,XX woman presenting with primary amenorrhea, elevated gonadotropins, and ovarian cysts. Again, compound heterozygosity was found for two different exon 10 mutations. One was a C1303T transition leading to cysteine rather than arginine, whereas the other was a G1310A transition leading to tyrosine rather than cysteine. A different phenotype was reported by Mullis et al. (107). Clitoral enlargement occurred at puberty, and there was no breast development. Multiple ovarian follicular cysts were present. FSH was elevated; estrone and estradiol were decreased. Estrogen and progesterone therapy resulted in a growth spurt, decreased FSH, decreased androstenedione and testosterone, breast development, menarche, and decreased follicular cysts. Compound heterozygosity existed.
M. Germ Cell Failure in Both Sexes (Nonsyndromic) In several sibships, both males (46,XY) and females (46,XX) have shown germ cell failure. Affected females show streak gonads, whereas males show germ cell aplasia
(Sertoli-cell-only syndrome). In two families, parents were consanguineous. In neither were somatic anomalies observed (108,109). These families demonstrate that a single autosomal gene may deleteriously affect germ cell development in both sexes, presumably acting at a site common to early germ cell development.
N. Germ Cell Failure in Both Sexes (Syndromic) In contrast to families described in the previous section, in two others, coexisting patterns of somatic anomalies suggest different genes. In both, parents were consanguineous. A1 Awadi et al. (110) reported germ cell failure and an unusual form of alopecia. Scalp hair persisted in the midline, but no hair was present on sides ("mane-like"). Mikati et al. (111) reported germ cell failure, microcephaly, short stature, mental retardation; and unusual facies (synophrys, abnormal pinnae, micrognathia, loss of teeth). The siblings reported by A1 Awadi et al. (110) were Jordanian; those reported by Mikati et al. (111) were Lebanese.
O. Myotonic Dystrophy (CTG n) Myotonic dystrophy is an autosomal dominant disorder characterized by muscle wasting (head, neck, extremities), frontal balding, cataracts, and male hypogonadism (80%) attributable to testicular atrophy. Female hypogonadism is very much less common, if increased at all. Despite frequent citations in texts, ovarian failure in myotonic dystrophy is poorly documented. Pathogenesis of myotonic dystrophy involves nucleotide expansion of CTG repeats in the 3' untranslated region of the causative gene, which is located on chromosome 19. Normally, 5-27 CTG repeats are present. Heterozygotes usually have at least 50 repeats; severely affected individuals show 600 or more. As in patients with FRAXA (FMR1) (discussed later), response to ovulation induction is poor. Sermon et al. (112) report fewer embryos per cycle in these patients than in patients without FRAXA who undergo standard assisted reproductive technology (ART) treatments. More recent reports show better results (113).
P. Galactosemia Galactosemias is caused by galactose-l-phosphate uridyl transferase (GALT) deficiency. Kaufman et al. (114) reported POF in 12 of 18 galactosemic women, and Waggoner et al. (115) reported ovarian failure in 8 of 47 (17%) females with galactosemia. Pathogenesis presumably involves galactose
42
JOE LEIGH SIMPSON
toxicity after birth, given that elevated fetal levels of toxic metabolites should be cleared rapidly in utero by maternal enzymes. Consistent with this idea, a neonate with galactosemia showed normal ovarian histology (116). Once postulated, there remains little reason to believe that POF is caused by heterozygosity for galactosemia. Not even all homozygotes for human galactosemia are abnormal. Null transgenic mice in which GALT is inactivated (knockout) are normal with respect to ovarian failure (117).
Q:. Carbohydrate-Deficient Glycoprotein (CDG) In type I carbohydrate-deficient-glycoprotein (CDG) deficiency, mannose 6-phosphate cannot be converted to mannose 1-phosphate. Thus, lipid-linked mannose-containing oligosaccharides, necessary for secretory glycoproteins, cannot be synthesized. The gene is located on 16p13, and the usual molecular perturbation is a missense mutation (118). In addition to neurologic abnormalities (119), ovarian failure is common. FSH is elevated, secondary sexual development fails to occur, and ovaries lack follicular activity (120,121).
R. Fragile X Syndrome (CGG n) Fragile X syndrome is a common form of X-linked mental retardation, caused by mutation of the FMRI gene, located on Xq27. The molecular basis involves repetition of the triplet repeat CGG. If more than 200 repeats exist, transcriptional silencing of a RNA-binding protein occurs. In normal males, the normal number of CGG repeats is less than 55. Males or heterozygous females with 55 to 199 repeats are said to have a premutation (122). During female (but not male) meiosis, the number of triplet repeats may increase (expand). A phenotypicaUy normal woman with a FRAXA premutation may have an affected son if the number of CGG repeats on the oocyte of the X she transmits to her male offspring expands during meiosis to greater than 200. Affected males show mental retardation, characteristic facial features, and large testes. Expansion will not occur if there are less than approximately 55 CGG repeats, although the precise threshold remains arguable. Females may also show mental retardation, but the phenotype is less severe than in males. Of relevance here is that 20% to 25% of females with the FRAXA premutation show POF, defined as menopause prior to 40 years of age. Schwartz et al. (123) found oligomenorrhea in 38% of premutation carriers versus 6% control subjects. Analyzing 1268 control subjects, 50 familial POF cases, and 244 sporadic POF cases, Allingham-Hawkins et al. (124), reported that 63 of 395 premutation carriers (16%) underwent menopause before 40 years of age; the frequency in control subjects was 0.4%. In the U.S., Sullivan
et al. (122) found 12.9% of premutation carriers (N = 250; great than 59 repeats) to have POF, versus 1.3% (2/157) of control subjects. Surprisingly, FSH was increased in premutation carriers age 30 to 39 years, but not in carriers of other ages. The number of repeats significantly correlated with risk of POF, within a specified range. The risk slightly increased up to 79 repeats, but was much higher for 80 to 99 repeats. Yet there was no further increased risk for 100 or more repeats. The reason for the plateau is not clear. However, this observation is consistent with females with the full mutation not showing POF (124). FMRI testing is recommended in Europe as part of the evaluation for premature ovarian failure (125). If oocyte or ovarian slice cryopreservation becomes practical, population screening might be justified for fertility preservation.
S. Genes Postulated on Basis of Animal Models In many mouse mutants, knockout genes cause germ cell deficiencies, in either males and females or both. The widely varying modes of action of these genes make it clear that germ cell failure in humans will not necessarily be predicted simply on the basis of ostensible gene action. Among genes that seem promising for investigation are Folliculogenesis specific basic helix-loop-helix (FIGLA) (126); connexin 37 (CX37), also called gap junction protein, alpha 4, 37kDa (GJZI4) (87); G protein-coupled receptor 3 (GPR3) (127); and NOBOX oogenesis homeobox (NOBOX) (128), or ovarian homeobox (OBOX) (129). The only one of these genes for which a search for perturbations has been reported in human POF is NOBOX, for which Zhao et al. (130) found no mutations in 30 Japanese patients.
T. Undefined Autosomal Genes Using microarrays, Arraztoa et al. (131) found 95 genes to be expressed at high levels in primordial monkey oocytes. Each gene could be plausibly entertained as a candidate gene for ovarian failure. This citation is but a single example of the many candidate genes that can be expected to be derived from this broadly applicable approach.
X. H O W O F T E N IS PREMATURE OVARIAN FAILURE GENETIC? As already discussed individually, premature ovarian failure can result from several different genetic mechanisms. These include (1) X-chromosomal abnormalities; (2) autosomal recessive genes causing the various types of XX gonadal dysgenesis; and (3) autosomal dominant genes whose action is restricted to POE The former two topics have been
43
CHAPTER 3 Genetic Programming in Ovarian Development and Oogenesis considered in detail earlier, so we shall focus here only on the third. However, prior to doing so it is useful to recall the role the former two etiologies play in P O E
A. X - C h r o m o s o m a l
Abnormalities
Not only complete ovarian failure but premature ovarian failure occurs in X abnormalities. Many mosaic individuals are so mildly affected that they are never detected clinically. At least 10% to 15% of 45,X/46,XX indMduals menstruate, compared with fewer than 5% of45,X indMduals (16). Spontaneous menstruation occurs in about half of all 46,X, del(X)(p11) and 46,X,del(X)(p21 or 22) individuals, who often present with secondary amenorrhea and premature ovarian failure. Deletions or X autosomal translocations involving regions Xp22 and Xq26 are more likely to be associated with premature ovarian failure than complete ovarian failure. Recall also that women with the F R A X A premutation ( F M R 1 ) show an increased frequency of premature ovarian failure, a phenomenon that may or may not be the result of perturbations of the terminal Xq ovarian maintenance genes.
B. Autosomal Recessive POF In some families we have noted that the propositus may have 46,XX gonadal dysgenesis and streak gonads, but a sibling ovarian hypoplasia with some oocytes. These sibships suggest that some m u t a n t genes responsible for causing XX gonadal dysgenesis (see Table 3.3) are capable of exerting variable expressivity. Thus, these autosomal recessive mutations may be manifested as less severe ovarian pathology. In the Finnish cases of F S H R mutations ascertained by Aittomaki (61,62), P O F not infrequently coexisted in the same kindred as complete ovarian failure. Genes causing familiar XX gonadal dysgenesis genes may therefore also be responsible for familial premature ovarian failure.
C. Autosomal Dominant POF Probably distinct from conditions discussed earlier is idiopathic P O F transmitted in more than one generation (132,133). This suggests autosomal dominant inheritance. These families could also have subtle X deletions, F R A X A premutations, or any of a number of perturbations in autosomal genes listed earlier. In his 1984 study, Mattison et al. (134) examined five families. These families were probably ascertained from a very large population base, raising the concern that the familial aggregates could have been observed by chance or on the basis of polygenic factors. In none were ovarian antibodies present.
Defining POF as cessation of menses for 6 months or longer, Vegetti et al. (135) ascertained 81 Italian women under age 40. Of these, 10 were excluded on the basis of presumptive known etiology (5 abnormal karyotypes, 3 previous ovarian surgeries, 1 prior chemotherapy, 1 galactosemia). Pedigree analysis was performed. Of the remaining 71, 23 (31%) had an affected female relative. Subjects with a positive family history were an older median age (37.5 years) then those without such a history (31 years). Pattern of inheritance was autosomal dominant inheritance. Transmission occurred through both male and female members. Neither BPE nor fragile X syndromes were observed clinically.
References 1. Jirasek J. Principles of reproductive embryology. In: JL Simpson, ed. Disorders of sexual differentiation: etiology and clinical delineation. San Diego: Academic Press, 1976; 51 - 110. 2. Cussen LJ, MacMahon RA. Germ cells and ova in dysgenetic gonads of a 46-XY female dizygotic twin. Am J Dis Child 1979;133:373-375. 3. Greenberg F, Gresik MV, Carpenter RJ, et al. The Gardner-SilengoWachtel or genito-palato-cardiac syndrome: male pseudohermaphroditism with micrognathia, cleft palate, and conotruncal cardiac defect. AmJMed Genet 1987;26:59-64. 4. Evans EP, Ford CE, Lyon ME Direct evidenceof the capacityof the XY germ cell in the mouse to become an oocyte. Nature 1977;267: 430-431. 5. Bardoni B, Zanaria E, Guioli S, et al. A dosage sensitive locus at chromosome Xp21 is involved in male to female sex reversal. Nat Genet 1994;7:497-501. 6. Yu RN, Ito M, Saunders TL, et al. Role of Ahch in gonadal development and gametogenesis.Nat Genet 1998;20:353-357. 7. Cramer DW, Xu H, Harlow BL. Family history as a predictor of early menopause. Fertil Sterff 1995;64:740-745. 8. CramerDW, Harlow BL, Barbieri RL, Ng WG. Galactose-l-phosphate uridyl transferase activityassociatedwith age at menopause and reproductive history. Fertil Steri11989;51:609-615. 9. Torgerson DJ, Thomas RE, Reid DM. Mothers and daughters menopausal ages: is there a link? EurJ Obstet Gynecol Reprod Bio11997;74: 63-66. 10. Treloar SA, Do KA, Martin NG. Genetic influences on the age at menopause. Lancet 1998;352:1084-1085. 11. Snieder H, MacGregor AJ, Spector TD. Genes control the cessation of a woman's reproductive life: a twin study of hysterectomy and age at menopause.J Clin Endocrinol Metab 1998;83:1875-1880. 12. Treloar SA, Martin NG. Age at menarche as a fitness trait: nonadditive genetic variance detected in a large twin sample. Am J Hum Genet 1990;47:137-148. 13. Murabito JM, Yang Q~ Fox C, Wilson PW, Cupples LA. Heritability of age at natural menopause in the Framingham Heart Study.J Clin Endocrinol Metab 2005;90:3427- 3430. 14. van Asselt KNI, Kok HS, Pearson PL, et al. Heritability of menopausal age in mothers and daughters. Fertil Steri12004;82:1348-1351. 15. Towne B, Czerwinski SA, Demerath EW, et al. Heritability of age at menarche in girls from the Fels Longitudinal Study. Am J Phys dnthropo12005;128:210-219.
16. Simpson JL. Gonadal dysgenesis and abnormalities of the human sex chromosomes: current status of phenotypic-karyotypic correlations. Birth Defects OrigArtic Ser 1975;11(4):23-59.
44 17. Simpson JL. Disorders of sexual differentiation: etiology and clinical delineation. New York: Academic Press, 1976. 18. Loughlin SA, Redha A, McIver J, et al. Analysis of the origin of Turner's syndrome using polymorphic DNA probes. J Med Genet 1991;28:156-158. 19. Mathur A, Stekol L, Schatz D, et al. The parental origin of the single X chromosome in Turner syndrome: lack of correlation with parental age or clinical phenotype. A m J Hum Genet 1991;48:682-686. 20. Lorda-Sanchez I, Binkert F, Maechler M, Schinzel A. Molecular study of 45,X conceptuses: correlation with clinical findings. A m J M e d Genet 1992;42:487-490. 21. James RS, Dalton P, Gustashaw K, et al. Molecular characterization of isochromosomes ofXq.Ann Hum Genet 1997;61:485-490. 22. James RS, Coppin B, Dalton P, et al. A study of females with deletions of the short arm of the X chromosome. Hum Genet 1998;102: 507-516. 23. Ogata T, Matsuo N. Turner syndrome and female sex chromosome aberrations: deduction of the principal factors involved in the development of clinical features. Hum Genet 1995;95:607-629. 24. Speed RM. Oocyte development in XO foetuses of man and mouse: the possible role of heterologous X-chromosome pairing in germ cell survival. Chromosoma 1986;94:115-124. 25. Speed RM. The possible role of meiotic pairing anomalies in the atresia of human fetal oocytes. Hum Genet 1988;78:260-266. 26. Migeon B R, Jelalian K. Evidence for two active X chromosomes in germ cells of female before meiotic entry. Nature 1977;269:242-243. 27. Simpson JL. Pregnancies in women with chromosomal abnormalities. In: Schulman JD, Simpson JL, eds. Genetic diseases in pregnancy. New York: Academic Press, 1981;439-471. 28. Dewhurst J. Fertility in 47,XXX and 45,X patients. J Med Genet 1978;15;132-135. 29. Singh DN, Hara S, Foster HW, Grimes EM. Reproductive performance in women with sex chromosome mosaicism. Obstet Gynecol 1980;55:608-611. 30. Ishikawa M, Hidaka E, Wakui K, et al. [Habitual abortion and high frequency of low frequent X chromosome monosomy mosaicism: detection by interphase FISH analysis of buccal mucosa cells and lymphocytes]. Rinsho Byori 2000;48:955-959. 31. Sybert VP. Turner syndrome. In: Cassidy SB, Allanson J, eds. Management of genetic syndromes. Hoboken, NJ: John Wiley and Sons, 2005;589-605. 32. Hook EB, Hamerton JL. The frequency of chromosome abnormalities detected in consecutive newborn studies: differences between studies - - results by sex and by severity of phenotype involvement. In: Hook EB, Porter IH, eds. Population cytogenetic studies in humans. New York: Academic Press, 1977;63 - 79. 33. Simpson JL. Phenotypic-karyotypic correlations of gonadal determinants: Current status and relationship to molecular studies. In: Sperling K, Vogel F, eds. Human genetics. (Proceedings of 7th International Congress, Berlin, 1986). Heidelberg, Berlin: Springer-Verlag, 1987; 224-232. 34. Simpson JL. Ovarian maintenance determinants on the X chromosome and on autosomes. In: Coutifaris C, Mastroianni L, eds. New horizons in reproductive medicinem--Proceedings of the IXth World Congress on Human Reproduction, Philadelphia, 1996. New York: Parthenon Publishing Group, 1997;439-444. 35. Simpson JL. Genetics of female infertility. In: Filicori M, Flamigni C, eds. Proceedings oft he conference, Treatment of Infertility: The New Frontiers. Boca Raton, FL: Communications Media for Education, 1998;37-52. 36. Simpson JL. Genetics of oocyte depletion. In: Lobo RA, ed. Perimenopause." Serono Symposia USA, Norwdl, Massachusetts. NewYork: Springer, 1997;36-45.
JoE LEIGH SIMPSON 37. Fraccaro M, Maraschio P, Pasquali F, Scappaticci S. Women heterozygous for deficiency of the (p21 leads to pter) region of the X chromosome are fertile. Hum Genet 1977;39:283-292. 38. Simpson JL, Lebeau MM. Gonadal and statural determinants on the X chromosome and their relationship to in vitro studies showing prolonged cell cycles in 45,X; 46,X,del(X)(pll); 46,X, del(X)(q13); and 46,X, del(X)(q22) fibroblasts. Am J Obstet Gyneco11981;141:930- 940. 39. Wilson MG, Modebe O, Towner JW, Frasier SD, Lin MS. UllrichTurner syndrome associated with interstitial deletion of Xpll.4 leads to p22.31.AmJMed Genet 1983;14:567-576. 40. Herva R, Kaluzewski B, de la Chapelle A. Inherited interstitial del(Xp) with minimal clinical consequences: with a note on the location of genes controlling phenotypic features. Am J Med Genet 1979;3:43-58. 41. Krauss CM, Turksoy RN, Atkins L, et al. Familial premature ovarian failure due to an interstitial deletion of the long arm of the X chromosome. N EnglJ Med 1987;317:125 - 131. 42. Fitch N, de Saint V, Richer CL, Pinsky L, et al. Premature menopause due to a small deletion in the long arm of the X chromosome: a report of three cases and a review.AmJ Obstet Gyneco11982;142:968-972. 43. Tharapel AT, Anderson KP, Simpson JL, et al. Deletion (X) (q26.1--+q28) in a proband and her mother: molecular characterization and phenotypic-karyotypic deductions. Am J Hum Genet 1993;52: 463-471. 44. Jones MH, Furlong RA, Burkin H, et al. The Drosophila developmental gene fat facets has a human homologue in Xpll.4 which escapes X-inactivation and has related sequences on Yq11.2. Hum Mol Genet 1996;5:1695-1701. 45. Zinn AR, Ross JL. Molecular analysis of genes on Xp controlling Turner syndrome and premature ovarian failure (POF). Semin Reprod Med 2001;19:141-146. 46. Luoh SW, Bain PA, Polakiewicz RD, et al. Zfx mutation results in small animal size and reduced germ cell number in male and female mice. Development 1996;124:2275-2284. 47. Galloway SM, McNatty K_P, Cambridge LM, et al. Mutations in an oocyte-derived growth factor gene (BMP15) cause increased ovulation rate and infertility in a dosage-sensitive manner. Nat Genet 2000;25: 279-283. 48. Dube JL, Wang P, Elvin J, et al. The bone morphogenetic protein 15 gene is X-linked and expressed in oocytes. Mol Endocrinol 1998;12: 1809-1817. 49. Di Pasquale E, Beck-Peccoz P, Persani L. Hypergonadotropic ovarian failure associated with an inherited mutation of human bone morphogenetic protein-15 (BMP15) gene. AmJHum Genet 2004;75:106-111. 50. Dixit H, Deendayal M, Singh L. Mutational analysis of the mature peptide region of inhibin genes in Indian women with ovarian failure. Hum Reprod 2004;19:1760-1764. 51. Castrillon DH, Wasserman SA. Diaphanous is required for cytokinesis in Drosophila and shares domains of similarity with the products of the limb deformity gene. Development 1994;120:3367-3377. 52. Bione S, Sala C, Manzini C, et al. A human homologue of the Drosophila melanogaster diaphanous gene is disrupted in a patient with premature ovarian failure: evidence for conserved function in oogenesis and implications for human sterility.Am JHum Genet 1998;62:533-541. 53. Bione S, Toniolo D. X chromosome genes and premature ovarian failure. Semin Reprod Med 2000;18:51-57. 54. Prueitt RL, Ross JL, Zinn AR. Physical mapping of nine Xq translocation breakpoints and identification of XPNPEP2 as a premature ovarian failure candidate gene. Cytogenet Cell Genet 2000;89:44-50. 55. Van De Velde-Staquet MF, Breynaert R, Walbaum R, et al. [Progeny of mothers with trisomy 21 (apropos of a case)]. J Genet Hum 1973;21:187-206. 56. Assche EV, Bonduelle M, Tournaye H, et al. Cytogenetic of infertile men. In: Steirteghem AV, Devroey P, Liebaers I, eds. Genetics and assisted human conception. Oxford, UK: Oxford University Press, 1996;1-26.
CHAPTER 3 Genetic Programming in Ovarian Development and Oogenesis 57. Simpson JL, Christakos AC, Horwith M, Silverman FS. Gonadal dysgenesis in individuals with apparently normal chromosomal complements: tabulation of cases and compilation of genetic data. Birth Defects OrigArtic Ser 1971;7(6):215-228. 58. Simpson JL. Gonadal dysgenesis and sex chromosome abnormalities. Phenotypic/karyotypic correlations. In: Vallet HL, Porter IH, eds. Genet# mechanisms of sexual development. New York: Academic Press, 1979;365-405. 59. Matthews CH, Borgato S, Beck-Peccoz P, et al. Primary amenorrhoea and infertility due to a mutation in the beta-subunit of follicle-stimulating hormone. Nat Genet 1993;5:83-86. 60. Layman LC, Lee EJ, Peak DB, et al. Delayed puberty and hypogonadism caused by mutations in the follicle-stimulating hormone beta-subunit gene. N EnglJ Med 1997;337:607-611. 61. Aittomaki K. The genetics of XX gonadal dysgenesis.AmJHum Genet 1994;54:844-851. 62. Aittomaki K, Lucena JL, Pakarinen P, et al. Mutation in the folliclestimulating hormone receptor gene causes hereditary hypergonadotropic ovarian failure. Cell 1995;82:959-968. 63. Layman LC, Amde S, Cohen DP, Jin M, Xie J. The Finnish folliclestimulating hormone receptor gene mutation is rare in North American women with 46,XX ovarian failure. Fertil Steri11998;69:300-302. 64. Liu JY, Gromoll J, Cedars MI, La Barbera AR. Identification of allelic variants in the follicle-stimulating hormone receptor genes of females with or without hypergonadotropic amenorrhea. Fertil Steri11998;70: 326-331. 65. Simoni M, Gromoll J, Nieschlag E. The follicle-stimulating hormone receptor: biochemistry, molecular biology, physiology, and pathophysiology. Endocr Rev 1997;18:739-773. 66. da Fonte Kohek MB, Batista MC, Russell AJ, et al. No evidence of the inactivating mutation (C566T) in the follicle-stimulating hormone receptor gene in Brazilian women with premature ovarian failure. Fertil Steri11998;70:565-567. 67. de la Chesnaye E, Canto P, Ulloa-Aguirre A, Mendez JR No evidence of mutations in the follicle-stimulating hormone receptor gene in Mexican women with 46,XX pure gonadal dysgenesis. Am JMed Genet 2001;98:125-128. 68. Beau I, Touraine P, Meduri G, et al. A novel phenotype related to partial loss of function mutations of the follicle stimulating hormone receptor. J Clin Invest 1998;102:1352-1359. 69. Touraine P, Beau I, Gougeon A, et al. New natural inactivating mutations of the follicle-stimulating hormone receptor: correlations between receptor function and phenotype. MolEndocrino11999;13:1844-1854. 70. Sultan C, Lumbroso S. LH receptor defects. In: Kempers RD, Cohen KJ, Haney AF, Younger JB, eds. Fertility and reproductive medicine--proceedings of the XVI World Congress on Fertility and Sterility. Amsterdam: Elsevier Science, 1998;769-782. 71. Matzuk MM, Lamb DJ. Genetic dissection of mammalian fertility pathways. Nat Cell Bio12002;4:$1:41-49. 72. Latronico AC, Anasti J, Arnhold IJ, et al. Brief report: testicular and ovarian resistance to luteinizing hormone caused by inactivating mutations of the luteinizing hormone-receptor gene. N Engl J Med 1996;334:507-512. 73. Shelling AN, Burton KA, Chand AL, et al. Inhibin: a candidate gene for premature ovarian failure. Hum Reprod 2000;15:2644-2649. 74. Marozzi A, Porta C, Vegetti W, et al. Mutation analysis of the inhibin alpha gene in a cohort of Italian women affected by ovarian failure. Hum Reprod 2002;17:1741-1745. 75. Gui LM, Joyce IM. RNA interference evidence that growth differentiation factor-9 mediates oocyte regulation of cumulus expansion in mice. Biol Reprod 2005;72:195-199. 76. Juengel JL, Hudson NL, Heath DA, et al. Growth differentiation factor 9 and bone morphogenetic protein 15 are essential for ovarian follicular development in sheep. Biol Reprod 2002;67:1777-1789.
45
77. Dong J, Albertini DF, Nishimori K, et al. Growth differentiation factor-9 is required during early ovarian folliculogenesis. Nature 1996;383: 531-535. 78. Dixit H, Rao LK, Padmalatha V, et al. Mutational screening of the coding region of growth differentiation factor 9 gene in Indian women with ovarian failure. Menopause 2005;12:749-754. 79. Di Pasquale E, Beck-Peccoz P, Persani L. Hypergonadotropic ovarian failure associated with an inherited mutation of human bone morphogenetic protein-15 (BMP15) gene. Am J Hum Genet 2004;75: 106-111. 80. Takebayashi K, Takakura K, Wang H, et al. Mutation analysis of the growth differentiation factor-9 and -9B genes in patients with premature ovarian failure and polycystic ovary syndrome. Fertil Steril 2000;74: 976-979. 81. Kovanci E, Rohozinski J, Simpson JL, et al. Growth differentiating factor-9 mutations may be associated with premature ovarian failure. Fertil Steri12007. In Press. 82. Perrault M, Klotz P, Housset E. Deux cas de syndrome de Turner avec surdi-mutite dans une meme fratrie. Bull Soc ivied Hop Paris 1951;67:79-84. 83. Josso N, de Grouchy J, Frezal J, Lamy M. Le syndrome de Turner familial etude de deux families avec caryotypes XO or XX. Ann Pediatr 1963;10:163-167. 84. Christakos AC, Simpson JL, Younger JB, Christian CD. Gonadal dysgenesis as an autosomal recessive condition. Am J Obstet Gynecol 1969;104:1027-1030. 85. Pallister PD, Opitz JM. The Perrault syndrome: autosomal recessive ovarian dysgenesis with facultative, non-sex-limited sensorineural deafness. Am J Med Genet 1979;4:239-246. 86. Nishi Y, Hamamoto K, Kajiyama M, Kawamura I. The Perrault syndrome: clinical report and review. Am J Med Genet 1988;31: 623-629. 87. Simon AM, Goodenough DA, Li E, Paul DL. Female infertility in mice lacking connexin 37. Nature 1997;385:525-529. 88. Skre H, Bassoe HH, Berg K, Frovig AG. Cerebellar ataxia and hypergonadotropic hypogonadism in two kindreds. Chance concurrence, pleiotropism or linkage? Clin Genet 1976;9:234-244. 89. De Michele G, Filla A, Striano S, Rimoldi M, Campanella G. Heterogeneous findings in four cases of cerebellar ataxia associated with hypogonadism (Holmes' type ataxia). Clin Neural Neurosurg 1993;95: 23-28. 90. Linssen WH, Van den Bent MJ, Brunner HG, Pods PJ. Deafness, sensory neuropathy, and ovarian dysgenesis: a new syndrome or a broader spectrum of Perrault syndrome? Am J Med Genet 1994;51: 81-82. 91. Gottschalk ME, Coker SB, Fox LA. Neurologic anomalies of Perrault syndrome. Am J Med Genet 1996;65:274-276. 92. Fryns JP, Van Lingen C, Devriendt K, Legius E, Raus P. Two adult females with a distinct familial mental retardation syndrome: nonprogressive neurological symptoms with ataxia and hypotonia, similar facial appearance, hypergonadotrophic hypogonadism, and retinal dystrophy. JMed Genet 1998;35:333-335. 93. Amor DJ, Delatycki MB, Gardner RJ, Storey E. New variant of familial cerebellar ataxia with hypergonadotropic hypogonadism and sensorineural deafness. Am J Meal Genet 2001;99:29-33. 94. Maximilian C, Ionescu B, Bucur A. [Two sisters with major gonadal dysgenesis, dwarfism, microcephaly, arachnodactyly, and normal karyotype 46,XX].J Genet Hum 1970;18:365-378. 95. Q.uayle SA, Copeland KC. 46,XX gonadal dysgenesis with epibulbar dermoid. AmJMed Genet 1991;40:75-76. 96. Hisama FM, Zemel S, Cherniske EM, Vladutiu GD, Pober BR. 46,XX gonadal dysgenesis, short stature, and recurrent metabolic acidosis in two sisters. Am JMed Genet 2001;98:121-124.
46 97. Zlotogora J, Sagi M, Cohen T. The blepharophimosis, ptosis, and epicanthus inversus syndrome: delineation of two types. Am J Hum Genet 1983;35:1020-1027. 98. Panidis D, Rousso D, Vavilis D, Skiadopoulos S, Kalogeropoulos A. Familial blepharophimosis with ovarian dysfunction. Hum Reprod 1994;9:2034-2037. 99. Crisponi L, Deiana M, Loi A, et al. The putative forkhead transcription factor FOXL2 is mutated in blepharophimosis/ptosis/epicanthus inversus syndrome. Nat Genet 2001;27:159-166. 100. De Baere E, Dixon MJ, Small KW, et al. Spectrum of FOXL2 gene mutations in blepharophimosis-ptosis-epicanthus inversus (BPES) families demonstrates a genotype-phenotype correlation. Hum Mol Genet 2001;10:1591-1600. 101. Harris SE, Chand AL, Winship IM, et al. Identification of novel mutations in FOXL2 associated with premature ovarian failure. Mol Hum Reprod 2002;8:729- 733. 102. Araki S, Chikazawa K, Sekiguchi I, et al. Arrest of follicular development in a patient with 17 alpha-hydroxylase deficiency: folliculogenesis in association with a lack of estrogen synthesis in the ovaries. Fertil Steri11987;47:169-172. 103. Rabinovici J, Blankstein J, Goldman B, et al. In vitro fertilization and primary embryonic cleavage are possible in 17 alpha-hydroxylase deficiency despite extremely low intrafollicular 17 beta-estradiol. J Clin Endocrinol Metab 1989;68:693-697. 104. Simpson ER. Genetic mutations resulting in estrogen insufficiency in the male. Mol Cell Endocrino11998;145:55-59. 105. Ito Y, Fisher CR, Conte FA, Grumbach MM, Simpson ER. Molecular basis of aromatase deficiency in an adult female with sexual infantilism and polycystic ovaries. Proc Natl Acad Sci U S A 1993;90: 11673-11677. 106. Conte FA, Grumbach MM, Ito Y, Fisher CR, Simpson ER. A syndrome of female pseudohermaphrodism, hypergonadotropic hypogonadism, and multicystic ovaries associated with missense mutations in the gene encoding aromatase (P450arom). J Clin Endocrinol Metab 1994;78:1287-1292. 107. Mullis PE, Yoshimura N, Kuhlmann B, et al. Aromatase deficiency in a female who is compound heterozygote for two new point mutations in the P450arom gene: impact of estrogens on hypergonadotropic hypogonadism, multicystic ovaries, and bone densitometry in childhood. J Clin Endocrinol Metab 1997;82:1739-1745. 108. Smith A, Fraser IS, Noel M. Three siblings with premature gonadal failure. Fertil Steri11979;32:528-530. 109. Granat M, Amar A, Mor-Yosef S, Brautbar C, Schenker JG. Familial gonadal germinative failure: endocrine and human leukocyte antigen studies. Fertil Steri11983;40:215-219. 110. A1Awadi SA, Farag TI, Teebi AS, et al. Primary hypogonadism and partial alopecia in three sibs with mullerian hypoplasia in the affected females. Am J Med Genet 1985;22:619-622. 111. Mikati MA, Najjar SS, Sahli IF, et al. Microcephaly, hypergonadotropic hypogonadism, short stature, and minor anomalies: a new syndrome. Am J Med Genet 1985;22:599-608. 112. Sermon K, De Vos A, Van DV, et al. Fluorescent PCR and automated fragment analysis for the clinical application of preimplantation genetic diagnosis of myotonic dystrophy (Steinert's disease). Mol Hum Reprod 1998;4:791- 796. 113. Verlinsky YKA, Kuliev A. Practicalpreimplantation genetic diagnosis. London: Springer, 2005. 114. Kaufman FR, Kogut MD, Donnell GN, et al. Hypergonadotropic hypogonadism in female patients with galactosemia. N Engl J Med 1981;304:994-998. 115. Waggoner DD, Buist NR, Donnell GN. Long-term prognosis in galactosaemia: results of a survey of 350 cases. J Inherit Metab Dis 1990;13:802-818. 116. Levy HL, DriscoU SG, Porensky RS, Wender DE Ovarian failure in galactosemia. N EnglJ Med 1984;310:50.
JoE LEIGH SIMPSON 117. Leslie ND, Yager KL, McNamara PD, Segal S. A mouse model of galactose-l-phosphate uridyl transferase deficiency. Biocbem Mol Med 1995;59:7-12. 118. Bjursell C, Stibler H, Wahlstrom J, et al. Fine mapping of the gene for carbohydrate-deficient glycoprotein syndrome, type I (CDG1): linkage disequilibrium and founder effect in Scandinavian families. Genomics 1997;39:247-253. 119. Matthijs G, Schollen E, Pardon E, et al. Mutations in PMM2, a phosphomannomutase gene on chromosome 16p13, in carbohydratedeficient glycoprotein type I syndrome (Jaeken syndrome). Nat Genet 1997;16:88-92. 120. de Zegher F, Jaeken J. Endocrinology of the carbohydrate-deficient glycoprotein syndrome type 1 from birth through adolescence. Pediatr Res 1995;37:395-401. 121. Kristiansson B, Stibler H, Wide L. Gonadal function and glycoprotein hormones in the carbohydrate-deficient glycoprotein (CDG) syndrome.Acta Paediatr 1995;84:655-659. 122. Sullivan AK, Marcus M, Epstein MP, et al. Association of FMR1 repeat size with ovarian dysfunction. Hum Reprod 2005;20:402-412. 123. Schwartz CE, Dean J, Howard-Peebles PN, et al. Obstetrical and gynecological complications in fragile X carriers: a multicenter study. Am J Med Genet 1994;51:400-402. 124. Allingham-Hawldns DJ, Babul-Hirji R, Chitayat D, et al. Fragile X premutafion is a significant risk factor for premature ovarian failure: the International Collaborative POF in Fragile X study--preliminary data. Am J Med Genet 1999;83:322- 325. 125. Foresta C, Ferlin A, Gianaroli L, Dallapiccola B. Guidelines for the appropriate use of genetic tests in infertile couples. EurJ Hum Genet 2002;127:4645-4654. 126. Soyal SM, Amleh A, Dean J. FIGalpha, a germ cell-specific transcription factor required for ovarian follicle formation. Development 2000; 127:4645 - 4654. 127. Ledent C, Demeestere I, Blum D, et al. Premature ovarian aging in mice deficient for Gpr3. Proc Natl Acad Sci U S A 2005;102: 8922-8926. 128. Rajkovic A, Pangas SA, Ballow D, Suzumori N, Matzuk MM. NOBOX deficiency disrupts early folliculogenesis and oocyte-specific gene expression. Science 2004;305:1157-1159. 129. Rajkovic A, Yan C, Yan W, Klysik M, Matzuk MM. Obox, a family of homeobox genes preferentially expressed in germ cells. Genomics 2002;79:711 - 717. 130. Zhao XX, Suzumori N, Yamaguchi M, Suzumori K. Mutational analysis of the homeobox region of the human NOBOX gene in Japanese women who exhibit premature ovarian failure. Fertil Steril 2005;83:1843-1844. 131. Arraztoa JA, Zhou J, Marcu D, et al. Identification of genes expressed in primate primordial oocytes. Hum Reprod 2005;20:476-483. 132. Starup J, Sele V. Premature ovarian failure. Acta Obstet Gynecol Scand 1973;52:259-268. 133. Austin GE, Coulam CB, Ryan RJ. A search for antibodies to luteinizing hormone receptors in premature ovarian failure. Mayo Clin Proc 1979;54:394-400. 134. Mattison DR, Evans MI, Schwimmer WB, et al. Familial premature ovarian failure. Am J Hum Genet 1984;36:1341 - 1348. 135. Vegetti W, Grazia TM, Testa G, et al. Inheritance in idiopathic premature ovarian failure: analysis of 71 cases. Hum Reprod 1998;13:1796-1800. 136. Pober BR, Zemel S, Hisama FM. 46,XX gonadal dysgenesis, short stature and recurrent metabolic acidosis in two sisters. Am J Hum Genet 1998;63:Al17. 137. Hamet P, Kuchel O, Nowaczynski W, et al. Hypertension with adrenal, genital, renal defects, and deafness. A new familial syndrome. Arch Intern Med 1973;131:563-569. 138. Simpson JL, Elias S. Genetics in obstetrics and gynecology, 3rd ed. Philadelphia: WB Saunders, 2003.
CHAPTER 3 Genetic Programming in Ovarian Development and Oogenesis 139. Nance MA, Berry SA. Cockayne syndrome: review of 140 cases. Am J Med Genet 1992;42:68-84. 140. Sugarman GI, Landing BH, Reed WB. Cockayne syndrome: clinical study of two patients and neuropathologic findings in one. Clin Pediatr 1977;16:225-232. 141. Martsolf JT, Hunter AG, Haworth JC. Severe mental retardation, cataracts, short stature, and primary hypogonadism in two brothers. A m J M e d Genet 1978;1:291-299. 142. Harbord MG, Baraitser M, Wilson J. Microcephaly, mental retardation, cataracts, and hypogonadism in sibs: Martsolf's syndrome. JMed Genet 1989;26:397-400. 143. Hennekam RC, van de Meeberg AG, van Doorne JM, Dijkstra PF, Bijlsma JB. Martsolf syndrome in a brother and sister: clinical features and pattern of inheritance. EurJ Pediatr 1988;147:539- 543. 144. Weemaes CM, Hustinx TW, Scheres JM, et al. A new chromosomal instability disorder: the Nijmegen breakage syndrome. Acta Paediatr &and 1981;70:557-564. 145. Conley ME, Spinner NB, Emanuel BS, Nowell PC, Nichols WW. A chromosomal breakage syndrome with profound immunodeficiency. Blood 1986;67:1251-1256. 146. Chrzanowska KH, Kleijer wJ, Krajewska-Walasek M, et al. Eleven Polish patients with microcephaly, immunodeficiency, and chromosomal instability: the Nijmegen breakage syndrome. Am J Med Genet 1995;57:462-471. 147. Goto M, Tanimoto K, Horiuchi Y, Sasazuki T. Family analysis of Werner's syndrome: a survey of 42 Japanese families with a review of the literature. Clin Genet 1981;19:8-15. 148. Hall JG, Pallister PD, Clarren SK, et al. Congenital hypothalamic hamartoblastoma, hypopituitarism, imperforate anus and postaxial polydactyly--a new syndrome? Part l: clinical, causal, and pathogenetic considerations. Am J M e d Genet 1980;7:47- 74. 149. Starr DG, McClure JP, Connor JM. Non-dermatological complications and genetic aspects of the Rothmund-Thomson syndrome. Clin Genet 1985;27:102-104. 150. Zadik Z, Levin S, Prager-Lewin R, Laron Z. Gonadal dysfunction in patients with ataxia telangiectasia. Acta Paediatr Scand 1978;67: 477-479.
47
151. Waldmann TA, Misiti J, Nelson DL, Kraemer KH. Ataxiatelangiectasis: a multisystem hereditary disease with immunodeficiency, impaired organ maturation, x-ray hypersensitivity, and a high incidence of neoplasia. Ann Intern Med 1983;99:367- 379. 152. German J. Bloom's syndrome. I. Genetical and clinical observations in the first twenty-seven patients. A m J H u m Genet 1969;21:196-227. 153. German J, Bloom D, Passarge E. Bloom's syndrome XI. Progress report for 1983. Clin Genet 1984;25:166-174. 154. German J. Bloom syndrome: a mendelian prototype of somatic mutational disease. Medicine 1993;72:393 - 406. 155. Naguib KK, Sundareshan TS, Bahar AM, et al. Fertility with deletion Xq25: report of three cases; possible exceptions for critical region hypothesis. Fertil Steri11988;49:917-919. 156. Massa G, Vanderschueren-Lodeweyckx M, Fryns JP. Deletion of the short arm of the X chromosome: a hereditary form of Turner syndrome. EurJ Pediatr 1992;151:893-894. 157. Veneman TF, Beverstock GC, Exalto N, Mollevanger P. Premature menopause because of an inherited deletion in the long arm of the X-chromosome. Ferti! Steri! 1991;55:631-633. 158. Schwartz C, Fitch N, Phelan MC, et al. Two sisters with a distal deletion at the Xq26/Xq27 interface: DNA studies indicate that the gene locus for factor IX is present. Hum Genet 1987;76:54-57. 159. Zinn AR, Ouyang B, Ross JL, et al. Del (X)(p21.2) in a mother and two daughters with variable ovarian function. Clin Genet 1997;52:235-239. 160. Zinn AR, Tonk VS, Chen Z, et al. Evidence for a Turner syndrome locus or loci at Xp11.2-p22.1. AmJHum Genet 1998;63:1757-1766. 161. Marozzi A, Dalpra L, Ginelli E, Tibiletti MG, Crosignani PG. FRAXA premutations are not a cause of familial premature ovarian failure. Hum Reprod 1999;14:573-575. 162. Davison RM, Q.uilter CR, Webb J, et al. A familial case of X chromosome deletion ascertained by cytogenetic screening of women with premature ovarian failure. Hum Reprod 1998;13:3039-3041. 163. Susca F, Aoam R, Vucubim M, Louerro G, Guanti G. Xq deletion and premature ovarian failure. Hum Reprod 1999;14:236. 164. Simpson JL, Rajkovic A. Ovarian differentiation and gonadal failure. Am JMed Genet 1999;89:186-200.
This Page Intentionally Left Blank
~HAPTER ~
Basic Biology: Ovarian Anatomy and Physiology GREGORY F. ERICKSON
Universityof California, San Diego, School of Medicine, La Jolla, CA 92093
R. JEFFREY C H A N G Universityof California, San Diego, School of Medicine, La Jolla, CA 92093
I. STATEMENT OF T H E PROBLEM
approximately 65%--that is, from approximately 25% per transfer in women 30 years of age or younger to approximately 9% per transfer in women over the age of 36 (6,7). A similar age-related decrease in female fecundity has been found using therapeutic donor insemination (8) and gamete intrafallopian tube (GIFT) transfer (9). The low fecundity rate continues through approximately 44 years, after which viable pregnancies almost never occur (9-11). These facts, together with the delay in childbearing by women in developed countries, have set the stage for an increase in reproductive problems and disorders attributable to female aging, in particular infertility. A notable consideration is that women between 36 and 44 years of age can exhibit regular menstrual cycles (12). This maintenance of ovarian cyclicity is important because it argues that the decrease in fecundity in these older women is not the result of failure of aged ovaries to produce dominant follicles. Presumably the cyclical activity reflects the ability of these dominant follicles to undergo the physiologic changes that typically occur during selection, ovulation, luteinization, and luteolysis. One of the main fines of evidence in support of this theory is that near normal quantifies of androgen (13), estrogen, and progesterone (14,15) appear to be secreted from aged dominant follicles over the menstrual cycle. These findings indicate that dominant follicles of older women in their late reproductive years are fully capable of expressing near normal
Today, the menopause occurs in most women at about 51 years of age. Demographic studies have demonstrated that the mean life expectancy of women in developed countries (1) has increased from an estimated 45 years in 1850 to 82 years in 1998 (Fig. 4.1). This is an important observation because it indicates that most women today may live up to one-third of their lives postmenopausally; that is, they will live approximately 30 years after the menopause. Clinicians can therefore expect to extend care to increasingly larger numbers of women with advanced reproductive age in whom ovarian dysfunction will be a major cause of infertility and morbidity. If one considers that the vast majority of fertility and gynecologic problems in the aging woman are a direct consequence of the age-related decrease in ovarian reserve (OR), it becomes apparent that the disappearance of primordial follicles is in one of the critical events in the life of all women. One reproductive feature most adversely affected by the age-related decrease in OR is fecundity. The basis for this age-related change is the failure of dominant follicles to release eggs that can undergo normal embryonic development (2-5). This decrease becomes particularly evident in patients older than 36 years of age undergoing in vitro fertilization (IVF), when pregnancy rates (PR) fall sharply, by T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
49
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
ERICI,:SONANn CHANG
50
FIGURE 4.1 Changesin the life expectancyand age of the menopausein women overthe past 150 years. (Reprintedfrom Nachtigall LE. The aging woman. In: SciarraJJ, ed. Gynecology and obstetrics, vol. 1. Philadelphia:J.B. Lippincott, 1995; 2, with permission.)
patterns of steroidogenic activity. By contrast, studies with oocytes from aged dominant follicles have demonstrated the existence of alterations that contribute negatively to pregnancy. For example, investigators have recently found that aneuploidy increases significantly in embryos that develop from oocytes isolated from the mature follicles of women after 35 years (16). Thus, one is led to the conclusion that (a) the endocrine and gametogenic function of dominant follicles can become dissociated in women after 36 years of age and (b) the aberrant expression of cellular responses in the egg would appear to be the basis for the age-related decrease in fecundity. Understanding how the developmental potential of the aged oocyte is altered independently of changes in granulosa and theca cell function is a fundamental question in ovary research. Although relatively little is known about this problem, an interesting role for OR has been suggested from clinical studies that indicate that basal ovarian follicle number, not oocyte age, is the main determinant predicting pregnancy in older women (17,18). That is, the likelihood of pregnancy is highest in older women with sonographic evidence of six or more ovarian follicles compared with those with less than six follicles. Given this relationship, it is not unreasonable to propose that the selective deteriorative changes that occur in the oocytes of older women are either correlated with or casually connected to a significant decrease in OR, rather than aging itself. A fundamental question is: How does this occur?
units of the ovary. Morphologically, each primordial follicle is composed of an outer single layer of squamous epithelial cells, which are termed granulosa or follicle cells, and a small (approximately 15 Ixm in diameter) immature oocyte arrested in the dictyotene stage of meiosis; both cell types are enveloped by a thin, delicate membrane called the basal lamina or basement membrane (Fig. 4.2). By virtue of the basal lamina, the granulosa and the oocyte exist in a microenvironment in which direct contact with other cells does not occur. Although small capillaries are occasionally observed in proximity to primordial follicles, these follicles do not have an independent blood supply (1). The mean diameter of a nongrowing primordial follicle is 29 lxm (19). All the primordial follicles present in a woman's ovaries are formed before birth. Developmentally, the primordial follicles are formed in the cortical cords of the fetal ovaries between the fifth and ninth months of gestation (1). During this period, all the germ cells are stimulated to initiate meiosis. Because the oocytes in the primordial follicles have entered meiosis, almost all oocytes that are capable of participating in reproduction during a woman's life are formed at birth; that is, the human ovaries acquire a lifetime quota of eggs before birth. Soon after primordial follicle formation, some are recruited (activated) to initiate growth. As successive recruitment proceeds over time, the size of the pool of primordial follicles becomes progressively smaller (Fig. 4.3). Between the times of birth and menarche the number of primordial follicles (and thus oocytes) decreases from several million to severn hundred thousand (see Fig. 4.3). As a woman ages, the number of primordial follicles (OR) continues to decline, until at menopause they are difficult to find.
II. T H E PRIMORDIAL FOLLICLE Before addressing this question, we must understand the basic biology of the primordial follicles. The primordial follicles represent a pool of nongrowing follicles from which all dominant preovulatory follicles are selected (1). Thus, primordial follicles are, in a real sense, the fundamental reproductive
FIGURE 4.2
Electronmicrographof a human primordialfollicleshowing oocyte nucleus (N), Balbiani body (*),and granulosa or follicle cells (arrowbeads). (Reprinted from ref. 1, with permission.)
CHAPTER4 Basic Biology: Ovarian Anatomy and Physiology
51 FOLLICULAR DESTINY
7.0. 2,000,000 Total Primordial Follicles
6.0 "~
ATRESIA U
4.0
3.0
2.0 Z
~.o 0-6
400
LH, FSH -r
Ovulation ~
Corpus Lutet~
O.
g
9 tJ], ~ 5
AGE(months pc.) [ BIRTH
10
20
30
40
50
AGE ( y e a r s )
FIGURE 4.3 Changes in the total number of oocytes in the human ovaries during aging. In the embryo at early to mid-gestation, the number of oocytes increases to almost 7 million. Shortly thereafter, the number falls sharply to about 2 million at birth. The enormous loss (approximately 70%) of oocytes in the embryo between 6 and 9 months is caused by apoptosis. The store of eggs continues to diminish with age, until no oocytes are detected in the ovaries at about 50 years of age. (Reprinted from Baker TG. Radiosensitivity of mammalian oocytes with particular reference to the human female. AmJ Obstet Gyneco11971;110:746-761, with permission.)
III. THE ADULT OVARY In this section, we deal with the anatomy and physiology of folliculogenesis as it occurs in normal women during the reproductive years. We focus our attention on the manner in which the developmental program is expressed in a recruited primordial follicle as it develops to the ovulatory stage or dies by atresia. An underlying principle of the human ovaries is that of the 2 million primordial follicles, only 400 or so will complete their development and undergo ovulation and corpus luteum formation; all others (99.9%) will die by atresia after recruitment (Fig. 4.4). Therefore, the very essence of folliculogenesis is selection.
A. Anatomy The adult human ovary is a mass of follicles, luteal tissue, blood vessels, nerves, and connective tissue elements, all of which form a relatively heterogeneous assemblance of histologic units. It is the continuous and progressive change in follicles and corpora lutea that give rise to the cyclical
FIGURE4.4 Folliculogenesis is a highly selective process. Of the 2 million primordial follicles at birth, only 4 or so are brought to ovulation and luteinization by FSH and LH. (Reprinted from Soules MR, Bremner WJ. The menopause and climacteric: endocrinologic basis and associated symptomatology. JAm Geriatr Sac 1982;30:547-561, with permission.)
changes in the menstrual cycle. During the reproductive years, the normal human ovaries are oval-shaped bodies that each measure 2.5 to 5.0 cm in length, 1.5 to 3 cm in width, and 0.6 to 1.5 cm in thickness (1). The medial edge of the ovary is attached by the mesovarium to the broad ligament, which extends from the uterus laterally to the wall of the pelvic cavity. The surface of the ovary is covered by an epithelial layer of cuboidal cells resting on a basement membrane. This layer, termed the germinal or serous epithelium, is continuous with the peritoneum. Underlying the serous epithelium is a layer of dense connective tissue termed the tunica albuginea. The ovary is organized into two principal parts: a central zone, the medulla, which is surrounded by a particularly prominent peripheral zone, the cortex (Fig. 4.5). Embedded in the connective tissue of the cortex are the follicles containing the female gametes, oocytes. The number and size of the follicles vary depending on the age and reproductive state of the female. The existence of follicles of different sizes (primordial, primary, secondary, tertiary [Graafian], and atretic) reflects specific changes associated with their growth, development, and fate. At the end of the follicular phase of the menstrual cycle, the Graafian follicle that reaches maturity distends the ovarian surface to form a stigma, which eventually ruptures with release of the ovum (see Fig. 4.5). After ovulation, the wall of the ovulating follicle develops into an endocrine structure, the corpus luteum. If implantation does not occur, the corpus luteum deteriorates and eventually becomes a nodule of dense connective
52
ERICKSON AND C H A N G
always contain a pool of small Graafian follicles from which a prospective dominant follicle can be selected. Once selected, a dominant follicle typically grows and develops to the preovulatory state. Those follicles that are not selected become committed to die by atresia. 1. ENDOCRINOLOGY OF FOLLICULOGENESI8
FIGURE 4.5 Diagram summarizing the anatomy and histology of the human ovary during the reproductive years. The development of the follicles and corpus luteum occurs within the cortex; the spiral arteries, hilus cells, and autonomic nerves are located in the medulla. (Reprinted from ref. 1, with permission.)
tissue termed the corpus albicans. At the medial border of the cortex is a mass of loose connective tissue, the medulla. This tissue contains a network of convoluted blood vessels and associated nerves that pass through the connective tissue toward the cortex (see Fig. 4.5). A distinct group of testosterone-producing cells, the hilus cells, lies in the medulla at the hilum of the ovary (20). The arterial supply to the ovary originates from two principal sources: One, the ovarian artery, arises from the abdominal aorta; the other is derived from the uterine artery (1). These two vessels, which enter the mesovarium from opposite directions, form an anastomotic trunk and become a common vessel called the ramus ovaricus artery. At frequent intervals this artery gives rise to a series of primary branches, which enter the hilum like teeth on a rake. In the hilum, numerous secondary and tertiary branches are given off to supply the medulla (see Fig. 4.5).
B. Physiology Typically, the human ovaries produce a single dominant follicle that releases a mature egg into the oviduct to be fertilized at the end of the follicular phase of the menstrual cycle. Formation of each dominant follicle begins with recruitment of several primordial follicles into a pool of growing follicles destined to participate in a subsequent ovulatory cycle. It is not known exactly how recruitment occurs, but it appears to be controlled by local ovarian regulatory factors by autocrine/paracrine mechanisms. In a broad sense, all growing follicles can be divided into two groups, healthy and atretic, according to whether or not apoptosis (programmed cell death) is present in granulosa cells (21,22). As a consequence of successive recruitments, the ovaries appear to
Regardless of age, follicle growth and development are brought about by the combined action of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) on the follicle cells. FSH and LH bind to specific high-affinity receptors on the membranes of the granulosa and theca interstitial cells respectively. The interaction of these ligands with their receptors activates signal transduction pathways that stimulate mitosis and differentiation responses in the granulosa and theca cells (23,24). Physiologically, these signaling pathways act in parallel to regulate the expression of specific genes in a precise quantitative and temporal fashion. There are two major endocrine responses associated with foUiculogenesis. The first is that the actions of FSH and LH stimulate the production of large amounts of estradiol by the dominant follicle. This important gonadotropin-dependent response is called the two-gonadotropin, two-cell concept (Fig. 4.6). Because the estradiol response appears to be specific to a dominant follicle, the levels of plasma estradiol can be used as a marker for the growth and viability of the dominant follicle. The second is the marked increase in the production of inhibin B by FSH (25). With respect to aging, observations support the possibility that localized changes in inhibin B production may play a role in the accelerated loss of OR. We will return to this issue later. 2. CHRONOLOGY
In women, foUiculogenesis is a very long process (23). In each menstrual cycle, the dominant follicle that is selected to ovulate originates from a primordial follicle that was recruited to grow about 1 year earlier. Whatever the course of development or the final destiny (atresia or ovulation), all follicles undergo various progressive changes (Fig. 4.7). The very early stages of foUiculogenesis (class 1, primary and secondary; class 2, early tertiary) proceed very slowly. Consequently, it requires 300 days or more for a recruited primordial follicle to complete the preantral or hormone-independent period. The basis for the slow growth is the very long doubling time (approximately 250 hours) of the granulosa cells. When follicular fluid begins to accumulate at the class 2 stage, the Graafian follicle begins to expand relatively rapidly (see Fig. 4.7). As the antral (hormone-dependent) period proceeds, the Graafian follicle passes through the small (classes 3, 4, 5), medium (classes 6, 7), and large (class 8) stages. A dominant follicle that survives to the ovulatory stage requires about 40 to 50 days to complete the whole antral period. Selection of the
CHAPTER4 Basic Biology: Ovarian Anatomy and Physiology
53
FIGURE 4.7 The chronology offolliculogenesis in the human ovary. Folliculogenesis is divided into two major periods, preantral (gonadotropin independent) and antral (FSH dependent). In the preantral period, a recruited primordial follicle develops to the primary/secondary (class 1) and early tertiary (class 2) stage, at which time cavitation or antrum formation begins. The antral period includes the small Graafian (0.9-5 mm, class 4 and 5), medium Graafian (6-10 mm, class 6), large Graafian (10-15 mm, class 7), and preovulatory (16-20 mm, class 8) follicles. Time required for completion of preantral and antral periods is approximately 300 days and 40 days, respectively. Number of granulosa cells, go, follicle diameter, ram; percentage of atresia indicated. (Reprinted from Gougeon A. Dynamics of follicular growth in the human: a model from preliminary results. Hum Reprod 1986;1:81-87, with permission.) FIGURE 4.6 The two-gonadotropin, two-cell concept of follicle estrogen production. (Reprinted from Erickson GE Normal ovarian function. Clin Obstet Gyneco11978;21:31-52, with permission.)
dominant follicle is one of the last steps in the long process of foUiculogenesis. The dominant follicle, which is selected from a cohort of class 5 follicles, requires approximately 20 days to develop to the stage where it undergoes ovulation. Those follicles that are not selected become atretic. Atresia can occur at each stage of Graafian follicle development, but atresia appears most prominent in follicles at the class 5 stage (see Fig. 4.7). 3. SELECTION
The dominant follicle that will ovulate its egg in the next cycle is selected from a cohort of healthy, small Graafian follicles (4.7 + 0.7 mm in diameter) at the end of the luteal phase of the menstrual cycle (23). Morphologically, each cohort follicle contains a fully grown egg, about 1 million granulosa cells, a theca interna containing several layers of theca interstitial cells, and a band of smooth muscle cells in the theca externa (Fig. 4.8). Selection is a quintessential aspect of ovary physiology. It is characterized by a high sustained rate of granulosa mitosis. Shortly after the mid-luteal phase, the granulosa cells in all the cohort class 4 and 5 follicles show a sharp increase (approximately twofold) in the rate of granulosa mitosis (23).
FIGURE 4.8 Diagrammatic representation of a Graafian follicle. (Reprinted from Erickson GE Primary cultures of ovarian cells in serum-free medium as models of hormone-dependent differentiation. Mol Cell Endocrino11983;29:21-49, with permission from Elsevier.)
The first indication that the prospective dominant follicle has been selected is that the granulosa cells of the chosen follicle continue dividing at a fast rate while proliferation slows in the nondominant cohort follicles. Because this distinguishing event is seen at the late luteal phase, it has been concluded that selection occurs at this point in the cycle. As mitosis and follicular fluid accumulation continue (see Fig. 4.7), the dominant follicle grows rapidly during the follicular phase, reaching 6.9 + 0.5 mm at days 1 to 5, 13.7 + 1.2 mm at days 6 to 10, and 18.8 + 0.5 mm at days 11 to 14. In
54
ERICKSON AND CHANG
nondominant follicles, growth and expansion proceed more slowly, and with time, atresia becomes increasingly more evident (see Fig. 4.7). Rarely does an atretic follicle reach more than 9 mm in diameter, regardless of the stage in the cycle. FSH is obligatory for follicle selection, and no other ligand by itself can serve in this regulatory capacity. Physiologically, the mechanism of selection is causally connected to a small but significant secondary rise in plasma FSH during the early follicular phase of the menstrual cycle (the primary FSH rise being the midcycle preovulatory surge of FSH and LH). The secondary rise in plasma FSH begins a few days before the progesterone and estradiol levels reach baseline values at the end of luteal phase, and it continues through the first week of the follicular phase (Fig. 4.9). The importance of the secondary rise in FSH is demonstrated by the fact that the dominant follicle will undergo atresia if the FSH levels are decreased during this time of the menstrual cycle. Consequently, the secondary rise in FSH is obligatory for the selection of a dominant follicle that will ovulate in the next cycle. One of the major consequences of the secondary FSH rise is
(a) ~so
8o
g
that a critical threshold level of FSH accumulates in the follicular fluid of the chosen follicle. In normal class 5 to 8 follicles, the mean concentration of follicular fluid FSH increases from approximately 1.3 mlU/mL (approximately 58 ng/mL) to approximately 3.2 mlU/mL (approximately 143 ng/mL) through the follicular phase (26). In contrast, the levels of FSH are low or undetectable in the microenvironment of the nondominant cohort follicles. Thus, the selection and continued growth of a dominant follicle involves a progressive increase in the concentration of FSH within its microenvironment. Once activated, the dominant follicle becomes dependent on FSH for its survival. The regulation of FSH levels in follicular fluid is totally obscure. The FSH triggers a marked activation of mitosis and differentiation of the granulosa cells, which in turn is reflected in a progressive increase in estradiol and inhibin B synthesis and follicular fluid accumulation (see Fig. 4.9). One of the effects of the increased estradiol and inhibin B production is that the secondary rise in FSH is suppressed (see Fig. 4.9). When this occurs, the concentration of FSH falls below threshold levels and the development of the nondominant follicles stops. It is noteworthy that mitosis in these atretic follicles can be markedly stimulated by treatment with gonadotropin during the early follicular phase. Thus, if FSH levels within the microenvironment are increased, the nondominant follicle could perhaps be rescued from atresia.
C. The Role of FSH
_
(b)
0
,
,
0
1500
5O -J
~
4o .
|174
30
~
2o
lOOO{
TTT
m
e~
500
o. O
(c)
A
"
i
-
9
40-
.
.
.
.
.
.
i
i
i
-
20"
..,I
-,-TTT'r
0 15
T
30"
--J
.I.
-
The major FSH-dependent changes that occur during the development of the dominant follicle are summarized in Fig. 4.10. In women, the granulosa cells are the only cell types known to express FSH receptors. It follows, therefore, that FSH-mediated effects in the dominant follicle are at the level of the granulosa cells. In dominant follicles, the FSH-induced differentiation of the granulosa cells involves three major responses, namely increased steroidogenic potential, mitosis, and LH receptors.
10
.-1"11
u)
v T
10
5
"~'4
0 "-i4
0
Days relative to midcycle LH peak
FIGURE 4.9 The endocrinology of the luteal-follicular transition in women. Data are mean + standard error (SE) of daily serum concentrations of FSH, LH, estradiol, progesterone, and immunoreactive inhibin in women with normal cycles. Note the secondary rise in plasma FSH in the late luteal phase (2 days before menses). (Reprinted from ref. 62, with permission from The Endocrine Society.)
1. STEROIDOGENIC POTENTIAL
The FSH ligand interacts with its receptor on the granulosa cells, and the binding event is transduced into an intracelMar signal via the heterodimeric G proteins (see Fig. 4.10). The FSH-bound receptor activates the ~xGstimulatory (cxGs), which activates adenylate cyclase to generate increases in cyclic adenosine 3',5'-monophosphate (c_AMP), which triggers protein kinase A (PICA) to phosphorylate cAMP-responsive element-binding protein (CREB) or other related DNAbinding proteins. After phosphorylation, these proteins bind to upstream DNA regulatory elements called cyclic-AMP response elements (CRE), where they regulate gene transcription. In this regard the FSH signal mechanisms stimulate
CHAPTER4 Basic Biology: Ovarian Anatomy and Physiology
FSH
55
Adenylate
Cyclase
~lJill
I c~"o';ste;o', l
,,=occ
I Ggneno,one I ~.... [ Progesterone
IWli GTPs~ GDP GDP
~
P~:tein ~ ir(~
i Go;is ]
II
C-AMP ATP
...........
[ Andr~176
]
4,,=oa,o. [Es--~ron__e~
I Estrad,o,
I
LHReceptor FIGURE 4.10 Schematicmodel of the FSH signal transduction pathway in granulosa cells of a dominant follicle leading to increased steroidogenic potential, mitosis, and LH receptor expression.
the expression of specific genes that control the level of estradial production by the granulosa cells (27). The major steroidogenic genes induced by FSH include the P450 aromatase (P450 arom) a~d 17[3-hydroxysteroid dehydrogenase (17 ~3-HSD) (see Fig. 4.10). The temporal pattern of expression of these genes has an important role in generating the normal pattern of estradiol production by the dominant follicle during the follicular phase of the cycle (see Fig. 4.9). FSH also acts on the granulosa cells to increase its potential for luteinization as reflected by in vitro experiments with cultured granulosa cells from human follicles at different stages of development (28). The mechanisms by which this progesterone potential remains suppressed during folliculogenesis in viva remains unknown, but a putative FSH-dependent luteinizing inhibitor has been proposed (28). Recent studies suggest that bone morphogenetic proteins (BMPs) may be physiologic mediators of endogenous progesterone production (29). Within the rat ovary, BMP messenger RNA (mRNA) is expressed in granulosa cells (BMP-2 and -6), theca cells (BMP3b,-4, and -7), and the oocyte (BMP-6 and -15 and GDF-9). In addition, BMP receptor mRNAs have been demonstrated in these tissues as well (30). In cultured rat granulosa cells, BMPs inhibit FSH-induced increases in progesterone production. As such, BMPs are considered to be luteinizing inhibitors. This concept is supported by in viva data demonstrating that the expression of BMP ligands and receptors is low or absent in the corpus luteum during luteinization. Additionally, with the onset of luteolysis the BMP system reexpresses itself as BMP receptor, ALK-6, mRNA, which was
increased in the corpus luteum (31). These findings raise the possibility that BMP signaling pathways may act as paracrine or autocrine regulators of progesterone production and, at least in part, have a functional role in preventing premature luteinization. 2. MITOSIS
The granulosa cells in the dominant follicle have the ability to divide at a relatively rapid rate throughout the follicular phase of the cycle, increasing from about 1 • 106 cells at selection to more than 50 x 106 cells at ovulation (23,24). Despite its overall importance to ovarian physiology, it remains unclear how granulosa proliferation is controlled. There is evidence in humans that FSH stimulates the rate ofgranulosa cell division in viva and in vitro (see Fig. 4.10), but the mechanism by which FSH stimulates mitosis is not understood. 3. INDUCTION OF LH RECEPTOR
The ability of LH/human chorionic gonadotropin (hCG) to activate the ovulatory cascade in the dominant follicle is dependent upon the expression of a large number of LH receptors on the granulosa cells (1). Studies have clearly demonstrated an obligatory role of FSH in the induction of LH receptor (see Fig. 4.10). A key feature of LH receptor expression in the granulosa layer is that it is suppressed throughout most of folliculogenesis. That is, the number of LH receptors remains low in
56
ERICKSONAND CHANG
interstitial cells (TIC), theca lutein cells (TLC), and hilus cells (HC). The TIC and TLC are related to each other by a developmental sequence occurring during folliculogenesis and luteogenesis, a process called the cogenesis (35). The formation of the TIC and TLC involves a developmental process that encompasses both proliferation and differentiation (32,36). Because the cogenesis is accompanied by mitosis, it contributes to total interstitial mass and therefore total androgen potential. LH promotes androgen synthesis through activation of the LH/hCG receptor/cAMP-dependent protein kinase A (PKA) signal transduction pathway (see Fig. 4.11). The heterotrimeric guanine-nucleotide proteins (G proteins) act as transducers that couple LH/hCG bound receptors to adenylate cyclase, which forms the second messenger, cAME cAMP activates PKA, which in turn phosphorylates specific serine and threonine residues on substrate proteins. The phosphorylated proteins generate cytoplasmic and nuclear responses that can lead to increased steroidogenesis. Androstenedione is the principal steroid produced by TIC, and treatment with LH increases its production in a time- and dose-dependent manner (32). This concept explains in part the regulated production of androstenedione in normal women and its overexpression in women with chronically elevated levels of plasma LH. At the molecular level, activation of the LH signaling cascade leads to the stimulation of gene transcription, most notably P450c22 and P450c17 (37). The fact that the level of transcription and translation of these genes increases during folliculogenesis argues that LH-induced differential gene expression plays a physiologic role in androstenedione production by human TIC over the menstrual cycle.
granulosa cells during the early and intermediate stages of dominant follicle growth, but then increases sharply to very high levels at the preovulatory stage. The acquisition of LH receptors implies that when the LH ligand enters the microenvironment of the dominant follicle in the late follicular phase, it can act on the granulosa cells to regulate their function, perhaps even replacing FSH as the principal regulator of granulosa cytodifferentiation.
D. The Role of L H Two hormones, LH and insulin, regulate steroidogenesis in the theca interstitial tissue, and both function as stimulators of androgen production (32). Each hormone interacts with a transmembrane receptor, and the binding event is transduced into an intracellular signal that stimulates transcription and translation of specific steroidal genes (Fig. 4.11). Throughout the life of a woman, LH acts as a critical positive regulator of ovary androgen biosynthesis (32). The LH-receptor interactions in the interstitial cells are critically important in estradiol production by virtue of their ability to promote the production of androstenedione, the P450AROM substrate (see Fig. 4.6). The activation of the LH-receptor signaling pathway in the ovary interstitial cells results in the expression of a battery of genes leading to increased androgen synthesis (see Fig. 4.11). The role of LH in stimulating androgen production has been intensively studied in women because of its involvement in infertility and hyperandrogenism, such as in polycystic ovarian syndrome (PCOS) (33,34). There are three families of interstitial cells in the human ovaries (see Fig. 4.5), the theca
~-holesterol f
LI~IAI~
~ S t A R ~- -i~[Cholestero,
~ii~iie GDP
~9. ..................
GTP
c-AMP ATP
l ," " !P,,KAj.... ,
,41. s S
P450C22 ~ ~"
Insulin
,t...
~
i
/'1"
.... :.._.[.Substrate Po,| Protein_]
" ""
+~
"t"
"~"~...
"" P450c17
[ Pregnen~176 I3~-HSD ~ I Pr~176 I / i'"i 17-OHase~ ~r 17-OHase"/" 3~-HSD
t
17-20lyase~ ~ 17-20lyase l DHEA 1 31--I~ [Androstenedione I 3-HSD ~ 1713-HSD~ = ~ [ Testosterone I
en
FIGURE4.11 Diagramofthe LH and insulinsignaltransductionpathways in ovarianinterstitialcellsleadingto increasedandrostenedioneproduction.
57
CHAPTER4 Basic Biology: Ovarian Anatomy and Physiology It has been known for many years that the rate-limiting step in steroidogenesis involves the translocation of cholesterol from the outer mitochondrial membrane to the inner mitochondrial membrane where it metabolizes to pregnenolone by P450c22. This process is regulated by steroidogenic acute regulatory protein (STAR) (38). An important concept is that StAR is obligatory for LH-induced steroidogenesis. In the human ovary StAR is expressed in the interstitial cells, including the TIC (39). In addition to LH, there is convincing evidence that insulin signaling has a significant role in regulating TIC function in women (32). Insulin receptors with protein tyrosine kinase (PTK) activity have been demonstrated in normal human ovaries. In situ hybridization and immunohistochemical studies have revealed that insulin receptors are expressed in TIC of Graafian follicles (both dominant and cohort) (40). Insulin stimulates androgen production by isolated TIC, and the stimulation is believed to be mediated by the insulin receptor (41). Subsequent co-incubation with an insulin antibody was associated with a markedly subdued androgen response. Thus, activation of the insulin receptor signaling pathway can function alone to increase TIC androgen production, and importantly, the pathway can interact with the LH receptor pathway to further enhance the signals evoked by each receptor (see Fig. 4.11). The cross-talk between the insulin and LH receptor pathways appears to be clinically relevant because of the development of hyperandrogenism in women with insulin resistance and compensatory hyperinsulinemia. The link between insulin resistance and hyperandrogenemia may also involve serine phosphorylation of the insulin receptor as well as that of cytochrome P450c17, which is a microsomal enzyme normally expressed in ovaries and adrenal tissue (42,43). Serine phosphorylation of the insulin receptor occurs at the expense of tyrosine phosphorylation and causes impaired insulin action, thereby accounting for insulin resistance. By comparison, serine phosphorylation ofP450c17 upregulates 17-hydroxylase and 17-20 lyase activity in the ovary, which may explain the enhanced 17-hydroxyprogesterone response to gonadotropin stimulation in women with PCOS compared with normal women (44).
FIGURE 4.12 Comparison of the autocrine-paracrine and endocrine concepts. H, hormone. (Reprinted from Erickson GF. Nongonadotropic regulation of ovarian function: Growth hormone and IGFs. In: Filicori M, Flanigni C, eds, Ovulation induction: basic science and clinical advances, Excerpta Medica International Congress Series. Philadelphia: Elsevier, 1994; 73-84, with permission.)
act in an autocrine/paracrine manner to regulate the timing and degree of hormone-dependent folliculogenesis. This is the autocrine/paracrine or growth factor concept (Fig. 4.12). There are five different classes of growth factors (insulinlike growth factor [IGF], transforming growth factor-f3 [TGF13], TGF-o~, fibroblast growth factor [FGF], and cytokines); all five classes have been described within follicles of human ovaries (45). The principle that arises from all the evidence is that growth factors act by autocrine and paracrine mechanisms to cause plus and minus changes that determine whether a follicle lives or dies. The current challenges are to understand how specific families of growth factors exert control of ovarian functions and how these modulations are integrated into the overall pattern of physiology and pathophysiology during the life of the female.
IV. OVARY RESERVE As discussed earlier, the number of ovarian primordial follicles decreases with age from birth through the menopause (see Fig. 4.3). Importantly, recent studies (46,47) with human ovaries have established the concept that the rate of loss of OR (primordial follicles) is not constant during aging, with a significant accelerated decrease in OR occurring at about 37 years in most women (Fig. 4.13).
E. Intraovarian Control As discussed, the development of the dominant follicle is directed by the endocrine hormones FSH and LH. These ligands bind to receptors that are coupled to the cyclic AMP/PKA signal transduction pathways, which in turn are coupled to differential gene activity in a quantitative and temporal fashion. An important concept to emerge in the past decade is that growth factors, which are themselves products of the ovary, modulate (either amplify or attenuate) FSH and LH action. All growth factors are ligands that can
A. Regulation Morphometric analysis of normal ovaries has demonstrated that the rate of recruitment (initiation of primordial follicle growth) accelerates sharply in women at approximately 38 years of age. Consequently, there is a biexponential decrease in OR in women (46,47). It can be seen (Fig. 4.14) that the number of primordial follicles falls steadily for more than three decades, but when the pool of
ERICKSON AND CHANG
58
I B,RTH I
I MENARCHE !
lO 6
ACCELERATED
§
§
+§
"I
DECREASE,,','
§
§
§
ili
+
It +
lO 5
0
12
37
m+ +
51
+
4.13 Major events that occur in a women's life that impact on fertility and fecundity. (Reprinted from Erickson GE Basic biology: ovarian anatomy and physiology.In: Lobo R, KelseyJ, Marcus R, eds., Menopause: biology andpathobiology. San Diego: Academic Press, 2000; 13-31, with permission.)
~
1 i
I0
+
=.
§
"6
i §
m ~
10 3
I
%,
§ +
Z
=1,:,
l ~ ~ m
i
o
2
|
,
~
m ii o
i0
primordial follicles reaches a critical number of approximately 25,000 at 37.5 +_ 1.2 years, the rate of loss of primordial follicles accelerates approximately twofold. Consequently, the OR is reduced to 1000 primordial follicles at approximately 51 years of age (46,47), which corresponds to the median age at natural menopause (48). If the earlier rate of decrease in primordial follicles persisted, menopause would not be expected until the female reached 71 years of age. Notably, in this natural process the number of primordial follicles within the ovaries of any given woman who reaches 38 years of age is variable; that is, important individual differences in OR exist. As seen in Fig. 4.14, some women reach the critical threshold of 25,000 primordial follicles in their late twenties, whereas others do not reach this threshold until their forties. It seems, therefore, that age alone has limited predictive value for accurately determining a woman's OR. The significance of this variability is demonstrated by the fact that women who continue to menstruate regularly after the age of 45 have 10 times more primordial follicles than do those with irregular menses (49). Further, a greater number of primordial follicles is functionally coupled with a higher pregnancy rate in older women (6,13,17,18). It can be argued, therefore, that OR determines the number of maturing Graafian follicles, which in turn determines menstrual activity, which in turn determines fecundity. In a real sense, OR may be of greater importance than a woman's chronologic age in predicting fertility. If we accept this argument, then OR, not age, would be the fundamental factor in determining the decrease in fecundity at approximately 38 years. Hence the question: W h a t is the underlying mechanism for the accelerated recruitment at approximately 38 years of age? Although the answer to this question is not known, it is reasonable to assume that regulatory molecules are involved. In this regard, there are two possibilities: the decrease of one or more necessary inhibitors and the increase in one or more stimulators. Despite its physiologic importance, very little is known about the mechanisms of recruitment in any species. Evidence from animal studies indicates that the rate of re-
-
§247247
AGE IN YEARS FIGURE
m§
+
o
J
Age (years)
FIGURE 4.14. The age-related decrease in the total number of primordial follicles (PF) within both human ovaries from birth to the menopause. As a result of recruitment (initiation of PF growth), the number of PF decreases progressivelyfrom approximately1,000,000 at birth to 25,000 at 37 years. At 37, the rate of recruitment increases sharply and the number of PF declines to 1000 at menopause, at about 51 years. (Reprinted from ref. 46, with permission.)
cruitment of primordial follicles can be influenced by several regulatory factors (Table 4.1). One particularly interesting result with aging rats is that the rise in plasma FSH following unilateral ovariectomy is associated with a significant reduction in the number of primordial follicles within the ovaries; the most striking feature is that the effect was observed only in old rats (50). These studies support the potentially important concept that increased levels of FSH might function to accelerate the rate of recruitment during aging. Hence the question: To what extent is the accelerated loss of OR in aging women a consequence of increased circulating FSH? This is an important question because we know that a significant elevation of plasma F S H is observed in women when the loss of O R is accelerated at approximately 38 years (11,17,18), and the increased plasma F S H corresponds to the time that fecundity drops. Nonetheless, the question of whether the age-related increase in FSH in women is casually connected to the stimulation of recruitment remains unanswered. TABLE 4.1
Known Modulators of Primordial Follicle Number in Laboratory Animals
Regulator FSH Thymus removal Starvation GH/PRL Morphine sulfate Epidermal growth factor
Effect on OR
Reference
Decrease Decrease Increase Increase Increase Increase
42 43 44 45 46 47
59
CHAPTER4 Basic Biology: Ovarian Anatomy and Physiology Experiments in mice show that the rate of recruitment can be modulated by several factors, including the thymus, restricted food intake, prolactin (PRL) and growth hormone (GH), opiates, and epidermal growth factor (EGF) (see Table 4.1). Experiments with neonatal mice indicate that thymectomy leads to a dramatic loss of primordial follicles by apoptosis (51). Because apoptotic primordial follicles are rarely seen in aging women (46,47), it seems unlikely that a thymus factor plays an important role in the accelerated loss of OR at 38 years. In another study, a 50% reduction in food intake was found to increase the number of primordial follicles, suggesting starvation may increase OR (52). This could be potentially important in humans, but the question of whether starvation elicits a similar effect in women needs to be carefixUy examined. Studies using the sterile Snell dwarf mouse indicate that their ovaries contain significantly more primordial follicles than those of the wild type. Precisely how this occurs is uncertain, but it has been theorized that the endocrine state resulting from chronically low GH and/or PRL might be involved (53). Finally, experiments done in mature mice have shown that the administration of either morphine sulfate (54) or EGF (55) leads to a sustained reduction in the rate of primordial follicle recruitment, followed by an increase in OR. These studies, albeit limited, support the concept that the rate of recruitment can be modulated, either increased or decreased, by regulatory elements. Although the clinical significance of these animal data is unknown, they raise the intriguing idea that it may be possible to slow down the rate of recruitment. If true, these data could have important implications for increasing OR, which in turn could have important implications for fecundity in older women.
the mechanisms underlying accelerated recruitment and decreased fecundity in women after 36 years of age. Therefore, to understand the physiologic mechanism underlying the agerelated increase in FSH, one needs to understand the structure of inhibin and age-related changes in inhibin production in women. Inhibin is a member of the TGF-[3 superfamily (56). Inhibin molecules are composed of two heterodimeric proteins, a common cx subunit and one of two distinct [3 subunits termed [3A and [3B (Fig. 4.15). The two subunits (ci and [3A or [3B) are held together by disulfide bonds producing two different inhibins, termed inhibin A and inhibin B, respectively. By contrast, the activins are built of two types of the [3 subunits generating dimeric proteins called activin A ([3A, [3A), AB ([3A, [3B), or B ([3B, [3B). It should be mentioned here that the differential regulation of cx subunit expression might be expected to have profound effects on the levels of inhibin and activin produced by the ovary; that is, a high and low level of o~ subunit expression would be expected to result in relatively high and low levels ofinhibin and activin production, respectively (see Fig. 4.15). It is now clear that a monotropic rise in FSH occurs in women during aging (12). The rise in FSH, which precedes that of LH by almost 10 years, becomes detectable after 36 years of age (14). A detailed examination of the FSH and LH levels in young and old women over the cycle (57) revealed that serum FSH, but not LH, is significantly elevated in older women throughout the menstrual cycle (Fig. 4.16). It is certainly of interest that the increase in plasma FSH coincides with the accelerated loss in OR. Presumably, some alteration has occurred in the negative feedback mechanism of FSH production in aging women, which is reflected in an increase in plasma FSH levels. The most likely explanation for this observation is that aging in women leads to a significant decrease in inhibin production. Direct evidence that the changing FSH profiles in aging women are accompanied by a concomitant decrease in plasma inhibin during the follicular phase of the cycle has been reported (58). The strong evidence that inhibin exerts a negative
B. Endocrine Parameters At the present time, there is considerable interest in the hypothesis that increased plasma FSH levels causal to decreased ovary inhibin A and B production may be involved in Inhibin A
Inhibins
s
Activin A
Activins
Inhibin B
$ s
s s
Activin AB s $
Activin B $ s
FIGURE 4.15 A model of the inhibin and activin molecules.(Reprinted from Erickson GE Basic biology: ovarian anatomy and physiology.In: Lobo R, KelseyJ, Marcus R, eds., Menopause:biology andpathobiology. San Diego: AcademicPress,2000; 13-31, with permission.)
60
ERICKSON AND CHANG 40.
LH
35-
~
---.,e,.,..--
Okka'
LH
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm=l h M G
(n,,,l 1)
................................................. C C
---.=.---- YoungerLH (n=t't)
4000 -
30
AGE < 30
25
hCG
/
INHIBIN .-
_
r
30 - 3 4
"~ 20
3000 - >
35
C
2000
lo 5 . . . .
15
i
. . . .
i
. . . .
i
. . . .
i
. . . .
il
lOO0
. . . .
,!,
_
I
FSH
I
I
I
I
i
12 6000 _J 0
g :I:
E
,
ESTRADIOL
30
o
30 - 3 4
~-
o
9 35
ca. M . _o
AGE
<
4000
co .,..., (D
o,
laJ E t_
--'-"'-
-' 5
-
i
-I0
. . . .
i
-5
. . . .
i
0
. . . .
i
. . . .
5
i
10
-"
-
-
2000 -
_
-
15
C y d e Days Relative to LH Surge
0
FIGURE 4.16 The daily serum FSH and LH levels throughout the menstrual cycle of 11 women in each group (mean + SE). (Reprinted from ref. 57, with permission from The Society for Gynecologic Investigation.)
feedback effect on pituitary FSH secretion in animals (59,60) supports the theory that decreased ovary inhibin production might be responsible for the increased FSH levels in women after 36 years, which in turn might be responsible for the decreased fecundity that can occur at this time. Direct evidence to support this theory has come from studies (15) showing that women age 35 years or older produce less inhibin in response to exogenous gonadotropin than women less than 35 years (Fig. 4.17). By contrast, no significant differences in plasma estradiol (and progesterone) are detectable in these women (see Fig. 4.17). Studies in normally cycling women have revealed a selective decrease in plasma inhibin levels during the follicular phase of the menstrual cycle, beginning at approximately 36 years of age (15,58). It should be noted that the fully processed dimeric inhibin A has been shown to be the predominant circulating form in women before and after treatment with menopause gonadotropin (61). Thus, a functional link between aging in women and decreased expression of ovary inhibin A is suggested (see Fig. 4.9). However, inhibin studies (62) in normal women over the cycle indicate that important changes in inhibin B can be detected during the luteal-follicular transition (Fig. 4.18). Indeed, Klein et al. (63)
I
I
-5
-4
-3
1
-2
I
-1
I
O
I
§
17*
D a y of t r e a t m e n t
FIGURE 4.17 The selective decrease in the inhibin response during ovarian hyperstimulation with human menopause gonadotropin in aging women. The inhibin levels, but not estradiol levels, are significantly lower in women 35 years or older. (Reprinted from ref. 15, with permission from The Endocrine Society.)
have presented evidence for a role of decreased inhibin B in the monotropic FSH rise in aging women. From all these data, it seems reasonable to propose that a decreased ability of the ovaries to produce inhibin B (and perhaps inhibin A as well) may be the underlying cause of the monotropic rise in FSH in women after approximately 35 years of age. The question of whether these changes in inhibin and FSH negatively affect the egg is not understood. However, the data fit with a prediction that the decrease in inhibin, which begins around the time of the accelerated recruitment at 37 years of age, may be involved, directly or indirectly, in the mechanisms that cause poor oocyte quality in aging women. What cells in the ovary are responsible for inhibin production? Studies using in situ hybridization and immunohistochemistry have documented the tissue-specific expression of the el, f3A, and 13B subunits of inhibin in normal human ovaries over the menstrual cycle (64,65). Yamoto et al. (64) found that the three inhibin subunit proteins are selectively expressed in the granulosa cells of
61
CI-IAVTV.V,4 Basic Biology: Ovarian Anatomy and Physiology
(=) T
40-
- 75
_=. z=
lo .~ 0
"
l
... l
v
,
I
,
, " 0
(b) 76-
l
-8OO
5v
4 32-
1" JL
~
"-
-~
-2oo
""
~
i
:L
i-
o
Days relative to FSH peak
FIGURE 4.18 Plasmaconcentrations ofinhibin A and B and estradiol and FSH during the luteal-follicular transition in normal cyclingwomen. Data were aligned with respect to the day of the intercycle FSH peak (mean __+SE). (Reprinted from ref. 62, with permission from The Endocrine Society.)
growing follicles; however, there are important differences between the preantral and Graafian follicle stages. Here, the most striking difference is that granulosa cells in preantral follicles express relatively high levels of the [3A and [3B proteins, but the o~ subunit protein appears undetectable (64). By contrast, the granulosa ceils in the healthy antral follicles express all three subunit proteins, and the levels of el, [3A, and [3B proteins become very high in the dominant follicle, particularly at the preovulatory stage (64). By virtue of the different pattern of (x subunit expression during folliculogenesis, it would appear that granulosa cells in preantral follicles (primary, secondary, and early tertiary) produce activin, whereas those in healthy GraafJan (antral) follicles produce inhibin. The question of whether the dominant follicle produces activin remains to be answered. An in vivo study on inhibin secretion by human ovaries presents a compelling case that the entire pool of healthy Graafian follicles synthesize and secrete inhibin (66). First, the concentration of inhibin is higher in the ovarian veins than the peripheral circulation during the normal menstrual cycle. This finding indicates that plasma inhibin comes from the ovaries. Second, the amount of inhibin secreted during the late follicular phase is similar in the veins of both ovaries (66). Therefore, it seems likely that the Graafian follicles in both ovaries (dominant and nondominant) secrete inhibin during the follicular phase. These data implicate all follicles
(healthy and atretic) in the production of inhibin during the follicular phase of the menstrual cycle; that is, peripheral inhibin levels reflect the total number of developing Graafian follicles. It is possible that this concept could have relevance to the lower levels of inhibin seen in older women. It has been shown that women exhibit an age-related decrease in the total number of Graafian follicles in the ovaries. Collectively, these observations offer a plausible explanation for the reduced levels of circulating inhibin after 36 years of age; that is, fewer Graafian follicles in turn results in lower plasma inhibin levels, which in turn lead to increased FSH levels. Another possible explanation is that a decrease in the expression of the ci and/or the [3 subunits in the granulosa cells plays a role in the lower level of circulating inhibin in women after 36 years of age. This possibility implies that an age-related defect or alteration occurs in the granulosa cells, which leads to the underexpression of inhibin (but not estradiol) after approximately 35 years. Indeed there is evidence from studies with cultured granulosa lutein cells to support this idea (67,68)
V. ACCELERATED LOSS IN OR: THE ACTIVIN HYPOTHESIS It is now of interest to discuss the potential causal connection between the monotropic rise in FSH and the mechanisms underlying the accelerated loss of OR at approximately 38 years of age. The existence of [3A and [3B proteins in the granulosa cells of secondary and early tertiary follicles argues that these proteins serve some function in preantral folliculogenesis in humans. The fact that the subunit protein appears undetectable in these follicles suggests that they may dimerize to form activin (64). Therefore, one could hypothesize that activin may be an autocrine/ paracrine regulator of preantral folliculogenesis. This is of interest because activin appears to be a potent inducer of FSH receptors in granulosa cells (69). Furthermore, it has recently been shown that activin accelerates folliculogenesis (70,71). These observations support the possibility that activin may play a part in the accelerated loss of OR through increasing granulosa FSH sensitivity, which could in turn may play a role in the pathogenesis of the egg in old dominant follicles by causing a premature overexpression of its development. How might this occur? Three different isoforms of activin (see Fig. 4.15) have been isolated from porcine follicular fluid (72-74) and shown to be disulfide-linked homodimers of the inhibin [3A subunit (activin A, Mr 24,000), the [3B subunit (activin B, Mr 22,000), or a heterodimer composed of a [3A and [3B subunit protein (activin AB, Mr 23,000). The isoforms are present in equimolar concentrations in follicular fluid pooled from all antral follicles (74). So far,
62
ERICKSON AND C H A N G
there is no evidence for activin in follicular fluid of dominant follicles. Originally, activin was found to be an stimulator of FSH secretion in vitro (72,73) and in vivo (74-78); however, subsequent studies demonstrated that activin exerts a wide range of positive and negative effects in many different target cells (79). Activin achieves these effects by binding to a novel family of transmembrane receptors with protein serine/threonine kinase activity (80). In women, plasma levels of free activin are low and do not change substantially during the cycle (81). Thus, it seems likely that activin regulates follicular function physiologically by autocrine/paracrine mechanisms. It has been shown that developing follicles indeed produce and respond to activin. As discussed earlier, the [3A and [3B subunits are selectively expressed in human granulosa cells of healthy follicles between the secondary and preovulatory stages (64,65). It seems likely that in the absence of the oL subunit, activin may function in initiating or maintaining the growth and development of preantral follicles during the gonadotropin independent stages of folliculogenesis. Studies in the rat have shown that FSH can stimulate activin expression in granulosa cells in vivo (82,83), and convincing evidence that rat granulosa cells from preantral follicles actually secrete dimeric activin has been reported (84). Further, the mRNAs for activin receptor subtype II (Act R II and Act R IIB) have been identified in rat follicles (85,86), being present in the oocytes and granulosa cells (87). Moreover, specific binding of radiolabeled activin to these cells has been
demonstrated (88-90). Collectively, these results support the hypothesis that human granulosa cells in preantral follicles may produce and respond to activin, and importantly, this process may be amplified by FSH. Much of our understanding of the biologic effects of activin in the ovary has come from studies in laboratory animals. There is evidence suggesting that the autosecretion of activin may play a role in regulating follicle growth and development. Most striking is the observation that activin is a potent stimulator of FSH receptor expression in rat granulosa cells (86,91). Thus far, activin is the only ligand known to induce FSH receptors. This may have relevance to the acquisition of FSH receptors in the granulosa cells, which occurs early in preantral follicle development, at the primary and secondary stages (92,93). Another important effect of activin is that it can prevent FSH-induced receptor downregulation (94). Therefore, the concept emerging is that activin produced by the granulosa cells themselves might play an important physiologic role in the induction and maintenance of FSH receptors in the granulosa cells during folliculogenesis (Fig. 4.19). How might this situation impact OR and fecundity in aging women? Because FSH stimulates activin production and the FSH levels are elevated in women after 36 years of age, one could postulate that these two elements might act synergistically to accelerate the rate of granulosa cytodifferentiation and folliculogenesis in aging ovaries with respect to OR. Accordingly, one could propose the following cas-
~ ~ T B
~CL&SB
AAAAAA r / GTP cAMP
ATP
~176
Dw~t~,]~iTIATION
~ GqpJun~o~ ~ ~H ~ r
~mmTRAL G S , ~ t ~ S A
CELL
FIGURE 4.19 The activin autocrine hypothesis for accelerated granulosa cytodifferentiation and folliculogeneis. (Reprinted from Erickson GF. Dissociation of endocrine and gametogenic ovarian function. In: Lobo R, ed, Perimenopause. Serono Symposia, New York: Springer-Verlag, 1997; 101-118, with kind permission of Springer Science and Business Media.)
CHAPTER4 Basic Biology: Ovarian Anatomy and Physiology cade process. The granulosa cells in preantral follicles synthesize, secrete, and respond to intrinsic activin. One major response to the autosecretion of activin is the expression and maintenance of FSH receptors. The relatively high level of FSH after 36 years has a stimulatory effect on the autocrine activin mechanism. This results in a synergistic interaction between the two signal transduction pathways, which leads to accelerated growth and differentiation responses in the granulosa cells. In this hypothesis, the relatively high amounts of activin could have a strong stimulatory effect on oocyte development in the presence of high FSH. These potent stimulatory effects are then theorized to produce an "overripe" egg lacking a normal meiotic spindle in the aged dominant follicle. Clearly, further work is needed to test the validity of this new hypothesis. Nonetheless, this idea is consistent with the data of Gougeon et al. (47) showing an accelerated loss of developing preantral follicles in women after 37 years. Further, the facts that activin and FSH interact to markedly increase LH receptor (95) and estradiol production (96) in rat granulosa cells and that activin can accelerate meiotic maturation (97) and promote antrumlike formation (71) (i.e., accelerate FSH-dependent granulosa cytodifferentiation) are consistent with this hypothesis. There is evidence that activin can influence physiologic responses in vivo. Doi et al. (98) found that FSH action in vivo can be amplified by exogenous activin. That is, injected activin enhances follicle growth, FSH receptor, number, and estradiol production in intact and hypophysectomized immature rats. These observations are important because they demonstrate that the positive effects of activin on granulosa and other follicle processes observed in vitro will also occur in vivo. Therefore, these results further support the contention that interactions between the autocrine growth factor, activin, and the elevated FSH levels might have a strong stimulatory effect on granulosa cells, which leads to the acceleration of follicle growth and development after 36 years. Interestingly, the fact that the length of the follicular phase of the menstrual cycle is significantly shorter in older women (17,18) fits with this prediction. It should be mentioned that negative effects of activin on folliculogenesis have also been reported. Foremost is the study by Woodruff et al. (90), who showed that activin injection into the ovary bursa of immature rats caused oocyte degeneration, granulosa pyknosis, and decreased mitosis. Therefore, it is also possible that high levels of activin might induce atresia and trigger oocyte demise in the rat.
VI. N E W DATA ON THE EFFECTS OF ACTIVIN Our recent results regarding activin action in adult cycling rats are relevant to this new hypothesis (70). We found that the administration of recombinant human activin A to rats
63 produced dramatic structure/function changes in folliculogenesis. The most striking results are as follows. First, activin stimulated a twofold increase in the number of large Graafian follicles during the follicular phase of the cycle. The data suggested that activin increased the size of the pool of early tertiary preantral follicles and their growth and development to the preovulatory stage (70). Interestingly, nearly all these large follicles contained apoptotic granulosa cells, and therefore they were classified as atretic (see Fig. 4.4). Based on these results, we conclude that activin provides a multifunctional stimulus in vivo that includes both the stimulation and inhibition of follicle cell activities. Second, these large atretic follicles ovulated prematurely, approximately 24 hours earlier than normal. Histologically, the ovulatory changes evoked by activin paralleled those described for normal physiologic ovulation: thecal swellings, the initiation of germinal vesicle breakdown, cumulus expansion, stigma formation, release of egg cumulus complexes, and morphologic luteinization of the follicle wall (70). These observations provide the first evidence that a ligand, namely activin, can significantly shorten (by 25%) the length of the follicular phase of the normal estrous cycle. This necessarily implies that dominant follicle development and ovulation were accelerated in response to activin administration. Third, we found that the activin-exposed eggs in the oviducts and in the large ovulating follicles were arrested in metaphase I and appeared degenerate (see Fig. 4.5). This finding confirms and extends other studies showing that activin acts in the rat ovary to negatively affect oocyte quality. There is evidence that Act RII receptors are strongly expressed in the rat oocytes (87) and that activin can accelerate meiotic maturation in isolated rat oocytes (97). Therefore, this negative action of activin might be mediated by the activin signaling pathway present in the rat egg. The mechanisms and the physiologic/pathophysiologic implications for the multifunctional actions of activin remain to be elucidated. Nevertheless, our observations support the proposition that the autosecretion of activin may contribute to the acceleration of follicle development, which could result in the premature ovulation of overripe eggs in cycling women by autocrine/paracrine mechanisms.
VII. CONCLUSION From the preceding discussion, it is clear that the primary problem in the dominant follicle that leads to reduced fecundity in older women is the susceptibility of the egg to meiotic nondis]unction and aneuploidy. A potentially important theory to explain the problem was developed in this discussion. The evidence indicating that an age-related decrease in the production of ovary inhibin leads to a monotropic rise in FSH, which in turn is reflected in the acceleration of the loss of OR by virtue of accelerating the rate of
64
ERICKSON AND CHANG
recruitment, was discussed. Further, it was suggested that specific interactions between granulosa-derived activin and increased FSH receptor and ligand may act synergistically to further accelerate the rate of granulosa and oocyte cytodifferentiation. This functional response might then lead to accelerated development of the dominant follicle, which in turn is reflected in the age-related shortening of the follicular phase. At the level of the oocyte, these changes are reflected in an increased potential for aneuploidy.
References 1. Erickson GE The ovary: basic principles and concepts. In: Felig P, Baxter JD, Broadus AE, Frohman LA, eds. Endocrinology and metabolism, 3rd ed. New York: McGraw-Hill, 1995;973-1015. 2. Sauer MV, Paulson RJ, Lobo RA. A preliminary report on oocyte donation extending reproductive potential to women over 40. N EnglJ Med 1990;323:1157-1160. 3. Navot D, Bergh PA, Williams MA, et al. Poor oocyte quality rather than implantation failure as a cause of age-related decline in female fertility. Lancet 1991;337:1375-1377. 4. Sauer MV, Paulson RJ, Lobo RA. Pregnancy after age 50: application of oocyte donation to women after natural menoapause. Lancet 1993 ;341:321-323. 5. Sauer MV, Miles RA, Dahmoush L, et al. Evaluating the effect of age on endometrial responsiveness to hormone replacement therapy: a histologic ultrasonographic, and tissue receptor analysis.JAssist Reprod Genet 1993;10:47-52. 6. Padilla SL, Garcia JE. Effect of maternal age and number of in vitro fertilization procedures on pregnancy outcome. Fertil Steril 1989; 52:270-273. 7. Piette C, de Mouzon J, Bachelot A, Spira A. In-vitro fertilization: influence of women's age on pregnancy rates. Hum Reprod 1990;5: 56-59. 8. Cecos F, Schwartz D, Mayaux MJ. Female fecundity as a function of age. N EnglJ Med 1982;306:404-406. 9. Q.asim SM, Karacan M, Corsan GH, Shelden R, Kemmann E. Highorder oocyte transfer in gamete intrafallopian transfer patients 40 or more years of age. Fertil Steri11995;64:107-110. 10. Penzias AS, Thompson IE, Alper MM, Oskowitz SP, Berger MJ. Successful use of gamete intrafallopian transfer does not reverse the decline in fertility in women over 40 years of age. Obstet Gynecol 1991;77: 37-39. 11. Wood C, Calderon I, Crombie A. Age and fertility: results of assisted reproductive technology in women over 40 years.J Assist Reprod Genet 1992;9: 482-484. 12. Sherman BM, Korenman SG. Hormonal characteristics of the human menstrual cycle throughout reproductive fife. J Clin Invest 1975;55: 699-706. 13. Navot D, Rosenwaks Z, Margofioth EJ. Prognostic assessment of female fecundity. Lancet 1987;2:645-647. 14. Lee SJ, Lenton EA, Sexton L, Cooke ID. The effect of age on the cyclical patterns of plasma LH, FSH, estradiol and progesterone in women with regular menstrual cycles. Hum Reprod 1988;3:851-855. 15. Hughes EG, Robertson DM, Handelsman DJ, et al. Inhibin and estradiol responses to ovarian hyperstimulation: Effects of age and predictive value for in vitro fertilization outcome. J Clin Endocrinol Metab 1990;70:358-364. 16. Munne S, Alikani M, Tomldn G, Grifo J, Cohen J. Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil Steri11995;64:382-391.
17. Frattarelli JL, Levi AJ, Miller BT, Segars JH. A prospective assessment of the predictive value of basal antral follicles in in vitro fertilization cycles. Fertil Steri12003 ;80:350-3 5 5. 18. Nahum R, Shifren JL, Chang Y, et al. Antral follicle assessment as a tool for predicting outcome in IVY--is it a better predictor than age and FSH? JAssist Reprod Genet 2001;18:151-155. 19. Gougeon A, Chainy GBN. Morphometric studies of small follicles in ovaries of women at different ages. J Reprod Ferti11987;81:433-442. 20. Erickson GF, Magoffin DA, Dyer CA, Hofeditz C. The ovarian androgen producing cells: A review of structure/function relationships. Endocr Rev 1985;6:371-399. 21. Hsueh AJ, Billig H, Tsafriri A. Ovarian follicle atresia: a hormonally controlled apoptotic process. Endocr Rev 1994;15:707-724. 22. Erickson GE Defining apoptosis: players and systems. J Soc Gynecol Investig 1997;4:219-228. 23. Gougeon A. Regulation of ovarian follicular development in primates: facts and hypotheses. Endocr Rev 1996;17:121-155. 24. McNatty KP, Moore-Smith D, Osathanondh R, Ryan KJ. The human antral follicle: functional correlates of growth and atresia. Ann Bid Anim Biochim Biophys 1979;19:1547-1558. 25. Groome NP, Ilfingworth PJ, O'Brien M, et al. Detection of dimeric inhibin throughout the human menstrual cycle by two-site enzyme immunoassay. Clin Endocrinol 0xf1994;40:717-723. 26. McNatty KP, Hunter WM, McNeilly AS, Sawers RS. Changes in the concentration of pituitary and steroid hormones in the follicular fluid of human graafian follicles throughout the menstrual cycle.J Endocrino11975;64:555-571. 27. Richards JS. Hormonal control of gene expression in the ovary. Endocr Rev 1994;15:725-751. 28. Channing CE Influences of the in vivo and in vitro hormonal environment upon luteinization of granulosa cells in tissue culture. Recent Prog Horm Res 1970;26:589-622. 29. Shimasaki S, Zachow RJ, Li D, et al. A functional bone morphogenetic protein system in the ovary. Proc Natl Acad Sci U S gt 1999;96: 7282-7287. 30. Shimasaki S, Moore RK, Otsuka F, Erickson GF. The bone morphogenetic protein system in mammalian reproduction. Endocr Rev 2004;25:72-101. 31. Yi SE, LaPolt PS, Yoon BS, et al. The type I BMP receptor BmprIB is essential for female reproductive function. Proc Natl Acad Sci U S A 2001 ;98:7994-7999. 32. Erickson GE Normal regulation of ovarian androgen production. Sem Reprod Endocrino11993;11:307-312. 33. Erickson GF, Yen SSC. The polycystic ovary syndrome. In: Adashi E, Leung PKC, eds. The ovary. New York: Raven Press, 1993;561-579. 34. Erickson GE PCO: the ovarian connection. In: Adashi EY, Rock JA, Rosenwaks Z, eds. Reproductive endocrinology, surgery, and technology. New York: Raven Press, 1995;1141-1160. 35. Magoffin DA, Erickson GF. Control systems of theca-interstitial cells. In: Findlay JK, ed. Molecular biology of the female reproductive system. New York: Academic Press, 1994;39-65. 36. Erickson GF, Magoffin DA, Dyer C, Hofeditz C. The ovarian androgen producing cells: a review of structure/function relationships. Endocr Rev 1985;6:371-399. 37. Miller WL. Molecular biology of steroid hormone synthesis. Endocr Rev 1988;9:295-318. 38. Clark BJ, Wells J, King SR, Stocco DM. The purification, cloning, and expression of a novel luteinizing hormone-induced mitochondrial protein in MA-10 mouse Leydig tumor cells. J Bid Chem 1994;269: 28314-28322. 39. Kiriakidou M, McAllister JM, Sugawara T, Strauss JFr. Expression of steroidogenic acute regulatory protein StAR in the human ovary.J Clin Endocrinol Metab 1996;81:4122-4128.
CHAeTEI~ 4 Basic Biology: Ovarian Anatomy and Physiology 40. E1-Roeiy A, Chen X, Roberts VJ, et al. Expression of insulin-like growth factor-I (IGF-I) and IGF-II and the IGF-I, IGF-II, and insulin receptor genes and localization of the gene products in the human ovary. J Clin Endocrinol Metab 1993;77:1411-1418. 41. Nestler JE, Jakubowicz DJ, de Vargas AF. Insulin stimulates testosterone biosynthesis by human thecal cells from women with polycystic ovary syndrome by activating its own receptor and using inositolglycan mediators as the signal transduction system. J Clin Endocrinol Metab 1998;83:2001-2005. 42. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev 1997;18: 774-800. 43. Li M, Youngren JF, Dunaif A, et al. Decreased insulin receptor (IR) autophosphorylation in fibroblasts from patients with PCOS: effects of serine kinase inhibitors and IR activators. J Clin Endocrinol Metab 2002;87: 4088-4093. 44. Ehrmann DA, Barnes RB, Rosenfield RL. Polycystic ovary syndrome as a form of functional ovarian hyperandrogenism due to dysregulation of androgen secretion. Endocr Rev 1995;16:322-353. 45. Adashi E, Leung PCK. The ovary: comprehensive endocrinology. New York, Raven Press, 1993. 46. Faddy MJ, Gosden RG, Gougeon A, Richardson SJ, Nelson JE Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause. Hum Reprod 1992;7:1342-1346. 47. Gougeon A, Ecochard R, Thalabard JC. Age-related changes of the population of human ovarian follicles: increase in the disappearance rate of non-growing and early-growing follicles in aging women. Biol Reprod 1994;50:653-663. 48. McKinlay SM, Brambilla DJ, Posner JG. The normal menopause transition. Maturitas 1992;14:103-115. 49. Richardson SJ, Senikas V, Nelson JF. Follicular depletion during the menopausal transition: evidence for accelerated loss and ultimate exhaustion. J Clin Endocrinol Metab 1987:65:1231-1237. 50. Meredith S, Dudenhoeffer G, Butcher RL, Lerner SP, Walls T. Unilateral ovariectomy increases loss of primordial follicles and is associated with increased metestrous concentration of follicle-stimulating hormone in old rats. Biol Reprod 1992;47:162-168. 51. Lintern-Moore S. Effect of athymia on the initiation of follicular growth in the rat ovary. Biol Reprod 1977;17:155-161. 52. Lintern-Moore S, Everitt AV. The effect of restricted food intake on the size and composition of the ovarian follicle population in the Wistar rat. Biol Reprod 1978;19:688-691. 53. Howe E, Lintern-Moore S, Moore GP, Hawkins J. Ovarian development in hypopituitary Snell dwarf mice. Biol Reprod 1978;19:959-964. 54. Lintern-Moore S, Supasri Y, Pavasuthipaisit K, Sobhon R Acute and chronic morphine sulfate treatment alters ovarian development in prepuberal rats. Biol Reprod 1979;21:379-383. 55. Lintern-Moore S, Moore GPM, Panaretto BA, Robertson D. Follicular development in the neonatal mouse ovary: effect of epidermal growth factor. Acta Endocrinol Copenh 1981;96:123-126. 56. Massagu~ J, Attisano L, Wrana JL. The TGF-b family and its composite receptors. Trends Cell Bio11994;4:172-178. 57. Klein NA, Battaglia DE, Clifton DK, Bremner WJ, Soules MR. The gonadotropin secretion pattern in normal women of advanced reproductive age in relation to the monotropic FSH rise.J Soc GynecolInvestig 1996;3:27-32. 58. Lenton EA, DeKretser DM, Woodward AJ, Robertson DM. Inhibin concentrations throughout the menstrual cycles of normal, infertile, and older women compared with those during spontaneous conception cycles. J Clin Endocrinol Metab 1991;73:1180-1190. 59. Rivier C, Vale W, Rivier J. Studies of the inhibin family of hormones: a review. Horm Res 1987;28:104-118. 60. Rivier C, Vale W. Immunoneutralization of endogenous inhibin modified hormone secretion and ovulation rate in the rat. Endocrinology 1989;125:152-157.
65 61. Muttukrishna S, Fowler PA, Groome NP, et al. Serum concentrations of dimeric inhibin during the spontaneous human menstrual cycle and after treatment with exogenous gonadotrophin. Hum Reprod 1994; 9:1634-1642. 62. Groome NP, Illingworth PJ, O'Brien M, et al. Measurement of dimeric inhibin B throughout the human menstrual cycle. J Clin Endocrinol Metab 1996;81:1401-1405. 63. Klein NA, Illingworth PJ, Groome NP, et al. Decreased inhibin B secretion is associated with the monotropic FSH rise in older, ovulatory women: a study of serum and follicular fluid levels of dimeric inhibin A and B in spontaneous menstrual cycles. J Clin Endocrinol Metab 1996;81:2742-2745. 64. Yamoto M, Minami S, Nakano R, Kobayashi M. Immunohistochemical localization of inhibin/activin subunits in human ovarian follicles during the menstrual cycle. J Clin Endocrinol Metab 1992;74:989-993. 65. Roberts VJ, Barth S, E1-Roeiy A. Expression of inhibin/activin subunits and follistatin messenger ribonucleic acids and proteins in ovarian follicles and the corpus luteum during the human menstrual cycle. J Clin Endocrinol Metab 1993;77:1402-1410. 66. Illingworth PJ, Reddi K, Smith KB, Baird DT. The source of inhibin secretion during the human menstrual cycle. J Clin Endocrinol Metab 1991;73:667-673. 67. Seifer DB, Gardiner AC, Lambert-Messerlian, G, Schneyer AL. Differential secretion of dimeric inhibin in cultured luteinized granulosa cells as a function of ovarian reserve. J Clin Endocrinol Metab 1996;81:736-739. 68. Seifer DB, Gardiner AC, Lambert-Messerlian G, Schneyer A. Differential secretion of dimeric inhibin in cultured luteinized granulosa cells as a function of ovarian reserve.J Clin Endocrinol Metab 1996;81: 736-739. 69. Dain LB, Stein P, Krimer ARD, et al. Progesterone production in cultured human granulosa cells: correlation with follicular fluid hormone levels. Fertil Steri11991;58:1093-1098. 70. Erickson GF, Kokka S, Rivier C. Activin causes premature superovulation. Endocrinology 1995;136:4804-4813. 71. Li R, Phillips DM, Mather JR Activin promotes ovarian follicle development in vitro. Endocrinology 1995;136:849-856. 72. Vale W, Rivier J, Vaughan J, et al. Purification and characterization of an FSH releasing protein from porcine ovarian follicular fluid. Nature 1986;321:776-779. 73. Ling N, Ying S-Y, Ueno N, et al. Pituitary FSH is released by a heterodimer of the b-subunits from the two forms of inhibin. Nature 1986;321:779-782. 74. Nakamura T, Asashima M, Eto Y, et al. Isolation and characterization of native Activin B.JBiol Chem 1992;267:16385-16389. 75. Schwall R, Schmelzer CH, Matsuyama E, Mason AJ. Multiple actions of recombinant activin-A in vivo. Endocrinology 1989;125: 1420-1423. 76. Rivier C, Vale W. Effect of recombinant activin-A on gonadotropin secretion in the female rat. Endocrinology 1991;129:2463-2465. 77. Carroll RS, Kowash PM, Lofgren JA, Schwall RH, Chin WW. In vivo regulation of FSH synthesis by inhibin and activin. Endocrinology 1991 ;129:3299-3304. 78. Woodruff TK, Krummen LA, Lyon RJ, Stocks DL, Mather JR Recombinant human inhibin A and recombinant human activin A regulate pituitary and ovarian function in the adult female rat. Endocrinology 1993;132:2332-2341. 79. DePaolo LV, Bicsak TA, Erickson GF, Shimasaki S, Ling N. Follistatin and activin: a potential intrinsic regulatory system within diverse tissues. Proc Soc Exp Biol Med 1991;198:500-512. 80. Mathews LS. Activin receptors and cellular signaling by the receptor serine kinase family. Endocr Rev 1994;15:310-325. 81. Demura R, Suzuki T, Tajima S, et al. Human plasma free activin and inhibin levels during the menstrual cycle. J Clin Endocrinol Metab 1993; 76:1080-1082.
66 82. Meunier H, Cajander SB, Roberts VJ, et al. Rapid changes in the expression of inhibin or-, ~A-, and ~B-subunits in ovarian cell types during the rat estrous cycle. Mol Endocrino11988;2:1352-1363. 83. Meunier H, Roberts VJ, Sawchenko PE, et al. Periovulatory changes in the expression of inhibin or-, [3A-, and ~B-subunits in hormonally induced immature female rats. Mol Endocrino11989;3:2062-2069. 84. Miyanaga K, Erickson GF, DePaolo LV, Ling N, Shimasaki S. Differential control of activin, inhibin, and follistatin proteins in cultured rat granulosa cells. Biochem Biophys Res Commun 1993;194:253-258. 85. Feng ZM, Madigan MG, Chen CLC. Expression of type II activin receptor genes in the male and female reproductive tissues of the rat. Endocrinology 1993;132:2593-2600. 86. Nakamura M, Minegishi T, Hasegawa Y, et al. Effect of an activin A on follicle-stimulating hormone (FSH) receptor messenger ribonucleic acid levels and FSH receptor expressions in cultured rat granulosa cells. Endocrinology 1993;133:538-544. 87. Cameron VA, Nishimura E, Mathews LS, et al. Hybridization histochemical localization of activin receptor subtypes in rat brain, pituitary, ovary, and testis. Endocrinology 1994;134:799-808. 88. LaPolt PS, Soto D, Su JG, et al. Activin stimulation of inhibin secretion and messenger RNA levels in cultured granulosa cells. Mol Endocrino11989;3:1666-1673. 89. Xiao S, FindlayJK. Interaction between activin and follicle-stimulating hormone-suppressing protein and their mechanisms of action on cultured rat granulosa ceils. Mol Cell Endocrino11991;79:99-107. 90. Woodruff TK, Krummen L, Mceray G, Mather JP. In situ ligand binding of recombinant human [1251] activin-A and recombinant human [1251] inhibin-A to the adult rat ovary. Endocrinology 1993; 133:2998-3006.
ERICKSON AND CHANG 91. Xiao S, Robertson DM, Findlay JK. Effects of activin and folliclestimulating hormone (FSH)-suppressing protein/follistatin on FSH receptors and differentiation of cultured rat granulosa cells. Endocrinology 1992;131:1009-1016. 92. Presl J, Pospisil J, Figarov~i V, Krabec Z. Stage-dependent changes in binding of iodinated FSH during ovarian follicle maturation in rats. Endocrinol Exp 1974;8:291-298. 93. Zeleznik AJ, Schuler HM, Reichert LE. Gonadotropin-binding sites in the Rhesus monkey ovary: role of the vasculature in the selective distribution of human chorionic gonadotropin to the preovulatory follicle. Endocrinology 1981;109:356-362. 94. Nimrod A, Lamprecht SA. Hormone-induced desensitization of cultured rat granulosa ceils to FSH. Biochem Biophys Res Commun 1980; 92:905-911. 95. Nakamura K, Nakamura M, Igarash IS, et al. Effect of activin on luteinizing hormone-human chorionic gonadotropin receptor messenger ribonucleic acid in granulosa cells. Endocrinology 1994;134:2329-2335. 96. Mir6 F, Smyth CD, Hillier SG. Development-related effects of recombinant activin on steroid synthesis in rat granulosa cells. Endocrinology 1991;129:3388-3394. 97. Itoh M, Igarashi M, Yamada K, Hasegawa Y, et al. Activin A stimulates meiotic maturation of the rat oocyte in vitro. Biochem Biophys Res Commun 1990;166:1479-1484. 98. Doi M, Igarashi M, Hasegawa YU, et al. In vivo action of activin-A on pituitary-gonadal system. Endocrinology 1992;130:139-144.
~HAPTER
Endocrine Changes During the Perimenopause HENRY G. HELENA
BURGER PrinceHenry's Institute, Clayton, Victoria 3168, Australia.
J. TEEDE
Jean Hailes Research Group, Monash Institute for Health Services Research, Clayton, Victoria 3168, Australia.
The estrogen deficiency state after menopause was recognized clinically more than 100 years ago. Corresponding understanding of the stable endocrine physiology in postmenopausal women is largely complete with high gonadotropin, low sex steroid, and low inhibin levels. The dynamic hormonal fluctuations that control fertile cycles during the middle reproductive years also are well understood. However, for the years of transition from the fertile, ovulatory cycles of the middle reproductive years to the stable postmenopausal estrogen deficiency state, our understanding is still evolving. Until recently, gradually declining estrogen levels accompanied by rising gonadotropins were thought to characterize the period known as the perimenopause, but conventional thinking has been challenged as the endocrine physiology of the perimenopause received increasing attention. Wide variations in hormonal profiles exist between and within individuals, and declining levels of the inhibins appear to play a pivotal role in maintaining estrogen levels until just before menopause by permitting increased levels of gonadotropins. The perimenopause is the phase extending from the onset of symptoms of the ensuing menopause to 1 year after the final menstrual period (FMP), with a median age of onset of 45.5 to 47.5 years and an average duration of 5 years established in longitudinal studies (1,2). Perimenopausal women with a high incidence of clinical symptoms (1) seek T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
medical consultation more frequently than premenopausal or postmenopausal women. Dysfunctional uterine bleeding is most common during perimenopause, culminating in peak rates of hysterectomy. Changes in the skeletal and cardiovascular systems have been observed even during early perimenopause. With aging of the substantial generation of "baby boomers," increasing numbers of women are becoming perimenopausal. An accurate understanding of the endocrine changes occurring during this phase of the reproductive life cycle has significant therapeutic and diagnostic implications.
I. E N D O C R I N E DYNAMICS: T H E N O R M A L R E P R O D U C T I V E CYCLE The normal hormonal dynamics of the hypothalamicpituitary-ovarian axis control reproductive physiology during the middle reproductive years. An understanding of this control provides a background for subsequent observations throughout the perimenopause. The pituitary is regulated by pulsatile secretion of gonadotropin-releasing hormone (GnRH) from the hypothalamus. Luteinizing hormone (LH) and follicle-stimulating hormone (FSH), produced by the pituitary in response to GnRH, regulate ovarian function. These gonadotropins are
67
Copyright 9 2007 by Elsevier, Inc. All rights of reproduction in any form reserved.
BURGER A N D TEEDE
68 subject to predominantly negative feedback by the sex steroids estradiol and progesterone. With FSH, a dimeric glycoprotein, regulation is more complex, because it has a constitutive secretory component, as shown when isolated pituitary gonadotrophs are maintained in long-term cell culture in the absence of GnRH. They continue to produce significant amounts of FSH, but LH secretion rapidly drops to undetectable levels. LH is thus entirely GnRH dependent. FSH is subject to negative feedback control that is mediated by the inhibins and sex steroids. Ovarian follicular activity is reflected by production of sex steroids and peptide hormones (i.e., inhibin and activin). The sex steroids include estradiol produced by the follicle, progesterone produced by the corpus luteum after the maturation of the dominant ovarian follicle, and androgens, primarily testosterone and androstenedione, secreted by the theca interna and the ovarian stroma. Appreciation of the pivotal role played by the ovarian glycoproteins inhibin and activin is a relatively recent development. The function of the inhibins includes paracrine regulation of the gonads and pituitary and closed long-loop negative feedback on FSH at the level of the pituitary (3). Inhibin is a dimeric glycoprotein produced in the granulosa cells of the ovary (3). It has been documented to increase in puberty, fluctuate across the menstrual cycle, and become undetectable after menopause. Two major and distinct inhibin subtypes, inhibins A and B, are composed of a common subunit and one of two [3 subunits, [3a and [3b.The physiologic roles of the two inhibins are distinct by virtue of their 13 subunits. The two subtypes display functional, structural, and molecular differences (4). Most studies of the physiology of inhibin have employed a heterologous radioimmunoassay, the Monash assay, developed in Melbourne. It is nonselective, detecting inhibins A and B and inactive free ~ subunits (5). Subsequent work demonstrated that the Monash assay largely parallels the patterns seen with inhibin A. Specific, two-site assays have been developed for the measurement of inhibins A and B, and their physiology in the menstrual cycle has been documented (4,6) (Fig. 5.1). Only inhibin B is found in male plasma, but inhibin A and inhibin B occur in women. In women, inhibin B is produced mainly by the granulosa cells of the cohort of developing follicles, and inhibin A is a product of the dominant follicle and the corpus luteum, as is estradiol. Peripheral plasma levels of inhibin A increase progressively in the later part of the follicular phase, rising to a midcycle peak corresponding to the LH and FSH peak (see Fig. 5.1). They then fall, only to rise again to their peak levels in the luteal phase, parallel to the patterns of estradiol and progesterone. Inhibin B peaks in the early follicular phase, then declines before a midcycle peak and falls to low levels in the luteal phase (4,6). Data suggest that inhibin B may be less biologically active than inhibin A, although this area is still being researched (7).
A
200
80-
~, 60_i
<= .=_ .=
B
~"
~=o
40-
1004 m
20-
0
"0
5O
-1500
20
, 4i
e.g.. t-
9
_,,,,,
10 a.
0
0
C a~
g+ ~-. :i: ..a
is
,0
20
5
10 0
, -14
l t -7 0 7 Days relative to midcycle LH peak
: 0 14
FIGURE 5.1 Plasma concentrations ofinhibin A (A), inhibin (B), estradiol, and progesterone, and LH and FSH (C) during the menstrual cycle. Data displayed with respect to the day of the midcycle LH peak. (From ref. 4, with permission.)
At a functional level, the major role of inhibin is the negative feedback regulation of pituitary FSH secretion (3). FSH administration stimulates inhibin production (8), and inhibin levels are inversely correlated with FSH concentrations (9,10) (Fig. 5.2). lnhibin B is thought to be the predominant peptide regulatory feedback factor for FSH in the follicular phase, along with estradiol, and inhibin A may be more important in the luteal phase (4). Activins, FSHstimulating peptides discovered during the purification of inhibin, are formed from the dimerization of two inhibin 13 subunits. Activins primarily act in paracrine regulatory functions in the ovary and in the pituitary, where activin B is responsible for the autonomous component of FSH secretion (11).
II. E N D O C R I N E D Y N A M I C S : THE PERIMENOPAUSE Although significant progress has been made in recent years, the hormonal features of the menopausal transition are still being clarified. Current understanding encompasses
69
CHAPTER 5 Endocrine Changes During the Perimenopause 4.5 4.0 3.5
I
00 v
3.0
O _J
2.5
i m
2.0
r -1.5/_,m 1.5
.--
m
2.0
m
a
.......... m-o''" o
.....
a
~ .........
2.5
m I
l:::)m
m
....
~
. . . . . . .
0
1
m --a
~
i
a
a
3
am
3.0 3.5 4.0 Log (Inhibin B)
m
m
[] '"
m []
4.5
. . . . . .
,
5.0__
FIGURE 5.2
Relationshipsbetween follicular phase log (FSH) and log (inhibin B) in two age groupsof regularlycyclingwomen,20 to 39 years([21) and 40 to 50 yearsold (m). (From ref. 10, with permission.) the exponential decline of oocyte numbers as menopause approaches, with fluctuating ovarian hormone production and altered feedback regulation on the pituitary as a result of approaching ovarian failure. The role of declining inhibin levels is significant, with reduced negative feedback on the pituitary resulting in increased FSH production. Knowledge about the endocrine changes occurring during the perimenopause is based on evidence from several different study designs, and interpreting the results of these studies has also been difficult because of the variety of definitions used for this phase of reproductive life. Many studies have related observed changes to age rather than cycle pattern or symptoms. Several longitudinal studies encompassing serum sex steroid, gonadotropin, and inhibin profiles exist, although most data are cross-sectional. Traditional concepts about the endocrine changes characterizing the perimenopause included gradually declining estrogen levels and rising gonadotropins. However, increasing evidence suggested that estrogen and FSH levels rise during the perimenopause. Reyes studied ovulating women between 20 and 50 years old. FSH levels increased with age, but estradiol did not decline before menopause (12). Lee et al. had a similar finding, with a rising FSH concentration but no decline in levels of estradiol in a study of 94 regularly cycling women 24 to 50 years of age (13). FSH rose as a function of increasing age (especially in the early follicular phase and at midcycle), with a minimal rise in LH and no change in estradiol or progesterone levels. Rannevik et al. (14) studied 160 women from the Malm6 perimenopausal project over a 12-year period. Estradiol was measured for 7 years prior to the FMP and was stable until 6 months prior to the FMR
Only 1 to 6 months before the FMP did the estradiol (E2) level fall slightly, but it fell substantially by 1 to 6 months after FMR Longcope et al. (15) also showed that E2 levels were preserved until a few months before the onset of amenorrhea. Progesterone levels fell progressively from 4 years prior to the onset of amenorrhea. Fitzgerald observed that perimenopausal women had the most variability in ovarian steroid profiles but found that mean serum estrogens were no different than those of younger women (16). The maintenance of estrogen levels for as long as possible can reduce undesirable health outcomes. Some of the longitudinal data are based primarily on urinary steroid profiles. Several of these studies have documented elevated estrogen production in the setting of multiple developing follicles in perimenopausal women. In the 1970s, Brown analyzed urinary steroid profiles in 85 "climacteric" women and demonstrated that estrogen levels fluctuated, producing periods of hyperestrogenism (17). Unfortunately, little recognition was given to these findings because they were not published extensively and were thought to be incompatible with the conventional thinking of a slowly progressive decline in estrogen. Metcalf documented elevated FSH with high estrogen levels on 32 separate occasions in 14 perimenopausal women (18). Santoro studied daily urine profiles ofperimenopausal women between the ages of 47 and 50 years who showed greater estrone conjugate excretion than did women in their middle reproductive years (19). Variability was evidenced by episodes of marked hyperestrogenism, and elevated FSH was documented in many of the women studied. These findings are consistent with erratic follicular development and occasions of multiple follicles developing at any one time (20). Recent observations have documented the profiles of urinary FSH, LH, estrone glucuronide, and pregnanediol glucuronide in normal volunteers at various stages of the menopausal transition; previous findings have been confirmed. In particular, it has been shown that the cycle day of the first significant follicular phase rise of estrogen excretion (and hence estradiol secretion into the circulation) is correlated inversely with FSH levels in late-reproductive-age, regularly cycling women. As cycles become irregular and the ovaries more resistant to gonadotropin stimulation, the relationship changes. Higher FSH levels are then associated with a longer interval to the increase in estrogen excretion (21,22). These fluctuating hormone profiles can explain the variable symptoms, including those consistent with transient estrogen excess and the high incidence of dysfunctional uterine bleeding. An increase in the rate of follicle development with elevated FSH and estrogen levels has also been found in adolescents with anovulatory dysfunctional uterine bleeding (23). In one study (24), daily serum samples were obtained for one full cycle in cycling women early (19 to 34 years) and
BURGERAND TEEDE
70 later (42 to 49 years) in reproductive life. Insulinlike growth factor-1 (IGF-1) and growth hormone levels in these two groups were studied. The rationale was based on observations that estradiol influences both levels and that growth hormone and IGF-1 levels decrease with age and are low after menopause. However, IGF-1 levels were not lower in older-reproductive-age women in this study, perhaps consistent with maintained estrogen levels over these years, because it was again demonstrated that the older group of women had elevated FSH and elevated estradiol concentrations (Fig. 5.3). Blake et al. (24) stated, "It is important to highlight the estrogen increase because it may represent a harbinger of impending perimenopause, with therapeutic and diagnostic relevance." Given the accumulating evidence, it became increasingly obvious that the proposed declining
A 1
,
L
[~]Younger _J
300
Older**
~
p=o.tt3
j
E
iX -...- 200 0 "0 ,,,,,,..
r
p=O, 124
p=O.1t4
_
T
00 1 0 0 u.I
B
Early Late Follicular
Ovulatory
Mid Luteal
20 [ ~ Younger** 15
Older
_J v 32 co I.u
10
5
Early
Late
Follicular
Ovulatory
Mid
Luteal
FIGURE 5.3 Estradiol (A) and FSH (B) by age group and cycle stage in younger women (19 to 34 years) and older women (42 to 47 years). (From ref. 24, with permission.)
estrogen stimulus for rising gonadotropins appeared unsustainable. It was necessary to postulate that other factors were responsible for the observed rise in FSH. This dilemma was touched on in the 1970s, when the observations about estrogen and FSH were first made. Sherman and Korenman observed shorter cycle length in older cycling women because of a shortened follicular phase (25). They observed elevated FSH levels and proposed that reductions in an "inhibin-like substance," similar to that found in men, could be the primary stimulus for FSH elevation. Although longitudinal studies had not been completed, further evidence to support this theory of declining inhibin before menopause was sought. Most of these studies used the nonspecific Monash assay. In vitro evidence for a reduced inhibin reserve with age (26) was supported by studies of in vivo inhibin production in women undergoing in vitro fertilization. Hughes (27) demonstrated that inhibin responses to gonadotropin hyperstimulation were significantly lower in women older than 35 years compared with those of younger women, whereas the estradiol responses were similar in all groups. Cross-sectional studies demonstrated declining inhibin concentrations with increasing age and elevated FSH levels (9,10,28,29). The Melbourne Women's Midlife Health Project, for example, is based on a cross-sectional survey of a randomly selected, community-based sample of 2001 Melbourne women between the ages of 45 and 55 at the time of their initial interview (May 1991) (28). A longitudinal study of 438 of these women has examined many aspects of the menopausal transition. The first-year data have been analyzed crosssectionally in terms of menstrual cycle history and therefore menopausal status. Twenty-seven percent of subjects had reported no change in menstrual frequency or flow, and 23% reported a change in flow only. Nine percent reported a change in frequency without a change in flow, and 28% reported a change in frequency and flow. By the time the first-year blood sample was obtained, 13% reported at least 3 months of amenorrhea. Only those who had experienced 3 months of amenorrhea showed a statistically significant fall in estradiol, for which the geometric mean concentration was 42% of the group with no change in cycle. There was a broad spread of estradiol levels, with some greater than 1500 pmol/L, suggesting granulosa cell hyperstimulation by elevated FSH levels. Immunoreactive inhibin was significantly lower in those who had experienced a change in frequency and flow and lower still in those with 3 months of amenorrhea. Overall, the data suggested a fall in inhibin levels earlier than the fall in estradiol, consistent with the hypothesis that a declining inhibin concentration provides a mechanism that allows FSH to rise and thereby maintain intact early-follicular-phase estradiol concentrations. The situation has been further clarified by comparison of early-follicular-phase hormone levels in normally ovulating women between the ages of 20 and 45, selected to show an
CHAPTER5 Endocrine Changes During the Perimenopause increase in serum FSH, with those in women between the ages of 20 and 25 years (30). A fall in the level of inhibin B in the follicular phase of the cycle appeared to account for the age-associated rise in FSH, with estradiol levels slightly higher in the older women. Other studies have confirmed these findings (31,32). For inhibin decline to be a primary stimulus for FSH elevation and maintenance of estrogen levels, as we hypothesized, it is necessary to postulate that the secretion of estrogen and inhibin may reflect different aspects of granulosa cell function (33). Estradiol plays a central role in female reproductive function and has been described previously as the physiologic basis of the fertile period (34). Teleologically, it could be hypothesized that preservation of E2 secretion would be desirable for as long as possible in the human female. Consequences of the loss of E2 include the development of estrogen deficiency symptoms and undesirable health outcomes such as loss of bone and increased susceptibility to atherosclerosis and myocardial infarction. Given the confirmation that inhibin B production by the ovary declines significantly as a function of increasing age in regularly cycling women, while E2 production is preserved, it can be postulated that the decline in inhibin B levels leads to a rise in FSH and that it is the increased FSH drive that allows E2 secretion to be maintained as ovarian follicle numbers fall. There is a marked follicular-phase decrease in inhibin B levels in early perimenopause, defined on the basis of a reported change in menstrual cycle frequency (35), when FSH levels start to rise but inhibin A and estradiol concentrations remain unchanged (Fig. 5.4). Only in late perimenopause (i.e., after more than 3 months of amenorrhea) do estradiol and inhibin A levels also fall, with a marked rise in FSH. This pattern is supported by the studies demonstrating different estrogen and inhibin responses to ovarian hyperstimulation (27). The endocrine characteristics of the irregular cycles that characterize the menopausal transition have been described in a comprehensive longitudinal study of 12 Swedish women, in whom hormonal measurements were made in blood samples collected three times weekly for a month, annually until their final menstrual period (36,37). Two major types of cycles were identified following the onset of menstrual irregularitymcycles that had the endocrine features associated with normal ovulation, and cycles that were either anovulatory or showed markedly delayed ovulation, in which FSH levels were raised and estradiol and inhibin levels were low or normal. Normal ovulatory cycles occurred during cycle irregularity, suggesting that the women might still be fertile. There was no systematic change in hormone levels in such ovulatory cycles over the years preceding the FMP. The Melbourne Women's Midlife Health Project has also pubfished the endocrine data from its longitudinal study in which the progressive rise in FSH and the progressive fall in estradiol and the inhibins in the 1 to 2 years on each side of the
71
FICURE 5.4 Geometric mean levels (with lower 95% confidence intervals) of FSH, immunoreactive inhibin (IR-INH), inhibin A, inhibin B, and estradiol as a function of menopausal status in a group of 110 women from the Melbourne Women's Midlife Health Project. Stage 1, premenopausal: no change in cycle frequency; stage 2, early perimenopausal: change in cycle frequency; stage 3, late perimenopausal: 3 to 11 months of amenorrhea; stage 4, postmenopausal: more than 12 months of amenorrhea. Values with the same superscript (* or t) are not statistically different; for values with differing superscripts, p < 0.05. (From ref. 35, with permission.)
72
BURGER AND TEEDE
FMP were documented. FSH levels at the time of the FMP were approximately 50% of those ultimately achieved postmenopausally, while estradiol levels were approximately 50% of reproductive age follicular phase levels (38) (Fig. 5.5). The overall fall in inhibin B levels correlates with physiologic changes occurring in the ovary as the number of follicles, the source of inhibin B production, declines dramatically. Early autopsy studies (39) counted follicle numbers in ovaries of women 7 to 44 years old. Gougeon (40) obtained ovaries after oophorectomy combined with hysterectomy from women 19 to 52 years old. Subsequent studies completed by Richardson et al. on oophorectomy specimens estimated follicle numbers in women 44 to 55 years old (41). Combining these studies and applying mathematical analysis, Richardson demonstrated a steady decline in follicle numbers from the early years until 40 years of age. After 40, there was an exponential acceleration of follicle loss. Faddy and Gosden developed a mathematical model based on histologic analysis of ovaries of 52 cycling women (42). The model predicted a decline in the number of oocytes developing to a stage at which two layers of granulosa cells surround the oocyte, from five per day at 24 to 25 years to one per day at 49 to 50 years. They hypothesized that the accelerated rate of follicular depletion was attributable to an increase in the rate of atresia of primordial follicles. This idea was consistent with the observed increasing percentages of cycles that are anovulatory as women approach menopause (43,44). The primary event that stimulates accelerated follicular depletion is not understood and remains a topic of ongoing research. Fewer FSH receptors are found on ovarian follicles in the years preceding menopause, possibly suggesting that there is a disturbance of follicular maturation and function before exhaustion of follicles (45). Altered hypothalamic-pituitary sensitivity to hormonal factors may also contribute to perimenopausal endocrine physiology. Several observations support this theory. Estrogen suppression of the hypothalamic-pituitary axis is less complete in the perimenopausal woman, with hormone re"120 11250 3~176
-E
200-
o 150 t
.o
OOl
/\
A
V\
~ ~o~.=.=."
w p L = ,..w J ' - - r ' P "
\
/
= "~ kloo -80
r"
6o
~...~~__,
0 i .......................... i ............ ! ............!............ i........... ~ .......... i............ ! -3 -2 -1 0 1 2 3 -4 Years around Menopause
40
I
@
-~
-r
09 U
20
..........~.................:0 4 5
FIGURE 5.5 Geometric means of FSH and E2 in relation to the FMR The horizontal axis represents time (years) with respect to the FMP (0); negative (positive) numbers indicate time before (after the FMP). (From ref. 38, with permission.)
placement therapy only partially suppressing gonadotropin production. A longitudinal study compared perimenopausal women with dysfunctional uterine bleeding (DUB) to regularly menstruating control subjects (46). Basal and stimulated serial steroid and gonadotropin levels were analyzed. In contrast to controls, exogenous estradiol did not have a consistent suppressive effect on gonadotropins in women with DUB. This finding and the documentation of elevated estrogens in the presence of elevated gonadotropins, as well as the failure of elevated estrogens to induce LH surges, indicate that feedback may be altered (47). In light of the erratic and unreliable basal FSH levels during the perimenopause, other stimulatory procedures have been employed, primarily to predict fertility in women of older reproductive age. The clomiphene citrate challenge (CCC) was used to determine fertility potential in older reproductive women. In a study by Gindoffet al. (48), responses of FSH to CCC were assessed according to menstrual cycle pattern. Younger cycling women, older cycling women, irregularly cycling women older than 35, and menopausal women were compared. The perimenopausal and postmenopausal women showed baseline elevated FSH levels. Sustained FSH elevation after CCC was obvious in the perimenopausal women. Basal estrogen levels were higher, but stimulated estrogen responses were lower in perimenopausal women. The FSH responses to stimulation with GnRH did not differ in any of the groups of cycling women (49). With fluctuating but generally rising FSH, variable estradiol, and falling inhibin levels, what are the patterns of progesterone and androgens? Progesterone is produced primarily in the corpus luteum after ovulation. Ultimately the level of progesterone declines as menopause approaches. Anovulatory cycles occur with increasing frequency as menopause approaches, with 3% to 7% of cycles anovulatory in otherwise healthy women 26 to 40 years old, but 12% to 15% in those between the ages of 41 and 50 (43). Anovulation may occur because of increased rates of follicle atresia. Ultimately, the level of progesterone declines as menopause approaches (19). Urinary pregnanediol levels were found to decline in perimenopausal women in studies completed in the 1970s (17). Trdvoux et al. (50) also pointed out that the frequency of nondetectable levels of progesterone gradually increased as the FMP approached. These authors noted that in 11.5% of their cases, the endometrium was found to be secretory, despite a P level below 3 nmol/L. The discordance was seen most frequently in the period from 3 years to I year before the FMP. Rannevik et al. (51) reported that the frequency of cycles with P values indicative of ovulation (concentrations greater than 10 nmol/L) decreased from about 60% to less than 10% during the 6 years preceding the FMP. Ovulatory P levels were found in 62.2% of women 72 to 61 months premenopausal and 4.8% who were 6 to 0 months premenopausal, while all serum levels of P were less than 2 nmoVL postmenopausally.
CHAPTER 5 Endocrine Changes During the Perimenopause This reduction in ovulatory cycles observed in the setting of complex endocrine changes contributes to reduced fertility. It can also be noted that the endocrine changes observed in the late perimenopause and early postmenopause do not explain adequately the timing of the FMP, which appears to be an event determined by uterine factors (18,37). Androgens are primarily produced by ovarian theca and stroma cells and by the adrenals. The major circulating androgens and preandrogens in women are testosterone and dehydroepiandrosterone sulphate (DHEAS), respectively. Several studies have now clearly demonstrated that there is a fall (approximately 50%) in testosterone levels between the ages of 20 and 45 (52,53), whereas there is no significant fall associated with the menopausal transition itself (53,54) (Fig. 5.6). This observation is consistent with continued androgen secretion by the perimenopausal and postmenopausal ovary in the face of falling estradiol production (55). Free testosterone falls in parallel with total testosterone. Much of the confusion
1.61.41.2-J
0
1-
E v
0.8 -
O
0.6
o
73 regarding androgen levels in relation to menopause may be the result of studies that compared testosterone levels in older postmenopausal women with levels in women in their 20s. It appears that the major fall in testosterone levels occurs between 20 and 40. DHEAS falls progressively with increasing age (53), and the rate of fall is not affected significantly by the menopause. Because of the extensive conversion of DHEAS intracellularly to D H E A and testosterone, peripheral circulating levels provide only a very approximate index of tissue exposure to androgens in women. Based on the foregoing evidence, it is hypothesized that a fall in inhibin levels (perhaps mainly inhibin B) occurs with reproductive aging because of declining follicle numbers, allowing a rise in FSH, which leads to accelerated follicle development and increased estrogen secretion in perimenopausal women. Progesterone levels fall as more cycles become anovulatory, and androgens show little or no change. These endocrine changes appear to vary between and within individuals, and substantial changes have been observed from one cycle to the next. We previously concluded that for, estrogen and FSH profiles, the "most noteworthy characteristic is significant hormonal variability" (33). The only conclusion that can be made with confidence concerning hypothalamic-pituitary-ovarian function in individual perimenopausal women is that it is unsafe to generalize and not very useful to measure hormone levels at individual time points (67).
III. CLINICAL SEQUELAE OF E N D O C R I N E CHANGES IN T H E PERIMENOPAUSE
-
0.40.2 0
i
!
I
!
.............. i ................ ! ............... i ........................ I
25-
20_J o
E15-
I.~ 0 10I.L _
"
I
'
I
I
........ I
18-24 25-34 3 5 ~ 4 45-54 55-64 65-75 Age group (yrs) FIGURE 5.6 M e a n total (Testo) and free testosterone (Free 7') concentrations as a function of age in normal women. (Adapted from ref. 53, with permission from the author.)
The clinical features and management of the perimenopausal woman are covered elsewhere in this book, but it is appropriate to highlight the clinical observations with direct reference to the observed endocrine changes. Women in the perimenopausal years are more likely to seek medical consultation than their premenopausal or postmenopausal counterparts. Vasomotor symptoms, including hot flushes, night sweats, disturbances of sexuality, and psychologic symptoms, are markedly increased in the perimenopause (57,58). Vasomotor symptoms demonstrate the most instability, probably reflecting fluctuating hormone profiles. In one review (59), the increased frequency of migraines in the perimenopause was also attributed to fluctuating hormone profiles. Negative psychologic symptoms are most common and a sense of well-being is least prevalent during the perimenopausal period (60). Symptoms of estrogen excess and deficiency often occur in the same individual over time. Breast tenderness, menorrhagia, migraine, nausea, shorter cycle length, and a shorter follicular phase (17,19,46), all features of estrogen excess, have been documented in the perimenopause.
74 The severity of premenstrual dysphoria symptoms also increase in late reproductive age. Along with significant symptoms related to hormonal fluctuations, women also contend with irregular and heavyflow menstrual cycles. Fitzgerald et al. (16) described agerelated changes in the reproductive cycle. Initially, cycles are shorter, with a shorter follicular phase progressing to increased cycle length and then to irregularity. DUB, occurring with persistent elevation of unopposed estrogens (61), occurs most frequently in perimenopausal women, who have the greatest maximal thickness of endometrium (16) and have the highest incidence of hysterectomy. Menorrhagia occurs in 20% of perimenopausal women, compared with 9% of women in other phases of reproductive life, and differences in uterine vessel structure have been documented in perimenopausal women (62). Ironically, in 1974, on the basis of his studies of ovarian estrogen secretion in women with DUB, Fraser and Baird (61) suggested that hyperestrogenism resulted from multiple follicles developing because of abnormal gonadotropin release, which is consistent with today's thinking.
IV. H O R M O N E - I N D U C E D CHANGES IN THE PERIMENOPAUSE Estrogen deficiency induces a rapid phase of bone turnover in the early postmenopausal period that contributes to osteoporosis in later life. The changes in bone turnover in the perimenopause evoke controversy about whether significant alterations precede the decline in estrogen levels. In a prospective study of bone loss in perimenopausal women (part of the Melbourne Women's Midlife Health Project), Guthrie et al. (63) showed that maximal rates of bone loss occurred in women who became postmenopausal during the period of observation. The researchers found a statistically significant increased rate of loss in late but not early perimenopausal women. Estradiol levels are preserved in early but not late perimenopausal women. These authors have subsequently demonstrated that the fall in bone mineral density during the menopausal transition is correlated only with the fall in estradiol levels (64). Nilas and Christiansen (65) reported radial bone loss in women during perimenopause associated with increased FSH but not with reduced estrogen levels. In a study by Steinberg et al. (66), crosssectional changes in bone mass density (BMD) in the perimenopause correlated inversely with FSH levels, but estrogen and testosterone levels correlated directly with lower BMD. Slemenda et al. (67) undertook a longitudinal study over 3 years and demonstrated that women in the later phase of the menopausal transition with elevated FSH levels had reduced radial BMD but that those in the early transitional phase with irregular cycles and normal FSH levels did not have reduced BMD.
BURGER AND TEEDE
The opportunity to improve peak bone mass has passed in women reaching the menopausal transition, and further bone loss should be prevented to avoid later osteoporotic fractures. Given the emerging evidence that bone turnover increases and BMD falls preceding menopause, it may be timely to consider osteoporosis and skeletal health earlier than conventionally thought. Alterations have also been demonstrated in lipid profiles during the perimenopause, with progressive decreases in high-density lipoproteins and increases in total cholesterol preceding menopause by several years (68). Cardiovascular health across the menopausal transition has now been well evaluated in longitudinal cohort studies with interesting results (68,69). The changes in cardiovascular risk factors appear primarily to be related to age and body mass index (BMI), rather than associated with the hormonal changes over the menopausal transition (69,71). In the comprehensive Melbourne Women's Midlife Health Project cohort study, where 438 women were followed across the transition for what is now 12 years, the only metabolic parameter related to menopause per se was the high-density lipoprotein level (69). All other lipids, blood pressure, and BMI were related to age, not final menstrual period (68,71). Over the initial 6 years of the study, 16% of women developed impaired fasting glucose, an observation related to BMI rather than menopause (70). Serum insulin levels increased over these years but again were related to BMI (70). This data strongly supports the importance of lifestyle advice in this population, but primarily based on age rather than on menopausal status.
V. ENDOCRINE PROFILE ASSESSMENT O F T H E P ERIMEN O PAU SAL WOMAN In the setting of variable hormone profiles and symptoms, how should a physician approach the assessment of a perimenopausal woman? The studies of perimenopausal women have found the variations in FSH, inhibin, and estrogen levels to be transient and therefore unreliable in diagnosing approaching menopause or in predicting the stage of menopausal transition for any woman (56). Caution in interpreting hormone profile results is recommended, but if FSH assays are to be used, early follicular phase samples are the most reliable (56). A more rational alternative to hormone profile testing would be acquisition of individual longitudinal symptom data on women who present with perimenopausal complaints. The importance of daily menstrual pattern documentation was emphasized by Treloar (2). The Melbourne Women's Midlife Health Project has demonstrated that hormone profiles correlate well with symptoms and cycle features (72). Daily symptom diaries in our hands have proven more useful than isolated steroid profiles, and they can be helpful in self-education for women and provide an assessment tool for clinicians.
75
CHAPTER 5 Endocrine Changes During the Perimenopause
VI. C O N C L U S I O N A l t h o u g h the h o r m o n a l changes in perimenopausal w o m e n are still being clarified, existing data show an exponential decline in oocyte number, with consequent declining inhibin B levels, fluctuating and often elevated estradiol, and generally increasing F S H levels. This h o r m o n a l milieu may be accompanied by significant symptoms and menstrual disturbances, increased bone turnover, and lipid profile changes. Improved understanding of the endocrine changes characterizing the perimenopause can assist the clinician in accurate assessment and m a n a g e m e n t of patients. T h e ideal endocrine assessment is likely to be individual, longitudinal, and sympt o m based and is best achieved by the use of a perimenopause diary rather than isolated endocrine profiles.
References 1. McKinlay SM, Brambilla DJ, Posner JG. The normal menopause transition. Maturitas 1992;14:103-115. 2. Treloar AE. Menstrual cyclicity and the premenopause. Maturitas 1981;3:249-264. 3. Burger HG. Inhibin. Reprod Med Rev 1992;1:1-20. 4. Groome NP, Illingworth PJ, O'Brien M, et al. Measurement of dimeric inhibin B throughout the human menstrual cycle. J Clin Endocrinol Metab 1996;81:1401-1405. 5. Burger HG. Clinical utility ofinhibin measurements.J Clin Endocrinol Metab 1993;76:1391-1396. 6. Groome NP, Illingworth PJ, O'Brien M, et al. Detection of dimeric inhibin A throughout the human menstrual cycle by two-site enzyme immunoassay. Clin Endocrino11994;40:717- 723. 7. Robertson DM, Cahir N, Findlay JK, Burger HG, Groome N. The biological and immunological characterization of inhibin A and B forms in human follicular fluid and plasma. J Clin Endocrinol Metab 1997;82:889-896. 8. Hee J, MacNaughton J, Bangah M, et al. FSH induces dose-dependent stimulation of immunoreactive inhibin secretion during the follicular phase of the human menstrual cycle.J Clin Endocrinol Metab 1993;76: 1340-1343. 9. MacNaughton J, Bangah M, McCloud PM, et al. Age related changes in follicle stimulating hormone, luteinizing hormone, estradiol and immunoreactive inhibin in women of reproductive age. Clin Endocrinol 1992;36:339-345. 10. Burger HG, Dudley EC, Mamers P, et al. Early follicular phase serum FSH as a function of age: the roles ofinhibin B, inhibin A and estradiol. Climacteric 2000;3:17 - 24. 11. Corrigan AZ, Bilezikjian LM, Carroll RS, et al. Evidence for an autonomous role of activin B within rat anterior pituitary cultures. Endocrinology 1991;128:1682-1684. 12. Reyes FL, Winter JS, Faiman C. Pituitary ovarian relationships preceding the menopause. A cross-sectional study of serum follicle-stimulating hormone, luteinizing hormone, prolactin, estradiol and progesterone levels. Am J Obstet Gyneco11977;129:557-564. 13. Lee SJ, Lenton EA, Sexton L, Cooke ID. The effect of age on the cyclical patterns of plasma LH, FSH, estradiol and progesterone in women with regular menstrual cycles. Hum Reprod 1988;3:851-855. 14. Rannevik G, Carlstrom K, Jeppsson B, et al. A prospective long-term study in women from pre-menopause to post-menopause: changing profiles of gonadotrophins, oestrogens and androgens. Maturitas 1986;8: 297-307.
15. Longcope C, Franz C, Morello C, et al. Steroid and gonadotropin levels in women during the perimenopausal years. Maturitas 1986;8: 189-196. 16. Fitzgerald CT, Self MW, Killick SR, Bennett DA. Age related changes in the female reproductive cycle. BrJ Obstet Gynaeco11994;101:229-233. 17. BrownJB, Harrisson P, Smith MA, Burger HG. Correlations between the mucus symptoms and the hormonal markers offertility throughout reproductive life. Melbourne: Ovulation Method Research Centre of Australia,
Advocate Press, 1981. 18. Metcalf MG, Donald RA, Livesey JH. Pituitary-ovarian function in normal women during the menopausal transition. Clin Endocrinol 1981;14:245-255. 19. Santoro N, Brown JR, Adel T, Skurnick JH. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab 1996;81:1495-1501. 20. MetcalfMG, Donald RA. Fluctuating ovarian function in a perimenopausal woman. N Z MedJ 1979;89:45-47. 21. Miro F, Parker SW, Aspinall LJ, et al. Relationship between folliclestimulating hormone levels at the beginning of the human menstrual cycle, length of the follicular phase and excreted estrogens: the FREEDOM study.J Clin Endocrinol Metab 89:3270- 3275. 22. Miro F, Parker SW, Aspinall LJ, et al. FREEDOM study: origins and consequences of the elongation of the human menstrual cycle during the menopausal transition: the FREEDOM study. J Clin Endocrinol Metab 2004;89:4910-4915. 23. Baird DT. Anovulatory dysfunctional uterine bleeding in adolescence. In: Flamigni C, Venturoli S, Givens JR, eds. Adolescence in females. Chicago: Year Book, 1985;273-285. 24. Blake EJ, Adel T, Santoro N. Relationships between insulin-like growth hormone factor-1 and estradiol in reproductive aging. Fertil Steri11997;6 7:697- 701.
25. Sherman BM, Korenman SG. Hormonal characteristics of the human menstrual cycle throughout reproductive life. J Clin Invest 1975;55: 699- 706. 26. Pellicer A, Mari M, de los Santos MJ, et al. Effects of aging on the human ovary: the secretion of immunoreactive a-inhibin and progesterone. Fertil Steri11994;61:663-668. 27. Hughes EG, Robertson DM, Handelsman DJ, et al. Inhibin and estradiol responses to ovarian hyperstimulation: effects of age and predictive value for in vitro fertilization outcome.J Clin Endocrinol Metab 1990;70: 358-364. 28. Burger HG, Dudley EC, Hopper JL, et al. The endocrinology of the menopausal transition: a cross-sectional study of a population-based sample. J Clin Endocrinol Metab 1995;80:3537- 3545. 29. Lenton EA, Kretser DM, Woodward AJ, Robertson DM. Inhibin concentrations throughout the menstrual cycles of normal, infertile and older women compared with those during spontaneous conception cycles. J Clin Endocrinol Metab 1991;73:1180-1190. 30. Klein NA, Illingworth PJ, Groome NP, et al. Decreased inhibin B secretion is associated with the monotropic FSH rise in older, ovulatory women: a study of serum and follicular fluid levels of dimeric inhibin A and B in spontaneous menstrual cycles. J Clin Endocrinol Metab 1996;81:2742-2745. 31. Welt CK, McNicholl DJ, Taylor AE, HallJE. Female reproductive aging is marked by decreased secretion of dimeric inhibin. J Clin Endocrinol Metab 1999;84:105-111. 32. Danforth DR, Arbogast LK, Mroueh, J, et al. Dimeric inhibin: a direct marker of ovarian aging. Fertil Steri11998;70:119-123. 33. Burger HG. Inhibin and steroid changes in the perimenopause. In: Lobo R, ed. The perimenopause. New York: Springer-Verlag, 1998. 34. Burger HG. The physiological basis of the fertile period. In: Harrison RF, Thompson BW, eds, Fertility and sterility. Lancaster, MTP Press Limited, 1984;51-58.
76 35. Burger HG, Cahir N, Robertson DM, et al. Serum inhibins A and B fall differentially as FSH rises in perimenopausal women. Clin Endocrinol 1998;48:809-813. 36. Landgren BM, Collins A, Csemiczky G, et al. Menopause transition: annual changes in serum hormonal patterns over the menstrual cycle in women during a nine-year period prior to menopause.J Clin Endocrinol Metab 2004;89:2763-2769. 37. Burger HG, Robertson D, Baksheev L, et al. The relationship between the endocrine characteristics and the regularity of menstrual cycles in the approach to menopause. Menopause 2005;12:267-274. 38. Burger HG, Dudley EC, Hopper JL, et al. Prospectively measured levels of serum follicle stimulating hormone, estradiol, and the dimeric inhibins during the menopausal transition in a population-based cohort of women. J Clin EndocrinolMetab 1999;84:4025-4030. 39. Block E. Q.uantitative morphological investigations of the follicular system in women: variations in different ages.dcta/Inat 1952;14:108-123. 40. Gougeon A. Caractere qualitative and quantatif de la population folliculaire dans liavaire humain adulte. Contracept Fertil Sex 1984;12: 527-535. 41. Richardson SJ, Senikas Y, Nelson JE. Follicular depletion during the menopausal transition: evidence for accelerated loss and ultimate exhaustion. J Clin EndocrinolMetab 1987;65:1231. 42. Faddy MJ, Gosden RG, Gougeon A, et al. Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause. Hum Reprod 1992;7:1342-1346. 43. Doring GK. The incidence of anovulatory cycles in women. J Reprod Ferti11969;6:77- 81. 44. Metcalf MG. Incidence of ovulatory cycles in women approaching the menopause. J Biosoc Sci 1979;11:39-48. 45. Vihko KK. Gonadotropins and ovarian gonadotropin receptors during the perimenopausal transition period. Maturitas 1996;2:$19- $22. 46. Van Look PF, Lothian H, Hunter WM, et al. Hypothalamic-pituitaryovarian function in perimenopausal women. Clin Endocrinol 1977;7: 13-31. 47. Weiss G, Skurnik JH, Godsmith LT, Santoro NF, Park SJ. Menopause and hypothalamic-pituitary sensitMty to estrogen. J/IMA 2004;292: 2991-2996. 48. Gindoff PR, Schmidt PJ, Rubinow DR. Responses to clomiphene citrate challenge test in normal women through perimenopause. Gynecol Obstet Invest 1997;43:186-191. 49. Schmidt PJ, Gindoff PR, Baron DA, Rubinow DR. Basal and stimulated gonadotropin levels in the perimenopause. Am J Obstet Gynecol 1996;175:643-650. 50. Trtvoux R, De Brux J, Castainier M, et al. Endometrium and plasma hormone profile in the peri-menopause and post-menopause. Maturitas 1986;8:309-326. 51. Rannevik G, Jeppsson S, Johnell O, et al. A longitudinal study of the perimenopausal transition: altered profiles of steroid and pituitary hormones, SHBG and bone mineral density. Maturitas 1995;21: 103-113. 52. ZumoffB, Strain GW, Miller LK, Rosner W. Twenty-four hour mean plasma testosterone concentration declines with age in normal premenopausal women. J Clin Endocrinol Metab 1995;80:1429-1430. 53. Davison SL, Bell R, Donath S, et al. Androgen levels in adult females: changes with age, menopause, and oophorectomy. J Clin Endocrinol Metab 2005;90:3847- 3853.
BURGER AND TEEDE
54. Burger HG, Dudley EC, Cui J, et al. A prospective longitudinal study of serum testosterone dehydroepiandrosterone sulfate and sex hormone binding globulin levels through the menopause transition. J Clin Endocrinol Metab 2000;85:2832-2938. 55. Judd HL. Hormonal dynamics associated with the menopause. Clin Obstet Gyneco11976;19:775-788. 56. Burger HG. Diagnostic role of follicle-stimulating hormone (FSH) measurements during the menopausal transition--analysis of FSH, oestradiol and inhibin. EurJ Endocrino11994;130:38-42. 57. Dennerstein L, Smith AM, Morse C, et al. Menopausal symptomatology: the experience of Australian women. Med J Aust 1993;159: 232-236. 58. Mitchell ES, Woods NE Symptom experiences of mid-life women: observations from the Seattle Midlife Women's Health Study. Maturitas 1996;25:1-10. 59. MacGregor EA. Menstruation, sex hormones and migraine. Neurol Clin 1997;15:125-141. 60. Hunter M, Battersby R, Whitehead M. Relationships between psychological symptoms, somatic complaints and menopausal status. Maturitas 1986;8:217-228. 61. Fraser IS, Baird DT. Blood production and ovarian secretion rates of estradiol 1713 and estrone in women with dysfunctional uterine bleeding. J Clin EndocrinolMetab 1974;39:564- 569. 62. Abberton KM, Taylor NH, Healy DL, Rogers PA. Vascular smooth muscle alpha actin distribution around endometrial arterioles during the menstrual cycle: increased expression during the perimenopause and lack of correlation with menorrhagia. Hum Reprod 1996;11:204-211. 63. Guthrie JR, Ebeling PR, Hopper LJ, et al. A prospective study of bone loss in menopausal Australian-born women. OsteoporosInt 1998;19: 165-173. 64. Guthrie JR, Lehert P, Dennerstein L, et al. The relative effect of endogenous estradiol and androgens on menopausal bone loss: a longitudinal study. Osteoporosint 2004;15:881- 886. 65. Nilas L, Christiansen C. Bone mass and its relationship to age and the menopause, y Clin EndocrinolMetab 1987;65:697- 702. 66. Steinberg RK, Freni-Titulaer W, Depuey EG, Miller DT, et al. Sex steroids and bone density in premenopausal and perimenopausal women. J Clin EndocrinolMetab 1989;69:533 - 539. 67. Slemenda C, Hui SL, Longcope C, Johnston CC. Sex steroids and bone mass: a study of changes about the time of menopause. J Clin Invest 1987;80:1261-1269. 68. Do K-A, Green A, Guthrie, J, et al. A longitudinal study of risk factors for coronary heart disease across the menopausal transition, diner J Epidemio12000;151:5 84- 593. 69. Guthrie JR, Dennerstein L, Taffe JR, et al. The menopausal transition: a 9-year prospective population-based study. The Melbourne Women's Midlife Health Project. Climacteric2004;7:375-389. 70. Guthrie JR, Ball M, Dudley EC, et al. Impaired fasting glycaemia in middle-aged women: a prospective study. International Journal of Obesity & Related Metabolic Disord 2001;25:646-651. 71. Guthrie JR, Taffe JR, Lehert P, et al. Association between hormonal changes at menopause and the risk of a coronary event: a longitudinal study. Menopause 2004;11:315-322. 72. Guthrie JR, Dennerstein L, Taffe J, et al. Hot flushes during the menopause transition: a longitudinal study in Australian-born women. Menopause 2005;12:460-467.
~HAPTER (
Epidemiology of Menopause: Demographics, Environmental Influences, and Ethnic and International Differences in the Menopausal Experience ELLEN
GAIL A.
B. GOLD
GREENDALE
Division of Epidemiology, Department of Public Health Sciences, University of California, Davis, CA 95616
Divisionsof Geriatrics/General Internal Medicine, Department of Internal Medicine, David Geffen School of Medicine, Universityof California at Los Angeles,Torrance, CA 90095
I. I N T R O D U C T I O N
menopause experience may have been affected by the nature of the samples of women studied. In addition, a number of methodologic issues arise, which must be considered in conducting and comparing studies of the menopausal transition. Therefore, this chapter will begin with a discussion of the methodologic issues, to be followed by a review of factors studied to date that have been suspected or shown to affect the nature of the transition.
Although menopause is a universal phenomenon among women, the timing of the onset and the signs and symptoms of the perimenopause, menopausal transition, and final menstrual period are not (1). Most of our knowledge and perceptions of the experience of menopause have derived from studies largely of white women, and many have been studies of clinic-based, rather than population-based, samples of women. Thus, until recently, much of the picture of the
A. Methodologic Concerns The authors are indebted to the following collaborators in the study of the natural history of the menopause for their contributions to this work: Drs. Barbara Abrams, Shelley Adler, Gladys Block, Maradee Davis, Bruce Ettinger, Bill Lasley, Marion Lee, Marianne O'Neill Rasor, Steven Samuels, Helen Schauffler, Barbara Sommer, and Barbara Sternfeld. TREATMENT OF THE POSTMENOPAUSAL W O M A N
Most studies of the menopausal transition have been cross-sectional, rather than longitudinal, in design, providing an opportunity for distortion of the true picture of the menopausal experience, particularly for understanding
77
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
78
GOLD AND GREENDALE
risk factors that precede, rather than accompany or follow, the menopause transition. Further, definitions of menopause have varied depending on number of months of amenorrhea, and scales and time frames for assessing symptoms have varied from study to study. Studies have also varied in whether and which factors have been included in controlling simultaneously for the effects of multiple confounding variables. 1. AGE AT MENOPAUSE The analysis of age at natural menopause in a number of studies has been calculated only as a simple mean, rather than using the less-biased survival or multivariable regression analysis approaches, which include more information and observations, because women are included but withdrawn or censored when they experience surgical menopause or are still premenopausal (2). Also, accuracy of reporting of age at menopause can vary by whether or not menopause was natural and by duration from time of menopause to time of the interview about menopause (3). Further, in some studies reporting age at menopause, it is unclear if the age at the final menstrual period is being reported, which appears to be more frequent, or if the age at cessation of menses plus 1 year of amenorrhea, the World Health Organization's definition of menopause (4), is what is reported, apparently a more rare occurrence (5). Finally, in future studies, an accurate picture of the true age at menopause may become even more difficult to discern as more women are prescribed and take oral contraceptives or hormone replacement medication prior to the final menstrual period. 2. PRESENTATION AND SYMPTOMS OF MENOPAUSE A number of methodologic issues may influence reporting and thus the prevalence of menopausal symptoms observed in different studies. First, a lack of consistency in symptoms included, in scales used to assess them, or in the time frame for assessment may lead to differences in ascertainment. Second, a failure to recognize colloquial or culturally specific expressions for certain symptoms, even for hot flashes (which itself can be considered a Western colloquialism), may account for some of the differences in prevalence among different populations. An extension of this problem is failure to recognize and address cultural sensitivities in reporting symptoms, which may also be affected by who asks the questions and may result in underreporting of specific or even most symptoms. Third, most studies do not incorporate hormonal measures and may use inadequate questions about menstrual bleeding so that perimenopausal status cannot be well established. Further, a lack of comparability among studies with regard to the age group of women studied may also be reflected in the resulting differ-
ences in prevalence rates of symptoms. Finally, in some studies, age is used as a surrogate for menopausal status, so that menopausal status is presumed based on age, rather than on menstrual function, thus introducing misclassification of menopause-related symptoms and possible lack of comparability with studies that have asked more completely about menstrual function. In recent years, more information has begun to appear regarding differences in the timing and presentation of menopause experienced by samples of women of different socioeconomic, ethnic, and lifestyle backgrounds, resulting in a fuller and more varied picture and also greater insights (and questions) regarding the physiology underlying the menopausal experience. This field of investigation will benefit and greater understanding will result if this trend continues, along with increased standardization of instruments and methods so that a clearer picture of the menopausal experience will emerge.
B. S u m m a r y o f U n d e r l y i n g P h y s i o l o g y Although the physiologic changes that accompany the menopause are described in detail elsewhere in this volume (Chapter 5), a brief summary is mentioned here to provide relevant context to the signs and symptoms of the menopause experience and the factors that affect them that are discussed in this chapter. The cessation of menstruation that defines menopause reflects cessation of ovulation due to a loss of ovarian follicles. This loss results in a number of endocrine changes, particularly the decline in ovarian production of estradiol, the most biologically active form of estrogen (6,7), as well as increased circulating concentrations of follicle-stimulating hormone (FSH) and decreased concentrations of inhibin, which normally inhibits the release of FSH (6). Age at menopause may be more sensitive to varying rates of atresia of ovarian follicles (8) than to the absolute number of oocytes depleted (9). As circulating estrogen concentrations decline in the perimenopause, variations in the timing of menstrual bleeding and in the nature of bleeding may occur. Menstruation may occur at irregular intervals due to irregular maturation of residual follicles, with diminished responsiveness to gonadotropin stimulation, or to anovulatory uterine bleeding after estrogen withdrawal without evidence of corpus luteum function (10). As menstrual cycles become increasingly irregular, uterine bleeding may occur after an inadequate luteal phase or without ovulation or evidence of corpus luteum function (10), usually indicated by a short luteal phase, characteristic of women over the age of 40 years (11,12). Such cycles may be associated with insufficient FSH in the follicular phase, in turn resulting in lower luteal phase estrogen and progesterone secretion. The absence of the corpus luteum, resulting in estrogen
79
CHAPTER 6 Epidemiology of Menopause secretion (even hyperestrogenicity [11,13]) unopposed by progesterone, may lead to profuse blood loss. Conversely, relatively low concentrations of estrogen that may accompany the menopause transition also may lead to intermittent spotting. Thus, the nature and timing of bleeding may vary both within and between women, but relatively little is known about host, environmental, or lifestyle factors that may affect such variation. Extraovarian sources, such as adipose tissue, convert androstenedione to testosterone and estrone postmenopausally (14-17), although postmenopausal estrone production is only about one-third that of premenopausal production (8,18) but may be elevated in perimenopausal women in both the follicular and luteal phases (11). This peripheral production of estrone has been related to the amount of body fat (15,19) but has not been consistently related to age, height, or years since menopause. Also, in the perimenopause, FSH concentrations increase (7) without a concomitant increase in luteinizing hormone (LH) (20). Reduced concentrations of estrogen and progesterone and increases in FSH (20) affect the central nervous system (21-24), result in vasomotor instability, and may lead to the characteristic hot flashes or flushes in many perimenopausal and postmenopausal women. Two longitudinal studies have shown that vasomotor symptoms in the perimenopause and postmenopause are related to serum estrogen levels (21,25), but one cross-sectional study did not observe a significant relationship (26). Hot flashes may also be more prevalent in women who experience irregular menses prior to menopause than in women who experience an abrupt cessation of menses (27). Other cross-sectional and longitudinal studies have reported that the prevalence of symptoms appears to peak during the perimenopause transition and may decrease somewhat after menopause (28-30). In the international literature, some have reported that about 40% to 70% of perimenopausal and 60% to 80% of menopausal white women experience hot flashes (30,31), and a substantial majority of these report that they are moderate or severe (32,33). Discrepancies in the prevalence of hot flashes at least partially reflect inconsistencies in research methods and study populations (34). Other symptoms in hormone-receptive tissues may also occur in the perimenopause and postmenopause (23). Changes in the vagina and vulva may result in atrophy, pruritus, dryness, bleeding, and dyspareunia. Estimates of the prevalence of vaginal dryness range from 12% to 34%, depending on the age group ofwomen studied (21,32,35 - 38). The embryology of the urinary and genital systems is shared, and the urethral epithelium and submucosa are affected by estrogen (39,40). Although about a quarter of midlife women report some form of incontinence, its frequency does not appear to be related to menopausal status as determined from menstrual changes (41) or serum FSH
or estradiol concentrations, even though rates of incontinence increase during the time of menopause and then decline thereafter (42). Although mood changes and sleep disturbances may also occur at this time, the causal time sequence of vasomotor symptoms, mood changes, and sleep disturbances and the factors that influence their occurrence or perception have not been clarified. Thus, for example, it is not known if hormonal changes affect mood and sleep independent of their effect on vasomotor instability. A variety of physical symptoms, such as headaches, joint pain, aches in the back of the neck and shoulders (29), constipation, and dizzy spells may increase during the perimenopausal and postmenopausal years (43). However, the empirical data are inconclusive in identifying which, if any, are more prevalent during different stages of the menopause transition. For example, some investigators have found that women report significantly more frequent joint pain and dizzy spells when perimenopausal than they did when premenopausal (25,29), and others found a greater proportion of perimenopausal and postmenopausal women reporting aches and pains, but not dizzy spells, compared with premenopausal women (44). Still others have found no association between any somatic symptoms and menopausal status (45). Although some factors have been identified that are associated with early age at menopause and risk of experiencing symptoms, the relation of many has not been examined, most have not been examined in relation to duration of the perimenopause, and the endocrinologic effects of known risk factors in perimenopausal women still remain to be explored adequately.
II. D E M O G R A P H I C CHARACTERISTICS
A. Age at Natural Menopause and Onset of the Perimenopause Age at natural menopause has traditionally been defined as the age at the final menstrual bleeding, which is followed by at least 12 months of amenorrhea (4). The intrinsic public health interest in age at menopause results from the suggestions of some researchers that the age at which natural menopause occurs may be a marker of aging and health (46-48). Later age at natural menopause has been associated with reduced risk of cardiovascular disease (49-53), atherosclerosis (54), osteoporosis (55), and fracture (56) but an increased risk of breast (57,58), endometrial, and ovarian (49,59,60) cancer. Recent research results indicate that later menopause is associated with longer overall survival and greater life expectancy (49).
GOLD AND GREENDALE
80
Cross-sectional data indicate that endocrine changes characteristic of the onset of the perimenopause begin at around 45 years (61). Whereas the median age at menopause in white women from industrialized countries is between 50 and 52 years and of the perimenopause is 47.5 years (28,62-65), with slight evidence of increasing age at menopause over time (29,65-68), these onsets may vary by race and ethnicity (see Ethnic and International Differences later in this chapter) and may be affected by lifestyle factors (discussed in Environmental Influences).
(64,69,70,81,82), an observation that is also consistent with the theory that OCs delay depletion of oocytes. However, the finding is not wholly consistent across studies, because one study reported that this delay became nonsignificant after a time-dependent adjustment for when OCs were used (64), and another study reported that OC users had a significantly earlier natural menopause than nonusers, although this association was not consistent across 5-year age groups (62).
B. Presentation and Symptoms of Menopause 1. SOCIOECONOMIC STATUS
Lower social class, as measured by the woman's level of education completed or by her own or her husband's occupation, has been associated in more than one study with an earlier age at menopause (63-66,68-71). One study found that education was more strongly associated than occupation (64). Most studies that have examined the relation of marital status have found that single women have menopause at an earlier age and that this association cannot be explained by nulliparity (64,72,73).
2. MENSTRUAL AND REPRODUCTIVE HISTORY
Age at menopause may be a marker for hormonal status or changes earlier in life (74). In a landmark longitudinal study of largely white, well-educated women, those whose median menstrual cycle length between the ages of 20 and 35 years was less than 26 days were reported to have menopause 1.4 years earlier than women with cycle lengths between 26 and 32 days, whereas a later natural menopause (mean = 0.8 year later) was observed in women with cycle lengths of 33 days or longer (75). In addition, variability in cycle length of 9 or more days was also associated with a later age at menopause in this and other studies (64,76), although one early study reported an earlier menopause in women with irregular menses (65). Increasing parity, particularly in women of higher socioeconomic status (SES), has also been associated with later age at menopause (62-64,66,68-70,73,74,77), consistent with the theory that menopause occurs after sufficient depletion of oocytes (77). Although some studies report no familial relationship, one study has reported that age at menopause is positively associated with maternal age at menopause (69), and one recent study has shown genetic control of age at menopause in a study of twins (78). Age at menarche has been fairly consistently observed n o t to be associated with age at menopause, after adjusting for parity and cycle length (64-66,72,79-81), as have prior spontaneous abortion, age at first birth, and history of breastfeeding (64,80,81). Women who have used oral contraceptives (OCs) have also been reported to have a later age at menopause
1. SOCIOECONOMIC STATUS
Although the majority of white women experience vasomotor symptoms (hot flushes or flashes or night sweats), the prevalence of symptom reporting varies greatly by socioeconomic status. Estimates of the incidence of vasomotor symptoms in menopausal white women range in population studies in the United States and worldwide from 24% to 93% (30,31). Less educated women report more hot flashes and irritability than more educated women (29,30,44,83-87). One large cross-sectional study reported increased prevalence of all symptoms associated with difficulty paying for basics (29). One relatively small cross-sectional study found that women who reported mood changes or irritability that they believed were related to the menopausal transition were significantly more likely to report more everyday complaints (88). This was in contrast to women who reported hot flashes, sweating, or headaches associated with the menopausal transition who did not report more everyday complaints. Homemakers have been shown to report hot flashes for a longer period than employed women (89), although working women of lower SES report more stress and tension during menopause (89,90), and worsening work stress has been associated with increased reporting of vasomotor symptoms, general health symptoms, and sexual difficulties (44). One cross-sectional study has reported an association of occupational solvent exposure and forgetfulness during midlife (91). 2. MENSTRUAL AND REPRODUCTIVE HISTORY The relationship of menstrual characteristics to the probability of experiencing menopausal symptoms largely remains unexplored, and reproductive history has been examined in a few studies but with somewhat inconsistent results. One cross-sectional study reported that women having natural menopause before age 52 years had a significantly greater rate of reporting hot flashes (86). In another study, multiparous women had a lower prevalence of hot flashes than nulliparous women or women with abrupt cessation of menses (27). However, yet another study showed menopausal symptoms to be associated positively with increasing parity (92), and some studies report no differences in symptom reporting
81
CHAPTER 6 Epidemiology of Menopause frequency by parity (86,93,94). Three longitudinal studies and one retrospective study have shown that reporting of menopausal vasomotor symptoms was significantly more frequent among women who reported experiencing premenstrual tension or symptoms before menopause (30,83,95,96), a finding that may be related to higher FSH levels in menstruating women with hot flashes (97). One of these studies also reported that women with vasomotor symptoms were significantly more likely to report that their mothers also had vasomotor symptoms than women without symptoms (95). Most studies show symptoms to be more prevalent in hysterectomized women (29,44) and among women who experience an early menopause. In summary, later age at menopause may be a marker of health and longevity (46-48). Studies have fairly consistently shown that lower socioeconomic status (63-66,69,70), single marital status (64,72,73), regular menstrual cycles (64,76), nonuse of oral contraceptives (64,69,70,81,82), and lower parity (62-64,66,67,70,73,74,77) are associated with earlier menopause. Age at menarche (64-66,72,79-81), prior spontaneous abortion, age at first birth, and prior breastfeeding (64,80,81) are not associated with age at menopause. Lower socioeconomic status (30,44,83-87) and history of premenstrual tension (30,83,95,96) are associated with greater menopausal symptom reporting. However, the relation of parity to prevalence of symptoms has been inconsistent across studies.
III. ETHNIC AND INTERNATIONAL DIFFERENCES A. Age at Natural Menopause and at Onset of Perimenopause 1. ETHNIC DIFFERENCES
African-American (76) and Latina (70,98) women have been observed to have natural menopause about 2 years earlier than white women, despite their increased average body mass relative to white women (see Body Mass and Composition later). However, one small study in Nigeria reported the average age at menopause to be 52.8 years (99), nearly 2 years later than that generally reported for white women in industrialized nations. Mayan women, despite their high parity, have been reported to experience menopause at about age 45 years (100). Further, Mexican-American women may have shorter bleeding periods and follicular phase lengths (101). In contrast, Asian women tend to have similar age at menopause to Caucasian women (70,102), although Thai women have been reported to have a lower median age at menopause, at age 49.5 years, despite their high parity (79), and Filipino Malay women have been reported to have an average age at menopause of 47 to 48 years (103).
2. INTERNATIONALDIFFERENCES A number of reports tend to indicate that women living in developing countries (including Indonesia, Singapore, Pakistan, Chile, and Peru) experience menopause several years earlier than those in developed countries (82,104-107). Some work has also indicated that women living in urban areas have a later menopause than women in rural areas (108). Women living at high altitude in the Himalayas or in the Andes of Peru have been shown to undergo menopause 1 to 1.5 years earlier than those living at lower altitudes or in less rural areas (82,109-111). It is unclear if these geographic differences in the age at natural menopause reflect socioeconomic, environmental, racial/ethnic, or lifestyle differences and whether and how these affect physiology.
B. Presentation and Symptoms of Menopause 1. ETHNIC DIFFERENCES
Although the majority of Caucasian women experience menopausal symptoms (30,31,33), the reported frequency is much lower in most Asian women who have been studied (30,33,112-117); however, one retrospective study reported no difference in symptom prevalence between Japanese and Caucasian menopausal women in Hawaii (118). Further, some estimates of the prevalence of hot flashes have varied in similar Asian populations, such as from 23% (79) to 60% in Japanese and Chinese women (30) to 69% (119) in Thai women. Among Filipino Malay women age 40 to 55 years, reporting of vasomotor or circulatory symptoms occurred in 63%, and nervous or psychologic symptoms (particularly headache or irritability) were reported by 79%, although only 31% consulted a physician, a rate that was higher among women with a vocational or college education (103). It is unclear whether these ethnic differences in symptom frequencies are due to differences in cultural perceptions of menopause and reporting symptoms (120); diet (121); physical activity or body mass (122); differences in use of herbs or plant-estrogen-containing products (123) or in the use of acupuncture (which lowers excretion of the vasodilating neuropeptide calcitonin gene-related peptidelike immunoreactivity) (124) between Asian and Caucasian women; or differences in serum estradiol levels. Serum estradiol levels have been observed to be significantly lower in Asian women in relatively nonsystematic studies that did not indicate adequate control of the day of the menstrual cycle on which blood was drawn for estrogen assays (122,125). This finding has also been demonstrated in longitudinal studies that have not only controlled for menstrual cycle day of the blood draw but also for other factors, such as smoking and body mass index, that affect circulating estradiol levels (126). However, recent work (26,30) suggests that
82
GOLD AND GREENDALE
racial/ethnic differences in vasomotor-symptom reporting in midlife women persist even after adjustment for the significant effects of differences in hormone levels, educational attainment, body mass index, smoking, dietary factors, and psychosocial characteristics. Mayan women report no hot flashes (100), despite hormone profiles similar to Western women (127). On the other hand, African-American and Hispanic women have been reported to have a higher prevalence of vaginal dryness than Caucasian women (21,29,86). Some researchers believe that differences in the prevalence of symptom reporting reflect negative cultural stereotypes of aging and of the menopause experience (128,129) and are related to mental health (36). However, others believe that because some studies report a frequency of hot flashes, night sweats, and vaginal dryness in such places as Indonesia and Southeast Asia, for example, that is similar to that seen in Western countries, the latter view may be too simplistic (130). Rather, cultural values of menopause, as well as climate, dietary habits, and lifestyle, may also be related. In summary, ethnicity appears to be related to both age at menopause and symptom reporting. African-American (76) and Latina (70,98) women have an earlier menopause than Caucasian or some Asian (103) women, although not all Asian women (79,103). Women in less developed countries also experience menopause earlier (82,104-107). Additionally, Mayan (100) and Asian (30,79) women report fewer hot flashes, whereas African-American and Hispanic women have a higher prevalence of vaginal dryness than Caucasian women (21,29,86).
IV. ENVIRONMENTAL INFLUENCES A. Age at Natural Menopause and Onset of the Perimenopause 1. SMOKING
Perhaps the single most consistently shown (micro-) environmental effect on age at menopause is that women who smoke stop menstruating 1 to 2 years earlier than comparable nonsmokers (62,63,66,68-70,76,131-136) and have a shorter perimenopause (28). In some studies, heavy smokers have been observed to have an earlier menopause than light smokers, suggesting a dose response effect of smoking on atrophy of ovarian follicles (69,135-139). Former smokers have only a slightly earlier age at menopause than those who never smoked, and increased time since quitting diminishes the difference (137,140). The polycyclic aromatic hydrocarbons in cigarette smoke are known to be toxic to ovarian follicles (141,142) and thus could result in premature loss of ovarian follicles and early menopause in smokers. The fact that former smokers have
only a slightly earlier menopause than nonsmokers is not wholly consistent with this but could reflect a duration and thus a dose effect. Because drug metabolism is enhanced in smokers (143), estrogen also may be more rapidly metabolized in the livers of smokers, possibly leading to an earlier decline in estrogen levels (144). Smoking has also been observed to have antiestrogenic effects (145). Greater prevalence of hysterectomy among premenopausal smokers than nonsmokers (137,146) does not appear to account for the earlier menopause in smokers (147). Although one study reported that nonsmoking women whose spouses smoked had an age at menopause resembling that of smokers (148), very little is known about the effect of passive smoke exposure on age at menopause. 2. BODY MASS AND COMPOSITION
A number of studies have examined the relation of body mass to age at menopause, and the findings have been rather inconsistent. Some studies have reported both increased body mass (indicated by weight for height) and upper body fat distribution (indicated by waist-to-hip ratio) to be positively associated with later age at menopause (62,68,144,146) and increased sex hormone concentrations (149), although other studies report no significant association of body mass with age at menopause (63,64,70,76,150,151). Some studies have found a relationship between weight (139) or increased upper body fat distribution (150) and earlier age at menopause, particularly in smokers. One study reported earlier menopause in women on weight reduction programs or who had gained more than 26 pounds between the ages of 20 and 45 years (76). Some of these discrepant findings may be explained by differences in study design (cross-sectional or retrospective versus prospective) or analysis (e.g., inadequate or varying control of confounding variables or survival analysis versus comparison of crude means). In general, the better-designed and -analyzed studies show no relationship. Although body mass and composition may be related to age at menopause and risk of developing symptoms, they are also related inversely to physical activity, alcohol consumption, and education and positively related to infertility and parity (152). Further research is needed to examine the independent contribution or interactive effect of body mass and composition and these other factors on the age at and course of menopause, using appropriate longitudinal study design and data analysis techniques that control for the effects of multiple confounding variables simultaneously.
3. PHYSICAL ACTIVITY
Exercise results in changes in a number of endocrine parameters (estradiol, progesterone, prolactin, LH, and FSH), both during and after intense physical activity
83
CHAPTER 6 Epidemiology of Menopause (153-155), with concentrations of these hormones tending to be lower at rest (153,154,156). Athletes experience a later age at menarche and increased incidence of anovulation (157) and amenorrhea (158) and, in those who menstruate, a shortened luteal phase and reduced mean and peak progesterone levels (152,156). Although exercise is associated with decreased concentrations of reproductive hormones and frequency of ovulation, few studies have examined the effect of exercise on age at menopause, although one study of modest size has reported no relationship (76). 4. OCCUPATIONAL/ENVIRONMENTALFACTORS
Although almost nothing is known about the effects of occupational or other environmental factors on age at and course of menopause, occupational exposures and stressors (such as shift, hours worked, hours spent standing, and heavy lifting) have been shown to increase risk of adverse pregnancy outcomes (159-162) and to affect menstrual cycle length and variability and fecundability (163-166). In addition, a number of environmental exposures, such as dichlorodiphenyltrichloroethane (DDT) and polychlorinated biphenyls, have been shown to have estrogenic activity and may be associated with an increased risk of breast cancer (167,168), although this association has not been consistently observed (169,170). Thus, it is reasonable to expect that occupational and environmental exposures may be related to endocrine disruption that is reflected in altered age at menopause. One recent study showed a modest effect on age at menopause in women in Seveso, Italy, exposed to 2,3,7,8-tetrachlorobenzo-p-dioxin (TCDD) during a chemical plant explosion in 1976; T C D D is a halogenated compound that may affect ovarian function (171). Another study showed that exposure to 1,1-dichloro2,2-bis(p-chlorophenyl) ethylene (DDE) was also associated with earlier menopause (172). It is estimated that 40 million women in the United States alone, and several hundred million worldwide (4), will have experienced the menopausal transition between 1990 and 2010, due to the aging of the baby boomer generation (173). Currently, approximately 70% of American women have worked outside the home (174). Thus, this period in reproductive epidemiologic research presents a prime opportunity to learn more about the effects of occupational and environmental exposures on the menopause transition in these women. 5. DIET
A study from Papua New Guinea has suggested that malnourished women have cessation of menses about 4 years earlier than well-nourished women (175). This is consistent with other studies showing that women with greater weight (108,139) and height (72) may have a later age at menopause. Vegetarians have also been observed to
have an earlier age at menopause in one study (176). Inclusion of meat in the diet of vegetarians has been observed to increase the episodic releases of LH and FSH and the length of the menstrual cycle (177). Thus, meat may modify the interaction of hormones along the hypothalamicpituitary-ovarian axis. At least one study has reported that increased meat or alcohol consumption is significantly associated with later age at menopause after adjusting for age and smoking (69). Dietary fiber (whose intake tends to be inversely related to meat intake) may interrupt enterohepatic circulation of sex hormones, leading to the lower estrogen concentrations observed in vegetarian women (178). Premenopausal women administered soy have shown increased plasma estradiol concentrations and follicular phase length, delayed menstruation, and suppressed midcycle surges of LH and FSH (179). In postmenopausal women fed soy, FSH and LH did not decrease significantly, nor did sex hormone-binding globulin (SHBG) increase, and little change occurred in endogenous estradiol or body weight, although a small estrogenic effect on vaginal cytology was observed (180). The role of dietary phytoestrogens, fat, protein, and other nutrients in affecting age at menopause and risk and severity of menopausal symptoms in perimenopausal and postmenopausal women remains to be systematically studied.
B. Presentation and Symptoms of Menopause 1. SMOKING
Smokers have lower serum estradiol and estrone concentrations among postmenopausal women (181) and lower urinary estrogen among premenopausal women (182). These hormonal effects may be related to the findings, in the few studies that have examined it, that smokers report more hot flashes (Table 6.1) and irritability than nonsmokers (29,44,84), as well as more change in sexual desire (85). Current smoking has been associated with increased reporting of symptoms during the menopausal transition in a number of studies (29,30,44, 85-87,183-189). However, one study reported no significant increase in hot flashes in smokers (86) but showed that thin women who smoked premenopausally had the greatest increase in hot flashes. Relatively little information is available about the relation of passive smoke exposure or whether smoking affects severity or frequency of symptoms, but one crosssectional study has shown a significant association of reporting of vasomotor symptoms with higher reported amounts of passive exposure to cigarette smoke (26). 2. BODYMASS AND COMPOSITION
Much of the early clinical literature suggested that higher body weight might reduce the probability of experiencing symptoms, particularly hot flashes (93,190,191), as a result of
84
GOLD AND GREENDALE TaBLe 6.1
Publication year and first author (citation) 1994 Schwingl (86)
1995 Collins
(85)
1997 Kuh (44)
1997 Leidy (84)
1998 Staropoli 088)
2000 Gold (29)
Observational Studies of Smoking (Active and Passive) and Vasomotor Symptoms
Sample and design characteristics
VMS: how measured and frequency
Cross-sectional control group of 334 from populationbased study of reproductive cancers Ages 45-98 80 black, 254 white --- 9 years postmenopausal Cross-sectional Age 48 Recruited through Swedish population register 73% of 1324 premenopausal Longitudinal for 11 years Population-based, socially stratified cohort of 1213 Assessed at age 47 Pre-, peri-, and postmenopausal Cross-sectional, community-based sample of 152 from employees, health clinics, and exercise groups Ages 4 0 - 6 0 Pre-, peri-, and postmenopausal Cross-sectional clinic-based sample of 233 Ages 4 5 - 6 4
Cross-sectional Ages 40-55 Community-based sample of 16,065 Pre-, peri-, and postmenopausal
Smoking: how measured
Results"
Hot flashes since periods stopped Yes vs. no
Smoking at menopause Ever, current, age starting, number of years since quitting, total years smoking Starting or stopping smoking within 3 years of menopause
OR = 1.9,p = 0.03 for smokers with BMI < 24.2, ns for all others Adjusted for age at menopause, age at menarche, education, periods usually irregular, alcohol use
Frequency of hot flashes, excessive sweating, waking up sweating Combined three symptoms into VMS; any vs. none Degree bothered by hot flashes, cold or night sweats, and trouble sleeping over last 12 months Grouped three symptoms together as any vs. no VMS Current hot flashes and sweating Any vs. none
Any current vs. none
[3 = 0.09 (OR = 1.09), p = 0.003 Adjusted for menopausal status, education, history of premenstrual syndrome
Current, former, or nonsmoker
OR = 1.6, 95% CI 1.2, 2.2 Adjusted for life and work stress; menopausal status; education; and history of health problems, anxiety, and depression
Smoking habits at interview
OR = 1.1, ns Unadjusted
Frequency and severity of perimenopausal hot flashes at time periods stopped or since periods became irregular Any vs. none
Smoking at menopause and ever-smoking Pack-years = number of cigs/day multiplied by number of years of smoking
Hot flashes, night sweats in prior 2 weeks Combined into VMS, any vs. none
Current and former smoking Packs per day
OR = 1.6, 95% CI 1.1, 2.1 for smoking at menopause Adjusted for maternal history of hot flashes, BMI, use of oral contraceptives, ethnicity, income, education Unadjusted OR for + 41 packyears = 2.5, 95% CI 1.2, 4.9 OR = 1.50, 95% CI 1.28, 1.76 for < 10 cigs/day; OR = 1.65, 95% CI 1.42, 1.92 for 10-19 cigs/day; OR = 1.68, 95% CI 1.46, 1.94 for >-- 20 cigs/day Adjusted for age, education, parity, menopausal status, marital status, BMI, race/ ethnicity, difficulty paying for basics, physical activity
VMS, vasomotor symptoms m in most cases refers to hot flashes and night sweats or sweating but in some cases only refers to hot flashes, which in most studies are highly correlated with the former symptoms; OR, odds ratio; CI, confidence interval; ns, nonsignificant; BMI, body mass index; cigs, cigarettes. aNote that because controlled covariates differ among all studies, adjusted point estimates are not directly comparable.
85
CHAPTER 6 Epidemiology of Menopause
TABLE 6.1 Publication year and first author (citation) 2000 Dennerstein (183)
2003 Li (189)
2003 Whiteman (187)
2004 Gold (26)
Observational Studies of Smoking (Active and Passive) and Vasomotor Symptoms - - cont'd Sample and design characteristics
VMS: how measured and frequency
Longitudinal for 7 years Ages 45-55 years at baseline Population-based from RDD in Australia 172 went from pre- to postmenopause Cross-sectional Population-based sample of 6917 from Lund, Sweden Ages 50-64 years Pre- and postmenopausal Cross-sectional Population-based sample 209 African American, 874 non-African American Ages 40-60 Pre-, peri-, and postmenopausal Cross-sectional Ages 42-52 years Community-based sample of 2823: 750 African American; 1 1 4 8 Caucasian; 218 Chinese; 239 Hispanic; 198 Japanese Pre- and early perimenopausal
2005 Guthrie (185)
Longitudinal Age 45-55 years at baseline 350 communitybased women
2005 Ford (186)
Longitudinal Age 24-44 years at baseline 660 communitybased Caucasian women
ln, natural logarithm; no. = number.
Smoking: how measured
Results
"Bothersome" hot flashes in previous 2 weeks
History of smoking Pack-years
OR = 7.0, 95% CI 1.6, 30.2 for -> 10 pack-years Adjusted for number of symptoms, occupation, estradiol at late perimenopause
Hot flashes/sweats bothersome or interfered with quality of life Yes vs. no
Nonsmoking, -< 14 cigs/ day, -> 15 cigs/day
Ever hot flashes and severity and frequency Moderate or severe vs. not and daily vs. not
Ages smoked and average amount smoked Current, former, amount currently smoked, and pack-years of smoking
Number of days in past 2 weeks reporting hot flashes, cold sweats, night sweats Summed any vs. none for each of 3 symptoms for possible 0, 1, 2, or 3 symptoms
Current and former smoking in packs per day Passive smoke exposure in total person-hours/ week at home, at work, or in other public/social settings
Frequency of bothersome hot flashes in previous 2 weeks 83% of women in cohort reported bothersome hot flashes Hot flushes or flashes and night sweats 18% at baseline, 50% at 9 years
Current smoking (yes/no)
OR = 1.55, 95% CI 0.95, 2.54 for -> 15 cigs/day in postmenopausal women adjusted for age, education, menopausal status, employment, oophorectomy, alcohol consumption, increased weight, history of cancer OR = 1.9, 95% CI, 1.3, 2.9 for moderate/severe hot flashes; OR = 2.2, 95% CI 1.4, 3.7 for daily hot flashes Adjusted for age, BMI, race, menopausal status, hormone therapy use, herbal supplement use, nulliparity, tubal ligation OR = 1.0, 95% CI 0.8, 1.4 for In (no. cigs + 1); OR = 1.2, 95% CI 1.0, 1.3 for in (no. hours of passive smoke exposure + 1) Adjusted for age, ethnicity, in BMI, menopause status, In alcohol intake, education, in fat intake, In dietary fiber intake, genistein intake, In calories, history of premenstrual symptoms, use of over-the-counter pain medication, physical activity, comorbidities, stress Adjusted OR = 1.65, p = 0.005 for current smoking Adjusted for age, menopausal status, in estradiol, In FSH, exercising every day
Current, former, never
Adjusted OR = 2.8 (95% CI 1.5, 5.3) for current smokers Adjusted for age, logBMI, IogFSH, testosterone, logestradiol, smoking, menopausal status, use of oral contraceptives, use of hormone therapy, marital status, parity continued
86
GOLD AND GREENDALE
TABLE 6.1 Publication year and first author (citation) 2006 Gold (30)
Observational Studies of Smoking (Active and Passive) and Vasomotor Symptoms - - cont'd
Sampleand design characteristics
VMS: how measured and frequency
Longitudinal Age 42-52 years and pre- or early perimenopausal at baseline Community-based sample of 2784: 930 African American; 1543 Caucasian; 284 Hispanic; 250 Chinese; 281 Japanese
Number of days in past 2 weeks reporting hot flashes, cold sweats, night sweats Any vs. no VMS Any VMS - 6 days in past 2 weeks vs. no or any VMS < 6 days 23-43% any VMS among premenopausal to 58-82% any VMS in late perimenopause
higher circulating estrogen levels in heavier postmenopausal women due to peripheral production of estrone in adipose tissue. However, most recent studies have not confirmed this in perimenopausal women (Table 6.2). One study showed no relation of body mass index (BMI) to reporting of hot flashes in nonsmokers (86). Another study reported that women with more lower body fat reported more hot flashes (192). However, several studies have recently reported significantly higher BMI in women reporting hot flashes, pins and needles, backaches, aches/stiffness in joints, shortness of breath, and fluid retention (21,26,29,30,186,193-195). Additionally, two populationbased studies have reported no significant increase in weight at the menopause (196,197), and one large study reported no increase in waist-to-hip ratio with menopause (150), although a larger, more recent study has shown increased body mass index with the menopausal transition (198). 3. PHYSICALACTIVITY
Serum concentrations of estradiol, progesterone, prolactin, LH and FSH all tend to increase during and after intense exercise (153-155), whereas resting values tend to be lower in athletes (154,156). The findings from various studies regarding the effect of physical activity on reporting of symptoms, particularly vasomotor symptoms, have been inconsistent (Table 6.3), perhaps due to differences in techniques in assessing physical activity and in sample sizes. Midlife women who participate in an exercise program have been reported in some studies to experience less frequent and less severe vasomotor symptoms, despite the fact that lower estrogen concentrations are associated with higher levels of physical activity (199,200). However, this has not been consistent in other cross-sectional, case-control, or longitudinal cohort studies and in some intervention studies, some of which have found no association of
Smoking: how measured
Results
Current and former smoking in packs per day Passive smoke exposure in total person-hours/ week at home, at work, or in other public/social settings
Adjusted OR = 1.63 (95% CI 1.25, 2.12) for current smoking in overall cohort; ORs ranged from 1.14 to 3.09 by race/ethnicity Adjusted for menopausal stares, age, body mass index, education, history of premenstrual symptoms, site, symptom sensitivity, baseline anxiety, baseline depressive symptoms
physical activity with symptoms (85,201-207); others have found a protective effect (29,185,208). Because the onset of hot flashes is accompanied by lower circulating levels of plasma [3-endorphins (209) and physical actMty increases secretion of endogenous opioid peptides, particularly ~-endorphins (210), exercise may prevent symptoms. Exercise also appears to have antidepressant effects (211,212) and thus may also be associated with increased well-being and with fewer midlife psychologic symptoms, including negative mood and change in sexual desire (85). 4. DIET
A number of dietary factors are considered to play a role in production, metabolism, and excretion of estrogen; in phases of the menstrual cycle; and in severity of menopausal symptoms. Vegetarian women have been shown to have lower plasma estrone and estradiol concentrations, perhaps due to lower saturated fat intake (213). Further, Asian women, who consume less fat, excrete two to four times as much estrogen and have substantially lower plasma estrone and estradiol concentrations than Caucasian women (122,125). The relation of fat, alcohol, protein, or other nutrient (such as antioxidant) intake to risk of experiencing menopausal symptoms has not been well studied. Nonetheless, some reports have indicated that alcohol may be estrogenic and may contain phytoestrogens (214) and that alcohol intake is inversely associated with levels of SHBG (149,215,216). However, at least one case-control (86) and one longitudinal study (30) found no association of alcohol consumption with vasomotor symptom reporting, although amounts consumed were relatively low (Table 6.4). One cross-sectional study (194) did report a significant positive association of hot flashes with number of alcoholic drinks consumed per week.
87
CHAPTER 6 Epidemiology of Menopause TABLE 6.2
Observational Studies of Body Mass Index and Vasomotor Symptoms BMI: how measured and BMI of study participants
Publication year and first author (citation)
Sample and design characteristics
VMS: how measured and frequency
1994 Schwingl (86)
Control group from population-based study of reproductive cancers 344 postmenopausal women Mean age 58 years,--- 9 years postmenopausal Subsample from population-based breast cancer screening project in Netherlands N = 3273 age 40-44 years N = 601 age 54-69 years
Recalled HF when menses stopped Outcome: any vs. no VMS
Not specified
Thin (BMI < 24) smokers, higher odds of VMS OR = 1.9 (p = 0.03)
Self-reported Yes/no, preceding year Prevalence VMS: 18% of younger group 32% of older group Self-reported Any vs. none, time frame not specified
Measured height and weight BMI groups: < 22 (low) 22-25 (medium) > 25 (high)
Medium vs. low OR 1.3 (1 - 1.7) High vs. low OR 1.7 (1.3-2.2) Adjusted for age, menopause status, and waist/hip ratio Prevalence VMS: Low: 50% Medium: 70% High: 74% Unadjusted OR = 1.09 per BMI unit Adjusted for age, professional status, FSH, estradiol
1996 den Tonkelaar (193)
1997 Chiechi (195)
Convenience sample from outpatient menopause clinic N = 181 (age not reported)
1998 Wilbur (21)
24-cell randomly selected quota sample (professional status, race, age) Postmenopausal
Self-reported Any vs. none, past 2 weeks
2001 Freeman (194)
Community-based sample 218 Caucasian; 218 African-American women with cycles 22- 35 days Mean age 41 years Community-based sample 750 African American; 1148 Caucasian; 218 Chinese; 239 Hispanic; 198 Japanese; 2823 total Pre- and early perimenopausal mean age 46 years
Structured interview HF past month Frequency Outcome: any vs. no VMS Prevalence VMS: 27% Self-report questionnaire Hot flashes, cold sweats in past 2 weeks Any vs. none of each, summed (range
2004 Gold (26)
Measured height and weight BMI groups: < 24 (low) 24-27 (medium) > 27 (high) Measured height and weight Mean BMI = 28 Mean BMI by VMS: VMS present = 34 V-MS absent = 21 Measured height and weight Analyses per unit BMI
Measured height and weight Overall sample Median BMI = 28.5
0-3)
V-MS,vasomotor symptoms; BMI, body mass index (kg/squared meters); OR, odds ratio; CI, confidence interval. aNote that because controlled covariates differ among all studies, adjusted point estimates are not directly comparable.
Results a
OR = 1.04 per BMI unit Adjusted for menopause symptoms, FSH, anxiety, cycle day
OR = 1.2 (p < 0.05) log BMI 75th percentile compared to 25th percentile Adjusted for ethnicity, menopause stage, education, smoking (active and passive), alcohol, comorbidity, perceived stress, physical activity continued
GOLD AND GREENDALE
88
TABLE 6.2
Observational Studies of Body Mass Index and Vasomotor Symptoms - - cont'd
Publication year and first author (citation)
Sample and design characteristics
VMS: how measured and frequency
BMI: how measured and BMI of study participants
2005 Ford (186)
Longitudinal Ages 24-44 years at baseline 660 communitybased Caucasian women
Hot flushes or flashes and night sweats 18% at baseline, 50% at 9 years
Measured height and weight Mean 26.8 (+_ 6.1) at baseline
2006 Gold (30)
Longitudinal Ages 42-52 years and preor early perimenopausal at baseline Community-based sample of 2784:930 African American; 1543 Caucasian; 284 Hispanic; 250 Chinese; 281 Japanese
Number of days in past 2 weeks reporting hot flashes, cold sweats, night sweats Any vs. no VMS Any VMS -> 6 days in past 2 weeks vs. no or any VMS < 6 days 23-43% any VMS among premenopausal to 58-82% any VMS in late perimenopause
Measured height and weight Overall sample median BMI = 28.5 at baseline
Plant sterols have also been under study in recent research with regard to their effects on circulating hormones, menstrual cycles, and menopausal symptoms. Pbytoestrogen is a term that includes classes of compounds that are nonsteroidal and either of plant origin or derived from metabolism of precursors in plants eaten by humans (217). The main classes of compounds are isoflavones and lignans. They structurally resemble estradiol and have been shown to have weak estrogenic actMty, to compete with estradiol for binding to estrogen receptors in tissues (218,219), and when ingested to have estrogenic and antiestrogenic effects, depending on the concentrations of circulating endogenous estrogens and estrogen receptors (220,221). In rats, the most potent of these, coumestrol, suppressed estrous cycles but did not behave as a typical antiestrogen (222). Soy products are rich in phytoestrogens, which have been detected in high concentrations in the plasma or urine of indMduals who consumed soy or other phytoestrogens (223). Other less concentrated dietary sources of phytoestrogens include rice, corn, alcohol, cereal bran, whole wheat, and beans (224). In Japanese women, phytoestrogen excretion is 100 times higher and endogenous estrogen excretion is 100 to 1000 times higher than in American and Finnish women (225). Differences in phytoestrogen intake may be a (partial)
Results Adjusted OR per unit logBMI = 6.5 (95% CI 1.9, 21.6) Adjusted for age, logFSH, testosterone, logEstradiol, smoking, menopausal status, use of oral contraceptives, use of hormone therapy, marital status, parity Adjusted OR = 1.03 (95% CI 1.01, 1.04) per unit increase in BMI in overall cohort; adjusted ORs ranged from 1.02 to 1.05 by race/ethnicity Adjusted for menopausal status, age, smoking, education, history of premenstrual symptoms, site, symptom sensitivity, baseline anxiety, baseline depressive symptoms
explanation for the differences in frequencies of menopausal symptoms observed in Asian and Caucasian women, although this is not currently known and has not been confirmed in one cross-sectional (26) and one longitudinal (30) study. Urinary excretion of phytoestrogens and the concentration of plasma SHBG have been positively associated with dietary intake of fiber, which has been inversely related to plasma percentage of free estradiol (226). In postmenopausal women whose diets were supplemented with soy or wheat flour (which contain less potent enterolactones), statistically signific a n t - 4 0 % and 25%, respectivelymreductions in hot flashes were observed, while vaginal cell maturation was unchanged and FSH decreased (227). In addition, in a small randomized trial of a 12-week phytoestrogen-rich diet, postmenopausal women on the diet showed significantly increased SHBG, significant reduction in hot flashes and vaginal dryness, and significant increases in serum concentrations of phytoestrogens, though no significant change in estradiol (217). Other observational and intervention studies, examining different types of phytoestrogens and different dosages, have found a protective effect on vasomotor symptoms (228-233), but other studies have found no protective effect (234-238). In summary, environmental factors do influence the menopausal transition. Active smoking has been consistently
89
CHAPTER 6 Epidemiology of Menopause
TABLE 6.3
Observational and Interventional Studies of Physical Activity and Vasomotor Symptoms
Publication year and first author (citation)
Sample and design characteristics
VMS: how measured and frequency
1998 Ivarsson (208)
Population-based cohort Cross-sectional analyses 739 postmenopausal No oophorectomy 35% were HT users
1990 Wilbur (204)
Volunteer sample of 386 women from a bone health study Cross-sectional Pre-, peri-, and postmenopausal No HT No oophorectomy Subsample of 214 women from a community-based survey Cross-sectional Perimenopausal No HT use No oophorectomy 90% Caucasian Randomly selected peri- and postmenopausal women from large HMO case control (N = 171) No HT use No oophorectomy All Caucasian
Light discomfort from flashes, moderate hot flashes, severe hot flashes 33.7% light, 37.5% moderate, 13.5% severe Self-reported over "past few months" Prevalence of VMS 30% on overall sample
1999 Li (205)
1999 Sternfeld (201)
PA: how ascertained
Results
Intensity-based sport and recreational PA Strenuous compared to sedentary
RR = 0.26 (0.10, 0.71) Only unadjusted analyses reported
Usual PA in multiple domains during prior year Ergometer measured fitness
No association between self-reported PA or measured fitness and VMS
Self-reported VMS in past year Prevalence of VMS low; mean score of 0.8 corresponds to rare on scale
Usual current and long-term PA, recreational
No association between current or prior PA and VMS
Self-reported VMS Cases had at least 1 VMS/day in 3 months following FMP Controls had VMS less than 1x/week in 3 mos after FMP
Self-reported habitual PA in multiple domains, in prior year
Usual current and long-term PA, three domains Vigorous exercise for greater than 3 hours/week Exercise every day
No association between vigorous recreational PA and VMS: OR = 1.03 (0.97, 1.1) for 50-unit increment in PA No association between PA in any domain and VMS No independent effect of PA on VMS PA + HT was not related to lower VMS than HT alone Adjusted OR = 0.94, ? = 0.01 Adjusted for age, menopausal status, In estradiol, In FSH, current smoking
2003 Li (189)
Substudy of 239 women from community survey Cross-sectional All postmenopausal HT permitted
Self-reported VMS in past year
2005 Guthrie (185)
Longitudinal Age 45-55 years at baseline 350 community-based women
Frequency of bothersome hot flashes in previous 2 weeks 83% of women in cohort reported bothersome hot flashes
HT, hormone therapy; VMS, vasomotor symptoms; PA, physical activity; RR, relative risk; OR, odds ratio.
associated with a 1- to 2-year earlier menopause (26,63,66,69, 70, 76,132-136), in a dose response relationship, although the role of passive smoke exposure, shown in one study to be associated with vasomotor symptoms (26), is uncertain. Findings regarding the relations of body weight and body composition to age at menopause have been inconsistent. However, a number of recent studies have found a significant positive association of body mass index to vasomotor symptom reporting, contrary to the expectation based on early
observations in menopausal women of reduced symptom reporting due to increased estrone levels in heavier women as a result of conversion from androstenedione in peripheral fat. The relations of diet, physical activity, and occupational or other environmental factors to age at menopause largely have not been investigated. Active smoking has been associated in numerous studies with increased symptom reporting during the menopause transition (29,30,44,84,85,87,181,183,187189). The findings regarding a relation of physical activity to
G O L D AND GREENDALE
90
TABLE 6.4 Publication year and first author (Citation)
Observational Studies of Alcohol Use and Vasomotor Symptoms
Sample and design characteristics
VMS: how measured and frequency
Alcohol: how ascertained
Results
Control group from populationbased study of reproductive cancers 344 postmenopausal women Mean age 58 years -~ 9 years postmenopausal Community-based sample 218 Caucasian; 218 AfricanAmerican women with cycles 22-35 days Mean age 41 years
Recalled HF when menses stopped Outcome: any vs. no HF
Ever (any) vs. never use
OR = 1.3 (/5 = 0.13) Adjusted for age, BMI, smoking
Structured interview HF past month Outcome: any vs. no HF Prevalence of HF: 27%
Number of drinks per
OR = ~.~ (p =
2004 Gold (26)
Community-based sample 750 African-American; 1148 Caucasian; 218 Chinese; 239 Hispanic; 198 Japanese; 2823 total Pre- and early perimenopausal Mean age 46 years
Self-report questionnaire Hot flashes, cold sweats in past 2 weeks Any vs. none of each, summed (range 0 - 3)
Food frequency questionnaire % of calories from alcohol (converted to grams) Median intake 6 grams of alcohol
2006 Gold (30)
Longitudinal Age 42-52 years and pre- or early perimenopausal at baseline Community-based sample of 2784:930 African American; 1543 Caucasian; 284 Hispanic; 250 Chinese; 281 Japanese
Number of days in past 2 weeks reporting hot flashes, cold sweats, night sweats Any vs. no VMS Any VMS >- 6 days in past 2 weeks vs. no or any VMS < 6 days 23-43% any VMS among premenopausal to 58-82% any VMS in late perimenopause
Food frequency questionnaire % of calories from alcohol (converted to grams) Median intake 6 grams of alcohol
1994 Schwingl
(86)
2001 Freeman (194)
week
Average drinks per week: HF group = 3; no HF group = 1.5
0.0002) adjusted for age, race, other symptoms, FSH, anxiety, BMI cycle day No effect of alcohol (median of nonzero alcohol use vs. no alcohol use) Adjusted for age, BMI, smoking, comorbidity, menopause stage No effect of alcohol Adjusted for menopausal status, age, education, smoking, body mass index, history of premenstrual symptoms, site, symptom sensitivity, baseline anxiety, baseline depressive symptoms
VMS, vasomotorsymptoms;HF, hot flushes; OR, odds ratio; BMI, body mass index. symptom reporting have been inconsistent, but the best designed studies suggest no effect. Phytoestrogen intake has been related to reduced frequency or severity of hot flashes in some studies (217,227-233), but this result has not been consistent (26,30,234-238), quite possibly due to the different types and dosages of phytoestrogens that have been examined. The role of alcohol consumption in relation to vasomotor symptoms has been inconsistent across the few studies that have examined it, and the role of other dietary factors is only beginning to be explored.
V.
CONCLUSIONS
Despite important methodologic differences and the limitations in the study designs used and the populations studied in the accumulating literature on the menopausal experience, an interesting and complex picture is emerging. A number of
demographic (e.g., education, employment, race/ethnicity), menstrual and reproductive, and lifestyle (e.g., smoking and diet) factors appear to be important determinants of the age at which natural menopause occurs and to have meaningful relationships to the varied symptom experiences of women. African American and Latina race/ethnicity, smoking, lower parity, vegetarian diet and undernutrition, and lower socioeconomic status have been found fairly consistently to be associated with earlier menopause, an indicator of reduced longevity. Symptom reporting varies by race/ethnicity, with less reporting of vasomotor symptoms in most Asian populations and increased reporting of vasomotor symptoms and vaginal dryness in African-American and Hispanic women. History of premenstrual tension or symptoms, smoking, and lower socioeconomic status have been associated with increased symptom reporting. However, a number of the relationships are inconsistent (e.g., the role of body mass and composition, diet, and physical
91
C~taVTER 6 Epidemiology of Menopause activity), possibly due to varying methodologic approaches and limitations, and others remain largely unexplored (e.g., passive smoke exposure and occupational and other environmental exposures). Therefore, much remains to be learned about how these factors affect hormones at the physiologic level and thus determine the onset of the perimenopause, the timing of the final menstrual period, and the occurrence of the constellation of symptoms that are associated with the menopause transition. Furthermore, increased understanding of the underlying physiologic bases of these influences needs to include potential racial/ethnic differences in physiologic responses to lifestyle factors and other environmental exposures, as well as increased understanding of the cultural contexts, cultural differences, and cultural sensitivities that affect the presentation and experience of the menopausal transition. Increasing knowledge about these relationships ultimately offers w o m e n and their health care providers enhanced choices and alternatives, based on deeper understanding, to deal with the individual presentations of menopause.
References 1. Avis NE, Kaufert PA, Lock M, McKinlay SM, Vass K. The evolution of menopausal symptoms. Ba,Tlieres Clin Endocrinol Metab 1993;7:17- 32. 2. Cramer DW, Xu H. Predicting age at menopause. Maturitas 1996;23: 319-326. 3. Hahn RA, Eaker E, Rolka H. Reliability of reported age at menopause. AmJEpidemio11997;146:771-775.
4. World Health Organization. Research on the menopause in the 1990s. Geneva, Switzerland: World Health Organization, 1996. 5. Sowers MF, LaPietra MT. Menopause: its epidemiology and potential association with chronic diseases. Epidemiol Rev 1995;17:287-302. 6. Gosden RG. Biology of the menopause."the causesand consequencesof ovarian ageing. London: Academic Press, 1985. 7. Burger HG, Dudley EC, Hopper JL. The endocrinology of the menopausal transition: a cross-sectional study of a population-based sample. J Clin Endocrinol Metab 1995;80:3537-3545. 8. SouleMR, BremnerVVJ.The menopause and climacteric: endocrinologic basis and associated symptomatology.JAm Geriatrics Soc 1982;30:547. 9. Thomford PJ, Jelovsek FR, Mattison DR. Effect ofoocyte number and rate of atresia on the age of menopause. Repro Toxico11987;1:41-51. 10. Sherman BM, West JH, Korenman SG. The menopausal transition: analysis of LH, FSH, estradiol and progesterone concentrations during menstrual cycles of older women.J Clin Endocrinol Metab 1976;42:629. 11. Santoro N, Rosenberg-BrownJ, Adel T, SkurnickJH. Characterization of reproductive hormonal dynamics in the perimenopause.J Clin Endocrinol Metab 1996;81:1495-1501. 12. Upton GV. The perimenopause: physiologic correlates and clinical management. J Repro Med 1982;27:1. 13. Shideler SE, DeVane GW, Kalra PS, Benirschke K, Lasley BL. Ovarian pituitary hormone interactions during the menopause. Maturitas 1989;11:331-339. 14. Grodin JM, Siiteri PK, MacDonald PC. Extraglandular estrogen in the postmenopause. In: Ryan KJ, Gibson DC, eds. The menopause and aging. Washington DC: U.S. Government Printing Office, 1973;15. 15. Nimrod A, Ryan KJ. Aromatization of androgens by human abdominal and breast fat tissue.J Clin EndocrinolMetab 1975;40:367. 16. Kase N. Estrogen replacement therapy for the menopause. In: Diamond MC, Korenbrot CC, eds. Hormonal contraceptives, estrogens and human welfare. New York: Academic Press, 1978;75.
17. HershcopfR, Bradlow HL. Obesity, diet, endogenous estrogens and the risk of hormone-sensitive cancer. Am J Clin Nutr 1987;45:283-289. 18. Johansson AG, Forslund A, Sjodin A. Determination of body composition: a comparison of dual energyx-ray absorptiometry and hydrodensitometry. Am J Clin Nutr 1984;57:373-376. 19. Cutler WB, Garcia CR. The medical management of menopause and perimenopause. Philadelphia: JB Lippincott, 1984. 20. MetcalfMG, Donald RA, LiveseyJH. Pituitary-ovarian function before, during and after the menopause: a longitudinal study. Clin Endocrinol 1982;17:489-494. 21. Wilbur JE, Miller AM, Montgomery A, Chandler E Sociodemographic characteristics, biological factors, and symptom reporting in midlife women. Menopause 1998;5:43- 51. 22. Greene JG. The cross-sectional legacy: an introduction to longitudinal studies of the climacteric. Maturitas 1992;14:95-101. 23. Barbo D. The physiology of the menopause. Med Clin N Am 1987;71:11. 24. World Health Organization. Research on the menopause. Geneva, Switzerland: World Health Organization, 1981. 25. Matthews KA, Wing RR, Kuller LH, Meilahn EN, Plantinga P. Influence of the peri-menopause on cardiovascular risk factors and symptoms of middle-aged healthy women. Arch Int Med 1994;154:2349-2355. 26. Gold EB, Block G, Crawford S, et al. Lifestyle and demographic factors in relation to vasomotor symptoms: baseline results from the Study of Women's Health Across the Nation. Am J Epidemiol 2004;159: 1189-1199. 27. Batrinos ML, Panitsa-Faflia C, Pitoulis S, et al. The clinical features of the menopause and its relation to the length of pregnancies and lactation. Maturitas 1979;1:261-268. 28. McKinlay SM, Brambilla DJ, Posner JG. The normal menopause transition. Maturitas 1992;14:103-115. 29. Gold EB, Sternfeld B, Brown C, et al. The relation of demographic and lifestyle variables to symptoms in a multi-racial/ethnic population of women aged 40-55 years. Am J Epidemio12000;152:463-473. 30. Gold EB, Colvin A, Avis N, et al. Longitudinal analysis of vasomotor symptoms and race/ethnicity across the menopausal transition: Study of Women's Health Across the Nation (SWAN). Am J Public Health 2006;96:1226-1235. 31. Kronenberg F. Hot flashes: epidemiologyand physiology.Ann NYAcad Sci 1990;592:52- 86. 32. Hagstad A, Janson PO. The epidemiology of climacteric symptoms. Acta Obstet Gynecol &and Supp11986;134:159.
33. Samsioe G, Bryman I, Ivarsson E. Some anthropological aspects of the climacteric syndrome.Acta Obstet Gynecol Scand Supp11985;130:5- 7. 34. Dennerstein L. Well-being, symptoms and the menopausal transition. Maturitas 1996;23:147-157. 35. Stadberg E, Mattsson LA, Milsom I. The prevalence and severity of climacteric symptoms and the use of different treatment regimens in a Swedish population. Acta Obstet Gynecol Scand 1997;76:442-448. 36. O'Connor VM, DelMar CB, Sheehan M, et al. Do psychosocial factors contribute more to symptom reporting by middle-aged women than hormonal status? Maturitas 1995;20:63-69. 37. OldenhaveA, Jaszman LJ, Haspels AA. Impact of climacteric on wellbeing: a survey based on 5213 women, 39 to 60 years old. Am J Obstet Gyneco11993;168:772- 780. 38. Hammar M, Berg G, Fahraeus L, Larsson-Cohn U. Climacteric symptoms in an unselected sample of Swedish women. Maturitas 1984;6: 345-350. 39. Everett HS. Urology in the female.AmJ Surg 1991;52:521-659. 40. Asmussen M, Miller A. Clinical gynaecological urology. Oxford, UK: Blackwell Scientific Publications, 1983;21. 41. Waetjen LE, Liao S, Johnson WO, et al. Factors associated with prevalent and incident urinary incontinence in a cohort of midlife women: a longitudinal analysis of data: study of women's health across the nation.AmJEpid 2007;165:309-318.
92 42. Versi E, Cardozo L, Studd J, et al. Urinary disorders and the menopause. Menopause 1995;2:89- 95. 43. Neugarten BL, Kraines RJ. "Menopausal symptoms" in women of various ages. PsychosomMed 1965;27:266-273. 44. Kuh DL, Wadworth M, Hardy R. Women's health in midlife: the influence of the menopause, social factors and health in earlier life. BrJ Obstet Gynaeco11997;104:923- 933. 45. Mikkelsen A, Holte A. A factor-analytic study of climacteric symptoms. Psychiatry SocialSci 1982;2:35-39. 46. Cooper GS, Sandler DP. Age at natural menopause and mortality. Ann Epidemio11998;8:229-235. 47. Wise AM, Krajnak KM, Kashon ML. Menopause: the aging of multiple pacemakers. Science 1996;273:67-70. 48. Snowdon DA, Kane RL, Beeson WL, et al. Is early natural menopause a biologic marker of health and aging? Am J Public Health 1989;79: 709-714. 49. Ossewaarde ME, Bots ML, Verbeek ALM, et al. Age at menopause, cause-specific mortality and total life expectancy. Epidemiology2005;16: 556-562. 50. De Kleijn MJ, van der Schouw YT, Verbeek AL, et al. Endogenous estrogen exposure and cardiovascular mortality risk in postmenopausal women. Am J Epidemio12002;155:339- 345. 51. Van der Schouw YT, van der Graaf Y, Steyerberg EW, et al. Age at menopause as a risk factor for cardiovascular mortality. Lancet 1996;347:714-718. 52. Jacobsen BK, Nilssen S, Heuch I, et al. Does age at natural menopause affect mortality from ischemic heart disease? J Clin Epidemio11997;50: 475-479. 53. Hu FB, Grodstein F, Hennekens CH, et al. Age at natural menopause and risk of cardiovascular disease.Arch Item Med 1999;159:1061-1066. 54. Joakimsen O, Bonaa KH, Stensland-Bugge E, et al. Population-based study of age at menopause and ultrasound assessed carotid atherosclerosis: the Tromso Study.J Clin Epidemio12000;53:525-530. 55. Kritz-Silverstein D, Barrett-Connor E. Early menopause, number of reproductive years, and bone mineral density in postmenopausal women. Am J Public Health 1993;83:983- 988. 56. Van Der Voort DJ, Van Der Weijer PH, Barentsen R. Early menopause: increased fracture risk at older age. OsteoporosInt 2003;14:525-530. 57. Kelsey JL, Gammon MD, John EM. Reproductive factors and breast cancer. EpidemiolRev 1993;15:36-47. 58. MonninkhofEM, van der Schouw YT, Peeters PH. Early age at menopause and breast cancer: are leaner women more protected? A prospective analysis of the Dutch DOM cohort. Breast Cancer Res Treat 1999;55:285-291. 59. De Graaff J, Stoke LA. Age at menarche and menopause of uterine cancer patients. EurJ ObstetBynecolReprod Bio11978;8:187-193. 60. Franceschi S, La Vecchia C, Booth M, et al. Pooled analysis of 3 European case-control studies of ovarian cancer. II. Age at menarche and at menopause. IntJ Cancer 1991;49:57-61. 61. Trevoux R, DeBrux J, Castaneir M, et al. Endometrium and plasma hormone profile in the peri-menopause and post-menopause. Maturitas 1986;8:309. 62. Greendale G, Hogan P, Kritz-Silverstein D, et al. (for the PEPI trial investigators). Age at menopause in women participating in the postmenopausal estrogen/progestins interventions (PEPI) trial: an example of bias introduced by selection criteria. Menopause 1995;2:27-34. 63. Luoto R, Laprio J, Uutela A. Age at natural menopause and sociodemographic status in Finland. Am J Epidemio11994;139:64- 76. 64. Stanford JL, Hartge P, Brinton LA, et al. Factors influencing the age at natural menopause. J Chron Dis 1987;40:995-1002. 65. Magursky V, Mesko M, Sokolik L. Age at the menopause and onset of the climacteric in women of Martin district, Czechoslovakia. Int J Ferti11975;20:17- 23.
GOLD AND GREENDALE 66. van Noord PAH, Dubas JS, Dorland M, et al. Age at natural menopause in a population-based screening cohort: the role of menarche, fecundity, and lifestyle factors. Fertil Steri11997;68:95-102. 67. Flint M. Is there a secular trend in age of menopause. Maturitas 1978;1:133-139. 68. Rodstrom K, Bengtsson C, Milsom I, et al. Evidence for a secular trend in menopausal age: a population study of women in Gothenburg. Menopause 2003;10:538- 543. 69. Torgerson DJ, Avenell A, Russell IT, Reid DM. Factors associated with onset of menopause in women aged 45-49. Maturitas 1994;19: 83-92. 70. Gold EB, Bromberger J, Crawford S, et al. Factors associated with age at menopause in a multi-ethnic population of women. Am J Epidemiol 2001;153:865-874. 71. Lawlor IDA, Ebrahim S, Smith GD. The association of socio-economic position across the life course and age at menopause: the British Women's Heart and Health Study. Br J Obstet Gynecol 2003;110: 1078-1087. 72. Brand PC, Lehert PH. A new way of looking at environmental variables that may affect the age at menopause. Maturitas 1978;1: 121-132. 73. McKinlay S, Jefferys M, Thompson B. An investigation of the age at menopause. J Biosoc Sci 1972;4:161-173. 74. Whelan EA, Sandler DP, McConnaughey DR, Weinberg CR. Menstrual and reproductive characteristics and age at natural menopause. Am J Epidemio11990;131:625-632. 75. Treloar AE, Boynton RE, Behn BG, Brown BW. Variation of the human menstrual cycle through reproductive life. Int J Fertil (Part 2) 1966;12:77-126. 76. Bromberger JT, Matthews KA, Kuller LH, et al. Prospective study of the determinants of age at menopause. Am J Epidemiol 1997;145: 124-133. 77. Soberon J, Calderon JJ, Goldzieher JW. Relation of parity to age at menopause. Am J Obstet Gyneco11966;96:96-100. 78. Snieder H, MacGregor AJ, Spector ID. Genes control cessation of a woman's reproductive life: a twin study of hysterectomy and age at menopause. J Clin EndocrinolMet 1998;83:1875-1880. 79. Chompootweep S, Tankeyoon M, Yamarat K, et al. The menopausal age and climacteric complaints in Thai women in Bangkok. Maturitas 1993;17:63-71. 80. Parazzini F, Negri E, LaVecchia C. Reproductive and general lifestyle determinants of age at menopause. Maturitas 1992;15:141 - 149. 81. van Keep PA, Brand PC, Lehert PH. Factors affecting the age at menopause. J Biosoc Sci Supp11979;6:37-55. 82. Gonzales GF, Villena A. Age at menopause in central Andean Peruvian women. Menopause 1997;4:32- 38. 83. Avis N, Crawford S, McKinlay S. Psychosocial, behavioral, and health factors related to menopause symptomatology. J Womens Health 1997;3:103-120. 84. Leidy LE. Menopausal symptoms and everyday complaints. Menopause 4:154-160. 85. Collins A, Landgren BM. Reproductive health, use of estrogen and experience of symptoms in peri-menopausal women, a populationbased study. Maturitas 1995;20:101-111. 86. Schwingl PJ, Hulka BS, Harlow SD. Risk factors for menopausal hot flashes. Obstet Gyneco11994;84:29-34. 87. Dennerstein L, Smith AM, Morse CA, et al. Menopausal symptomatology: the experience of Australian women. MedJ Australia 1993;159; 232-236. 88. Hunter MR, Battersby R, Whitehead MI. Relationships between psychological symptoms, somatic complaints and menopausal status. Maturitas 1986;8:217-228.
CHAPTER 6 Epidemiology of Menopause 89. Scaglia H, Medina M, Pinto-Ferreira AL, et al. Pituitary LH and FSH secretion and responsiveness in women of old age. Acta Endocrino11976;81:673-679. 90. van Keep PA, Kellerhals JM. The impact of sociocultural factors on symptom formation. Some results of a study on aging women in Switzerland. Psychother Psychosom 1974;23:251-263. 91. Green RS, Gold EB, Samuels SJ, Dosemeci M. The relation of occupational solvent exposure to symptom reporting in a sample of Caucasian and Chinese midlife women.J OccupEnviron Med2005;47: 410-423. 92. Kirchengast S. Relations between fertility, body shape and menopause in Austrian women.JBiosoc Sci 1992;24:555-559. 93. Sherman BM, Wallace RB, Bean JA, Chang Y, Schlabaugh L. The relationship of menopausal hot flushes to medical and reproductive experience. J Geronto11981;36:306- 309. 94. McKinlay SM, Jefferys M. The menopausal syndrome. Brit J Prey Med 1974;28:108-115. 95. Skarsgard C, Bjors E, Nedstrand E, et al. Do women's premenstrual symptoms and their mother's climacteric history predispose to their own vasomotor symptoms? Menopause 1996;3:133-139. 96. Hunter M. The south-east England longitudinal study of the climacteric and postmenopause. Maturitas 1992;14:117 - 126. 97. Studd J, Chakravarti S, Oram D. The climacteric. In: Greenblatt RG, Studd J, eds. Clinics in obstetrics and gynecology, vol 1: The menopause. Philadelphia: WB Saunders, 1977. 98. Alvarado G, Rivera R, Ruiz R, et al. Characteristicas del patron de sangrado menstrual en un grupo de mujeres normales de Durango. Ginecol Obstetr Mex 1988;56:127-133. 99. Otolorin EO, Adeyefa I, Osotimehin BO, et al. Clinical, hormonal and biochemical features of menopausal women in Ibadan, Nigeria. A f r J M e d Sci 1989;18:251-255. 100. Beyene Y. Cultural significance and physiological manifestations of menopause, a bicultural analysis. Culture Ailed Psychiat 1986;10: 47-71. 101. World Health Organization. A prospective multicenter trial of the ovulation method of natural family planning. III. Characteristics of the menstrual cycle and of the fertile phase. Fertil Steril 1983;40: 773-778. 102. Boulet M. The menopause and the climacteric in seven Asian countries. In: Sixth International Congress on the Menopause. New Jersey: Parthenon, 1990. 103. Ramoso-Jalbuena J. Climacteric Filipino women: a preliminary survey in the Philippines. Maturitas 19:183-190. 104. McCarthy T. The prevalence of symptoms in menopausal women in the Far East: Singapore segment. Maturitas 1994;19:199-204. 105. Samil RS, Wishnuwardhani SD. Health of Indonesian women, citydwellers of perimenopausal age. Maturitas 1994;19:191-197. 106. Wasti S, Robinson SC, Akhtar Y, Khan S, Badaruddin N. Characteristics of menopause in three groups in Karachi, Pakistan. Maturitas 1993;16:61-69. 107. Blumel J, Cubillos M, Brandt A, et al. Some clinical aspects of menopause. Rev Chil Obstet Gyneco11988;53:278-282. 108. MacMahon B, Worcester J. Age at menopause, United States 1960-1962. Vital Health Stat 11 1966;19:1-19. 109. Kapoor AK, Kapoor S. The effects of high altitude on age at menarche and menopause. JBiometeor 1986;30:21-26. 110. Beall CM. Ages at menopause and menarche in a high altitude Himalayan population. Ann Hum Bio11983; 10:365 - 370. 111. Flint MP. Menarche and menopause in Rajput women. PhD dissertation, City University of New York, 1974. 112. Boulet MJ, Oddens BJ, Lehert P, et al. Climacteric and menopause in seven south-east Asian countries. Maturitas 1994; 19:157-176.
93 113. Haines CJ, Chung TKH, Leung DHY. A prospective study of the frequency of acute menopausal symptoms in Hong Kong Chinese women. Maturitas 1994;18:175-181. 114. Tang GWK. The climacteric of Chinese factory workers. Maturitas 1994;19:177-182. 115. Lock M. Contested meanings of the menopause. Lancet 1991;337: 1270-1272. 116. Liu C-H. Medical care-seeking behavior among climacteric women in Taiwan. In: Sixth International Congress on the Menopause. New Jersey: Parthenon, 1990. 117. Lock M, Kaufert P, Gilbert P. Cultural construction of the menopausal syndrome: the Japanese case. Maturitas 1988;10:317- 332. 118. Goodman MJ, Stewart CJ, Gilbert F. Patterns of menopause: a study of certain medical and physiological variables among Caucasian and Japanese women living in Hawaii. J Geronto11977;32:291-298. 119. Sukwatana P, Meekhangvan J, Tamrongterakul T, et al. Menopausal symptoms among Thai women in Bangkok. Maturitas 1991;13: 217-228. 120. U.S. Congress, Office of Technology Assessment. The menopause, hormone therapy and women's health (OTA-BP-BA-88). Washington, DC: U.S. Government Printing Office, 1992. 121. Hu JF, Zhao XH, Jia JB, et al. Dietary calcium and bone density among middle aged and elderly women in China. Am J Clin Nutr 1993;58:219-227. 122. Wang DY, Key TJ, Pike MC, Boreham J, Chen J. Serum hormone levels in British and rural Chinese females. Breast Cancer Res Treat 1991;$41-45. 123. Sheehy, G. The Silent Passage, Menopause. New York: Random House, p. 2, 1992. 124. Wyon Y, Lindgren R, Lundeberg T, Hammar M. Effects of acupuncture on climacteric vasomotor symptoms, quality of life, and urinary excretion of neuropeptides among postmenopausal women. Menopause 1995;2:3-12. 125. Key TJ, Chen J, Wang DY, Pike MC, Boreham J. Sex hormones in women in rural China and in Britain. BrJ Cancer 1990;59:631-636. 126. Randolph JF, Sowers MF, Gold EB, et al. Reproductive hormones in the early menopausal transition: relationship to ethnicity, body size, and menopausal status. J Clin Endocrinol Metab 2003;88: 1516-1522. 127. Martin MC, Block JE, Sanchez SD, Arnaud CD, Beyene Y. Menopause without symptoms: the endocrinology of menopause among rural Mayan Indians. Am J Obstet Gyneco11993;168:1839-1845. 128. Avis NE, McKinlay SM. A longitudinal analysis of women's attitudes toward the menopause: results from the Massachusetts Women's Health Study. Maturitas 1991;13:65. 129. Lock M. Attitudes of aging: Japanese experience and perceptions of menopause. Cult Med Psychiat 1986;10:23-46. 130. Oddens BJ. The climacteric cross-culturally: the International Health Foundation South-east Asia study. Maturitas 1994;19:155-156. 131. Palmer JR, Rosenberg L, Wise LA, et al. Onset of natural menopause in African American women. Am J Public Health 2003;93:299- 306. 132. McKinlay SM, Bifano NL, McKinlay JB. Smoking and age at menopause in women. Ann Intern ivied 1985;103:350-356. 133. Andersen FS, Transbol I, Christiansen C. Is cigarette smoking a promoter of the menopause. Acta Med Scand 1982;212:137-139. 134. Hiatt RA, Fireman BH. Smoking, menopause, and breast cancer. JNatl Cancer Inst 1986;76:833-838. 135. Hartz AJ, Kelber S, Borkowf H, et al. The association of smoking with clinical indicators of altered sex steroids - - a study of 50,145 women. Pub Health Rep 1987;102:254-259. 136. Brambilla DJ, McKinlay SM. A prospective study of factors affecting age at menopause. J Clin Epidemio11989;42:1031-1039.
94 137. Adena MA, Gallagher HG. Cigarette smoking and the age at menopause. Ann Human Bio11982;9:121 - 130. 138. Jick H, Porter J, Morrison AS. Relation between smoking and age of natural menopause. Lancet 1977;1:1354-1355. 139. Willett W, Stampfer MJ, Bain C, et al. Cigarette smoking, relative weight and menopause. Am J Epidemio11983;117:651-658. 140. Midgett AS, Baron JA. Cigarette smoking and the risk of natural menopause. Epidemiology 1990;1:464-480. 141. Mattison DR, Thorgierssen SS. Smoking and industrial pollution and their effects on menopause and ovarian cancer. Lancet 1978;1:187. 142. Essenberg JM, Fagan L, Malerstein AJ. Chronic poisoning of the ovaries and testes of albino rats and mice by nicotine and cigarette smoke. WestJ Surg Obstet Gyneco11951;59:27-32. 143. Hart P, Farrell GC, Cooksley WGE, Powell LW. Enhanced drug metabolism in cigarette smokers. Br MedJ 1976;3:147. 144. Lindquist O, Bengtsson C. Menopausal age in relation to smoking. Acta Med &and 1979;205:73-77. 145. Michnovicz J, Hershcopf R, Naganuma H, et al. Increased 2hydroxylation of estradiol as a possible mechanism for the antiestrogenic effect of cigarette smoking. N Engl J Med 1986;315: 1305-1309. 146. Daniell HWP. Smoking, obesity, and the menopause. Lancet 1978;2:373. 147. Krailo MD, Pike MC. Estimation of the distribution of the age at natural menopause from prevalence data. Am J Epidemiol 1983;117: 356-361. 148. Everson RB, Sandier DP, Wilcox AJ, et al. Effect of passive exposure to smoking on age at natural menopause. Br MedJ 1986;293:792. 149. London S, Willett W, Longcope C, McKinlay S. Alcohol and other dietary factors in relation to serum hormone concentrations in women in climacteric. A m J Clin Nutr 1991;53:166-171. 150. den Tonkelaar I, Seidell J. Fat distribution in relation to age, degree of obesity, smoking habits, parity and estrogen use: a cross-sectional study of 11,825 Dutch women participating in the DOM project. Int J Obesity 1990;14:753-761. 151. Kaufman DW, Slone D, Rosenberg L, Miettinen OS, Shapiro S. Cigarette smoking and age at natural menopause. Am J Public Health 1980;70:420-422. 152. Kaye S, Folsom A, Prineas RJ, Potter JD, Gapstur SM. The association of body fat distribution with lifestyle and reproductive factors in a population study of postmenopausal women. Int J Obesity 1990;14: 583-591. 153. Cummings SR, KelseyJ, Nevitt MC, O'Dowd KJ. Epidemiology of osteoporosis and osteoporotic fractures. EpidemiolRev 1985;7:178-208. 154. Bonen A, Ling WH, Belcastro AN, Simpson AA. Profiles of selected hormones during menstrual cycles of teenage athletes. J Appl Physiol 1981;50:545-551. 155. Jurkowski JE, Joanes NL, Walker C, et al. Ovarian hormonal responses to exercise.JAppl Physio11978;44:109-114. 156. Loucks AB, Mortola LF, Girtoon L, Yen SSC. Alterations in the hypothalamic-pituitary-ovarian and the hypothalamic-pituitaryadrenal axes in athletic women. J Clin Endocrinol Metab 1989;68: 402-411. 157. Bernstein L, Ross RK, Lobo RA, et al. The effects of moderate physical activity on menstrual cycle patterns in adolescence: implications for breast cancer prevention. BrJ Cancer 1987;55:681-685. 158. Loucks AB, Horvath SM. Athletic amenorrhea: a review. Med Sci Sports Exer 1985;17:56-72. 159. Mamelle N, Laumon B, Lazar E Prematurity and occupational activity during pregnancy. Am J Epidemio11984;119:309- 322. 160. McDonald AD, McDonald JC, Armstrong B, et al. Fetal death and work in pregnancy. BrJ Ind IVied 1988;45:148-157. 161. Beaumont JJ, Swan SH, Hammond SK, et al. Historical cohort investigation of spontaneous abortion in the Semiconductor Health Study: Methods and analyses of risk in fabrication overall and in fabrication work groups. Am J Ind Med 1995;28:735-750.
GoLD AND QREENDALE 162. Swan SH, Beaumont JJ, Hammond SK, et al. Historical cohort study of spontaneous abortion among fabrication workers in the Semiconductor Health Study; agent-level analysis.Am JInd Meal 1995;28:751- 770. 163. Messing K, Saurel-Cubizolles MG, Bourgine M, Kaminski M. Menstrual cycle characteristics and work condition of workers in poultry slaughterhouses and canneries. ScandJ Work Environ Health 1992;18: 302-309. 164. Ng TP, Foo SC, Young T. Menstrual function in workers exposed to toluene. BrJ Ind Med 1992;49:799-803. 165. Eskenazi B, Gold EB, Samuels SJ, et al. Prospective assessment of fecundability of female semiconductor workers. Am J Ind Med 1995;28: 817-832. 166. Gold EB, Eskenazi B, Hammond SK, et al. Prospectively assessed menstrual cycle characteristics in female wafer-fabrication and nonfabrication semiconductor employees.Am J randMed 1995;28:799- 816. 167. Falck Jr F, Ricci Jr A, WolffMS, et al. Pesticides and polychlorinated biphenyl residues in human breast lipids and their relation to breast cancer. Arch Environ Health 1992;47:143-146. 168. WolffMS, Toniolo PG, Lee EW, et al. Blood levels of organochlorine residues and risk of breast cancer.JNatl CancerInst 1993;85:648-652. 169. Krieger N, Wolff MS, Hiatt RA, et al. Breast cancer and serum organochlorines: a prospective study among white, black and Asian women. J Natl Cancer Inst 1994;86:589 - 599. 170. Hunter DJ, Hankinson SE, Laden F, et al. Plasma organochlorine levels and risk of breast cancer. N EnglJ Med 1997;337:1253-1258. 171. Eskenazi B, Warner M, Marks AR, et al. Serum dioxin concentrations and age at menopause. Environ Health Perspect2005;113:858-862. 172. Cooper GS, Savitz DA, Millikan R, Kit TC. Organochlorine exposure and age at natural menopause. Epidemiology 2002;13:729-733. 173. Skolnick AA. At third meeting, menopause experts make the most of insufficient data. JAMA 1992;268:2483- 2485. 174. Meisenheimer JR. How do immigrants fare in the US labor market? Monthly Labor Rev 1992;3 - 19. 175. Scragg RFR. Menopause and reproductive span in rural Nuigini. Proc Annu Syrup Papua New Guinea Med Soc 1973;126-144. 176. Baird DD, Trlavsky FA, Anderson JjB. Do vegetarians have earlier menopause? Proc Soc Epidemiol Res 1988;907- 908. 177. Hill PB, Garbaczewski L, Daynes G, Gaire KS. Gonadotrophin release and meat consumption in vegetarian women. Am J Clin Nutr 1986;43:37-41. 178. Adlercreutz H, Mousavi Y, Loukovaara M, Hamalainen E. Lignans, isoflavones, sex hormone metabolism and breast cancer. In: Hochberg R, Naftolin F, eds. The new biology of steroid hormones. New York: Raven Press, 1992;145-154. 179. Cassidy A, Bingham S, Setchell KDR. Biological effects of a diet of soy protein rich in isoflavones on the menstrual cycle of premenopausal women. Am J Clin Nutr 1994;60:333- 340. 180. Baird DD, Umbach DM, Lansdell L, et al. Dietary intervention study to assess estrogenicity of dietary soy among postmenopausal women. J Clin Endocrinol Met 1995;80:1685-1690. 181. Jensen J, Christiansen C, Rodbro E Cigarette smoking, serum estrogens, and bone loss during hormone-replacement therapy early after menopause. NEnglJMed 1985:313:973-975. 182. MacMahon B, Trichopoulos D, Cole P, Brown J. Cigarette smoking and urinary estrogens. NEnglJ Med 1982;307:1062-1065. 183. Dennerstein L, Dudley EC, Hopper JL, et al. A prospective populationbased study of menopausal symptoms. Obstet Gynecol 2000;96: 351-358. 184. Hunter MS. Predictors of menopausal symptoms: psychosocial aspects. Baillieres Clin Endocrinol Metab 1993;7:33-45. 185. Guthrie JR, Dennerstein L, Taffe JR, et al. Hot flushes during the menopause transition: a longitudinal study in Australian-born women. Menopause 2005;12:160-167. 186. Ford K, Sowers MF, Crutchfield M, et al. A longitudinal study of the predictors of prevalence and severity of symptoms commonly associated with menopause. Menopause 2005;12:308-317.
CHAPTER 6 Epidemiology of Menopause 187. Whiteman MK, Staropoli C, Langenberg P, et al. Smoking, body mass, and hot flashes in mid-fife women. Obstet Gyneco12003;101:264. 188. Staropoli CA, Flaws JA, Bush TL, Mouton AW. Predictors of menopausal hot flashes. J WomensHealth 1998;7:1149-1155. 189. Li C, Samsioe G, Borgfeldt J, et al. Menopause-related symptoms: what are the background factors? A prospective population-based cohort study of Swedish women (the Women's Health in Lund area study). Am J Obstet Gyneco12003;189:1646-1653. 190. Campagnoli C, Morra G, Belforte P, et al. Climacteric symptoms according to body weight in women of different socio-economic groups. Maturitas 1981;3:279-287. 191. Erlik Y, Meldstrum DR, Judd HL. Estrogen levels ofpostmenopausal women with hot flushes. Obstet Gyneco11981;59:403-407. 192. Morton K, Moore M. Predictors of hot flash severity. Ann NYAcad &i 1976;592:457-458. 193. den Tonkelaar I, Seidell JC, van Noord PAH. Obesity and fat distribution in relation to hot flashes in Dutch women from the DOMproject. Maturitas 1996;23:301 - 305. 194. Freeman EW, Sammel MD, Grisos JA, et al. Hot flashes in the late reproductive years: risk factors for Africa American and Caucasian women. J WomensHealth Gend Based Med 2001;10:67- 76. 195. Chiechi LM, Ferreri R, Granieri M, et al. Climacteric syndrome and body-weight. Clin Exp Obstet Gyneco11997;24:163-166. 196. Hjortland M, McNamara P, Kannel WB. Some atherogenic concomitants of menopause: the Framingham study. Am J Epidemiol 1976;103:304-311. 197. Wing R, Matthews K, Kuller LH, et al. Weight gain at the time of menopause. Ann Intern Med 1991;151:97-102. 198. Matthews KA, Abrams B, Crawford S, et al. Body mass index in midlife women: relative influence of menopause, hormone use, and ethnicity. IntJ Obes Relat Metab Disord 2001;25:863-873. 199. Wallace JP, Lovell S, Talano C, et al. Changes in menstrual function, climacteric syndrome, and serum concentrations of sex hormones in pre- and post-menopausal women following a moderate intensity conditioning program. Med Sci Sports Exer 1982;14:154. 200. Hammar M, Berg G, Lindgren R. Does physical exercise influence the frequency of postmenopausal hot flushes? Acta Obstet Gynecol &and 1990;69:409-412. 201. Sternfeld B, Quesenberry Jr CP, Husson G. Habitual physical activity and menopausal symptoms: a case-control study. J Womens Health 1999;8:115-123. 202. Slaven L, Lee C. Mood and symptom reporting among middle-aged women: the relationship between menopausal status, hormone replacement therapy, and exercise participation. Health Psycho11997;16: 203-208. 203. Guthrie JR, Smith AMA, Dennerstein L, Morse C. Physical activity and the menopause experience: a cross-sectional study. Maturitas 1995;20:71-80. 204. Wilbur J, Dan A, Hedricks C, Holm K. The relationship among menopausal status, menopausal symptoms, and physical activity in midlife women. Faro Community Health 1990;13:67-78. 205. Li S, Holm K, Gulanick M, et al. The relationship between physical activity and perimenopause. Health Care Women Int 1999;20: 163-178. 206. Li S, Holm K. Physical activity alone and in combination with hormone replacement therapy on vasomotor symptoms in postmenopausal women. WestJNurs Res 2003;25:274-288. 207. Aiello EJ, Yswi Y, Tworoger SS, et al. Effect of a year-long moderate intensity exercise intervention on the occurrence and severity of menopause symptom in postmenopausal women. Menopause 2004;11: 382-388. 208. Ivarsson T, Spetz A-C, Hammar M. Physical exercise and vasomotor symptoms in postmenopausal women. Maturitas 1998;29:139-146. 209. Tepper R, Neri A, Kaufman H, Schoenfeld A, Ovadia J. Menopausal hot flushes and plasma beta-endorphins. Obstet Gynecol 1987;70: 150-152.
95 210. Harber VJ, Sutton JR. Endorphins and exercise. Sports Med 1984;1: 154-171. 211. Wilbur J, Holm K, Dan A. The relationship of energy expenditure to physical and psychologic symptoms in women at midlife. Nurs Outlook 1992;40:269-276. 212. Dunn AL, Dishman RK. Exercise and the neurobiology of depression. Exer Sports Sci Rev 1991;19:41-98. 213. Armstrong BK, Brown JB, Clarke HT, et al. Diet and reproductive hormones: a study of vegetarian and nonvegetarian post-menopausal women. JNatl CancerInst 1981;67:761-767. 214. Gavaler JS, Rosenblum ER, Deal SR, Bowie BT. The phytoestrogen congeners of alcoholic beverages: current status. Proc Soc Exp BiolMed 1985;208:98-102. 215. Katsouyanni K, Boyle P, Trichopoulos D. Diet and urine estrogens among postmenopausal women. Oncology 1991;48:490-494. 216. Reichman ME, Judd JT, Longcope C, et al. Effects of alcohol consumption on plasma and urinary hormone concentrations in premenopausal women.JNatl CancerInst 1993;85:722-727. 217. Brzezinski A, Adlercreutz H, Shaoul R, et al. Short-term effects of phytoestrogen-rich diet on postmenopausal women. Menopause 1997;4:89-94. 218. Shutt DA, Cox ILl. Steroid and phyto-estrogen binding to sheep uterine receptors in vitro. J Endocrino11972;52:299 - 310. 219. Martin PM, Horwitz KB, Ryan DS, McGuire WL. Phytoestrogen interaction with estrogen receptors in human breast cancer ceils. Endocrinology 1978;103:1860-1867. 220. Cassidy A, Bingham S, Carlson J, Setcheil KDR. Biological effects of plant estrogens in premenopausal women. FASEBJ 1993;A866. 221. Cassidy A, Bingham S, Setchell KDR. Biological effects of a diet of soy protein rich in isoflavones on the menstrual cycle of premenopausal women. Am J Clin Nutr 1994;60:33 - 40. 222. Whitten PL, Lewis C, Russell E, Naftolin E Potential adverse effects of phytoestrogens. J Nutr 1995;125:771S- 776S. 223. Axelson M, Kirk DN, Farrant RD, et al. The identification of the weak estrogen equol in human urine. BiochemJ 1982;210:353-357. 224. Coward L, Barnes NC, Setchell KDR, Barnes S. The isoflavones genistein and diadzein in soybean foods from American and Asian diets. JAgric Food Chem 1993;41:1961-1967. 225. Adlercreutz H, Fotsis T, Bannwart C, et al. Determination of urinary lignans and phytoestrogen metabolites, potential anti-estrogens and anticarcinogens, in urine of women on various habitual diets.J Steroid Biochem 1986;25:791-797. 226. Adlercreutz H, Hockerstedt K, Bannwart C. Effect of dietary components, including lignans and phytoestrogens, on enterohepatic circulation and liver metabolism of estrogens and on sex hormone binding globulin (SHBG). J Steroid Biochem 1987;27:1135-1144. 227. Murkies AL, Lombard C, Strauss BJ, et al. Dietary flour supplementation decreases post-menopausal hot flushes: effect of soy and wheat. Maturitas 1995;21:189-195. 228. Albertazzi P, Pansini F, Bonaccorsi G, et al. The effect of dietary soy supplementation on hot flushes. Obstet Gyneco11998;91:6-11. 229. Harding C, Morton M, Gould V, et al. Dietary soy supplementation is estrogenic in menopausal women. Am J Clin Nutr 1998;68:1532S. 230. Washburn S, Burke GL, Morgan T, Anthony M. Effect of soy protein supplementation on serum lipoproteins, blood pressure and menopausal symptoms in perimenopausal women. Menopause 1999;6: 7-13. 231. Scambia G, Mango D, Signorile PG, et al. Clinical effects of a standardized soy extract in postmenopausal women: a pilot study. Menopause 2000;7:105-111. 232. Upmalis DH, Lobo R, Bradley L, et al. Vasomotor symptom relief by soy isoflavone extract tablets in postmenopausal women: a multicenter, double-blind, randomized placebo controlled study. Menopause 2000;7:236-242.
96 233. Faure ED, Chantre P, Mares E Effects of a standardized soy extract on hot flushes: a multicenter, double-blind, randomized placebocontrolled study. Menopause2002;9:329-334. 234. Q uella SK, Loprinzi CL, Barton DL, et al. Evaluation of soy phytoestrogens for the treatment of hot flashes in breast cancer survivors: a North Central Cancer Treatment Group trial. J Clin Onco12000;18: 1068-1074. 235. St. Germain A, Peterson CT, Robinson JG, Alekel DL. Isoflavonerich or isoflavone-poor soy protein does not reduce menopausal symptoms during 24 weeks of treatment. Menopause2001;8:17-26.
G O L D AND GREENDALE
236. Woods MN, Swine R, Kronenberg G. Effects of a dietary soy bar on menopausal symptoms. Am J Clin Nutr 1999;68(suppl):1533S. 237. Burke GL, Legault C, Anthony M, et al. Soy protein and isoflavone effects on vasomotor symptoms in peri- and post-menopausal women: the Soy Estrogen Alternative Study. Menopause2003;10:147-153. 238. Verhoeven MO, van der Mooren MJ, van de Weijer PHM, et al. Effect of a combination of isoflavones and Actaea racemosa Linnaeus on climacteric symptoms in healthy symptomatic perimenopausal women: a 12-week randomized, placebo-controlled, double-blind study. Menopause2005;12:412-426.
SECTION 11
Ovarian Senescence and Options In the rodent, there is good evidence that the brain (hypothalamus) contributes to the decline in reproductive function. In the human, however, the process is almost entirely that of ovarian failure, with loss of oocytes through atresia until the point of depletion around the time of menopause. This was reviewed in the previous section by Erickson and Chang. The clinical issues that result from ovarian failure in younger women are the focus of this section. Ovarian failure, which occurs in women prior to age 40, is termed premature ovarian failure. Robert W. Rebar discusses this entity in terms of diagnosis, etiology, and treatment. In terms of treatment, providing hormonal support for these young women can really be seen as "replacement therapy," which is a term we no longer use for hormonal therapy after the menopause. Substantial bone loss and increased cardiovascular morbidity has been documented in women with premature ovarian failure who have not received any hormonal "replacement." How these women may be able to conceive is also discussed here and is the subject of Chapter 8 by Mark V. Sauer and Prati Vardhana. Fertility problems due to an "egg factor" are among the most difficult to treat. This often includes chronologically young women, often in their 30s, who have poor ovarian reserve and cannot produce adequate, healthy, and mature oocytes even with maximal gonadotropic stimulation and in vitro fertilization. Many of these women consider egg donation. With the ovarian aging process and accelerated atresia come problems with the egg cytoplasm, including waning mitochondrial function. Although some enthusiasm has been garnered with using donor cytoplasm (cytoplasmic transfer) or nuclei transfer (transfer of the nucleus from an "older" egg into a donor enucleated egg), these techniques are not available and may not be advisable. Also, although in the rodent there is evidence that ovarian stem cells exist that could potentially lead to regeneration of the oocyte pool, this probably does not occur in the human. Thus, egg donation remains the most practical alternative for chronologically young women, many prior to the age of natural menopause, who wish to bear their own biological children. Remarkably, the success of this process (egg donation), as reviewed by Mark Sauer, is the most successful of all our treatments for fertility, with pregnancy rates in the range of 50% to 60% per initiated cycle.
This Page Intentionally Left Blank
tAPTEf
Premature Ovarian Failure ROBERT W. REBAR
American Society for Reproductive Medicine, Birmingham, AL 35216
I. I N T R O D U C T I O N
is true even if 4 or more months of amenorrhea and menopausal symptoms are added to criteria for establishing the diagnosis (3,4). Thus, the term premature ovarian failure is medically inaccurate and misleading to patients and their caregivers as well.
Premature ovarian failure (POF) remains an enigmatic condition. It is even known by several different names. Although most commonly called premature ovarianfailure, it is sometimes referred to as premature menopause, hypergonado-
tropic hypogonadism, hypergonadotropic amenorrhea, primary ovarianfailure, andprimary ovarian insufficiency.We shall use the term premature ovarianfailure throughout this chapter.
II. CLINICAL FEATURES OF PREMATURE OVARIAN FAILURE
P O F generally describes a syndrome consisting of (a) primary or secondary amenorrhea; (b) hypoestrogenism; (c) hypergonadotropinism (with circulating levels of folliclestimulating hormone [FSH] typically greater than 30 mlU/ mL and consistent with ovarian failure); and (d) age under 40 years at the time of onset. Menopause is generally defined as the permanent cessation of menses and normally occurs at about the age of 51 years (1); thus, by definition P O F is distinct from normal menopause. Moreover, at one time it was believed that circulating levels of FSH in the menopausal range provided prima facie evidence for ovarian follicle depletion and thus permanent cessation of ovarian function (2); however, it is now clear that such is not the case (3). Individuals with P O F may ovulate and even conceive spontaneously years after the diagnosis is established (4). More problematic is the fact that definitive criteria for diagnosis have not even been established. However, an operational definition in common use states that affected women should have an interval of at least 4 months of amenorrhea in association with menopausal levels of serum FSH on at least two occasions (4-6). The fallacy of using elevated FSH levels alone to establish a diagnosis of irreversible ovarian failure has already been noted, and the same
To define the clinical spectrum of women with POF, we compiled data from 115 sequential affected women seen between 1978 and 1988 (4). These data have been confirmed by the more than 300 additional women we have seen with this disorder since that report (unpublished). A number of interesting differences and similarities between those with primary and those with secondary amenorrhea were noted in the published series (Table 7.1). In more than 75% of the patients, symptoms of estrogen deficiency, most commonly estrogen deficiency or dyspareunia due to vaginal dryness, were evident, but these symptoms were far more common in women with secondary amenorrhea. Failure to develop mature secondary sex characteristics and chromosomal abnormalities were far more common in those with primary amenorrhea. Chromosomal abnormalities were present in more than half the women with primary amenorrhea, who tended to have deletions of all or part of one X chromosome, whereas those with secondary amenorrhea more commonly had an additional X chromosome. In contrast, chromosomal abnormalities were present in only about 13% of those with secondary amenorrhea who were tested.
TREATMENT OF THE POSTMENOPAUSAL W O M A N
99
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
ROBERT W. REBAR
100 TABLE 7.1
Features of Women with Primary and Secondary Amenorrhea
Description
Primary amenorrheaa
Symptoms of estrogen deficiency Incomplete sexual development Karyotypic abnormalities Immune abnormalities Decreased spinal bone densityb Progestin-induced withdrawal bleeding Pregnancies before diagnosis Evidence of ovulation after diagnosis Pregnancies after diagnosis
20 90 55 20 50 20 0 0 0
Secondary amenorrheaa 85 8 13 20 60 50 35 25 8
aApproximatepercentagesbased on data from Rebar and Connolly(4). UComparedwith age-matched normal controls. Four of the women with secondary amenorrhea and normal karyotypes in our series reported a family history of early menopause prior to the age of 40. Since publication of this series, other investigators have documented the importance of family history. Premutations in the FMR1 gene, now recognized as sometimes associated with a neurodegenerative disorder (7,8), are present in 14% of women with familial POF, as well as in about 2% of women with isolated POF (9). Easily detected immune disturbances were present in approximately 20% of the patients. Thyroid abnormalities were most common, with five women having Hashimoto's thyroiditis, two developing primary hypothyroidism, one developing subacute thyroiditis, and one having Graves' disease. Three asymptomatic patients had high titers of antimicrosomal antibodies. One of the women had vitiligo and hypoparathyroidism; one had Addison's disease; and one additional woman had insulin-dependent diabetes mellitus. Because there is no control population with which to compare the affected women and because autoimmune disturbances are so common in women, it is impossible to conclude with assurance that immune disorders are more common in women with POE Although numerous investigators have suggested the possibility, it would be irresponsible to conclude that immune disturbances and POF are causally linked. The association of POF with adrenal insufficiency (Addison's disease) has been recognized for several years (10,11), and lymphocytic oophoritis has been documented in a handful of patients with this complex. More recently, testing for adrenal antibodies using an indirect immunofluorescence assay (which is commercially available) revealed that 4% of those with normal karyotypes and spontaneous POF had steroidogenic cell autoimmunity (12,13). A relatively small number of the women in our series, all with secondary amenorrhea, had received chemotherapy with alkylating agents, and in some cases radiation therapy as well, before developing hypergonadotropic amenorrhea. The effects of alkylating agents and radiation therapy on ovarian function have been recognized for several years (14,15), and
it is now recognized that the incidence of permanent ovarian failure increases as the age of the woman at the time of therapy increases. The number of women presenting with POF after curative treatment for any of a variety of malignancies is increasing; in this population, too, the POF is not always permanent. Spinal bone density, as evaluated by dual photon absorptiometry, was less than 90% (range 62% to 105%, mean 85.7%) of the mean value observed in age-matched controls in 16 of the 26 women who underwent such testing. The association of POF with osteopenia is now well recognized (16). Progestin-induced withdrawal bleeding occurred in just less than 50% of the women tested in this series. Withdrawal bleeding even occurred in two of the nine individuals with primary amenorrhea who were challenged. Moreover, there was no correlation between the response to exogenous progestin and subsequent ovulation. None of the women with primary amenorrhea ever ovulated or conceived with her own oocytes. In contrast, more than one-third of the women with secondary amenorrhea were pregnant at least once before developing hypergonadotropic amenorrhea, and almost one-quarter had evidence of ovulation after the diagnosis of POF was established. Yet only one-ninth (8.2%) of those with secondary amenorrhea later conceived spontaneously. Twenty-five of the patients with secondary amenorrhea were treated with clomiphene citrate in an effort to induce ovulation, but only four (16%) ovulated as determined by serial ultrasound and serum progesterone levels. Because each of the four who ovulated had evidence of spontaneous episodic ovulation before therapy, it is unclear if the clomiphene actually induced ovulation or if ovulation occurred in association with clomiphene by chance alone. Nineteen women had gonadotropin secretion suppressed either with exogenous estrogen and progestin (n = 14) or with gonadotropinreleasing hormone agonist (n = 5) for 1 to 3 months. Ovulation induction was then attempted with exogenous gonadotropins. Only two of the patients (among those suppressed with agonist) ovulated, and only one conceived. These data
CHAPTER 7 Premature Ovarian Failure are consistent with a later controlled trial documenting that ovulation induction is unlikely to be successful (5). Twelve women with secondary amenorrhea underwent ovarian biopsies, with apparent viable oocytes noted in seven of the specimens. Yet two of the eight subsequent pregnancies occurred in women with no follicles found in the biopsy specimens. Fully seven of these eight pregnancies occurred while the patients were taking exogenous estrogen; the remaining pregnancy in this series occurred following administration of clomiphene. Five of the eight pregnancies ended in normal-term live births, two ended in spontaneous abortion, and one ended in elective abortion. Only three patients with primary amenorrhea underwent biopsy of gonadal tissue: The two with 46,XY karyotypes had dysgerminoma. The one additional patient had fibrous streaks. These observations confirm that POF is a heterogeneous disorder. Moreover, they stress the importance of measuring circulating levels of FSH in all women who present with amenorrhea. Progestin-induced withdrawal bleeding is now an outmoded and inappropriate tool in the evaluation of women with amenorrhea.
III. PREVALENCE OF PREMATURE OVARIAN FAILURE Estimation of the prevalence of POF in the general population is difficult. In one study, 7% of 300 consecutive women presenting with amenorrhea had POF (17). An estimate based on several studies concluded that 0.3% of reproductiveage women (approximately 130,000) have POF (18). Still another study suggested that 5% to 10% of women with secondary amenorrhea have POF (19). Based on 1950 data, the risk of experiencing menopause prior to the age of 40 was calculated as 0.9% in Rochester, Minnesota (20). Perhaps the major point of these estimates is that POF is sufficiently common that most clinicians will see individuals who present with this disorder.
IV. ETIOLOGY OF PREMATURE OVARIAN FAILURE De Moraes-Ruehsen and Jones (17) suggested three possible explanations for the early completion of atresia that they believed existed in women with hypergonadotropic amenorrhea and premature ovarian failure: (a) decreased germ cell endowment, (b) accelerated loss of oocytes (atresia), and (c) postnatal germ cell destruction. Because none of these possibilities can be true in individuals in whom many follicles still remain, some block to gonadotropin action in ovarian follicles must exist in such affected women. Given older data that even postmenopausal women may have a few remaining ovarian follicles (21,22) and evidence
101 that follicle number decreases rapidly in the last several months before the menopause (23,24), occasional ovulations and rare pregnancies may occur in women with this disorder who are just experiencing an "early" but "normal" menopausal transition (1). Among the various causes of POF that have now been identified, it is clear that some are present only in those who have no oocytes, whereas others may include the potential for ovulation and spontaneous pregnancy. Given current knowledge, it is impossible to develop a classification scheme for POF that does not include some overlap, but a suggested classification is presented in Table 7.2.
A. Cytogenetic Abnormalities Involving the X Chromosome 1. STRUCTURALALTERATIONS OR ABSENCE OF AN X CHROMOSOME Individuals with the various forms of gonadal dysgenesis, with or without the stigmata of Turner syndrome, typically present with hypergonadotropic amenorrhea, regardless of the extent of pubertal development and the presence or absence of associated anomalies or stigmata. It is well recognized that cytogenetic abnormalities of the X chromosome can impair TABLE 7.2
Tentative Classification of Premature Ovarian Failure
I. Cytogenic abnormalities involving the X chromosome A. Structural alterations or absence of an X chromosome B. Fragile X premutations C. Trisomy X with or without mosaicism II. Enzymatic defects A. Steroidogenic enzyme defects 1. 17ot-hydroxylasedeficiency 2. 17,20-desmolase deficiency 3. 20,22-desmolase deficiency 4. Aromatase deficiency B. Galactosemia III. Other genetic alterations IV. Defective gonadotropin secretion or action A. Receptor or postreceptor defects 1. FSH receptor mutations 2. LH receptor mutations B. Secretion of biologically inactive gonadotropin V. Environmental insults A. Chemotherapeutic agents B. Ionizing radiation C. Viral infection D. Surgical injury or extirpation VI. Immune disturbances A. In association with other autoimmune disturbances (15-20% of cases) B. Isolated C. In association with congenital thymic aplasia VII. Idiopathic LH, luteinizinghormone.
102
ROBERT W. REBAR
ovarian development and function. Studies of 46,,3( individuals and those with various X chromosomal deletions have confirmed that two intact X chromosomes are necessary for maintenance of oocytes (25). The gonads of 45,X fetuses contain the normal complement of oocytes at 20 to 24 weeks of fetal age, but these rapidly undergo atresia so that essentially none are present at birth (26). Primary or secondary amenorrhea typically occurs in women with deletions in either the short or the long arm of one X chromosome (25). Structural abnormalities of the X chromosome also can have a negative impact on ovarian function and are present in some women with POF (25,27). The association of submicroscopic deletions of Xq26-27 indicates that even subtle molecular defects in the X chromosome can impact on ovarian function and be associated with POF (28). Although women with stigmata of Turner syndrome are evident on physical examination, individuals with many forms of gonadal dysgenesis may not have any such stigmata. Women with pure gonadal dysgenesis, who generally present with primary amenorrhea and sexual infantilism, are of normal height and do not have the somatic abnormalities associated with Turner syndrome (25,29). Such individuals have either a 46,XX or 46,XY karyotype. In the extremely rare disorder of mixed gonadal dysgenesis, a germ cell tumor or a testis accounts for one gonad, with an undifferentiated streak, rudimentary, or no gonad accounting for the other (30). Such individuals are generally mosaic, with the 45,X/46,XY karyotype reported most commonly. Almost all affected individuals are raised as females, with mild to moderate masculinization occurring at puberty. Abnormal genitalia may be noted at puberty. Because of the malignant potential of intraabdominal gonads with a Y chromosomal component (31-33), the gonads should be removed. 2. TRISOMY X WITH OR WITHOUT MOSAICISM
An excess of X chromosomes also may be found in some women who develop POF (34). Affected individuals typicaUy develop normal secondary sex characteristics and only later develop POE Reports of the triple-X syndrome associated with immunoglobulin deficiency (35) and Marfan syndrome (36), together with the observation that control of T-cell function may be related to the X chromosome (37), suggest a possible association between immunologic abnormalities and triple-X females with POE
B. Enzymatic Defects 1. 170s
DEFICIENCY
The rare women with deficiency of the 17cx-hydroxylase enzyme are identified easily because of the associated findings of primary amenorrhea, sexual infantilism, hypergonadotropinism, hypertension, hypokalemic alkalosis, and
increased circulating concentrations of deoxycorticosterone and progesterone (38-41). Ovarian biopsies have revealed numerous large follicular cysts with complete failure of orderly follicular maturation (40). 2. GALACTOSEMIA
Women with galactosemia may develop amenorrhea with elevated gonadotropin levels even when treatment with a galactose-restricted diet begins at an early age (42-44). The etiology of the ovarian failure in this disorder is unresolved, but pregnant rats fed a 50% galactose diet deliver pups with significantly reduced numbers of oocytes, apparently because of decreased germ cell migration to the genital ridges (45). There is also evidence that excess galactose inhibits follicular development in the rat ovary (46). 3. AROMATASE DEFICIENCY
Several case reports of individuals with documented mutations in the CYP19 (aromatase P450) gene have been described in detail (47-52). Estrogen biosynthesis was virtually absent in all these patients and associated with a number of anticipated and unanticipated findings. It is clear that aromatase deficiency is an autosomal recessive condition manifested in 46,XX individuals by female pseudohermaphroditism with clitoromegaly and posterior labioscrotal fusion at birth; enlarged cystic ovaries in association with elevated FSH levels during childhood; lack of pubertal development in association with further enlargement of the clitoris, normal development of pubic and axillary hair, and continued existence of enlarged multicystic ovaries during the teenage years; and severe estrogen deficiency, virilization, and enlarged multicystic ovaries in association with markedly elevated levels of gonadotropins in adulthood. Administration of exogenous estrogen results in prompt lowering of circulating gonadotropin levels. Consistent with the diagnosis of POF is the observation that many closely packed primordial follicles were present in an ovarian biopsy specimen obtained from an affected 17 month old (51), but a biopsy specimen from a 13 year old showed excessive atresia (52). Affected 46,XY individuals are normal at birth and during childhood but develop eunuchoid proportions and osteoporosis as they continue to grow during adulthood. It has become clear that estrogen is essential for epiphyseal closure. Because of the absence of aromatase enzyme in the placenta, mothers of affected children develop reversible virilization during the second half of pregnancy.
C. Other Genetic Alterations It is becoming clear that mutations to any of several different genes can result in POE As noted previously, perhaps the most important mutations identified thus far involve the
CHAPTER 7 Premature Ovarian Failure FMR1 gene; mutations result in the fragile X syndrome
(characterized by impaired intellectual functioning, certain mild physical changes, social anxiety, language difficulties, and problems with balance), and premutations can result in POF and a neurodegenerative disorder (7,8). Other rare genetic causes of familial POF for which routine genetic testing in sporadic cases is not now clinically warranted include mutations involving FSHR (the FSH receptor), FOXL2 (a forkhead transcription factor associated with the blepharophimosis/ptosis/epicanthus inversus syndrome), I N H A (inhibin alpha gene), EIF2B (a family of genes associated with central nervous system leukodystrophy and ovarian failure), BMP15 (bone morphogenetic protein 15), PMM2 (phosphomannomutase 2), and AIRE (autoimmune polyendocrinopathy candidiasis ectodermal dystrophy syndrome) (53-61). It is likely that other genetic mutations that lead to POF will be identified in the future.
D. Defective Gonadotropin Secretion or Action Data from a variety of sources now indicate that abnormal structure, secretion, metabolism, or action of gonadotropins forms the basis for early ovarian failure in some women. We have reported altered forms of immunoreactive luteinizing hormone (LH) and FSH in urinary extracts from women with POF compared to those from oophorectomized and postmenopausal women (62), suggesting that metabolism or excretion of gonadotropins is altered in some cases. As already noted, genetic mutations in receptor structure manifest in altered FSH action and POF (53,54). Also reported are individuals with POF and evidence of intermittent follicular activity who appear to have low-molecular-weight FSH receptor-binding activity that antagonizes normal FSH binding (63).
E. Environmental Insults Destruction of oocytes by any of several environmental insults, including ionizing radiation, various chemotherapeutic agents, certain viral infections, and even cigarette smoking, may accelerate follicular atresia (64). 1. RADIATION Approximately 50% of individuals who receive 400 to 500 rads to the ovaries over 4 to 6 weeks, as commonly occurs in treatment for Hodgkin's disease, will develop permanent ovarian failure (15,65). For any given dose of radiation, the older the woman, the greater the likelihood of her developing hypergonadotropic amenorrhea. It appears that 800 rads is sufficient to result in permanent sterility in all women (65).
103 That the amenorrhea following radiation therapy is not always permanent was reported as long ago as 1939 (66). The transient nature of the hypergonadotropic amenorrhea in some women suggests that some follicles may be damaged but not destroyed by relatively low doses of radiation. This information has led to the practice of transposing the ovaries to the pelvic sidewalls, today often by laparoscopy, to minimize the dose of radiation to which they are exposed; one review has concluded that transposition in women under age 40 results in preservation of ovarian function in 88.6% of cases (67,68). 2. CHEMOTHERAPEUTICAGENTS
It is clear that chemotherapeutic agents, particularly alkylating agents, may produce either temporary or permanent ovarian failure (14,15,69-72). In general, the younger the woman at the time of therapy, the more likely it is that ovarian function will not be compromised by chemotherapy. It appears that the greater the number of oocytes present in the ovaries at the time of therapy, the more likely it is that normal ovarian function will persist. The frequency of congenital anomalies does not appear to be increased in the children of women previously treated with chemotherapeutic agents (73). There is the suggestion, however, that one agent, dactinomycin, may be associated with an increased risk of congenital heart disease, and further studies in this area are clearly needed. 3. VIRALAND OTHER AGENTS
Although several viruses are believed to have the potential to cause ovarian failure, confirming that this is the case in women is difficult. The best documented series includes three presumptive cases of"mump oophoritis" that preceded ovarian failure, including cases in a mother and her daughter in which the mother had documented mumps parotiditis and abdominal pain during pregnancy just prior to the delivery of a daughter who later suffered from hypergonadotropic amenorrhea (74). Although there is no evidence that cigarette smoking will lead to premature menopause, data do exist documenting that cigarette smokers experience menopause several months before nonsmokers (75).
F. I m m u n e D i s t u r b a n c e s Several autoimmune abnormalities are known to be associated with hypergonadotropic amenorrhea (Table 7.3). As is characteristic for other autoimmune disturbances, the ovarian "failure" may wax and wane, and pregnancies may occur, at least early in the disease process. We reviewed 380 cases of premature ovarian failure in the literature and noted that 17.5% had an autoimmune disorder present in addition to the ovarian failure. Unfortunately,
104 TABLE 7.3
ROBERT W. REBAR
Possible Autoimmune Disorders Associated with Premature Ovarian Failurea Percentage of women with POF affected
Thyroid Polyendocrinopathy type I Polyendocrinopathy type II Unspecified Other (misceUaneous)b Myasthenia gravis Adrenal Multiple endocrinopathy (unclassified) Diabetes mellitus Pernicious anemia Systemic lupus erythematosus
6.8 5.3 5.0 3.9 2.9 2.4 2.1 1.6 0.8 0.5 0.5
aBased on data from reic. 89:119 out of 380 patients (31.3%) surveyed had autoimmune disease. bIncluding one case each with asthma, Crohn's disease, glomerulonephritis, idiopathic thrombocytopenia, purpura, and vitiligo.
autoimmune dysfunction is also common in women without ovarian failure, and it is not clear that the incidence is actually increased in women with POE However, additional evidence that some cases of POF may have an autoimmune etiology is provided by sporadic case reports documenting return of ovarian function following either immunosuppressire therapy or recovery from an associated autoimmune disease (76-79). In a few cases, lymphocytic infiltrates suggesting autoimmune dysfunction have been observed in ovarian biopsy specimens (79). In fact, autoimmune lymphocytic oophoritis was originally reported in association with adrenal insufficiency (Addison's disease) (10,11). It is now clear that the women with POF who have steroidogenic cell autoimmunity have lymphocytic oophoritis as the mechanism for the ovarian failure. One review reported that all patients with histologically confirmed lymphocytic autoimmune oophoritis had adrenal antibodies when tested using an indirect immunofluorescence assay (10). There is now evidence that antibodies to the 21-hydroxylase enzyme measured by a commercially available immunoprecipitation assay are generally in good agreement with results testing for adrenal cortex antibodies by indirect immunofluorescence, although a variety of antibodies to steroidogenic enzymes can be detected in patients with steroidogenic cell autoimmunity (80). Testing for adrenal antibodies by indirect immunofluorescence will identify the 4% of women with spontaneous POF who have steroidogenic cell autoimmunity and are at risk for adrenal insufficiency, a potentially fatal disorder (12,13). When POF occurs in association with adrenal insufficiency, the ovarian failure presents first about 90% of the time (81). It appears that a few women presenting with POF
will have asymptomatic adrenal insufficiency; these individuals are at risk of developing adrenal crisis. Adrenal antibodies will identify women who may have occult adrenal insufficiency at the time of initial presentation as well as those who should be followed closely for the subsequent development of adrenal insufficiency (82,83). Still other immune abnormalities have been identified in some patients with POE Enhanced release of leukocyte migration inhibition factor (MIF) by peripheral lymphocytes has been observed following exposure of the lymphocytes to crude ovarian proteins (84,85). A significant association of early ovarian failure with HLA-DR3 has been noted (86), perhaps suggesting a genetic susceptibility in some individuals. Several years ago, complement-dependent cytotoxic effects, as documented by inhibition of progesterone production and cell lysis, were observed when sera from 9 of 23 patients were added to cultured granulosa cells (87). Indirect immunofluorescence of ovarian biopsy specimens from some patients has revealed antibodies reacting with various ovarian components (88). Circulating immunoglobulins to ovarian proteins have been detected by immunochemical techniques by several investigators (89). Utilizing a solid-phase, enzyme-linked immunosorbent assay, we have detected antibodies to ovarian tissue in 22% ofkaryotypically normal women with POF (90). The most welldocumented evidence of autoantibodies to ovarian tissue comes from a study of two patients with POF and myasthenia gravis who had circulating immunoglobulin G that blocked binding of FSH to ovarian cell surface receptors (91). However, presently there is no test to detect ovarian specific antibodies that has proven clinical utility (90,92,93). It has become clear that there are links between the immune and reproductive systems. It has been known for several years that congenitally athymic girls dying before puberty have ovaries devoid of oocytes on autopsy (94). We have shown that congenitally athymic mice, well known to have premature ovarian failure, have lower gonadotropin concentrations prepubertally than do their normal heterozygous littermates (95). These hormonal alterations, as well as the accelerated loss of oocytes, can be prevented by thymic transplantation at birth (96). It is essential to recognize that ovarian development occurring during the first few weeks of life in the mouse occurs in utero in women and in nonhuman primates. Thus, thymic ablation in fetal rhesus monkeys in late gestation is associated with a marked reduction in oocyte number at birth (97). One possible explanation for the association of thymic aplasia and ovarian failure may be found in our observation that peptides produced by the thymus gland can stimulate release of gonadotropin-releasing hormone (GnRH) and consequently LH (98). That gonadotropins are required for normal ovarian development is supported by the observation that fetal hypophysectomy in rhesus monkeys leads to newborns having no oocytes in their ovaries (99).
CHAPTER 7 Premature Ovarian Failure
From a theoretical point of view, identifying women with an autoimmune etiology for their POF is important because affected patients might be treated effectively early in the disease process, before all viable oocytes are destroyed.
G. Idiopathic Premature Ovarian Failure Although the diagnosis of "idiopathic" causes of POF should be a diagnosis of exclusion, presently no definitive etiology is identified in most patients with POE It is likely that additional causes of POF will be recognized as investigation of the human genome continues.
H. Resistant Ovary Syndrome"
An Outdated Term As originally defined, the resistant ovary or "Savage" syndrome was found in young amenorrheic women with (a) elevated peripheral gonadotropin levels, (b) normal but immature follicles in the ovaries, (c) a 46,XX karyotype, (d) mature secondary sex characteristics, and (e) decreased sensitivity to stimulation with exogenous gonadotropin (100). It is clear from this consideration of the causes of POF, however, that these criteria might easily be the result of several different etiologies. Moreover, regardless of the etiology, these features may be common to the vast majority of individuals who present with hypergonadotropic amenorrhea at some time during the disease process prior to the final loss of all oocytes. Thus, the term resistant ovary is outdated and should no longer be used.
V. EVALUATION OF INDIVIDUALS WITH HYPERGONADOTROPIC AMENORRHEA Young women with hypergonadotropic amenorrhea should be evaluated to identify (a) specific, potentially treatable causes and (b) other potentially dangerous associated disorders. It is important to make the diagnosis in a timely manner. One report found that more than one-half of patients who presented with secondary amenorrhea saw three or more clinicians before laboratory testing established the diagnosis (101). In general, young women who experience loss of regular menses for 3 or more consecutive months merit appropriate evaluation. A thorough history and physical examination are warranted. Clinical evaluation of the vaginal mucosa and cervical mucus may help determine if any endogenous estrogen is present. In general, laboratory evaluation should include measurement of basal levels ofprolactin, FSH, and thyroid-stimulating hormone (TSH) after pregnancy is ruled out.
105 FSH levels are generally greater than 30 mlU/mL in women who do not have functioning gonads. If the FSH level is greater than 20 mIU/mL in the initial measurement and the patient is less than 40 years old, then the measurement of FSH should be repeated and serum estradiol should be measured as well to confirm hypogonadism. In addition, the measurement of basal LH levels may help determine if any functional follicles are present. In general, if the estradiol concentration is greater than 50 pg/mL or if the LH concentration (in terms of milli-International Units per milliliter) is greater than the FSH concentration on any occasion, then at least a few viable oocytes still must be present. Irregular uterine bleeding, indicative of continuing estrogen production, also suggests the presence of remaining functional oocytes. The presence of identifiable follicles on transvaginal ultrasonography also can be used to identify women with remaining oocytes. Because about one-half the women with POF will experience withdrawal bleeding in response to a progestin challenge (4), this test should not be used as a substitute for measuring basal FSH levels. When ovarian failure presents as primary amenorrhea, approximately 50% of individuals will have an abnormal karyotype (4). However, most cases of spontaneous POF present as secondary amenorrhea. In our series, we found an abnormal karyotype in only 13% of women 30 years of age or younger who developed secondary amenorrhea. Still it would seem prudent to obtain a karyotype in women with onset of hypergonadotropic amenorrhea prior to age 30 to identify those with various forms of gonadal dysgenesis, individuals with mosaicism, those with trisomy X, and those with a portion of a Y chromosome. If a Y chromosome is present, gonadal extirpation is warranted because of the increased risk of malignancy (31-33). Chromosomal evaluation also may be warranted to rule out familial transmission in women who develop hypergonadotropic amenorrhea after the birth of daughters. Approximately 6% of women with 46,XX spontaneous POF have premutations of the FMR1 gene, the gene responsible for the fragile X syndrome, the most common cause of familial mental retardation, with the risk being greater if there is a family history of POF (9). Thus, a good argument can be made for screening for this abnormality. In addition, testing for adrenal antibodies by indirect immunofluorescence is warranted. If antibodies are detected, a corticotropin stimulation test is warranted to identify women with adrenal insufficiency. Because of the frequency of autoimmune thyroid disease in women with POF, it is also reasonable to measure TSH and thyroid-stimulating immunoglobulins in women with POE Ovarian biopsy is not justified in women with hypergonadotropic amenorrhea and a normal karyotype. It is not clear how the results would alter therapy. In one series, one of the two patients who eventually conceived had no oocytes present on biopsy (18). Similarly, we reported that two of
106 eight subsequent pregnancies among 97 women with secondary hypergonadotropic amenorrhea occurred in women with no follicles present in ovarian tissue obtained at laparotomy (4). As noted (19), if five sections of an ovarian biopsy are examined and each is 6 Ixm thick, then the presence of follicles is sought from a sample representing less than 0.15% of a 2 • 3 x 4 cm ovary. Thus, the absence of follicles on biopsy may not be representative of the remainder of the ovary. Evaluation of bone density appears warranted in women with POF because of the high incidence of osteopenia (4). Periodic assessment may be warranted, regardless of therapy, to assess the rate of bone loss. Similarly, monitoring patients for the development of autoimmune endocrinopathies may be warranted even if all the tests are normal when the patient is first evaluated. Development of other disorders after diagnosis of POF does occur (4), and there is little knowledge about the natural history of the development of associated autoimmune disorders.
VI. TREATMENT Perhaps the first challenge in providing appropriate treatment to women with POF is informing the patient of the diagnosis in a sensitive manner. The manner in which the diagnosis is delivered can affect the degree of emotional trauma experienced, especially if the woman already sought assistance because of infertility or anticipates having children in the future. It may be best to schedule a separate visit to review the findings and discuss treatment options when the diagnosis is suspected. It is important to explain that remissions and spontaneous pregnancies can occur, and that POF differs from the normal menopause in important ways. Because the diagnosis can be emotionally devastating, it may be important to provide ongoing psychologic support. Referral to an organization such as the POF Support Group (www.pofsupport.org) should also be considered. It is reasonable to treat all young women with POF with exogenous estrogen whether they are interested in childbearing or not. The accelerated bone loss often accompanying this disorder may well be prevented by the administration of exogenous estrogens (16). In addition, spontaneous pregnancies can occur in patients with this disorder when they are taking exogenous estrogen--even in the form of combined oral contraceptive agents (4,102). However, the possibility of spontaneous pregnancy after the diagnosis is established appears to be less than 10%. The pregnancy rate is low despite the fact that about one-fourth of women ovulate after the diagnosis of hypergonadotropic amenorrhea is made (4). Because of the possibility of pregnancy, women who do not desire pregnancy and are sexually active should be advised to use barrier contraception, even if they are taking oral contraceptives. Women should be advised to contact
ROBERT W. REBAR
their physician if they develop any signs or symptoms consistent with pregnancy or if they do not have withdrawal bleeding while taking exogenous estrogen and progestin. Why it is that these women can ovulate and conceive while taking oral contraceptives has not been determined. Although exogenous estrogen may be provided in the form of either estrogen-progestin therapy or oral contraceptives, therapy with exogenous estrogen and progestin is more physiologic. It is important to remember that these young women may require twice as much estrogen as do postmenopausal women to alleviate signs and symptoms of hypoestrogenism. Findings documenting significant risks of exogenous estrogen when administered to postmenopausal women do not apply to these patients, for whom estrogen therapy is truly replacement. At present there just are no data assessing the risks in young women. There is controversy as to how best to provide progestin. Although many clinicians now provide progestin continuously along with the estrogen, this is less physiologic than utilizing the progestin sequentially. Some clinicians administer the progestin for 12 to 14 days each month, whereas others administer it less frequently. Because many patients prefer having fewer menses, I typically administer progestin every other month and commonly use micronized progesterone. I prefer administering estradiol-17[3 by patch in an effort to provide therapy that is as physiologic as possible. There is no evidence, however, that any one form of estrogen replacement therapy is safer or more efficacious than any other. Several isolated case reports suggested that ovarian suppression with estrogen or a GnRH agonist followed by stimulation with human menopausal gonadotropin might be efficacious in inducing ovulation and allowing pregnancy (103-106). Most of these reports originated from patients treated by just one group of physicians. Larger series suggest that the possibility of successful ovulation induction and pregnancy is small indeed and appears no greater than what occurs spontaneously in these patients (4,5,107). The majority of the data indicate that there is little point in attempting to induce ovulation in women with POE In vitro fertilization involving oocyte donation clearly provides individuals with hypergonadotropic amenorrhea with the greatest likelihood of bearing children. The first successful case of oocyte donation in humans was reported in 1984. A young woman with ovarian failure was given oral estradiol valerate and progesterone pessaries to prepare the endometrium for transfer of a single donated oocyte following fertilization with her husband's sperm (108). Following several series documenting success with oocyte donation (109-111), use of oocyte donation has become widespread, and success rates are generally greater than those observed with traditional in vitro fertilization. Thus, oocyte donation offers the possibility of pregnancy to all women with POF as long as a normal uterus is present. One cautionary note, however, is warranted: Recent data indicate that the risk of
CI-IAeTER 7 Premature Ovarian Failure
aortic rupture is increased during pregnancy in women with Turner syndrome (112,113). An echocardiogram to exclude dilation of the aortic root is indicated in women with Turner syndrome who are contemplating pregnancy. However, even if the findings are normal, the risk of aortic rupture may be increased because the structure of the aortic wall is abnormal. Thus, such women must be counseled carefully before oocyte donation is contemplated. If the patient does become pregnant, careful cardiac monitoring is warranted throughout the course of the pregnancy. A strong case exists for recommending adoption for such women, as well as for those unwilling to undergo the laborious procedures involved in oocyte donation.
References 1. Soules MR, Sherman S, Parrott E, et al. Executive summary: stages of Reproductive Aging Workshop (STRAW). Fertil Steril 2001;76: 874-878. 2. Goldenberg RL, Grodin JM, Rodbard D, Ross GT. Gonadotropins in women with amenorrhea. The use of plasma follicle-stimulating hormone to differentiate women with and without ovarian follicles. Am J Obstet Gyneco11973;116:1003-1012. 3. Rebar RW, Erickson GF, Yen SS. Idiopathic premature ovarian failure: clinical and endocrine characteristics. Fertil Steri11982;37:35-41. 4. Rebar RW, Connolly HV. Clinical features of young women with hypergonadotropic amenorrhea. Fertil Steri11990;53:804-810. 5. Nelson LM, Kimzey LM, White BJ, Merriam GR. Gonadotropin suppression for the treatment of karyotypically normal spontaneous premature ovarian failure: a controlled trial. Fertil Steri11992;57:50-55. 6. Anasti JN. Premature ovarian failure: an update. Fertil Steri11998;70: 1-15. 7. Hagerman RJ, Hagerman PJ. The fragile X premutation: into the phenotypic fold. Curr @in Genet Dev 2002;12:278-283. 8. Hagerman RJ, Leavitt BR, Farzin F, et al. Fragile-X-associated tremor/ ataxia syndrome (FXTAS) in females with the FMR1 premutation. Am JHum Genet 2004;74:1051-1056. 9. Sherman SL. Premature ovarian failure in the fragile X syndrome. Am J Med Genet 2000;97:189-194. 10. Hoek A, Schoemaker J, Drexhage HA. Premature ovarian failure and ovarian autoimmunity. Endocr Rev 1997;18:107-134. 11. Irvine WJ, Chan MM, Scarth L, et al. Immunological aspects of premature ovarian failure associated with idiopathic Addison's disease. Lancet 1968;2:883 - 887. 12. Bakalov VK, Vanderhoof VII, Bondy CA, Nelson LM. Adrenal antibodies detect asymptomatic auto-immune adrenal insufficiency in young women with spontaneous premature ovarian failure. Hum Re])rod 2002; 17:2096 - 2100. 13. Falorni A, Laureti S, Candeloro P, et al. Steroid-cell autoantibodies are preferentially expressed in women with premature ovarian failure who have adrenal autoimmunity. Fertil Steri12002;78:270-279. 14. Siris ES, Leventhal BG, Vaitukaitis JL. Effects of childhood leukemia and chemotherapy on puberty and reproductive function in girls. NEnglJMed 1976;294:1143-1146. 15. Damewood MD, Grochow LB. Prospects for fertility after chemotherapy or radiation for neoplastic disease. Fertil Steri11986;45(4):443-459. 16. Metka M, Holzer G, Heytmanek G, Huber J. Hypergonadotropic hypogonadic amenorrhea (World Health Organization III) and osteoporosis. Fertil Steri11992;57:37-41.
107 17. De Moraes-Ruehsen M, Jones GS. Premature ovarian failure. Fertil Steri11967;18:440-461. 18. Aiman J, Smentek C. Premature ovarian failure. Obstet Gyneco11985;66: 9-14. 19. Alper MM, Garner PR, Seibel MM. Premature ovarian failure. Current concepts. J Reprod Med 1986;31:699-708. 20. Coulam CB, Adamson SC, Annegers JE Incidence of premature ovarian failure. Obstet Gyneco11986;67:604-606. 21. CostoffA, Mahesh VB. Primordial follicles with normal oocytes in the ovaries of postmenopausal women. J Am Geriatr Soc 1975;23: 193-196. 22. Hertig AT. The aging ovary-preliminary note. J Clin EndocrinolMetab 1944;4:581. 23. Richardson SJ, Nelson JE Follicular depletion during the menopausal transition. Ann N YAcad Sci 1990;592:13 - 20; discussion 44- 51. 24. Richardson SJ, Senikas V, Nelson JE Follicular depletion during the menopausal transition: evidence for accelerated loss and ultimate exhaustion. J Clin Endocrinol Metab 1987;65:1231-1237. 25. Simpson JL, Rajkovic A. Ovarian differentiation and gonadal failure. AmJMed Genet 1999;89:186-200. 26. Singh RP, Cart DH. The anatomy and histology of XO human embryos and fetuses. Anat Rec 1966;155:369-383. 27. Rebar RW, Erickson GF, Coulam CB. Premature ovarian failure. In: Gondos B, Riddick D, eds. Pathology of infertility. New York: Thieme Medical Publishers, 1987. 28. Krauss CM, Turksoy RN, Atkins L, et al. Familial premature ovarian failure due to an interstitial deletion of the long arm of the X chromosome. N EnglJ Med 1987;317:125 - 131. 29. Espiner EA, Veale AM, Sands VE, Fitzgerald PH. Familial syndrome of streak gonads and normal male karyotype in five phenotypic females. N EnglJ Med 1970;283:6-11. 30. Davidoff F, Federman DD. Mixed gonadal dysgenesis. Pediatrics 1973;52:725-742. 31. Manuel M, Katayama PK, Jones HW Jr. The age of occurrence of gonadal tumors in intersex patients with a Y chromosome. A m J Obstet Gyneco11976;124:293- 300. 32. Schellhas HE Malignant potential of the dysgenetic gonad. I. Obstet Gynecol 1974;44:289- 309. 33. Schellhas HE Malignant potential of the dysgenetic gonad. II. Obstet Gyneco11974;44:455-462. 34. Villanueva AL, Rebar RW. Triple-X syndrome and premature ovarian failure. Obstet Gyneco11983;62(3 Suppl):70S-73S. 35. SillsJA, Brown JK, Grace E, et al. XXX syndrome associated with immunoglobulin deficiency and epilepsy.J Pediatr 1978;93:469-471. 36. Smith TF, Engel E. Marfan's syndrome with 47,XXX genotype and possible immunologic abnormality. South MedJ 1981;74:630-632. 37. Purtilo DT, DeFlorio D Jr, Hutt LM, et al. Variable phenotypic expression of an X-linked recessive lymphoproliferative syndrome. NEnglJMed 1977;297:1077-1080. 38. Biglieri EG, Herron MA, Brust N. 17-hydroxylation deficiency in man.J Clin Invest 1966;45:1946-1954. 39. Goldsmith O, Solomon DH, Horton R. Hypogonadism and mineralocorticoid excess. The 17-hydroxylase deficiency syndrome. N EnglJ Med 1967;277:673-677. 40. Mallin SR. Congenital adrenal hyperplasia secondary to 17-hydroxylase deficiency. Two sisters with amenorrhea, hypokalemia, hypertension, and cystic ovaries.Ann Intern Med 1969;70:69-75. 41. Miller WL. Steroid 17alpha-hydroxylase deficiency--not rare everywhere. J Clin EndocrinolMetab 2004;89:40-42. 42. Hoefnagel D, Wurster-Hill D, Child EL. Ovarian failure in galactosaemia. Lancet 1979;2:1197. 43. Kaufman FR, Kogut MD, Donnell GN, et al. Hypergonadotropic hypogonadism in female patients with galactosemia. N Engl J Med 1981;304:994-998.
108 44. Guerrero NV, Singh RH, Manatunga A, et al. Risk factors for premature ovarian failure in females with galactosemia. J Pediatr 2000;137: 833-841. 45. Chen YT, Mattison DR, Feigenbaum L, Fukui H, Schulman JD. Reduction in oocyte number following prenatal exposure to a diet high in galactose. Science 1981;214:1145-1147. 46. Liu G, Shi F, Blas-Machado U, et al. Dietary galactose inhibits GDF-9 mediated follicular development in the rat ovary. Reprod Toxicol 2006;21:26-33. 47. Shozu M, Akasofu K, Harada T, Kubota Y. A new cause of female pseudohermaphroditism: placental aromatase deficiency. J Clin Endocrinol Metab 1991;72:560-566. 48. Portrat-Doyen S, Forest MG, Nicolino M, Morel Y, Chatelain PC. Female pseudohermaphroditism (FHP) resulting from aromastase (P450arom) deficiency associated with a novel mutation (R457) in the CYP19 gene. Horm Res 1996;46(suppl):14-20. 49. Ludwig M, Beck A, Wickert L, et al. Female pseudohermaphroditism associated with a novel homozygous G-to-A (V370-to-M) substitution in the P-450 aromatase gene. J Pediatr Endocrinol Metab 1998;11: 657-664. 50. Mullis PE, Yoshimura N, Kuhlmann B, et al. Aromatase deficiency in a female who is compound heterozygote for two new point mutations in the P450arom gene: impact of estrogens on hypergonadotropic hypogonadism, multicystic ovaries, and bone densitometry in childhood. J Clin Endocrinol Metab 1997;82:1739-1745. 51. Conte FA, Grumbach MM, Ito Y, Fisher CR, Simpson ER. A syndrome of female pseudohermaphrodism, hypergonadotropic hypogonadism, and multicystic ovaries associated with missense mutations in the gene encoding aromatase (P450arom). J Clin Endocrinol Metab 1994;78:1287-1292. 52. Morishima A, Grumbach MM, Simpson ER, Fisher C, Qin K. Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. J Clin Endocrinol Metab 1995;80:3689-3698. 53. Aittomaki K, Lucena JL, Pakarinen P, et al. Mutation in the folliclestimulating hormone receptor gene causes hereditary hypergonadotropic ovarian failure. Cell 1995;82:959-968. 54. Aittomaki K, Herva R, Stenman UH, et al. Clinical features of primary ovarian failure caused by a point mutation in the follicle-stimulating hormone receptor gene.J Clin Endocrinol Metab 1996;81:3722-3726. 55. Crisponi L, Deiana M, Loi A, et al. The putative forkhead transcription factor FOXL2 is mutated in blepharophimosis/ptosis/epicanthus inversus syndrome. Nat Genet 2001;27:159-166. 56. Bodega B, Porta C, Crosignani PG, Ginelli E, Marozzi A. Mutations in the coding region of the FOXL2 gene are not a major cause of idiopathic premature ovarian failure. Mol Hum Reprod 2004;10:555-557. 57. De Baere E, Dixon MJ, Small KW, et al. Spectrum of FOXL2 gene mutations in blepharophimosis-ptosis-epicanthus inversus (BPES) families demonstrates a genotype-phenotype correlation. Hum Mol Genet 2001;10:1591-1600. 58. Fogli A, Rodriguez D, Eymard-Pierre E, et al. Ovarian failure related to eukaryotic initiation factor 2B mutations.Am J Hum Genet 2003;72: 1544-1550. 59. Di Pasquale E, Beck-Peccoz P, Persani L. Hypergonadotropic ovarian failure associated with an inherited mutation of human bone morphogenetic protein-15 (BMP15) gene. Am J Hum Genet 2004;75: 106-111. 60. Ahonen P, Myllarniemi S, Sipila I, Perheentupa J. Clinical variation of autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED) in a series of 68 patients. N Engl J Med 1990;322: 1829-1836. 61. Nagamine K, Peterson P, Scott HS, et al. Positional cloning of the APECED gene. Nat Genet 1997;17:393-398.
ROBERT W. REBAR 62. Silva de Sa MF, Matthews MJ, Rebar RW. Altered forms of immunoreactive urinary FSH and LH in premature ovarian failure. Infertility 1988;11:1-11. 63. Sluss PM, Schneyer AL. Low molecular weight follicle-stimulating hormone receptor binding inhibitor in sera from premature ovarian failure patients. J Clin Endocrinol Metab 1992;74:1242-1246. 64. Verp M. Environmental causes of ovarian failure. Semin ReprodEndocrinol 1983;1:101-111. 65. Ash P. The influence of radiation on fertility in man. Br J Radiol 1980;53:271-278. 66. Jacox H. Recovery following human ovarian irradiation. Radiology 1939;32:538-545. 67. Tulandi T, Al-Took S. Laparoscopic ovarian suspension before irradiation. Fertil Steri11998;70(2):381-383. 68. Bisharah M, Tulandi T. Laparoscopic preservation of ovarian function: an underused procedure. Am J Obstet Gyneco12003;188:367- 370. 69. Homing SJ, Hoppe RT, Kaplan HS, Rosenberg SA. Female reproductive potential after treatment for Hodgkin's disease. N Engl J Med 1981;304:1377-1382. 70. Koyama H, Wada T, Nishizawa Y, Iwanaga T, Aoki Y. Cyclophosphamide-induced ovarian failure and its therapeutic significance in patients with breast cancer. Cancer 1977;39:1403-1409. 71. Stillman RJ, Schiff I, Schinfeld J. Reproductive and gonadal function in the female after therapy for childhood malignancy. Obstet Gynecol Surv 1982;37:385-393. 72. Whitehead E, Shalet SM, Blackledge G, et al. The effect of combination chemotherapy on ovarian function in women treated for Hodgkin's disease. Cancer 1983;52:988-993. 73. Green DM, Zevon MA, Lowrie G, Seigelstein N, Hall B. Congenital anomalies in children of patients who received chemotherapy for cancer in childhood and adolescence. N EnglJ Med 1991;325:141-146. 74. Morrison JC, Givens JR, Wiser WL, Fish SA. Mumps oophoritis: a cause of premature menopause. Fertil Steri11975;26:655-659. 75. Jick H, Porter J. Relation between smoking and age of natural menopause. Report from the Boston Collaborative Drug Surveillance Program, Boston University Medical Center. Lancet 1977;1:1354-1355. 76. Bateman BG, Nunley WC Jr, Kitchin JD 3rd. Reversal of apparent premature ovarian failure in a patient with myasthenia gravis. Fertil Steri11983;39:108-110. 77. Coulam CB, Kempers RD, Randall RV. Premature ovarian failure: evidence for the autoimmune mechanism. Fertil Steri11981;36:238-240. 78. Lucky AW, Rebar RW, Blizzard RM, Goren EM. Pubertal progression in the presence of elevated serum gonadotropins in girls with mukiple endocrine deficiencies.J Clin Endocrinol Metab 1977;45:673 -678. 79. Rabinowe SL, Berger MJ, Welch WR, Dluhy RG. Lymphocyte dysfunction in autoimmune oophoritis. Resumption of menses with corticosteroids. Am J Med 1986;81:347- 350. 80. Chen S, Sawicka J, Betterle C, et al. Autoantibodies to steroidogenic enzymes in autoimmune polyglandular syndrome, Addison's disease, and premature ovarian failure.J Clin EndocrinolMetab 1996;81:1871-1876. 81. Turkington RW, Lebovitz HE. Extra-adrenal endocrine deficiencies in Addison's disease. Am JMed 1967;43:499-507. 82. Betterle C, Volpato M, Rees Smith B, et al. I. Adrenal cortex and steroid 21-hydroxylase autoantibodies in adult patients with organ-specific autoimmune diseases: markers of low progression to clinical Addison's disease. J Clin Endocrinol Metab 1997;82:932-938. 83. Betterle C, Volpato M, Pedini B, et al. Adrenal-cortex autoantibodies and steroid-producing cells autoantibodies in patients with Addison's disease: comparison of immunofluorescence and immunoprecipitation assays. J Clin Endocrinol Metab 1999;84:618-622. 84. Edmonds M, Lamki L, Killinger DW, Volpe R. Autoimmune thyroiditis, adrenalitis and oophoritis. Am J Med 1973;54:782- 787.
CHAPTER 7 Premature Ovarian Failure 85. Pekonen F, Siegberg R, Makinen T, Miettinen A, Yli-Korkala O. Immunological disturbances in patients with premature ovarian failure. Clin Endocrinol (Oxf) 1986;25:1-6. 86. Walfish PG, Gottesman IS, Shewchuk AB, et al. Association of premature ovarian failure with HLA antigens. Tissue Antigens 1983;21: 168-169. 87. McNatty KP, Short RV, Barnes EW, Irvine WJ. The cytotoxic effect of serum from patients with Addison's disease and autoimmune ovarian failure on human granulosa cells in culture. Clin Exp Immunol 1975;22:378-384. 88. Muechler EK, Huang KE, Schenk E. Autoimmunity in premature ovarian failure. IntJ Ferti11991;36:99-103. 89. LaBarbera AR, Miller MM, Ober C, Rebar RW. Autoimmune etiology in premature ovarian failure. Am J Reprod Immunol Microbiol 1988;16:115-122. 90. Kim JG, Anderson BE, Rebar RW, LaBarbera AR. A biotinstreptavidin enzyme immunoassay for detection of antibodies to porcine granulosa cell antigens. J Immunoassay 1991;12:447-464. 91. Chiauzzi V, Cigorraga S, Escobar ME, Rivarola MA, Charreau EH. Inhibition of follicle-stimulating hormone receptor binding by circulating immunoglobulins. J Clin Endocrinol Metab 1982;54:1221 - 1228. 92. Wheatcroft NJ, Salt C, Milford-Ward A, Cooke ID, Weetman AR Identification of ovarian antibodies by immunofluorescence, enzymelinked immunosorbent assay or immunoblotting in premature ovarian failure. Hum Reprod 1997;12:2617- 2622. 93. Novosad JA, Kalantaridou SN, Tong ZB, Nelson LM. Ovarian antibodies as detected by indirect immunofluorescence are unreliable in the diagnosis of autoimmune premature ovarian failure: a controlled evaluation. BMC Womens Health 2003;3:2. 94. Miller ME, Chatten J. Ovarian changes in ataxia telangiectasia. Acta Paediatr &and 1967;56:559-561. 95. Rebar RW, Morandini IC, Erickson GF, Petze JE. The hormonal basis of reproductive defects in athymic mice: diminished gonadotropin concentrations in prepubertal females. Endocrinology 1981;108:120-126. 96. Rebar RW, Morandini IC, Benirschke K, Petze JE. Reduced gonadotropins in athymic mice: prevention by thymic transplantation. Endocrinology 1980;107(6):2130-2132. 97. Healy DL, Bacher J, Hodgen GD. Thymic regulation of primate fetal ovarian-adrenal differentiation. Bid Reprod 1985;32:1127-1133. 98. Rebar RW, Miyake A, Low TL, Goldstein AL. Thymosin stimulates secretion of luteinizing hormone-releasing factor. Science 1981;214: 669-671. 99. Gulyas BJ, Hodgen GD, Tullner WW, Ross GT. Effects of fetal or maternal hypophysectomy on endocrine organs and body weight in infant rhesus monkeys (Macaca mulatta): with particular emphasis on oogenesis. Biol Reprod 1977;16:216-227.
109 100. Jones GS, De Moraes-Ruehsen M. A new syndrome of amenorrhea in association with hypergonadotropism and apparently normal ovarian follicular apparatus. Am J Obstet Gyneco11969;104:597-600. 101. Alzubaidi NH, Chapin HL, Vanderhoof VH, Calls KA, Nelson LM. Meeting the needs of young women with secondary amenorrhea and spontaneous premature ovarian failure. Obstet Gynecol 2002;99:720-725. 102. Alper MM, Jolly EE, Garner PR. Pregnancies after premature ovarian failure. Obstet Gyneco11986;67(3 suppl):59S-62S. 103. Check JH, Chase JS. Ovulation induction in hypergonadotropic amenorrhea with estrogen and human menopausal gonadotropin therapy. Fertil Steri11984;42:919-922. 104. Check JH, Chase JS, Spence M. Pregnancy in premature ovarian failure after therapy with oral contraceptives despite resistance to previous human menopausal gonadotropin therapy. Am J Obstet Gynecol 1989;160:114-115. 105. CheckJH, Chase JS, Wu CH, Adelson HG. Ovulation induction and pregnancy with an estrogen-gonadotropin stimulation technique in a menopausal woman with marked hypoplastic ovaries. Am J Obstet Gyneco11989;160: 405 - 406. 106. Check JH, Wu CH, Check ML. The effect of leuprolide acetate in aiding induction of ovulation in hypergonadotropic hypogonadism: a case report. Fertil Steri11988;49:542-543. 107. Ledger WL, Thomas EJ, Browning D, Lenton EA, Cooke ID. Suppression of gonadotrophin secretion does not reverse premature ovarian failure. BrJ Obstet Gynaeco11989;96:196-199. 108. Lutjen P, Trounson A, Leeton J, et al. The establishment and maintenance of pregnancy using in vitro fertilization and embryo donation in a patient with primary ovarian failure. Nature 1984;307:174-175. 109. Chan CL, Cameron IT, Findlay JK, et al. Oocyte donation and in vitro fertilization for hypergonadotropic hypogonadism: clinical state of the art. Obstet Gynecol Surv 1987;42:350-362. 110. Sauer MV, Paulson R. Oocyte donation for women who have ovarian failure. Contemp Obstet Gyneco11989:125-135. 111. Rebar RW, Cedars MI. Hypergonadotropic forms of amenorrhea in young women. Endocrinol Metab Clin North Am 1992;21:173-191. 112. Karnis ME Zimon AE, Lalwani SI, et al. Risk of death in pregnancy achieved through oocyte donation in patients with Turner syndrome: a national survey. Fertil Steri12003;80:498-501. 113. Practice Committee of the American Society for Reproductive Medicine. Increased maternal cardiovascular mortality associated with pregnancy in women with Turner syndrome. Fertil Steril 2005;83:1074-1075.
This Page Intentionally Left Blank
2HAPTER
Reproductive Options for Perimenopausal and Menopausal Women M A R K V. S A U E R PRATI
VARDHANA
Columbia University Medical Center, New York, NY 10032 Columbia University Medical Center, New York, NY 10032
Since 1990, there has been a substantial increase in the number of perimenopausal and menopausal women interested in fertility care. This rise followed the publication of successful pregnancies in frankly menopausal women undergoing oocyte donation, which focused international attention on the reproductive problems of women in their 40s and 50s (1). Presently, more than 9000 cases of oocyte donation are performed in the United States, mostly for women of advanced reproductive age. In 2003, of the 122,872 assisted reproductive technology (ART) cycles performed, 12%, or approximately 14,000 cycles, utilized donated eggs or embryos, a percentage that has been steadily increasing since 1995 (Fig. 8.1). Similarly, more than 12,000 cases of in vitro fertilization are initiated annually in women older than 40 years of age, as reported by the Society for Assisted Reproductive Technology (SART) and the Centers for Disease Control and Prevention (CDC) (2). Twenty percent of women undergoing ART cycles in 2003 were over the age of 40 (Fig. 8.2). However, despite the rising enthusiasm for fertility care, success rates in older women using their own oocytes have not significantly improved and remain very low compared with pregnancy rates observed in younger patients (Fig. 8.3). Poor TREATMENT OF THE POSTMENOPAUSAL WOMAN
outcomes are a result of natural ovarian senescence, which directly contributes to reduced fertility and pregnancy wastage in this population. Consequently, a woman's age is the most important factor affecting the chances of a live birth when her own eggs are used. Unless the aging oocyte is replaced, most efforts at assisted reproduction are destined to fail once perimenopausal signs and symptoms are present. The term fecundity refers to a woman's normal ability to reproduce. A fecundability rate is often used to describe the monthly conceptions that occur among sexually active couples within a population. Fecundability rates have been calculated for many different populations and vary slightly according to cultural, religious, and sexual practices. However, a feature common to all groups is the inevitable decline in fecundity that accompanies aging. Women most often conceive and deliver their children while in their twenties. Typically, fertility rates decline during the fourth decade of life, reaching a nadir by the time women enter their early forties. The inevitable loss of fertility is readily apparent when reviewing the birth rates of "natural populations." Natural populations are composed of individuals who do not practice contraception. Within 111
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
1
1
2
S
A
u
E
New treatment procedures 0.1% (163 cycles)
R
AND VARDHANA
50
Frozen-donor 3.6% I
40
(4,4q
Fresh-donor 8 (9,859 cycles)
3O
t=
8
~1. 20
Frozen-nondonor 14.2 (17,517 cycles)
10
<23 24
Fresh-nondonor 7, (91,032 cycles)
26
28
30
32
34
36
38
40
42
44 >45
Age (years)
Pregnancy rate ~
*Totaldoesnot equal 100%due to rounding.
* For
FIGURE 8.1 Types of ART procedures in the United States, 2003. (Data from www.cdc.gov/ART/ART2003.)
natural populations, fertility remains relatively stable until women reach approximately 30 years of age, when a significant fall occurs (Fig. 8.4) (3). By age 35, delivery rates are reduced by one-half, and by age 45, live births are diminished by 95% from values seen in the same population at age 25 (4). Comparing figures from natural populations to delivery rates in the U.S. population at large, similar trends are apparent (5). Less than 2% of all live births occur in women older than 40 years. By age 47, this is further reduced to a mere 0.01% of deliveries (6). Although the lay press has focused on sensational births occurring in menopausal women, few individuals are actually able to successfully deliver a healthy baby beyond the age of 45 without assistance from oocyte donation.
consistency, all
rates are b a s e d o n
Live birth rate ~+.- Singleton live birth rate r
started.
FIGURE 8.3 Pregnancy rates, live birth rates, and singleton live birth rates for women of different ages who had ART procedures using fresh nondonor eggs or embryos in 2003. (Data from www.cdc.gov/ART/ ART2003.)
Further complicating the decreasing fertility rate of older women is the exponential rise in the incidence of aneuploidy noted in the embryos of conception cycles. This phenomenon leads to an elevated rate of miscarriage and an increase in the number of observed anomalies in the delivered offspring. For example, at age 25, only 10% of clinically diagnosed pregnancies end in spontaneous abortion (7). By age 45, the spontaneous abortion incidence is 40% to 50%. Increased rates of spontaneous abortion occurring with advancing maternal age is further demonstrated by reviewing studies of women undergoing artificial 600
........................
" ~
g
.
_
.
.
~
E
10 55
84184
8
7 cO6 O
o5 13_
20-24
4
30-34
35-39
4~4
4~g
Mother*s Age at Oelive~
2 1 0
25-29
<23 24
26
28
30
32 34 36 Age (years)
38
40
42
44 >45
FIGURE 8.2 Age distribution of women who had ART cycles using fresh nondonor eggs or embryos, 2003. (Data from www.cdc.gov/ART/ ART2003.)
FIGURE 8.4 Fertility rates in natural populations, and recent United States populations, note a dramatic fall beginning approximately at age 30 years to almost negligible levels at age 45 years. [black four pointed star], Hutterites (United States 20th Century); ", Burgeoisie Geneva 17th Century; [white circle], Burgeoisie Geneva 16th Century; [black small square], French Village 17th Century; [white square], Iranian Village 20th Century; [white up pointing small triangle], United States (1955); [black up pointing small triangle], United States (1981). Maroulis GB. Affect of aging on fertility and pregnancy. Semin ReprodEndocrino11991;9:165-175.
CHAPTER 8 Reproductive Options for Perimenopausal and Menopausal Women insemination using donor sperm (8). Even in ART cycles, miscarriage rates approach almost 30% in women using their own eggs at the age of 40 (Fig. 8.5). Pregnancy wastage is thought to result principally from random mutations within resting oocytes. Throughout life, human eggs are suspended in development at the diplotene stage of meiosis I. The oocytes residing in the ovaries during the perimenopause have been present since before birth. The protracted process of oocyte aging seems to exert its deleterious effects primarily on the cell nucleus, as evidenced by the positive correlation of maternal age with chromosomal aberrations. As a result, trisomy is witnessed in only 0.1% of newborns of 25-year-old mothers, rising to 10% as maternal age reaches 45 years (9). The meiotic competence of in vitro matured human oocytes is influenced adversely by age, with an increased frequency of errors in chromosome segregation at the first meiotic division (Table 8.1) (10). Studies performing preimplantation genetic diagnosis (PGD) using fluorescence in situ hybridization (FISH) in IVF embryos have documented the increase in chromosomal aneuploidies with advancing maternal age (Fig. 8.6) (11). These findings agree with observations that a high percentage of abortuses in women of advanced reproductive age are chromosomally abnormal (12). In a teleologic sense, preimplantation and postimplantation losses protect the species from unwanted genetic mutations while simultaneously minimizing the health risks posed by pregnancy in the older individual. Although less well defined, adverse reproductive events also occur in men older than 55 years of age. Advanced paternal age has been associated with trisomies and the iso-X syndrome (13,14). Other reproductive hazards include an increased incidence of chronic genitourinary ailments, particularly prostatitis and epididymitis, which affect fertilization in vivo and in vitro. Chronic infections are difficult or impossible to eradicate using antibiotic therapy. Even when infections are successfully treated, lingering inflammation in the genitourinary tract may produce leukocytospermia, 70 60 50 =
o
40
a.
30
10 <25 25
~!ryo~ 4(
S,ow--' ,iiii'.::ilill ~-Fragm~2n~ted Embryos. ArrestedEmbry~"~oq
......................................................................................................................................................... ~ J
>., 3 5::~ 0
30
= 25 c
20 ..
o15 "~1o rv 26-34 yrs 35-39 yrs *p=O.O05comparedto 26-34 yr old group
40-47 yrs
FIGURE 8.6 Aneuploidyincreases with maternal age as determined by fluorescence in situ hybridization assays of preimplanted embryos. (From Munn4 S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil Steri11995;64:382-391.)
which is known to inhibit fertilization in vitro (15). Because older females are more likely to partner with older males, these additional fertility risks further confound successful outcomes. When pregnancies do occur in women of advanced reproductive age, the obstetric risks are also increased. Delayed childbearing is associated with adverse perinatal outcomes as observed in the Swedish Medical Birth Register (16). This review focused on the obstetric experience of 173,715 nulliparous Nordic women who were 20 years or older. Mothers older than 40 years experienced an approximately one-and-a-half- to twofold increased risk for growth retardation, preterm birth, and late fetal and early neonatal deaths compared with women younger than 25. Stillbirth rates also rise sharply after age 40 (17). As a population, women older than 35 are considered "high risk" because of their increased likelihood for developing gestational diabetes, hypertension, preterm labor, and growth retardation (9). Even in donor oocyte pregnancies in menopausal women over 50, there is a high rate of antenatal complications, with the majority being gestational hypertension (18). Therefore, careful obstetric surveillance is necessary in older women attempting pregnancy. TABLE 8.1 Aberrations in Metaphase II Spindle Formation and Chromosomal Alignment Related to the Age of the Patient
20
0
%
113
Patient age (years) 27
29
31
33 35 37 Age (years)
39
41
43>43
Miscarriagerates amongwomenwho had ART cyclesusing flesh nondonoreggs or embryos,by age of woman, 2003.
Oocytes with aberrations
<35
11%
-->35
71%
FIGURE 8.5
Adapted from reir 10.
1
1
4
S
A
U
The prevalence of childlessness and infertility in older couples is difficult to ascertain. However, a progressive increase in the number of childless couples has been described. It is estimated that only 5% of childless couples in their early 20s wish to begin a family, compared with more than 60% of couples in their 40s (2). Biologically, women may be best suited to reproduce while in their 20s. However, psychosocially, many young individuals are neither in a position to raise a child nor desirous to begin a family. Demographic data from the United States indicates that the majority of women utilizing ART procedures to achieve pregnancy are over the age of 30, and oftentimes, over 40 years (5). Reasons given for this delay include the pursuit of educational and vocational goals, later marriages, an increased prevalence of divorce, and the widespread availability of effective birth control. Many women are electing to begin a family later in life as a result of second marriages. Unfortunately, many individuals are unaware of the change in fecundity status that naturally occurs with advanced age and suddenly find themselves unable to conceive despite having had little or no problem in the past. In one survey, among women older than 40 interested in oocyte donation two-thirds had never delivered a baby and 51% were recently divorced (19). In general, women seeking fertility care who are older than 40 have poor reproductive outcomes (20-24) (Fig. 8.7). Registries that track and tally success rates for assisted reproduction have reported similar findings from various parts of the world (25-27). The pregnancy rates logged in the medical literature actually overestimate the likelihood of achieving pregnancy, since manywomen entering treatment are dropped from therapy because of poor responses to controlled ovarian hyperstimulation and are not included in tallies. In essence, evolution has precluded many modern women from having children after the age of 40 and certainly by age 50, when most individuals experience complete cessation of ovulatory function. Unlike other mammals, the human ovaries have largely exhausted their supply of oocytes by the time menopause occurs (28). This diminution inevitably occurs despite the fact that less than 0.001% of the ovary's original number of oocytes are actually ovulated. Histologic studies reveal that, regardless of chronologic age, only a few thousand eggs remain by menopause (29). The compensatory rise in stimulating pituitary gonadotropin that normally accompanies ovarian failure is unable to recruit eggs from this surviving cohort. Cadaver studies indicate a decline in follicular mass with advancing age, with accelerated rates of follicular atresia occurring during the last decade of reproductive life. Similarly, ovaries removed from healthy women of various ages and analyzed for the presence of gametes demonstrate an accelerated depletion of oocytes as menopause approaches. Curiously, the largest turnover of eggs occurs before birth, with a steady decline in number noted, from approximately 7 million oocytes at 20 weeks' gestation to about 2 million at the time of delivery (30). At the time
E
R
AND VARDHANA
% 35--
30-
25-
201
15
10
0
20
I I ', I I ', I I I I I I I I I I i I I I I I I I I I ', 25 30 35 40 45
Age
FIGURE 8.7 Pregnancy rates for assisted reproduction fall to very low levels after the age of 40, as demonstrated by the French National Register on in vitro Fertilization ( H V N A T ) data of 5500 cycles.
of menarche there are approximately 300,000 eggs, and by menopause, few primordial follicles remain (28) (Fig. 8.8).
I. NATURAL FERTILITY IN THE PERIMENOPAUSE Despite the low incidence of pregnancy in women of advanced reproductive age, spontaneous conceptions do occur even in the face of elevated gonadotropins. However, reports of healthy older women undergoing artificial insemination demonstrate reduced fecundity, with cumulative pregnancy rates approximating 40% (31,32). Spontaneous abortions are common and occur in as many as one-half of the clinical pregnancies reported for these women. A high percentage of losses result from aneuploidy. Anomalies in live births are also increased. For instance, Down syndrome occurs in 0.5 to 0.7 per thousand live births in 25-year-old mothers, but the number rises to 75.8 to 152.7 per thousand live births by age 49 (33). Of 2404 amniocenteses performed
CHAPTER 8 Reproductive Options for Perimenopausal and Menopausal Women
FIGURE 8.8 Decline in oocytes from birth to menopause. Faddy MJ, Gosden RG, Gougeon A, et al. Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause. Hum Reprod 1992;7:1342-1346; Erickson GF. Ovarian anatomy and physiology. In: Lobo RA, KelseyJ, Marcus R. Menopause."biology and pathobiology. San Diego, CA: Academic Press, 2000:13-32.
because of advanced maternal age, 2.4% were discovered to be aneuploid, and 50% of these were trisomy (34). As menopause approaches, menstrual cycles rhythm generally decreases in length as a result of oocyte loss, earlier follicular recruitment and shortening of the follicular phase (35-37). As a measure of reproductive reserve, serum follicle-stimulating hormone (FSH) levels drawn on the third day of the menstrual cycle have been used as prognostic indicators before in vitro fertilization (38). Levels above 15 mIU/mL are associated with decreased success rates, and when greater than 25 mIU/mL, pregnancy rarely occurs. Elevated values are observed with increasing frequency when evaluating women older than 40 years of age and are common in most women older than 45. An increase in early follicular phase FSH coincides with the period in which diminished fecundity rates are witnessed. The elevated gonadotropins represent compensatory stimulation resulting from a progressively dwindling number of functioning follicles. This rise in FSH is caused in part by a decreased secretion of ovarian inhibin B, a glycoprotein heterodimer produced by the granulosa cells of the developing antral follicles and a decrease in inhibin A, secreted by the corpus luteum (39-43). The loss of negative feedback inhibition triggers the rise in FSH levels (39,40,44). It appears that a state of
115
"reproductive menopause" exists up to 10 years before the cessation of menses heralds the onset of the "endocrine menopause" (Fig. 8.9) (45). Traditional therapies designed to enhance fertility during this transition period are likely to fail, and live births occur in fewer than 5% of treatment cycles (46,47). The identification and accurate measurement of inhibin has provided further insight into the effect of age on follictdogenesis. Inhibin correlates with follicular function and granulosa cell competence and is decreased with age (41,48). Inhibin measured in the follicular fluid of hyperstimulated ovaries aspirated for purposes of in vitro fertilization reflects correlations among the number of recruited follicles, oocytes retrieved, and embryos produced. Not surprisingly, inhibin was reduced in women experiencing a poor response to ovarian hyperstimulation (49). As a serum marker, it may be a more sensitive prognostic indicator of ovarian reproductive competence than levels of serum FSH. After menopause, inhibin is undetectable in serum samples. In vitro studies of cultured luteinized granulosa cells have shown that in patients with low basal FSH values (< 6 IU/L), inhibin secretion is twofold higher than patients with high basal FSH levels (> 10 IU/L) (50). In recent years, new promising markers for assessing reproductive potential have emerged. Another method used to measure ovarian reserve involves ultrasound determination of the number of antral follicles, or developing oocytes, in the early follicular phase. A detectable decline in follicle count precedes the decrease in ovarian steroid hormones and rise in gonadotropins. Sonographic studies confirm that the antral follicle count declines with chronologic age, likely a result of a diminution in the primordial follicle reserve (51). Antral follicles in the ovary can be visualized by transvaginal ultrasound at a size of 2 to 10 mm. Measurements are taken in the early follicular phase (cycle days 2, 3, or 4) (51). Below the age
90 80
E o
_
. 9. . . . . . . . . .
Entry into menopausal transition
::"
Spontaneous menopause
_
...""
~
/"
60 5O 40 30 2O 10
dn
"t
I .... 1..... I
34
36
38
I
40
42
44
I
46
I
48
I
50
I
52
I
54
I
56
I
58
60
Age (yr)
FIGURE 8.9 Rising follicle-stimulating hormone values that mark the beginning of the reproductivemenopauseprecedes actual cessation of bleeding or endocrine menopause. (From ref. 45, with permission.)
1
1
6
S
A
u
~
of 30, the number of antral follicles is much greater than after 40, with women over age 40, usually revealing less than 10 total antral follicles (51,52). The rate of decline of antral follicle counts rapidly increases after the age of 37, reflecting the diminution of the remaining primordial follicle pool (51). Antral follicle counts have been shown to be reproducible over multiple cycles and correlate with age and response to ovarian stimulation during IVF (Fig. 8.10). Therefore, some consider it to be the best marker of ovarian reserve (53,54). Recently, serum antimfillerian hormone (AMH) levels have been introduced as a novel measure of ovarian reserve. Anti-Mfillerian hormone, also called Mfillerian inhibitory substance (MIS), is a product of the granulosa cells within the preantral and antral follicles. Serum MIS levels on cycle day 3 decrease progressively with age and become undetectable after menopause (55). In addition, MIS is related to the number of antral follicles and to the ovarian response to controlled ovarian hyperstimulation. During IVF cycles, higher day 3 serum MIS values are associated with a greater number of retrieved oocytes (56). In a study at one center of 56 women with normal day 3 FSH levels, patients with a poor response to IVF (less than six oocytes retrieved) had significantly lower follicular and luteal phase MIS compared with high responders (20 or more oocytes retrieved) (57). Because MIS levels represent the primordial pool of nongrowing FSH-independent follicles that may respond to exogenous gonadotropins, measuring MIS levels could help to predict ovarian response in assisted reproduction cycles, especially in older women with early follicular FSH levels within the normal range (56). Numerous studies have shown that serum MIS concentrations show the best correlation with diminished
. . . . . . . . . . . . . . . . .
O O O
N 25o
o
~
o
O
o
Oo ~
15-
o
-8ot~,~-- "-~ o ~ R
Z
0
o o-
6 o
O-5 0
.....
i 5
I 10
I 15
I 20
I 25
I 30
35
Antral follicle count (diameter of 2 - 5 m m ) FIGURE 8.10 Scattergram with 95% confidence bands for correlation analysis of number of oocytes retrieved and antral follicle count detected at the beginning of treatment cycles in women who underwent ARTs. Chang MY, Chiang CH, Hsieh TT, Soong YK, Hsu KH. Use of the antral follicle count to predict the outcome of assisted reproductive technologies. Ferti!
Steri11998;69:505-510.
.
R
AND VARDHANA
ovarian reserve. When women are studied prospectively over time by assessing their ovarian reserve, MIS has been shown to best reflect the process of reproductive aging and the continuous decline of the oocyte/follicle pool as compared with FSH, inhibin B, and antral follicle count (58). As such, serum MIS is becoming a standard test performed in many fertility centers prior to the starting controlled ovarian hyperstimulation. The diminished capacity to conceive and carry babies to term typically has been blamed on the ovaries and uterus. A large drop in available oocytes for recruitment and ovulation accompanies senescence in the ovary. Higher rates of chromosomally abnormal eggs are present in the remaining pool for selection. However, in laboratory animals, the uterus also undergoes age-related changes responsible for diminished implantation and pregnancy rates. As a result, older animals eventually are unable to achieve pregnancy despite the transfer of embryos produced by younger donors. The number of implantations per animal and the proportion of mice found to have any implantation a week postconception declined significantly by 9 months of age (59). Similarly, older mice were observed to have fewer implantations sites and subsequently were noted to be twice as likely to resorb an early pregnancy compared with younger animals (60). Correlates using other laboratory animals also exist and include a reduction in litter size with age observed in the hamster (61). In younger animals, 31% of the losses resulted from preimplantation death, and 69% were postimplantation events. However, in senescent hamsters, a reversed scenario was observed, with preimplantation death occurring in 64% and postimplantation losses seen in 36% of cases. The higher overall loss rates of older animals were thought to be a consequence of less viable ova. When older hamsters received ovarian grafts from younger animals, they were better able to support transferred blastocysts, implying the aging corpus luteum also plays a role in the early establishment and maintenance of pregnancy (62). Although the number of implants is significantly reduced in mated female hamsters, the number of decidual reactions per uterine horn is the same in older animals and younger females (63). This may reflect delayed or abnormal patterns of embryonic development in the blastocysts of older individuals. Abnormal concepti may be secondary to impaired oocyte quality or may perhaps be secondary to effects of the oviductal environment that alters early preimplantation development. Many possible factors influence the relationship between the conceptus and the endometrial environment. These include the rate and normal pattern of development of ova within the older female reproductive tract, delayed uterine receptivity to blastocyst implantation secondary to a decreased capacity of older uterine tissues to take up steroids, and the less efficient uterine response to a decidualizing stimulus as seen in aging mice, rats, and hamsters (64). Similarly, the age-related decline in human fertility may partly be caused by uterine factors (65). Uterine receptivity
CHAPTER 8 Reproductive Options for Perimenopausal and Menopausal Women is best measured by comparing individual embryo implantation rates in humans of various ages undergoing in vitro fertilization and embryo transfer. Women younger than 30 years of age approach rates as favorable as 20% per embryo transferred, decreasing to 9% for women 36 years of age and older (66). After age 40, individual embryo implantation rates are less than 5% (67). Alterations in uterine blood flow also occur with declining levels of estradiol, which may adversely affect the local endometrial environment (68). The identification of estrogen receptors in the wall of human uterine arteries supports this hypothesis (69). Fibrotic changes found in the muscularis of the uterine artery further underscore important physiologic changes accounting for decreases in local blood flow (70). Approximately one-half of spontaneously aborted pregnancies are chromosomally normal, implying that a local endomyometrial factor may be responsible for the loss. Whether this is a primary target organ event or secondary to the inability of the aging corpus luteum to support the pregnancy remains conjectural. Unfortunately, it is not possible to dissociate the oocyte from the influence of the local environment when studying normal postfertilization development and implantation. Oocyte and embryo donation to older women, using gametes obtained from younger individuals, provides an ideal opportunity to ascertain the contribution of each of these two variables independently from each other. When prescribed hormone replacement, the endometria of menopausal women between 40 and 60 years of age exhibit normal histologic and steroid receptor response (70). Likewise, biopsied endometria of younger women with ovarian failure are indistinguishable from older women on similar replacement regimens. When adequate doses of exogenous estrogen and progesterone are delivered, most uteri respond appropriately and demonstrate normal endometrial morphology, regardless of the patient's age or diagnosis. The high rate of implantation and delivered pregnancies reported in women of advanced reproductive age undergoing oocyte donation is testimony to the receptivity of a uterine environment supplemented with hormone replacement. Success rates with oocyte donation are three to five times higher than those in older women using their own eggs who are undergoing standard in vitro fertilization (Fig. 8.11) (67).
II. DEVELOPMENT OF OOCYTE AND EMBRYO D O N A T I O N AND ITS APPLICATION IN OLDER W O M E N The ability to transfer preimplantation embryos conceived in vitro or in viva from one female to another has been performed in more than a dozen different mammals
117
FIGURE 8.11 Live birth rates for ART cycles using flesh embryos from donor eggs versus ART cycles using a woman's own eggs among women of different ages. (Data from www.cdc.gov/ART/ART2003.)
(71). First demonstrated in the rabbit, this method was popularized by the animal husbandry business and revolutionized the cattle breeding industry (72). More than 100,000 calves are born annually as a result of refinements in this technique. Transfer of zygotes to recipient animals, synchronized to the menstrual cycle of donors, results in the establishment of pregnancy in 25% to 50% of transfer cycles. Modification of these techniques led to the establishment of the first embryo donation pregnancy in humans in 1983 (73). Early attempts focused on the nonsurgical obtainment of embryos conceived in viva from spontaneously ovulating cycles using uterine lavage. However, natural cycles produced morphologically normal embryos in only 25% of attempts, and as a result, pregnancies in recipients occurred in only 10% to 12% of initiated cycles (74). In addition, the use of uterine lavage of embryos risked the transmission of infectious agents, causing this method to fall out of favor. Unlike the experience in cattle, efforts to maximize the yield of embryos per lavage by superovulating human donors were unsuccessful and resulted in a high rate of complications (75,76). Controlled ovarian hyperstimulation of oocyte donors followed by in vitro fertilization and embryo transfer to women with premature ovarian failure was reported in 1983 and resulted in a successful pregnancy in one of seven women undergoing treatment (77). During the next 10 years, with the advent of transvaginal oocyte aspiration and advances in laboratory techniques, increasing numbers of studies employing a variety of transfer techniques were published showing high rates of successful implantation and pregnancy (78-82).
118
Surprisingly, in women with hypergonadotropic hypogonadism receiving sex steroid replacement and donated oocytes fertilized in vitro, clinical pregnancy rates and individual embryo implantation rates are significantly increased over values normally seen in women attempting in vitro fertilization using their own oocytes (47). This is attributed to the provision of large numbers of high-quality oocytes for in vitro fertilization produced by young, fertile donors, combined with the enhanced endometrial environment provided by the orderly deliverance of controlled doses of estrogen and progesterone. In this manner, embryo quality is maximized and endometrial receptivity simultaneously enhanced (83). By 1990, reports of successful pregnancies in menopausal women beyond the age normally considered to be "premature" appeared (67,84). Reluctance to transfer embryos to women older than 40 was based on the general belief that a uterine factor precluded implantation and pregnancy in aging animals. However, preliminary trials of oocyte donation in menopausal women demonstrated similar success rates for implantation and pregnancy as younger recipients. Obstetric outcomes were favorable, and miscarriage rates were reduced below that normally seen in older mothers (47). As mentioned earlier, donor eggs or embryos were used in approximately 12% of all ART cycles carried out in 2003. The majority of these cycles were performed in women over the age of 39 (2) (Fig. 8.12). Oocyte donation from young donors overcomes the problems of diminished ovarian reserve and increased aneuploidy risk that accompanies advancing age and results in significantly higher pregnancy rates than standard IVF regimens. Women over 45 years of age, even as old as 55, may achieve pregnancy rates on par with young women using their own eggs (18,85,86). Recipient age does not adversely affect cycle outcome when donor
FIGURE 8.12 Percentageof ART cycles using donor eggs, by ART patient's age, 2003.
SAUER AND VARDHANA
oocytes are used, with fertilization rates, embryo implantation rates, and ongoing pregnancy rates similar to younger cohorts (87). Donor egg and embryo transfer provides the most reasonable reproductive option for older women who are either perimenopausal or menopausal and remains the most successful fertility treatment for patients of advanced reproductive age.
III. CHILDBEARING BY THE PERIMENOPAUSAL A N D MENOPAUSAL W O M A N Most women interested in oocyte donation appear to be perimenopausal, usually between the ages of 40 and 50 years (1,2). Older patients traditionally have the worst prognosis for fertility using natural or assisted reproductive techniques with their own eggs (Fig. 8.13). Poor response to ovarian hyperstimulation and IVF has been well documented in older women. Numerous studies have demonstrated a decline in IVF success with advancing age, especially above age 40 (88). With advancing age, the number of oocytes retrieved and embryos obtained, implantation, and viable pregnancy rates rapidly decline. The response to ovarian stimulation is diminished, requiring large dosages of gonadotropins and yielding high cycle cancellation rates (89). Often, poor response to stimulation with gonadotropins is the earliest sign of ovarian aging and is seen prior to any hormonal alterations or menstrual cycle changes (90-92). However, using oocyte donation, women of advanced reproductive age have pregnancy rates similar to younger women.
Demographic differences are apparent in older recipients seeking fertility care compared with women with premature ovarian failure (19). They are commonly employed in fulltime vocational pursuits and in many cases are highly educated. A large percentage of patients have delayed childbearing in order to achieve professional goals. In addition, these women have often remarried, are likely to have undergone an elective termination of pregnancy, and usually have been a recipient of extensive infertility care prior to attempts at oocyte donation (93). Oocyte donation remains one of the most efficient and safe assisted reproductive techniques (1). Endometrial biopsies of recipients age 50 to 60 years suggests that the uterus maintains its ability to respond normally if given pharmacologic doses of sex steroid (18,70). Analysis of egg donation cycles shows no significant differences in rates of implantation or ongoing pregnancy in older women as compared with younger women receiving donated embryos. These rates, however, are significantly higher than the rates in infertile women of similar age undergoing standard in vitro fertilization using their own eggs. This suggests that the
CHAPTER 8 Reproductive Options for Perimenopausal and Menopausal Women
*For consistency, all rates are based on cycles started.
FIGURE 8.13 Pregnancyrates, live birth rates, and singleton live birth rates for ART cyclesusing fresh nondonor eggsor embryosamongwomen aged 40 and older,2003. (Data from www.cdc.gov/ART/ART2003.)
endometrium retains its ability to respond to gonadal steroids and provides a receptive environment for embryo implantation and gestation even in older women (67,94,95). Despite the success of oocyte donation in perimenopausal and postmenopausal women, there is still some debate as to what the upper age limit for donor IVF should be (96). Concerns include the unknown risks to the elderly gravidarum, issues of longevity in the delivering parents, and the "unnatural" method inherent to the process of establishing the gestation (97,98). The recommendation to extend therapy to women over the age of 45 has been made with cautious reservation by the American Society for Reproductive Medicine and with the proviso that recipients be adequately screened medically and psychologically (99). As many as 20% to 30% of potential recipients of advanced reproductive age may fail the screening process and ultimately be precluded from treatment (47). However, after comprehensive medical screening, women found to be emotionally and physically fit have performed well in their attempts at pregnancy, and outcomes have been good.
IV. SCREENING AND PREPARATION OF POTENTIAL RECIPIENTS Screening potential candidates for oocyte donations has centered on testing their overall health. Although the probabilities for establishing pregnancy in recipients may be dramatically altered using oocyte donation, obstetric risks are age related and therefore significantly increased in the older population. Testing cardiovascular health is important,
119
because the stress of pregnancy on the heart is significant. For this reason, baseline assessments to uncover diabetes, hyperlipidemia, and exercise intolerance are also indicated. A generalized search for occult malignancies has uncovered several cancers, including breast, uterine, and cervical carcinomas; lymphoma; and melanoma. Table 8.2 lists the surveillance tests most often used in screening older recipients. Donors and recipients follow a standard synchronization regimen with donors undergoing ovarian downregulation using leuprolide acetate followed by gonadotropins. In many perimenopausal patients, ovarian function is typically still intact, because ovulatory cycles usually continue throughout the 5- to 10-year transition period that defines the perimenopause. Therefore, for purposes of donor synchronization and to avoid an untimely premature ovulation in perimenopausal recipients, which would create a progestational endometrium, it is preferable to downregulate the pituitary of cycling women with a gonadotropin-releasing hormone (GnRH) agonist before prescribing standard hormone replacement. Women undergoing donor IVF have successfully used a variety of hormone replacement regimens. Recipients are treated with oral micronized estradiol (E2) for several days before donor-initiated gonadotropin
TABLE 8.2
Screening Examination Required of Couples of Advanced Reproductive Age Requesting Oocyte Donation
If over 40:
Electrocardiogram Mammogram Chest roentgenogram 2-hour glucose tolerance test If over 45:
Treadmill test If over 50 or premature ovarian failure:
Bone densitometry All women of advanced reproductive age: Blood chemistry panel Sensitive thyroid-stimulating hormone (TSH) Complete blood count with platelets Papanicolaou examination Cervical cultures (gonorrhea/Chlamydia) Rubella Urinalysis and culture Blood type and Rh Hemoglobin electrophoresis Genetic testing (if applicable for certain ethnic groups) Infectious disease, both spouses
Human immunodeficiency virus (HIV- 1/2) Syphilis (RPR) Hepatitis A, B, C screen Human lymphotrophic virus (HTLV-1/2) Reproductive
Transvaginal ultrasound of pelvis Hysterosalpingogram or saline hysterosonogram Semen analysis
120 therapy. Most commonly, oral estradiol, micronized or in the valerate form, has been prescribed. Estrogen has also been delivered transdermally with good results. Advantages of this method include maintenance of physiologic sex steroid levels and lessened hepatic effects (100). However, many patients develop rashes and irritation at the patch sites, and multiple patches must be worn simultaneously to achieve adequate levels of estradiol. Delivery may be accomplished in a sequential step-up fashion to mimic the normal fluctuations in serum estradiol levels (101) or delivered as a fixed continuous dose (102). Pregnancy rates are similar using either approach. When following the step-up approach, synchronization with the donor undertaking ovarian hyperstimulation requires the recipient begin medication 4 to 5 days in advance of the donor's injecting gonadotropins. Pharmacologic levels of circulating sex steroid result from prolonged use of medicinal estrogen, and despite claims of "physiologic" replacement, serum values of estrone, estrone sulfate, and estrone glucuronide are known to be grossly elevated (103). Progesterone is needed to decidualize the endometria in preparation for embryo implantation. Recipients begin taking progesterone on the morning of the day before the donor's egg retrieval and continue progesterone until after the pregnancy is established. A variety of formulations have been used successfully (104,105). Most commonly, intramuscular progesterone given twice daily or daily vaginal progesterone has been recommended. The need to maintain replacement steroids throughout the first trimester commonly leads to local irritation and inflammation at the injection sites. Mternate regimens include suppositories and gel-based or encapsulated micronized progesterone. Embryo transfers occurs 72 hours after retrieval, or 5 days after retrieval in the case of a blastocyst transfer. E2 and progesterone are continued until either a negative pregnancy test results or until 10 to 14 weeks of gestational age (program dependent) in women achieving pregnancy (106) (Fig. 8.14). Morphologic analyses of endometrial biopsy specimens taken from women using hormone replacement therapy have uncovered certain unique characteristics. Although histologically close to normal, mid-luteal samples typically demonstrate a delayed pattern of glandular maturation, exhibited in up to 25% of samples (70). When endometria are resampled later in the cycle (day 26 to 28), biopsy results are usually normal, implying a catch-up phenomenon. Transvaginal ultrasound images denote a homogeneous echodense pattern, with a thickness approaching 8 to 10 mm. However, pregnancies have occurred with measures as thin as 4 mm and as thick as 23 mm (107). Sex steroid receptors for progesterone and estrogen are within normal limits for luteal endometria. In many cases, women might not have been taking hormone replacement therapy and therefore require a priming cycle to develop a full endometrial response. This occurs in
SAUERAND VARDHANA
FIGURE8.14 Schematicof cyclesynchronizationusing a GnRH agonist in both donor and recipient. GnRH agonists are used to down-regulatethe pituitary of recipients with evidence of ovarian activityprior to beginning oral estradiol. Oral estradiol is prescribed to the recipient 4-5 days in advanceof the donor startinggonadotropin injections.Progesteroneis administered starting the day after hCG injection in the donor, and I day prior to aspirating oocytes. Embryo transfer is performed 3 days following oocyte retrieval. Serum pregnancytesting occurs 12 days post-transfer. Pregnant patients are maintained on estradiol and progesterone through 12 weeksof gestational age. SauerMV, Cohen MA. Egg/embryodonation. In: Gardner D, WeissmanA, Howles C, Shohan Z, eds. Textbook of assisted reproductive techniques laboratory and clinicalperspectives, 2nd ed. London: Taylor and Francis, 2001:843-853.
approximately 5% of new cases, regardless of the recipient's age. A mock cycle enables the discovery of such indMduals and permits adjusting for a hyperplastic glandular pattern of response that occasionally occurs (2%) (108). A practice transfer using an embryo transfer catheter is performed at the time of biopsy to measure the length and contour of the cavity and to ensure that a transcervical embryo transfer can easily be accomplished.
V. OBSTETRIC MANAGEMENT AND DELIVERY CONSIDERATIONS Pregnancies resulting from oocyte donation are considered high risk by obstetricians. Although several studies have concluded that the outcomes of pregnancies following oocyte donation are favorable, there are potential obstetric risks such as gestational hypertension, which can be more complicated in older women. Furthermore, the occurrence of multiple gestations is common and is seen in 20% to 40% of live births (Fig. 8.15). Pregnancies initially are documented using serum beta human chorionic gonadotropin ([3hCG) measurements and transvaginal ultrasound. Ultrasound examinations performed early in the gestation document the number of implantation sites and delineate normal embryonic growth. Commonly, supernumerary implantations occur and often fail to develop normally. In many cases, abnormal implantation sites are absorbed without incident. Other times, their collapse results in vaginal bleeding. Ultrasound is useful for identifying patients at risk. Visualizing the early pregnancy appears to facilitate the patient's acceptance of the pregnancy, many of whom initially express difficulty in believing the pregnancy
121
CHAPTER 8 Reproductive Options for Perimenopausal and Menopausal Women
*Number of fetuses not known because the pregnancy ended in early miscarriage.
FIGURE 8.15 Risk of having multiple-fetus pregnancy and multipleinfant live birth from ART cyclesusing fresh donor eggs. (Data from www. cdc.gov/ART/ART2003.) actually occurred. Serial measures of estradiol and progesterone are neither helpful nor indicated, because pharmacologic doses of hormone are delivered daily and serum levels do not reflect the tissue response (103,109). Referral to specialists in maternal and fetal medicine is appropriate given the age of patients. Regardless of their prior health, pregnant women older than 40 should be considered high-risk patients. Reports describe a tendency for hypertensive complications in women after oocyte donation (110). Monitoring for diabetes and early signs of hypertension is important. Serial growth assessment provides early evidence of fetal growth retardation. Late-occurring events that complicate pregnancies in the older women, particularly stillbirth, may best be avoided by an aggressive approach to delivery (111). With full knowledge of the gestational age of these mothers, attempts to induce labor near term (38 to 39 weeks) should be considered iudicious.
lative bodies in the United States will enact restrictive measures to preclude this application of the method. Increasing numbers of women in their 40s are seeking fertility care. Most of these patients fail to become pregnant using their own eggs, and many elect a trial of oocyte donation. For older patients, oocyte donation may actually represent the only viable option for parenting, because adoption services are also difficult to secure for couples of advanced reproductive age. Vigilant surveillance is imperative in the screening of these individuals. To maximize success, a thorough health assessment is mandatory. Oocyte donation was not intended to be used indiscriminately (118). Many abnormalities are discovered, some of which do not preclude treatment, but others may dictate exclusion. This necessitates a more primary care approach by the fertility specialist and requires the development of more discriminatory criteria from that usually practiced when dealing with younger patients. Concerns have been raised regarding the potential harm to the "fabric of society" brought on by allowing older individuals the chance to become parents. However, precedent already exists in that many children have been reared by grandparents, and the grandparents take on most of the parenting role in many cultures. Society has been accepting of older men and younger wives starting families. Typically, in countries where restrictions on oocyte donation exist, no laws preclude males from using donated sperm or older men from procreating with their younger wives. Precluding healthy women from granting themselves a successful alternative for reproduction while allowing their male counterparts access to such opportunity should be considered sexist and prejudicial. As in most cases of social evolution, as increasing numbers of cases accumulate, it is likely that acceptance will follow.
VI. FUTURE APPLICATIONS References Interest has focused on the means for rejuvenating the oocytes of older women using cytoplasm infused from younger donor oocytes (112). Similarly, nuclear transplantation techniques may allow enucleated donor oocytes to be reconstituted with the genetic material of older recipients (113), allowing women an opportunity to perpetuate their genetic lineage. It has also been suggested that women should store or bank their oocytes (oocyte cryopreservation), similar to men and their sperm, to avoid age-related infertility later in life (114). However, given the low rate of success in using cryopreserved oocytes (115), promoting this approach for routine clinical use remains highly controversial. The widespread use of oocyte donation for increasing numbers of women of advanced reproductive age was inevitable. Despite attempts in several countries to limit or prohibit oocyte donation (116,117), it is unlikely that legis-
1. SauerMV. Treatingwomen of advancedreproductiveage.In: Principlesof oocyteand embryodonation. New York: Springer-Verlag,1998;271-292. 2. Assisted reproductive technology in the United States and Canada: 2003 results generated from the American Society for Reproductive Medicine/Society for Assisted Reproductive Technology Register. www.cdc.gov/ART/ART2003, 2005;389. 3. Maroulis G. Effects of aging on fertility and pregnancy. Semin Reprod Endocrino11991;9:165.
4. Henry L. Some data on natural fertility. Eugenics 1961;8:81. 5. Center for Disease Control and Prevention. 2003 Assisted Reproductive Technology (ART) Report. www.cdc.gov/ART/ART2003. 6. HansenJR Older maternal age and pregnancyoutcome: a reviewof the literature. Obstet GynecolSurv 1986;41:726-742. 7. Stein ZA. A woman's age: childbearing and child rearing.Am J Epidemio11985;121:327-342.
8. Virro MR, Shewchuk AB. Pregnancy outcome in 242 conceptions after artificial insemination with donor sperm and effects of maternal age on the prognosis for successful pregnancy. Am J Obstet Gynecol 1984;148:518-524.
122 9. Leveno KJ. Pregnancy after 35. In: Cunningham FG, MacDonald PC, Gant NF: Williams obstetrics, 18th ed., vol suppl 2. Norwalk, CT: Appleton-Century-Crofts, 1989:1-12. 10. Volarcik K, Sheean L, Goldfarb J, et al. The meiotic competence of in-vitro matured human oocytes is influenced by donor age: evidence that folliculogenesis is compromised in the reproductively aged ovary. Hum Reprod 1998;13:154-160. 11. Benadiva CA, Kligman I, Munne S. Aneuploidy 16 in human embryos increases significantly with maternal age. Fertil Steril 1996;2: 248-255. 12. Lauritsen JG. Aetiology of spontaneous abortion. A cytogenetic and epidemiological study of 288 abortuses and their parents. Acta Obstet Gynecol Scand Supp11976;52:1-29. 13. Magenis RE, Chamberlin J, Cruz FF, Gerald SG. Trisomy 21 clown's syndrome. NICHD Mental Retardation Research Center Series. Baltimore: University Park Press, 1981. 14. Lenz W. Epidemiology of congenital malformations.Ann NYAcad Sci 1965;123:228-236. 15. Hellstrom WJG, Neal DE. Diagnosis and therapy of male genital tract infections. Infertil Reprod Med Clin North Am 1992;3:399. 16. Cnattingius S, Forman MR, Berendes HW, et al. Delayed childbearing and risk of adverse perinatal outcome. A population-based study. JAMA 1992;268:886- 890. 17. Public Health Statistics 1978-84. Wisconsin Department of Health and Human Services, 1985. 18. Sauer MV, Paulson RJ, Lobo RA. Pregnancy after age 50: application of oocyte donation to women after natural menopause. Lancet 1993;34 1(8841):321-323. 19. Sauer MV, Paulson RJ. Demographic differences between younger and older recipients seeking oocyte donation. J Assist Reprod Genet 1992;9: 400-404. 20. Medical Research International. In vitro fertilization/embryo transfer in the United States: 1985 and 1986 results from the National IVF/ET Registry. Fertil Steri11988;49:212-215. 21. Medical Research International and the Society of Assisted Reproductive Technology. In vitro fertilization/embryo transfer in the United States: 1987 results from the National IVF-ET Registry. Fertil Steril 1989;51:13-19. 22. Medical Research International and the Society for Assisted Reproductive Technology. In vitro fertilization-embryo transfer in the United States: 1988 results from the IVF-ET Registry. Fertil Steril 1990;53: 13-20. 23. Medical Research International and the Society for Assisted Reproductive Technology. In vitro fertilization-embryo transfer (IVF-ET) in the United States: 1989 results from the IVF-ET Registry. Fertil Steri11991;55:14-22. 24. Medical Research International and the Society for Assisted Reproductive Technology. In vitro fertilization-embryo transfer (IVF-ET) in the United States: 1990 results from the IVF-ET Registry. Fertil Steri11992;57:15-24. 25. Templeton A, Morris JK. Reducing the risk of multiple births by transfer of two embryos after in vitro fertilization. N Engl J Med 1998; 339:573-577. 26. Dicker D, Goldman JA, Burton AH, Dicker RC. Age and pregnancy rates in in vitro fertilization. J In Vitro Fert Embryo Transf 1991;8: 141-144. 27. Piette C, de Mouzon J, Bachelot A, Spira A. In vitro fertilization: influence of women's age on pregnancy rates. Hum Reprod 1990;5: 56-59. 28. Richardson SJ, Senikas V, Nelson JE Follicular depletion during the menopausal transition: evidence for accelerated loss and ultimate exhaustion. J Clin Endocrinol Metab 1987;65(6):1231-1237. 29. Block E. Q.uantitative morphological investigations of the follicular system in women; variations at different ages. Acta Anat (Basel) 1952;14: 108-123.
SAUER AND VARDHANA 30. Baker TG. A quantitative and cytological study of germ cells in human ovaries. Proc R Soc Lond B Bid Sci 1963;158:417-433. 31. Stovall DW, Toma SK, Hammond MG, Talbert LM. The effect of age on female fecundity. Obstet Gyneco11991;77:33- 36. 32. Schwartz D, Mayaux MJ. Female fecundity as a function of age: results of artificial insemination in 2193 nulliparous women with azoospermic husbands. Federation CECOS. NEnglJ Med 1982;306:404-406. 33. Huck E. Rates of chromosomal abnormalities at different maternal ages. Obstet Gyneco11981;58:282. 34. Golbus MS, Loughman WD, Epstein CJ, et al. Prenatal genetic diagnosis in 3000 amniocenteses. N EnglJ Med 1979;300:157-163. 35. Santoro N, Brown JR, Adel T, Skurnick JH. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab 1996;81:1495-1501. 36. Klein NA, Soules MR. Endocrine changes of the perimenopause. Clin Obstet Gyneco11998;41:912-920. 37. Macklon NS, Fauser BC. Ovarian reserve. Semin Reprod Med 2005;23:248-256. 38. Scott RT, Toner JP, Muasher SJ, et al. Follicle-stimulating hormone levels on cycle day 3 are predictive of in vitro fertilization outcome. Fertil Steri11989;51:651-654. 39. Klein NA, Illingworth PJ, Groome NP, et al. Decreased inhibin B secretion is associated with the monotropic FSH rise in older, ovulatory women: a study of serum and follicular fluid levels of dimeric inhibin A and B in spontaneous menstrual cycles. J Clin Endocrinol Metab 1996;81:2742- 2745. 40. Klein NA, Battaglia DE, Fujimoto VY, et al. Reproductive aging: accelerated ovarian follicular development associated with a monotropic follicle-stimulating hormone rise in normal older women. J Clin Endocrinol Metab 1996;81:1038-1045. 41. Danforth DR, Arbogast LK, Mroueh J, et al. Dimeric inhibin: a direct marker of ovarian aging. Fertil Steri11998;70:119-123. 42. Groome NP, Illingworth PJ, O'Brien M, et al. Measurement of dimeric inhibin B throughout the human menstrual cycle. J Clin Endocrinol Metab 1996;81:1401-1405. 43. Santoro N, Adel T, Skurnick JH. Decreased inhibin tone and increased activin A secretion characterize reproductive aging in women. Fertil Steri11999;71:658-662. 44. Soules MR, Battaglia DE, Klein NA. Inhibin and reproductive aging in women. Maturitas 1998;30:193-204. 45. Natchtigall R. Assessing fecundity after age 40. Contemp Obstet Gynecol 1991;36:11. 46. Wood C, Calderon I, Crombie A. Age and fertility: results of assisted reproductive technology in women over 40 years. JAssist Reprod Genet 1992;9:482-484. 47. Sauer MV, Paulson RJ, Lobo RA. Reversing the natural decline in human fertility. An extended clinical trial of oocyte donation to women of advanced reproductive age. JAMA 1992;268:1275-1279. 48. Hughes EG, Robertson DM, Handelsman DJ, et al. Inhibin and estradiol responses to ovarian hyperstimulation: effects of age and predictive value for in vitro fertilization outcome. J Clin Endocrinol Metab 1990;70:358-364. 49. Jacobs SL, Metzger DA, Dodson WC, Haney AE Effect of age on response to human menopausal gonadotropin stimulation.J Clin Endocrinol Metab 1990;71:1525-1530. 50. Seifer DB, Gardiner AC, Lambert-Messerlian G, Schneyer AL. Differential secretion of dimeric inhibin in cultured luteinized granulosa cells as a function of ovarian reserve.J Clin Endocrinol Metab 1996;81: 736-739. 51. Scheffer GJ, Broekmans FJ, Dorland M, et al. Antral follicle counts by transvaginal ultrasonography are related to age in women with proven natural fertility. Fertil Steri11999;72:845- 851. 52. Kline J, Kinney A, Kelly A, Reuss ML, Levin B. Predictors of antral follicle count during the reproductive years. Hum Reprod 2005;20: 2179-2189.
CHAPTER 8 Reproductive Options for Perimenopausal and Menopausal W o m e n 53. Vladimirov IK, Tacheva DM, Kalinov KB, Ivanova AV, Blagoeva VD. Prognostic value of some ovarian reserve tests in poor responders. Arch Gynecol Obstet 2005;272:74- 79. 54. Ng EH, Chan CC, Tang OS, Ho PC. Antral follicle count and FSH concentration after clomiphene citrate challenge test in the prediction of ovarian response during IVF treatment. Hum Reprod 2005;20: 1647-1654. 55. Fanchin R, Schonauer LM, Righini C, et al. Serum anti-Mullerian hormone is more strongly related to ovarian follicular status than serum inhibin B, estradiol, FSH and LH on day 3. Hum Reprod 2003;18: 323-327. 56. Seifer DB, MacLaughlin DT, Christian BP, et al. Early follicular serum mullerian-inhibiting substance levels are associated with ovarian response during assisted reproductive technology cycles. Fertil Steril 2002;77:468-471. 57. Eldar-Geva T, Ben-Chetrit A, Spitz IM, et al. Dynamic assays of inhibin B, anti-Mullerian hormone and estradiol following FSH stimulation and ovarian ultrasonography as predictors of IVF outcome. Hum Reprod 2005;20:3178-3183. 58. van Rooij IA, Broekmans FJ, te Velde ER, et al. Serum anti-Mullerian hormone levels: a novel measure of ovarian reserve. Hum Reprod 2002; 17:3065 - 3071. 59. Harman SM, Talbert GB. The effect of maternal age on ovulation, corpora lutea of pregnancy, and implantation failure in mice. J Reprod Ferti11970;23:33- 39. 60. Holinka CF, Tseng YC, Finch CE. Reproductive aging in C57BL/6J mice: plasma progesterone, viable embryos and resorption frequency throughout pregnancy. Biol Reprod 1979;20:1201 - 1211. 61. Thorneycroft IH, Soderwall AL. The nature of the litter size loss in senescent hamsters. Anat Rec 1969; 165:343 - 348. 62. Blaha GC. The influence of ovarian grafts from young donors on the development of transferred ova in aged golden hamsters. Fertil Steril 1970;21:268-273. 63. Maibenco HC, Krehbiel RH. Reproductive decline in aged female rats. J Reprod Ferti11973;32:121-123. 64. Wener MA, BJ, Gordon JW. The effects of aging on sperm and oocytes. Semin ReprodEndocrino11991;9:231. 65. Levran D, Ben-Shlomo I, Dor J, et al. Aging of endometrium and oocytes: observations on conception and abortion rates in an egg donation model. Fertil Steri11991;56:1091-1094. 66. Sauer MV, Paulson RJ. Oocyte donation to women with ovarian failure. Contemp Obstet Gyneco11989;34:125. 67. Sauer MV, Paulson RJ, Lobo RA. A preliminary report on oocyte donation extending reproductive potential to women over 40. N EnglJ Med 1990;323:1157-1160. 68. de Ziegler D, Bessis R, Frydman R. Vascular resistance of uterine arteries: physiological effects of estradiol and progesterone. Fertil Steril 1991;55:775-779. 69. Perrot-Applanat M, Groyer-Picard MT, Garcia E, Lorenzo F, Milgrom E. Immunocytochemical demonstration of estrogen and progesterone receptors in muscle cells of uterine arteries in rabbits and humans. Endocrinology 1988;123:1511-1519. 70. Sauer MV, Miles RA, Dahmoush L, et al. Evaluating the effect of age on endometrial responsiveness to hormone replacement therapy: a histologic ultrasonographic, and tissue receptor analysis. JAssist Reprod Genet 1993;10:47-52. 71. Buster JE, Sauer MV. Nonsurgical donor ovum transfer: new option for infertile couples. Contemp Obstet Gyneco11986;28:39. 72. Seidel G E Jr. Superovulation and embryo transfer in cattle. Science 1981;211:351-358. 73. Bustillo M, Buster JE, Freeman AG, et al. Nonsurgical ovum transfer as a treatment in infertile women. Preliminary experience. JAMA 1984;251:1171-1173. 74. Sauer MV, Bustillo M, Gorrill MJ, et al. An instrument for the recovery of preimplantation uterine ova. Obstet Gyneco11988;71:804-806.
123
75. Sauer MV, Anderson RE, Paulson RJ. A trial of superovulation in ovum donors undergoing uterine lavage. Fertil Steri11989;51:131-134. 76. Carson SA, Smith AL, Scoggan JL, Buster JE. Superovulation fails to increase human blastocyst yield after uterine lavage. Prenat Diagn 1991;11:513-522. 77. Lutjen P, Trounson A, Leeton J, et al. The establishment and maintenance of pregnancy using in vitro fertilization and embryo donation in a patient with primary ovarian failure. Nature 1984;307:174-175. 78. Sauer MV, Paulson RJ, Macaso TM, Francis MM, Lobo RA. Oocyte and pre-embryo donation to women with ovarian failure: an extended clinical trial. Fertil Steri11991;55:39-43. 79. Rosenwaks Z. Donor eggs: their application in modern reproductive technologies. Fertil Steri11987;47:895- 909. 80. Navot D, Laufer N, Kopolovic J, et al. Artificially induced endometrial cycles and establishment of pregnancies in the absence of ovaries. N EnglJ Med 1986;314:806-811. 81. Abdulla HI, Baber R, Kirkland A, et al. A report on 100 cycles ofoocyte donation: factors affecting the outcome. Hum Reprod 1990;5:1018. 82. Asch RH, Balmaceda JP, Ord T, et al. Oocyte donation and gamete intrafallopian transfer in premature ovarian failure. Fertil Steril 1988;49:263-267. 83. Paulson RJ, Sauer MV, Lobo RA. Factors affecting embryo implantation after human in vitro fertilization: a hypothesis. AmJ Obstet Gynecol 1990; 163:2020- 2023. 84. Serhal PF, Craft IL. Oocyte donation in 61 patients. Lancet 1989;1: 1185-1187. 85. Sauer MV, Paulson RJ, Ary BA, Lobo RA. Three hundred cycles of oocyte donation at the University of Southern California: assessing the effect of age and infertility diagnosis on pregnancy and implantation rates. JAssist Reprod Genet 1994;11:92-96. 86. Yaron Y, Amit A, Brenner SM, et al. In vitro fertilization and oocyte donation in women 45 years of age and older. Fertil Steril 1995;63: 71-76. 87. Legro RS, Wong IL, Paulson RJ, Lobo RA, Sauer MV. Recipient's age does not adversely affect pregnancy outcome after oocyte donation. Am J Obstet Gyneco11995;172:96-100. 88. Elizur SE, Lerner-Geva L, Levron J, et al. Factors predicting IVF treatment outcome: a multivariate analysis of 5310 cycles. Reprod Biomed Online 2005;10:645-649. 89. Lass A, Croucher C, Dully S, et al. One thousand initiated cycles of in vitro fertilization in women > or = 40 years of age. Fertil Steri11998; 70:1030-1034. 90. Lashen H, Ledger W, Lopez-Bernal A, Barlow D. Poor responders to ovulation induction: is proceeding to in-vitro fertilization worthwhile? Hum Reprod 1999;14:964-969. 91. Kailasam C, Keay SD, Wilson P, Ford WC, Jenkins JM. Defining poor ovarian response during IVF cycles, in women aged <40 years, and its relationship with treatment outcome. Hum Reprod2004;19:1544-1547. 92. Cahill DJ, Prosser CJ, Wardle PG, Ford WC, Hull MG. Relative influence of serum follicle stimulating hormone, age and other factors on ovarian response to gonadotropin stimulation. Br J Obstet Gynaecol 1994;101:999-1002. 93. Sauer MV, Ary B R, Paulson RJ. The demographic characterization of women participating in oocyte donation: a review of 300 consecutively performed cycles. IntJ GynaecolObstet 1994;45:147-151. 94. Remohi J, Gartner B, Gallardo E, et al. Pregnancy and birth rates after oocyte donation. Fertil Steri11997;67:717-723. 95. Paulson RJ, Hatch IE, Lobo RA, Sauer MV. Cumulative conception and live birth rates after oocyte donation: implications regarding endometrial receptivity. Hum Reprod 1997;12:835- 839. 96. Sauer MV, Paulson RJ. Understanding the current status of oocyte donation in the United States: what's really going on out there? Fertil Steri11992;58:16-18.
124
SAUER AND VARDHANA
97. Taylor PJ, Gomel V. "Abraham laughed." Int J Fertil 1992;37: 202-203. 98. Seibel M. To everything there is a season. Contemp Obstet Gynecol 1992;37:153-156. 99. Guidelines for gamete and embryo donation. The American Society for Reproductive Medicine. Fertil Steri12002;77(suppl 5). 100. Steingold KA, Matt DW, DeZiegler D, et al. Comparison of transdermal to oral estradiol administration on hormonal and hepatic parameters in women with premature ovarian failure. J Clin Endocrinol Metab 1991;73:275-280. 101. Sauer MV, Stein AL, Paulson RJ, MoTet DL. Endometrial responses to various hormone replacement regimens in ovarian failure patients preparing for embryo donation. IntJ Gynaecol Obstet 1991;35:61-68. 102. Leeton J, Rogers P, Cameron I, Caro C, Healy D. Pregnancy results following embryo transfer in women receiving low-dosage variablelength estrogen replacement therapy for premature ovarian failure.
J In Vitro Fert Embryo Transf1989;6:232-235. 103. Cassidenti D, Miles RA, Vijod A, et al. Comparing responses to varying hormone replacement regimens prior to embryo donation: a histologic, serologic, and receptor analysis. In: Proceedingsof the 38th Annual Meeting of the Societyfor GynecologicInvestigation. San Antonio, 1992. 104. Sauer MV. Progesterone therapy: modern uses and treatment alternatives. Contemp Obstet Gyneco11997;42:4- 7. 105. Sauer M. Hormone replacement prior to embryo donation in women with ovarian failure. Female Patient 1991;16:15. 106. Pena JE, Chang PL, Chan LK, et al. Supraphysiological estradiol levels do not affect oocyte and embryo quality in oocyte donation cycles. Hum Reprod 2002;17:83 - 87. 107. Remohi J, Ardiles G, Garcia-Velasco JA, et al. Endometrial thickness and serum oestradiol concentrations as predictors of outcome in oocyte donation. Hum Reprod 1997;12:2271-2276.
108. Sauer MV, Paulson RJ, Moyer DL. Assessing the importance of endometrial biopsy prior to oocyte donation. J Assist Reprod Genet 1997;14:125-127. 109. Miles RA, Paulson RJ, Lobo RA, et al. Pharmacokinetics and endometrial tissue levels of progesterone after administration by intramuscular and vaginal routes: a comparative study. Fertil Steri11994;62:485-490. 110. Abdulla HI, Billett A, Kan AK, et al. Obstetric outcome in 232 ovum donation pregnancies. BrJ Obstet Gyneco11998;105:332. 111. Sauer MV, Paulson RJ. Establishment of consecutive pregnancies in a menopausal woman following oocyte donation. Gynecol Obstet Invest 1991;32:118. 112. Cohen J, Scott R, Schimmel T, Levron J, Willadsen S. Birth of infant after transfer of anucleate donor oocyte cytoplasm into recipient eggs. Lancet 1997;350:186-187. 113. Takeuchi T, Ergun B, Huang TH, et al. Preliminary experience of nuclear transplantation in human oocytes (Abstract 0-233). In: Pro-
ceedings of the 54th Annual Meeting of the American Societyfor Reproductive Medicine. San Francisco, 1998. 114. Oktay K, Newton H, Aubard Y, Salha O, Gosden RG. Cryopreservation of immature human oocytes and ovarian tissue: an emerging technology? Fertil Steri11998;69:1-7. 115. Tucker MJ, Wright G, Morton PC, Massey JB. Birth after cryopreservation of immature oocytes with subsequent in vitro maturation. Fertil Steri11998;70:578-579. 116. Peinado JA, Russell SE. The Spanish law governing assisted reproduction techniques: a summary. Hum Reprod 1990;5:634-636. 117. Mori T. National regulation of and achievements in assisted reproduction in Japan. Japan Society of Obstetrics and Gynecology.JAssist Reprod Genet 1992;9:293-298. 118. Sauer MV. Motherhood at any age? Egg donation was not intended for everyone. Fertil Steri11998;69:187-188.
SECTION III
The Perimenopause This section of the book deals with a very difficult and arbitrary time in the reproductive life span of a woman. The perimenopause encompasses the time immediately before and the first few years after the last menstrual period, which defines menopause. The temporal relationship of such changes around menopause is quite variable and has been described in the Stages of Reproductive Aging Workshop (STRAW). These stages have been depicted in a figure in Chapter 1, by Speroff, Barnhart, and Gonzalez, as well as in Chapter 11, by Peck, Chervenak, and Santoro, in this section. The variability and unpredictability of menstrual cycles around this time is a common concern for many women who may wish to consider various treatment options. In the Chapter 9, Nancy King Reame describes in detail some of the neuroendocrine changes that occur and the gradual dyssynchronization between the ovary and the hypothalamic-pituitary axis. This chapter should be read together with Chapter 5, by Burger and Teede, which outlines the other hormonal changes occurring during the perimenopause. In Chapter 10, Hale and Fraser discuss the important issue of menstrual irregularity during the perimenopause, a major concern for many women. They discuss why this occurs and possible treatments. In the following chapter, Peck, Chervenak, and Santoro discuss various treatment options for perimenopausal women after reviewing some of the normal changes that occur. The consideration of treatment here includes more than dealing with menstrual irregularity and may involve issues such as with hot flushes. Finally, David Archer discusses the issue of contraception. Some women are concerned about pregnancy until they reach menopause. Contraception for "older" women is, therefore, an important issue to consider, and one in which various risks and benefits need to be examined.
This Page Intentionally Left Blank
2HAPTER !
Neuroendocrine Regulation of the Perimenopause Transition NANCY KING REAME
School of Nursing, Columbia University, New York, NY 10032
I. D E F I N I T I O N OF PERIMENOPAUSE
interval between is associated with a discordant rise in follicle-stimulating hormone (FSH) versus leuteinizing hormone (LH)--that is, the monotropic FSH rise, progressive loss of regular menstrual cyclicity, and depletion of responsive follicles from the ovary (3). This classical view of a finite primordial follicle pool has been challenged recently by John Tilley and colleagues (4,5), who showed in a knockout mouse model that germline stem cells can repopulate a germ cell-depleted postnatal ovary and renew the primordial follicle pool. However, it is unclear to what extent this process would apply to the human menopause. A working model of the progressive changes in FSH and cycle regularity, independent of age, has been developed as the framework for staging the reproductive aging in women (by the Stages of Reproductive Aging [STRAW] workshop) (6). According to STRAW guidelines, five stages precede the final menstrual period (FMP): the early, mid-, and late stages of the reproductive interval, followed by the early and late stages of the menopausal transition. The postmenopause is divided into an early stage (first 12 months after FMP) and late stage (the years beyond the first 12 months after FMP). The perimenopause transition overlaps these arbitrary divisions and extends from the time of onset of highly variable cycle lengths and ends 12 months after the final menstrual period (see Figure 11.1). The median age at onset of menstrual irregularity is about 47 years (7). Despite the inevitability of menopause for all women who live long enough, the events leading up to and following the last menses are highly variable in duration and
Given the increasing life span of women in the United States, the number of years spent in the postmenopausal state is significant. Thus, the cessation of menses and the resulting hypoestrogenism may have important health consequences for the quality of life of a large and growing proportion of the population. Urinary continence, bone metabolism, cardiovascular function, memory and cognition, the synchrony of daily biorhythms, and the aging process itself have all been shown to be influenced by estrogen. Moreover, compared with other organ systems, the female reproductive system is unique in that it undergoes spontaneous senescence at a relatively young age, thus making it an excellent model for the study of the aging process free of chronic disease. The concept of the perimenopause was first introduced by Treolar and colleagues in 1967 (1), when they conducted a cross-sectional analysis of several hundred menstrual cycle calendars obtained from women across the reproductive life span. In that study, the critical marker of aging was menstrual irregularity, defined as a change in genital bleeding to either longer or shorter flow intervals. However, menstrual cycle irregularity is now known to be a late feature of the transition, brought about by neuroendocrine events that occur well before cyclic ovulation is disrupted. Reproductive aging is a continuum that begins with a steep decline in fertility by age 35, long before the final menstruation (menopause) occurs at age 51 on average (2). The TREATMENT OF THE POSTMENOPAUSAL W O M A N
127
Copyright 9 2007 by Elsevier,Inc. AU rights of reproduction in any form reserved.
128
NANCY KING REAME
magnitude from woman to woman, thus contributing to the decadewide span in the timing of age at menopause (45 to 55 years). In a study of ovarian function in 17 women in this age range, the number of follicles present in the ovary varied from more than 1000 in those still cycling, to a few hundred in perimenopause, to none in the postmenopausal state (8). Moreover, although the mean duration of the perimenopausal transition is approximately 4 years, nearly 10% of women experience no changes in cycle regularity, with the persistence of regular cycles until the final menses (9). The variability in the timing, intensity, and severity of the accompanying symptoms is also striking (10). In a study of 463 Japanese women ages 41 to 60 seeking care for menopausal complaints, 22% had regular cycles, 28% were oligomenorrheic, and 50% were postmenopausal (11). Such variability in the menopause experience is believed to result from a variety of external and internal influences, such as a woman's body habitus, smoking status, psychologic well-being, and genetic makeup (12,13). As data are accruing from longitudinal studies of large cohorts of women across the menopause transition, such as the Study of Women Across the Nation (SWAN) (10,14), the importance of both ovarian and nonovarian contributors to the regulation of reproductive aging should slowly emerge. This chapter will summarize evidence that aging effects at all three levels of the hypothalamic-pituitaryovarian (HPO) axis may contribute to the progressively pronounced hypersecretion of pituitary gonadotropins that occurs at the close of the reproductive years. Best understood of the contributors to the aging-related alterations in gonadotropin-releasing hormone (GnRH) secretion are the ovarian causes.
circulating FSH versus LH, which is especially prominent in the early follicular phase. This discordant increase in FSH occurs gradually across the mid-reproductive years, becoming pronounced in women over 40 (15,16). The etiology of this rise of one gonadotropin but not the other, at a time when ovulatory function remains preserved (17-19), has been the topic of intense investigation. The prevailing view is that the monotropic increase in FSH results from a greater sensitivity to declining ovarian feedback on the HPO axis compared with LH. Although a growing body of evidence supports the existence of a separate FSH-releasing factor (20), this factor has yet to be isolated, nor has it been implicated in the monotropic rise. The enhanced levels of FSH in the early follicular phase of older cycling women appears to be important for maintenance of estradiol levels because older women demonstrate a higher estrogen but less suppressed FSH response compared with younger women when challenged with a GnRH agonist with FSH add-back (21). At the time of the final menstrual period, the ovary is reduced to approximately 50% of its premenopausal volume primarily due to a dramatic reduction in follicular mass (22,23). The timing of these changes in morphology is not well understood. Although accelerated follicular depletion has already begun by age 40 (24), estradiol levels often remain normal or elevated and ovulatory cycles are maintained for another decade on average (Fig. 9.1). Moreover, estrogen concentrations are still detectable by ovarian vein sampling in amenorrheic women with elevated plasma gonadotropin levels (25).
Young
[]
Older
9
II. OVARIAN D E T E R M I N A N T S OF REPRODUCTIVE AGING
~
._!
~D ~. "1._1
6
A. Monotropic Rise in FSH The female reproductive cycle in premenopausal women is a highly interactive, exquisitely timed process involving dynamic secretion of hormones and regulatory peptides from the entire HPO axis. It is now generally accepted that GnRH is the stimulating factor for both LH and FSH and any divergence in their secretion can be explained by (a) differential sensitivities of LH and FSH to variations in the dose or frequency of pulsatile GnRH secretion; (b) the gonadal hormonal milieu, including sex steroid and FSH regulatory peptides; (c) and alterations in the pituitary tone of activin or follistatin. Studies have now clarified that the hormonal harbinger of reproductive aging is the disproportionate increase in
..I
6
~"103 !,
3 o
_"•
30
.~
20 g
300
c.._..
200
ID
u.l lOI Follicular
Luteal
Follicular
Luteal
o
FIGURE 9.1 Mean concentrations of gonadotropins and sex steroids in young (mean age = 27.9 + 2 years) and older (mean age = 43.7 + I year) ovulatory women. Group differences: LH, .p < 0.05; FSH, p = < 0.001. (From ref. 42.)
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition
B. FSH Regulatory Peptides As reviewed in Chapter 5 and summarized later, it is now well documented that diminished input from nonsteroidal factors, notably inhibin B from the declining pool of small follicles, is responsible for the dampened inhibitory tone of ovarian feedback on FSH secretion. The inhibin cx-~3A and ~3B protein subunits are members of the transforming growth factor-f3 (TGF-~3) family of peptides. Produced in the ovary and pituitary, they are encoded by distinct genes and dimerize to give rise to inhibin A (~-I3A), inhibin B (cx-~3B), activin A (~3A-~3A), activin AB (~3A-~B), and activin B (~3B-f3B). Inhibins and activins have opposing effects on FSH secretion: inhibins suppress FSH, and activins stimulate FSH production (26). Follistatins, monomeric proteins distinct from both inhibins and activins, have functional overlap with inhibins in suppressing FSH release, through their action as binding proteins for both activins and inhibins (27). Inhibin A (from the mature follicle and corpus luteum) and B (from small antral follicles) are secreted in reciprocal, mirror image patterns of each other across the ovulatory menstrual cycle, acting in tandem to regulate the cyclic profile of FSH secretion (28). Beginning in the early 1990s with the introduction of the dimeric inhibin assays, reduced inhibin B concentrations have been consistently demonstrated in conjunction with enhanced serum FSH elevations in the early follicular phase of cycles in women of advanced reproductive age (29-32). Moreover, daily blood sampling studies have shown lower concentrations of dimeric inhibins throughout the cycle in older women with raised FSH versus younger control subjects (33,34). This hormone pattern has also been confirmed across the menopause transition in a population-based cohort (19). It has been proposed that not all women of advanced reproductive age demonstrate this physiologic loss of FSH restraint. In the single study where 16 women over age 40 were stratified by day 3 FSH values and divided into normal (< 8 mlU/mL; n = 10) or raised FSH groups (> 8 mlU/ mL; n -- 6), no abnormalities in inhibin B in the follicular phase or inhibin A in the luteal phase were evident in the "normal FSH" older group compared with younger control subjects (35). Because ultrasonographic scanning was not performed, it is not known to what extent ovarian morphology may have influenced the findings or served as a confound in group assignment. The inverse relationship between FSH and inhibin B has been shown to persist after pituitary downregulation, suggesting that the low levels of inhibin B in older women are independent of the previous luteal phase inhibin or steroid secretion (36). The pool of developing follicles may be more sensitive to the age-related effects on inhibin production than the corpus luteum, because the decline in inhibin B is already apparent in women as young as age 35, prior to detectable changes in luteal phase inhibin A (34). However, in ovulatory women ages 40
129
to 45, inhibin A was demonstrated to be significantly higher in the presence of lower inhibin B during the intercycle phase compared with women in their 20s (37). However, to presume that a singular deficiency in inhibin activity could account for age-related elevations in FSH fails to consider the other component peptides known to mediate FSH regulation. In addition to the inhibitory role of estradiol and inhibins, studies have now documented the negative influence of follistatin and the positive roles of activin in regulating FSH release. With the development of highly specific assays for most of these regulatory peptides, investigations are now possible of the combined steroid and nonsteroidal ovarian milieu during reproductive aging. Noteworthy is the finding that in postmenopausal women, circulating levels of both follistatin (38) and activin (39) are elevated. To what extent changes in FSH regulatory proteins seen during reproductive aging reflect changes in ovarian production of these regulators is unknown, because these peptides are also produced in multiple other sites, including the pituitary (40,41). We and others have extended the inhibin hypothesis of ovarian aging to examine whether changes in the overall tone of ovarian feedback mediate the age-related rise in FSH. When measured simultaneously, the secretory profiles of the inhibins and activin A have been demonstrated to be altered across the menstrual cycle of women over age 40, despite normal or elevated concentrations of estrogen. Compared with younger-aged controls, activin A demonstrates higher (33,42) or normal (35) concentrations that do not change across the cycle, whereas the overall inhibin tone is reduced in the presence of unchanged follistatin, suggesting the potential for a net stimulatory effect on FSH regulation (33,42,43). In parallel with these endocrine changes, the follicular fluid concentrations of both total follistatin and total activin A are elevated in the dominant follicles of older women, suggesting an enhanced paracrine input to HPO regulation (44). Although the endocrine role of these hormones (inhibin, activin and follistatin) in regulating reproduction has been highlighted, there is evidence to show that their local action is likely to be important. For example, GnRH stimulates both activin and follistatin production by the pituitary, with activin in turn stimulating FSH secretion. Activin also has specific actions on the ovary to stimulate FSH receptors on developing follicles. This altered proportion of dimers in older women has been explained by a reduction in inhibin ci subunit available from the depleted ovarian pool to combine with the [3 subunits to form inhibins, thus favoring activin formation and, in turn, enhanced FSH secretion (33). In a group of perimenopausal women with irregular cycles, the suppression of FSH during hormone replacement therapy was associated with a suppression of activin A in the face of unchanged inhibin B and follistatin (45). As activins appear capable of
130
NANCY KING REAME
direct pituitary stimulation of FSH secretion and are produced in the pituitary (46), this finding is compatible with the view that estradiol's negative feedback action on FSH release may involve indirect paracrine as well as direct endocrine mechanisms (47,48). Because peripheral concentrations of activin A, total follistatin, and free follistatin do not appear to be influenced by cycle phase (42), yet are altered in polycystic ovarian syndrome (49), we examined the relative importance of ovarian function to FSH modulation by simultaneously assessing ovarian morphology (presented in Table 9.1) and the concentrations of FSH regulatory peptides in cycling women of differing ages compared with that of women after spontaneous menopause or ovariectomy while receiving estrogen replacement therapy (50). Findings from this study in which all regulators of activin availability were considered in unison in women with and without ovarian function confirm earlier speculation that the bulk of circulating activin A and follistatin involved in the regulation of FSH comes from nonovarian sources. Failure to restore FSH in the ovariectomized women receiving estrogen therapy (ET) to perimenopausal concentrations adds to the growing body of evidence that declining levels of inhibin B and consequent change in activin signaling play a primary role in the monotropic rise in FSH prior to menopause. The inverse relationship between age and follistatin 288 (the follistatin isoform for which a human assay is available) raises the possibility that follistatin may also have some role in facilitating the age-related increase in FSH. As seen in Fig. 9.2, inhibin B was highest in the women with regular cycles, reduced in perimenopausal women, and below assay sensitivity in both postmenopausal groups. Perimenopausal women presented with larger mean ovarian volume versus older cycling (OC)
TABLE 9.1
women and higher estradiol but fewer follicles. Inhibin A was barely detectable in the perimenopausal group (assay sensitivity = 8 pg/mL). In contrast, activin A was similar across groups, and a nonovarian aging effect explained differences in follistatin (rho = -0.31;p = 0.006). Given the constancy of follistatin levels during the menstrual cycle in the face of dynamically changing levels of FSH (51,52), earlier studies have raised questions about an endocrine role for follistatin in the control of FSH regulation. However, considering that both follistatin and inhibin regulation of FSH are mediated by altered activin signaling (53,54), and the importance of their relative equilibrium for FSH control, the directionality of changes in these regulators in concert points to an endocrine-mediated action for activin. Viewed in this context, activin signaling as a function of the endocrine tone of these regulators is likely to be inversely related to FSH regulation. Confirmation of this premise awaits availability of sensitive assays for measuring free activin in the circulation.
1. ANTIMfJLLERIAN HORMONE AS A NOVEL PREDICTOR OF OVARIAN AGING
Antimtillerian hormone (AMH), also known as mfillerian inhibiting substance (MIS), is a member of the T G F 13 family produced by granulosa cells of early developing follicles (55). Transgenic studies indicate that A M H knockout mice are born with a normal number of follicles but develop an accelerated depletion of ovarian reserve (56). In humans, it has been shown in several longitudinal studies to be a promising predictor of the menopausal transition and ovarian aging in general, even in younger women. The results of de Vet and colleagues (57) suggest that changes in serum A M H levels occur relatively early in
Transvaginal Ultrasound Findings in Women of Similar Age and BMI but Varying Menopause Stage*
Endometrium (mm) Combined ovarian volume (mm) Follicles/subject (2-10 mm) No. subjects with follicles > 10 mm No. follicles > 10 mm
Younger cycling (YC) n = 16 23.9 + 0.9 years
Older c y c l i n g (OC) n = 17 45.9 ___0.8 years
Perimenopausal (PERI) n = 21 49.0 _+ 0.6 years
Postmenopausal (PM) n = 10 46.2 _+ I year
p value
5.4 __ 0.4 1316.3 _+ 132.2
6.0 _+ 0.8 970.7 _+ 129.8a
5.9 _+ 1.2 1550.8 +_ 148.5b
3.1 ---+0.2 a'b 513.6 _+ 81.6a,b,c
0.025 0.001
13.0 __ 2.8
1.8 __ 0.7 a
0.9 __ 0.4a
0.8 +_ 0.5 ~
0.001
0
0
7
1
0
0
11
3
*Cyclingwomen were studied on cycle day 5 _+ 1. Young,cyclingwomen served as control subjects. Values are mean _+ SE. ap 0.02 vs. YC. bp 0.03 VS. OC.
~p < 0.001 vs. PER1.
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition
131
the sequence of events associated with ovarian aging, at a time prior to a prominent change in inhibin B or FSH. When several markers of ovarian reserve were measured at 4-year intervals in normal women who were approaching the menopausal transition, serum level of A M H was the best predictor of the onset of cycle irregularity, compared with inhibin B, antral follicle count, and FSH (58). Thus, the idea has emerged that AMH, as a substance produced in the early stages of follicle development, may be a better reflection of ovarian reserve (of the stock of primordial follicles) than FSH and inhibin B, which are more distally related. A contributory role for A M H in the monotropic rise of FSH has yet to be determined.
C. O v a r i a n A g i n g Effects on L H
FIGURE 9.2 Mean concentrations o f F S H peptide modulators in younger, cycling women compared with four groups of women over age 40 differing by ovarian function. YC, young cycling; OC, old cycling; PER[, perimenopausal; PM, spontaneous menopause; OVX+ET, surgical menopause and receiving estrogen therapy. Data are mean + SEM. Estradiol values, expressed in pmol/L, are as follows: YC = 95.8; O C = 113.8; PERI = 227.5; P M = 24.5; O V X + ET = 297.4. (Adapted from ref. 50.)
The current view is that the regulation of LH secretion is relatively resistant to the incipient decline in ovarian reserve, with perhaps only a subtle rise in concentrations until just prior to menopause, when the ovarian pool of responsive follicles is dramatically reduced along with a steep decline in estradiol and gonadal inhibin (8). In a study of 94 subjects ages 24 to 50 years, a significant increase in mean FSH secretion was detected by age 35, with no aging effects observed in LH secretion until age 45 (15). A later study by the same authors of 500 regularly cycling women demonstrated the onset of the monotropic rise in FSH levels as early as age 28 years, followed by a more modest increase in LH by age 35 (16). This finding of a delayed and more subtle rise in LH in relatively young women led the authors to conclude that an increase in both FSH and LH concentrations could be used as the earliest endocrine markers of reproductive aging. The previously undetected rise in LH prior to age 40 may have been uncovered in this cohort due to the very large sample size. A more recent longitudinal study of 400 cycling women ages 35 to 45 demonstrated that although the progressive rise in FSH observed over the 12-month study was highly correlated with smoking status, the corresponding increase in LH was inversely related to body mass index (BMI), suggesting different etiologies in these observed aging effects (59). The fact that ovariectomy in premenopausal women is associated with a dramatic increase in LH levels to four- to sixfold within 1 to 4 weeks of surgery (60) has served as evidence that ovarian inhibition is the major input to the GnRH-mediated regulation of LH secretion. It should be noted that in the absence of negative feedback signals (estradiol and inhibin) after ovariectomy, the discordance in gonadotropin secretion is particularly evident, with FSH levels rising earlier and remaining persistently higher than LH, suggesting that the unrestrained, endogenous GnRH pulse generator may directly or indirectly favor FSH secretion.
132 Whether inhibin might have an effect on LH secretion is unlikely, as administration of inhibin A to rhesus monkeys in the early follicular phase suppressed FSH but not LH levels (61). In contrast, infusion of activin A into monkeys results in enhanced secretion of both FSH and LH at baseline and after GnRH challenge (62). A gonadotrophin surge attenuating factor (GnSAF) has been proposed as a substance that is produced by ovaries undergoing ovulation induction and that attenuates the endogenous LH surge by reducing the pituitary response to GnRH (63). Mthough not fully characterized, higher bioactivity of GnSAF has been detected in small growing follicles than in preovulatory follicles (64) as well as in the circulation of women during the early and mid-follicular phase compared with the late follicular phase of the normal menstrual cycle (65). In postmenopausal women treated with steroids to simulate the normal menstrual cycle, the LH response to GnRH began earlier in the simulated follicular phase compared with that seen in follicular phase controls (66). The investigators concluded that in functioning ovaries, GnSAF in the early to midfollicular phase of the cycle antagonizes the sensitizing effect of E2 on the pituitary. How such a proposed mediator of the HPO activity might be altered during the menopause transition is not known.
III. DYNAMIC GONADOTROPIN SECRETION IN YOUNG WOMEN It is well established that a pulsatile pattern of gonadotropin-releasing hormone (GnRH) is essential for physiological gonadotrope function. As reviewed by Marshall (67), much has been learned about the precise nature of the pulsatile rhythm that regulates the fertile menstrual cycle through the assessment of pulsatile LH as a surrogate marker of GnRH pulse generator activity. LH episodes originate from periodic activation of the pituitary gonadotrope by intermittent hypothalamic GnRH stimulation. The release magnitude of episodic gonadotropin secretion is defined, among other determinants, by the pituitary responsiveness and the capacity of GnRH to prime the gonadotrope. The activation of gene expression for gonadotropin subunits is governed by the periodicity of GnRH signals to the gonadotropes. GnRH pulse frequency dictates both the absolute levels and the ratio of LH and FSH release, so that more rapid pulse frequencies promote LH secretion and slower frequencies favor FSH. The synchronized firing of the GnRH neurons are modulated by both afferent excitatory and inhibitory inputs. The inhibitory system includes the endogenous opioids, ~/-aminobutyric acid (GABA), and dopamine, whereas the excitatory inputs include nitric oxide, neuropeptide Y, and norepinephrine.
NANCY KING REAME
A. T h e M e n s t r u a l Cycle Numerous cross-sectional (68-70) as well as longitudinal studies (71) of pulsatile LH characteristics during the ovulatory menstrual cycle have demonstrated that LH pulse frequency increases during the follicular phase from approximately one pulse every 90 to 100 minutes to hourly pulses at the time of the LH surge. LH pulse frequency is maintained during the LH surge but slows during the luteal phase, with one pulse every 2 to 6 hours, varying in amplitude. These characteristic changes in LH pulsatility are regulated in large part by steroid-mediated effects on the hypothalamic opioid system, which in turn act as a fluctuating brake on the GnRH pulse generator. Changes in LH pulsatility, rather than FSH, are used to infer GnRH secretion in the peripheral circulation as FSH exhibits a reduced and delayed release by the pituitary, as well as a slower metabolic clearance rate when compared with LH. Thus, the ability to change GnRH amplitude and slow frequency appears to be a critical requirement for the maintenance of cyclic ovulatory function. The dynamic gonadotropin secretory patterns in relation to HPO function have now been well documented for many of the reproductive endocrine disorders and shown to differ from that of the ovulatory menstrual cycle (Table 9.2). This body of knowledge provides important context for examining dynamic changes in the perimenopause years. B. G o n a d o t r o p i n C h r o n o b i o l o g y and Sleep Effects Superimposed on the cyclic changes in pulsatile gonadotropin secretion that occur as a function of the menstrual cycle are the effects of circadian rhythms and sleep. As reviewed by Van Cauter (72), the accumulated evidence suggests that under the influence of the central pacemaker of the suprachiasmatic nucleus (SCN), all hormones of the hypothalamopituitary axis are influenced by both sleep (irrespective of the time of day) and circadian rhythmicity (regardless of the sleep-wake status). Despite a large body of evidence in animals for a putative role, the importance of this SCN-mediated function in human reproduction and particularly menopause is unclear. In terms of GnRH activity, most data have been obtained from men or young, cycling women in whom circadian excursions appear to influence pulse amplitude, while sleep affects pulse frequency (72). In contrast, a recent study by Lavoie et al. (73) of 22 postmenopausal women, where sleep was prevented and environmental conditions were controlled, demonstrated the absence of circadian rhythms for pulsatile LH, FSH, or its glycoprotein free c~ subunit (FAS) despite the maintenance of diurnal fluctuations in cortisol and thyroid-stimulating hormone (TSH). Based on prior
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition
TABLE 9.2
133
Dynamic Gonadotropin Activity in Various Reproductive States in Adult Women
Hormonal component
Postmenopause
Central opioid tone FSH (mlU/ml) LH (mIU/ml) LH pulse interval (min) LH pulse amplitude Estradiol Androgens
Absent >30 >30 60-100 High Very low Low
Cycle day 6 Absent <10 <10 60-100 Low Low Very low
HA Present <10 <10 > 100 Variable Very low Low
PCO Absent <10 >10 60 High Elevated Elevated
HA, hypothalamicamenorrhea;PCO, polycysticovarian syndrome. Modified from ReameNE. In: Lobo RA, ed. Perimenopause.New York: Springer-Verlag,2000;161.
evidence that diverse outputs from the circadian clock exist and differentially influence pituitary function (74), these investigators proposed that the GnRH axis may be uniquely regulated by the SCN relative to other hormonal axes. Findings from hamster studies suggest that circadian regulation of LH release may be mediated through direct input of the suprachiasmatic nucleus. Studies of circadian clock genes in GT1-7 cells (immortalized GnRH-secreting GT1 neurons) demonstrated that perturbation of circadian clock function by transient expression of a dominant-negative clock gene disrupts normal ultradian patterns of GnRH secretion, significantly decreasing mean pulse frequency (75). This finding suggests that an endogenous clock in GnRH neurons assists in the control of normal patterns of pulsatile GnRH secretion. LH is known to demonstrate a nocturnal, sleep-dependent decline in basal concentrations that is especially prominent in the early follicular phase (76). Using a 20-minute sampling frequency, Soules et al. (77) demonstrated that this nocturnal inhibition was due to a slowing of LH pulse frequency during the early morning between 1 AM and 5 AM, with a corresponding increase in LH pulse amplitude. Hypothalamic opioid activity appears to be involved, because pulse frequency is increased after naloxone administration (78), whereas dopaminergic blockade has no effect (79). Moreover, the response to opioid blockade in the early follicular phase is only observed with sleep, in contrast to the marked expression of day-time opioid suppression on LH pulsatility that occurs in the mid-luteal phase (80). During the early follicular phase, LH and FSH responses to 25 Ixg GnRH are markedly blunted when assessed in awake subjects at night; but when administered during sleep, only LH blunting is abolished without a corresponding effect on FSH, suggesting that increased pituitary sensitivity is not involved (81). Hall and colleagues (82) have recently demonstrated that the sleep-entrained slowing of LH pulses is specifically related to deep and REM sleep, whereas brief wake episodes within sleep are stimulatory. The physiologic relevance of sleep-related changes in LH in the early follicular
phase is not known, but it has been suggested to be important for the maintenance of adequate cyclicity and normal folliculogenesis (83).
C. Synchronicityof Other Hormones with Nighttime Gonadotropin Secretion Changes in reproductive hormone release coincident with sleep may represent manifestations of entrained links between CNS regulation and endocrine function. Circadian and sleep-entrained variabilities in the release of other pituitary tropic hormones, such as TSH (84) and prolactin (85), have been well documented. Leptin, a gene product of the obesity (OB) gene, is mainly secreted by the white adipose tissue but also expressed in a variety of organs including the hypothalamus and pituitary (86). It acts by binding to specific leptin receptors, which are expressed in GnRH neurons, in pituitary cells, and in preovulatory follicles, suggesting a possible endocrine/paracrine role in reproduction. Circulating leptin levels display a circadian rhythm that is similar to those for prolactin, thyrotropin, and melatonin but inverse to adrenocorticotropic hormone (ACTH) and cortisol. In rats, the surgical ablation of the SCN abolishes the leptin diurnal secretory pattern (87). It has been demonstrated that the ultradian fluctuations in leptin show pattern synchrony with those of LH in young women during the follicular phase, a time when nocturnal slowing of LH pulses was evident (88). The investigators postulated that in addition to its trophic effects, leptin may contribute to reproductive axis organization by regulating the minute-to-minute oscillations in the levels of LH and estradiol in the critical period before ovulation. Thus, at night as leptin levels rise, the pulsatility profile of LH changes from high frequency/low amplitude to low frequency/high amplitude, becoming synchronous with leptin pulsatility. Such a view would help explain the disruption of HPO function that is characteristic of states of low leptin synthesis, such as anorexia nervosa. Recently, conclusive evidence of leptin's
134
NANCYKING REAME
significant neuroendocrine role in reproduction was provided in a study of leptin replacement therapy administered to women with hypothalamic amenorrhea related to eating disorders. When replacement levels of leptin were achieved after 2 weeks, mean LH concentrations rose and LH pulse frequency was restored, followed by an increase in the number of dominant follicles, ovarian volume, and estradiol levels over a period of 3 months (89). Although melatonin controls seasonal reproductive cyclicity in some mammalian species, its role as a pacemaker of the human reproductive axis is controversial. Cagnacci et al. (90,91) speculated that melatonin may play a role in timing diurnal LH modifications, as melatonin administration enhanced basal daytime LH secretion, LH pulse amplitude, and responses to a low-dose GnRH challenge during the follicular phase, but not in the luteal phase. However, others (92) have not found a consistent melatonin effect on follicular phase LH levels. In addition, the circadian rhythm of melatonin secretion does not change significantly across the menstrual cycle, and supraphysiologic melatonin concentrations did not decrease the midcycle LH surge response (93-95). In men, long-term administration of daily lowdose supplements of melatonin has no effect on a number of pituitary and gonadal hormones (96). 1. SEASONALVARIATION
Seasonal variability in pulsatile LH secretion was suggested by the findings of Martikainen et al. (97), who studied normal volunteers in Finland for 6 daytime hours of the mid-follicular phase during peak differences in seasonal daylight (December versus May). Although mean concentrations, pulse frequency, and amplitude of both LH and FSH did not differ by season, the area under the curve was significantly higher during the winter. A seasonal effect has also been observed for the timing of the LH surge, with ovulation more likely to occur in the morning during the spring and in the evening during autumn and winter (98). Levels of nighttime LH have been shown to be higher in the summer at mid-cycle, at a time when the nocturnal rise in melatonin was reduced (99).
D. FSH Chronobiology Whether FSH secretion changes over a 24-hour period remains controversial. When measured every 15 minutes, a robust circadian rhythm has been described in young women, which is diminished after menopause (100). In that study, a cosine rhythm with a nighttime decline in transverse mean FSH was observed in the early and late follicular phase despite no evidence of circadian rhythmicity in LH or estradid. Although diminished compared with the early follicular phase, the comparable timing of the FSH acrophase in
the late follicular and mid-luteal phases, and its presence in the postmenopause group, prompted the investigators to propose a central, rather than peripheral, feedback mechanism for the circadian rhythmicity. The authors concluded that their findings provided further evidence for a dissociation in the hypothalamic regulation of pituitary LH and FSH secretion in women. Other studies have failed to confirm a circadian rhythm in FSH (101). Using an analytical technique to define discrete secretory bursts measured at 10-minute intervals over 24 hours, FSH secretion was maximal during the late follicular phase (high estradiol) and in postmenopausal women unrestrained by estrogen (102). Although FSH and LH secretory bursts demonstrated a significant concordance rate of 25%, the relatively high rate of nonconcordance prompted the investigators to propose that distinct mechanisms other than a single releasing hormone probably operate to differentially regulate the secretion of each gonadotropin. Diurnal variations in LH and FSH were not described.
IV. G O N A D O T R O P I N CHANGES DURING THE PERIMENOPAUSE
A. Older Cycling Women The fact that FSH is elevated in normal cycling women over age 40 without concomitant decreases in ovarian steroids would also support the possibility of an aging change in the GnRH signal. Recently, attention has turned from the mechanisms that give rise to the early increase in FSH to the causes of the more subtle alterations in pulsatile LH secretion that lead to the magnified secretory profile of the postreproductive years. Collectively, the data are conflicting. Our studies (103) of daytime pulsatile LH secretion, where blood was sampled every 10 minutes across three phases of the same menstrual cycle, showed a gradual rise in pulse frequency with advancing age (Fig. 9.3). In contrast, in studies by others, LH pulsatile secretion in older cycling women was observed to be similar (104) or reduced (105) compared with younger controls. The use of less frequent sampling (every 20 minutes) in the former study and the use of a less conventional pulse analysis methodology in the latter may in part account for the differences in findings. In our studies, the enhancement of both LH pulse frequency and amplitude occurred prior to any overt reductions in cyclic estradiol or progesterone concentrations and was phase dependant. Using an intensive sampling protocol (every 10 minutes for 8 daytime hours), subjects ages 40 to 50 years demonstrated shorter cycles, higher mean FSH across all 3 study days, and higher mean LH in the follicular and late luteal phase compared with the youngest age group (20 to 34 years). In keeping with earlier findings of a gradual rise in
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition
basal L H levels with age (16), we observed a strong correlation between increasing age and higher transverse mean L H in the follicular phase (r = 0.42, p = 0.008) and late luteal studies (r - 0.60, p --- 0.0001). Unlike previous studies, our ability to document this age-related increase in a sample of only 32 women is probably related to the large number of data points per study (n = 49 over 8 hours) used to calculate mean concentrations. Additionally, the age effects we observed were most evident in the late luteal phase, a time that is typically less represented in daily sampling studies due to the high variability of cycle length in older women. In the women over 40 years, individual secretory patterns of L H and FSH across the menstrual cycle were highly variable. Figure 9.4 depicts examples of individual secretory patterns for L H and FSH (open circles) from three older subjects to highfight the variability present on cycle day 6 of a presumed ovulatory cycle. (Ovulation was presumed based on the presence of a mid-cycle urinary L H surge and progesterone values ranging between 5 and 10 ng/mL when measured every 30 minutes during the mid-luteal (ML) study.) In five subjects, elevated FSH secretion persisted across the cycle in the presence of normal changes in pulsatile L H secretion; five others exhibited a failure to slow L H pulse frequency and increase amplitude in the luteal phase with or without enhanced FSH secretion. The remaining 6 subjects exhibited similar gonadotropin profiles to those observed in the youngest age group. In addition to the cross-sectional cohort, we had the rare opportunity to restudy an individual at multiple time points across the climacteric. Figure 9.5 presents gonadotropin patterns across two ovulatory menstrual cycles of the same subject studied a decade apart. In this particular woman, despite a strikingly similar gonadotropin profile in the follicular phase at both time points, there is a shorter luteal phase by age 45 as evidenced by an earlier decline in progesterone concentrations. The markedly lower basal estradiol values in the later study may account for the failure of FSH to suppress in the luteal phase. The absence of largeamplitude L H pulses in the midluteal phase by age 45 was a common finding in the cross-sectional study.
FIGURE 9.3 Effects of age on pulsatile LH secretory characteristics. Age groups are in years. All subjects were studied acrossthe same menstrual cycle.Foll, earlyfollicular phase;ML, midlutealphase;LL, late luteal phase, *= p < 0.05; ***= p < 0.001. (From ref. 103.)
41.3
39.7
32.3
31.2
38.4
37.1
135
34.1
22.8
17.7
67.3
43.0
62.2 9
40
9
49.5 9
9
55.1 9
52.2
9
9
..I 30
u
"r _J
20
10
Age 43 yrs 0
800
"
'"'
~
1000
'
-
'
9
1200
-
!
1400
-
iI
1600
-
w
1800
Age 45 yrs ~"
800
-
ir
1000
-
"w
1200
-
' J
"'
1400
!
1600
-
/
1800
Age 46 yrs =iF
800
-"
"w
1000
'-
'!
1200
~
-
|
1400
-
~
'""
1600
|
1800
FIGURE9.4 Variationsin dynamicgonadotropinsecretion on cycleday 6 in three ovulatorysubjects overage 40 with similarestradiol concentrations (pg/mL listed above each graph). Open circles, FSH; closedcircles, LH; *, LH pulse.
136
NANCY KING REAME 07 80
,
9,
,
9
,
0.3 70
10.0 250
14.0 225
14.0 174 .
9.0 Prog (ng/mt} 167 E2 (l:,g/ml) ,
9
Age 35
28 day cycle ,,,I
5
915
i Z
tlk
10
14
=, 5-
9
'"
"
""
.....
""
l
-,r
'
l
I'
I"
Oay 4
"
I"
"
I
--
"
Day 19
02 62
0.2 43
14.7 30
"
"
Day 25 14.0 44
2.2 18
0.4 Pro9 (no/ml) 10 E2 (log/ml) :
+
LH
.J
i'z,
.
u~ 1o
.
.
Age 45
.
26 day cycle
UL
14 Z
5
0 900 "
' " '1 3 0 0
1100
; " '
1 O0
1700
'
900
'
1100
; ' "1 5 0 0
1 O0
Day 6
Day 20
;
1 O0
" " '
900
1100
+
1 3' 0 0
"
1 5' 0 0 "
1 ;0 0
"
Day 26
FIGURE 9.5 Longitudinalassessment of gonadotropin secretory patterns in a regularlycyclingwoman studied at a 10-year interval.The x axis is clockhours. Day refers to menstrualcycleday (day 1 = first day of menses). (From Reame NE. Gonadotropin changesin the perimenopause.In: Lobo R, ed. Perimenopause. Springer-Verlag,New York, 1997;165, with permission.)
Taken together, these data suggest that (a) the agerelated increase in FSH concentrations in ovulatory women, although more pronounced, is associated with phasedependent enhancement of pulsatile LH secretion; (b) the higher LH concentrations are brought about by changes in both pulse frequency and amplitude; and (c) these age effects preempt overt reductions in cyclic estradiol or progesterone concentrations. Studies underway in our laboratory are addressing whether the perimenopausal gonadotropin secretion occurs first at night, as in puberty, and if menopause is heralded by a suppression of the sleep-induced changes in pulsatility. Our preliminary data obtained in a small group of ovulatory, older cycling women indicate that LH pulse slowing during sleep persists in the early follicular phase but amplitude fails to rise (106). To what extent the impairment in sleep quality that occurs in middle age, regardless of gender, plays a role in this altered LH response is unclear. We have shown that regardless of the stage of the menopausal transition, middle-aged women demonstrate a significant reduction in sleep efficiency during a hospital sleep challenge compared with young, cycling control subjects (107).
B. Age-Related Oligomenorrhea Transitory episodes of elevated gonadotropins, hypoestrogenism, and irregular cycles have been a common feature observed in studies of hormonal patterns in the perimenopause (18,108). As the menopause approaches, these occasional postmenopausal patterns become more prolonged, with less frequent bouts of gonadal steroid cyclicity. To assess the effect of intermittent follicular activity on the H P O axis, we conducted a series of studies in perimenopausal women. Eligibility criteria included a change in menstrual cycle regularity in the past year, 45 years of age or older, the onset of hot flashes or other estrogen-deficiency symptoms, and a basal serum FSH value of 20 mIU/mL or greater. Figure 9.6 presents data from three 8-hour studies of dynamic LH secretion conducted in the same subject on cycle day 6 (top panel), day 26 (middle panel), and day 33 (bottom panel) of a 43-day menstrual cycle. The typical gonadotropin profile of a postmenopausal woman observed in the first study is dramatically altered during an episode of abnormal follicular development as evidenced by the ultrasound detection of an ovarian cyst.
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition 9.4
50
--I
13.5
0.02
=,
15.3
0.02
0.02
40
,,,,,
30
20
10
-
,
800
-
,
1000
- '"
1200
,
-
,
1400
' -
'l
1600
1800
EF 76.7
83.6
0.02
50
104.9
0.02
0.02
40
3O
20
10
. 800
,
..
1000
. 12;0
,'
9
1400
;
.
16
0
I
1800
ML 216.9
265.8
1.37
50"
244.7
1.00
E2
1.10
(PG/ML)
PROG
(NG/k&)
40
30
2O
I0
9 800
9
.' 1000
-
,
-
1200
l 1400
" -
l 1600
9
'l 1800
LL
FIGURE 9.6 Secretory patterns of LH (closedcircles) and FSH (open circles) in a 52-year-old subject on cycle day 6 (EF, top panel), day 26 (ML, middle panel) and day 33 (LL, bottom panel) of a 43-day menstrual cycle. Steroid values are presented across the top of each panel.
Under the influence of the twentyfold increase in estradiol concentrations, there is a marked suppression of FSH to near normal concentrations. The reversal of the L H / F S H ratio in the presence of reduced LH pulse amplitude and frequency is reminiscent of the normal luteal
137
phase despite the absence of significant progesterone exposure. These data serve to highlight the range of H P O secretory activity present during the perimenopause years and the need for caution when attempting to document menopausal status from a single hormone determination at any given time. A cross-sectional, epidemiologic study of perimenopausal women in Australia revealed much overlap and high variability in single FSH and estradiol determinations for age and BMI-matched women with regular cycles versus those with irregular menstrual function (109). In the SWAN study, urinary LH surges were detected in only about half of women with cyclic changes in estrogens, suggesting loss of hypothalamic/pituitary sensitivity to estrogen feedback (110). Three types of anovulatory patterns were uncovered in the subgroup of 160 participants in whom luteal activity was absent as measured by daily urinary measures of pregnanediol glucuronide over 50 days or one complete menstrual cycle: those with estrogen increases with an LH surge, those with estrogen increases without an LH surge, and those with neither. Because antral follicle counts or measurement of inhibins were not assessed in parallel, it is not known to what extent diminished ovarian reserve may have accounted for these differences.
C. GNRH Stimulation Testing as a Probe of Pituitary Aging Conflicting evidence exists about age effects on pituitary reserve in perimenopausal women when assessed with a GnRH challenge test. An early study demonstrated that the gonadotropin responses to high-dose (100 ug) and low-dose (10 ug) GnRH are similar in hypogonadal women and in women during the early follicular phase but are maximally different during the late follicular phase (111). This difference is presumed to reflect the effect of rising estradiol levels in the young women and, in turn, increasing pituitary reserve relative to pituitary sensitivity to GnRH (112). To determine whether menstrual cycle irregularity during the perimenopause may be related to increased gonadotropin bioactivity, Schmidt et al. (113) performed GnRH challenge tests using a 100-ug dose in the early to mid-follicular phase. The perimenopausal group was compared with young, regular-cycling women, a group of older, cycling women, and a postmenopausal group. Although the perimenopause was associated with magnified gonadotropin levels similar to those of the postmenopausal group at both baseline and after stimulation, only postmenopause women demonstrated increased LH bioactivity. Because E2 in the early to midfollicular phase in the perimenopausal group was not lower than in the young cycling women and there were no group differences in androgens, the authors concluded that
138
NANCY KING REAME
steroidal feedback differences did not explain the enhanced LH biologic/immunological ratio in the postmenopausal years. Gonadal peptides were not assessed. To what extent the 5- to 10-year age difference between the perimenopausal/ postmenopausal groups and the older cycling women may have contributed to the magnified GNRH response and bioactive secretion was not explored. Using a GnRH dose of 25 ug as a measure of near maximal pituitary release capacity (sensitivity), Fujimoto and colleagues (114) compared gonadotropin responses in young and older cycling women in the early follicular phase (Fig. 9.7). They showed that the percentage of change in both FSH and LH was higher in the young versus older group despite similar levels of estradiol and inhibin, suggesting a diminished pituitary responsiveness in the older cohort.
V. THE NEUROREPRODUCTIVE AXIS IN THE POSTMENOPAUSE In women, disproportionately higher levels of FSH versus LH are the norm in the immediate postmenopausal years, with only gradual declines in both gonadotropins occurring after the seventh decade (115). The pattern of pulsatile LH secretion is characterized by relatively high pulse amplitude and a regular frequency similar to that of the follicular phase (116). This unrestrained frequency is believed to be at near maximal rate, as estrogen treatment has only minor enhancing effects on frequency that remain well within the range of the follicular phase (117). In perimenopausal women undergoing oophorectomy, LH and FSH receptor density in ovarian tissue was found to correlate inversely with serum gonadotropin concentrations; in postmenopausal women, where serum levels were highest, neither FSH nor LH receptors in the ovaries were detectable (118). Although the hypothalamic content of GnRH is lower
GnRH Administration
~o
u P<0.05
:
~i
100
0 -100 !
-20
--0
20
40 60 Time (minutes)
80
100
120
FIGURE 9.7 Percent change in mean LH after a GnRH challenge in older and younger women during the early follicular phase of a normal menstrual cycle. (From ref. 114.)
in postmenopausal women (119), GnRH gene expression in the medial basal hypothalamus is increased after menopause without any reductions in area or neuron number compared with that of younger controls (120). Such data suggest that the increase in gene expression is secondary to the ovarian failure and is not the result of nonspecific aging.
A. Dynamic Gonadotropin Secretion The bulk of neuroendocrine studies of postmenopausal women have been undertaken not for the purpose of assessing aging mechanisms per se, but rather to simulate aspects of the underlying LH pulse signal free of the influence of endogenous ovarian hormones--that is, a "castrate" model. Despite the widely held view that LH and FSH secretions after menopause change to a uniform picture of high-amplitude, high-frequency secretion (approximately hourly), estimates of pulsatile LH secretion have varied markedly from study to study, with reported mean pulse frequencies ranging from 60 to 120 minutes (68,116,121-124) and mean LH concentrations ranging from a low of 19 mIU/mL (39) to values exceeding 75 mIU/mL in the two subjects studied by Yen and colleagues (68). Within the same study, individual variability of pulsatility patterns is high. In the nine subjects studied by Couzinet (124), pulse frequency ranged from four pulses per 8 hours to eight pulses per 8 hours, with the range spread equally across the sample. Mean LH concentrations averaged 28.8 IU/L. High intergroup variability in LH pulse patterns was also noted by Rossmanith et al. (116), who reported a mean pulse frequency of four pulses per 8 hours, a frequency similar to the luteal phase, for their group of seven subjects. Thus, these data, when examined closely, demonstrate a level of variability in LH pulsatile secretion similar to that reported across the highly variable menstrual cycle of young women. A number of differences in study design, methodologies, and sample selection may contribute to the variable findings in LH pulse profiles after menopause. Previous studies have used assay methods with varying sensitivities, different criteria to define LH pulses, and heterogeneous study groups. To ensure an adequately depleted ovarian milieu, most studies have required subjects to be at least 2 years postmenopausal and without a current history of estrogen replacement therapy. However, the minimum selection criteria have been interpreted broadly by different investigators: studies have included heterogeneous samples of women ranging in gynecologic age from 2 to 20 years beyond menopause and histories of estrogen replacement therapy as recent as 6 to 8 weeks before study. In addition, the type of menopause (surgical versus spontaneous) has not been controlled for, so that samples ranging in size from 5 to 15 subjects have included both oophorectomized, premenopausal women in their early 40s and spontaneously menopausal women in their 60s, adding
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition further to the heterogeneity of populations. Although it has been assumed that these differences in subject traits and characteristics are benign with respect to the study paradigm, the applicability of the findings to the understanding of the normal transition to the early postmenopause state is unclear. Because its half-life has been estimated to be two- to fourfold shorter, the glycoprotein FAS, although tightly correlated with LH secretion, has been proposed as a more sensitive marker of GnRH secretion at fast pulse frequencies, such as in postmenopausal women (125). FAS may also be more resistant to downregulation than LH. GnRH agonist administration to women after menopause results in persistent elevation of FAS despite suppression of LH levels (126). More importantly perhaps, Hall and colleagues (127) have proposed that the quality of the LH architecture may change as a function of menopause. Disappearance of endogenous LH after GnRH receptor blockade is prolonged in postmenopausal women, compared with young, cycling control subjects. The disappearance of FAS was not altered, suggesting that age differences in LH relate to LH microheterogeneity rather than renal clearance factors (127). Like FAS, uncombined and biologically inactive LH[3 may also be secreted in dissociated free form. Using a specific immunoassay capable of distinguishing the [3 subunit from the dimeric form ofLH, Couzinet and co-workers characterized its activity in a variety of reproductive states and conditions (128,129). Although a high concordance ofLH, LH[3, and FAS pulses was observed in postmenopausal women similar to that seen during the mid-cycle LH surge (129), the regulation of LH[3 and FAS was different. In postmenopausal women treated with GnRH agonist, there was a parallel suppression of LH and LH[3, whereas plasma FAS levels increased, suggesting to the investigators a tighter control of the 13 subunit by pulsatile GnRH than for FAS. How the GnRH pulse generator continues to age throughout the long duration of postmenopausal life has been a subject of more recent interest. Rossmanith and coworkers (130) showed a decline in LH pulse frequency with aging when comparing naturally postmenopausal women of younger (ages 49 to 57) and older ages (78 to 87 years), although other investigators have not (131,132). When Hall and colleagues used a 5-minute sampling protocol, they were able to demonstrate a 35% reduction in FAS pulse frequency between the fifth and eighth decades with aging, providing evidence of slowing of the hypothalamic GnRH pulse generator activity (133). In contrast, this same group of investigators showed that gonadotropin-negative feedback regulation by exogenous steroids is maintained (or even increased) in late postmenopausal women (134). Using a submaximal GnRH antagonist to provide an estimate of the overall amount of endogenous GnRH, Hall and co-workers examined the degree of inhibition of LH as a measure of hypothalamic aging and the effects of sex steroid therapy (pituitary responsiveness) in both
139
younger (45 to 55 years) and older (70 to 80 years) postmenopausal women. Percent inhibition of LH following submaximal GnRH receptor blockade decreased with age, implying an increase in GnRH secretion with age. With estrogen treatment there was an increase in the percentage of inhibition of LH in response to submaximal GnRH receptor blockade, and a further increase with estrogen plus progesterone, indicating a progressive decrease in endogenous GnRH secretion with gonadal steroid feedback. Mean LH and FSH levels were lower at baseline in older compared with younger postmenopausal women. However, the effect of gonadal steroid feedback on endogenous GnRH secretion was similar in younger and older postmenopausal women. FAS pulse frequency was unchanged with E2 administration but decreased dramatically with addition of progesterone in both older and younger postmenopausal women. The investigators concluded that the aging-related decrease in GnRH pulse frequency in concert with the increase in overall amount of GnRH secreted implied that the quantity of GnRH secreted per pulse increases with age. Moreover, gonadal steroid negative feedback effects on the hypothalamic and pituitary components of the reproductive axis appear to be maintained through the eighth decade of life in normal women. Taken together, these studies show an overall increase in gonadotropin concentrations from the premenopausal to the perimenopausal or early postmenopausal period, followed by a reduction in gonadotropin or FAS release from the early to the late postmenopausal period, including lower hormone concentrations and decreased pulse frequency.
B. Gonadotropin Chronobiology after Menopause Although the amplitude of many diurnal rhythms is dampened with aging, such as those of temperature, TSH, melatonin, prolactin, growth hormone (GH), and cortisol (72), it is not clear to what extent these changes reflect a true diminishment of circadian activity versus a reduction in the strength of the environmental cues imposed on the circadian pacemaker. Based on studies in middle-aged rats, it has been proposed that disorders in circadian regulators of the GnRH pulse regulator may contribute to the age-related alterations in gonadotropin secretory profiles (135). In humans, the SCN remains functional and does not decrease in size even with advanced age (136). Studies of aging changes in the diurnal rhythmicity of the gonadotropins have produced conflicting findings. Rossmanith and Lauritzen (83) compared 24-hour pulsatile patterns of LH in postmenopausal women to those of women at three different menstrual cycle phases (Fig. 9.8). In addition to the sleep-entrained rise in LH in the early follicular phase, when secretory profiles were fitted to cosinor
140
NANCY KING REAME
EFP
SLEEP
l
30
~-
9
20
.J lO o
080O
LFP
1300
1700
21 O0
ee
20
0500
SLEEP
' 9
"o
9
9
9~
... ~.
0100
~
~
, 9
e o e
'
0800
.
~
-J lO
MLP
""
SLEEP
1
20
"
10
.J
o
9 0800
A,, 1300
9
9
9
1700
2100
0100
9 0500
~l 0800
SLEEP
PMW 140 120
100 .J
~ so
5
5o 4O
s ,800
9 i300
9 1700
9 2100
9 0100
05;0
0800
FIGURE 9.8 Diurnal variation in LH pulsatile secretion. Studies of women in the early follicular phase (EFP), late follicular phase (LFP), and midluteal phase (MLP) and of a postmenopausal woman (PMW) are presented. Dashed line, cosine function; arrow, acrophase time. Asterisks denote significant pulses. (From
ref. 83.)
functions, diurnal excursions in LH secretion were observed in 14 of the 16 early follicular phase women, 11 of the 14 late follicular phase women, 12 of the 15 mid-luteal phase women, and 7 of the 8 postmenopausal women. However, the postmenopausal women demonstrated the smallest deviation from mesor levels (the value around which the oscillation occurs), and a marked shift was observed in the acrophase time
from early afternoon in the cycle to early morning hours after menopause. This dampened rhythmicity in pulsatile LH secretion after menopause was attributed to the loss of sex steroid sensitivity of opioid action (83). FSH secretion was not assessed. Another study using similar clinical protocol and analytical methods uncovered a blunted circadian rhythm in FSH secretion in postmenopausal women compared with younger cycling control subjects (100). In that study, a cosine rhythm with a nighttime decline in transverse mean FSH was observed in the early and late follicular phase and was markedly attenuated after menopause. In contrast to the findings of the former study (83), these investigators observed no evidence of circadian rhythmicity in LH or estradiol under similar conditions. In a study of 24-hour excursions of gonadotropin secretion in euthyroid, healthy middle-aged women (137), we compared sleep-entrained changes in gonadotropin secretion in three groups of middle-aged women who were ovulatory (OC n = 14; mean age = 45 + 1 year), postmenopausal (n = 9; mean age = 49 + 1), or receiving estrogen replacement therapy after ovariectomy (Ovx n = 11; mean age - 47 + 1). EEG sleep monitoring was performed and percentage of sleep efficiency (percentage of time in bed asleep) did not differ across groups. We modified the conventional circadian pattern analysis method to include an 8-, 12-, and 24-hour period for both sine and cosine analysis so as not to arbitrarily constrain the peak and nadir of the excursions to 12 hours. With this approach, we confirmed a diminished excursion in both LH and FSH circadian curves in the naturally postmenopausal group when compared with a cycling group of women of similar age. In the postmenopausal group, FSH declined with sleep but LH did not. No diurnal differences in mean estradiol were observed in any study group. Because of the sleep-entrained onset of larger but fewer LH pulses in the young women, the presence of a circadian rhythm was modest. The 24hour circadian patterns of the surgically menopausal group were similar to that of the older control subjects in that a significant decline in mean nighttime LH and FSH was seen, due to the absence of large amplitude pulses. These data suggest that estradiol exposure is necessary but not sufficient to maintain the sleep-entrained, circadian profile of LH pulsatility in the follicular phase in young cycling women (Fig. 9.9). The estrogen-specific effect on L H during sleep was also confirmed prospectively in three ovariectomized women studied before and 12 weeks after estrogen replacement: ET suppressed 24-hour transverse mean LH concentrations, with even further reductions during sleep due to pulse slowing, without a compensatory rise in amplitude (138) (Fig. 9.10). A recent experimental study has demonstrated intriguing evidence that these diurnal rhythms in
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition
141
FIGURE 9.9
The diurnal patterns of LH pulsatility observed in the early follicular phase of the menstrual cycle and in a woman receiving estrogen therapy after oophorectomy (right). Estrogen therapy fails to restore the sleep-entrained onset of high-amplitude pulses, resulting in a significant fall in the corresponding mean LH.
(left)
gonadotropin secretion can be abolished when sleep is prevented and constant light and activity is maintained over a 24-hour interval (139). These effects are specific to the HPO axis, because the typical circadian effects on both cortisol and TSH persisted; moreover, no age effects were apparent between younger (45 to 55 years) and older (70 to 80 years) postmenopausal women. Such data suggest that the age-related changes in gonadotropin secretory activity are not due to a dampening of the amplitude of circadian rhythms.
FIGURE 9.10 Estrogen therapy (ET) effects on wake-sleep change in mean LH in three ovariectomized women after 12 weeks of treatment.
C. Ovarian Steroid Activity of the Postmenopausal Ovary Postmenopausal women with intact ovaries have been shown to have 40% greater testosterone levels and 10% greater androstenedione levels than age-matched women who had previously undergone oophorectomy (140,141). Based on classic work by Judd and colleagues (142), who measured ovarian vein and peripheral plasma hormone concentrations before and after ovariectomy, it has been presumed that the follicle-depleted ovary was the major sources of testosterone after menopause. These studies as well as others showing dramatic testosterone suppression after GnRH agonist treatment led Adashi (143) to conclude that the postmenopausal ovary, rather than a defunct endocrine gland in "end-organ failure," is gonadotropin dependent and responsive to LH. This view has recently been challenged by negative findings in studies of androgen-specific steroidogenic enzyme activity in postmenopausal human ovaries, where transcript levels of 17~-hydroxylase (CYP17) were shown to be negligible (144) or undetectable (145). In contrast, when all steroidogenic enzymes necessary for androgen biosynthesis were measured in unison using the more sensitive technique of oligonucleotide microarray analysis, CYP17 was clearly detectable along with other augmentors of the conversion of 21-carbon (C21) steroids to C19 androgens (146). The additional finding of an absence of CYP19 expression led the
142
NANCY KING REAME
investigators to conclude that although the postmenopausal ovary does not synthesize estrogens de novo, it does possess the full array of enzymes necessary to contribute to androgen production via the As steroid synthesis pathway. With the advent of the more sensitive immunofluorometric assays (IFMA), the low concentrations of testosterone normally present in perimenopausal women are now within the range of assay sensitMty, thus making it feasible for the first time to reliably characterize its dynamic secretion. As a further test of the hypothesis that the climacteric ovary is a gonadotropin-responsive, androgen-producing gland, we assessed the relationship between pulsatile LH secretion and episodic release of testosterone (T) in comparison to cortisol secretion in hypogonadal females (147). Figure 9.11 presents plasma concentrations of LH, testosterone, and cortisol sampied over 8 daytime hours from a 50-year-old postmenopausal women with an FSH value greater than 50 mlU/mL and an estradiol level of 5 pg/mL. In this indMdual, testosterone secretion was episodic and the concordance between T and LH pulses was 71% (p < 0.01), but no significant cross-correlation of secretory patterns was observed. In contrast, there was no significant relationship between pulses of T and cortisol, although a strong positive cross-correlation was observed at Postmenopause
0.3
o
30
0
11 ,
0.2
N
E
E r
20
o
E I
0.1
...i
101
O
O 9LHT |
|
n
i
20
0.0 0.3
E
15 0.2
~=
v O
.-~ 10
etll
r..) O
0.1
09
.,
11
,. 13
A
15 ,
F
17 I
0.0
Clock Time FIGURE 9.11 Episodic secretion of testosterone (closed circles) in a postmenopausal woman and its relationship to L H (open circle) and cortisol (F: open triangle). Pulse concordance with LH: p < 0.01. Cross-correlation with cortisol: 5 = 0.58,p < 0.01.
a 0 time lag (r = 0.58; p = < 0.01). These preliminary data suggest that although the adrenal gland may serve as the rhythm generator for basal testosterone secretion, pituitary LH contributes to the pulsatile release ofT in postmenopausal women.
VI. B RAIN AGING AND REPRODUCTIVE SENESCENCE There is currently renewed debate over the relative contribution of the ovaries and the hypothalamic-pituitary unit to the initiation of the human menopause (135). In middleaged rodents, estropause appears to be largely independent of ovarian controls (149). As reviewed by Wise and colleagues (149,150), heterochronic ovarian transplant studies clearly implicate the hypothalamic-pituitary axis as a primary mediator of both the monotropic FSH rise and reproductive aging in the rat (151). Ovulation can be restored in senescent ovaries when transplanted to the kidney capsule of young females (152), and CNS-acting agents can reinitiate estrous and ovulation in aged animals (153). Conversely, although ovarian transplants from young donors to old recipients in the mouse will double the number of cyclic ovulations, they fail to prevent cycle lengthening (154). Taken together, such data support a clear influence of the hypothalamic-pituitary system in the onset of reproductive senescence in rodents. Drawing on data from their work, Phyllis Wise and colleagues (155) have proposed a competing hypothesis on the trigger for the human menopause. It is their view that preemptive aging changes in the brain lead to the alterations in folliculogenesis, gonadal peptide activity, and gonadotropin augmentation. Several lines of evidence from her laboratory suggest that changes in a variety of neurotransmitter systems that regulate GnRH secretion and possibly circadian, diurnal, and ultradian oscillation are altered with age and may contribute to reproductive senescence. The observation that changes in pulsatile LH release can be detected in middleaged rats that showed no deterioration in the regularity of their estrous cycles suggested that aging of the hypothalamic pulse generator occurs early during the transition to acyclicity and may play a causative role in age-related transition. The change in LH pulsatility has been shown to correlate with changes in the diurnal pattern of activity or gene expression of norepinephrine, serotonin, and f3-endorphin. Based on these findings, they propose that multiple pacemakers in the brain are likely to govern the orchestration of complex neurochemical events that give rise to reproductive cyclicity. The accelerated loss of follicles in women after age 35 is proposed to reflect an age-related desynchronization in the rhythmicity of pulsatile GnRH secretion (150). Specifically these investigators postulate that a progressive deterioration of the 24-hour rhythmicity of the biologic clock located
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition in the SCN of the hypothalamus leads to an uncoupling of the coordinated neurosecretory inputs that govern the activity of the GnRH pulse generator. Such insults would lead to a dampening of the GnRH pulse frequency and in turn the preferential increase in the release of FSH over LH. Recently, this hypothesis has been modified in light of new evidence from this laboratory in the rodent model (156) that the integrity of the biologic clock does not deteriorate in a unified manner; instead, aging differentially influences various components of the SCN. Although the 24-hour rhythm in messenger RNA (mRNA) expression of vasoactive intestinal polypeptide (VIP) disappeared by the time animals were middle aged, the mRNA rhythm and content of arginine vasopressin (AVP), a better marker of SCN integrity, was totally unaffected by age. Such data reflect the complexity of the neurorhythmic controls governing reproduction. The idea of derangements in the circadian controls of GnRH secretion as a factor in reproductive aging clearly merits further examination given the growing body of evidence for age-related declines in other neuroendocrine systems mediated by hypothalamic function. For example, studies have demonstrated diminished function of the somatotropic axis of premenopausal women (157,158) and diurnal cortisol secretion in postmenopausal women (72). Prolactin is pulsatile and magnified with sleep in postmenopausal women but dampened overall due to a lower pulse frequency compared with normal-cycling younger women (159). These lines of evidence suggest that subtle aging deficits in hypothalamic function may exist much earlier than previously believed, but to what extent these alterations are relevant to the initiation of menopause remains to be determined. Although nonhuman primates undergo the menopause transition much later in life compared with women (160), the regulatory controls of ovulatory function are strikingly similar (161), and recent studies in monkeys have helped shed light on features of neuroreproductive aging (162,163). Using a push-pull cannula inserted into the hypothalamic portal circulation, Gore and colleagues (163) characterized pulsatile GnRH release and confirmed earlier findings (162) of enhanced LH pulsatility in aging rhesus monkeys after spontaneous menopause. Compared with young monkeys sampled during the early follicular phase, robust GnRH and LH pulses are maintained with an overall elevated pulse amplitude but no change in pulse frequency. These results contrast markedly with those in rats, where a decrease in pulsatile GnRH and LH release occurs with aging in conjunction with a loss of positive feedback to estradiol. Moreover, rats experience reproductive senescence in midlife, whereas rhesus monkeys undergo this process much later in life. However, given striking parallels in several of the neuroendocrine mechanisms controlling reproduction, a recent NIH consensus conference concluded that
143
the nonhuman primate model carries some important advantages for the study of the human menopause (164).
VII. FUTURE STUDIES In summary, there is heightened interest in the role of central aging deficits in the etiology of the menopause. A fundamental question is whether GnRH secretion increases at the time of menopause, and if so, whether this is mediated by declines in the integrity of central circadian pacemakers. This hypothesis provides important new directions for studies of the HPO axis at menopause. An understanding of the factors that interact and initiate the process of hypoestrogenism in aging women is needed to develop strategies for alleviating the negative aspects of the menopause and to better comprehend the process of biologic aging. As reviewed by Yin and Gore (165), studies of aging changes in the GnRH neurosecretory systems of rats, monkey, and humans have demonstrated alterations in GnRH gene expression and the morphology and ultrastructure of the GnRH cell body and neuroterminal. The use of advanced molecular and structural biology techniques may provide us with improved resolution and a broader perspective on the study of GnRH neural activity in the future. Interactions among GnRH neurons, environmental factors, and genes will be fruitful areas for further studies on the mechanisms of reproductive aging. With the continuing development and refinement of highly specific assays it will be possible in the near future to systematically assess the component roles of the inhibins, activins, and follistatin as local and central mediators of aging changes in gonadotropin secretion (Fig. 9.12). Whether antimtillerian hormone will prove to be the best predictor of ovarian reserve will also be known. Such information should greatly add to our understanding of the ovulatory process and the abnormalities associated with premature menopause and infertility. Although seldom reported, it is presumed that the majority of neuroendocrine studies of the menopause transition have been limited to predominantly white, middle-class samples of patients or volunteers. How race, socioeconomic status, and lifestyle factors (e.g., smoking history, diet, body fat characteristics, exercise) may collectively or independently govern the nature and timing of perimenopause events is unknown. In 1992, the National Institute on Aging launched a multisite longitudinal study of the hormonal and systemic effects of the natural menopause transition in AfricanAmerican, Hispanic, Asian-American, and Caucasian women as a way to address the limited information available about determinants of the menopause experience especially for women of color. Now after more than a decade of study, results suggest that there are significant differences across racial and ethnic groups in menopause symptomatology, such as
144
NANCY KING REAME
FIGURE 9.12 Proposed model for aging effects on the HPO axis during the early follicular phase. Age effects at both the level of central pacemakers and the ovary may act together to promote the rise in FSH, enhanced LH pulsatility, and hyperestrogenism. Other potential mediators include NPY (central) and AMH (ovary).
hot flash severity, that are not accounted for by body mass index, smoking, or socioeconomic factors (10,14). To what extent genetic influences on hormone synthesis, secretion, and metabolism are involved in such variation remains to be determined. Finally, Te Velde and Pearson (166) call for a paradigm shift in perspective to better understand the biologic basis of functional aging. They propose combining a systems approach with reductionist methods in order to move from questions of correlational relationships to causal relationships. The challenge to future investigators is to design experimental paradigms with a biologic systems approach in mind for incorporation into a neuroendocrinology model of aging. Such an approach will allow identification of pivotal points in reproductive aging pathways that lend themselves to intervention.
References 1. TreolarAE, BoyntonRE, Behn BG, BrownBW. Variationof the human menstrual cyclethrough reproductivelife. IntJ Ferti11967;12:77-126. 2. Menken J, Trussell J, Larsen U. Age and infertility. Science 1986;26: 1389-1394. 3. Sherman S. Defining the menopause transition. Am J Med 2005; 118(suppl 2):3S-7S. 4. Johnson J, Canning J, Kaneko T, Pru JK, Tilly JL. Cermline stem cells and follicular renewal in the postnatal mammalian ovary.Nature 2004; 428:145-150. 5. Johnson J, Bagley J, Skaznik-Wikiel M, et al. Oocyte generation in adult mammalian ovaries by putative germ cells in bone marrow and peripheral blood. Cell 2005;122:303-315. 6. Soules, MR, Sherman S, Parrott E, et al. Executive summary: stages of Reproductive Aging Workshop (STRAW). Ferti! Steri! 2001;76: 874-878. 7. McKinley SM. The normal menopause transition: an overview. Maturitas 1996;23:137-145.
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition 8. Richardson SJ, Senikas V, Nelson JE Follicular depletion during the menopausal transition: evidence for accelerated loss and ultimate exhaustion at menopause.J Clin Endocrinol Metab 1987;65:1231-1237. 9. McKinlay SM, Brambilla DJ, Posner JG. The normal menopause transition. Am J Human Bio11992;4:37-46. 10. Gold EB, Sternfeld B, Kelsey JL, et al. The relation of demographic and lifestyle factors to symptoms in a. multi-racial/ethnic population of women 40-55 years of age. AmJEpidemio12000;152:463-473. 11. Ushiroyama T, Ikeda A, Ueki M. Evidence for attenuation of gonadotropin pulse frequency in hypergonadotropic women with estradiol secretion in the menopausal transition. PsychoneuroEndocrino11999;24:85-97. 12. Harlow BL, Wise LA, Otto MW, Soares CN, Cohen LS. Depression and its influence on reproductive endocrine and menstrual cycle markers associated with perimenopause: the Harvard Study of Moods and Cycles. Arch Gen Psychiatr 2003;60:29- 36. 13. Santoro N. The menopausal transition. Am J Med 2005;118 (suppl 2):8-13. The NIH State-of-the-Science Conference on Management of Menopause-Related Symptoms, March 21-23, 2005. 14. Randolph JF Jr, Sowers M, Gold EB, et al. Reproductive hormones in the early menopausal transition: relationship to ethnicity, body size, and menopausal status. J Clin Endocrinol Metab 2003; 88:1516-1522. 15. Lee SJ, Lenton EA, Sexton L, Cooke ID. The effect of age on the cyclical patterns of plasma LH, FSH, estradiol and progesterone in women with regular menstrual cycles. Human Reprod 1988;3:851- 855. 16. Ebbiary A, Lenton EA, Cooke ID. Hypothalamic-pituitary aging: progressive increase in FSH and LH concentrations throughout the reproductive life in regularly menstruating women. Clin Endocrinol 1994;41:199-206. 17. Reame NE, Kelch RP, Beitins IZ, et al. Age effects on FSH and pulsatile LH secretion across the menstrual cycle of premenopausal women. J Clin Endocrinol Metab 1996;81:1512-1518. 18. Santoro N. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab 1996;81:1495-1501. 19. Burger HG, Dudley EC, Hopper JL, et al. Prospectively measured levels of serum follicle-stimulating hormone, estradiol, and the dimeric inhibins during the menopausal transition in a population-based cohort of women. J C/in EndocrinolMetab 1999;84:4025-4030. 20. Padmanabhan V, McNeilly AS. Is there an FSH-releasing factor? Reproduction 2001;121:21 - 30. 21. Hanson KR, Thyer AC, Sluss PM, et al. Reproductive aging and ovarian function: is the early follicular phase FSH rise necessary to maintain adequate secretory function in older ovulatory women? Hum Reprod 2005;20:89- 95. 22. Faddy MN, Gosden RG, Gougeon A, Richardson SJ, Nelson JE Accelerated disappearance of ovarian follicles in mid-life: implications for forecasting menopause. Hum Reprod 1992;1342-1346. 23. Gougeon A, Ecochard R, Thalabard JC. Age-related changes of the population of human ovarian follicles: increase in the disappearance rate of non-growing and early-growing follicles in aging women. Biol Reprod 1994;50:653-663. 24. Scheffer GJ, Broekmans FJ, Dorland M, et al. Antral follicle counts by transvaginal ultrasonography are related to age in women with proven natural fertility. Fertil Steri11999;72:845- 851. 25. Ushiroyama T, Yoshikawa M, Saeki M, Okuda K, Sugimoto O. Hypergonadotropinemia with estradiol secretion in peri- and postmenopausal period. Acta Obstet Gynecol &and 1989;68:139-143. 26. Ying SY. Inhibins, activins and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr Rev 1988; 9:267-293. 27. Phillips DJ, de Kretser DM, Padmanabhan V. Follistatin: from puberty to menopause. In: Muttukrishna S, Ledger W, eds. Inhibin, activin and follistatin in human reproductive physiology. London: Imperial College Press, 2001;6141-6164.
145
28. DePaulo LV, Bicsak TA, Erickson GF, Shimasaki S, Ling N. Follistatin and activin: a potential intrinsic regulatory system within diverse tissues. Proc Soc Exp BiolMed 198:500-512, 1991. 29. Lenton EA, de Kretser DM, Woodward AJ, Robertson DM. Inhibiin concentrations throughout the menstrual cycles of normal, infertile, and older women compared with those during spontaneous conception cycles. J Clin Endocrinol Metab 1991;73:1180 - 1190. 30. MacNaughton J, Bangah M, McCloud P, Hee J, Burger H. Agerelated changes in follicle stimulating hormone, luteinizing hormone, estradiol and immunoreactive inhibin in women of reproductive age. Clin Endocrinol (Oxf) 1992;36:339- 345. 31. Klein, N, Illingwortth PJ, Groome NP, et al. Decreased inhibin B secretion is associated with the monotropic FSH rise in older, ovulatory women: a study of serum and follicular fluid levels of dimeric inhibin A and B in spontaneous menstrual cycles. J Clin Endocrinol Metab 1996;81:2742-2745. 32. Soules MR, Battaglia DE, Klein NA. Inhibin and reproductive aging in women. Maturitas 1998;30:193 - 204. 33. Santoro N, Add T, Skurnick, JH. Decreased inhibin tone and increased activin A secretion characterize reproductive aging in women. Fertil Steri11999;71:658-662. 34. Welt CK, McNicholl DJ, Taylor AE, Hall JE. Female reproductive aging is marked by decreased secretion of dimeric inhibin. J Clin Endocrinol Metab 1999;84:105-111. 35. Muttukrishna S, Child T, Lockwood GM, et al. Serum concentrations of dimeric inhibins, activin A, gonadotropins and ovarian steroids during the menstrual cycle in older women. Hum Reprod 2000;15:549-556. 36. Klein NA, Harper AJ, Houmard BS, Sluss PM, Soules MR. Is the short follicular phase in older women secondary to advanced or accelerated dominant follicle development? J Clin Endocrinol Metab 2002; 87:5746-5750. 37. Klein NA, Houmard BS, Hansen KR, et al. Age-related analysis of inhibin A, inhibin B and activin A relative to the intercycle monotropic follicle-stimulating hormone rise in normal ovulatory women. J Clin Endocrinol Metab 2004;89:2977-2981. 38. Watkatsuke M, Shintani Y, Abe M, et al. Immunoradiometric assay for follistatin: serum immunoreactive follistatin levels in normal adults and pregnant women. J Clin Endocrinol Metab 1996;81:630. 39. Harada K, Shintani Y, Sakkamoto Y, et al. Serum immunoreactive activin A levels in normal subjects and patients with various diseases. J Clin Endocrinol Metab 1996;81:2125. 40. Padmanabhan V, Sharma TE Neuroendocrine vs. paracrine control of follicle-stimulating hormone. Arch Med Res 2001;32:533- 543. 41. Bilezikjian LM, Blount AL, Corrigan AZ, et al. Actions of activins, inhibins and follistatins: implications in anterior pituitary function. Clin Exp Pharmacol Physio12001;28:244-248. 42. Reame, NE, Wyman TL, Phillips DJ, deKretser DM, Padmanabhan V. Net increase in stimulatory input resulting from a decrease in inhibin B and an increase in activin A may contribute in part to the rise in follicular phase follicle-stimulating hormone of aging cyclic women. J Clin Endocrinol Metab 1998;83:3302-3307. 43. Robertson DM, Burger HG. Reproductive hormones: aging and the perimenopause. Acta Obstet Gynecol &and 2002;81:612 -616. 44. Klein NA, Battalia DE, WoodruffTK, et al. Ovarian follicular concentrations of activin, follistatin, inhibin, insulin-like growth factor, IGFbinding protein 2, IGFBP-3, and vascular endothelial growth factor in spontaneous menstrual cycles of normal women of advanced reproductive age. J Clin Endocrinol Metab 2000;85:4520-4525. 45. Reame, NE, Zuliani GC, Lukacs J, et al. Circulating levels of activin decrease in response to hormone replacement therapy. Menopause 1998;5:267. 46. Mather JP, Woodruff TK, Krummen LA. Paracrine regulation of reproductive function by inhibin and activin. Proc Soc Exp Biol Med 1992;201:1-15.
146 47. Padmanabhan V, West C. Endocrine, autocrine, and paracrine actions of inhibin/activin/foUistatin on follicle-stimulating hormone. In: Muttukrishna S, Ledger W, eds. Inhibin, activin and follistatin in human reproductivephysiology. London: Imperial College Press, 2001;61-90. 48. Padmanabhan V, Sharma TP. Neuroendocrine vs paracrine control of follicle-stimulating hormone.Arch Med Res 2001;32:533-543. 49. Norman RJ, Milner CR, Groome NP, Robertson DM. Circulating follistatin concentrations are higher and activin concentrations are lower in polycystic ovarian syndrome. Hum Reprod 2001;16:668-672. 50. Reame N, Lukacs J, Olton P, et al. Differential effects of aging on Activin A and its binding protein, follistatin, across the menopause transition. Fertil Steri12007; in press. 51. McConnell DS, Wang Q~ Sluss PM, et al. A two-site chemiluminescent assay for activin-free follistatin reveals that most follistatin circulating in men and normal cycling women is in an activin-bound state. J Clin Endocrinol Metab 1998;83:851- 858. 52. Khoury RH, Wang Q F, Crowley WF Jr, et al. Serum follistatin levels in women: evidence against an endocrine function of ovarian foUistatin. J Clin Endocrinol Metab 1995;80:1361-1368. 53. Martens JW, de Winter JP, Timmerman MA, et al. Inhibin interferes with activin signaling at the level of the activin receptor complex in Chinese hamster ovary cells. Endocrinology 1998;138:2928-2936. 54. Lebrun JJ, Vale WW. Activin and inhibin have antagonistic effects on ligand-dependent heteromerization of the type I and type II activin receptors and human erythroid differentiation. Mol Cell Bid 1997; 17:1682-1691. 55. Visser JA, de Jong FH, Laven JSE, Themmen APN. Anti-Miillerian hormone: a new marker for ovarian function Reproduction 2006;131:1-9. 56. Durlinger P, Kramer B, Karels FH, et al. Control of primordial follicle recruitment by anti-Mullerian hormone in the mouse ovary. Endocrinology 1999;140:5789-5796. 57. de Vet A, Laven JS, de Jong FH, Themmen APN, Fauser BC. Antimullerian hormone serum levels: a putative marker for ovarian aging. Fertil Steri12002;77:357- 362. 58. van Rooij IA, Broekmans FJ, Scheffer GJ, et al. Serum antimullerian hormone levels best reflect the reproductive decline with age in normal women with proven fertility: a longitudinal study. Fertil Steril 2005;83:979-987. 59. Cramer DW, Barbieri RL, Fraer AR, Harlow BL. Determinants of early follicular phase gonadotropins and estradiol concentrations in women of late reproductive age. Hum Reprod 2002;17:221-227. 60. Monroe SE, Jaffe RB, Midgley AR. Regulation of human gonadotropins. XIII. Changes in serum gonadotropins in menstruating women in response to oophorectomy.J Clin Endocrinol Metab 1972;34:420-422. 61. Molskness TA, Woodruff TK, Hess DL, Dahl KD, Stouffer RL. Recombinant human inhibin-A administered early in the menstrual cycle alters concurrent pituitary and follicular, plus subsequent luteal, function in rhesus monkeys J Clin Endocrinol Metab 1996; 81:4002- 4006. 62. McLachlan RI, Dahl KD, Bremner WJ, et al: Recombinant human activin A stimulates basal and GnRH-stimulated FSH and LH release in the adult male macaque, Macaca fascicularis. Endocrinology 1989; 125:2787-2789. 63. Messinis IE, Templeton AA. Evidence that gonadotrophin surgeattenuation factor exists in man. J Reprod Ferti11991;92:217- 223. 64. Fowler PA, Sorsa T, Harris WT, Knight PG, Mason HD. Relationship between follicle size and gonadotrophin surge-attenuating factor (GnSAF) bioactivity during spontaneous cycles in women. Hum Reprod 2001;16:1353-1358. 65. Martinez F, Barri PN, Coroleu B, et al. Women with poor response to IVF have lowered circulating gonadotrophin surge-attenuating factor (GnSAF) bioactivity during spontaneous and stimulated cycles. Hum Reprod 2002;17:634-640. 66. Dafopoulos K, Kotsovassilis CG, Milingos S, et al. Changes in pituitary sensitivity to GnRH in estrogen-treated post-menopausal women:
NANCY KING REAME
67.
68.
69. 70.
71.
72.
73.
74. 75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
evidence that gonadotrophin surge attenuating factor plays a physiological role. Hum Reprod 2004;19:1985-1992. Marshall J. Hormonal regulation of the menstrual cycle and mechanisms of ovulation. In: DeGroot LJ and Jameson JL, eds. Endocrinology, vol 3, 5th ed. Philadelphia: Elsevier, 2006;2911-2922. Yen SSC, Tsai CC, Naftolin F, Vandenberg G, Ajabor L. Pulsatile patterns of gonadotropin release in subjects with and without ovarian function. J Clin Endocrinol Metab 1972;34:671-675. Santen RJ, Bardin CW. Episodic luteinizing hormone secretion in man.J Clin Invest 1973;52:2617. Backstrom CT, McNeilly AS, Leask RM, Baird DT. Pulsatile secretion of LH, FSH, prolactin, oestradiol and progesterone during the human menstrual cycle. Clin Endocrinol 1982;17:29. Reame NE, Sauder SE, Kelch RP, Marshall JC. Pulsatile gonadotropin secretion during the human menstrual cycle: evidence for altered frequency of gonadotropin-releasing hormone secretion. J Clin Endocrinol Metab 1984;59:328. Van Cauter E, Copinschi, G. Endocrine and Other Biologic Rhythms. In: DeGroot LJ, Jameson JL, Endocrinology, vol 1, 5 ed. Philadelphia: Elsevier, 2006;341 - 372. Lavoie HB, Marsh EE, Hall JE. Absence of apparent circadian rhythms of gonadotropins and free oL-subunit in postmenopausal women: evidence for distinct regulation relative to other hormonal rhythms. J Biol Rhythms 2006;21:58-67. Perreau-Lenz S, Pevet P, Buijs RM, Kalsbeek A. The biological clock: the bodyguard of temporal homeostasis. ChronobiologyInt 2004;21:1-25. Chappell PE, White RS, Mellon PL. Circadian gene expression regulates pulsatile gonadotropin-releasing hormone (GnRH) secretory patterns in the hypothalamic GnRH-secreting GT1-7 cell line. J Neurosci 2003 ;23:11202 - 11212. Kapen S, Boyar R, Hellman L, Weitzman ED. The relationship of luteinizing hormone to sleep in women during the early follicular phase: effects of sleep reversal and a prolonged 3 hour sleep-wake schedule. J Clin Endocrinol Metab 1976;43:1-31, 37. Soules MR, Steiner RA, Cohen NL, Bremner WJ, Clifton DK. Nocturnal slowing of pulsatile luteinizing hormone secretion in women during the follicular phase of the menstrual cycle. J Clin Endocrinol Metab 1985;61:43-49, 1985. Rossmanith WG, Yen SSC. Sleep-associated decrease in LH pulse frequency during the early follicullar phase of the menstrual cycle: evidence for an opioidergic mechanism. J Clin Endocrinol Metab 1987; 67:713-718, 1987. Rossmanith WG, Mortola JF, Yen SC. Effects of dopaminergic blockade on the sleep-associated changes of LH pulsatility in early follicular phase women. Neuroendocrinology 1988;48:634-639. Yen SSC, Q.uigley ME, Reid RL, et al. Neuroendocrinology of opioid peptides and their role in the control of gonadotropin and prolactin secretion. Am J Obstet Gyneco11985;152:485-493. Rossmanith WG, Boscher S, Kern W, Fehm HL. Impact of sleep on the circadian excursions in the pituitary gonadotropin responsiveness of early follicular phase women. J Clin Endocrinol Metab 1993;76: 330-336. Hall JE, Sullivan JP, Richardson GS. Brief wake episodes modulate sleep-inhibited luteinizing hormone secretion in the early follicular phase. J Clin Endocrinol Metab 2005;90:2050-2055. Rossmanith WG, Lauritzen C. The luteinizing hormone pulsatile secretion: diurnal excursions in normally cycling and postmenopausal women. GynecolEndocrino11991;5:249-265. Greenspan SL, Klibanski A, Schoenfeld D, Ridgway EC. Pulsatile secretion of thyrotropin in man. J Clin Endocrinol Metab 1986; 63:661-668. Sassin JF, Frantz AG, Weitzman ED, Kapen S. Human prolactin: 24-hour pattern with increased release during sleep. Science 1972;177: 1205-1207.
CHAPTER 9 Neuroendocrine Regulation of the Perimenopause Transition 86. Bluher S, Mantzoros CS. The role ofleptin in regulating neuroendocrine function in humans. JNutr 2004;134:2469S-2474S. 87. Kalsbeek A, Fliers E, Romijn JA, et al. The suprachiasmatic nucleus generates the diurnal changes in plasma leptin levels. Endocrinology 2001 ;142:2677- 2685. 88. Licinio J, Negrao AB, Mantzoros C, et al. Synchronicity of frequently-sampled, 24-h concentrations of circulating leptin, luteinizing hormone and estradiol in healthy women. ProcNatlAcad Sci 1998; 95:2541-2546. 89. Welt CK, Chan JL, Bullen J, et al. Recombinant human leptin in women with hypothalamic amenorrhea. N Eng J Med 2004; 351:987-997. 90. Cagnacci A, Paoletti Am, Soldani R, et al. Melatonin enhances the luteinizing hormone and follicle-stimulating hormone responses to gonadotropin-releasing hormone in the follicular, but not in the luteal menstrual phase. J Clin Endocrinol Metab 1995;80:1095-1099. 91. Cagnacci A, Soldani R, Yen SSC. Exogenous melatonin enhances luteinizing hormone levels of women in the follicular but not in the luteal menstrual phase. Fertil Steri11995;63:996-999. 92. Terzolo M, RevelliA, Guidetti D, et al. Evening administration of melatonin enhances the pulsatile secretion of prolactin but not of LH and TSH in normally cycling women. Clin Endocrino11993:39:185-189. 93. Brzezinski A, Lynch HJ, Seibel MM, et al. The circadian rhythm of plasma melatonin during the normal menstrual cycle and in amenorrheic women.J Clin Endocrinolo Metab 1988;66:891-895. 94. Zimmermann RC, Schroder S, Baars S, Schumacher M, Weise HC. Melatonin and the ovulatory luteinizing hormone surge. Fertil Steril 1990;54:612-618. 95. Berga SL, Yen SS. Circadian pattern of plasma melatonin concentrations during four phases of the human menstrual cycle. Neuroendocrinology 1990;51:606-612. 96. Luboshitsky R, Levi M, Shen-Orr Z, et al. Long-term melatonin administration does not alter pituitary-gonadal hormone secretion in normal men. Hum Refrod 2000;15:60-65. 97. Martikainen H, Ruokonen Ai, Tomas C, Kauppila A. Seasonal changes in pituitary function: amplification ofmidfollicular luteinizing hormone secretion during the dark season. Fertil Steri11996;65:7188- 7120. 98. Testart J, Frydman R, Roger M. Seasonal influence of diurnal rhythms in the onset of the plasma luteinizing hormone surge in women. J Clin Endocrinol Metab 1982;55:374- 377. 99. Kivela A, Kauppila A, Ylostalo P, Vakkuri O, Leppaluoto J. Seasonal, menstrual and circadian secretions of melatonin, gonadotropins and prolactin in women, gcta Physiol Scand 1988;132:321-327. 100. Mortola JF, Laughlin GA, Yen SSC. A circadian rhythm of serum follicle stimulating hormone in women. J Clin Endocrinol Metab 1992; 75:861 - 864. 101. Veldhuis JD, Iranmanesh A, Johnson ML, Lizarralde G. Twenty-four hour rhythms in plasma concentrations of adenohypophyseal hormones are generated by distinct amplitude and/or frequency modulation of underlying pituitary secretory bursts. J Clin Endocrinol Metab 1990;71:1616-1623. 102. Booth RA Jr, Weltman JY, Yankov V-I, et al. Mode of pulsatile FSH secretion in gonadal hormone-sufficient and -deficient women. J Clin Endocrinol Metab 1996;81:3208-3214. 103. Reame NE, Kelch RP, Beitins IZ, et al. Age effects on FSH and pulsatile LH secretion across the menstrual cycle of premenopausal women. J Clin Endocrinol Metab 1996;81:1512-1518. 104. Klein NA, Battaglia DE, Clifton DK, Bremner WJ, Soules MR. The gonadotropin secretion pattern in normal women of advanced reproductive age in relation to the monotropic FSH rise. J Soc Gynecol Invest 1996;3:27-32. 105. Matt DW, Kauma SW, Pincus SM, Veldhuis JD, Evans WS. Characteristics of luteinizing hormone secretion in younger vs older premenopausal women, gm J Obstet Gyneco11998;178:504- 510.
147
106. Reame NE, Lukacs J, Ansbacher R, Chilimigras J, Padmanabhan V. The hypothalamic GnRH pulse generator is altered in ovulatory, premenopausal women: evidence from 24hr pulsatile LH studies. Fertil Steri12002;76:$96 (0255). 107. Lukacs J, Chilimigras J, Cannon JR, Dormire S, Reame NE. Agerelated sleep impairments are independent of ovarian function in midlife women in response to a hospital sleep challenge. J Women's Health 2004;13:333-340. 108. Santoro N, Crawford SL, Allsworth JE, et al. Assessing menstrual cycles with urinary hormone assays. Am J Physiol Endocrinol Metab 2003 ;284:E521 - E530. 109. Burger HG, Dudley EC, Hopper JL, et al. The endocrinology of the menopausal transition: a cross-sectional study of a population-based sample. J Clin Endocrinol Metab 1995;80:3537-3545. 110. Weiss G, Skurnick JH, Goldsmith LT, Santoro N, Park, SJ. Menopause and hypothalamic-pituitary sensitivity to estrogen.JAMg 2004; 292:2991-2996. 111. Wang CF, Lasley BL, Lein A, Yen SSC. The functional changes of the pituitary gonadotrophs during the menstrual cycle. J Clin Endocrinol Metab 1976;42:718- 728. 112. HoffJD, Lasley BL, Wang CF, Yenn SSC. The two pools of pituitary gonadotropin: regulation during the menstrual cycle.J Clin Endocrinol Metab 1977;44:302-312. 113. Schmidt PH, Gindoff PR, Baron DA, Rubinow DR. Basal and stimulated gonadotropin levels in the perimenopause. Am J Obstet Gyneco11996;175:643-650. 114. Fujimoto VY, Klein NA, Battaglia DE, Bremner WJ, Soules MR. The anterior pituitary response to a gonadotropin-releasing hormone challenge test in normal older reproductive-age women. Fertil Steril 1996;65:539-544. 115. Rossmanith WG, Liu CH, Laughlin GA, et al. Relative changes in LH pulsatility during the menstrual cycle: using data from hypogonadal women as a reference point. Clin Enocrinol 1990;32: 647-660. 116. Rossmanith WG, Handke-Vesely A, Wirth U, Scherbaum WA. Does the gonadotropin pulsatility of postmenopausal women represent the unrestrained hypothalamic-pituitary activity? Eur J Endocrinol 1994;130:485-493. 117. Popovic V, Spremovic-Radjenovic S, Eric-Marinkovic J, Grossman A. Effect of sodium valproate on luteinizing hormone secretion in preand postmenopausal women and its modulation by naloxone infusion. J Clin Endocrinol Metab 1996;81:2520-2524. 118. Vihko KK, Kujansuu E, Morsky P, Huhtaniemi I, Punnonen R. Gonadotropins and gonadotropin receptors during the perimenopause. EurJ Endocrinol 1996;134:357- 361. 119. Parker CR, Porter JC. Luteinizing hormone-releasing hormone and thyrotropin-releasing hormone in the hypothalamus of women: effects of age and reproductive status. J Clin Endocrinol Metab 1984; 58:488-491. 120. Rance NE, Uswandi SW. Gonadotropin-releasing hormone gene expression is increased in the medial basal hypothalamus of postmenopausal women. J Clin Endocrinol Metab 1996;81:3540-3546. 121. Veldhuis JD, Evans WS, Rogol AD, Thorner MO, StumpfE Actions of estradiol on discrete attributes of the luteinizing hormone pulse signal in man.J Clin Invest 1987;79:769-776. 122. Cagnacci A, Melis GB, Paoletti AM, et al. Influence of oestradiol and progesterone on pulsatile LH secretion in postmenopausal women. Clin Endocrinol (Oxf) 1989;31:541-550. 123. D'Amico JF, Greendale GA, Lu JK, Judd HL. Induction ofhypothalamic opioid activity with transdermal estradiol administration in postmenopausal women. Fertil Steri11991;55:754-758. 124. Couzinet B, Lahlou N, Thomas G, et al. Effects of gonadotrophin releasing hormone antagonist and agonist on the pulsatile release of gonadotrophins and alpha-subunit in postmenopausal women. Clin Endocrinol (Oxf) 1991;34:477-483.
148 125. Hayes FJ, McNichoU DJ, Schoenfeld D, Marsh EE, Hall JE. Free alpha subunit is superior to luteinizing hormone as a marker of gonadotropin-releasing hormone despite desensitization at fast pulse frequencies. J Clin EndocrinolMetab 1999;84:1028-1036. 126. Oppenheim DS, Bikkai H, Crowley WF Jr, Klibansld A. Effects of chronic GnRH analogue administration on gonadotrophin and alpha-subunit secretion in post-menopausal women. Clin Endocrinol 1992;36:559-564. 127. Sharpless JL, Supko, JG, Martin KA, Hall JE. Disappearance of endogenous luteinizing hormone is prolonged in postmenopausal women. J Clin EndocrinolMetab 1999;84:688-694. 128. Chanson P, Pantel J, YoungJ, et al. Free luteinizing-hormone [3-subunit in normal subjects and patients with pituitary adenomas. J Clin Endocrinol Metab 1997;82:11397-11402. 129. Couzinet B, Pantel J, Chanson P, et al. Measurement of plasma free luteinizing hormone g-subunit in women J Clin Endocrinol Metab 2000; 85:2293 - 2296. 130. Rossmanith WG, Scherbaum WA, Lauritzen C. Gonadotropin secretion during aging in postmenopausal women. Neuroendocrinology 1991;54:211-218. 131. Lambalk CB, deBoer L, Schoute E, Popp-Snyders C, Schoemaker J. Postmenopausal and chronologic age have divergent effects on pituitary and hypothalamic function in episodic gonadotropic secretion. Clin Endocrino11997;46:439-443. 132. Santoro N, Banwell T, Tortoriello D, et al. Effects of aging and gonadal failure on the hypothalamic-pituitary axis in women.AmJ Obstet Gyneco11998;178:732-741. 133. Hall JE, Lavoie HB, Marsh EE, Martin KA. Decrease in gonadotropinreleasing hormone (GnRH) pulse frequency with aging in postmenopausal women. J Clin EndocrinolMetab 2000;85:1794-1800. 134. Gill S, Sharpless JL, Rado K, Hall JE. Evidence that GnRH decreases with gonadal steroid feedback but increases with age in postmenopausal women. J Clin EndocrinolMetab 2002;87:2290-2296. 135. Wise PM, Krajnak KM, Kashon ML. Menopause: the aging of multiple pacemakers. Science 1996;273:67-70. 136. Swab DE Aging of the human hypothalamus. Horm Res 1995;43:8-11. 137. Reame NE, Lukacs J, Ansbacher R, Carlson N, Padmanabhan V. Circadian patterns in gonadotropin secretion at menopause: evidence that loss of estrogen, not aging is responsible for the dampened rhythm in both LH and FSH. Menopause 2002;9:515. 138. Reame N, Lukacs J, Chilimigras J, et al. Estrogen restores the nighttime suppression of LH secretion after menopause but has no role in the FSH diurnal rhythm. Proceedings of the Annual Meeting of the Endocrine Society, ENDO 2002, San Francisco, June 21, 2002. 139. Lavoie HB, Marsh EE, Hall JE. Absence of apparent circadian rhythms of gonadotropins and free alpha-subunit in postmenopausal women: evidence for distinct regulation relative to other hormonal rhythms. J Biol Rhythms 2006;21:58-67. 140. Laughlin GA, Barrett-Connor E, Kritz-Silverstein D, Von Muhlen D. Hysterectomy, oophorectomy, and endogenous sex hormone levels in older women: the Rancho Bernardo Study.J Clin EndocrinolMetab 2000;85:645-651. 141. Davison SL, Bell R, Donath S, Montalto JG, Davis SR. Androgen levels in adult females: changes with age, menopause, and oophorectomy. J Clin EndocrinolMetab 2005;90:3847-3853. 142. Judd HL, Lucas WE, Yen SSC. Endocrine function of the postmenopausal ovary: concentrations of androgens and estrogens in ovarian and peripheral vein blood. J Clin Endocrinol Metab 1974;39: 1020-1024. 143. Adashi EY. The climacteric ovary as a functional gonadotropindriven androgen-producing gland. Fertil Steri11994;62:20-27. 144. Couzinet B, Meduri G, Lecce MG, et al. The postmenopausal ovary is not a major androgen-producing gland. J Clin Endocrinol Metab 2001 ;86:5060- 5066.
NANCY KING REAME 145. Jabara S, Christenson LK, Wang CY, et al. Stromal cells of the human postmenopausal ovary display a distinctive biochemical and molecular phenotype. J Clin EndocrinolMetab 2003;88:484-492. 146. Havelock JC, Rainery WE, Bradshaw KD, Carr BR. The postmenopausal ovary displays a unique pattern of steroidogenic enzyme expression. Hum Reprod 2006;21:309- 317. 147. Reame NE, Rolfes-Curl A, Foster C, Padmanabhan V. Testosterone secretion during the perimenopause: evidence for episodic secretion and responsiveness to pulsatile LH. Menopause 1995;2:247(S-9). 148. Gore AC, Oung T, Yung S, Flagg RA, Woller MJ. Neuroendocrine mechanisms for reproductive senescencein the female rat: gonadotropinreleasing hormone neurons. Endocrine2000;13:315- 323. 149. Wise PM, Scarbrough K, Lloyd J, et al. Neuroendocrine concomitants of reproductive aging. Exp Geronto11994;29:275-283. 150. Wise PM. Menopause and the brain. ScientificAmerican, specialissue on Women'sHealth: a lifelongguide 1998 (summer);9:79-81. 151. Sopelak VM, ButcherRL. Contribution of the ovary versus hypothalamus-pituitary to termination of estrous cycles in aging rats using ovarian transplants. Biol Reprod 1982;27:29. 152. Ascheim E Relation of neuroendocrine system to reproductive decline in female rats. In: Meittes J, ed. Neuroendocrinology of aging. New York: Plenum Press, 1983;73-101. 153. Quadri SK, Kledzik GS, Meites J. Reinitiation ofestrous cycles in old constant-estrous rats by central-acting drugs. Neuroendocrinology, 1973;11:248-255. 154. Nelson JF, Felicio LS. Hormonal influences on reproductive aging in mice. Ann NYAcad Sci 1990;592:8-12. 155. Wise PM. New understanding of the complexity of the menopause and challenges for the future. In: Bellino FL, ed. Proceedings of the International Symposium on the Biology of Menopause. Norwell, MA: Springer, 2000;1- 8. 156. Wise PH, Smith MJ, Dubal DB, et al. Neuroendocrine modulation and repercussions of female reproductive aging. Rec Prog Hormone Res 2002;57:235-256. 157. Wilshire GB, Loughlin JS, Brown JS. Diminished function of the somatotropic axis in older reproductive-aged women. ] Clin Endocrinol Metab 1995;80:608"613. 158. Cano A, Catelo-Branco C, Tarin J. Effect of menopause and different combined estradiol-progestin regimens on basal and growth hormonereleasing hormone-stimulated serum growth hormone, insulin-like growth factor-I, insulin-like growth factor binding protein (IGFBP)-I, and IGFBP-3 levels. Fertil Steri11999;71:261-267. 159. Katznelson L, Risldnd PN, Saxe VC, Klibansld A. Prolactin pulsatile characteristics in postmenopausal women. ] Clin Endocrinol Metab 1998;83:761-764. 160. Shideler SE, Gee NA, Chen J, Lasley BL. Estrogen metabolites and follicle-stimulating hormone in the aged macaque female. Biol Reprod 65:1718-1725, 2001. 161. Wildt L, Hausler A, Marshall G, et al. Frequency and amplitude of gonadotropin-releasing hormone stimulation and gonadotropin secretion in the rhesus monkey. Endocrinology 1981;109:376-385. 162. Woller MJ, Everson-Binotto G, Nichols E, et al. Aging-related changes in release of growth hormone and luteinizing hormone in female rhesus monkeys. J Clin EndocrinolMetab 2002;87:5160- 5167. 163. Gore AC, Windsor-Engnell BM, Terasawa E. Menopausal increases in pulsatile gonadotropin-releasing hormone release in a nonhuman primate (Macaca mulatta). Endocrinology 2004;145:4653-4659. 164. Bellino FL, Wise PM. Nonhuman primate models of menopause workshop. Biol Reprod 2003;68:10-18. 165. Yin W, Gore AC. Neuroendocrine control of reproductive aging: roles of GnRH neurons. Reproduction 2006;131:403-414. 166. Te Velde ER, Pearson PL. The variability of reproductive ageing. Human Reprod Update 2002;8:141-154.
; H A P T E R 1(
Changes in the Menstrual Pattern During the Menopause Transition GEORGINA E. HALE
Department of Obstetrics and Gynaecology,University of Sydney NSW 2006, Australia
IAN S. FRASER
Department of Obstetrics and Gynaecology,University of Sydney NSW 2006, Australia
I. I N T R O D U C T I O N
the older woman. Intermenstrual bleeding is often associated also with postcoital bleeding. Intermenstrual and postcoital bleeding are often associated with important genital tract pathology, and a high index of suspicion is required to ensure that such pathology is not overlooked in this age group (1). Increasing variability in the volume of menstrual blood loss also occurs. In this age group, heavy menstrual bleeding is much more common than it is in the mid-reproductive years, and it may be associated with disturbances of ovulatory or endometrial function or with pelvic pathology. A moderately high index of suspicion may be required to distinguish "normal" patterns during the menopausal transition from those that may indicate pelvic pathology. During the gearing-down phase of the menopausal transition, with variable and unpredictable menstrual patterns, other symptoms of "menopause," such as hot flashes and night sweats, may also develop many months or even years before the menopause finally occurs. The menstrual changes predominantly are due to changes in ovarian steroid secretion consequent on oocyte and follicle depletion (2) but are also influenced by age-related and pathology-related changes in the uterus.
The menopause is strictly defined as the last natural menstrual period a woman will ever experience and hence is only determined in retrospect. Most authorities agree that menopause can be deemed to have occurred when no menstrual bleeding has been noted for 1 year in a woman of the appropriate age, especially if supported by the presence of vasomotor or other symptoms. This permanent cessation of menses is preceded by a variable phase, generally lasting 2 to 5 years, usually called the menopause transition, perimenopause, or climacteric. This phase of progressive gearing down of the ovaries also continues for 2 to 3 years beyond the menopause. The menopausal transition is characterized by increasing irregularity and unpredictability of the menstrual cycle. There is an increase in the incidence of short and long follicular phases, defective ovulation, anovulation, and highly erratic cycles. This pattern of increasing irregularity needs to be distinguished from a pattern of continuing regular menstrual cycles with superimposed episodes of intermenstrual bleeding. Different patterns of intermenstrual bleeding can be quite confusing when they occur in TREATMENT OF THE POSTMENOPAUSAL WOMAN
149
Copyright 9 2007by Elsevier,Inc. All rightsof reproductionin anyformreserved.
150
HALE AND FRASER
The stages of the menopause transition have now been more objectively characterized according to the Stages of Reproductive Aging Workshop (STRAW) criteria (3), which include recognition of increasing cycle irregularity (Table 10.1).
A. M e n s t r u a l P a t t e r n s D u r i n g the M e n o p a u s e T r a n s i t i o n We are indebted to the huge database of prospectively collected, long-term menstrual charts established by Dr. Alan Treloar, through the Tremin Trust at the University of Minnesota, for our present understanding of the changes in menstrual patterns preceding menopause. The initial volunteers for this study were recruited in the 1930s, and numbers were progressively added over the next 40 to 50 years. Volunteers were asked to record the onset of every menstrual period from recruitment until menopause, with noted pauses for pregnancy. This unique database is now supervised by the College of Nursing at the University of Utah. The Treloar program initially reported on the menstrual patterns of 2702 women followed for many years (25,825 total woman-years of menstrual experience) of whom 120 had reached menopause at the time of publication (4). Subsequent longerterm and more detailed reviews of the same cohort provided data on 324 women reaching menopause (5,6). The mean age of menopause in this group was 49.5 years, and the women had been menstruating for an average of 35.9 years. The findings of this study were supported by the large independent menstrual database collected by Vollman (7). The studies of Treloar and Vollman both demonstrate an increasing irregularity of cycles as menopause approaches. This is specifically seen as a sharp increase in cycle variability in 10% to 15% of women 6 years before menopause, which includes increased numbers of both and short and long cycles. There was a sharp increase in
TABLE 10.1
variability in a further 30% of women between 3 and 2 years before menopause, although the majority of women demonstrate little change in most cycles until the last 1 to 2 years. Similar increases in variability precede menopause in Indian women (8), and probably occur in a similar manner in all ethnic groups, although the ages at which change occurs may vary slightly. The mean and centile variations in cycle length during the years leading up to menopause are illustrated in Fig. 10.1. These data demonstrate that the mean cycle length increases from 26 days at 4 years before menopause to 27 days at 2 years before and 28 days at 1 year before the final menstrual period (4). During the same time frame, the 10th percentile of women experience a shortening of cycle length from 21 days through 17 days to 16 days. By contrast, women in the 80th percentile experience an increase from 29 days through 40 days up to 57 days. This figure gives a clear impression of the rapidly increasing intermenstrual intervals in some women as menopause approaches, and also of the contrasting shortening of intervals in others. The consolidated data in Fig. 10.1 do not clearly illustrate the fact that some women will experience highly erratic and unpredictable variations in cycle length during this phase of life (Fig. 10.2). One of the women illustrated in Fig. 10.2 has experienced hugely varying intermenstrual intervals over 9 years, with extreme variations between 26 and 191 days in the 2 years before menopause. The other woman in Fig. 10.2 has exhibited minimal variation except in the last 2 years, when the maximal variations were between 19 and 81 days, with only one cycle longer than 38 days. The data from individual women can be used to calculate the percentage probability of menopause having already occurred according to the duration of amenorrhea at different ages (9) (Table 10.2). For example, a woman who has a first episode of amenorrhea greater than 180 days between the ages of 45 and 49 years has a 45.5% chance of having already
STRAW Criteria from Late Reproductive Age through the Menopausal Transition to the Final Menstrual Period Menopausal transition (perimenopause)
Stage
Late reproductive age
Menstrual cycle
Regular
FSH
Elevated in e a r l y follicular phase
FSH, follicle-stimulatinghormone.
Early
Late
Final menstrual period
Variable cycle length (> 7 days different from normal) Elevated in early follicular phase
Long cycles with episode of amenorrhea (> 60 days) Elevated (at any stage)
Continuing amenorrhea Markedly elevated
CHAPTER 10 Changes in the Menstrual Pattern During the Menopause Transition
151
Im
-ii MF.NS~
INT(RV~.S
! ' ii
" "
;
-
!
---in-i"
....... ....q
I"-. . . . .
1
,, .... , ....
-:: . . . . . :~ ..... i. . . . .
....
"ii~
! "
~'~
~-~.~-i
",--.L..
il 0
| 0
9 i 2
.
1
9 1
4
6
MENSTRUAL YEAR
9 I 210
9 i 22
9 f ~h4.
9 1 ~
,,
1 aO
,,
I 30
~ I C A L
9 I _a=
,
I !!4
AGE
,
L 36.
.
i ~
9 I 40
-0
-6
-4
-2
PR(-MENOPAUSAL YEAR
FIGURE 10.1 Contoursfor the distribution of mean intervals between successivemenstrual onsets at different ages.The right end of the graph shows menstrual intervals in the 9 yearspreceding the onset of menopause. (From ref. 4, with permission.)
reached menopause. On the other hand, she also has a 54.5% chance of having one or more additional menses before menopause supervenes. Others have found that 70% of women experienced significant oligomenorrhea before menopause, whereas 18% experienced heavy or frequent, irregular bleeding, and only 12% had fairly regular cycles up to the onset of postmenopausal amenorrhea (10). The important message is that the actual time of occurrence of menopause is difficult to predict ahead of this event in individual women. The increasingly erratic nature of the menstrual cycle as menopause approaches is associated with an increasing incidence of shortened follicular phases, lengthened follicular phases, defective luteal function, and anovulation (7,11-13). Data from a large cross-sectional basal body temperature chart study demonstrated that the incidence of defective luteal function increases from 8% of cycles at 31 to 35 years to 36% at 40 to 50 years of age (10). This study also demonstrated that anovulation increases from 8% at 31 to 35 years up to 16% of cycles at 45 to 50 years, and the increase is mainly noticeable in the late menopausal transition. This information is mirrored by the demonstration of a marked increase in the incidence of anovulation associated with cystic glandular hyperplasia of
the endometrium in older women, peaking at the age of around 50 years (14,15). This condition could be regarded as one end of the spectrum of anovulatory dysfunctional uterine bleeding, and up until recently, it was graced with the specific name of m e t r o p a t h i a hemorrhagica. A dramatic decline in fertility precedes menopause by about 10 years but this does not correlate directly with any changes in the menstrual pattern (13). Individuals may still be fertile during periods of considerable irregularity, and spontaneous pregnancies, albeit rare, can occur after the age of 50. Indeed, the longstanding substantiated record of the oldest mother to give birth after a spontaneous conception is Ruth Ellen Kistler at 57 years and 129 days, but the oldest reported mother so far is Rossanna Dalla Corta from Italy, who gave birth in 1994 at the age of 63. It is likely that further births to women in their late 50s and 60s will now occur because of the application of assisted reproductive technologies using donor oocytes (16). The incidence of natural quinquagenarian births varies widely from culture to culture, with the highest known rate in the mid 1970s being in Albania (at 55 per 10,000 births). In these days of assisted reproductive technologies, the majority of births occurring to women in their 50s and 60s
152
HALE AND FRASER
t
~0,
. . . . -'o I , #6007:
I
MIm~clto
,
50,
o
ol
I , . . . . .
~'
,
i
I
I
, . . . . . . I I " I
, I I I
I , '
, ,
. . . . . .
I
40,-
I , . . . . . . .
. . . . . .
t 30,-~',
. . . .
20
9 l
l I I
/
I
/
lOt
L
21
,
,
,
'I
'
'
'
'
'
'
I
I
I
I
I
I
I
.
I
. . . . . .
I I o
-,-
.
l
I-
.
i
"
o
.
l l I
"
I ,
l
. . . . . .
I
- . . . . .
'
I l I o '
"
~
I
a I o
I
.
.
'8 ' I . I I i "1
,;
, ........ I I I I r . . . . . j 25
.z
--I .
.
'
.
I
I
I
~
I
I
l
I I I I
'
2o,
---
I I
"4 ......
!I
i . . . . . . I I I I "1 ,
,
t
I
,
,---
I I I I
. I
I I !
14
,
-'
.
.
.
|
,'l. . . . . . .
|"
.
.
9
u -'"
Ii | ~
,
.
:
f/
.
.
. "
'"
, . . . . . I I I I -1 . . . . . . i "
.
.
.
I I I
.
.
o
.
" ~ " ' - ' " ' - '
, I " I , 'I " r ,
.
I
.
.
I I I
i
,
I
I
i
I
o
I
,
,......
,-
I
I I I I
.
.
. . . . . .
.
-
. . . . . . .
.
I "I I
.
"X
.
9
' I I I I "1 . . . . . . ,
--L~J
i
" - "
~
~o-'l| ':If--"
l
l
I
i ~ l ' . . . . .
-I
'
. . . _ . _
; ....
. . . . . . . . .
.
"~i
.
-.
. . . .
I
I
-.I
.
"
n
. . . . . .
I
i--,-|---
I I I t r . . . . . , 31
I
,,
.
.
.
.
-"
I I I i
I "1 I i
~ . . . . . . ,
9
.
,
fl V:I1 V 8 o I
I o I
I
il' -l-n II
,...... ,
,I'I-''-, ,II
' / I
'
"-
I . I l. . . . . .
I I I
:
9
i
I
I
I I l I
'I
|
I
I o u
/Ij I/I
f l l , V-l-, [ / /
I_
---
u,,,~o.TM. ~
,,. vo'~. ~,,~,:
--
I
',
.
I -,
I ",
I
I
U "
YEAR
,
I I I
.
--
.
11
O
.
. . . . .
, ....... . I . I I . I r----2 , Z7
7/ .
I I I
o '
'
-
I
o .I t
,
I .'o-"
~ I
I
I
,
. . . . . .
I I
..
I
'
......
, o
I o I. . . .
O I I
'.
'.
:
' l [ ,-V ....... I /
' ' ' . . . . . . ' n " ,
i I
I i
o
....
,
I
,--
I
'
I I
I
iI
I
- - - - - ~
.
,
!I
I
io~
I
.
.
, ' ......... I I I I !'. . . . 23
. . . . . .
"--" e ,
,
'
[
......
,
I
,,
I
I , . . . . .
,
T '
,'. . . . . .
50,, . . . . . .
40,
'
I
I ,--I ,
I ,
- -
I ....
o 0. . . . . . I ,
o i
. . . . . .
MENSTRUAL
601 . . . . . .
I , . . . . . . II ,
, . . . . . .
. . . . ,
',
I , . . . . . . l o
. . . . . .
I
. . . .
I I
O ,
' . . . . . .
.
)"
17
I
I
I
U)o
~
I
I
. . . . . .
o I i
i e , - - - - - - ~ - - - - - - i
'.
. . . . .
I
r I
- -16-
, - ~ -
. . . . .
I
O o r . . . . . I
le
a I "1 . . . . . . I
M(NSTRUAL
I r . . .M . . i
! "1 . . . . . .
22
......
_._L__.
24
-_
,
:
j
YEAR
FIGURE 10.2 Menstrual intervals in two women during the 11 years leading up to menopause, demonstrating extreme differences in menstrual experience during the menopausal transition. (From res 4, with permission.)
are now due to the use of donated oocytes from younger women. Nevertheless, flesh corpora lutea have been found at laparotomy even up to 3 years after the age of apparent spontaneous menopause (17). It is clear from the data of WaUace et al. (9) that return of menses may occur after prolonged periods of perimenopausal amenorrhea, and that these cycles may occasionally be ovulatory (12). TABLE 10.2 Percentage Probability of Menopause Having Already Occurred According to Age and to Duration of First Episode of Amenorrhea Age (years) Duration of first amenorrheic episode (days)
45-49
50-52
> 53
90-119 180-209 270-299 360 +
12.8 45.5 74.1 89.5
30.4 65.2 86.3 93.6
47.4 71.9 88.6 95.5
Modified from ref. 9.
B. Postmenopausal Bleeding Episodes The Treloar data demonstrate that even in women over the age of 52 with 1 year of amenorrhea, 4.5% will have at least one more episode of menstruation (9). Vaginal bleeding occurring more than 1 year after the menopause is conventionally defined as postmenopausal bleeding and must be investigated because of an incidence of 10% to 20% of underlying genital tract malignancy. Nevertheless, a majority of such episodes of bleeding occur in the absence of pathology and are probably associated with growth and atresia of an evanescent ovarian follicle and rarely even ovulation.
C. Menstrual Blood Loss The volume of menstrual blood loss is unrealistic to objectively measure in clinical practice, and there are no published longitudinal studies of menstrual blood loss through the perimenopause. Our preliminary data indicate that measured menstrual blood loss does not increase
CHAPTER 10 Changes in the Menstrual Pattern During the Menopause Transition
through the menopause transition when the woman experiences normal ovulatory cycle endocrine patterns. However, bleeding may become heavier when the woman experiences distorted ovarian follicular patterns of increased estradiol secretion. Cross-sectional data are fragmentary but suggest an overall tendency toward an increase in the volume of blood loss as menopause approaches. These cross-sectional studies have demonstrated a trend towards increasing volumes of objectively measured blood loss as women grow older (18,19). In the Swedish study from Gothenburg, the mean monthly volume rose from 28.4 mL at 15 years of age to 62.4 mL at 50 years (18). Although this mean rise could have been accounted for by substantial increases in a small proportion of the total group, Rybo and colleagues demonstrated a highly significant and progressive rise in menstrual blood loss in 33 individual women from 36 mL to 68 mL at intervals over 12 years between the ages of 38 and 50 years (20) (Fig. 10.3). Perception and tolerance play a major role in the reporting of"heaviness" of menstrual bleeding (18,21). A major proportion (more than 50%) of the volume of the menstrual flow is made up of an endometrial transudate rather than whole blood (22).
FIGURE 10.3 Objective measurements of menstrual blood loss in 33 Swedish women studied on three occasions over a 12-year period between the ages of 38 and 50 years. (Modified from ref. 20.)
153
This proportion does not appear to change with age and probably greatly influences women's perception of the absolute volume of their flow. An increasing proportion of women develop excessively heavy menstrual bleeding as they move into the late reproductive years (14,18). The incidence of ovulatory and anovulatory disturbances of bleeding patterns increase during this phase, and heavy bleeding due to pelvic pathology is also more common. Few women with heavy bleeding due to disturbances of endometrial or ovulatory function in this age group are consistently anoxatlatory as implied by American definitions (23), and most have ovulatory cycles from time to time (24). Some women also experience genuinely excessive bleeding, with measured menstrual blood loss in excess of 80 mL per month, when they are placed on treatment with standard regimens of hormone replacement therapy following the menopause (25,26) (Table 10.3).
D. Causes of Menstrual Changes in the Menopause Transition The main physiologic explanation for changes in the regularity of menstrual cycles during the menopause transition is loss of ovarian follicle numbers (27,28). As the follicle pool decreases, the levels of inhibin B produced by the preantral and antral follicle decreases (29,30). This in turn causes an increase in FSH secretion, particularly during the follicle phase of the cycle. High FSH levels cause a disturbance in normal cyclical follicle growth and high and often erratic levels of estradiol (31). As the number of follicles deplete, the less ovarian follicles will be able to respond to high FSH levels, and the more elongated the cycles will become (28). The irregularities in estradiol levels and the inconsistent cyclical progesterone levels contribute to irregularities in bleeding, and there is some evidence that high and prolonged unopposed estradiol secretion is associated with increased menstrual blood loss (32,33). Alternating ovulatory and anovulatory cycles are also likely to contribute to increased menstrual blood loss through increasing the likelihood of disordered endometrial proliferation and altered hemostats mechanisms during menstrual flow (34). Various genital tract pathologies are common in this age group, and the likelihood of finding particular pathology will depend on the nature of the bleeding disturbance (Table 10.4). In one study of 500 perimenopausal women attending a single clinic, 20% gave a history of menorrhagia, metrorrhagia or intermenstrual bleeding; 9% of these had a genital tract malignancy, and 14% had endometrial hyperplasia (10). Q.ginn et al. have reported that 38% of women found to have a premenopausal endometrial carcinoma presented with regular menorrhagia, whereas 29% presented with irregular bleeding and 33% with both irregular and very heavy
154
HALE AND FRASER
TABLE 10.3 Measured Menstrual Blood Loss During Programmed Withdrawal Bleeds in Postmenopausal Women Treated with Three Different Regimens of Combined Sequential Hormone Replacement Therapy Range ofMBL (mL)
Median MBL (mL) Number of women Rees" Sporrong 1a Sporrong 2a
50 23 23
3 mo
6 mo
12 mo
3 mo
6 mo
12 mo
% with MBL > 80 mL
26
23
17 ~ ~
1-313 0-582 0-346
2-256 ~ --
1-106 ~ ~
13 13 13
22
16
~
From refs. 25 and 26. apreparations used: Rees: estradiol valerate 2 mg, levonorgestrel 75 Ixg; Sporrong 1: estradiol valerate 2 mg, levonorgestre150 Ixg; Sporrong 2: estradiol valerate 2 mg, medroxyprogesterone 10 mg.
bleeding (35). Benign pelvic causes of these menstrual symptoms are also very common, with uterine myomata, adenomyosis, endometriosis, and endometrial polyps accounting for at least 50% of cases of menorrhagia. The remainder are due to ovulatory and anovulatory dysfunctional uterine bleeding. The need for precision in diagnosis and evaluation has become increasingly important in recent years with the development of a widening range of options for medical and surgical management (36,37).
E. N e e d for Investigation a n d T r e a t m e n t The key requirement here is to determine when an alteration in the pattern of bleeding in a woman who is the perimenopausal age group is the result of serious pelvic TABLE 10.4 Patterns and Causes of Menstrual Disturbance in the Menopause Transition
Oligomenorrhea, amenorrhea, short cycles, hypomenorrhea: Can be part of the natural changes in pattern and volume of menstruation that occur with declining ovarian function Range of pathologic causes occasionally may be found with these symptoms at this stage of life Intermenstrual Needing (with or without postcoital bleeding): Usually associated with recognizable pelvic pathology (predominantly surface lesions of the genital tract): Endometrial disease--polyps, leiomyomata, endometrifis, carcinoma Adenomyosis or endometriosis Cervical diseasempolyps, cervicitis, ectropion, carcinoma
Abnormally heavy bleeding: Pelvic diseasemleiomyomata, adenomyosis, endometriosis, endometrial polyps, endometrial adenocarcinoma, myometrial hypertrophy, atriovenous malformations, and other rarities Systemic diseasemdisorders of hemostasis, hypothyroidism, systemic lupus erythematosus, rarities Dysfunctional uterine bleeding--anovulatory, ovulatory (acute or chronic) Irregular bleeding: Hypothalamicmpituitary anovulatory disturbances Endometrial or cervical carcinoma
pathology, the most serious, of course, being genital tract cancer. Table 10.4 summarizes the situations where genital tract cancer is a possibilitymthat is to say, almost every type of menstrual bleeding disturbance apart from the typical patterns seen before and during the menopause transition (where shortening of cycles occurs through the late reproductive age, followed by increasingly erratic short and long cycles and ultimately oligomenorrhea). This means that a high index of suspicion needs to be exercised when the bleeding pattern is frequent, prolonged, intermenstrual, or excessively heavy. The following approaches to investigation need to be considered: 9 Good quality transvaginal ultrasound scanning (sometimes with sonohysterography) to define pelvic pathologies involving the endometrium, myometrium, or ovaries (see Table 10.4). 9 Hysteroscopy and dilation and curettage (or endometrial biopsy) to assess endometrial and endocervical surface lesions 9 Full blood count to exclude significant anemia; other blood tests are only infrequently helpful (there is considerable controversy about the value of a rise in serum follicle-stimulating hormone [FSH] in predicting menopausal transition; if a serum FSH level is to be of any help, it probably needs to be timed on day 2 to 3 of menses) 9 Prospective menstrual charting (may help to define the nature of the menstrual disturbance) Treatment is basically aimed at management of the cause of the menstrual disturbance. Modern techniques have allowed an unprecedented degree of precision in diagnosis, and this continues to improve with ongoing research. This precision should help to ensure that treatment is guided reasonably precisely to the options appropriate for the defined cause. If underlying pathology is confirmed, then active therapy should clearly be directed to the specific cause. This will generally require surgery of some type, but certain pathologies can be managed satisfactorily by
CHAPTER 10 Changes in the Menstrual Pattern During the Menopause Transition medical therapy. This topic is too large for comprehensive review in this chapter. The remaining group of conditions includes those where no specific underlying cause has been detected. These conditions are often poorly defined, and the terminologies vary greatly from one country to another. They generally come under the ambit of terms such as anovulatory dysfunctional uterine bleeding and ovulatory dysfunctional uterine bleeding(38).
The broad approaches to treatment are threefold: 9 Observation. One approach is to keep the patient under
observation but institute no initial active therapy.
155
pattern of flow. Increasing volume of flow is common as women grow older, and this may be perceived by the woman as being abnormal. This symptom commonly causes social distress and concern about cancer and is relatively infrequently associated with progressive development of iron deficiency and anemia. It is a critically important clinical skill to know when to investigate the unpredictable menstrual changes of women during the menopausal transition in order to determine the possible presence of serious underlying pathology. The physician must remain alert to ensure early detection of the 40% to 50% of endometrial adenocarcinomas that manifest before menopause is reached.
9 Medical management. There is limited good-quality evi-
dence concerning management of dysfunctional uterine bleeding in the menopause transition, but much anecdote. Hormonal therapies are used most widely, with the combined oral contraceptive pill being the most popular (in spite of limited good-quality evidence of efficacy [39]). There is increasing evidence for the benefits of the levonorgestrel intrauterine system (Mirena, Schering) in this situation, and this may indeed be the hormonal therapy of choice for many women encountering problems with dysfunctional uterine bleeding through the menopause transition (40). An alternative effective medical therapy for those women with excessively heavy menstrual bleeding through the menopause transition is tranexamic acid (Cyklokapron, Pharmacia), an oral antifibrinolytic preparation (41). 9 Surgical management. This has been the traditional approach to the management of bleeding disturbances in the menopause transition, and hysterectomy carried out by any one of a variety of different routes has been the most widely recommended procedure (42). The issue of conservation or simultaneous removal of the ovaries needs consideration in this situation, and this usually needs to be individualized by discussion with the patient. Alternative surgical approaches may include one of the effective endometrial ablation techniques (43). The general public possesses an increasing awareness of alternative treatments for menstrual disturbances among this age group, and many women are now choosing the L N G - I U S or endometrial ablation in preference to hysterectomy (44,45).
II. C O N C L U S I O N During the menopausal transition, menstrual intervals generally tend to increase, but many cycles show shortened follicular phases or anovulation, especially in the late transition. The most striking feature of the menstrual cycle in women during the menopausal transition is its unpredictability, with erratic and major variations in menstrual intervals, volume, and
References 1. Fraser IS, Petrucco OM. The management ofintermenstrual and postcoital bleeding and an appreciation of the issues arising out of the recent medico-legal case of O'Shea v Sullivan and Macquarie Pathology. ANZJOG 1996;36:67- 73. 2. Richardson SJ, Senikas V, Nelson JE Follicular depletion during the menopausal transition: evidence for accelerated loss and ultimate exhaustion. J Clin Endocrinol Metab 1987;65:1231-1237. 3. SoulesMR, Sherman S, Parrott E, et al. Stages of Reproductive Aging Workshop (STRAW). Fertil Steri12001;76:874-878. 4. TreloarAE, Boynton RE, Behn BG, Brown BW. Variationof the human menstrual cyclethrough reproductive life. IntJFerti11967;12:77-126. 5. Treloar AE. Menarche, menopause and intervening fecundity. Hum Bid 1974;46:89-107. 6. Treloar AE. Menstrual cyclicityand the premenopause. Maturitas 1981; 3:49-64. 7. Vollman RE The menstrual cycle. Philadelphia: WB Saunders, 1977. 8. Jeyaseelan L, Antonisamy B, Rao PS. Pattern of menstrual cycle length in South Indian women: a prospective study. Soc Bid 1992;39: 306-309. 9. Wallace RB, Sherman BM, BeauJA, Treloar AE, Schlabaugh L. Probability of menopause with increasing duration of amenorrhoea in middle-aged women. Arn J Obstet Gyneco11979;135:1021-1024. 10. Seltzer VL, Benjamin F, Deutsch S. Perimenopausal bleeding patterns and pathologic findings.JAm WornAssoc 1990;45:132-134. 11. Doring GK. The incidence ofanovular cycles in women.JReprod Fertil 1969; 6(suppl):77- 81. 12. Sherman BM, Korenman SG. Hormonal characteristics of the menstrual cycle throughout reproductive life.J Clin Invest 1975;55:699-706. 13. Burger HG, Robertson DM, Baksheev L, et al. The relationship between the endocrine characteristics and the regularityof the menopause transition. Menopause 2005;12:267-274. 14. Schr6der R. Endometrial hyperplasia in relation to genital function. Arn J Obstet Gyneco11954;68:294-309.
15. Fraser IS, Baird DT. Endometrial cystic glandular hyperplasia in adolescent girls. J Obstet Gynaecol Brit Cwlth 1972;79:1009-1015. 16. Matthews P, ed. The Guinness book of records. Dublin: Guinness Publishing, 1996;57. 17. Novak ER. Ovulation after fifty. Obstet Gyneco] 1970;36:903-910. 18. Hallberg L, Hogdahl AM, Nilsson L, Rybo G. Menstrual blood loss--a population study.Acta Obstet Gynecol &and 1966;45:320-351. 19. Cole SK, BillewiczWZ, Thomson AM. Sources of variation in menstrual blood loss.J Obstet GynaecolBrit Cwlth 1971;78:933-939. 20. Rybo G, Leman J, Tibblin E. Epidemiology of menstrual blood loss. In: Baird DT, Michie EA, eds. Mechanism of menstrual bleeding. New York: Raven Press, 1983:181-193.
156 21. Fraser IS, McCarron G, Markham R. A preliminary study of factors influencing perception of menstrual blood loss volume. Am J Obstet Gyneco11984;149:788- 793. 22. Fraser IS, McCarron G, Markham R, Resta T. Blood and total fluid content of menstrual discharge. Obstet Gyneco11985;65:194-198. 23. Cowan BD. Dysfunctional uterine bleeding: clues to efficacious approaches. In: Alexander NJ, d'Arcangues C, eds. Steroid hormones and uterine bleeding. Washington: AAAS Press, 1992:9-15. 24. Fraser IS, Baird DT. Blood production and ovarian secretion rates of oestradiol-17[3 and estrone in women with dysfunctional uterine bleeding. J Clin Endocrinol Metab 1974;39:564- 570. 25. Rees MCP, Barlow DH. Q.uantitation of hormone replacement-induced withdrawal bleeds. BrJ Obstet Gynaeco11991;98:106-107. 26. Sporrong T, Rybo G, Vilbergson G, Crona N, Mattson LA. An objective and subjective assessment of uterine blood loss in postmenopausal women on hormone replacement therapy. Br J Obstet Gynaecol 1992; 99:399-401. 27. Van Zonneveld P, Scheffer GJ, Broekmans FJ, et al. Do cycle disturbances explain the age-related decline of female fertility? Cycle characteristics of women aged over 40 years compared with a reference population of young women. Hum Reprod 2003;18:495-501. 28. O'Connor KA, Holman DJ, Wood JW. Menstrual cycle variability and the perimenopause. Am J Hum Bid 2001;13:465-478. 29. Welt CK, McNicholl DJ, Taylor AE, HallJE. Female reproductive aging is marked by decreased secretion of dimeric inhibin. J Clin Endocrinol Metab 1999;84:105-111. 30. Burger HG, Cahir N, Robertson DM, et al. Serum inhibins A and B fall differentially as FSH rises in perimenopausal women. [erratum appears in Clin Endocrinol (Oxf) 1998, Oct;49:550]. Clin Endocrinol 1998;47:809-813. 31. Miro F, Parker SW, AspinaU LJ, et al. Origins and consequences of the elongation of the human menstrual cycle during the menopausal transition: the FREEDOM Study.J Clin Endocrin Metab 2004;89:4910-4915. 32. Brown JB, Kellar R, Matthews GD. Preliminary observations on urinary oestrogen excretion in certain gynaecological disorders. J Obstet Gynaecol Brit Empire 1959;66:177-211. 33. Balfinger CB, Browning MC, Smith AH. Hormone profiles and psychological symptoms in perimenopausal women. Maturitas 1987;9:235-251. 34. Ferenczy A. Pathophysiology of endometrial bleeding. Maturitas 2003;45:1-14.
H A L E AND FRASER
35. Q.uinn M, Neale BJ, Fortune DW. Endometrial carcinoma in premenopausal women: a clinico-pathological study. Gynaecol Oncol 1985;20: 298-306. 36. Hickey M, Fraser IS. Mechanisms and management of dysfunctional uterine bleeding. In: Fraser IS, Jansen RPS, Lobo R, Whitehead MI, eds. Estrogens and progestogens in clinicalpractice. London: Churchill Livingstone, 1998;419-436. 37. Cameron IT, Fraser IS, Smith SK, eds. Clinical disorders of the endometrium and menstruation. Oxford: Oxford University Press 1998. 38. Fraser IS, Inceboz US. Defining disturbances of the menstrual cycle. In: O'Brien PMS, Cameron IT, MacLean AB, eds. Disorders of the menstrual cycle. London: RCOG Press 2000:141-152. 39. Fraser IS, McCarron G. Randomised trial of two hormonal and two prostaglandin-inhibiting agents in women with a complaint of menorrhagia. Aust N Z J Obstet Gynaeco11991;31:66-72. 40. Andersson K, Rybo G. Levonorgestrel-releasing intrauterine device in the treatment of menorrhagia. BrJ Obstet Gynaeco11991;97:690-704. 41. Milsom I, Andersson K, Andersch B, Rybo G. A comparison of flurbiprofen, tranexamic acid and a levonorgestrel-releasing intrauterine contraceptive device in the treatment of idiopathic menorrhagia. Am J Obstet Gyneco11991;164:879-883. 42. Clarke A, Black N, Rowe P, Mott S, Howie K. Indications for and outcomes of total abdominal hysterectomy for benign disease: a prospective cohort study. Br J Obstet Gynaeco11995 ;102:611- 620. 43. Lethaby A, Hickey M. Endometrial destruction techniques for heavy menstrual bleeding: a Cochrane review. Hum Reprod 2003;17: 2795-2806. 44. Hurskainen R, Teperi J, Aalto AM. Q.uality of life and cost effectiveness of levonorgestrel-releasing intrauterine system versus hysterectomy for treatment of menorrhagia: a randomised controlled trial. Lancet 2001; 357:273-277. 45. Bourdiez P, Bongers MY, Mol BWJ. Treatment of dysfunctional uterine bleeding: patient preferences for endometrial ablation, a levonorgestrel releasing intrauterine device or hysterectomy. Fertil Steril 2004;82:160-166.
-IAPTER 1~
Decisions Regarding
Tre atme nt D uring the Menopause Transition ALISON C . PECK The Fertility Institutes, Encino, CA 91436 JUDI
L. CHERVENAK Department of Obstetrics and Gynecology,DMsion of Reproductive Endocrinology and Infertility, Montefiore Medical Center, Albert Einstein College of Medicine, New York, NY 10461
N A N E T T E SANTORO
Divisionof Reproductive Endocrinology, Department of Obstetrics, Gynecology& Women's Health, Albert Einstein College of Medicine, Bronx, NY 10461
The U.S. population is aging. We are seeing an increase in the number of elderly people and an improvement in survival at advanced ages (1). As the post-World War II baby boom generation reaches age 65, which will occur between the years 2010 and 2030, the most rapid increase in the elderly population in history is expected to occur. As life expectancy increases, women will spend more of their lives in the postmenopausal period. It is therefore critical to identify and correct risk factors that could adversely affect health and quality of life. The menopausal transition is a stage in a woman's life during which she has an opportunity to reduce her risk factors in order to maximize the quality of the rest of her life. Natural menopause is traditionally defined as 12 consecutive months of amenorrhea. A variety of terms and definitions have been used in medical literature to define the period before menopause when a woman's hormonal milieu is associated with irregular menstrual cycles and increased episodes of amenorrhea. Commonly accepted terminology TREATMENT OF THE POSTMENOPAUSAL W O M A N
for this time has included premenopause, perimenopause, menopausal transition, and climacteric. In July 2001, the Stages of Reproductive Aging Workshop (STRAW) assembled to standardize the staging system for reproductive aging and develop a consensus on the nomenclature for the premenopause (2). With the final menstrual period (FMP) as the anchor point, the staging system encompasses five stages before the FMP and two stages after. Stages - 5 to - 3 are the reproductive interval; stages - 2 to - 1 represent the menopausal transition; and stages + 1 to +2 encompass the postmenopause (2). The revised nomenclature is as follows (Fig. 11.1): 9 Menopause." The anchor point that is defined after 12 months of amenorrhea following the FMP, which reflects a near complete but natural decrease in ovarian hormone secretion. 9 Menopausal transition: Stages - 2 (early) and - 1 (late) encompass the menopausal transition and are defined 157
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
158
PECK E T
by menstrual cycle and endocrine changes. The menopausal transition begins with variation in menstrual cycle length in a woman who has a monotropic folliclestimulating hormone (FSH) rise and ends with the FMP, recognized only after 12 months of amenorrhea. 9 Postmenopause: Stages + 1 (early) and +2 (late) encompass the postmenopause. The early postmenopause is defined as 5 years since the FMP. (The participants agreed that this interval is relevant because it encompasses a further dampening of ovarian hormone function to a permanent level, as well as a phase of accelerated bone loss.) Stage + 1 was further subdivided into segment a, the first 12 months after the FMP, and segment b, the next 4 years. Stage +2 has a definite beginning but its duration varies, because it ends with death. Further divisions may be warranted as women live longer and more information is accumulated. 9 Perimenopause." Perimenopause literally means "about or around the menopause." It begins with stage - 2 and ends 12 months after the FMP. The climacteric is a popular but vague term used synonymously with perimenopause. Generally speaking, the term menopausal transition is preferred over perimenopause and climacteric. Smoking is the greatest independent risk factor for earlier menstrual irregularity and earlier menopause. Smoking causes an earlier menopause by about 1 to 2 years (3,4). Another strong indicator for an earlier age at menopause is a maternal history of early menopause (5). The likelihood of a premature menopause has also recently been shown to vary by ethnicity (6). Premature (before age 40) and early (before age 45) menopause are more common in African-American (1.4%), Hispanic (1.4%), and Caucasian (1%) women and much less common in Chinese (0.5%) and Japanese (0.1%) women. Once a woman older than 45 has had 1 year of amenorrhea, she has less than a 10% likelihood of ever menstruating
AL.
again (7). However, there is no clear-cut transition period from the premenopausal to the postmenopausal state. Cessation of menstrual cyclicity occurs spontaneously at some point during this transition. Regarding the hormonal milieu of women traversing the menopause, Metcalf et al. (8) could not identify any hormonal differences between the irregular cycles of the perimenopausal woman and the immediately postmenopausal woman, except that no detectable progesterone was ever produced after a woman's FMR The preponderance of evidence now suggests that significant changes are occurring in a woman's hormonal environment during the menopausal transition. During the menopausal transition, there is an increase in the proportion of anovulatory cycles. However, the mechanisms responsible for the menopausal transition's anovulation remain unclear. The anovulatory cycles occurring during the menopausal transition appear similar to those occurring in adolescence and may reflect an inability to produce a luteinizing hormone (LH) surge after exposure to estrogen (9). Central changes in the hypothalamicpituitary axis may affect gonadotropin secretion. In the Study of Women's Health across the Nation (SWAN), Weiss and colleagues found a relative hypothalamic-pituitary insensitivity to estrogen in aging women that was manifested by positive and negative feedback mechanisms. Anovulatory cycles with estrogen peaks similar to those that result in LH surges in younger women did not result in LH surges in older reproductive-age women, indicating a lack of estrogen-positive feedback on LH secretion in this older population. In addition, levels of follicular-phase estrogen that cause negative feedback of LH in normal ovulatory women fail to suppress LH secretion in some older women (10). A lack of response to an estradiol challenge with an LH surge in perimenopausal women with dysfunctional uterine bleeding has also been described
FIGURE 11.1 Stages/Nomenclature of Normal Reproductive Aging in Women.
Recommendations of Stages of Reproductive Aging Workshop (STRAW), Park City, UT, July 2001. (Adapted from ref. 2.)
159
CHAPTER 11 Decisions Regarding Treatment During the Menopause Transition (9,11,12). However, abnormalities in ovarian steroid or peptide secretion may also play a role. During the menopausal transition, ovarian function is highly variable. Length and quality of menses varies as anovulatory cycles become more common. In women approaching menopause, Landgren et al. have reported an increasing proportion of cycles with prolonged follicular phases that are either due to delayed ovulation or anovulation (13). Hormone levels may fluctuate widely during this time, and as estrogen levels decrease, the inherent protective effects of estrogen on bone may also decrease. Thus, these hormonal changes associated with aging may have detrimental effects that must be recognized, addressed, and ameliorated whenever possible.
I. C H A N G E S A S S O C I A T E D WITH AGIN G Many of the physiologic changes associated with menopause occur or begin before the last menstrual period (14) and may be associated with somatic aging. Somatic aging is reflected by decreases in somatotropic axis function, adrenal androgen production, and loss of bone mineral density after peak bone mass has been attained. Several hormonal systems have age-related changes that may interact with reproductive aging (9). The most important factor regulating the pace of the menopausal transition is ovarian function. Ovarian follicular depletion is the ultimate causative factor for menopausal transition and menopause. However, the commencement of the menopausal transition may also result from aging-associated changes in other systems, such as the hypothalamus, pituitary, and uterus. Because none of these systems has been fully examined in the human, their contribution to the onset of menopause remains unclear.
II. C H A N G E S IN T H E H O R M O N A L ENVIRONMENT
A. Progestogenic Changes Normal (15-18) and decreased (19,20) corpus luteum production of progesterone has been observed in the menopausal transition. In the Daily Hormone Study, a substudy of SWAN, older women (49 years or older) had lower totalcycle integrated progesterone compared with younger women (ages 18 to 32) (21). Clinically, it would be very helpful to have further clarification regarding progesterone levels in the menopausal transition. If decreased progesterone levels are associated with increased estradiol, then this may predispose women to dysfunctional uterine bleeding and endometrial hyperplasia.
B. Estrogenic Changes Although progesterone is no longer produced after a woman's final menstrual period, there exists a brief time when small amounts of estrogen may still be produced. Metcalf et al. (8) observed that although elevations in FSH and LH are common before the final menses, episodes of significant estrogen production are not uncommon in the first year after the final menstruation. Midcycle estrogen concentrations in the menopausal transition have been shown to be normal or increased (15,19,20,22), whereas androgens have been observed to be normal or decreased, independent of major changes in sex hormone-binding globulin (23,24). Hyperestrogenemia may be a feature of the early menopausal transition, but cycles in the late menopausal transition may have decreased levels of estrogen (9,19). During the menopausal transition, estradiol levels do not gradually decrease but instead fluctuate greatly around the normal range until menopause, when no more responsive follicles are present (25). Thus, as a woman ages, there is not a downward spiral in the estrogenic milieu, but instead, a "roller coaster" in estrogen production (9). This important feature of the menopausal transition is clinically frustrating, because patients may complain of waxing and waning symptoms for which therapy must be customized. It is important to seriously consider a patient's complaints of irregular bleeding because the fluctuations in estrogen levels, associated with periods of hyperestrogenemia, may predispose a woman to endometrial hyperplasia with its potential sequelae. Ultrasound monitoring and biopsy may be necessary in these patients. Finally, fluctuations in estradiol during the menopausal transition may be due to a decrease in the aging ovary's ability to regulate FSH secretion. It has been suggested that decreasing negative feedback of estrogen on the hypothalamic-pituitary axis, reflecting a decrease in the number of ovarian follicles with age, leads to a rise in FSH. However, increases in FSH in normally cycling older reproductive age women are not accompanied by a decrease in estradiol. For this reason, attention has been directed to the concurrent fall in inhibin during this reproductive stage (26).
C. Changes in Inhibin It has been suggested that a cycle day 3 serum FSH is an indirect bioassay of dimeric inhibin production at the level of the granulosa cell (27). Because inhibins are products of granulosa cells, they have been proposed to be menopausal markers and have been used to measure ovarian reserve. Different patterns of circulating inhibins A and B observed during the human menstrual cycle suggest that they may
160
PECK ET AL.
have different physiologic roles. Inhibin A is believed to be a product of the dominant follicle and corpus luteum. Inhibin B is a product of smaller, preovulatory follicles and is the dominant inhibin in the follicular phase of the cycle. Decreased production of both these peptides has been reported in women during the menopausal transition, leading to decreased "restraint" of FSH secretion (28). More recent data by Burger et al. imply that falling follicle numbers with consequent falls in the levels of inhibin B are the primary trigger for the rising FSH level observed in the early menopausal transition. It is not until late in the menopausal transition that inhibin A and estrogen levels fall in relation to the decrease in the number of ovarian follicles (29). Women approaching menopause, studied during the follicular phase of their menstrual cycle, had elevated FSH levels and low inhibin/FSH ratios, providing further evidence supporting the basis for the early increase in serum FSH and decrease in inhibins (13). Furthermore, women with low cycle day 3 serum concentrations of inhibin B have been shown to demonstrate a poorer response to ovulation induction than women with high day 3 inhibin B (28). It also appears that greater circulating amounts of FSH are needed to initiate folliculogenesis in reproductively aged women with "decreased follicular reserve." When all these mechanisms are active, enhanced FSH secretion may cause an "overshoot" of estradiol production and the process of folliculogenesis that results in hyperestrogenemia (9,19). In support of this concept, it appears that women in their 40s are more likely to have naturally occurring twin pregnancies than are younger women, implying that multiple folliculogenesis may be more common in this age group (30). Thus, although reproductive efficiency is markedly decreased in the menopausal transition, hormonal secretion patterns may be occasionally exuberant. The early menopausal transition is therefore conceptualized as an endocrine state of compensated failure.
D. Androgenic Changes The three major sources for circulating androgens during the reproductive period are the ovary, the adrenal cortex, and peripheral conversion of circulating androstenedione and dehydroepiandrosterone (DHEA) to testosterone. The premenopausal ovary produces 25% of circulating testosterone, 60% of circulating androstenedione, and about 20% of circulating DHEA. The adrenal cortex produces 40% of circulating androstenedione, 25% of testosterone, and almost all D H E A and D H E A sulfate (DHEAS). In the postmenopausal period, 50% of circulating testosterone levels result from peripheral conversion of androstenedione (31). Androgen production from the postmenopausal ovary is controversial. In the study of Laughlin et al. plasma
testosterone and androstenedione levels were 40% and 10% lower, respectively, in oophorectomized women than in intact postmenopausal women (32). Although Laughlin et al. and others (32-34) have reported on the ovaries being a critical source of androgens throughout the lifespan of older women, Couzinet and colleagues concluded that the climacteric ovary is not a major androgen-producing gland (35). They observed similar plasma androgen levels in naturally postmenopausal and oophorectomized women. In addition, they performed immunohistochemistry in steroidogenicaUy active cells in postmenopausal and premenopausal ovaries. They found that steroidogenic enzymes mandatory for androgen biosynthesis, P-450 SCC, 3 [3-HSD, P-450 C17, and aromatase, were absent in the cells of postmenopausal ovaries compared with more than 20% in cells of younger control subjects. It is interesting that cross-sectional studies of oophorectomized women in more than one worldwide sample indicate lower androgens, but direct examination of the ovary suggests that the source of these circulating androgens may well not be ovarian. Further research is needed to clarify this apparent discrepancy. Both adrenal and ovarian androgen levels decline after age 20. By age 40, serum androgen levels are approximately half those found at age 20 (36). Most of the marked decrease in circulating C19 steroids and the resulting androgen metabolites occurs between ages 20 to 30 and 50 to 60 years. Smaller changes are seen after age 60 years (37). However, changes in the androgenic environment can be affected by other factors associated with aging. For example, insulin and insulin-like growth factor 1 (IGF-1) can both act as stimulants of androgens by the ovarian stroma and theca tissues. In normally menstruating women, there is a preovulatory increase in intrafollicular and peripheral androgens. At midcycle, peripheral androstenedione and testosterone increase by 15% to 20% (38). Several speculations exist regarding the role of the midcycle rise in androgens. It may help accelerate follicular atresia so that at ovulation, there is a single dominant follicle (39). Alternatively, it may be involved in the stimulation of libido: It has been shown that femaleinitiated sexual activity occurs most often at midcycle (40). If androgens are important in the production of a single dominant follicle and in stimulating libido, then the agerelated decrease in androgens may be associated with the increased incidence of multiple pregnancy and decreased libido that has been reported in older reproductive-aged women. If this association holds true, then androgen replacement may be useful for restoring libido in symptomatic older women. Among the adrenal androgens, DHEAS is most abundant hormone in the body. However, DHEAS is not biologically active unless it is converted to testosterone or estradiol. In the early 20s, D H E A production is maximal. With increasing age, its secretion is greatly decreased. The decrease is accelerated after menopause (25). In the elderly,
CHAPTER 11 Decisions Regarding Treatment During the Menopause Transition concentrations of DHEAS are only about 10% of those in younger persons (41). The age-associated decrease in DHEAS is independent ofcortisol (42). Decreased DHEAS also appears to be independent of reproductive aging and instead represents a somatic aging event. Studies to support this belief still need to be performed (9). Because the adrenal cortex androgens, D H E A and DHEAS, have such low intrinsic biologic activity unless converted to more active androgens, they only recently have been considered to be potentially important in immunocompetence and general well-being (43). Their role in the menopause transition has yet to be fully established. Lasley et al. have noted an increase in D H E A S associated with the late menopausal transition (44). Relationships between DHEAS levels and cardiovascular morbidity and mortality that have been reported for men are not true for women. In women, although higher levels of DHEAS were associated with several major cardiovascular disease risk factors, they were not related to risk of fatal cardiovascular disease (45,46). The role of DHEAS and a rationale for its supplementation in the menopausal transition remains to be elucidated.
E. S o m a t o t r o p i c Axis C h a n g e s Growth hormone (GH), under hypothalamic regulation by growth hormone-releasing hormone (GHRH), is a pulsatile hormone released from the anterior pituitary. Somatostatin, on the other hand, inhibits GH secretion. With aging, there is a decrease in GH secretion. It remains to be elucidated whether the decrease in GH results from increased release of somatostatin, decreased levels of GHRH, decreased sensitivity to GHRH, or a combination of these factors (9). Significant gender differences exist in GH secretion. In women, estrogen appears to play an important role in GH secretion. There is a positive association between estrogen status and GH concentrations. Thus, in a decreased estrogenic environment, such as that found in menopause, there is decreased GH secretion (47). Age itself may be a more important factor affecting concentrations of GH than estrogen alone. Recent studies have shown that decreased somatotropic axis activity is detectable before any changes occur in menstrual cyclicity or evidence of ovarian failure is present. Older, regularly cycling women (ages 42 to 46) secrete less GH in the daytime than do younger, regularly cycling controls (ages 19 to 34). This was found to occur in the older women despite higher estradiol levels on the day of sampling (when compared with their younger control subjects ). Older reproductive-age women had twice the early follicular phase concentration of estradiol as their younger controls (48).
161
Lower IGF-1 levels have been observed to be associated with elevated estradiol and decreased G H levels in older reproductive-age women (48). The mechanism by which changes in IGF-1 and G H affect perimenopausal physiology is not fully understood. Whether or not functional changes in the somatotropic axis and hormonal environment affect sensitivity to insulin remains to be shown.
F. M e t a b o l i c C h a n g e s In many women, emergence of the metabolic syndrome presents concurrent with estrogen deficiency. The metabolic syndrome encompasses insulin resistance, central adiposity, dyslipidemia, hypertension, hypercoagulability, and a proinflammatory state (49). It is not considered one disease entity but rather a constellation of related risk factors that together increase the risk of cardiovascular disease (50). Postmenopausal women have a 60% increased risk of the metabolic syndrome (51), and approximately half of cardiovascular events in women are related to the metabolic syndrome (52). Although the etiology is unknown, many believe the underlying pathophysiology is related to increased visceral obesity and insulin resistance (53). During the menopausal transition, insulin sensitivity decreases, especially when there is weight gain (54-56). Wing et al. noted a direct association between weight gain and insulin resistance in women during the menopausal transition (55). A prospective study of 485 middle-aged women aged 42 to 50 years showed that after 3 years, the average weight gain was 2.25 to 4.19 kg. However, there were no significant differences between the amounts of weight gain in premenopausal versus postmenopausal women (2.07 vs. 1.35 kg, respectively) (55). Although it is commonly believed that women gain weight with menopause, studies have shown that increases in body mass index (BMI) are not independent of normal aging (57). Even so, body fat composition does change across the menopause transition. Estrogen promotes the accumulation of gluteofemoral fat (58), whereas the loss of estrogen with menopause is associated with an increase in central fat (59). In the Melbourne Women's Midlife Health Project, 102 middleaged women were evaluated during the menopausal transition to find that the free testosterone index was the major hormonal change associated with central adiposity (60). Finally, in SWAN, the authors confirmed that changes in menopausal status were not associated with weight gain or increases in weight circumference and found that maintaining or participating in regular exercise can help prevent or diminish these gains (61). Thus, aging has been associated with decreased GH and IGF-1 levels, decreased insulin sensitivity, increased insulin resistance (62), and weight gain (43). It is important for
162
PECK ET AL.
women, especially during the menopausal transition and postmenopausal period, to control their weight to minimize their already age-associated increased risk for diseases such as cardiovascular disease.
G. G o n a d o t r o p i n Level C h a n g e s During the menopausal transition, there is an increase in FSH that has been attributed to a loss of ovarian inhibin. This relationship appears to be supported by available immunoassay data (15,17,63). Although serum levels of FSH progressively increase with age, much overlap exists regarding the level and its association with the timing of menopause. Therefore, although measuring FSH may be useful in an infertility setting, it is a poor predictor of the timing of menopause for any individual woman (23,24). Longitudinal studies have shown that the increase in FSH occurs as early as the early 40s in normal women (64). Along with the elevation in FSH, there is a lesser, but still significant, rise in perimenstrual levels of LH (64). Because the age at which the rise in FSH first appears may not necessarily correlate with onset of the menopausal transition or menopause, monitoring of serum FSH or LH for menopause status has limited usefulness. Gonadotropin receptor level changes have also been noted to occur during the menopausal transition. A Finnish study (65) investigated FSH, LH, and 17[3-estradiol levels in women during the menopausal transition before elective abdominal hysterectomy and salpingo-oophorectomy and measured ovarian FSH and LH receptor content. Higher serum gonadotropin levels were found in women with fewer gonadotropin receptors. Postmenopausal women had no detectable FSH or LH receptor levels. High serum gonadotropin levels in women during the menopausal transition suggest the presence of low or undetectable levels of ovarian gonadotropin receptors. The authors proposed that measurement of gonadotropin receptor levels might be a useful indicator of ovarian status during the menopause transition. No absolute predictors of ovarian function exist to date. The marked variations in excreted FSH and LH during a typical ovulatory cycle indicate that there is no simple measure of the effect of age on ovarian function. As mentioned earlier in the chapter, the inhibins have been used to measure ovarian reserve, but inhibin B does not appear to surpass FSH (66). Attention has recently focused on available, stable markers of ovarian follicular development. Antral follicles and follicles in earlier stages produce mtillerian-inhibiting substance (MIS) or antimtillerian hormone (AMH). van Rooij et al. measured MIS, FSH, inhibin B, and estradiol in 81 women ages 25 to 46, who
were cycling at the first of two time points approximately 4 years apart (67). Fourteen of the 81 women developed irregular menses by the second blood draw, indicating the onset of the menopausal transition. MIS was the hormone most closely associated with the onset of the transition. Thus, although it is difficult to recognize early ovarian failure in the clinical setting (64), MIS may be more reflective of the primordial follicle pool than other previously studied markers of ovarian reserve (68) and a promising predictor for the occurrence of the menopausal transition (67).
H. Cycle L e n g t h C h a n g e s As age increases, there is a significant decrease in length of the follicular phase. Although average follicular phase length in women ages 18 to 24 is 15 ___2 days, the average in women 40 to 44 years is 10 ___4 days (16). This follicular phase shortening appears to result from accelerated folliculogenesis during the menopausal transition. This subsequently causes a 3-day decrease in the intermenstrual interval in most women (19). However, this change in length of the follicular phase is not necessarily continuous. Before the menopausal transition, with increasing age there is a decrease in mean menstrual cycle length. During the menopausal transition, cycle length becomes highly variable (69). Menstrual cycles during the menopausal transition are unpredictably irregular or "irregularly irregular," and there is no apparent orderly progression between the extremes of short and long cycles (69,70). The average follicular phase decrease by 3 to 4 days is clinically useful because it precedes obvious, clinically detectable endocrine changes (16). Patients with very frequent cycles or very heavy bleeding often present as a diagnostic and therapeutic challenge. For many such women, hormonal therapy with low-dose oral contraceptives may be useful. SWAN was the first study to evaluate daily menstrual cycle characteristics of a community-based cohort of women in the early stages of the menopausal transition. Santoro et al. noted important differences in cycle characteristics related to age, body mass index, and ethnicity. As reported previously, older age was associated with greater cycle variability, such as longer and more irregular cycles (71). Interestingly, women from all ethnic groups with BMIs greater than 25 kg/m 2 were less likely to have cycles with evidence of luteal activity and more likely to have longer total cycle lengths, a longer follicular phase length, and a shorter luteal phase length (71). Furthermore, after adjusting for BMI, Chinese and Japanese women had lower whole-cycle estradiol excretion compared with African-American, Caucasian, and Hispanic women, possibly indicating that Chinese and Japanese women are approaching menopause more slowly than other ethnic groups (71).
CHAPTER 11 Decisions Regarding Treatment During the Menopause Transition
III. C O M M O N COMPLAINTS AND HEALTH RISKS ASSOCIATED WITH THE MENOPAUSAL TRANSITION A. Hot Flushes The incidence of hot flushes is about 10% before the menopausal transition. However, during the menopausal transition, the incidence greatly increases, reaching a peak at menopause of about 50%. By about 4 years postmenopause, the incidence decreases to about 20% (72). A populationbased study of subjective hot flush reporting in pre-, peri-, and postmenopausal women revealed that 13% of premenopausal, 37% of perimenopausal, and 62% of postmenopausal women (as well as 15% of women on hormonal therapy) complained of at least one hot flush in the 2 weeks prior to the study. Although FSH levels were higher in the women with at least one hot flush per day, estradiol levels were higher in women with one or no hot flushes per week. These investigators concluded that hot flush frequency was associated with increasing FSH and decreasing estradiol levels (73). In the Melbourne Women's Midlife Health Project, Dennerstein et al. reported that vasomotor symptoms, vaginal dryness, and breast tenderness appeared to be specifically related to hormonal changes in women as they progressed through the menopausal transition (73). Patients often present with a primary complaint of vasomotor symptoms. For these patients, hormonal therapy may ameliorate their symptoms. Choice of hormonal therapy should be customized for each patient. Low-dose oral contraceptives in nonsmoking candidates or standard or lowdose estrogen therapy (pill or patch) are potential choices. Of note, a twofold increase in the overall estimated risk of myocardial infarction is associated with low-dose oral contraceptive use (74). Tanis et al. found that oral contraceptive use in women with co-morbidities, including hypertension, smoking, diabetes, and hypercholesterolemia, are at considerable risk of myocardial infarction (75), and therefore careful selection of appropriate candidates is crucial when initiating therapy.
B. Cardiovascular Changes During the menopausal transition, as women age, their risk for cardiovascular disease increases. In fact, the leading cause of death for women in the United States, beginning at age 40, is cardiovascular disease. After age 50, women have the same rate of cardiovascular disease as 40-year-old men, and eventually they have the same or higher rates as men in older age. After age 50, women have greater rates of hypertension than do men (76,77). The possible sequelae of changes in weight and IGF-1 levels may have great clinical
163
impact because they are predictive of cardiovascular disease (55,56). The menopausal transition presents an opportunity for a woman to mitigate her risk factors for cardiovascular disease (through weight control, diet, and exercise). After menopause, the incidence of coronary heart disease increases, probably secondary to multiple mechanisms. Risk factors for cardiovascular disease include high cholesterol and other alterations in the lipid profile, abnormal glucose tolerance, hypertension, insulin resistance, smoking, and obesity. After the menopause, higher cholesterol, triglycerides, total/high-density lipoprotein cholesterol, insulin levels, and body weight are present (55,56,78,79). Risk factors for cardiovascular disease can be affected by hormonal fluctuations. Estrogen may have cardioprotective effects independent of its effects on lipids, including improved pulsatility index, vasodilation, improved blood flow, and inhibition of atheromatous plaque formation (80-83). Guthrie et al. investigated hormone levels at menopause in the Melbourne Women's Midlife Health Project. In this longitudinal observational study of middle-aged Australianborn women, high BMI, an increase in BMI, low estradiol levels, a decrease in estradiol levels, and high free testosterone levels were associated with increased risk of a coronary event, whereas frequent exercise lowered the risk (84). Lifestyle changes can vastly ameliorate cardiovascular risk factors. These changes include exercise, weight loss, careful diet, blood pressure monitoring, stress reduction, and cessation of smoking. The menopausal transition presents an ideal time for modification of risk factors so that a woman will maximize not only her years during the menopausal transition but also her postmenopausal years.
C. Bone Mineral Density Changes In the menopausal transition, decreased ovarian function is associated with altered calcium metabolism and decreased trabecular bone mass (85). Perimenopausal and postmenopausal women have significant bone loss in all skeletal sites, especially trabecular bone. Sex steroids appear to play an important role in maintaining integrity of the skeleton throughout a woman's life (86). During the menopausal transition, changes in bone mineral density (BMD) are influenced by endogenous estradiol levels. Mean estradiol levels of 69 pg/mL at the femoral neck and 89 pg/mL at the lumbar spine on dual-energy x-ray absorptiometry (DEXA) are the optimal levels for preventing postmenopausal bone loss. Slemenda and colleagues (86) found that premenopausal and postmenopausal bone loss are significantly associated with decreased concentrations of androgens. Hence, the literature supports the use of androgens with estrogen therapy in postmenopausal women to stimulate bone formation and prevent bone loss (87,88). However, recent data show that the absolute level of androgens and changes in
164 these levels have no influence on bone loss in middle-aged women (89). Bone mass measurements, such as those used to predict postmenopausal fracture risk, may also be predictive of traumatic fractures in the menopausal transition. Fractures in women approaching menopause can be weakly but significantly predicted by bone mass quantification (especially of the lumbar spine) using DEXA of the spine and hip. One study involving 1000 perimenopausal women who had screening DEXA noted a 2% incidence of stress fractures in women in the 2 years before screening (90). Traditionally it has been shown that women of different ethnicities have varying degrees of fracture rates and BMD. Caucasian women have been reported to have lower BMD than African Americans (91,92) and higher BMD than Asian women (93). Recent data from SWAN have challenged this belief. In SWAN, after adjusting for ethnic differences in bone size and lifestyle variables, lumbar spine and femoral neck BMD were highest in African Americans, and there were no significant differences among Caucasian, Japanese and Chinese women (94). In women weighing less than 70 kg, lumbar spine BMD was similar in AfricanAmerican, Chinese, and Japanese women and was lowest in Caucasian women, which may explain why Caucasian women have higher fracture rates than other groups (94). The use of urinary bone markers, such as urine C- and N-telopeptides (NTX), hydroxyproline, free deoxypyridinoline, calcium and pyridinium cross-link excretion, and serum bone markers, such as osteocalcin (OC) and bone-specific alkaline-phosphatase, to assess bone turnover is appealing because of its ease of use and noninvasiveness. However, there is no consensus regarding the optimal biochemical markers of bone turnover, and different markers of either osteoblast activity or bone resorption can give qualitatively different results (95). Nevertheless, serum OC and urinary NTX are among the most widely used assays and are more responsive to changes in estrogen status than other markers (96). SWAN evaluated the ethnic variation in bone turnover, employing these markers, in 2313 premenopausal or early perimenopausal women of varying ethnicities. Finkelstein et al. reported serum osteocalcin levels 11% to 24% higher in Caucasians compared with African-American, Chinese, and Japanese women (97). Urinary NTX levels were significantly higher in African Americans and Caucasians than in Chinese women, which is consistent with the findings of other investigators (98,99). Interestingly, SWAN showed that ethnic patterns of adult bone turnover do not parallel ethnic patterns of BMD. Finally, there is significant variation in bone turnover in postmenopausal women in different geographic areas (100), and regional variation in bone turnover independent from ethnicity was documented in SWAN, with higher levels of serum OC and urinary NTX in women from the Midwest and Northeast than in women from California (97). Further
PECK E T
AL.
studies are necessary to establish the role of urinary bone markers in the clinical diagnosis of osteopenia and in the care of women approaching menopause. Prevention of bone loss should be encouraged early in all women. Weight-bearing exercise, calcium and vitamin D supplementation, and hormone therapy should be discussed with all women during and after the menopausal transition. If a patient has not already considered ways to reduce her risk for bone loss, the menopausal transition can serve as an "alarm" so that she can take action to maximize her bone density.
IV. H O R M O N E THERAPY IN THE MENOPAUSAL TRANSITION During the menopause transition, hormone therapy (HT) may reduce symptoms such as hot flushes and difficulty sleeping. One study of 32 women ages 42 to 47 years, with irregular anovulatory cycles and symptoms associated with menopause, involved administration of a 6-month course of transdermal estradiol patches (0.05 mg/day for 21 days) and oral progestogens (10 mg/day for 10 days) (10l). Menopausal symptoms were relieved during therapy; there were decreases in serum levels of FSH and LH and an increase in serum estradiol. After 6 months of therapy, FSH and LH concentrations were significantly lower than they were before HT. If a patient reports hot flushes, difficulty sleeping, or other complaints associated with the menopause transition, H T is a viable option. H T will alleviate symptoms such as hot flushes and thus immediately improve her quality of life and will also reduce her risk factors for osteoporosis and cardiovascular disease, two major causes of morbidity and mortality in the older woman. On the other hand, it is important to advise the patient that standard H T regimens are not adequate for contraception. If a sexually active woman has less than 1 year of amenorrhea before beginning HT, she is at a low but real risk for a possibly unwanted pregnancy. She should be advised and encouraged to consider using other forms of contraception such as barrier methods. Alternatively, very-low-dose (20 tag) ethinyl estradiol-containing oral contraceptives are often appealing and well tolerated, with an excellent safety profile in a nonsmoking, older, reproductive-aged woman. Low-dose oral contraceptives may be safely continued up to menopause in women without co-morbidities that increase their risk for myocardial infarction, as long as they are aware of the risks. When to switch a patient from oral contraceptives to hormone therapy presents a clinical dilemma. At present, there does not exist a simple biochemical test that definitively predicts the onset of menopause. Without conclusive clinical data, it is our policy to prospectively establish a date
CHAPTER 11 Decisions Regarding Treatment During the Menopause Transition
at which oral contraceptive use will be stopped and H T use begun. For most women, age 5 1 - the average age at natural menopause m i s a comfortable age at which to make this transition. This transition should be done in partnership with the patient and must take into account the fact that hormone therapy is not an adequate contraceptive.
V. SUMMARY During the menopausal transition, a woman may present with "irregularly irregular" menses, hot flushes, dysfunctional uterine bleeding, difficulty sleeping, mood changes, and osteoporosis/osteopenia. These and other complaints may result from periods of hyperestrogenemia, normal or hypoestrogenemia, decreased androgen levels, and decreased levels of GH and IGF-1 seen in the menopausal transition. The treatment of women in the menopausal transition may present a clinical challenge secondary to the lack of a neatly organized transition period and to the variation that exists among women and within each woman. Therefore, it is important that we become familiar with the woman in the menopausal transition and her needs. The menopausal transition presents an ideal time for reduction of risk factors that may affect quality of life, not just during the menopausal transition but also for her postmenopausal years.
References 1. Manton KG, Soldo BJ. Dynamics of health changes in the oldest old: new perspectives and evidence. Milbank Q 1985;63:206-285. 2. Soules MR, Sherman S, Parrott, E, et al: Executive summary: Stages of Reproductive Aging Workshop (STRAW). Fertil Steril 2001;76: 874-878. 3. Brambilla D, McKinlay S. A prospective study of factors affecting age at menopause. J Clin Epidemio11989;42:1031. 4. McKinlay S, Bifano N, McKinlay J. Smoking and age at menopause in women. Ann Intern Med 1985;103:350. 5. Cramer DW, Xu H, Barlow BL. Family history as a predictor of early menopause. Fertil Steri11995;64:740. 6. Luborsky JL, Meyer P, Sowers ME Gold EB, Santoro N. Premature menopause in a multi-ethnic population study of the menopause transition. Hum Reprod 2003;18:199-206. 7. Wallace R, Sherman, Bean J, et al. Probability of menopause with increasing duration of amenorrhea in middle-aged women. Am J Obstet Gyneco11979;135:1021. 8. Metcalf MG, Donald RA, Livesey JH. Pituitary-ovarian function before, during and after menopause: a longitudinal study. Clin Endocrinol 1982;17:489. 9. Santoro N. Hormonal changes in the perimenopause. Clin Consult Obstet Gyneco11996;8:2. 10. Weiss G, Skurnick JH, Goldsmith LT, Santoro NF, Park SJ. Menopause and hypothalamic-pituitary sensitivity to estrogen. JAMA 2004;292:2991-2996. Erratum in: JAMA 2005;293:163. 11. Fraser IS, Baird DT. Blood production and ovarian secretion rates of estradiol 17-beta and estrone in women with dysfunctional uterine bleeding. J Clin Endocrinol Metab 1974;39:564.
165
12. Van Look P, Lothian H, Hunter WM, et al. Hypothalamic-pituitary ovarian function in perimenopausal women. Clin Endocrinol 1977;7: 13-31. 13. Landgren B, Collins A, Csemiczky G, et al. Menopause transition: annual changes in serum hormonal patterns over the menstrual cycle in women during a nine-year period prior to menopause.J Clin Endocrinol Metab 2004;89:2763-2769. 14. McKinlay SM. The normal menopause transition: An overview. Maturitas 1996;23:137. 15. Sherman BM, West JH, Korenman SG. The menopausal transition: analysis ofLH, FSH, eatradiol and progesterone concentrations during menstrual cycles of older women. J Clin Endocrinol Metab 1976; 42:629. 16. Lenton EA, Landgren B, Sexton L, et al. Normal variation in the length of the follicular phase of the menstrual cycle: Effect of chronological age. BrJ Obstet Gyneco11984;91:681. 17. Sherman BM, Korenman SG. Hormonal characteristics of the human menstrual cycle throughout reproductive life. J Clin Invest 1975; 55:669. 18. Lenton EA, Sexton L, Lee S, et al. Progressive changes in LH and FSH and LH:FSH ratios in women throughout reproductive life. Maturitas 1988;100:35. 19. Santoro N, Brown JR, Adel T, et al. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab 1996;81:1495. 20. Reyes F1, Winters JS, Faiman C. Pituitarymovarian relationship preceding the menopause: a cross-sectional study of serum FSH, LH, prolactin, estradiol and progesterone levels. Am J Obstet Gynecol 1977;129:557. 21. Santoro N, Lasley B, McConnell D, et al. Body size and ethnicity are associated with menstrual cycle alterations in women in the early menopausal transition: the Study of Women's Health across the Nation (SWAN) daily hormone study. J Clin Endocrinol Metab 2004;89: 2622-2631. 22. Shideler S, DeVane O, Kaira F, et al. Ovarian-pituitary hormone interactions during the perimenopauae. Maturitas 1989;11:331. 23. Longcope C, Baker S. Androgen and estrogen dynamics: relationships with age, weight and menopausal status. J Clin Endocrinol Metab 1993;76:601. 24. Longcope C, Franz C, MoreUo C, et al. Steroid and gonadotropin levels in women during the perimenopausal years. Maturitas 1986;8:189. 25. Speroff L, Glass R, Kase N, eds. Clinicalgynecologic endocrinology and infertility, 5th ed. Baltimore: Williams & Wilkins, 1994. 26. Lee SJ, Lenton, EA, Sexton L, et al. The effect of age on the cyclical patterns of plasma LH, FSH, oestradiol and progesterone in women with regular menstrual cycles. Human Reprod 1988;3:851-855. 27. Seifer DB, Gardiner AC, Lambert-Messserljan G, et al. Differential secretion of dimen inhibin in cultured luteinized granulosa cells as a function of ovarian reserve. J Clin Endocrinol Metab 1996;81: 736-739. 28. Seifer DB, Lambert-Messerlian G, Hogan JW, et al. Day 3 serum inhibin B is predictive of assisted reproductive technologies outcome. Fertil Steri11997;67:110-114. 29. Burger HG, Cahir N, Robertson DM, et al. Serum inhibins A and B fall differently as FSH rises in perimenopausal women. Clin Endocrinol 1998;48:809-813. 30. MacGillivray I. Epidemiology of twin pregnancy. Semin Perinato! 1986;10:4-8. 31. Longcope C. Adrenal and gonadal steroid secretion in normal females. J Clin Endocrinol Metab 1986;15:213. 32. Laughlin GA, Barrett-Connor E, Kritz-Silverstein D, yon Muhlen D. Hysterectomy, oophorectomy, and endogenous sex hormone levels in older women: the Rancho Bernardo Study. J Clin Endocrinol Metab 2000;85:645-651.
1
6
6
P
E
33. Davis S. Androgen replacement in women: a commentary. J Clin Endocrinol Metab 1999:84;1886-1891. 34. Davison S, Bell R, Donath S, Montalto J, Davis S. Androgen levels in adult females; changes with age, menopause and oophorectomy. J Clin EndorinolMetab 2005;90:3847- 3853. 35. Couzinet B, Meduri G, Leece MG, et al. The postmenopausal ovary is not a major androgen-producing gland. J Clin Endocrinol Metab 2001 ;86:5060- 5066. 36. Zumoff B, Strain GW, Miller LK, et al. 24-hour mean plasma testosterone concentration declines with age in normal premenopausal women. J Clin Endocrinol Metab 1995;80:1429. 37. Labrie FF, Belanger A, Cusan L, et al. Marked decline in serum concentrations of adrenal C19 sex steroid precursors and conjugated androgen metabolites during aging. J Clin Endocrinol Metab 1997;82: 2396-2402. 38. Judd LH, Yen S. Serum androstenedione and testosterone levels during the menstrual cycle.J Clin Endocrinol Metab 1973;36:475 39. Mushayandebvu T, Castracane D, Santoro N, et al. Evidence for diminished mid-cycle ovarian androgen production in older reproductive aged women. Fertil Steri11996;65:721. 40. Adams DB, Gold AR, Burr AD. Rise in female-initiated sexual activity at ovulation and its suppression by oral contraceptives. N EnglJ Med 1978;299:1145. 41. Orentrelch N, Brind JL, Rizer RL, et al. Age changes and sex difference in serum dehydroepiandrosterone sulfate concentrations throughout adulthood. J Clin Endocrinol Metab 1984;59:551. 42. Parker LN, Odell WD. Decline of adrenal androgen production measured by radioimmunoasaay of urinary unconjugated dehydroepiandrosterone. J Clin Endocrinol Metab 1978;47:600. 43. Buster JE, Casson PR, Straughn AB, et al. Postmenopausal steroid replacement with micronized dehydroepiandrosterone: preliminary oral bioavailability and dose proportionality studies. Am J Obstet Gynecol 1992;166:1163. 44. Lasley BL, Santoro N, RandolfJF, et al. The relationship of circulating dehydroepiandrosterone, testosterone, and estradiol to stages of the menopausal transition and ethnicity. J Clin Endocrinol Metab 2002; 87:3760-3767. 45. Barrett-Connor E, Goodman-Gruen D. DHEAS does not predict cardiovascular death in postmenopausal women. The Rancho Bernardo Study. Circulation 1995;91:1757-1760. 46. Barrett-Connor E, Goodman-Gruen D. The epidemiology of DHEAS and cardiovascular disease. Ann NYAcad Sci 1995;774:259-270. 47. Ho KY, Evans WS, Blizzard RM, et al. Effects of sex and age on the 24-hour profile of growth hormone secretion in man: Importance of endogenous estradiol concentration. J Clin Endocrinol Metab 1987; 64:51. 48. Wilshire G, Loughlin J, Brown J, et al. Diminished function of the somatotropic axis in older reproductive-aged women. J Clin Endocrinol Metab 1995;80:608. 49. Cart MC. The emergence of the metabolic syndrome with menopause. J Clin Endocrinol Metab 2003;88:2404-2411. 50. Lakka HM, Laaksonen DE, Lakka TA, et al. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 2002;288:2709-2716. 51. Park YW, Zhu S, Palaniappan L, et al. The metabolic syndrome: prevalence and associated risk factor findings in the U.S. population from the Third National Health and Nutrition Examination Survey, 1988-1994. Arch Intern Med 2003;163:427-436. 52. Wilson PW, Kannel WB, Silvershatz H, D'Agostino RB. Clustering of metabolic factors and coronary heart disease. Arch Intern Med 1999;159:1104-1109. 53. Despres JE Abdominal obesity as important component of insulinresistance syndrome. Nutrition 1993;9:452-459.
C
K
ET AL.
54. Wing RR, Matthews KA, Kuller LH, et al. Environmental and familial contributions to insulin levels in middle-aged women. JAMA 1992;268:1890. 55. Wing RR, Matthews KA, Kuller LH, et al. Weight gain at the time of menopause. Arch Intern Med 1991;151:97. 56. Wing RR, Kuller LH, Bunker C, et al. Obesity, obesity-related behaviors and coronary heart disease risk factors in black and white premenopausal women. IntJ Obesity 1994;13:511. 57. Crawford SL, Casey VA, Avis NE, McKinlay SM. A longitudinal study of weight and the menopause transition: results from the Massachusetts Women's Health Study. Menopause 2000;7:96-104. 58. Krotkiewski M, Bjorntorp P, Sjostrom L, Smith U. Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution.J Clin Invest 1983;72:1150-1162. 59. Poehlaman ET, Toth MJ, Gardner AW. Changes in energy balance and body composition at menopause: a controlled longitudinal study. Ann Intern Med 1997;123:673-675. 60. Guthrie JR, Dennerstein L, Taffe JR, et al. Central abdominal fat and endogenous hormones during the menopausal transition. Fertil Steril 2003;79:1335-1340. 61. Sternfeld B, Wang H, Quesenberry CP Jr, et al. Physical activity and changes in weight and waist circumference in midlife women: findings from the Study of Women's Health Across the Nation. Am JEpidemiol 2004;160:912- 922. 62. Proudler AJ, Felton CV, Stevenson JC. Aging and the response of plasma insulin, glucose and C-peptide concentrations to intravenous glucose in postmenopausal women. ClinJ Sci 1992;83:489. 63. Buckler HM, Evans CA, Mantora H, et al. Gonadotropin, steroid and inhibin levels in women with incipient ovarian failure during anovulatory and ovulatory rebound cycles. J Clin Endocrinol Metab 1991; 72:116. 64. Metcalf MG, Livesey JH. Gonadotropin excretion in fertile women: effect of age and the onset of the menopausal transition. J Endocrinol 1985;105:357. 65. Vihko KK, Kujansuu E, Morsky P, et al. Gonadotropins and gonadotropin receptors during the perimenopause. Eur J Endocrinol 1996; 134:357. 66. Hall JE, Welt CK, Cramer DW. Inhibin A and inhibin B reflect ovarian function in assisted reproduction but are less useful at predicting outcome. Hum Reprod 1999:14:409-415. 67. van Rooij IAJ, den Tonkelaar I, Broekmans FJM, et al. Anti-mullerian hormone is a promising predictor for the occurrence of the menopausal transition. Menopause 2004;11:601-606. 68. Santoro N. Can a blood test predict the onset of menopause? Menopause 2004;11:585-586. 69. Treloar A, Bounton A, Benn R, et al. Variation of the human menstrual cycle through reproductive life. lntJ Ferti11967;12:77. 70. MetcalfMG. The approach of menopause: a New Zealand study. N Z MedJ 1988;101:103. 71. Santoro N, Lasley B, McConnell D, et al. Body size and ethnicity are associated with menstrual cycle alterations in women in the early menopausal transition: the Study of Women's Health across the Nation (SWAN) Daily Hormone Study. J Clin Endocrinol Metab 2004;89: 2622-2631. 72. McKinlay SM, Brambilla DJ, Posner JG. The normal menopause transition. Am J Hum Bio11992;4:37. 73. Guthrie JR, Dennerstein L, Hopper JL, et al. Hot flushes, menstrual status and hormone levels in a population-based sample of midlife women. Obstet Gyneco11996;88:437-430. 74. Tanis BC, Rosendall FR. Venous and arterial thromboembolism during oral contraceptive use: risk and risk factors. Semin VascMed 2003; 3:69-84.
CHAPTER 11 Decisions Regarding Treatment During the Menopause Transition 75. Tanis BC, van den Bosch MA, Kemmeren JM, et al. Oral contraceptives and the risk of myocardial infarction. N Engl J Med 2001;345: 1787-1793. 76. Castelli WP. Menopause and cardiovascular disease. In: Eskin BA, ed. The menopause-comprehensive management. New York: McGraw-Hill, 1994:117-136. 77. Kannel WB. Metabolic risk factors for coronary heart disease in women: perspective from the Framingham Study. Am HeartJ 1987;114: 413-419. 78. Razay G, Heaton KW, Bolton CR. Coronary heart disease risk factors in relation to the menopause. N Z J M e d 1992;85:889. 79. Matthews K, Meilahn E, Kuller LH, et al. Menopause and risk factors for coronary heart disease. NEnglJMed 1989;321:641. 80. Steinleitner A, Stanczyk FZ, Levin JN. Decreased in vitro production of 6 keto-prostaglandin F1 by uterine arteries from postmenopausal women. Am J Obstet Gyneco11989;161:1677. 81. Wren BG. Hypertension and thrombosis with postmenopauaal estrogen therapy. In: Studd JW , Whitehead MI, eds. The menopause. Oxford: Blackwell Scientific, 1989;181-189. 82. Hussman F. Long-term metabolic effects of estrogen therapy. In: Greenblatt RB, Heithecker R, eds. A modern approach to the perimenopausal years." nero developments in bioscience. New York: W de Gruyter, 1986:163-175. 83. Adams MR, Clarkson TB, Koritnik DR, et al. Contraceptive steroids and coronary artery atherosclerosis in cynomolgus macaques. Fertil Steri11987;144:41. 84. Guthrie JR, Taffe JR, Lehert P, Burger HG, Dennerstein L. Association between hormonal changes at menopause and the risk of coronary event; a longitudinal study. Menopause 2004;11:315-322. 85. Garton M, Martin J, New S, et al. Bone mass and metabolism in women aged 45-55. Clin Endocrino11996;44:536. 86. Slemenda C, Longcope C, Peacock M, et al. Sex steroids, body mass, and bone loss. A prospective study of pre, peri and postmenopausal women.J Clin Invest 1996;97:14. 87. Barrett-Connor E, Young R, Notelovitz M, et al. A two-year, doubleblind comparison of estrogen-androgen and conjugated estrogens in surgically menopausal women: effects on bone mineral density, symptoms and lipid profiles. J Reprod Med 1999;44:1012-1020. 88. Davis SR, McCloud P, Strauss BJG, et al. Testosterone enhances estradiol's effects on postmenopausal bone density and sexuality. Maturitas 1995;21:227-236. 89. Guthrie JR, Lehert P, Dennerstein L, et al. The relative effect of endogenous estradiol and androgens on menopausal bone loss: a longitudinal study. OsteoporosInt 2004;15:881-886.
167
90. Stewart A, Torgeson DJ, Reid DM. Prediction of fractures in perimenopausal women: A comparison of DEXA and broadband ultrasound attenuation. Ann Rheum Dis 1996;55:140. 91. Trotter M, Broman GE, Peterson RR. Densities of bone of White and Negro skeletons. J BoneJoint Surg 1960;42A:50-58. 92. Kleerekoper M, Nelson DA, Peterson EL, et al. Reference data for bone mass, calciotropic hormones, and biochemical markers of bone remodeling in older (55-75) postmenopausal white and black women. J Bone Miner Res 1994;9:1267-1276. 93. Russell-Aulet M, Wang J, Thornton JC, Colt EWD, Pierson RN. Bone mineral density and mass in a cross-sectional study of white and Asian women. J Bone Miner Res 1993;8:575-582. 94. Finkelstein JS, Mei-Ling TL, Sowers M, et al. Ethnic variation in bone density in premenopausal and early perimenopausal women: effects of anthropometric and lifestyle factors. J Clin EndocrinolMetab 2002;87:3057- 3067. 95. Rosen CJ, Chesnut CH, Mallinak NJS. The predictive value of biochemical markers of bone turnover for bone mineral density in early postmenopausal women treated with hormone replacement or calcium supplementation. J Clin Endocrinol Metab 1997;83:1904-1910. 96. Prestwood KM, Pillbeam CC, Burleson JA. The short term effects of conjugated estrogen on bone turnover in older women. J Clin Endocrinol Metab 1994;79:366-371. 97. Finkelstein JS, Sowers M, Greendale GA, et al. Ethnic variation in bone turnover in pre- and early perimenopausal women: effects of anthropometric and lifestyle factors. J Clin Endocrinol Metab 2002;87:3051 - 3056. 98. Henry YM, Eastell R. Ethnic and gender differences in bone mineral density and bone turnover in young adults; effect of bone size. Osteoporos Int 2000;11:512-517. 99. Bell NH, Williamson BT, Hollis BW, Riggs BL. Effects of race on diurnal patterns of renal conservation of calcium and bone resorption in premenopausal women. OsteoporosInt 2001;12:43-48. 100. Cohen FJ, Eckert S, Mitlak B H. Geographic differences in bone turnover: data from a multinational study in healthy postmenopausal women. Calcif Tissue Int 1998;63:277-282. 101. DeLeo V, Lanzetta D, D'Antona D, et al. Transdermal estrogen replacement therapy in normal perimenopausal women: effects on pituitary ovarian function. Contraception 1996;10:49.
This Page Intentionally Left Blank
2HAPTER
11
Use of Contraceptives
for Older Women DAVID E A R C H E R
Department of Obstetrics and Gynecology, CONRAD Clinical Research Center, Eastern Virginia Medical School, Norfolk, VA 23507
Despite an age-related decline in fecundity, women over the age of 35 have a fertility rate of 335 per 1000 age 35 to 39 and 25 per 1000 age 45 plus (1) (Fig. 12.1). Sexually active women over the age of 35 require an acceptable, effective contraceptive (1). The perimenopausal transition is marked by irregular ovulation and clinically is apparent with changes in the woman's menstrual cycle. Despite this reduced incidence of ovulation, these women are still at risk for pregnancy. The use of hormonal steroids in combination oral contraceptives can provide contraception and menstrual cycle control for these women (2). Menopausal symptoms, which frequently occur in the perimenopause, can be improved with the use of oral contraceptives (3,4). Older women have a variety of contraceptive options available to them based on their lifestyle and individual preferences. The final decision on which contraceptive method to recommend and utilize should be based on the patient's history and physical findings, current frequency of coital activity, and prior contraceptive experiences. The information obtained by the physician or health care provider concerning these three parameters will allow for a frank discussion of the risks and benefits of each contraceptive option. This dialogue will allow the patient to reach a decision on the best method for her. Currently available contraceptive options are listed in Table 12.1. TREATMENT OF THE POSTMENOPAUSAL W O M A N
This review will discuss each of these methods with the intent that they will be used for women over the age of 35 years. Combination oral contraceptives (COCs, containing both an estrogen and a progestin) should be considered for those individuals over 35 years of age who do not smoke. The use of oral contraceptives in women who smoke cigarettes and are over the age of 35 is not recommended because of the reported increase in cardiovascular disease. The incidence is reported to be 400 cases per 100,000 women who smoke and are over the age of 35 (5-7). The use of COCs in women with controlled hypertension or diabetes mellitus should be individualized. These medical conditions are felt to represent relative contraindications to the use of COCs. Women with cardiovascular risks (Table 12.2) should be individually evaluated. Controlled dyslipidemia is not a contraindication to the use of COC. Women with diabetes who have retinopathy or nephropathy should not use COC (8). A progestin-only contraceptive is recommended for women with coronary artery disease, cerebrovascular disease, and congestive heart failure (9).
I. ORAL CONTRACEPTIVES Combination oral contraceptives are one of the best options because of their ease of administration and known benefits for reduction in the incidence of functional ovarian 169
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
1
7
0
D
A
V
I TABLE 12.2
D
F. ARCHER
Risk Factors for Cardiovascular Disease
Hypertension Smoking cigarettes Diabetes mellitus Pregnancy-induced hypertension Dyslipidemias Family history of early onset of cardiovascular disease Obesity Androgen excess states
FIGURE 12.1
Fertility rates for women as a function of age. (From res 1.)
cysts, pelvic inflammatory disease, ovarian cancer, and endometrial cancer (2,10). The reluctance to use them in older women has in part been due to the perception that they increase the risk of cardiovascular events in older women. As described later, this is not the case, making COCs a first choice for these women. COCs have been available since 1960. They consist of an orally active estrogen, usually ethinyl estradiol (EE), and a synthetic progestin. All the formulations on the U.S. market
TABLE 12.1
Available Contraceptive Options for Older Women
Combination estrogen plus progestin steroidal contraceptives Oral Vaginal Transdermal Progestin-only methods Subcutaneous implants (not available in the United States at the time of writing) Injectable 3-month interval Oral progestin-only pills Intrauterine devices Copper containing Levonorgestrel releasing Barrier devices Condoms: male and female Spermicides Diaphragms Other Diaphragm-like devices Symptothermal or rhythm methods
have undergone clinical trials or are generic versions of the original compound. All marketed products have been found to be effective in preventing pregnancy. There has been a steady reduction in the concentration of estrogen and progestin in COCs since their introduction. This has been driven by the concern that some of the adverse side effects are related to the dose of the estrogen. Secondly, the philosophy of using the least amount of a medication that can be proven to be effective is an important consideration. At the present time there are several COC formulations on the U.S. market that contain low doses of ethinyl estradiol (20 ~g) (Table 12.3). All these formulations have pregnancy rates (Pearl Indices) of less than 2 pregnancies per 100 women per year (Pearl Index). Associated with the reduction in the estrogen dose has been the development of new progestational compounds. The first progestins marketed in the United States were 19-nor steroids derived from testosterone. Although ethynodid diacetate was the first marketed progestin, it is converted to norethindrone after ingestion, which appears to be the biologically active form (11) (Fig. 12.2). Norethindrone has been further modified to create what are called gonane progestins, with a methyl group at position 18 of the molecule. Norgestrel or its active isomer levonorgestrel is the principal gonane progestin. The last 15 years have seen further modification in the steroidal configuration of levonorgestrel to yield three progestins known as norgestimate, desogestrel, and gestodene. Figure 12.2 has drawings of the steroid configuration of these orally active progestational agents.
TABLE 12.3 Combination Oral Contraceptives Containing 20 Izg of Ethinyl Estradiol Progestin concentration Name Loestrin Mircette Alesse YAZ
Progestin
(~g)
Norethindrone Acetate Desogestrel Levonorgestrel Drospirenone
1.000 0.150 0.100 3.0
CHAPTER 12 Use of Contraceptives for Older Women
171
Chemical Derivatives of Testosterone OH
T
Norethindmne
0
"
~
(~H CinCH
v
OH
Ethi
.
CinCH
H
N
ICH
"
Chemical Derivatives of
Levonorgestrel Levonorgestre! \ OH
C CH
Desogestrel
\
OH ...CinCH
Norgestimate ~
OAcetate CnCH
HEN"" ~
V
FIGURE 12.2 Orally active progestational agents derived from testosterone. Estranes are derivatives of norethindrone, while gonanes are related to levonorgestrel.
A recent introduction is the progestin drospirenone (DRSP), which is derived from spironolactone. This compound has a different molecular configuration than the classical 19-nor steroids used in COCs (Fig. 12.3) (see color insert).
A. Efficacy Oral combination contraceptives are designed to prevent pregnancy. Efficacy is assessed using the Pearl Index, which measures the number of pregnancies that occur in a known number of women during 1 year of use of a contraceptive agent (12). It is calculated by dividing the number of unintended
pregnancies by the number of months of use of the particular method whose efficacy is being measured, and multiplying the result by 1200. The current method uses the number of cycles of use (rather than months), and the result is multiplied by 1300 (on the basis that the average cycle length is 28 days). The contraceptive effectiveness of COCs can also be calculated using a life table analysis (13). Women over the age of 35 have been shown to have a decreasing fecundity (1,14). It is obvious that this reduction does not reach zero, and these women are at risk for pregnancy. According to data from the CDC, unintended pregnancies are a major problem in older women, where the use of therapeutic abortion is 400 per 1000 live births (Fig. 12.4).
1
7
2
D
A
V
I
D
E ARCHER
Spironolactone and Drospirenone
FIGURE 12.3
Drospirenonea progestational agent derived from spironolactone.
is the inhibition of ovulation; second is alteration in the receptivity of the cervical mucus for sperm penetration; and third is a change in the endometrium that may alter or inhibit implantation (16-18).
2. REDUCED INCIDENCE OF ECTOPIC PREGNANCY
FIGURE 12.4 Abortion rates by age groups in the United States by year. (From http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5212al.htm#tab4/)
These data underscore the need for contraceptive methods for the older woman.
B. Benefits of Oral Contraceptives in Women over 35 1. CONTRACEPTION
The principal benefit of COCs is the ability to allow a couple to determine when they want a pregnancy to occur. The utilization of oral contraceptives has been limited in older women because of the risk of cardiovascular events (discussed later). Current evidence indicates that for women with no cardiovascular risk, COCs are a highly effective and safe method (15). A low-dose estrogen-containing (35, 30, or 20 txg) COC should be the principal method used. Contraception is achieved in several ways. The principal method
Associated with the reduction in the occurrence of pregnancy is a concomitant reduction in ectopic pregnancy (10,19,20). This is obvious because inhibition of ovulation is one of the major mechanisms of action of COCs. Thus, the reduction in intrauterine pregnancies should be mirrored with a similar decline in ectopic pregnancies. COCs have been shown to reduce the incidence of functional ovarian cysts (10,12,20,21). In fact, COCs are indicated for the treatment of functional ovarian cysts, although one to two cycles may be required before the ovarian cyst regresses.
3. PELVIC INFLAMMATORY DISEASE
Use of COCs has been reported to decrease the incidence of pelvic inflammatory disease (PID) (10,12,20,21). The occurrence and incidence of cervical colonization with Neisseriagonorrhea and Cblamydiatracbomatisis not reduced, but the rates of clinical PID have been found to be decreased in women using COCs (20).
4. ENDOMETRIAL AND OVARIAN NEOPLASIA COCs have been found to reduce the incidence of endometrial cancer by at least 40% (22-25). This protective effect is apparent after 1 year of use. The reduced incidence
CHAPTER 12 Use of Contraceptives for Older Women lasts for 15 years. This duration is the length of the followup studies reported to date. The mechanism of action is thought to be due to the progestin altering the endometrium. In support of this hypothesis is the finding of an enhanced reduction in the incidence of endometrial adenocarcinoma in women who have used a low-dose estrogen with a potent progestin. COC reduce the incidence of ovarian neoplasia by 40% (20,26). The duration of this effect is also 15 years following 1 or more years of use of COCs. Recent data indicate that reduction in the incidence of ovarian cancer is also applicable to women who have a strong history of inheritable breast and ovarian cancer (27). These women often have BRCA1 or BRCA2 present on genotyping. The inhibition of ovulation is felt to be the principal mechanism whereby ovarian neoplasia is interdicted by COCs.
5. IRREGULAROR EXCESSIVEUTERINE BLEEDING
COCs have been used for the treatment of irregular or heavy uterine bleeding associated with anovulation, uterine fibroids, and women presenting with dysfunctional uterine bleeding (28). Clinical evidence of the COC efficacy for each of these clinical conditions is sparse. COCs have been shown to reduce the amount of bleeding in normal women. Menorrhagia and irregular menstrual intervals are also amenable to intervention with COCs (21). The data on the efficacy of COC as a treatment menorrhagia is scant (29).
6. BONE MINERAL DENSITY
COCs have been reported to enhance bone mineral density (BMD) in older women (30-34). This attribute would make them a logical choice for perimenopausal women who need to have a "bone bank" of increased bone mineral density prior to menopause.
C. A d v e r s e E v e n t s A s s o c i a t e d with Oral Contraceptives 1. MYOCARDIALINFARCTION
Myocardial infarction has been reported to be increased in women over the age of 35 who use a COC and who smoke (7,8). Recent epidemiologic studies have indicated that in healthy women who do not smoke, there is no increased risk of myocardial infarction (8,35,36). The Transnational study at one time showed that desogestrel-containing COCs would actually reduce the incidence of myocardial infarction in women compared with other progestins (37,38). In women over the age of 35 who do not have any cardiovascular risk, COCs are indicated as an effective means of contra-
173 ception (see Table 12.2). This author would recommend that the 20-1xg preparations be used if starting women on COCs at this age. The risk of myocardial infarction related to smoking and oral contraceptive use decreases after stopping smoking. Several studies have found that the risk is diminished 1 or more years after smoking cessation (39,40). Women who have successfully stopped smoking for more than 1 year do not appear to have an increased risk of myocardial infraction, and COCs can be used in these women.
2. CEREBROVASCULARDISEASE Cerebrovascular disease, or stroke, has been suggested to be increased in women who use COCs (41). This increase in the incidence has been linked to the dose of estrogen, with a reduction in the incidence of stroke associated with a decrease in estrogen dose. The overall incidence of stroke is rare in women under the age of 45 (42,43). A significant increase in stroke is associated with advancing age. Smoking is a risk factor for stroke in both premenopausal and postmenopausal women. Recent data have found an increased incidence of both thrombotic and hemorrhagic stroke in women who smoke (43). In women without risk factors, the current or past use of COCs was not associated with an increase in the incidence of either thrombotic or hemorrhagic stroke (41,43). The World Health Organization ( W H O ) has reported an increased risk of ischemic stroke in nonsmokers with normal blood pressure using oral contraceptives (44). There was no significant increase in hemorrhagic stroke in this population (45). Women receiving treatment for diabetes mellitus or hypertension were found to have a significant increase in the relative risk for stroke (43,45). These data indicate that other disease states may contribute significantly to the cause of stroke in women and not specifically the hormones in COCs.
3. VENOUSTHROMBOEMBOLISM COCs have been linked with the occurrence of deep vein thrombosis (DVT) and pulmonary embolism (PE) (46,47). The incidence of venous thrombosis increases with age (48). The occurrence of DVT has been linked to the dosage of EE in the COCs (49,50). Reductions in the ethinyl estradiol dose from more than 50 Ixg to the current formulations containing 20 to 35 Ixg have reduced the incidence and relative risk from venous thromboembolism (VTE). The reports from three recent studies indicate that the relative risk for DVT is approximately 2.0 for COCs with less than 35 Ixg of ethinyl estradiol (46,47,51). These results translate into an actual incidence of 2 to 4 cases of DVT per 10,000 women at risk per year. Controversy has arisen from these papers in regard to an increased incidence of VTE in women using third-generation COCs, such as norgestimate, desogestrel,
1
7
4
D
A
and gestodene. Compared with second-generation COCs, such as levonorgestrel, the epidemiologic studies have found a relative risk of 1.5 or 2.0 for VTE in young women using third-generation COCs compared with second-generation COCs (levonorgestrel). Recent publications have discussed the potential biases in each of these studies in order to explain the increased relative risk with the newer progestational COCs (47,52-54). One of the biases identified is the healthy user effect (55,56). Physicians will not or do not change COC formulations in women without problems. A second bias is the attrition of susceptibles. This hypothesis states that there is a pool of women who are more likely to develop VTE when placed on COCs. This group of women could develop a DVT or PE after having a VTE diagnosed. The patient who develops a DVT on COCs is no longer a candidate for COCs. New starters on COCs can include a pool of susceptible women within the larger cohort of women who are not susceptible to VTE. Therefore, use of newer formulations of COCs can have a larger at-risk group of women than the older formulations because of their inclusion as new starters. These findings could result in the apparent increased incidence of DVT in current users of desogestrel- or gestodenecontaining products. These points have been extensively debated in the literature. At present, the most important point that has been made is that there is no biologic plausibility for the results of an increase in VTE incidence in women on third-generation progestational agents. Estrogens have always been implicated in the etiology of VTE. The low-dose 20-1~g ethinyl estradiol products available have an even lowered incidence of VTE (49,50). 4. BREAST CANCER
The role of hormonal agents in the development of breast cancer has been debated for some time. The use of COCs in North America has been estimated to have involved 85% of the female population at some time in their life. The most recent review by the Centers for Disease Control and Prevention (CDC) has not shown a significant increase in the incidence of breast cancer in current or past users of COCs (57). The incidence of breast cancer was the same in both white and black women, and the age at initiation of COC was not associated with an increased incidence (57).
5. WEIGHT GAIN
One of the principal reasons for discontinuation of COCs is the real or perceived weight gain. Clinical trials that are performed for the introduction or licensing of COCs have failed to show any significant weight gain in participants during the clinical trial (58,59). A recent Cochrane review found equivocal evidence for COCs' effect
V
I
D
17.ARCHER
on weight but concluded that if weight gain occurred, no large effect was apparent (60). Despite this, there remains a strong feeling that COCs will cause weight gain. The most common reasons for weight gain are lack of exercise and overeating. Counseling on diet and exercise is required to interdict weight gain, but its effectiveness in modifying behavior is limited.
6. HEADACHES
Headache is a ubiquitous symptom and is reported in about 10% of women participating in clinical trials of COCs (61). This figure always seems high, but this is due to the lack of an appropriate control group. Menstrual-related headaches and other symptoms are those that are associated with the onset of menstruation and are cyclic and repetitive in nature (62). The management of these headaches can be a problem. Recently, the extended use of COC for 42 days or longer in a continuous fashion has been described (63). This may be used for the management of menstrual migraines, and other menstrual symptoms (64,65). The consumer and health care practitioner should be aware that continuous use of COC can result in breakthrough bleeding in these women (64,66-68). When breakthrough bleeding occurs, the COC should be stopped for 3 to 5 days. After this, continuous COC can be reinitiated, and this may enhance compliance. There are no known successful interventions for the irregular bleeding that occurs with continuous-use COCs. A second option is to use a transdermal delivery system of estradio1-17b (E2) as a bridge during the placebo or pill free days of the COC cycle. This will theoretically prevent the fall in serum E2 levels, which are thought to be the stimulus for the headache. Efficacy data are limited for this approach which has largely been driven by anecdotal reporting (69). The recent introduction of COCs with a reduced pill-free interval (Mircette [Organon, Inc, West Orange, NJ] and Loestrin-24 [Warner Chilcott, Rockaway, NJ]) may be useful in this instance, but there are no data indicating efficacy for menstrual migraine with this addition of several more days of hormonal steroids. Migraine headache is also associated with the withdrawal of the estrogen in the COC preparation. The significance of migraine headaches is their association with stroke, and the effect of COCs has been debated (70,71). The migraine literature indicates that there is an increased incidence of migraine headache in COC users. The COC literature is sparse in this regard, but a recent review indicates that migraine headache sufferers do not have an increased incidence of occurrence of headaches while using COCs (72). A trial of COC is indicated in women with a history of migraine headaches. This can be viewed as a therapeutic trial. In women with no increase in frequency or intensity of their migraine headaches, continued use of COCs is indicated. The use of progestin-only contraceptives, either oral, injec-
175
CHAPTER 12 Use of Contraceptives for Older Women
tion, or implantation, may be indicated in women who continue to have frequent or severe migraine headaches.
There are two different IUDs on the U.S. market--the copper-containing IUD, known as Paraguard (Ortho Pharmaceutical, Raritan, NJ) and the medicated device containing levonorgestrel, Mirena (Berlex, Montville, NJ).
II. INTRAUTERINE DEVICES Intrauterine devices offer a significant alternative for older women. They are a highly effective means of contraception and have few systemic side effects. Intrauterine devices (IUDs) have been linked to an increased incidence of peMc inflammatory disease (PID) (73-75). The early articles from the C D C implied that there was an increased risk of PID in users of IUDs (73). The more recent review of the CDC data does not support an increase in the incidence of PID in monogamous couples (76). The increased incidence is found in divorced or unmarried women and is strongly linked to the number of sexual partners. A large multicenter, multinational trial reported by the World Health Organization also indicated that there is no increase in the incidence of PID in women using IUDs. There is evidence for an increase in what appears to be PID in the first 3 to 4 weeks following insertion of the IUD (77). These data have led to published reports indicating that prophylactic antibiotics could or should be used at the time of insertion of the IUDs. There is no consensus on the use of antibiotics at the time of IUD insertion. At the present time in the United States there is no compelling evidence that concurrent antibiotics with insertion of the IUD prevent the occurrence of infection immediately following the insertion of the IUDs. Intrauterine devices do not have a systemic contraceptive action. They appear to exert their effects locally, either by altering gamete function or changing the endometrial receptivity. The bulk of the data support an effect on gametes with either no or few sperm cells detected in the fallopian tubes of women using IUDs (78,79). This finding is also reflected in the lack of fertilized oocytes or embryos present in the fallopian tubes of women using an IUD. At one time it was hypothesized that one of the mechanisms of action of the IUD was to inhibit implantation or to be an abortifacient. There is no evidence that there is any abortifacient activity associated with IUDs. Published reports have failed to find any evidence of an increased occurrence of either early clinical miscarriages or evidence of human chorionic gonadotropin (hCG) in serum or urine as an indicator of pregnancy (78). Recently we have shown that the administration of pentoxifylline (Trental) to rats bearing an IUD can reverse the contraceptive effect of the IUDs (80). We believe that this effect is due to the action of pentoxifylline altering the function of the endometrial leukocytes that are present in the intrauterine lumen and the endometrial stroma of women and animals bearing an IUD. Our current hypothesis is that intrauterine cytokines are involved as part of the mechanism of IUD action.
A. Copper-Containing IUDs The copper-containing (Cu) IUD is a highly effective contraceptive with a Pearl Index of between 1.0 and 2.5 pregnancies per 100 women per year of use. The variability of the Pearl Index is related to the population under investigation. The mechanism of action of the Cu IUDs is both a foreign body reaction of the IUD's frame (silicone) and the release of elemental copper, which is a known spermicide. The load or amount of copper on the devices has been variable, but the Paraguard has 380 square millimeters of surface area of copper wound on the arms and the stem of the IUD. This system has an effective duration of action of 10 years (81- 84). Side effects include uterine cramping and, in some instances, expulsion of the device. Another side effect is an increase in the amount and duration of menstrual flow. Nonsteroidal anti-inflammatory agents have been shown to be effective in decreasing the amount of bleeding associated with IUDs (76). The studied doses are ibuprofen 800 mg three times a day and Anaprox 500 mg three to four times a day.
B. Levonorgestrel-Releasing Intrauterine Device The release of levonorgestrel from the stem of the IUD (Mirena [Berlex laboratories, Montville, NJ]) is designed to enhance its contraceptive efficacy. The Mirena has been described as a levonorgestrel intrauterine system (LNG-IUS) delivering levonorgestrel to the endometrium. Levonorgestrel concentrations are high in the endometrium, which shows progestationaX dominance on endometrial histology (85,86). The use effectiveness of the LNG-IUS is very high, with a Pearl Index of 0.2. The duration of efficacy is currently 5 years, which makes this a cost-effective method. Clinical trials have found that there is a decrease in the amount of menstrual blood loss in women using the intrauterine system (87). Menstrual cramps are also decreased with this system.
III. INJECTABLE CONTRACEPTION Injectable preparations of a progestin have been available for 30 years. The prototype is Depo-Provera, a crystalline suspension of medroxyprogesterone acetate. A long-acting form of norethindrone enanthate (Noristerate [Schering AG]) is marketed outside the United States. Noristerate is
1
7
6
D
A
administered on an every 2-month basis. Depot medroxyprogesterone acetate (DMPA) is administered at 11- to 12-week intervals. The advantage of DMPA is the fact that it only contains a progestational steroid and has no estrogen component. The contraceptive efficacy is high, on the order of less than 1 pregnancy per 100 women per year of use (Pearl Index) (88). The advantages for older women are the lack of estrogenic side effects. Progestin-only methods may be used in women at risk for cardiovascular problems. Women who have liver dysfunction may be candidates for DMPA, but this decision should be individualized. Limited data suggest that progestin-only contraception should or could be used in women with migraine headaches. Conclusive medical evidence in this regard is lacking. One significant drawback is the pharmacokinetic profile of DMPA. In some women, medroxyprogesterone acetate has been found to remain in the circulation for more than 18 months (76). This extended duration of action has resulted in a delay in the return of ovulation and, of course, fertility. This potential problem should be discussed with women who may want to become pregnant in the future. The overall return of fertility (fecundity) is comparable to that of women stopping other forms of reversible contraception based on the cumulative pregnancy rate. There is a delay of between 3 to 6 months in pregnancy rates immediately following DMPA use, but by the end of 24 months, pregnancy rates are comparable between women who used D P M A and users of other forms of reversible contraceptives. The disadvantages for older women are the increased risk of irregular bleeding that occurs in the first year of use. A decrease in bone mineral density has been observed in current users, which appears to improve after discontinuation of the method (89,90). Irregular bleeding is common in all progestin-only methods of contraception. The cause of the irregular bleeding is unknown. The pattern of bleeding is irregular spotting and staining without premenstrual or prodromal symptoms (91). The bleeding usually consists mainly of spotting; rarely is there an associated anemia. This irregular bleeding does not imply that there is endometrial pathology. Endometrial biopsies have demonstrated only an atrophic or progestationally suppressed endometrium. Treatment usually consists of simple reassurance and observation. In some cases where there is concern, the use of an oral estrogen has been reported. There appears to be some efficacy in the use of ethinyl estradiol or estrone sulfate to stop the bleeding or spotting, but the improvement is minimal (92). By the end of the first year of use of DMPA, more than 50% of the women are not bleeding (93). A decrease in bone mineral density in women using DMPA has been reported (94). Peripheral serum estradiol levels were lower than those found in the early follicular phase. This study was small and requires further follow up. Extensive follow up of users of DMPA has been performed
V
I
D
F. ARCHER
by the W H O because of concerns over potential breast cancer. The report of a multinational, multicenter trial has shown only an increase in breast cancer in the first 3 months of use, and there is no increase in breast cancer with longterm (over 10 years) use of DMPA (95,96). There appears to be a significant reduction in the incidence of endometrial cancer in DMPA users (97,98). This latter fact may enhance the utility of DMPA for older women. There is always concern about changes in mood in older women using medroxyprogesterone acetate orally. Reports of these changes are for the most part anecdotal. The same effect on mood or depression has been sought in younger women using DMPA. No significant change in depression scores was found in a prospective study of DMPA users (99,100). Overall, DMPA is a reasonable alternative for contraception in older women. It has the advantage of limited motivation for the consumer. The need to return for repeat injections on a regular basis may be a significant issue for some individuals, making this a less desirable alternative for contraception in women over the age of 35 years.
IV. IMPLANTABLE CONTRACEPTIVES There were no implantable contraceptive devices marketed in the United States at the end of 2006.
A. Levonorgestrel Implants Levonorgestrel implants (Norplant [Wyeth-Ayerst, Philadelphia]) are a six-capsule system with each capsule containing 30 mg of levonorgestrel. A newer two-rod device, known as Jadelle, also contains levonorgestrel. The contraceptive efficacy is due to increasing the length of the rods to enhance delivery of the steroid (101,102). Levonorgestrel implants were developed by the Population Council beginning in 1968 to 1970. Several different progestational agents were evaluated for efficacy at that time, but levonorgestrel was selected due to its release rate through silastic membrane (103). Clinical trials have indicated that levonorgestrel implants have a very low contraceptive failure rate, with a Pearl Index of less than 0.5 pregnancies per 100 women per year (104). Levonorgestrel implants can be used in older women because it contains no estrogen, as mentioned earlier for DMPA. The advantages of the Norplant contraceptive devices are the fact that the consumer requires little motivation after the insertion, and the duration of contraceptive efficacy is up to 5 years. As with all progestin-only contraceptives, side effects include irregular uterine bleeding (101,105). This is principally light bleeding and spotting and is similar to that described for DMPA. Thirty percent of women will experience
177
CHAPTER 12 Use of Contraceptives for Older Women
irregular bleeding and spotting within the first 3 months after the insertion of Norplant and Jadelle (105 - 107). With continued use of levonorgestrel implants, occurrences of menstrual bleeding decrease (101). Other side effects that have been associated with the levonorgestrel system include weight gain, depression, acne, and loss of scalp hair (101,108-110). No one has reported alopecia, although a heavy loss of hair has been anecdotally reported. The principal contraindications to levonorgestrel implants are active liver disease and undiagnosed uterine bleeding. Both these conditions should be evaluated prior to insertion, along with any other laboratory evaluation that is indicated. Overall, the levonorgestrel implant systems have a very positive safety profile. They have not been linked to any risk of cardiovascular disease (101,107,110). There is no evidence of an increased risk of venous thromboembolism with progestin-only contraceptive methods (107).
B. Etonorgestrel Implant Etonorgestrel, the metabolite of desogestrel (3-keto desogestrel) or etonogestrel, has been used in a ethylene vinyl acetate delivery system. This product is a single-rod implant with a duration of action of 3 years (111,112). The contraceptive efficacy was found to be high during clinical trials (113-115). Endometrial bleeding that is irregular in occurrence and duration was a common side effect. Recent attempts to control the endometrial bleeding and spotting using short-term therapeutic interventions resulted in only minimal improvements (116).
V. TRANSDERMAL CONTRACEPTION The ability to deliver ethinyl estradiol and norgestimate through the skin using enhancers has resulted in the development of a once-a-week transdermal contraceptive, OrthoEvra (Ortho-McNeil Pharmaceutical, Raritan, NJ) (117). The efficacy of the transdermal contraceptive is comparable to oral contraceptives (118). The compliance with a oncea-week administration was higher in younger women compared with COCs, resulting in an increased efficacy in the 18-to-25-year-old population (119). There is a need for reliability of the adhesive characteristics of the transdermal system. Patch adherence was maintained under various degrees of heat and wetness (120). Concerns have been raised over the risk of venous thrombosis with transdermal contraception. There was no evidence of an increased adverse event profile in the clinical trials (121). According to data from the manufacturer, there is an increase of 60% in the area under the curve for ethinyl estradiol compared with a 35-~g COC preparation.
Peak blood levels of ethinyl estradiol were less than that found with a COC (122). A recent nested case-controlled study found that both COC and transdermal contraceptive systems had a similar incidence of venous thromboembolism (123). An increased risk of venous thrombosis is present with all estrogen-containing hormonal preparations.
VI. STEROIDAL VAGINAL CONTRACEPTIVE Vaginal rings delivering steroids have been investigated for many years as a different route of delivery for contraceptive steroids (124). The first development was of progestinonly vaginal rings that were associated with a high incidence of irregular bleeding (125). Combination rings using ethinyl estradiol and a progestin were effective and had a better bleeding profile (126). The ethinyl estradiol and etonorgestrel vaginal contraceptive ring (NuvaRing [Organon, Inc, Roseland, NJ]) found inhibition of ovulation with good cycle control in clinical trials (127-129). A significant advantage for compliance is the ability to leave the vaginal ring in place for 21 days, removing it only to have a withdrawal bleeding episode (127). There does not appear to be a significant local reaction to the vaginal contraceptive ring (130).
VII. BARRIER CONTRACEPTIVES A variety of barrier-type contraceptives are available on the U.S. market. Physical barriers include the diaphragm, other diaphragm-like devices (Lea's contraceptive, Femcap), and male and female condoms. Listed under barriers, although not truly a physical barrier, are spermicides, usually containing nonoxynol-9 (N-9), that are available in a variety of forms, including films, suppositories, and tablets. In general, barrier contraceptives have been associated with a higher pregnancy rate than hormonal contraceptive methods. The use effectiveness rates range from 7 to 15 pregnancies per 100 women per year with a variety of barrier contraceptives (12,131). With highly motivated couples who consistently use barrier contraceptives in a reliable fashion, pregnancy rates as low as 4 to 5 pregnancies per 100 women per year have been reported. Overall, barriers offer an advantage to older women from several standpoints. They are coitally related, and from this standpoint they only require use at the time of intercourse. In a couple with declining frequency of coital activity, this may be something that is attractive to them. Secondly, they do not involve hormonal medication, an intrauterine device, or an implant. From this standpoint, they are totally consumer controlled
1
7
8
D
A
and consumer driven. The appropriate use of the barrier contraceptive is dependent on the motivation of the consumer and her partner. Barrier contraceptives depend on the utilization of an associated spermicide to enhance their efficacy. This is true for the diaphragm, the cervical cap, and, in some instances, male and female condoms. Male and female condoms have been impregnated with N-9 as an additional means of reducing pregnancy rates. The spermicidal product nonoxynol-9 is a surface-active detergent that has been shown to lyse cell membranes, resulting in spermicidal immobilization or death (132). The concentrations of N-9 in spermicide vary between preparations. Marketed preparations and their spermicidal content and concentration are shown in the accompanying table (Table 12.4). Barrier contraceptives, in order to be effective, should be inserted prior to penile penetration of the vagina. In general, for spermicides, an interval of between 5 and 15 minutes has been recommended between application and vaginal penetration. Clinical trials to confirm the efficacy of this time interval have not been reported. Some spermicides on the U.S. market have never been tested in clinical trials. Our experience using a variety of marketed spermicidal preparations has shown that the spermicide effectively reduces the number of motile sperm seen in the postcoital test of normal couples to less than one sperm per high-powered microscopic field (HPF) (133,134). There is no good correlation between the postcoital test results and the reduction in fecundity at this juncture. However, the postcoital test has been used along with in vitro spermicidal function as surrogates to indicate a high level of contraceptive efficacy.
A. Diaphragms Diaphragms have been available for many years on the U.S. market and have principally been made out of latex. A coil spring is present in the outer margin as a means of holding the diaphragm in place. All diaphragms require fitting by a health care provider, and fitting changes should be performed after vaginal delivery. The fitting is done best by using a ringlike device to measure the normal dimensions of the upper vagina. Diaphragm sizes are in millimeters, such that a number 70 diaphragm is a 70 mm in transverse diameter. All marketed diaphragms, whether they are spring or arc-spring types, rely on the application of the spermicide within the concavity and around the edges of the diaphragm to enhance their contraceptive efficacy. There are no studies that this author is aware of indicating that the diaphragm alone without a spermicidal product has any contraceptive efficacy. Recent clinical trials of two new barrier devices, the Lea's contraceptive and the Femcap, have been performed with and
V
I
D
F. ARCHER
TABLE 12.4 Firms Manufacturing or Distributing Contraceptives Worldwide, 1993 and 1994 Company Aladan Ansell Berlex
Boehringer Ingelheim Bristol-Myers Squibb Carter-Wallace CCC (Canada)" Cervical Cap Ltd. Chartex (United Kingdom) Cilar (UK)a Dongkuk Trading (Korea)a Finishing Enterprisesa Gedeon Richter (Hungary) Gruenenthal Hyosung (Korea)a Jenapharm (Germany) Kinsho Mataichi (Japan) a Leiras Oy Pharmaceuticals Lexis London Rubber Magnafarma Mayer Mead Johnson Medimpex a (Hungary and USA) Menarini Milex Products Inc.
National Sanitary Okamoto, USA
Product manufactured or distributed Condoms Condoms (including one with nonoxynol-9) Oral contraceptives (Tri-Levlen, Levlen) Intrauterine system (Mirena) Oral contraceptives Oral contraceptives Condoms (including one spermicidally lubricated) IUDs Cervical cap (Prentif; manufactured by Lamberts/Dalston England) Female condom (Femidom) Oral contraceptives Condoms IUDs Emergency postcoital contraceptive (Postinor) Oral contraceptives Condoms Oral contraceptives Spermicides Progestin-releasing IUD (Mirena) Norplant (manufacture) Oral contraceptives (NEE) Condoms Diaphragms Oral contraceptives Condoms Oral contraceptives (Ovcon) Oral contraceptives/raw materials Oral contraceptives Diaphragms (Omniflex, Wide-seal) Jellies and creams (Shur Seal Jel has nonoxynol-9) Condoms Condoms
aFirm supplies to UNFPA procurement.Where the firm is listed with more than one product line and is a UNFPA source,the product supplied is also markedwith an ~. SOURCES: Frost and Sullivan.U.S. contraceptive and fertility product markets.New York, 1993. United Nations PopulationFund (UNFPA). 1993 procurement statistics. New York, 1993.
179
CHAPTER 12 Use of Contraceptives for Older Women TABLE 12.4 Firms Manufacturing or Distributing Contraceptives Worldwide, 1993 and 1994--cont'd Company Organon
(Akzo) (Netherlands) a Ortho Pharmaceuticala
Parke-Davis (Warner-Lambert) Polifarma Reddy Health Care RFSU of Sweden Roberts Rugby Labs Safetex Schering AG (Germany)a Schering Plough (USA) Schmid
Searle (Monsanto) Seohung (Korea)a Syntex Thompson Medical Upjohn a (Upjohn Belgium)a Warner-Chilcott Whitehall Wisconsin Pharmacal Wyeth-Ayerst (American Home Products) Wyeth-Pharma (Germany)a Wyeth (France)a
Product manufactured or distributed Oral contraceptives (Marvelon, Desogen, Jenest) Implant (IMplanon) Vaginal Ring (NuvaRing) IUD (Multiload) (manufacturing subsidiary, Bangladesh) Oral contraceptives (Loestrin; Ortho-Cept, Ortho-Cyclen, Other Tri-Cyclen, Ortho-Novum, Modicon) Diaphragms (Allflex, Ortho Diaphragm) Spermicidesa (Gynol [octoxynol]) IUDs Oral contraceptives Oral contraceptives Condoms Condoms Oral contraceptives Oral contraceptives (Genora) Condoms Oral contraceptivesa Injectablesa Spermicides Condoms (including spermicidal condoms) Spermicides Diaphragms (distributed by GynoPharma) Oral contraceptives (Demulen) Condoms Oral contraceptives (TriNorinyll, Devcon, Norinyl, Brevicon) Spermicides Injectable (Depo-Provera/
DMPA) Oral contraceptives (Nelova) Sponge (Today)b Female condom (Reality) Oral contraceptives (Lo-Ovral, Nordette, Triphasil) (joint venture, Egypt, production) Norplant (marketing, distribution) Oral contraceptives Oral contraceptives
bWhitehall decided to discontinuethe Todaysponge because of the costs of bringing the plant up to U.S. Food and Drug Administration specifications.
without spermicide. The postcoital test was used as the measure with the Femcap, which was tested with and without spermicide, and compared with the diaphragm with spermicide (133). There was no difference in the number of sperm in the cervical mucus (average 0.1 sperm per HPF) in each treatment arm. Similar results were found in a phase I trial of the Lea's contraceptive when it was compared with and without spermicide to a diaphragm with spermicide (135). No sperm were present in the cervical mucus of the women with a diaphragm or Lea's contraceptive with spermicide. Only two sperm were found in one woman in the group using the Lea's contraceptive alone without spermicide. The contraceptive efficacy of the Lea's contraceptive was evaluated further in a clinical trial with and without spermicide. There were more pregnancies in the nonspermicide group, but the difference was not statistically different. The overall pregnancy rate was not different from the historical pregnancy rate of diaphragm users (136). Neither of these barrier devices is available on the U.S. market as of this time.
B. Condoms Both male and female condoms are available on the U.S. market. In the past, male condoms were made of latex, but recently they have been made of a polyurethane material that is designed to increase strength, reduce breakage, and enhance heat transmission and therefore increase sensation with coital activity (137-139). A variety of condom designs and colors have been used in order to enhance consumer utilization in both developing and developed countries. The addition of a spermicide to the male condom should enhance its contraceptive efficacy, but there are no identified studies that address this issue. It should be pointed out that nonoxynol-9, as well as latex, can result in allergic reactions in consumers. This reaction can take the form of a local irritation, or burning, and in rare instances an anaphylactoid reaction has been reported. Female condoms, on the other hand, are available and are made out of a polyurethane material (140). This device has a baglike structure with an inner ring that is designed to anchor in the upper portion of the vagina, and an outer ring that protrudes outside the vaginal introital area. This was designed to protect the perineum. The objective of the female condom is to allow a woman-controlled method that will prevent pregnancy and reduce the heterosexual transmission of sexually transmitted diseases, such as gonorrhea, chlamydia, syphilis, and possibly human immunodeficiency virus (HIV) (141). The female condom has as its major advantage this protective effect. Its major disadvantage is its price and the fact that it is only utilized for one time and then is disposed of.
1
8
0
D
A
V
C. S p e r m i c i d e s Spermicides are principally nonoxynol-9 products in the United States. They come in a variety of formulations and packaging as shown in Table 12.4. Other agents that have been used as spermicides are octoxynol-9, menfegol, and benzalkonium chloride. At the present time, spermicides, like condoms, are available over the counter, but many couples are unaware of their availability, effectiveness, and utility. Use effectiveness with spermicides are on the order of condoms and diaphragms, with a wide range of reported pregnancy rates (131). Many spermicides are packaged in such a manner that they can be carried in the purse or in a small wallet without contributing undue bulk. At the present time, the U.S. Food and Drug Administration is undertaking a review of currently marketed spermicides to document their efficacy. In general, spermicides are to be utilized prior to the insertion of the penis. They are only effective for a one-time use, and a second episode of coital activity requires a second application of the product. Side effects associated with spermicide use have been an increased incidence of vaginitis and an increase in urethritis (142-147). Benefits have been a reduction in the transmission of sexually transmitted diseases and the fact that it is a local product without systemic effects (148-150).
VIII. S U M M A R Y The contraceptive products that are available for young women are also available for women over the age of 35. However, the physician or health care provider should take into consideration the woman's medical history, physical findings, and past contraceptive utilization before recommending or prescribing any of the contraceptive options as listed in this chapter. Overall, in highly motivated individuals, even barrier contraceptives can reduce the pregnancy rate down to below 5 to 6 pregnancies per 100 women per year. The health care provider should be aware of the fact that women over the age of 35 still have a pregnancy potential and require contraception.
References 1. Heffner LJ. Advanced maternal age--how old is too old? N EnglJ Med 2004;351:1927-1929. 2. Kaunitz AM. Oral contraceptive use in perimenopause. Am J Obstet Gyneco12001;185(2 suppl):S32- 37. 3. Blumel JE, Castelo-Branco C, Binfa L, Aparicio R, Mamani L. A scheme of combined oral contraceptives for women more than 40 years old. Menopause 2001;8:286-289. 4. Ylikorkala O. Drospirenone, a progestin with a unique cardiovascular profile, for safe contraception and treatment of menopausal symptoms. Climacteric2005;8(suppl 3):1 - 3.
I
D
E ARCHER
5. W H O scientific group meeting on cardiovascular disease and steroid hormone contraceptives. Wkly E.pidemiolRec 1997;72(48):361-363. 6. Petitti DB, Sidney S, Q.gesenberry CP. Oral contraceptive use and myocardial infarction. Contraception 1998;57:143-155. 7. Farley TM, Collins J, Schlesselman JJ. Hormonal contraception and risk of cardiovascular disease, an international perspective. Contraception 1998;57:211-230. 8. ACOG Committee on Practice Bulletins--Gynecology. ACOG practice bulletin no. 73. Use of hormonal contraception in women with coexisting medical conditions. Obstet Gyneco12006;107:1453-1472. 9. Cardiovascular disease and steroid hormone contraception. Report of a W H O scientific group. World Health Organ Tech Re.p Ser 1998; 877:1-89. 10. Burkman RT. Noncontraceptive clinical benefits of oral contraceptives. Int J Ferti11989;34(suppl):50- 55. 11. Thorneycroft IH. Cycle control with oral contraceptives: a review of the literature.Am J Obstet Gyneco11999;180(2 Pt 2):280-287. 12. Hatcher R, Trussell J, Stewart F, et al. Contraceptive technology, 16th rev ed. Manchester, NH: Irvington Publishers, 1994. 13. Farley TM. Life-table methods for contraceptive research. Star IVied 1986;5:475-489. 14. Schwartz D, Mayaux MJ. Female fecundity as a function of age: results of artificial insemination in 2193 nulliparous women with azoospermic husbands. Federation CECOS. N EnglJ Med 1982;306:404-406. 15. Thorneycroft IH. Update on androgenicity. Am J Obstet Gynecol 1999;180(2 Pt 2):288-294. 16. Grimes DA, Godwin AJ, Rubin A, Smith JA, Lacarra M. Ovulation and follicular development associated with three low-dose oral contraceptives: a randomized controlled trial. Obstet Gyneco11994;83:29-34. 17. Killick SR, Fitzgerald C, Davis A. Ovarian activity in women taking an oral contraceptive containing 20 microg ethinyl estradiol and 150 microg desogestrel: effects of low estrogen doses during the hormonefree interval. Am J Obstet Gynecol1998;179: S18- $24. 18. Lete I, Morales E Inhibition of follicular growth by two different oral contraceptives (monophasic and triphasic) containing ethinylestradiol and gestodene. EurJ Contrace.ptRe.prodHealth Care 1997;2:187-191. 19. Drife J. Benefits and risks of oral contraceptives. Adv Contrace.pt 1990;6 (suppl): 15 - 25. 20. Mishell DR Jr. Noncontraceptive benefits of oral contraceptives. J Reprod Med 1993;38(12 suppl):1021-1029. 21. The ESHRE Capri Workshop Group: noncontraceptive health benefits of combined oral contraception. Hum Reprod Update2005;11:513-525. 22. La Vecchia C, Tavani A, Franceschi S, Parazzini F. Oral contraceptives and cancer. A review of the evidence. Drug Saf1996;14:260-272. 23. Schlesselman JJ. Risk of endometrial cancer in relation to use of combined oral contraceptives. A practitioner's guide to meta-analysis. Hum Re.prod 1997;12:1851-1863. 24. Vessey MP, Painter R. Endometrial and ovarian cancer and oral contraceptives--findings in a large cohort study. BrJ Cancer 1995;71: 1340-1342. 25. Jick SS, Walker AM, Jick H. Oral contraceptives and endometrial cancer. Obstet Gyneco11993;82:931-935. 26. Grimes DA, Economy KE. Primary prevention of gynecologic cancers. d m J Obstet Gyneco11995;172(1 Pt 1):227-235. 27. Narod SA, Risch H, Moslehi R, et al. Oral contraceptives and the risk of hereditary ovarian cancer. Hereditary Ovarian Cancer Clinical Study Group. NEnglJMed 1998;339(7):424-428. 28. Shaw RW. Assessment of medical treatments for menorrhagia. Br J Obstet Gynaeco11994;101(suppl 11):15-18. 29. Iyer V, Farquhar C, Jepson R. Oral contraceptive pills for heavy menstrual bleeding. CochraneDatabase Syst Rev 2000:CD000154. 30. Gambacciani M, Spinetti A, Cappagli B, et al. Hormone replacement therapy in perimenopausal women with a low dose oral contraceptive preparation: effects on bone mineral density and metabolism. Maturitas 1994;19:125-131.
CHAPTER 12 Use of Contraceptives for Older W o m e n 31. Liu SL, Lebrun CM. Effect of oral contraceptives and hormone replacement therapy on bone mineral density in premenopausal and perimenopausal women: a systematic review. BrJ Sports Med 2006;40: 11-24. 32. Gambacciani M, Ciaponi M, Cappagli B, et al. Effects of low-dose, continuous combined hormone replacement therapy on sleep in symptomatic postmenopausal women. Maturitas 2005;50(2):91-97. 33. Kleerekoper M, Brienza RS, Schultz LR, Johnson CC. Oral contraceptive use may protect against low bone mass. Henry Ford Hospital Osteoporosis Cooperative Research Group. Arch Intern Med 1991;151: 1971-1976. 34. Tuppurainen M, Kroger H, Saarikoski S, Honkanen R, Alhava E. The effect of previous oral contraceptive use on bone mineral density in perimenopausal women. OsteoporosInt 1994;4:93-98. 35. Rosenberg L, Palmer JR, Sands MI, et al. Modern oral contraceptives and cardiovascular disease. Am J Obstet Gyneco11997;177:707- 715. 36. Sidney S, Petitti DB, Q.uesenberry CP Jr, et al. Myocardial infarction in users of low-dose oral contraceptives. Obstet Gynecol 1996;88: 939-944. 37. Lewis MA, Heinemann LA, Spitzer WO, MacRae KD, Bruppacher R. The use of oral contraceptives and the occurrence of acute myocardial infarction in young women. Results from the Transnational Study on Oral Contraceptives and the Health of Young Women. Contraception 1997;56:129-140. 38. Lewis MA, Spitzer WO, Heinemann LA, et al. Third generation oral contraceptives and risk of myocardial infarction: an international casecontrol study. Transnational Research Group on Oral Contraceptives and the Health of Young Women. Brit MedJ 1996;312:88-90. 39. Rosenberg L, Palmer JR, Shapiro S. Decline in the risk of myocardial infarction among women who stop smoking. N EnglJ Med 1990;322: 213-217. 40. McElduff P, Dobson A, Beaglehole R, Jackson R. Rapid reduction in coronary risk for those who quit cigarette smoking. Aust N Z J Public Health 1998;22:787- 791. 41. Lidegaard O, Kreiner S. Cerebral thrombosis and oral contraceptives. A case-control study. Contraception 1998;57:303 - 314. 42. Petitti DB, Sidney S, Q.uesenberry CP Jr, Bernstein A. Incidence of stroke and myocardial infarction in women of reproductive age. Stroke
1997;28:280-283. 43. Petitti DB, Sidney S, Bernstein A, et al. Stroke in users of low-dose oral contraceptives. N EnglJ Med 1996;335:8-15. 44. Ischaemic stroke and combined oral contraceptives: results of an international, multicentre, case-control study. W H O Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Lancet 1996;348:498-505. 45. Haemorrhagic stroke, overall stroke risk, and combined oral contraceptives: results of an international, multicentre, case-control study. W H O Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Lancet 1996;348:505-510. 46. Venous thromboembolic disease and combined oral contraceptives: results of international multicentre case-control study. World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Lancet 1995;346:1575-1582. 47. Spitzer WO, Lewis MA, Heinemann LA, Thorogood M, MacRae KD. Third generation oral contraceptives and risk of venous thromboembolic disorders: an international case-control study. Transnational Research Group on Oral Contraceptives and the Health of Young Women. Brit MedJ 1996;312:83 - 88. 48. Walker AM. Newer oral contraceptives and the risk of venous thromboembolism. Contraception 1998;57:169-181. 49. Lidegaard O, Edstrom B, Kreiner S. Oral contraceptives and venous thromboembolism: a five-year national case-control study. Contraception 2002;65:187-196. 50. Lidegaard O, Kreiner S. Contraceptives and cerebral thrombosis: a five-year national case-control study. Contraception 2002;65:197-205.
181 51. Jick H, Jick SS, Gurewich V, Myers MW, Vasilakis C. Risk of idiopathic cardiovascular death and nonfatal venous thromboembolism in women using oral contraceptives with differing progestagen components. Lancet 1995;346:1589-1593. 52. Lewis MA, Heinemann LA, MacRae KD, Bruppacher R, Spitzer WO. The increased risk of venous thromboembolism and the use of third generation progestagens: role of bias in observational research. The Transnational Research Group on Oral Contraceptives and the Health of Young Women. Contraception 1996;54:5-13. 53. Lewis MA. The epidemiology of oral contraceptive use: a critical review of the studies on oral contraceptives and the health of young women. Am J Obstet Gyneco11998;179:1086-1097. 54. Suissa S, Blais L, Spitzer WO, et al. First-time use of newer oral contraceptives and the risk of venous thromboembolism. Contraception 1997;56:141-146. 55. Spitzer WO. The 1995 pill scare revisited: anatomy of a non-epidemic. Hum Reprod 1997;12:2347-2357. 56. Spitzer WO. Bias versus causality: interpreting recent evidence of oral contraceptive studies, d m J Obstet Gyneco11998;179(3 Pt 2):$43-50. 57. Marchbanks PA, McDonald JA, Wilson HG, et al. Oral contraceptives and the risk of breast cancer. NEnglJMed 2002;346:2025-2032. 58. Rosenberg M. Weight change with oral contraceptive use and during the menstrual cycle. Results of daily measurements. Contraception 1998;58:345-349. 59. Reubinoff BE, Grubstein A, Meirow D, et al. Effects of low-dose estrogen oral contraceptives on weight, body composition, and fat distribution in young women. Fertil Steri11995;63:516-521. 60. Gallo MF, Grimes DA, Schulz KF, Helmerhorst FM. Combination estrogen-progestin contraceptives and body weight: systematic review of randomized controlled trials. Obstet Gyneco12004;103(2):359-373. 61. Archer DF, Maheux R, DelConte A, O'Brien FB. A new low-dose monophasic combination oral contraceptive (Alesse) with levonorgestrel 100 micrograms and ethinyl estradiol 20 micrograms. North American Levonorgestrel Study Group (NALSG). Contraception 1997;55:139-144. 62. Sulak PJ, Scow RD, Preece C, Riggs MW, Kuehl TJ. Hormone withdrawal symptoms in oral contraceptive users. Obstet Gyneco12000;95: 261-266. 63. Sulak PJ, Cressman BE, Waldrop E, Holleman S, Kuehl TJ. Extending the duration of active oral contraceptive pills to manage hormone withdrawal symptoms. Obstet Gyneco11997;89:179-183. 64. Edelman A, Gallo ME Nichols MD, et al. Continuous versus cyclic use of combined oral contraceptives for contraception: systematic Cochrane review of randomized controlled trials. Hum Reprod 2006;21: 573-578. 65. Edelman AB, Gallo MF, Jensen JT, et al. Continuous or extended cycle vs. cyclic use of combined oral contraceptives for contraception. Cochrane Database Syst Rev 2005;3:CD004695. 66. Kwiecien M, Edelman A, Nichols MD, Jensen JT. Bleeding patterns and patient acceptabifity of standard or continuous dosing regimens of a lowdose oral contraceptive: a randomized trial. Contraception2003;67:9-13. 67. Anderson FD, Halt H. A multicenter, randomized study of an extended cycle oral contraceptive. Contraception 2003;68:89-96. 68. Miller L, Hughes JP. Continuous combination oral contraceptive pills to eliminate withdrawal bleeding: a randomized trial. Obstet Gynecol 2003;101:653-661. 69. Macgregor EA, Hackshaw A. Prevention of migraine in the pill-free interval of combined oral contraceptives: a double-blind, placebocontrolled pilot study using natural oestrogen supplements, y Fam Plann Reprod Health Care 2002;28:27-31. 70. Curtis KM, Mohllajee AP, Peterson HB. Use of combined oral contraceptives among women with migraine and nonmigrainous headaches: a systematic review. Contraception 2006;73:189-194. 71. Mohllajee AP, Curtis KM, Martins SL, Peterson HB. Does use of hormonal contraceptives among women with thrombogenic mutations increase their risk of venous thromboembolism? A systematic review. Contraception 2006;73:166-178.
1
8
2
D
A
72. Mattson RH, Rebar RW. Contraceptive methods for women with neurologic disorders. Am J Obstet Gyneco11993;168(6 Pt 2):2027-2032. 73. Lee NC, Rubin GL, Borucki R. The intrauterine device and pelvic inflammatory disease revisited: new results from the Women's Health Study. Obstet Gyneco11988;72:1-6. 74. Kronmal RA, Whitney CW, Mumford SD. The intrauterine device and pelvic inflammatory disease: the Women's Health Study reanalyzed. J Clin Epidemio11991;44:109-122. 75. Grodstein F, Rothman KJ. Epidemiology of pelvic inflammatory disease. Epidemiology 1994;5:234-242. 76. Archer DE Reversible contraception for the woman over 35 years of age. Curr Opin Obstet Gyneco11992;4:891- 896. 77. Farley TM, Rosenberg MJ, Rowe PJ, Chen JH, Meirik O. Intrauterine devices and pelvic inflammatory disease: an international perspective. Lancet 1992;339:785- 788. 78. Ortiz ME, Croxatto HB, Bardin CW. Mechanisms of action of intrauterine devices. Obstet GynecolSurv 1996;51(12 suppl):S42-51. 79. Spinnato JA 2nd. Mechanism of action of intrauterine contraceptive devices and its relation to informed consent. Am J Obstet Gyneco11997;176: 503-506. 80. Ramey JW, Starke ME, Gibbons WE, Archer DE The influence of pentoxifyUine (Trental) on the antifertility effect of intrauterine devices in rats. Fertil Steri11994;62:181-185. 81. Chi IC. The multiload I U D - - a U.S. researcher's evaluation of a European device. Contraception 1992;46:407-425. 82. Kimmerle R, Weiss R, Berger M, Kurz KH. Effectiveness, safety, and acceptability of a copper intrauterine device (CU Safe 300) in type I diabetic women. Diabetes Care 1993;16:1227-1230. 83. Rosenberg MJ, Foldesy R, Mishell DR Jr, et al. Performance of the TCu380A and Cu-Fix IUDs in an international randomized trail. Contraception 1996;53:197-203. 84. Sivin I, Shaaban M, Odlind V, et al. A randomized trial of the Gyne T 380 and Gyne T 380 Slimline Intrauterine Copper devices. Contraception 1990;42:379-389. 85. Nilsson CG, Haukkamaa M, Vierola H, Luukkainen T. Tissue concentrations of levonorgestrel in women using a levonorgestrel-releasing IUD. Clin Endocrinol (Oxf) 1982;17:529-536. 86. Silverberg SG, Haukkamaa M, Arko H, Nilsson CG, Luukkainen T. Endometrial morphology during long-term use of levonorgestrelreleasing intrauterine devices. IntJ GynecolPatho11986;5:235-241. 87. Lethaby AE, Cooke I, Rees M. Progesterone or progestogen-releasing intrauterine systems for heavy menstrual bleeding. CochraneDatabase Syst Rev 2005:CD002126. 88. Kaunitz AM. Long-acting contraceptive options. Int J Fertil Menopausal Stud 1996;41:69- 76. 89. Cundy T, Ames R, Home A, et al. A randomized controlled trial of estrogen replacement therapy in long-term users of depot medroxyprogesterone acetate. J Clin EndocrinolMetab 2003;88:78- 81. 90. Tang OS, Tang G, Yip P, Li B, Fan S. Long-term depot-medroxyprogesterone acetate and bone mineral density. Contraception 1999;59: 25-29. 91. Fraser IS, Jansen RE Why do inadvertent pregnancies occur in oral contraceptive users? Effectiveness of oral contraceptive regimens and interfering factors. Contraception 1983;27:531 - 551. 92. Said S, Sadek W, Rocca M, et al. Clinical evaluation of the therapeutic effectiveness of ethinyl oestradiol and oestrone sulphate on prolonged bleeding in women using depot medroxyprogesterone acetate for contraception. World Health Organization, Special Programme of Research, Development and Research Training in Human Reproduction, Task Force on Long-acting Systemic Agents for Fertility Regulation. Hum Reprod 1996;ll(suppl 2):1-13. 93. Fraser IS. A survey of different approaches to management of menstrual disturbances in women using injectable contraceptives. Contraception 1983;28:385-397.
V
I
D
E ARCHER
94. Cromer BA, Blair JM, Mahan JD, Zibners L, Naumovski Z. A prospective comparison of bone density in adolescent girls receiving depot medroxyprogesterone acetate (Depo-Provera), levonorgestrel (Norplant), or oral contraceptives.J Pediatr 1996;129:671-676. 95. Skegg DC, Noonan EA, Paul C, et al. Depot medroxyprogesterone acetate and breast cancer. A pooled analysis of the World Health Organization and New Zealand studies. JWMA 1995;273:799-804. 96. Long-term use of hormonal contraceptive DMPA not linked to breast cancer. Prog Hum Reprod Res 1995:5. 97. Depot-medroxyprogesterone acetate (DMPA) and risk of endometrial cancer. The WHO Collaborative Study of Neoplasia and Steroid Contraceptives. Int J Cancer 1991;49:186-190. 98. Kaunitz AM. Depot medroxyprogesterone acetate contraception and the risk of breast and gynecologic cancer. J Reprod Med 1996;41 (5 suppl):419-427. 99. Westhoff C, Wieland D, Tiezzi L. Depression in users of depomedroxyprogesterone acetate. Contraception 1995;51:351 - 354. 100. Westhoff C, Truman C, Kalmuss D, et al. Depressive symptoms and Depo-Provera. Contraception 1998;57:237-240. 101. Sivin I, Alvarez F, Mishell DR Jr, et al. Contraception with two levonorgestrel rod implants. A 5-year study in the United States and Dominican Republic. Contraception 1998;58:275-282. 102. Sivin I, Wan L, Ranta S, et al. Levonorgestrel concentrations during 7 years of continuous use of Jadelle contraceptive implants. Contraception 2001;64:43-49. 103. Lifchez AS, Scommegna A. Diffusion of progestogens through Silastic rubber implants. Fertil Steri11970;21:426-430. 104. Sivin I, Viegas O, Campodonico I, et al. Clinical performance of a new two-rod levonorgestrel contraceptive implant: a three-year randomized study with Norplant implants as controls. Contraception 1997;55:73-80. 105. Archer DF, Philput CA, Weber ME. Management of irregular uterine bleeding and spotting associated with Norplant. Hum Reprod 1996;ll(suppl 2):24-30. 106. Diaz J, Faundes A, Olmos P, Diaz M. Bleeding complaints during the first year of norplant implants use and their impact on removal rate. Contraception 1996;53:91-95. 107. Sivin I. Risks and benefits, advantages and disadvantages oflevonorgestrel-releasing contraceptive implants. Drug Saf2003;26:303- 335. 108. DugoffL, Jones OW 3rd, Allen-Davis J, Hurst BS, SchlaffWD. Assessing the acceptability of Norplant contraceptive in four patient populations. Contraception 1995;52:45-49. 109. Tang GW, Lo SS. Levonorgestrel intrauterine device in the treatment of menorrhagia in Chinese women: efficacy versus acceptability. Contraception 1995;51:231-235. 110. Sivin I, Mishell DR Jr, Darney P, Wan L, Christ M. Levonorgestrel capsule implants in the United States: a 5-year study. Obstet Gynecol 1998;92:337-344. 111. Bennink HJ. The pharmacokinetics and pharmacodynamics of Implanon, a single-rod etonogestrel contraceptive implant. EurJ Contracept Reprod Health Care 2000;5(suppl 2):12-20. 112. Makarainen L, van Beek A, Tuomivaara L, Asplund B, Coelingh Bennink H. Ovarian function during the use of a single contraceptive implant: Implanon compared with Norplant. Fertil Steril 1998;69: 714-721. 113. Funk S, Miller MM, Mishell DR Jr, et al. Safety and efficacy oflmplanon, a single-rod implantable contraceptive containing etonogestrel. Contraception 2005;71:319-326. 114. Le J, Tsourounis C. Implanon: a critical review..~Inn Pbarmacotber 2001;35:329-336. 115. Zheng SR, Zheng HM, Qian SZ, Sang GW, Kaper RF. A randomized multicenter study comparing the efficacy and bleeding pattern of a single-rod (Implanon) and a six-capsule (Norplant) hormonal contraceptive implant. Contraception 1999;60:1-8.
CHAPTER 12 Use of Contraceptives for Older Women 116. Weisberg E, Hickey M, Palmer D, et al. A pilot study to assess the effect of three short-term treatments on frequent and/or prolonged bleeding compared to placebo in women using Implanon. Hum Reprod 2006;21:295- 302. 117. Abrams LS, Skee DM, Natarajan J, Wong FA, Lasseter KC. Multiple-dose pharmacokinetics of a contraceptive patch in healthy women participants. Contraception 2001;64:287-294. 118. Zieman M, Guillebaud J, Weisberg E, et al. Contraceptive efficacy and cycle control with the Ortho Evra/Evra transdermal system: the analysis of pooled data. Fertil Steri12002;77(2 suppl 2):$13-18. 119. Archer DF, Bigrigg A, Smallwood GH, et al. Assessment of compliance with a weekly contraceptive patch (Ortho Evra/Evra) among North American women. Fertil Steri12002;77(2 suppl 2):$27-31. 120. Zacur HA, Hedon B, Mansour D, et al. Integrated summary of Ortho Evra/Evra contraceptive patch adhesion in varied climates and conditions. Fertil Steri12002;77(2 suppl 2):$32-35. 121. Sibai BM, Odlind V, Meador ML, et al. A comparative and pooled analysis of the safety and tolerability of the contraceptive patch (Ortho Evra/Evra). Fertil Steri12002;77(2 suppl 2):$19-26. 122. Abrams LS, Skee D, Natarajan J, Wong FA. Pharmacokinetic overview of Ortho Evra/Evra. Fertil Steri12002;77(2 suppl 2):$3-12. 123. Jick SS, KayeJA, Russmann S, Jick H. Risk of nonfatal venous thromboembolism in women using a contraceptive transdermal patch and oral contraceptives containing norgestimate and 35 microg of ethinyl estradiol. Contraception 2006;73:223-228. 124. Ballagh SA. Vaginal ring hormone delivery systems in contraception and menopause. Clin Obstet Gyneco12001;44:106-113. 125. Mishell DR Jr, Lumkin M, Jackanicz T. Initial clinical studies of intravaginal rings containing norethindrone and norgestrel. Contraception 1975;12(3):253-260. 126. Alvarez-Sanchez F, Brache V, Jackanicz T, Faundes A. Evaluation of four different contraceptive vaginal rings: steroid serum levels, luteal activity, bleeding control and lipid profiles. Contraception 1992;46(4): 387-398. 127. Roumen E Contraceptive efficacy and tolerability with a novel combined contraceptive vaginal ring, NuvaRing. Eur J Contracept Reprod Health Care 2002;7 (suppl 2):19-24; discussion 37-39. 128. Roumen FJ, Apter D, Mulders TM, Dieben TO. Efficacy, tolerability and acceptability of a novel contraceptive vaginal ring releasing etonogestrel and ethinyl oestradiol. Hum Reprod 2001;16:469-475. 129. Mulders TM, Dieben TO. Use of the novel combined contraceptive vaginal ring NuvaRing for ovulation inhibition. Fertil Steril 2001; 75:865 - 870. 130. Veres S, Miller L, Burington B. A comparison between the vaginal ring and oral contraceptives. Obstet Gyneco12004;104:555-563. 131. TrusseUJ. Efficacy of barrier contraceptives. In: Mauck CK, Cordero M, Gabelnick HL, Spieler JM, Rivera R, eds. Barrier contraceptives, current status andfuture prospects. New York: Wiley-Liss;1994:22. 132. Doncel GE Chemical vaginal contraceptives: preclinical evaluation. In: Mauck CK, Cordero M, Gabelnick HL, Spieler JM, Rivera R, eds. Barrier contraceptives, current status andfuture prospects. New York: Wiley-Liss;1994:147-162.
183 133. Mauck CK, Baker JM, Barr SP, Johanson W, Archer DE A phase I study of Femcap used with and without spermicide. Postcoital testing. Contraception 1997;56:111-115. 134. Mauck CK, Baker JM, Barr SP, Johanson WM, Archer DE A phase I comparative study of three contraceptive vaginal films containing nonoxynol-9. Postcoital testing and colposcopy. Contraception 1997;56: 97-102. 135. Archer DF, Mauck CK, Viniegra-Sibal A, Anderson FD. Lea's Shield: a phase I postcoital study of a new contraceptive barrier device. Contraception 1995;52:167-173. 136. Mauck C, Glover LH, Miller E, et al. Lea's Shield: a study of the safety and efficacy of a new vaginal barrier contraceptive used with and without spermicide. Contraception 1996;53:329-335. 137. Farr G, Katz V, Spivey SK, et al. Safety, functionality and acceptability of a prototype polyurethane condom. Adv Contracept 1997;13: 439-451. 138. Frezieres RG, Walsh TL, Nelson AL, Clark VA, Coulson AH. Breakage and acceptability of a polyurethane condom: a randomized, controlled study. Faro Plann Perspect 1998;30:73-78. 139. Rosenberg MJ, Waugh MS, Solomon HM, Lyszkowski AD. The male polyurethane condom: a review of current knowledge. Contraception 1996;53:141-146. 140. Farr G, Gabelnick H, Sturgen K, Dorflinger L. Contraceptive efficacy and acceptability of the female condom. Am J Public Health 1994;84: 1960-1964. 141. Elias CJ, Coggins C. Female-controlled methods to prevent sexual transmission of HIV. Aids 1996;10(suppl 3):$43-51. 142. Berer M. Adverse effects of nonoxynol-9. Lancet 1992;340:615-616. 143. Feldblum PJ. Self-reported discomfort associated with use of different nonoxynol-9 spermicides. Genitourin Med 1996;72:451-452. 144. McGroarty JA, Reid G, Bruce AW. The influence of nonoxynol9-containing spermicides on urogenital infection. J Urol 1994;152: 831-833. 145. Roddy RE, Cordero M, Cordero C, Fortney JA. A dosing study of nonoxynol-9 and genital irritation. Int J STD AIDS 1993;4: 165-170. 146. Stafford MK, Ward H, Flanagan A, et al. Safety study of nonoxynol-9 as a vaginal microbicide: evidence of adverse effects. JAcquir Immune Defic Syndr Hum Retroviro11998;17:327- 331. 147. Steiner MJ, Cates W Jr. Condoms and urinary tract infections: is nonoxynol-9 the problem or the solution? Epidemiology 1997;8: 612-614. 148. Howett MK, Neely EB, Christensen ND, et al. A broad-spectrum microbicide with virucidal activity against sexually transmitted viruses. Antimicrob Agents Chemother 1999;43:314-321. 149. Roddy RE, Zekeng L, Ryan KA, et al. A controlled trial of nonoxynol 9 film to reduce male-to-female transmission of sexually transmitted diseases. N EnglJ IVied 1998;339:504- 510. 150. Cook RL, Rosenberg MJ. Do spermicides containing nonoxynol-9 prevent sexually transmitted infections? A meta-analysis. Sex Transm Dis 1998;25:144-150.
This Page Intentionally Left Blank
SECTION IV
Changes OccurringAfter Menopause In this section on changes occurring after menopause, the intent is to provide a review of some of the major changes affecting women after the perimenopausal period described in the previous section. Missing here, but of sufficient importance to warrant separate dedicated sections, are the concerns for cardiovascular health, cancer, and osteoporosis. An in-depth understanding of many of the physiologic changes that occur will be valuable for clinicians in counseling women and offering possible treatments. Robert R. Freedman begins with an in-depth discussion of hot flushes, the classic symptom of postmenopausal women. Next, Frederick Naftolin and colleagues review the effects of sex steroids on the brain and how changes in these levels may affect brain chemistry and symptomatology. This theme is continued with Barbara B. Sherwin's discussion of psychologic functioning, which should be paired with David R. Rubinow's discussion of such changes in perimenopausal women in Chapter 24. Mark Brincat discusses collagen in Chapter 16 and describes the importance of collagen content for the skin and skeleton and how menopause and sex steroids may influence collagen content. In the next chapter, H. Irving Katz describes skin changes after menopause and differentiates between normal aging and the influence of hormones. G6ran Samsioe next describes the importance of changes in the urinary system after menopause and how this affects quality of life. In this edition a new section has been created that will delve more deeply into female urology and pelvic support, which are of great importance to older women. Next Gloria Bachmann describes vulvovaginal complaints after menopause and atrophic changes that accompany the changes in the hormonal milieu after menopause. On a related but different topic, Lorraine Dennerstein describes sexuality around the time of menopause and the possible influence of hormones.
This Page Intentionally Left Blank
~ H A P T E R 1;
Menopausal Hot Flashes ROBERT R.
FREEDMAN
Departments of Obstetrics and Gynecology and Psychiatry and Behavioral Neurosciences, Wayne State University School of Medicine, Detroit, MI 48201
I. EPIDEMIOLOGY OF H O T FLASHES
answers to these questions are unknown and represent important avenues for further research.
Hot flashes are the most common symptom of the menopause and occur in the vast majority of postmenopausal women. Their prevalence among naturally menopausal women has been reported to be 68% (1) to 82% (2) in the United States, 60% in Sweden (3), and 62% in Australia (4), with a median age of onset of approximately 51 years (5). Among ovariectomized women, the prevalence of hot flashes is approximately 90% (2,6). In one study, Feldman et al. (2) found that 64% of women experienced hot flashes for 1 to 5 years, and Kronenberg reported the median length of the symptomatic period to be 4 years. Studies of risk factors for menopausal hot flashes have found few strong effects. There is some evidence that heavier women are more likely to report hot flashes because increased body fat raises core body temperature (7). No significant association has been found between the report of hot flashes and socioeconomic status, age, race, parity, age at menarche, age at menopause, or number of pregnancies (5). Cultural factors do affect the reporting of hot flashes. Compared with Western women, women from Indonesia report hot flashes at rates of only 10% to 20% (8,9), Chinese women at rates of 10% to 25% (10), and Mexican Mayan women not at all (11). Reasons for these findings are not known. Perhaps women from rural and non-Western cultures demonstrate physiologically defined hot flashes as frequently as Western women but are acculturated in some way to not report them. Or they may actually have fewer physiologically defined flashes. This could be due to factors such as diet, because some foods such as yams and soy products contain substantial amount of phytoestrogens, which may help ameliorate hot flashes (12). The TREATMENT OF THE POSTMENOPAUSAL WOMAN
II. DESCRIPTIVE PHYSIOLOGY A. Self-Reported Data Kronenberg (5) conducted an extensive questionnaire study of hot flashes in 506 women ranging in age from 29 to 82. Of those reporting current symptoms, 87% had daily hot flashes, and one-third of these reported more than 10 per day. Hot flashes generally lasted 1 to 5 minutes, with about 6% lasting more than 6 minutes. About 40% of the women recognized a premonition that a hot flash was about to begin. The experience of a hot flash was most often described as sensations of heat, sweating, flushing, chills, clamminess, and anxiety. Sweating was reported most often in the face, head, neck, and chest, but rarely in the lower body.
B. Physiologic Data 1. SKINTEMPERATURE AND BLooD FLow Peripheral vasodilation, as evidenced by increased skin temperature, occurs during hot flashes in all areas that have been measured. These areas include the fingers, toes, cheek, forehead, forearm, upper arm, chest, abdomen, back, calf, and thigh (13-16). Finger blood flow (14) and hand, calf, and forearm blood flow (17) also increase during hot flashes. Thermographic measurements during hot flashes yielded data similar to those obtained with skin temperature (18).
187
Copyright 9 2007 by Elsevier,Inc. M1 rights of reproductionin any form reserved.
1
8
8
R
O
B
2. SWEATINGAND SKIN CONDUCTANCE
Sweating and skin conductance, an electrical measure of sweating, also increase during hot flashes (Fig. 13.1). Molnar (13) obtained sweat prints with iodized paper during hot flashes and reported profuse sweating on the forehead and nose, moderate sweating on the sternum and adjacent areas, and little or none on the cheek and leg. The total body sweat rate was estimated to be about 1.3 g/min. In our laboratory, we measured sweat rate and skin conductance simultaneously from the sternum (16). Sweat rate was recorded by capacitance hygrometry using a 3.5 cm diameter plastic chamber attached over the sternum. Compressed air, regulated at 200 mL/min, was dried over CaCO2 and passed through the chamber. Skin conductance level was also recorded from the sternum using a 0.5-volt constant voltage circuit and disposable Ag/AgC1 electrodes. Both measures increased significantly during 29 hot flashes recorded in 14 women (Fig. 13.2). Measur-
E
R
T
R. FREEDMAN
able sweating occurred during 90% of the flashes, and there was a close time correspondence between both measures. 3. CORE BODYTEMPERATURE
Homeotherms regulate core body temperature between upper thresholds, where sweating and peripheral vasodilation occur, and a lower threshold, where shivering occurs. If core body temperature were elevated in women with hot flashes, their symptoms of sweating and peripheral vasodilation could be explained. However, measurements of esophageal (13), rectal (13), and tympanic (15) temperatures were not elevated prior to hot flashes. These studies all found declines of about 0.3~ following hot flashes, probably due to increased heat loss (peripheral vasodilation) and evaporative cooling (sweating). However, esophageal and rectal temperatures have long thermal lag times and might respond too slowly to appear along with the rapid peripheral events of the hot flash (19). Addi-
FIGURE 13.1 Peripheralphysiologice v e n t s of the hot flash. (Data from ref. 16. Drawingby Jeri Pajor.)
CHAPTER 13 Menopausal Hot Flashes
189
FIGURE 13.3
Radiotelemetry pill.
had not significantly changed. These results were replicated during a daytime study in the laboratory (16).
4. METABOLIC RATE
FIGURE 13.2 Time course of skin conductance and sweating in 29 hot flashes. (From res 16.)
tionaUy, it has been shown that tympanic temperature does not reliably measure core body temperature because it is affected by peripheral vasodilation and sweating (20). We therefore conducted several studies in which we measured core body temperature using an ingested radiotelemetry pill, which has a faster response time than the esophageal and rectal methods (Fig. 13.3). The pill is swallowed 90 minutes before an experiment, to allow its egress from the stomach, and the signals are detected and stored in a small digital recorder. The typical transit time through the gut is about 24 to 72 hours, during which the recorder samples the data every 30 seconds. Hot flashes are recorded on a separate device, using sternal skin conductance level as the marker. In the first study, 10 symptomatic women were recorded using ambulatory monitoring for 24 hours (21). Of 77 hot flashes recorded, 46 (60%) were preceded by small but significant increases in core body temperature. In a second study, conducted during sleep in a temperature-controlled laboratory, 37 hot flashes occurred in 8 postmenopausal women (22). Significant core temperature elevations preceded 24 of the flashes (65%), whereas rectal temperature
Elevations in core body temperature can be caused by increased metabolic rate (heat production) and by peripheral vasoconstriction (decreased heat loss). In the last study (16) we sought to determine if either of these factors accounted for the core body temperature elevations preceding hot flashes. Twenty-nine flashes were recorded in 14 postmenopausal women. Significant elevations in metabolic rate (about 15%) occurred but were simultaneous with sweating and peripheral vasodilation and did not precede the core temperature elevations. Peripheral vasoconstriction did not occur. Thus, increased metabolic rate and peripheral vasoconstriction did not account for the core body temperature elevations in these women. 5. HEART RATE
Modest increases in heart rate, about 7 to 15 beats/min (13,23,24), occur at approximately the same time as the peripheral vasodilation and sweating.
III. OBJECTIVE MEASUREMENT OF HOT FLASHES
A. In the Laboratory 1. FINGER TEMPERATURE
Temperature from the dorsum of one finger was proposed as the first physiologic marker for menopausal hot flashes (25). In 7 symptomatic women, 41 skin temperature elevations greater than 1~ occurred within approximately
1
9
0
R
O
B
1 minute of the subjective hot flash. However, the duration of the temperature elevations averaged 31 minutes, whereas the duration of subjective flushing was 2.3 minutes. Also, precise definitions of the onset and offset of the temperature elevations were not reported. 2. SKIN CONDUCTANCE
Subsequently, skin conductance recorded from the sternum was investigated as a hot flash marker. Tataryn et al. (26) found that 98% of 128 subjective flashes in 8 postmenopausal women were accompanied by elevations in sternal skin conductance, compared with 82% for finger temperature and 81% for decreased tympanic temperature. All these changes were significantly reduced by estrogen administration in 4 of the women. However, the precise characteristics of the skin conductance responses were not defined. Our laboratory subsequently sought to determine these characteristics (24). Sternal skin conductance level, finger temperature, and heart rate were recorded for 4 hours in 11 postmenopausal and 8 premenopausal women. Twentynine subjective hot flashes were indicated by pushbutton in the first group. All these were accompanied by an increase in sternal skin conductance >-2 gmho/30 sec. One skin conductance elevation occurred without a button press. All skin conductance elevations occurred within 66 seconds of the button press. No skin conductance elevations occurred in the premenopausal women. Thus, there was a concordance of 95% between the skin conductance criterion and the subjects' reports. Significant elevations in skin temperature and heart rate occurred during the flashes but were not as sensitive or specific as the skin conductance elevations. We replicated these findings in 18 symptomatic and 8 asymptomatic postmenopausal women (27). There was a concordance of 80% between the sternal skin conductance criterion (2 }amho/30 sec) and the subjective reports (button press) in 15 flashes recorded in the symptomatic women. No events occurred in the asymptomatic women. Our findings were then independently replicated by another laboratory (28). In two separate studies of 20 symptomatic women, a concordance of 90% was obtained between the sternal skin conductance criterion and subjective reports. Measurements of finger temperature and blood flow were less predictive and did not improve the concordance rate when added to the skin conductance measure.
E
R
T
R. FREeDMaN
24 hours. Using the same basic circuit and electrodes, we found a concordance of 86% between the skin conductance criterion and button presses in 43 flashes recorded in seven symptomatic women (24). No such changes occurred in the eight premenopausal women. We replicated these findings in a second study (27). A concordance of 77% was obtained in 149 flashes recorded in 10 symptomatic women. Twelve skin conductance responses occurred in 8 putative asymptomatic women, representing a false response rate of about 8%. These ambulatory monitoring procedures were then successfully used to demonstrate the efficacy of a behavioral treatment for hot flashes in two subsequent studies (29,30). More recently, a smaller solid-state recorder has become available (UFI Biolog) that will continuously record hot flashes for up to 7 days. However, the skin conductance level (SCL) electrodes must be replaced every 1 or 2 days. Use of this device in symptomatic breast cancer survivors resulted in a concordance rate of 70% between SCLdetected hot flashes and event marks (31).
C. Provocation Techniques For laboratory investigations, it would be useful to reliably provoke hot flashes as opposed to waiting for them to occur during extended recording periods. Sturdee (23) observed that peripheral warming provoked objective and subjective hot flashes in seven of eight symptomatic women. We therefore sought to operationally define this procedure. Two 40- • circulating water pads maintained at 42~ were placed on the torso of 11 supine symptomatic women in a 23~ room (24). Eight hot flashes occurred within 30 minutes. A concordance of 73% was obtained between the skin conductance criterion (2 btmho/30 sec) and subjective report (button press). These findings were replicated in a subsequent study in 14 symptomatic women with a concordance of 84%. In this study, 25 hot flashes occurred during a 45-minute heating period. No objective or subjective responses occurred in eight asymptomatic women.
IV. ENDOCRINOLOGY A. Estrogens
B. Ambulatory Monitoring To evaluate treatment studies it would be useful to have a method that could be used outside the laboratory over longer periods. We therefore developed methods for recording sternal skin conductance on ambulatory monitors for
Because hot flashes accompany the decline of estrogens in the vast majority of naturally and surgically menopausal women, there is little doubt that estrogens play a role in the genesis of hot flashes. However, estrogens alone do not appear responsible for hot flashes because there is no correlation between the presence of this symptom and plasma (32),
CHAPTER 13 Menopausal Hot Flashes urinary (33), or vaginal (33) concentrations. No differences in unconjugated plasma estrogen concentrations were found in symptomatic versus asymptomatic women (34). AdditionaUy, clonidine significantly reduces hot flash frequency without altering circulating estrogen values (35). Prepubertal girls have low estrogen production without hot flashes, and hot flashes occur in the last trimester of pregnancy when estrogen production is high. Nevertheless, estrogen administration in hormone replacement therapy virtually eliminates hot flashes (36,37).
B. Gonadotropins Because gonadotropins become elevated at menopause, their possible role in the initiation of hot flashes has been investigated. Although no differences in leuteinizing hormone (LH) concentrations were found between women with and without hot flashes (38), a temporal association was found between LH pulses and hot flash occurrence (39,40). However, subsequent investigation revealed that women with a defect of gonadotropin-releasing hormone (GnRH) secretion (isolated gonadotropin deficiency) had hot flashes but no LH pulses and women with abnormal input to GnRH neurons (hypothalamic amenorrhea) had some LH pulses but no hot flashes (41). Additionally, hot flashes occur in hypophysectomized women, who have no LH release (42); in women with pituitary insufficiency and hypoestrogenism (43); and in women with LH release suppressed by GnRH analog treatment (44,45). Thus, LH cannot be the basis for hot flashes.
C. Opiates It was observed that alcohol-induced flushing in subjects taking chlorpropamide, a drug that stimulates insulin release and lowers blood glucose, was related to opiate receptor activation (46). Lightman et al. (47) subsequently found that naloxone infusion significantly reduced hot flash and LH pulse frequencies in six postmenopausal women. However, DeFazio et al. (48) attempted to replicate this study and found no effects. Tepper et al. (49) found that plasma [3endorphin concentrations decreased significantly before menopausal hot flashes, whereas Genazzani et al. (50) found significantly increased values preceding hot flashes. Thus, there is no consistent evidence of the involvement of an opioidergic system in menopausal hot flashes.
D. Catecholamines There is considerable evidence that norepinephrine plays an important role in thermoregulation mediated, in part through ot2-adrenergic receptors (51). Injection of norepi-
191 nephrine into the preoptic hypothalamus causes peripheral vasodilation, heat loss, and a subsequent decline in core body temperature (51). Additionally, there is considerable evidence that gonadal steroids modulate central noradrenergic activity (52). Studies of plasma norepinephrine have not found increased concentrations prior to or during hot flashes (14,39). However, brain norepinephrine content cannot be measured in plasma, due to the large amounts derived from peripheral organs (53). We therefore measured plasma 3-methoxy-4-hydroxyphenylglycol (MHPG), the main metabolite of brain norepinephrine, to determine if central norepinephrine concentrations were elevated during hot flashes (29). We studied 13 symptomatic and 6 asymptomatic postmenopausal women who were supine, with an intravenous (IV) line in place, in a 23~ room. Blood samples were drawn at the beginning and end of a 60-minute period and during a hot flash, if one occurred. The same procedures were followed during a 45-minute heating period. Basal M H P G levels were significantly higher in the symptomatic women (p < 0.0001) and increased significantly during resting and heat-induced flashes. There were no hot flashes or significant M H P G changes in the asymptomatic women, whose blood drawing times were yoked to those of six symptomatic women. However, approximately 50% of the free M H P G that enters the blood is metabolized peripherally to vanillylmandelic acid (VMA), and VMA formation can compete with M H P G production (54). Thus, fluctuations in peripheral VMA formation could potentially distort measurements of plasma MHPG. Therefore, we measured both compounds simultaneously before and after hot flashes in 14 symptomatic women (16). Plasma M H P G concentrations increased significantly (p < 0.02) between the preflash (3.7 ng/mL _+ 1.4) and postflash (5.1 _+ 2.3) blood samples, whereas VMA levels did not significantly change (6.2 + 1.8 ng/mL versus 6.1 + 2.5). However, more recent research (53) has shown that only a minority of plasma M H P G is derived from brain, with the majority coming from skeletal muscle. Therefore, the measurement of plasma MHPG, in and of itself, cannot be used to indicate central nervous system (CNS) activation but does represent whole-body sympathetic activation. Clonidine, an ot2-adrenergic agonist, reduces central noradrenergic activation and hot flashes (55-57). Yohimbine, an cx2-adrenergic antagonist, increases central noradrenergic activation. We sought to determine if clonidine would ameliorate hot flashes and if yohimbine would provoke them in controlled laboratory conditions (58). Nine symptomatic postmenopausal women, ages 43 to 63 years, served as subjects. Six asymptomatic women, ages 46 to 61 years, served as a comparison group. All women were in good health and had been amenorrheic for 2 years or more. In two blind laboratory sessions, subjects received either intravenous clonidine HC1 (l~tg/kg) or placebo followed by a 60-minute waiting period and then by 45 minutes of
192
FIGURE 13.4 A hot flash, indicated by a sternal skin conductance response, occurred after intravenous infusion of 0.032 mg/kg yohimbine in a menopausal women with hot flashes. No responses occurred in the matched placebo session or in an asymptomatic woman given higher doses. (From ref. 58.) peripheral heating. In two additional blind sessions, subjects received yohimbine HC1 (0.032 to 0.128 mg/kg IV) or placebo. Clonidine significantly (p = 0.01) increased the length of heating time needed to provoke a hot flash compared with placebo (40.6 ___ 3.0 minutes versus 33.6 ___ 3.6) and reduced the number of hot flashes that did occur (2 versus 8) (Fig. 13.4). In the symptomatic women, six hot flashes occurred during the yohimbine sessions and none during the corresponding placebo sessions, a statistically significant difference (p < 0.015). No hot flashes occurred in the asymptomatic women during either session (Fig. 13.5). These data support the hypothesis that oL2_adrenergic receptors within the central noradrenergic system are involved in the initiation of hot flashes and are consistent with the idea that brain norepinephrine is elevated in this process. Animal studies have shown that yohimbine increases norepinephrine release by blocking inhibitory presynaptic oL2-adrenergic receptors (59). These autoreceptors mediate the turnover of norepinephrine through a feedback mechanism, and a reduction in their number or sensitivity would result in increased norepinephrine release (60). This mechanism is consistent with human studies showing that yohimbine elevates and clonidine reduces plasma levels of M H P G (61). Therefore, the yohimbine provocation and clonidine inhibition of hot flashes in symptomatic women may reflect a deficit in inhibitory o~2-adrenergic receptors not seen in asymptomatic women. Additionally, the injection of clonidine into the hypothalamus reduces body temperature and activates heat conservation mechanisms, effects that are blocked by yohimbine (62). Thus, oL2-adrenoceptors in the hypothalamus may be responsible for the events of the hot flash that are characteristic of a heat dissipation response. There is considerable evidence demonstrating that estrogens modulate adrenergic receptors in many tissues (52). It is possible, therefore, that hypothalamic oL2-adrenergic
ROBERT R. FREEDMAN
FIGURE 13.5 The occurrence of a hot flash during body heating was delayed after l~lg/kg clonidine, compared with placebo. No hot flashes occurred in an asymptomaticwoman. (From ref. 58.)
receptors are affected by the estrogen withdrawal associated with the menopause. As noted earlier, a decline in inhibitory presynaptic cx2 receptors would lead to increased central norepinephrine levels, and this is consistent with evidence from animal studies.
V. THERMOREGULATION AND H O T FLASHES Increased thermosensitivity at menopause has been noted in the literature for many years and is reflected in reports of increased hot flash frequency and duration during warm weather (63,64). Peripheral heating has been demonstrated to provoke hot flashes in most of our symptomatic subjects (24), and this has been found by others as well (23). As noted earlier, core body temperature (Tc) in homeotherms is regulated by hypothalamic centers between the thresholds of Tc for sweating and peripheral vasodilation and shivering (Fig. 13.6). According to this mechanism, the heat dissipation responses of hot flashes (sweating, peripheral vasodilation) would be triggered if body temperature were elevated or the sweating threshold lowered. We previously demonstrated that peripheral heating induced hot flashes in symptomatic but not asymptomatic postmenopausal women nor in premenopausal women (24,27). These data suggested that the sweating threshold was reduced in symptomatic postmenopausal women. Considerable research in humans and animals has shown that conditions that alter the sweating threshold tend to alter the shivering threshold in the same direction (51). We therefore tested to see if the Tc shivering threshold was reduced in symptomatic women, similar to their reduction in sweating threshold. We found that the shivering threshold was elevated rather than reduced in symptomatic compared with
193
CHAPTER 13 Menopausal Hot Flashes Small core body temperature (T~) elevations acting within a reduced thermoneutral zone trigger HFs in symptomatic postmenopausal women.
~sYMPT_ _.O_~_T.~ HF) _ , i ~ ' - - " ~ . _ . . . . / Sweating Threshold ~.~ ~, ThermoneutralZone
ASYMFr_O_MA_TIC= Sweating Threshold
....
Shivering Threshold Shivering Threshold
+
I
Sweating Thresho!d. . . . . . . . . . . . . . . . . . Thermoneutral Zone . . . . . . . . . . . .
; ....
T . . .YOH9 . . . . .......... .". BRAIN l MHPG? i
NON
HI: I
CLON
......
NE
t
T
5-HT
t
T
FIGURE 13.6 We have shown that the thermoneutral zone is narrowed in symptomaticwomen. Elevated brain norepinephrine (NE) in animals reduces this zone. Yohimbine (YOH) elevatesbrain norepinephrine and should reduce this zone. Conversely, clonidine (CLON) should widen it.
....
SSRI?
l
Shivering Threshold
asymptomatic women (65). This result implies that the thermoneutral zone is narrowed in postmenopausal women with hot flashes. This hypothesis would explain the ability of small Tc elevations, as we found with the telemetry pill, to trigger the heat loss mechanisms of the hot flash (sweating, cutaneous vasodilation) and would also explain the shivering observed following many of them. We therefore measured the thermoneutral zone in symptomatic and asymptomatic postmenopausal women, hypothesizing a reduction in the former group. We studied 12 symptomatic and 8 asymptomatic postmenopausal women (66). We measured body temperature using a rectal probe, the ingested telemetry pill, and a weighted average of rectal and skin temperatures and determined the sweating and shivering thresholds for each. In a subsequent session, we raised body temperature to the sweating threshold using exercise. The symptomatic women had significantly smaller interthreshold zones than the asymptomatic women on all three measures of body temperature (Table 13.1). Sweat rates were significantly higher in the former group. During exercise, all the
symptomatic and none of the asymptomatic women demonstrated hot flashes. We subsequently studied the effects of clonidine and estrogen on the Tc sweating threshold. In the first study, 12 symptomatic postmenopausal women and 7 asymptomatic women received IV clonidine (2btg/kg) or placebo during separate laboratory sessions in which the Tc sweating threshold was determined. Clonidine significantly raised this threshold in the symptomatic women but lowered it in the asymptomatic women (67). In the second study, 24 symptomatic women were randomly assigned to receive 1 mg/day of 17 ~3-estradiol (E2) orally or a placebo, double-blind, for 90 days. E2 significantly raised the sweating threshold and reduced laboratory-recorded hot flashes, whereas placebo had the opposite effects. Neither group demonstrated changes in plasma M H P G , basal body temperature, or Tc elevations (68). Animal studies have shown that increased brain norepinephrine narrows the width of the interthreshold zone (51). Conversely, clonidine reduces norepinephrine release, raises the sweating threshold and lowers the shivering threshold in
TABLE 13.1 Sweating Thresholds, Shivering Thresholds, and Interthreshold Zones for Rectal Temperature, Telemetry Pill Temperature, and Mean Body Temperature (means + S.E.) Sweating Symptomatic Asymptomatic Pvalue
37.4 + 0.06 37.7 + 0.05 0.001
Symptomatic Asymptomatic P value
37.2 _+ 0.09 37.5 _+ 0.14 0.008
Symptomatic Asymptomatic P value
37.2 +_ 0.07 37.6 _+ 0.04 0.0003
P values for group differences, unpaired T-tests.
Rectal temperature (~ Shivering 37.4 _+ 0.06 37.3 __ 0.16 NS Telemetry pill temperature (~ 37.2 _+ 0.15 37.1 ___0.09 NS Mean body temperature (~ 36.4 _+ 0.06 36.1 + 0.18 0.02
Interthreshold 0.0 + 0.06 0.4 _+ 0.18 0.005 0.0 ___0.11 0.4 _+ 0.18 0.005 0.8 _+ 0.09 1.5 _+ 0.20 0.0006
1
9
4
R
O
B
E
R
human studies (69). Thus, we suggest that elevated brain norepinephrine narrows the thermoregulatory interthreshold zone in symptomatic postmenopausal women (see Fig. 13.6). This zone was so small as to be virtually zero using our methods. We propose that small elevations in core body temperature trigger hot flashes when the sweating threshold is crossed. Core body temperature falls following hot flashes, and patients often report shivering at this time. This likely represents the point where the shivering threshold is crossed, although this has not been directly measured.
VI. CIRCADIAN R H ~ H M S The circadian rhythm of Tc is well known, and similar variations in other thermoregulatory parameters, such as heat conductance and sweating, have also been shown. These patterns suggest that the thermoregulatory effector responses of hot flashes might also demonstrate temporal variations. A previous study showed circadian rhythmicity of self-reported hot flashes in some menopausal women, but no physiologic data were collected (70). We recruited and screened 10 symptomatic and 6 asymptomatic postmenopausal women (21). Each received 24-hour ambulatory monitoring of sternal skin conductance level to detect hot flashes as well as ambient temperature, skin temperature, and Tc. The last measure was recorded using the ingested radiotelemetry pill. Cosinor analysis demonstrated a circadian rhythm ~ < 0.02) of hot flashes with a peak around 1825 hours (Fig. 13.7). This rhythm lagged the circadian rhythm of Tc in symptomatic women by about 3 hours. Tc values of the symptomatic women were lower than those of the asymptomatic women (p < 0.05) from 0000 to 0400 and at 1500 and 2200 hours. The majority of hot
12-
F---i o--o
HF Frequency . . . .
..,-:::"
Core Temperature~Symptomatic
",,,
~
~8
D,
= 6
.~c"" ~':,?.,,....
37.1
Q -.,el Core Temptm~re, A s y m p t o m a U c ~
10
2~ ,,
- 37,2
,.o. u::; .......( ................ : / t
~
.
.-
,.
"-..
.... \""'~""~.
~,
///" :
I-] i'",,~
~,
36.7
i
~.~ 8
,,~\.
~ E
/
\
//
i
i
Z 2
I !
!
!
i
0
2
4
6
i 8
10
I
I
12
14
Hour
i i!
1
!
!
I!
16
18
:i
i , , :i~
I 20
'"
~.2 36.1
22
FIGURE 13.7 Hot flash frequency and core body temperature over 24 hours. Hot flash frequency in 10 symptomatic women, shown as bars. Best-fit cosine curve for hot flash frequency is shown as a dashed line ( . . . . ). A solid line ( O - - O ) with best-fit cosine curve ( ) represents 24-hour core temperature data for 10 symptomatic women. A dotted line (~1.......•) with best-fit cosine curve ( ..... ) represents 24-hour core temperature data in six asymptomatic women. (From ref. 21.)
T
R. FREEDMAN
flashes were preceded by elevations in To, a statistically significant effect (p < 0.05). Hot flashes began at significantly (p < 0.02) higher levels of Tc (36.82 ___0.04~ compared with all nonflash periods (36.70 ___0.005~ These data are consistent with the hypothesis that elevated Tc serves as part of the hot flash triggering mechanism.
VII. H O T FLASHES AND SLEEP Many epidemiologic studies have found increased reports of sleep disturbance during the menopausal transition (71-74). It is generally believed that hot flashes produce arousals and awakenings from sleep, leading to fatigue and, possibly, impaired performance. However, this notion is challenged by two recent laboratory investigations (75,76). In one stu@ symptomatic and asymptomatic postmenopausal women and premenopausal women of similar ages were recorded under controlled laboratory conditions (75). They were screened to eliminate those with any drug use; sleep, physical, or mental disorder; or BMI greater than 30. There were no group differences whatsoever on any sleep stage measure, sleep or fatigue questionnaires, or performance test. When hot flashes occurred (mean - 5.2 ___ 2.9 SD/night), they tended to follow, rather than precede, arousals and awakenings. These data provide no evidence that hot flashes produce sleep disturbance in symptomatic postmenopausal women. These findings are strongly supported by those of a large, recent epidemiologic investigation (76). The Wisconsin Sleep Cohort Study measured sleep quality by complete laboratory polysomnography and by self-reports in a probability sample of 589 pre-, peri-, and postmenopausal women. Sleep quality was not worse in perimenopausal or postmenopausal women nor in symptomatic versus asymptomatic women on any measure. Thus, whereas the majority of epidemiologic studies find increased reports of sleep disturbance during menopause, this is not supported by laboratory investigations. This apparent contradiction may be partially resolved by a recent study conducted in our laboratory. Eighteen symptomatic and 6 asymptomatic postmenopausal women and 12 eumenorrheic women of similar ages were recorded on warm (30~ ambient), neutral (23~ and cold (18~ nights. When data were examined for the entire night, the same findings reported above were obtained: There were no significant differences among the groups and no evidence of hot flash-induced sleep disturbance. However, when data were examined by halves of the night, a different picture emerged. We divided the data because most rapid eye movement (REM) sleep occurs in the second half of the night, and it has been previously reported that thermoregulatory effector responses (e.g., hot flashes) are suppressed during REM. These analyses showed that, during the first half of the night, the women
195
CHAPTER 13 Menopausal Hot Flashes
with hot flashes had significantly more arousals and awakenings than the other two groups and the 18~ ambient temperature significantly reduced the number of hot flashes. These effects did not occur in the second half of the night. In the first half of the night most hot flashes preceded arousals and awakenings. In the second half this pattern was reversed. Because the previous laboratory studies did not analyze data by halves of the night, the discrepancy with the epidemiologic studies may be partially explained. Further research on the role of menopause and hot flashes in sleep disturbance is needed.
VIII. TREATMENT OF H O T FLASHES
A. Hormone Therapy Until recently, hormone therapy was the gold standard of treatment for hot flashes. However, recent studies from the Women's Health Initiative have shown increased risks for breast cancer, coronary heart disease (CHD), thromboembolism, stroke, and dementia for estrogen plus progesterone treatment (77) and an increased risk of stroke with no reduction of C H D risk for estrogen alone (78). In light of the altered risk/benefit ratios for these treatments, they are now being given at lower doses. These issues are discussed elsewhere in this volume.
B. Lifestyle Modifications Because hot flashes are triggered by Tc elevations (66,79) and are more frequent in warm environments (64), procedures to reduce Tc and ambient temperature are recommended. Dressing in layers, drinking cold drinks, and using fans and air conditioning should be attempted. Weight loss may be helpful through reduction of insulation from body fat (7). Smoking cessation may also be beneficial (80-82).
C. Behavioral Treatments Because the thermoneutral zone may be narrowed due to elevated sympathetic activity, relaxation procedures to reduce this activation have been employed. Paced respiration (slow, deep, abdominal breathing) has been shown to reduce hot flash frequency by about 50% from baseline, when implemented to symptom onset. In the first study, women given this procedure plus muscle relaxation exercises showed significant amelioration of objective, laboratory-recorded hot flashes relative to the control procedure (alpha wave electroencephalography [EEG] biofeedback) (83). In the second study, women received paced respiration, muscle re-
laxation, or alpha EEG biofeedback (29). Only the women in the paced respiration group showed significant declines in hot flash frequency, measured by 24-hour ambulatory monitoring of sternal skin conductance. These results were replicated in a third study (30), which did not find declines in measures of sympathetic activation: plasma catecholamines, M H P G , and platelet ci2-receptors. Therefore, the mechanism of action of paced respiration upon hot flashes is not yet known. Two subsequent studies also found significant amelioration of subjective hot flashes using relaxation procedures (84,85).
D. Exercise Physical exercise has also been used as a potential treatment for hot flashes. There have been three randomized clinical trials (RCT) and three other studies. The largest RCT (n = 173) (86) compared a moderate-intensity exercise intervention with a stretching control group over I year. Exercise significantly increased the severity of hot flashes with no change in their occurrence. A Japanese study (87) compared 20 women in a 12-week education and exercise program with 15 no-treatment controls. There were no significant effects on hot flashes. A Swedish study (88) compared 15 women in a three-times-per-week exercise program with 15 women receiving oral estradiol. There was no change in hot flash frequency in the exercise group but a 90% decline in the estradiol group. A large (n = 1323), population-based, retrospective study in Linkoping, Sweden (89) found no significant effect of moderate exercise (1 to 2 hours a week) on hot flash occurrence. A case-control study (n = 171) (90) at a health maintenance organization (HMO) in California also found no effects of exercise on hot flashes. A retrospective, population-based study in Lund, Sweden (n = 6917) (91) found that vigorous exercise (more than 3 hours a week) was associated with significant reductions in hot flash frequency and intensity in a small number of women (4%), but this was confounded by other factors. Taken together, these studies do not demonstrate significant, positive effects of physical exercise on menopausal hot flashes. Our finding that exercise triggers hot flashes in the laboratory may, in part, explain these results (66).
E. Phytoestrogens Isoflavones or phytoestrogens possess estrogenic properties and are found in soy products and red clover. Black cohosh is another plant-derived substance used to treat hot flashes (Remifemin). A recent review of 22 controlled studies (92), 12 on soy and 10 on other botanical compounds, found no consistent improvement of hot flashes relative to placebo.
1
9
6
R
O
B
F. Antidepressants Several recent studies have found efficacy for certain antidepressants in the treatment of hot flashes. Paroxetine (93), a selective serotonin reuptake inhibitor (SSRI), was shown to decrease hot flash composite scores by 62% (12.5 mg/day) and 65% (25.0 mg/day)in 165 women reporting two to three hot flashes a day. The placebo response rate was 37.8%. Fluoxetine is another SSRI used to treat hot flashes. In a study of 81 breast cancer survivors (94), a crossover analysis showed a reduction in hot flash frequency of about 20% over the placebo condition. Venlafaxine, a serotonin/norepinephrine reuptake inhibitor (SNRI), has also shown efficacy in treating hot flashes. In a study of 229 women (95), venlafaxine reduced hot flash scores by 60% from baseline at 75 and 150 mg/day and 37% at 37.5 mg/day compared with 27% for placebo. Side effects of these antidepressants include nausea, dry mouth, somnolence, decreased appetite, and insomnia. As noted earlier, clonidine ameliorates hot flashes by raising the Tc sweating threshold. Two small placebocontrolled studies found that oral clonidine reduced hot flash frequency by 46% (66) and transdermal clonidine reduced it by 80% (96). Two larger studies of breast cancer survivors on tamoxifen showed smaller, but significant reductions in hot flash frequency for oral (97) and transdermal clonidine (98) compared with placebo. Side effects of clonidine include hypotension, dry mouth, and sedation.
E
R
T
R. FREEDMAN
ingested telemetry pill. We then found that the thermoneutral zone, within which sweating, peripheral vasodilation, and shivering do not occur, is virtually nonexistent in symptomatic women but normal (about 0.4~ in asymptomatic women. Thus, we believe that small temperature elevations preceding hot flashes acting within a reduced thermoneutral zone constitute the triggering mechanism. We also demonstrated that central sympathetic activation is elevated in symptomatic women, which, in animal studies, reduces the thermoneutral zone. Clonidine reduces central sympathetic activation, widens the thermoneutral zone, and ameliorates hot flashes. Estrogen virtually eliminates hot flashes and widens the thermoneutral zone, but the pathway through which this occurs is not known. Behavioral relaxation procedures reduce hot flash frequency to the same extent as clonidine (about 50%), but their mechanism of action is also not understood. The role of hot flashes and menopause in sleep disturbance is currently unresolved.
References 1. Neugarten BL, Kraines RJ. "Menopausal symptoms" in women of various ages. PsychosomMed 1965;27:266-273. 2. Feldman BM, Voda A, Gronseth E. The prevalence of hot flash and associated variables among perimenopausal women. Res Nurs Hlth 1985;8:261-268. 3. Hagstad A, Janson PO. The epidemiologyof climacteric symptoms. dcta Obstet Gynecol&and Supp11986;134:59-65.
G. Gabapentin Finally, gabapentin is an anticonvulsant of unknown mechanism, which was fortuitously found to ameliorate hot flashes in some patients. A controlled study of 59 women (99) found a reduction of hot flash frequency of 45% versus 29% for placebo. Side effects of this compound include dizziness and peripheral edema.
IX. SUMMARY Hot flashes are the most common symptom associated with menopause, although prevalence estimates are lower in some rural and non-Western areas. The symptoms are characteristic of a heat-dissipation response and consist of sweating on the face, neck, and chest, as well as peripheral vasodilation. Although hot flashes clearly accompany the estrogen withdrawal at menopause, estrogen alone is not responsible because levels do not differ between symptomatic and asymptomatic women. Until recently it was thought that hot flashes were triggered by a sudden, downward resetting of the hypothalamic set point, because there was no evidence of increased core body temperature. However, we recently obtained such evidence, using a rapidly responding
4. Gutherie JR, Dennerstein L, Hopper JL, Burger HG. Hot flushes, menstrual status, and hormone levels in a population-based sample of midlife women. Obstet Gyneco11996;88:437-442. 5. KronenbergE Hot flashes:epidemiologyand physiology.~Inn NY/Icad Sci 1990;592:52-86. 6. Chakravarti S, Collins WP, Newton JR, Oram DH, Studd JWW. Endocrine changes and symptomatologyafter oophorectomyin premenopausal women. BrJ Obstet Gyneco11977;84:769-775. 7. Freedman RR. Hot flash trends and mechanisms. Menopause 2002;9: 151-152. 8. Van Keep PA, HumphreyM. Psycho-socialaspects of the climacteric. In: Van Keep PA, Greenblatt RB, A1-Beaux-FernetM, eds. Consensus on menopause research. Lancaster, England: MTP Press, 1976;5-8. 9. Flint M, Samil RS. Cultural and subcultural meanings of the menopause. /Inn N Y/Icad &i 1990;592:134-148. 10. Tang G. Menopause: the situation in Hong Kong Chinese women. In: Berg G, Hammar M, eds. The modern management of the menopause, ed 8. New York: Parthenon Press, 1993;47-55. 11. Beyene Y. Cultural significance and physiological manifestations of menopause a biocultural analysis. Cult Med Psychiatry 1986;10:47-71. 12. Murkies AL, Wilcox G, Davis SR. Phytoestrogens.J Clin Endocrinol Metab 1998;83:297-303. 13. Molnar GW. Body temperature during menopausalhot flashes.Jdppl Physiol Resp Environ Ex Physio11975;38:499-503.
14. KronenbergF, Cote LJ, Linkie DM, Dyrenfurth I, DowneyJA. Menopausal hot flashes: thermoregulatory, cardiovascular, and circulating catecholamine and LH changes.Maturitas 1984;6:31-43. 15. Tataryn IV, Lomax P, BajorekJG, et al. Postmenopausalhot flushes: a disorder of thermoregulation.Maturitas 1980;2:101-107. 16. Freedman RR. Biochemical, metabolic, and vascular mechanisms in menopausal hot flushes. Fertil Steri11998;70:1-6.
CHAPTER 13 Menopausal Hot Flashes 17. Ginsburg J, Swinhoe J, O'Reilly B. Cardiovascular responses during the menopausal hot flush. BrJ Obstet Gynaeco11981;88:925-930. 18. Sturdee DW, Reece BL. Thermography of menopausal hot flushes. Maturitas 1978;201-205. 19. Molnar GW, Read RC. Studies during open heart surgery on the special characteristics of rectal temperature. J Appl Physiol 1974;36: 333-336. 20. Shiraki K, Nobuhide K, Sagawa S. Esophageal and tympanic temperature responses to core blood temperature changes during hyperthermia. J Appl Physiol 1986;61:98-102. 21. Freedman RR, Norton D, Woodward S, Cornelissen G. Core body temperature and circadian rhythm of hot flashes in menopausal women. J Clin Endocrinol Metab 1995;80:2354-2358. 22. Freedman RR, Woodward S. Core body temperature during menopausal hot flushes. Menopause 1996;65:1141-1144. 23. Sturdee DW, Wilson KA, Pipili E, Crocker D. Physiological aspects of menopausal hot flush. Br MedJ 1978;2:79-80. 24. Freedman RR. Laboratory and ambulatory monitoring of menopausal hot flashes. Psychophysiology 1989;26:573-579. 25. Meldrum DR, Shamonki IM, Frumar AM, et al. Elevations in skin temperature of the finger as an objective index of postmenopausal hot flashes: Standardization of the technique.AmJ Obstet Gyneco11979;135: 713-717. 26. Tataryn IV, Lomax P, Meldrum DR, et al. Objective techniques for the assessment of postmenopausal hot flashes. Obstet Gyneco11981;57: 340-344. 27. Freedman RR, Woodward S, Norton D. Laboratory and ambulatory monitoring of menopausal hot flushes: Comparison of symptomatic and asymptomatic women. J Psychophysio11992;6:162-166. 28. de Bakker IPM, Everaerd W. Measurement of menopausal hot flushes: validation and cross-validation. Maturitas 1996;25:87-98. 29. Freedman RR, Woodward S. Behavioral treatment of menopausal hot flushes: evaluation by ambulatory monitoring. Am J Obstet Gynecol 1992;167:436-439. 30. Freedman RR, Woodward S, Brown B, Javaid JI, Pandey GN. Biochemical and thermoregulatory effects of behavioral treatment for menopausal hot flashes. Menopause 1995;2:211-218. 31. Carpenter JS, Andrykowski MA, Freedman RR, Munn R. Feasibility and psychometrics of an ambulatory hot flash monitoring device. Menopause 1999;6:209-215. 32. Askel S, Schomberg DW, Tyrey L, Hammond CB. Vasomotor symptoms, serum estrogens, and gonadotropin levels in surgical menopause. A m J Obstet Gyneco11976;165-169. 33. Stone SC, Mickal A, Rye F, Rye PH. Postmenopausal symptomatology, maturation index, and plasma estrogen levels. Obstet Gynecol 1975;45: 625-627. 34. Hutton JD, Jacobs HS, Murray MAF, James VHT. Relation between plasma esterone and estradiol and climacteric symptoms. Lancet 1978;1:671-681. 35. Schindler AE, Muller D, Keller E, Goser R, Runkel E Studies with clonidine (Dixarit) in menopausal women. Arch Gynecol 1979;227: 341-347. 36. Johnson, S. Menopause and hormone replacement therapy. Med Clin North Am 1998;82:297-320. 37. Kenemans P, Barentsen R, Van de Weijer E Practical HRT. Medical Forum International, 2nd ed., 1996, 11-18. 38. Campbell S. Intensive steroid and protein hormonal profiles on postmenopausal women experiencing hot flashes and a group of controls. In: Campbell S, ed. Management of the menopause and post-menopausal years. London: MTP Press, 1976. 39. Casper RF, Yen SSC, Wilkes MM. Menopausal flushes: a neuroendocrine link with pulsatile luteinizing hormone secretion. Science 1979;205: 823-825. 40. Tataryn IV, Meldrum DR, Lu KH, Frumar AM, Judd HL. LH, FSH, and skin temperature during menopausal hot flush. J Clin Endocrinol Metab 1979;49:152-154.
197 41. Gambone J, Meldrum DR, Laufer L, et al. Further delineation ofhypothalamic dysfunction responsible for menopausal hot flashes. J Clin Endocrinol Metab 1984;59:1092-1102. 42. Mulley G, Mitchell RA, Tattersall RB. Hot flushes after hypophysectomy. Br MedJ 1977;2:1062. 43. Meldrum DR, Erlik Y, Lu JKH, Judd HL. Objectively recorded hot flushes in patients with pituitary insufficiency.J Clin Endocrinol Metab 1981;52:684-687. 44. Casper RF, Yen SSC. Menopausal flushes: effect of pituitary gonadotropin desensitization by a potent luteinizing hormone releasing factor agonist. J Clin Endocrinol Metab 1981;53:1056-1058. 45. DeFazio J, Meldrum DR, Laufer L, et al. Induction of hot flashes in premenopausal women treated with a long-acting GnRH agonist. J Clin Endocrinol Metab 1983;56:445-448. 46. Leslie RDG, Pyke DA, Stubbs WA. Sensitivity to enkephalin as a cause of non-insulin dependent diabetes. Lancet 1979;1:341-343. 47. Lightman SL, Jacobs HS, Maguire AK, McGarrick G, Jeffcoate SL. Climacteric flushing: clinical and endocrine response to infusion of naloxone. BrJ Obstet Gynaeco11981;88:919-924. 48. DeFazio J, Verheugen C, Chetkowski R, et al. The effects of naloxone on hot flashes and gonadotropin secretion in postmenopausal women. J Clin EndocrinolMetab 1984;58:578-581. 49. Tepper R, Neri A, Kaufman H, Schoenfield A, Ovadia J. Menopausal hot flushes and plasma g-endorphins. Obstet Gyneco11987;70:150--152. 50. Genazzani AR, Petraglia F, Facchinetti F, et al. Increase of proopiomelanocortin-related peptides during subjective menopausal flushes. Am J Obstet Gyneco11984;149:775-779. 51. Brfick K, Zeisberger E. Adaptive changes in thermoregulation and their neuropharmacological basis. In: Sch6nbaum E, Lomax P, eds. Thermoregulation:physiology and biochemistry. New York: Pergamon Press, 1990; 255-307. 52. Insel PA, Motulskey HJ. Physiologic and pharmacologic regulation of adrenergic receptors. In: Insel PA, ed. Adrenergic receptors in man. New York: Marcel Dekker, 1987;201-236. 53. Lambert GW, Kaye DM, Vaz M, et al. Regional origins of 3-methoxy-4-hydroxyphenylglycol in plasma: effects of chronic sympathetic nervous activation and devervation, and acute reflex sympathetic stimulation. Jgluto Nerv Sys 1995;55:169-178. 54. Kopin IJ, Blombery P, Ebert MH, et al. Disposition and metabolism of MHPG-CD3 in humans: plasma MHPG as the principal pathway of norepinephrine metabolism and as an important determinant of CSF levels of MHPG. In: Usdin E, ed. Frontiers in biochemicalandpharmacological research in depression. New York: Raven Press, 1984;57-68. 55. Clayden JR, Bell JW, Pollard E Menopausal flushing: double blind trial of a non-hormonal medication. Br MedJ 1974;1:409-412. 56. Laufer LR, Erlik Y, Meldrum DR, Judd HL. Effect of clonidine on hot flushes in postmenopausal women. Obstet Gynecol 1982;60: 583-589. 57. Schmitt H. The pharmacology of clonidine and related products. In: Gross F, ed. Handbook ofexperimentalpharmacology, vo139: Antihypertensive agents. New York: Springer-Verlag, 1977;299-396. 58. Freedman RR, Woodward S, Sabharwal SC. e~2-Adrenergic mechanism in menopausal hot flushes. Obstet Gyneco11990;76:573-578. 59. Golberg M, Robertson D. Yohimbine: A pharmacological probe for study of the e~2-adrenoceptor.Pharmacol Rev 1983;35:143-180. 60. Starke K, Gothert M, Kilbringer H. Modulation of neurotransmitter release by presynaptic autoreceptors. Physiol Rev 1989;69:864-989. 61. Chamey DS, Heninger GR, Sternberg DE. Assessment of c~2-adrenergic autoreceptor function in humans: Effects of oral yohimbine. L ~ Sci 1982; 30:2033-2041. 62. Zacny E. The role of ot2-adrenoceptors in the hypothermic effect of clonidine in the rat.J Pharm Pharmaco11982;34:455-456. 63. Molnar GW. Menopausal hot flashes: their cycles and relation to air temperature. Obstet Gyneco11981;57(suppl):52-55. 64. Kronenberg F, Barnard RM. Modulation of menopausal hot flashes by ambient temperature. J Therm Bio11992;17:43-49.
1
9
8
R
O
B
65. Freedman RR, Woodward S. Altered shivering threshold in postmenopausal women with hot flashes. Menopause 1995;2:163-168. 66. Freedman RR, Krell W. Reduced thermoregulatory null zone in postmenopausal women with hot flashes. Am J Obstet Gynecol 1999;181: 66-70. 67. Freedman RR, Dinsay, R. Clonidine raises the sweating threshold in symptomatic but not in asymptomatic postmenopausal women. Fertil Steri12000; 74:20-23. 68. Freedman RR, Blacker CM. Estrogen raises the sweating threshold in postmenopausal women with hot flashes. Fertil Steri12002;77:487-490. 69. Delaunay L, Bonnet F, Liu N, et al. Clonidine comparably decreases the thermoregulatory thresholds for vasoconstriction and shivering in humans. Anesthesiology 1993;79:470-474. 70. Albright DL, Voda AM, Smolensky MH, Hsi B, Decker M. Circadian rhythms in hot flashes in natural and surgically induced menopause. ChronobiolInternat 1989;6:279-284. 71. Baker A, Simpson S, Dawson D. Sleep disruption and mood changes associated with menopause.J Psychosom Res 1997;43:359-369. 72. Kuh DL, Wadsworth M, Hardy R. Women's health in midlife: the influence of the menopause, social factors and health in earlier life. BrJ Obstet Gynaeco11997;104:923-933. 73. Owen JF, Matthews KA. Sleep disturbance in healthy middle-aged women. Maturitas 1998;30:41-50. 74. Kravitz HM, Ganz PA, Bromberger J, et al. Sleep difficulty in women in midlife: a community survey of sleep and the menopause transition. Menopause 2003;10:19-28. 75. Freedman RR, Roehrs TA. Lack of sleep disturbance from menopausal hot flashes. Fertil Steri12004;82:138-144. 76. Young T, Rabago D, Zgierska A, Austin D, Finn L. Objective and subjective sleep quality in premenopausal, perimenopausal, and postmenopausal women in the Wisconsin cohort study. Sleep 2003;26: 667-672. 77. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestins in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. J_//MA 2002;288:321-333. 78. The Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. J_dMA 2004;291:1701-1712. 79. Freedman RR. Core body temperature variation in symptomatic and asymptomatic postmenopausal women: brief report. Menopause 2002;3: 399-401. 80. Gold EB, Sternfield B, KelseyJL, et al. Relation of demographic and lifestyle factors to symptoms in a multi-racial/ethnic population of women 40-55 years of age. Am J Epidemio12000;152:463-473. 81. Whiteman MK, Staropoli CA, Lengenberg PW, et al. Smoking, body mass, and hot flashes in midlife women. Obstet Gynecol 2003;101: 264-272. 82. Jessen AB, Toubro S, Astrup A. Effect of chewing gum containing nicotine and caffeine on energy expenditure and substrate utilization in men. Am J Clin Nutr 2002;77:1442-1447. 83. Germaine LM, Freedman RR. Behavioral treatment of menopausal hot flashes: evaluation by objective methods. J Consult Clin Psychol 1984;52:1072-1079.
E
R
T
R. FREEDMAN
84. Irvin JH, Domar AD, Clark C, Zuttermeister PC, Friedman R. The effects of relaxation response training on menopausal symptoms. J Psychosom Obstet Gyneco11996;17:202-207. 85. Wijima K, Melin A, Nedstrand E, Hammar M. Treatment of menopausal symptoms with applied relaxation: a pilot study.J Behavior Tber Exper Psychiatr 1997;28:251-261. 86. Aiello EJ, Yasui Y, Tworoger SS, et al. Effects of a year-long, moderateintensity exercise intervention on the occurrence and severity of menopause symptoms in postmenopausal women. Menopause 2004;11: 382-388, 2004. 87. Ueda M. A 12-week structured education and exercise program improved climacteric symptoms in middle-aged women. J Physiol AnthropolAppl Human Sci 2004;23:143-148. 88. Lindh-Astrand L, Nestrand E, Wyon Y, Hammar M. Vasomotor symptoms and quality of life in previously sedentary postmenopausal women randomized to physical activity or estrogen therapy. Maturitas 2004;48:97-105. 89. Ivarsson T, Spetz A-C, Hammar M. Physical exercise and vasomotor symptoms in postmenopausal women. Maturitas 1998;29:139-146. 90. Sternfield B, Q.uesenberry CP, Husson G. Habitual physical activity and menopausal symptoms: a case-control study. J Womens Health 1999;8:115-123. 91. Li C, Samsioe G, Borgfeldt C, et al. Menopause-related symptoms: what are the background factors? A prospective population-based cohort study of Swedish women (The Women's Health in Lund Area Study). Am J Obstet Gyneco12003;189:1646-1653. 92. Kronenberg F, Fugh-Berman A. Complementary and alternative medicine for menopausal symptoms: a review of randomized, controlled trials. Ann Intern Med 2002;137:805-813. 93. Stearns V, Beebe KL, Iyengar M, Dube E. Paroxetine controlled release in the treatment of menopausal hot flashes: a randomized controlled trial. JAMff 2003;289:2827-2834. 94. Loprinzi CL, Sloan JA, Perez EA, et al. Phase III evaluation of fluoxetine for treatment of hot flashes.J Clin Onco12002;20:1578-1583. 95. Loprinzi CL, Kugler JW, Sloan JA, et al. Venlafaxine in management of hot flashes in survivors of breast cancer: a randomized controlled trial. Lancet 2000;356:2059-2063. 96. Nagamani M, Kelver ME, Smith ER. Treatment of menopausal hot flashes with transdermal administration of clonidine. Am J Obstet Gyneco11987;156:561-565. 97. Pandya KJ, Raubertas RF, Flynn PJ, et al. Oral clonidine in postmenopausal patients with breast cancer experiencing tamoxifen induced hot flashes: a University of Rochester Cancer Center Community Clinical Oncology Program study. Ann Intern Med 2000;132: 788-793. 98. Goldberg RM, Loprinzi CL, O'Fallon JR, et al. Transdermal clonidine for ameliorating tamoxifen-induced hot flashes. J Clin Onco11994;12: 155-158. 99. Guttuso T Jr, Kurlan R, McDermott MP, Keiburtz K. Gabapentin's effects on hot flashes in postmenopausal women: a randomized controlled trial. Obstet Gyneco12003;101:337-345.
~HAPTER 1 ~
Clinical Effects of Sex Steroids on the Brain IVALDO DA SILVA Gynecology Department, Federal University of $5o Paulo, Brazil 04038-031 FREDERICK NAFTOLIN
Department of Obstetrics and Gynecology, NewYork University School of Medicine, New York, NY 10016
early symptoms of menopause are hot flushes (vasomotor episodes [VMEs]), mood/cognition changes, and sleep disorders. Although the occurrence of hot flushes is usually transient (3 to 5 years), they are important in two ways: First, hot flushes are often intense and debilitating (2). Second, they indicate the presence of uncompensated, clinically relevant estrogen deficiency that may herald the development of other complications of estrogen deficiency, such as vascular disease, bone loss, and metabolic syndromes, as noted elsewhere in this book. The central and peripheral nervous systems, which are estrogen target tissues, undergo anatomic and biochemical remodeling throughout life. These changes may be subtle; but the brain's responses to sex hormones ~ estrogen, progesterone, and androgen ~ have important roles in the modulation of brain function (3). These hormones affect neurons, glia, and microglia (i.e., brain macrophages) in many areas in the brain, not just in the portions of the hypothalamus and preoptic area involved solely with autonomic function. The decline of sex steroids, particularly estrogen, during the postreproductive decades is accompanied by changes in eating, metabolism, and sleep; behavior; mood; sexuality; locomotor activity; immune response; memory; and cognitive function (4). The list continues to grow and now includes ischemic vascular and dystrophic brain problems (5).
I. CLINICAL EFFECTS OF SEX STEROIDS O N T H E BRAIN Over the past century, better medical, economic, and sociocultural conditions in our society have doubled life expectancy for women. Symptomatology, mental health, and degenerative brain diseases, especially stroke and dementia due to vascular conditions or Alzheimer's disease, are part of aging and of menopause. They have become increasingly important with the rise in the aging population. This importance will only grow during the coming years, so that a better understanding and proactive stance become more important in the 21st century. Menopause is an important period of transition to dependence on locally formed estrogen. During the reproductive period, the chief source of estrogen is the ovary. Although extra gonadal estrogen formation furnishes estrogen in most tissues, including the brain and blood vessels (1), this is a relatively small source of estradiol in the woman's economy. At menopause, the loss of ovarian estrogen and continued secretion of androgens by the ovary and adrenal gland make extra gonadal estrogen the chief source of estrogen. This is tissue dependent and insufficient to block the appearance of menopausal symptoms in the great majority of women. The most common TREATMENT OF THE POSTMENOPAUSAL WOMAN
199
Copyright 9 2007by Elsevier,Inc. All rightsof reproductionin any formreserved.
200
DA SILVAAND NAFTOLIN
Although these brain changes originate in the central nervous system (CNS), there are reports that hormonal changes also influence peripheral nervous system (PNS) functions such as sensory function, fine-touch perception, two-point discrimination, hearing, smell, and vision (6). These are less well documented and do not exclude primary effects of sex steroids on the tissues and organs in which the PNS is embedded and from which it derives its blood supply. For completeness, we have listed major groups of brain functions that may deteriorate during the menopause (Table 14.1). This table gives a powerful demonstration of the brain's integrative function and estrogen's role. Because of space limitations, the remainder of this chapter concentrates on brain functions in which sex steroids are best known to play a role and that may be preserved by hormone treatment. An overview of the brain, its functional construction, and the mechanism of sex steroid actions on the brain precedes these considerations.
FIGURE 14.1 Brain development along the neural tube results in functional and anatomic mini-organs linked by tracts. The adult brain is the result of folding the neural tube and its derivatives to fit the cranium.
II. THE BRAIN A N D SEX STEROIDS A. Anatomic-Functional Correlations The brain is a linear ensemble of structures that develop along the neural tube and that are roughly proportional in size to the functional requirements of individual species (Fig. 14.1). Humans have a large prefrontal cortical area, which serves to serve cognition, memory, and mood. This area and the central, sensory cortical areas are linked with other regions through axonal connections running along the area beneath the original neural tube (i.e., adult ventricular system). These axons form "tracts" that connect distant areas of the brain, allowing signals to be processed between brain areas and eventually stored, used, or discarded (Fig. 14.2). The hippocampus, which primarily processes incoming
TABLE 14.1
Brain Functions Affected During Menopause
Autonomic Gonadotrophins Sleep Vasomotor episodes Libido Mood Metabolic regulation Cognition Sensory perception Memory Voluntary motor function Immunologic function Sexually dimorphic function and dysfunction (presumed to be sex steroid related)
information from the environment, body sensors, and other brain areas and stores short-term memory, is located in, and connected to, the temporal cortex. There are also major reciprocal connections with the hypothalamus and cortical areas in which long-term memory is stored and other cognitive functions occur. Memory and cognitive centers along the visual pathway contribute to the final inflow path to the hippocampus. The interaction of all these structures results in optimal function of the brain. Evidence for this arrangement shows that age-related dystrophy affecting the hippocampal neurons first results in a deficit of short-term memory and then is generalized to most brain functions (7). Because the brain is an active metabolic tissue, it requires a massive blood flow, which is also hormone-sensitive (8). Although most attention has focused on its neurons, the brain is mainly composed of glial cells, particularly astroglia. Therefore, reported discordance in size and in areas of the brain most likely reflects glial differences; neurons are interspersed in the mass of glial cells, usually clustering into groups called nuclei. However, closeness is not critical, as the neurons interact through arborization of their axonal (afferent) processes, which connect to the dendritic (efferent) tree of target neurons via cell specializations known as synapses. The information then passes to the neuronal cell body, or soma. The cellular actions are similar to the general metabolism that goes on in nonneural cells; the main difference is in the arborization of communicating axons and dendrites that allows both local and long distance communication. The astroglia, like the neurons, are sex steroid sensitive. They can form metabolites through the steroid degradation process. Generally, it is accepted that androgens are aromatized by neurons (9) and ring-A reduced by astroglia (10).
CHAPTER 14 Clinical Effects of Sex Steroids on the Brain
201
FIGURE 14.2 Each region of the brain has an important role in specific brain functions. Optimal brain activity is maintained by means of the integration of different areas by neural tracts. ARC, arcuate nucleus; POA, preoptic area; SO, supraoptic nucleus; PVN, paraventricular nucleus; VMN, ventromedial nucleus.
Evidence has been presented indicating that astroglia can be activated to aromatize androgens (11). The astroglia are becoming better understood as a second communication system in the brain, but the regulation of this calciumdriven communication is not well understood, nor are the roles of aging and steroids resolved (12). Neuronal processes connect through myriad synapses, cellcell interfaces that perform the specialized function of furnishing neurotransmitters to the synaptic cleft and passing them back and forth (uptake and reuptake). Neurotransmitter expression is regulated by many substances, including sex steroids (13). The formation and maintenance of synapses is estrogen regulated (14), especially in the areas where there are estrogen receptors (ERs). Sex steroids also regulate the number and function of the neurotransmitter receptors that translate the messages carried into the synapses by the neurotransmitters. As mentioned earlier, with aging, there is a shift in the regulatory balance between circulating steroids and locally formed steroids. During menopause, the follicular estrogen decreases, leaving a greater burden of steroid supply to peripheral conversion by brain and other issues. Generally, estradiol is synaptogenic and promotes communication between neurons. This is best recognized in animal studies that have shown that estradiol induces the growth of specialized outpouchings called spines along the afferent dendrites of neurons. This has been especially studied in the information processing part of the brain, the bilateral hippocampuses (15). The addition/lengthening of spines on the dendrites furnishes "space" for the axons to place more boutons and form information-carrying synapses. Thus, the effect of estrogen is largely to bring more information into the brain and speed processing, especially in the hippocampus. This
may be the mechanism for estrogen's effect on memory (16-19). The glial cells have a major role in all these activities. The astroglia are the embedment of the neurons. Synapses must penetrate sheets ofglial processes to make their connections. The astroglia respond to neurotransmitters and other products in the area of the neuronal cell bodies and along the neurites (neural processes) and synapses (10). The glia buffer the leakage of neural products (e.g., neurotransmitters, cytokines, free radicals). In this way, the glia cells form a protective or reparative barrier between neurons. We and others have shown that the astroglia are extremely physically active, slinging out processes and shuttering the space vacated by changing synapses (20). This is regulated by estradiol (14). A specialized form of glia, the oligodendroglia, is present in CNS and PNS. These glial cells wrap axons with myelin, ensuring rapid, efficient neurotransmission. In animals, the oligodendroglia have also been shown to metabolize cholesterol and other B-ring unsaturated steroids, such as pregnenolone and cognate progestins (21). The major metabolic products are allopregnanolone and alloprogesterone, which are presently being studied as sedative, anxiolytic neuroactive compounds (21,22). The role of these compounds in menopausal mood changes is only now being explored (23). Another type of glia, the microglia, constitutes the brain's macrophages. The microglia express ER(x, make estrogen, and produce cytoldnes, growth factors (molecules that regulate cellular responses to insult), and immune-checkpoint proteins in response to estrogen. We have proposed that in this way the microglia form the (estrogen-sensitive) immunologic brain barrier to maintain homeostasis and avoid inflammation that could damage nearby neurons and other cells. Maintaining homeostasis could avoid brain dystrophy
202
DA SILVAAND NAFTOLIN
(10,24). We also proposed that this "immune brain barrier" (IBB) (12) helps to maintain the brain as an immunologically privileged organ. The IBB is mediated by the expression of the activational response to antigens (25) and the production of the immune-checkpoint proteins, Fas ligand (FasL), and CD40 in microglia, astrocytes, and neurons (24). As activated T cells enter the brain, they may contact processes of the glia limitans or perivascular microglia (26), thereby receiving an estrogen-regulated Fas/FasL-mediated death signal (25,27,28). These actions of estrogen that limit the inflammatory response in the brain may keep harmful inflammation in check during the postreproductive period, but this would require the presence of estrogen or androgenic precursors. The outcome of repeated inflammation is dystrophy, so there are implications for the advent and course of dementia in aging. Finally, there are direct interactions with the proteins found in Alzheimer's dementia (AD). Estrogen has been associated with regulation of the presence of insoluble amyloid [3 in the brain. This is the inflammatory form of a commonly circulating protein (29). In addition, estradiol has been shown to inhibit the hyperphosphorylation of tau, a microtubule associated protein (MAP) that is bound up in the tangles that escape degenerated cells in AD (30).
B. Estrogen and General Brain Function Estrogen appears to affect all brain cells via direct cellular effects and indirect effects (estrogen-sensitive cells regulating connected non-ER-bearing cells). Estrogen affects neurons and glia and also regulates the brain's blood vessels. Estrogen has been shown to influence most brain functions by regulating biochemical and anatomic conditions and modulating the uptake and turnover of neurotransmitters, neuronal enzyme activity, and the expression of steroid receptors on the brain. Most estrogen actions on brain cells are ER mediated.
C. Estrogen, Estrogen Receptors, and the Brain Because Doisy and Butenandt identified and determined the formula for estrogen (31), the term estrogen has many times been redefined because of the discovery of many classes of compounds that are estrogenic and also for the broadening description of specific biochemical actions of estrogenic compounds. After the development of the primary steroidal estrogens (e.g., estradiol, estrone, estriol), the first nonsteroidal, synthetic estrogen, diethylstilbestrol (DES), was produced. Description of the estrogenic effects of plant estrogen (i.e., phytoestrogen) soon followed, and a large group of nonsteroidal compounds were also found to have estrogenic actions. Shortly thereafter,
TABLE 14.2 Clinical Effects of Estradiol (E2), Raloxifene (RLX), Tamoxifen (TMX), and Genistein (GEN) E2 Vasomotor events
Sexuality Cognitive
effects SSRI synergy Brain fMRI
TMX
GEN
t -~?
t-~?
t
lr- t
Gonadotrophins Sleep regulation
RLX
1[~
t- t t- t t- t t- t
t- t t- t?
It-,?
t , strong evidence; ) , weak evidence.
primary estrogens and DES became clinically available. The next step was the commercial development: steroidal estrogens, nonsteroidal estrogens, estrogen agonists or antagonists, and antiestrogens were drawn from the previously described compounds or their congeners. Researchers became interested in studying differences in mechanism of action of these classes of compounds. The development of compounds with new agonist or antagonist properties was accompanied by descriptive terms that appear to add little to the fundamental understanding of estrogen action: phytoestrogen, xenoestrogens, estrogen-like endocrine disrupters, and selective estrogen receptor modulators (SERMs) (32). A composite of the known clinical effects of the prototypical SERMs in clinical practice follows. Note the large gaps in knowledge (Table 14.2). With increased knowledge of estrogen actions on the brain, quantitative and qualitative inconsistencies were apparent in clinical situations. This may be resolved by the discovery of a second, specific ER, ER-[3, (33,34), which has a regional distribution and specificity of action in the brain that differs from ER-oL, although ER-ci and ER-13 may be found together in specific brain areas (Fig. 14.3). Estrogen regulates the neural activity through genomic effects that are regulated by ERs. DNA binding regulates transcription of RNA for new protein synthesis and expression. Because the ligand ERs dimerize before DNA binding, and there are two individual ERs (ER-ci and ER-[3), the formation of homodimers or heterodimers is possible (35). The resulting transcripts can be agonistic or antagonistic to transcription depending on the ligands (36), the receptor type (37), co-transcription factors, and possible effects of homodimer or heterodimer formation (Fig. 14.4) (37). Although receptors mediate the bulk of steroid actions in the brain, some actions appear to not require receptors. These have been shown experimentally to include changes in cell membrane channel permeability (38). Other, rapid actions of sex steroids may be cause by direct effects through several possibilities, such as early-intermediate gene activation (e.g., cFOS) or neurotrophin-driven actions (38).
CHAPTER 14 Clinical Effects of Sex Steroids on the Brain
203
FIGURE 14.3 Distributionof estrogen receptors ER-cxand ER-f3 mRNA in the rat brain.
Studies mapping estrogen and progestin receptors (PRs) in the brain have shown that ERs and PRs are co-localized in many areas, including the hypothalamus, hippocampus, amygdala, and limbic forebrain system (39,40). Studies showing that ER is present in areas previously thought to be devoid of ER, such as the cortex, are especially promising for explaining effects of estrogen (41,42). In fact, the distribution of steroid receptors is regionalized in a manner similar to the regionalization of function in the brain (see Figs. 14.2 and 14.3). For example, ER levels are high in the hypothalamus, where estrogen-dependent actions regulate neuroendocrine functions such as gonadotropin-releasing hormone (GnRH) control, sexual behavior, feeding behavior, and vasomotor stability. Because of diverse axon pathways and the synapses that these axons form with distant neurons, effects of small numbers of neurons often have disproportionate effects on brain function. For example, a small number of estrogen-sensitive acetylcholine neurons send axons throughout the brain, allowing indirect effects of estrogen on distant neurons. Because of the possibility of connections between ER-positive and ER-negative neurons, very few of the brain's neural networks could be insensitive to estrogen. This is shown in a diagrammatic manner in Fig. 14.4 (43). It also follows that cell death or dystrophy among relatively distant neurons can have widely felt consequences.
D. Sex Steroids and Brain Phenotype The presence of morphologic or functional sexual dimorphism in brain areas could give clues to estrogen action on the brain. In animals, hormonal effects during early prenatal and postnatal development induce sexual differentiation of many organ systems, including the brain (44). These sex differences carry over into adulthood, when estrogen affects
FIGURE 14.4 Possibilities for dimerization ofliganded receptors as they regulate DNA transcription.
females and males differently. In addition to classic signs of menopause in women (e.g., hot flushes), the incidence of depression and Alzheimer's dementia are higher among women (45-47), supporting the hypothesis that gender or hormonal balance is involved. In monkey models, neurite formation and synaptogenesis have been proven to be targets of estrogen in developing and adult subjects (48,49).
1. BIOCHEMICAL EFFECTS OF ESTROGEN IN THE BRAIN
Neurons are responsive to estrogen. Regulation of fiber growth and branching; synaptic plasticity; dendritic spines; synaptogenesis; glial cell function; blood flow; regulation of neurotransmitters/neuropeptides, neurotransmitter receptors, and neurotrophic factors; and clearance of proteins (e.g., [3-amyloid) have all been traced to estrogen. The enzyme aromatase (i.e., estrogen synthetase) is present in many brain areas (49). During the normal reproductive cycle, physiologically important levels of estrogen enter the circulation and reach the brain, thereby being the important regulator as
204
DA SILVAAND NAFTOLIN
opposed to locally formed estrogen. In the postreproductive years, the circulating estrogen falls and androgens are relatively maintained. This emphasizes local estrogen formation in specific brain areas as having functional importance.
III. HORMONES AND BRAIN FUNCTIONS OF CLINICAL IMPORTANCE TO THE MENOPAUSE A. V a s o m o t o r E p i s o d e s a n d O t h e r Brain D y s f u n c t i o n at the Start o f the C l i m a c t e r i c Vasomotor episodes (VMEs), usually referred to as hot flushes or hot flashes, are the most common symptom reported by climacteric women and are the primary cause for seeking medical advice during this period. The rate of reporting of VMEs is not constant throughout the world; for example, 50% to 85% of women in North America and Europe report VMEs at the time of cessation of ovarian function. The basis for this discrepancy is not understood, but evolutionary and cultural aspects appear to play important roles (50,51). The regular and dose-related diminution of VMEs in symptomatic women described elsewhere in this book attest to the role of estrogen in the occurrence of VMEs. Thermoregulatory centers in the hypothalamus are largely responsible for the control of vasomotor tone and VMEs (see Fig. 14.2). This is discussed in detail elsewhere in this book. VMEs are only one of many estrogen-dependent signs and symptoms of brain dysfunction that are prominent during the climacteric. Interestingly, these manifestations are generally not conceived as "brain symptoms." Rather, they are usually lumped in the category of "menopausal symptoms," thereby losing their significance as harbingers of brain dysfunction and perhaps permanent brain disease. This is a massively underdocumented area in need of much research.
FIGURE 14.5 Illustrationof the widespreadinfluenceof few acetylcholineand estrogen-receptor-containingneuronson the brain.
VMEs can have marked effects on personal and professional performance, especially when they are associated with other estrogen-deficiency brain dysfunction, such as sleep cycle disintegration (52), cognitive disorders (53), and neuroendocrine dysfunction (54,55). Most women have VMEs only for a limited time, 3 to 5 years, although estrogen levels remain low. Some compensatory changes in the brain or elsewhere must underlie the recession of symptoms. To explore this link, we studied VMEs in postpartum women and found that suckling-related VMEs continued past the reestablishment of ovarian cycles, indicating the presence of an intermediary neuronal network such as oxytocin in the chain of VME-related events (56). In any case, the role of estrogen cannot be doubted, because any time women (or men) undergo estrogen withdrawal, they have VMEs and they respond to estrogen replacement (57,58). The therapy for VMEs is estrogen. Although other approaches have been tried, they are largely unsuccessful and do not act to avoid other long-term consequences of ovarian failure. Although phytoestrogens or other forms of estrogen have been used, their efficacy and usefulness against longterm estrogen deficiency remains to be proven (59). As discussed elsewhere, numerous other treatments may substitute for estrogen, but they are secondary treatments.
B. Sleep D i s o r d e r s a n d the C l i m a c t e r i c The hypothalamus contains nuclei involved in sleep regulation (see Fig. 14.5). This hypothalamic area is associated with the circadian clock located in the suprachiasmatic nucleus. The periaqueductal gray matter is near and is involved in sleep regulation. In addition to ERs, the periaqueductal gray matter has a high PR content. Sleep is a reparative brain function. Sleep architecture is influenced by internal and external signals, including estrogen and progesterone. Loss of these hormones results in disintegrating sleep patterns, as measured clinically by
CHAPTER 14 Clinical Effects of Sex Steroids on the Brain electroencephalogram recordings (60). Sleep is clinically divided into two major states: rapid eye movement, (REM) sleep and non-rapid eye movement (NREM) sleep. During sleep, NREM and REM alternate or cycle. The average individual falls asleep within 10 minutes. The first part of sleep normally is an NREM phase, which is followed after 70 to 90 minutes by REM. The onset of sleep and the first REM period is defined as REM latency. Sleep can be disturbed by many external and internal factors. For example, sleep disturbances may be associated with VMEs in menopausal women (52). Associated complaints include sleeplessness, sweating to the point of needing to change clothes during the night, and impairment of one's daily life because of a lack of rest. Our group has shown that another cause of failed sleep, sleep apnea, occurs in menopausal women and responds to hormone therapy (HT). Sleep apnea is important because, in addition to lost sleep, it is thought to be a precursor to nocturnal cardiac arrhythmias and myocardial infarction (61). REM latency is increased and sleep efficiency decreased in menopausal women experiencing VMEs compared with those without VMEs (62,63). Because VMEs and sleep disorders go hand in hand, estrogen is the first-line pharmacologic approach for both disorders (52), and earlier studies showed that estrogen was effective in reducing sleep disorders and hot flushes in menopausal women (68). A pilot study investigated the effects of estrogen replacement therapy (ERT) on the rates of cycling alternate patterns of sleep (CAPS) and nocturnal hot flushes in postmenopausal women. It confirmed the previously described findings. Estrogen decreases the hot flushes and sleeps disturbances, reducing the rate of CAPS (65). Moreover, estrogen treatment of hypogonadal women decreased sleep latency, reduced waking episodes, and prolonged REM sleep (66). Others investigators' analyses of hot flushes and night sweats as the cause of psychosocial, behavioral, and health factors have suggested nonpharmacologic therapy (63), but this should only be tried in cases of failed HT. The key issues are the long-term effects of estrogen deficiency.
C. M o o d a n d the C l i m a c t e r i c Mood is a generic term with many aspects, such as feeling of worth, aggression, and psychomotor activity. Many brain regions are involved in establishing and maintaining mood. The highlighted brain areas in Fig. 14.2 are most directly associated with mood. Many of the same areas are simultaneously involved in the development and level of mood and cognition. Both the climacteric and premenstrual syndrome (PMS) are associated with symptoms such as irritability, anxiety, fatigue, depression, and sleep disturbances, which raises the consideration of hormonal causes. Depression also is more
205 prevalent among women (67), a discordance that could be related to developmental sex differences but more likely reflects the neurotransmitter and hormonal environment. Cyclic affective disorders seem to be more often expressed by females (68); after puberty the rate increases rapidly in girls compared with boys (69). Despite these variations, there is no evidence that depression in women is related only to abnormalities of gonadal function. Women are subject to changes in mood and to hormonal shifts. It is therefore not surprising that the two have been associated, and many therapeutic attempts have been made at connecting them. However, the results of these treatments are equivocal and appear to be specific to the individual being treated. Moreover, these patients often receive mood-altering drugs and HT.
1. ANATOMIC BASIS FOR SEX STEROIDS AND MOOD
Our understanding of the anatomic basis of CNS function is an amalgam of animal studies and clinical observations. The main anatomic areas related to the regulation of mood are limbic brain structures, including the amygdala, hippocampus, parahippocampal gyrus (part of the temporal lobe), thalamus, mamillary body, septum pellucidum, cingulate cortex, and cingulum. The hypothalamus has many connections to these areas and is considered by some to be part of the limbic system. However, because of its neuroendocrine activity, we prefer to consider it a distinct region that contributes to limbic function. All these structures interact by neural networks, together affecting brain functions related to emotion and mood. Similarly, areas of the temporal cortex, including the amygdala, have been related to specific emotions such as joy, fear, and anger. Sex steroids directly affect all these areas. Estrogen receptors are very dense in regions of the limbic system and hypothalamus. ER-ci and ER-[3 have been localized in these regions and may be involved in estrogen's regulation of mood (41,42). 2. FUNCTIONAL BASIS FOR MOOD
The problem of composing a mechanistic description or even a solution for mood changes based on anatomic or neuroendocrine differences in clinical subjects or patients is extraordinarily complex. Animal models are very poor substitutes for human mood or behavior. Most studies have required grossly (i.e., clinically diagnosable) abnormal mood to even show effects of hormone treatment. Less disturbed mood is often identified after its correction, when H T is employed for some other reason. Because of the inability to find endocrine changes that predict or even explain mood disorders (70), we have considered that there is a "substrate" of brain function (the totality of neuronal and nonneuronal cells, connections, and milieu interior) on which is laid the effects of many neurotransmitter and neuroregulatory substances, including sex steroids. Although the complex equation that is represented by
206 "mood" may well respond to hormone replacement (71), especially when the hormone deficiency is direct and abrupt (72-74), numerous psychotropic agents, especially the selective serotonin reuptake inhibitors (SSRIs), may have effects on the same substrate. These effects may be direct or indirect. For example, the SSRIs have been shown to cause rapid changes in serotonin metabolism but only gradual improvement in dysphoria. Some intermediate shift in the substrate seems likely; in such cases, catecholamines may be the final pathway of positive results. In a similar manner, sex steroids may shift the substrate, improving dysphoria and other moods.
3. EFFECTS OF SEX STEROIDS ON MOOD
Hormonal fluctuations have been associated with mood changes and perhaps mood disorders. The effects of changing estrogen levels are likely through changes in neurotransmitter systems. For example, when an estrogen deficit is evident, changes have been reported in the cholinergic, catecholaminergic, and serotoninergic systems (75). Variations in serotonin function are related to mood and depression. Antiserotonin drugs have been shown to induce depression in some humans (76). Estrogen has been shown to influence the midbrain serotoninergic system. Serotonin activity and serotonin receptors are increased by estrogen, while monoamine oxidase (MAO) levels are decreased. Menopause may be accompanied by a decrease in serotonin levels, which is reversed by H T (77). Other neurotransmitter systems may be involved in estrogen's effect on SSRI efficacy. For example, estrogen has also been shown to have a dose-dependent effect the dopamine system. Estrogen increases dopamine transmission and D2 dopamine receptors. Estrogen acts on the neurotransmitters' receptors and synapses involving in all these systems is fast. In contradistinction to the proposed secondary effect of SSRIs, above, estrogens affects on dopamine levels and receptors are very rapid (78). The role of individual sex steroids in mood is of great interest and is still a mystery. Extreme deficiencies of androgen or estrogens respond well to replacement therapy. Less clear are the effects of progesterone or the 19-nor progestins, both of which are employed in H T for their antiestrogenic actions on the endometrium. Despite the perceived clinical wisdom that they cause mood changes, blinded, cross-over studies employing controls have not confirmed a connection between depression and the use of progestins (79,80). Placebo-controlled clinical studies have tested the effects of ERT on mood symptoms in women during the postpartum period and in natural and surgical menopause. Generally, women treated with high-dose estrogen therapy have reported improved symptoms, particularly an improved sense of well-being (71). On the other hand, contradicting studies have reported no effects of hormone replacement on
DA SILVAAND NAFTOLIN
women's mood (69,81). These studies evaluated various estrogen regimens: estrogen alone and combined sequentially with progestin. Negative moods and psychologic symptoms were expressed by women in therapy with low-dose estrogen plus progestin compared with estrogen and placebo (82). Studies in hysterectomized women also reported a decrease in depressive symptoms after treatment with estrogen therapy (72,73,83- 85). ' Mood is a complex term that includes clinical mania, depression, and other disorders. The basis of mood disorders is not clear, but they are often related to or affected by hormone status. However, the long-term effect of repeated progestin administration on such chronic conditions as the neural dystrophies is not well known. Considerations for treatment should follow these guidelines: 1. Treatment should only be for indications. 2. All drugs have side effects, and patients should be accordingly monitored. 3. All efforts made to minimize dosage have thus far been rewarded. 4. Because the basis for success and failure of sex steroids in affecting mood is unclear, other possible underlying causes may appear during treatment. 4. DEPRESSION AND THE CLIMACTERIC Although mood disorders may be troublesome, clinical depression is recognized because of its disabling nature. Severely disordered sleep, psychomotor retardation, and feelings of worthlessness characterize depression and may be associated with suicidal ideation. Clinical depression is a severe dysphoria and requires a rigorous evaluation, followup, and a treatment plan. Considerable evaluation of the possible relationship between thyroid status and adrenocorticoid status and depression has occurred; however, less attention has been given to sex steroids and clinical depression. Despite dropping the diagnosis of"involutional melancholia," longitudinal studies continue to report increased rates of clinical depression during the perimenopause (86-89). Mood changes are influenced by various factors, such as a history of depression or PMS. However, it is hard to evaluate the role of the clinical history, and the literature is ambiguous (90). In any case, it seems that severe psychiatric illnesses commonly recur (91,92). Because of reported improvement in clinically depressed postmenopausal women on treatment with conjugated equine estrogen (93), many attempts at hormonal treatment of depressive symptoms have been undertaken. Most of the studies showed positive results in patients with depressed mood or mild depressive symptoms but did not provide new information about the effect of H T on major depression (71). A
CHAPTER 14 Clinical Effects of Sex Steroids on the Brain report appeared regarding women previously affected by "postpartum depression," a clinical depression marked by psychomotor retardation and sleep disorder. High doses of transdermal estradiol were administered. In this preliminary study, a striking improvement in depression was reported compared with those receiving placebo. During the first month of therapy, about 50% of women reported beneficial effect on depressive symptoms. Those are encouraging results because one-half of the patients reported a rapid effect on mood symptoms in contradistinction to the usual 2-week delay seen with SSRIs (94). Unfortunately, the researchers did not disclose the previous treatment status of their patients. Further and better described studies are needed. Patient selection and the test instruments play important roles in the outcomes and interpretation of studies on depression. Because improvement of depression with psychotropic agents, especially the SSRIs, has been dramatic, it is likely that hormones will remain an adjunct therapy. The first diagnostic measures must distinguish between mood disorders and major depressive illnesses. A dysphoric mood can be treated with estrogens at the outset. Psychotropic agents may then be added. It is important to add tricyclics or SSRIs slowly.Their effects may take some time to be fully felt. Possible synergistic actions between ERT and psychotropic drugs must be kept in mind (95). Estrogen may synergize with antidepressants. In the case of major depression, because ERT has not been shown to regularly or sufficiently improve the symptoms, we consider ERT as second-line adjunctive therapy. In all cases of major depression, a psychiatric consultation must be attained. No evidence supports a role for progestins or androgens in the treatment of major depression.
D. Cognitive F u n c t i o n and the Climacteric Many brain regions are involved in the cognitive process. Although this section focuses on memory and the limbic system, other cognitive functions may bypass the limbic system (see Fig. 14.2). For example, the visual cognitive system apparently has multiple memory and cognitive way-stations that contribute to the complete cognitive process. Cognition is the mental process by which knowledge is acquired or used, and it depends on several elements of the brain functioning harmoniously. These include the intake of information, processing, and distribution of action (including autonomic function) and memory.
ON
1. ANATOMIC BASIS FOR SEX STEROIDS' EFFECTS COGNITION
The areas of the brain involved in cognition are mainly the cerebral cortex, the temporal lobes, and the limbic system. New information enters the brain through the sensory
207 system (i.e., peripheral and cranial nerves) and is processed through the sensory cortex. Each of these areas has been shown to contain ERs (ER-[3 > ER-ot) (41,42) and therefore can be expected to be estrogen-sensitive. Study results have supported the effect of estrogen on cognition (96); however, because ER distribution is selective, it is reasonable to expect that estrogens can affect cells in specific regions of cortex. This high degree of selectivity results in a multitude of cognitive functions that may respond to estrogen. It also complicates evolution of estrogen effects on cognition (Fig. 14.9). More than just the estrogenic environment plays a role in cognitive decline related to aging. For example, from the endocrine side, neuronal loss has been associated with stress, possibly through adrenocorticosteroids (97). Much work remains to complete the understanding of the process, and derailment of memory and estrogen replacement represents a disproportionately large clinical therapeutic area that will shrink as more information appears. 2. EFFECT OF SEX STEROIDS ON COGNITION AND MEMORY
Verbal memory has also been positively correlated with endogenous estrogen levels during the luteal phase of the ovarian cycle and in hormone treated menopausal women (98). Menopausal women undergoing estrogen treatment often remark on the beneficial effects on memory and cognitive functions, even though they did not initially complain of deficits. Several studies have evaluated the effects of estrogen replacement therapy on cognitive function in menopausal women. Although these studies are heterogeneous in their experimental designs and evaluations, in general they support a role for HT, specifically ERT, in maintaining several types of short-term memory and cognition (96,99,100). Other investigators have evaluated treatments using estrogen or estrogen plus androgen versus placebo in women who had undergone total abdominal hysterectomy (TAH) and bilateral salpingooophorectomy (BSO), showing beneficial effects of the drugs on memory (53,101). Similarly, using GnRH analog to induce artificial menopause produced a decline in verbal memory score during the GnRH treatment that was reversed by treatment with conjugated equine estrogen (98). In our own double-blind study, we have shown (unpublished data) that ERT improves verbal and cognitive reading skills (102). Evaluating cognition and memory is a complex task in the usual clinical situation. Most often, the improvement is noticed in retrospect. Although the picture of estrogen's effect on memory is promising, further studies comparing like aspects of memory and excluding confounding side effects of estrogen, such as improved sleep, are needed.
208
DA SILVAAND NAFTOLIN
3. DEMENTIA AND THE CLIMACTERIC
Severe loss of intellectual function and short-term memory in aging subjects are the hallmarks of dementia. AD and vascular dementia are the most frequent diagnoses; the former is more common in the United States. Dementia is estimated to increase approximately 5% per year in women older than 65 years of age, reaching 50% in women older than 85 (103). Alzheimer's dementia is a neurodegenerative disease associated with neuronal cell loss and with development of degenerative lesions called amyloid plaques and neurofibrillary tangles. These lesions are present in relatively larger numbers than seen in normal aging. It is critical to distinguish AD from vascular dementia. The two conditions may coexist, and failure to establish the diagnosis is among the problems surrounding the evaluation of such studies as the subset analysis in the Woman's Health Initiative (104). As in AD, the occurrence of repeated small strokes results in short-term memory loss. On the other hand, long-term memory is more durable and continues to function when not required for storage or retrieval of impaired of shortterm memory and cognition loss. The dementia is progressive and distinguished mainly by a decline in memory followed by a gradual disintegration of intellectual function and orientation, language, judgment, and problem solving.
Ultimately, other vital functions begin to fail, leaving the patient unable to care for herself or himself and subject to terminal wasting or intercurrent illness. All treatment should aim to avoid late progress of AD. Prevention of early AD can also accomplish the same objective (105). It is critical to understand this end point's importance; anything that delays the transition from self-sufficiency is of greater importance than improvement of individual brain activities such as cognitive skill scores. This is the therapeutic goal in AD cases, not complete rehabilitation. Vascular dementia is caused by multiple small infarctions and may occur without a major stroke. Preliminary results indicate a vasodilation effect of estrogen and a vasoconstrictive effect of progestins on cerebral vessels (106).
4. ESTROGEN AND DEMENTIA
Several factors support a protective role of estrogen in dementia, and there is evidence that diminished estrogen, as found in menopausal women, may contribute to the neurodegenerative process associated with dementia (Fig. 14.6). AD has been shown to have an age-corrected rate between 1.4 to 3 times higher in women than in men (45-47). Women with AD have a lower body mass index, which is consistent with estrogen production in adipose tissue weigh less than those without AD, and obese women have a
FIGURE 14.6 Thisillustrates the manner bywhich decreased estrogen could allow activation ofmicroglia to progress past homeostatic killing of the occasional inflammatory cell that enters the area before it causes a snowball of inflammation and self-propelling pathology and neuronal damage/loss. (Modified from ref. 24.)
209
CHAPTER 14 Clinical Effects of Sex Steroids on the Brain higher production rate of estrogen from endogenous androgen in adipose tissue (107). Estrogen may have specific effects on AD, for example, we have shown that estradiol diminishes tau hyperphosphorylation, the apparent basis of the role of tau in the development of microtubule tangles in AD lesions (30). Animal studies also support a protective role for estrogen. In 1998, evidence regarding a possible mechanism through which estrogen exerts an antineurodegenerative role in the brain was reported in a multicenter study. It showed that the metabolism of the Alzheimer's [3-amyloid precursor protein may be regulated by 1713-estradiol in neuroblastoma cells and in primary cell cultures derived from human and rat neocortexes (108). These data emphasize the possible role that estrogen replacement therapy could play in the prevention and delay of the onset of AD. Table 14.3 reviews a group of commonly quoted epidemiologic studies (105,107,109-111), largely supporting lower dementia (AD) incidence in women taking estrogen. Among these, the prospective study by Tang et al. is the most interesting. This group of about 1200 women received neurologic examinations before beginning the study. Although their estrogen usage was not controlled and the case-control method was being used, the study strongly supports a delay by H T of AD diagnosis, which may be dose-related. In the Tang study of ERT patients, the diagnosis of AD was delayed by about 2 years. However, once
TaBLe 14.3
diagnosed, the progression of AD was similar to the progress of the control subjects (105). We have already pointed out the importance of even a short delay of the onset and course of dementia in this elderly population. The apparent similarity in disease progress between H T users and control subjects is consistent with a collateral protective effect by estrogen. Similar implications may be drawn from the only other longitudinal prospective study that tested the neurologic status before hormones were started (112). Not all researchers agree that estrogen delays the incidence of AD. Brenner et al. reported that several oral estrogen preparations were not related to the risk of AD. A negative correlation between ERT and cognitive function was evaluated. They compared women who never used estrogen with women currently under treatment or treated in the past, showing no evident difference on cognitive performance (113). We have mentioned the W H I subgroup analysis earlier, and reintroduce it only to indicate that the results are not valuable due to the age of the subjects at the start of hormone treatment, the lack of minimal cognitive deficit preceding the diagnosis of dementia, and the lack of determination of the role of vascular dementia in the study (114). Although studies using animal models support estrogen's reduction of degenerative processes of aging in the CNS, further clinical prospective evaluation of the role played by estrogen in the onset of AD-like lesions must be accomplished.
Epidemiologic Studies on Estrogen Use and Dementia
Type of study
Number of patients
Case-control study of community elderly women Case-control study of retirement community
143 with AD; 7% taking ERT 92 controls; 18% taking ERT 138 with AD or other dementia; 38% taking ERT 550 controls; 46% taking estrogens 93 with AD; 12% taking ERT 65 with V-D; 11% taking ERT 148 controls whom 20 taking ERT 167 women developed AD from 1124 women, 13% taking estrogens
Study
Year
Henderson et al.
1994
Paganini-Hill et al.
1994
Mortel et al.
1995
Case-control study of friends and relatives
Tang et al.
1996
Prospectivecohort study of community-based study of aging northern Manhattan, New York
Kawas et al.
1997
Prospective cohort of community-dwelling women
Follow-up No report
9 years
14 years
1 - 5 years
34 women developed AD from 16 years 472 post/perimenopause women, 34% taking ERT
AD, Alzheimer's disease; ERT, estrogen replacementtherapy;VD, vascular dementia.
Outcome Postmenopausal ERT may be associated with decreased risk of AD. Increase incidence of AD in elderly women may correlate with estrogen deficiency. Lack of ERT is associated with increased risk of dementia.
AD onset was later in women who had taken estrogen compared with women who never used it. Reduced risk of AD for women who had reported the use of estrogen.
210
DA SILVAAND NAFTOLIN
5. EFFECTS OF SEX STEROIDS ON DEMENTIA
Only a few limited studies exist regarding the treatment of AD with estrogen. In the main, these reports encourage the use of estrogen to improve cognitive functions in AD patients. Honjo et al. showed that six of seven women treated for 6 weeks with ERT, after which ERT was stopped (ERT was stopped after 3 weeks for two women), reported an improvement in cognitive functions compared with nontreated women (115). Later, these results were supported by the only placebo-controlled study reported, but this was also a very limited study (116). Based on the promise of these studies, a large prospective trial of I year of treatment of well-diagnosed women with AD has now been reported that failed to show any improvement of cognition or other measures (117). The combination of the lack of an effect in this large randomized, controlled clinical trial plus the induction of thrombotic disease in elderly women (118) contraindicates the use of estrogen in these cases.
E. Effects o f P r o g e s t e r o n e a n d Progestins on the Brain Progesterone is a 21-carbon steroid normally formed during the metabolic transformation of cholesterol to androgens and estrogens. Because of pharmacodynamic problems, the clinical formulations often use progesterone substitutes (i.e., progestins), which are 19-nor androgens. Progesterone and progestins interact with the androgen, progesterone, and adrenocorticoid receptors. It is not surprising that progestins affect the brain, however, the exact outcomes and their causes are presently under examination. Our review of this topic is abbreviated because it is in a state of flux and new reports often disagree. This may be due to the great differences in available progestin compounds. These include progesterone, androgenic progestins, such as hydroxyprogesterone and 19-nor progestins, and drospirenone, which is an aldosteronederived compound that has no androgenic action. It is important to focus prospective studies on cognitive effects rather than making them subsets of studies designed to test antiestrogenic actions. It is important to understand that the main reason to use progesterone or progestins is because of their antiestrogen qualities (119), although increasing evidence documents direct effects on neurons by progesterone or progestins. The onset of menopause is characterized by a marked reduction of estrogen accompanied by the lack of the luteal (progesterone) phase. As a result, most postmenopausal progesterone comes from the adrenal cortex. The balance of its effect is to antagonize estrogen action. The use of progesterone or progestins may be a disadvantage if the (beneficial) effects of estrogen are opposed throughout
the bo@ However, because estrogen treatment has been associated with an increased rate of endometrial cancer and breast tenderness, progesterone or progestins are prescribed to downregulate the ER-mediated actions. Receptor-mediated and non-receptor-mediated progesterone effects are not confined to the reproductive organs (120). For example, progesterone crosses the blood-brain barrier. Robel and Baulieu indicated that progesterone and its metabolites may also be formed in the brain (121), but in the case of women, this remains a work in progress. 5ot-reductase of progestins occurs in the brain. Tetrahydroprogestin (THP) has been shown to interact with ~/-aminobutyric acid (GABA) receptors in animal brain studies (22). Specific gila cells (i.e., oligodendroglia) perform this metabolism. Astrocytes and some neurons contain 5ot-reductase, which may contribute to T H P formation and action (122). Regional differences in brain steroid concentrations have been shown in women postmortem, emphasizing the local role that may be played by progesterone, its metabolites, and T H P in brain function. Brain T H P variations have been found during autopsies on fertile women during the luteal phase compared with postmenopausal control women (123). Moreover, high plasma levels of progesterone and THP have been correlated with increased fatigue and confusion in healthy women taking an oral dose of micronized progesterone (124). Additional mechanisms by which progesterone and its metabolites can act on the brain have been proposed, largely on the basis of studies in rats. Neuroendocrine and behavioral function can be mediated by progesterone receptors and protein synthesis (125). This action of progesterone may also downregulate ER (119). However, progesterone and its ring-A-reduced metabolites may also affect areas of the CNS where PR expression is low or absent, indicating non-receptor-mediated progesterone actions. Rapid alterations of CNS excitability, producing behavioral effects not attributable to intracellular receptors, have been reported seconds after the parenteral introduction of many steroids (38). This emphasizes the possible sedative roles played by these steroids in physiologic conditions such as pregnancy and stress (126), perhaps through increased GABA-inhibitory synaptic neurotransmission in the brain. This effect is mediated by the interaction between reduced progesterone and the GABA-A receptor complex through activation of receptor-gated C1- channels. In humans, progesterone has been reported to have sedative and anesthetic effects (127). In this regard, administered progesterone had anxiolytic and hypnotic properties (124). These actions are similar to those of the benzodiazepines and barbiturates that are known to interact with GABA-A receptors (128). Although progesterone and progestins have been reported to have similar effects on mood when used in contraception (129-131), such effects remain less documented in postmenopausal women during HT.
CHAPTER 14 Clinical Effects of Sex Steroids on the Brain Despite widely-held beliefs, contradictory findings have been reported in the assessment of the effects of progesterone/ progestins during HT. Two double-blind, placebo-controlled, cross-over trials with postmenopausal women reported no psychologic effects of medroxyprogesterone administered alone (89) or sequentially with transdermal estrogen (79). Absence of negative effects in anxiety and depression has also been reported by other studies employing sequential and combined H T (132). When synthetic progestins were used in sequentially combined transdermal therapeutic systems, no adverse affects were reported regarding mood (133). Despite these studies, the popular anecdotal gynecologists' bias remains that mood changes are often associated with progestins during HT. This is often one of the major causes of poor compliance and is a serious problem because it obstructs the use of liT. It is important to develop objective findings of negative effects of progesterone or progestins on mood. Some studies have confirmed negative effects of progestin added sequentially to estrogen during H T (82,134). However, because negative psychologic changes are not reported by all women, a relation between the dose of progesterone or progestin preparations and rate of mood disturbances has been suggested as an explanation (133). Despite a growing body of evidence indicating that circulating or locally formed progestins may act through the GABA receptor system to affect mood or alertness, the picture remains unclear in HT. In practice we employ minimum systemic progestin treatment, favoring local progestins for protection of reproductive tissue hyperplasia. Although women continue to have a bias against progestins, more evidence is required before this is justified. The recent availability of effective very-low-dose, estrogenonly therapy (monotherapy), plus the evidence that estrogen alone may not affect the incidence of papilloserous endometrial cancer (135) or breast cancer (104), is supportive of monotherapy (136). Moreover, the clinical availability of micronized progesterone and the spironolactone derivative drospirenone (137) are both promising and will have to be carefully evaluated before the final word is written regarding the role of progestins in HT.
E Effects o f A n d r o g e n o n the B r a i n Although there is much data on the role of androgens in behavioral studies on laboratory animals, the clinical literature remains unclear in this area. However, because the removal of the ovaries results in the loss of the only major source of directly secreted testosterone in the body, the case for the use of androgens, particularly methyl testosterone in the United States and tibolone in other countries, remains substantial (138-140). Overall, numerous reports indicate that feelings of well-being and sexual desire are improved in
211 postovariectomy (73) subjects who receive ERT plus androgens. This is the subject of Chapter 55 in this book. Although the ovaries continue to secrete testosterone following menopause, it has recently become of interest to evaluate the use of testosterone and similar compounds in physiologically menopausal women. At this point, the published data are not sufficient to determine the value of this practice. The main indications have been minimal sexual dysfunction and quality-of-life issues. This remains a vexing area, mainly due to the lack of well-accepted diagnostic tools. This area will continue to be of interest, however, and future research will determine the place for androgen treatment in normally menopausal women. Attention has turned toward the possible benefits of dehydroepiandrosterone ( D H E A ) a n d dehydroepiandrosterone sulfate (DHEAS), which are the major 19-carbon secretions from the adrenal gland. Although animal data indicate that D H E A and DHEAS may also play a role in the brain (141), this remains to be shown in humans. Many experiments using animal models support roles for D H E A and its metabolites and congener acting in the brain and tissues (141,142), but these await confirmation in clinical situations (143). These compounds are not overtly androgenic, however, they may be peripherally converted into androstenedione and then into the potent androgens testosterone and dihydrotestosterone. DHEA-derived androstenedione and testosterone may also be converted into estrogen. D H E A and DHEAS decrease gradually in men and women after a peak at early adulthood (144). Beneficial effects on energy longevity, cardiovascular diseases, cancer, and immune response have been reported (145). The absence of proven D H E A receptor in humans, along with pharmacologic regimens that have excluded transformation to active androgens and estrogens, may cause difficulty in considering D H E A or DHEAS as primary preventive or therapeutic agents in hormone replacement and brain function.
References 1. Burger HG. Physiologicalprinciples of endocrine replacement: estrogen. Horm Res 2001;56(suppl 1):82-85. 2. RatkaA, Miller V, Brown K, et al. MenopausalVasomotorSymptoms (MVS) surveyfor assessmentof hot flashes.J WomensHealth (Larchmt) 2006;Jan-Feb:15. 3. Maclennan AH, Henderson VW, Paine BJ, et al. Hormone therapy, timing of initiation, and cognitionin women aged older than 60 years: the REMEMBERpilot study.Menopause 2006;13:28-36. 4. BachmannGA. Menopausalvasomotorsymptoms: a reviewof causes, effects and evidence-based treatment options.J Reprod Med 2005;50: 155-165. 5. McCullough LD, Hurn PD. Estrogen and ischemic neuroprotection: an integrated view. Trends Endocrinol Metab 2003;14:228-235. 6. Barsom SH, Mansfield PK, Koch PB, Gierach G, West SG. Association betweenpsychologicalstress and menstrualcyclecharacteristics in perimenopausal women. WomensHealth Issues 2004;14:235-241.
212 7. Morris JC, Price A1. Pathologic correlates of nondemented aging, mild cognitive impairment, and early-stage Alzheimer's disease.dMdNeurosci 2001;17:101-118. 8. Gibson CL, Gray LJ, Murphy SP, Bath PM. Estrogens and experimental ischemic stroke: a systematic review,d Cereb Blood Flow Metab 2006;Jan 25. 9. Naftolin E Brain aromatization of androgen, d Reprod Med 1994;39: 257-261. 10. Silva I, Mor G, Naftolin E Estrogen and the aging brain. Maturitas 2001;38:95-100. 11. Azcoitia I, Sierra A, Veiga S, Garcia-Segura LM. Aromatase expression by reactive astroglia is neuroprotective.Ann NYAcad Sci 2003;1007: 298-305. 12. Hung AJ, Stanbury MG, Shanabrough M, et al. Estrogen, synaptic plasticity and hypothalamic reproductive aging. Exp Gerontol 2003; 38:53-59. 13. Reichlin S. Neuroendocrinology. In: Wilson J, Foster D, eds. Williams textbook of endocrinology, 8 ed. Philadelphia: VVB Saunders, 1995: 13-19. 14. Naftolin F, Leranth C, Horvath T, Garcia-Segura L. Potential neuronal mechanisms of estrogen actions in synaptogenesis and synaptic plasticity. CellMol Neurobio11996;16:213-223. 15. Woolley CS, Weiland NG, McEwen BS, Schwartzkroin PA. Estradiol increases the sensitivity of hippocampal CA1 pyramidal cells to NMDA receptor-mediated synaptic input: correlation with dendritic spine density. JNeurosci 1997;17:1848-1859. 16. Espeland MA, Rapp SR, Shumaker SA, et al. Conjugated equine estrogens and global cognitive function in postmenopausal women: Women's Health Initiative Memory Study. JHMA 2004;23:2959-2968. 17. Maki PM. Estrogen effects on the hippocampus and frontal lobes. Int J Fertil l,Vomens Med 2005;50:67-71. 18. Craig MC, Maki PM, Murphy DG. The Women's Health Initiative Memory Study: findings and implications for treatment. Lancet Neurol 2005;4:190-194. 19. Maki PM. Potential importance of early initiation of hormone therapy for cognitive benefit. Menopause 2006;13:6-7. 20. Olmos G, Naftolin F, Perez J, Tranque P, Garcia-Segura LM. Synaptic remodeling in the rat arcuate nucleus during estrus cycle. Neuroscience 1989;32:663-667. 21. Negri-Cesi P, Poletti A, Celotti E Metabolism of steroids in the gain: a new insight into the role of the 5-alpha-reductase and aromatase in the brain differentiation and functions. J. Steroid Biochem Mol Bill 1996;58:455-466. 22. Paul SM, Purdy RI--I.Neuroactive steroids. FHSEBJ 1992;6:2311-2322. 23. Pluchino N, Genazzani AD, Bernardi F, et al. Tibolone, transdermal estradiol or oral estrogen-progestin therapies: effects on circulating allopregnanolone, cortisol and dehydroepiandrosterone levels. Gynecol Endocrino12005;20:144-149. 24. Mor G, Nilsen J, Horvath T, et al. Estrogen and microglia: a regulatory system that affects the brain. JNeurobio11999;40:484-496. 25. Bechmann I, Mot G, Nilsen, et al. FasL (CD95L, ApolL) is expressed in the normal rat and human brain-evidence for the existence of an immunological brain barrier. Glia 1999;27:62- 74. 26. Bechmann I, Goldmann J, Kovac AD, et al. Circulating monocytic cells infiltrate layers of anterograde axonal degeneration where they transform into microglia. FASEBJ 2005;19:647-649. 27. Mor G, Sapi E, Abrahams VM, et al. Interaction of the estrogen receptors with the Fas ligand promoter in human monocytes. J Imrnunol 2003;170:114-122. 28. Nilsen J, Mot G, Naftolin E Estrogen-regulated developmental neuronal apoptosis is determined by estrogen receptor subtype and the Fas/ Fas ligand system. J Neurobio12000;43:64- 78. 29. Honjo H, Iwasa K, Kawata M, et al. Progestins and estrogens and Alzheimer's disease. J Steroid Biochem ]ViolBio12005;93:305-308.
DA SILVA AND NAFTOLIN 30. Alvarez De La Rosa M, Silva I, Nilsen J, et al. Estradiol prevents neural tau hyperphosphorylation characteristic of Alzheimer's disease. Ann NYAcad Sci 2005;1052:210-224. 31. Korak KS, Migliaccio S, Basis VL. Estrogens. In: Munson P, Mueller R, Breese G, eds. Principles of pharmacology: basic concepts and clinical implications. New York: Chapman & Hall, 1994;809-826. 32. MacGregor J, Tonetti D, Craig Jordan V. The complexity of selective estrogen receptor modulation: the design of a postmenopausal prevention maintenance therapy. In: Linsday R, Dempster D, Craig Jordan V, eds. Estrogens and antiestrogens. Philadelphia: Lippincott-Raven Publishers, 1997. 33. Mosselman S, Polman J, Dijkema R. ER beta: identification and characterization of a novel human estrogen receptor. FEBS Lett 1996;392: 49-53. 34. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA. Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc NatlAcad Sci USA 1996;93:5925-5930. 35. Giguere V, Tremblay A, Tremblay GB. Estrogen receptor beta: reevaluation of estrogen and antiestrogen signaling. Steroids 1995;63: 335-339. 36. Kuiper G, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 1997:3:863-870. 37. Paech K, Webb P, Kuiper GG, et al. Differential ligand activation of estrogen receptors ER alpha and ER beta at API sites. Science 1997;277:1508-1510. 38. Landers J, Spelsberg T. New concepts in steroid hormone action: transcription factors, proto-oncoeenes and the cascade model for steroid regulation of gene expression. In: Stein G, Stein J, Lian L, eds. Critical review ofeukaryotic gene expression. Boca Raton. FL: CRC Press, 1992. 39. MacLusky NJ, McEwen BS. Progestin receptors in the developing rat brain and pituitary. Brain Res 1980;189:262-268. 40. McEwen B, Davis P, Gerlach J, et al. Progestin receptors in the brain and pituitary gland. In: Bardin C, Milgrom E, Mauvais-Jarvis I, eds. Progesterone andprogestin. New York: Raven Press, 1953. 41. Shughrue PL, Lane MV, Merchenthaler I. Comparative distribution of estrogen receptor-alpha and -beta mRNA in the rat central nervous system. J Cornp Neuro11997;388:507- 515. 42. Laflamme N, Nappi RE, Drolet G, Labrie C, Rivest S. Expression and neuropeptidergic characterization of estrogen receptor (ER alpha and beta) thought the rat brain: anatomical evidence of district role of each subtype. JNeurobio11998;36:357- 378. 43. Honjo H, Tamura T, Matsumoto Y, et al. Estrogen as a growth factor to central nervous cells. Estrogen treatment promotes development of acetyl-cholinesterase-positive basal forebrain neurons transplanted in the anterior eye chamber. J Steroid Biochern Mol Bid 1992;41:633-635. 44. MacLusky NJ, Naftolin E Sexual differentiation of central nervous system. Science 1981;211:1294-1302. 45. Rocca WA, Amaducci LA, Schoenberg BS. Epidemiology of clinically diagnosed Alzheimer's disease. Ann Neuro11986;19:415- 424. 46. Jorm AF, Korten AE, Henderson AS. The prevalence of dementia: a quantitative integration of the literature. Acta Psychiatry &and 1987;76:465-479. 47. Hofman A, Rocca WA, Brayne C, et al. The prevalence of dementia in Europe: a collaborative study of 1930-1990 findings. Eurodem Prevalence Research Group. IntJEpidemio11991;20:736-748. 48. Naftolin F, Leranth C, Perez J, Garcia-Segura LM. Estrogen induced synaptic plasticity in adult primate neurons. Neuroendocrinology 1993; 57:935-939. 49. Jakab RL, Horvath TL, Leranth C, Harada N, Naftolin E Aromatase immunoreactivity in the rat brain: gonadectomy sensitive hypothalamic neurons and an oil responsive "limbic ring" of the lateral septum-bed nucleus-amygdala complex.J SteroidBiochem Mol Bid 1993:44:481-498.
CHAPTER 14 Clinical Effects of Sex Steroids on the Brain 50. Rybo G, Westemberg H. Symptoms in the postmenopausema population study: a preliminary report. Acta Obstet Gynecol&and 1971;9:25. 51. Thompson B, Hart S, Durno D. Menopausal age and symptomatology in general practice.J Biases Sci 1973;5:71-82. 52. Erlik Y, Tataryn LV, Meldrum DR, et al. Association of waking episodes with menopausal hot flash. JAMA 1981 ;245:1741 - 1744. 53. Phillips SM, Sherwin BB. Effects of estrogen in memory function in surgically menopausal women. Psychoneuroendocrinology 1992:17: 485-495. 54. Rosenberg J, Larsen SH. Hypothesis: pathogenesis of postmenopausal hot flush. Med Hypotheses 1991;35:349-350. 55. Meldrum DR, Tataryn LV, Frumar AM, et al. Gonadotropins, estrogens, and adrenal steroids during the menopausal hot flash. J Clin Endocrinol Metab 1950;50:685-689. 56. Greene R, Kletzky O, Cabus E, Naftolin F. Similarities betweenperipheralperfusion augmentation changes induced by postpartum lactation and postmenopausal hotflushes. Presented at the 41st Annual Meeting of the Society for Gynecologic Investigation: Chicago, 1994. 57. Smith J. Management of hot flushes due to endocrine therapy for prostate carcinoma. Oncology 1996;10:1319-1322. 58. Hardt W, Schmidt-Gollwitzer M. Sustained gonadal suppression in fertile women with the LHRH agonist buserelin. Clin Endocrinol 1983;19:613-617. 59. Tham DM, Gardner C, Haskell W. Clinical review 97: potential health benefits of dietary phytoestrogens: a review of the clinical, epidemiological, and mechanistic evidence. J Clin Endocrinol Metab 1998:83:2223 - 2235. 60. Kupfer D. The sleep LEG in diagnosis and treatment of depression. In: Rush A, Altshuler K, eds. Depression, basic mechankms diagnosk and treatment. New York: Guilford, 1986. 61. Keefe DL, Watson R, Naftolin E Hormone replacement therapy alleviated sleep apnea in menopausal women: a pilot study. Menopause 1999;6:196-200. 62. Shaver J, Giblin E, Lents M, Lee K. Sleep patterns and stability in perimenopausal women. Sleep 1988;11:556-561. 63. Avid NE, Crawford SL, McKinlay SM. Psychosocial, behavioral, and health factors related to menopause symptomatology. Womens Health 1997;3:103-120. 64. Thomson J. Double blind study on the effect of estrogen on sleep, anxiety and depression in perimenopausal women: preliminary results. Proc R Soc Med 1976;69:829-830. 65. Scharf MB, McDannold MD, Stover R, Zaretsky N, Berkowitz DV. Effects of estrogen replacement therapy on rated of cyclic alternating patterns and hot-flush events during sleep in postmenopausal women: a pilot study. Clin Ther 1997;19:304-311. 66. SchiffI, Regestein Q~Tulehinisky D, et al. Effects of estrogens on sleep and psychological slate of hypogonadal women. JAMA 1979;212:2405. 67. Weissman M, Bland R, Canino G. Cross-national epidemiology of major depression and bipolar disorder. JAMA 1996;276:293-299. 68. Parry B. Reproduce factors affecting the course of affective illness in women. Psychiatr d/in North Am 1989;12:207-220. 69. Kessler RC, McGonagle KA, Swartz M, Blazer DG, Nelson CB. Sex and depression in the National Comorbidity Survey. I: Lifetime prevalence, chronicity and recurrence. JAffect Discord 1993;9:85-96. 70. Saks BR, Frank JB, Lowe TL, et al. Depressed mood during pregnancy and the puerperium: Clinical recognition and implications for clinical practice. Am J Psychiatry 1985;142:728- 731. 71. Zweifel JE, O'Brien WH. A meta-analysis of the effect of hormone replacement therapy upon depressed mood. Psychoneuroendocrinology 1997;22:189-212. 72. Sherwin BB, Gelfand MM. Effects of parenteral administration of estrogen and androgen on plasma hormone levels and hot flushes in the surgical menopause. AmJ Obstet Gyneco11984;148:552-557.
213 73. Sherwin BB, Gelfand MM. Differential symptoms response to parenteral estrogen and/or androgen administration in surgical menopause. AmJ Obstet Gyneco11985;151:153-160. 74. Resnick SM, Maki PM, Rapp SR, et al. Effects of combination estrogen plus progestin hormone treatment on cognition and affect. J Clin Endocrinol Metab 2006;91:1802-1810. 75. Schmidt P, Roca C, Ribinow D. Perimenopausal depression. In: Lobo R, ed. Perimenopause. New York: Springer, 1997:246-254. 76. Keefe D, Naftolin E Brain neurochemistry and mood. In: Smith S, Schiff I, eds. Modern management ofpremenstrual syndrome. New York: W W Norton, 1992;55-70. 77. Slopien R, Meczekalski B, Warenik-Szymankiewics A. Relationship between climacteric symptoms and serum serotonin levels in postmenopausal women. Climacteric2003;6:53-57. 78. Jacobs PA, Hyland ME. An evaluation of the benefits of taking hormone replacement therapy with other prescription drugs. Maturitas 2003;46:273-281. 79. Kirkham C, Hahn PM, Van Vugt DA, Carmichael JA, Reid RL. A randomized double blind placebo-controlled cross-over trial to assess the side effects of medroxyprogesterone acetate in hormone replacement therapy. Obstet Gyneco11991;78:93-97. 80. Prior JC, Alojado N, McKay DW, Vigna YM. No adverse effects of medroxyprogesterone treatment without estrogen in postmenopausal women: double blind placebo controlled crossover trial. Obstet Gynecol 1994;83:24-28. 81. Myers L, Dixen J, Morissette D, Carmichael M, Davidson JM. Effects of estrogen, androgen, and progestin on sexual psychobiology and behavior in postmenopausal semen. J Clin Endocrinol Metab 1990;70: 1124-1131. 82. Sherwin B. The impact of different doses of estrogen and progestin on mood and sexual behavior in postmenopausal women. J Clin Endocrinol Metab 1991;72:336-343. 83. Maki P, Hogervorst E. The menopause and HRT. HRT and cognitive decline. Best Pract Res Clin EndocrinolMetab 2003;17:105-122. 84. Schmidt PJ. Mood, depression, and reproductive hormones in the menopausal transition. Am J Meat2005;118:54- 58. 85. Richards M, Rubinow DR, Daly RC, Schmidt PJ. Premenstrual symptoms and perimenopausal depression. Am J Psychiatry 2006;163: 133-137. 86. Avi NE, Crawford S, Stellato R, Longcope C. Longitudinal study of hormone levels and depression among women transitioning through menopause. Climacteric 2001;4:243-249. 87. Hunter MS. Psychological and somatic experience of the menopause: a prospective study. PsychosomMed 1990;52:357- 367. 88. Hay AG, Bancroft J, Johnstone EC. Affective symptoms in women attending a menopause clinic. BrJ Psychiatry 1994;164:513-516. 89. Schmidt P, Roca C, Ribinow D. Perimenopausal depression. In: Lobo R, ed. Perimenopause. New York: Springer, 1997;246-254. 90. Bremner JD, Vythilingam M, Ng CK, et al. Regional brain metabolic correlates of alpha-methylparatyrosine-induced depressive symptoms: implications for the neural circuitry of depression. JAMA 2003;289: 3125-3134. 91. Parry B. Reproductive related depression in women: phenomena of hormonal kindling? In: Hamilton J, Harberger P, eds. Postpartumpsychiatric illness:a picturepuzzle. Philadelphia: University of Pennsylvania Press, 1992. 92. Yorkers K. The association between premenstrual dysphoric disorder and other mood disorder. J Clin Psychiatry 1997;58(suppl):19-25. 93. Klaiber EL, Brovermen DM, Vogel W, Kobayashi Y. Estrogen therapy for severe persistent depression in women. Arch Gen Psychiatry 1979;36:550-554. 94. Henderson AF, Gregoire AJ, Kumar RD, Studd JW. Treatment of severe postnatal depression with oestradiol skin patches. Lancet 1991;338:816-817.
214 95. Schneider LS, Small GW, Hamilton SH, Bystritsky A, Nemeroff CB, Meyers B S. Estrogen replacement and response to fluoxetine in a multicenter geriatric depression trial. Fluoxetine collaborative study group. Am J Geriatr Psychiatry 1997;5:97-106. 96. Wickelgren I. Estrogen stakes claim to cognition. Science 1997;276: 675-678. 97. Lupien SJ, McEwen BS. The acute effects of corticosteroids on cognition: integration of animal and human studies. Brain Res Rev 1997; 24:1-27. 98. Davis SR. 'Tkdd back" estrogen reverses cognitive deficits induced by a gonadotropin-releasing hormone agonist in women with leiomyomata uteri. J Clin EndocrinolMetab 1997;82:702- 703. 99. Haskell SG, Richardson ED, Horwitz RI. The effects of estrogen replacement therapy on cognitive function in women: a critical review of the literature. J Clin Epidemio11997;50:1249-1264. 100. Sherwin BB. Estrogen and cognitive functioning in women. Proc Soc Exp Bid Med 1998;217:17-22. 101. Sherwin B. Estrogen and/or androgen replacement therapy and cognitive functioning in surgically menopausal women. Psychoneuroendocrinology 1988;13:345-355. 102. Shaywitz SE, Naftolin F, Zelterman D, et al. Better oral reading and short-term memory in midlife, postmenopausal women taking estrogen. Menopause 2003;10:420-426. 103. Bechman DL, Wolf PA, Linn R, et al. Prevalence of dementia and probable senile dementia of Alzheimer type in the Framingham study. Neurology 1992;42:115-119. 104. Stefanick ML, Anderson GL, Margolis KL, et al. Effects of conjugated equine estrogens on breast cancer and mammography screening in postmenopausal women with hysterectomy. JAMA 2006;295: 1647-1657. 105. Tang MX, Jacobs D, Stern Y, et al. Effect of oestrogen during menopause on risk age at onset of Alzheimer's disease. Lancet 1996; 348:429-432. 106. Schneck MJ, Sarrel PI, Albakri E, et al. Oral progesterone therapy impairs cerebrovascular reactivity. Stroke 1995;26:724. 107. Henderson VW, Paganini-Hill A, Emanuel CK, Dunn ME, Buckwalter JG. Estrogen replacement therapy in older women. Arch Neurol 1994;51:896- 900. 108. Yamamoto KR. Steroid receptor regulated transcription of specific genes and gene networks. Annu Rev Genet 1985;19:209-252. 109. Paganini-Hill A, Henderson VW. Estrogen deficiency and risk of Alzheimer's disease in women. Am JEpidemio11994;140:256-261. 110. Mortel KF, Meyers JS. Lack of postmenopausal estrogen replacement therapy and the risk of dementia. J Neuropsychiatry Clin Neurosci 1995;7:334-337. 111. Kawas C, Resnick S, Morrison A, Brookmeyer R, Corrada M, Zonderman A, Bacal C, Lingle DD et al. A prospective study of estrogen replacement therapy and the risk of developing Alzheimer's disease: the Baltimore Longitudinal Study of Aging. Neurology 1997;48:1517-1521. 112. Zandi PP, Carlson MC, Plassman BL, et al. Hormone replacement therapy and incidence of Alzheimer disease in older women: the Cache County Study. JAM_d 2002;288:2123-2129. 113. Brenner DE, Kukull WA, Stergachis A, et al. Postmenopausal estrogen replacement therapy and risk of Alzheimer's disease: a population based case-control study. Am J Epidemio11994;140:262-267. 114. Shumaker SA, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women's Health Initiative Memory Study. JAMA 2004;291:2947-2958. 115. Honjo H, Ogino Y, Naitoh K, et al. In vivo effects by estrone sulfate on the central nervous system-senile dementia (Alzheimer's type). J Steroid Biochem 1989;34:521-525.
IDA SILVA AND NAFTOLIN 116. Honjo H, Ogino Y, Naftolin F, et al. An effect of conjugated estrogen to cognitive impairment in women with senile dementia Alzheimer's type: a placebo-controlled double blind study. Inter Menopause Soc 1993;1:167-171. 117. Mulnard RA, Cotman CW, Kawas C, et al. Estrogen replacement therapy for treatment of mild to moderate Alzheimer disease: a randomized controlled trial. Alzheimer's Disease Cooperative Study. JAMA 2000;283:1007-1015. Erratum in: JAMA 2000;284:2597. 118. Langer RD, Pradhan AD, Lewis CE, et al. Baseline associations between postmenopausal hormone therapy and inflammatory, haemostatic, and lipid biomarkers of coronary heart disease. The Women's Health Initiative Observational Study. Thromb Haemost 2005;93:1108-1116. 119. Clarke CL, Sutherland RL. Progestin regulation of cellular proliferation. Endocr Rev 1990;11:266-301. 120. Graham J, Clarke C. Physiological action of progesterone in target tissues. Endocr Rev 1997:18:502-519. 121. Robel P, Baulieu EE. Neurosteroids: biosynthesis and function. Crit Rev Neurobio11995;9:383- 394. 122. Poletti A, Rabuffetti M, Celotti F. The 5a-reductase in the rat brain. In: Genazzani A, Petraglia P, Purdy R, eds. The brain sourceand target ofsex steroids hormones. London: The Parthenon Publishing Group, 1996. 123. Bixo M, Andersson A, Winblad B, Purdy RH, Backstrom T. Progesterone, 5alpha-pregnane-3,20-dione and 3alpha-hydroxy-5alphapregnane-20-one in specific regions of the human female brain in different endocrine states. Brain Res 1997;764:173-178. 124. Freeman EW, Purdy RH, Coutifaris C, Rickets K, Paul SM. Anxiolyric metabolites of progesterone: correlation with mood and performance measures following oral progesterone administration to healthy female volunteers. Neuroendocrinology 1993;58:478-484. 125. Yamamoto KR. Steroid receptor regulated transcription of specific genes and gene networks. Annu Rev Genet 1985;19:209-252. 126. Purdy RH, Morrow AL, Moore PH Jr, Paul SM. Stress-induced elevations of aminobutyric acid type A receptor-active steroids in the rat brain. ProcNatlAcad Sci USA 1991;88:4553-4557. 127. Selye H. The anesthetic effect of steroid hormones. Proc Soc Eur Bid 1941;46:116-121. 128. Majewska MD, Harrison NL, Schwartz RD, Barker JL, Paul SM. Steroids hormone metabolites are barbiturate-like modulators of GABA receptor. Science 1986;232:1004-1007. 129. Kane FJ Jr. Evaluation of emotional reactions to oral contraceptive uses. Am J Obstret Gyneco11976;126:968-972. 130. Wagner KD. Major depression and anxiety disorder associated with Norplant.J Clin Psychiatry 1996;57:152-157. 131. Jensvold M. Nonpregnant reproductive-age women. Part II. Exogenous sex steroid hormones and psychopharmacology. In: Jenseold M, Halbreich L, Hamilton J, eds. Sex, gender and hormones. Washington, DC: American Psychiatric Press, 1996. 132. Siddle NC, Fraser D, Whitehead MI, et al. Endometrial, physical and psychological effects ofpostmenopausal oestrogen therapy with added dydrogesterone. Br J Obstet Gynaeco11990;97 :1101-1107. 133. Whitehead MI, Hillard TC, Crook D. The role and use of progestogens. Obstet Gyneco11990;75:$59- $76. 134. Holst J, Backstrom T, Hammarback S, von Schoultz B. Progesteron addition during estrogen replacement therapy-effects on vasomotor symptoms and mood. Maturitas 1989;11:13-20. 135. Naftolin F, Silver D. Is progestogen supplementation of ERT really necessary? Menopause 2002;9:1 - 2. 136. Richman S, Edusa V, Fadiel A, Naftolin F. Low-dose estrogen therapy for prevention of osteoporosis: working our way back to monotherapy. Menopause 2006;13:148-155.
CHAPTER 14 Clinical Effects of Sex Steroids on the Brain 137. Stevenson JC. A new hormone replacement therapy containing a progestogen with anti-mineralocorticoid activity. J Br Menopause Soc 2006;12(suppl 1):8-10. 138. Warnock JK, Swanson SG, Borel RW, et al. Combined esterified estrogens and methyltestosterone versus esterified estrogens alone in the treatment of loss of sexual interest in surgically menopausal women. Menopause 2005;12:374- 384. 139. Somboonporn W, Davis S, Seif MW, Bell R. Testosterone for periand postmenopausal women. Cochrane Database Syst Rev 2005;19: CD004509 140. Liu JH. Therapeutic effects of progestins, androgens, and tibolone for menopausal symptoms. Am J Med 2005;118(suppl 2):88- 92. 141. Ren X, Noda Y, Mamiya T, Nagai T, Nabeshima T. A neuroactive steroid, dehydroepiandrosterone sulfate, prevents the development of morphine dependence and tolerance via c-los expression linked to the extracellular signal-regulated protein kinase. Behav Brain Res 2004;152:243-250.
215 142. Ma]ewska MD. Neuronal actions of dehydroepiandrosterone. Possible role in brain development aging, memory and affect.Ann NYAcad Sci 1995;774:111-120. 143. Morales AJ, Nolan JJ, Nelson JC, Yen SS. Effect of replacement dose of dehydroepiandrosterone in men and women of advancing age. J Clin Endocrinol Metab 1995;80:2799. 144. Ravaglia G, Forti P, Maioli F, et al. Dehydroepiandrosterone-sulfate serum levels and common age-related diseases: results from a crosssectional Italian study of a general elderly population. Exp Gerontol 2002;37:701-712. 145. Dayal M, Sammel MD, Zhao J, et al. Supplementation with DHEA: effect on muscle size, strength, quality of life, and lipids. J Womens Health (Larchrnt) 2005;14:391-400.
This Page Intentionally Left Blank
~ H A P T E R 1_
Impact of the Changing Hormonal Milieu on Psychological Functioning BARBARA
B.
SHERWIN
Department of Psychology, McGill University, Montreal, Q.uebec, Canada H3A 1B1
The steady increase in female life expectancy over the past century means that women now live one-third of their lives beyond cessation of their ovarian functioning. Therefore, quality-of-life issues related to aging in women have assumed increasing importance in the minds of health professionals and of women themselves. Prominent among the complaints of perimenopausal and postmenopausal women are changes in behavior. More specifically, changes in mood, in memory, and in sexual functioning are frequently reported around the time of menopause. Although it must be acknowledged that sociocultural factors, individual factors, and environmental factors likely converge to influence the experience of the menopause for each individual woman, there is also accumulating evidence that changes in the hormonal milieu may underlie some of the psychological symptoms reported at this time. Moreover, to the extent that specific sex hormones may enhance specific aspects of psychological functioning, it follows that their administration, after the menopause, may serve to maintain these functions, thereby preserving the quality of
ACKNOWLEDGMENT: The preparation of the manuscript for this chapter was supported by research funds associated with the CIHR Distinguished Scientist Award to B. B. Sherwin. TREATMENT OF THE POSTMENOPAUSAL W O M A N
217
life in aging women. This chapter reviews the mechanisms of action of the sex steroids on the central nervous system and the clinical literature that bears on the relationship between sex hormones and behavior with the goal of providing a clearer understanding of the role of hormones in psychological functioning in menopausal women.
I. E P I D E M I O L O G Y OF PSYCHOLOGICAL SYMPTOMS Prior to the attempt to explain the manner in which changes in the endocrine milieu may precipitate psychological symptoms at menopause, it is important to establish that these symptoms do indeed occur in a considerable number of women. Several epidemiologic studies undertaken on random samples of the population in the United States (1-2) as well as in Sweden (3) and England (4) have failed to document an increase in depressive symptoms around the time of menopause compared with other times during the lifespan. Inconsistent with these reports are the results of other studies of nonclinical populations that have found increased incidences of depressive symptoms in perimenopausal and postmenopausal women (5-8). Interestingly, in a longitudinal survey of 2500 middle-aged women in Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
218
Massachusetts, those who had undergone a surgical menopause had high and clinically significant depression scores, whereas the naturally menopausal women did not (9). Evidence from studies undertaken in menopause clinics tells yet a different story. Of 100 menopausal women who sought medical care in one clinic, 79% had physical symptoms such as hot flushes and 65% had varying degrees of depression (10). Similarly, in a British study, 89% of all women who attended a menopause clinic had depressive symptoms (11). Thus, it would seem that whereas the occurrence of depressive symptoms is far from a ubiquitous phenomenon at the time of menopause, many women do, in fact, develop these symptoms, particularly during the perimenopause. Moreover, there is some evidence that vulnerability toward the development of depressive symptoms may be greater in women who undergo a surgical menopause and those who had experienced prior episodes of depression. Complaints of memory and concentration difficulties are frequently voiced around the time of menopause. Indeed, the inclusion of items to assess memory and concentration abilities in one of the earliest menopausal symptom checklists (1) serves as acknowledgment that subjective complaints of changes in cognitive functioning have long been associated with the menopause. Unfortunately, no epidemiologic studies are available to shed light on the frequency and severity of these changes around the time of menopause. Survey data point out the considerable frequency of a variety of sexual problems in menopausal women. In a nonclinical sample, Kinsey et al. (12) documented a 53% and a 48% decrease in the frequencies of coitus and orgasm respectively in the postmenopause. Longitudinal studies of menopausal women in Sweden that found a 52% decrease in sexual interest and a 20% decrease in orgasm frequency also established statistically that these decrements in sexual functioning were related to menopause and not to aging per se (13). Numerous survey studies of menopausal women have confirmed decreases in sexual interest of 33% (14), 85% (15), and 39% (16). These epidemiologic data serve to point out that between one-third and one-half of menopausal women recruited from the general population complain of a problem in one or more aspects of sexual functioning.
II. NEUROBIOLOGIC EFFECTS OF ESTROGEN, ANDROGEN, AND PRO GESTIN In order to account for changes in sex hormone production as determinants of psychological symptoms in menopausal women, it is necessary to review briefly some of the known effects of these hormones on the brain. Estrogen has both direct and inductive effects on neurons. Direct effects of estrogen on the brain take place fairly rapidly. For
BARBARAB. SHERWIN
example, estrogens alter the electrical activity of the hypothalamus (17). On the other hand, inductive effects of estrogen are both delayed in onset and prolonged in duration. Its mode of action here is assumed to occur via its induction of ribonucleic acid (RNA) and protein synthesis by means of genomic mechanisms, which, in turn, cause changes in levels of specific gene products such as neurotransmitter synthesizing enzymes (18). In rats, autoradiographic studies have demonstrated that neurons that contain specific nuclear receptors for estrogen are found in specific areas of the brain, predominantly in the pituitary, hypothalamus, limbic forebrain (including the amygdala and lateral septum), and cerebral cortex (19). Specific cytosolic receptors for testosterone are found mainly in the preoptic area of the hypothalamus, with smaller concentrations in the limbic system and in the cerebral cortex (19). Within the past 15 years, the direct modulation of neural development and neural circuit formation by estrogen has been demonstrated in both the neonatal and adult brain. Specifically, estrogen has a facilitatory effect on synapse formation in the hypothalamus in both adult and aged female rats (20). Estrogen can also exert a stimulatory influence on synaptogenesis in the ventromedial nucleus, in the lateral septum, and in the midbrain central gray area of adult rats (21). Moreover, androgen plays a significant role in regulating synaptic remodeling in the androgen-sensitive spinal motoneuron pools (22). Thus, the sex steroids seem to play a critical role in the organization and reorganization of neural circuitry, driving neuroendocrine and behavioral functions both during development and in adulthood. The idea that estrogen interacts with genomic mechanisms to regulate levels of specific gene products led to discoveries of numerous neurochemical effects of this sex steroid. Among the most relevant of its mechanisms of action with respect to the development of depression at the time of menopause is that estrogen increases the rate of degradation of monoamine oxidase (MAO), the enzyme that catabolizes the neurotransmitter serotonin (23), whose deficiency is thought to be one causal factor in depression. One action of potential importance for memory functions is the induction by estrogen of choline acetyltransferase (CAT), the enzyme that is needed to synthesize acetylcholine (24) and whose deficiency is a hallmark of Alzheimer's disease (25). Moreover, estrogen positively effects neuronal architecture by increasing spine density in the CA 1 pyramidal cells in the hippocampus (26) and it has been suggested that estrogen's trophic and plastic effects may be caused, in part, by neurotrophic factors (27). Estrogen also may act as an antioxidant (28). Finally, estrogen enhances adrenergic function, which has also been linked to cognitive abilities. In one study, the administration of clonidine, an ~-adrenergic agonist, was associated with dosedependent decreases in plasma norepinephrine levels and with reduced mental performance scores (29).
CHAPTER 15 Impact of the Changing Hormonal Milieu on Psychological Functioning Functional imaging techniques have also been used to examine estrogenic effects on brain areas involved in cognitive functioning. For example, in a magnetic resonance imaging (MRI) study, estrogen-induced changes in brain activation patterns occurred during the administration of both verbal and nonverbal memory tests (30). In a positron emission tomography (PET) study that examined alterations in regional cerebral blood flow (rCBF), estrogen use was associated with increased rCBF in the right hippocampus, parahippocampal gyms, and middle left temporal gyms, all regions implicated in learning and memory (31). In summary, neurophysiologic studies undertaken to localize function of estrogen and testosterone provide evidence that both sex steroids are found in areas of the brain that are thought to subserve emotion, cognition, and sexuality. Moreover, findings from basic neuroscience show that estrogen can profoundly affect the concentration of neurotransmitter synthesizing and catabolizing enzymes, which, in turn, influence the brain concentrations and synaptic availability of certain neurotransmitters. These mechanisms of estrogenic action on brain chemistry provide possible explanations for the purported influence of this hormone on mood and memory. In contradistinction to the stimulatory effects of estrogen on various brain mechanisms, progestins have potent anesthetic properties (32). Indeed, the administration of large doses of progestins to humans induces dizziness, drowsiness, and even deep sleep (33). Moreover, whereas estrogen decreases MAO activity in the amygdala and hypothalamus (24), progestins increase it (34), thereby resulting in lower concentrations of brain serotonin, which may predispose to dysphoric moods. These neurochemical effects of progestins may explain why some women on estrogen-progestin replacement regimens experience unpleasant behavioral side effects, as will be discussed in a subsequent section.
III. SEX S T E R O I D S A N D M O O D The most common strategy for investigating possible effects of estrogen on mood is to test menopausal women before and after a trial of hormone replacement therapy. The findings of many early studies are difficult to interpret because of their methodologic shortcomings: Some studies were not blinded (35) or contained women who had malignant disease (36) or concurrent psychiatric illness (37). Moreover, none of these studies measured circulating levels of the sex hormones coincident with psychological testing. Several investigations undertaken in nonpsychiatric populations of postmenopausal women have reported changes in affect as a function of circulating sex hormone levels. In a prospective randomized controlled trial (RCT) (38), premenopausal women who were scheduled for total abdominal hysterectomy (TAH) and bilateral salpingo-oophorectomy
219
(BSO) for benign disease were tested 1 month before surgery. Following TAH and BSO, these women received estrogen, androgen, an estrogen-androgen combined preparation, or placebo administered intramuscularly once a month for 3 months. During the fourth postoperative month, all women received placebo and, 1 month later, were crossed over to another treatment they had not already received for an additional 3 months. In addition to the placebo group, another group of women who required TAH only were included in order to control for the surgical procedure itself. Blood was sampled at each test time for assay of circulating levels of the sex hormones. In both treatment phases, women who received placebo had higher depression scores when compared with those treated with any of the active hormone preparations and compared also with the ovary-intact group (Fig. 15.1). Furthermore, depression scores of the hormonetreated women increased during the placebo month between the two treatment phases coincident with their significant decrease in plasma estradiol and testosterone levels at that time, indicating that they had more positive moods while they were receiving hormone therapy. In a cross-sectional study, Sherwin (39) investigated otherwise healthy women who had undergone TAH and BSO for benign disease approximately 4 years earlier. Women who had been receiving estrogen or an estrogentestosterone combined drug intramuscularly once a month for the previous 2 years had more positive moods than a matched control group of women who had remained untreated since their TAH and BSO 4 years earlier. In an RCT of asymptomatic surgically menopausal women, depression scores decreased in those given either 0.625 mg or
A_J U ,<
TAH ond BSO ~
18
CROSS OVER I I
,,I[
tn 14,1 at
14
i
12
I,
o
U tlt z O
u') tn I,M a_ LU
f~ z < uJ :E
E-A
108
i E A
O'
!
BASELINE
|
TREATMENTI
PLACEBO
I
TREATMENT2
FIGURE 15.1 Mean depression scores _ SEM. The treatment groups
are: E-A(A), E(0), A(EI),PL(O), and hysterectomyCON(Q). BSO,bilateral salpingo-oophorectomy; CON, hysterectomized control group with intact ovaries; PL, placebo group; TAH, total abdominal hysterectomy. (From ref. 38.)
220 1.25 mg conjugated equine estrogen (CEE) for 3 months, whereas no mood changes occurred in patients who had received placebo (36). As in the Sherwin et al. (1988) study (39), mood scores fell within the normal range of values during both the pretreatment and the posttreatment test times. These findings confirmed the association between positive moods and serum levels of estradiol and testosterone in healthy, nondepressed women. It is important to note that the positive association between mood and sex hormone levels seen in these studies (38,39) occurred when the two variables fluctuated within the normal range. That is, when hormones were withdrawn during the placebo month, depression scores increased in all women, indicating more dysphoric moods. However, although depression scores increased in these surgically menopausal women when they received a placebo, they did not become clinically depressed; that is, their scores on the depression scale were higher than they were when they were receMng hormone therapy but they were not in the clinically depressed range. What they complained of subjectively when their hormone levels were low were feelings of sadness or being "blue," but the severity of their symptoms fell short of the diagnostic criteria for a major depressive disorder (MDD). Recently, several investigators have attempted to determine whether exogenous estrogen would be effective as the sole treatment for M D D in perimenopausal or postmenopausal women. When perimenopausal women (23% with a diagnosis of M D D and 77% with minor depression) received either 0.05 mg estradiol transdermaUy or placebo for 3 weeks, a full or partial therapeutic response was seen in 80% of women treated with estradiol and in 22% of those treated with placebo (40). However, other symptoms commonly seen in depressed indMduals, such as disturbed sleep, lack of energy, somatic preoccupation, and feelings of unreality, failed to improve with 3 weeks of estrogen treatment. Therefore, although estrogen treatment significantly improved depressed mood in a considerable number of these depressed women, it failed to alleviate the entire constellation of symptoms that makes up the depressive syndrome. In another recent RCT, 50 perimenopausal women with M D D (52%), dysthymic disorder (22%), or minor depressive disorder (26%) were randomly treated with transdermal patches of 100 txg estradiol or placebo for 12 weeks (41). Then, the treatments were discontinued and the women were reassessed 4 weeks later. Remission of depression occurred in 68% of women treated with estradiol compared with a 20% remission rate in the placebo group. Moreover, in the estrogen responders, remission occurred irrespective of the type of depression. As might be expected, hot flush frequency increased 4 weeks after discontinuation of estrogen but mood scores remained significantly below scores recorded at pretreatment baseline. Others have also found that the beneficial effect of
BARBARAB. SHERWIN
estrogen on affective symptoms was direct and did not occur secondary to the alleviation of hot flushes (38,39), suggesting that vasomotor symptoms and mood are independent and differentially mediated. Although estrogen has been shown to be moderately effective for the treatment of depression in perimenopausal women, there is evidence that its efficacy does not extend to the treatment of depression in older postmenopausal women. For example, when mild to moderately depressed women who had undergone menopause approximately 17 years earlier were randomized to treatment with either 100 Ixg estradiol or placebo transdermaUy for 8 weeks, there was no difference in remission rate between the estrogen and placebo groups (42). When both perimenopausal and postmenopausal women were treated with 100 txg estradiol transdermally in an open-label trial, 6 of the 9 perimenopausal women but only 2 of the 11 postmenopausal women achieved remission of their depression after 4 weeks of treatment (43). Therefore, although the use of estrogen as a sole agent is somewhat effective for the treatment of depressive episodes in perimenopausal women, at the present time the evidence suggests that this does not hold true for postmenopausal women and that estrogen alone should not be used as treatment for depression in this older population. Other investigators have hypothesized that estrogen might be an effective adjunctive treatment in women with affective disorders who respond partially, or not at all, to treatment with antidepressants. In a retrospective study of premenopausal and postmenopausal women with MDD, all patients received 20 mg fluoxitine per day for 12 weeks (44). No differences in rates of remission or partial response occurred between women who were and who were not taking estrogen, indicating that estrogen failed to augment response to the antidepressant. However, the study was retrospective and women had not been randomly assigned to estrogen therapy (oral conjugated equine estrogen, doses not reported). In an open-label study of 16 perimenopausal women with MDD, 10 were not taking any antidepressants, whereas 6 women were receiving approximately 30 mg fluoxitine daily at the start of the study (45). All women were given 0.3 mg oral esterified estrogens daily for 8 weeks. Both the estrogen alone and the estrogen plus fluoxetine groups experienced a significant reduction in depression scores after the first week of treatment. This suggested that estrogen alone is effective for the treatment of depression in perimenopausal women and effective also as adjunctive treatment for women who had no response or only a partial response to antidepressants alone. However, the design limitations of this study suggest that the findings be interpreted cautiously. Finally, in another openlabel study, women whose M D D had improved but had not remitted following treatment with 100 txg transdermal estradiol for 4 weeks were given 20 mg citalopram daily in addition to the estradiol (46). Remission of depression occurred in 92% of these postmenopausal women (mean age, 51 years)
CHAPTER 15 Impact of the Changing Hormonal Milieu on Psychological Functioning 7.0
7.0 CEE 0.625mg + MPA 5mg
6.5
M
= o o
"o o o E c M G)
6.5
6.0
6.0
5.5
5.5
V. SEX STEROIDS AND SEXUALITY
CEE 0.625mg + Placebo
5.0
4.5
4.5
4.0
4.0
3.5 7.0
3.5
P1234123412341234
7.0
CEE 1.25mg + MPA 5rag
6.5
,, ,. ~..Jr
,~'~
T-.z
6.0
5.5
5.5
5.0
5.0
4.5
4.5 4.0
4.0 3.5
. . . . . . . . . . . . . . . . . P 234123412341234
6.5
6.0
i
I
i,,
I
I
I
I
I,,
I
i
i
i
P1234123412341234 3 6 9 L.__.._.._._ month . . . . .
i
i
i
12
i
I
3.5
CEE 1.25mg + Placebo i
i
i
i
i
I
i
i
i
i
i
I
P1234123412341234 3 6 9 I . month.
I
I,I
221
I
12
I
I
FIGURE 15.2 Meanpsychologicalsymptomscores (_+SD).CEE, conjugated equine estrogen; MPA, medroxyprogesterone acetate; P, pretreatment. (From ref. 50.)
following 8 weeks of combined estrogen plus citalopram treatment. Although interesting and potentially important, the conclusion that citalopram augmented the antidepressant effect of estrogen in these women is tempered by the fact that the study was open-label and had no control group. There is by now a fair amount of anecdotal and clinical evidence that some considerable proportion of menopausal women on combined or sequential estrogen-progestin replacement regimens complain of dysphoric moods and irritability, which are thought to be causally related to the progestin. Indeed, women who received subcutaneous implants of estradiol and norethisterone (47), percutaneous estradiol plus lynestrenol (48), and ethinyl estradiol plus levonorgestrel (49) all experienced a dampening of mood compared with those treated with the same dose of estrogen alone. In a prospective study, we recently compared the effect of adding 5 mg medroxyprogesterone acetate (MPA) or placebo from days 15 to 25 of the treatment cycle to either 0.625 mg or 1.25 mg CEE taken from days 1 to 25 of the cycle (50). Women who received 0.625 mg CEE and 5 mg MPA had more negative moods (Fig. 15.2) and more psychological symptomatology during treatment compared with those who were taking 1.25 mg CEE and placebo. These findings suggested that the effects ofprogestin on the central nervous system are reflected in an increase in psychological symptomatology, which can be attenuated by a higher estrogen/progestin dose ratio.
IV. ESTROGEN AND COGNITIVE FUNCTIONING There is now a considerable literature on the efficacy of estrogen on cognitive functions in postmenopausal that is addressed comprehensively elsewhere in this book.
In the attempt to assess the impact of changing sex hormone levels on sexuality in postmenopausal women, it is important to consider that human sexual behavior is comprised of identifiably distinct but interrelated processes. Davidson et al. (51) suggested that the components of human sexual behavior could be conceptualized under two major headings. The first category subsumes those behaviors associated with libido or sexual motivation, such as sexual desire, sexual fantasies, and satisfaction or pleasure, whereas potency refers to pelvic vasocongestion, orgasmic contractions, and possible extragenital responses. In humans, sexual dysfunctions are frequently limited to one distinguishable aspect of the behavioral sequence. It is also important to recall that during reproductive life, the ovary produces approximately 25% of total circulating testosterone (T), 60% of androstenedione, and 20% of dehydroepiandrosterone (52). Several investigators have reported that plasma testosterone levels are lower in naturally postmenopausal women compared with younger women whose blood was sampled during the follicular phase of the menstrual cycle (53,54). A more definitive determination of residual ovarian production of steroids after the menopause has been accomplished by measuring the concentration of steroids in blood samples taken from the ovarian artery and the ovarian vein at the time of abdominal surgery. The presence of a concentration gradient across the ovary indicates that the ovary is secreting steroids. In two such studies, it was found that the ovary continues to secrete appreciable amounts of T in about 50% of postmenopausal women (55,56). In a third investigation, the ovarian vein values of testosterone were significantly lower in postmenopausal women than in a control group of young women during the periovulatory phase of the cycle (57). More recently, several longitudinal studies have provided more explicit information on the timing of the decrease in androgen levels with regard to age in women. In the Study of Women's Health across the Nation (SWAN), there was a 25% decrease in T levels between 42 and 50 years of age but no further decline during the menopausal transition (58). Likewise, the Melbourne Women's Midlife Health Project, which followed women through the menopausal transition, also failed to find a decrease in total T levels at that time (59). More recently, in a cross-sectional Australian study of 1423 women ages 18 to 75 years, both total and free T levels declined steeply with age starting in the early reproductive years, but there was no independent effect of menopausal status on T levels (60). However, women age 55 years or older who reported bilateral oophorectomy and were not taking hormones had significantly lower total T and free T levels than age-matched women with intact ovaries. The available evidence therefore suggests that the decline in T levels in women begins in the early reproductive years and continues with increasing age,
222
BARBARAB. SHERWIN
whereas the postmenopausal ovary continues to secrete T in approximately 50% of women, albeit in concentrations that are significantly lower than menstrual cycle values in younger women. Of course, serum T levels decrease significantly within 24 to 48 hours following TAH and BSO (52-54). Because the integrity of the tissues of the female reproductive tract is dependent on estrogen, degenerative changes in these structures ensue when levels of estrogen decrease after the menopause. It is to be expected, therefore, that some of these changes may adversely affect sexual functioning. For example, decreased vaginal lubrication and atropic vaginitis may result in dyspareunia. Moreover, low levels of estrogen may decrease blood flow to the reproductive organs, resulting in diminished pelvic vasocongestion. Indeed, blood flow to the vulva increased by 50%, as measured by Doppler technology, when estrogen was administered to postmenopausal women (61). Thus, the known effects of estrogen on peripheral tissues make it likely that the decrease in circulating estrogen at the time of menopause may cause specific averse effects on sexual functioning such as decreased vaginal lubrication and vasocongestion and dyspareunia. These are functions related to sexual potency as defined by Davidson et al. (51) and are usually reliably relieved by the administration of exogenous estrogen. On the other hand, there is by now compelling evidence that libido, or sexual motivation, in women is dependent on androgens. Early evidence for this notion came from clinical reports of the libido-enhancing effects of testosterone given to women who had breast cancer (62,63). Similarly, a radical decrease in libido was also observed in women who were deprived of endogenous androgen production following their bilateral oophorectomies and adrenalectomies undertaken in the attempt to halt the course of metastatic breast cancer (64,65). During the past decade, androgenic effects on female sexuality have been investigated using more rigorous methodology. Prospective studies of subcutaneous im-
plantation of pellets containing estradiol and testosterone to postmenopausal women have been carried out in Britain and Australia. When women who were preselected for the complaint of loss of libido received implants of both sex steroids, the symptom was reversed in two-thirds after 3 months of treatment (66,67). In another double-blind implant study, postmenopausal women who complained of loss of libido that had not been relieved by oral estrogen reported significant symptomatic relief following treatment with a combined estrogen-testosterone implant (68). In contrast, no change in libido occurred in the groups that had randomly received implants of estrogen alone. Several prospective investigations of the intramuscular administration of an estrogen-androgen preparation to surgicaUy menopausal women provide additional support for the androgenic enhancement of libido in women (69). Patients who received a combined estrogen-androgen drug reported higher levels of sexual desire and arousal and a greater frequency of sexual fantasies compared with those treated with estrogen alone or placebo following their surgery. These findings were subsequently confirmed in surgically menopausal women who had been treated with the estrogen-androgen combination long term compared with matched groups treated long term with estrogen alone and a third group that remained untreated following their TAH and BSO at least 2 years earlier (70) (Fig. 15.3). More recently, three RCTs have reported on the efficacy of a transdermal T patch in addition to oral estrogen for the treatment of diminished sexual desire in surgically menopausal women. In the first such study, 75 healthy women (mean age, 47 years) who had undergone TAH and BSO approximately 5 years earlier were selected because they reported impaired sexual functioning despite receiving at least 0.625 mg CEE daily for at least 2 months before entry into the study (71). Sexual behavior was quantified by means of the 22-item Brief Index for Sexual Functioning for Women
FIGURE 15.3 Mean + standard error of mean (SEM) scoresfor the immediate and delayedparagraph recall tests from the Wechsler Memory Scale at the three test times. LAD, leuprolide acetate depot. (From Sherwin BB, Tulandi T. "Add-back" estrogen reverses cognitive deficits induced by a gonadotropin-releasing hormone agonist in womenwith leiomyomatauteri.J Clin Endocrinol Metab 1996;81:2545-2449.)
CHAPTER 15 Impact of the Changing Hormonal Milieu on Psychological Functioning (BISFW) at screening and again after 12 weeks of treatment with their usual dose of CEE and, in random order, placebo, 150 p~g/dayT, and 300 I,g/day T transdermally for 12 weeks each. Despite a considerable placebo response, the 300 p~g but not the 150 ~g dose o f T resulted in increases in scores for frequency of sexual activity and pleasure-orgasm as measured by the BISFW. Again, at the higher but not at the lower dose of transdermal T the percentages of women who had sexual fantasies, who masturbated, or who engaged in sexual intercourse at least once a week increased two to three times compared with baseline. Positive well-being, depressed mood, and composite scores of the Psychological General Well-Being Index also improved significantly in women when they were receiving the 300 ~g testosterone patch compared with the placebo patch. It is noteworthy that mean bioavailable T levels were 2.0 _+ 1.4 ng/dL at pretreatment (normal female range: 1.6-12.7 ng/dL) and rose to 7.1 _+ 4.1 ng/dL with the 150-p~g patch and to 11.4 + 9.5 ng/dL with the 300 b~g testosterone patch. Therefore, the enhancement of sexual behavior reported by women who received treatment with the 300 btg testosterone patch was achieved when the serum levels ofbioavailable T it induced were at the upper limit of the normal female range. Two recent large, multicenter, randomized, controlled trials provided further support for these findings using the same transdermal T patch. In a double-blind, placebocontrolled trial, healthy women who had developed low sexual desire following TAH and BSO despite treatment with oral estrogen therapy randomly received either a placebo, a 150 lag/day, a 300 lag/day, or a 450 lag/day T patch for 24 weeks in addition to oral estrogen (72). Women who received the 300 lag/day testosterone patch but not those who received the 150 or 450 lag patches experienced greater increases in sexual desire and in frequency of satisfying sexual activity that those treated with placebo. In the second study, 533 estrogen-treated surgically menopausal women with hypoactive sexual desire disorder randomly received treatment with either a placebo or a 300 tag/day testosterone patch for 24 weeks (73). Once again, the 300 lag testosterone patch significantly increased satisfying sexual activity and sexual desire while decreasing personal distress. Taken together, these studies on the efficacy of the combined use of the T transdermal patch and oral estrogen therapy provide strong support for the conclusion that 300 lag/day of T administered transdermally increases sexual desire and the incidence of satisfying sexual encounters in surgically menopausal women. A recent study investigated the efficacy of a transdermal testosterone cream in premenopausal women complaining of low libido. Thirty-four women (mean age, 39 years) who had normal menstrual cycles and were complaining of low libido were randomized to receive either T cream (10 mg) or placebo cream daily for two double-blind, 12-week treatment periods separated by a single-blind, 4-week washout
223
period (74). Scores on the Sabbattesberg Sexual Self-Rating Scale increased significantly following treatment with the T cream but not with placebo cream. Total T levels were within the lower third of the normal female range before treatment and the Free Androgen Index (FAI) increased to values above the upper limit of the normal female range after treatment with the T cream. However, no increased incidence of acne or hirsutism appeared to result from these high T levels for the 12-week duration of the treatment phase. These findings suggest that T cream may be a safe and efficacious treatment for complaints of low libido in premenopausal women, although more information from long-term trials is needed on this drug. Taken together, the findings from both the subcutaneous implant pallet studies, the prospective investigations that used intramuscular hormonal preparations, and the RCTs of the transdermal T patches allow the conclusion that the addition ofT to an estrogen replacement regimen is associated with an enhancement of sexual desire, interest, and enjoyment of sex in postmenopausal women. These findings also support the contention that in women, as in men (75), testosterone has its major impact on the motivational or libidinal aspects of sexual behavior, such as desire and fantasies, and not on peripheral physiologic responses. Studies on nonhuman primates support the conclusion that testosterone exerts this effect on sexual desire via mechanisms that impact directly on the brain rather than by an effect on peripheral tissues (76).
VI. SUMMARYAND CLINICAL IMPLICATIONS Although the intent of this chapter was to elucidate possible psychological consequences of endocrine changes at the menopause, it is important to mention that nonbiologic factors also influence symptomatology at this time. For example, cross-sectional studies have found that women of lower socioeconomic status are more likely to have psychological symptoms around the time of menopause (77-78). More over, negative attitudes toward the menopause, poor social support, poor marital relations, stressful life events, and recent bereavements also correlate positively with degree of psychological symptoms (79). It is therefore extremely important that the health care provider make a reasonable effort to determine whether the psychological symptoms presented by a given patient can be more rightfully ascribed wholly or in part to the hormonal changes she is experiencing rather than to concurrent stresses in her personal life, for which hormone replacement therapy will be largely ineffective. Like for all other patients whose psychological distress is substantially influenced by adverse personal and social factors, menopausal women with such a history should he referred for psychiatric or psychological
224 evaluation. On the other hand, the research findings reviewed here provide considerable evidence that the decrease in the production of the sex steroids at the time of menopause can alter brain structure and function in ways that can precipitate psychological symptoms at this time. For example, estrogen influences neurotransmitter activity in several ways to affect mood (19-24). Although the specific mechanisms of estrogenic action on neurochemical and neurophysiologic function have not been clearly elucidated, there is by now sufficient empirical grounds to conclude that estrogen enhances mood, whereas its relative absence dampens mood in many women. That having been said, it is important to recall that the administration of estrogen in doses that are conventionally used to treat postmenopausal women will likely alleviate "minor psychiatric symptoms" such as irritability, short-lasting mood swings, crying spells, and feelings of sadness that typically appear during the perimenopause. However, there is insufficient evidence to support the idea that physiologic doses of estrogen are an effective first-line treatment for episodes of major depressive disorders in perimenopausal and postmenopausal women. As during all other life stages, major affective disorders in postmenopausal women usually need to he treated with psychotropic medications and/or psychotherapy. Interestingly, the evidence that estrogen potentiates the effect of some antidepressants (80) raises the possibility that estrogen-treated postmenopausal women may respond to lower doses of antidepressant drugs. A corollary of this observation is that postmenopausal women whose depression is refractory to antidepressant treatment may benefit by the addition of estrogen to their psychopharmacologic regimen, although there is currently no evidence from controlled trials to support the suggestion. Finally, with respect to disturbances in sexual functioning that are frequently reported by menopausal women, the simplest and most direct approach is to reverse, by means of estrogen administration, the genital tissue and vascular changes that inevitably ensue under conditions of estrogen deprivation. Estrogen replacement therapy ought to restore the integrity of the vaginal tissues and, concomitantly, alleviate symptoms such as decreased vaginal lubrication and dyspareunia. However, there is now compelling evidence that T and not estrogen is primarily responsible for the maintenance of sexual interest and desire in women. When it can he determined that the onset of this complaint was associated in time with the perimenopausal phase in a naturally menopausal woman or with the postoperative period in a surgically menopausal woman, treatment with a combined estrogen-testosterone preparation is highly effective in restoring sexual desire. Indeed, the recent position statement of the North American Menopause Society recommends that postmenopausal women with decreased sexual desire associated with personal distress and with no other identifiable cause may be candidates for T therapy (81). However,
BARBARAB. SHERWIN
when lack of sexual desire has been lifelong or when this symptom considerably predated the menopause, the expectation that combined estrogen-testosterone therapy itself will reverse the symptom is doubtful and sexual counseling should also be sought. Historically, both gynecologists and menopausal women have attributed an astonishing array of psychological symptoms to the endocrine changes that characterize the menopause. In large part, this was due to the faculty logic that the occurrence of two events in temporal contiguity implies a cause-and-effect relationship. Two things have happened during the past few decades that allow for more refined and precise formulations concerning the relationships between changing gonadal hormone levels and psychological symptoms. First, recent advances in neuroscience have described mechanisms whereby the sex hormones influence brain morphology and function that serve to explain hormonal influences on specific symptoms such as mood. Second, increased methodologic rigor applied to clinical studies of hormone-behavior relationships has enhanced the reliability of these research findings and, consequently, has served to clarify possible associations between specific hormones and specific symptoms. New theoretical knowledge that bears on the manner in which sex hormones affect brain function is being used, as this chapter hopefully points out, to provide empirically based clinical guidelines for the evaluation and treatment of specific psychological symptoms that may be associated with the menopause. Finally, it should be acknowledged that the development of psychological symptoms at the time of menopause is neither universal nor inevitable. The fact that several epidemiologic studies of nonclinical populations failed to find any increase in psychological symptoms at the time of menopause attests to the fact that some women do not experience any symptoms at all, or perhaps only to a trivial extent. On the other hand, there is overwhelming evidence that the majority of women who seek medical help around the time of menopause from family practitioners, gynecologists, or menopause clinics have prominent psychological symptomatology. It is these substantial numbers of women who stand to benefit from our increasing knowledge of the psychotropic properties of the sex hormones.
References 1. NeugartenBL, Kraines RJ. Menopausal symptoms in women of various ages.Psycbosom Med 1965;27:266-273. 2. McKinlay SM, Brambilla DJ, Pasner JG. The normal menopausal transition. Mauritas 1992;14:103-115. 3. H/illstr6mT, Samuelsson S. Mental health in the climacteric:the longitudinal study of women in Gothenburg. Acta Obstet Genecol Scand 1985;130:13-18. 4. McKinlay SM, Jeffreys M. The menopausal syndrome. Br J Prey Soc Med 1974;28:108-115.
CHAPTER 15 Impact of the Changing Hormonal Milieu on Psychological Functioning 5. Bungay GT, Vessey ME McPherson CK. Study of symptoms in middlelife with special reference to menopause. Br MedJ 1980;2:181-183. 6. Greene JG, Cooke DJ. Life stresses and symptoms at the climacteric. BrJ Psychiatry 1980;136:486-491. 7. Ballinger CB. Psychiatric morbidity and the menopause: screening of a general population sample. Br MedJ 1975;3:344-346. 8. Hunter M, Battersby R, Whitehead M. Relationships between psychological symptoms, somatic complaints, and menopausal status. Maturitas 1986;8:217-228. 9. McKinlay JB, McKinlay SM, Brambilla D. The relative contribution of endocrine changes and social circumstances to depression in mid-aged women. J Health Soc Behav 1987;28:345-363. 10. Anderson E, Hamburger S, Liu JG, Rebar RW. Characteristics of menopausal women seeking assistance.Am J Obstet Gyneco11987;156:428-433. 11. Montgomery JC, Brincat M, Tapp A, et al. Effect of oestrogen and testosterone implants on psychological disorders in the climacteric. Lancet 1987;(Feb 7):297-299. 12. Kinsey AC, Pomeroy WB, Martin CE, Gebhard PH. Sexual behavior in the human female. Philadelphia: WB Saunders, 1953. 13. Hiillstr6m T. Sexuality in the climacteric. Clin Obstet Gynecol 1977;4:227-239. 14. H~illstr6m T, Samuelsson S. Changes in women's sexual desire in middle life; the longitudinal study of women in Gothenburg. Arch Sex Behav 1990;19:259-268. 15. McCoy NL, Davidson JM. A longitudinal study of the effects of menopause on sexuality. Maturitas 1985;7:203-210. 16. Sarrel PM, Whitehead MI. Sex and menopause: defining the issues. Maturitas 1985;7:217-224. 17. Kelly MJ, Moss RL, Dudley CA, Fawcett CE The specificity of the response of preoptic-septal area neurons to estrogen: 17-13 estradiol vs 17-oL estradiol and the response of extrahypothalamic neuron. Exp Brain Res 1977:30:43-52. 18. O'Mally BW, Means AR. Female steroid hormones and target cell nuclei. Science 1974;183:610-620. 19. McEwen BS. The brain as a target organ of endocrine hormones. In: Kreiger DT, Hughes JS, eds. Neuroendocrinology. Sunderland, MA: Sinauer Assoc, 1980;33-42. 20. Matsumoto A, Arai Y, Osanai M. Estrogen stimulates neuronal plasticity in the deafferenated hypothalamic nucleus in aged female rats. Neurosci Res 1985;2:412-418. 21. Miyakawa M, Arai Y. Synaptic plasticity to estrogen in the lateral septurn of the adult male and female rats. Brain Res 1987;436:184-188. 22. Matsumoto A, Micevych PR, Arnold AR Androgen regulates synaptic input to motoneurons of the adult rat spinal cord. J Neurosci 1988;8: 4168-4176. 23. Luine VN, McEwen BS. Effect of estradiol on turnover of type A monoamine oxidase in brain. J Neurochem 1977;28:1221-1227. 24. Luine VN, Khylchevskaya RI, McEwen BS. Effect of gonadal steroids on activities of monoamine oxidase and cholineacetylase in rat brain. Brain Res 1975;86:293-306. 25. Bartus RT, Dean RL, Beer B, Lippa AS. The cholinergic hypothesis of memory dysfunction. Science 1982;217:408-417. 26. Wooley CS, Gould E, Frankfurt M, McEwen BS. Naturally occurring fluctuation in dendritic spine density on adult hippocampal pyramidal neurons. J Neurosci 1990; 10:4035 - 4039. 27. Brinton RD. Cellular and molecular mechanisms of estrogen regulation of memory function and neuroprotection against Alzheimer's disease. Learn Mem 2001;8:121-133. 28. Behl C. Estrogen can protect neurons: modes of action. J SteroidBiochem Mol Bid 2002;83:195-197. 29. Kugler J, Seus R, Krauskopf R, Brecht HM, Raschig A. Differences in psychic performance with guanfacine and clonidine in normotensive subjects. BrJ Clin Pharmaco11980;10:795-805.
225
30. Shaywitz SE, Shaywitz BA, Pugh KR. Effect of estrogen on brain activation patterns in postmenopausal women during working memory tasks. JAM_d 1999;281:1197-1202. 31. Maki PM, Resnick SM. Longitudinal effects of estrogen replacement therapy on PET cerebral blood flow and cognition. Neurobiol Aging 2000;21:373-383. 32. Holzbauer M. Physiological aspects of steroids with anesthetic properties. Med Bio11976;54:227-242. 33. Merryman W, Boidman R, Barnes L, Rothchild I. Progesterone "anesthesia" in human subjects. J Clin 1954;14:1567-1569. 34. Holzbauer M, Yondin MBH. The oestrous cycle and monoamine oxidase activity. BrJ Pharmaco11973;48:600-608. 35. George GCW, Bearemont PVJ, Beardwood DJ. Effects of exogenous estrogen on minor psychiatric symptoms in postmenopausal women. S Afr MedJ 1973;47:2337-2344. 36. Chakravarti S, Collens WP, Newton JR. Endocrine changes and symptomatology after oophorectomy in premenopausal women. BrJ Obstet Gynaeco11977;74:769-775. 37. Stuckler RC, Booth R, Cecutti A. The role or estrogen replacement in the climareric syndrome. PsycholMed 1977;7:631-639. 38. Sherwin BB, Gelfand MM. Sex steroid and affect in the surgical menopause; a double-blind cross-over study. Psychoneuroendocrinology 1985;10:325-335. 39. Sherwin BB. Affective changes with estrogen and androgen in replacement therapy in surgically menopausal women.JAffect Disord 1988;14: 177-187. 40. Schmidt PJ, Nieman L, Danaceau MA, et al. Estrogen replacement in perimenopause related depression; a preliminary report. Am J Obstet Gyneco12000;183:414-420. 41. Soares CN, Almeida OP, Joffe H, et al. Efficacy of estradiol for the treatment of depressive disorders in perimenopausal women. Arch Gen Psychiatry 2001;58:529-534. 42. Morrison MF, Kallan MJ, Ten Have T, et al. Lack of efficacy of estradiol for depression in postmenopausal women: a randomized, controlled trial. Biol Psychiatry 2004;55:406-412. 43. Cohen LS, Soares CN, Poitras JR, et al. Short-term use of estradiol for depression in perimenopausal and postmenopausal women. Am J Psychiatry 2003;160:1519-1522. 44. Amsterdam J, Garcia-Espana F, Fawcett J, et al. Fluoxetine efficacy in menopausal women with and without estrogen replacement. J Affect Disord 1999;55:11 - 17. 45. Rasgon NL, Altshuler LL, Fairbanks LA, et al. Estrogen replacement therapy in the treatment of major depressive disorder in perimenopausal women. J Clin Psychiatry 2002;63(suppl) 7):45-48. 46. Soares CN, Poitras JR, Prouty J, et al. Efficacy of citalopram as monotherapy or as an adjunctive treatment to estrogen therapy for perimenopausal and postmenopausal women with depression and vasomotor symptoms. J Clin Psychiatry 2003;64:473-479. 47. Schneider MA, Brotherton PL, Hailes J. The effect of exogenous oestrogens on depression in menopausal women. MedJAustr 1977;2:162-170. 48. Holst J, Backstr6m T, Hammerbiich S, yon Schoultz B. Progestogen addition during oestrogen replacement therapy--effects on vasomotor symptoms and mood. Maturitas 1989;11:13-19. 49. Dennerstein L, Burrows GD, Hyman G, Sharpe K. Hormone therapy and affect. Maturitas 1979;1:247-254. 50. Sherwin BB. The impact of different doses of estrogen and progestin on mood and sexual behavior in postmenopausal women. J Clin Endocrinol Metab 1991;72:336-343. 51. Davidson JM, Gray GD, Smith ER. The sexual psychoendocrinology of aging. In: Meites J, ed. Neuroendocrinology of aging. New York: Plenum Press; 1983;221-258. 52. Longcope C. Adrenal and gonadal steroid secretion in normal females. J Clin Endocrinol Metab 1986;15:213-228.
226 53. Longcope C. Steroid production in pre- and postmenopausal women. In: Greenblatt R, Mahesh V, McDonough E, eds. The menopausal syndrome. New York: Medcom Press; 1974:6-11. 54. Vermeulen A. The hormonal activity of the postmenopausal ovary. J Clin EndocrinolMetab 1976;42:247-253. 55. Longcope C, Hunter R, Franz C. Steroid secretion by the postmenopausal ovary. Am J Obstet Gyneco11980;138:564-570. 56. Lucisano A, Acampora MG, Russo N, et al. Ovarian and peripheral plasma levels of progestogens, androgens and oestrogens in postmenopausal women. Maturitas 1984;6:45-53. 57. Botella-Llusia J, Oriol-Bosch A, Sanchez-Garrido E Testosterone and 17-[3-oestradiol secretion of the human ovary. Maturitas 1979; 2:7-12. 58. Randolph JF, Sowers M, Gold EB, et al. Reproductive hormones in the early menopausal transition. J Clin Endocrinol Metab 2003;88: 1516-1522. 59. Burger HG, Dudley EC, Cui J, Dennerstein L, Hooper J. A prospective longitudinal study of serum testosterone, dehydroepiandrosterone sulphate, and sex hormone-binding globulin levels through the menopause transition. J Clin EndocrinolMetab 2000;85:2832-2838. 60. Davidson SL, Bell R, Donath S, Montalto JG, Davis SR. Androgen levels in adult females: changes with age, menopause and oophorectomy. J Clin Endocrinol Metab 2005;90:3847-3853. 61. Sarrel PM. Progestogens and blood flow. Int ProcJ 1989;1:266-271. 62. Foss GL. The influence of androgens on sexuality in women. Vitam Horm 1947;5:317-391. 63. Kennedy BJ. Effects of massive doses of sex hormones on libido, ivied Asp Hum Sex 1973;7:67-75. 64. Drellich MG, Waxenberg SE. Erotic and affectional components of female sexuality. In: Masserman J, ed. Science and psychoanalysis. New York: Grune & Stratton; 1966;192-217. 65. Waxenberg SE, Drellich MG, Sutherland AM. Changes in female sexuality after adrenalectomy. J Clin Endocrino11959;19:193-202. 66. Burger HG, Hailes J, Menlaus M, et al. The management of persistent menopausal symptoms with oestradiol-testosterone implants; clinical, lipid and hormonal results. Maturitas 1984;6:351-358. 67. Cardozo L, Gibb DMF, Tuck SM, et al. The effects of subcutaneous hormone implants during the climacteric. Maturitas 1984;5:177-184. 68. Burger H, Hailes J, Nelson J, Menelaus M. Effect of combined implants of oestradiol and testosterone on libido in postmenopausal women. Lancet 1987;294:936-937.
BARBARAB. SHERWIN 69. Sherwin BB, Gelfand MM, Brender W. Androgen enhances sexual motivation in females; a prospective cross-over study of sex steroid administration in the surgical menopause. PsychosomMed 1985;7:339-351. 70. Sherwin BB, Gelfand MM. The role of androgen in the maintenance of sexual functioning in oophorectomized women. Psychosom Med 1987;49:397-409. 71. Shifren JL, Braunstein GD, Simon JA. Transdermal testosterone treatment in women with impaired sexual function after oophorectomy. N EnglJ Med 2000;343:682-688. 72. Braunstein GD, Sundwall DA, Katz M. Safety and efficacy of a testosterone patch for the treatment of hypoactive sexual desire disorder in surgically menopausal women: a randomized placebo-controlled trial. Arch Int Med 2005;165:1582-1589. 73. Buster JF, Kingsberg SA, Aguirre O. Testosterone patch for low sexual desire in surgically menopausal women: a randomized trial. Obstet Gyneco12005;105:944- 952. 74. Goldstat R, Esther B, Tran J. Transdermal testosterone therapy improves well-being, mood and sexual function in premenopausal women. Menopause 2003;10:390- 398. 75. Bancroft J, Wu FCW. Changes in erectile responsiveness during androgen replacement therapy. Arch Sex Behav 1983;12:59-66. 76. Everitt BJ, Herbert J. The effects of implanting testosterone propionate in the central nervous system on the sexual behavior of female rhesus monkeys. Brain Res 1975;86:109-120. 77. Polit D, Larocco S. Social and psychological correlates of menopausal symptoms. PsychosomMed 1980;42:335-345. 78. Schneider M, Brotherton P. Physiological, psychological correlates of situational stresses in depression during the climacteric. Maturitas 1979;1:153-158. 79. Greene JG. Bereavement and social support at the climacteric. Maturitas 1983;5:115-124. 80. Price WA, Giannini AJ. Antidepressant effects of estrogen. J Clin Psychiatry 1985;46:506-510. 81. The role of testosterone therapy in postmenopausal women: position statement of the North American Menopause Society. Menopause 2005;12:497-511.
]HAPTER 1(
Connective Tissue Changes in the Menopause and with Hormone Replacement Therapy MARK B R I N C A T
Department of Obstetrics and Gynecology, St. Luke's Hospital Medical School, University of Malta, Msida MSD 06, Gwardamangia, Malta
RAY GALEA
Department of Obstetrics and Gynecology, St. Luke's Hospital Medical School, University of Malta, Msida MSD 06, Gwardamangia, Malta
BARON
Department of Obstetrics and Gynecology, St. Luke's Hospital Medical School, University of Malta, Msida MSD 06, Gwardamangia, Malta
YVES MUSCAT
I. I N T R O D U C T I O N
body. Both show deterioration with estrogen deficiency. This deterioration can be prevented and even reversed with appropriate and adequate estrogen replacement (2-5).
Collagen is the major protein component of most connective tissues of vertebrates. In mammals it constitutes about 25% to 35% of their total protein. It is present in virtually every animal tissue, providing an extracellular framework for all metazoan animals. It is a highly conserved, evolutionarily ancient family of molecules. Type I collagen, which constitutes 90% of the total collagen in the body, is the most important (1). It is predominant in skin and bone. In addition, skin contains an amount of the very similar type III collagen. Bone and skin are the two organs that together contain some 80% of all the connective tissues found in the human TREATMENT OF THE POSTMENOPAUSAL WOMAN
II. B O N E The collagens in bone are mainly type I collagen, with type V making up a minor component. The noncollagen proteins include plasma proteins, proteoglycans, osteocalcin, osteopontin, osteonectin, and bone sialoprotein. Collagen makes up the greatest proportion, as it constitutes approximately 35% of dry defatted bone mass. The organic matrix of bone acts rather like internal girders and confers on bone 227
Copyright 9 2007by Elsevier,Inc. All rights of reproductionin any formreserved.
228 its tensile strength. It has been suggested (2) that a decline in this organic matrix is the primary pathologic event leading to osteoporosis (6,7). Bone loss is age related in both sexes, but in women there is an acceleration in the rate of bone loss following the menopause. It has been shown (8) that by the age of 70 years women lose up to 50% of their bone mass, whereas a man would be expected to lose only 25% by the age of 90. A genetic component has been implicated in the prevalence of osteoporotic fractures. An epidemiologic study performed by our group has shown significant differences of hip and Colles' fracture incidences across various countries and populations. Muscat Baron et al. showed that the incidence of radial hip fractures in Mediterranean countries was one-half to one-third that of Northern European populations (9). This epidemiologic pattern may be due to the genetic variation of collagen type and content. Bone mineral density, a major determinant of osteoporotic fracture risk, has a strong genetic component, and several candidate gene polymorphisms involving collagen and estrogen have been implicated in the regulation of this process. Associations between the bone mineral density and polymorphisms in the collagen type I alpha I gene (COLIA1), estrogen receptor (ER) alpha, and the resultant bone mineral density have been shown. The COLIA1 gene has been found to be associated with the prevalence of osteoporotic fractures (10). Genetic variations explain as much as 70% of the variance of bone mineral density in the population. Several quantitative trait loci for bone mineral density have been identified by genome screening in humans and more than 100 candidate gene polymorphisms tested for association with bone mineral density and fracture. Variants in the vitamin D receptor, COLIA1, ER alpha (ER oO, interleukin 6 (11_,6),and LDL receptor-related protein 5 (LRPS) genes were all found to be significantly associated with differences in bone mineral density or fracture risk in multiple replication studies (11). The genetic influence on collagen content may not only limit itself to the onset of osteoporosis but may also be related to the development of osteoarthrosis, another common debilitating condition associated with the aging process. Linkage and linkage disequilibrium of several candidate genes, such as ER alpha and COL1/11, have been shown to be related to extracellular inorganic pyrophosphate transport and osteoporotic and osteoarthritic phenotypes (12,13).
BRINCAT ET AL.
neck, and at Ward's triangle), and skin thickness was measured using high-frequency (22.5 MHz) ultrasound. Skin thickness correlated with the collagen content in the skin. These measurements were carried out in a group of untreated postmenopausal women and a group of untreated postmenopausal women who had sustained an osteoporotic fracture. The results showed that the skin thickness and bone density parameters were much lower in women who had sustained osteoporotic fractures compared with control subjects (Fig. 16.1). Women with fractures had mean bone mass values that were some 20% below the mean values of controls. Skin thickness varied within a narrower range (mean difference 4%), but significant differences between controls and women who had sustained an osteoporotic fracture were also consistently noted. When the skin thickness measurements were combined with bone density parameters, the accuracy of predicting an osteoporotic fracture increased, as shown in Fig. 16.2 (14) (see color insert). The effect of estrogen on osteoblasts (bone-forming cells) may be direct, or it may rely on an alteration of the coupling phenomenon of bone formation and bone resorption (15). Estrogen replacement affects osteoporosis in two ways: It can prevent it, or it can correct it once it has occurred. It has been postulated that estrogen replacement therapy leads to an increase in bone mass by increasing the collagen levels in bone. Indeed, the observation (4) that women on estrogen replacement therapy had a higher dermal skin collagen content than those who were not on estrogen replacement therapy had led to the idea that a similar increase in bone collagen content in women on estrogen replacement therapy occurred. Possibly, this increase
III. B O N E A N D SKIN Our group investigated the relationship between age, bone mass, and skin thickness in a biophysical study of postmenopausal women that looked at differences in various bone density measurements and in skin thickness. Bone was measured using a dual-energy x-ray absorptiometer (between the second and fourth lumbar vertebrae, at the femur
Osteoporotic Fractures vs Controls p<0.001
FIGURE 16.1 Mean skin thickness in 411 untreated postmenopausal women as control subjects and 119 untreated postmenopausalwomenwho had suffered an osteoporoticbone fracture.
CHAPTER 16 Connective Tissue Changes in the Menopause and with Hormone Replacement Therapy
229
FIGURE 16.2 Bone density at L 2 - L 4 plotted against skin thickness in 1500 untreated postmenopausal women as control subjects (dots) and 250 untreated postmenopausal women who had suffered an osteoporotic bone fracture (squares). The accuracy of combining skin thickness measurements with bone density parameters in predicting bone fractures is shown.
in collagen content leads to an increase in bone strength and thus a diminution of osteoporotic fractures. When estrogen treatment is started in the menopausal period and continued into later life, the best result for bone density is obtained. However, estrogen therapy initiated later in life also provides benefits (16).
IV. COLLAGEN MARKERS In order to look at the collagen changes occurring with estrogen therapy, our group undertook another study (17). This was a biophysical/biochemical study involving a group of untreated postmenopausal women and a group of women who were on estrogen replacement therapy. In addition to having the same bone density and skin thickness measurements as in the previous study (mentioned earlier), the women in this study also had collagen marker estimations. In this study we looked at the excretion of procollagen IC end-terminal peptides (PCICP), a marker for bone formation, and pyridinium cross-links, a marker for bone resorption. When the untreated postmenopausal control subjects were compared with women who had been on hormone replacement therapy, the latter showed a drastic increase in osteoclastic activity (pyridinium cross-link excretion). The mean decrease was 27.2%. There was also a decrease in bone formation, osteoblastic activity (PCICP) being decreased by a mean of 11.3%. All these differences (p < 0.001) indicated that bone remodeling had readjusted in patients on estrogen replacement therapy. The women in this study who were taking hormone replacement therapy had only been on this treatment for a short time (mean 6 months). This may imply that changes occur rapidly. It is interesting to note that, according to the above calculations based on bone collagen markers, a mean positive change of 15.9% occurred in bone remodeling in women on estrogen replacement therapy.
This value is of the same order as the mean percentage change in bone mass of the same women, measured using the biophysical method. This showed a mean positive change at the L 2 - L 4 region of 16.1% (5,17) (Fig. 16.3). This could indicate that these markers are indeed sensitive and can be used for monitoring postmenopausal bone loss as well as accurately titrating the response to treatment, both when such treatment is prophylactic and when it is used therapeutically in the management of established osteoporosis. The excretion of pyridinium cross-links of collagen in the first-morning-void urine sample showed some intraindividual variation (18), and this may be a limitation of its usefulness. Some authors (19) claim that the intact amino terminal peptide of type I procollagen, a protein set free from the other end on the same procollagen molecule, is a more dynamic marker for bone metabolism than PCICP (the carboxyl-terminal propeptide of type I procollagen). These authors therefore recommend that it be used as a marker reflecting the effect of estrogen on bone collagen formation during estrogen replacement therapy. Increasing evidence is now available (20) that collagen markers can become a useful tool in predicting osteoporosis and fracture risks, in addition to bone mineral density measurements. Early changes in collagen markers can predict long-term changes in bone density with therapy. Serum levels of type I carboxyl-terminal pyridinoline cross-linked telopeptide may be useful in detecting changes in bone resorption after oophorectomy and suggest that women are at greater risk for bone resorption after oophorectomy than after physiologic menopause (21). Short-term changes in biochemical markers of bone turnover at 6 months appear to predict bone density changes at the spine and hip at 3 years in elderly women on alendronate, hormone replacement therapy, or combination therapy. The greatest decrease in urinary N-telopeptide showed a 10.1% increase in lumbar bone density and a 6.1% increase in hip
230
BRINCAT ET AL.
FIGURE 16.3 A: Mean procollagen value in 92 untreated postmenopausal women and 68 postmenopausal women on hormone replacement therapy. An 11.3% difference in bone formation was seen in treated women compared with untreated women. B: Mean value of pyridinium crosslinks in 129 untreated and 76 treated women. A 27.2% difference in bone resorption was seen in treated women compared with untreated women. C: Mean bone density measured at the L 2 - L 4 vertebrae in 140 untreated women and 83 treated women. A 16.1% difference in bone density was seen in treated women compared with untreated women.
bone density (22). Women with bone loss receiving hormone replacement therapy exhibit significantly lower C-telopeptide, N-telopeptide levels than those receiving no therapy (23). Selective estrogen receptor modulators (SERMs) such as raloxifene reduce bone turnover, as evidenced from biochemical markers, and increase bone density, although to a lesser extent than conjugated oestrogens (24). Additional information is needed to define specific situations and cut points to allow marker results to be used with confidence in making therapeutic decisions about indMdual patients (25).
V. SKIN The skin is one of the largest organs in the body, containing a sizeable proportion of total body collagen. This organ is composed of a population of cells of diverse embryonic
origin that, under normal conditions, exist side by side in complete harmony, forming a complex mosaic. Atrophy of the dermis after the menopause is due to a decrease in the dermal skin collagen content. The amounts of hydroxyproline and glycosylated hydroxylysine (26,27)in type I collagen and of immature and reducible cross-links decrease with age (28). To what extent these changes are fundamental to the aging process is still unknown. This decrease is not only arrested but reversed by hormone replacement therapy (29). There is now strong evidence, as seen from the National Health and Nutrition Examination Survey (30), that estrogen use prevents dry skin and skin wrinkling, thus extending the potential benefits of postmenopausal estrogen therapy to include protection against selected age- and menopause-associated dermatologic conditions. Estrogens have in fact been shown to enhance the dermal content of water, glycosaminoglycans (GAGs), and
CHAPTER 16 Connective Tissue Changes in the Menopause and with Hormone Replacement Therapy collagen. Both estrogen and androgen receptors have been identified on dermal fibroblasts (31). Estrogen could increase the rate of collagen production (32) by altering the degree of polymerization of GAGs. Estrogens increase hydroscopic qualities (33), probably through the enhanced synthesis of dermal hyaluronic acid (34). Collagenous fibrils were found to be less fragmented in the dermis of women treated with estrogens (35). Studies have been carried out on skin changes in various connective-tissue and endocrine disorders. Black (36) and Shuster and colleagues (37) looked at the relationship between skin thickness and skin collagen in women with systemic sclerosis, osteoporosis of mixed etiology, and hirsutism and found a good correlation between the two parameters. In a small study, Black (36) demonstrated changes in the collagen content and thickness of the skin in women with osteoporosis (mixed etiology) treated with androgens, when compared with women with osteoporosis who had not been on this treatment. Shuster and coworkers (37) found an increase in skin collagen in women with hirsutism, but the increase was not statistically significant. Reports on the skin collagen changes with age are conflicting. Shuster and Bottoms (38,39) showed that the best way of expressing skin collagen content was by measuring the collagen content of a skin biopsy of 1 mm 2 of skin surface. This method takes into account the possibility of changes in the total mass of the dermis. They reported a reduction in total skin collagen with age, but this was not confirmed by Reed and Hall (40). Shuster and Black found that the amount of collagen present in skin was at all ages higher in males than in females. Hall and associates (41) confirmed that skin collagen content was higher in males than in females but once again could not confirm a significant decline of skin collagen with age. The skin tends to get thinner with increasing age past the menopause, and this is reversed with adequate estrogen replacement therapy (6), as first noted by Albright and coworkers (42) in 1940. They reported that old women with osteoporotic fractures had a higher incidence of thin skin. McConkey and colleagues (43) also showed that osteoporosis was more common in women with transparent skin than in women with thick, opaque skin. The dermal skin itself, composed as it is of connective tissue, including the predominant protein collagen as well as elastin (small amounts) and GAGs, also has more than one variable that is affected by estrogen lack or its replacement. In work in rats, for example (34), castrated rats that were given estrogens had a 70% increase in their GAG content in 2 weeks. Similar increases in women would lead to skin thickness increases that far outstrip what would be expected from collagen content increases alone. Most of the studies have measured the thickness of the dermis, because this is the layer that is mainly composed of connective tissue. Subcutaneous tissue is an added variable.
231
Studies using Harpenden's calipers, which also include subcutaneous fat, have been conflicting (44,45). Ultrasound can be used to measure skin thickness. Ultrasound examination has been used by dermatologists when assessing dermal malignancies. Good reproducibility has been obtained by using high-frequency (22.5 MHz) ultrasound to determine the thickness of the skin. Excluding the subcutaneous tissues, various studies have now shown that skin thickness increases with estrogen replacement therapy. Skin changes and bone changes with estrogen are linked (6,7). In animals, estrogens appear to alter the vascularization of the skin (46), and a change in the connective tissue in the dermis occurs, as is reflected by increased mucopolysaccharide incorporation, hydroxyproline turnover, and alterations in ground substance. In addition to the increased dermal turnover of hyaluronic acid, the dermal water content is enhanced with estradiol therapy (24). Rauramo and Punnonen (47) observed that oral estrogen therapy in castrated women caused thickening of the epidermis for 3 months and that this persisted for 6 months. Punnonen (48), using two different strengths of estrogen in castrated women, showed a statistically significant thickening of the epidermis after 3 months with both strengths, but this thickness persisted only with the lower dose. A third of patients on the higher dose started showing significant thinning of the epidermis, possibly because the treatment was too strong, therefore exerting a corticosteroid effect. Other authors were also able to show beneficial skin changes, using both topical estradiol (49) and estradiol implants (50). Topical estradiol gel has also been shown to increase skin collagen content, as measured by skin hydroxyproline content. Skin blister fluids were assayed, and an increase of both PICP and PINP characterized the gel. In comparing a group of patients who had been treated with estradiol and testosterone implants from 2 to 10 years with a group of untreated postmenopausal women, it was shown that the treated group had highly significantly greater skin collagen content than the untreated group. An optimal skin collagen level was obtained after 2 years of treatment with an optimal estrogen regimen. Levels that are too high or too low give lower levels of collagen (51). The same conclusions were also reported in relation to the epidermis (52). The decline in skin collagen content after the menopause occurs at a much more rapid rate in the initial postmenopausal years than in the later years. Some 30% of skin collagen is lost in the first 5 years after the menopause (51), with an average decline of 2.1% per postmenopausal year over a period of 20 years. The increase in skin collagen content after 6 months of sex hormone therapy depends on the collagen content at the start of treatment (41). In women with a low skin collagen content, estrogens are initially of therapeutic and later of prophylactic value, whereas in those with mild loss of collagen content in the early menopausal years
232 estrogens are of prophylactic value only. Therefore a deficiency in skin collagen can be corrected but not overcorrected. Skin collagen content has been shown to have a strong correlation with dermal skin thickness, as measured radiologically (51). With the use of 100-mg subcutaneous estradiol implants, significant increases in skin thickness and metacarpal index occurred over a 1-year period. Most of the increase occurred in the first 6 months. Brincat and Castelo Branco and their colleagues (49,51,53,54) have shown that, following the menopause, skin collagen content and skin thickness are higher in women on estrogen replacement therapy compared with age-matched women receiving no treatment. Prospective studies have shown that skin thickness, skin collagen, and bone mass increase in postmenopausal women who start estrogen replacement therapy. Estrogen has been shown to reduce collagen turnover and improve collagen quality (53,55). When postmenopausal women were treated with topical estrogens (56), it was found that, after treatment for 6 months, the elasticity and firmness of the skin had markedly improved and the wrinkle depth and pore sizes had decreased by 61% to 100%. Furthermore, the skin moisture was also increased and the measurement of wrinkles, using skin profilometry, revealed significant, or even highly significant, decreases of wrinkles. On immunohistochemistry of the same subjects, significant increases of type III collagen labeling were combined with increased numbers of collagen fibers at the end of the treatment period. This was also confirmed by Creidi et al. in their study (57). The mechanical properties of the skin have been shown to be improved with estrogen replacement therapy in other studies (58). Brincat and coworkers (50,59,60) found significant correlations among the skin (dermal) collagen, skin thickness (measured radiologically), and metacarpal index, both in postmenopausal women who receiving estrogen replacement therapy and in untreated postmenopausal women. The common factor was the connective tissue present in all three sites. These findings were independent of the woman's age, number of years since the menopause, original skin thickness, or metacarpal index. Skin thickness as measured by high-resolution ultrasound showed correlations with bone densitometry and hormonal levels (61). Significant correlations between skin thickness and bone density were noted (_p < 0.0001). In this study, skin ultrasound, in conjunction with bone densitometry, was again suggested as a useful form of screening for osteoporosis. In another study, Holland et al. (62) determined the effects of percutaneous estradiol (E2) implants on the collagen composition and maturity of the bone and skin of osteoporotic postmenopausal women. Cortical bone taken from transiliac biopsies showed a significant increase in the mature crosslinks of both hydroxylysylpyridinoline (p < 0.01) and lysylpyridinoline (p < 0.01) and a significant reduction in the
BRINCAT ET AL.
percentage of collagen (p < 0.001). Trabecular bone exhibited a similar pattern, with lysylpyridinoline increasing significantly (p < 0.05). Skin biopsies demonstrated a significant reduction in the immature cross-link hydroxylysinorleucine (p < 0.01) but no significant change in the percentage of collagen content or the mature cross-link histidinohydroxylysinonorleucine. The median increases in bone density were 11.5% at the lumbar spine and 4.34% at the femoral neck. These findings suggest a reduction in the turnover of bone collagen following estrogen replacement therapy, possibly leading to the increased concentration of the mature collagen cross-links. The resultant formation of a mature collagen fiber with increased strength and elasticity may help reduce the risk of furore osteoporotic fractures (63). Attempts to identify a value of skin thickness that could be used as a threshold in screening for menopausal osteoporosis have been under way for some time (51).
VI. CAROTID ARTERIES Since the Women's Health Initiative Study (WHI), the cardioprotective effect of hormone replacement therapy in postmenopausal women has been put into question (63). However, the average age of the women recruited (63.5 +_ 7 years) and other variables influencing cardiovascular disease may mitigate the study's conclusion that hormone replacement therapy is not cardioprotective (64). The cardioprotective effect of hormone replacement therapy has been suggested for many studies (65). This has been attributed to the favorable serum lipoprotein profile brought about by this treatment in postmenopausal women. However, only a 20% to 25% reduction in coronary artery plaque formation can be accounted for by this mechanism, and other mechanisms of cardioprotection have been suggested (66). One such mechanism could be due to postmenopausal connective tissue changes that mimic those in skin and bone. Using high-frequency ultrasound, our group looked at the differences in the layers of one of the major blood vessels, the carotid artery. One study investigated differences in the externa, the media, and the intima of the carotid artery in two groups of postmenopausal women receiving estrogen replacement therapy and an untreated group. The results showed that the media, the layer containing the greatest amount of collagen, was thicker in the estrogen-treated groups than in the untreated group. The intima of the untreated group was also shown to be significantly thicker than that of the treated groups, possibly due to reduced collagen deposits in the intimal atherosclerotic plaque (Fig. 16.4). This signifies that there was less atherosclerotic plaque in the treated groups than in the untreated group (67). The relationship among ER alpha expression, collagen content, and distensibility of the human uterine artery has also been studied (68,69). A negative correlation between ER
CHAPTER 16 Connective Tissue Changes in the Menopause and with Hormone Replacement Therapy alpha content and collagen content (r = 0.76; n = 13) and a positive correlation between collagen content and passive tension (r - 0.72; n = 13) in the uterine artery were detected. A high content of ER alpha correlates with a lower collagen concentration. This suggests that estrogen through activation of ER alpha protects against inappropriate vascular (intima) collagen accumulation, making the vessel more distensible. If these arterial changes induced by hormone replacement therapy also occur in the coronary arteries (and some suggestion of this has been made), then these may partially explain the cardioprotective effect this treatment has in young (less than 60 years old) postmenopausal women. Early evidence suggests that SERMs have the same effect. Raloxifene has been shown to have the same effect as estrogens on several surrogate markers.
VII. INTERVERTEBRAL DISCS Intervertebral discs have a high collagen content together with hydrophilic glycosaminoglycans. Under the age of 35 years, the intervertebral discs contain the more flexible collagen types II, IV, and 1X, besides the ubiquitous type I collagen (70). With the aging process and possibly with increasing years beyond the menopause, collagen types I, II1, and VI predominate at the expense of collagen types II, IV, and IX. These changes in the collagen content may have significant effects on the shape and biomechanics of the intervertebral disc. In turn, alterations in disc shape and mechanical properties may have an impact on vertebral body deformity and eventual fracture.
233
Muscat Baron et al. recently studied the lumbar intervertebral disc height in three groups of women. The three groups of women consisted of premenopausal women (n = 25), untreated postmenopausal women (n= 33), and hormonetreated postmenopausal women (n = 44). The lumbar intervertebral disc height was shown to be significantly higher in the premenopausal group (height of three lumbar discs: 2.16 _+ 0.1 cm) and the hormone-treated group (disc height 2.2 _+ 0.12 cm) compared with the untreated postmenopausal women (disc height 1.86 _+ 0.06 cm) (p < 0.0001) (71). A similar study of intervertebral disc height measurements with a larger number of women, including a group of postmenopausal women with radiographically confirmed osteoporotic fractures, has been completed by our group. The results revealed a similar pattern as in the previous study. The premenopausal women and hormone-treated women had disc heights of 2.01 _+ 0.09 cm and 2.15 _+ 0.08 cm, respectively. These latter results were significantly greater than the untreated postmenopausal group (height of three lumbar discs 1.82 _+ 0.06 cm) and, furthermore, higher in the osteoporotic fracture group (1.58 + 0.1 cm) (p < 0.0001)(Fig. 16.5).
VIII. CONCLUSION Sex steroids play a vital role in the integrity of connective tissue of an individual. This constitutes yet another strong argument for long-term hormonal or hormone-like therapy.
FIGURE 16.4 Mean ___SE of carotid artery wall, media (A), and intima/media (B) thickness in 51 untreated postmenopausal women as control subjects, 45 postmenopausalwomen receivingoral hormone replacementtherapy (HRT), and 32 postmenopausalwomenwith estradiol implants.
234
BRINCAT ET AL.
FIGURE 16.5 Intervertebral disc height in four groups of postmenopausal women. The disc heights measured belong to the intervertebral discs found between the vertebrae 12th thoracic and 4th lumbar. Fracture grp vs Untreated grp p<0.0001. Fracture grp vs Premenopausal grp p<0.0001. Fracture grp vs HRT grp p<0.0001. Untreated grp vs Premenopausal grp p<0.0001
References 1. Murry RK, Keeley FW. The extracellular matrix. In: Murray RK, Granner DK, Mays PA, Rodwell VW, eds. Harper's biochemistry, 24th ed. London: Appleton & Lange, 1996;667-685. 2. Punnonen R, Vilska S, Rauramo L. Skinfold thickness and long term postmenopausal hormone therapy. Maturitas 1984;5:259-262. 3. Dunn LB, Damesyn M, Moore AA, Reuben DB, Greendale GA. Does estrogen prevent skin aging? Results from the First National Health and Nutrition Examination Survey (NHANESI). Arch Dermatol 1997; 133:339-342. 4. Schmidt JB, Binder M, Demschik G, BieglmayerC. Reiner A. Treatment of skin aging with topical estrogens. IntJ Dermato11996:35L:669-674. 5. The Writing Group for the PEPI Trial. Effects of hormone therapy on bone mineral density: results from the PostmenopausalEstrogen/Progestin Interventions (PEPI) trial.JAm MedAssoc 1996;276:1389-1396. 6. Brincat M, Moniz CF, Kabaln S, et al. Decline in skin collagen content and metacarpal index after the menopause and its prevention with sex hormone replacement. BrJ Obstet Gynaeco11987;94:126-129. 7. Brincat M, Kabalan S, Studd JWW, et al. A study of the relationship of skin collagen content, skin thickness and bone mass in the postmenopausal woman. Obstet Gynaeco11987;70:840-845. 8. Gordon GS. Prevention of bone loss and fractures in women. Maturitas 1984;6:225-242. 9. Muscat Baron Y, Brincat M, Galea R, Muscat Baron A. An epidemiologic study of postmenopausal osteoporotic fractures in a Mediterranean country. Calcified Tissue Int 1994;54:365-369. 10. Bandres E Pombo I, Gonzalez-Huarriz M, Rebollo A, Lopez G, Garcia-Foncillas J. Association between bone mineral density and polymorphisms of the VDR, Er alpha, COL1A1 and CTR genes in Spanish postmenopausal women. J Endocrinol Invest 2005;28:312- 321. 11. Ferrari SL, Rizzoli R. Gene variants for osteoporosis and their pleiotropic effects in aging. MolAspects Med 2005;26:14_5-167. 12. Livshits G. Genetic epidemiology of skeletal system ageing in apparently healthy human population. Mech Ageing Dev 2005;126: 269-279. 13. Ravn P, Warming L, Christau S, Christiansen C. The effect of cartilage of different forms of application of postmenopausal estrogen therapy: comparison of oral and transdermal therapy. Bone 2004;35:1216-1221. 14. Brincat M, Galea R, Muscat Baron Y. Keynote lecture. Italian Menopause Society, Rome, 1994. 15. Vaananen HK, Harkonen PL. Estrogen and bone metabolism. Maturitas 1996;23(suppl):565- 569. 16. Schneider DL, Barrett Connor EL, Morton DJ. Timing of postmenopausal estrogen for optimal bone mineral density.The Rancho Bernardo Study. J A m MedAssoc 1997;277:543-547.
17. Brincat M, Galea R, Muscat Baron Y, Xuereb A. Changes in bone collagen markers and in bone density in hormone treated and untreated postmenopausal women. Maturitas 1997;27:171 - 177. 18. Ginty F, Flynn A, Cashman K. Inter and intra individual variations in urinary excretion of pyridinium crosslinks of collagen in healthy young adults. EurJ Clin Nutr 1998;52:71-73. 19. Suvanto Luukkonen E, Risteli L, Sundstrom H, et al. Comparison of three serum assays for bone collagen formation during postmenopausal estrogen-- progestin therapy. Clin Chim Acta 1996;266:105 - 116. 20. Melton L J 3rd, Khosla S, Atknson EJ, O'Fallon WM, Riggs BL. Relationship of bone turnover to bone density and fractures. J Bone Miner Res 1997;12:1083-1091. 21. Ohta H, MaNta K, Komukai S, Nozawa S. Bone resorption versus estrogen loss following oophorectomy and menopause. Maturitas 2002;30:27- 33. 22. Greenspan SSL, Resnick NM, Parker RA. Early changes in biochemical markers of bone turnover are associated with long term changes in bone mineral density in elderly women on alendronate, hormone replacement therapy, or combination therapy: a three year, double blind, placebo controlled randomised clinical trial. Clin Endocrinol Metab 2005 ;90:2762 - 2767. 23. Hamwi A, Ganem AH, Grebe C, et al. Markers of bone turnover in postmenopausal women receiving hormone replacement therapy. Clin Chem Lab Med 2001;39:414-417. 24. Prestwood KNI, Gunness M, Muchmore DB, et al. A comparison of the effects of raloxifene and estrogen on bone in postmenopausal women. J Clin Endocrinol Metab 2000;85:2197-2202. 25. Watts NB. Clinical utility of biochemical markers of bone remodelling. Clin Chem 1999;45:1359-1368. 26. Barnes MJ, Constable BJ, Morton LF, et al. Age related variations in hydroxylation of kysine and proline on collagen. Biochem J 1974;139: 461-468. 27. Murai A, Miyahara T, Shiozawa S. Age related variations in glycosylation of hydroxyproline in human and rat skin collagens. Biochem Biophys Acta 1975;404:345-348. 28. Risteli J, Kivirikko KI. Intracellular enzymes of collagen biosynthesis in rat liver as a function of age and in hepatic injury induced by dimethylnitrosamine: changes in prolyl hydroxylase, lysyl hydroxylase, collagen galactosyltransferase and collagen glucosyltransferase activities. Biochem J 1976:158:361-367. 29. Hall DA, Reed FB, Noki G, Vince JD. The relative effects of age and cortico steroid therapy on the collagen profiles of dermis from subjects with rheumatoid arthritis. Age Ageing 1974;3:15-22. 30. Dunn LB, Damesyn M, Moore AA, Reuben DB, Greendale GA. Does estrogen prevent skin aging? Results from the First National Health and Nutrition Examination Survey (NHANES I). Arch Dermatol 1997; 133:339-342.
CHaea'v.R 16 Connective Tissue Changes in the Menopause and with H o r m o n e Replacement Therapy
31. Stumpf WE, Sur M, Joshi SE. Estrogen target cells in the skin. Experimentia 1976;30:196. 32. Boucek RJ, Noble NL, Woessner JF Jr. Properties of fibroblasts. In: Page IH ed. Connective tissue thrombosis and atherosclerosis. New York: Academic Press, 1959;193-211. 33. Danforth DN, Vies A, Breen M, et al. The effect of pregnancy and labor on the human cervix: changes in collagen, glycoproteins and glucosaminoglycans. Am J Obstet Gyneco11971;120:641-651. 34. Grosman N, Hindberg E, Schn J. The effect of oestrogenic treatment on the acid mucopolysaccharide pattern in skin of mice. Acta Pharmacol Toxico11971 ;30:458-464. 35. Goldzieher MA. The effects ofoestrogens on the senile skin.J Gerontol 1946;1:196. 36. Black MM. A modified radiographic method for measuring skin thickness. BrJ Dermato11969;1:661. 37. Shuster J, Black MM, McVitie E. The influence of age and sex on skin thickness, skin collagen and density. BrJDermato11975;93:639. 38. Shuster S, Bottoms E. Senile degeneration of skin collagen Clin Sci 1963;25:487-491. 39. Shuster S, Bottoms E. Effect of ultraviolet light on skin collagen. Mature 1963;199:192-193. 40. Reed VB, Hall DA. In: Frische R, Hartmann F eds. Connective tissues--biochemistry and pathophysiology. Berlin: Springer Verlag, 1974;290. 41. Hall DA, Reed FB, Noki G, Vince JD. The relative effects of age and corticosteroid therapy on the collagen profiles of dermis from subjects with rheumatoid arthritis. Age Ageing 1974;3:15-22. 42. Albright F, Bloomberg E, Smith PH. Postmenopausal osteoporosis. Tram AssocAm Phys 1940;55:298- 305. 43. McConkey B, Fraser GR, Bligh AS, Whitely M. Transparent skin and osteoporosis. Lancet 1963;1:693-695. 44. Varila E, Rantala I, Ikarinem A, et al. The effect of topical estradiol on skin collagen ofpostmenopausal women. BrJ Obstet Gynaeco11995;102: 985-989. 45. Tan CY, Stratham B, Marks R, Payne PA. Skin thickness measurements by pulsed ultrasound: its reproducibility, validation and variabil-
ity. BrJ Dermato11994:96:1392-1394. 46. Goodrich SM, Wood JE. The effect of oestradiol 1713 on peripheral venous distensibility and velocity of venous blood flow. Am J Obstet
Gyneco11966;96:407-411. 47. Rauramo L, Punnonen R. Wirking einer oralen estrogentherapie mit oestriolsuccinat auf die haut hastierter. Frauen Haut Gerchluts Kr 1969;44:463-470. 48. Punnonen R. Effect of castration and peroral therapy on skin. Acta
Obstet Gynaecol &and Supp11973;21:1-4. 49. Brincat M, Moniz CJ, Studd JWW, et al. Long term effects of the menopause and sex hormones on skin thickness. Br J Obstet Gynaecol 1985;92:256-259. 50. Brincat M, Moniz CJ, Kabalan S, et al. Decline in skin collagen content and metacarpal index after the menopause and its prevention with sex hormone replacement. BrJ Obstet Gynaecol 1987;94: 126-129. 51. Brincat M, Moniz CF, Studd JWW, et al. Sex hormones and skin collagen content in postmenopausal women. Br Med J 1983;287: 1337-1338. 52. Shahrad P, Marks RA. Pharmacological effect of oestrone on human epidermis. BrJ Dermato11977;97:383- 386. 53. Castelo-Barcia C, Pons F, Gratacos E, et al. Relationship between skin collagen and bone changes during ageing. Maturitas 1994;18:199-206. 54. Castelo Branco C, Duran M, Gonzalez Merlo J. Skin collagen changes related to age and hormone replacement therapy. Maturitas 1992; 15:113-119. 55. Brincat M. Skin collagen mimics metacarpal index reflecting the effect of the menopause and HRT on osteoporosis. In: Christiansen C, et al, eds. Osteoporosis, vol 1. Proceedings of the Copenhagen International Symposium on Osteoporosis. 1984.
235
56. Schmidt JB, Binder M, Demschik G, Bieglmayer C, Reiner A. Treatment of skin aging with topical estrogens. IntJDermato11996;35:669-674. 57. Creidi P, Faivre B, Agache P, et al. Effect of a conjugated estrogen (Premarin) cream on ageing facial skin. A comparative study with a placebo cream. Maturitas 1994;19:211 - 223. 58. Pverard GE, Letawe L, Dowlati A, Pierard-Franchimant L. Effect of hormone replacement therapy for menopause on the mechanical properties of skin.JAm Soc 1995;43:662-664. 59. Brincat M, Studd W W , Moniz CF, Parsons V, Darby AJ. Skin thickness and skin collagen mimic an index of osteoporosis in the postmenopausal woman. In: Christiansen C, et al, eds. Osteoporosis, vol 1.
Proceedings of the Copenhagen International Symposium on Osteoporosis. 1984:353-355. 60. Brincat M, Studd JWW, Moniz CF, Parsons V, Darby AJ. Skin thickness measurement. A simple screening method for determining patients at risk ofpostmenopausal osteoporosis. In: Christiansen C, et al, eds. Osteoporosis, vol 1. Proceedings of the Copenhagen International Symposium on Osteoporosk.1984. 61. Pverard GE, Letawe L, Dowlati A, Pierard Franchimant L. The effect of hormone replacement therapy for menopause on the mechanical properties of skin.JAm Soc 1995;43:662-664. 62. Holland EE Studd JWW, Mansell JP, Leather AT, Bailey AH. Changes in collagen composition and cross links in bone and skin of osteoporotic postmenopausal women treated with percutaneous estradiol implants. Obstet Gyneco11994;83:180-183. 63. Brincat M, Studd JWW, Moniz CE Parsons V, Darby AJ. Skin thickness and skin collagen mimic an index of osteoporosis in the postmenopausal woman. In: Christiansen C et al, eds. Osteoporosis, vol 1.
Proceedings of the Copenhagen International Symposium on OsteoDorosis. 1984:353-355. 64. Roussouw JE, Anderson GL, Prentince RL, et al. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomised controlled trial. JAMA 2002;17:288:321 - 333. 65. Shapiro S. Effects of HRT on the risks of breast cancer and cardiovascular disease: the validity of the epidemiological evidence. Gynecol Obstet Ferti12004;32:382- 390. 66. Grodstein F, Stampfer MJ, Manson JF, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N EnglJ Med 1996;15:335:453-461. 67. Stevenson JC. The metabolic and cardiovascular consequences of ART. BrJ Clin Pract 1995;49:87-90. 68. Muscat Baron Y, Brincat M, Galea R. Carotid artery wall thickness in women treated with hormone replacement therapy. Maturitas 1997; 27:47-53. 69. Baron YM, Galea R, Brincat M. Carotid artery wall changes in estrogen treated and untreated postmenopausal women. Obstet Gynecol 1998;91:982-986. 70. Nerlich A, Sclheiher E, Boos N. Immunohistological marker for age related changes in the lumbar spine. Spine 1997;22:2781-2795. 71. Muscat Baron Y, Brincat MP, Galea R, Calleja N. Intervertebral disc height in treated and untreated postmenopausal women. Human Reproduction 2005:20:3566-3570.
This Page Intentionally Left Blank
~HAPTER 1 ;
Menopause and the Skin HARRY IRVING K A T Z JANET H I L L PRYSTOWSKY
Department of Dermatology, University of Minnesota, Minneapolis, MN 55455 St. Luke's-Roosevelt Hospital Center, New York, NY 10021
I. I N T R O D U C T I O N
fractional portions of what occurs in the skin of postmenopausal women. Therefore, Table 17.3 can be used as an overview summarization of results from numerous studies that may help to form a larger picture from, at times, conflicting data that exit on this topic. The potentially important roles of androgen or progesterone monotherapy supplementation require more study, for menopausal skin will only be discussed within the context of their use with estrogen and not alone. Several review articles on the topic of hormone replacement therapy and postmenopausal skin changes have been published and give additional overviews on this subject (13,14,16,18-22). Skin changes such as dyspigmentation, solar keratosis, and skin malignancies that are secondary to a lifetime of sun exposure or other extrinsic etiologies not directly related to menopause are beyond the scope of this chapter. Instead, skin changes addressed in detail in this chapter include dry skin, sweating, skin atrophy (wrinkles), wound healing, vitamin D3 production, and overall skin appearance, which are relevant to the lives of menopausal women (8,9,20,21). Each selected topic will include a brief description of the skin structures and functions that are altered by the aging process and the subsequent effects following estrogen or HRT supplementation.
Human skin is a hormonally responsive (1,2), highly integrated three-tiered organ system that serves many homeostatic functions (3,4), including protective barrier activities (5), immune surveillance (3), thermoregulation (6,7), vitamin D production (8,9), and psychosocial personal needs (5,10,11), as are summarized in Table 17.1. Multifunctional hormone expression in the skin includes estrogen, androgen, progesterone, and vitamin D receptors (1,2,4,9). Estrogen affects skin cellular profiferation, differentiation, and function (1,2,12-15, 15a). Furthermore, aromatase and type I 1713-hydroxysteroid dehydrogenase enzymes that catalyze the formation of estradiol are also found in skin, making it a major source of peripheral estrogen generation for postmenopausal women (1). Epidermal keratinocytes, dermal fibroblasts, blood vessels, and hair follicles also have estrogen receptors (1,2,15b). Therefore, it is expected that numerous skin changes occur in menopausal women (14,16-18, 18a). A summary of menopausal and other age- or environment-related skin changes are noted in Table 17.2. The discussion in this chapter will concentrate primarily on selected structural and functional alterations of skin aging that are generally favorably altered with estrogen-based hormone replacement therapy (HRT). Unfortunately, many of the published studies on the subject are observations of HRT users versus nonHRT users in a single center, involving a small number of subjects that have differing experimental designs, varied end points, and assorted treatment regimes. Data derived from such studies may be incomplete but document incremental TREATMENT OF THE POSTMENOPAUSAL WOMAN
II. DRY SKIN Dry skin complaints are common among menopausal women, especially when the environmental ambient humidity is low (23). Dry skin has a noticeably dull color, rough 237
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
238
KATZ AND PRYSTOWSKY
TABLE 17.1
Functional Properties of the Skin
Keratinocyte-derived epidermis (118) 9 Differentiating keratinocytes evolve into corneocytes and other ingredients needed for the constantly renewing, nonviable but enzymatically active skin surface stratum corneum (SC) (5,27) 9 Provides maintenance of gradient against endogenous insensible water loss (TEWL) in an external environment ('-~ 0.5 L/clay) (5,27) 9 Forms a tough, protective impediment against exogenous exposures such as chemicals, toxins, and microbes (5,5,29) 9 Maintains water-holding capacity at the SC skin surface with "natural moisturizing factor" (NMF) to maintain SC extendibility (plasticity) and provides a hydrated milieu for enzymatic corneocyte cleavage during desquamation (5,25,27) 9 Provides ultraviolet light protection by melanin and urocanic acid (3) 9 Biosensor role to restore perturbed barrier homeostasis (e.g., TEWL, humidity, pH) (5,27) 9 Immune surveillance (3) 9 Vitamin D precursor generation (9) 9 Esthetics (10,11)
Fibroblast-derived dermis (118) 9 Provides undulating, harmoniously interactive cohesive and semipermeable membrane at the dermal-epidermal junction 9 Generates the extracellular matrix (i.e., collagen, elastic tissue, and glycosaminoglycans) 9 Skin appendageal structures 9 Source for stem cells to regenerate epidermal keratinocytes 9 Provides hair follicles for esthetics, ultraviolet light protection, and thermoregulation 9 Contains sebaceous glands for antimicrobial and thermoregulatory activities 9 Contains apocrine sweat glands for scents 9 Eccrine sweat glands for thermoregulafion by evaporative cooling 9 Provides efferent and afferent microvasculamre connections to vascular system 9 Has neurosensory functions (e.g., pain, cold, heat, pressure, and vibration) 9 Lymphatic channels provide debris removal 9 Affects thermoregulation 9 Cutaneous blood flow in concert with sweating participates in thermoregulation and is essential for thermal homeostasis (6,40,41)
Adipocyte-derived normal subcutaneous tissue (118) 9 Offers protective buffer for underlying muscular or bony structures and thermal insulation 9 Provide a source of autogenous stem cells (119) 9 Source of estrogen (only source during menopause), lepfin, angiotensinogen, plasminogen activator inhibitor-I, and transforming growth factor-[3 (120)
surface, is flaky, and, if severe, may develop fissures (24-26). Dry skin can be pruritic, painful, or exacerbate other dermatoses. Excessively dry skin can also cause perturbations of the skin's barrier function, leading to inflammation. Dry skin is caused by a defect in the water-holding capacity of the most superficial layer of the epidermis, called the stratum corneum
TABLE 17.2 Summaries of Cumulative Chronologic, Environmental, and Hormonally Related Skin Aging Changes (Clinical Consequences)
Epidermis (5,13,14,18,20,21,27,105,121) 9 Decreased stratum corneum water-holding capacity (dry skin, barrier perturbation, irritant contact dermatitis) 9 Decreased keratinocyte density (epidermal atrophy, decreased epidermal healing, increased susceptibility to injury, impaired healing rate) 9 Decreased Langerhans cells epidermal density (altered immune surveillance) 9 Decreased melanocyte density (skin pallor, increased ultraviolet light susceptibility) 9 Decreased previtamin D generation (vitamin D deficiency) 9 Development of benign, premalignant, and malignant lesions
Dermis (13,14,18,20,21,41,90,93,121) 9 Decreased dermal-epidermal junction cohesion (predisposition to shearing injuries) 9 Decreased fibroblast density and synthetic activity with collagen, elastic tissue, and glycosaminoglycans (atrophy, increased laxity, decreased elasticity, decreased tensile strength, impaired wound healing rate) (69) 9 Impaired lymphatic dermal debris clearance (impaired wound healing) 9 Altered vascular response or blood flow (limited tissue perfusion; impaired thermoregulation, wound healing, and vasodilatation; and increased blood flow during hot flushes) 9 Increased or too robust dermal inflammatory response (impaired wound healing) 9 Altered sensory perception (increased susceptibility to injury)
Subcutaneous tissue (11,17,121,122) 9 Subcutaneous tissue loss (increased susceptibility to injury, less thermal insulation) 9 Subcutaneous tissue accumulation and dimpling of thighs or buttock (cellulite, unaesthetic appearance, increased susceptibility to leg ulcers)
Skin appendages (13,14,18,20,21,105) 9 Altered hair growth (gray hair, decreased scalp hair, increased facial hair) 9 Altered sweat glands activity (altered thermoregulatory activities, less heat or cold adaptability) 9 Altered fingernails and toenails (visible deformities, nail plate thickening, increased susceptibility to onychomycosis, decreased nail strength) 9 Increased sebaceous gland size and decreased sebum production (dry skin, decreased innate immunity)
(SC), which abuts the skin surface (24,27). A normaUy functioning SC is an interactive subdivision of the epidermis that is vital to skin defenses and overall well-being, as summarized in Table 17.1 (5). The primary function of the SC is to provide a barrier for homeostatic protection of underlying structures such that humans can survive in a hostile external terrestrial environment (26-28). Components of the SC are derived from the keratinocytes that populate the underlying viable epidermis. Differentiat-
239
CHAeTE~ 17 Menopause and the Skin TABLE 17.3
Study Results for H R T Effects on the Skin in Postmenopausal Women
Epidermis kinetics and stratum corneum (SC) hydration studies (i.e., dry skin and wound healing) Primary author Callens (36) Dunn (23) Guinot (32) Haapasaari (67) Jemec (33) Pierard-Franchimont (31) Sator (63) Schmidt (34,35) Son (15)
Effect of estrogen or combined HRT in postmenopausal women No change in dry skin scores or SC hydration $ Odds of dry skin from a cross-sectional analysis of a national probability sample-based study in multiple U.S. community sites 1" SC hydration in HRT users > 5 years No change in SC hydration (early menopause) No short-term (7-week) change in SC hydration with topical estradiol gel 1" Water-holding capacity (p < 0.01) 1" SC hydration (p < 0.05) depending on the site Clinical improvement but changes in SC hydration were not significant for topical estrogen creams 1" Epidermal thickness (p < 0.05) and keratinocyte proliferation with topical estradiol preparation
Dermis--collagen content, skin thickness, and elastic tissue studies (i.e., atrophy, sagging, wrinkling, or wound healing) Primary author Brincat (50-52,58,64) Brincat (65) Callens (36) Castelo-Branco (53) Chen (62) Creidi (59) Dunn (23) Fuchs (123) Haapasaari (67) Henry (75) Jemec (33) Maheux (60) Meschia (61) Pierard (72) Pierard-Franchimont (73) Sator (63) Sauerbronn (57) Sawas (54) Schmidt (34,35) Smeets (77) Son (15) Sumino (79) Varila (55) Wolff (71)
Effect of estrogen or combined HRT in postmenopausal women Several studies showing ]" skin collagen content or 1" skin thickness 1" Skin collagen content of abdominal skin biopsy specimens (p < 0.001) with estrogen cream 1" Skin thickness over breast (p -< 0.01) and forearm (p -< 0.05) 1" Skin collagen content (p < 0.05) 1" Skin thickness in HRT group (p < 0.01) 1" Skin thickness (p = 0.013) and $ fine wrinkling (p = 0.012) of the face with an estrogen cream $ Odds of wrinkles from a cross-sectional analysis of a national probability sample-based study in multiple U.S. community sites 1" Epidermal thickness with topical estradiol cream (p = 0.0046) No changes in collagen or elastic fiber content (early menopause) Slows facial aging process but does not prevent wrinkling No short-term (--- 7 week) improvement of elasticity with topical estradiol gel 1" Skin thickness (p = 0.01) and 1" dermal thickness (p < 0.05) from greater trochanter areas but not subumbilical or thyroid sites ]" Skin thickness (p < 0.001) Slows facial sagging or slackness but does not prevent wrinkling Mitigates facial distensibility and elasticity in those subjects most compliant after 5 years $ Skin thickness (p -< 0.05) and 1" elasticity (f < 0.05) ]" Collagen content (p < 0.05) but no changes in elastic tissue 1" Collagen type III content (p < 0.01) ,l, Facial wrinkles for topical estrogen creams (p < 0.002) No significant correlation in skin thickness and bone mineral density 1" Procollagen (p < 0.01) and elastic tissue precursors (p < 0.05) with enhanced transforming growth factor-J3 levels using topical estradiol preparation 1" Skin elasticity (p < 0.05) ]" Procollagen I (p < 0.03) surrogate for collagen synthesis and collagen content (p < 0.02) with topical estradiol gel ,], Skin rigidity and $ average wrinkle scores of face (p < 0.05)
ing daughter keratinocytes provide cornified cells and other ingredients (i.e., keratohyalin granules, lamellar body lipids, keratin filament bundles, amino acids, cytokines, and enzymes) for the constantly renewing, nonviable but enzymatic active stratum corneum (5,27). The SC is composed of a selectively permeable "brick and mortar" barrier-like structure having overlapping layers of tightly compacted, protein-
enriched cornified cells called corneocytes ("the bricks") held to each other by cell-to-cell attachments called corneodesmosomes and hydrophobic lipids extraceUular domains ("the mortar") (5,27,28).
The SC forms a tough protective impediment against exogenous exposures (e.g., chemicals, toxins, microbes, etc.) (3,24,25,29). Under normal circumstances, the SC is
240 almost completely impermeable to endogenous water loss and maintains a water vapor gradient against excessive insensible water loss or transepidermal water loss (TEWL) (3,26,27). The integrity of the SC is determined experimentally by use of a noninvasive instrument known as an evaporimeter that measures T E W L from the skin surface. An intact SC barrier has low T E W L values (approximately 0.5 L per day), whereas a perturbed SC, which can be found in many inflammatory skin diseases, has T E W L levels that are 2 to 10 times higher than normal. Maintenance of the SC water balance is accomplished by virtue of matured corneodesmosomes-bound corneocytes and the physical conformation of long, tight, tortuous intercellular lamellar hydrophobic lipid (mostly ceramide) domains (5,27). Additionally, the presence of a sufficient amount of an innate "natural moisturizing factor" (NMF) to maintain proper hydration is important for homeostasis of the SC. N M F is formed by hydrolysis of filaggrin (derived from keratinocyte keratohyalin granules) within corneocytes. N M F is a mixture of amino acids, lactate, urea, and other ingredients that are hygroscopic (i.e., have the ability to absorb water from exogenous and endogenous sources). Evolutionarily, N M F at the SC surface allows for the storage of sufficient amounts of water (optimally at 30% to 50%) to maintain flexibility (plasticity) and a hydrated milieu for orderly enzyme-mediated desquamation with imperceptible corneocyte cleavage leaving a visibly smooth skin surface appearance (3,24,25,27). Ambient humidity is an important influence on the degree of SC hydration (5). Low-humidity environments stimulate keratinocyte proliferation (i.e., more NMF) to promote greater SC hydration. Determination of the actual SC water content is technically difficult to measure. However, hydration of the SC can be estimated by use of a noninvasive instrument that measures electrical capacitance. Electrical capacitance is dependent on the dielectric constant of SC constituents. Proportionally, water has a much higher dielectric constant than the other SC components. Electrical capacitance is measured in arbitrary units and is used as an indirect measure of SC hydration. Additional functions of the epidermis are also noted in Table 17.1. With dry skin, the SC less pliable and corneocytes are shed in visible clumps or flakes, especially if the water content is less than 10% (27). Severe dry skin can cause intense pruritus, compromise barrier function, and lead to secondary microbial invasion or inflammation. It is believed that N M F is lower or functionally deficient in aged skin (27,30). A National Health and Nutrition Examination Survey that included 3875 postmenopausal women from multiple USA communities found that the odds of having dry skin are less if women have been on HRT compared with those women who have not been on HRT (23). In addition, a small pilot biophysical measurement study
KATZ AND PRYSTOWSKY
evaluated the skin water-holding capacity in menopausal women (n = 30) to determine whether transdermal estrogen (n = 15) had an effect on water-holding capacity of the skin (31). Skin capacitance and transepidermal water losses were measured on normal-looking skin and at plastic occlusive stress test sites. Investigators demonstrated that the water-holding capacity of the stratum corneum was significantly increased at the plastic occlusion stress test site in those women receiving transdermal estrogen compared with menopausal women not receiving transdermal estrogen (31). Also recently, an SC hydration study was conducted in 106 Caucasian menopausal women, with otherwise healthy facial skin, who took HRT for less than 1 year, 1 to 5 years, more than 5 years, or who had never taken HRT. Electrical capacitance measurements reflecting SC hydration were significantly higher in those on HRT (especially for those who had been on H R T more than 5 years) compared with those never on H R T (32). However, other authors (33-36) have noted no significant change in epidermal hydration and dry skin, as is summarized in Table 17.3. Regardless of whether or not postmenopausal women are on HRT, more practical adjuncts are useful to prevent dry skin, which include precautions regarding the use of skin-care items that paradoxically might predispose the SC to dry out (37,38). The use of harsh soaps, astringents or other dehydrating skin-care products or exposure to ultraviolet light can lower the N M F and may compromise SC barrier function (25). Postmenopausal patients will typically have fewer dry skin symptoms if they use less soap and bath less frequently to compensate for the agerelated decrease in N M F from their skin. Cleansers with mild synthetic surfactants with added moisturizing ingredients known as syndets cause minimal barrier perturbation and may be used instead of regular soap (38). Treatment with moisturizing topical agents that contain hygroscopic ingredients such as glycerin, lactic acid, lactate or urea will tend to help the stratum corneum retain water (39). Furthermore, agents with alpha hydroxy acids (i.e., lactic acid or ammonium lactate) or products with urea may aid in the exfoliation or shedding corneocytes, promoting smoother appearing skin. Emollients containing glycol stearate or glyceryl stearate provide partial SC occlusion that helps hydration to improve the appearance of the skin surface (39). Other occlusive moisturizers such as petrolatum or lanolin are helpful because they leave a thin film of oil on the surface of the skin that decreases evaporation of water from the underlying epidermis. Avoiding excessive ultraviolet light exposure during the hours of 10 AM to 4 PM, wearing protective apparel to shade exposed skin, and frequent application of broad-spectrum UVA/UVB sunscreen agents can help to prevent photodamage.
CHAPTER 17 Menopause and the Skin
III. SWEATING Widespread cutaneous sweating is usually evoked by elevation of body temperature (e.g., secondary to fever or exercise) mediated predominantly through the thermoregulatory sympathetic nervous system's elaboration ofacetylcholine (see Table 17.1). Cutaneous vasodilation with attendant increased blood flow and evaporative sweating helps to dissipate heat (6,40,41). Eccrine sweat glands and dermal blood vessels are hormonally responsive (2). In contrast, decreased body temperature causes a noradrenergic medicated vasoconstriction with reduction in cutaneous blood flow that conserves body heat. Episodic, unpredictable, recurrent, transient day or night precipitous sweating (hyperhidrosis) is one component of the unpleasant physical manifestation of the menopausal hot flush syndrome (42,43). Hot flushes are a common hallmark of perimenopause and menopause. Abnormal sweating is preceded or accompanied by the skin sensation of warmth or intense heat and a flushed appearance. Menopausal sweating usually lasts for only minutes, but if it occurs frequently or is intense, it can be embarrassing and may even soil clothing. The incidence of hot flushes is variable from one country to another, but it has been estimated to occur in more than 50% of perimenopausal and menopausal women in the United States (43). The exact mechanism underlying menopausal hot flushes is unknown. However, recent data suggest that hot flushes are a multifactorial syndrome that involves deficiency of neuroendocrine factors involved in thermoregulation (44). Endogenous estrogen levels alone are not predictive of hot flush frequency or severity (43). The hypothalamus is considered an important focal point in the evolution of hot flushes (44). The hypothalamic thermoregulatory system acts like a thermostat located within the brain. Hypothalamic thermoregulatory pathways initiate core body temperature (To), peripheral vascular, neuromuscular, and sudoriferous responses. An increase in Tc can cause vasodilation and sweating. In contrast, a decrease in T~ can cause vasoconstriction and shivering. A neuroendocrine agent such as estrogen is thought to modulate hypothalamic thermoregulatory control within a normal homeostatic thermoneutral temperature range (45). Data suggest that perimenopausal and menopausal women who experience hot flushes have a narrow thermoneutral temperature range. Hence, a small elevation in core body temperature (i.e., a few tenths of a degree) can result in sweating that would not elicit such a response in otherwise nonclimacteric women (7). Therefore, lower sweating thresholds and high sweat rates characterize postmenopausal women with hot flushes (7). Exogenous estrogen therapy can raise sweating thresholds by expanding narrow thermoneutral zones to a more normal temperature range but does not affect core body temperature fluctuations
241 (7,44). Menopausal hyperhidrosis may be mild in intensity or occur infrequently and require no therapy other than lifestyle changes such as a cool environment. However, many other women do have considerable discomfort and require treatment interventions (46). As is well known, systemic estrogen-based hormone replacement therapy is a very effective treatment for menopausal vasomotor and sudoriferous instability. Although topical antisweat remedies such as antiperspirants and aluminum chloride preparation may be useful for localized sweating, they are impractical if sweating is widespread. Alternatively, serotonin reuptake inhibitors and clonidine have been used systemically for menopausal hot flushes (43,46-48).
IV. ATROPHY The dermis provides the structural support and the microvasculature repository supplying nourishment for the epidermis, as is summarized in Table 17.1. The fibroblast is the key cell for the dermis. Fibroblasts synthesize proteins such as collagen and elastic tissues along with hyaluronic acidcontaining glycosaminoglycans that are the major constituents for the dermal extracellular matrix (ECM). Transforming growth factor-~3 (TGF- f3) is emerging as a very important cytokine for homeostasis of the dermal E C M (49). TGF-I3 is part of a multifunctional cytokine isoform family that upregulates fibroblast synthesis of the extracellular matrix proteins and downregulates the expression of matrix metalloproteinases that can degrade matrix proteins (15). Matrix metalloproteinases (MMPs) are a family of proteolytic enzymes such as collagenase and elastase that can degrade collagen and elastic tissue, respectively. Tissue matrix metalloproteinase inhibitor (TMMPI) dampens the proteolytic activity of MMR Collagen is formed from procollagen that is secreted by fibroblasts into the extracellular matrix. In skin collagen content experiments, hydroxyproline measurements are used as a surrogate for collagen content of the dermis. Collagen tissue (the majority as type I and a minority as type III) makes up the bulk of the fibrous protein elements (approximately 95%) of the dermis, providing the main mass and tensile strength for the skin. Sparse but especially organized elastic tissue along with collagen provides the venue for stretched skin to rebound to its original form (i.e., elasticity). Glycosaminoglycans, primarily through the presence of the hyaluronic acid, imbibe endogenous water that accounts for skin turgor (firmness). An altered balance of TGF-~3 levels, fibroblast synthetic activity, M M P enzyme activation, or presence of T M M P I can affect the support function of dermal ECM, leading to attendant wrinkling or sagging that are hallmark of aged skin. Examples of intrinsically aged skin can be found over the buttock or another cutaneous surface that has received
242
KATZ AND PRYSTOWSKY
minimal sun exposure. Intrinsic atrophy is clinically noted by the "cigarette paper" crinkling of the skin surface as is depicted in Fig. 17.1, A (see color insert) in an elderly woman's arm. Dermal atrophy results principally from a loss of collagen (50-57), which accounts for the bulk of the decrease in overall skin thickness (36,51,52,58-63) in estrogen-deficient postmenopausal women. Brincat et al. (51) found that the collagen content in untreated postmenopausal women declined as much as 30% during the first 5 years after onset of menopause. This was reflected by increased urinary hydroxyproline excretion. Furthermore, the decline in collagen in the untreated women was related to the cohort's menopausal age and not their chronologic age. In another study, Brincat et al. (52) examined skin collagen content and skin thickness in postmenopausal women. The skin collagen content and skin thickness showed similar declines between 1% to 2% per year after the menopause. Additional series of studies by Brincat and associates (50,52,58,64,65) found that decreased skin collagen con-
FIGURE 17.1
tent levels in postmenopausal women were incrementally increased by estrogen alone or estrogen plus testosterone supplementation. Other studies that are summarized in Table 17.3 (23,53-55,63) have confirmed that collagen loss and the attendant decreased skin thickness occurring in postmenopausal women is responsive (but not overresponsive) to estrogen alone or estrogen plus progesterone supplementation. Furthermore, hormonal increases in collagen content are proportional to the amount of collagen present when the supplementation is started (52,66). However, if the postmenopausal cohort chosen to be studied has been estrogen deficient only a short time (i.e., less than 1 year), then the relative changes may not be apparent or reach statistical significance, as noted in the negative studies by Haapasaari (33,67) and summarized in Table 17.3. Therefore, hormonal supplementation can be considered as a potential preventative therapy for skin atrophy in postmenopausal women who have adequate dermal collagen or as a therapy for those with decreased dermal collagen (14).
(a) Skin tear due to fragilityof forearm skin in an 87-year-old female without hormone replacement therapy. (b) Lesion taped closed to increase rate of healing by approximatingthe epidermal edges of the full-thicknesswound. (c) Wound healed, with linear scar.
CHAPTER 17 Menopause and the Skin In the real world, most postmenopausal women may not be overtly aware of the consequences of skin atrophy, as noted in Fig. 17.1 (see color insert), but they are concerned about wrinkles, which are an early sign of dermal atrophy (23). Wrinkles are foldlike surface contour depressions indicating underlying structural atrophic alterations within the dermis (68-70). Fine wrinkles are superficial and disappear by stretching the skin but deeper wrinkles do not disappear with clinical technique. Contet-Audonneau, Jeanmaire, and Pauly (69) biopsied 157 wrinkles taken from exposed facial and sun-protected areas of abdominal skin in 46 subjects of elderly men and women. Biopsy specimens underwent histochemistry, immunohistochemistry, electron microscopy, and quantification by image analysis in addition to classic standard staining techniques. The results confirmed that beneath a wrinkle there are more pronounced atrophic changes than the nonwrinkled adjacent surrounding skin. These changes were present in both the covered and exposed skin wrinkles but to a greater degree in the photodamaged facial skin biopsies (69). Dunn et al. (23) performed a cross-sectional analysis in a National Health and Nutrition Examination Survey that included 3875 postmenopausal women from multiple U.S. communities and found that the odds of having wrinkles are less if women have been on HRT compared with those women who have not been on HRT. Table 17.3 summarizes results of several studies of the effects of hormonal replacement therapy on facial wrinkles. Schmidt and associates found that topical estrogen-containing creams significantly improved wrinkles (34,35). In a very small recent study by Wolff and associates (71) it was found that the aggregate total clinical wrinkle score (but not individual scores) from 11 selected facial areas of 9 postmenopausal women treated with HRT was significantly better than that of 11 postmenopausal women not treated with HRT. Sagging and loss of elasticity also indicate further deterioration occurring in the extracellular matrix of the dermis (72,73). Evidence for the causal role of menopause in elastic fiber changes was also found by a light and electronmicroscopic study of biopsy specimens from sun-protected buttock skin from three women who had premature menopause (ages 30 to 37) (74). The authors found that elastic fibers had degenerative changes comparable to that seen in postmenopausal women two decades older. The results suggested that premature estrogen deprivation might be the cause of alterations in skin elastic fibers (74). The results of several biophysical studies on skin extensibility and elasticity in postmenopausal women treated with topical or systemic HRT are summarized in Table 17.3. Separate studies by Pierard et al. (72), Pierard-Franchimont et al. (73), and Henry et al. (75) have demonstrated that skin biophysical changes (i.e., increased laxity, increased extensibility, and decreased elasticity) occur in postmenopausal women and that oral estrogen with sequential progestin therapy can mitigate or slow facial skin aging progression but not prevent wrinkling per se.
243 Decreased skin collagen content and skin thickness parallel decreases of bone density in postmenopausal women. Both bone and skin have abundant type I collagen that declines during menopause. In support of these findings, a group of investigators found that skin-fold thickness of the back of the hand correlated with the presence of osteoporosis. Thus, the lower the skin fold thickness, the greater the probability of underlying osteoporosis (76). The anecdotal clinical observation of"thin skin and weak bones" is generally but not universally supported by many reports in the literature, as summarized in Table 17.3 (22,52,66,77). A further relationship of the loss of dermal extracellular matrix (elastic tissue and collagen) to a bone is supported Sumino and associates (78), who tested the potential association between skin elasticity of the right forearm using a suction device linked to a computer and x-ray absorptiometry for bone mineral density (BMD) of the lumbar spine in 38 otherwise healthy Japanese postmenopausal women. Their results demonstrated a positive correlation between the loss of skin elasticity and decreased BMD in these postmenopausal women (78). Furthermore, in another similar skin elasticity study by Sumino and associates (79) that was performed in 176 postmenopausal women and in 45 matched premenopausal cohorts, they found a significant negative correlation existed between skin elasticity and years in menopause. After menopause, there was a 0.55% decline per year in skin elasticity. In a subset of this study, the authors treated 12 subjects prospectively with HRT and demonstrated a 5% increase in skin elasticity after 1 year of treatment. Hence, the hypoestrogenic effects on skin thickness (i.e., collagen content and elastic tissue) have been demonstrated to increase by small but important increments in women treated with hormone replacement therapy compared with women who were not treated. These adjunctive benefits may be helpful in preventing or slowing the progression of hormone deficiency-related skin changes contributing to skin atrophy, wrinkles, and skin fragility in postmenopausal women. Furthermore, menopausal changes in skin and bone appear to be temporally and physiologically correlated.
V. W O U N D HEALING Aged skin is reported not to heal as speedily as more youthful skin, but generally the final result is qualitatively akin to that in younger subjects (80-84). However, wound healing in an elderly person occurs in a milieu tempered by confounding factors other than age that may, albeit temporarily, slow the process (85). In clinical practice, it is usually the chronic rather than the acute wound that presents a therapeutic challenge (83,85). Although many chronic wounds occur in elderly individuals, it does not necessarily mean that the failure to heal is purely related to age alone but more likely to underlying comorbid conditions such as vas-
244 cular insufficiency, diabetes mellims, sustained localized skin pressure, or debilitation (85). In contrast, a commonly occurring acute wound in elderly individuals is the result of shear trauma involving age-related collagen-deficient thinned skin with a flattened dermal-epidermal junction, as is summarized in Table 17.2 and depicted in Fig. 17.1, A (see color insert), that can be repaired with paper strips (Fig. 17.1, B) (see color insert) because sutures frequently tear the skin further. When these acute wounds heal, they form linear scars on the forearms and legs of women (Fig. 17.1, C) (see color insert). Skin wounds undergo orchestrated healing phases consisting of inflammation, vascular proliferation, cellular proliferation, and remodeling that can be influenced by the local milieu, systemic encumbrances, and hormonal factors (83,85). The cellular elements of wound repair include macrophages, inflammatory cells, vascular endothelial cells, fibroblasts, and keratinocytes. As noted in Table 17.2, aged skin may have an impaired blood supply (83) and fluid drainage (86-88) and lowered innate immune defenses (88-90) as compared with more youthful skin. In addition, aged individuals tend to have elevated levels of proinflammatory cytokines such as interleukin 1 (IL-1), interleukin 6 (IL-6), and tumor necrosis factor-or (TNF-o~) that stimulate an influx of active acute and chronic inflammatory cells with their inherent collection of degrading proteolytic enzymes to a wound that may already be deficient in components of the extracellular matrix (84,91,92). If extensive, cytokine overproduction may paradoxicaUy provoke immunosuppression that may also complicate the complex healing process (84). Hypoestrogenemia in aged menopausal females may account in part for the cytokine-induced hyperinflammation, abundant proteolytic milieu, and impaired dermal EMC content that may adversely slow wound healing (14,16,84,93). A case-cohort study of elderly (->65 years old) patients (n = 44,195) in a United Kingdom General Practice Research Database that provides retrospectively gathered information suggests that HRT usage in elderly women may reduce their susceptibility for chronic wounds (94). Elderly postmenopausal women (n = 4944) who had taken HRT were less likely to develop venous leg ulcers (age-adjusted relative risk 0.65 [95% CI 0.61-0.69]) or pressure ulcers (0.68 [0.62-0.76]) than those who did not use HRT. Ashcroft and associates (12,80,93) have performed wound healing investigations that provide more direct insight on the molecular and enzymatic role of estrogen in human wound healing. In a small skin biopsy wound study of healthy postmenopausal females, they found a reduced rate of healing associated with low levels of TGF-[3 (80). As mentioned earlier, TGF-~ is a multifunctional cytokine family that upregulates fibroblast synthesis of the extraceUular matrix proteins (collagen and elastic fibers) (15) and downregulates the expression of matrix metaUoproteinases that degrade matrix proteins (15). A second, larger randomized, double-blind study by Ashcroft and associates (93) consisted of performing
KATZANDPRYSTOWSKY 4 mm punch biopsies of upper arm sites through either estrogen or placebo patches (applied just before wounding and left on for 24 hours) in 36 elderly (mean age > 70) healthy men (n = 18) and women (n = 18). After 7 and 80 days postwounding, the healing or healed sites were excised to determine the degree of macroscopic wound area (image analysis) and microscopic cross-section wound sizes, collagen contents, inflammation, relative wound strengths, and enzymatic responses during healing. The results revealed that the estrogen patch sites were more healed at day 7 and were stronger at day 80 than the placebo patch sites in both men and women. There were fewer neutrophils, decreased elastase levels, and less fibronectin degradation in the estrogen patch wound sites than the placebo sites. Additional study by Ashcroft and associates (12) involving the wounding of ovariectomized mice suggest that macrophage migration inhibitory factor (MIF) is a proinflammatory cytokine that is upregulated during hypoestrogenemia and contributes to excessive wound inflammation with delayed healing. In the presence of estrogen, MIF is downregulated with less inflammation and more prompt healing. Further human confirmation on the role of estrogen, inflammation, and wound healing actMty in aging skin comes from recent work by Son and associates (15). These authors biopsied buttock target areas of 16 elderly (68 to 82 years) Korean men (n = 8) and women (n = 8) following the application of topical 1713-estradiol or vehicle under occlusion three times per week for 2 weeks. Skin biopsy specimens were analyzed using immunohistochemistry staining, Western blotting, reverse transcriptase polymerase chain reaction (RT-PCR), and enzyme-linked immunosorbent assay (ELISA). Topical 1713-estradiol significantly increased the synthesis and content of dermal collagen along with reducing the expression of matrix metalloproteinases in aged skin. Additional in vitro cellular and molecular studies on this topic also suggest that estrogen facilitates keratinocyte proliferation for reepithelization (15,95,96) and dampens excess inflammation with its attendant proteolytic degradation of the wound's newly formed extracellular matrix (4,97). Therefore, even though human studies are admittedly relatively few in number, estrogen alone or in combination with a progestin appears to be beneficial in regards to either the prevention of wounds (maintenance of collagen and skin thickness) or promotion of a dermal extracellular matrix environment that is more conducive to healing than is usually found with untreated hypoestrogenemia (14).
A. Vitamin D Although vitamin D is not usually directly considered estrogen dependent per se, its endogenous production is dependent on a well-functioning epidermis that could be compromised in postmenopausal women, as is noted in Tables 17.1 and 17.2. Vitamin D is essential for regulation
245
CHAPTER 17 Menopause and the Skin and important for cellular proliferation and differentiation functions in many organ systems, including bones and the skin (9). Aging and, in particular, menopause can place stress on the ability of climacteric women to maintain healthy skin that can be exacerbated by vitamin D deficiency (98-100). It turns out that the skin is a major source of the precursors for vitamin D generation. This takes on further significance because of the public health pronouncements regarding the use of sunscreens and other methods to avoid the hazards of excessive chronic ultraviolet skin exposure (101). However, it is equally true that the epidermis is involved in a photochemical reaction that initiates endogenous production of vitamin D production within the confines of the skin (8,9). Ultraviolet light B (UVB; approximately 297 nm) exposure generates the photochemical conversion of 7-dehydrocholesterol to previtamin D3 within keratinocytes. Previtamin D3 ultimately undergoes further transformation by the liver and kidney, resulting in hormonally active calcitriol (lcx, 25-dihydroxyvitamin D3). Calcitriol has many important functions, including regulation of calcium metabolism and immunomodulatory activity, along with modulation of cellular activities in the epidermis and bones (8,102). It is estimated that more than 90% of vitamin D requirement is formed through sunlight exposure (103). However, a high prevalence of significant seasonal hypovitaminosis D3 has been documented to occur in otherwise healthy postmenopausal women who have had low UVB exposure (99). Inadequate vitamin D due to insufficient UVB may be a contributing cause of osteoporosis (99). Hypovitaminosis D3 can also affect the differentiation of the epidermis, resulting in perturbations of stratum corneum barrier function, delays in wound healing, or exacerbation of other inflammatory dermatoses (9). Therefore, short bursts (5 to 10 minutes) of limited UVB exposure three times a week and increased dietary or supplemental vitamin D3 are recommended to prevent hypovitaminosis (9,100).
VI. P S Y C H O L O G I C E F F E C T S OF MENOPAUSAL CHANGES RELATED T O SKIN A P P E A R A N C E ~ S E L F - I M A G E Hypoestrogenemia occurs during perimenopause. The phenotypic effects of estrogen deficiency may affect psychosocial behavior, such as how others judge a perimenopausal woman's age (10,104). In such a study, Wildt and Sir-Petermann (104) made age estimates for each of 100 women within 1 minute of initially entering a women's clinic. Subsequently, these women had serum estradiol determination performed. The results of the study suggested that the accuracy of age estimation strongly correlated with estradiol levels. Women with high estradiol levels
appeared younger while women with low levels appeared older than their actual chronologic age. The effects of perimenopause and menopause occur within the overall background of progressive inexorable intrinsic alterations (i.e., chronologic aging that occurs in everyone with the passage of time) of the covered integument that underlies additional extrinsic changes, principally due to cumulative chronic ultraviolet light irradiation (i.e., photoaging) over exposed skin (26,105,106). Therefore, the impact of skin aging in postmenopausal women includes not only the limitations in skin homeostatic physiologic functions mentioned earlier, but also the overt changes in physical appearance due to the cumulative effects of chronologic, environmental, ultraviolet light, and hormonal deficiency factors on the integument, as are summarized in Table 17.2 (14). All these are important issues because women may survive one-third of their lives in the postmenopausal state (107). Preserving physical appearance is a multibillion-dollar business. Among older women, attractiveness promotes a better self-image that benefits their mental, physical, and social well-being (11). Last but not the least, women with aging facial skin frequently expresses concern because these changes are a visible and frequent reminder that they are getting older every time they look at themselves in the mirror (11). Referral to a knowledgeable dermatologist or cosmetic surgeon can assist patients with the problems associated with the changes in their skin appearance and, at the same time, screen for precancerous and cancerous lesions. Sunscreens, cosmetics, topical retinoids, cx- and [3-hydroxy acids, chemical peels, dermabrasion, botulinum toxin injections, soft-tissue fillers, laser resurfacing, and cosmetic surgery are among the many approaches to rejuvenating aging skin for women in the postmenopausal phase of their life (105,106,108-114). These agents and procedures principally help ameliorate the cumulative aging changes that accompany menopause.
VIII. T O P I C A L H O R M O N E REPLACEMENT THERAPY In recent years the role of HRT in menopausal and postmenopausal women has been assessed. Systemic HRT therapy is no longer recommended for the prevention of chronic conditions per se (115). Therefore, systemic HRT use solely for postmenopausal skin aging changes noted in this presentation is n o t recommended (13,71,116,116a). Skin benefits, at present, will only be available to those menopausal women who have other climacteric medical needs requiring systemic HRT. This is a conundrum for most women because estrogen is a relatively small molecule that can easily be incorporated into an acceptable cream or gel topical delivery vehicle (117). In fact, as noted in Table 17.3, investigational low-dose topical estrogen products (creams or gels) have
246 generally, but not always, demonstrated increases in dermal collagen, skin thickness, and rates of wound healing and reduction of facial wrinkles (see studies by Brincat [52], Jemec [33], Creidi [59], Callens [36], Schmidt [34], Schmidt [35], Varila [55], Fuchs [123], and Son [15]). Unfortunately, for apparent safety reasons, there is no approved prescription topical estrogen product for use for hypoestrogenic skin. However, it would be highly desirable to have a safe and effective topical estrogenic preparation that has been robusily tested and can be delivered to a limited skin area in such a way as to provide local but not systemic effects. This would necessitate double-blind, randomized, rigorous scientific studies to support all claims of safety and efficacy. Liability concerns and governmental regulatory requirements (117) would be considerable, and it would take many years to achieve; however, the reward of a safe and effective topical estrogen for use on estrogen-deficient skin might be worth the effort by obviating the need for systemic HRT.
VII. SUMMARY Human skin is a three-tiered hormonally responsive integrated organ system to maintain homeostasis through preservation of a constantly renewing, intact, functioning stratum corneum barrier. Epidermal keratinocytes, dermal fibroblasts, blood vessels, and skin appendages have estrogen receptors. Menopausal skin effects occur within the overall background of progressive inexorable intrinsic alterations (i.e., chronologic aging) that are superimposed changes over extrinsic changes, principally due to cumulative ultraviolet exposure. The principal way to discern skin alterations related to hypoestrogenemia occurring in menopausal women is through study of the amelioration that occurs after estrogen-based hormonal replacement. HRT studies have not generally been scientifically vigorous but indicate mixed clinical effects regarding improvements in dry skin, SC hydration, and barrier preservation, as shown in Table 17.3. Dry skin complaints are common and generally can easily be treated with small alterations in personal lifestyle habits. Significant menopausal-associated sweating can be managed with HRT or other systemic therapies. The scientific data regarding the role of HRT for menopausal changes in the dermis (ECM) are in the direction of being possibly beneficial. Incremental increases of the dermal extracellular matrix and a wound microenvironment are statistically significant in small scale studies. HRT helps to maintain or restore collagen and skin thickness, possibly dampening of the pace of development of fine wrinkles, increasing skin strength, improving acute wound healing rates, and perhaps preventing skin breakdown in elderly postmenopausal women. On the molecular level, HRT limits proteolytic effects of excessive inflammation by generating TGF-[3 to promote a wound environment that is favorable to healing.
KATZ AND PRYSTOWSKY
Oral vitamin D3 supplementation can help to augment deficiency that may occur due to decreased photochemical conversion of 7-dehydrocholesterol to previtamin D3 in the aged epidermis. The maintenance of skin function and appearance is important and plays an influential role in the quality of life for women who can spend up to one-third of their lives in a hypoestrogenic state. Ideally, a safe and effective topical hormone replacement preparation without systemic adverse effects would be a highly desirable treatment for hypoestrogenic skin (124).
References 1. Nelson LR, Bulun SE. Estrogen production and action. J Am Acad Dermato12001;45:$116- $124. 2. Thornton MJ. The biological actions of estrogens on skin. Exp Dermatol 2002;11:487- 502. 3. Chuong CM, Nickoloff BJ, Elias PM, et al. What is the 'true' function of skin? Exp Dermato12002;11:159-187. 4. Kanda N, Watanabe S. Regulatory roles of sex hormones in cutaneous biology and immunology. J Dermatol Sci 2005;38:1-7. 5. Elias PM. Stratum corneum defensive functions: an integrated view. J Invest Dermato12005;125:183-200. 6. DiPasquale DM, Buono MJ, Kolkhorst FW. Effect of skin temperature on the cholinergic sensitivity of the human eccrine sweat gland. Jpn J Physio12003;53;427-430. 7. Freedman RR, Subramanian M. Effects of symptomatic status and the menstrual cycle on hot flash-related thermoregulatory parameters. Menopause 2005;12:156-159. 8. Lehmann B, Querings K, Reichrath J. Vitamin D and skin: new aspects for dermatology. Exp Dermato12004;13(suppl)4:11-15. 9. Holick ME Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers, and cardiovascular disease. AmJ Clin Nutr 2004;80:1678S- 1688S. 10. Kligman AM, Koblenzer C. Demographics and psychological implications for the aging population. Dermatol Clin 1997;15:549-553. 11. Koblenzer CS. Psychosocial aspects of beauty: how and why to look good. Clin Dermato121:473-475. 12. Ashcroft GS, Mills SJ, Lei K, et al. Estrogen modulates cutaneous wound healing by downregulating macrophage migration inhibitory factor. J Clin Invest 2003;111:1309-1318. 13. Sator PG, Schmidt JB, Rabe T, Zouboulis CC. Skin aging and sex hormones in women--clinical perspectives for intervention by hormone replacement therapy. Exp Dermato12004;13(suppl)4:36-40. 14. Brincat M, Baron YM, Galea R. Estrogens and the skin. Climacteric 2005;8:110-123. 15. Son ED, Lee JY, Lee S, et al. Topical application of 17beta-estradiol increases extracellular matrix protein synthesis by stimulating TGFbeta signaling in aged human skin in vivo. J Invest Dermatol 2005;124:1149-1161. 15a. Hall G, Phillips TJ. Estrogen and skin: the effects of estrogen, menopause, and hormone replacement therapy on the skin. JAm Acad Dermato12005;53:555-568. 15b. Ohnemus U, Uenalan M, Inzunza J, Gustafsson JA, Paus R. The hair follicle as an estrogen target and source. EndocrRev 2006;27:677-706. 16. Ashcroft GS, Ashworth JJ. Potential role of estrogens in wound healing. d m J Clin Dermato12003;4:737-743. 17. Cooke PS, Naaz A. Role of estrogens in adipocyte development and function. Exp Bid Med (Maywood) 2004;229:1127-1135. 18. Hall GK, Phillips TJ. Skin and hormone therapy. Clin Obstet Gynecol 2004;47:437-449.
CHAPTER 17 Menopause and the Skin 18a. Verdier-Sevrain S, Bonte F, Gilchrest B. Biology of estrogens in skin: implications for skin aging. Exp Dermato! 2006;15:83-94. 19. Shah MG, Maibach HI. Estrogen and skin. An overview. Am J Clin Dermato! 2001;2:143-150. 20. Raine-Fenning NJ, Brincat, MR Muscat-Baron Y. Skin aging and menopause: implications for treatment. Am J Clin Dermato!2003;4:371 - 378. 21. Anonymous. Skin. Obstet Gyneco!2004;104:92S-96S. 22. Shuster S. Osteoporosis, a unitary hypothesis of collagen loss in skin and bone. Med Hypotheses 2005;65:426-432. 23. Dunn LB, Damesyn M, Moore AA, Reuben DB, Greendale GA. Does estrogen prevent skin aging? Results from the First National Health and Nutrition Examination Survey (NHANES I).Arch Dermato! 1997;133: 339-342. 24. Loden M. Role of topical emollients and moisturizers in the treatment of dry skin barrier disorders. Am J Clin Dermato12003;4:771 - 788. 25. Johnson AW. Overview: fundamental skin care--protecting the barrier. Dermato! Ther 2004;17(suppl 1):1-5. 26. Marks R. The stratum corneum barrier: the final frontier. J Nutr 2004; 134:2017 S - 2021S. 27. Rawlings AV, Matts PJ. Stratum corneum moisturization at the molecular level: an update in relation to the dry skin cycle.JInvest Dermato! 2005;124;1099-1110. 28. Elias PM, Ghadially R. The aged epidermal permeability barrier: basis for functional abnormalities. C/in Geriatr Med 2002;18:103-120, vii. 29. Candi E, Schmidt R, Melino G. The cornified envelope: a model of cell death in the skin. Nat Rev Mol Cell gio! 2005;6:328-340. 30. Takahashi M, Tezuka T. The content of free amino acids in the stratum corneum is increased in senile xerosis. Arch Dermato! Res 2004;295: 448-452. 31. Pierard-Franchimont C, Letawe C, Goffin V, Pierard GE. Skin waterholding capacity and transdermal estrogen therapy for menopause: a pilot study. Maturitas 1995;22:151-154. 32. Guinot C, Malvy D, Ambroisine L, et al. Effect of hormonal replacement therapy on skin biophysical properties of menopausal women. Skin Res Techno12005;11:201-204. 33. Jemec GB, Serup J. Short-term effects of topical 17 beta-oestradiol on human postmenopausal skin. Maturitas 1989; 11:229 - 234. 34. Schmidt JB, Binder M, Macheiner W, et al. Treatment of skin ageing symptoms in perimenopausal females with estrogen compounds. A pilot study. Maturitas 1994;20:25 - 30. 35. Schmidt JB, Binder M, Demschik G, Bieglmayer C, Reiner A. Treatment of skin aging with topical estrogens. IntJDermato11996;35:669-674. 36. Callens A, Vaillant L, Lecomte P, et al. Does hormonal skin aging exist? A study of the influence of different hormone therapy regimens on the skin of postmenopausal women using non-invasive measurement techniques. Dermatology 1996; 193:289- 294. 37. Scheman AJ. Sensitive skin products: 2003. Skinmed 2003;2:374-375. 38. Subramanyan K. Role of mild cleansing in the management of patient skin. Dermatol Ther 1996;17(supp1 1):26-34. 39. Rawlings AV, Canestrari DA, Dobkowski B. Moisturizer technology versus clinical performance. Dermatol Ther 2004;17(suppl 1):49- 56. 40. Inoue Y, Shibasaki M, Ueda H, Ishizashi H. Mechanisms underlying the age-related decrement in the human sweating response. EurJAppl Physiol OccupPhysio11999;79:121-126. 41. Kenney WL, Munce TA. Invited review: aging and human temperature regulation. JAppl Physio12003;95:2598-2603. 42. Heinemann K, Saad E Sweating attacks: key symptom in menopausal transition only for women? Eur Uro12003;44:583-587. 43. Anonymous. Treatment of menopause-associated vasomotor symptoms: position statement of The North American Menopause Society. Menopause 2004;11:11-33. 44. Notelovitz M. Hot flashes and androgens: a biological rationale for clinical practice. Mayo Clin Proc 2004;79:$8-$13. 45. Freedman RR, Roehrs TA. Lack of sleep disturbance from menopausal hot flashes. Obstet GynecolSurv 2005;60:106-107.
247 46. Bachmann GA. Menopausal vasomotor symptoms: a review of causes, effects and evidence-based treatment options. J Reprod Med 2005;50: 155-165. 47. Freedman RR. Pathophysiology and treatment of menopausal hot flashes. Semin Reprod Med 2005;23:117-125. 48. Anonymous. Treatment of menopausal vasomotor symptoms. Obstet Gyneco12005;105:1137-1139. 49. Ignotz RA, Endo T, Massague J. Regulation of fibronectin and type I collagen mRNA levels by transforming growth factor-beta. JBio! Chem 1987;262:6443 -6446. 50. Brincat M, Moniz CF, Studd JVV, et al. Sex hormones and skin collagen content in postmenopausal women. Br Meat J (Clin Res Ed) 1983;287:1337-1338. 51. Brincat M, Moniz CJ, Studd, JW, et al. Long-term effects of the menopause and sex hormones on skin thickness. Br J Obstet Gynaecol 1985;92:256-259. 52. Brincat M, Versi E, Moniz CF, et al. Skin collagen changes in postmenopausal women receiving different regimens of estrogen therapy. Obstet Gyneco11987;70:123-127. 53. Castelo-Branco C, Duran M, Gonzalez-Merlo J. Skin collagen changes related to age and hormone replacement therapy. Maturitas 1992;15: 113-119. 54. Savvas M, Bishop J, Laurent G, Watson N, Studd J. Type III collagen content in the skin of postmenopausal women receiving oestradiol and testosterone implants. BrJ Obstet Gynaeco11993;100:154-156. 55. Varila E, Rantala I, Oikarinen A, et al. The effect of topical oestradiol on skin collagen of postmenopausal women. Br J Obstet Gynaecol 1995;102:985-989. 56. Affinito P, Palomba S, Sorrentino C, et al. Effects of postmenopausal hypoestrogenism on skin collagen. Maturitas 1999;33:239-247. 57. Sauerbronn AV, Fonseca AM, Bagnoli VR, Saldiva PH, Pinotti JA. The effects of systemic hormonal replacement therapy on the skin of postmenopausal women. IntJ Gynaeco! Obstet 2000;68:35-41. 58. Brincat M, Yuen AW, Studd JVV, et al. Response of skin thickness and metacarpal index to estradiol therapy in postmenopausal women. Obstet Gyneco11987;70:538-541. 59. Creidi P, Faivre B, Agache P, et al. Effect of a conjugated oestrogen (Premarin) cream on ageing facial skin. A comparative study with a placebo cream. Maturitas 1994;19:211 - 223. 60. Maheux R, Naud F, Rioux M, et al. A randomized, double-blind, placebo-controlled study on the effect of conjugated estrogens on skin thickness. Am J Obstet Gyneco11994;170:642-649. 61. Meschia M, Brusch F, Amicarelli FEA. Transdermal hormone replacement therapy and skin in postmenopausal women: A placebo controlled study. Menopause 1994;1:79- 82. 62. Chen L, Dyson M, Rymer J, Bolton PA, Young SR. The use of highfrequency diagnostic ultrasound to investigate the effect of hormone replacement therapy on skin thickness. Skin Res Techno! 2001;7: 95-97. 63. Sator PG, Schmidt JB, Sator MO, Huber JC, Honigsmann H. The influence of hormone replacement therapy on skin ageing: a pilot study. Maturitas 2001;39:43- 55. 64. Brincat M, Moniz CF, Kabalan S, et al. Decline in skin collagen content and metacarpal index after the menopause and its prevention with sex hormone replacement. BrJ Obstet Gynaeco11987;94:126-129. 65. Brincat M, Versi E, O'Dowd T, et al. Skin collagen changes in postmenopausal women receiving oestradiol gel. Maturitas 1987;9:1-5. 66. Castelo-Branco C, Pons F, Gratacos E, et al. Relationship between skin collagen and bone changes during aging. Maturitas 1994;18: 199-206. 67. Haapasaari KM, Raudaskoski T, Kallioinen M, et al. Systemic therapy with estrogen or estrogen with progestin has no effect on skin collagen in postmenopausal women. Maturitas 1997;27:153-162. 68. Lapiere CM. The ageing dermis: the main cause for the appearance of 'old' skin. BrJ Dermato11990;122 (suppl 35):5-11.
248 69. Contet-Audonneau JL, Jeanmaire C, Pauly G. A histological study of human wrinkle structures: comparison between sun-exposed areas of the face, with or without wrinkles, and sun-protected areas. Br J Dermatol 1999;140:1038-1047. 70. Bosset S, Barre P, Chalon A, et al. Skin ageing: clinical and histopathologic study of permanent and reducible wrinkles. EurJ Dermatol 2002;12:247-252. 71. Wolff EF, Narayan D, Taylor HS. Long-term effects of hormone therapy on skin rigidity and wrinkles. Fertil Steri12005;84:285-288. 72. Pierard GE, Letawe C, Dowlati A, Pierard-Franchimont C. Effect of hormone replacement therapy for menopause on the mechanical properties of skin. JAm Geriatr Soc 1995;43:662-665. 73. Pierard-Franchimont C, Cornil F, Dehavay J, et al. Climacteric skin ageing of the face--a prospective longitudinal comparative trial on the effect of oral hormone replacement therapy. Maturitas 1999;32:87-93. 74. Bolognia JL, Braverman IM, Rousseau ME, Sarrel PM. Skin changes in menopause. Maturitas 1989; 11:295- 304. 75. Henry F, Pierard-Franchimont C, Cauwenbergh G, Pierard GE. Agerelated changes in facial skin contours and rheology. JAm Geriatr Soc 1997;45:220-222. 76. Orme SM, Belchetz PE. Is a low skinfold thickness an indicator of osteoporosis? Clin Endocrinol (Oxf) 1994;41:283-287. 77. Smeets AJ, Kuiper JW, van Kuijk C, Berning B, Zwamborn AW. Skin thickness does not reflect bone mineral density in postmenopausal women. OsteoporosInt 1994;4:32-35. 78. Sumino H, Ichikawa S, Abe M, et al. Effects of aging and postmenopausal hypoestrogenism on skin elasticity and bone mineral density in Japanese women. EndocrJ 2004;51:159-164. 79. Sumino H, Ichikawa S, Abe M, et al. Effects of aging, menopause, and hormone replacement therapy on forearm skin elasticity in women. JAm Geriatr Soc 2004;52:945-949. 80. Ashcroft GS, Dodsworth J, van Boxtel E, et al. Estrogen accelerates cutaneous wound heating associated with an increase in TGF-betal levels. Nat Meal 1997;3:1209-1215. 81. Ashcroft GS, Horan MA, Herrick SE. Age-related differences in the temporal and spatial regulation of matrix metaUoproteinases (MMPs) in normal skin and acute cutaneous wounds of healthy humans. Cell Tissue Res 1997;290:581-591. 82. Ashcroft GS, Horan MA, Ferguson MW. Aging alters the inflammatory and endothelial cell adhesion molecule profiles during human cutaneous wound healing. Lab Invest 1998;78:47-58. 83. Gosain A, DiPietro LA. Aging and wound healing. WorldJ Surg 2004;28:321-326. 84. Kovacs EJ. Aging, traumatic injury, and estrogen treatment. Exp Geronto12005;40:549- 555. 85. Thomas DR. Age-related changes in wound healing. Drugs Aging 2001;18:607-620. 86. Gniadecka M, Serup J, Sondergaard J. Age-related diurnal changes of dermal oedema: evaluation by high-frequency ultrasound. BrJDermato11994;131:849-855. 87. Ryan T. The ageing of the blood supply and the lymphatic drainage of the skin. Micron 2004;35:161-171. 88. Reed MJ, Ferara NS, Vernon RB. Impaired migration, integrin function, and actin cytoskeletal organization in dermal fibroblasts from a subset of aged human donors. Mech Ageing Dev 2001;122:1203-1220. 89. Horan MA, Ashcroft GS. Ageing, defence mechanisms and the immune system. Age Ageing 1997;26(suppl 4):15-19. 90. Ogrin R, Darzins P, Khalil Z. Age-related changes in microvascular blood flow and transcutaneous oxygen tension under basal and stimulated conditions. J GerontolA Bid Sci Med Sci 2005;60:200-206. 91. Pfeilschifter J, Koditz R, Pfohl M, Schatz H. Changes in proinflammatory cytokine activity after menopause. Endocr Rev 2002;23:90-119. 92. Menger MD, Vollmar B. Surgical trauma: hyperinflammation versus immunosuppression? Langenbecks Arch Surg 2004;389:475-484.
KATZ AND PRYSTOWSKY 93. Ashcroft GS, Greenwell-Wild T, Horan MA, Wahl SM, Ferguson MW. Topical estrogen accelerates cutaneous wound healing in aged humans associated with an altered inflammatory response. Am J Patho11999;155:1137-1146. 94. Margolis DJ, Knauss J, Bilker W. Hormone replacement therapy and prevention of pressure ulcers and venous leg ulcers. Lancet 2002;359: 675-677. 95. Verdier-Sevrain S, Yaar M, Cantatore J, Traish A, Gilchrest BA. Estradiol induces proliferation of keratinocytes via a receptor mediated mechanism. FASEBJ 2004;18:1252-1254. 96. Kanda N, Watanabe S. 17 [3-estradiol enhances heparin-binding epidermal growth factor-like growth factor production in human keratinocytes. Am J Physiol Cell Physio12005;288:C813- C823. 97. Kanda N, Watanabe S. 17 [3-estradiol inhibits MCP-1 production in human keratinocytes. J Invest Dermato12003 ;120:105 8 -1066. 98. MacLaughlin J, Holick ME Aging decreases the capacity of human skin to produce vitamin D3.J Clin Invest 1985;76:1536-1538. 99. Bhattoa HP, Bettembuk P, Ganacharya S, Balogh A. Prevalence and seasonal variation of hypovitaminosis D and its relationship to bone metabolism in community dwelling postmenopausal Hungarian women. OsteoporosInt 2004;15:447-451. 100. Reginster JY. The high prevalence of inadequate serum vitamin D levels and implications for bone health. Curr Med Res Opin 2005;21: 579-586. 101. Nola I, Kotrulja L. Skin photodamage and lifetime photoprotection. Acta Dermatovenerol Croat 2003;11:32-40. 102. Kira M, KobayashiT, Yoshikawa K. Vitamin D and the skin.JDermatd 2003;30:429-437. 103. Reichrath J, Girndt M, Tilgen W, Q.uerings K. UV-exposition and vitamin D: how much sunlight do we need? Exp Dermato12005;14:153 (abstract). 104. Wildt L, Sir-Petermann T. Oestrogen and age estimations of perimenopausal women. Lancet 1999;354:224. 105. Lockman A, Lockman D. Skin changes in the maturing woman. Clin Fam Pract 2002;4:113-134. 106. Stratigos AJ, Katsambas AD. The role of topical retinoids in the treatment of photoaging. Drugs 2005;65:1061-1072. 107. Wines N, Willsteed E. Menopause and the skin. AustralasJDermatol 2001;42:149-148. 108. Said S, Meshkinpour A, Carruthers A, Carruthers J. Botulinum toxin A: its expanding role in dermatology and esthetics. AmJ Clin Dermatol 2003;4:609-616. 109. Obagi S. Correction of surface deformities: Botox, soft-tissue fillers, lasers and intense pulsed fight, and radiofrequency.Atlas OralMaxillofac Surg Clin North Am 2004;12:271-297. 110. Friedman O. Changes associated with the aging face. FacialHast Surg Clin North Am 2005;13:371-380. 111. DiBernardo BE, Cacciarelli A. Cutaneous lasers. Clin Plast Surg 2005;32:141-150. 112. Weiss J. A review of clinical experience and recommendations for improving patient care. Cutis 2005;75:32-38. 113. Samuel M, Brooke PC, Hollis S, Griffiths, CE. Interventions for photodamaged skin. CochraneDatabase Syst Rev 2005;CD001782. 114. Thornfeldt CR. Cosmeceuticals: separating fact from voodoo science. Skinmed 2005;4:214-220. 115. Anonymous. Summaries for patients. Hormone therapy to prevent chronic conditions in postmenopausal women: recommendations from the U.S. Preventive Services Task Force. Ann Intern Med 2005; 142:159. 116. Schmidt JB. Perspectives of estrogen treatment in skin aging. Exp Dermato12005;14:156. 116a. Verdier-Sevrain S, Bonte F, Gilchrest B. Biology of estrogens in skin: implications for skin aging. Exp Dermato12006;15:83-94.
CHAPTER 17 Menopause and the Skin 117. Draelos ZD. Topical and oral estrogens revisited for antiaging purposes. Fertil Steri12005;84:291-292. 118. Kanitakis J. Anatomy, histology and immunohistochemistry of normal human skin. EurJ Dermato12002;12:390-399. 119. Rodriguez AM, Elabd C, Amri EZ, Ailhaud G, Dani C. The human adipose tissue is a source of multipotent stem cells. Biochimie 2005;87:125-128. 120. Fruhbeck G. The adipose tissue as a source of vasoactive factors. Curr Med Chem CardiovascHematolAgents 2004;2:197-208. 121. Montagna W, Carlisle K. Structural changes in ageing skin. Br J Dermato11990;122(suppl 35):61- 70.
249 122. Rossi AB, Vergnanini AL. Cellulite: a review. J Eur Acad Dermatol Venereo12000;14:251- 262. 123. Fuchs KO, Solis O, Tapawan R, Paranjpe J. The effects of an estrogen and glycolic acid cream on the facial skin of postmenopausal women: a randomized histologic study. Cutis 2003;71:481-488. 124. Naunton M, A1 Hadithy AF, Brouwers JR, Archer DE Estradiol gel: review of the pharmacology, pharmacokinetics, efficacy, and safety in menopausal women. Menopause 2006;13:517-527.
This Page Intentionally Left Blank
;HAPTER 1
Urogenital Symptoms around the Menopause and Beyond GORAN SAMSIOE
Departmentof Obstetrics & Gynecology,Lund University Hospital, 221 85 Lund, Sweden
I. I N T R O D U C T I O N
with age, typically 20% to 30% in young adults, peaking around middle age (prevalence 30% to 40%), and rising steadily in old age (30% to 50%) (3,4). Female urinary incontinence leads to significant morbidity, resulting in impaired quality of life and social and occupational life and adversely affecting women's physical, psychologic, sexual, and domestic well-being. Not surprisingly, symptoms of depression and anxiety increase in patients with U-I; such symptoms complicate medical illnesses and management and have major economic consequences. Despite the burden of UI symptoms, only some 25% of affected women consult a doctor (5-7), often because of their misconception that the condition is a normal consequence of childbirth and aging, their lack of knowledge about possible therapy (6,8), or shame and embarrassment about discussing incontinence (6,9). Several questionnaire-based surveys have been carried out to delineate the nature and occurrence of urogenital problems in women, especially after the age of 50 years (1-8) (Fig. 18.1). Results are variable due mainly to different definitions of incontinence and sometimes to a selection among women that were studied. From observational studies it can be inferred that urinary incontinence is two to three times more prevalent in women than in men and that prevalence increases with age. If data from various population-based studies are pooled, incontinence in women
Advancing age is associated with a decline in a large number of physiologic processes. The urogenital system is no exception. Symptoms and signs from urologic structures and the female sex organs often parallel each other as they originate from a common fetal structure, the urogenital sinus. Symptoms and signs of lost urogenital integrity increase with age but could be modulated by hormonal status as well as lifestyle and environmental factors (Table 18.1). By introducing measures that have an impact on lifestyle, environment, and the hormonal situation, it is possible to influence the urogenital aging process and thereby reduce disabling symptoms and signs in elderly women. The aging process as well as hormonal insufficiency will ultimately occur in all women. Lifestyle changes often accompany these two processes. There is a huge inter-individual as well as intra-individual sensitivity to these changes and symptoms and signs of urogenital aging are therefore highly variable within an individual as well as between individuals. Urinary incontinence is defined by the International Continence Society (ICS) as "any involuntary leakage of urine" (1). Urinary incontinence (UI) is currently an underdiagnosed and underreported condition (2) that is especially common in women. Estimates of the prevalence of UI vary and increase T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
251
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
252
G6RANSAMSIOE TABLE 18.1
Some Common Symptoms of Urogenital Aging
Urologic symptoms Frequent micturition Recurrent cystitis Urethritis Dysuria Urge incontinence Stress incontinence Mixed type of incontinence
Vaginal symptoms Vaginal dryness Loss of lubrication Dyspareunia Vaginitis Vaginal discharge Vulvar itching and burning
seems to increase from 3% to 5% at age 20, to 8% to 9% at age 30, to 12% to 15% at age 50. We published one study in women under 65 years, with a detailed questionnaire on urogenital problems concomitant with a personal interview of each subject (9). Among these 3000 women we can confirm that the overall prevalence of poor control of micturition occurred in about one-third of the cases. However, it should be pointed out that some women complained of this only in specific social situations, such as concomitantly with coughing due to bronchitis. Incontinence was associated with increased parity, overweight, and hysterectomy, confirming previous data, but the condition was also associated with a family history of diabetes. As can be expected, there is a difference between the prevalence and incidence of various symptoms related to urogenital ageing, depending on the design of a given study. In comparison with osteoporosis and cardiovascular disease, much less attention has been paid to problems incurred by urogenital aging. Given the present composition of the population in modern Western countries, 8% of the total population experiences urogenital problems (8). In the United States alone, 20 million women suffer from
these socially disabling symptoms. The prevalence figures obtained by questionnaire surveys are often higher than the clinical experience suggests. Many women feel embarrassed and do not readily address problems related to lost of urogenital integrity, unless specifically asked. Methodologically sound studies are needed to document the economic and personal impact of UI in order to develop better guidelines for the allocation of health care resources and research funding to this major public health problem. Furthermore, although there are several examples of randomized clinical trials and epidemiologic studies that provide evidence about the prevalence of UI, the efficacy and safety of UI treatments, the direct costs of treatment, the characteristics of the patients seeking treatment for UI, and the impact of UI on health-related quality of life (HRQoL), the results cannot be compared or combined across the different studies, countries, and patient populations because the studies have used different methodologies, and were carried out in different settings, using different definitions. The economic costs for incontinence have been estimated to be $8 to $10 billion per annum, corresponding to 2% to 3% of the total costs of the health care system. There can be little doubt that incontinence is the major urogenital problem encountered by women. Consequently this chapter will concentrate on the pathophysiology and treatment of this disabling disorder. Other problems will be mentioned in brief.
A. Paraurethral Collagen and the Menopause Menopause seems to induce a paraurethral connective tissue which is enriched with collagen which has different mechanical properties due to a higher degree of cross-linking (10). A concomitant decrease in the proteoglycan/coUagen ratio suggests that this tissue has some increase in load-bearing ability but has decreased in elasticity (10).
B. Paraurethral Collagen and Stress Urinary Incontinence in Women of Reproductive Age
FIGURE 18.1 Prevalenceofvarious types of femaleurinaryincontinence by age from a community-basedsurvey
A higher collagen concentration in combination with higher cross-linking has been demonstrated in women with stress-incontinence compared with continent control subjects (11). These biochemical changes are accompanied by ultrastructural differences in terms of significantly larger collagen fibril diameters. These alterations focus on a possible disturbance in the fibrilization process (11). It is possible that the pathophysiology of stress urinary incontinence in women is different between women of reproductive age and women after menopause (11). In women of reproductive age with severe stress incontinence, significant changes in the extracellular matrix of the paraurethral
253
CHAPTER 18 Urogenital Symptoms around the Menopause and Beyond connective tissue have been shown (11). These changes result in a stiffer, less supportive paraurethral tissue with no major changes in biochemistry or ultrastructure.
III. MECHANISMS OF C O N T I N E N C E A. Urine Storage and Micturition
II. INNERVATI ON
A. Parasympathetic Pathways The parasympathetic innervation to the bladder is derived from the sacral parasympathetic nucleus located at $2-$4, which in turn receives input from the pontine micturition center in the central nervous system (CNS). Preganglionic axons travel in the pelvic nerve to ganglia located within the pelvic plexus, to vesical ganglia (located on the serosal surface of the bladder), or to intramural ganglia in the bladder wall. Postganglionic axons leave the ganglia to supply the smooth muscle of the lower urinary tract. Activation of parasympathetic postganglionic nerves contracts the smooth muscle of the detrusor during micturition. The role of the parasympathetic nervous system in urethral function remains uncertain (12).
B. Sympathetic Pathways The preganglionic sympathetic axons are derived from T 1 0 - 1 2 and leave the spinal cord via the splanchnic nerves. The fibers are conveyed to the inferior mesenteric ganglion, where they either synapse with postganglionic cell bodies or pass directing through without synaptic contact. They pass via the pelvic plexus or directly to the bladder and urethra via the hypogastric nerve. Some preganglionic axons pass down the sympathetic chain to the lumbosacral level of the spinal cord, where they either synapse with cell bodies of the sympathetic chain ganglia or pass directly in the pelvic nerve (12). A substantial part of urethral tone, an important factor for maintenance of intraurethral pressure, is mediated through the stimulation of c~-adrenoreceptors by released noradrenaline (13).
C. SensoryAfferent Innervation The hypogastric pathway is implicated in sensations associated with bladder filling and bladder pain. The pudendal pathway is involved with the sensation that micturition is imminent. The sensory fibers traveling along the hypogastric nerve are responsible for a feeling of bladder filling and bladder pain. Axons located in the pudendal nerve are implicated in a sensation that micturition is imminent. Sensory afferents travel in the pelvic nerve to the dorsal horn of the sacral spinal cord (14) (Fig. 18.2).
The function of the bladder and urethra is to store urine and, when appropriate, for the evacuation of urine. The bladder forms a reservoir and accommodates to increasing volumes of urine, by the outlet region increasing its resistance. Several factors contribute to the ability of the outlet region to maintain continence. The proximal portion of the urethra lies within the abdominal cavity, and this position allows the pressure to rise along with any rise in intraabdominal pressure by passive transmission. The periurethral striated muscle seems to play an important role in maintaining continence during coughing and sneezing (15), but the rhabdosphincter and the properties of the smooth muscle of the urethra also contribute to maintaining continence. The lamina propria is also believed to "seal" the female urethra, thereby contributing to urethral closure (16). The previously described meshwork of detrusor muscle is ideally suited for emptying the spherical bladder. Urethral opening is not only a consequence of bladder contractions, but it is also facilitated by a drop in urethral pressure which proceeds the contraction (15). The factors contributing to this drop in pressure are still largely unknown. All pelvic floor muscles connected to the urogenital system are intimately involved in the mechanisms of urinary continence. Disturbance of any part of this delicate system may contribute to the development of incontinence. However, when one element is abnormal, other mechanisms may be able to compensate and maintain continence (17,18). The micturition cycle can be separated into three phases: bladder filling, activation of the micturition reflex, and bladder emptying (Fig. 18.3). During filling (stretch), passive bladder wall properties are activated and build up intramural tension, which in turn triggers bladder afferent nerves, leading to a desire to void. Activation of the micturition reflex, in the meaning of initiation of voiding, is a condition of nervous control. Emptying (contractility) is a neuromuscular activity. When the micturition reflex has been activated, pressure falls in the urethra and the pelvic floor becomes silent, as evidenced by the electromyelogram (EMG) pattern. The detrusor muscle contracts, the bladder neck opens, and urine is passed. The bladder emptying is related to the various morphologic and functional aspects of detrusor contractility, including also intramural changes. Stress urinary incontinence (SUI) is caused by a deficiency of the closure pressure in the urethra concomitant, with an increase in intraabdominal pressure in the absence of detrusor activity (17). There are probably several types of stress incontinence. Each type of stress incontinence reflects the malfunction of one or a battery of anatomic components of the supportive and the sphincteric system.
254
G~RaN SAMSIOE
Nerves, neurotransmitters and receptors involved in continence/voiding
FIGURE 18.2 This diagram showswhich neurotransmitters and receptors are involvedin the peripheral motor innervation of the lower urinary tract (LUT). During bladder filling, noradrenaline (NA) release from the sympathetic hypogastric nerve results in bladder relaxationby stimulating f32/f33-adrenoceptorsin the detrusorsmooth muscleand in closureof the bladder neck by stimulating cxl-adrenoceptors.The urethral striated muscle remains closedby release of acetylcholine (ACb) from the somaticpudendal nerve,which results in stimulationof nicotinic receptors (Nic) in the urethral striated muscle and subsequent contraction.The pelvicfloor striated muscles contract due to release of ACh from the somatic sacral nerve. During voiding, the release of NA and ACh from these nerves is inhibited, which leads to relaxation of the bladder neck, urethral striated muscle, and pelvic floor muscles.At the same time, the parasympatheticpelvic nerve releases ACh. This stimulates M2/M3 muscarinic/cholinergic receptors in the detrusor smooth muscle,which results in bladder contraction and emptying.
Until recently the predominating theory behind the occurrence of stress incontinence has been the so-called pressure transmission theory (17). In stress-incontinent women, this theory suggests that the intraabdominal pressure increase (caused by coughs, jumps, etc.)will not be properly transmitted from the intraabdominal pressure compartment (in which the bladder is located) down to the urethra and therefore the bladder pressure will exceed the urethra pressure during a short period, allowing urine to leak. The "integral theory" suggests that continence is maintained by a complicated interplay between specific urogenital structures, aiming to stop the flow of urine at the level of the mid-urethra during situations of increased intra-abdominal pressure. Consequently, a surgical procedure should aim at creating a functional pubourethral ligament supporting the mid-urethra and reinforcing the suburethral vaginal wall (19). Another theory emphasizes a strong anatomic support of the urethra from below, implying that the urethra may be closed by forces acting from above, such as the abdominal pressure transmitted at stress situations (18,19). Accumulating evidence indicates that a more integrated, comprehensive view regarding the different structures, both
inside and outside the urethra, is needed to explain the mechanisms of continence (18,19). In addition, biochemical changes inclusive of enzyme activities may well be different in continent and incontinent patients. This is an area that is only recently being exploited. Defects in the supporting tissues that both actively and passively stabilize the urethra lead to poor control of micturition. In addition, the neuromuscular function is of pivotal importance, and impaired secretion, uptake, or synthesis of neurotransmitters such as noradrenaline and serotonin may also contribute to the pathophysiology.
IV. TREATMENT MODALITIES FOR FEMALE URINARY INCONTINENCE From a clinical standpoint it is often wise to differentiate between stress and urge incontinence, as treatments may be quite different. However, a huge overlap between the two conditions occurs, and therefore it is sometimes difficult to treat them separately. Some types of treatments could be regarded as common to both stress and urge
CHAPTER 18 Urogenital Symptoms around the Menopause and Beyond
255 TABLE 18.2
Scandinavian Guidelines for Assessment of Female Incontinence
Patient history
Type and severity of incontinence Effect on quality of life Other diseases/drug affecting the lower urinary tract Drinking and voiding habits Pelvic mass Mucosal atrophy Genital descensus Blood, glucose, protein Infection Clarifies patient history Basis for behavioral advice Objective incontinence Q.uantification of incontinence
Pelvic examination Urine test Frequency/Volume chart Stress test Pad test FIGURE 18.3 The innervation of the LUT by several afferent sensory and efferent peripheral autonomic and somatic nerves is controlled by the central nervous system. The coordinated alternated activity of the bladder, urethra, and pelvic floor during urine storage and voiding is regulated by afferent sensory and efferent autonomic and somatic peripheral nerves. These involve different neurotransmitters that act on different receptors in the target organs (bladder, urethra, and pelvic floor). The activity of the peripheral nerves is coordinated by the pontine micturition center (PMC) in the brainstem of the central nervous system (CNS). The PMC itself is under conscious control of the cerebral cortex. During bladder filling, the sympathetic fibers of the hypogastric nerve ensure relaxation of the detrusor and closure of the bladder neck. The urethral striated muscle remains closed due to somatic pudendal nerve activity. The somatic sacral nerve ensures that the pelvic floor muscles are also contracted and as such support the compression of the urethra and prevention of leakage. During voiding, the activity of the hypogastric, pudendal, and sacral nerves is inhibited, whereas the parasympathetic pelvic nerve is stimulated. This results in relaxation of the bladder neck, urethral striated muscle, and pelvic floor muscle and contraction of the detrusor and, therefore, bladder emptying.
incontinence, such as pelvic floor exercise, (local) estrogen administration, or electrostimulation. Table 18.2 summarizes recommendations for the clinical evaluation of women with incontinence. Multiple surgical techniques have been devised, perhaps suggesting the difficult nature of this disorder and the less than satisfactory success of treatment (20). Details will be described in Chapter 51.
A. Pharmacologic Treatment of Stress Incontinence In urodynamic investigations, it has been found that the intraurethral pressure is low in women with stress incontinence. Accordingly, data suggest that at least two (x-adrenergic receptor agonists, norephedrine and midodrine, increase intraurethral pressure and were effective also in clinical trials. It should be pointed out that the lowered intraurethral pressure is only a minor contributor to the overall clinical picture,
Adapted from Mouritsen L, Lose G, Ulmsten U, et al. Consensus of basic assessment of female incontinence, dcta Obstet Gyneco! &and 1997;(suppl 166):59-60.
which limits the use of any pharmacologic agent only to milder cases.
B. Serotonin-Norepinephrine Reuptake Inhibitor (SNRI) Preparations One strategy for improving urinary continence is to augment the function of the urethral rhabdosphincter through neuropharmacology. Targeting serotonin and norepinephrine (or noradrenaline) receptors in Onuf's nucleus has been shown to augment the function of the urethral rhabdosphincter by increasing pudendal nerve efferent activity. It is proposed that the ability of serotonin and norepinephrine to enhance the effects of glutamate (the primary excitatory neurotransmitter for pudendal sphincter motor neurons), while having no direct effects of their own, allows facilitation of rhabdosphincter activity during urine storage while allowing complete relaxation during micturition. Indeed duloxetine, a dual serotonin (5-HT)-norepinephrine reuptake inhibitor (SNRI), potentiates these physiologic effects of endogenous serotonin and norepinephrine (by inhibiting the reuptake of these neurotransmitters in the presynaptic element) and thereby enhances the central nervous system's natural continence control mechanisms. Duloxetine, 80 mg per day (40 mg twice daily), decreased the incontinence episode frequency (IEF) and improved incontinence-related quality of life (I-QOL) independent of baseline incontinence severity and also in patients awaiting surgery. In the trial in patients awaiting surgery, onset of action was closely monitored and all
256
G6RAN SAMSIOE
patients who responded to duloxetine did so within 1 to 2 weeks. The decrease in IEF and improvement in I-Q_OL were not due to more frequent voiding, as the mean time between voids increased. Nausea was the most common treatment adverse event. This was mostly experienced early after the start of duloxetine (usually within the first few days) and was usually mild or moderate and nonprogressive in severity. The majority of patients reporting nausea continued treatment with duloxetine, and in most of these patients the nausea resolved within 1 to 4 weeks. It can therefore be concluded that duloxetine 40 mg twice daily is a new and promising pharmacologic treatment approach for women with SUI (21).
C. Urge I n c o n t i n e n c e Drug treatment for female urinary incontinence requires a thorough knowledge of the differential diagnosis and pathophysiology of incontinence, as well as of the pharmacologic agents employed. Pharmacotherapy has to be tailored to suit the incontinence subtype and should be carefully balanced according to efficacy and side effects of the drug. Pharmacologic treatments of incontinence due to detrusot overactivity (urge incontinence) include anticholinergic agents such as tertiary amines. Tricyclic agents such as imipramine have been shown to be effective in decreasing incontinence frequency (22). Strategies in pharmacologic treatment of incontinence due to urethral sphincter insufficiency are designed to increase bladder output resistance. Drugs include those with direct oL-adrenergic agonist activity, which may be enhanced by concomitant estrogen treatment. Also [3-adrenergic blocking drugs, which might allow unopposed stimulation of o~-receptor-mediated contracted muscle responses, are of theoretical interest (23). Antimuscarinic drugs have a long-standing tradition in the treatment of urge incontinence (24). Atropine and related substances are tertiary amines easily absorbed from the gastrointestinal (GI) tract. However, these substances also readily penetrate into the CNS, and central nervous side effects are therefore a limiting factor in clinical usage. The quartery amines such as propantheline, emepronium, and tolterodine display less of these effects. However, oral bioavailability is variable and fairly low (5% to 10%). Mouth dryness, the most common of these side effects, limits the use of these drugs. Tolterodine and metabolites seems to have some selectivity for the bladder compared with the salivary glands (24). Calcium channel blockers such as nifedipine are theoreticaUy interesting. However, peripheral vasodilation is often pronounced and effects on the bladder are still debatable. Hence these drugs are at present not a viable alternative for the clinician (25).
Oxybutynin displays mainly antimuscarinic effects, although a direct relaxing effect on muscular tissues has been demonstrated. Again low oral bioavailability and a fairly high percentage of typically antimuscarinic side effects, such as mouth dryness, constipation, and visual disturbances, limits its use. Side effects are dose dependent, and lower than recommended oral doses may render side effects which are minimal (26). Recently a transdermal patch was introduced that seemed to widen the therapeutic window and increase compliance. 13-receptor agonists have also been used because such agents can relax human bladder muscle cells and increase bladder capacity.Therapeutic effects in patients have been obtained by terbutaline but often at the expense oftachycardia. It can be concluded that the therapeutic effect is less distinct than with anticholinergics, and the specific indications, if any, of this family of drugs needs to be established (27). Overactive bladder (OAB) is the term used to describe the symptom complex of urinary frequency and urgency, with or without urge incontinence. Although antimuscarinic drug therapy has proven to be effective in the management of patients with symptoms of the OAB syndrome, compliance with medication is often affected by the bothersome antimuscarinic adverse effects of dry mouth, constipation, somulence, and blurred vision. The development of bladder selective M(3)-specific antagonists offers the possibility of increasing efficacy while minimizing adverse effects. At present, there are but a few M(3)-specific antagonists available, such as solifenacin and darifenacin. Darifenacin is a novel muscarinic M(3) selective receptor antagonist developed for the once-daily treatment of overactive bladder, a chronic, debilitating, and highly prevalent condition affecting adults of all ages. Preclinical research has confirmed the pharmacologic profile of darifenacin as a potent antimuscarinic agent with up to 59-fold higher affinity for M(3) receptors than other muscarinic receptor subtypes and selectivity for the bladder over other tissues expressing these receptors. Extensive research in large, randomized, placebo-controlled trials have demonstrated that darifenacin, at doses of 7.5 and 15 mg once daily, is efficacious in the treatment of overactive bladder, improving the core symptoms of urinary urgency, urge incontinence, increased micturition frequency, and bladder capacity. In addition, post hoc analyses have shown that many patients can achieve clinically meaningful continence levels, such as 90% or greater reduction in incontinence episodes or 7 or more consecutive dry days. These results are supported by significant improvements in quality of life, which have paralleled the overactive bladder symptom reductions. Both fixed and flexible darifenacin dosing regimens produce these beneficial effects, which extend to the more vulnerable population of older patients. Hence, in conjunction with data showing that this agent has a good safety and tolerability profile, these findings indicate that darifenacin may provide an effective
CHAPTER 18 Urogenital Symptoms around the Menopause and Beyond alternative pharmacotherapy for the treatment of patients with overactive bladder. Solifenacin, with a flexible dosing regimen, showed greater efficacy to tolterodine in decreasing urgency episodes, incontinence, urge incontinence, and pad usage and increasing the volume voided per micturition. More solifenacin-treated patients became continent and reported improvements in perception of bladder condition assessments. The majority of side effects were mild to moderate in nature, and discontinuations were comparable and low in both groups. Another novel and intriguing approach is the use ofbotulinum toxin type A. This regimen should be saved for patients with otherwise refractory idiopathic detrusor overactMty. Preliminary results (28) are encouraging, with subjective marked improvements but also improvements in objectively measured variables.
D. Estrogen Loss of estrogen leads to atrophy of the urethral mucosa, which in turn results in a widening of the urethra and thereby a decrease in resistance to flow. Control of micturition calls for increased muscular activity that is hard to achieve, as estrogen deficiency reduces blood flow through the urogenital tissues, thereby impeding muscular function. With a decline in blood flow, important functions of the immune system may also be affected and, in combination with the wide urethra, facilitate microbial migration into the urinary bladder. This, together with the vaginal reservoir of urinary pathogenic microorganisms, makes it understandable that there is a high frequency of symptomatic and asymptotic bacteriuria, as well as recurrent cystitis (29). In stress incontinence, anatomic changes are the main player and consequently a surgical approach remains the first-line treatment of this condition even if estrogens improve the condition somewhat. Albeit limited, data support the notion that estrogen effects in stress incontinence may be mediated via an influence on collagen and neuromuscular metabolism, which are of importance for the supportive function of the pelvic floor. This may well include an effect on intraurethral pressure. However, use of hormone therapy does not seem to prevent urinary incontinence. In fact, observational studies (30,31) as well as randomized controlled trials suggest the opposite, inclusive of the estrogen-only arm of the W H I study (32). Consequently, neither hormone therapy nor estrogen monotherapy could be used as prophylaxis for urinary incontinence in postmenopausal women. Whether there is a place for hormones in the treatment of incontinence is still open. In elderly women, incontinence is a major reason for a permanent admission to nursing homes. The use of
257
estrogen replacement therapy, especially low-dose topical applications, may substantially reduce both incontinence and recurrent urinary tract infections, thereby making it possible to live outside an institution. This is a major aspect, which affects both quality of life and the costs of caring for those with incontinence. In addition, the use of sanitary protection may be limited. Women with incontinence often suffer socially, as they are afraid of situations in which they cannot reach a toilet almost immediately. That is why many elderly women avoid long travel on buses and visits to the theater or other cultural events and often remain confined to their own homes, which has a serious impact on their quality of life (33). As mentioned earlier, urogenital atrophy does not begin until endogenously produced estrogens are markedly lower than those required for endometrial proliferation. This observation opens a therapeutic window (see Fig. 18.1). It is possible to target urogenital atrophy while minimizing the risk of endometrial proliferation. Estrogen aimed at controlling urogenital symptoms therefore could and should be given at much lower doses than is common when treating other climacteric complaints. Vaginal estrogens can also be given without a progestogen. For estradiol, daily doses of 50 to 100 b~g are recommended in systemic patches to ensure relief of vasomotor symptoms, whereas 8 to 10 btg of vaginally delivered estradiol seems sufficient to combat urogenital symptoms and signs. In other words, only 10% to 15% of the dose required to alleviate hot flushes is needed to achieve clinical improvement of urogenital estrogen deficiency symptoms. Instead of using ultra-low-dose estradiol, it is also possible to administer estriol, which is a short-acting estrogen. Estriol is available in tablet, cream, and vaginal pessary forms. Vaginal administration is preferred, as the therapeutic window with tablets is fairly narrow and may lead to endometrial proliferation and possibly an increased risk of cancer (34). A number of clinical trials have shown the clinical efficacy of estriol as well as of ultralow doses of estradiol in treating lower urinary tract infections (35,36), various forms of incontinence (37,38), and vaginal problems (39). Marked changes may occur within the lower urogenital tract with little or no influence on endocrine parameters such as sex steroid hormone levels, Sex Hormone Binding Globulin (SHBG), or gonadotropin levels. Even lipid metabolic parameters, such as triglycerides, are not influenced (40). Systemic absorption via the vaginal route is negligible after 3 to 4 weeks of treatment. Hence, low-dose estrogen treatment offers no protection for osteoporosis or cardiovascular disease but also carries no increased risk for breast cancer or any other side effect. Consequently, neither absolute nor relative contraindication exists for low-dose estrogen therapy via the vaginal route. However, in the old elderly, long-standing low-dose estrogen therapy may have some systemic effect (34,41).
258
G6RAN SAMSIOE
In 1991, well-established vaginal delivery systems of estrogens aimed at the treatment of urogenital estrogen deficiency symptoms were made available as over-the-counter (OTC) preparations in Sweden. Experience with vaginal estrogens as OTC preparations have been positive; no negative effects have been encountered. With low-dose OTC vaginal estrogens, the costs of treatment and prevention of estrogen deficiency symptoms of the urogenital tract can be limited to the cost of the medication itself. There is no need for compulsory follow-up programs. Trained nurses may give counseling and answer questions about low-dose estrogen treatment by vaginal administration in elderly women. As in the vagina, estrogen stimulates maturation of the urethral mucosa, as evidenced by an increase in superficial cells and a decrease in parabasal cells (42-45). Furthermore, several studies have shown a correlation between cell maturation index in the urethra, with subjective improvement of urinary stress incontinence during estrogen therapy (46-49). However, these two events may well be coincidental, and further studies are needed to suggest a functional relationship. General recommendations may be found in Figure 18.4
E. Genital Symptoms In terms of genital symptoms of menopause, the symptom which occurs first is a feeling of dryness, dependent on atrophy of the mucus-producing glands of the vaginal wall.
In the vagina, atrophy yields a thin mucosa susceptible not only to infections but also to stress through intercourse. Petechial bleedings are not uncommon and will lead to vaginal obliteration, which is most pronounced in the fundal region of the vagina. Itching and a feeling of burning are well-known symptoms in the vulva and vagina as a result of atrophic vaginitis. Also, atrophy of the minor lips can lead to burning sensations and itching in the xatlva, but it should be remembered that several dermatologic conditions such as lichen tuber, psoriasis, eczema, and lichen sclerosus yield similar vulvar symptoms, and differential diagnosis must always be considered. Infections such as candidosis may be superimposed and should also be considered in this context. Vaginal symptoms are especially overt in women who have a male partner capable of and interested in sexual activity. There is a general belief that sexual activity is elderly women is low, but in the study by Barlow et al. in women ages 55 to 74, about half reported being sexually active. Among these, 23% of women in general and 14% of women between 70 and 74 reported frequent intercourse--that is, once per week or more. There was a difference between countries, which may reflect cultural differences, as intercourse was relatively less common in the United Kingdom and Germany and higher in Italy and Denmark. Of those women who were sexually active in the past year, 22% reported a loss of interest. However, the most common cause for lost sexual activity (66%) was loss of partner. Male factors inclusive of impotence were reported by 10% of women. In this report by the women there was also a striking geographic difference in lost male capacity, as 13% of Danish women reported this problem but only 6% of women in Italy and the United Kingdom. Sexual activity is dependent not only on the vaginal function but also on libido.
E Libido
FIGURE 18.4 The definition of the three main types ofUI (SUI, UUI,
and MUI) were developedby the International Continence Society(ICS) and adopted by the International Consultation on Incontinence (ICI). They differentiate three main types/symptomsof UI: Stressurinaryincontinence (SUI) is defined as the involuntaryleakage (of urine) on effort or exertion, or on sneezing or coughing.Urge urinary incontinence (UUI) is defined as the involuntaryleakage (of urine) accompaniedby or immediately precededby urgency.Mixed urinaryincontinence (MUI) is defined as the involuntaryleakage (of urine) associatedwith urgency and also with exertion, effort, sneezing and coughing.
There can be little doubt that sex steroids contribute to brain dimorphism and to neuroplasticity as well as psychoplasticity. Estrogen is a prudent activator of sexual attraction, as the hormone causes and maintains secondary sex characteristics and is included in nocturnal activation of the neurovegetative pathways that lead to vaginal and clitoral congestion during rapid eye movement (REM) phases of sleep (50,51). Estrogens have been shown to activate pheromone secretion in sweat and sebaceous glands that contributes to the scent of women, which is of great importance in sexual attraction (52). For women with an active sex life, the loss of lubrication and glandular functions severely impairs sexual desire. Treatment of this condition improves quality of life, not only for the woman but also for her partner.
CHAPTER 18 Urogenital Symptoms around the Menopause and Beyond Human sexuality is rooted in three main dimensions: biologic, motivational-affective, and cognitive. Human sexunity is mainly relational. Couple satisfaction and premenopausal quality of sexuality are strong predictors also for the quality (and quantity) of postmenopausal sex life (53). Hormonal influence is of importance for the level of sexual arousal and the quality of physical and physiologic response. However, hormones cannot influence the direction of sexual arousal. Optimal estrogenic influence could be considered a very important factor for female sexuality, but it is the androgenic influence that enhances vital energy, libido, and sexual activity both in men and women (54). It should be remembered in this context that peak concentrations of androgens during the menstrual cycle coincides with the estradiol peak at ovulation (55). Administration of combined oral contraceptives but also hormone replacement therapy (HRT) may increase prolactin. Prolactin may inhibit libido via the dopaminergic system in the central nervous system. Progesterone and most progestogens seem to have a mild inhibiting effect. Oxytocin is also of importance. It peaks in plasma at orgasm, and in animals it is responsible for sexual satiety after extended mating. Loss of sexual desire is often a symptom of depression, and it may accompany several chronic diseases both psychiatric and somatic (56). The relation of estrogen treatment to libido is again complex. The vasointestinal peptide (VIP) is a neurotransmitter that is closely linked to variations in sex steroids, and estrogens are a potent stimulator. VIP in turn translates sexual desire into a vascular response that leads to liquid transudation in the vagina (50). By increasing number of glands as well as glandular function, estrogen contributes further to the lubrication of the vagina, thereby rendering dyspareunia less common. On the other hand, systemic estrogen treatment may increase the level of SHBG and thereby decrease the free androgen fraction, which may have a negative impact on libido. But estrogen also has a positive effect on general well-being and mood swings, as well as sleep disturbances. By doing so, a general interest in sex is conceivably increased, but this should be regarded as an indirect effect of the estrogens. Local estrogen treatment in the forms of a vaginal ring, a vaginal tablet, pessaries, or cream improves a local function only. Hence, no effects could be encountered on general well-being or mood. The vaginal lamina propria contains a plexus of large but thin veins. One of the important physiologic response factors for sexual arousal is vasocongestion of the vaginal wall. Normal function of this plexus could well be of importance for vaginal physiology (50). It is also well known that sexually active women have fewer vaginal problems than those who do not have a partner (57). Local stimulatory factors, directly or indirectly, contribute to vaginal integrity.
259
Sexual activity is dependent on the psychologic and physiologic integrity of both the male and the female counterpart. Impotence in men gradually increases with age, and erectile dysfunction is a common problem. Unfortunately, men are less prone than women to consult doctors for sexual dysfunction. Chronic prostatitis can often be treated, and other measures to help or ameliorate erectile dysfunction are available today. In order to have a good sex life, a couple must be able to communicate their problems to one another. Understanding of a problem of the counterpart is of great importance to the success of other modalities of treatment of sexual dysfunction both in men and in women. Using the McCoy Sex Scale, Nathorst B66s and Hammar recently reported (58) that a continuous combined regimen containing 2 mg of estradiol and 1 mg of norethisterone acetate significantly improved several parameters of importance for sexual function. The same was true for the preparation tibolone, which is a steroid with weak estrogenic and progestogenic but also androgenic properties. The difference in effect of the two preparations was small. The result of this study conceivably suggests that androgen treatment could be used for improving sexual function in women. Similar results have been obtained also by Sherwin et al. in combination with estrogens and androgens (59). Also tibolone could be used in women with endogenously deprived androgen influence. This not uncommon in oophorectomized women, as the postmenopausal ovary is responsible for a substantial part of female androgen production throughout life. Embarrassing increase of libido is a side effect sometimes encountered in H R T with androgenic progestogens, especially with advancing age. Of interest is also the fact that the stimulatory effect of oxytocin on libido, which has been well documented in animals, also seems to be true for humans. Indeed H R T influences oxytocin; estrogens stimulate and progestogens attenuate estrogen effects (56).
References 1. Abrams P, Cardozo L, Fall M, et al. The standardisation of terminology in lower urinary tract function: report from the standardisation sub-committee of the international continence society. Urology 2003;61:37-49. 2. Culligan pJ, Heit M. Urinary incontinence in women: evaluation and management. Am Faro Physician 2000;62:2433-2444, 2447, 2452. 3. Hunskaar S, Burgio K, Diokno AC, et al. Epidemiology and natural history of urinary incontinence (UI). In: Abrams P, Cardozo L, Khoury S, Wein A, eds. Incontinence. 2nd International consultation on incontinence, 2nd ed. Plymouth: Health Publication Ltd, 2002:165-201. 4. Hannestad YS, Rortveit G, Sandvik H, Hunskaar S. A communitybased epidemiological survey of female urinary incontinence: the Norwegian EPINCONT study.J Clin Epidemio12000;53:1150-1157. 5. Hannestad YS, Rortveit G, Hunskaar S. Help-seeking and associated factors in female urinary incontinence. &and J Prim Health Care 2002;20:102-107.
260 6. Shaw C, Tansey R, Jackson C, Hyde C, Allan R. Barriers to help seeking in people with urinary symptoms. Family Practice 2001;18:48-52. 7. Kinchen KS, Burgio K, Diokno AC, Fultz NH, Bump R, Obenchain R. Factors associated with women's decisions to seek treatment for urinary incontinence. J Women'sHealth 2003;12:687-698. 8. Norton PA, MacDonald LD, Sedgwick PM, Stanton SL. Distress and delay associated with urinary incontinence, frequency, and urgency in women. BMJ 1988;297:1187-1189. 9. Margalith I, Gillon G, Gordon D. Urinary incontinence in women under 65: quality of life, stress related to incontinence and patterns of seeking health care. Quality of Life Research 2004;13:1381-1390. 10. Hu TW. Impact of urinary incontinence on health care costs. J A m Geriatr Soc 1990;38:292-295. 11. Ekelund P, Grimby A, Milsom I. Urinary incontinence: social and financial costs high. Br MedJ 1993;306:1344-1346. 12. Falconer C, Ekman-Ordeberg G, Ulmsten C, et al. Changes in paraurethral connective tissue at menopause are counteracted by estrogen. Maturitas 1996;24:197-204. 13. Falconer C. Metabolic and functional aspects of collagen in stress urinary incontinence. Menopause Rev 1998;3:17- 23. 14. Lincoln J, Burnstock G. Autonomic innervation of the urinary bladder and urethra. In: Maggi CA, ed. Nervous control of the urogenital system. Chur, UK: Harwood Academic Press, 1993:33-68. 15. Andersson KE. Pharmacology of lower urinary tract smooth muscles and penile erectile tissues. Pharmacol Rev 1993;45:253- 308. 16. Morrison JBE Sensations arising from the lower urinary tract. In: Torrens M, Morrison JBF, eds.: The physiology of the lower urinary tract. Berlin, Germany: Springer-Verlag, 1987:89-131. 17. DeLancey JO. Structural support of the urethra as it relates to stress urinary incontinence: the hammock hypothesis. Am J Obstet Gynecol 1994;170:1713 - 1720. 18. Tanagho EA, Miller ER. Initiation of voiding. Br J Urol 1970;42: 175-183. 19. Rud T, Andersson KE, Asmussen H, et al. Factors maintaining the intraurethral pressure in women. Invest Uro11980;17:343-347. 20. Petros PE, Ulmsten U. An integral theory of female urinary incontinence. Acta Obstet Gynecol &and 1990;69(suppl):153. 21. Cammu H,Van Nylen M, Blockeel C, Kaufman L, Amy JJ. Who will benefit from pelvic floor muscle training for stress urinary incontinence. Am J Obstet Gyneco12004;191:1152 - 1157. 22. Vierhout ME, Lose G. Preventive vaginal and intra-urethral devices in the treatment of female urinary stress incontinence. Curr Opin Obstet Gynecol 1997;9:325- 328. 23. Thyssen H, Lose G. New disposable vaginal device (continence guard) in the treatment of female stress incontinence. Acta Obstet Gynecol Scand 1996;75:170-173. 24. Grodstein F, Lifford K, Resnick NM, Curhan GC. Postmenopausal hormone therapy and risk of developing urinary incontinence. Obstet Gyneco12004;103:254-260. 25. Moghaddas F, Lidfeldt J, Nerbrand C, Jemstr6m H, Samsioe G. Prevalence of urinary incontinence in relation to self-reported depression, intake of serotonergic antidepressants and hormone therapy in middle-aged women: a report from the Women's Health in the Lund Area Study. Menopause 2005;12:318-324. 26. Hen&ix SL, Cochrane BB, Nygaard IE, et al. Effects of estrogen with and without progestin on urinary incontinence. JAMA 2005;293: 935-948. 27. Samsioe G. Urogenital ageing--a hidden problem. Am J Obstet Gynecol 1998;178:S245- $249. 28. Weiderpass E, Baron JA, Adami HO, et al. Low-potency oestrogen and risk of endometrial cancer: a case-control study. Lancet 1999;353: 1824-1828. 29. Brandberg A, Mellstr6m D, Samsioe G. Low dose oral estriol treatment in elderly women with urogenital infections. Acta Obstet Gynecol &and 1987;33(suppl 140):33-38.
G6RAN SAMSIOE 30. Raz R, Stamm W. A controlled trial of intravaginal estriol in postmenopausal women with recurrent urinary tract infections. N Eng!J Med 1993;329:753-756. 31. Samsioe G, Jansson I, Mellstr6m D, Svenborg A. The occurrence, nature and treatment of urinary incontinence in a 70 year old population. Maturitas 1985;7:335-343. 32. Fantl JA, Cardozo L, McClish DK. Estrogen therapy in the management of urinary incontinence in postmenopausal women: a metaanalysis. First report of the Hormones and Urogenital Therapy Committee. Obstet Gyneco! 1994;1:12-18. 33. Barentsen R, van de Weijer PHM, Schram JHN. Continuous low dose estradiol released from a vaginal ring versus estriol vaginal cream for urogenital atrophy. EurJ Obstet Gyneco11997;71:73-80. 34. Sindberg Eriksen P, Rasmussen H. Low-dose 17 [3-ostradiol vaginal tablets in the treatment of atrophic vaginitis: a double-blind placebo controlled study. EurJ Obstet GynecolReprod Bio! 1992;44:137-144. 35. Mettler L, Olsen PG. Long-term treatment of atrophic vaginitis with low-dose oestradiol vaginal tab lets. Maturitas 1991; 14:23 - 31. 36. Crona N, Samsioe G, Svanborg C. A double-blind, cross-over study of the effects of orally administered estriol (Triovex) and placebo on lipid and lipoprotein metabolism in postmenopausal women. J Clin Exp Geronto! 1982;4:137-150. 37. Naessen T, Berglund L, Ulmsten U. Bone loss in elderly women prevented by ultralow doses of parenteral 17 beta-estradiol. Am J Obstet Gyneco11997;177:115-119. 38. Klutke JJ, Bergman JA. Hormonal influence on the urinary tract. Uro! Clin North Am 1995;22:629-639. 39. Bhatia NN, Bergman A, Karram MM. Effects of oestrogens on urethral function in women with urinary incontinence. Am J Obstet Gynecol 1989;260:176-181. 40. Iosif CS, Batra SW, Astedt B. Estrogen receptors in the human female lower urinary tract. A m J Obstet Gyneco11981;141:817- 820. 41. Griebling TL, Nygaard IE. The role of estrogen replacement therapy in the management of urinary incontinence and urinary tract infection in postmenopausal women. Endocrinol Metab Clin North Am 1997;26: 347-360. 42. Smith E Estrogens and the urogenital tract. Studies on steroid hormone receptors and a clinical study on a new estradiol releasing vaginal ring. dcta Obstet Gynecol &and Supp11993;157:1-26. 43. Sundberg A, Rosen J, Gustafsson A, Carlstrom K. Cytosol estrogen receptors in the urogenital tissues in stress incontinent women. Acta Obstet Gynecol &and 1981;60:585-586. 44. Ahlstrom K, Sandahl B, Sjoberg B, et al. Effect of combined treatment with phenyl propanolamine and estriol, compared with estriol treatment alone, in postmenopausal women with stress urinary incontinence. Gynecol Obstet Invest 1990;30:37-43. 45. Fantl JA, Cardozo L, McClish DK, Estrogen therapy in the management of urinary incontinence in postmenopausal women: a metaanalysis. First report of the Hormones and Urogenital Therapy Committee. Obstet Gyneco11994;83:12-18. 46. Baldessarini KJ. Drugs in the treatment of psychiatric disorders. In: Gilman et al, eds. The pharmacological basis of therapeutics, 7th ed. New York: McMillan, 1985;387-445. 47. Cardozo LD, Stanton SL. An objective comparison of the effects of parenterally administered drugs in patients suffering from detrusor instability. J Uro11979;122:58-59. 48. Stahl MMS, Ekstr6m B, Sparf B, Mattiasson A, Andersson KE. Urodynamic and other effects of tolterodine: a novel antimuscarinic drug, for treatment of detrusor overactivity. Neurouro! Urodyn 1995;14: 647-655. 49. Forman A, et al. Effects of nifedipine on the smooth muscle of the human urinary tract in vitro and in vivo. Acta PharmacolToxico11978;43:11 - 118. 50. Yarker YE, Goa KL, Fitton A. Oxybutynin: a review of its pharmacodynamic and pharmacokinetic properties, and its therapeutic use in detrusor instability. Drug Aging 1995;6:243-262.
CHAPTER 18 Urogenital Symptoms around the Menopause and Beyond 51. Lindholm P, Lose G. Terbutaline (Bricanyl) in the treatment of female urge incontinence. UrolInt 1986;41:158-160. 52. Ghonheim GM, Van Leeuwen JS, Elser DM. A randomized controlled trial of duloxetine alone, pelvic floor muscle training alone, combined treatment and no active treatment in women with stress urinary incontinence. J Uro12005;173:1647-1653. 53. Graziottin A. Sexuality and libido. In: Studd JWW, ed. The year book oft be Royal College of Obstetricians and Gynecologists. London: RCOG Press, 1966:235-244. 54. Levin RJ. The mechanisms of human female sexual arousal. Ann Rev Sex Res 1992;3:1-48. 55. Arimondi C, Vannelli GB, Balboni GC. Importance of olfaction in sexual life: morphofunctional and psychological studies in man. Biorned Res (India) 1993;4:43-52.
261
56. MacPhee DC, Johnson SM, Van der Veer MM. Low sexual desire in women: the effect of marital therapy. J Sex Marital Tber 1995;21: 159-182. 57. McCoy NL. Menopause and sexuality. In: Beard MK, ed. Optimizing hormone replacement therapy. Minneapolis: McGraw-Hill, 1995: 32-36. 58. Nathorst-B66s J, Hammar H. Effect on sexual fife: a comparison between tibolone and a continuous estradiol-norethisterone acetate regimen. Maturitas 1997;26:15-20. 59. Sherwin BB, Gelfand MM, Brender W. Androgen enhances sexual motivation in females: a prospective, crossover study of sex steroid administration in the surgical menopause. PsychosomMed 1985;47:339-351.
This Page Intentionally Left Blank
-~ ( ~ H A P T E R 1~
VulvovaginalComplaints GLORIA BACHMANN
RU-FONG
j.
CHENG
Department of Obstetrics/Gynecology & Medicine, UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ 08901
Departmentof Obstetrics, Gynecology,& Reproductive Sciences,UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ 08901
ELISHEVA ROVNER Women's Health Institute, UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ 08901
I. I N T R O D U C T I O N Estrogen loss from follicular depletion in the menopausal ovary is the major cause of vulvovaginal dysfunction in the older woman, as it accounts for most of the anatomic, cytologic, physiologic, and bacteriologic genital changes that occur. As the overall life expectancy of women increases, so does the amount of time that women will spend in the postmenopausal state of their lives. Although hot flushes and night sweats can dramatically herald the loss of estrogen production by the ovaries, these symptoms usually abate over time regardless of whether or not an acute intervention was utilized. Vulvovaginal symptoms, in contrast to vasomotor disturbances, are usually progressive in nature and do not ameliorate with time from the menopausal transition. Women may start to experience subtle signs of changing estrogen levels, such as vaginal dryness, early in the perimenopausal years. Although atrophic changes of the vagina and vulva are unlikely to resolve spontaneously, they are reversed when estrogen therapy is initiated. Data also suggest that continued sexual exchange may help to maintain vaginal integrity as well. Atrophic changes of the vagina and vulva are the most frequent causes of genital complaints in the menopausal woman, but other causes should be considered. Similar symptoms can be experienced TREATMENT OF THE POSTMENOPAUSAL WOMAN
263
by postmenopausal women suffering from infection, trauma, foreign body, inflammatory conditions, and benign and malignant tumors.
II. VULVOVAGINAL C O M P L A I N T S Numerous studies have investigated the prevalence of vulvovaginal symptoms. In an interview study of 500 postmenopausal women attending health centers in Istanbul, Turkey, Oskay et al. found that the most common complaints were urinary incontinence (68.8%), vaginal dryness (43.2%), vaginal itching and infection (36.2%), and difficulty voiding (11.6%). For women who were sexually active, 83.6% reported decrease in sexual desire and 68.8% reported difficulty in sexual arousal (1). Pastore et al., using data from the Women's Health Initiative, a U.S.-based study of postmenopausal women, found the most prevalent urogenital symptoms were vaginal dryness (27%), vaginal irritation or itching (18.6%), vaginal discharge (11.1%), and dysuria (5.2%) (2). In a study of noninstitutionalized Dutch women ages 50 to 75 years, Van Geelen et al. found a 27% overall prevalence of vaginal symptoms. However, only one out of every three women sought help for these symptoms (3). Table 19.1 summarizes the common presentations of vulvovaginal complaints. Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
264
BACHMANNET AL. TABLE 19.1
Common Presentations
Vaginal dryness and irritation Vulvovaginal pruritus Vaginal pressure Malodorous vaginal discharge Dyspareunia Postcoital bleeding Dysuria Urinary frequency and urgency Frequent urinary tract infections Urinary stress incontinence
During the perimenopausal and early menopausal years, vulvovaginal symptoms may be attributed to infectious/ inflammatory, vulvodynia, or psychologic causes. Women may go through several treatment regimens with various antifungal agents, antibiotics, steroids, vitamins, and antidepressants without significant relief of symptoms. Many older women self-medicate with home remedies for relief of symptoms and seek medical assistance only at the point where the vulva is excoriated from chronic scratching. Inflammatory conditions of the vulva, such as contact dermatitis, squamous hyperplasia, or lichen sclerosis, should be ruled out before a hormonal etiology is diagnosed, as these conditions also produce symptoms of severe pruritus, soreness, pain, and dyspareunia like atrophic vaginitis does. Vaginal pressure in the older woman may result from atrophic changes in the genital region alone or in conjunction with uterine prolapse, cystocele, or rectocele. Although vaginal dryness can be bothersome to sexually abstinent women, it is most often voiced as a significant problem in sexually active women who find coital activity uncomfortable because of inadequate lubrication. As the atrophic symptoms worsen over time, pelvic changes due to loss of estrogen become more pronounced, and the symptom complex is referred to as atrophic vaginitis. In addition to vulvar dystrophies, vulvovaginal complaints in postmenopausal women not related directly to estrogen loss also include a variety of systemic and chronic conditions such as diabetes, autoimmune diseases, and urinary incontinence. Table 19.2 summarizes the differential diagnosis for vulvovaginal complaints. An in-depth discussion of vulvar intraepithelial neoplasia (VIN), which can be an important cause of vulvovaginal complaints, can be found elsewhere.
III. E S T R O G E N LOSS Estrogen plays an important role in genital health, especially vulvovaginal health. Estrogen stimulates the maturation of the vaginal epithelium. The mature epithelium produces glycogen, which breaks down to glucose in the vagina. Lactobacilli metabolize this glucose and produce lactic acid, which some data suggest is responsible for the acidic pH of the vagina (4). Lactobacilli have been shown
TABLE 19.2
Differential Diagnosis
Acute: Bacterial vaginosis Candidiasis Pudendal nerve injury Trauma Foreign body Intermediate: Ulcers: Cicatricial pemphigoid Pyoderma gangrenosum Decubitus ulcer Apthous ulcer Papule/nodule/tumor: Bartholin's gland Hidradenoma Apocrine gland Genital wart Pustule/abscess: Bartholin's gland Scabies/pediculosis Chronic: Vulvodynia Dermatoses: Squamous cell hyperplasia Lichen sclerosis Lichen simplex chronicus Lichen planus Chronic eczema Other: Seborrheic dermatitis Psoriasis Contact dermatitis Tinea cruris Medical disorders Psychiatric disorder
to control the growth of vaginal pathogens through the production of hydrogen peroxide (5-6). As the amount of estrogen decreases, the quantity of lactobacilli decrease as well. This in turn leads to a shift of vaginal pH toward alkalinity, which allows for colonization of the vagina by fecal flora and other pathogens. Estrogen loss also leads to a change in both quality and quantity of vaginal secretions. Other changes that occur with estrogen loss include the following (7-11): 9 Vulva loses collagen, adipose, and water-retaining ability. 9 Vagina shortens and narrows. 9 Vaginal walls become thinner and less elastic, lose rogation, and become pale. 9 Vaginal surface becomes friable with petechiae, ulcerations, and bleeding often occurring after minimal trauma. (As this cycle is repeated, adhesions develop between touching surfaces.) 9 Prepuce of clitoris atrophies; the clitoris loses its protective covering and is more easily irritated. 9 Vaginal secretions change in quality and quantity.
265
CHAPTER 19 Vulvovaginal Complaints
IV. SEXUAL COMPLAINTS Although the incidence of sexual problems in the general population is high and increases with age, older women and men are often reluctant to seek medical services for their sexual difficulties because they may consider these dysfunctions a natural part of aging (12-14). Although psychologic and sociocultural factors play a role in the cause of sexual problems, in the older population anatomic and physiologic causes of sexual dysfunctions are more common as compared with younger individuals. Dyspareunia, a common sexual problem in the postmenopausal female, is usually related to estrogen loss to the genitalia, with the most common symptom initially being lack of adequate vaginal lubrication with sexual arousal. Over time, dyspareunia, and postcoital bleeding may also result from estrogen loss to the vagina. Loss of sexual desire also becomes an issue for older women, especially those who are surgically menopausal, because of loss of gonadal production of androgens. Women without significant testosterone level changes may also report loss of libido and be reluctant to pursue their sexual desire because of fear of coital pain from local atrophic changes of the vagina (15). In addition, because the postmenopausal vagina has decreased basal levels of vaginal capillary blood flow and oxygen, it does not reach the level of engorgement that the estrogen-primed vagina does during sexual arousal. Estrogen therapy results in an increase in vaginal lubrication, blood flow, and oxygen levels (16-17). In aging women, dyspareunia is usually of the insertional type rather than the thrusting-induced type seen in younger women with pelvic pathology. Hallstrom (18) reported coital pain in 8% of 231 postmenopausal women he studied and Eskin noted that dyspareunia was the second most frequent complaint of postmenopausal women requesting gynecologic services (19). Bachmann et al. (20) found that 30% of postmenopausal women not on hormone therapy experience dyspareunia. Menopausal patients should be given supportive guidance and factual information about changes in sexual function. Even when urogenital factors have been reversed by estrogen therapy, the loss of sexual desire and arousal may persist. Some women, especially those who have been oophorectomized, may benefit from the addition of an androgen to improve sexual desire (21-24). The level of sexual function for many older women will be determined by the activity of their partner. Vulvovaginal changes and associated symptoms in the woman may affect the sexual performance of the man as well. Medical problems and cancer become more prevalent during the later years and the impact of these diseases on sexual function should be addressed by the clinician. When couples experience problems that are chronic and not improved by nonhormonal or hormonal therapy, more extensive counseling and education should be offered. If information on alter-
native sexual activities and practices still do not alleviate the problem, a sex specialist should be considered (25).
V. EVALUATION Although many urogenital complaints in the postmenopausal woman can be attributed to estrogen loss, all complaints should be evaluated with a pelvic examination to rule out other causes. Medical history should include questions about exogenous agents such as soaps, perfumes, powders, deodorants, use of panty liners or perineal pads, spermicides, and lubricants, all of which may cause or exacerbate symptoms. Inquiries should be made into gynecologic history, especially sexual dysfunction during the reproductive years. A review of systems should focus on recurrent urinary tract infections and urinary incontinence, and social history should screen for any handicap that would interfere with sexual activity, as well as a history of domestic violence and substance abuse.
A. Physical Examination A pelvic examination should be performed, which may include diagnostic testing as dictated by findings. Inspection of the vulva may reveal obvious lesions, dermatoses, erythema, lacerations, or inflammation. More subtle findings on the vulva are changes in connective tissue support and turgor, elasticity, dryness, and loss of pubic hair (26). The labia may show shrinkage or fusion of the labia minora, and there may be a change in caliber of the introital opening or stenosis. Vulvar biopsy with a 3- to 5-mm Keyes punch or shave or excisional biopsy should be performed whenever there is an abnormal-appearing lesion or mass (27). Examination should also include inspection of the urethra for a polyp, urethral caruncle, eversion of the urethral mucosa, and ecchymoses. On vaginal examination, the epithelial rugae, color, and moisture can be noted. If there is discharge, the color and odor should be evaluated. Any lesions such as fissures, ecchymoses, telangiectasias, and ulcerations should be documented and evaluated properly, and the total depth of the vagina can be noted. And finally, one should evaluate for any evidence of organ prolapse, such as cystocele, rectocele, and enterocele. An example of atrophic vaginitis is seen in Fig. 19.1 (see color insert). Cervical cancer screening with a Pap test is recommended every 1 to 3 years depending on prior screening results and indMdual risk (28). A wet mount for infection can reliably diagnose bacterial vaginosis, Candida, and Tricbomonas.
B. Evaluation The problem inherent with evaluation of vulvovaginal symptoms is that they are largely subjective and descriptive. Laboratory markers provide some objective measure for
266
BACHMANN ET AL.
FIGURE 19.1 This photo demonstrates classic changes in an atrophic vulva, with narrowing of the introitus and inflamed mucosal surfaces.There is also loss of labial and vulvar fullness.
Vulva No Lesions or Discharge Diffuse Changes
Trial of Hormone Therapy
Perineal Hygiene Remove Irritants NonNormonal
3<
Improvement 2-4 weeks
Continue Hormone therapy
Vulvar
No improvement 2-4 weeks
Topical Glucocorticoid
reporting results of research and clinical studies but have limited clinical usefulness. Many studies have been performed utilizing the vaginal maturation index, a ratio of parabasal, intermediate, and superficial squamous cells noted on a cytologic smear of ceils from the upper onethird of the vagina. This vaginal scraping is usually performed at the time the Pap smear is taken. The vaginal maturation index provides a means of evaluating chronic hormonal influence on the vaginal vault. Estrogen stimulates the development of superficial squamous cells on the vaginal maturation index, and an atrophic pattern shows predominance of parabasal cells, with a spectrum of intermediate cells in the middle. A wet prep with or without a culture for pathogens can be performed if there is evidence of an abnormal discharge. Vaginal pH obtained by moistening of a pH test strip (29) has been shown to be an effective measure of hormonal influence. Pelvic ultrasound can be performed to evaluate the endometrial lining or stripe if there is postmenopausal vaginal bleeding or postcoital spotting. An endometrial stripe measurement of 4 to 5 mm is consistent with loss of adequate estrogen stimulation (30). Figure 19.2 presents an overview of vulvovaginal complaint management.
Lesion
Biopsy
Vaginal
Wet Prep pH Possible Culture
J Treat
Based
on Pathology
Discharge
No Infection Elevated pH
Atrophic Vaginitis
Trial of Hormone therapy
N Infection
Candida
Ant ifungal
BY
Trich
Metronida zole
FIGURE 19.2 Managementofvulvovaginal complaints in menopausalwomen. BV,,bacterial vaginosis; Trich, Trichomonas.
CHAPTER 19 Vulvovaginal Complaints
267
VI. TREATMENT Postmenopausal women require an individualized management plan, which may consist of supportive guidance, involvement of a caretaker if the patient is bedridden or a resident in a nursing facility, and treatment of any inflammatory conditions and other etiologies, as well as removing genital irritants. Vulvar hygiene, including the avoidance of heavily scented products, use of loose-fitting cotton undergarments, cleansing with mild soap, and thoroughly drying the perineal area after bathing or swimming and other lifestyle modifications, should be included in the counseling of women with vaginal atrophy (31). Smoking causes increased metabolism of estrogen; thus, decreasing or discontinuing tobacco use may also decrease symptoms of atrophy. Coital activity, whether through intercourse or masturbation, increases blood flow to pelvic organs. Treatment of complaints not related to estrogen loss should always be offered before offering hormonal treatments. Although vulvar dystrophy has a low progression rate to invasive malignancy, 30% of vulvar cancers are associated with dystrophy (32). The treatment for hyperplastic dystrophy is topical glucocorticoids such as triamcinolone and betamethasone. Lichen sclerosis is best treated with a high-potency topical glucocorticoid, such as clobetasol propionate 0.05% ointment (Cormax, Temovate), once daily for 6 to 12 weeks, then one to three times per week as maintenance. There is generally no role for hormonal creams (estrogen, progesterone, or testosterone) in the treatment of these conditions. Figures 19.3 and 19.4 give an overview of treatment options.
A. Hormonal therapy 1.
SYSTEMIC
Systemic estrogen therapy and high-dose vaginal estrogen therapies are known to have a stimulatory effect on the endometrium. Whenever unopposed estrogen replacement therapy is given, there should be concern about its trophic
effects on the endometrium, which can result in proliferation, hyperplasia, or carcinoma. Therefore, when only vaginal atrophy is being treated, low-dose vaginal estrogen is the recommended treatment when estrogen is used. In addition, local vaginal estrogen therapy can provide most women with faster and more complete relief of genitourinary complaints than systemic preparations (33-34). Women taking systemic estrogen therapy for vasomotor symptoms or other menopausal symptoms who have not had satisfactory resolution of vulvovaginal changes may benefit from the additional use of local vaginal treatment as well. 2. LOCAL Three FDA-approved vaginal estrogen delivery systems are available for the treatment of vaginal atrophymcreams, a vaginal ring, and a vaginal tablet. With estrogen cream, treatment should commence with a loading dose of 2 to 4 g (one-half to one full applicator) of vaginal estrogen cream daily for 2 weeks. After the initial therapeutic response is achieved with the vaginal cream, the frequency and dose can be titrated down to meet the requirements of the patient. Smaller doses used one to three times per week are usually sufficient to maintain the beneficial results. It is important to note that the systemic absorption of vaginally applied estrogen cream is significantly greater when the vagina is atrophic than when the normal physiology of the vagina has been restored (34). An estradiol-releasing silicone vaginal ring is another option for women who prefer to have a pessary-like device in their vagina for up to 3 months that slowly releases low-dose 17[3-estradiol (35-39). The estrogen ring, when placed in the top of the vaginal vault, continuously releases 5 to 10 g/day of estradiol for up to 90 days (40). The estradiol-releasing ring has been shown to improve significantly patient symptoms such as vaginal dryness, vaginal pruritus, dyspareunia, frequency, and urgency. Ring use also results in improvement of vaginal cytology maturation, decrease in vaginal pH to premenopausal levels, and improvement in physician assessment
Hormonal Estrogen vaginal ~ m
(Premarin | Estrace|
to 1 applicator full daily x 2 weeks, then titrate to lowest, effective dose
Estradiol vaginal tablets (Vagifem |
Nonhormonal Replens | Bioadhesive vaginal moisturizing gel as needed
RepHresh |
25 micrograms per vagina daily x 2 weeks,
Bioadhesive vaginal polymer gel every three
then 1 tablet vaginally 2 times a week
days
Estradiol releasing silicone vaginal ring (Estring | 5 to I 0 micrograms per day for 90 days. For
KY Products | (Jelly, Ultra Gel, Silk-E) Water based vaginal lubricants as needed
some women, this may require a healthcare professional to insert
and replace.
FIGURE
19.3 Treatmentofvulvovaginalcomplaints due to atrophy.
268
BACHMANN ET AL.
Condition Hyperplastic Dystrophy
Lichen Sclerosis
FIGURE 19.4
Treatment Topical glucocorticoids (triamcinolone or betamethasone) 9 Once Daily x Two Weeks 9 Maintenance therapy not necessary. High-potency topical glucocorticoids (clobetasol propionate) 9 Once daily x 6-12 weeks 9 Then 1-3 times/week for maintenance
Treatment of non-atrophy-related conditions.
of vaginal epithelial integrity (27,41-42). One study using the low-dose estradiol-releasing vaginal ring showed no vaginal bleeding (41), whereas another showed no sign of stimulation on progesterone challenge test and endometrial stripe thickness as assessed by pelvic sonogram (39). Two other studies using the estrogen ring showed the rate of endometrial stimulation, as evidenced by progesterone challenge test and endometrial biopsy, to be equivalent to lowdose conjugated estrogen vaginal cream (33-34). Another option is the slow-release hydrophilic estrogen tablet that contains 25 p~g of 1713-estradiol inserted into the vagina twice weekly. In one study, this delivery system for estrogen relieved vaginal symptoms with only 6% (2 of 31) of patients demonstrating an endometrial response (e.g., a prestudy atrophic endometrium changing to a poststudy weakly proliferative one) (40).
B. Nonhormonal Alternatives to hormone therapy for relief of coital discomfort may also be effective for some women. Vaginal moisturizers and lubricating jellies are available, and those with bioadhesive polymer gels have a longer effect than those that are water based. Bioadhesive moisturizing gel has demonstrated a beneficial effect, as noted in one study that reported improved vaginal maturation index, increased vaginal moisture, and increased the quantity of vaginal fluid. Used every 3 days, bioadhesive polymer gel with an acidifying agent has been reported to reduce elevated vaginal pH to acidic levels. Water-based lubricating jellies may also reduce coital discomfort. Many of these products come in a range of preparations, including jelly, gel, moisturizer, and warming formulations (31).
C. Botanicals Usage of botanicals for symptomatic relief of menopausal symptoms has also become popular among some subsets of women. Clinicians need to be aware of this to counsel
patients effectively. According to the North American Menopause Society, more than 30% of women use acupuncture, natural estrogen, herbal supplements, or plant estrogens (43). Soy, chasteberry, ginseng, and vitamin D have all been studied for their use in the treatment of postmenopausal vulvovaginal symptoms. Soy isoflavones are a type of phytoestrogen, which is a plant-based compound with weak estrogenic activity. There is conflicting data with regard to vaginal maturation index, although there may be some effect on estrogen-sensitive tissue in the genital tract. Some soy products may be useful in the treatment of vaginal dryness; however, the sources and doses need to be established (44). Chasteberry, also known as vitex, Chaste tree, Monk's pepper, agnus castus, Indian spice, sage tree hemp, and tree wild pepper, contains hormonelike substances that competitively bind receptors and produce antiandrogenic effects. However, claims of efficacy are poorly documented (44). Ginseng was evaluated by the American Botanicals Council. When they examined 54 different ginseng products they found varying quantities of ginseng present, including little to no ginseng. Other analyses of ginseng products had similar findings of variable quantities of active ingredient. In addition, there was significant contamination of specimens with caffeine, pesticides, and lead (44). Therefore, at this time, ginseng cannot be recommended as a treatment for menopause. Vitamin D is a supplement that shows some promise for use of menopausal vulvovaginal symptoms. Vitamin D is involved with regulation of growth and differentiation of cells, including squamous epithelium in the vagina. In a recent study, vitamin D supplementation caused significant positive changes in the vaginal maturation index (45). This could be a direction for future research with the use of botanicals in menopause.
VII. CONCLUSION Atrophic changes of the vulva, vagina, and lower urinary tract can detract from the quality of life of menopausal women. The patient may initially attribute the symptoms of atrophy to other vulvovaginal conditions. Embarrassment arising from cultural or religious beliefs can prevent the patient from openly discussing problems related to sexual function. Sensitive questioning and thorough evaluation will correctly identify the pattern of changes as those caused by the loss of estrogen. Hormonal and nonhormonal treatments can provide patients with relief and a return to their previous level of function. Because these atrophic changes respond favorably to estrogen therapy, no reason exists for older women to suffer from urogenital atrophy symptoms.
CHAPTER 19 Vulvovaginal Complaints
References 1. Oskay UY, Beji NK, Yalcin O. A study on urogenital complaints of postmenopausal women aged 50 and over. Acta Obstet Gynecol Scand 2005;84:72-78. 2. Pastore LM, Carter RA, Hulka BS, et al. Self-reported urogenital symptoms in postmenopausal women: women's health initiative. Maturitas 2004;49:292-303. 3. Van Geelen JM, Van de Weijer PH, Arnolds HT. Urogenital symptoms and resulting discomfort in non-institutionalized Dutch women aged 50-75 yrs. Int UrogynecolJ Pelvic Floor Dysfunc 2000;11"9-14. 4. Steward-Tule DES. Evidence that vaginal lactobacilli do not ferment glycogen. Am J Obstet Gyneco11964;88:676-679. 5. Eschenbach DA, Davick PR, Williams BL, et al. Prevalence of hydrogen peroxide-producing lactobacillus species in normal women and women with bacterial vaginosis. J Clin Microbio11989;27:251-256. 6. Klebanoff SJ, Hillier SL, Eschenbach DA, et al. Control of the microbial flora of the vagina by H202-generating lactobacilli. J Infect Dis 1991;164:94-100. 7. Milsom I, Arvidsson I, Ekelund P, et al. Factors influencing vaginal cytology, pH and bacterial flora in elderly women. Acta Obstet Gynecol Scand 1993;72:286-291. 8. Hoh I, Manaka M. Analysis of human vaginal secretions by SDSpolyacrylamide gel electrophoresis. Forensic Sci Int 1988;37:237-242. 9. Paavonen J. Physiology and ecology of the vagina. Scand J Infect Dis Supp11983;40:31- 35. 10. Semmens JP, Tsai CC, Semmens EC, et al. Effects of estrogen therapy on vaginal physiology during menopause. Obstet Gyneco11985;66:15-18. 11. Oriba HA, Maibach HI. Vulvar transepidermal water loss (TEWL) decay curves. Effect of occlusion, dilapidation, and age. ilcta Derm Venereol 1989;69:461 - 465. 12. Bachmann GA, Leiblum SR, Grill J. Brief sexual inquiry in gynecologic practice. Obstet Gyneco11989;73:425-427. 13. Dank E, Anderson C, Rubinstein DR. Frequency of sexual dysfunction in animal couples. NEnglJMed 1978;299:111-115. 14. Wise TN, Rabins PY, Gahnsley J. The older patient with a sexual dysfunction. J Sex Marital Ther 1984;10:117-121. 15. Bachmann GA, Leiblum SR, Bernard S, et al. Correlates of sexual desire in postmenopausal women. Maturitas 1985;7:211-216. 16. Hoon PW. Physiologic assessment of sexual response in women: the unfulfilled promise. Clin Obstet Gyneco11984;27:767-779. 17. Semmens J, Wagner G. Estrogen deprivation and vaginal function in postmenopausal women. JAMA 1982;248: 445. 18. Hallstrom T. Sexuality in the climacteric. Clin Obstet Gynecol 1977; 4:227. 19. Eskin B. Sex and the gynecologic patient. In: Oaks WW, Melchiode GA, Ficher I, eds. Sex and the life cycle. New York: Grune & Stratton, 1976. 20. Bachmann GA, Leiblum SR, Kemmann E, et al. Sexual expression and its determinants in the postmenopausal woman. Maturitas 1984;6:19-29. 21. Sherwin BB, Gelfand MM, Brender W. Androgen enhances sexual motivation in females: a prospective, crossover study of sex steroid administration in the surgical menopause. Psychosomivied 1985;47:339-351. 22. Sherin BB, Gelfand MM. Differential symptom response to parenteral estrogen and/or androgen administration in the surgical menopause. A m J Obstet Gyneco11985;151:153-160. 23. Sherwin BB. Affective changes with estrogen and androgen replacement therapy in surgically menopausal women. J~4ffect Disord 1988;14: 177-187. 24. Burger H, Hailes J, Nelson J, et al. Effect of combined implants of oestradiol and testosterone on libido in postmenopausal women. Br Med J 1987;295:936-937.
269 25. Schover LR, Evans RE, Van Eschenbach AC. Sexual rehabilitation in a cancer center: diagnosis and outcome in 384 consultations. Arch Sex Behav 1987;16:445- 461. 26. Bachmann GA, Nevandusky NS. Diagnosis and treatment of atrophic vaginitis. Am Fam Physician 2000;61:3090- 3096. 27. Amin SH, Kuble CL, Fitzpatrick LA. Comprehensive exam of the older woman. Mayo Clin Proc 2003;78:1157-1185. 28. Nevin JE, Pharr ME. Preventative care for the menopausal woman. Prim Care Clin Office Pract 2002;29:583-597. 29. Carranza-Lira S, Fragoso-Diaz N, Luz A, et al. Vaginal dryness assessment in postmenopausal women using PH test strip. Mauritas 2003;45: 55-58. 30. Smith-Bindman R, Kerlikowske K, Feldstein VA, et al. Endovaginal ultrasound to exclude endometrial cancer and other endometrial abnormalities. JAA~4 1998;280:1510-1517. 31. SOGC Clinical Practice Guidelines. The detection and management of vaginal atrophy. IntJ Gynecol Obstet 2005;88:222-228. 32. Messinger-Rapport BJ, Thacker HL. Prevention for the older woman: a practical guide to hormone replacement therapy and urogynecologic health. Geriatrics 2001;56:32-42. 33. Bachmann G, Notelovitz M, Nachtigall I, et al. A comparative study of a low-dose estradiol vaginal ring and conjugated estrogen cream for postmenopausal urogenital atrophy. Prim Care Update Ob/Gyns 1997;4: 109-115. 34. Ayton RA, Darling GM, Murkies AI, et al. A comparative study of safety and efficacy of continuous low dose oestradiol released from a vaginal ring compared with conjugated equine oestrogen vaginal cream in the treatment of postmenopausal urogenital atrophy. Br J Obstet Gyneco11996;103:351-358. 35. Barentsen R, van de Wei]er PH, Schram JH. Continuous low dose estradiol released from a vaginal ring versus estriol vaginal cream for urogenital atrophy. EurJ Obstet GynecolReprod Bid 1997;71:73-80. 36. Henriksson L, Stjernquist M, Boquist I, et al. A comparative multicenter study of the effects of continuous low-dose estradiol released from a new vaginal ring versus estriol vaginal pessaries in postmenopausal women with symptoms and signs of urogenital atrophy. Am J Obstet Gyneco11994;171:624-632. 37. Smith P, Heimer G, Lindskog M, et al. Oestradiol-releasing vaginal ring for treatment of postmenopausal urogenital atrophy. Maturitas 1993;16:145-154. 38. Mettler I, Olsen R Long-term treatment of atrophic vaginitis with low-dose oestradiol vaginal tablets. Maturitas 1991;14:23-31. 39. Foidart J, Vervliet. j, Buytaert R Efficacy of sustained-release vaginal oestriol in alleviating urogenital and systemic climacteric complaints. Maturitas 1991;13:99-107. 40. Gabrielsson J, Wallenbeck L, Birgerson L. Pharmacokinetic data on estradiol in light of the estring concept. Acta Obstet Gynecol &and Suppl 1996;163:26-31. 41. Pschera H, Hjerpe A, Carlstrom K. Influence of the maturity of the vaginal epithelium upon the absorption of vaginally administered estradiol-17 beta and progesterone in postmenopausal women. Gynecol Obstet Invest 1989;27:204-207. 42. Schmidt G, Andersson SB, Nordle O, et al. Release of 17-beta-oestradiol from a vaginal ring in postmenopausal women: pharmacokinetic evaluation. GynecolObstetInvest 1994;38:253-260. 43. Kaufert P, Boggs P, Ettinger B, et al. Women and menopause: beliefs, attitudes, and behaviors. The North American Menopause Society 1997 Menopause Survey. Menopause 1997;5:197-202. 44. ACOG practice bulletin. Use of botanicals for management of menopausal symptoms. Obstet Gyneco12001;97(suppl):1 - 11. 45. Yildirim B, Kaleli B, Duzcan E, et al. The effects of postmenopausal vitamin D treatment on vaginal atrophy. Maturitas 2004;49:334-337.
This Page Intentionally Left Blank
; H A P T E R 2(
Sexuality: Clinical Implications from Epldemlologlc Studies LORRAINE
DENNERSTEIN
Office for Gender & Health, Department of Psychiatry, The University of Melbourne, Parkville 3010, Australia
I. I N T R O D U C T I O N
This chapter uses population-based studies to document the changes that occur in the different domains of women's sexual function in midlife and how these relate to aging, hormonal changes of the menopausal transition, and other psychosocial and physical factors. Populationbased studies provide information on the prevalence and type of changes in sexual function and their relationship to menopausal hormonal changes and other possible determinants. These studies are complementary to clinical trials that provide evidence of hormone effects on specific parameters of sexual function in the groups studied. Population-based studies enable the study of women in their own naturalistic setting. These investigations also assess the effect of factors other than hormonal changes. The results can then be generalized to the ethnic group and location studied. Methodologic limitations of these studies include the fact that most population-based studies of female sexual function have failed to include validated measures of female sexual function. Measures used have mostly been single questions addressed at particular domains of function, or respondents have been asked to report their sexual problems or difficulties. Studies of large sample sizes often ask fewer questions with less assurance of a reliable answer compared with smaller studies, which may be able to collect more
The gradual extension of life span during the 20th century continues, with the life expectancy for American Caucasian women now 80.5 years and, for African-American women, 76.1 years. The average American woman may spend as many or more years in the postmenopausal phase of her life as she does in the fertile phase between menarche and menopause. By 2000 there were an estimated 45 million postmenopausal women in the United States, 2 million of whom had undergone surgical menopause. The number of postmenopausal women worldwide will grow significantly in the coming decades (1). Women hope to maintain a high level of health and quality of life in their postmenopausal years. Yet sexual complaints are among the most frequently reported health concerns by women attending menopause clinics (2). This indicates that menopausal status (and underlying hormonal change) may be linked to adverse effects on sexuality. There are a number of other possible explanations for deteriorating sexual functioning at this phase of life. These include length of relationship, chronologic aging, other physical health problems, loss of partner, partner's health and medication usage, and the many psychosocial stressors associated with midlife. TREATMENT OF THE POSTMENOPAUSAL WOMAN
271
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
272 detailed data. Most studies of the relationship between menopause and sexual functioning were carried out before the recent interest in refining classifications for sexual dysfunction (3). Important components of the classification systems for sexual dysfunction are the separation of desire, arousal, orgasmic, and pain disorders and the requirement that the person should express distress. The majority of scales developed to measure sexual functioning prior to the consensus conference (3) did not include any measure of distress. Other requirements for study of the menopausal transition include that the age of subjects at baseline should be young enough such that measures are obtained before major change occurs in the hypothalamic-pituitary-ovarian axis. Any use of hormone therapy (HT) and surgery that may compromise ovarian function, such as hysterectomy, should be documented. New definitions for the latter part of reproductive functioning, including menopause, were reached at the Staging of Reproductive Aging Conference in Park City, UT in 2001. Consensus was reached for a seven-stage model (4). This model consists of early, peak, and late reproductive phases; early and late menopausal transition phases; and early and late postmenopause phases. Features of early menopausal transition are the onset of variable cycle lengths with increasing follicle-stimulating hormone (FSH), whereas late menopausal transition is characterized by at least two skipped menstruations with elevated FSH and increased likelihood of vasomotor symptoms. The final menstrual period, which marks the transition to postmenopause, is recognized after 12 months of amenorrhoea. The early postmenopausal phase comprises the first 5 years after final menstrual period, and the late postmenopausal phase comprises the following years. Most epidemiologic studies have not directly measured hormonal status of the women surveyed but instead used menstrual status as a proxy for hormonal status. Also crucial to the inferences that can be drawn are the type of study (cross-sectional versus longitudinal) and statistical techniques that can unravel the complex interrelationships between outcomes and determinants and disentangle the effects of aging from those of menopause. Cross-sectional studies allow us to identify differences among age ranges, reproductive status groups, and ethnic groups. A major advantage of longitudinal studies is that of less reliance on retrospective data, providing that the recall period in question is kept short. Longitudinal studies of samples derived from the general population are in the best position to sort out whether a change in sexual function associated with the menopausal transition occurs, and if so, whether this reflects aging, health status, or hormonal or psychosocial factors. A major advantage of longitudinal studies is the ability to control for the effect of prior level of sexual function.
LORRAINE DENNERSTEIN
II. EFFECTS OF AGING Aging and length of the relationship are known to affect sexual functioning of both men and women, and these variables are often confounded. For examples, James (1983) (5) used cross-sectional and longitudinal data to show that coital rate halved over the first year of marriage and then took another 20 years to halve again. A number of investigators have reported that an additional decrement in aspects of sexual functioning occurs in mid-age. Pfeiffer and Davis (1972) (6), using cross-sectional data from the Duke University study, found a pattern of declining sexual activity in both men and women, but the decrement was larger for women than for men of the same age. The sharpest decline in sexual interest for women occurred around the mean of age of menopause (7). Often studies such as the Swedish studies Hallstrom (1977) (8) and Hallstrom and Samuelsson (1990) (9) and British studies (Hawton, Gath, and Day 1994) (10) also report declining sexual function with increasing age and increased sexual dysfunction with age. Osborn et al. (1988) (11) and Laumann et al. (1994) (12) report, in their U.S.based national study, increased sexual inactivity with age. In a multinational cross-sectional study carried out concurrently in Europe and the United States (WISHES), using validated measures of sexual function and sexual distress, declines in all aspects of sexual function were evident with age. Sexually related distress, however, decreased with age, so there was no increase in sexual dysfunction with age (13). In the Melbourne Women's Midlife Health Project longitudinal study, we used a detailed sexuality questionnaire, the Personal Experiences Q.uestionnaire, adapted from the McCoy Female Sexuality Q.uestionnaire (14) and subjected to a number of internal and external validation studies (15-18). The scale provides domain scores for frequency of sexual thoughts (Libido), arousal, enjoyment and orgasm (Sexual Responsivity), Frequency of Sexual Activities, and Dyspareunia. Feelings about Partner and Partner Problems are also measured. Using data collected annually for 6 years, we found the following factors significantly diminished (/5 < 0.001) over the time period studied: Feelings for Partner, Sexual Responsivity, Frequency of Sexual Activities, and Libido. The following factors significantly increased (/5 < 0.001) over the time period studied: Vaginal Dryness/ Dyspareunia and Partner Problems (19).
III. AGING VERSUS MENOPAUSAL STATUS A number of studies have found an additional decrement in aspects of sexual function in midlife coinciding with mean age of menopause or menopausal status (where measured).
CHAPTER 20 Sexuality: Clinical Implications from Epidemiologic Studies
Studies that failed to find effects of menopausal status have been limited by small sample sizes, wide age bands, and lack of validated measures. Very few studies have followed the same cohort prospectively across the menopausal transition. Longitudinal studies allow us to disentangle the effects of aging from those of menopausal hormonal changes (which are inevitably confounded) as well as to measure the powerful effects of psychosocial factors, including the individual's own prior level of sexual function and changes in sexual partnerships. The Melbourne Women's Midlife Health Project used the Personal Experiences Q.uestionnaire to calculate a total score of sexual function from the domains of Libido, Sexual Responsivity, and Frequency of Sexual Activities. Scores -<7 indicate low sexual function, similar to women with sexual dysfunction (17). From early to late menopausal transition, the percentage of women with scores of low sexual function rose from 42% to 88% (20). By the postmenopausal phase there was a significant decline in Sexual Responsivity, Frequency of Sexual Activities, Libido, and the total score and a significant increase in Dyspareunia and Partner Problems in sexual performance (Figs. 20.1 and 20.2). Increasing Dyspareunia and decreasing Libido and Responsivity correlated with decreasing estradiol but not with androgens (20).
273
Melbourne Women's Midlife Health Project (21), the most important factors (in decreasing order of importance) influencing libido, arousal, and orgasm (together termed Sexual Response) were prior level of Sexual Response, loss or gain of a sexual partner, feelings toward the partner, and then estradiol level (R2= 0.65) (Fig. 20.3). Relationship factors were less important for Dyspareunia, which was predicted by prior level of Dyspareunia and estradiol level (R2 =0.53). Frequency of Sexual Activities was not influenced by estradiol level but was predicted by prior level of Sexual Activities, change in partner status, feelings for partner, and level of Sexual Response (R2=0.52). The minimum effective estradiol level needed to increase Sexual Response by 10% (700 pmol/L estradiol) is twice that needed to decrease Dyspareunia. Endogenous testosterone (the free testosterone index) and dehydroepiandrosterone sulfate (DHEAS) were not related to sexual function domains in this sample of women passing through the natural menopausal transition. The only additional factor to affect sexual function domains was well-being, measured by the Affectometer2. Wellbeing was itself affected by lifestyle factors, stress, daily hassles, vasomotor symptoms, sleep, and self-rated health (22).
V. ETHNICITY IV. RELATIVE IMPORTANCE OF HORMONAL TO RELATIONAL AND OTHER PSYCHOSOCIAL FACTORS Further analyses assist us in understanding the relative impact of hormonal changes of the menopausal transition on women's sexuality. Using statistical techniques including structural equation modeling on 8 years of data from the
0,5 "
CHANGE IN SPEQ TOTAL SCORE
,0
.
.
.
.
.
.
.
.
.
.
The Melbourne sample consisted of Caucasian women. It is not clear whether there will be any ethnic variation in response to declining estradiol at menopause. Using cross-sectional data from baseline (prior to or early in the menopausal transition), the Study of Women's Health Across the Nation (SWAN) study reported substantial ethnic differences in sexual domains (23). After controlling for a wide range of variables, AfricanAmerican women reported higher frequency of sexual intercourse than Caucasian women; Hispanic-American women reported lower physical pleasure and arousal; and ChineseAmerican and Japanese-American women reported more pain and less desire and arousal than Caucasian women, although the only significant difference was in arousal (23). European studies have also reported substantial differences among countries in domains of sexual function, such as frequency of sexual intercourse, but nevertheless found a similar pattern of decline in the effects of menopause on desire, arousal, and orgasm across all countries (24).
-0,5
VI. SURGICAL MENOPAUSE -1.0 ..................................... I
!1
MENOPAUS~ PHASE
FIGURE 20.1 Changein total score on the Short Personal Experiences Questionnaire (SPEQ). Data are means;the bar indicates confidenceintervals. I, changefrom earlyto late menopausaltransition; II, changefrom late menopausal transition to postmenopause. (From ref. 20.)
The discussion thus far has focused on changes related to the natural menopausal transition. Surgical menopause, in which both ovaries are removed, would be predicted to have more deleterious effects on sexual function because all ovarian estrogen and androgen will be removed. Using a validated measure for hypoactive sexual desire disorder (HSDD)
274
LORRAINE DENNERSTEIN
FIGURE 20.2 Changes in domains on the Short Personal Experiences Q.uestionnaire (SPEQ). Data are means; the bar indicates confidence intervals. I, change from early to late menopausal transition; II, change from late menopausal transition to postmenopause. (From ref. 20.)
FIGURE 20.3 Optimized model of relationship between hormonal and relationship factors and sexual functioning. (From ref. 21.)
in women aged 18 to 70 years in four European countries, the WISHES study found that women who had undergone surgical menopause had significantly increased risk of H S D D compared with similarly aged women with intact reproductive organs. Younger women (age < 50 years) were the most affected (25).
VII. SUMMARYAND CLINICAL IMPLICATIONS There is a decline in all aspects of female sexual function with age. A further incremental decline in most aspects of sexual function occurs as women pass through the menopausal
transition. This further decline is related to falling estradiol levels. Other factors, such as prior sexual function and partnerrelated factors, have larger effects on women's sexual function than do hormonal factors. However, when relationship factors are stable, declining estradiol has noticeable effects. Not all women with low sexual function are distressed about it, and distress about sexual function declines with age. Women who have undergone surgical menopause with removal of ovaries will be more at risk of sexual dysfunction than those with intact reproductive organs. Thus, this analysis demonstrates that hormonal change is but one aspect of the many factors that affect sexual function. When a mid-aged woman reports sexual problems, the clinician should take a detailed history involving the
CHAPTER 20 Sexuality: Clinical Implications from Epidemiologic Studies w o m a n and her partner, as individuals and together. Given the range of factors affecting sexual functioning and the significantly more powerful effect of partner factors compared with that of h o r m o n a l factors, a broadly based biopsychosocial approach should be used. T h e clinician should specifically ask about b o t h e r s o m e symptoms known to be responsive to HT. T h e s e symptoms should be treated because they affect m o o d as well as cause distress in their own right. Consideration should be given to supplementation with estrogen if the w o m a n has other indicators of estrogen deficiency (hot flashes, genital atrophic changes, oophorectomy) and if, according to the woman, the deterioration in sexual function coincides with the menopausal transition, natural or induced. Bilaterally o o p h o r e c t o m i z e d w o m e n may benefit from the addition of testosterone to estrogen supplementation. W i t h each w o m a n , the clinician should discuss contraindications (absolute and relative) to the use of either h o r m o n e and risks and side effects of HTs. However, h o r m o n a l prescription alone is rarely enough. Particular attention should be given to the assessm e n t of the relationship with the partner, as well as other stressors in the woman's life. T h e midlife transition allows the opportunity for positive change in sexual relationships if couples increase their intimacy at this time. Finally, steroids such as estrogen, progestins, and androgens have subtle but important effects on female sexual functioning. These effects can be over-ridden by powerful psychosocial factors, such as a new relationship or past experiences which may influence sexual function. A n individual approach is needed to sort out the relative role of these factors for any particular w o m a n so that appropriate therapy can be planned.
References 1. Dennerstein L, Hayes R. Confronting the challenges: epidemiological study of female sexual dysfunction and the menopause. J Sex Med 1985;2:118-132. 2. Sarrel PM, Whitehead MI. Sex and menopause: defining the issues. Maturitas 1985;7:217-224. 3. Basson R, Berman J, Burnett A, et al. Report of the International Consensus Development Conference on Female Sexual Dysfunction: definitions and classifications.J Uro12000;163:888- 893. 4. Soules MR, Sherman S, Parrott E, et al. Executive summary: stages of reproductive aging workshop (STRAW). Fertil Steri12001;76:874- 878. 5. James WH. Decline in coital rates with spouses' ages and duration of marriage.JBiosoc &i 1983;15:83-87.
275
6. Pfeiffer E, Davis GC. Determinants of sexual behavior in middle and old age.JAm Geriatr Soc 1972;20:151-158. 7. Pfeiffer E, Verwoerdt A, Davis GC. Sexual behaviour in middle life. Am J Psychiatry 1972;128:1262-1267. 8. Hallstrom T. Sexuality in the climacteric. Clin Obstet Gynaeco11977;4: 227-239. 9. Hallstrom T, Samuelsson S. Changes in women's sexual desire in middle life: the longitudinal study of women in Gothenburg. Arch Sex Behav 1990;19:259-268. 10. Hawton K, Gath D, Day A. Sexual function in a community sample of middle-aged women with partners: effects of age, marital, socioeconomic, psychiatric, gynaecological, and menopausal factors. Arch Sex Behav 1994;23:375-395. 11. Osborn M, Hawton K, Gath D. Sexual dysfunction among middle aged women in the community. BMJ 1988;296:959-962. 12. Laumann EO, Gagnon RTM, Michaels S. The social organization of sexuality: sexual practices in the United States. Chicago: University of Chicago Press, 1994. 13. Hayes RD, Dennerstein L, Bennett C, et al. Low sexual desire, distress due to sexual desire, and aging. Proceedings of the ISSWSH Annual Meeting, Atlanta, GA, 2004, p. 36. 14. McCoy NL, Matyas JR. Oral contraceptives and sexuality in university women. Arch Sex Behav 1996;25:73-90. 15. Dennerstein L, Dudley EC, Hopper JL, Burger H. Sexuality, hormones and the menopausal transition. Maturitas 1997;26:83-93. 16. Dennerstein L, Lehert P, Dudley E. Short scale to measure female sexuality: adapted from McCoy Female Sexuality questionnaire. J Sex Marital Ther 2001;27:339- 351. 17. Dennerstein L, Anderson-Hunt M, Dudley E. Evaluation of a short scale to assess female sexual functioning. J Sex Marital Ther 2002a;28:389-397. 18. Dennerstein L, Lehert P. Modelling mid-aged women's sexual functioning: a prospective, population-based study.J Sex Marital Ther 2004a;30: 173-183. 19. Dennerstein L, Lehert P, Burger H, Garamszegi G, Dudley E. Menopause and sexual functioning. In: Studd J, ed. The management oft he menopause."millennium review. London: Parthenon, 1999:203-210. 20. Dennerstein L, Randolph J, Taffe J, Dudley E, Burger H. Hormones, mood, sexuality and the menopausal transition. Fertil Steri12002b;77: S42-848. 21. Dennerstein L, Lehert P, Burger H. The relative effects of hormones and relationship factors on sexual functioning of women through the natural menopausal transition. Fertil Steri12005a;84:174-180. 22. Dennerstein L, Lehert P, Burger H. Sexuality.Am J Med 2005b;118: 59-63. 23. AvisN, Zho MS, Johannes CB, et al. Correlatesof sexual function among multi-ethnic middle-aged women: results from the Study of Women's Health Across the Nation (SWAN). Menopause2005;12:385-398. 24. Dennerstein L, Lehert P. Women's sexual functioning, lifestyle, midage and menopause in 12 European countries. Menopause 2004b;11: 778-785. 25. Dennerstein L, Koochald P, Barton I, Graziottin A. Hypoactive sexual desire disorder in menopausal women: a survey of western European women. J Sex Med 2006;3:212-222.
This Page Intentionally Left Blank
SECTION V
Brain Function In the previous section (Chapter 14), Frederick Naftolin and colleagues outlined the important effects of sex steroids on the brain. Abundant steroid receptors and local brain production of steroids make it quite clear that important effects are mediated by estrogens, progestogens, and androgens. Since a woman's brain is exposed to fairly high concentration of sex steroids from the time of puberty, what happens after menopause? Although some local production of steroids still continues, there is a major decrease in this exposure. This section begins with an overview of effects of menopause and cognition. What actually happens? From a clinical vantage point, many women enter menopause with the concern that they will "lose their minds." Pauline M. Maki first describes what is known regarding menopause and cognition, and the influence of sex steroids. Jacobs and Sano next review the epidemiology of the major forms of dementia, Alzheimer's disease. Next, Victor W. Henderson reviews the literature on the effects of hormones on Alzheimer's disease. Throughout these discussions relating to cognition, the various authors have pointed out that the time of initiation of hormonal therapy may be critical in determining the risk of developing Alzheimer's disease. Finally in this section, David R. Rubinow examines the issue of depression and the influence of estrogen. Once again, the issue of treating women in the perimenopause and later is discussed.
This Page Intentionally Left Blank
;HAPTER 2:
Menopause and Cognition PAULINE M. M A K I
Departments of Psychiatry and Psychology, Center for Cognitive Medicine, University of Illinois at Chicago, Chicago IL 60612
Most women expect to experience some degree of cognitive decline as they age. However, the frequency of cognitive complaints among women reporting to menopausal clinics raises questions about whether these complaints worsen during the menopausal transition. Memory lapses and wordfinding difficulty can be sources of concern, particularly when they interfere with daily life. Cognitive abilities including concentration, memory, planning, and strategizing are important not only for effective functioning in everyday activities, but also for optimal and consistent performance in the workplace at a time when many women are at the peak of their careers. This chapter presents a brief review of the effects of the menopausal transition, menopausal symptoms, and menopausal hormone therapy (HT) on cognitive test performance and the brain systems subserving cognitive functions.
monal changes or menopausal stage but instead related the complaints to stress, health problems, and age. Notably, a follow-up analysis found that complaints did vary by menopausal stage and were more frequent in the early and middle transitional stages than in the late transition and postmenopausal stages (3). In the Study of Women's Health Across the Nation (SWAN), 44% of early and late perimenopausal women and 41% of postmenopausal women endorsed "forgetfulness" on a menopausal symptom inventory, compared with 31% of premenopausal women (4). Complaints of forgetfulness were related to age > 44 years, less than a high school education, lack of full-time employment, and low physical actMty. This rate is similar to that observed in a telephone survey of women aged 45 to 54 years residing in Durham County, North Carolina (n = 581), in which 44% reported memory complaints (5). That study found memory complaints to be a significant predictor of depressive symptoms. In general, the studies suggest that memory complaints among midlife women are common--affecting about 40% to 67% of womenmand are related to age, physical well-being, and psychologic well-being. Several studies provide objective support for the perception that age influences cognitive abilities at midlife. A cross-sectional study revealed a stronger relationship between age and reaction time on speeded cognitive tests in 33 postmenopausal women (mean age 52 years) compared with 24 younger women of reproductive age (mean age 34 years) (6). That finding suggests that menopause might accelerate declines processing speed. Similar evidence comes from a study examining the correlation between age and verbal memory in young (ages 16 to 47 years) and older (ages 55 to 89 years) men and women (7). A negative correlation be-
I. M E M O R Y A N D T H E MENOPAUSAL TRANSITION Memory complaints are common among midlife women reporting to menopause clinics and those in the general population. For example, 67% of women in a largely Hispanic clinical population reported memory complaints (1). This rate is similar to reports from population-based studies. For example, in the Seattle Midlife Women's Health Study, 62% of participants reported declines in memory function from an earlier period in life (2). Specific complaints included difficulty recalling words or numbers, forgetting events and actions, and difficulty concentrating. Women generally did not attribute these declines in memory to horTREATMENT OF THE POSTMENOPAUSAL W O M A N
279
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
280 tween age and memory performance was observed in younger and older men and older women, but not in younger women. These data suggest that for women, age begins to influence verbal memory sometime in midlife, whereas for men, age influences performance earlier in life. Biologic plausibility for these effects comes from neuroimaging data examining effects of age on the hippocampus, a brain area that subserves memory performance (7a). Women showed greater hippocampal glucose metabolism compared with men early in life, but not later in life. Although there is some evidence of an objective increase in the rate of cognitive and brain aging at midlife, there is little evidence that this is specifically related to menopause. No effect of menopausal stage was observed on a word-list memory test in a cross-sectional analysis of 326 women in the Melbourne Midlife Women's Health Study (8). Similar results were obtained in a longitudinal analysis of two cognitive measures, working memory and psychomotor speed, in 868 women from the Chicago SWAN site (9). To date, the only published study to report an effect of menopausal stage on cognitive function found declines in performance on a test of psychomotor speed, attention, and mental flexibility (i.e., Trails B) among postmenopausal women compared with premenopausal and perimenopausal women, even after adjusting for age and education (10). This study was based on a cross-sectional study of a large sample of 1270 Chinese women who completed a comprehensive neuropsychologic test battery. The findings suggest that postmenopausal women might experience some declines in tasks mediated by the frontal lobes. SWAN investigated the effects of menopausal stage on a comprehensive cognitive test battery given over several years. Forthcoming data from that study will provide unique insights into changes in memory and other cognitive abilities across the menopausal transition.
II. E S T R O G E N A N D OBJECTIVE M E A S U R E S O F M E M O RY A recent review of randomized clinical trials of H T and cognitive function suggested that the best evidence in support of a favorable effect comes from studies examining estrogen therapy (ET) in women younger than 65 (11). Seven of nine trials in younger women found an advantage for ET on at least one cognitive test (12). Most of those trials reported improvements on tests of memory (12-15), particularly verbal memory, and attention (16,17). The variability across treatment studies in terms of treatment duration (3 days to 6 months), time since menopause (immediately following surgical menopause to 8 years), and history of prior ET use (0% to 68%) suggests that these effects may be robust. However, the studies included only small samples (12 to 60 women) and frequently enrolled women with moderate menopausal symptoms. The two randomized trials that did not enroll women with menopausal complaints failed to find
PAULINEM. MAKI benefit with ET. One trial included all asymptomatic women and only assessed attention (18). The other included a sample in which 68% had previously used ET and 48% had undergone surgical menopause an average of 8 years previously (19). Insights into the effects of combined estrogen plus progesterone treatment (EPT) in younger women is surprisingly limited to one randomized trial, which involved only 49 women and showed improved verbal memory with ET (estradiol valerate) and improved numerical memory with EPT (ET plus dinogest) (20,21). In summary, there is little evidence of a detrimental effect of ET on cognition in younger women and more evidence of a beneficial effect on memory and attention, particularly in symptomatic women. The lack of studies of EPT and cognition is particularly troublesome because many naturally menopausal women with intact uteri who seek therapies for menopausal symptoms will use a progestogen. Randomized clinical trial data from younger women contrast with data from older women and suggest that timing of initiation may be an important consideration in determining hormone effects on cognition. Women initiating conjugated equine estrogens plus medroxyprogesterone acetate (CEE/ MPA) later in life are at risk for cognitive declines, particularly on measures of verbal memory and global cognitive status. These effects may be most evident among those with poor cardiovascular or cognitive status at baseline. For example, although six of six trials in younger women showed beneficial effects of ET on verbal memory (12-15,20,22), the two trials of EPT in older women suggested negative, if any, effects (23,24). One study, a large trial of CEE/MPA in older women with coronary artery disease (23), reported a trend for a negative effect of treatment on verbal memory and a significant negative, but clinically insignificant, effect on verbal fluency. None of the smaller randomized trials of EPT in older women reported an effect of treatment on verbal memory or other cognitive domains (24,25). A study of ET in women older than 65 years also reported no effect of ET on cognitive function, although high estradiol levels posttreatment were associated with improvements in verbal memory (26). In contrast to the studies on verbal memory, one small clinical trial of ET in older women suggested beneficial effects on memory for pictures, memory for spatial locations, and mental rotations (27), outcomes not included in the larger studies. Observational studies suggest that H T protects against the typical longitudinal declines in figural memory that occur with aging (28,29). Overall, these studies raise further questions about the importance of the particular cognitive domain under investigation, the particular formulation of estrogen under study, and the role of progesterone in predicting cognitive effects. Data from the Women's Health Initiative Memory Study (WHIMS), an ancillary study to the Women's Health Initiative (WHI), revealed negative effects of CEE alone and CEE/MPA in older women, although the effects were limited to women who performed poorly at baseline. W H I M S
CHAPTER 21 Menopause and Cognition involved more than 7000 women age 65 and older and provided cognitive data only from a brief measure of cognitive stares, the Modified Mini-Mental State Exam (3MSE). The first report on this data was included in a quality of life study and indicated no effect of CEE/MPA on 3MSE scores (30). In contrast, a later report from W H I M S reported that both CEE/MPA and CEE alone were associated with decreased cognitive stares (>2 standard deviations below the mean) over 4- and 5-year follow-ups, respectively (31). Notably, this negative effect was observed only in women who scored below the normal range at baseline. W H I M s included more comprehensive cognitive evaluations only in women showing cognitive impairment during dementia work-ups. The W H I Study of Cognitive Aging (WHISCA) was initiated to determine whether H T influenced the rate of cognitive aging in W H I volunteers who did not show cognitive impairment at baseline. W H I S C A included a comprehensive neuropsychologic assessment, including measures of verbal and figural memory. Those data, which are currently under review, will lend insights into hormone effects on more specific cognitive domains. One important clinical issue is whether H T has the same effect on cognition in older, long-term users as in older women who recently initiated HT. Although it is uncommon, particularly after the WHI, for women over age 65 to initiate HT, it is important to recognize the absence of support for an overall cognitive benefit of H T in this group and evidence of negative effects on certain outcomes. No long-term randomized clinical trials of H T have addressed this question. However, a recent study reported data on current cognitive function in a sample of 343 women who were randomized to received H T or placebo in clinical trials 5, 11, or 15 years earlier (32). Women randomized to H T for 2 to 3 years showed a decreased risk of cognitive impairment compared with those randomized to receive placebo. There was also a trend toward a decreased risk of cognitive impairment among those who received long-term HT. This suggests that longterm H T might not be detrimental to cognitive function if treatment is initiated early in the menopause. That hypothesis has also been tested in studies of Alzheimer's disease and is supported by longitudinal data (33). However, in the absence of firm data to the contrary, it seems prudent to caution older women about the W H I M S finding and the known risks of H T on cognitive decline and dementia.
III. BIOLOGIC PLAUSIBILITY FOR ESTROGEN EFFECTS ON THE BRAIN: NEUROIMAGING STUDIES Neuroimaging studies allow for the in vivo examination of hormone effects on brain areas that subserve particular cognitive functions. Current methods provide indirect measures of neural activity by using changes in blood flow or
281 glucose metabolism to estimate changes in neural activity. Two neuroimaging techniques are commonly used to study the neural targets of HT. Functional magnetic resonance imaging (fMRI) is a noninvasive neuroimaging method that localizes brain areas showing changes in the level of blood oxygenation. This method is known as blood oxygenation level-dependent (BOLD) contrast. Positron emission tomography (PET) identifies brain areas showing changes in neural activity with a radioactive tracer, which is typically attached to a water or glucose molecule to measure blood flow and glucose metabolism, respectively. These techniques lend unique insights into H T and cognition because they offer evidence of biologic plausibility in humans. Images can be acquired during cognitive tests to provide insights into behavioral changes with clinical relevance. In light of the data suggesting possible beneficial effects of H T on objective tests of memory, neuroimaging studies of verbal and figural memory lend important insights into the biologic plausibility of hormone effects on brain regions subserving memory. Two studies examined patterns of brain activity with PET during performance of memory tasks that are similar to standardized neuropsychological tests (34,35). Both studies drew on samples of elderly, long-term H T users who initiated treatment at the time of the menopausal transition and continued to use HT, and both studies used groups of well-matched controls. The first of those studies was a cross-sectional comparison of H T users and nonusers (34). Results showed differences in patterns of brain activation during the memory tasks in regions that have been implicated in memory processing in human and animal studies. Specifically, H T altered blood flow in the right inferior parietal region and right parahippocampal gyrus during the figural memory task and in the right parahippocampal gyrus, precuneus, inferior frontal cortex, and dorsal frontal gyms during the verbal memory task. Importantly, H T users in this neuroimaging study showed superior memory on standardized neuropsychological tests of verbal and figural memory. Those behavioral findings corroborated findings from larger-scale studies involving the same instruments (28,36). The localization of hormone effects to the hippocampus and frontal lobes is important; changes in the function of synapses in the frontal lobes underlie many of the cognitive changes observed in normal aging, whereas changes in the hippocampus underlie changes in memory in preclinical and clinical dementia (37). The human imaging data also parallel findings from animal models. A recent study showed evidence of estrogen-induced memory enhancement (38) and formation of spines and synapses in frontal lobes in ovariectomized rhesus monkeys (39). Similarly, estrogeninduced changes in hippocampal morphology and connectivity have been shown to relate to memory enhancements in ovariectomized rhesus monkeys (40). A follow-up longitudinal study focused on PET measures of brain aging, specifically changes in regional blood
282
PAULINEM. MAKI
FIGURE 21.1 HT users show significant increases in hippocampal blood flow over a 2year interval. (Reprinted from Neurobiologyof Aging, Volume 21, Maki PM, Resnick SM. Longitudinal effects of estrogen replacement therapy on PET cerebral blood flow and cognition, 373-383, 9 2000 with permission from Excerpta Medica, Inc.)
FICURE 21.2 Estrogen modulates brain activation in frontal lobes in postmenopausal women. (From JAMA, April 7, 1999, Page 1199, Copyright 9 1999 American Medical Association. All rights reserved.)
flow over time, in HT users and nonusers (35) (Fig. 21.1) (see color insert). Results indicated significant differences between the two groups in a number of brain regions, with most changes indicating areas of increased blood flow over time in women receiving HT. The largest of these differences were in the right hippocampus, entorhinal cortex, parahippocampal gyrus, and left middle temporal gyms. These regions show the opposite effect (i.e., decreased blood flow [or metabolism]) in individuals at increased risk for Alzheimer's disease (41-43), suggesting potential protective effects among early initiators of HT. The lack of randomized trial data to address the effects of early initiation and continued use of HT on brain function suggests a need for caution in clinical application of these findings. Randomized clinical trial data from a short-term, doubleblind, placebo-controlled crossover study of ET (1.25 mg CEE/day for 21 days) also suggested significant effects of ET on brain function. This study examined activation patterns during figural and verbal working memory tests in 46 postmenopausal women (mean age 51 years) using fMRI (44). During storage of verbal material, ET increased activa-
tion in anterior frontal regions and the inferior parietal lobule bilaterally and decreased activation in the inferior parietal lobule and right superior temporal gyms. During storage of nonverbal material, ET decreased activation in the inferior parietal lobule. Across the two tests combined, ET led to an exaggeration of a commonly observed pattern of brain activation across encoding and retrieval tests. Although performance on the tests did not change with treatment, a subsequent report on neuropsychologic test performance in this study group found improved verbal memory and oral reading with treatment (22). This investigation is consistent with previous reports of a female advantage (45) and a beneficial effect of HT (46) on working memory tasks mediated by the frontal lobes. The frontal lobes subserve a number of other cognitive tasks, including "executive functions" (e.g., planning, strategizing, mental flexibility, inhibition), but there is less support for hormone effects on those tasks (47). Thus, animal and human studies demonstrate that HT alters the function of frontal areas that change with age and affects some, but not all, frontally mediated tasks (Fig. 21.2) (see color insert).
CHAPTER21 Menopause and Cognition
IV. SELECTIVE ESTROGEN RECEPTOR MODULATORS A N D COGNITIVE F U N C T I O N The Multiple Outcomes of Raloxifene Evaluation (MORE) study was the largest randomized clinical trial to date of any estrogenic agent on cognitive function and supported suggestions from H T studies of a potential beneficial effect on verbal memory. Specifically, raloxifene (60 or 120 mg daily) was associated with a significant improvement in verbal memory scores and prevented abnormal age-related declines in verbal memory in elderly women age 70 and older (48). A follow-up study in this cohort examined the effects of raloxifene on mild cognitive impairment, a preclinical form of dementia, and found a decreased risk with the 120mg dose only (49). Whether raloxifene is an agonist or antagonist in the hippocampus of human females is unknown; however, fMRI studies show a change in activation in the parahippocampal gyrus during encoding with raloxifene (50). This targeting of the parahippocampal gyrus is similar to that observed with H T (34,35). The Study of Tamoxifen and Raloxifene (STAR), a randomized clinical trial comparing raloxis and tamoxifen on the risk for breast cancer, includes a cognitive substudy to compare the effects of the two selective estrogen receptor modulators (SERMs) on measures of memory that were shown in MORE and studies of l i T to be sensitive to the effects of estrogen. Breast cancer patients treated with antiestrogens demonstrate significant reductions in verbal memory (51). Insights into the cognitive-sparing versus cognitive-impairing effects of SERMs are an important issue for women who take SERMS for the prevention and treatment of breast cancer.
V. INDIRECT EFFECTS OF ESTROGEN ON COGNITIVE FUNCTION H T might affect cognitive change indirectly by improving menopausal symptoms that interfere with cognitive function. However, surprisingly few studies have shown that symptom amelioration is related to cognitive enhancement. In the MORE trial, raloxifene, which induces hot flushes, improved cognition, and hot flush severity and frequency were unrelated to cognitive test performance (48). The opposite is also the case: treatments that have no effect on hot flushes have been shown to improve cognition. Clinical trials with phytoestrogens, for example, report improvements in memory and executive function, with no improvements in hot flushes (52,53). It is likely that indirect effects are most
283 pronounced in symptomatic women, with greater indirect effects during the early menopause compared with later in life (54). H T improves subjective and objective measures of sleep quality (55,56), ameliorates hot flushes (57), and improves mood in certain women (58). Notably, the studies in younger women that showed beneficial effects of ET on verbal memory were primarily in samples of symptomatic women, with two negative studies involving samples of asymptomatic women. Ongoing studies are investigating whether therapies that target these symptoms, even those without appreciable estrogenic action (e.g., antidepressants and black cohosh), have indirect benefits on cognition.
VI. SUMMARY Cognitive complaints are common in midlife women. Our understanding of objective changes in cognitive function across the menopausal transition is limited by the lack of longitudinal studies, although cross-sectional studies suggest no effect. There is support from randomized clinical trials in younger, mostly symptomatic women of a beneficial effect of ET on verbal memory and attention, but the sample sizes in these trials were small. There are little data on the effects of estrogen in combination with progesterone on cognitive function, and the findings with ET alone may not generalize to EPT. Older women, particularly those with cardiovascular risk factors and cognitive dysfunction, experience a decline in verbal memory and cognitive status with EPT. Even in older women, H T may have beneficial effects on other tasks of memory, including figural memory and possibly working memory. Raloxifene provides cognitive benefits even in older women. Neuroimaging studies support the biologic plausibility of beneficial effects of H T and raloxifene on brain regions subserving memory functioning, including the hippocampus and frontal lobes. There is a pressing need for studies that examine cognitive effects of different SERMs, different formulations of estrogen and progesterone, and different regimens (e.g., cyclic versus continuous) to address clinically important issues in the treatment of menopausal symptoms. It is important to keep in mind that H T is not approved for the treatment of cognitive complaints or dysfunction in any patient group. This is justified in light of the welldemonstrated negative effects of H T on cognition in elderly women and the lack of large-scale studies supporting the use of H T for cognition in younger women. It is clinically prudent to caution women who initiate H T for the treatment of menopausal symptoms about the risk of dementia and cognitive decline observed in older women in W H I M S (59), while recognizing that the absolute risk of dementia in younger women is very low.
284
PAULINE M. MAKI
References 1. Schnatz PF, Banever AE, Greene J F, O'Sullivan DM. Pilot study of menopause symptoms in a clinic population. Menopause 2005;12: 623-629. 2. Mitchell ES, Woods NE Symptom experiences of midlife women: observations from the Seattle Midlife Women's Health Study. Maturitas 1996;25:1-10. 3. Woods NF, Mitchell ES, Adams C. Memory functioning among midlife women: observations from the Seattle Midlife Women's Health Study. Menopause 2000;7:257-265. 4. Gold EB, Sternfeld B, Kelsey JL, et al. Relation of demographic and lifestyle factors to symptoms in a multi-racial/ethnic population of women 40-55 years of age. Am JEpidemio12000;152:463-473. 5. Bosworth HB, Bastian LA, Kuchibhatla MN, et al. Depressive symptoms, menopausal status, and climacteric symptoms in women at midlife. Psychosom Med 2001;63:603-608. 6. Halbreich U, Lumley LA, Palter S, et al. Possible acceleration of age effects on cognition following menopause. J Psychiatr Res 1995;29: 153-163. 7. Kramer J, Yaffe K, Lengenfelder J, Delis D. Age and gender interactions on verbal memory performance. J Int Neuropsychol Soc 2003;9: 97-102. 7a. Murphy DG, DeCarli C, McIntosh AR, et al. Sex differences in human brain morphometry and metabolism: an in vivo quantitative magnetic resonance imaging and positron emission tomography study on the effect of aging. Arch Gen Psychiatry 1996;53:585-594. 8. Henderson V, Guthrie J, Dudley E, Burger H, Dennerstein L. Estrogen exposures and memory at midlife: A population-based study of women. Neurology 2003;60:1369-1371. 9. Meyer PM, Powell LH, Wilson RS, et al. A population-based longitudinal study of cognitive functioning in the menopausal transition. Neurology 2003;61:801-806. 10. Fuh JL, Wang SJ, Lu SR, Juang KD, Lee SJ. Alterations in cognitive function during the menopausal transition. JAm Geriatr Soc 2003;51: 431-432. 11. Maki PM. A systematic review of clinical trials of hormone therapy on cognitive function: effects of age at initiation and progestin use. Ann N YAcad Sci 2005;1052:182-197. 12. Krug R, MoUe M, Dodt C, Fehm HL, Born J. Acute influences of estrogen and testosterone on divergent and convergent thinking in postmenopausal women. Neuropsychopharmacology 2003;28:1538-1545. 13. Hackman BW, Galbraith D. Replacement therapy and piperazine oestrone sulphate ("Harmogen") and its effect on memory. Curr Med Res Olin 1976;4:303-306. 14. Sherwin BB. Estrogen and/or androgen replacement therapy and cognitive functioning in surgically menopausal women. Psychoneuroendocrinology 1988;13:345-357. 15. Phillips SM, Sherwin BB. Effects of estrogen on memory function in surgically menopausal women. Psychoneuroendocrinology 1992;17: 485-495. 16. Vanhulle G, Demol R. A double-blind study into the influence of estriol on a number of psychological tests in post-menopausal women. In: van Keep PA, Greenblatt M, Albeaux-Fernet M, eds. Consensus on the menopause research. London, MTP Press, 1976:94-99. 17. Fedor-Freybergh P. The influence of oestrogens on the wellbeing and mental performance in climacteric and postmenopausal women. Acta Obstetrica Gynecologica Scandinavica 1977;64:1-91. 18. Ditkoff EC, Crary WG, Cristo M, Lobo RA. Estrogen improves psychological function in asymptomatic postmenopausal women. Obstet Gyneco11991;78:991-995. 19. Polo-Kantola P, Portin R, Polo O, et al. The effect of short-term estrogen replacement therapy on cognition: a randomized, double-blind, crossover trial in postmenopausal women. Obstet Gyneco11998;91:459-466.
20. Linzmayer L, Semlitsch HV, Salem B, et al. Double-blind, placebocontrolled psychometric studies on the effects of a combined estrogenprogestin regimen versus estrogen alone on performance, mood and personality of menopausal syndrome patients. Arzneimittelforschung 2001;51:238-245. 21. Saletu B, Anderer P, Gruber D, et al. Hormone replacement therapy and vigilance: double-blind, placebo-controlled EEG-mapping studies with an estrogen-progestogen combination (Climodien, Lafamme) versus estrogen alone in menopausal syndrome patients. Maturitas 2002;43:165-181. 22. Shaywitz SE, Naftolin F, Zelterman D, et al. Better oral reading and short-term memory in midlife, postmenopausal women taking estrogen. Menopause 2003;10:420-426. 23. Grady D, Yaffe K, KristofM, et al. Effect ofpostmenopausal hormone therapy on cognitive function: the Heart and Estrogen/Progestin Replacement Study. Am J Med 2002;113:543-548. 24. Binder EF, Schechtman KB, Birge SJ, Williams DB, Kohrt WM. Effects of hormone replacement therapy on cognitive performance in elderly women. Maturitas 2001;38:137-146. 25. Goebel JA, Birge SJ, Price SC, Hanson JM, Fishel DG. Estrogen replacement therapy and postural stability in the elderly. Am J Otol 1995;16:470-474. 26. Wolf OT, Kudielka BM, Hellhammer DH, et al. Two weeks of transdermal estradiol treatment in postmenopausal elderly women and its effect on memory and mood: verbal memory changes are associated with the treatment induced estradiol levels. Psychoneuroendocrinology 1999;24:727-741. 27. Duka T, Tasker R, McGowan JE The effects of 3-week estrogen hormone replacement on cognition in elderly healthy females. Psychopharmacology (Berl) 2000;149:129-139. 28. Resnick SM, Metter EJ, Zonderman AB. Estrogen replacement therapy and longitudinal decline in visual memory: a possible protective effect? Neurology 1997;49:1491-1497. 29. Smith YR, Giordani B, Lajiness-O'Neill R, Zubieta JK. Long-term estrogen replacement is associated with improved nonverbal memory and attentional measures in postmenopausal women. Fertil Steril 2001;76:1101-1107. 30. Hays J, Ockene J, Brunner R, et al. Effects of estrogen plus progestin on health-related quality of life. N EnglJ Med 2003;348:1839-1854. 31. Espeland M, Rapp S, Shumaker S, et al. Conjugated equine estrogens and global cognitive function in postmenopausal women: Women's Health Initiative Memory Study. JAMW 2004;291:2959-2968. 32. Bagger YZ, Tanko LB, Alexandersen P, Qin G, Christiansen C. Early postmenopausal hormone therapy may prevent cognitive impairment later in life. Menopause 2005;12:12-17. 33. Zandi PP, Carlson MC, Plassman BL, et al. Hormone replacement therapy and incidence of Alzheimer's disease in older women: the Cache County Study. JAMA 2002;288:2123-2129. 34. Resnick SM, Maki PM, Golski S, Kraut MA, Zonderman AB. Estrogen effects on PET cerebral blood flow and neuropsychological performance. Horm Behav 1998;34:171-184. 35. Maki PM, Resnick SM. Longitudinal effects of estrogen replacement therapy on PET cerebral blood flow and cognition. Neurobiol Aging 2000;21:373-383. 36. Maki P, Zonderman A, Resnick S. Enhanced verbal memory in nondemented elderly women receiving hormone-replacement therapy. Am J Psychiatry 2001;158:227-233. 37. Morrison JH, Hof PR. Life and death of neurons in the aging brain. Science 1997;278:412-419. 38. Rapp PR, Morrison JH, Roberts JA. Cyclic estrogen replacement improves cognitive function in aged ovariectomized rhesus monkeys. J Neurosci 2003;23:5708-5714. 39. Tang Y, Janssen WG, Hao J, et al. Estrogen replacement increases spinophilin-immunoreactive spine number in the prefrontal cortex of female rhesus monkeys. Cereb Cortex 2004;14:215-223.
CHAPTER 21 Menopause and Cognition 40. Adams MM, Morrison JH. Estrogen and the aging hippocampal synapse. Cereb Cortex 2003;13:1271-1275. 41. Johnson KA, Jones K, Holman BL, et al. Preclinical prediction of Alzheimer's disease using SPECT. Neurology 1998;50:1563-1571. 42. Kennedy AM, Frackowiak RS, Newman SK, et al. Deficits in cerebral glucose metabolism demonstrated by positron emission tomography in individuals at risk of familial Alzheimer's disease. Neurosci Lett 1995;186:17-20. 43. Reiman EM, Caselli RJ, Yun LS, et al. Preclinical evidence of Alzheimer's disease in persons homozygous for the epsilon 4 allele for apolipoprotein E. N EnglJ Med 1996;334:752-758. 44. Shaywitz SE, Shaywitz BA, Pugh KR, et al. Effect of estrogen on brain activation patterns in postmenopausal women during working memory tasks. JAMA 1999;281:1197-1202. 45. Duff SJ, Hampson E. A sex difference on a novel spatial working memory task in humans. Brain Cogn 2001;47:470-493. 46. Duff SJ, Hampson E. A beneficial effect of estrogen on working memory in postmenopausal women taking hormone replacement therapy. Hormones Behavior 2000;38:262-276. 47. Lacreuse A, Wilson ME, Herndon JG. Estradiol, but not raloxifene, improves aspects of spatial working memory in aged ovariectomized rhesus monkeys. NeurobiolAging 2002;23:589-600. 48. Yaffe K, Krueger K, Sarkar S, et al. Cognitive function in postmenopausal women treated with raloxifene. N Engl J Med 2001;344: 1207-1213. 49. Yaffe K, Krueger K, Cummings SR, et al. Effect of raloxifene on prevention of dementia and cognitive impairment in older women: the Multiple Outcomes of Raloxifene Evaluation (MORE) randomized trial. Am J Psychiatry 2005;162:683-690. 50. Simone JM, Neele SJ, Rombouts SA, et al. Raloxifene affects brain activation patterns in postmenopausal women during visual encoding. J Clin Endocrinol Metab 2001;86:1422-1424.
285 51. Shilling V, Jenkins V, Fallowfield L, Howell T. The effects of hormone therapy on cognition in breast cancer. J Steroid Biochem Mol Biol 2003;86:405-412. 52. Duffy R, Wiseman H, File SE. Improved cognitive function in postmenopausal women after 12 weeks of consumption of a soya extract containing isoflavones. Pharmacol Biochem Behav 2003;75:721-729. 53. Woo J, Lau E, Ho SC, et al. Comparison on pueraria lobata with hormone replacement therapy in treating the adverse health consequences of menopause. Menopause 2003;10:352-361. 54. Maki PM. Methodological pitfalls in the study of estrogen on cognition and brain function. In: Genzannni A, ed. Hormone replacement therapy and neurologicalfunction. Boca Raton, FL: CRC Press, 2003. 55. Polo-Kantola P, Erkkola R, Helenius H, Irjala K, Polo O. When does estrogen replacement therapy improve sleep quality? AmJ Obstet Gynecol 1998;178:1002-1009. 56. Antonijevic IA, Stalla GK, Steiger A. Modulation of the sleep electroencephalogram by estrogen replacement in postmenopausal women. Am J Obstet Gyneco12002;182:277-282. 57. Utian WH, Shoupe D, Bachmann G, Pinkerton JV, Pickar JH. Relief of vasomotor symptoms and vaginal atrophy with lower doses of conjugated equine estrogens and medroxyprogesterone acetate. Fertil Steril 2001;75:1065-1079. 58. Soares CN, Almeida OP, Joffe H, Cohen LS. Efficacy of estradiol for the treatment of depressive disorders in perimenopausal women: a double-blind, randomized, placebo-controlled trial. Arch Gen Psychiatry 2001;58:529-534. 59. Shumaker S, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women's Health Initiative Memory Study. JAMA 2004;291:2947-2958.
This Page Intentionally Left Blank
2 H A P T E R 21
9
9
9
C o gnltlVe H e alth in th e Postmenopausal W o m a n DIANE
M. JACOBS
Consultant, San Diego, CA; Department of Psychiatry, Mount Sinai School of Medicine, New York, NY 10468
M A R Y S A N O Alzheimer's Disease Research Center, Department of Psychiatry, Mount Sinai School of Medicine, James J. Peters VA Medical Center, Bronx, NY 10468
Women in the United States and other western societies now live an average of 80 or more years, a consequence of which is that most will experience menopause and live for an additional 30 years beyond this event. As the "baby boom" generation approaches menopause and the population continues to age, health-related consequences of menopause will have an increasingly greater public health impact. Maintaining cognitive health and preventing Alzheimer's disease (AD) and other forms of dementia are important medical concerns of postmenopausal women and should be important clinical goals of their health care providers. Intact cognition is essential for optimal social and occupational functioning, especially in light of the cognitive demands of our increasingly technological society. Cognitive complaints are common among menopausal and postmenopausal women, although objective evidence of menopause-related cognitive dysfunction is limited. This chapter reviews the current state of knowledge on the effects of menopause and hormone replacement on cognition in postmenopausal women. We review the evidence for a role of menopause and hormone replacement therapy (HRT) on dementia and cognitive functioning in older as well as younger postmenopausal women.
ter surgical menopause may have a direct effect on cognitive functioning (1). Three months postoperatively, test scores of women undergoing oophorectomy declined significantly on measures of attention, abstract reasoning, clerical speed and accuracy, and immediate memory for prose passages. Women who had a hysterectomy but whose ovaries were retained showed no such change in postoperative cognitive performance. Further, scores of women post-oophorectomy who received postoperative HRT also were unchanged. Similarly, in a later report, Phillips and Sherwin (2) demonstrated a significant post-oophorectomy decline on a measure of verbal paired word associate learning in women receiving a placebo, but no such decline was observed among women receiving hormone replacement. Pharmacologic suppression of ovarian function with gonadotropin-releasing hormone (GnRH) agonists also resulted in decreased memory ability; however, this effect was reversed by coincident administration of estradiol (3). The evidence for an effect of natural menopause on cognition has been less compelling than the results from studies following surgical menopause. Although cognitive complaints are common among women during and after the menopausal transition, with more than 40% of women between the ages of 48 and 55 complaining of forgetfulness (4), neuropsychologic investigations frequently have failed to find evidence of an objective deficit related to the menopause. One recent investigation, a birth cohort study of 1261 women, did find weak evidence of an adverse effect of natural menopause on cognitive function, but this effect was
I. MENOPAUSE, ESTROGEN, AND COGNITION Early evidence for a link between menopause and cognition came from studies of women undergoing surgical menopause. Findings of Sherwin and colleagues suggested that the dramatic changes in levels of ovarian hormones afTREATMENT OF THE POSTMENOPAUSAL WOMAN
This work was supported by Federal Grants AG15922.
287
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
288
largely explained by premenopausal cognitive functioning (5). Cognition in this cohort was not associated with vasomotor or psychologic symptoms, and there was no evidence of any effects of HRT use. The authors concluded that common environmental or genetic factors may influence both the timing of natural menopause and lifetime cognitive function through long-term or lifelong hormonal mechanisms. Cross-sectional studies, however, have failed to find an association between reproductive period, as a surrogate measure of endogenous estrogen exposure, and cognitive performance (6). Epidemiologic studies and small randomized clinical trials have suggested that H R T may improve cognitive functioning in postmenopausal women, especially women with menopausal symptoms (see LeBlanc 2001 for a systematic review and meta-analysis [7]). In addition, observational studies have suggested a role for HRT in the prevention of AD. Recent findings from the Women's Health Initiative (WHI), however, suggest that--at least in women age 65 or older--HRT actually increases the risk of dementia (8,9), stroke (10,11) and may have a deleterious effect on cognition (12,13). Although these findings cast into doubt the putative beneficial effects of HRT on cognition, questions remain regarding the cognitive efficacy of different estrogen preparations, the role of progesterone, and impact of age, menopausal symptoms, and the timing and duration of therapy on cognitive outcomes.
II. D E M E N T I A A N D COGNITIVE F U N C T I O N I N G IN W O M E N AGED 65 A N D OLDER Dementia, defined by profound loss of memory and other cognitive functions sufficient to interfere with social and occupational functioning, is a serious and growing medical problem. AD is the most common cause of dementia, followed by cerebrovascular disease/stroke. The prevalence and incidence of AD, as well as its cost to society, increases exponentially with age. It has been estimated that starting at age 65, the prevalence of AD doubles with every 5 years of age, with prevalence estimates reaching 20% in those over age of 90 (14). While there is little evidence of sex differences in age-specific rates of dementia, the prevalence of AD in women is twice that of men, in part due to their greater longevity (15). It has been recognized that even prior to the diagnosis of AD, cognitive deficits, in particular significant memory impairment, occur. While the economic impact of this prodromal state is not well documented, it is easy to imagine the potential for significant functional problems in an increasingly technologic society. Without considering the impact of
JACOBSAND SANO this prodromal state, estimates of the total cost of AD range from $60 billion to $90 billion annually (16). A growing interest in women's health and an awareness of the aging of society stimulated many research initiatives to explore the relationship of hormones and dementia. Early evidence from epidemiologic investigations suggested that postmenopausal HRT may play a role in attenuating the devastation of AD and memory loss in aging. Although initial case-control studies yielded conflicting results (1726), subsequent large-scale prospective cohort studies consistently showed a lower risk of developing AD among women who had used HRT during the postmenopausal period (27-29). LeBlanc and colleagues (7) combined the results of 2 cohort and 10 case-control studies by metaanalysis. Results yielded a summary risk ratio of 0.66 (95% confidence interval [CI], 0.53-0.82); women with a history of HRT use were 34% less likely to develop AD than those who had never used HRT. Most, but not all, women in these studies used oral conjugated equine estrogen (CEE). Data on dose, duration, and timing of treatment in relation to the menopause and the cognitive outcome were scant, and results were conflicting.
A. The Women's Health Initiative Memory Study The promising results from observational studies of the effects of HRT on dementia risk spurred an ancillary study to the W H I hormone therapy trials" the Women's Health Initiative Memory Study (WHIMS) (30). W H I M S was nested within the W H I placebo-controlled trial of combination HRT with 0.625 mg daily of CEE and 2.5 mg of medroxyprogesterone acetate (MPA) for women with a uterus or CEE alone for those who had hysterectomy. W H I M S participants were aged 65 or older and were free of probable dementia at baseline, although about 6% did appear to have cognitive impairment. No other inclusion/exclusion criteria were required. Participants consented to annual assessment with a brief global measure of mental status, the Modified Mini Mental State Exam (3MSE) (31). Those who scored below preset cut points on follow-up mental status screening were referred for more detailed neuropsychologic assessment and for a clinical dementia evaluation by a physician. Primary outcome measures of the W H I M S trial were incident dementia and mild cognitive impairment (MCI). Criteria for MCI required poor performance (tenth percentile or lower) in at least one area of cognitive function, some functional impairment, decline from baseline functioning, absence of dementia, and no evidence of a psychiatric disorder or medical condition that could account for the decline in cognitive function. Secondary analyses examined global cognitive function as assessed by the 3MSE.
CHAPTER 22 Cognitive Health in the Postmenopausal Woman
1. RESULTS FROM THE ESTROGEN PLUS PROGESTIN TRIAL
From 4894 eligible participants in the W H I estrogen plus progestin trial, W H I M S recruited 4532 women (92.6%). In contrast to the promising findings from observational studies, results from W H I M S revealed a twofold increased risk of dementia for women in the estrogen plus progestin group (8). This increased risk translated into an additional 23 cases of dementia per 10,000 women per year. AD was the most common cause of dementia in both the HRT and placebo groups, although the specific effect for AD was not significant. Treatment groups did not differ in their risk of developing MCI. Rapp and colleagues (12) assessed global cognitive function in the same W H I M S cohort by examining scores on the 3MSE. Results revealed that mean total scores in both groups improved slightly over time. Women in the estrogen plus progestin group had smaller average increases in total score compared with women who received placebo; however, these differences were small in magnitude and not clinically important. There was, however, a small increased risk of clinically meaningful cognitive decline in the estrogen plus progestin group such that more women had a substantial and clinically significant decline (->2 standard deviations [SDs]) on mental status testing compared with the placebo group. 2. RESULTS FROM THE UNOPPOSED ESTROGEN TRIAL AND POOLED DATA Of 3200 age-eligible participants in the W H I CEE (i.e., estrogen-alone) trial, 2947 (92.1%) were enrolled in WHIMS. Results from this trial did not reveal any significant effect of treatment group on incidence of dementia. When, however, data from the CEE trial were pooled with data from the CEE/MPA trial, as per the original W H I M S protocol, a significant increased risk of dementia was found for those on active drug (hazards ratio [HR], 1.76; 95% CI, 1.19-2.60; p = 0.005) (9). This result remained significant after excluding participants with baseline mental status scores below the screening cut point. Incident MCI was not associated with HRT in the CEE study or in the combined trial analyses. Women assigned to CEE, however, were about 40% more likely to be diagnosed with either MCI or probable dementia at some point during the trial. The effects of CEE on global cognitive functioning as assessed by the 3MSE were similar to those observed with CEE/MPA. Women assigned to CEE therapy scored, on average, slightly but significantly lower compared with women assigned to placebo across the 5.4 years of follow-up (13). Again, this overall difference was so small that it was deemed clinically irrelevant. However, compared with
289 women assigned to placebo, women assigned to CEE were significantly more likely to experience a clinically significant decline on mental status testing. The most dramatic declines were observed among women who scored poorly on the 3MSE at their screening visit, suggesting that HRT may have accelerated an extant disease process (e.g., cerebrovascular disease, neuropathologic changes) in these women. This finding parallels that seen in women with frank dementia (described below).
B. H R T
as a T r e a t m e n t
for A D in P o s t m e n o p a u s a l W o m e n There have been several randomized clinical trials of CEE in women already diagnosed with mild to moderate AD (32-34). In each of these trials, CEE failed to improve cognition or slow the rate of cognitive decline among women with AD. Duration of follow-up ranged from 4 to 12 months. Furthermore, there was some evidence of deleterious cognitive effects. Based on these findings, use of CEE to treat AD in postmenopausal women is not recommended. In several small placebo-controlled studies, Asthana and colleagues found that transdermal 1713-estradiol improved attention and memory functioning in postmenopausal women with AD (35,36). As these studies had very small sample sizes (12 and 20 participants) and the duration of treatment was short (8 weeks), the therapeutic potential of transdermal 17 [3-estradiol awaits the completion of larger multicenter studies with longer treatment durations.
C. S u m m a r y o f F i n d i n g s in W o m e n O v e r A g e 65 Despite suggestions from observational studies that HRT may play a role in maintaining cognitive health and preventing AD in older women, findings from the W H I M S have demonstrated clearly that--at least in women age 65 and oldermthere were no cognitive benefits of CEE with or without MPA. Women randomized to active therapy scored lower on a measure of global cognitive functioning and were more likely to develop dementia or MCI. The increased risk of cognitive decline in women randomized to HRT may be due to the vascular effects of estrogens (e.g., increases in thrombin, fibrinolysis, triglycerides, and C-reactive protein) and associated vascular disease in the brain; the high incidence of stroke in both the CEE and CEE/MPA trials supports this claim (10,11). It is unclear whether the results from the W H I M S will generalize to other hormone preparations; however, caution seems warranted in prescribing any HRT to healthy older women for the purpose of preventing cognitive decline.
290 It is possible that initiating hormone therapy at age 65 or older is too late to prevent AD as the underlying disease processes (e.g., neurodegeneration), which clearly develop before clinically significant symptoms are recognized, may already be underway. In a large, prospective cohort study, Zandi and colleagues (29) observed that among women with a mean baseline age of 74.5 years, those who reported prior HRT use had a significantly reduced risk of developing AD, whereas no such risk reduction was observed among current HRT users unless duration of use exceeded 10 years. This finding suggests that there may be a critical window of time during which HRT exposure may reduce the risk of AD. Randomized clinical trials have shown that CEE does not improve cognitive function or slow the rate of cognitive decline among women diagnosed with mild to moderate AD, further lending support to the idea that CEE will not alter the course of neurodegeneration once the process has begun. Thus, the model for preventing AD with H R T may be similar to that reported in osteoporosis, wherein the primary positive benefit of HRT on bone loss is found in the immediate perimenopausal/postmenopausal period (37).
III. COGNITIVE FUNCTIONING IN YOUNGER POSTMENOPAUSAL WOMEN Although findings from W H I M S clearly indicate that women aged 65 and older should not be treated with CEE with or without MPA to enhance cognition or prevent dementia, questions remain about whether these findings can be generalized to younger postmenopausal women, including women seeking relief from climacteric symptoms. Cognitive complaints are common among perimenopausal and recently postmenopausal women; however, a causal relationship between cognitive disturbances and the menopause has not been clearly demonstrated. Studies to date have not adequately separated the effects of aging on cognition from the potential effects of menopause. To further complicate this area of research, other symptoms associated with menopause (e.g., sleep disturbance) may indirectly contribute to cognitive symptoms. There have been a number of randomized controlled trials examining the effects of unopposed estrogen (typically estradiol or CEE) on cognitive functioning in younger postmenopausal women (1,2,38-44; see LeBlanc et al. [7] for review). Many of these studies had small sample sizes, and the duration of use was uniformly short, ranging from 21 days to 6 months. Methodologic differences in study design (i.e., crossover versus separate experimental and placebo groups); preparation and dose of HRT; cognitive outcome measures; type of menopause (i.e., surgical versus natural);
JAcoBs AND SANO whether or not women with menopausal symptoms were included; and prior estrogen use made it impossible to combine these trials quantitatively in a meta-analysis. Nevertheless, some trends were apparent. First, no deleterious effects on cognitive functioning were observed. Second, H T did not consistently enhance cognitive test performance in women who were not experiencing menopausal symptoms. Among women with menopausal symptoms, however, HRT improved cognitive test performance, especially on tests of verbal memory and attention, as well as abstract reasoning and motor speed (7). Although not examined directly, it is possible that the cognitive improvement observed in symptomatic women receiving HRT may be attributed to a lessening of other menopausal symptoms (e.g., hot flushes, sleep disturbance, mood changes). Nevertheless, it appears that cognitive improvementmat least in the domains of verbal memory and attentionmmay be an additional benefit of HRT for some women. To date, only two studies have examined the effects of estrogen opposed with progestin on cognition in younger women, and they were based on the same cohort of 49 women (45,46). These studies were designed to compare three groups" unopposed estrogen (estradiol valerate), opposed estrogen (estradiol valerate plus Dienogest), and placebo. Results revealed that women randomized to unopposed estrogen performed better than the other two groups on a test of associate verbal learning, whereas those receiving opposed estrogen performed better on a test of numeric memory. The dearth of studies examining cognitive effects of opposed estrogen therapy in younger women is troubling because relatively young women without hysterectomy are a prevalent and clinically relevant portion of the postmenopausal population.
IV. SELECTIVE ESTROGEN RECEPTOR MODULATORS (SERMs) AND COGNITION Based on the impression that estrogen was associated with a beneficial effect on cognition, there has been interest in determining if SERMs may also have a cognitive benefit. These agents are designed to target some, but not all, populations of estrogen receptors. Perhaps the best known of these agents is raloxifene, which stimulates estrogen receptors in bone but appears to have no deleterious effect on breast or uterus. In large long-term trials with a primary outcome of fractures, studies of cognition have been added to the protocols to determine whether a cognitive benefit exists. A recent report from the Multiple Outcome of Raloxifene Evaluation (MORE) study, which included 5153 women with a mean age of 66.6 years, suggested that a high dose (120 mg/day) of raloxifene was associated with a
CHAPTER 22 Cognitive Health in the Postmenopausal Woman
reduced risk of MCI, with a borderline beneficial effect on the risk of AD (47). Interestingly, the lower dose (60 mg/ day) had no beneficial effect on cognition, and there was no evidence of a dose response curve. The beneficial effect of high-dose raloxifene, although small, is promising and worthy of further study. However, given the inability to observe a consistent treatment effect on cognition and the contradictory dose response, the biologic mechanism of this effect is unclear. In any case, this finding supports an earlier study that showed no negative effect of raloxifene on cognitive function (48).
V. UNANSWERED QUESTIONS AND FUTURE DIRECTIONS The results from W H I M S answered the important clinical question of whether HRT (i.e., CEE or CEE/MPA) delays the onset of dementia or cognitive decline in women age 65 and older. However, a number of questions regarding the effects of hormone replacement on cognition in women remain unanswered. As discussed previously, the effect of the timing of initiation of treatment on cognitive outcomes is unclear. The cohort in W H I M S may have been too old to appreciate potential cognitive benefits of HRT. It is possible that there is a critical period during which hormone replacement confers a neuroprotective effect that serves to decrease the risk of dementia or cognitive decline years or decades later. The results reported by Zandi and colleagues (29) (i.e., a decreased risk of AD among past users of HRT but not among current users) support the idea of such a critical period. The menopausal transition, with its relatively rapid depletion of estrogen, would seem to be a likely candidate for such a critical interval. Another important observation is that the deleterious effect of estrogen on cognition appears to be most robust in the presence of extant cognitive impairment or flank dementia. Perhaps estrogen confers additional risk in the presence of neural compromise, such as cell loss and plaque and tangle burden associated with AD. It would be valuable to understand the mechanisms of this risk. For example, vascular or coagulation effects of estrogen may be responsible for additional cognitive compromise when AD is present. It is possible that potential neurotropic effects of estrogen are negated by an increase in vascular disease. Given these risks, future trials designed to examine potential benefits of estrogen on cognitive outcomes should exclude participants with evidence of cognitive impairment at the screening assessment. Alternately, there may be value in examining estrogen in combination with treatments that could protect against vascular risk. It is unclear what effects, if any, other hormone preparations, dosages, or modes of administration have on cognitive health. As described previously, estradiol has shown some
291 positive results in trials of younger women; however, results from these studies should be interpreted cautiously in terms of both the risks and benefits of treatment because sample sizes were generally small and the duration of treatment short. Long-term, large-scale, randomized controlled trials with younger postmenopausal women are desperately needed to address how to best treat this clinically relevant group. Unresolved issues to be addressed in future studies include the cognitive effects of various forms of hormone replacement (e.g., estradiol versus CEE); route of administration (oral versus transdermal); the timing of initiation of therapy (perimenopausal versus postmenopausal); and the type of treatment regimen (continuous versus cyclic). Additional factors that can potentially affect therapeutic response to HRT include hysterectomy status, prior therapy with estrogen, and lifetime exposure to endogenous estrogens. Further research also is needed on the effects of progestational agents in hormone therapies on cognitive health. There has been speculation that progestins in general, and MPA in particular, may have deleterious effects on cognitive functioning. Results from W H I M S support this theory: There was no increased risk of dementia or MCI with CEE alone, but CEE/MPA was associated with a twofold increased risk of dementia.
VI. SUMMARYAND CONCLUSIONS In contrast to promising results from observational studies suggesting that HRT attenuates age-related cognitive decline and decreases the risk of AD, results from W H I M S clearly demonstrate that CEE and CEE/MPA do not benefit cognitive health in women age 65 or older. These drugs should not be used to maintain cognitive function in these women. Relatively small trials suggest that HRT may exert a cognitive benefit in younger postmenopausal women, particularly those reporting menopausal symptoms. Cognitive benefit is most often seen on measures of attention and verbal memory. It appears that these impairments, or at least subjective complaints of cognitive dysfunction, may be among the menopausal symptoms ameliorated by HRT in some women. However, corroboration of these results awaits larger controlled trials. It therefore seems clinically prudent to prescribe HRT only for its approved indications and to caution younger women initiating HRT about the W H I M S findings. Further research is needed on the effects of timing of HRT (i.e., initiation of treatment relative to the menopause). Some investigators have suggested a critical period during which initiation of HRT may decrease the risk of cognitive decline years or decades later. Further research also is needed on the cognitive effects of other hormone preparations, doses, and modes of administration as well as the effects of the various progestational agents used in combination therapy.
292
JacoBs AND SANO References
1. Sherwin BB. Estrogen and/or androgen replacement therapy and cognitive functioning in surgically menopausal women. Psychoneuroendocrin 1988;13:345-357. 2. Phillips SM, Sherwin BB. Effects of estrogen on memory function in surgically menopausal women. Psychoneuroendocrin 1992;17:485-495. 3. Sherwin BB, Tulandi T. "Add-back" estrogen reverses cognitive deficits induced by a gonadotropin-releasing hormone agonist in women with leiomyomata uteri. J Clin Endocrinol Metab 1996;81:2545-2549. 4. Gold EB, Sternfeld B, Kelsey JL, et al. Relation of demographic and lifestyle factors to symptoms in a multi-racial/ethnic population of women 40-55 years of age. Am J Epidemio12000;152:463-473. 5. Kok HS, Kuh D, Cooper R, et al. Cognitive function across the life course and the menopausal transition in a British birth cohort. Menopause 2006;13:19-27. 6. Low LF, Anstey KJ, Jorm AF, Rodgers B, Christensen H. Reproductive period and cognitive function in a representative sample of naturally postmenopausal women aged 60-64 years. Climacteric 2005;8: 380-389. 7. LeBlanc ES, Janowsky J, Chan BK, Nelson HD. Hormone replacement therapy and cognition: systematic review and meta-analysis. JAMA 2001;285:1489-1499. 8. Shumaker SA, Legault C, Rapp SR, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651-2662. 9. Shumaker SA, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women's Health Initiative Memory Study. JAMA 2004;291:2947-2958. 10. Wassertheil-SmoUer S, Hendrix SL, Limacher M, et al. Effect of estrogen plus progestin on stroke in postmenopausal women: the Women's Health Initiative: a randomized trial. JAAgA 2003;289:2673-2684. 11. Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMA 2004;2919:1701-1712. 12. Rapp SR, Espeland MA, Shumaker SA, et al. Effect of estrogen plus progestin on global cognitive function in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003 ;289:2663-2672. 13. Espeland MA, Rapp SR, Shumaker SA, et al. Conjugated equine estrogens and global cognitive function in postmenopausal women: Women's Health Initiative Memory Study. JAMA 2004;291:2959-2968. 14. Jorm AE The epidemiology of Alzheimer's disease and related disorders. London: Chapman and Hall, 1990. 15. Brookmeyer R, Gray S, Kawas C. Projections of Alzheimer's disease in the United States and the public health impact of delaying disease onset. Am J Public Health 1998;88:1337-1342. 16. Ernst RL, Hay JW. The US economic and social costs of Alzheimer's disease revisited. AJPH 1994; 84:1261-1264. 17. Heyman A, Wilkinson WE, Stafford JA, et al. Alzheimer's disease: a study of epidemiological aspects. Ann Neuro11984;15:335-341. 18. Amaducci LA, Fratiglioni L, Rocca WA, et al. Risk factors for clinically diagnosed Alzheimer's disease: a case-control study of an Italian population. Neurology 1986;36:922-931. 19. Broe GA, Henderson AS, Creasey H, et al. A case-control study of Alzheimer's disease in Australia. Neurology 1990;40:1698-1707. 20. Graves AB, White E, Koepsell TD, et al. A case-control study of Alzheimer's disease. Ann Neuro11990;28"766-774. 21. Brenner DE, Kukull WA, Stergachis A, et al. Postmenopausal estrogen replacement therapy and the risk of Alzheimer's disease: a populationbased case-control study. Am J Epidemio11994;140:262-267.
22. Henderson VW, Paganini-Hill A, Emanuel CK, Dunn ME, Buckwalter JG. Estrogen replacement therapy in older women: comparisons between Alzheimer's disease cases and nondemented control subjects. Arch Neuro11994;51:896-900. 23. Mortel KF, Meyer JS. Lack of postmenopausal estrogen replacement therapy and the risk of dementia. J Neuropsycbiatry Clin Neurosci 1995;7:334-337. 24. Paganini-Hill A, Henderson VW. Estrogen replacement therapy and risk of Alzheimer disease. Arch Intern Med 1996;156:2213-2217. 25. Waring SC, Rocca WA, Petersen RC, et al. Postmenopausal estrogen replacement therapy and risk of AD: a population-based study. Neurology 1999;52:965-970. 26. Harwood DG, Barker WW, Loewenstein DA, et al. A cross-ethnic analysis of risk factors for AD in white Hispanics and white nonHispanics. Neurology 1999;52:551-556. 27. Tang MX, Jacobs D, Stern Y, et al. Effect of oestrogen during menopause on risk and age at onset of Alzheimer's disease. Lancet 1996;348:429-432. 28. Kawas C, Resnick S, Morrison A, et al. A prospective study of estrogen replacement therapy and the risk of developing Alzheimer's disease: the Baltimore Longitudinal Study of Aging. Neurology 1997;48:1517-1521 [published correction appears in Neurology 1998;51:654]. 29. Zandi PP, Carlson MC, Plassman BL, et al. Hormone replacement therapy and incidence of Alzheimer disease in older women: the Cache County Study. JAMA 2002;288:2123-2129. 30. Shumaker SA, Reboussin BA, Espeland MA, et al. The Women's Health Initiative Memory Study (WHIMS): a trial of the effect of estrogen therapy in preventing and slowing the progression of dementia. Control Clin Trials 1998;19:604-621. 31. Teng EL, Chui H. The Modified Mini-Mental State (3MS) examination.J Clin Psychiatry 1987;48:314-318. 32. Henderson VW, Paganini-HiU A, Miller BL, et al. Estrogen for Alzheimer's disease in women: randomized, double-blind, placebo-controlled trial. Neurology 2000;54:295-301. 33. Mulnard RA, Cotman CW, Kawas C, et al. Estrogen replacement therapy for treatment of mild to moderate Alzheimer disease. JACMA 2000;283:1007-1015. 34. Wang PN, Liao SQ~ Liu RS, et al. Effects of estrogen on cognition, mood, and cerebral blood flow in AD: a controlled study. Neurology 2000;54:2061-2066. 35. Asthana S, Baker LD, Craft S, et al. High-dose estradiol improves cognition for women with AD: results of a randomized study. Neurology 2001;57:605-612. 36. Asthana S, Craft S, Baker LD, et al. Cognitive and neuroendocrine response to transdermal estrogen in postmenopausal women with Alzheimer's disease: results of a placebo-controlled, double-blind, pilot study. Psycboneuroendocrinology 1999;24:657-677. 37. Beck TJ, Stone KL, Oreskovic TL, et al. Effects of current and discontinued estrogen replacement therapy on hip structural geometry: the Study of Osteoporotic Fractures. J Bone Miner Res 2001;16:2103-2110. 38. Vanhulle G, Demol R. A double-blind study into the influence of estriol on a number of psychological tests in post-menopausal women. In: van Keep PA, Greenblatt RB, Albeaux Fernet M, eds. Consensus on the menopause research. Lancaster, United Kingdom: MTP, 1976. 39. Hackman BW, Galbraith D. Replacement therapy and piperazine oestrone sulphate ("Harmogen") and its effect on memory. Curr Med Res Opin 1976;4:303-306. 40. Fedor-Freybergh E The influence of oestrogens on the wellbeing and mental performance in climacteric and postmenopausal women. Acta Obstet Gynecol &and Supp11977;64:1-91. 41. Ditkoff EC, Crary WG, Cristo M, Lobo RA. Estrogen improves psychological function in asymptomatic postmenopausal women. Obstet Gyneco11991;78:991-995.
CHAPTER 22 Cognitive Health in the Postmenopausal W o m a n 42. Polo-Kantola P, Portin R, Polo O, et al. The effect of short-term estrogen replacement therapy on cognition: a randomized, double-blind, cross-over trial in postmenopausal women. Obstet Gynecol 1998;91: 459-466. 43. Shaywitz SE, Shaywitz BA, Pugh KR, et al. Effect of estrogen on brain activation patterns in postmenopausal women during working memory tasks. JAMff 1999;281:1197-1202. 44. Shaywitz, SE, Naftolin F, Zelterman D, et al. Better oral reading and short-term memory in midlife, postmenopausal women taking estrogen. Menopause 2003;10:420-426. 45. Linzmayer L, Semlitsch HV, Salem B, et al. Double-blind, placebocontrolled psychometric studies on the effects of a combined estrogenprogestin regimen versus estrogen alone on performance, mood and personality of menopausal syndrome patients. Arzneimittelforschung 2001;51:238-245.
293 46. Salem-Zyhlarz G, Anderer P, Gruber G, et al. Insomnia related to postmenopausal syndrome and hormone replacement therapy: sleep laboratory studies on baseline differences between patients and controls and double-blind, placebo-controlled investigations on the effects of a novel estrogen-progestogen combination (Climodien, Lafamme)versus estrogen alone.J Sleep Res 2003;12:239-254. 47. Yaffe K, Krueger K, Cummings SR, et al. Effect of raloxifene on prevention of dementia and cognitive impairment in older women: the Multiple Outcomes of Raloxifene Evaluation (MORE) randomized trial. Am J Psychiatry 2005;162:683-690. 48. Nickelsen T, Lufkin EG, Riggs BL, Cox DA, Crook TH. Raloxifene hydrochloride, a selective estrogen receptor modulator: safety assessment of effects on cognitive function and mood in postmenopausal women. Psychoneuroendocrinology1999;24:115-128.
This Page Intentionally Left Blank
2 H A P T E R 2;
The Role of Sex Steroids i9n A 1 z h e l"m e r
'sD" lsease:
Prevention and Treatment VICTOR W. HENDERSON
Departments of Health Research and Policy, and of Neurology and Neurological Sciences, Stanford University, Stanford, CA 94305
Dementia, among the most feared accompaniments of aging, represents the loss of mental abilities sufficient to interfere substantially with the ability to conduct one's daily affairs. Dementia has multiple causes; of these, Alzheimer's disease (AD) is by far the most prevalent in most countries (1,2). In AD, cognitive loss begins insidiously and progresses gradually over about a decade. An early and consistent feature of AD is impairment in episodic memory (3), namely, an impairment in the ability to learn and recall consciously acquired new information. Episodic memory deficits are often not as prominent in some of the other causes of dementia. Most patients with AD also show difficulties with attention, language, visuospatial skills, judgment, and abstract reasoning. Behavioral symptoms such as depression, apathy, agitation, or delusions also occur (4). Many patients experience an early phase characterized primarily by difficulty restricted to episodic memory and without limitation in the ability to conduct usual daily actMties. The term mild cognitive impairment (MCI) is increasingly applied in this setting (5). Key histopathologic features of AD include neurofibrillary tangles within vulnerable neurons and neuritic plaques (6). Tangles are largely composed of tau, a microtubule-associated protein that has been excessively phosphorylated. Plaques, which accumulate in the neuropil between neuronal cell bodies, often contain a central core of [3-amyloid, a polypeptide proteolytically derived from a larger precursor protein. The relation between [3-amyloid and tau remains unresolved. Within the plaque, an inflammatory process is suggested by the presence of microglia and reactive astrocytes, TREATMENT OF THE POSTMENOPAUSAL WOMAN
together with cytokines, complement proteins, and acute phase reactants (7). The pathogenesis of AD is unknown, although it is apparent that different genetic and nongenetic factorsmonly some of which have been identifiedmcan in isolation or combination culminate in the characteristic pathology and symptoms of AD. Point mutations in genes on chromosomes 14, 1, and 21 (encoding the presenilin proteins and the amyloid precursor protein) are expressed as autosomal dominant traits and lead to dementia symptoms before about 60 years of age (8). These mutations do not play an important role in later-onset illness, in which it is likely that a number of "susceptibility" genes influence risk. The best described of these is the chromosome 19 gene that encodes apolipoprotein E. This lipid transport protein exists in three common polymorphisms, with an increased susceptibility to AD conferred by possession of the e 4 allele. Gender modifies risk, with the e 4 allele of apolipoprotein E increasing risk for women more than men (9). The prevalence of AD increases with advancing age (1,2). There is observational evidence that other factors modify risk. Elevated risk is seen in people with low educational achievement (10) or previous head injury (11) or with conditions associated with vascular disease, such as hypertension (12,13), elevated serum cholesterol (13), and elevated serum homocysteine (14). AD risk might be reduced by use of anti-inflammatory drugs (15) and statins (16), modest alcohol consumption (17), by physical exercise (18), or participation in mentally stimulating activities (19). 295
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any formreserved.
296 Serum estradiol concentrations may (20) or may not be reduced in women with AD (21,22), but a woman's exposures to estrogen in the form of estrogen-containing hormone therapy (HT) is relevant to AD susceptibility and treatment. Less is known about potential roles of other sex steroids (progesterone and testosterone), dietary estrogens, and selective estrogen receptor modulators. Estrogen has important effects on brain function that might be predicted to modify the risk of developing AD, and HT has been considered as treatment for AD. The relation between HT and AD susceptibility has been addressed in observational research as well as in a large primary prevention trial. This body of clinical research has clear clinical implications regarding HT and AD in some situations, but other important clinical issues remain unanswered.
I. E S T R O G E N A N D T H E BRAIN Menopause represents the permanent cessation of menses due to the loss of ovarian follicular function (23). Natural menopause, with the nearly complete loss of ovarian estrogen production, occurs at an average age of 51 years, and about 40% of a woman's adult life occurs after the menopause. The brain is an important target organ for estrogen and other sex steroids, and subsets of neurons express receptors for estrogen, androgen, and progesterone. Two types of estrogen receptorswalpha and betamare expressed within specific regions of the human brain (24). The beta receptor is expressed in higher concentrations in the hippocampus and cerebral cortex, a fact that implies that actions mediated by the beta receptor could play a role in cognitive function (24). Receptors located on the plasma membrane, which help regulate intracellular signaling cascades, mediate some rapid estrogen actions that do not require genomic activation (25). Antioxidant properties of estrogen may not require receptor binding (26). Central nervous system consequences of menopausal estrogen loss remain an area of intense study and also of some controversy. Many actions of estrogen are potentially relevant to cognition and dementia (Table 23.1) (27). Estrogen modulates growth proteins (28,29), enhances the outgrowth of nerve processes (30,31), and promotes plasticity of nerve synapses (32). Within the hippocampus, estrogen enhances long-term potentiation, a physiologic process believed to contribute to the formation of episodic memories (32). Corticosteroid secretion that occurs in response to stress can deleteriously affect hippocampal neurons (33). Basal cortisol levels are reported to be increased in AD (34,35), and estrogen may mitigate the stress response in older women (36). Estrogen may protect against neuronal damage due to toxicity induced by excitatory neurotransmitters, [3-amyloid, or ischemia (37,38), and estrogen may prolong viability by
VICTORW. HENDERSON
TABLE 23.1 Estrogen Actions That May Be Relevant to Alzheimer's Disease Effects on neuronal growth, differentiation, and survival Enhanced long-term potentiation Blunted stress response Neuroprotection against oxidative stress, excitatory neurotoxicity, [3-amyloid toxicity, ischemia, and apoptosis Pro-inflammatory and anti-inflammatory effects Effects on thrombosis and coagulation Effects on neurotransmitter systems, including acetylcholine, noradrenalin, and serotonin Increased cerebral blood flow and glucose transport Increased expression of apolipoprotein E in the brain Decreased formation of [3-amyloid and hyperphosphorylated tau protein Adapted from reference27. protecting against apoptosis (39,40). Oxidative damage is prominent in the AD brain (41), and antioxidant properties of estrogens (26,37) may also play a protective role. The role of inflammation in AD pathogenesis and treatment is controversial (7,42,43), but estrogen effects on inflammarion--both pro-inflammatory and anti-inflammatory-may be relevant to this disorder. Estrogen reduces microglia activation (44) and regulates monocyte survival (45); oral estrogens elevate serum concentrations of C-reactive protein, an acute phase inflammatory marker (44,46). Whether these inflammatory effects are clinically relevant to AD is difficult to surmise. Prothrombotic (47) properties of some estrogens may be harmful, possibly contributing to increased cerebrovascular disease (48-50), and postmortem evidence of vascular pathology is commonly noted in the AD brain (51). Estrogen influences several neurotransmitter systems, including acetylcholine. A coherent body of experimental and clinical evidence indicates the importance of acetylcholine in memory and attention. The major class of medications approved for AD treatment elevate brain levels of this transmitter by inhibiting the enzyme acetylcholinesterase. Neurofibrillary tangle formation is prominent in cholinergic neurons of the basal forebrain area (52). These neurons express receptors for estrogen (primarily the alpha receptor) (53), and in the laboratory estrogen enhances cholinergic function after ovariectomy (54,55). Estrogen interacts with monoaminergic neurotransmitters, which are thought to be important in cognition and mood. Neurons in the locus coeruleus (origin of noradrenergic fibers) and the raphe region of the brainstem (origin of serotonergic fibers) are affected by AD pathology, and subpopulations of these neurons express estrogen receptors (56,57). Other estrogen actions increase cerebral blood flow (58), augment glucose transport into the brain (59), increase brain expression of apolipoprotein E (60), reduce the formation of
CHAPTER 23 The Role of Sex Steroids in Alzheimer's Disease: Prevention and Treatment [3-amyloid in the brain (61,62), and diminish hyperphosphorylation of the tau protein (63). With the exception of some estrogen effects on inflammation and coagulation, most actions would appear to work in the direction of improving cognition and reducing AD risk or symptoms.
II. E S T R O G E N A N D C O G N I T I O N Men and women perform similarly on most cognitive tasks, but on average, men tend to score better on tests of visuospatial and mathematic reasoning skills, and women tend to score better on verbal tasks (64). Estrogen modulates neural activity that occurs during cognitive processing (65,66). During the performance of certain cognitive tasks, gender-associated differences in patterns of neuronal activation are observed, suggesting analogous differences in functional organization within the brain (67). Such differences may in part be mediated by long-term and short-term effects of exposures to sex steroids. Putative estrogen effects on cognition are also implied by performance fluctuations during a woman's menstrual cycle (68-70), cognitive consequences of suppression of endogenous estrogen production (71,72), and better verbal memory in transsexual men treated with estrogen (73). A common interpretation of these studies is that estrogen helps maintains skills at which women tend to excel (e.g., certain verbal abilities), although not all analytic results are consistent with this simple formulation.
A. Hormone Therapy and Cognitive Abilities in Midlife Many women complain of forgetfulness or poor memory around the time of the menopausal transition (74-77), and it is reasonable to consider whether hormonal changes associated with the menopause have a direct effect on memory capabilities. However, forgetfulness is a common symptom at any age (74), and memory symptoms encompass a variety of cognitive complaints (76). For younger women in whom ovarian estrogen production is abruptly halted (e.g., after oophorectomy), there is experimental evidence that shortterm use of H T can improve verbal (but not nonverbal) memory, such as that measured by the number of details recalled from a paragraph story (72,78). In contrast, available evidence suggests that natural menopause per se does not have an important effect on episodic memory (79) or other cognitive skills (80), and serum estrogen levels at midlife are unrelated to memory scores (79). Memory symptoms during middle-age are linked to perceived stress, mood, and physical health (81), and such factors may influ-
297
ence a woman's perception of her memory. It is also possible that psychometric testing has been insensitive to what is actually happening in a woman's life. It may be difficult, for example, to quantify problems that led someone to enter a room and then not recall why she went into the room in the first place.
B. Hormone Therapy and Cognitive Abilities in Old Age Serum estrogen levels in older women are generally not associated with cognitive abilities (82,83), although estradiol levels were positively related to memory skills in two studies (84,85). Results of observational research involving older women are inconsistent in delineating a putative cognitive role for HT. Current H T use (86), prior (but not current) H T use (87), or ever-use (88,89) are reported to protect against cognitive decline; other investigators provide no compelling evidence that H T preserves cognitive function (90,91). One study ofpostmenopausal women found a modest beneficial association between current use of unopposed estrogen and rate of cognitive decline but a modest detrimental association for combined H T with estrogen and progestogen (92). In another large cohort, there was little cognitive benefit when current H T users were compared with never users, and long-term users were at increased risk of cognitive decline, particularly among women who initiated H T at older ages (93). Potential biases in these observational studies particularly include the healthy user bias, namely, women who begin H T tend to be healthier in the first place than women who do not begin H T (94,95). Results from large randomized clinical trials provide a more consistent picture for H T initiation among older postmenopausal women. For generally healthy older women (mean age 81 years) (96), older women with known coronary heart disease (mean age 71 years) (97), and older women with cerebrovascular disease (mean age 70 years) (98), H T does not substantially influence episodic memory or other cognitive skills (Table 23.2). H T also does not have a substantial effect on global cognition among postmenopausal women who are 65 years of age or older (99,100). This conclusion is based on the Women's Health Initiative Memory Study (WHIMS), a multicenter randomized clinical trial that is described in more detail later. Over a mean follow-up of approximately 5 years, scores on the modified Mini-Mental State examination, a screening instrument that provides a global measure of cognition, remained approximately equivalent between W H I M S participants assigned to H T and those assigned to placebo. In both groups, scores tended to increase over time, presumably representing effects of practice. The magnitude of increase was somewhat less in women assigned to HT, but
VICTOR W. HENDERSON
298 TABLE 23.2 Hormone Therapy in Older Postmenopausal Women without Dementia: Larger Clinical Trials Assessing Episodic Memory*
Author and year
Clinical status
Binder et al., 2001 ( 9 6 )
Healthy
Grady et al., 2002 ( 9 7 )
History of coronary heart disease History of stroke or transient ischemic attack
Viscoli et al., 2005 ( 9 8 )
Number of women
Mean treatment duration
Significance: Episodic memory
Significance: Other cognitive measures
52
9 months
NS
NS
1063
50 months
NS
NSt
461
38 months
NS
NS:i:
*Episodic memorytasks were verbal paired word associate learning for Binder et al. (96), word list recall for Grady et al. (97), and recall of names of pictures used in a naming test for Viscoli et al. (98). ]-Women in the placebo group performed better on one of four other cognitive tasks (categoryfluency). tPost hoc analyses restricted to women with a baseline Mini-Mental State examination score of at least 28 showed less decline on this task among women in the active treatment group, but no between-group differences on five other tasks, including the episodic memory task. N S, not significant.
the average between-group difference was small and not clinically important (99,100).
III. H O R M O N E T H E R A P Y A N D THE PREVENTION OF ALZHEIMER'S DISEASE: OBSERVATIONAL STUDIES Nearly 20 case-control and cohort studies have examined the relation between a woman's use of H T and her subsequent risk of developing AD. In general, most reported associations indicate a reduction in risk; no observational study has reported significantly increased risk. Overall risk is modestly reduced by about one-third (101,102). Beneficial associations between a woman's use of H T and AD risk are reported from the Leisure World retirement community (103); a New York City (Manhattan) community-based cohort (104); the Baltimore Longitudinal Study of Aging (105); the Italian Longitudinal Study on Aging (106); a study in Rochester, Minnesota (107); a European population-based study of early-onset AD (108); a study in Cache County, Utah (109); and the Multi-Institutional Research in Alzheimer Genetic Epidemiology study (110). In contrast to these generally positive studies, no protective association of H T was observed among women enrolled in a health maintenance organization in Seattle (111) or among newly diagnosed cases of AD in a general practice database in the United Kingdom (112). Several studies report greater risk reductions among women who are long-term H T users (103,104,107,109). In Cache County, Utah, a protective association was found for past H T use but not current use (109), and in the MultiInstitutional Research in Alzheimer Genetic Epidemiology,
the protective association was found among younger postmenopausal women but not among older women (110). H T protection, when observed, is not significantly modified by a woman's apolipoprotein E genotype (104, 108,110).
IV. H O R M O N E THERAPY AND THE PREVENTION OF ALZHEIMER'S DISEASE: WHIMS Meta-analyses of observational studies suggest that the overall AD risk reduction associated with H T use is about one-third (101,102). The clinical relevance of these protective associations, however, is called into question by findings from W H I M S , a large clinical trial designed to assess cognitive outcomes after H T (113,114). Discordant findings regarding H T and AD prevention could reflect unappreciated bias or confounding in the observational studies, but other factors come into play. The parent study for W H I M S was the Women's Health Initiative, a multicenter, randomized, double-blind, placebocontrolled trial that assessed effects of a commonly prescribed oral H T on cardiovascular disease, breast cancer, and other defined outcomes (50,115). Active treatment was with conjugated equine estrogens (0.625 mg/day) given with or without a progestogen (medroxyprogesterone acetate 2.5 mg/day) as a continuous combined preparation. Depending on hysterectomy status, active treatment was with the combined H T preparation (estrogen plus progestogen [E+P] trial for women with a uterus) or with estrogen alone (E-alone trial for women who had undergone hysterectomy). The W H I M S trials were halted early because more adverse health out-
CHAPTER 23 The Role of Sex Steroids in Alzheimer's Disease: Prevention and Treatment
299
TABLE 23.3 Incident Dementia in the Women's Health Initiative Memory Study
WHIMS trial Estrogen plus progestogen trial Estrogen-alone trial
Number of women treated
Mean treatment duration
Rate per 10,000 person years, hormone therapy group
Rate per 10,000 person years, placebo group
Hazard ratio (95% confidence interval)
4532
4.1 years
45
22
2.05 (1.21-3.48)
2947
5.2 years
37
25
1.49 (0.83-2.66)
In the estrogen plus progestogen trial for women with a uterus, hormone therapywas conjugated equine estrogens, 0.625 mg/day,plus medroxyprogesterone acetate, 2.5 mg/day (113). In the estrogen-alone trial for women without a uterus, active treatment was with conjugated equine estrogens, 0.625 mg/day (114). WHIMS, Women's Health Initiative Memory Study (113,114).
comes occurred among women assigned to active treatment than among women assigned to placebo (E + P trial) or because no overall health benefit was seen in for women in the active treatment group (E-alone trial) (50,115). W H I M S assessed the incidence of dementia and other cognitive outcomes among a subset of participants in the Women's Health Initiative. The W H I M S trials are considered to be primary prevention trials because women were generally healthy, even if not necessarily in optimal health. W H I M S , which ultimately involved more than 7400 women, was restricted to community-dwelling women between the ages of 65 and 79 years, inclusive, at the time of treatment allocation (113,114). Sixty-one women developed dementia during the course of the E + P trial, as did 47 women during the E-alone trial. Separate outcomes were not reported for specific subtypes of dementia. In the E + P trial, the hazard ratio (a measure of relative risk) for dementia was doubled for women assigned to active treatment compared with controls (Table 23.3). In the E-alone trial, the hazard ratio increased by about one-half for women assigned to active treatment (see Table 23.3). In pooled analyses from both W H I M S trials, the hazard ratio for H T assignment was 1.76 (95% confidence interval [CI] 1.19 to 2.69) (114). In all treatment arms, older women and women with lower baseline cognitive scores were more likely to develop dementia. It is notable that increased risk was apparent within several years of randomization (113,114), especially since the preclinical phase of AD is believed to begin a decade or more before the onset of overt dementia (116). It is speculated that H T effects on vascular disease may have contributed to a rapid appearance of dementia in the W H I M S trials (114,117). The W H I M S trials also considered M C I as a secondary outcome, defined largely on the basis of low scores on neuropsychologic tests in the absence of dementia. This definition of M C I is somewhat different from that used in other studies, which have usually emphasized poor episodic memory in older persons without dementia (5). For W H I M S participants assigned to HT, the risk of M C I was somewhat increased, but the difference in risk between H T and pla-
cebo groups was not significant (hazard ratio 1.25, 95% CI 0.97 to 1.60) (113,114).
A. Discordant Findings between Observational Studies and WHIMS As summarized previously, most observational studies of AD risk have implied a protective association for H T use. Indeed, even among participants in the W H I M S trials, women who had used H T prior to W H I M S enrollment were less likely to develop dementia than women who had never used HT, regardless of whether they were randomized to H T or placebo during the W H I M S trials (113,114). Although W H I M S authors did not provide separate risk estimates for AD or other specific dementia causes, AD was more common among W H I M S participants allocated to H T than to placebo. The numbers were small (33 women in the two H T groups developed AD versus 21 in the two placebo groups [113,114]), but it is probably unreasonable to assume that W H I M S findings do not pertain to AD. Are discordant findings between observational studies and the W H I M S trials due to a healthy user bias in the former? Women who use H T on average enjoy better health, are better educated, and lead healthier lifestyles than women who do not use H T (95,118). These differences by themselves may influence AD risk. Yet another possibility is that key characteristics of women in the W H I M S trial differed from those of women in observational trials. Several of these differences have been discussed (119), but the principal difference concerns the age when women initiated HT. Most observational reports included postmenopausal women of various ages, but in these reports H T use was usually analyzed as ever-use versus never-use. H T is most often prescribed for vasomotor symptoms around the time of menopause, taken for several years, and then discontinued. Thus, most H T use in observational studies occurred at a relatively young age. However, H T use in the W H I M S trials occurred relatively later in life, at age 65 or older. Two observational studies using computerized pharmacy databases failed to
300 show an association between H T and AD risk; interestingly, the databases had not been established early enough to document early H T use in many instances (111,112). The biologic effects of estrogen sometimes vary according to age or time since menopause; for example, the effects of H T on bone fracture (120) or atherosclerosis progression (121-123). Some investigators have considered whether H T effects on AD risk might also vary depending on usage during a critical window, defined by young age or temporal proximity to menopause (124). In partial support of this conjecture, one observational report found that past H T use, but not current H T use, was linked to a reduction in AD risk (109). A second report found that a woman's use of l i T was associated with reduced risk, but risk reduction was restricted to younger postmenopausal women (110).
V. H O R M O N E T H E R A P Y A N D ALZHEIMER'S DISEASE T R E A T M E N T Qgite apart from any potential role in AD prevention, H T has been considered as a possible treatment for women with symptoms of this illness. This consideration is supported by a strong biologic rationale for why estrogen might benefit processes implicated in AD pathogenesis (see Table 23.1). Several randomized, double-blind, placebo-controlled, parallelgroups trials have evaluated effects of estrogen in women with AD. Mean ages of clinical trial participants range from 72 to 83 years. Results are not fully congruent. Findings from several smaller trials seemed encouraging (125-127), but those from larger trials indicate that H T does not benefit women with dementia due to AD (128-131). Three small, randomized controlled trials followed participants for 3 to 8 weeks; findings from these short-term studies suggest that estrogen might benefit women with AD (125-127). The first of these, published in 1993 in Japan, reported 14 women randomized to conjugated estrogens or placebo (125). At 3 weeks, improvement was noted in the estrogen group on each of three outcome measures; between-group differences were significant on one of these, the revised Hasegawa's Dementia Scale (125). In two separate 8-week trials, a University of Washington group used transdermal estradiol to study 12 (126) and 20 (127) women with AD. Outcomes in both trials included results from a comprehensive neuropsychologic battery. In the initial report, significant differences favoring the estrogen group were seen in a small subset of measures; namely, number of self-corrections in the interference condition of the Stroop Color Interference task and cued delayed recall on the Buschke Selective Reminding Test (126). In the follow-up study, Stroop and Buschke tests were defined as primary outcomes. This time, women in the estrogen group were faster in completing the interference condition of the Stroop and outperformed placebo subjects on the total recall mea-
VICTORW. HENDERSON
sure (immediate plus delayed recall) of the Buschke selective reminding test (127). Four somewhat longer-term randomized controlled trials followed women for 12 to 52 weeks. In these, investigators concluded that estrogen was probably ineffectual for AD (128-131). A multi-site U.S. study assessed 36 women allocated to conjugated estrogens or placebo. After 16 weeks, treatment groups did not differ on the primary outcome, the cognitive portion of the Alzheimer's Disease Assessment Scale (128). A 12-week trial conducted in Taiwan compared 47 women receiving conjugated estrogens or placebo; there were no differences on three primary outcomes (Cognitive Ability Screening Instrument, Clinical Dementia Rating, and Clinician Interview-Based Impression of Change) (129). In the longest study, a multicenter trial of 120 women without a uterus, participants received conjugated estrogens or placebo. Intent-to-treat analyses after 1 year revealed that treatment groups did not differ on the Clinical Global Impression of Change. Eighty percent of estrogen users and 74% of placebo users showed decline on this measure (130). Finally, a multicenter French study randomized 117 women to transdermal estradiol or placebo (131). Women in the active-treatment arm received oral progesterone, and women in both arms received a cholinesterase inhibitor. Cholinesterase inhibitors, which increase levels of ace@choline in the brain, are effective for AD treatment, and one rationale for the use of a cholinesterase inhibitor in the French study was that estrogen increases cholinergic markers in the brain (54). Moreover, post hoc analyses of a prior AD treatment trial using a cholinesterase inhibitor suggested that women already taking H T and randomized to the cholinesterase inhibitor performed significantly better than women not taking H T and randomized to the cholinesterase inhibitor (132). Nevertheless, French investigators reported that after 28 weeks, estradiol and placebo treatment groups did not differ on the primary outcome, the cognitive portion of the Alzheimer's Disease Assessment Scale (131).
VI. O T H E R C O M P O U N D S
A. Other Sex Steroids: Progesterone and Testosterone Subsets of neurons express receptors for progesterone (133) or androgen (134), as well as for estrogen (24). All are members of a nuclear receptor superfamily and function as ligand-activated transcription factors, which bind to DNA and modulate expression of target genes. Neither testosterone, the predominant circulating androgen, nor estradiol is synthesized by neurons or glia, although testosterone can be converted within the brain to estradiol through the action of aromatase. Other steroids, referred to as neurosteroids (135), can be synthesized within the central nervous system. Pro-
CHAPTER 23 The Role of Sex Steroids in Alzheimer's Disease: Prevention and Treatment gesterone is a neurosteroid, as is dehydroepiandrosterone, described later. For women with a uterus, treatment with unopposed estrogen increases risk of endometrial cancer (136), and H T therefore usually includes a progestogen given cyclically or continuously. Intracellular progesterone receptors are distributed within distinct brain areas (133), and depending on the experimental model, progesterone can enhance, modify, or oppose estrogen actions in the brain. Different progestogens may affect the brain differently. In vitro, neuroprotective effects of progesterone differ from those of medroxyprogesterone acetate, the progestogen used in the W H I M S trial (137). Limited clinical data raise the possibility that the addition of a progestogen adversely influences putative cognitive benefits of estrogen. In a prospective cohort study of older Japanese Americans without dementia, there was a modest beneficial association between current unopposed estrogen use and the rate of change on a global cognitive measure, but a modest detrimental association between current estrogen used with a progestogen and cognitive change (92). In a prospective cohort study of female nurses, cognitive performances were about the same for unopposed estrogen users and users of combined estrogen-progestogen HT, but the risk of substantial cognitive decline was somewhat more elevated among long-term use of combined H T (93). However, in the W H I M S trials, when effects of H T (E-alone or E + P ) were compared with those of placebo on a global cognitive measure, both H T groups performed about the same relative to their respective placebo controls (99), suggesting that cognitive actions of progestogenmat least in this settingmare probably unimportant. Furthermore, both E-alone and E + P increased dementia risk in the W H I M S trials (113,114). Although the estimate of increased risk was somewhat greater with E + P than with E-None, the two risk estimates themselves did not differ significantly from each other (114). Again, this finding suggests that progestogen in this setting was not critical in determining whether women developed dementia. Testosterone, like estrogens, has neurotrophic effects (138) and can reduce formation of the [3-amyloid peptide (139). In healthy men, testosterone is thought to enhance cognitive performance that relies on visuospatial skills (140,141), and low serum concentrations of testosterone may be associated with the subsequent development of AD in men (142). The serum concentration of testosterone is reduced after oophorectomy (143) but does not change appreciably across the natural menopausal transition (143,144). Oral HT, however, can decrease free testosterone by increasing levels of sex hormone-binding globulin (145,146), a binding protein for both testosterone and estradiol. There is evidence that testosterone may mediate sexual desire and arousal (147), but for postmenopausal women, research on androgens and
301
other aspects of brain function is limited. In a small, randomized crossover trial of surgically menopausal women, 3 months of testosterone improved several aspects of cognition (148). Testosterone levels are positively (83,84,149) as well as negatively (85) associated with some aspects of cognitive performance in postmenopausal women, but this relation has not been carefully examined in the setting of randomized clinical trials. Among older women, low testosterone is not linked to cognitive decline (150). The relation between androgens and AD has not been assessed in women.
B. Dehydroepiandrosterone Dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEA-S) are of potential interest in AD. Produced primarily by the adrenal glands rather than the gonads, D H E A is also a neurosteroid. D H E A itself has androgenic properties, and its metabolites include both testosterone and estradiol. Blood levels of D H E A and D H E A S decline dramatically with age (143,144). Reduced D H E A S levels were associated with functional limitations in a large cohort of older women (151) but were not associated with cognitive decline in another cohort of elderly women (152). In a randomized crossover trial of healthy middle-age and elderly men and women, active treatment with D H E A enhanced feelings of psychologic well-being (153), but a small 2-week randomized trial found no cognitive effect of D H E A in healthy older men and women (154). Serum concentrations of D H E A or DHEA-S are not strongly related to AD. In one study, concentrations of D H E A were reported to be normal in AD (155), but in another study, levels were increased in women with AD (35). DHEA-S concentrations in AD patients are variably reported as normal (156-158) or reduced (34,155,159). Low DHEA-S levels may (160) or may not (151) predict incipient dementia. In a small, randomized, placebo-controlled trial of men and women with AD, no benefit of D H E A was noted after treatment for 6 months (161).
C. Selective Estrogen Receptor Modulators Selective estrogen receptor modulators (SERMs) lack the basic steroid structure of sex hormones but possess tissuespecific estrogenic effects. SERMs induce unique conformational changes in the estrogen receptor (162). Raloxifene, used for osteoporosis prevention, and tamoxifen, used in breast cancer prevention, are well-known SERMs. Within the brain, agonist or antagonist profiles of tamoxifen and raloxifene differ from each other and vary within different regions of the brain (163). In secondary analyses from a multinational randomized placebo-controlled clinical trial involving postmenopausal women (the Multiple Outcomes
302
VICTOR W. HENDERSON
of Raloxifene Evaluation study), high-dose raloxifene reduced the likelihood that study participants would develop cognitive impairment (164). In contrast, the possibility that tamoxifen could impair cognitive function was suggested in one observational study (165). Neurologic effects of SERMs are an increasingly important area of consideration as these compounds are developed and marketed.
D. Phytoestrogens A variety of dietary compounds found in plants have estrogenic, anti-estrogenic, and S E R M properties. These phytoestrogens are found in a wide range of grains, fruits, and vegetables. Estrogenic effects vary considerably and do not necessarily mimic those of estradiol (166). Some phytoestrogens are quite potent and are consumed in large quantities. For example, genistein, found in high concentration in tofu and other soy products, has a strong affinity for the estrogen beta receptor but only a weak affinity for the alpha receptor (167). Clinical trials have begun to consider potential cognitive effects of phytoestrogens (168). Potential effects of phytoestrogens on A D are unknown, although in the Honolulu Heart Study, tofu consumption in midlife was associated with poor cognitive performance in late life (169).
VII. CONCLUSIONS AND RECOMMENDATIONS There is a strong biologic rationale for the use of estrogen-containing H T for the prevention and treatment of A D (see Table 23.1). However, research findings to date have been disappointing. For treatment, clinical trial evidence indicates that beginning H T in the presence of dementia due to A D does not improve symptoms or slow progression (128-131). There is no evidence of substantial harm in this setting, but the specific issue of H T discontinuation for women with A D has not been addressed. A more important public health concern is H T use by healthy women without dementia. There is no clinical trial evidence that initiating H T after about age 64 improves memory (see Table 23.2) or other cognitive abilities. In addition, for the older postmenopausal woman the W H I M S trials provide strong evidence that H T elevates dementia risk within several years of initiation (113,114). However, it is controversial whether W H I M S results should be applied to younger women who were not eligible to enroll in W H I M S protocols. Observational studies, in which H T exposures tended to occur in younger women, indicate that H T use might reduce A D risk (101,170). However, potential biases in observational studies remain a strong concern. H T use has declined since the W H I results were announced, but millions of women continue to use H T
(171,172), and many are confused by what they have heard concerning HT, cognition, and dementia. Fortunately, the incidence of A D is very low during the midlife period, when women and their physicians might consider H T for moderate to severe vasomotor symptoms of the menopause (173). If H T elevates A D risk in this younger age group (an inference based on generalizations from W H I M S findings), the absolute risk of short-term usage is likely to be low. Further research is needed, however, before it can be known whether potential cognitive risks and benefits of midlife H T use might differ substantially from those in later life.
References 1. Hebert LE, Scherr PA, Bienias JL, Bennett DA, Evans DA. Alzheimer disease in the US population: prevalence estimates using the 2000 census.Arch Neuro12003;60:1119-1122. 2. Launer LJ, Andersen K, Dewey ME, et al. Rates and risk factors for dementia and Alzheimer's disease: results from EURODEM pooled analyses. Neurology 1999;52:78-84. 3. Small BJ, Fratiglioni L, Viitanen M, Winblad B, B~ickman L. The course of cognitive impairment in preclinical Alzheimer disease.Arch Neuro12000;57:839-844.
4. LyketsosCG, Lopez O, Jones B, et al. Prevalence of neuropsychiatric symptoms in dementia and mild cognitive impairment: results from the cardiovascularhealth study.JAMA 2002;288:1475-1483. 5. Petersen RC, Doody R, Kurz A, et al. Current concepts in mild cognitive impairment.Arch Neuro12001;58:1985-1992. 6. Mirra SS, Heyman A, McKeel D, et al. Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer's disease. Neurology 1991;41:479-486. 7. NeuroinflammationWorking Group. Inflammation and Alzheimer's disease. NeurobiolAging 2000;21:383-421. 8. Bird TD. Genetic factors in Alzheimer's disease. N Engl J Med 2005;352:862-864. 9. Farrer LA, Cupples LA, Haines JL, et al. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease.JAMA 1997;278:1349-1356. 10. Katzman R. Education and the prevalence of dementia and Alzheimer's disease. Neurology 1993;43:13-20. 11. MortimerJA, French LR, Hutton JT, Schuman LM. Head injury as a risk factor for Mzheimer's disease. Neurology 1985;35:264-267. 12. Skoog I, Lernfelt B, Landahl S, et al. 15-year longitudinal study of blood pressure and dementia. Lancet 1996;347:1141-1145. 13. KivipeltoM, Helkala EL, Laakso MP, et al. Midlife vascular risk factors and Alzheimer's disease in later life: longitudinal, population based study. BMJ 2001;322:1447-1451. 14. Seshadri S, Beiser A, Selhub J, et al. Plasma homocysteine as a risk factor for dementia and Alzheimer's disease. N Engl J Med 2002; 346:476-483. 15. in t' Veld BA, Ruitenberg A, Hofman A, et al. Nonsteroidal antiinflammatory drugs and the risk of Alzheimer's disease. N Engl J Med 2001;345:1515-1521. 16. Wolozin B, Kellman W, Ruosseau P, Celesia GG, Siegel G. Decreased prevalence of Alzheimer disease associatedwith 3-hydroxy-3methyglutaryl coenzyme A reductase inhibitors. Arch Neurol 2000;57:1439-1443. 17. Mukamal KJ, Kuller LH, Fitzpatrick AL, et al. Prospective study of alcohol consumption and risk of dementia in older adults. JAMA 2003;289:1405-1413.
CHAPTER 23 The Role of Sex Steroids in Alzheimer's Disease: Prevention and Treatment 18. Laurin D, Verreault R, Lindsay J, MacPherson K, Rockwood K. Physical activity and risk of cognitive impairment and dementia in elderly persons. Arch Neuro12001;58:498-504. 19. Wilson RS, Mendes De Leon CF, Barnes LL, et al. Participation in cognitively stimulating activities and risk of incident Alzheimer disease. JAMA 2002;287:742-748. 20. Manly JJ, Merchant CA, Jacobs DM, et al. Endogenous estrogen levels and Alzheimer's disease among postmenopausal women. Neurology 2000;54:833-837. 21. Hogervorst E, Smith AD. The interaction of serum folate and estradiol levels in Alzheimer's disease. Neuroendocrinol Lett 2002;23: 155-160. 22. Geerlings MI, Launer LJ, de Jong FH, et al. Endogenous estradiol and risk of dementia in women and men: the Rotterdam Study. Ann Neuro12003;53:607-615. 23. North American Menopause Society. Menopausepractice.'a clinician's guide. Cleveland, OH: North American Menopause Society, 2004 24. Osterlund MK, Gustafsson J-A, Keller E, Hurd YL. Estrogen receptor [3 (ER[3) messenger ribonucleic acid (mRNA) expression within the human forebrain: distinct distribution pattern to ERalpha mRNA. J Clin Endocrinol Metab 2000;85:3840-3846. 25. Edwards DE Regulation of signal transduction pathways by estrogen and progesterone. Annu Rev Physio12005;67:335-376. 26. Behl C, Skutella T, Lezoualch F, et al. Neuroprotection against oxidative stress by estrogens: structure-activity relationship. Mol Pharmacol 1997;51:535-541. 27. Henderson VW. Hormone therapy and the brain: a clinical perspective on the role of estrogen. New York: Parthenon, 2000. 28. Miranda RC, Sohrabji F, Toran-Allerand CD. Presumptive estrogen target neurons express mRNAs for both the neurotrophins and neurotrophin receptors: a basis for potential developmental interactions of estrogen with neurotrophins. Mol Cell Neurosci 1993;4:510-525. 29. Shughrue PJ, Dorsa DM. Estrogen modulates the growth-associated protein GAP-43 (neuromodulin) mRNA in the rat preoptic area and basal hypothalamus. Neuroendocrinology 1993;57:439-447. 30. Toran-Allerand CD. Organotypic culture of the developing cerebral cortex and hypothalamus: relevance to sexual differentiation. Psychoneuroendocrinology 1991;16:7-24. 31. Brinton RD, Tran J, Proffitt P, Montoya M. 17 [3-estradiol enhances the outgrowth and survival of neocortical neurons in culture. Neurochem Res 1997;22:1339-1351. 32. Foy MR, Henderson VW, Berger TW, Thompson RE Estrogen and neural plasticity. Curr Dir Psychol Sci 2000;9:148-152. 33. Fuchs E, Flugge G, Ohl F, et al. Psychosocial stress, glucocorticoids, and structural alterations in the tree shrew hippocampus. Physiol Behay 2001;73:285-291. 34. Armanini D, Vecchio F, Basso A, et al. Alzheimer's disease: pathophysiological implications of measurement of plasma cortisol, plasma dehydroepiandrosterone sulfate, and lymphocytic corticosteroid receptors. Endocrine 2003;22:113-118. 35. Rasmuson S, Nasman B, Carlstrom K, Olsson T. Increased levels of adrenocortical and gonadal hormones in mild to moderate Alzheimer's disease. Dementia 2002;13:74-79. 36. Lindheim SR, Legro RS, Bernstein L, et al. Behavioral stress responses in premenopausal and postmenopausal women and the effects of estrogen. AmJ Obstet Gyneco11992;167:1831-1836. 37. Goodman Y, Bruce AJ, Cheng B, Mattson ME Estrogens attenuate and corticosterone exacerbates excitotoxicity, oxidative injury, and amyloid [3-peptide toxicity in hippocampal neurons. J Neurochem 1996;66:1836-1844. 38. Alkayed NJ, Murphy SJ, Traystman RJ, Hurn PD. Neuroprotective effects of female gonadal steroids in reproductively senescent female rats. Stroke 2000;31:161-168.
303
39. Pike CJ. Estrogen modulates neuronal Bcl-xL expression and [3-amyloid-induced apoptosis: relevance to Alzheimer's disease. J Neurochem 1999;72:1552-1563. 40. Alkayed NJ, Goto S, Sugo N, et al. Estrogen and Bcl-2: gene induction and effect of transgene in experimental stroke. J Neurosci 2001;21:7543-7550. 41. Zhu X, Lee HG, Casadesus G, et al. Oxidative imbalance in Alzheimer's disease. Mol Neurobio12005;31:205-217. 42. Aisen PS, Schafer KA, Grundman M, et al. Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial. JAMA 2003;289:2819-2826. 43. Wyss-Coray T, Yan F, Lin AH, et al. Prominent neurodegeneration and increased plaque formation in complement-inhibited Alzheimer's mice. ProcNatlAcad &i U S A 2003;99:10837-10842. 44. Vegeto E, Belcredito S, Etteri S, et al. Estrogen receptor-alpha mediates the brain antiinflammatory activity of estradiol. ProcNatlAcad &i U S A 2003;100:9614-9619. 45. Mor G, Sapi E, Abrahams VM, et al. Interaction of the estrogen receptors with the Fas ligand promoter in human monocytes. Jlmmunol 2003;170:114-122. 46. Cushman M, Legault C, Barrett-Connor E, et al. Effect of postmenopausal hormones on inflammation-sensitive proteins: the Postmenopausal Estrogen/Progestin Interventions (PEPI) Study. Circulation 1999;100:717-722. 47. Rosendaal FR, Helmerhorst FM, Vandenbroucke JP. Female hormones and thrombosis, drterioscler Thromb Vasc Biol 2002;22: 201-210. 48. Gillum LA, Mamidipudi SK, Johnston SC. Ischemic stroke risk with oral contraceptives. JAM/t 2002;284:72-78. 49. Wassertheil-Smoller S, Hen&ix S, Limacher M, et al. Effect of estrogen plus progestin on stroke in postmenopausal women: the Women's Health Initiative: a randomized trial. JAMA 2003;289:2673-2684. 50. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMd 2004;291:1701-1712. 51. Jellinger KA, Attems J. Incidence of cerebrovascular lesions in Alzheimer's disease: a postmortem study. Acta Neuropathol (Bed) 2002; 105:14-17. 52. Coyle JT, Price DL, DeLong MR. Alzheimer's disease: a disorder of cortical cholinergic innervation. Science 1983;219:1184-1190. 53. Shughrue PJ, Scrimo PJ, Merchenthaler I. Estrogen binding and estrogen receptor characterization (ERR and ER[3) in the cholinergic neurons of the rat basal forebrain. Neuroscience2000;96:41-49. 54. Luine V. Estradiol increases choline acetyltransferase activity in specific basal forebrain nuclei and projection areas of female rats. Exp Neuro11985;89:484-490. 55. Gibbs RB. Effects of gonadal hormone replacement on measures of basal forebrain cholinergic function. Neuroscience2000; 101:931-938. 56. Herbison AE, Simonian SX, Thanky NR, Bicknell RJ. Oestrogen modulation of noradrenaline neurotransmission. Novartis Found Syrup 2000;230:74-85. 57. Bethea CL, Lu NZ, Gundlah C, Streicher JM. Diverse actions of ovarian steroids in the serotonin neural system. Front Neuroendocrinol 2002;23:41-100. 58. Slopien R, Junik R, Meczekalski B, et al. Influence of hormonal replacement therapy on the regional cerebral blood flow in postmenopausal women. Maturitas 2002;46:255-262. 59. Cheng CM, Cohen M, Wang J, Bondy CA. Estrogen augments glucose transporter and IGF1 expression in primate cerebral cortex. FASEB J 2001;15:907-915. 60. Stone DJ, Rozovsky I, Morgan TE, et al. Astrocytes and microglia respond to estrogen with increased apoE mRNA in vivo and in vitro. Exp Neuro11997;143:313-318.
304 61. Petanceska SS, Nagy G, Frail D, Gandy S. Ovariectomy and 1713estradiol modulate the levels of Alzheimer's amyloid [3 peptides in brain. Neurology 2000;54:2212-2217. 62. Zheng H, Xu H, Uljon SN, et al. Modulation of A[3 peptides by estrogen in mouse models. J Neurochem 2002;80:191-196. 63. Alvarez-De-La-Rosa M, Silva I, Nilsen J, et al. Estradiol prevents neural tau hyperphosphorylation characteristic of Alzheimer's disease. Ann NYAcad Sci 2005;1052:210-224. 64. Kimura D. Sex and cognition. Cambridge, MA: MIT Press, 1999. 65. Berman KF, Schmidt PJ, Rubinow DR, et al. Modulation of cognition-specific cortical activity by gonadal steroids: a positron-emission tomography study in women. Proc Natl Acad Sci U S A 1997;94: 8836-8841. 66. Shaywitz SE, Shaywitz BA, Pugh KR, et al. Effect of estrogen on brain activation patterns in postmenopausal women during working memory tasks. JAMA 1999;281:1197-1202. 67. Shaywitz BA, Shaywitz SE, Pugh KR, et al. Sex differences in the functional organization of the brain for language. Nature 1995;373: 607-609. 68. Hampson E. Variations in sex-related cognitive abilities across the menstrual cycle. Brain Cogn 1990;14:26-43. 69. Bibawi D, Cherry B, Hellige JB. Fluctuations of perceptual asymmetry across time in women and men: effects related to the menstrual cycle. Neuropsychologia 1995;33:131-138. 70. Maki PM, Rich JB, Rosenbaum RS. Implicit memory varies across the menstrual cycle: estrogen effects in young women. Neuropsychologia 2002;40:518-529. 71. Varney NR, Syrop C, Kubu CS, et al. Neuropsychologic dysfunction in women following leuprolide acetate induction ofhypoestrogenism. JAssist Reprod Genet 1993;10:53-57. 72. Sherwin BB, Tulandi T. "Add-back" estrogen reverses cognitive deficits induced by a gonadotropin-releasing hormone agonist in women with leiomyomata uteri. J Clin EndocrinolMetab 1996;81:2545-2549. 73. Miles CRG, Sanders G, Hines M. Estrogen and memory in a transsexual population. Horm Behav 1998;34:199-208. 74. Neugarten BL, Kraines RJ. "Menopausal symptoms" in women of various ages. PsychosomMed 1965;27:266-273. 75. Matthews KA, Wing RR, Kuller LH, Meilahn EN, Plantinga E Influence of the perimenopause on cardiovascular risk factors and symptoms of middle-aged healthy women. Arch Intern Med 1994;154:2349-2355. 76. Mitchell ES, Woods NE Midlife women's attributions about perceived memory changes: observations from the Seattle Midlife Women's Health Study. J Women Health Gender-BasedMed 2001;10:351362. 77. Obermeyer CM, Reher D, Alcala LC, Price K. The menopause in Spain: results of the DAMES (Decisions At MEnopause) study. Maturitas 2005;52:190-198. 78. Phillips SM, Sherwin BB. Effects of estrogen on memory function in surgically menopausal women. Psychoneuroendocrinology 1992;17: 485-495. 79. Henderson VW, Dudley EC, Guthrie JR, Burger HG, Dennerstein L. Estrogen exposures and memory at midlife: a population-based study of women. Neurology 2003;60:1369-1371. 80. Meyer PM, Powell LH, Wilson RS, et al. A population-based longitudinal study of cognitive functioning in the menopausal transition. Neurology 2003;61:801-806. 81. Woods NF, Mitchell ES, Adams C. Memory functioning among midlife women: observations from the Seattle Midlife Women's Health Study. Menopause 2000;7:257-265. 82. Portin R, Polo-Kantola P, Polo O, et al. Serum estrogen level, attention, memory and other cognitive functions in middle-aged women. Climacteric 1999;2:115-123.
VICTORW. HENDERSON 83. Barrett-Connor E, Goodman-Gruen D. Cognitive function and endogenous sex hormones in older women. JAm Geriatr Soc 1999;47: 1289-1293. 84. Drake EB, Henderson VW, Stanczyk FZ, et al. Associations between circulating sex steroid hormones and cognition in normal elderly women. Neurology 2000;54:599-603. 85. Hogervorst E, De Jager C, Budge M, Smith AD. Serum levels of estradiol and testosterone and performance in different cognitive domains in healthy elderly men and women. Psychoneuroendocrinology 2004;29:405-421. 86. Resnick SM, Metter EJ, Zonderman AB. Estrogen replacement therapy and longitudinal decline in visual memory. A possible protective effect? Neurology 1997;49:1491-1497. 87. Matthews K, Cauley J, Yaffe K, Zmuda JM. Estrogen replacement therapy and cognitive decline in older community women.JAm Geriatr Soc 1999;47:518-523. 88. Jacobs DM, Tang MX, Stern Y, et al. Cognitive function in nondemented older women who took estrogen after menopause. Neurology 1998;50:368-373. 89. Carlson MC, Zandi PP, Plassman BL, et al. Hormone replacement therapy and reduced cognitive decline in older women: the Cache County study. Neurology 2001 ;57:2210-2216. 90. Barrett-Connor E, Kritz-Silverstein D. Estrogen replacement therapy and cognitive function in older women. JAMA 1993;269:2637-2641. 91. de Moraes SA, Szldo M, Knopman D, Park E. Prospective assessment of estrogen replacement therapy and cognitive functioning: atherosclerosis risk in communities study. Am J Epidemiol 2001;154: 733-739. 92. Rice MM, Graves AB, McCurry SM, et al. Postmenopausal estrogen and estrogen-progestin use and 2-year rate of cognitive change in a cohort of older Japanese American women: the Kame Project. Arch Intern Med 2000;160:1641-1649. 93. Kang JH, Weuve J, Grodstein E Postmenopausal hormone therapy and risk of cognitive decline in community-dwelling aging women. Neurology 2004;63:101-107. 94. Hemminki E, Matin M, Topo E Selection to postmenopausal therapy by women's characteristics. J Clin Epidemio11993;46 :211-219. 95. Matthews KA, Kuller LH, Wing RR, Meilahn EN, Plantinga E Prior to use of estrogen replacement therapy, are users healthier than nonusers? Am J Epidemio11996;143:971-978. 96. Binder EF, Schechtman KB, Birge SJ, Williams DB, Kohrt WM. Effects of hormone replacement therapy on cognitive performance in elderly women. Maturitas 2001;38:137-146. 97. Grady D, Yaffe K, KristofM, et al. Effect ofpostmenopausal hormone therapy on cognitive function: the Heart and Estrogen/Progestin Replacement Study. AmJMed 2002;113:543-548. 98. Viscoli CM, Brass LM, Kernan WN, et al. Estrogen therapy and risk of cognitive decline: results from the Women's Estrogen for Stroke Trial (WEST). Am J Obstet Gyneco12005;192:387-393. 99. Rapp SR, Espeland MA, Shumaker SA, et al. The effect of estrogen with progestin treatment on global cognitive function in postmenopausal women: results from the Women's Health Initiative Memory Study. JAMA 2003;289:2663-2672. 100. Espeland MA, Rapp SR, Shumaker SA, et al. Conjugated equine estrogens and global cognitive function in postmenopausal women: Women's Health Initiative Memory Study. JAMA 2004;291:29592968. 101. Hogervorst E, Williams J, Budge M, Riedel W, Jolles j. The nature of the effect of female gonadal hormone replacement therapy on cognitive function in post-menopausal women: a meta-analysis. Neuroscience 2000;101:485-512. 102. LeBlanc ES, Janowsky J, Chan BKS, Nelson HD. Hormone replacement therapy and cognition: systematic review and meta-analysis. JAMA 2001;285:1489-1499.
CHAPTER 23 The Role of Sex Steroids in Alzheimer's Disease: Prevention and Treatment 103. Paganini-Hill A, Henderson VW. Estrogen replacement therapy and risk of Alzheimer's disease. Arch Intern Med 1996;156:2213-2217. 104. Tang M-X, Jacobs D, Stern Y, et al. Effect ofoestrogen during menopause on risk and age at onset of Alzheimer's disease. Lancet 1996;348:429-432. 105. Kawas C, Resnick S, Morrison A, et al. A prospective study of estrogen replacement therapy and the risk of developing Alzheimer's disease: the Baltimore Longitudinal Study of Aging. Neurology 1997;48:1517-1521. 106. Baldereschi M, Di Carlo A, Lepore V, et al. Estrogen-replacement therapy and Alzheimer's disease in the Italian Longitudinal Study on Aging. Neurology 1998;50:996-1002. 107. Waring SC, Rocca WA, Petersen RC, et al. Postmenopausal estrogen replacement therapy and risk of AD: a population-based study. Neurology 1999;52:965-970. 108. Slooter AJ, Bronzova J, Witteman JC, et al. Estrogen use and early onset Alzheimer's disease: a population-based study. J Neurol Neurosurg Psychiatry 1999;67:779-781. 109. Zandi PP, Carlson MC, Plassman BL, et al. Hormone replacement therapy and incidence of Alzheimer's disease on older women: the Cache County stud> JAMA 2002;288:2123-2129. 110. Henderson VW, Benke KS, Green RC, Cupples LA, Farrer LA. Postmenopausal hormone therapy and Alzheimer's disease risk: interaction with age. J Neurol Neurosurg Psychiatry 2005;76:103-105. 111. Brenner DE, Kukull WA, Stergachis A, et al. Postmenopausal estrogen replacement therapy and the risk of Alzheimer's disease: a population-based case-control study. Am J Epidemio11994;140:262-267. 112. Seshadri S, Zomberg G L, Derby LE, et al. Postmenopausal estrogen replacement therapy and the risk of Alzheimer's disease. Arch Neurol 2001;58:435-440. 113. Shumaker SA, Legault C, Rapp SR, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study (WHIMS). JAMH 2003;289:2651-2662. 114. Shumaker SA, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women's Health Initiative Memory Study. JAMA 2004;291:2947-2958. 115. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321-333. 116. Elias MF, Beiser A, Wolf PA, et al. The preclinical phase of Alzheimer's disease: a 22-year prospective study of the Framingham cohort. Arch Neuro12000;57:808-813. 117. Henderson VW. Hormone therapy and Alzheimer's disease: benefit or harm? Expert Opin Pharmacother 2004;5:389-406. 118. Derby CA, Hume AL, Barbour MM, et al. Correlates of postmenopausal estrogen use and trends through the 1980s in two southeastern New England communities. Am J Epidemio11993; 137:1125-1135. 119. Henderson VW. Estrogen-containing hormone therapy and Alzheimer's disease risk: understanding discrepant inferences from observational and experimental research. Neuroscience 2006;138:10311039. 120. Torgerson DJ, Bell-Syer SE. Hormone replacement therapy and prevention of nonvertebral fractures: a meta-analysis of randomized trials. JAMA 2001;285:2891-2897. 121. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis. A randomized, double-blind, placebo-controlled trial. Ann Intern Ailed 2001;135:939-953. 122. Hodis HN, Mack WJ, Azen SP, et al. Hormone therapy and the progression of coronary-artery atherosclerosis in postmenopausal women. N EnglJ Med 2003;349:535-545.
305
123. Mack WJ, Slater CC, Xiang M, et al. Elevated subclinical atherosclerosis associated with oophorectomy is related to time since menopause rather than type of menopause. Fertil Steri12004;82:391-397. 124. Resnick SM, Henderson VW. Hormone therapy and risk of Alzheimer disease: a critical time. JAMA 2002;288:2170-2172. 125. Honjo H, Ogino Y, Tanaka K, et al. An effect of conjugated estrogen to cognitive impairment in women with senile dementia-Alzheimer's type: a placebo-controlled, double-blind study. J Jpn Menopause Soc 1993;1:167-171. 126. Asthana S, Craft S, Baker LD, et al. Cognitive and neuroendocrine response to transdermal estrogen in postmenopausal women with Alzheimer's disease: results of a placebo-controlled, double-blind, pilot study. Psychoneuroendocrinology 1999;24:657-677. 127. Asthana S, Baker LD, Craft S, et al. High-dose estradiol improves cognition for women with AD: results of a randomized study. Neurology 2001;57:605-612. 128. Henderson VW, Paganini-Hill A, Miller BL, et al. Estrogen for Alzheimer's disease in women: randomized, double-blind, placebocontrolled trial. Neurology 2000;54:295-301. 129. Wang PN, Liao SO,~Liu RS, et al. Effects of estrogen on cognition, mood, and cerebral blood flow in AD: a controlled study. Neurology 2000;54:2061-2066. 130. Mulnard RA, Cotman CW, Kawas C, et al. Estrogen replacement therapy for treatment of mild to moderate Alzheimer disease: a randomized controlled trial. JAMA 2000;283:1007-1015. 131. Rigaud AS, Andre G, Vellas B, Touchon J, Pere JJ. No additional benefit of HRT on response to rivastigmine in menopausal women with AD. Neurology 2003;60:148-149. 132. Schneider LS, Farlow MR, Henderson VW, Pogoda JM. Effects of estrogen replacement therapy on response to tacrine in patients with Alzheimer's disease. Neurology 1996;46:1580-1584. 133. Schumacher M, Robert E Progesterone: synthesis, metabolism, mechanisms of action, and effects in the nervous system. In: Pfaff DW, Arnold AP, Etgen AM, Fahrbach SE, Rubin RT, eds. Hormones, brain and behavior, vol 3. San Diego: Academic Press, 2002:683-745. 134. Simerly RB, Chang C, Muramatsu M, Swanson LW. Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: an in situ hybridization study. J Comp Neuro11990;294:76-95. 135. Dubrovsky BO. Steroids, neuroactive steroids and neurosteroids in psychopathology. Prog Neuropsychopharmacol Bid Psychiatry 2005;29: 169-192. 136. Grady D, Gebretsadik T, Kerlikowske K, Ernster V, Petitti D. Hormone replacement therapy and endometrial cancer risk: a metaanalysis. Obstet Gyneco11995;85:304-313. 137. Nilsen J, Brinton RD. Impact of progestins on estrogen-induced neuroprotection: synergy by progesterone and 19-norprogesterone and antagonism by medroxyprogesterone acetate. Endocrinology 2002;143:205-212. 138. Leranth C, Hajszan T, MacLusky NJ. Androgens increase spine synapse density in the CA1 hippocampal subfield of ovariectomized female rats. J Neurosci 2004;24:495-499. 139. Gouras GK, Xu H, Gross RS, et al. Testosterone reduces neuronal secretion of Alzheimer's b-amyloid peptides. Proc Natldcad Sci U S d 2000;97:1202-1205. 140. Janowsky JS, Oviatt SK, Orwoll ES. Testosterone influences spatial cognition in older men. t?ehav Neurosci 1994;108:325-332. 141. Silverman I, Kastuk D, Choi J, Phillips K. Testosterone levels and spatial ability in men. Psychoneuroendocrinology 1999;24:813-822. 142. Moffat SD, Zonderman AB, Metter EJ, et al. Free testosterone and risk for Alzheimer disease in older men. Neurology 2004;62:188-193. 143. Davison S, Bell R, Donath S, Montalto J, Davis SR. Androgen levels in adult females: changes with age, menopause and oophorectomy. J Clin Endocrinol Metab 2005;90:3847-3853.
306 144. Burger HG, Dudley EC, Cui J, Dennerstein L, Hopper JL. A prospective longitudinal study of serum testosterone, dehydroepiandrosterone sulfate, and sex hormone-binding globulin levels through the menopause transition.J Clin EndocrinolMetab 2000;85:2832-2838. 145. Serin IS, Ozcelik B, Basbug M, et al. Long-term effects of continuous oral and transdermal estrogen replacement therapy on sex hormone binding globulin and free testosterone levels. Eur J Obstet Gynecol Reprod Bio12001;99:222-225. 146. Slater CC, Zhang C, Hodis HN, et al. Comparison of estrogen and androgen levels after oral estrogen replacement therapy. J ReprodMed 2001;46:1052-1056. 147. North American Menopause Society. The role of testosterone therapy in postmenopausal women: position statement of the North American Menopause Society. Menopause 2005;12:497-511. 148. Sherwin BB. Estrogen and/or androgen replacement therapy and cognitive functioning in surgically menopausal women. Psychoneuroendocrinology 1988;13:345-357. 149. Wolf OT, Kirschbaum C. Endogenous estradiol and testosterone levels are associated with cognitive performance in older women and men. Horm Behav 2002;41:259-266. 150. Yaffe K, Lui LY, Grady D, et al. Cognitive decline in women in relation to non-protein-bound oestradiol concentrations. Lancet2000;356: 708-712. 151. Berr C, Lafont S, Debuire B, Dartigues J-F, Baulieu E-E. Relationships of dehydroepiandrosterone sulfate in the elderly with functional, psychological, and mental status, and short-term mortality: a French community-based study. Proc Natl Acad Sci U S A 1996;93:1341013415. 152. Yaffe K, Ettinger B, Pressman A, et al. Neuropsychiatric function and dehydroepiandrosterone sulfate in elderly women: a prospective study. Biol Psychiatry 1998;43:694-700. 153. Morales AJ, Nolan JJ, Nelson JC, Yen SSC. Effects of replacement dose of dehydroepiandrosterone in men and women of advancing age. J Clin EndocrinolMetab 1994;78:1360-1367. 154. Wolf OT, Neumann O, Hellhammer DH, et al. Effects of a two-week physiological dehydroepiandrosterone substitution on cognitive performance and well-being in healthy elderly women and men. J Clin Endocrinol Metab 1997;82:2363-2367. 155. Yanase T, Fukahori M, Taniguchi S, et al. Serum dehydroepiandrosterone (DHEA) and DHEA-sulfate (DHEA-S) in Alzheimer's disease and in cerebrovascular dementia. EndocrJ 1996;43:119-123. 156. Birkenhtiger-Gillesse EG, Derksen J, Lagaay AM. Dehydroepiandrosterone sulphate (DHEAS) in the oldest old, aged 85 and over. Ann NYAcad Sci 1994;719:543-552. 157. Schneider LS, Hinsey M, Lyness S. Plasma dehydroepiandrosterone sulfate in Alzheimer's disease. Biol Psychiatry 1992;31:205-208. 158. de Bruin VM, Vieira MC, Rocha MN, Viana GS. Cortisol and dehydroepiandosterone sulfate plasma levels and their relationship to aging, cognitive function, and dementia. Brain Cogn 2002;50:316-323.
VICTORW. HENDERSON 159. Murialdo G, Barreca A, Nobili F, et al. Relationships between cortisol, dehydroepiandrosterone sulphate and insulin-like growth factor-I system in dementia. J EndocrinolInvest 2001;24:139-146. 160. Hillen T, Lun A, Reischies FM, et al. DHEA-S plasma levels and incidence of Alzheimer's disease. Biol Psychiatry 2000;47:161-163. 161. Wolkowitz OM, Kramer JH, Reus VI, et al. DHEA treatment of Alzheimer's disease: a randomized, double-blind, placebo-controlled study. Neurology 2003;60:1071-1076. 162. McDonnell DP. The molecular pharmacology of SERMs. Trends EndocrinolMetab 1999;10:301-311. 163. Zhou W, Koldzic-Zivanovic N, Clarke CH, et al. Selective estrogen receptor modulator effects in the rat brain. Neuroendocrinology 2002;75:24-33. 164. Yaffe K, Krueger K, Cummings SR, et al. Effect of raloxifene on the prevention of dementia and cognitive impairment in older women: the Multiple Outcomes of Raloxifene Evaluation (MORE) randomized trial. AmJ Psychiatry 2005;162:680-693. 165. Paganini-Hill A, Clark LJ. Preliminary assessment of cognitive function in breast cancer patients treated with tamoxifen. Breast CancerRes Treat 2000;64:165-176. 166. Patisaul HB, Dindo M, Whitten PL, Young LJ. Soy isoflavone supplements antagonize reproductive behavior and estrogen receptor ixand [3-dependent gene expression in the brain. Endocrinology 2001;142:2946-2952. 167. Kuiper GG, Lemmen JG, Carlsson B, et al. Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 1998;139:4252-4263. 168. File SE, Hartley DE, Elsabagh S, Duffy R, Wiseman H. Cognitive improvement after 6 weeks of soy supplements in postmenopausal women is limited to frontal lobe function. Menopause 2005;12: 193-201. 169. White LR, Petrovitch H, Ross GW, et al. Brain aging and midlife tofu consumption. JAm CoIINutr 2000;19:242-255. 170. Yaffe K, Sawaya G, Lieberburg I, Grady D. Estrogen therapy in postmenopausal women. JAMA 1998;279:688-695. 171. Hersh AL, Stefanick ML, Stafford RS. National use of postmenopausal hormone therapy: annual trends and response to recent evidence. JAMA 2004;291:47-53. 172. Buist DS, Newton KM, Miglioretti DL, et al. Hormone therapy prescribing patterns in the United States. Obstet Gynecol 2004; 104:1042-1050. 173. Kokmen E, Beard CM, O'Brien PC, Kurland LT. Epidemiology of dementia in Rochester, Minnesota. Mayo Clin Proc 1996;71: 275-282.
~HAPTER 2 z
Estrogens and Depression i n Women DAVID R. RUBINOW
CATHERINE A. ROCA PETER
J.
SCHMIDT
Department of Psychiatry, University of North Carolina, Chapel Hill, NC 27599 Behavioral Endocrinology Branch, National Institute of Mental Health, Bethesda, MD 20892 Behavioral Endocrinology Branch, National Institute of Mental Health, Bethesda, MD 20892
I. I N T R O D U C T I O N
In this article we will review the following: (1) the effects of estrogen on the central nervous system (CNS); (2) the relationship between perimenopause/postmenopause and depression; (3) the effect of estrogen on mood in humans; and (4) the clinical evaluation and management of depression during the perimenopause.
A role of estrogen in the regulation of mood can be inferred from descriptions of the neuromodulatory effects of estrogens, reports of the onset of depression or other mood disorders in association with the perimenopause and menopause, and observations by several authors of antidepressantlike effects of estrogen in hormone replacement regimens. During the late nineteenth century, early medical observers such as George Savage of the Bethlem Royal Hospital in London, England reported the appearance of depression in association with the perimenopause (1,2). Contemporaneously, the organotherapists reported the beneficial effects of the extract of ovarian tissue in women with perimenopauserelated hot flushes (3) and in some (but not all) women with perimenopause-related psychiatric illness (4). Subsequently, several investigators have examined the effects of estrogen replacement on mood, with mixed results (see review by Rosenthal [5]). Although the role of estrogen in the treatment of hot flushes has been consistently documented, the relationship between perimenopause-related changes in estrogen and depression (as well as estrogen's therapeutic role in depression) remains to be established. In contrast, a considerable literature documents the widespread and important neuroregulatory effects of estrogen in animals, suggesting a neurobiologic basis for several recent reports of the salutary effects of estrogen on mood (alone or in combination with traditional antidepressants). TREATMENT OF THE POSTMENOPAUSAL WOMAN
II. B A C K G R O U N D
A. Impact of Depression Major and minor depressions are the two most prevalent forms of acute depressive illness. Major depression has an estimated lifetime prevalence of 17% and affects approximately twice as many women as men (6,7). The exact prevalence of minor depression is unclear due to differences in the diagnostic criteria used across studies; however, its prevalence is thought to approximate that of major depression (8,9). Recently, major depression was identified as a leading source of disease-related disability in developed countries, and it is predicted to be a leading cause of disability worldwide by the year 2020 (second only to heart disease) (10). Minor depressions, by definition, have fewer and less severe symptoms than major depressions (11,12). Nonetheless, they are associated with disability comparable to that of major depression (13-15). In fact, major depressions of moderate severity are not distinguished from minor 307
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
308
RUBINOW ET AL.
depressions by family history (16,17), course (i.e., both major and minor depressions occur in subjects over their lifetime) (11,16), or biologic characteristics (18,19). In addition to the functional disability directly attributed to major and minor depressions, adverse medical sequelae of major depression have been identified, including increased risks for osteoporosis, the metabolic syndrome, and cardiovascular disease (20-24). Finally, if depression exists as a comorbid condition, it may increase both the morbidity and mortality of several medical illnesses, particularly heart disease (25-28).
B. Criteria for the Diagnosis of Major and Minor Depressions Despite its prevalence and associated disability, depression is underdiagnosed by health care providers and, therefore, also undertreated (29-33). Standardized criteria for diagnosing both major and minor depressions have been developed (34) to distinguish depressive symptoms, which may be multidetermined, from depressive syndromes, which have particular familial patterns, biologic features, and treatment response characteristics. The American Psychiatric Association's Diagnostic and Statistical Manual, Fourth Edition (DSM-IV) (34) specifies selected core symptoms of depression (five for major depression and three for minor depression), which must persist for at least a 2-week period, be associated with clinically significant distress or impairment in social or occupational functioning, and not be caused by medications, a medical condition (e.g., hypothyroidism), or bereavement. Structured diagnostic interviews are employed in research settings to establish the presence of a diagnosis of depression (e.g., Structured Clinical Interview for DSM-IV [SCID] (35)), and modifications of these diagnostic instruments may be used to screen patients for the presence of depression in medical settings (36,37). For example, recent studies employing one of these screening scales, the Primary Care Evaluation of Mental Disorders (PRIME-MD) (37), have identified rates of mood disorder in women of approximately 31% in primary care clinics (38) and 13% in gynecologic clinics (39). As stated earlier, mood disorders are frequently underdiagnosed by the health care provider, and in one of these studies, the current episode of mood disorder had not been recognized previously in 80% of the women (39).
C. Pathophysiology of Depression Traditionally, depression was considered to be caused by an underlying dysregulation of one or more of the classic neurotransmitter systems, and many commonly used antide-
pressant drugs were observed to influence these same systems. Reports of abnormal levels of these neurotransmitters and their metabolites in peripheral body fluids (i.e., cerebrospinal fluid [CSF], urine, plasma) and peripheral cells in depression supported this hypothesis (40). More recently, the acute depletion of either the neurotransmitters serotonin or noradrenaline/dopamine was observed to induce depression in antidepressant-treated men and women (41) as well as changes in the pattern of activation in the prefrontal cortex (42,43). Postmortem and in vivo radioligand imaging studies have identified alterations in neurotransmitter receptor levels or function in depressed subjects compared with asymptomatic controls, including decreased serotonin 1A receptors (postmortem and in vivo) and decreased alpha-2 and [3-adrenergic receptors (postmortem) (44). Mood stabilizers (e.g., lithium, valproate, carbamazepine) and antidepressants have different effects on monoamine function; however, they do influence several of the signal transduction pathways regulated by traditional neurotransmitters and antidepressants. For example, in vitro studies have reported that both mood stabilizers and antidepressants alter the function or levels of several targets of these pathways: cyclic adenosine monophosphate (camp) levels and cAMP response element binding protein (CREB) (increased), brain-derived neurotrophic factor (BDNF) (increased), extracellular signal-regulated kinase-mitigen activated protein kinase (ERK-MAPK) activity (increased), bcl-2 (increased), Wnt cascade-glycogen synthesis kinase-3 beta (GSK-3 beta) (decreased), and beta catenin (increased) (44). The well-documented regulatory roles of these molecules in cell survival and neuroplasticity were recognized and subsequently integrated into hypotheses about the pathophysiology of depression. Abnormalities in regional brain activity (e.g., blood flow and metabolism) also supported the concept that neuroplasticity could be an integral part of both the pathophysiology and treatment of mood disorders. For example, brain imaging studies have observed abnormalities in depressed patients in the activity of several brain regions considered relevant in mood regulation. Specifically, regional blood flow and metabolism in depression have been reported to be increased in the amygdala and pregenual area of the anterim cingulate gyrus and decreased in the dorsomedial and dorsoanterolateral prefrontal cortex. Decreased blood flow and metabolism have been described in the subgenual anterioi cingulate, although both are believed to represent actual increases when corrected for the decreased cortical volume in this region, which is also observed in depression. Structural imaging studies have confirmed abnormalities in the volume of similar brain regions, and postmortem studies have identified both glial and neuronal cell loss in some of these same brain regions in patients with affective disorders compared with controls (45,46).
CHAPTER 24 Estrogens and Depression in Women
D. The Course and Treatment of Depression The occurrence of a single episode of depression will increase an individual's risk of developing recurrent episodes of depression (47). Estimates of the duration of a major depressive episode range from an average of 4 to 8 months (33,48,49), with one study reporting that 50% of episodes of major depression remit within 3 months regardless of treatment (49). However, in approximately 20% of patients with major depression, the episodes become chronic with duration greater than 24 months. Treatments for both major and minor depression include psychotherapy (time-limited, focused, cognitive, or interpersonal therapy), antidepressant medications, and other somatic therapies (e.g., electroconvulsive therapy [ECT] for severe major depression) (50). Reported response rates to therapy in major depression vary across studies; however, approximately 50% of ambulatory subjects with major depression will respond after treatment with either timelimited psychotherapy or antidepressant medication (51). Nonresponders continue to experience symptoms and require additional treatment, which may include adjunctive therapy or an alternate treatment strategy.
III. M E T H O D O L O G I C PROBLEMS IN INVESTIGATIONS OF THE RELATIONSHIP BETWEEN DEPRESSION, MIDLIFE, AND REPRODUCTIVE AGING Methodologic problems in previous studies included the manner in which both reproductive status and mood syndromes were defined and characterized. Only limited confidence could be placed in the conclusions of studies that employed neither endocrinologically meaningful evaluations of ovarian aging nor standardized criteria for measuring the presence of depression.
A. Characterizing Reproductive Status Several criteria have been employed to define the reproductive status of women participating in studies of the relationship between the menopausal transition and depression. First, an age window of 45 to 55 years has been used to select perimenopausal subjects. Although the average age of the menopause is 51 years, there is considerable individual variation in the age of onset of the menopause, ranging from the early forties to the late fifties. Adopting an age window as the sole selection criteria, obviously, will lead to the selec-
309 tion of a sample of women, all of whom are at midlife but who differ in their reproductive status: some premenopausal, some perimenopausal, and some postmenopausal. Second, investigators have employed age combined with self-reports of changes in menstrual cycle length and have defined the menopause as 1 year of amenorrhea and the perimenopauseas menstrual cycle irregularity. However, self-reports of menstrual cycle irregularity cannot be used to reliably define the menopause transition. Treloar (52) observed that menstrual cycle irregularity is not confined to the perimenopause and may occur frequently during other periods of reproductive life. Moreover, Kaufert et al. (53) observed that among a sample of middle-aged women with menstrual cycle irregularity, as many women returned to normal menstrual cycle function as entered the menopause (defined by 6 months of amenorrhea) during a 3-year period of follow-up. Finally, there is a 5% to 10% probability that a woman could have menstrual bleeding even after 12 months of amenorrhea (54). The third criterion commonly used to define reproductive status has been the presence of elevated plasma gonadotropin (i.e., follicle-stimulating hormone [FSH]) levels in the context of low plasma estradiol levels. Studies have identified that the perimenopause has distinct endocrine characteristics, with the early perimenopause (e.g., high gonadotropin levels and increased estradiol secretion) differing from the late perimenopause (e.g., high gonadotropin levels and decreased estradiol secretion). Consequently, investigators have developed more precise criteria for the phases of reproductive aging, thus facilitating the collection of more homogeneous samples to better illuminate the interaction between the stage of ovarian aging and specific physiologic or behavioral end points. For example, the Stages of Reproductive Aging Workshop (STRAW) criteria for reproductive, perimenopausal (menopausal transition), and postmenopausal years, developed by Soules et al. (55), define the early perimenopause to include women with menstrual cycle irregularity (defined as a variable cycle length that differs from normal by more than 7 days) and elevated plasma FSH levels (defined as levels >- 2 standard deviations above mean values in women of reproductive age). During the late perimenopause, elevated FSH levels occur in conjunction with two or more skipped cycles and a period of amenorrhea lasting at least 60 days. This could include up to 1 year of amenorrhea, at which time the woman has entered the early postmenopause.
B. Defining Depression An equally important methodologic issue is the need for investigators to define and evaluate the presence of depression as a syndrome (a condition meeting standardized diagnostic criteria) rather than an unintegrated set of symptoms.
310 Mood symptoms differ from mood syndromes in their instruments of detection, their duration, and their impact. In general, a structured diagnostic interview (e.g., SCID [35]) is the most reliable method for assessing the presence or past occurrence of a major or minor depression. In contrast, the cross-sectional scales employed in many studies measure the severity of depressive-like symptoms but do not assess either the longitudinal persistence of a core group of depressive symptoms or the level of functional impairment, both important constituents of a depressive syndrome. Additionally, cross-sectional rating scales, such as the Center for Epidemiological Studies-Depression (CES-D), have restricted sampling intervals, consisting of a 1- to 2-week window of time prior to the date when the scale is completed. Thus, if 50% of major depressive episodes have remitted by 3 months, as reported previously (49), then the administration of a cross-sectional rating on a yearly basis, for example, could not reliably assess the occurrence of depressions during the previous year. Finally, in addition to employing standardized psychiatric diagnostic interviews for establishing the presence of current and past episodes of depression, the type of mood syndromes should be defined. For example, both major and minor depressions exist, yet many studies focus only on the presence of major depression, with the exclusion of minor depression justified on the basis of its lesser significance. However, this decision to exclude minor depression may no longer be justified given recent evidence documenting the substantial disability associated with minor depressions (see earlier discussion). Similarly, a cross-sectional rating scale "cut-off" score of 16 or greater is employed as a proxy for the presence of clinically significant depressive symptoms and as a selection criterion for both trials of antidepressant medications and some epidemiologic studies of midlife depression (56-58). However, clinical significance also has been identified for considerably lower cut-off scores on these same cross-sectional ratings. For example, Beck Depression cutoff scores of 10 or greater (below the customary cut-off of 16) are associated with a significantly increased mortality in patients with heart disease (59).
IV. EFFECTS ON ESTROGEN ON THE CENTRAL NERVOUS SYSTEM Estrogen, like other members of the steroid hormone family (e.g., progesterone, androgen, glucocorticoid, mineralocorticoid), is synthesized from cholesterol by a relatively small group of enzymes with multiple sites of action along the hormone synthetic cascade. Upon arrival at its target cell, estrogen diffuses into the cell and binds its receptor, which is located in the nucleus. The estrogen receptor is a member of a group of structurally related transcriptional factors; in other words, proteins that bind DNA or interact with one another to regulate transcription of mRNA. Amino
RUBINOW ET AL.
acid sequence and structure-function analyses of steroid receptors have identified a number of domains (i.e., sequences of amino acids) within the receptor protein, each with a particular function related to ligand binding or activation of the target genes (for review, see [60-62]). Two separate isoforms of the estrogen receptor have been identified: alpha (c~) and beta ([3) (63,64). In contrast to identified isoforms of progesterone and androgen receptors, which are posttranscriptional products, oL and [3 forms of the estrogen receptor are encoded by two separate genes, located on chromosomes 6 and 14, respectively (65). Structurally, both the oL and [3 estrogen receptors share considerable amino acid sequence homology in both the DNA binding (96%) and ligand binding (58%) domains (63). The two forms of the estrogen receptor are differentially distributed in a tissuespecific manner throughout the body, including the CNS (66-68), where estrogen receptor [3 is abundant. The significance of the presence of these different forms of the estrogen receptor is in part revealed by observations that they are differentially modulated by estradiol, elicit different transcriptional responses, and modify each other's activity (69,70). In human brains, region-specific concentrations of both estradiol and estrogen receptors ~ and [3 have been observed in postmortem studies (71-73). The actions of estrogen within the CNS are potent and widespread. As transcriptional regulators, activated estrogen receptors direct or modulate the synthesis of the synthetic and metabolic enzymes as well as the receptor proteins for many neurotransmitters and neuropeptides (74). In this fashion, estrogen regulates the activity of virtually all elements of serotonergic and cholinergic neurotransmission. For example, estrogen modulates the synthesis of serotonin (75), serotonin uptake (by regulation of the serotonin transporter) (76,77), serotonin receptor transcription (78) and density (79), and the response to serotonin stimulation (80). The neuroregulatory potential of estrogen and its receptor(s) is only partially conveyed by consideration of classical, hormone-activated, receptor-mediated effects. First, many of the transcriptional actions of estrogen are directed by tissuespecific transcriptional co-activators or co-repressors (81). Second, several compounds have been observed to activate the estrogen receptor in the absence of estrogen: phytoestrogens (e.g., genistein and daidzein); selective estrogen receptor modulators (SERMs) (e.g., raloxifene [82]), and classical neurotransmitters (e.g., dopamine) activating the estrogen receptor via the D1 receptor (62,83). Third, in some tissues the unoccupied estrogen receptor may repress genomic transcription (84), perhaps through protein-protein interactions (81). Fourth, estrogen has acute, nongenomic actions by which it modulates signal transduction and ion channel function. Thus, estradiol has been shown to acutely modulate non-N-methyl-D-aspartate (non-NMDA) glutamate receptor function in the hippocampus (85), calcium and potassium channel activity (86-88), and the MAP kinase and PI3-K/ Akt (protein kinase B) signaling cascades (89-94). Addition-
CHAPTER 24 Estrogens and Depression in Women
ally, Mobbs et al. (95) have shown that estrogen induces an isoform of phospholipase C alpha in rat brain, while we have demonstrated that estrogen in women both decreases the level and increases the inactivation of the G protein Gi alpha 2 (Manji et al., unpublished data), consistent with the observed ability of estrogen to uncouple the D2 receptor from its signal transduction mechanisms (96). Similarly, Mize et al. (97) demonstrated that estradiol uncouples 5-HT1A receptors from their associated G protein and activates protein kinases A and C in the forebrain and hippocampus by cell membrane mechanisms (97). Fifth, some actions of estrogen on brain appear to be context and developmental stage dependent. Toran-Allerand (98) has shown that estrogen has reciprocal interactions with CNS growth factors that may mediate the response to estrogen stimulation throughout development: estradiol stimulates its own receptor early in development, inhibits it during adulthood, and stimulates it again in the context of brain injury. These interactions between estradiol and growth factors are of further interest given recent reports that BDNF levels are decreased by stress and are elevated by antidepressants (99-101). The extent to which the aforementioned effects underlie or contribute to differential effects of estrogen on brain physiology, behavior, and mood across individuals is unclear but is of considerable potential importance in the etiology and treatment of depression during the perimenopause. Several types of data from studies in humans have converged to suggest that estradiol may affect mood through effects on the serotonergic system. First, a number of sexual dimorphisms have been reported in serotonergic measures in humans, including CSF 5-HIAA (increased in pain syndromes and depression in women) (102-104), 5HT2 receptor-binding capacity in brain (decreased in women) (105), whole brain serotonin synthesis (decreased in women) (106), and response to tryptophan depletion (increased in women) (107-109). Second, the neuroendocrine response to stimulation with serotonin agonists has been reported to vary, albeit inconsistently, according to menstrual cycle phase. For example, stimulated prolactin secretion during the luteal phase (compared with the follicular phase) has been observed to be augmented following mchlorophenylpiperazine (m-CPP) (110) and buspirone (111), blunted following L-tryptophan (112), and unaffected following d-fenfluramine (107). Third, several serotonergic measures may change following estrogen replacement. Sherwin et al. (113) reported the upregulation of platelet 3Himipramine binding sites (a measure of the serotonin transporter) when estrogen was administered to hypogonadal women. These data parallel observations of estradiol-stimulated upregulation of the serotonin transporter mRNA in the raphe of rats (114) but not macaques (76). In contrast, an earlier study by Best et al. (115) did not observe an estrogenrelated change in platelet imipramine binding sites or 5HT2 receptor binding in hypogonadal women. Estrogen replacement increased the prolactin response to m-CPP (but
311 not significantly) in postmenopausal women (116), although no effect of estradiol replacement was observed on m-CPP stimulated neuroendocrine response in young women compared with the response seen during gonadotropin-releasing hormone (GnRH) agonist-induced hypogonadism (117). Clearly, the nature (indirect) and multiplicity (e.g., serotonin synthesis, receptor concentration, and reuptake; response to agonist) of the serotonergic measures employed in an already scanty literature preclude any inference about the role of serotonin in possible psychotropic effects of estrogen. In addition to classic neurotransmitter systems, the activities of several neural signaling systems implicated in depression are modulated by ovarian steroids. Specifically, ovariectomy has been reported to decrease, and estradiol to both increase and decrease, BDNF levels in the forebrain and hippocampus (118,119). Estrogen also increases CREB activity (120) and trkA (121) and decreases GSK-3[3 (Wnt pathway) (122) in the rat brain, in a region-specific manner and in directions similar to those of mood-stabilizing drugs. Thus, the therapeutic potential of gonadal steroids in depression is suggested not only by their widespread actions on neurotransmitter systems but also by certain neuroregulatory actions shared by both estrogen and traditional therapies for depression (i.e., antidepressants, ECT).
V. WHAT IS THE EVIDENCE OF AN ASSOCIATION BETWEEN THE MENOPAUSAL TRANSITION AND DEPRESSION? The majority of women do not develop depression during the menopausal transition, and the perimenopause is not uniformly associated with changes in a woman's mood. In fact, epidemiologic studies examining gender and age-related differences in the 6-month to 1-year prevalence of major depression reported no increased prevalence of major depression in women at midlife (age range approximately 45 to 55 years) (47,123). Nonetheless, although the postmenopause is not associated with an increased risk for developing depression in women (57,124,125), depressive symptoms have been observed more frequently in perimenopausal women compared with postmenopausal women in several longitudinal, community-based studies (126,127) as well as in women attending gynecology clinics (128-130). Similarly, community-based surveys of the prevalence of major or minor depression also suggest that the menopausal transition is a time of increased vulnerability for depression to develop (131). The Study of Women's Health Across the Nation (SWAN) (132) employed a measure of"psychological distress" as a proxy for the syndrome of depression by requiring that core depressive symptoms (sadness, anxiety, and irritability) persist for at least 2 weeks. SWAN's initial
312 cross-sectional survey observed that perimenopausal women reported significantly more "psychological distress" than either premenopausal or postmenopausal women (defined by self-reported menstrual cycle status) (132). Two recent studies found results similar to those of SWAN. First, Freeman et al. (56) identified an increased risk for significant depression (defined by elevated CES-D scale scores and the PRIME MD [37]) during the perimenopause compared with the premenopause or postmenopause. Moreover, this association remained after adjusting for several variables, including past history of depression, severe premenstrual syndrome, poor sleep, and hot flushes. Levels of depression were increased relative to those found in the postmenopausal women; however, only 3% of the sample (approximately 10 women) was followed through to the postmenopausal phase. In a recent prospective study (133), 29 asymptomatic, premenopausal women were followed until 6 to 12 months after their last menstrual period. A fourteenfold increase in the rate of onset of depression was observed during the 24 months surrounding the final menstrual period relative to the 31 premenopausal years used as a comparison time period, suggesting an increased risk of depression in women during both the late perimenopause and the early postmenopause relative to the premenopause. The majority of the women in this study remained asymptomatic throughout the perimenopause. Nonetheless, these data suggest that events surrounding the final menstrual period may predispose some women to develop depression. Although several factors could precipitate depression in these women, endocrine events are suggested by the stage of the perimenopause (i.e., late perimenopause) during which depressions appeared. The late perimenopause is characterized by more prolonged hypogonadism than the early perimenopause, during which estradiol secretion may be increased. Indeed, the late perimenopause may represent the phase of the perimenopause most correctly characterized by estradiol "withdrawal" relative to either the postmenopause or the early perimenopause (134,135). Thus, the timing of appearance of the depressions observed suggest an endocrine mechanism related to the perimenopause (estradiol withdrawal and/or recent onset of prolonged hypogonadism) in the pathophysiology of perimenopausal depression.
VI. WHAT FACTORS INFLUENCE THE RISK FOR DEPRESSION DURING THE MENOPAUSAL TRANSITION? As described in the previous section, several epidemiologic studies have surveyed the presence of depressive symptoms in women at midlife and identified rates of depressive symptoms ranging from 8% to 40% (125,136,137). However, the samples in these studies consisted of women at midlife who were in different phases of reproductive aging,
RUBINOW E T
AL.
and symptoms often were assessed independent of the presence of clinically meaningful depressive syndromes; these findings, therefore, are not directly translatable into either prevalence figures or risk factors for depressive syndromes associated with the reproductive endocrine changes characterizing the perimenopause. Nonetheless, these studies of women who become depressed during midlife identified several variables associated with depression, including the following: previous episodes of depression (130,138), longer duration of the perimenopause (defined by menstrual cycle irregularity) (125), presence of hot flushes (125,127), retrospective reports of premenstrual dysphoria (PMD) or postpartum depression (PPD) (129,138-141), stressful life circumstances (57,58,142,143), complaints of poor health (58), history of smoking (129,140), disturbed sleep (56,136,144), reduced parity, and unmarried status (140). Many of these factors also are associated with an increased risk of developing depression during other stages of life (i.e., past history of depression, stressful life events, reports of PMD or PPD, smoking and sleep disturbance) and therefore are not specific to depression during the perimenopause. As discussed earlier, recent studies have identified the menopausal transition to be an independent risk factor for depression at midlife in women (56,132,137,145). These findings suggest that, in some women, ovarian aging and the events surrounding the perimenopause increase the vulnerability to develop depression. Finally, several proposed risk factors such as insomnia, increased stress, and complaints of poor health may be symptoms, but are not necessarily a cause, of a current depressive episode. In the study by Schmidt et al. (133), 9 women followed prospectively who developed depression during the perimenopause were not distinguished from those not developing depression by any variable previously reported in association with the onset of perimenopausal or midlife depression. There was no increased onset of depression in women with a past history of depression (diagnosed by structured clinical interview). In fact, of the depressions that were observed, a similar percentage occurred in women with no prior history of depression (6/20) as in those with a past history (3/9) (small sample size notwithstanding). Nor did the three women with histories of PPD develop perimenopauserelated depression, suggesting that the presence of one episode of a reproductive endocrine-related mood disorder (i.e., PPD) does not predict the uniform occurrence of depression during a subsequent period of hormonal change (the perimenopause). Similarly, prior studies employing retrospective reports of the onset of PMD have suggested that it is both an accompaniment and (possibly) a predictor of depression during the perimenopause (129,138-140,146). However, in the study by Schmidt et al. (the first to prospectively evaluate, using daily symptom ratings, self-reports of the onset of PMD in women entering the perimenopause), PMD only rarely accompanied perimenopausal depression. In fact, in those women who became depressed during the perimeno-
CHAPTER 24 Estrogens and Depression in Women pause, only one woman met criteria for PMD in the 4 years before the final menstrual period, and two additional women met criteria for significant premenstrual cyclicity intermittently in three to five menstrual cycles over the course of the 4 years prior to the final menstrual period. Nonetheless, recent cross-sectional data (147) suggest a higher than expected co-occurrence of prospectively confirmed PMD and perimenopausal depression. Women with perimenopausal depression (n = 70) (and who were not amenorrheic) were significantly more likely to meet criteria for PMD than asymptomatic perimenopausal controls (n = 35) (21% compared with 3%). However, despite the higher rate of PMD, the majority of women with perimenopausal depression did not meet criteria for PMD, consistent with the longitudinal observations. Thus, PMD is neither a uniform accompaniment nor a necessary antecedent of perimenopausal depression. Nevertheless, the presence of PMD in perimenopausal depression may identify a subgroup of women in whom mood disturbance is directly linked to reproductive events. Hot flushes also are frequently reported to accompany depression in the perimenopause (56,125,127,136,148) and are viewed as potentially causal. Thus, hot flushes would be hypothesized to disturb sleep and, by so doing, contribute to daytime mood symptoms, consistent with the domino theory. For this relationship to be tenable, one would expect that reported hot flushes would occur prior to the onset of depression (i.e., as a precipitant of depression). Nonetheless, data from both the SWAN study (132,137) and Freeman and colleagues (56) demonstrated that hot flushes and the perimenopause were independent risk factors for depression. Thus, both hot flushes and perimenopause-related ovarian events may impart separate risks for developing depression at midlife. In our prospective study, hot flushes were not uniformly present in all women and, when present, did not necessarily precede the depression. Eight of the nine women experiencing depression reported the onset of hot flushes at some point during the perimenopause, consistent with prior studies (56,148); however, the timing of the relationship varied considerably, ranging from several years before to several years after the onset of depression. Only four (44%) of the women who developed a depression during the perimenopause reported the onset of hot flushes proximate to the development of their depression. Thus, hot flushes appear to be neither a necessary nor a sufficient accompaniment of depression during the perimenopause, and perimenopausal depression cannot be dismissed as epiphenomenal to hot flushes. Finally, stressful life events are a frequent accompaniment of depression, and in some depressed subjects, stressful life events may contribute to the onset of depression (149-151). Similarly, stressful events have been reported in association with depressive symptoms at midlife (57,58,142,143,152,153) as well as in women with major and minor depression during the perimenopause (154). However, in the latter study, women with perimenopausal depression were not distin-
313 guished from asymptomatic perimenopausal women by reports of a greater number of exit events or personal losses. Additionally, in the prospective study by Schmidt et al. (133), the women who developed depression reported similar numbers of exit-type or unpleasant events in the 6 months preceding the onset of depression as they did at other times during the study. Indeed, exit events, if they do occur, are not necessarily associated with negative mood symptoms in women at midlife (155). Thus, although stressful events are an accompaniment of both midlife and perimenopausal depression, there is not evidence to support the concept that depressions at this time in a woman's life are caused by the "empty nest" syndrome. In summary, several factors, including perimenopausal reproductive status, are associated with developing depression at midlife. Many of these same factors are frequent accompaniments ofperimenopausal depression; however, none is uniformly present in these depressed women. Notably, a past history of depression, whether related to reproductive endocrine change (i.e., PPD or PMD) or not, fails to predict the onset of perimenopausal depression. As a caveat, however, our inability to identify predictors of the onset of depression may reflect the small sample sizes of depressed perimenopausal women examined. Future studies will clarify whether specific factors exist that predict or increase the risk of developing depression during the perimenopause, independent of those factors that increase a woman's risk for depression at other times across the life cycle.
VII. EFFICACY OF E S T R O G E N IN M O O D DISORDERS In 1934, a report of one of the first controlled trials of estrogen (theelin injections) was published in JAMA and described the superior efficacy of estrogen compared with placebo (155a). Subsequently, some (156-159), but not all (160,161), clinic-based, placebo-controlled trials demonstrated the salutary effects of estradiol replacement on mood symptoms in perimenopausal women with high scores on depressive symptom rating scales. These studies in perimenopausal women are consistent with reports that estradiol replacement enhances mood in women after the surgical menopause (162-164). Additionally, Ditkoff et al. (165) reported estradiol to have an acute mood-enhancing effect in relatively asymptomatic postmenopausal women. Traditional antidepressants do not generally elevate mood in asymptomatic subjects. Thus, the report by Ditkoff et al. (165) suggests that the mood-elevating effects of estrogen may be mediated by mechanisms other than those thought to be involved in the traditional antidepressant response. Finally, results of some (166-168), but not all (169), studies suggest that estradiol replacement in some women augments the mood-stabilizing and antidepressant effects of traditional psychotropic medications. It is noteworthy, how-
3
1
4
R
u
B
I
ever, that the superior efficacy of estradiol augmentation in the study by Schneider et al. (167) is seen relative to results with estradiol and placebo, not antidepressant alone. Two major confounds exist in studies that attempt to assess the mood-enhancing effects of estrogen: (1) the intimate relationship between estrogen deficiency-related vasomotor symptoms and mood disturbance and (2) the failure to study women who meet standardized diagnostic criteria for the syndrome of depression. For example, as noted earlier, the salutary effects of estrogen on mood symptoms could simply reflect the elimination of thermoregulatory dyscontrol and accompanying sleep disturbances. Moreover, depressive symptoms may be multidetermined, whereas depression, the syndrome, has specific biologic concomitants and treatment response characteristics. Recently, three double-blind, placebo-controlled trials, employing similar methodologies and identical preparations of estradiol (i.e., 17{3estradiol), have examined the psychotropic efficacy of estradiol in perimenopausal and postmenopausal women with major and minor depressions (170-172). First, the therapeutic efficacy of estradiol (1713-estradiol) was examined in a double-blind, placebo-controlled trial in 34 late perimenopausal women (defined by STRAW criteria [55]) who also met standardized diagnostic criteria for major and minor depression (170). After 3 weeks of estradiol, depression rating scale scores were significantly decreased compared with baseline scores and significantly lower than scores in the women receiving placebo. Moreover, those women initially randomized to placebo showed a similarly significant decrease in depression ratings after crossover to 3 weeks of estradiol. A full or partial therapeutic response was seen in 80% of subjects on estradiol and in 22% of those on placebo, consistent with the observed effect size in a meta-analysis of studies examining estrogen's effects on mood (173). The therapeutic response to estrogen was observed in both major and minor depression as well as in women with and without hot flushes. Finally, neither baseline nor post-treatment estradiol levels predicted the observed therapeutic response. These data suggest that estrogen's effect on depression is not solely a product of its ability to reduce the distress of hot flushes, consistent with recent community-based cross-sectional surveys (132). These findings also are consistent with data from Montgomery et al. (159) and Saletu et al. (175), which suggest the beneficial effects of estradiol on mood in perimenopausal women reporting depressive symptoms. A second randomized, double-blind, placebo-controlled study by Soares et al. (171) confirmed the observations of Schmidt et al. (170). Soares et al. reported a significant and beneficial effect of estradiol replacement compared with placebo in women with perimenopause-related major depression (as defined by the PRIME MD) (30) and additionally reported that baseline plasma estradiol levels did not predict response to estrogen treatment (171). In contrast, a recent
N
O
W
ET AL.
study in depressed women who were 5 to 10 years postmenopause, using a design similar to that employed in perimenopausal women (170,171), failed to observe a significant antidepressant effect of estradiol relative to placebo (172). The evidence that younger perimenopausal but not older postmenopausal women respond to estrogen therapy suggests that those mood disorders occurring in perimenopausal women are caused by changes in hormones (e.g., withdrawal or fluctuations) rather than prolonged ovarian steroid deficiency. Despite evidence for an acute antidepressant effect of estradiol in hypogonadal women, the therapeutic role of estrogen in perimenopause-related depression remains to be determined. The possibility that estrogen's acute antidepressant effects may not be maintained long term has been described by Butler and Lewis (176). In their report (176), a woman with first-onset depression during the perimenopause responded initially to estrogen replacement; however, she later experienced a relapse of her depression on estrogen and required treatment with traditional antidepressants. Trials evaluating the long-term efficacy of estrogen in the maintenance therapy of depression are required prior to recommending estrogen as a specific therapeutic agent for perimenopause-related depression. Similarly, no study has evaluated the efficacy of traditional antidepressants in women with depression during the perimenopause.
VIII. SUGGESTED EVALUATION AND MANAGEMENT OF PERIMENOPAUSERELATED DEPRESSION The management of mood and behavioral disturbances during the perimenopause requires determination of the symptoms experienced and the hormonal context in which they appear. As a complement to a careful evaluation, longitudinal monitoring of symptoms on a daily basis may provide invaluable information about the severity, stability, and pattern of symptom experience. Both affective and somatic symptoms (e.g., vaginal dryness, hot flushes,) should be followed. If not done previously, the presence of the perimenopause and hypoestrogenism should be documented with FSH and estradiol measures. As part of our operational criteria, we have required three of four serial FSH levels >14 IU/L for the perimenopause. Although estradiol levels <60 pg/ml are consistent with decreased ovarian function, levels >60 pg/ml may nonetheless appear in the presence of markedly elevated FSH levels that suggest ovarian insensitivty. The therapy selected for a major depressive disorder during the perimenopause will depend on the nature and severity of the somatic symptoms and the type of hormone replacement therapy (if present). In perimenopausal women presenting with depression, the presence of distressing signs of estro-
CHAPTER 24 Estrogens and Depression in Women gen deficiency, such as vaginal dryness or hot flushes, should lead to consideration of a trial of hormone replacement as the first approach, unless relative contraindications to estrogen replacement exist. Alternatively, if perimenopausal somatic symptoms are minimal despite laboratory evidence of hypoestrogenism and mood symptoms are moderate to severe, then the choice of hormone replacement versus antidepressant therapy may best be informed by such factors as past personal history of depression, family history of affective disorder, severity of affective symptoms, or the presence of contraindications to estrogen replacement. Finally, clinicians should appreciate that perimenopausal depression may present in the absence of typical perimenopause-related somatic symptoms, such as hot flushes or vaginal dryness. Thus, somatic symptoms are neither necessary nor sufficient for the production of depression during the perimenopause. Moreover, as demonstrated previously, the absence of hot flushes in depressed women who are endocrinologically confirmed to be perimenopausal does not predict the lack of efficacy of estrogen in the treatment of depression. The relationship between the onset of mood symptoms and the initiation of hormone replacement therapy should be determined for at least two reasons. First, mood and behavioral symptoms may appear with inadequate estrogen replacement and can remit with appropriate dosage adjustments or with a change to an alternative form of estrogen replacement. Adjustment of hormone replacement therapy is therefore recommended before adjunctive psychopharmacotherapy is considered. Second, mood and behavioral symptoms may directly result from the hormone replacement therapy. Specifically, cyclic mood and behavioral symptoms have been reported in association with sequential hormone replacement in some (177-179) but not all (180,181) studies and may remit with a change in the replacement regimen from sequential to continuous combination therapy. Alternatively, if addition of progesterone consistently precipitates adverse mood changes, estrogen replacement alone may be attempted if accompanied by appropriate monitoring of the endometrium. Symptoms of depression and loss of libido consequent to ovarian failure may nonetheless be responsive to antidepressant therapy, and this option should be considered (particularly in patients with more severe or disabling symptoms) regardless of the strength of the association between ovarian dysfunction and affective disorder.
IX. FUTURE DIRECTIONS Several new methodologic issues have been identified that may predict a differential response to changes in ovarian hormones (either endogenous or exogenous), including phase of reproductive aging (i.e., late perimenopause or early menopause versus 5 to 10 years postmenopause), presence of
315 menopausal symptoms, the duration of hypogonadism prior to receiving estrogen therapy, and genetic polymorphisms that underlie differences in steroid responsivity. A differential response to estradiol was originally reported by Appleby (182), with depressive symptoms in perimenopausal but not postmenopausal women responding to estradiol therapy under randomized, placebo-controlled conditions (observations confirmed by several recent randomized, controlled trials employing standardized psychiatric diagnostic interviews to establish the presence of depression [170-172]). Similarly, literature reviews and meta-analyses by Yaffe et al. (183) and LeBlanc et al. (184) concluded that the benefits of ovarian hormone therapy on cognitive function were limited to perimenopausal women compared with postmenopausal women and suggested that the beneficial effects of hormone therapy were secondary to the concurrent improvement in perimenopausal symptoms. Similarly, studies in both animals and humans suggest that a short duration of hypogonadism prior to initiation of estrogen therapy is associated with beneficial effects on measures of both cognition (185-187) and atherosclerotic plaque formation and cardiac function (188,189). These findings are consistent with the observed differences in treatment response in perimenopausal or recently menopausal versus older postmenopausal women (the former are more likely to be symptomatic) (128,129). Additionally, these findings introduce the concept of the critical window for the efficacy of estrogen therapy. For example, nonhuman primate studies have shown that initiation of estrogen is associated with cardioprotection when administered immediately after oophorectomy but not after 30 months (approximately 6 human years) (188). This critical window construct has been proposed as an explanation for some of the discrepant findings between the observational studies (many of which included younger, more symptomatic women) and the recent randomized controlled trials in the Women's Health Initiative (WHI) (190,191), which principally included older asymptomatic women (187,192). Finally, independent of stage of reproductive life or duration of hypogonadism, several studies employing both plasma lipid levels and cognitive outcome measures suggest that the effects of sex steroids may be influenced by the presence of polymorphisms in specific steroid receptors (193,194). Pursuit and refinement of these predictors of response to hormone therapy should define the optimal parameters for the use of hormone therapy in the treatment of perimenopausal mood disorders.
References 1. SavageGH. Some mental disorders associatedwith the menopause. Lancet 1893;ii:1128. 2. SavageGH. Some neuroses of the climacteric and summaryof climacteric cases at Bethlem, from July 1, 1888, to June 30, 1893. Tr Med Soc Lond 1893;xvii:31-47.
3
1
6
R
u
B
3. Fosbery WHS. Severe climacteric flushings successfully treated by ovarian extract. Br MedJ 1897;1:1039. 4. Easterbrook CC. Organo-therapeutics in mental diseases. Br MedJ 1900;2:813-823. 5. Rosenthal S. The involutional depressive syndrome. Am J Psychiatry 1968;124:21-35. 6. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psychiatry 1994;51:8-19. 7. Kessler RC, McGonagle KA, Nelson CB, et al. Sex and depression in the National Comorbidity Survey II: cohort effects.JAffective Disord 1994;30:15-26. 8. Johnson J, Weissman MM, Klerman GL. Service utilization and social morbidity associated with depressive symptoms in the community. JAMA 1992;267:1478-1483. 9. Kessler RC, Zhao S, Blazer DG, et al. Prevalence, correlates, and course of minor depression and major depression in the national comorbidity survey.JAffective Disord 1997;45:19-30. 10. Lopez AD, Murray CCJL. The global burden of disease, 1990-2020. Nature Med 1998;4:1241-1243. 11. Judd LL, Rapaport MH, Paulus MP, et al. Subsyndromal symptomatic depression: a new mood disorder? J Clin Psychiatry 1994;55: 18-28. 12. Rapaport MH, Judd LL, Schettler PJ, et al. A descriptive analysis of minor depression. Am J Psychiatry 2002;159:637-643. 13. Judd LL, Schettler PJ, Akiskal HS. The prevalence, clinical relevance, and public health significance of subthreshold depressions. Psychiatr Clin North Am 2002;25:685-698. 14. Broadhead WE, Blazer DG, George LK, et al. Depression, disability days, and days lost from work in a prospective epidemiologic survey. JAMA 1990;264:2524-2528. 15. Judd LL, Paulus MP, Wells KB, et al. Socioeconomic burden of subsyndromal depressive symptoms and major depression in a sample of the general populafion. Am J Psychiatry 1996;153:1411-1417. 16. Angst J, Merikangas K. The depressive spectrum: diagnostic classification and course. JAffective Disord 1997;45:31-40. 17. Kendler KS, Gardner CO Jr. Boundaries of major depression: an evaluation of DSM-IV criteria. Am J Psychiatry 1998;155:172-177. 18. Akiskal HS, Judd LL, Gillin C, et al. Subthreshold depressions: clinical and polysomnographic validation of dysthymic, residual and masked forms. JAffective Disord 1997;45:53-63. 19. Kumar A, Jin Z, Bilker W, et al. Late-onset minor and major depression: early evidence for common neuroanatomical substrates detected by using MRI. ProcNatlAcad Sci U SA 1998;95:7654-7658. 20. Barefoot JC, Schroll M. Symptoms of depression, acute myocardial infarction, and total mortality in a community sample. Circulation 1996;93:1976-1980. 21. Carney RM, Freedland KE, Miller GE, et al. Depression as a risk factor for cardiac mortality and morbidity: a review of potential mechanisms. J Psychosom Res 2002;53:897-902. 22. Gold PW, Chrousos GP. The endocrinology of melancholic and atypical depression: relation to neurocircuitry and somatic consequences. ProcAssocAm Physicians 1998;111:22-34. 23. Michelson D, Stratakis C, Reynolds J, et al. Bone mineral density in women with depression. NEnglJMed 1996;335:664-666. 24. Cizza G, Ravn P, Chrousos GP, et al. Depression: a major, unrecognized risk factor for osteoporosis? Trends Endocrinol Metab 2001;12: 198-203. 25. Carney RM, Rich MW, Freedland KE, et al. Major depressive disorder predicts cardiac events in patients with coronary artery disease. Psychosom ivied 1988;50:627-633. 26. Frasure-Smith N, Lesperance F, Talajic M. Depression following myocardial infarction: impact on 6-month survival.JAMA 1993;270: 1819-1825.
I
N
O
W
ET
AL,
27. Wassertheil-Smoller S, Shumaker S, Ockene J, et al. Depression and cardiovascular sequelae in postmenopausal women. The Women's Health Initiative (WHI). Arch Intern Med 2004;164:289-298. 28. Schulz R, Beach SR, Ires DG, et al. Association between depression and mortality in older adults. The Cardiovascular Health Study. Arch Intern Med 2000;160:1761-1768. 29. Sherboume CD, Wells KB, Hays RD, et al. Subthreshold depression and depressive disorder: clinical characteristics of general medical and mental health specialty outpatients. Am J Psychiatry 1994;151: 1777-1784. 30. Spitzer RL, Kroenke K, Linzer M, et al. Health-related quality of life in primary care patients with mental disorders: results from the PRIME-MD 1000 study.JAMA 1995;1274:1511-1517. 31. Wang PS, Berglund P, Kessler RC. Recent care of common mental disorders in the United States. J Gen Intern Med 2000;15:284-292. 32. Greden JF. The burden of recurrent depression: causes, consequences, and future prospects. J Clin Psychiatry 2001;62:5-9. 33. Kessler RC, Berglund P, Demler O, et al. The epidemiology of major depressive disorder. Results from the National Comorbidity Survey Replication (NCS-R). JAMA 2003;289:3095-3105. 34. Diagnostic and statistical manual of mental disorders,fourth edition. Washington, DC: American Psychiatric Association, 1994. 35. First MB, Spitzer RL, Gibbon M, et al. Structured clinical interview for DSM-IV Axis I disorders-patient edition. New York: Biometrics Research Department, New York State Psychiatric Institute, 1996. 36. Zimmerman M, Mattia JL. The reliability and validity of a screening questionnaire for 13 DSM-IV Axis I disorders (the Psychiatric Diagnostic Screening Questionnaire) in psychiatric outpatients. J Clin Psychiatry 1999;60:677-683. 37. Spitzer RL, Williams JBW, Kroenke K, et al. Utility of a new procedure for diagnosing mental disorders in primary care: the PRIMEMD 1000 study.JAMA 1994;272:1749-1756. 38. Linzer M, Spitzer R, Kroenke K, et al. Gender, quality of life, and mental disorders in primary care: results from the PRIME-MD 1000 study. Am J Med 1996;101:526-533. 39. Spitzer RL, Williams JBW, Kroenke K, et al. Validity and utility of the PRIME-MD patient health questionnaire in assessment of 3000 obstetric-gynecologic patients: the PRIME-MD patient health questionnaire obstetrics-gynecology study. Am J Obstet Gynecol 2000; 183:759. 40. Potter WZ, Manji HK. Catecholamines in depression: an update. Clin Chem 1994;40:279-282. 41. Booij L, Van der Does AJW, Riedel WJ. Monoamine depletion in psychiatric and healthy populations: review. Mol Psychiatry 2003;8: 951-973. 42. Neumeister A, Nugent AC, Waldeck T, et al. Neural and behavioral responses to tryptophan depletion in unmedicated patients with remitted major depressive disorder and controls. Arch Gen Psychiatry 2004;61:765-773. 43. Bremner JD, Vythilingam M, Ng CK, et al. Regional brain metabolic correlates of ~-methylparatyrosine-induced depressive symptoms: implications for the neural circuitry of depression. JAMA 2003;289: 3125-3134. 44. Manji HK, Drevets WC, Charney DS. The cellular neurobiology of depression. Nature Med 2001;7:541-547. 45. Drevets WC. Neuroimaging studies of mood disorders. Biol Psychiatry 2000;48:813-829. 46. Drevets WC. Neuroimaging and neuropathological studies of depression: implications for the cognitive-emotional features of mood disorders. Curr Opin Neurobio12001;11:240-249. 47. Kessler RC, McGonagle KA, Swartz M, et al. Sex and depression in the National Comorbidity Survey I: lifetime prevalence, chronicity and recurrence.JAffective Disord 1993;29:85-96.
CHAPTER 24 Estrogens and Depression in W o m e n 48. Patten SB, Lee RC. Refining estimates of major depression incidence and episode duration in Canada using a Monte Carlo Markov model. Med Decis Making 2004;24:351-358. 49. Spijker J, de Graaf R, Bijl RV, et al. Duration of major depressive episodes in the general population: results from The Netherlands Mental Health Survey and Incidence Study (NEMESIS). BrJ Psychiatry 2002;181:208-213. 50. Nemeroff CB, Owens MJ. Treatment of mood disorders. Nat Neurosci 2002;5:1068-1070. 51. Fava M, Kendler KS. Major depressive disorder. Neuron 2000;28: 335-341. 52. Treloar AE. Menstrual cyclicity and the pre-menopause. Maturitas 1981;3:249-264. 53. Kaufert PA, Gilbert P, Tate R. Defining menopausal status: the impact of longitudinal data. Maturitas 1987;9:217-226. 54. Guthrie J, Dennerstein L, Burger H. How reliably does 12-month amenorrhea define final menstrual period? Data from a longitudinal study. Letter to the editor. Climacteric 2002;5:92. 55. Soules MR, Sherman S, Parrott E, et al. Executive summary: Stages of Reproductive Aging Workshop (STRAW). Fertil Steril 2001; 76:874-878. 56. Freeman EW, Sammel MD, Liu L, et al. Hormones and menopausal status as predictors of depression in women in transition to menopause. Arch Gen Psychiatry 2004;61:62-70. 57. McKinlay JB, McKinlay SM, Brambilla D. The relative contributions of endocrine changes and social circumstances to depression in midaged women. J Health Soc Behav 1987;28:345-363. 58. Woods NF, Mitchell ES. Pathways to depressed mood for midlife women: observations from the Seattle midlife women's health study. Res Nurs Health 1997;20:119-129. 59. Frasure-Smith N, Lesperance F, Talajic M. Depression and 18-month prognosis after myocardial infarction. Circulation 1995;91:999-1005. 60. Williams GR, Franklyn JA. Physiology of the steroid-thyroid hormone nuclear receptor superfamily. Baillieres Clin Endocrinol Metab 1994;8:241-266. 61. Evans RM. The steroid and thyroid hormone receptor superfamily. Science 1988;240:889-895. 62. Tsai M-J, O'Malley BW. Molecular mechanisms of action of steroid/ thyroid receptor superfamily members, dnnu Rev Biochem 1994;63: 451-486. 63. Mosselman S, Polman J, Dijkema R. ER[3: identification and characterization of a novel human estrogen receptor. FEBS Lett 1996;392: 49-53. 64. Kuiper GGJM, Enmark E, Pelto-Huikko M, et al. Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natleqcad Sci U S A 1996;93:5925-5930. 65. Enmark EP-HM, Grandien K, Lagercrantz S, et al. Human estrogen receptor 13-gene structure, chromosomal localization, and expression pattern. J Clin Endocrinol Metab 1997;82:4258-4265. 66. Jones BM, Jones MK. Women and alcohol: intoxication, metabolism and the menstrual cycle. In: Greenblatt M, Schuckit M, eds. Alcoholism problems in women and children. New York: Grune and Stratton, 1976. 67. Li X, Schwartz PE, Rissman EF. Distribution of estrogen receptor[~-like immunoreactivity in rat forebrain. Neuroendocrinology 1997;66: 63-67. 68. Shughrue PJ, Komm B, Merchenthaler I. The distribution of estrogen receptor- [3 mRNA in the rat hypothalamus. Steroids 1996;61: 678-681. 69. Pettersson K, Grandien K, Kuiper GGJM, et al. Mouse estrogen receptor 13forms estrogen response element-binding heterodimers with estrogen receptor or. Mol Endocrino11997;11:1486-1496. 70. Clarke CH, Cunningham KA, Bland DA, et al. Sex hormone regulation of estrogen receptor (ERot and ER13) mRNA in female rat brain. Abstr 27th Annu Meeting Soc Neurosci 1997;23:1501.
317 71. Bixo M, Backstrom T, Winblad B, et al. Estradiol and testosterone in specific regions of the human female brain in different endocrine states. J Steroid Biochem Mol Bid 1995;55:297-303. 72. Kruijver FPM, Balesar R, Espila AM, et al. Estrogen-receptor-[3 distribution in the human hypothalamus: similarities and differences with ERc~ distribution. J Comp Neuro12003;466:251-277. 73. Taylor AH, A1-Azzawi E Immunolocalisation of oestrogen receptor beta in human tissues. J Ado/Endocrino12000;24:145-155. 74. Ciocca DR, Vargas Roig LM. Estrogen receptors in human nontarget tissues: biological and clinical implications. Endocr Rev 1995;16: 35-62. 75. Cohen IR, Wise PM. Effects of estradiol on the diurnal rhythm of serotonin activity in microdissected brain areas of ovariectomized rats. Endocrinology 1988;122:2619-2625. 76. Pecins-Thompson M, Brown NA, Bethea CL. Regulation of serotonin re-uptake transporter mRNA expression by ovarian steroids in rhesus macaques. Brain Res Ado/Brain Res 1998;53:120. 77. Fink G, Sumner BEH. Oestrogen and mental state. Nature 1996; 383:306. 78. Sumner BEH, Fink G. Estrogen increases the density of 5-hydroxytryptamine2a receptors in cerebral cortex and nucleus accumbens in the female rat. J Steroid Biocbem Mol Bio/1995;54:15-20. 79. Fischette CT, Biegon A, McEwen BS. Sex differences in serotonin 1 receptor binding in rat brain. Science 1983;222:333-335. 80. Matsuda T, Nakano Y, Kanda T, et al. Gonadal hormones affect the hypothermia induced by serotonin 1A (5-HT1A) receptor activation. Life Sci 1991;48:1627-1632. 81. Katzenellenbogen JA, O'Malley BW, Katzenellenbogen BS. Tripartite steroid hormone receptor pharmacology: interaction with multiple effector sites as a basis for the cell- and promoter-specific action of these hormones. Ado/Endocrino11996;10:119-131. 82. Yang NN, Venugopalan M, Hardikar S, et al. Identification of an estrogen response element activated by metabolites of 1713-estradiol and raloxifene. Science 1996;273:1222-1225. 83. Power RF, Mani SK, Codina J, et al. Dopaminergic and ligand-independent activation of steroid hormone receptors. Science 1991;254: 1636-1639. 84. Tzukerman M, Zhang X, Hermann T, et al. The human estrogen receptor has transcriptional activator and repressor functions in the absence of ligand. New Biologist 1990;2:613-620. 85. Wong M, Moss RL. Long-term and short-term electrophysiological effects of estrogen on the synaptic properties of hippocampal CA1 neurons. J Neurosci 1992;12:3217-3225. 86. Mermelstein PG, Becker JB, Surmeier DJ. Estradiol reduces calcium currents in rat neostriatal neurons via a membrane receptor. JNeurosci 1996;16:595-604. 87. Valverde MA, Rojas P, Amigo J, et al. Acute activation of Maxi-K channels (hSlo) by estradiol binding to the b subunit. Science 1999;285:1929-1931. 88. Zhang L, Sukhareva M, Barker JL, et al. Direct binding of estradiol enhances Slack (sequence like a calcium-activated potassium channel) channels' activity. Neuroscience 2005;131:275-282. 89. Watters JJ, Campbell JS, Cunningham MJ, et al. Rapid membrane effects of steroids in neuroblastoma cells: effects of estrogen on mitogen activated protein kinase signalling cascade and c-los immediate early gene transcription. Endocrinology 1997;138:4030-4033. 90. Singh M, $6t~16 G Jr, Guan X, et al. Estrogen-induced activation of mitogen-activated protein kinase in cerebral cortical explants: convergence of estrogen and neurotrophin signaling pathways. J Neurosci 1999;19:1179-1188. 91. Razandi M, Pedram A, Park ST, et al. Proximal events in signaling by plasma membrane estrogen receptors. J Bid Chemistry 2003;278: 2701-2712. 92. Zhang L, Li B, Zhao W, et al. Sex-related differences in MAPKs activation in rat astrocytes: effects of estrogen on cell death. MolBrain Res 2002;103:1-11.
318
93. Zhang L, Rubinow DR, Xaing G, et al. Estrogen protects against beta-amyloid-induced neurotoxicity in rat hippocampal neurons by activation of Akt. Neuro Report 2001;12:1919-1923. 94. Honda K, Sawada H, Kihara T, et al. Phosphatidylinositol 3-kinase mediates neuroprotection by estrogen in cultured cortical neurons. J Neurosci Res 2000;60:321-327. 95. Mobbs CV, Kaplitt M, Kow LM, et al. At the cutting edge: PLC-c~: a common mediator of the action of estrogen and other hormones? Mol Cell Endocrino11991;80:C187-C191. 96. Maus M, Bertrand P, Drouva S, et al. Differential modulation of D1 and D2 dopamine-sensitive adenylate cyclases by 1713-estradiol in cultured striatal neurons and anterior pituitary cells. J Neurochem 1989;52:410-418. 97. Mize AL, Young LJ, Alper RH. Uncoupling of 5-HT1A receptors in the brain by estrogens: regional variations in antagonism by ICI 182,780. Neuropharmacology2003;44:584-591. 98. Toran-Allerand CD. Developmental interactions of estrogens with the neurotrophins and their receptors. In: Micevych P, Hamer RP, eds. Neurobiological effects of sex steroid hormones. Cambridge: Cambridge University Press, 1994. 99. Smith MA, Cizza G. Stress-induced changes in brain-derived neurotrophic factor expression are attenuated in aged Fischer 344/N rats. NeurobiolAging 1996;17:859-864. 100. Duman RS, Heninger GR, Nestler EJ. A molecular and cellular theory of depression. Arch Gen Psychiatry 1997;54:597-606. 101. Siuciak JA, Lewis DR, Wiegand SJ, et al. Antidepressant-like effect of brain-derived neurotrophic factor (BDNF). Pharmacol Biochem Behav 1997;56:131-137. 102. Knotting Lv, Oreland L, H~iggendal J, et al. Relationship between platelet MAO activity and concentrations of 5-HIAA and HVA in cerebrospinal fluid in chronic pain patients.JNeural Transm [Gen Sect] 1986;66:37-46. 103. Young SN, Gauthier S, Anderson GM, et al. Tryptophan, 5-hydroxyindoleacetic acid and indoleacetic acid in human cerebrospinal fluid: interrelationships and the influence of age, sex, epilepsy and anticonvulsant drugs. J Neurol Neurosurg Psychiatry 1980;43:438-445. 104. Agren H, Mefford IN, Rudorfer MV, et al. Interacting neurotransmitter systems. A non-experimental approach to the 5-HIAA-HVA correlation in human CSE J Psychiatr Res 1986;20:175-193. 105. Biver F, Lotstra F, Monclus M, et al. Sex difference in 5HT2 receptor in the living human brain. NeurosciLett 1996;204:25-28. 106. Nishizawa S, Benkelfat C, Young SN, et al. Differences between males and females in rates of serotonin synthesis in human brain. Proc NatlAcad Sci U S A 1997;94:5308-5313. 107. Ellenbogen MA, Young SN, Dean P, et al. Mood response to acute tryptophan depletion in healthy volunteers: sex differences and temporal stability. Neuropsychopharmacology 1996;15:465-474. 108. Anderson IM, Parry-Billings M, Newsholme EA, et al. Dieting reduces plasma tryptophan and alters brain 5-HT function in women. PsycholMed 1990;20:785-791. 109. Walsh AES, Oldman AD, Franklin M, et al. Dieting decreases plasma tryptophan and increases the prolactin response to d-fenfluramine in women but not men.JAffective Disord 1995;33:89-97. 110. Su T-P, Schmidt PJ, Danaceau M, et al. Effect of menstrual cycle phase on neuroendocrine and behavioral responses to the serotonin agonist m-chlorophenylpiperazine in women with premenstrual syndrome and controls. J Clin EndocrinolMetab 1997;82:1220-1228. 111. Young EA. The role of gonadal steroids in hypothalamic-pituitaryadrenal axis regulation. Crit Rev Neurobio11995;9:371-381. 112. Bancroft J, Cook A, Davidson D, et al. Blunting of neuroendocrine responses to infusion of L-tryptophan in women with perimenstrual mood change. PsycholMed 1991;21:305-312. 113. Sherwin BB, Suranyi-Cadotte BE. Up-regulatory effect of estrogen on platelet 3H-imipramine binding sites in surgically menopausal women. Bid Psychiatry 1990;28:339-348.
RUBINOW ET AL. 114. McQ.ueen JK, Wilson H, Fink G. Estradiol-17~3 increases serotonin transporter (SERT) mRNA levels and the density of SERT-binding sites in female rat brain. Brain Res Mol Brain Res 1997;45:13-23. 115. Best NR, Barlow DH, Rees MP, et al. Lack of effect of oestradiol implant on platelet imipramine and 5-HT2 receptor binding in menopausal subjects. Psychopharmacology 1989;98:561. 116. Halbreich U, Rojansky N, Palter S, et al. Estrogen augments serotonergic activity in postmenopausal women. Bid Psychiatry 1995;37: 434-441. 117. Schmidt PJ, Raju J, Danaceau M, et al. The effects of gender and gonadal steroids on the neuroendocrine and temperature response to m-Chlorophenylpiperazine in Ieuprolide-induced hypogonadism in women and men. Neuropsychopharmacology2002;27:800-812. 118. Sohbrabji F, Miranda RC, Toran-Allerand CD. Estrogen differentially regulates estrogen and nerve growth factor receptor mRNAs in adult sensory neurons. JNeurosci 1994;14:459-471. 119. Cavus I, Duman RS. Influence of estradiol, stress, and 5-HT2A agonist treatment on brain-derived neurotrophic factor expression in female rats. Bid Psychiatry 2003;54:59-69. 120. Zhou Y, Watters JJ, Dorsa DM. Estrogen rapidly induces the phosphorylation of the cAMP response element binding protein in rat brain. Endocrinology 1996;137:2163-2166. 121. Sohrabji F, Greene LA, Miranda RC, et al. Reciprocal regulation of estrogen and NGF receptors by their ligands in PC12 cells. J Neurobid 1994;25:974-988. 122. Cardona-Gomez P, Perez M, Avila J, et al. Estradiol inhibits GSK3 and regulates interaction of estrogen receptors, GSK3, and betacatenin in the hippocampus. Mol Cell Neurosci 2004;25:363-373. 123. Weissman MM, LeafPJ, Tischler GL, et al. Affective disorders in five United States communities. PsycholMed 1988;18:141-153. 124. Kaufert PA, Gilbert P, Tate R. The Manitoba project: a re-examination of the link between menopause and depression. Maturitas 1992; 14:143-155. 125. Avis NE, Brambilla D, McKinlay SM, et al. A longitudinal analysis of the association between menopause and depression: results from the Massachusetts Women's Health Study. Ann Epidemiol 1994;4: 214-220. 126. Matthews KA. Myths and realities of the menopause. PsychosomMed 1992;54:1-9. 127. Hunter M. The South-East England longitudinal study of the climacteric and postmenopause. Maturitas 1992;14:117-126. 128. Stewart DE, Boydell K, Derzko C, et al. Psychologic distress during the menopausal years in women attending a menopause clinic. Int J Psychiatry Med 1992;22:213-220. 129. Dennerstein L, Smith AMA, Morse C, et al. Menopausal symptoms in Australian women. MedJAust 1993;159:232-236. 130. Hay AG, Bancroft J, Johnstone EC. Affective symptoms in women attending a menopause clinic. Br J Psychiatry 1994;164:513-516. 131. Maartens LWF, Knottnerus JA, Pop VJ. Menopausal transition and increased depressive symptomatology: a community based prospective study. Maturitas 2002;42:195-200. 132. Bromberger JT, Meyer PM, Kravitz HM, et al. Psychologic distress and natural menopause: a multiethnic community study. Am J Public Health 2001;91:1435-1442. 133. Schmidt PJ, Haq NA, Rubinow DR. A longitudinal evaluation of the relationship between reproductive status and mood in perimenopausal women. Am J Psychiatry 2004;161:2238-2244. 134. Santoro N, Brown JR, Adel T, et al. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab 1996;81:1495-1501. 135. Burger HG, Dudley EC, Hopper JL, et al. The endocrinology of the menopausal transition: a cross-sectional study of a population-based sample. J Clin EndocrinolMetab 1995;80:3537-3545.
CHAPTER 24 Estrogens and Depression in W o m e n 136. Dennerstein L, Dudley EC, Hopper JL, et al. A prospective populationbased study of menopausal symptoms. Obstet Gynecd 2000;96:351-358. 137. Avis NE, Stellato R, Crawford S, et al. Is there a menopausal syndrome? Menopausal status and symptoms across racial/ethnic groups. Soc Sci Med 2001;52:345-356. 138. Stewart DE, Boydell KM. Psychologic distress during menopause: associations across the reproductive cycle. Int J Psychiatry Med 1993;23:157-162. 139. Soares CD, Almeida OR Depression during the perimenopause.Arch Gen Psychiatry 2001;58:306. 140. Harlow BL, Cohen LS, Otto MW, et al. Prevalence and predictors of depressive symptoms in older premenopausal women: the Harvard study of mood and cycles.Arch Gen Psychiatry 1999;56:418-424. 141. Woods NF, Mitchell ES. Patterns of depressed mood in midlife women: observations from the Seattle midlife women's health study. Res Nurs Health 1996;19:111-123. 142. Dennerstein L, Lehert P, Burger H, et al. Mood and the menopausal transition. JNerv Ment Dis 1999;187:685-691. 143. Mitchell ES, Woods NF. Symptom experiences of midlife women: observations from the Seattle midlife women's health study. Maturitas 1996;25:1-10. 144. Owens JF, Matthews KA. Sleep disturbance in healthy middle-aged women. Maturitas 1998;30:41-50. 145. Bromberger JT, Assmann SF, Avis NE, et al. Persistent mood symptoms in a multiethnic community cohort of pre- and perimenopausal women. Am J Epidemio12003;158:347-356. 146. Freeman EW, Sammel MD, Rinaudo PJ, et al. Premenstrual syndrome as a predictor of menopausal symptoms. Obstet Gynecd 2004; 103:960-966. 147. Richards M, Rubinow DR, Daly RC, et al. Premenstrual symptoms and perimenopausal depression. Am J Psychiatry 2006;163:133-137. 148. Joffe H, Hall JE, Soares CN, et al. Vasomotor symptoms are associated with depression in perimenopausal women seeking primary care. Menopause 2002;9:392-398. 149. Maciejewski PK, Prigerson HG, Mazure CM. Sex differences in event-related risk for major depression. Psychol Med 2001;31:593604. 150. Ormel J, Oldehinkel AJ, Brilman EI. The interplay and etiological continuity of neuroticism, difficulties, and life events in the etiology of major and subsyndromal, first and recurrent depressive episodes in later life. Am J Psychiatry 2001;158:885-891. 151. Kendler KS, Kessler RC, Waiters EE, et al. Stressful life events, genetic liability, and onset of an episode of major depression in women. Am J Psychiatry 1995;152:833-842. 152. Greene JG, Cooke DJ. Life stress and symptoms at the climacterium. BrJ Psychiatry 1980;136:486-491. 153. Ballinger S, Cobbin D, Krivanek J, et al. Life stresses and depression in the menopause. Maturitas 1979;1:191-199. 154. Schmidt PJ, Murphy JH, Haq NA, et al. Stressful life events, personal losses, and perimenopause-related depression. Arch Womens Ment Health 2004;7:19-26. 155. Dennerstein L, Dudley E, Guthrie J. Empty nest or revolving door? A prospective study of women's quality of life in midlife during the phase of children leaving and re-entering the home. Psychol Med 2002;32:545-550. 155a.Werner AA, Johns GA, Hoctor EF, et al. Involutional melancholia: probable etiology and treatment. JAMA 1934;103:13-16. 156. Dennerstein L, Burows GO, Hyman GJ. Hormone therapy and affect. Maturitas 1979;1:247-259. 157. Zohar J, Shapira B, Oppenheim G, et al. Addition of estrogen to imipramine in female-resistant depressives. Psychopharmacol Bull 1985;21:705-706. 158. Brincat M, Studd JWW, O'Dowd T, et al. Subcutaneous hormone implants for the control of climacteric symptoms: a prospective study. Lancet 1984;1:16-18.
319 159. Montgomery JC, Brincat M, Tapp A, et al. Effect of oestrogen and testosterone implants on psychological disorders in the climacteric. Lancet 1987;1:297-299. 160. Strickler RC, Borth R, Cecutti A, et al. The role ofoestrogen replacement in the climacteric syndrome. PsycholMed 1977;7:631-639. 161. Coope J. Is oestrogen therapy effective in the treatment of menopausal depression? J R Coll Gen Pract 1981;31:134-140. 162. Sherwin BB, Gelfand MM. Sex steroids and affect in the surgical menopause: a double-blind, cross-over study. Psychoneuroendocrinology 1985;10:325-335. 163. Sherwin BB. The impact of different doses of estrogen and progestin on mood and sexual behavior in postmenopausal women. J Clin EndocrinolMetab 1991;72:336-343. 164. Sherwin BB. Affective changes with estrogen and androgen replacement therapy in surgically menopausal women. J Affective Disord 1988;14:177-187. 165. Ditkoff EC, Crary WG, Cristo M, et al. Estrogen improves psychological function in asymptomatic postmenopausal women. Obstet Gyneco11991;78:991-995. 166. Schneider LS, Small GW, Clary CM. Estrogen replacement therapy and antidepressant response to sertraline in older depressed women. Am J Geriatr Psychiatry 2001;9:393-399. 167. Schneider LS, Small GW, Hamilton SH, et al. Estrogen replacement and response to fluoxetine in a multicenter geriatric depression trial. AmJ Geriatr Psychiatry 1997;5:97-106. 168. Small GW, Schneider LS, Holman S, et al. Estrogen plus fluoxetine for geriatric depression. AbstrAPA 147th Ann Meeting 1994:203. 169. Amsterdam J, Garcia-Espana F, Fawcett J, et al. Fluoxetine efficacy in menopausal women with and without estrogen replacement. J Affective Disord 1999;55:11-17. 170. Schmidt PJ, Nieman L, Danaceau MA, et al. Estrogen replacement in perimenopause-related depression: a preliminary report. Am J Obstet Gyneco12000;183:414-420. 171. Soares CD, Almeida OP, Joffe H, et al. Efficacy of estradiol for the treatment of depressive disorders in perimenopausal women: a doubleblind, randomized, placebo-controlled trial. Arch Gen Psychiatry 2001;58:529-534. 172. Morrison MF, Kallan MJ, Ten Have T, et al. Lack of efficacy of estradiol for depression in postmenopausal women: a randomized, controlled trial. Biol Psychiatry 2004;55:406-412. 173. Zweifel JE, O'Brien WH. A meta-analysis of the effect of hormone replacement therapy upon depressed mood. Psychoneuroendocrinology 1997;22:189-212. 174. Nass R, Baker S. Androgen effects on cognition: congenital adrenal hyperplasia. Psychoneuroendocrinology1991; 16:189-201. 175. Saletu B, Brandstatter N, Metka M, et al. Double-blind, placebocontrolled, hormonal, syndromal and EEG mapping studies with transdermal oestradiol therapy in menopausal depression. Psychopharmacology 1995;122:321-329. 176. Butler RN, Lewis MI. Late-life depression: when and how to intervene. Geriatrics 1995;50:44-55. 177. Smith RNJ, Holland EFN, Studd JWW. The symptomatology of progestogen intolerance. Maturitas 1994;18:87-91. 178. Hammarback S, Backstrom T, Holst J, et al. Cyclical mood changes as in the premenstrual tension syndrome during sequential estrogenprogestagen postmenopausal replacement therapy. Acta Obstet Gynecol &and 1985;64:393-397. 179. Magos AL, Brewster E, Singh R, et al. The effects of norethisterone in postmenopausal women on oestrogen replacement therapy: a model for the premenstrual syndrome. BrJ Obstet Gynaeco11986;93: 1290-1296. 180. Kirkham C, Hahn PM, VanVugt DA, et al. A randomized, doubleblind, placebo-controlled, cross-over trial to assess the side effects of medroxyprogesterone acetate in hormone replacement therapy. Obstet Gyneco11991;78:93-97.
320 181. Prior JC, Alojado N, McKay DW, et al. No adverse effects of medroxyprogesterone treatment without estrogen in postmenopausal women: double-blind, placebo-controlled, crossover trial. Obstet Gyneco11994;83:24-28. 182. Appleby L, Montgomery J, Studd J. Oestrogens and affective disorders. In: Studd J, ed. Progressin obstetricsandgynaecology. Edinburgh: Churchill Livingstone, 1981. 183. Yaffe K, Sawaya G, Lieberburg I, et al. Estrogen therapy in postmenopausal women: effects on cognitive function and dementia. JAMA 1998;279:688-695. 184. LeBlanc ES, Janowsky J, Chan BKS, et al. Hormone replacement therapy and cognition: systematic review and meta-analysis. JAMA 2001 ;285:1489-1499. 185. Gibbs RB. Long-term treatment with estrogen and progesterone enhances acquisition of a spatial memory task by ovariectomized aged rats. NeurobiolAging 2000;21:107-116. 186. Zandi PP, Carlson MC, Plassman BL, et al. Hormone replacement therapy and incidence of Alzheimer disease in older women: the Cache County study. JAMA 2002;288:2123-2129. 187. Resnick SM, Henderson VW. Hormone therapy and risk of Alzheimer disease: a critical time. JAMA 2002;288:2170-2172. 188. Mikkola TS, Clarkson TB. Estrogen replacement therapy, atherosclerosis, and vascular function. CardiovascRes 2002;53:605-619.
RUBINOW ET AL. 189. Brownley KA, Hinderliter AL, West SG, et al. Cardiovascular effects of 6 months of hormone replacement therapy versus placebo: differences associated with years since menopause. Am J Obstet Gynecol 2004; 190:1052-1058. 190. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321-333. 191. Shumaker SA, Legault C, Rapp SR, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651-2662. 192. Grodstein F, Clarkson TB, Manson JE. Understanding the divergent data on postmenopausal hormone therapy. N EnglJ Med 2003;348: 645-650. 193. Herrington DM, Howard TD, Hawkins GA, et al. Estrogen-receptor polymorphisms and effects of estrogen replacement on high-density lipoprotein cholesterol in women with coronary disease. NEnglJ Med 2002;346:967-974. 194. Yaffe K, Lui LY, Grady D, et al. Estrogen receptor 1 polymorphism and risk of cognitive impairment in older women. Bid Psychiatry 2002;51:677-682.
SECTION VI
Bone Changes One of the major concerns after menopause is the development of osteoporosis and its sequelae, namely fractures and ensuing disability. Since the last Edition of this text, more options have become available, although some discourage the use of hormonal therapy, which remains an excellent choice in terms of efficacy, for the treatment of osteoporosis because of concerns of various risks. This section aims to cover the major areas of importance for the preservation of bone mass in postmenopausal women. There is clear overlap between chapters, but this may be useful, allowing each chapter to stand alone. First, Robert Lindsay and Felicia Cosman discuss the pathophysiology of bone loss as it occurs throughout life and particularly after menopause. An important new concept is that of bone strength, which is different from bone density per se. Bone mass, which primarily determines bone mineral density, is not a good predictor of fracture risk, while it is envisioned that bone strength may better reflect fracture risk along with other traditional risk factors. This topic will be introduced and discussed by Michael Kleerekoper. Biochemical markers of bone tumors (resorption and formation) have become popular. What is their role, and which markers should be measured? What follows is a discussion of prevention and treatment of osteoporosis in two separate chapters by John C. Stevenson and colleagues and Claus Christiansen and colleagues. While there is clear overlap here, the intent is for the first chapter to focus on prevention and the second on treatment. Finally, because of the importance and high prevalence of osteoarthritis after menopause, the chapter by Tank6 and colleagues attempts to put this in perspective for the reader. This section should also be read with the chapter on collagen by M. Brincat and colleagues, as this relates closely to the issue of bone health.
This Page Intentionally Left Blank
.~HAPTER 2.
Pathogenesis of Osteoporosis ROBE R T
LINDSAY
Helen Hayes Hospital, West Haverstraw, NY 10993; Columbia University College of Physicians and Surgeons, New York, NY 10032
FELICIA COSMAN
HelenHayes Hospital, West Haverstraw,NY 10993; ColumbiaUniversityCollege of Physicians and Surgeons, New York, NY 10032
II. PEAK BONE MASS A N D ITS DETERMINANTS
I. I N T R O D U C T I O N Osteoporosis is a disease characterized by reduced bone strength and increased susceptibility to fractures. Loss of bone strength encompasses both reduced bone mass and abnormal bone quality, which includes changes in bone microarchitecture, bone turnover, damage accumulation, and mineralization. Osteoporosis is generally asymptomatic until the clinical sequelae (fractures) occur. These fractures and their consequences cause pain, disability, deformity, and sometimes premature death. At any given level of trauma, an individual with the skeletal changes of osteoporosis is more likely to fracture than an individual with normal skeletal structure. The ability to measure bone mass noninvasively has allowed greater understanding of the pathogenesis of osteoporosis. We know now that the pathophysiology of osteoporosis is multifactorial and includes genetic, endocrine, and lifestyle influences. This chapter briefly reviews the role of these factors in the earlier gain and later loss of bone mass, as well as the cellular mechanisms responsible for the latter. The amount of tissue in the skeleton (bone mass) is the integral of the amount of bone accrued during growth and consolidation (peak bone mass) and the inevitable loss of bone tissue with aging and menopause in women. Therefore both peak bone mass and the rate and duration of bone loss determine fracture risk. Most clinicians consider the process of loss of mass and its accompanying architectural changes to be a prerequisite to the relationship between bone mass and fracture risk in older individuals. TREATMENT OF THE POSTMENOPAUSAL WOMAN
323
A. Genetic Factors Bone mass peaks sometime between the late teenage years and early 30s. Peak bone mass (PBM) is primarily determined by genetic factors, although lifestyle and environmental factors also influence the growth of the skeleton. At skeletal maturity, men have approximately 5% to 10% higher bone mass (measured by dual energy x-ray absorptiometry [DXA]) than do women (1). This difference appears around the time of puberty and is due to sexual dimorphism in periosteal bone apposition at puberty. This likely is stimulated by increased androgen production leading to greater periosteal apposition in males, whereas estrogen inhibits this process in females (2). There are also significant racial differences in PBM, with African Americans having 5% to 10% greater bone mass than Caucasians. Numerous genes regulate growth, including skeletal growth and thus peak bone mass and body size. Other genes control skeletal structure and perhaps density. Twin studies suggest heritability estimates for PBM of 50% to 80% for different skeletal sites (3). Many candidate genes have been identified, including those encoding collagen type 1-alpha; the estrogen (ER alpha) and vitamin D receptors; transforming growth factor beta; insulin-like growth factor; apolipoprotein E; interleukins (IL) including IL-1, IL-6, and IL-11; and bone morphogenetic protein. However, there is little consistency regarding the magnitude of importance of each of these genes Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
324
LINDSAY AND COSMAN
and the extent to which their contribution can be generalized to all populations (4). Recently, a so-called "high bone mass" gene has been identified as a mutation in the lowdensity lipoprotein receptor-related protein (LRP-5) (5). The extent to which this gene contributes to bone density in the general population is not known, although the importance of the pathway has become clear since the description of the relationship in this single family (5). During growth, the effect of genes is modulated by hormonal factors, which in themselves may be genetically mediated. For example, it has been suggested that 33% of the variance in PBM can be explained by variations in endogenous growth hormone production, even when adjusted for height (6).
B. Nutrition When Charles Dent said, "Senile osteoporosis is a pediatric disease," he meant that failure to achieve adequate PBM increases the risk of osteoporosis in later life. Nutrition is an important, modifiable factor that determines PBM. Several nutrients play key roles in skeletal development, including protein (calories), calcium, phosphorus, and vitamin D. Calcium is perhaps the nutrient that is commonly deficient in the diet of children and adolescents in the United States. Higher lifetime calcium intakes are inversely correlated with hip fracture rates, and milk consumption in youth is related to bone density in young adult women and men (7).
C. Exercise Bone adapts to the loads that are applied to it (Wolff's Law), and higher mechanical loads lead to increased bone density. This is particularly true during childhood and growth (8). High-impact forces applied to the skeleton appear to confer the greatest benefit. Weight-bearing exercise is associated with higher PBM in both genders. Gymnasts have greater bone density than runners, whereas swimmers and cyclists have lower bone density (9). In a recent study (10), which avoided the usual bias introduced by recruiting volunteers, a general, school-based exercise program for 2 years increased bone density and bone size in 7- to 9-yearold girls.
D. Menstrual Function Achieving maximal PBM requires a normal transition through puberty, and inadequate pituitary-ovarian function in teenage years adversely affects attainment of PBM. Reduced bone mass has repeatedly been demonstrated in women with amenorrhea due to different primary condi-
tions, including anorexia nervosa and hyperprolactinemia. This deficit is partially reversible if normal ovarian function is resumed, but the longer the amenorrhea, the less likely that bone mass can be completely regained (11). Excessive exercise with amenorrhea is also associated with lower bone mass than in eumenorrheic athletes. Depo medroxyprogesterone acetate (MPA), a commonly used birth control method that suppresses endogenous estrogen production, appears to be associated with a decrease in bone mass, which in some situations may be reversible (12).
III. THE PATHOGENESIS OF BONE LOSS Age-related bone loss is a universal phenomenon and occurs at almost all skeletal sites (except the skull, for example) in all races and in both genders. However, the cellular mechanism underlying bone loss differs between the sexes. The rate of age-related bone loss is influenced by factors similar to those modifying skeletal growth: genetic, endocrine, and environmental factors. The effects of each factor are mediated through modulation of the bone remodeling cycle.
A. Bone Remodeling and the Cellular Basis of Bone Loss Bone remodeling describes the process whereby old bone is continuously replaced by new tissue. The balance between the amount of bone removed and the amount that replaces it, when integrated over a number of remodeling cycles, determines whether there is net loss or gain of bone tissue. Bone remodeling is accomplished primarily by osteoclasts and osteoblasts, which together with other accessory cell types are called bone remodeling units (BRUs) (13). The three-dimensional geometry of the completed units of bone, called osteons or packets, is somewhat different in cortical and cancellous bone (Figs. 25.1 to 25.3) but the cellular processes are essentially the same. The first event to occur in the remodeling process is activation. During activation (occurring about 6 times per minute in the skeleton), mononuclear osteoclast precursors, which are derived from circulating monocytes, are attracted to the surface of bone and fuse to form multinucleated osteoclasts. Remodeling may be both a random and a targeted event. One enduring hypothesis argues that osteocytes act as mechanosensor cells that detect changes in the mechanical and perhaps chemical properties of the surrounding bone matrix and communicate them to the lining cells on the surface (14). Lining cells then appear to retract, leaving denuded bone surface that chemo-attracts osteoclast precursors. The differentiated osteoclasts degrade both the organic and inorganic components of the matrix. In
CHAPTER 25 Pathogenesis of Osteoporosis
325
FIGURE 25.1 Cross-sectionaldiagrams of evolving BRU in cancellous bone (upper) and cortical bone (lower).The arrowindicates the direction of movement through space. Note that the cancellous BRU is essentially one-half of the cortical BRU. (Reproduced with permission from Dempster DW. New concepts in bone remodeling. In: Seibel MJ, Robins SP, Bilezikian JP, eds. Dynamics of bone and cartilage metabolism. San Diego: Academic Press, 1999:261-273.)
FIGURE25.2 An intact bone packet in human cancellous bone and an intact Haversian systemin human cortical bone. Note the osteons that have been partially replaced by prior remodelingevents. (Reproducedwith permissionfrom Dempster DW. Bone remodeling.In: Coe FL, FavusMJ, eds. Disorders of bone and mineral metabolism, ed 2. Philadelphia, Lippincott,Williams & Wilkins, 2002:315-343.)
cancellous bone and on the endosteal and periosteal surfaces of cortical bone, osteoclasts move along the bone surface, leaving a trench behind them. In cortical bone, osteoclasts tunnel through the bone, creating a cylindric hole. Formation, activation, and activity of osteoclasts are regulated by local cytokines such as receptor activator of nuclear factorkappa beta ligand (RANKL), IL-1 and IL-6, and colonystimulating factors (CSFs), as well as by systemic hormones such as estrogens, parathyroid hormone, 1,25-dihydroxyvitamin D3, and calcitonin (15). The final common pathway for osteoclast activation is through RANK, the cognate receptor for RANKL, which is a member of the tumor necrosis factor (TNF) family. RANKL is secreted by osteoblasts and mononuclear cells and interacts with RANK, its recep-
tor, expressed on osteoclast precursor cells and other marrow stromal cells. The R A N K L / R A N K interaction promotes the differentiation, activation, and prolonged survival of osteoclasts. Osteoprotegerin (OPG), a member of the superfamily of T N F receptors, is a decoy receptor for RANKL and is also secreted by osteoblasts. The binding of RANKL to O P G results in inhibition of the differentiation and activity of osteoclasts. The rate of bone remodeling is ultimately controlled by the relative amounts of RANKL and O P G secreted close to the surface of bone. The resorption phase is followed by the reversal phase. The osteoclasts and mononucleated resorbing cells die by apoptosis and are replaced by osteoblasts as the cycle progresses into the formation phase (16). Teams of osteoblasts
326
FIGURE 25.3 Scanningelectron micrograph of a trabecular plate in hu-
man iliac cancellous bone. Osteoclasts have perforated the plate from the top and, a few millimeters to the left, from underneath. (Reproducedwith permission from Dempster DW. Bone remodeling.In: Coe FL, FavusMJ, eds. Disorders of bone and mineral metabolism, ed 2. Philadelphia, Lippincott, Williams & Wilkins, 2002:315-343.) follow osteoclasts as they traverse the bone surface or tunnel through the cortex. On the bone surfaces, they refill the trench with a new unit of lamellar bone, which is referred to as a hemi-osteon or, more commonly, a packet (see Fig. 25.2). Within the cortex, the osteoblasts refill the tunnel dug by osteoclasts and deposit concentric lamellae, starting on the outer wall and working inward. As a result, cross-sections of cortical osteons, or Haversian systems, are reminiscent of a cross-section of the trunk of a tree. Women experience slight but gradual trabecular thinning from the time of PBM until menopause, when an abrupt acceleration in the rate of bone loss occurs. This accelerated loss persists for about 5 to 10 years (Fig. 25.4) and is accompanied by a dramatic increase in bone turnover rate (17). The rapid bone loss experienced after menopause is due to
FIGURE 25.4 The relationship between cancellous bone volume and age in the human ileum in men (open circles) and women (closed circles, stippled). (Reproduced with permission.)
LINDSAY AND COSMAN
enhanced osteoclast activity and an overall increase in the rate of remodeling. An increased number of resorption sites are active per unit time, and the osteoclasts dig farther into the trabeculae and often perforate and ultimately remove them (Fig. 25.5). In contrast, in men aging results in a gradual decrease in the efficiency with which the osteoblasts refill the resorption cavities. Consequently, the thickness of the cancellous bone packets gradually decreases, which in turn leads to trabecular thinning with a linear reduction in cancellous bone volume with age (see Fig. 25.4). Although the individual trabeculae are thinner, trabecular connectivity is generally preserved (18). The consequence is a greater age-related loss of trabecular number and connectivity with age in women compared with men. The reason why osteoclasts gain the ability to perforate trabecular plates after menopause is thought to be due to an increase in their life span as a result of decreased estrogen levels and an accompanying increase in RANKL production, with perhaps a concomitant reduction in O P G production (19). Estrogen inhibits osteoclast production and promotes osteoclast apoptosis, whereas RANKL stimulates osteoclast recruitment and prolongs osteoclast life span. Bone loss by trabecular elimination is considerably more harmful to bone strength than loss of an equivalent amount of bone tissue through trabecular thinning (20). A significant increase in resorption cavity depth has been observed in early postmenopausal women, which together with simultaneous osteoclastic attack from both sides may contribute to complete trabecular perforation and elimination. Alternatively, particularly destructive onesided resorption can also perforate trabecula (see Figs. 25.5 and 25.6). At weight-bearing sites, including the vertebral bodies and the hip, trabeculae have a clear orientation with the weight-bearing struts aligned in the direction of the primary axis of loading. These trabeculae are in turn supported by trabeculae oriented perpendicular to the load axis, serving as
CHAPTER 25 Pathogenesis of Osteoporosis
FIGURE 25.5
OP
327
Scanningelectron micrographs of human cancellous bone.
Rx
/
FIGURE25.6 Theoreticexplanationfor the bone fragilityassociatedwith enhanced remodeling and why treatment with an agent that reduces bone turnover may be beneficial without changing bone mass significantly.In a normal subject (N), the two resorption cavities do not represent a threat to the stability of the trabecula because it is stabilized by the two horizontal trabeculae. In an osteoporotic individual (OP), the same two cavities significantly increase the chance of failure at these stress-concentrating loci. Reducing the number of resorption cavitiesby treatment (Rx) decreasesthe risk of failure.
cross-ties. Several studies have shown that preferential loss of these cross-ties occurs with aging (21), with consequent serious penalties for strength. According to Euler's theorem, the buckling load of a weight-bearing trabecula decreases inversely with the square power of its unsupported length. The loss of two cross-ties will triple the effective length of the trabecula but will decrease the load that the trabecula can support by a factor of nine. Thus, small changes in tissue mass can have devastating consequences for strength when the losses are strategically placed. In support of this concept, Ciarelli et al. (22) showed that trabecular bone in the proximal femur of women with hip fracture has significantly greater anisotropic (oriented) structure than controls, even
when the samples were matched for the total cancellous bone mass. The greater anisotropy was thought to be the result of preferential loss of cross-ties. Several studies have confirmed compromised cancellous bone microarchitecture in both men and women with osteoporosis-related fractures when compared with controls with the same bone mass (23). The loss of horizontal trabeculae leaves the load-bearing struts vulnerable to failure of a process that is compounded by the increased bone turnover that occurs after menopause. Each resorption cavity acts to concentrate stress at that trabecular site, creating a weak point and increasing the probability of failure at that site (24). This is analogous to filing a tiny scratch on a glass rod prior to breaking it. The mass of the rod barely changes, but when force is applied, the rod dutifully breaks exactly at the location of the scratch. The rapid reduction in the number of stress-concentrating resorption cavities by antiresorptive agents may be the explanation for the early decrease in fracture risk that is disproportionately greater than that predicted by the improvements in bone density. With aging, long bones increase in diameter, but the cortices become thinner. This is the end product of small differences in the amount of bone removed and formed in each remodeling cycle. This remodeling balance is negative on the inner endosteal surface but positive on the outer periosteal surface. Furthermore, more bone is lost at the endosteal surface than is gained at the periosteal surface. Although these differences may be quite small within each remodeling unit, the effects accumulate with age. Consequently, the endosteal surface drifts outward at a greater rate than the periosteal surface and, as a result, the cortex gets thinner. With age, women lose more bone from the endosteal surface than do men (25), and men may compensate for endosteal loss by laying down more bone on the periosteal surface (2). The result is a greater decline in the cross-sectional moment of inertia of long bones with age in women and consequently a
328
LINDSAY AND COSMAN
greater increase in fracture risk. Within the cortex, other agerelated changes magnify the effect of this cortical drift phenomenon, including a decline in the area occupied by circumferential lamellar bone, an increase in the number of Haversian systems and their fragments, a decrease in the radial closure rate of osteons, an increase in the diameter of the Haversian canals, a decrease in osteon wall thickness, and an increase in the number of resorption cavities that are aborted in the reversal phase and remain unfilled. Each of these contributes to an increase in cortical porosity that occurs to a greater extent in women than men (26). As with trabecular thinning in cancellous bone, this occurs mostly through an age-related decline in osteoblast recruitment and/or activity.
IV. E S T R O G E N DEFICIENCY Driving the changes that occur in remodeling is the transition from the estrogen-replete premenopausal state to the relatively estrogen-deficient postmenopausal state. Bone loss accelerates after a natural or surgical menopause. Some studies have shown that the rate of loss is dependent on the remaining estrogen supply. Rapid bone loss may continue for some years in most postmenopausal women but eventually slows as surfaces are lost. Bone remains sensitive to estrogen until old age, and estrogen intervention slows bone loss and reduces fracture risk. It appears as though the estrogen effect is mediated through modulation of RANKL and OPG, with evidence that estrogen downregulates RANKL and upregulates OPG.
protein has been found to reduce complications following hip fracture (30). Some elderly individuals may be deficient in intake of phosphate, which may also contribute to bone loss (31). In people who are elderly or chronically ill or who have malabsorption, low magnesium levels may accelerate bone loss (32). Several studies show a positive association between vitamin C and bone mass (33), and vitamin K intakes may be related to bone density levels and fracture risk (34). Excess vitamin A may cause bone loss, and high intakes have been associated with a twofold increase in the risk of hip fracture in some studies (35). Elevated serum homocysteine, perhaps associated with vitamin B deficiencies, may be associated with bone loss and increased fracture risk (36).
C. Alcohol and Smoking Excessive alcohol consumption increases the risk of osteoporosis by inhibiting bone formation and results in an increased risk of hip fracture (37). Nutritional deprivation in alcoholics contributes to the toxic effects of alcohol on bone and other organ systems. Alcohol abuse also increases the risk of falling. Smoking exerts a toxic effect on osteoblasts, leads to premature menopause, and reduces the production while accelerating the degradation of estrogen. As a result, smokers often have low bone mass (38). Smokers also generally weigh less than nonsmokers and are generally less physically active, further increasing the risk of fracture.
D. Physical Activity A. Nutrition: Calcium and Vitamin D It is clear that an adequate calcium intake throughout life is important for skeletal health. Some studies have shown that calcium plus vitamin supplementation reduces fracture risk among older individuals, and maintaining total calcium intake between 1000 and 1500 mg/day is prudent (27). Vitamin D deficiency or insufficiency (25-hydroxyvitamin D levels less than 20 ng/mL) is believed to play a role in fracture pathogenesis as a result of increasing rates of bone loss and possibly contributing to muscle weakness, impaired balance, and thereby increased risk of falling (28). A recent study suggested that approximately 50% of patients receiving therapy for osteoporosis had suboptimal circulating levels of 25-hydroxyvitamin D (29).
B. Other Nutritional Factors Low-protein intake can have deleterious effects in the elderly. Hip fracture patients are frequently malnourished with inadequate protein intake, and supplementation with
Immobilization results in rapid and profound bone loss. In contrast, weight-bearing and resistance exercise are associated with increased bone mass or slowed rate of bone loss (39). Lifetime physical activity is associated with reduced risk of fracture (40).
E. Chronic Diseases and Medications Many chronic diseases or conditions are associated with enhanced bone loss and increased risk ofosteoporosis (41,42). These include endocrine diseases (e.g., hyperthyroidism and hyperparathyroidism), rheumatologic conditions (e.g., rheumatoid arthritis and systemic lupus), chronic lung disease, gastrointestinal conditions associated with malabsorption (e.g., celiac disease or inflammatory bowel disease), eating disorders, neurologic diseases (e.g., Parkinson's disease, multiple sclerosis, and spinal cord injury), hematologic/oncologic diseases (most notably multiple myeloma), and organ transplantation. Bone loss associated with these diseases is due to a variety of reasons, including medication use, nutritional factors, diminished ability to perform physical activity, and
CHAPTER 25 Pathogenesis of Osteoporosis factors that directly affect bone remodeling. In addition, certain medications are known to alter bone metabolism and have detrimental effects on the skeleton. Glucocorticoids, for example, profoundly inhibit bone formation by p r o m o t i n g osteoblast apoptosis. T h e y also reduce calcium absorption, increase renal calcium excretion, and have multiple other adverse effects on bone and mineral metabolism. O t h e r medications that adversely affect the skeleton include antiepileptics, excessive thyroid hormone, chemotherapy, and gonadotropin-releasing h o r m o n e antagonists.
E Falling T h e majority of osteoporosis-related fractures are secondary to a fall (43,44). Approximately 30% of free-living (i.e., noninstitutionalized) elderly individuals fall at least once per year, and this n u m b e r is even higher in institutionalized individuals. T h e incidence of falls increases with age and, prior to age 70, is greater in w o m e n than men. A history o f falling greatly increases the risk o f subsequent falls. Risk factors for falling can be separated into intrinsic and extrinsic factors. Intrinsic factors include underlying disorders that impair balance and ambulation and may be associated with the clinical syndrome o f frailty. External factors include the use of medications such as anxiolytics, narcotics, antipsychotics, and hypoglycemic, hypotensive, and diuretic agents; alcohol abuse; and environmental hazards. T h e r e is also a strong association between inadequate vitamin D intake/production and fall risk (45).
References 1. Looker AC, Wahner HW, Dunn WL, et al. Proximal femur bone mineral levels of U.S. adults. OsteoDorosInt 1995;5:389-400. 2. Seeman E. Periosteal bone formation--a neglected determinant of bone strength. N EnglJ Med 2003;349:320-323. 3. Arden NK, Baker J, Hogg C, et al. The heritability of bone mineral density, ultrasound of the calcaneus and hip axis length: a study of postmenopausal twins. J Bone Miner Res 1996;11:530-534. 4. Eisman JA. Genetics of osteoporosis. Endocr Rev 1999;20:788-804. 5. Little RD, Recker RR, Johnson ML. High bone density due to a mutation in LDL-receptor-related protein 5. N EnglJ Med 2002;347:943944. 6. Rissell-Aulet M, Shapiro B, Jaffe CA, et al. Peak bone mass in young healthy men is correlated with the magnitude of endogenous growth hormone secretion. J Clin Endcrinol Metab 1998;83:3463-3468. 7. Welten DC, Kemper HCG, Post GB, Van Staveren WA. A metaanalysis of the effect of calcium intake on bone mass in young and middle aged females and males. J Nutr 1995;125:2802-2813. 8. Wallace BA, Cumming RG. Systematic review of randomized trials of the effect of exercise on bone mass in pre- and postmenopausal women. Calcif Tissue Int 2000;67:10-18. 9. Jaffe DR, Snow-Hatter C, Connolly DA, et al. Differential effects of swimming versus weight-bearing activity on bone mineral status of eumenorrheic athletes. J Bone Miner Res 1995;10:586-593.
329 10. Linden C, Ahlborg H G, Besjakov J, Gardsell P, Karlsson MK. A school curriculum-based exercise program increases bone mineral accrual and bone size in prepubertal girls: two year data from the pediatric osteoporosis prevention (POP) study. J Bone Miner Res 2006;21: 829-835. 11. Rencken ML, Chesnut CHIII, Drinkwater BL. Bone density at multiple skeletal sites in amenorrheic athletes. JABdA 1996;276:238-240. 12. Scholes D, LaCroix AZ, Ichikawa LE, Barlow WE, Ott SM. Change in bone mineral density among adolescent women using and discontinuing depot medroxyprogesterone acetate contraception. Arch Pediatr Adolesc Med 2005 159:139-144. 13. Dempster DW. Bone remodeling. In: Coe FL, Favus MJ, eds. Disorders of bone and mineral metabolism, ed 2. Philadelphia: Lippincott, Williams & Wilkins, 2002:315-343. 14. Rubin J, Rubin C, Jacobs CR. Molecular pathways mediating mechanical signaling in bone. Gene 2006;367:1-16. 15. Blair HC, Athanasou NA. Recent advances in osteoclast biology and pathological bone resorption. Histol Histopatbo12004;19:189-199. 16. Parfitt AM. Bone forming cells in clinical conditions. In: Hall BK, ed. Bone, vol 1: The osteoblast and osteocyte, Caldwell, NJ: Telford Press, 1990:351-429. 17. Recker R, Lappe J, Davies KM, Heaney R. Bone remodeling increases substantially in the years after menopause and remains increased in older osteoporosis patients. J Bone Miner Res 2004;19:1628-1633. 18. Weinstein RS, Hutson MS. Decreased trabecular width and increased trabecular spacing contribute to bone loss with aging. Bone 1987;8: 137-142. 19. Eghbali-Fatourechi G, Khosla S, Sanyal A, et al. Role of RANK ligand in mediating increased bone resorption in early postmenopausal women. J Clin Invest 2003;111:1221-1230. 20. Guo XE, Kim CH. Mechanical consequence of trabecular bone loss and its treatment: a three-dimensional model simulation. Bone 2002;30:404-411. 21. Thomsen JS, Ebbesen EN, Mosekilde LI. Age-related differences between thinning of horizontal and vertical trabeculae in human lumbar bone as assessed by a new computerized method. Bone 2002;31: 136-142. 22. Ciarelli TE, Fyhrie DP, Schaffler MB, Goldstein SA. Variations in three-dimensional cancellous bone architecture of the proximal femur in female hip fractures and controls. J Bone Miner Res 2000;15:32-40. 23. Dempster DW. Bone microarchitecture and strength. Osteoporos Int 2003;14:54-56. 24. Parfitt AM. Use ofbisphosphonates in the prevention of bone loss and fractures. Am J Med 1999;91:42S-46S. 25. Riggs BL, Melton LJ III, Robb RA, et al. Population-based study of age and sex differences in bone volumetric density, size, geometry, and structure at different skeletal sites.J Bone Miner Res 2004;19:1945-1954. 26. Bousson V, Peyrin F, Bergot C, et al. Cortical bone in the human femoral neck: three-dimensional appearance and porosity using synchrotron radiation. J Bone Miner Res 2004;19:794-801. 27. Chapuy MC, Arlot ME, Duboeuf F, et al. Vitamin D3 and calcium to prevent hip fractures in the elderly women. N EnglJ Med 1992;327: 1637-1642. 28. Bischoff-Ferrari HA, Willett WC, Wong JB, et al. Fracture prevention with vitamin D supplementation: a meta-analysis of randomized controlled trials. JAMA 2005;293:2257-2264. 29. Holick MF, Siris ES, Binldey N, et al. Prevalence of vitamin D inadequacy among postmenopausal North American women receiving osteoporosis therapy. J Clin Endocrinol Metab 2005;90:3215-3224. 30. Munger RG, Cerhan JR, Chiu BC. Prospective study of dietary protein intake and risk of hip fracture in postmenopausal women. Am J Clin Nutr 1999;69:147-152. 31. Heaney RP. Phosphorus nutrition and the treatment of osteoporosis. Mayo Clin Proc 2004;79:91-97.
330 32. Durlach J, Bac P, Durlach V, et al. Magnesium status and aging: an update. Magnes Res 1998;11:25-42. 33. Hall SL, Greendale GA. The relationship of dietary vitamin C intake to bone mineral density. Results from the PEPI study. CalcifTissue Int 1998;63:183-189. 34. Booth SI, Tucker KI, Chen H, et al. Dietary vitamin K intakes are associated with hip fracture but not with bone mineral density in elderly men and women. Am J Clin Nutr 2000;71:1201-1208. 35. Feskanich D, Singh V, Willett WC, Colditz GA. Vitamin A intake and hip fractures among postmenopausal women. JAMA 2002;287: 47-54. 36. McLean RR, Jacques PF, Selhub J, et al. Homocysteine as a predictive factor for hip fracture in older persons. N Engl J Med 2004;350: 2042-2049. 37. Felson DT, Kiel DP, Anderson JJ, Kannel WB. Alcohol consumption and hip fractures. The Framingham Study. Am J Epidemio11998;128: 1102-1110. 38. Bainbridge KE, Sowers M, Lin X, Harlow SD. Risk factors for low bone mineral density and the 6-year rate of bone loss among premenopausal and perimenopausal women. OsteoporosInt 2004;15:439-446.
LINDSAY AND COSMAN 39. Gregg EW, Pereira MA, Caspersen CJ. Physical activity, falls, and fractures among older adults: a review of the epidemiologic evidence. JAm Geriatr Sac 2000;48:883-893. 40. Jaglal SB, Kreiger N, Darlington G. Past and recent physical activity and risk of hip fracture. Am J Epidemiol 1993;138:107-118. 41. Kelman A, Lane NE. The management of secondary osteoporosis. Best Pract Res Clin R&umato12005;19:1021-1037. 42. Boling EP. Secondary osteoporosis: underlying disease and the risk for glucocorticoid-induced osteoporosis. Clin Ther 2004;26:1-14. 43. Close JC, Lord SL, Menz HB, Sherrington C. What is the role of falls? Best Pract Res Clin Rbeumato12005;19:913-935. 44. Gregg EW, Pereira MA, Caspersen CJ. Physical activity, falls, and fractures among older adults: a review of the epidemiologic evidence. JAm Geriatr Soc 2000;48:883-893. 45. Bischoff-Ferrari HA, Dawson-Hughes B, Willett WC, et al. Effect of vitamin D on falls: a meta-analysis. JAMA 2004;291:1999-2006.
2HAPTER 2(
Assessment of Bone Density and B one S trength MICHAEL
KLEEREKOPER
St. Joseph Mercy Hospital, Ann Arbor, MI; Department of Internal Medicine and Obstetrics and Gynecology, Wayne State University School of Medicine, Detroit, MI; St. Joseph Mercy Reichert Health Center, Ypsilanti, MI 48197
I. I N T R O D U C T I O N
the body (5). From attainment of this peak bone mass until menopause, BMC is constant, although some studies suggest a very slight decrease in BMC during the few years prior to menopause (6,7). At menopause there is a marked loss of BMC, which continues for approximately 5 to 7 years. At that time the rate of loss slows down to a rate known as age-related bone loss, a phenomenon that appears to be universal in all primates, both male and female. As BMC decreases, bone strength diminishes and risk of fracture increases. Fractures resulting from this low BMC are termed oste@orosis-relatedfractures. In general these are low- or minimal-trauma fractures, which are defined as fractures occurring after trauma equal to or less than a fall from a standing height. Although a decrease in BMC is the root cause of osteoporosis and osteoporosis-related fractures, it has become accepted practice in clinical medicine to report bone mineral density (BMD). Only Q C T can measure BMD directly because it is a three-dimensional (3D) technology. However, access to Q C T is limited, and few epidemiologic studies or clinical trials have been performed using Q C T (8-10). DXA is a two-dimensional (2D) technology such that only "areal" BMD can be measured. To obtain this areal BMD, the area of the skeleton through which the x-ray beams are passed is computed and BMC is divided by area results in areal BMD. While not strictly a physiologic measurement because density is by definition a 3D phenomenon, DXA is more widely available and has been the preferred method for epidemiologic studies, controlled clinical trials, and the everyday practice of medicine. Q CT is re-
Each day in the United States, 2500 to 3500 women become menopausal. Unless treated with estrogen (with or without progestin) for control of menopausal symptoms, each of these women will experience accelerated bone loss for about 5 to 7 years. After that time, the bone loss continues but at a slower rate. The consequence of this bone loss, if left unrecognized or untreated, is that 50% of these women are likely to develop one or more osteoporosisrelated fractures before the end of life (1). These fractures result directly from progressive loss of bone mass and bone density and loss of bone strength due to the microarchitectural changes that accompany the bone loss. This chapter details the approaches to assessing bone density and bone strength.
II. B O N E MASS A N D B O N E M I N E R A L DENSITY Bone mass, or more precisely, bone mineral content (BMC) can be measured directly with radiographic techniques, the most common methods being quantitative computerized tomography (QCT) and dual-energy x-ray absorptiometry (DXA). BMC measured by these methods has been shown to correlate well with directly measured bone ash weight (2-4). BMC increases with age until about age 25, although there are minor differences in the age at which this peak bone mass is reached at different skeletal sites in TREATMENT OF THE POSTMENOPAUSAL WOMAN
331
Copyright 9 2007by Elsevier,Inc. M1rights of reproductionin any form reserved.
332 stricted to radiology suites, whereas DXA is most often placed in physicians' offices. Despite the physical shortcomings of DXA, this method of assessing BMD has been demonstrated to correlate extremely well with fracture risk in untreated populations, with fracture risk doubling for each standard deviation decrease in BMD, no matter which skeletal site is measured (11,12). DXA-measured BMD varies markedly from site to site in the skeleton (13-16). The most common sites measured are the forearm (both radial midshaft, which is predominantly cortical bone, and the ultradistal radius, which is predominantly cancellous bone and a common site of fracture), the lumber spine from the first to the fourth lumbar vertebrae (L1-L4), and the proximal femur. Within the proximal femur, precise measurements can be made at the femoral neck (a very common site of hip fracture), the intertrochanteric region (another common site of fracture), and the total proximal femur. At each of these sites, BMD is normally distributed around a mean value. The distribution around the mean value at the radial midshaft is quite narrow, whereas it is quite broad at the femoral neck. To obviate these site-to-site differences in BMD and the range of BMD in healthy young premenopausal women (ages 25 to 49 in general), it has become common practice to report DXA BMD results in terms of standard deviations around the mean. By convention, these standard deviations are expressed as a T-score. In groups of women, T-scores at the different measurement sites are more closely correlated with each other than the raw BMD score. In the early 1990s a committee of experts convened by the World Health Organization (WHO) reviewed all the (then) available data and determined that the prevalence of a Tscore of-2.5 or lower closely approximated the prevalence of osteoporosis-related fractures in the community (17). They concluded that a T-score at -2.5 or below should be defined as osteoforosis (in much the same way that cut-points are used to define hypertension or hyperlipidemia). For reasons that are less clear, normal BMD was defined as a T-score of-1.0 or above. Women with a T-score between -1.0 and -2.5 were considered to have low bone mass. It is equally unclear why they chose to provide an alternative term, osteofenia, for this low bone mass group. In practice, osteofenia has become the more commonly used of these two terms. Because of site-to-site differences in BMD and T-scores, an individual subject may be found to have normal BMD at one site, low bone mass at another site, and osteoporosis at still another site. This is almost an inevitable consequence of skeletal anatomy and the indirect method of measuring BMD using DXA. Regrettably this has resulted in considerable confusion, such that many (probably most) bone density reports provide the patient with more than one diagnosis. This is an incorrect use of the technology, as a patient can only have one diagnosis, which should be based on the
MICHAEL KLEEREKOPER
lowest T-score obtained during the DXA study. It is this information that provides the most reliable DXA-based prediction of fracture risk. For older women (65 +) there is good epidemiologic data to develop a site-specific fracture risk (18). For women in the early menopausal years, this is not yet possible. The average age of menopause is close to 50, and the average age at which clinicians begin to see an increase in hip fracture rates is close to 75--an interval of 20 to 25 years between menopause and hip fracture. As the technology for hip DXA has been available for less than 20 years, it is simply not yet possible to have any confidence about site-specific fracture risk in younger postmenopausal women. As BMD decreases with advancing age, so do osteoporosisrelated fracture rates. However, the actual population of women over age 70 is substantially less than the population of women ages 50 to 70, and the incidence of osteoporosisrelated fractures is in fact greater in this younger group of women. In the first 10 years postmenopause, fracture of the distal forearm (CoUes' fracture) is most common; during the next decade, vertebral fracture is most common.
A. In Whom Should BMD Be Measured? Most professional organizations dealing with either osteoporosis as a disease or postmenopausal women as a group have developed clinical guidelines to address this question. The reader is advised to use the guidelines prepared by his or her professional organization. In general, the differences between these guidelines are small. In the United States, both Medicare and the United States Preventive Services Task Force (USPSTF) have determined that it is practical and cost effective to measure BMD in all women aged 65 or older and to obtain repeat measurements at 2-year intervals (19). The USPSTF recognizes that many risk factors accelerate bone loss in women under age 65, and each of the professional organizations that have developed guidelines has a similar approach. Although the selected risk factors vary among guidelines, the philosophy is that the more risk factors that are present, the earlier should the first BMD study be performed. The most compelling risk factor is a history of a previous fracture as an adult. Intellectually, this would seem to apply only to those who sustained a minimal trauma fracture, but in actuality this dictum seems to apply even to traumatic fractures. Other commonly applied risk factors include a family history of osteoporosis with fracture, premature menopause or a premenopausal history of prolonged amenorrhea, use of glucocorticosteroids, and cigarette smoking. A number of diseases associated with malnutrition or malabsorption are also important risk factors for fracture and indications for a BMD study (1). BMD change at menopause is rapid, but fracture risk is low such that if a decision is made to measure BMD at
CHAPTER 26 Assessment of Bone Density and Bone Strength menopause, a repeat study should not be performed for 2 to 3 years (20). As women age, the rate of bone loss slows but the risk of fracture increases. Here too the combination of rate of loss and risk of fracture is such that an interval of 2 years between serial studies is recommended (20). It is appropriate to make a strong comment about how serial studies should be performed. There are three manufacturers of DXA in the United States, and each has developed a different approach to optimizing B M D measurement on its system. Accordingly, the same person measured on the same day on each of these three DXA instruments is very likely to have minor differences in B M D and T-score. Because changes in B M D over time can be quite small, even in the treated state, considerable confusion and both patient and physician distress is the likely outcome if serial studies are performed at different facilities. This is true, albeit to a lesser extent, if the two studies were performed on exactly the same DXA make and model but at two different locations. Given the movement of people around the country and the changing locations dictated by health insurers, this will be a persistent problem, and at present, there is no recommended or accepted solution. It remains important for the physician (who may have no control over where the serial study is performed) to be aware of this issue and take it into consideration when discussing results with the patient. Another pitfall to consider when interpreting a DXA result is that it is a 2D technique that tends to "shortchange" slender women. Although it has been demonstrated in epidemiologic studies that "thinness" is indeed a risk factor for fracture, this generally refers to thinness that is disproportionate to the woman's height. Short, slender women will most likely have a low BMD, but that value will overestimate their risk of fracture. One more consideration is that the W H O criteria for "diagnostic" labeling based on B M D does not match most standard "laboratory" criteria. It is usual to label all values within 2 standard deviations of the mean value as being a normal result. The W H O opted to label values between -1.0 and-2.5 as abnormal. This ignores the statistics in that 16% of all normal subjects will be between 1 and 2 standard deviations below the mean. If the W H O criteria are strictly applied, 16% of normal woman would be categorized as having low bone mass or "osteopenia" (20). Caution is advised when applying this term to patients, many of whom regard osteopenia as a frightening condition with imminent risk of fracture.
B. How to Best Use DXA in Clinical Practice The simplest approach is to regard B M D as a risk factor for an adverse health outcome (fracture), just as a lipid profile is regarded as a risk factor for acute myocardial infarction and blood pressure as a risk factor for stroke. (It is worth
333 noting that B M D predicts fracture as well as blood pressure predicts the risk of stroke, and considerably better than total cholesterol predicts acute myocardial infarction.) Diagnostic labels should be used with caution, as should intervention decisions. Identified correctable risk factors should be corrected before more specific therapy is initiated. Nutritional and lifestyle advice should also be given. Once a decision to begin disease-specific therapy has been made, it is important to discuss expected outcomes. In particular, with those drugs that act by inhibiting bone loss (estrogen, calcitonin, raloxifene, and the bisphosphonates), antifracture protection may be provided even though serial B M D studies do not demonstrate any significant increase in BMD. The patient should be made aware of this before the second DXA study. The only time there should be concern that the therapy is not effective is when a significant decrease in B M D occurs between one study and the next. Even a fracture occurring on therapy is not an indication of treatment failure in the individual patient.
C. Bone Strength The most recent National Institutes of Health (NIH) consensus definition of osteoporosis is that it is "a systemic disease characterized by low bone mass and diminished bone quality" (21). Bone quality is a nebulous term including each of the many factors that can contribute to bone strength above and beyond bone mass. In practical terms there are only two parameters that are or are nearly ready for use in clinical practice: bone turnover and bone microarchitecture.
D. Bone Turnover Bone is a structural tissue and, in common with all structural materials, is subject to fatigue damage. Unique to the skeleton is the ability to self-repair and minimize the risk of fatigue damage. The process is known as either bone turnover or bone remodeling. Once peak adult bone mass has been attained, maintenance is achieved by removal of old bone (bone resorption) and replacement at the same site with new bone (bone formation). Resorption results from the action of osteoclasts and resorption from the action of osteoblasts. Resorption is rapid, with old bone being removed in about 10 days, but the repair process takes place over approximately 90 days. At menopause, estrogen deficiency unleashes a number of local cytokines that stimulate the recruitment and action of osteoclasts to an extent that osteoblast-mediated repair cannot keep up with resultant rapid bone loss. After 5 to 7 years, the intense osteoclastic activity slows down and the osteoblast-mediated repair begins to catch up, but never to a point where all bone loss is restored. After this 5- to 7-year period, when age-related bone loss takes control of
334
MICHAEL KLEEREKOPER
TABLE 26.1 Bone Turnover Markers (Available as Serum Based Assays) A. Resorption Cross-linked C-telopeptide of type I collagen (ICTP) Tartrate-resistant acid phosphatase 5b (TRACPSb) Amino-terminal telopeptide of collagen cross-links (NTx) Carboxy-terminal telopeptide of collagen cross-links (CTx) B. Formation Bone specific alkaline phosphatase (BSAP) Osteocalcin (OC) Carboxy-terminal propeptide of type I collagen (PICP) Amino-terminal propeptide of type I collagen (PINP)
the system, the osteoblast repair always lags behind the osteoclast-mediated bone loss, with a slow rate of bone loss continuing. Clearly, at some point a situation close to bone remodeling balance must occur or total loss of bone mass would result. This never happens, even in those conditions with the most rapid bone loss. The absolute and relative action of osteoclasts and osteoblasts can be measured by dynamic histomorphometry on bone biopsy specimens, but this invasive procedure is impractical for the many millions of women experiencing postmenopausal and age-related bone loss. A number of biochemical markers of bone resorption and bone formation have been developed and applied in clinical use (Table 26.1). There has been a great deal of reluctance to widely apply these biochemical tools to clinical practice because of a belief that these markers have too much variability to be applied to individual patients. To some extent this is true, but the techniques have evolved to a point where analytic variability has been minimized to closely match the analytic variability of many specialized biochemical tests. The biologic variability is a real phenomenon, but this too is not unique to the clinical biochemistry of the skeleton. These biochemical markers cannot be used to diagnose osteoporosis but are an important adjunct to BMD in "predicting" rates of future bone loss and in monitoring the responses to therapy. At menopause there is a marked increase
FIGURE 26.1 BMD change: healthypostmenopausal women with low and high bone turnover. BMD, bone mineral density; BAP,bone alkaline phosphatase; PICP, C-terminal propeptide of type 1 collagen; PINP, N-terminal propeptide of type 1 collagen; NTX, N-telopeptide breakdownproducts; CTX, C-telopeptide breakdownproducts;high turnover, bone marker levels abovethe upper limit of the premenopausal range. (From res 22, with permission.)
in markers of resorption and formation in many, but not all, women. A prospective 4-year study ofpostmenopausal women measured bone loss in those women in whom the markers were still within 2 standard deviations of the mean values in premenopausal women and compared the results with those women in whom the markers were already above the premenopausal reference interval (22). Those women with premenopausal levels for the markers had a BMD decrease of 1% or less over the 4 years. Those women with elevated markers had a 3% to 5% decrease in BMD during that same follow-up period. With each marker studied (there were seven in all), these differences were highly statistically significant and almost certainly of biologic significance as well (Fig. 26.1). In practical terms, for a woman in whom the DXA data does not clearly demonstrate that therapeutic intervention is immediately necessary, biochemical markers can be of considerable help in decision making. If the markers are still within the premenopausal reference interval, the anticipated loss in BMD over 2 years is low, and little irreversible skeletal harm is likely to ensue if the patient is followed without specific therapy. In contrast, a high value for the markers would seem to justify the use of specific therapy to prevent bone loss. The difficulties of using DXA to follow patients on therapy have already been addressed. Biochemical markers of bone turnover can be a very useful adjunct to monitoring the therapeutic response. In each of the published controlled clinical trials, there has been a decrease of 50% or more in the resorption markers within 3 months of initiation of therapy, after which time they remain at this lower level as long as effective therapy is continued (23-25). The nadir in formation markers is not seen until after 6 months of therapy. In an individual patient, if the expected decrease in markers are not seen, it is important to ensure that the patient is taking the therapy regularly and correctly. If so, then it is appropriate to look for causes for ongoing bone loss that is not halted by effective antiresorptive drugs. It is also worth noting that the early change in markers in clinical trials has generally been a better predictor of antifracture effectiveness of the therapy than changes in BMD.
CHAPTER 26 Assessment of Bone Density and Bone Strength
E. Bone Microarchitecture The process of menopause- and age-related bone loss results in the loss of connectivity of the elements of cancellous bone and in thinning of the cortical shell. This can be crudely assessed by plane radiographs but not to an extent that it is widely clinically applicable. Abnormal bone microarchitecture can also be assessed by 2D and 3D analysis of bone biopsy specimens. As noted, this is impractical in most clinical situations. Modalities for noninvasive in vivo assessment of bone microarchitecture are in various stages of development and clinical application. High-resolution peripheral Q CT (pQ_CT) has been developed for measurement of bone microarchitecture at the distal radius and distal tibia (Fig. 26.2, A) (see color insert), and preliminary data show that it has potential to predict future fracture occurrence (26). Highresolution MRI, termed virtual bone biopsy (VBB), has also been developed for assessment of microarchitecture at the distal radius and distal tibia (Figs. 26.2, B [see color insert], 26.3, and 26.4) (27,28). Short-term studies in postmenopausal women randomized to estrogen or placebo have demonstrated that within 1 year, microarchitectural disruption is seen in those women on placebo, with preservation of microarchitecture in those receiving estrogen (data only available as an abstract). Similar data have been punished for hypogonadal men receiving testosterone or placebo (29), and very recent data indicate the relationship between abnormal microarchitecture determined by VBB and the presence or absence of osteoporosis-related fractures (data only available as an abstract). These data are very encouraging but quite pre-
335 liminary because the technology for high-resolution p Q C T and VBB is not yet widely deployed. Nonetheless the available data demonstrate that microarchitectural changes can be seen earlier than changes seen with DXA and that smaller numbers of subjects are needed for serial studies than are needed for studies employing DXA. Although not directly clinically applicable, this latter finding will likely improve the ability to perform side-by-side comparison of available os-
FIGURE 26.3 Reproducibilityof VBB images.A: Repeat tibial scans in the patient with the highest Trabecular Volume/Bone Volume (TWBV). B: Repeat radius scans in the patient with the highest TWBV. (From ref. 28, with permission.)
FIGURE 26.2 High-resolution pQ.CT images of the distal radius. A: Radiograph showingthe site of imaging at the distal radius. The white line indicates the beginning of the joint space, and the two smaller lines to the left indicate the section of bone over which images are acquired. B: Representative cross-sectional images from the stack of CT slices from proximal (top left) to distal (bottom right). C: Representative 3D image. (From ref. 26, with permission.)
FIGURE 26.4 High-resolutionin vivo microscopyof the distal radius in a eugonadal man (left) and a hypogonadalman (right). (From ref. 29, with permission.)
336
MICHAEL KLEEREKOPER
teoporosis therapies and may make it more cost effective to conduct clinical trials of potential new therapies. It must be emphasized that, as with the biochemical markers, knowledge of bone microarchitecture is several years away from becoming a diagnostic tool for osteoporosis and risk of fracture, and a long-term prospective study in a large number of at-risk patients will be required for confirmation.
III. C O N C L U S I O N Q CT is the only technology that directly measures true volumetric BMD, but its availability is limited and there are few prospective studies of its true capabilities. DXA measures "areal" BMD, which is not a true physiologic measurement. However, despite this shortcoming, DXA has proven to be an invaluable clinical tool in the evaluation of osteoporosis. DXA has several shortcomings with respect to monitoring the progression or regression of osteoporosis. The technologies of biochemical markers of bone remodeling are now mature enough to be applied more routinely in clinical practice as an adjunct to DXA. Methods for detecting disruption in bone microarchitecture have been developed and refined, but the clinical data to determine their true clinical utility are not yet complete.
References 1. National Osteoporosis Foundation. Accessed March 23, 2007, from http://www.nof.org/osteoporosis/diseasefacts.htm. 2. Louis O, Van den Vinkel P, Covens P, Schoutens A, Osteaux M. Dualenergy x-ray absorptiometry of lumbar vertebra: relative contribution of body and posterior elements and accuracy in relation with neutron activation analysis. Bone 1992;13:317-320. 3. Hotchkiss CE. Use of peripheral quantitative tomography for densitometry of the femoral neck and spine in cynomolgus monkeys (Macaca fascicularis). Bone 1999;24:101-107. 4. Korver DR, Saunders-Blades JL, Nadeau KL. Assessing bone mineral density in vivo: quantitative computed tomography. Poult Sci 2004; 83:222-229. 5. Lorentzon M, Mellstrom D, Ohlsson C. Age of attainment of peak bone mass is site specific in Swedish men. J Bone Miner Res 2005; 20:1123-1227. 6. Toledo VA, Jergas M. Age-related changes in cortical bone mass: data from a German female cohort. Eur Radio12005;8:1-7. 7. Khan AA, Syed Z. Bone densitometry in premenopausal women: synthesis and review.J Clin Densitom 2004;7:85-92. 8. Kleerekoper M, Nelson DA, Flynn MJ, et al. Comparison of radiographic absorptiometry with dual-energy x-ray absorptiometry and quantitative computed tomography in normal older white and black women. J Bone Miner Res 1994;9:1745-1749. 9. Duboeuf F, Jergas M, Schott AM, et al. A comparison of bone densitometry measurements of the central skeleton in post-menopausal women with and without vertebral fracture. Br J Radiol 1995;68: 747-753. 10. Genant HK, Lang T, Fuerst T, et al. Treatment with raloxifene for 2 years increases vertebral bone mineral density as measured by volumetric quantitative computed tomography. Bone 2004;35:1164-1168.
11. Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 1996;312:1254-1259. 12. Schuit SC, Van der Klift M, Weel A, et al. Fracture incidence and association with bone mineral density in men and women: the Rotterdam study. Bone 2004;34:195-202. 13. K_rallEA, Dawson-Hughes B, Hirst K, et al. Bone mineral density and biochemical markers of bone turnover in healthy elderly men and women. J Gerontol Bid Sci Med Sci 1997;52:M61-M67. 14. Jorgensen HL, Warming L, Bjarnason NH, Andersen PB, Hassager C. How does quantitative ultrasound compare to dual x-ray absorptiometry at various skeletal sites in relation to the W H O diagnosis categories? Clin Physio12001;21:51-59. 15. Sahota O, Pearson D, Cawte SW, San P, Hosking DJ. Site-specific variation in the classification of osteoporosis, and the diagnostic reclassification using the lowest individual lumbar vertebra T-score compared with the L1-L4 mean, in early postmenopausal women. Osteoporos Int 2000;11:852-857. 16. Melton LJ III. The prevalence of osteoporosis: gender and racial comparison. Cakif Tissue Int 2001;69:179-181. 17. World Health Organization. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. World Health Organ Tech Rep Ser 1994;843:1-129. 18. Johnell O, Kanis JA, Oden A, et al. Predictive value of BMD for hip and other fractures. J Bone Miner Res 2005;20:1185-1194. 19. U.S. Preventive Services Task Force. Screening for osteoporosis in postmenopausal women: recommendations and rationale. Ann Intern Med 2002;137:526-528. 20. Kleerekoper M, Nelson DA. Is BMD testing appropriate for all menopausal women? IntJ Fertil Womens Med 2005;50:61-66. 21. http://www.nlm.nih.gov/pubs/cbm/osteoporosis/html. 22. Garnero P, Sornay-Rendu E, DuboeufF, Delmas PD. Markers of bone turnover predict postmenopausal forearm bone loss over 4 years: the OFELY study. J Bone Miner Res 1999;14:1614-1621. 23. Chestnut CH, Ettinger MB, Miller PD, et al. Ibandronate produces significant, similar antifracture efficacy in North American and European women: new clinical findings from BONE. Curr Med Res Opin 2005;21:391-401. 24. Johnell O, Scheele WH, Lu Y, et al. Additive effects of raloxifene and alendronate on bone density and biochemical markers of bone remodeling in postmenopausal women with osteporosis. J Clin Endocrinol Metab 2002;87:985-992. 25. Rosen CJ, Hochberg MC, Bonnick SL, et al. Treatment with onceweekly alendronate 70 mg compared with once-weekly risedronate 35 mg in women with postmenopausal osteoporosis: a randomized doubleblind study. J Bone Miner Res 2005;20:141-151. 26. Khosla S, Riggs BL, Atkinson EJ, et al. Effects of sex and age on bone microstructure at the ultradistal radius: a population-based noninvasive in vivo assessment. J Bone Miner Res 2006;21:124-131. 27. Gomberg BR, Saha PK, Wehrli FW. Topology-based orientation analysis of trabecular bone networks. Med Phys 2003;30:158-168. 28. Gomberg B R, Wehrli FW, Vasilic B, et al. Reproducibility and error sources of micro-MRI-based trabecular bone structural parameters of the distal radius and tibia. Bone 2004;35:266-276. 29. Benito M, Gomberg B, Wehrli FW, et al. Deterioration of trabecular architecture in hypogonadal men. J Clin Endocrinol Metab 2003;88: 1497-1502. 30. Benito M, Vasilic B, Wehrli FW, et al. Effect of testosterone replacement on trabecular architecture in hypogonadal men. J Bone Miner Res 2005 ;20:1785-1791.
2HAPTER 2,
Biochemical Markers of Bone Turnover RICHARD EASTELL AcademicUnit of Bone Metabolism, University of Sheffield, Metabolic Bone Centre, Sorby Wing, Northern General Hospital, Sheffield, UK $5 7AU
ROSEMARYA.
HANNON
AcademicUnit of Bone Metabolism, University of Sheffield, Metabolic Bone Centre, Sorby Wing, Northern General Hospital, Sheffield, UK $5 7AU
The past few years have seen an expansion of the use of biochemical markers of bone turnover in the care of postmenopausal women. Several new markers have become available and automated methods for measuring the markers have been developed, which means that most hospital clinical chemistry laboratories have the facilities to make these measurements. Measurement of markers of bone turnover is noninvasive and relatively inexpensive, and these markers can be measured repeatedly in the same subject. The major benefit of using bone turnover markers is that significant changes in response to treatment are observed within a few months, whereas significant changes in bone mineral density (BMD) are not observed for at least 12 months. In this chapter we review the most commonly used markers of bone turnover and discuss the use of these markers in the care of the postmenopausal woman.
I. MARKERS OF B O N E T U R N O V E R Biochemical markers of bone turnover are generally classified as markers of bone formation or markers of bone resorption (Table 27.1), reflecting osteoblast and osteoclast activity, respectively. The ideal marker is bone specific and reflects either bone formation or bone resorption, but not both. However, it must be kept in mind that bone formation and bone resorption are "coupled" events in most circumstances, such that any marker of bone formation or bone resorption can be used to characterize both the high turnT R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
337
over state after the menopause and the decrease in bone turnover in response to antiresorptive treatment.
A. Markers of Bone Formation 1. PROCOLLAGEN PROPEPTIDES
The procollagen propeptides, procollagen type I carboxyterminal propeptide (PICP) and procollagen type I aminoterminal propeptide (PINP), are released into the circulation during the extracellular conversion of procollagen to collagen. Type I collagen is a trimer comprising two identical oq (I) polypeptide chains and one o~2(I)polypeptide chain and is the major protein of bone matrix (90%) (1). It is synthesized by osteoblasts as a precursor, procollagen, and secreted into the extracellular space, where it undergoes a series of post-translational modifications including the cleavage of the propeptides by specific proteases (Fig. 27.1). The propeptides are released into the circulation, and the newly formed collagen molecules assemble as fibrils. PICP is a globular polypeptide with a molecular weight of 100 kDa, which contains both intrachain and interchain disulphide bridges. PINP is a smaller polypeptide with a molecular weight 35 kDa, with intrachain disulphide bridges but no interchain disulphide bridges. It is primarily a globular peptide but contains small amount of helix. Intact PINP appears to be a more sensitive marker of bone formation than PICP: the increases in PINP during the pubertal growth spurt, at the menopause and in Paget's Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
338
EASTELL
disease, are significantly greater than those for PICP (2). Likewise, the percentage decrease in PINP after one year of HRT was ~40% compared with ~20% in PICP in 47 postmenopausal women (1). There are several possible reasons for this difference in sensitivity. Although the contribution to circulating levels of the propeptides from nonosseous sources is thought to be small (2), differences in the patterns of cleavage of the propeptides from procollagen in different tissues may be responsible for intact PINP being more bone specific than PICP. PICP is completely cleaved and released in amounts equimolar to the amount of collagen synthesized in any tissue, whereas PINP is completely cleaved in collagen found in mineralized tissues but in soft tissues may be retained in fibers and released during fiber growth or collagen degradation in a degraded form (3) that is not detected by the intact PINP assay but is detected by the total PINP assay, which detects the intact molecule and other smaller fragments of the molecule. Differences in the clearance mechanisms of the two propeptides may also account for some of the differences in sensitivity. PICP is cleared by the mannose receptor of the liver endothelial cells, which are sensitive to hormones including estrogen, whereas PINP is cleared by the scavenger receptor of liver endothelial cells, which are not thought to be sensitive to circulating hormones (3-5). The greater sensitivity of PINP has meant that it is measured in preference to PICP in most studies.
TABLE 27.1 Bone Turnover Markers Specimen
Marker Bone resorption markers Cross-linked N-telopeptide of type I collagen (NTX) Cross-linked C-telopeptide of type I collagen (CTX)
Urine, serum Urine (cxotand [313 forms) Serum ([3[3 form) Serum
MMP-generated telopeptide of type I collagen (ICTP or CTX-MMP) Deoxypyridinoline, free and peptide bound (fDPD, DPD) Pyridinoline, free and peptide bound (fPYD, PYD) Hydroxyproline (OHP) Glycosyl hydroxylysine (GylHyl) Helical peptide (HelP) Tartrate resistant acid phosphatase 5b isoform specific for osteoclasts (TRACP 5b) Cathepsin K (Cath K) Osteocalcin fragments (uOC) Bone formation markers Osteocalcin (OC) Procollagen type I C-terminal propeptide (PICP) Procollagen type I N-terminal propeptide (PINP) Bone-specific alkaline phosphatase (bone ALP)
Urine, serum Urine serum Urine Urine, serum Urine Serum, plasma Urine, serum Urine Serum Serum Serum Serum
PROCOLLAGEN
MOLECULE ##
'I
,,
1
(150~1
,, ,,
I
/ G,obu,o,.
Oomo,n
I
I
/
/
=i
' I
,
, I
I
: 0
:
; I
I
:
,
~G,c
-Go,
.
-Go,
:
.
/l
Triple-Helical Domain / Nontriple- Helical Domain
1
,
,'
,
,
i
I
I
I
/
~_C-TERMINAL_.~ PROPEPTIDE
0
I
'
:
-
i
}'~'---'------ COLLAGEN MOLECULE 0 , ( 3000 A )
: 20~)
'I'G~
$
'
N-TERMINAL PROPEPTIDE
I I ( Mon)n 9 GIcNoc
/
"~
:
AND HANNON
III
v //
.
-Triple-Helical
Nontriple- Helical Domain
Domain
%
'
,
(,ooh)
'
I
1
I
Man)n ', I i[GIcNocn II
--
(
.~~_FI'"~,L., $
(,oo~) !
i
Nontriple - -elicol Domo0q
FIGURE 27.1 Extracellularcleavage of the propeptides of type I collagen, illustrating the origins of markers of bone formation, procollagen type I carboxyterminal propeptide (PICP) and procollagen type I aminoterminal propeptide (PINP). (From ref. 32, with permission.)
CHAPTER 27 Biochemical Markers of Bone Turnover 2.
OSTEOCALCIN
Osteocalcin (bone Gla protein) is the most abundant noncollagenous bone matrix protein, also found in dentin. It is a small protein of 49 amino acid residue and molecular weight 58 kDa, which is synthesized almost exclusively by highly differentiated osteoblasts and is therefore bone specific, an important property for a marker of bone formation (6). The majority of newly synthesized osteocalcin is incorporated into the bone matrix, but about 10% to 30% is released into the circulation. Osteocalcin is cleared by the kidneys and to a lesser extent by the liver. In renal failure, levels of osteocalcin and particularly its fragments are elevated (7). The function of osteocalcin is not fully understood. It has been suggested that it is involved in the regulation of coupling of bone resorption and bone formation (8). In a study of osteocalcin knockout mice, bone formation was increased but bone resorption was unaffected, which suggests that osteocalcin may be a limiting factor for bone formation (9). A major feature of the primary structure of osteocalcin is the y-carboxylated glutamic acid residues at positions 17, 21, and 24. The post-translational y-carboxylation of these residues is vitamin K dependent. Osteocalcin in bone is not completely carboxylated, and serum levels of the undercarboxylated form increase with age, especially in elderly women (10). The degree of carboxylation appears to be related to bone strength, since Vergnaud et al. (11) have shown that increased serum levels of undercarboxylated osteocalcin predict hip fracture risk independently of femoral neck BMD. Elderly women with hip fracture have lower levels of vitamin K1 and menquinone 7 and 8, two major components of vitamin I(2 (12). Although numerous immunoassays are available for the measurement of serum or plasma osteocalcin, comparability of the different assays is poor (13). This is probably due to the heterogeneity of the circulating fragments of osteocalcin and differences in the immunorecognition of these fragments by different assays. Intact osteocalcin is degraded in the circulation and in vitro (14). Only about 36% of circulating immunoreactive osteocalcin is the intact protein, 30% is a large N-mid-molecule fragment (residues 1-43), and the remainder consists of smaller fragments. The apparent stability (15) of immunoreactive osteocalcin is significantly greater when assays designed to measure the intact molecule and Nmid-molecule fragment are used than when assays designed to measure only the intact molecule are used. Assays that measure both the intact and N-mid-molecule fragment are therefore recommended, particularly in clinical practice. Recently, small fragments of osteocalcin have been isolated in urine (16), which are thought to be degradation products of bone matrix rather than byproducts of bone formation. This suggests that osteocalcin may also be a marker of bone resorption. Indeed it is now possibly best to regard osteocalcin as a marker of bone turnover rather than a specific marker of bone formation.
339 3. ALKALINE PHOSPHATASE
Total serum alkaline phosphatase has been used for many years as a marker of bone formation. Alkaline phosphatase is a membrane-bound orthophosphoric mono-ester phospho2-hydrolase with optimal activity at alkaline pH. Several different isoenzymes of alkaline phosphatase exist, coded for by four different genes: tissue nonspecific, intestinal, placental, and germ cell line. The tissue nonspecific gene codes for bone, liver, and kidney isoforms; the differences in these isoforms result from differences in the patterns of posttranslational glycosylation (17). In healthy adults, circulating alkaline phosphatase consists primarily of the bone isoform (~50%) and liver isoform (~50%). Only trace amounts of the intestinal and placental isoenzymes are found in normal adult serum. The precise function of the bone isoform remains unclear, but it is thought to be associated with mineralization. Total alkaline phosphatase is a useful marker of bone formation in diagnosis and monitoring of treatment in diseases characterized by major disturbances in bone turnover, such as Paget's disease, and some would argue that in most clinical situations, measurement of total alkaline phosphatase provides sufficient information (18). However, there is evidence to support the contrary argument that the bone alkaline phosphatase (bone ALP) is a more sensitive and more clinically useful marker in the measurement of more subtle changes in bone turnover, such as occur at the menopause (19,20). Garnero et al. (19) found a 77% increase in bone alkaline phosphatase after the menopause compared with a 24% increase in total alkaline phosphatase. Manual and automated immunoassays are available that are specific for the bone isoform. There is some crossreactivity with the liver isoform in most of these assays, but this is small (between 7% and 16%) and should only be a problem in patients with liver disease (19,21).
B. Markers of Bone Resorption 1. DEGRADATION PRODUCTS OF THE TELOPEPTIDE REGION OF TYPE I COLLAGEN
The majority of the markers of bone resorption are the degradation products of type I collagen. The most widely used of these are the peptide bound cross-links of type I collagen, the cross-linked carboxy and amino telopeptides of type I collagen (NTX and CTX), which are released into the circulation during bone resorption (22). A proportion of these are excreted into the urine, and the rest are degraded in the kidneys and eventually excreted in urine as free cross-links. Fig. 27.2 shows the origins of these markers in collagen. The pyridinium cross-links, pyridinoline and deoxypyridinoline, are nonreducible cross-links found in mature collagen, formed by the condensation of lysine and hydroxylysine residues in the telopeptide regions and specific lysine or hydroxylysine
340
EASTELL AND HANNON
N-Telopeptide ~ crosslink ~ ,, - - . ~ = ,.~,~. . . . .
H,N ..COOH .CH CH 9 91"12 /NH2 "! - ~ - - - - - ~ - H ~ CH'CH N ~--~=:1~.%~'.="=.%%i'.'.'.%~1 COOH I I
=.,I "-.-.-,=-.,,.,-=-.l.,.-.=.iF:= =
C A
~~='-=--=--' - -~-=-~"~--=C=---' - - - - - - -'*"~
930
~ "k,, ~ ' ' ~ -- C-Telopeptide crosslink
PH= ~H'OH pi-l= pH=
,,.CH~ H2N COOH
H,N ..COOH .CH CH ,Cl-lz ~ /NI-12 H ~ + CH'CH COOH PH ~H= pH= pH= 2
..CH HzN COOH
B
FIGURE 27.2 Schematic diagram illustrating the location of the N- and C-telopeptide cross-links in type I collagen and the structures of pyridinoline and deoxypyridinoline.
residues in the helical regions of adjacent collagen molecules. Free and peptide-bound forms of cross-finks are detectable in both serum and urine (23). In adult urine, 40% to 50% are free and 50% to 60% are bound (24), but the ratio of free to bound in both serum and urine may depend on the level of bone turnover; the higher the rate of bone turnover, the smaller the ratio of free to bound crosslinks (23,25). NTX can be measured in both urine and serum. The assay measures crosslinked N terminal telopeptides fragments of type I collagen which contain a specific 8 amino acid sequence from the ot2(I) N telopeptides o~2(26,27). The peptide component of CTX is an 8-amino acid sequence containing an aspartic acid, glycine peptide, which is highly susceptible to isomerization and racemization. This yields 4 isoforms of the telopeptide, the otL, [3L, otD, and [3D forms. The native L isoform is found in newly formed bone, whereas the other forms are found in increasing amounts as bone matures (28). Therefore a heterogeneous mix of the different isoforms of CTX molecules is found in urine and serum. Several assays have been developed to measure these different forms in urine (29). There is a radioimmunoassay that measures cio~ and cx[3 cross-linked molecules and also the single-chain, noncross-linked o~ forms of the peptide in urine, and an enzymelinked immunoassay (EHSA) that measures [3[3 and o~[3 and the single-chain, non-cross-linked [3 in urine. Recently two more specific EHSAs have been developed to specifically measure the cross-linked cici forms and the cross-linked [3[3 forms of CTX in urine. Neither assay measures single chain fragments of the Type I collagen telopeptides. The serum assay for CTX is probably currently the most widely used of the CTX assays is specific for the cross-linked [3[3 form. The free cross-links, pyridinoline and deoxypyridinoline, can be measured by immunoassay. Both the free and peptidebound cross-links can also be measured by High Performance Liquid Chromatography (HPLC), although this method has been used less since the development of immunoassays and automated methods.
Another C-terminal telopeptide of type I collagen that can be used as a marker of bone resorption is CTX M M P or ICTP (30). It has a longer peptide sequence than CTX and is released during the degradation of type I collagen by matrix metallopeptidases. It is a poor marker in osteoporosis, but may be useful in oncology as several studies have found highly elevated levels of ICTP in patients with metastatic bone disease (31). 2. COLLAGEN DEGRADATION PRODUCTS FROM THE HELICAL REClON OF TYPE I COLLACEN Hydroxyproline, possibly one of the longest-established markers of bone resorption, is a product of the degradation of the helical region of type I collagen. It represents about 13% of the amino acid content of collagen and is released into the circulation during bone resorption but is not reincorporated into new collagen (32). It circulates in either the free form (90%) or in peptide-bound forms. Most of the free hydroxyproline is filtered and reabsorbed by the kidney. The remainder is excreted in the urine together with the dialyzable and non-dialyzable peptide-bound forms. However, urinary hydroxyproline suffers from three major disadvantages as a marker of bone resorption: (1) a significant fraction of it is derived from nonosseous sources (33); (2) hydroxyproline is absorbed from dietary sources and diet must be restricted for 24 hours before samples are collected (34); and (3) most hydroxyproline is metabolized in the liver. It has now been largely superseded by the telopeptides. The other markers derived from the helical region are the galactosyl and glucosylgalactosyl hydroxylysines, which are products of the post-translational glycosylation of type I collagen. They are technically demanding to measure and are not widely used (35). However, an immunoassay has recently been developed that measures the so-called helical peptide, which is the 620-633 peptide of the (xI chain of type I collagen (36).
CHAPTER 27 Biochemical Markers of Bone Turnover 3. OSTEOCLASTICENZYMES
The other resorption markers are osteoclast enzymes.
a. Tartrate Resistant Acid Phosphatase Tartrate resistant acid phosphatase (TRAcP) 5b is an osteoclast enzyme. Serum TRAcP has been used for many years as a marker of bone resorption. However, until recently the kinetic methods used to measure TRAcP were not specific for the 5b isoform and detected TRAcP from other sources such as platelets, erythrocyte, and possibly osteoblasts (37). In the past few years, immunoassays have been developed that are specific for the 5b isoform (38). Using these 5b-specific assays, it has been shown that serum levels of TRAcP 5b reflect osteoclast number rather than osteoclast activity. b. Cathepsin K and Matrix Metalloproteinases Two newer markers of bone resorption are also osteoclast enzymes: cathepsin K and matrix metalloproteinase (MMP). Cathepsin K is a cysteine protease expressed by osteoclasts, which is essential for bone resorption (39). Preliminary evaluation has shown that serum levels are elevated in postmenopausal osteoporosis and in women with a history of fracture (40).
C. Summary There is now a large range of markers of bone turnover available, most of which can be measured reliably by simple ELISAs. Some may also be measured on automated analyzers. Different markers reflect different aspects of bone formation or resorption, and this must be considered when choosing markers for a particular study or clinical situation. In clinical research we usually recommend measuring two markers of bone formation and two markers of bone resorption, and in clinical practice, one marker of bone formation and one marker of bone resorption. PINP and bone alkaline phosphatase (ALP) are generally the most useful markers of bone formation in large clinical studies and clinical practice, and urinary NTX and serum CTX are the most useful markers of bone resorption. These markers tend to be more stable than some of the others, which is an important consideration, particularly in clinical practice.
II. VARIABILITY OF MARKERS Although biochemical markers are useful noninvasive tools that can be used repeatedly in the same patient to assess bone turnover, one of the major confounding factors for their use is their variability. The main components of overall intraindividual variability are analytical and biological. Analytical variability is small compared with biological variability for most assays (41,42). There are numerous
341 sources of variability such as age, gender, pubertal stage, menopausal status, diseases, drugs, fractures, menstrual, seasonal, day to day, and circadian. These sources of variability may be divided into two categories, the first being those that can be controlled by collecting samples under the appropriate conditions (43). These controllable sources of variability include circadian, menopausal, seasonal, or effect of exercise. The second category of sources of variability comprises the uncontrollable sources such as age, gender, diseases, drugs, and fractures. The effect of these uncontrollable sources of variability must be taken into account in the interpretation of results by the use of appropriate references ranges and a knowledge of the effects of diseases, drugs, and fractures on bone turnover. The intraindividual variability is greater for the urinary markers than for the serum markers. The intraindividual variability for serum markers is in the order of 10%, whereas for urinary markers it may be as much as 35% (41,44). This difference may be due to difficulties in timing and completeness of urine collections or to day-to-day changes in renal clearance or the renal conversion of peptide bound to free cross-links. Variability may also be affected by osteoporosis and tends to be higher for all markers in women with postmenopausal osteoporosis than in normal postmenopausal women (45). Intraindividual variability has considerable effect on the use of markers in research and in clinical practice. A useful concept that accounts for the intraindividual variability when deciding whether a true biologic change in a marker has occurred is the "least significant change." The least significant change is calculated from the intraindividual variation and represents a cut-off point. Any change between two measurements must be greater than this cut-off point to be biologically significant. Circadian rhythms can have a major effect on variability of the markers. Both markers of bone resorption and bone formation exhibit circadian rhythms, reaching a peak between 23:00 and 08:00 hours and falling to a nadir between 12:00 and 17:00 hours. The greatest circadian changes are seen in the markers of bone resorption. In young women the amplitude of the variation (i.e., the difference from peak to trough) is 70% of the mean value for 24 hours for total DPD (Fig. 27.3) and 63% for NTX (46). The amplitude of circadian rhythm free of free cross-links tends to be smaller than that of the telopepetides. Serum CTX, for example, exhibits a circadian rhythm with an amplitude of 80% of the 24 hour mean (47). The amplitude of the variation for bone formation markers range from 5% to 30% (48-52). These circadian rhythms are maintained into old age (55). In late postmenopausal women with osteopenia or osteoporosis, the circadian rhythm of some biochemical markers of bone turnover appears to be disturbed. The
342
FmURE 27.3 Circadian variation in excretion of DPD in 18 healthy premenopausal women. Results are shown as percentages of the 24-hour mean excretion in each subject. Each thin line represents one individual. The solid lines represent the mean _+ standard error of the mean (SEM) of the normalized values. Data are plotted at the midpoint of each 4-hour collection period. Subjects were recumbent with the lights off during the period indicated by the shaded bar. (Modified from ref. 46, with permission.)
peak in total DPD excretion is extended into the morning in women with osteoporosis compared with healthy agedmatched women (53). However, in women with osteopenia (forearm bone mineral content [BMC] more than 2 SDs below premenopausal mean BMC), the circadian rhythm in total DPD is the same as that for age-matched women (54). The nocturnal peak in PICP, but not in osteocalcin or bone alkaline phosphatase, is higher and extended in the osteopenic women (49). The implications of these circadian rhythms are considerable, both in investigative studies and in clinical practice. For example, if NTX is measured in a urine sample collected in the early morning (07:00) and in a second sample collected in the afternoon (15:00), there could be a difference in the measurements of 50%, which is comparable to the change in NTX expected after 6 months of hormone replacement therapy (HRT). Several recent studies have shown that much of the circadian rhythm, particularly that of CTX, is due to the effect of feeding (55,56). Menstrual and seasonal variations in markers are much weaker than the circadian variations. Markers of bone formation have been shown to be 15% higher in the luteal phase than in the follicular phase (57). Urinary levels of markers of resorption, NTX and total pyridinium cross-links, are somewhat higher during the earlier parts of follicular and luteal phases than in the later parts of these phases (58). ICTP, a serum marker of bone resorption, is 17% higher in the luteal phase than in these follicular phase (59).
EASTELL AND HANNON
Seasonal variation in markers of bone turnover has been observed in several but not all studies (60-65). In most studies that have demonstrated a significant seasonal rhythm, bone turnover tends to be lower in the summer than in the winter (61,62) and be related to seasonal changes in the vitamin D-parathyroid hormone (PTH) axis. However, in a recent multicenter European study of premenopausal and postmenopausal women, no wintertime increase in bone turnover was observed despite a significant seasonal decrease in vitamin D levels (64). The impact of fractures and drugs on intraindividual variability of markers is particularly important for postmenopausal women as they have an increased risk of fracture. Biochemical markers of bone formation and bone resorption increase up to 30% during the healing of a Colles' fracture, a characteristic osteoporotic fracture (66). Furthermore, the markers can remain elevated for up to 1 year after the fracture (67). In addition to drugs such as HRT and bisphosphonates taken to prevent bone loss, other drugs can affect levels of biochemical markers. Corticosteroids suppress bone formation without having a direct effect on bone resorption. Consequently, markers of bone formation, in particular osteocalcin, are reduced during corticosteroid treatment. There may be a small increase in markers of bone resorption, resulting from a secondary hyperparathyroidism induced by the effect of the corticosteroid on calcium absorption. Thiazide diuretics reduce bone turnover (68). Conversely, treatment with anticonvulsant drugs results in threefold increases in markers of bone resorption (69) and increases of up to 33% in markers of bone formation in women (70).
A. Summary A number of strategies can be followed to minimize variability: (1) samples, particularly for resorption markers, should be collected at a fixed time of day after an overnight fast; (2) two or more samples can be collected on different days within a given period and the mean level of the marker reported; and (3) patients should be asked about previous fractures and use of drugs.
III. CHANGES IN MARKERS AT MENOPAUSE, IN OSTEOPOROSIS, AND IN RESPONSE TO THERAPY A. Menopause Estrogen deficiency caused by ovariectomy or drug therapy, such as gonadotropin-releasing hormone (GnRH) agonist therapy, result in a rapid increase in the levels of
CXaVTER 27 Biochemical Markers of Bone Turnover
343
140 -
o
120 -
o
40
.
loo-
:,. 9
9
"~ == 80
Total D-Pyr
Total Pyr ~
;,~
, , ~ . ~ .~ . . . . . ~
~
l 9 ell
,:
~o
20
..
-
40
........
,.,:.
,
99
v
FIGURE 27.4 Individual values for four biochemical markers of bone resorption in 221 perimenopausal women and postmenopausal women, as well as postmenopausal women on hormone replacement therapy. For each marker, the solid line represents the group mean; the dashed line and dotted lines in the background represent, respectively, the mean +_1 SD for premenopausal women. A: Total pyridinoline and total deoxypyridinoline. B: N T X and free deoxypyridinoline. (From ref. 73, with permission.)
_o a0 ~=
9
.........
:
~:,
.
.
.
.
~
.... :~,.
- -~.~
. ....... ,lk'o........ o-.~oo:
20
P
9
o~~
.:"
Plrie
' 'r Post
1 HRT
0
t Peri
NTx
t Post
I HRT
Free D-Pyr
300 -
15
0
"C"
0
E 10
e 9
--.=200
UJ
o 03 O
z
9
ee
Oo
E .~-~100 J o " s
1,1~.
08 9 o
o:
:7.~ ~ 0
Peri
markers of bone turnover (71). Similarly the menopause is marked by an increase in levels of markers. The magnitude of the increase varies for the different markers, probably reflecting their specificity for bone or differences in their metabolism in low and high turnover states. The markers probably start to increase in the perimenopausal period before menstruation has ceased. Premenopausal women over the age of 40 years have higher levels of CTX and osteocalcin than younger women (72), which may be weakly associated with lower BMD. In women with altered menstrual patterns, where follicle-stimulating hormone (FSH) is elevated but estradiol is still at premenopausal levels, N T X is increased by 20% but formation markers are unchanged, compared with premenopausal levels (73) (Fig. 27.4). Perrien et al. have shown that the changes in bone turnover in premenopausal and perimenopausal women may be regulated by inhibins, particularly inhibin A, independently of FSH and estrogen levels (74). Once the estradiol levels decrease, markers of bone resorption and formation increase (75). In the first years of the menopause, mean levels of pyridinium cross-link excretion significantly increase and may even be doubled (26,76,77), but there is still considerable overlap with levels in premenopausal women. The increases in the telopeptides or total cross-links are greater than the increase in the free pyridinolines, possibly for reasons discussed previously (73,78). Garnero et al. (79)
Post
I
HlqT
I
Peri
I
Post
I
HRT
found increases of 80% to 100% in total cross-links measured by HPLC, 100% to 130% in telopeptides, and 50% in free deoxypyridinoline in a comparative study of 14 premenopausal women ages 33 to 44 years and 29 postmenopausal women ages 46 to 53 years who were within 3 years of the menopause. There is a considerably smaller increase in the serum markers of bone resorption, T R A C P and ICTP, of the order of 20% to 25% (80). Interestingly, cathepsin K appears to be lower in older women compared with premenopausal women (81). The differences in the increases in markers of bone formation in postmenopausal women may be related to the different aspects of bone formation that each marker reflects. It may also reflect a difference in tissue specificities of the markers. The increases in bone alkaline phosphatase and osteocalcin are in the order of 50% to 100% but may be as high as 150%. The increase in PICP is considerably less, around 20%. PINP, which may be the more bone-specific propeptide, shows a greater increase at the menopause (82). Whether the increase in bone turnover that occurs at the menopause is maintained into old age has been questioned (83), but we have found elevated levels of bone turnover in women into their eighth decade (84). Garnero et al. have shown that N T X and CTX and markers of bone formation remain elevated in women for 40 years after the menopause (85).
344
EASTELL AND HANNON
B. Osteoporosis Markers of bone resorption are significantly elevated in osteoporotic postmenopausal women as compared with normal postmenopausal women, but the markers of bone formation are much less elevated and may indeed be decreased (86,87). This pattern of changes suggests that a degree of imbalance of bone resorption and bone formation occurs in osteoporosis. We found a mean increase of 40% in the level of total deoxypyridinoline in a group of 63 women with postmenopausal osteoporosis compared with a group of 67 normal postmenopausal women (88). However, there was a considerable overlap of individual levels in the two groups. The heterogeneity of bone resorption in the osteoporotic group probably indicates different causes of the disease or may represent differences in the stage of the disease observed in the individuals in the study. Single measurements of total and free cross-links are unlikely to be useful in identifying osteoporosis in an individual postmenopausal woman. However, a recent study suggests that NTX can discriminate between normal postmenopausal, osteopenic, and osteoporotic women, as defined by the World Health Organization ( W H O ) criteria (89). Recently, Meier et al. have shown that the new marker of resorption, cathepsin K, is significantly increased in postmenopausal women with osteoporosis compared with healthy women (11.3 versus 3.1 nmol/L) (90).
In contrast to the antiresorptive drugs that suppresses bone turnover, the new anabolic treatment for osteoporosis, teriparatide, stimulates bone turnover, which when combined with a positive remodelling balance leads to an increase in bone mass (96). Both bone formation and bone resorption markers are increased in response to teriparatide (97). The procoUagen propetides peak 6 months after initiation of treatment: PINP is increased by 200% (98) and PICP by 60% (99), after which they return towards pretreatment levels. However, bone ALP remains elevated throughout treatment, as do markers of bone resorption, such as urinary NTX, which may be elevated by 200% after 12 months of treatment (99). Strontium ranelate has yet another mode of action, which is not fully understood; it results in small increases in bone formation and a small decrease in bone resorption. However, the changes in bone turnover markers in response to sodium ranelate are smaller than those seen in response to other treatments: serum CTX decreases by 12% and bone ALP increases by 8% (100).
IV. PREDICTION OF BONE LOSS, FRACTURE, AND RESPONSE TO THERAPY A. Bone Loss
C. Response to Therapy Currently bisphosphonates are the treatment of choice in postmenopausal osteoporosis. Bisphosphonates are antiresorptive agents that suppress osteoclast activity. Markers of bone resorption decrease rapidly in response to treatment. Typically there is a decrease of 50% to 70% in the telopeptides within the first 12 weeks of treatment (91,92). We have shown that there is a significant decrease in urinary NTX within the first 8 weeks of treatment with alendronate. The changes in other markers of bone resorption are somewhat smaller. Serum levels of TRACP5b are reduced just under 20% (93). The markers of bone formation are also significantly reduced in response to bisphosphonates. PINP is reduced by as much as 60% and bone ALP more modestly by about 40% (91). Although the efficacy of alendronate and risedronate in reducing fractures is similar, the decrease in bone turnover appears to be greater in patients treated with alendronate. Treatment with selective estrogen receptor modulators (SERMs) such as raloxifene have a smaller effect on markers of bone turnover than do bisphosphonates. Serum CTX decreases by 30% to 40% and TRACP 5b by only 10%. The response of the markers of bone formation is also smaller: PINP is decreased by 30% and bone ALP by 15% to 20% (94,95).
BMD in the postmenopausal woman is determined by peak bone mass and the amount of bone loss since the menopause. With increasing age, the contribution of bone loss to the resultant BMD becomes more significant (101). Furthermore, increased bone turnover becomes an increasingly important determinant of BMD as women grow older (86). In several studies, bone loss has been shown to correlate with markers of bone turnover, although this is not a universal finding. We have shown that in early postmenopausal women there is an inverse relationship between markers of bone turnover and bone loss (102). In other studies of women in the first 7 years after the menopause, a combination of resorption and formation markers or a combination ofpyridinoline with estradiol glucuronide and body mass index can predict up to 59% of the variance of bone loss at the forearm (103,104). Change in BMD at the spine and hip, the clinically relevant sites, can only be poorly predicted, if at all (105-107). At the forearm, high levels of PINP, osteocalcin, urinary NTX, and serum CTX are associated with rapid bone loss (108). In women over age 70, the correlation between baseline levels of biochemical markers and annual rate of change of BMD over a 3-year period at the total hip was moderate at best (106). Although most of these studies indicate that a relationship exists between high levels of bone turnover markers and increased bone loss, they
CHAPTER27 Biochemical Markers of Bone Turnover are not sufficiently sensitive to be used to predict bone loss in an indMdual postmenopausal woman.
B. Fracture Although biochemical markers of bone turnover may be able to predict bone loss and hence fracture risk, they may also predict fracture risk independently of BMD. High bone turnover per se can disrupt the trabecular architecture by increasing the incidence of trabecular perforation and buckling, thus reducing bone strength, without necessarily affecting BMD significantly. In retrospective and prospective studies, fracture risk may be associated with increased levels of markers of bone resorption but not bone formation (109). In a short prospective study of elderly French women, a 1SD increase in CTX and free deoxypyridinoline, adjusted for femoral neck BMD and gait, above the upper limit for premenopausal women, resulted in 2- and 1.7-fold increases, respectively, in risk of hip fracture over a 22-month followup period (110). Increased pyridinolines have also been shown to predict a history of fracture independent of BMD, age, and other markers of bone turnover (111). However, there is little convincing evidence to indicate that a single measurement of a biochemical marker of bone turnover can predict fracture risk in an individual woman, even over a short period of time. The combination of a biochemical marker and BMD may be a much more powerful predictor of fracture than BMD alone (112). Women with a low hip BMD (<2.5 SD below premenopausal mean) and high CTX (>2 SD above premenopasusal mean) have a considerably higher risk of hip fracture than those who have only one of these independant risk factors (109). Biochemical markers of bone turnover cannot substitute for serial BMD measurements but may be useful when considered in conjunction with BMD measurement.
C. Response to Therapy Early changes in bone turnover markers in response to treatment may be predictive of change in BMD and fracture risk. A number of studies have shown that changes in CTX and NTX after the start of treatment predict changes in lumbar spine BMD obtained 2.5 to 4 years later in elderly women treated with alendronate (113). Early change markers of bone turnover may predict change in hip BMD in elderly women treated with alendronate or HRT (114). Similar results have also been reported in the Danish cohort of the Early Postmenopausal Intervention Cohort (EPIC) Study for younger postmenopausal women treated with alendronate for the prevention of osteoporosis (115). Chen et al. have shown that increase in PICP at 1 month and PINP at 3 months was predictive of change at lumbar
345 spine BMD at 18 months in osteoporotic women treated with teriparatide (97). In addition, baseline levels of markers were also weakly predictive of response to treatment. Bauer et al. have recently published similar findings (116). In addition to predicting change in BMD, early changes in bone turnover markers may predict change in fracture risk. We have shown that magnitude of changes in urinary NTX and CTX 3 to 6 months after the start of risedronate treatment is related to the decrease in vertebral fractures in women with at least one previous vertebral deformity (117). In the Multiple Outcomes of Raloxifene Evaluation (MORE), a randomized, placebo-controlled trial of 7705 osteoporotic women treated with raloxifene for 3 years, change in serum osteocalcin was predictive of change in vertebral fracture risk and was a better predictor than change in femoral neck BMD in the women treated with raloxifene (118).
V. M O N I T O R I N G THERAPY For the individual woman, the change in BMD in the first year of antiresorptive treatment is not large enough, relative to the precision of the measurements, to determine whether she has responded to the treatment. Thus, it is often normal clinical practice to wait for 1 to 2 years before repeating the bone density measurement. The advantage of using markers of bone turnover to monitor response is that a significant change may be seen between 3 and 6 months after initiation of treatment. If markers are to be used in monitoring therapy, it is important to remember that the magnitude of the observed change per se may yield very little and possibly misleading information if it is not considered relative to the intraindividual variability of the marker concerned. For a woman who has two single measurements at baseline and after 6 months treatment, for example, the change must be greater than the least significant change (LSC) (described previously) for that marker if she is to be classified as a responder. An understanding of the time course of the changes in markers, particularly in the first 6 months, is also important in making decisions based on monitoring response with markers. In addition to being useful in assessing response to treatment, measuring markers of bone turnover during the first 6 months of treatment and reporting the results to the patient may encourage compliance (119).
VI. CONCLUSION Biochemical markers of bone turnover have proved to be useful, noninvasive, and relatively inexpensive tools for studying bone metabolism in population studies and are gradually becoming established in clinical practice. In the treatment of the individual postmenopausal woman, they
346
EASTELL AND HANNON
may be useful as adjuncts to B M D and other diagnostic tests. Their main use, however, is in monitoring response to treatment. The continued development of new markers of bone turnover will increase our knowledge of the pathophysiology of osteoporosis and other metabolic bone diseases, and after further evaluation, these new markers may find a place in the clinical care of postmenopausal women.
References 1. Suvanto-Luukkonen E, Risteli L, Sundstrom H, et al. Comparison of three serum assays for bone collagen formation during postmenopausal estrogen-progestin therapy. Clin Chim Acta 1997;266:105116. 2. Risteli J, Melkko J, Niemi S, Risteli L. Use of a marker of collagen formation in osteoporosis studies. CalcifTissue Int 1991;49:$24-$25. 3. Risteli J, Niemi S, Kauppila S, Melkko J, Risteli L. Collagen propeptides as indicators of collagen assembly. Acta Orthop &and Suppl 1995;266:183-188. 4. Melkko J, Hellevik T, Risteli L, Risteli J, Smedsrod B. Clearance of NH2-terminal propeptides of types I and III procollagen is a physiological function of the scavenger receptor in liver endothelial cells. J Exp Med 1994;179:405-412. 5. Smedsrod B, Melkko J, Risteli L, Risteli J. Circulating C-terminal propeptide of type I procollagen is cleared mainly via the mannose receptor in liver endothelial cells. BiochemJ 1990;271:345-350. 6. Price PA, Otsuka AA, Poser JW, Kristaponis J, Raman N. Characterization of a gamma-carboxyglutamic acid-containing protein from bone. Proc NatlAcad Sci U S A 1976;73:1447-1451. 7. Delmas PD, Wilson DM, Mann KG, Riggs BL. Effect of renal function on plasma levels of bone Gla-protein. J Clin Endocrinol Metab 1983 ;57:1028-1030. 8. Skjodt H, Gallagher JA, Beresford JN, et al. Vitamin D metabolites regulate osteocalcin synthesis and proliferation of human bone cells in vitro. J Endocrino11985;105:391-396. 9. Ducy P, Desbois C, Boyce B, et al. Increased bone formation in osteocalcin-deficient mice. Nature 1996;382:448-452. 10. Cairns JR, Price PA. Direct demonstration that the vitamin K-dependent bone Gla protein is incompletely gamma-carboxylated in humans. J Bone Miner Res 1994;9:1989-1997. 11. Vergnaud P, Garnero P, Meunier PJ, et al. Undercarboxylated osteocalcin measured with a specific immunoassay predicts hip fracture in elderly women: the EPIDOS Study. J Clin Endocrinol Metab 1997;82:719-724. 12. Hodges SJ, Pilkington MJ, Stamp TC, et al. Depressed levels of circulating menaquinones in patients with osteoporotic fractures of the spine and femoral neck. Bone 1991;12:387-389. 13. Masters PW, Jones RG, Purves DA, Cooper EH, Cooney JM. Commercial assays for serum osteocalcin give clinically discordant results. Clin Chem 1994;40:358-363. 14. Garnero P, Grimaux M, Seguin P, Delmas PD. Characterization of immunoreactive forms of human osteocalcin generated in vivo and in vitro. J Bone Miner Res 1994;9:255-264. 15. Blumsohn A, Hannon RA, Eastell R. Apparent instability of osteocalcin in serum as measured with different commercially available immunoassays. Clin Chem 1995;41:318-319. 16. Ivaska KK, Hellman J, Likojarvi J, et al. Identification of novel proteolytic forms of osteocalcin in human urine. Biochem Biophys Res Commun 2003;306:973-980.
17. Fishman WH. Alkaline phosphatase isozymes: recent progress. Clin Biochem 1990;23:99-104. 18. Woitge HW, Seibel MJ, Ziegler R. Comparison of total and bonespecific alkaline phosphatase in patients with nonskeletal disorder or metabolic bone diseases. Clin Chem 1996;42:1796-1804. 19. Garnero P, Delmas PD. Assessment of the serum levels of bone alkaline phosphatase with a new immunoradiometric assay in patients with metabolic bone disease.J Clin EndocrinolMetab 1993;77: 1046-1053. 20. Takahashi M, Kushida K, Hoshino H, et al. Comparison of bone and total alkaline phosphatase activity on bone turnover during menopause and in patients with established osteoporosis. Clin Endocrinol (Oxf) 1997;47:177-183. 21. Price CP, Milligan TP, Darte C. Direct comparison of performance characteristics of two immunoassays for bone isoform of alkaline phosphatase in serum. Clin Chem 1997;43:2052-2057. 22. Apone S, Lee MY, Eyre DR. Osteoclasts generate cross-linked collagen N-telopeptides (NTx) but not free pyridinolines when cultured on human bone. Bone 1997;21:129-136. 23. ColweU A, Eastell R. The renal clearance of free and conjugated pyridinium crosslinks of collagen. J Bone Miner Res 1996;1:1976-1980. 24. Seibel MJ, Woitge H, Scheidt-Nave C, et al. Urinary hydroxypyridinium crosslinks of collagen in a population-based screening for overt vertebral osteoporosis: results of a pilot study. J Bone Miner Res 1994;9:1433-1440. 25. Naylor KE, Jackson B, Eastell R. The renal clearance of free and peptide-bound deoxypyridinoline: response to pamidronate treatment of Paget's disease. J Bone Miner Res 2003;18:658-661. 26. Hanson DA, Weis MA, Bollen AM, et al. A specific immunoassay for monitoring human bone resorption: quantitation of type I collagen cross-linked N-telopeptides in urine. J Bone Miner Res 1992;7: 1251-1258. 27. Clemens JD, Herrick MV, Singer FR, Erye DR. Evidence that serum NTx (collagen-type I N-telopeptides) can act as an immunochemical marker of bone resorption. Clin Chem 1997;43:2058-2063. 28. Cloos PA, Fledelius C, Christgau S, et al. Investigation of bone disease using isomerized and racemized fragments of type I collagen. Calcif Tissue Int 2003;72:8-17. 29. Alexandersen P, Peris P, Guanabens N, et al. Non-isomerized C-telopeptide fragments are highly sensitive markers for monitoring disease activity and treatment efficacy in Paget's disease ofbone.JBone Miner Res 2005;20:588-595. 30. Garnero P, Ferreras M, Karsdal MA, et al. The type I collagen fragments ICTP and CTX reveal distinct enzymatic pathways of bone collagen degradation.J Bone Miner Res 2003;18:859-867. 31. Garnero E Markers of bone turnover in prostate cancer. Cancer Treat Rev 2001;27:187-192. 32. Prockop DJ, Kivirikko KI, Tuderman L, Guzman NA. The biosynthesis of collagen and its disorders (first of two parts). N EnglJ Med 1979;301:13-23. 33. Deacon AC, Hulme E Hesp R, et al. Estimation of whole body bone resorption rate: a comparison of urinary total hydroxyproline excretion with two radioisotopic tracer methods in osteoporosis. Clin Chim Acta 1987;166:297-306. 34. Gasser A, Celada A, Courvoisier B, et al. The clinical measurement of urinary total hydroxyproline excretion. Clin Claim Acta 1979;95: 487-491. 35. Al-Dehaimi AW, Blumsohn A, Eastell R. Serum galactosyl hydroxylysine as a biochemical marker of bone resorption. Clin Chem 1999;45:676-681. 36. Garnero P, Delmas PD. An immunoassay for type I collagen alpha 1 helicoidal peptide 620-633, a new marker of bone resorption in osteoporosis. Bone 2003;32:20-26. 37. Lain WK, Eastlund DT, Li CY, Yam LT. Biochemical properties of tartrate-resistant acid phosphatase in serum of adults and children. Clin Chem 1978;24:1105-1108.
CHAPTER 27 Biochemical Markers of Bone Turnover 38. Halleen JM, Matalo SL, Suominen H, et al. Tartrate-resistant acid phosphatase 5b: a novel serum marker of bone resorption. J Bone Miner Res 20001;15:1337-1345. 39. Saftig P, Hunziker E, Everts V, et al. Functions of cathepsin K in bone resorption. Lessons from cathepsin K deficient mice. ddv Exp Med Bid 2000;477:293-303. 40. Holzer G, Noske H, Lang T, Holzer L, Willinger U. Soluble cathepsin K: a novel marker for the prediction of nontraumatic fractures? y Lab Clin Med 2005;146:13-17. 41. Panteghini M, Pagani F. Biological variation in urinary excretion of pyridinium crosslinks: recommendations for the optimum specimen. Ann Clin Biochem 1996;33:36-42. 42. Panteghini M, Pagani F. Biological variation in bone-derived biochemical markers in serum. &and y Clin Lab Invest 1995;55: 609-616. 43. Hannon R, Eastell R. Preanalytical variability of biochemical markers of bone turnover. OsteoporosInt 2000;11:$30-$44. 44. Ju HS, Leung S, Brown B, et al. Comparison of analytical performance and biological variability of three bone resorption assays. Clin Cbem 1997;43:1570-1576. 45. Jensen JE, Sorensen HA, Kollerup G, Jensen LB, Sorensen OH. Biological variation of biochemical bone markers. ScandJ Clin Lab Invest Supp11994;219:36-39. 46. Blumsohn A, Herrington K, Hannon RA, et al. The effect of calcium supplementation on the circadian rhythm of bone resorption. J Clin Endocrino! Metab 1994;79:730-735. 47. Qyist P, Christgau P, Pedersen BJ, Schlemmer A, Christiansen C. Circadian variation in the serum concentration of C-terminal telopeptides of type I collagen (serum CTX): effects of gender, age, menopausal status, posture, daylight, serum cortisol and fasting. Bone 2002;31:57-61. 48. Nielsen HK, Brixen K, Mosekilde L. Diurnal rhythm in serum activity of wheat-germ lectin-precipitable alkaline phosphatase: temporal relationships with the diurnal rhythm of serum osteocalcin. Scand J Clin Lab Invest 1990;50:851-856. 49. Pedersen BJ, Schlemmer A, Rosenquist C, Hassager C, Christiansen C. Circadian rhythm in type I collagen formation in postmenopausal women with and without osteopenia. Osteoporos Int 1995;5: 472-477. 50. Eastell R, Simmons PS, Colwell A, et al. Nyctohemeral changes in bone turnover assessed by serum bone Gla-protein concentration and urinary deoxypyridinoline excretion: effects of growth and ageing. Clin Sci (Lond) 1992;83:375-382. 51. Schlemmer A, Hassager C, Pedersen BJ, Christiansen C. Posture, age, menopause, and osteopenia do not influence the circadian variation in the urinary excretion of pyridinium crosslinks. J Bone Miner Res 1994;9:1883-1888. 52. Hassager C, Risteli J, Risteli L, Jensen SB, Christiansen C. Diurnal variation in serum markers of type I collagen synthesis and degradation in healthy premenopausal women. J Bone Miner Res 1992;7: 1307-1311. 53. Eastell R, Calvo MS, Burritt MF, et al. Abnormalities in circadian patterns of bone resorption and renal calcium conservation in type I osteoporosis. J Clin Endocrinol Metab 1992;74:487-494. 54. Greenspan SL, Dresner-Pollak R, Parker RA, London D, Ferguson L. Diurnal variation of bone mineral turnover in elderly men and women. Calcif Tissue Int 1997;60:419-423. 55. Clowes JA, Allen HC, Prentis DM, Eastell R, Blumsohn A. Octreotide abolishes the acute decrease in bone turnover in response to oral glucose. J Clin Endocrinol Metab 2003;88:4867-4873. 56. Bjarnason NH, Henriksen EE, Alexandersen P, et al. Mechanism of circadian variation in bone resorption. Bone 2002;30:307-313. 57. Nielsen HK, Brixen K, Bouillon R, Mosekilde L. Changes in biochemical markers of osteoblastic activity during the menstrual cycle. J C/in Endocrino! Metab 1990;70:1431-1437.
347 58. Gorai I, Chaki O, Nakayama M, Minaguchi H. Urinary biochemical markers for bone resorption during the menstrual cycle. Cak~Tissue Int 1995;57:100-104. 59. Schlemmer A, Hassager C, Risteli J, et al. Possible variation in bone resorption during the normal menstrual cycle. Acta Endocrinol (Copenh) 1993;129:388-392. 60. Vanderschueren D, Gevers G, Dequeker J, et al. Seasonal variation in bone metabolism in young healthy subjects. Caki~ Tissue Int 1991;49:84-89. 61. Thomsen K, Eriksen EF, Jorgensen JC, Charles P, Mosekilde L. Seasonal variation of serum bone GLA protein. ScandJ Clin Lab Invest 1989;49:605-611. 62. Woitge HW, Scheidt-Nave C, Kissling C, et al. Seasonal variation of biochemical indexes of bone turnover: results of a population-based study. J Clin Endocrinol Metab 1998;83:68-75. 63. Woitge HW, Knothe A, Witte K, et al. Circaannual rhythms and interactions of vitamin D metabolites, parathyroid hormone, and biochemical markers of skeletal homeostasis: a prospective study. J Bone Miner Res 2000;15:2443-2450. 64. Blumsohn A, Naylor KE, Timm W, et al. Absence of marked seasonal change in bone turnover: a longitudinal and multicenter crosssectional study. J Bone Miner Res 2003;18:1274-1281. 65. Carnevale V, Modoni S, Pileri M, et al. Longitudinal evaluation of vitamin D status in healthy subjects from southern Italy: seasonal and gender differences. OsteoporosInt 2001;12:1026-1030. 66. Mallmin H, Ljunghall S, Larsson K. Biochemical markers of bone metabolism in patients with fracture of the distal forearm. C/in Orthop Relat Res 1993;295:259-263. 67. Ingle BM, Hay SM, Bottjer HM, Eastell R. Changes in bone mass and bone turnover following distal forearm fracture. Osteoporos Int 1999; 10:399-407. 68. Perry HM III, Jensen J, Kaiser FE, et al. The effects of thiazide diuretics on calcium metabolism in the aged. J Am Geriatr Soc 1993;41:818-822. 69. Ohishi T, Kushida K, Takahashi M, et al. Urinary bone resorption markers in patients with metabolic bone disorders. Bone 1994;15: 15-20. 70. Valimaki MJ, Tiihonen M, Laitinen K, et al. Bone mineral density measured by dual-energy x-ray absorptiometry and novel markers of bone formation and resorption in patients on antiepileptic drugs. J Bone Miner Res 1994;9:631-637. 71. Marshall LA, Cain DF, Dmowski WP, Chesnut CHIII. Urinary Ntelopeptides to monitor bone resorption while on GnRtt agonist therapy. Obstet Gyneco11996;87:350-354. 72. Ravn P, Fledelius C, Rosenquist C, Overgaard K, Christiansen C. High bone turnover is associated with low bone mass in both pre- and postmenopausal women. Bone 1996;19:291-298. 73. Ebeling PR, Atley LM, Guthrie JR, et al. Bone turnover markers and bone density across the menopausal transition. J Clin Endocrinol Metab 1996;81:3366-3371. 74. Perrien DS, Achenbach SJ, Bledsoe SE, et al. Bone turnover across the menopause transition: correlations with inhibins and FSH. J Clin Endocrinol Metab 2006;91:1848-1854. 75. Griesmacher A, Peichl P, Pointinger P, et al. Biochemical markers in menopausal women. ScandJ Clin Lab Invest Supp11997;227:64-72. 76. Hassager C, Colwell A, Assiri AM, et al. Effect of menopause and hormone replacement therapy on urinary excretion of pyridinium cross-links: a longitudinal and cross-sectional study. Clin Endocrinol (Oxf) 1992;37:45-50. 77. Uebelhart D, Gineyts E, Chapuy MC, Delmas PD. Urinary excretion of pyridinium crosslinks: a new marker of bone resorption in metabolic bone disease. Bone Miner 1990;8:87-96. 78. Gorai I, Taguchi Y, Chaki O, Nakayama M, Minaguchi H. Specific changes of urinary excretion of cross-linked N-telopeptides of type I collagen in pre- and postmenopausal women: correlation with other markers of bone turnover. CalcifTissue Int 1997;60:317-322.
348 79. Garnero P, Gineyts E, Arbault P, Christiansen C, Delmas PD. Different effects of bisphosphonate and estrogen therapy on free and peptide-bound bone cross-links excretion. J Bone Miner Res 1995; 10:641-649. 80. Hassager C, Risteli J, Risteli L, Christiansen C. Effect of the menopause and hormone replacement therapy on the carboxy-terminal pyridinoline cross-linked telopeptide of type I collagen. OsteoporosInt 1994;4:349-352. 81. Kerschan-Schindl K, Hawa G, Kudlacek S, Woloszczuk W, Pietschmann P. Serum levels ofcathepsin K decrease with age in both women and men. Exp Geronto12005;40:532-535. 82. Melkko J, Kauppila S, Niemi S, et al. Immunoassay for intact aminoterminal propeptide of human type I procollagen. Clin Chem 1996; 42:947-954. 83. Kelly PJ, Pocock NA, Sambrook PN, Eisman JA. Age and menopause-related changes in indices of bone turnover. J Clin Endocrinol Metab 1989;69:1160-1165. 84. Eastell R, Delmas PD, Hodgson SF, et al. Bone formation rate in older normal women: concurrent assessment with bone histomorphometry, calcium kinetics, and biochemical markers. J Clin Endocrinol Metab 1988;67:741-748. 85. Garnero P, Sornay-Rendu E, Chapuy MC, Delmas PD. Increased bone turnover in late postmenopausal women is a major determinant of osteoporosis. J Bone Miner Res 1996;11:337-349. 86. Valimaki MJ, Tahtela R, Jones JD, Peterson JM, Riggs BL. Bone resorption in healthy and osteoporotic postmenopausal women: comparison markers for serum carboxy-terminal telopeptide of type I collagen and urinary pyridinium cross-finks. Eur J Endocrinol 1994;131:258-262. 87. Kushida K, Takahashi M, Kawana K, Inoue T. Comparison of markers for bone formation and resorption in premenopausal and postmenopausal subjects, and osteoporosis patients. J Clin Endocrinol Metab 1995;80:2447-2450. 88. Eastell R, Robins SP, Colwell T, et al. Evaluation of bone turnover in type I osteoporosis using biochemical markers specific for both bone formation and bone resorption. OsteoporosInt 1993;3:255-260. 89. Schneider DL, Barrett-Connor EL. Urinary N-telopeptide levels discriminate normal, osteopenic, and osteoporotic bone mineral density. Arch Intern Med 1997;157:1241-1245. 90. Meier C, Meinhardt U, Greenfield JR, et al. Serum cathepsin K concentrations reflect osteoclastic activity in women with postmenopausal osteoporosis and patients with Paget's disease. Clin Lab 2006;52:1-10. 91. Rosen CJ, Hochberg MC, Bonnick SL, et al. Treatment with onceweekly alendronate 70 mg compared with once-weekly risedronate 35 mg in women with postmenopausal osteoporosis: a randomized double-blind study. J Bone Miner Res 2005;20:141-151. 92. Rizzoli R, Greenspan SL, Bone GIII, et al. Two-year results of onceweekly administration of alendronate 70 mg for the treatment of postmenopausal osteoporosis. J Bone Miner Res 2002;17:1988-1996. 93. Hannon RA, Clowes JA, Eagleton AC, et al. Clinical performance of immunoreactive tartrate-resistant acid phosphatase isoform 5b as a marker of bone resorption. Bone 2004;34:187-194. 94. Hansdottir H, Franzson L, Prestwood K, Sigurdsson G. The effect of raloxifene on markers of bone turnover in older women living in longterm care facilities. J A m Geriatr Soc 2004;52:779-783. 95. Reginster JY, Sarkar S, Zegels B, et al. Reduction in PINP, a marker of bone metabolism, with raloxifene treatment and its relationship with vertebral fracture risk. Bone 2004;34:344-351. 96. Riggs B L, Parfitt AM. Drugs used to treat osteoporosis: the critical need for a uniform nomenclature based on their action on bone remodeling. J Bone Miner Res 2005;20:177-184.
EASTELL AND HANNON 97. Chen P, Satterwhite JH, Licata AA, et al. Early changes in biochemical markers of bone formation predict BMD response to teriparatide in postmenopausal women with osteoporosis. J Bone Miner Res 2005;20:962-970. 98. McClung MR, San MJ, Miller PD, et al. Opposite bone remodeling effects of teriparatide and alendronate in increasing bone mass. Arch Intern Med 2005;165:1762-1768. 99. Dobnig H, Sipos A, Jiang Y, et al. Early changes in biochemical markers of bone formation correlate with improvements in bone structure during teriparatide therapy. J Clin Endocrinol Metab 2005;90:3970-3977. 100. Meunier PJ, Roux C, Seeman E, et al. The effects of strontium ranelate on the risk of vertebral fracture in women with postmenopausal osteoporosis. N E n g l J M e d 2004;350:459-468. 101. Hui SL, Slemenda CW, Johnston CC Jr. The contribution of bone loss to postmenopausal osteoporosis. OsteoporosInt 1990;1:30-34. 102. Rogers A, Hannon RA, Eastell R. Biochemical markers as predictors of rates of bone loss after menopause. J Bone Miner Res 2000;15: 1398-1404. 103. Uebelhart D, Schlemmer A, Johansen JS, et al. Effect of menopause and hormone replacement therapy on the urinary excretion of pyridinium cross-links. J Clin Endocrinol Metab 1991;72:367-373. 104. Mole PA, Walkinshaw MH, Robins SP, Paterson CR. Can urinary pyridinium crosslinks and urinary oestrogens predict bone mass and rate of bone loss after the menopause? Eur J Clin Invest 1992;22: 767-771. 105. Cosman F, Nieves J, Wilkinson C, et al. Bone density change and biochemical indices of skeletal turnover. Calcif Tissue Int 1996;58: 236-243. 106. Dresner-Pollak R, Parker RA, Poku M, et al. Biochemical markers of bone turnover reflect femoral bone loss in elderly women. CalcifTissue Int 1996;59:328-333. 107. Bauer DC, Sklarin PM, Stone KL, et al. Biochemical markers of bone turnover and prediction of hip bone loss in older women: the study of osteoporotic fractures. J Bone Miner Res 1999; 14:1404-1410. 108. Garnero P, Sornay-Rendu E, Duboeuf F, Delmas PD. Markers of bone turnover predict postmenopausal forearm bone loss over 4 years: the OFELY study. J Bone Miner Res 1999;14:1614-1621. 109. Hannon RA, Eastell R. Biochemical markers of bone turnover and fracture prediction. J Br Menopause Soc 2003;9:10-15. 110. Garnero P, Hausherr E, Chapuy MC, et al. Markers of bone resorption predict hip fracture in elderly women: the EPIDOS Prospective Study. JBone Miner Res 1996;11:1531-1538. 111. Melton LJ III, Khosla S, Atkinson EJ, O'Fallon WM, Riggs B L. Relationship of bone turnover to bone density and fractures. J Bone Miner Res 1997;12:1083-1091. 112. Sarkar S, Reginster JY, Crans GG, et al. Relationship between changes in biochemical markers of bone turnover and BMD to predict vertebral fracture risk. J Bone Miner Res 2004;19:394-401. 113. Greenspan SL, Rosen HN, Parker RA. Early changes in serum Ntelopeptide and C-telopeptide cross-linked collagen type 1 predict long-term response to alendronate therapy in elderly women. J Clin Endocrinol Metab 2000;85:3537-3540. 114. Greenspan SL, Resnick NM, Parker RA. Early changes in biochemical markers of bone turnover are associated with long-term changes in bone mineral density in elderly women on alendronate, hormone replacement therapy, or combination therapy: a three-year, doubleblind, placebo-controlled, randomized clinical trial. J Clin Endocrinol Metab 2005;90:2762-2767. 115. Ravn P, Thompson DE, Ross PD, Christiansen C. Biochemical markers for prediction of 4-year response in bone mass during bisphosphonate treatment for prevention of postmenopausal osteoporosis. Bone 2003;33:150-158.
CHAPTER 27 Biochemical Markers of Bone Turnover 116. Bauer DC, Garnero P, Bilezikian JP, et al. Short-term changes in bone turnover markers and bone mineral density response to parathyroid hormone in postmenopausal women with osteoporosis. J Clin Endocrinol Metab 2006;91:1370-1375. 117. Eastell R, Barton I, Hannon RA, et al. Relationship of early changes in bone resorption to the reduction in fracture risk with risedronate. JBone Miner Res 2003;18:1051-1056.
349 118. Sarkar S, Reginster JY, Crans GG, et al. Relationship between changes in biochemical markers of bone turnover and BMD to predict vertebral fracture risk. J Bone Miner Res 2004;19:394-401. 119. Clowes JA, Peel NF, Eastell R. The impact of monitoring on adherence and persistence with antiresorptive treatment for postmenopausal osteoporosis: a randomized controlled trial. J Clin Endocrinol Metab 2004;89:1117-1123.
This Page Intentionally Left Blank
; H A P T E R 2{
Interventions for Osteoporosis FERGUS S J . KEATING
GiggsHill Surgery,Thames Ditton, Surrey,UK KT7 0EB
JOHN C . STEVENSON
NationalHeart & Lung Institute, Imperial College London, Royal Brompton Hospital, London, UK SW3 6NP
I. THE MAGNITUDE OF THE DISEASE
tripling of the incidence of hip fracture may occur by 2040. A similar aging of the population is predicted in underdeveloped countries as infant and child mortality rates improve, and the consequent rise in osteoporosis incidence that will eventually ensue may impose an overwhelming financial burden on the health care systems of these countries.
Although previously underestimated as a major public health concern, osteoporosis is now recognized as one of the most important diseases facing women and men (to some lesser extent) of advancing age. Some 50% of women in the United Kingdom over 50 years of age (and 20% of men) will sustain an osteoporotic fracture (1). Postmenopausal osteoporosis accounts for the majority of the disease; however, although it is beyond the scope of this chapter, it should be remembered that other secondary causes account for a significant number of new cases of fracture each year. The extent of osteoporosis is highlighted by European statistics showing a yearly incidence of 60,000 hip fractures, 50,000 wrist fractures, and 120,000 vertebral fractures (1-3). In England and Wales alone, more than 1.14 million women have been diagnosed with postmenopausal osteoporosis following dual-energy x-ray absorptiometry (DEXA) scanning of the hip (4). In the United Kingdom, a woman has an approximate 17% lifetime risk of hip fracture (5), and the mortality 3 months post hip fracture is as high as 18% (6). Approximately 14,000 people will die each year in the United Kingdom as a result of hip fracture (7). The cost of treating all osteoporotic postmenopausal fractures in the United Kingdom is predicted to rise to s billion by 2020 (8). Moreover, the situation is only likely to worsen as the population of the developed world continues to age. In the United States, attributable costs for osteoporosis may be as high as $10 billion. Indeed, the growth in the elderly population of the United States is already exceeding the figures predicted, and if this trend were to continue, a T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
II. DEFINITION OF OSTEOPOROSIS Osteoporosis has been defined as "a disease characterized by low bone mass and micro-architectural deterioration of bone tissue, leading to enhanced bone fragility and a consequent increase in fracture risk" (9). Osteoporosis (porous bone) implies a histologic diagnosis of reduced bone tissue and deranged architecture but normal mineralization; however, the clinical diagnosis of the disease has been made traditionally on the occurrence of fragility fractures, principally at the femoral neck, spine, and distal forearm. The main obstacle to such a fracture-based diagnosis was that a patient previously described as normal became immediately osteoporotic on the occurrence of a fragility fracture, and this was inconsistent with our knowledge of the natural history of the disease. Such an approach also led inevitably to a delay in diagnosis of disease until an "end stage" was reached, and the opportunity was wasted to prevent the occurrence of the fracture. To overcome this dilemma in diagnosis, attempts have been made to define osteoporosis as a disease diagnosed at a certain low level of bone density, in the same way that hypertension as a disease is defined beyond a set level of blood pressure. It follows that for such a diagnosis to be viable, 351
Copyright 9 2007 by Elsevier, Inc. All rights of reproduction in any form reserved.
352
KEATING AND STEVENSON
TABLE 28.1 World Health Organization Classification of Osteoporosis Normal: Bone mineral density (BMD) not more than 1 standard deviation (SD) below the young adult mean Osteopenia: BMD between 1 and 2.5 SDs below the young adult mean Osteoporosis: BMD more than 2.5 SDs below the young adult mean Established (severe) osteoporosis: BMD more than 2.5 SDs below the young adult mean in the presence of one or more fragility fractures
accurate, noninvasive, and reproducible methods of measuring bone density must be available. The advent of DEXA fulfilled these criteria at minimal radiation exposure and is currently the most prevalent method of bone densitometry in the United Kingdom and United States. Although much debate continues as to the correct level of bone mineral density (BMD) below which to assign a diagnosis of osteoporosis, the World Health Organization (WHO) proposed a stratified classification of osteoporosis in 1994 to encompass both systems of diagnosis, and this is summarized in Table 28.1 (10). Thus an individual would be termed osteoporotic if his or her BMD were less than 2.5 standard deviations (SDs) below the young adult range, regardless of whether a fracture had occurred. A further classification including values of BMD between 1 and 2.5 SD below the young adult mean was also included, and such individuals were termed osteopenic (low bone mass) and said to be at higher risk of future osteoporosis than the normal population. The W H O is currently revising these definitions of osteopenia and osteoporosis to include factors other than only BMD, and these new recommendations are awaited.
A. L o w B o n e D e n s i t y a n d Fragility Fractures Prospective studies have shown that with declining BMD, the risk of fragility fractures increases progressively and continuously (11,12), and fracture risk increases by up to threefold for every SD decrease in BMD (13). At 2.5 SDs below the young adult mean, 30% ofpostmenopausal women will be identified as having osteoporosis using BMD measurements from the spine, hip, or forearm, and this is roughly the equivalent lifetime risk for fractures at these sites (13). Decreased bone mass is associated with increased fracture risk across all skeletal sites but is most significant in terms of mortality and morbidity at the hip. There is an almost exponential rise in hip fracture incidence after age 50 until at least the ninth decade, and incidence rates reach approximately 3% per year among Caucasian women over age 85 in northern Europe and the United States (14). The associated mortality with hip fracture is around 20% at 1 year post-
fracture, and this often follows an extended period of hospitalization. Morbidity is also high; 10% of women sustaining hip fracture become dependent in actMties of daily 1Mng, and almost 50% are rendered dependent upon long-term institutionalized nursing care (15). Although hip fracture is most common in the elderly, it is not an insignificant problem for younger postmenopausal women. In the age group 50 to 80 years, around 36% of classical osteoporotic fractures occur in those younger than age 65 (16,17). In the United Kingdom, this amounts to over 100,000 fractures per year in women aged 50 to 65 years, more than 9000 of which are hip fractures (16,17). Although mortality rates are lower for vertebral and wrist fractures, morbidity is significant. Even asymptomatic vertebral fractures may lead to a reduction in actMties of daily 1Mng through progressive skeletal deformities such as kyphoscoliosis, and persistent pain may impinge on the patient's life, both physically and psychologically. Painful, symptomatic vertebral fractures often necessitate periods of bed rest for as long as 6 months. Incidence rates for vertebral fractures are more difficult to report due to the large number of subclinical fractures and variation in the methodology employed to assess vertebral deformity, but the same pattern of age-related, exponentially increased incidence rate is also seen with vertebral fracture as with hip fracture. One study of a population in the United States showed an annual incidence of 0.5% at age 50, rising to 4% at age 85 (16). However, more data are available concerning the prevalence of vertebral fracture, and one study reported rising age-related prevalence rates reaching a maximum of 20% at age 70 in a population of 2063 women in Puerto Rico and Michigan (United States) (18). A prevalence of wedge fracture of around 60% in women older than 70 years and a prevalence of crush fracture of 10% has previously been shown (19). The mortality associated with vertebral fracture is often overlooked, but the 5-year mortality has been estimated at 20% to 30% (17). Distal forearm fractures also show a steep rise in incidence in women with increasing age, but the rise tends to occur at an earlier age, and by around 60 years of age, the incidence rate begins to level off to around 0.5% of women per year (20). It is thought that the classic fall onto the outstretched hand producing the distal radius fracture becomes less common with advancing age as decreased proprioceptive and reflex responses diminish the speed of this defensive mechanism. Consequently, more falls occur directly onto the lower limb, which may in part contribute to the increase in hip fracture observed with increasing age. Distal radius fracture accounts for 50,000 hospital admissions per year in the United States and more than 6 million restricted activity days in people over 45 years of age (21). Persistent morbidity is also frequent, with chronic pain, loss of function, neuropathies, and posttraumatic arthritis, and many patients report poor functional outcome at 6 months post-fracture.
CHAPTER 28 Interventions for Osteoporosis
III. GENETIC A N D ENVIRONMENTAL INFLUENCES ON BONE MASS Peak bone mass is reached soon after the end of linear growth and is largely genetically determined, although environmental factors can influence the eventual final peak bone mass achieved (Table 28.2). Twin studies and family studies of bone mineral density have shown that genetic factors are probably the single most important influence on bone mass and osteoporotic fracture risk. Studies have revealed that as much as 70% to 85% of the interindividual variation in bone mass is genetically determined (22). Several genes are thought to contribute to the determination of bone mass, rather than a single gene or pair of genes contributing major effects. However, at present the exact number and precise function of these genes remains unclear. Candidate gene studies have identified potential contributors to osteoporotic risk, and these are shown in Table 28.3. Much work is still outstanding as to the relative contributions, if at all, of these and other genes, and further elucidation of the human genome will inevitably produce further candidate genes that may be shown to contribute to bone density. Other evidence from identical twin studies has shown that variation in bone mass increases with age, suggesting that bone loss may be more independent of genetic factors (23)
TABLE 28.2
Factors Affecting Bone Mass
Genetic Racial Geographical Environmental 1. In attaining peak bone mass Hormonal Physical activity Diet 2. After peak mass achieved Smoking Alcohol Physical activity Hormonal Diet
TABLE 28.3 Gene studied
Vitamin D receptor Estrogen receptor Transforming growth f actor-~ Interleukin-6 Collagen type I genes Collagenase From ref. 23.
353 Studies of incidence rates of osteoporotic fractures show that racial and geographic factors are important determinants of fracture risk, with European and American women having generally higher fracture rates than women from developing nations (24). Within developed nations, racial variations in fracture rates are also observed; for example, African-American women in the United States display a hip fracture rate approximately one-third lower than that of American Caucasian women (25). Hip fracture rates among black African women are lower still than African-American women (24), despite the evidence from the limited number of studies available demonstrating that black African women have similar or slightly less bone density than American Caucasian women and considerably less bone mass than African-American women (26,27). This evidence would suggest that within populations, bone density variation is the most influential variable of fracture rates but that other variables play an important role in the observed differences of fracture rates among countries. Peak bone mass declines from around middle age, although earlier in the proximal femur (28), and both men and women lose about 0.3% to 0.5% of their bone mass every year. Again, adverse environmental factors can further decrease these losses. Women experience an additional accelerated period of bone loss following their menopause, which may be as much as tenfold higher than the premenopausal rate of bone loss, and for the first 5 to 10 years of postmenopausal life, this loss can be even higher still. Consequently, bone mass may fall to about half of its peak value by around 80 years of age. The accelerated rate of bone loss in postmenopausal women results from increased bone turnover due to the effect of estrogen deficiency on the bone remodeling process described in the next section.
IV. BONE REMODELING A N D OSTEOPOROSIS As with almost all adult bone diseases, osteoporosis develops from a disorder of the bone remodeling process. Adult skeletal bone consists of cortical (compact) and tra-
Some Candidate Genes for Studying the Genetic Contribution to Osteoporosis Rationale for study Vitamin D regulates bone cell differentiation, bone turnover, and calcium homeostasis by interaction with the vitamin D receptor Estrogen regulates bone turnover and skeletal growth by interaction with the estrogen receptor Present in bone matrix; is thought to couple bone resorption and bone formation Regulates osteoclast differentiation; may mediate some actions of sex hormones in bone Type I collagen is the most abundant protein in bone A degradative enzyme involved in resorption of bone matrix
354
KEATING AND STEVENSON
becular (cancellous) bone and is continuously repaired and reformed by a process known as remodeling. It appears that bone remodeling is an essential process in land-based vertebrates to maintain the integrity of the skeleton. Cortical bone is dense and forms some 85% of the total body bone, predominantly in the long shafts of the appendicular skeleton. It is laid down concentrically around central canals, termed Haversian systems, that contain blood, lymphatics, nerves, and connective tissue. Trabecular bone, although forming only 15% of the adult bone, is relatively prominent at the ends of long bones and in the inner parts of flat bones. It consists of interconnecting trabeculae interspersed by the bone marrow. In the lumbar spine, as much as 65% of the bone is of the trabecular type. Bone remodeling occurs at discrete sites, known as bone remodeling units, on the bone surface of both trabecular and cortical bone. The process basically involves the removal of mineralized bone and its replacement with newly formed and mineralized osteoid. Central to the process are the actions of osteoclasts and osteoblasts. Osteoclasts, derived from the hematopoietic precursors of the monocyte-macrophage lineage, perform the resorption of mineralized bone by acidification and proteolytic digestion, and osteoblasts, descended from bone marrow pluripotent stromal stem cells, are responsible for the formation and subsequent mineralization of bone matrix. Remodeling always follows in a sequential, closely coupled pattern of resorption followed by formation, and the process takes about 3 to 4 months to complete, with resorption occupying the first 10 days or so. Remodeling occurs in discrete packages on both cortical (in the Haversian systems) and trabecular bone (on the trabecular surfaces), and at any one time, multiple bone remodeling units are at different stages of the bone remodeling process. In the development of osteoporosis, an imbalance exists between resorption and formation, with net bone loss. The most important influences on this are the effects of aging and the loss of gonadal function (see later discussion).
A. Bone Resorption Bone resorption first involves activation of the quiescent bone surface, with retraction of the lining cells (specialized osteoblasts) and proteolytic digestion of the endosteal membrane, resulting in exposure of the mineralized bone ready for osteoclastic resorption. The underlying mechanisms governing activation are not well understood; however, it seems that matrix metalloproteinases (released from osteoblasts) are involved in the removal of the endosteal membrane. It is also possible that activation is prompted by mechanical stresses acting on specific sites, transmitted via the osteocytic-canalicular network of the bone (29). Activation is followed by recruitment of the osteoclast precursors to the exposed bone and differentiation of these precursors into the
mature osteoclast. A series of steps achieves this, commencing with fusion of the precursors into a multinucleated cell. It has been demonstrated that the higher the number of nuclei within the cell, the more effective the osteoclastic bone resorption (30). Recent work has shown that osteoclastic fusion may occur in a similar fashion to trophoblastic cell fusion in the placenta using (E)-cadherin, an hemophilic calcium-dependent cell attachment molecule (31,32). Following fusion of the precursors, attachment to the bone surfaces occurs via integrins on the osteoclast cell membrane that bind to bone matrix molecules through specific RGD (Arg-Gly-Asp) sequences (33). The osteoclast then undergoes polarization, with realignment of the intracellular organelles and the development of a ruffled border, a specialized area of the cell membrane in proximity to the bone surface that is associated with the expression of the vacuolar ATPase proton pump, src proto-oncogene, and proteolytic enzymes required in the resorptive process. After resorption, osteoclastic action is terminated by cell apoptosis, characterized by the morphologic changes of nuclear chromatin condensation and separation of the osteoclast from the bone surface. This apoptosis may be the common end pathway in the action of both estrogens and bisphosphonates on bone, and Parfitt has linked the excessive osteoclastic activity associated with estrogen deficiency to the absence of an apoptosis stimulus (34).
B. Bone Formation Bone formation commences with chemotaxis of osteoblasts and their precursors to the site of the resorptive defect, and it is thought that this occurs as a result of the action of local factors released by the resorbing bone. In vitro studies by Mundy and co-workers demonstrated that cells with osteoblastic properties were chemotactically attracted by local factors produced by the resorbing bone (35,36). Transforming growth factor [3 (TGF-[3) has been demonstrated to be chemotactic for bone cells (37) and is released from resorbing bone. Platelet derived growth factors (PDGFs) are also possible mediators in this process and have been demonstrated to be chemotactic for certain mesenchymal cells, monocytes, neutrophils, and smooth muscle cells (30). It is also possible that structural proteins such as collagen and bone GLA protein act as chemotactic stimulants to osteoblastic congregation at resorptive sites (35,36). Following chemotaxis, the osteoblasts undergo proliferation, and it is likely that growth factors such as the TGF-[3 superfamily, PDGF, insulin growth factors 1 and 2 (IGF-1 and -2), and human basic fibroblast growth factors (hbFGFs) are mediators in this process. The precursors then undergo differentiation into mature osteoblasts. Bonederived growth factors such as IGF-1 and bone morphogenetic protein (BMP)-2 have been suggested as mediators in
CHAPTER28 Interventions for Osteoporosis
this stage, and interestingly, TGF-[3 may actually inhibit osteoblast differentiation, which suggests that it acts as a trigger early in the process of osteoblast recruitment and proliferation but must be removed or inactivated prior to differentiation of the osteoblasts. Differentiated osteoblasts subsequently lay down the new osteoid in the resorption cavity, which is then mineralized. Cessation of osteoblast activity is the final stage of bone remodeling, and again local factors, possibly TGF-[3, are central to this process.
C. Control of Remodeling A complex interrelationship between systemic hormones, mechanical stresses, and locally derived cytokines, prostaglandins, and growth factors is responsible for the control of bone remodeling, and the major compounds involved in this control are shown in Table 28.4. Adults show resorption and re-formation of approximately 25% of their trabecular bone, but only around 3% of their cortical bone, each year. Trabecular bone has a higher surface-to-volume ratio, and as much as 85% of the surface of trabecular bone is in contact with the bone marrow. Thus, it follows that control of remodeling is primarily under local control (38). Local control of remodeling is primarily affected by a large number of cytokines and growth factors, some of which can exert influence on both osteoblasts and osteoclasts, and many of which act interdependently. For example, the precursor of mature osteoblasts, the stromal-osteoblastic precursor cell, is stimulated by both local factors, such as interleukin (IL)-I and tumor necrosis factor (TNF), and by systemic hormones, such as parathyroid hormone (PTH) and 1,25 dihydroxyvitamin D3 to produce in turn cytokines and growth factors, in particular IL-6 and IL-11, and these cytokines regulate the differentiation pathway of preosteoclasts. Indeed, the list of cytokines and colony stimulating factors involved in the development of osteoclasts, and therefore stimulation of their action, is large and includes IL-1, -3, -6, and -11; TNF; granulocyte-macrophage colony stimulating factor (GM-CSF); macrophage colony stimulating factor (MCSF); leukemia inhibitory factor; and stem cell factor (38). It seems that IL-6 may have a pivotal role in the osteoclastogenic process, in that it has been implicated in the stimulation of differentiation of the granulocyte-macrophage colony forming units (GM-CFU) into osteoclast precursors (39), also stimulates osteoclast formation, and, in combination with IL-1, stimulates bone resorption (38). It is also possible that IL-6 can stimulate mature osteoclasts, as receptors for IL-6 have been demonstrated on human osteoclastoma cells and were found to increase the resorptive activity of these cells (40). As observed previously, it has been shown that normal remodeling involves close coupling of the resorptive and formative processes. Here again, it is thought that local fac-
355 TABLE 28.4 Systematic Hormones and Local Factors Involved in the Control of Bone Remodeling Systemic Hormones Parathyroid hormone (PTH) Tri-iodothyronine Growth hormone (GH) Glucocorticoids 1,25 Dihydroxy-vitamin D3 Sex steroids Cytokines and Growth Factors Stimulate bone formation: Transforming growth factor-J3 (TGF-~) Insulin-like growth factors (IGFs) Platelet-derived growth factors (PDGFs) Fibroblast growth factors (FGFs) Bone morphogenetic proteins (BMPs) Stimulates bone resorption: Interleukin (IL)-I,-6,-8,-11 PDGFs Macrophage colony stimulating factor (M-CSF) Granulocyte/macrophage colony stimulating factor (GM-CSF) Epidermal growth factors (EGFs) Tumor necrosis factors (TNF) FGFs Leukemia inhibitory factor Inhibit bone resorption: Interferon-7 (IFN-(7) IL-4
tors are primarily involved in the regulation of this coupling. In particular, the TGF-[3 superfamily may be especially important. Bone resorption leads to the release of TGF-[3, as previously discussed, and this stimulates osteoblast precursor proliferation. The exposure, however, is transient, and subsequently the proliferating cells undergo differentiation into mature osteoblasts and express BMPs, which are autostimulatory upon osteoblasts, propagating the formation of mineralized bone.
V. THE TREATMENT OF POSTMENOPAUSAL OSTEOPOROSIS A. Estrogens In his classic observations of the link between osteoporosis and estrogen deficiency, Albright noted that 40 of 42 women with osteoporotic fractures were postmenopausal (41). It was first proposed that estrogen deficiency led to a decrease in osteoblast activity and a subsequent decrease in bone mass (42), but this explanation proved to be inadequate when a consistent decline in osteoblast activity was not shown in postmenopausal women (43). Various studies have showed that the cessation of ovarian function is associated
356 with an increase in bone turnover and a consequent decline in bone mass (44). The most usual cause of estrogen decline is the onset of menopause, either natural or surgical, and studies across all cultures have demonstrated bone loss. In addition, estrogen decline can result from ovarian failure secondary to anorexia, hyperprolactinemia, pituitary failure, and excessive exercise. Declining estrogen levels lead to an increase in the frequency of activation of remodeling cycles, with a subsequent increased rate of resorption and formation, as shown by the increased levels of markers for bone turnover in the serum and urine (serum alkaline phosphatase and osteocalcin; excretory products of type I collagen, such as hydroxyproline or pyridinoline cross-links of collagen) (45). An increase in the number of osteoclasts in trabecular bone is observed following the loss of ovarian activity. It has also been suggested that osteoclast aggression is increased in the estrogen deficient state, possibly due to the previously mentioned delay in osteoclast apoptosis at the termination of resorption, resulting in larger resorption cavities that can only be partially filled by osteoblast activity. Furthermore, with increased frequency of remodeling, the likelihood of simultaneous remodeling on opposite sides of trabeculae is increased, with a greater chance of complete transection of trabeculae and a consequent loss of template upon which further bone formation can occur, leading to the observed loss of bone mass and architecture characteristic of the disease. Fundamental differences exist between the patterns of bone loss seen with aging and loss of ovarian function. Postmenopausal bone loss is characteristically associated with excessive osteoclastic activity, as discussed previously, whereas bone loss associated with aging has been seen to be more related to a progressive decline in the number of osteoblasts available compared with the demand (46). The type of this bone loss also differs, with trabecular bone more affected in postmenopausal bone loss and cortical bone primarily affected by age-associated bone loss. Although the effects of estrogen deficiency on bone are well established, it has been difficult to describe the exact role of estrogens in modulating bone remodeling. It is known that pharmacologic doses of estrogens suppress the increased remodeling rates of postmenopausal women (47-49), and estrogen loss at menopause is associated with an apparent partial release from inhibition of skeletal resorption (50). Estrogen seems to act both directly and indirectly on bone. Eriksen et al. demonstrated evidence of estrogen acting through the classical estrogen receptor-mediated mechanism on cultured human osteoblast-like cells (51). Komm et al. showed that estrogen could act, via a direct receptormediated action, on osteoblast-like cells to enhance levels of type I procollagen and TGF-[3 mRNA levels (52) and argued that estrogen thus governed the transcriptional activity of the TGF-[3 gene in osteoblasts, which in turn could positively control the transcriptional activity of the type I
KEATING AND STEVENSON
collagen gene, thus providing a possible explanation for the observed effects of estrogen on bone remodeling. The indirect actions of estrogen on bone are mediated via cytokines and local growth factors. It has been proposed that TNF-cx and IL-1 both affect the initial step in estrogendeficiency bone loss. Both are released from the peripheral blood monocytes in excessive amounts in ovariectomized women and act as stimulators of bone resorption. This effect is antagonized by estrogen replacement. Further work on the TNF-ci cytokine has shown that permanent suppression of its activity can lead to protection against the increased bone turnover and bone loss associated with oophorectomy in mice (53). Further work on cytokines has shown that IL-6 production from bone marrow stromal and osteoblastic cell lines is inhibited by estrogen (54) and that this is mediated by an estrogen receptor-mediated inhibitory effect on the transcription gene for IL-6. Studies on ovariectomized mice showed that administration of 17~3-estradiol prevented the IL-6-mediated increases in osteoclast populations in trabecular bone. It can therefore be proposed that estrogen deficiency leads to in increase in expression of IL-6 and a consequent increased rate of osteoclastogenesis and therefore bone resorption. Furthermore, IL-6-deficient ovariectomized mice do not seem to undergo this bone loss associated with increased osteoclast numbers (38).
B. Hormone Replacement Therapy Epidemiologic and clinical data have consistently shown that hormone replacement therapy (HRT) results in a cessation of bone loss and an improvement in bone density with a subsequent reduction in risk of osteoporotic fracture (55-68). Earlier studies tended to use bone density measurements from the distal radius and the metacarpals, and therefore potential difficulty existed in attempting to extrapolate the results of these studies to the hip and spine, areas where trabecular bone is more prevalent. However, cross-sectional follow-up data from these earlier studies did show that similar effects were noted in the spine and hip (69). Lower doses than previously thought necessary are now proving effective (70-72), and a very low-dose transdermal estradiol product has recently been licensed for osteoporosis prevention by the U.S. Food and Drug Administration (FDA). The Women's Health Initiative (WHI) trials have confirmed the fracture reduction efficacy of HRT, including hip and spine fractures (73,74). These studies were of women aged 50 to 80 years, who were not known to have increased fracture risk and were supposed to be in general good health. With regard to the treatment of established osteoporosis, although it is likely to be effective, no appropriately sized studies to evaluate hip fracture prevention have been conducted with HRT in women with established osteoporosis. Whether very low-dose unopposed estrogen, with its minimal endo-
CHAPTER28 Interventions for Osteoporosis TABLE 28.5
Route Oral Transdermal Vaginal Subcutaneous Intramuscular
Estrogen Replacement Preparation
Tablets Patches, gels Creams/gels, rings, pessaries Implants
metrial stimulation and relative freedom from other side effects (72), will prove an effective and cheaper alternative for fracture prevention in these women remains to be determined. There is no doubting the efficacy of HRT for fracture prevention in postmenopausal women at increased risk but without previous fracture, and it may be best suited to younger postmenopausal women at increased risk. In such a population, HRT use for osteoporosis prevention would be cost effective (75). Hormone replacement can now be administered via a number of routes, and these are summarized in Table 28.5. There are several types of estrogens in common use in HRT preparations, and the most commonly used are conjugated equine estrogens (CEE), estradiol valerate, 17~-estradiol, and estrone sulphate. Studies have elucidated the optimum dosages of these estrogens for bone conservation, and the results are summarized in Table 28.6 (76-82). A higher dosage may be required for women who have undergone a premature menopause either due to surgery or for medical reasons, and more elderly women embarking on HRT may conserve bone on lower doses than listed, with the added benefit of minimizing the estrogenic side effects common in this group of patients. Unopposed estrogen use is advocated for women having undergone total hysterectomy, but additional progestogens are usually prescribed for women with an intact uterus embarking on HRT, to avoid the potential for endometrial hyperplasia and carcinoma. The addition of progestogen, either sequentially for 12 days each month, or as "bleed-free" continuous combined preparations, has been shown to abolish any increased risk of endometrial carcinoma compared with non-users of HRT (83). Both methods of combined HRT have been shown to produce bone-conserving effects, and there is some evidence to suggest that continuous combined HRT has a greater effect than sequential HRT (84,85). Progestogens derived from 19-norethisterone have been shown to be effective in reducing bone loss (84,86), possibly through their androgenic properties. Less-androgenic progestogens, such as dydrogesterone, have no effect on bone density (70). Although there is near unanimity among researchers on the beneficial effects of estrogen on bone density, far more debate exists regarding the timing and duration of treatment. HRT is not well tolerated in a substantial number of women,
357 TABLE 28.6 Commonly Used Estrogens and Standard Dosages Required to Preserve Bone Density Estrogen
Bone-Conserving Dose
Conjugated oral estrogens (72) Oral: estradiol valerate (77), 17~estradiol (71) Transdermal: 17~-estradiol (80) Percutaneous: 17~-estradiol (79) Subcutaneous: 17f3-estradiol (81-83)
0.3-0.625 mg daily 1-2 mg daily 50 gg daily 1.5 mg daily 25-50 mg every 6 months
Reference numbers are listed in parentheses.
and much attention has been drawn to the potential adverse effects of therapy, as discussed in the following section. Hormone replacement is most often commenced in the perimenopausal period, when symptomatic relief is often the most pressing indication for treatment, at least in the perception of the women affected. The duration of therapy has often been short term following the natural cessation of ovarian function and longer in the case of a surgically induced menopause. However, new data from a long-term prospective study suggest that even limited HRT use in women in the early menopause may result in later fracture reduction (87). 1. SAFETY OF H R T
In recent years, concerns regarding the safety of HRT have been expressed, first from large randomized clinical trials showing only minor risks (74,88), and secondly from an observational study (89) that appears to have overestimated the risk of breast cancer (90). This has led to certain regulatory authorities recommending that clinicians should only prescribe HRT for the prevention of osteoporosis "in postmenopausal women at high risk of future fractures who are intolerant of, or contraindicated for, other medicinal products approved for the prevention of osteoporosis" (90). Thus, it was recommended that HRT should no longer be prescribed freely for the prevention of osteoporosis. The concerns of the regulatory authorities about the safety of HRT have been challenged (90), and it has been suggested that their interpretation of the large clinical trials and one observational study, which they have primarily used as "evidence" for their recommendations, has been misguided. When given for the correct reasons, namely relief of menopausal symptoms or prevention of osteoporosis, the benefits of HRT far outweigh any risks.
2. RECOMMENDATIONS FOR H R T
A main indication for HRT remains the relief of postmenopausal symptoms, which brings about a major improvement in quality of life. No other therapy has proved
358
KEATING AND STEVENSON
more effective than HRT in this respect. It is also quite clear that HRT is an effective treatment for prevention of osteoporotic fractures in women without established disease. These women tend to be younger, because older women have an increasing prevalence of osteoporosis as defined by bone density criteria (91). HRT is less expensive than any of the alternatives, and thus its use for primary prevention should be actively encouraged, targeting postmenopausal women at high risk for fracture, such as those with osteopenia, family history of hip fracture, low body mass index, or history of corticosteroid use.
C. Bisphosphonates 1. PHYSIOLOGY
Bisphosphonates are stable analogues of pyrophosphate and share the same properties. Like pyrophosphate, they inhibit the precipitation of calcium phosphate from solution and inhibit crystal aggregation and the dissolution of crystals of hydroxyapatite. Bisphosphonates display a high affinity for hydroxyapatite and bone mineral, and this essentially limits their effects on bone. In a study using autoradiography with tritium labeled bisphosphonate, the highest concentration of the drug was found on the surface of the bone (92). Bisphosphonates were first recognized as inhibitors of bone resorption about 30 years ago (93,94). Bisphosphonates are effective in Paget's disease, hypercalcemia of malignancy, osteolytic bone metastasis, and osteoporosis. Gastrointestinal absorption of bisphosphonates is very low and if they are taken with or soon after even light meals, none will be absorbed. Approximately 20% of the circulating drug is retained by the skeleton (95), the rest being rapidly excreted unmodified by the kidney. Bisphosphonates bind strongly with hydroxyapatite crystals in the bone. Their skeletal half-life is long, perhaps up to 10 years or longer with certain bisphosphonates. The duration of biologic activity is very long in Paget's disease (months to years) but is shorter in patients with hypercalcemia of malignancy and even shorter in healthy postmenopausal women or patients with other metabolic bone disease (96). Little information has been published on the toxicity of bisphosphonates. Older studies examining etidronate have revealed little toxicity and no carcinogenicity, mitogenicity, or teratogenicity (97,98). Reproductive and developmental toxicity studies with alendronic acid and pamidronic acid have been conducted with rats and rabbits (99,100). No selective developmental toxicity was observed. However, at very high daily doses of pamidronate (10 times higher than the recommended human doses), a low incidence of shortened fetal long bones was noticed, associated with markedly low body weight. These effects are related to the mechanism of action of bisphosphonates. The drug prevents bone resorption in the pregnant animal, thus denying a source of calcium for
the growing fetus. In addition, it crosses the placenta, preventing fetal bone modeling. These findings suggest that bisphosphonates should be avoided in pregnancy unless absolutely indicated. 2. MODE OVACTION The mode of action of bisphosphonates has been studied extensively and appears to involve several possible mechanisms. At lower doses, bisphosphonates bind preferentially in the resorption pits under the osteoclasts; at higher doses, they may affect the mineralization associated with osteoblasts (101). The formation (adsorption) of calcium-bisphosphonate phase on the surface of the bone or enamel changes the bone physiochemical characteristics and can significantly reduce the rate of dissolution of the organic and inorganic substance. This leads to a shallowing of the resorption pits and a reduced rate of osteoclast migration and formation of new pits. At the cellular level, it appears that bisphosphonates are toxic to osteoclasts, leading to cellular death (102,103). As osteoclasts digest the bone surface of a bisphosphonate-treated subject, bisphosphonate is released and internalized by the osteoclast (104). Once internalized, bisphosphonates may interfere with the cytoskeleton, lead to loss of the ruffled border, or be directly cytotoxic (105). The micro-injection of bisphosphonates into isolated osteoclasts disrupts the cytoskeletal ring of action in polarized, resorbing osteoclasts (106). Bisphosphonates may also affect enzymes and then signal transduction cascades known to be responsible for osteoclast formation and activity (107). It is also possible that bisphosphonates affect bone resorption via release of a resorption inhibition factor of low molecular weight (<10,000 Da) from osteoblasts (108,109). 3. BISPHOSPHONATETREATMENT FOR OSTEOPOROSIS Etidronate was the first bisphosphonate developed and, when given early after the menopause, preserved bone mineral density (BMD) in the spine and femoral neck over a 2-year period compared with placebo (110,111). In another study (112), cyclical etidronate over 2 years significantly increased BMD in the lumbar spine compared with placebo across stratified postmenopausal age groups (i.e., 1-3, 4-6, and 7-10 years postmenopause). A 2-year study (113) in early postmenopausal women showed statistically significant preservation of spinal BMD with etidronate compared with placebo. There was no change in the femoral neck BMD in the etidronate group over baseline, but there was a 2% loss in the placebo group, the difference being significant. The results of the studies were consistent with the mechanism of action of antiresorptive agents, with the biochemical results showing significant suppression of markers of bone remodeling, indicative of reduction in bone resorption and bone formation.
CHaPTeR 28 Interventions for Osteoporosis Unfortunately, there is a lack of unequivocal evidence for reduction in fracture at nonvertebral sites with etidronate, and in recent years its use has been largely superseded in clinical practice by newer bisphosphonates. Alendronate is 100 times more potent than etidronate. It is poorly absorbed after oral administration, with bioavailability of only 0.75%, even when taken after an overnight fast and 2 hours before breakfast. In a 3-year study, oral alendronate increased BMD in the spine and reduced the incidence of fractures in women with postmenopausal osteoporosis (114). A further study showed that alendronate was effective in reducing the number of vertebral and other clinical fractures in postmenopausal women with low BMD and at least one existing vertebral fracture. The relative risk (RR) of sustaining a new vertebral fracture after 3 years of treatment with alendronate compared with those treated with placebo was 0.46. The risk of any clinical fracture was 0.72, and the risk of hip fracture and wrist fracture was 0.49 and 0.52, respectively (115).The efficacy of alendronate in preventing nonvertebral fractures was confirmed in a metaanalysis, with women over 65 years of age experiencing the greatest benefit (116). The risk reduction for hip and wrist fractures was 54% and 61%, respectively, but the risk reduction seen in the incidence of hip fractures was not statistically significant. Alendronate in half of the normal dosage (5 mg! day instead of 10 mg/day) was effective in preventing bone loss in early postmenopausal women. Two years of treatment resulted in a 3.5% gain in lumbar spine BMD. In this smaller dosage, alendronate was well tolerated, with a side effect profile similar to that of placebo (117). Risedronate has also been shown to reduce the risk of vertebral and hip fracture (the latter mainly in women with established osteoporosis) (118,119). Both alendronate and risedronate are available in daily and weekly oral formulations, the latter having been developed partially to address the issue of upper gastrointestinal disturbance commonly seen with oral administration of bisphosphonates. Bisphosphonate administration is undoubtedly tiresome for the patient because it requires dosage on an empty stomach, fasting for 2 hours postadministration, and avoiding lying down for at least 30 minutes postingestion. Of more concern, bisphosphonates have been reported to be associated with osteonecrosis of the jaws (120). Ibandronate is a newer bisphosphonate and has been shown to reduce vertebral fracture in women with postmenopausal osteoporosis when taken either daily or cyclically (121). It currently has a license for once-monthly treatment, and again this may partially improve compliance in patients in whom upper gastrointestinal side effects have been prominent. Its effect, if any, on hip fracture risk is currently incomplete. Ibandronate may also be administered every 3 months by intravenous injection. In the future, other new bisphosphonates may permit even less-frequent dosing. Zoledronate is being evaluated for 3-monthly administration orally and 12-monthly administration by intravenous infusion.
359
D. Calcium and Vitamin D 1. CALCIUM PHYSIOLOGY AND METABOLISM
More than 98% of the 0.7 to 1.5 kg of calcium in the average adult body lies in the skeleton, and only a minor proportion of the total calcium (about 0.5%) is exchangeable. The average daily dietary intake of calcium in American adults is 600 to 800 mg, less than half of which is absorbed. Most calcium is absorbed in the proximal small intestine. Calcium absorption increases in pregnancy and lactation and during rapid growth in children and decreases with advancing age. Maintenance of calcium balance is dependent on the efficiency of intestinal absorption. Vitamin D stimulates calcium absorption from the gut. PTH decreases the urinary loss of calcium by stimulating calcium reabsorption from the proximal tubule in the kidney. Intestinal disease or dietary calcium deprivation leads to decreased calcium bioavailability from the gut. The role of calcium in the prevention and treatment of osteoporosis has been controversial, with arguments presented both for and against the concept of calcium supplementation for all postmenopausal women (122,123). Evidence from an 18-month randomized controlled trial showed that increased dietary calcium (increased milk intake of 300 mL over the usual daily consumption of 150 mL) in adolescent girls led to increased bone mineral density (124). However, this study may have demonstrated an acceleration in the accrual of peak bone mass with calcium supplementation in adolescents rather than an overall increase in the peak mass achieved, and it is not necessarily applicable to a postmenopausal population. In adults following estrogen withdrawal, increased bone resorption occurs with increased plasma calcium levels, and consequently increased urinary calcium excretion, together with reduced PTH levels and calcitriol production. Estrogen administration lowers urinary calcium and hydroxyproline values and plasma alkaline phosphatase activity (125). These effects are largely mediated through the direct effect of estrogen on the bone via estrogen receptors on the bone cells and possibly via the direct effect of estrogen on tubular reabsorption of calcium via estrogen receptors in the kidneys (126). 2. CALCIUM TREATMENT
To counteract these losses in estrogen-depleted postmenopausal women, calcium supplementation may be considered. Calcium absorption decreases with age, and it has been estimated that 1000 mg/day of extra calcium is needed to counteract the decreased calcium absorption and increased calcium excretion (123). Cross-sectional casecontrolled studies have examined the link between calcium intake and hip fracture rates in men and women over the age of 50 (127-129). Women with lower calcium intake were at higher risk of hip fracture (adjusted RR 1.9) compared with controls.
360
Some double-blind, randomized, placebo-controlled studies (130-132), but not all (133), have generally confirmed the findings of case-controlled studies, showing that calcium supplementation between 500 to 1000 mg/day slows the rate of postmenopausal bone loss in women of at least 3 years postmenopause. The ages of the women participating in these studies were between 40 and 70 years. The serum parathyroid hormone concentrations, serum alkaline phosphatase concentration, and urinary hydroxyproline excretion tended to be lower in the calcium group. However, calcium and vitamin D supplementation may not prevent fractures in women without vitamin D deficiency (see later discussion). Calcium absorption is dependent on vitamin D (see later discussion) and can be inhibited by certain medications and foods. Optimal calcium intake may be achieved through diet, calcium-fortified foods, calcium supplements, or various combinations of these sources. Oral calcium supplementation uses calcium carbonate or citrate. In normal individuals, the absorption of calcium citrate is better than calcium carbonate (134). A dose-ranging study found that the calcium absorption curve is initially steep (between 0 mg and 500 mg) and then plateaus. It appears that the amount of calcium absorbed from 500 mg of calcium citrate is greater than that from 2000 mg of calcium carbonate (135). Not only is calcium better absorbed from its citrate salt, but calcium citrate leads to greater inhibition of P T H than does calcium carbonate (136). The better absorption of calcium citrate may be especially important in subjects with decreased stomach acidity. The secretion of gastric acid decreases with age, and as many as 50% of apparently healthy people over age 50 may have decreased secretion. The absorption of calcium citrate is independent of the presence of gastric acid in the stomach, whereas in persons with low gastric acid, the absorption of calcium carbonate is insignificant (137,138). High levels of calcium intake have several potential adverse effects. In dosages greater than 5000 mg/day, calcium toxicity with high serum calcium levels, severe renal damage, and ectopic calcium deposition (milk-alkali syndrome) may occur. This may also occur when calcium carbonate is overused as an antacid. Adverse effects have not usually been observed with doses less than 1500 mg/day. Ideally, calcium supplements should be taken between meals to maximize the absorption of calcium. Gastrointestinal side effects such as constipation are rare at dosages less than 1500 mg/day.
3. VITAMIN D PHYSIOLOGY AND METABOLISM
Vitamin D entails a range of fat-soluble compounds with similar activity on calcium and phosphate physiology, namely enhanced gastrointestinal absorption and an increase in bone deposition. The major biologic functions of vitamin D are to regulate calcium and phosphate absorption and me-
KEATING AND STEVENSON
tabolism and to maintain plasma calcium levels via bone resorption and formation. These actions help to form and maintain healthy bones. The compounds include vitamin D2 (ergocalciferol), pro-vitamin D3 (7-dehydrocholesterol), vitamin D3 (cholecalciferol), 25-hydroxycholecalciferol (25HCC; calcidiol), and 1,25-dihydroxycholecalciferol (1,25DHCC; calcitriol). Approximately 80% of the body's vitamin D comes from the action of ultraviolet-B (290-320 nm) sunlight on provitamin D3 in the skin, which results in the formation of vitamin D3. At body temperature, it takes about 24 hours for vitamin D3 to form in the skin. Aging decreases the capacity of the skin to produce vitamin D3, with a greater than fourfold decrease in production after age 70. The use of topical sunscreens, increasing latitude, early or later times of the day, when sunlight is weaker, and decreased area of skin exposure to light also negatively affect vitamin D3 production. Latitude is especially important, and in the winter months at latitudes >42 degrees, little or no vitamin D3 can be synthesized. When there is insufficient production of vitamin D3 in the skin, oral vitamin D intake becomes important. A few natural foods contain adequate amounts of vitamin D2 or D3, such as egg yolk and fishliver oils, but fortified products such as milk, infant formula, margarines, and breakfast cereals also contain vitamin D. Ergosterol within plant matter can be converted to vitamin D. Liver and liver products do contain plentiful sources of vitamin D but also are rich in vitamin A, and current advice is that intake of vitamin A-rich produce should be restricted to once weekly and avoided in pregnancy due to the potential risk of vitamin A toxicity. Vitamin D3 undergoes two hydroxylation steps before attaining full physiologic activity. The first step occurs in the liver, where calcidiol is produced, and the second step occurs in the kidney and depends on the body's requirements for calcium. If there is a tendency toward hypocalcemia, P T H stimulates the production of calcitriol. In the presence of normocalcemia or hypercalcemia, metabolism of calcidiol is directed by a 24-hydroxylase enzyme toward the formation of 24,25-DHCC, which is metabolically inactive. Calcitriol is the physiologically important metabolite, acting on its target tissues via steroid receptors. Calcitriol binds to the vitamin D receptor (VDR) and forms a complex. This complex interacts with a receptor accessory factor to form a heterodimer, which in turn interacts with the vitamin D response element in the nucleus. In the bone, this interaction leads to expression of mRNA and increased production of osteocalcin and osteopontin. In the intestine, calcium-binding protein is synthesized, leading to increased absorption of calcium and phosphate into the circulation (139,140). Although it is clear that vitamin D has an important role in normal bone homeostasis and calcium metabolism, its role in postmenopausal osteoporosis is less evident. As men-
CHAPTER28 Interventions for Osteoporosis tioned previously, peak bone mass is dependent on genetic and environmental factors, with the genetic component currently thought to be responsible for around 80% of the peak bone mass. It has been reported that the vitamin D genotype was responsible for as much as 75% of the total genetic component of bone density in healthy individuals. When a particular genotype was used as a predictor of low bone density (more than 2 SD below mean young normal density), it was found that women with B B genotype would reach this level 18.4 years after the menopause, compared with 29 years for the bb genotype and 22 years for the heterozygote genotype Bb (141). It appears that premenopausal individuals with the bb genotype not only have higher bone density but also have a different response to vitamin D administration than those with the B B genotype. When given vitamin D, those with the bb genotype respond with higher levels of osteocalcin and lower levels of parathyroid hormone (142). Although these findings may expand our knowledge of bone metabolism, they are by no means free of controversy. Other studies have found no correlation between low bone density and genotype (143) or between osteoporosis and genotype (144). The elderly are more prone to vitamin D deficiency, with poorer oral absorption of vitamin D and reduced formation in the skin further complicated by less exposure to sunlight, especially in those with poor mobility. In addition, older people may have less oral vitamin D intake, in part due to a lower intake of fortified milk. In one study, it was found that less than one in three brands of milk contained 80% to 120% percent of the amount of vitamin D stated on the label, and some of them contained none at all (145). Few studies have been performed using vitamin D alone to assess potential benefit in reducing fracture risk, and no studies have been performed in postmenopausal women alone. Extrapolation to a postmenopausal female population is therefore hazardous at best, but four large randomized studies have looked at vitamin D supplementation. The resuits of these studies were generally unfavorable, with one trial showing a non-significant increase in the risk of hip fractures in elderly men and women in primary care receiving a daily dose of 400 1U vitamin D (146), and a more recent trial showing a small, insignificant increase in all fractures with a large, borderline statistically significant increase in hip fractures with an annual injection of 300,000 IU of vitamin D (147). The Medical Research Council RECORD trial also studied vitamin D alone and found no evidence of benefit in the secondary prevention of low-trauma fractures (148). However, a trial of high-dose oral vitamin D (100,000 units) every four months in men showed a borderline statistically significant 22% reduction in osteoporotic fractures (149). In clinical studies examining the effects of calcium and vitamin D on BMD and fracture incidence, dual supplementation appeared to increase the BMD in men and women older than 65, and the incidence rate of nonvertebral
361 fractures was lower in the treatment group (150). Two large studies of elderly women residents in French nursing homes have also shown benefit (151,152), with the earlier work showing a 43% and 32% reduction in the number of hip fractures and total number of nonvertebral fractures, respectively, and the later work confirming this effect and also confirming the earlier study's finding of a reversal in secondary hyperparathyroidism. A community study in Denmark also showed a statistically significant 16% reduction in fractures (153). However, recent trial work in a primary care setting to assess the effect of dual therapy on women aged 70 years or older, with one or more risk factors for hip fracture, showed no benefit from calcium and vitamin D supplementation on the primary endpoint measured, all clinical fractures (154). These results are of concern in that the population studied was representative of the typical postmenopausal woman targeted for dual calcium and vitamin D supplementation in the community. The large benefit shown from dual supplementation in the nursing home studies mentioned previously may partly be explained by the tendency for this population to be less healthy, and more deficient in vitamin D and calcium, both from dietary intake and the lack of exposure to daylight, relative to a community population. A further, important study from the Medical Research Council RECORD Trial Group also showed no evidence of a benefit from combined calcium and vitamin D supplementation in secondary prevention of fractures (143). In studies of the elderly, the effect of vitamin D on bone density may be largely due to correcting subclinical vitamin D deficiency rather than affecting osteoporosis. There is legitimate concern about the ability of vitamin D to induce hypercalcemia, nephrocalcinosis, or renal stones. The calcium and vitamin D arm of the W H I randomized trial showed no overall fracture reduction and an increase in renal calculi (155). Calcium and vitamin D supplementation should be regarded at best as an adjunct to treatment. It can only be regarded as a treatment for those who are deemed to be deficient in calcium and vitamin D, mainly the housebound elderly.
E. P a r a t h y r o i d H o r m o n e ( P T H ) P T H is an 84 amino acid polypeptide chain produced in the cells of the parathyroid glands. Its function is to maintain the extracellular fluid (ECF) calcium concentration. However, the complete polypeptide chain is not necessary for the biologic action of the hormone. Synthetic fragments containing the amino-terminal sequence residues 1-34 exert the known biologic actions of the hormone. Any fall in plasma calcium concentration outside the physiologic range leads to increased P T H production and secretion. P T H acts directly on the bone and kidney, and indirectly on the intes-
362 tine through its effects on the synthesis of 1,25(OH)2D, to increase serum calcium concentration. P T H secretion leads to increased serum calcium via an increased rate of dissolution of bone mineral, a reduction in the renal clearance of calcium, and an increase in the efficiency of calcium absorption from the intestine. P T H exerts its effect through the mineral ion transport of the kidney and the bone and, by stimulating the renal 25-hydroxyvitamin D, 1-alphahydroxylase, increases intestinal calcium absorption. Once serum calcium increases, it suppresses further secretion of P T H via a direct feedback control mechanism. P T H has two main actions on bone, altering calcium availability and bone remodeling. The effect on calcium availability is rapid (within minutes) and is crucial for short-term calcium homeostasis. The effect on bone remodeling becomes apparent after some hours, with increased numbers of osteoclasts and a subsequent increase in the remodeling of bone. However, the exact effect of P T H on bone remodeling is not completely understood. Receptors for P T H are found on osteoblasts but not on osteoclasts. Thus, the action of P T H on osteoclasts is thought to be indirect, through cytokines released from osteoblasts to stimulate osteoclast differentiation from their precursors. P T H action on the bone depends on the mode of secretion. Chronic and greatly increased P T H secretion as in severe hyperparathyroidism leads to significant bone resorption and the well-defined bone disease osteitis fibrosa cystica. Chronic intermittent lower-dose hormone administration leads to an anabolic action on the skeleton with increased bone density. Animal data suggest a pronounced anabolic effect of P T H on the skeleton, with an increase in bone mass and bone strength (156,157).
1. P T H IN OSTEOPOROSIS
Early studies of P T H in the treatment of postmenopausal osteoporosis were variable. Reeve et al. (158) demonstrated no new fractures during one study of postmenopausal women with at least one previous osteoporotic vertebral fracture. However, the study was not randomized and there were no measurements of BMD. In a later 6-month randomized controlled trial, subcutaneous P T H administration was associated with preserved BMD in the lumbar spine, as compared with a 2.8% loss of BMD with placebo, when measured with DEXA scanning (159). However, no change in BMD was observed in the distal radius, and a slight decrease in femoral BMD was observed in the subjects, who were all under current treatment with a gonadotrophin releasing hormone (GnRH) analogue for endometriosis and fibroids. In an additional 3-year randomized controlled trial (160), subcutaneous P T H was given to postmenopausal women with established osteoporosis who were already receiving estrogen treatment. Compared with the control group, the BMD of patients taking both HRT and P T H
KEATING AND STEVENSON
showed an increase of 13% at the vertebrae and 2.7% at the hip, with significantly fewer vertebral fractures in the PTHtreated group. More recently, a recombinant human P T H (1-34), [rhPTH(1-34)], teriparatide, has been produced and licensed for the treatment of postmenopausal osteoporosis. Given by (self-administered) subcutaneous injection for 2 years in a randomized, placebo-controlled trial of postmenopausal women with previous osteoporotic fracture, teriparatide at 20-lag and 40-lag dosage reduced the RR of new incidences of vertebral fractures by 65% and 69%, respectively, compared with placebo and increased BMD in the spine by 9% and 13%, respectively, more than placebo (161). The RR of nonvertebral new fracture was also significantly reduced by 35% and 40%, respectively, compared with placebo, and hip BMD was significantly increased. The drug was well tolerated, but side effects of nausea and headache were more likely at the 40-btg dose. Some concerns were raised on the potential long-term safety of teriparatide given the findings of increased incidence of osteosarcoma in rats treated with high-dose regimens (161), but overall risk-benefit assessment has concluded that the findings from the rat studies were unlikely to predict an increased risk of osteosarcomas in human patients with osteoporosis receiving teriparatide therapy for 2 years or less (162). Further work on the cohort from the original study (161) has shown that teriparatide offers clinical benefit to patients across a broad range of age and disease severity (163). Furthermore, there is evidence that teriparatide can actually produce beneficial changes in bone structure and even restore trabecular connectivity (163a). Clearly, the administration of teriparatide via subcutaneous injection renders this an expensive and intensive therapy relative to alternative regimens, and although intensive monitoring for potential adverse effects remains important, its use in an appropriate setting seems justified by its beneficial effects on new fracture incidence.
F. Fluoride In the late 1930s, it was noticed that workers exposed to industrial fluoride had bones with abnormally high density. This observation led to the first clinical experiment with sodium fluoride, which was given to patients with Paget's disease of bone or patients with osteoporosis. It resulted in positive calcium balance (164). Fluoride is present in the food chain from a variety of sources, the most common of which are fluorinated water, toothpaste, and mouthwash. Absorption occurs rapidly in the stomach, mainly as hydrofluoric acid. Absorption decreases with age, and the presence of cations such as calcium, magnesium, and aluminium can decrease absorption by 20% to 30%. Fluoride given in large
CHAPTER28 Interventions for Osteoporosis dosages (25 to 100 mg/day) often leads to a chemical gastritis, but this is uncommon when slow-release tablets are used. Fluoride acts both on bone cells and bone mineral, increasing the recruitment of osteoblasts and decreasing the activity of osteoclasts. It displaces the hydroxyl group of hydroxyapatite to form fluoroapatite. These crystals of fluoroxyapatire are more densely packed and are less susceptible to the action of osteoclasts (165), but the fluoroxyapatite structure may be more brittle.
1. EFFECT o r FLUORIDE TREATMENT ON OSTEOPOROSIS
Riggs et al. performed a 4-year randomized trial using 75 mg/day of fluoride in postmenopausal women with established osteoporosis (166). This led to a striking improvement in bone density, increasing by 35% at the spine, 12% at the femoral neck, and 10% at the femoral trochanter, although a 4% decrease in the BMD at the radial shaft was noted. Fewer vertebral fractures were noted in the fluoridetreated patients, but this effect did not reach clinical significance. However, fewer nonvertebral fractures were seen in the placebo group, the effect of which was clinically significant. It was concluded that the newly formed bone had greater crystallinity but decreased elasticity and thus had inadequate strength. The authors were particularly troubled by the higher rate of long bone fracture in the humerus, femur, and tibia of fluoride-treated patients, as well as the high rate of side effects such as nausea, vomiting, epigastric pain, and leg pains. They thus felt that the fluoride-calcium combination was not suitable in the treatment of osteoporosis. In a later randomized controlled study, lasting almost 5 years, Pak et al. (167) administered 50 mg/day of slowrelease sodium fluoride and 400 mg/day of calcium to women with postmenopausal osteoporosis. They reported a yearly increase in bone density of 4% to 5% at the lumbar spine and 2.38% at the femoral neck, but no change at the radial shaft. The treatment group experienced significantly fewer new vertebral fractures, and the incidence of side effects, including gastrointestinal upset and microfractures, was not different between the groups. It was thought that the intermittent dosing of slow-release fluoride, of lower dosage than previously used, contributed to the limitation of side effects seen with this study. Two editorials were published as a response to the interim and full analyses of this study, concluding that slow-release sodium fluoride, 50 mg/ day, was a safe and effective method of increasing bone density and preventing first osteoporotic fractures (168,169). Although the use of fluoride had previously been endorsed by the U.K. Consensus Group (170), its use now is extremely limited in clinical practice and is mainly confined to the research setting, where it is being studied as part of combination therapies for osteoporosis.
363
G. Calcitonin Calcitonin is a peptide hormone that acts as a physiologic antagonist to parathyroid hormone. Calcitonin is produced by the parafollicular (C-cells) in the thyroid gland. Naturally occurring calcitonin is a 32 amino acid peptide, but there is considerable variability in sequence among species. Calcitonin from salmon is 10 to 100 times more potent than mammalian forms in lowering serum calcium in animals. The secretion of calcitonin, which has a circulating halflife of 2 to 15 minutes, is influenced directly by blood calcium levels, with an increase in calcium causing an increase of calcitonin secretion, leading to a hypocalcemic effect in situations of high bone turnover. Calcitonin acts by inhibiting osteoclast-mediated bone resorption and also by stimulating renal calcium clearance. It also reduces the number of active osteoclasts and the rate at which new osteoclasts are formed. Receptors for calcitonin are found on osteoclasts (171,172) and renal tubular cells. Other calcitonin-binding receptors are thought to be present in the brain, gastrointestinal tract, and immune system, and the presence of these may help to explain the analgesic effects exerted directly on cells in the hypothalamus and related structures by the hormone. Human calcitonin is not in general clinical use at present. Salmon, eel, and pork calcitonin have been used instead. Salmon caMtonin (salcatonin) differs from human calcitonin in 16 amino acid residues. It is 40 to 50 times more potent and has a longer circulating half-life than human calcitonin. Salcatonin is available in injectable or intranasal formulations (173). Following intranasal administration, peak plasma concentrations are seen 30 to 40 minutes later, compared with 15 to 25 minutes after intravenous or subcutaneous administration. Intranasal administration provides approximately 25% to 50% of the biologic activity of the same dose administered through injection. The elimination half-life of salcatonin is around 45 to 90 minutes. The administration of salcatonin may cause nausea, vomiting, facial flushing, tingling of the hands, and an unpleasant taste. Rarely, serious allergic reactions may occur, including bronchospasm, angioedema, or even anaphylactic shock. Nasal irritation may also occur. Approximately 40% to 70% of patients receiving intranasal or parenteral salcatonin therapy for more than 6 months will develop specific antibodies. However, these antibodies do not appear to decrease the effect of salcatonin on bone (174). The precise role of calcitonin in bone loss is not entirely clear. Levels are elevated during growth, pregnancy, and lactation, suggesting that a physiologic role of caMtonin is prevention of unwanted bone resorption (175). The role of endogenous secretion of calcitonin in postmenopausal osteoporosis is also not fully established. Plasma calcitonin is lower in women than in men (176) and lower in women with osteoporosis (177,178), although not all studies have
364
KEATING AND STEVENSON
found this to be the case (179). However, pharmacologic doses of calcitonin directly inhibit osteoclast activity. 1. THERAPEUTIC USE OF CALCITONIN IN POSTMENOPAUSALWOMEN Calcitonin given for I to 2 years to early postmenopausal women, within 5 years of their final menstruation, resulted in an increase in BMD of the lumbar spine. Doses of 50 IU daily given by subcutaneous injection led to an increase in vertebral BMD (180). Intranasal calcitonin appears to require a minimum dose of 200 IU daily (181,182). The effect on vertebral BMD is greater in older women, and little effect is seen on the proximal femur (182). Older women with established osteoporosis showed similar results and decreased vertebral fracture rates, which varied between 35% to 60% after 24 months of treatment (183,184). A metaanalysis reviewed 18 clinical trials involving calcitonin (185). The pooled positive change was found to be 1.97% in vertebral BMD and 0.32% in proximal femur BMD, with an aggregated number of vertebral fractures prevented by the treatment in the order of 59.2 per 1000 patient years. However, calcitonin was not shown to be effective in the prevention of bone loss in one study of perimenopausal women, aged 40 years or older, with irregular menstrual cycles (186). The largest randomized placebo-controlled fracture study of calcitonin showed a significant reduction in vertebral fractures with 200 IU daily by intranasal administration (187). However, the study compared doses of 100 IU, 200 IU, and 400 IU with placebo, and the 400-IU group failed to demonstrate any significant reduction in fracture. Even more surprising, there was a significant reduction in nonvertebral fractures seen only in the 100-IU group. It seems likely that small numbers of events make the findings very difficult to evaluate.
H. Tibolone Tibolone (LMal, Org OD 14) is a synthetic molecule that has been used to treat vasomotor symptoms of the menopause. It is structurally related to norethisterone and possesses a combination of estrogenic, progestogenic, and androgenic activities. In animal studies it was shown to have 1% to 10% of the estrogenic activity of ethinylestradiol, 12% of the progestogenic activity of norethisterone, and 2% of the androgenic actMty of methyltestosterone (188). Tibolone is effective in treating vasomotor symptoms (188). It has been shown that tibolone is effective in preventing accelerated bone loss in early postmenopausal women (189) and oophorectomized women (190,191). Tibolone is effective in the treatment of established osteoporosis. A 2-year study of tibolone treatment in a group of women with osteoporosis and a mean age of 69 led to an 8% increase in lumbar bone density
(192). Similarly positive results were seen in another study, which also revealed that a lower dose of 1.25 mg/day was as effective as the standard dose of 2.5 mg/day in the preservation of bone density (193). Final results are awaited from a fracture prevention trial of tibolone, but it is known that it reduced vertebral fractures by around 50% (194). An increase in stroke with tibolone was also recorded in this study, but further results need to be assessed before conclusions about the true relevance of this can be determined. Tibolone is otherwise safe and well tolerated (183), and its side effect profile is similar to HRT, with the most frequently reported symptoms being vaginal bleeding and discharge, pruritus vulvae, breast tenderness, acne, and hirsutism. Other side effects include weight changes, ankle edema, headache, and visual disturbances. Endometrial atrophy occurs in 90% of users (195), but should breakthrough bleeding occur after 3 months of treatment, investigation would be merited (196). The metabolic effects oftibolone have also been studied, and administration led to a decrease in total cholesterol, HDL, LDL, triglycerides, Apo-A1, Apo-B, and Lp(a) (197,198). Coagulation and fibrinolysis were also favorably affected, with decreased fibrinogen and PAI-1 and increased plasminogen levels (199). Its long-term effects on cardiovascular disease development remain unknown, as are any effects on other tissues such as brain and breast.
I. Androgens Age-related decrease in androgen production starts premenopausally, and androgen levels fall by half between the ages of 20 and 40 (200). After the menopause, the process continues. The age-related decline is particularly noticeable for dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS). Following natural menopause, the levels of D H E A and DHEAS are 40% and 30% less than premenopausal values, respectively, and those of androstenedione and testosterone are 50% and 70% less than their premenopausal values, respectively (201). After oophorectomy, the levels of testosterone and androstenedione fall by 50% in previously premenopausal women and by 50% and 21%, respectively, in previously postmenopausal women (202). Androgen receptors have been found in osteoblastic cells (203). Androgens directly stimulate human bone cell proliferation and differentiation (204). Testosterone is a precursor of estrogen, and one study of testosterone implants revealed that 3 months after implant insertion, estrogen levels had increased from 59.8 pmol/L to 120 pmol/L (205). An additional study of postmenopausal women with adrenocortical failure (Addison's disease) found 3 to 10 times lower levels of androstenedione, DHEA, DHEAS, and testosterone compared with controls and significantly lower radial bone density, both distally and at the midshaft (206). A
CHAPTER28 Interventions for Osteoporosis number of studies have been performed using parenteral testosterone (as subcutaneous implants or injections) and oral methyltestosterone. Two years of treatment with estrogen and methyltestosterone led to a 3.4% increase in spine BMD compared with the estrogen-only control group. The treatment was well tolerated, with acne and increased hair growth being the most common side effects (207). Methyltestosterone at daily doses of 1.25 to 2.5 mg has been found to be safe and effective and to avoid liver dysfunction. However, total cholesterol, HDL, triglycerides, and apolipoprotein A-1 decreased in estrogen/methyltestosterone groups compared with the baseline (208). A further metabolic problem with oral androgens is that they induce insulin resistance. Androgens have a positive skeletal effect. In one study, concurrent 100-mg testosterone implants and 75-mg estradiol implants led to a 5.7% increase in bone density at the spine and a 5.2% increase at the femoral neck compared with controls on CEE and progestogens (209), although the design and control of the study were less than ideal. Testosterone administration in the form of 50-mg subcutaneous implants every 6 months is well tolerated, with signs of androgen excess being exceedingly rare provided that treatment is properly monitored. General experience shows that if implants are withheld when serum testosterone is above the normal range, side effects can be avoided (210). In further studies, treatment of postmenopausal women with nandrolone decanoate injections led to increased vertebral BMD (211,212) and increased BMD at the distal radius (213,214) and resulted in reduced fracture rates (212). Women administered with testosterone have also reported increased well-being and libido (215,216). Non-oral delivery of androgens and the avoidance of potent testosterone esters would appear to be prudent.
j. Exercise Bone mass and muscle mass are positively correlated (217). Bone gain is also positively correlated with physical activity (218). Static loads are inefficient whereas dynamic loads are efficient in increasing the BMD. Changes in the strain may be more important than the absolute magnitude of the load; the higher the rate and frequency of changes in the strain, the higher the bone density produced. Studies in animals have demonstrated that increasing strain magnitude initiates a dose-dependent elevation of bone mass (219). During locomotion, the magnitude and rate of limb bone strains are higher with heel strike when running or landing from a jump. Gymnastic exercise is thus more effective in increasing the bone mass compared with cycling or swimming (220). Young gymnasts (mean age 21 years) gained about 2% in lumbar spine BMD and 1.6% in femoral neck BMD in an 8-month study, indicating that BMD of the clinically relevant sites of the lumbar spine and femoral neck
365 can respond dramatically to mechanical loading characteristic of gymnastic-type exercise in college-aged women (221). In professional tennis players, the BMD of the playing arm was found to be as much as 30% greater than the non-playing arm (222). Former elite athletes have demonstrated 12.1% higher BMD in the hip and 8.7% higher BMD in the spine compared with controls many years after stopping competitive training (223). In the same study, active controls (defined as those vigorously exercising for more than 1 hour per week) had higher bone density than inactive ones. The mechanism of action of exercise on the bone is not clear. The theory of mechanotransabduction, which concerns the flow of fluid through interstices of bone, provides a possible explanation, suggesting that loading causes fluid flow that affects the cell directly either by local deformation or by some electrical effect related to streaming potentials. Low-rate strains would cause relatively sluggish fluid movements, whereas high-rate strains could cause rapid local fluid movements with potentially greater effectiveness in initiating cellular events (224). Evidence that exercise is important in preserving bone density has also been derived from observation of paralyzed individuals. Studies involving paraplegics and tetraplegics have showed conclusively that there is a significant demineralization of the skeleton over time (225). Paraplegic patients able to stand in leg braces showed higher BMD in the proximal femur than those unable to do so, demonstrating the beneficial effect of mechanical loading on the bone in osteoporosis secondary to paraplegic conditions (226). Confirmation that loading is important in maintaining bone density has also been derived from studies on astronauts. After only 3 weeks in space, experimental dogs and rats began to lose bone mass. Similar findings were noted in astronauts after 24 days of flight, with losses of between 14% and 18% in the calcaneus (227). The effect of exercise seems to be independent of estrogen status in women. In premenopausal women aged between 35 and 45, high-impact exercise (including aerobic and step exercises and high-impact jumping exercises) performed 3 times per week over a period of 18 months led to a significant increase in BMD at the femoral neck compared with controls (228). Exercise also improved muscle performance and dynamic control in these subjects. However, female athletes who become estrogen deficient lose bone significantly, and thus exercise alone is not sufficient to prevent estrogen-deficiency bone loss.
1. EFFECT OF EXERCISE ON THE BONE DENSITY OF MENOPAUSAL WOMEN
It seems reasonable to extend the previous findings and physiologic considerations to the prevention of osteoporosis and fractures in postmenopausal women because prevention of fractures involves not only maximizing bone density but
366
KEATING AND STEVENSON
also addressing the indirect factors that contribute to falls (i.e., coordination, muscle strength, and balance). High-intensity strength training over 1 year led to improvements in muscle mass, strength, and balance and to bone density gains in estrogen-depleted women aged between 50 and 70. The gain in the femoral neck and the lumbar spine was 0.9% and 1%, respectively, in the exercise group, compared with losses of 2.5% and 0.019%, respectively, in the control group (229). The Leisure World Study examined the risk of hip fracture in a cohort of men and women with a mean age of 73 years. The study was prospective, and 8600 women were included in the analysis. Women engaging in active exercise for more than I hour per day had a significantly decreased risk of hip fracture compared with those exercising for less than half an hour or not at all (RR 0.6) (230). A further group of men and women over 70 years of age based in the community were assigned to either an intervention group (which included exercise) or a control group. Those in the intervention group suffered fewer fails and fewer fractures compared with the control group (231). It seems, therefore, that exercise is effective through all age groups. A population-based approach encouraging moderate exercise from childhood to old age may contribute to the preservation of bone mass, as well as improving muscle strength, agility, and balance. The consequent decreased risk of falls (232) together with improvement in bone density should logically lead to fewer fractures and thus less morbidity and mortality.
K. Selective Estrogen Receptor Modulators (SERMs) A search for safer and more acceptable alternatives to HRT led to the evaluation of several compounds, grouped together as "antiestrogens." Initially, these compounds were used in the treatment of hormonally responsive cancers, such as breast cancer. However, antiestrogens have a wide spectrum of actMty, from being pure estrogen antagonists to mixed agonists/antagonists. The most familiar are the triphenylethylenes, such as clomiphene, tamoxifen, 3-hydrotamoxifen (toremifene), 4-iodo-pyrrolidintamoxifen (droloxifene), and TAT-59. Clomiphene and tamoxifen are in widespread clinical use, and recently toremifene (Fareston) has been approved for the treatment of advanced breast cancer. Other agents with mixed agonist/antagonist properties include benzothiophenes such as raloxifene (formerly keoxofene) and structurally distinct molecule such as ormeloxifene (centchroman). ICI 182 780 has been found to be a pure estrogen antagonist and is now under clinical investigation for the treatment of tamoxifen-resistant breast cancer. The term selective estrogen receptor modulator (SERM) was proposed to define more precisely these compounds that can
bind to and activate estrogen receptors (ERs) but have effects on target tissues that may vary from those of estradiol (233). As currently used, SERM classification includes all compounds previously described, excluding pure antiestrogens such as ICI 182 780. Research actMty increased markedly following the discovery that tamoxifen-treated postmenopausal women with breast cancer had lower cholesterol and preserved bone mass compared with untreated control patients. Intensive research is currently aimed at understanding the mechanism of the selectivity of estrogen agonist and estrogen antagonist behavior by SERMs. The ER itself lies at the center of this mechanism. Emerging evidence from several groups indicates the presence of multiple transcription pathways for ligand-bound ERs and the existence of multiple ER subtypes. The ER contains multiple transcription activating functions (e.g., AF-1 and AF-2) that account for some of the tissue-selective effects of SERMs via different co-activator and co-repressor elements. At the cellular and molecular level, antiestrogens can act in several ways. Some antiestrogen compounds bind the ER but fail to activate it; some activate the ER, and the complex binds to the acceptor site but fails to fully regulate gene transcription or expression. There may also be alternative pathways to the classical one through which drugs may initiate cellular response. It appears that at least some SERMs may act by binding to the cell DNA without the DNA binding domain of the ER complex (234).
VI. EFFECTS OF SERMS ON POSTMENOPAUSAL OSTEOPOROSIS Tamoxifen is a nonsteroidal ER antagonist/agonist widely used for palliative and adjuvant treatment of breast cancer. Preliminary clinical studies demonstrated the efficacy of tamoxifen in the treatment of advanced breast cancer (235,236). It was originally believed that the antiestrogenic effect of tamoxifen would lead to bone loss, but subsequent work demonstrated that tamoxifen did not appear to have an antiestrogenic effect on bone (237). The effect of tamoxifen on bone depends on the menopausal stares of the woman. In postmenopausal women, 20 to 40 mg of tamoxifen administered daily led to bone preservation or increased bone density of between 0.61% to 1.71% at 2 years (238,239). In premenopausal women, slight bone loss was shown in one study (240). The bone-preserving effects of long-term tamoxifen treatment (20 mg daily) are likely to be only modest at best. At the end of a 5-year study, tamoxifen-treated patients had a bone density 2% above baseline, whereas untreated patients had a 2.7% loss compared with baseline (241). Another study found no increase in BMD in postmenopausal women treated with 40 mg tamoxifen for
CHAPTER28 Interventions for Osteoporosis 7 years in comparison with the baseline (242). Despite being a bone-preserving agent in one observational study, tamoxifen does not appear to offer protection against and may even increase the risk of fracture (243). Unfortunately, in addition to general side effects such as flushing, tamoxifen's usefulness as a postmenopausal treatment for preserving bone density would be severely limited by its estrogen agonist effect on the endometrium, where increased incidences of hyperplasia and carcinoma have been observed. Another SERM, raloxifene, was also investigated as a potential therapy for breast cancer. In vitro studies demonstrated the ability of raloxifene to prevent estrogen binding to the ER. It exhibited anti-tumor activity in carcinogeninduced rodents (244) and also in estradiol-dependent proliferation of MCF-7 breast cancer cells (245). However, it demonstrated a lack of anti-tumor effect in a small trial in postmenopausal women with breast cancer resistant to tamoxifen (246). Through a number of animal studies, it became evident that raloxifene has a SERM profile distinct from that of tamoxifen. Most animal studies were performed on a rat model. Raloxifene preserved both bone tissue in ovariectomized rats and the biomechanical properties of the bone (load to fracture ratio in the vertebra and shear to failure ratio in the femoral neck) (247). Raloxifene is primarily an antiresorptive drug. Like estrogen, it reduces bone remodeling in estrogen-deficient early postmenopausal women and induces a positive calcium balance shift. In one study, 60 mg/day of raloxifene was compared with 0.625 mg/day CEE and cyclicallyadministered MPA 5 mg daily for 14 days per month. Raloxifene resulted in improved intestinal calcium absorption, decreased urinary calcium excretion, and reduced bone resorption (248). Delmas et al. (249) initially reported results from a clinical study, which was a prospective randomized study of 601 postmenopausal women. The study subjects were given 60 mg of raloxifene for 24 months. At this dose, raloxifene led to an increase of 2.0% to 2.4% of BMD in the spine and hip, with no effect on endometrial thickness. Raloxifene was well tolerated with no significant difference in the adverse events between groups. However, the withdrawal rate in the study was high at 25%, and when measured as absolute gain in BMD rather than gain over placebo, the increase in BMD was less dramatic. However, it may be that reduction in bone turnover, rather than simple increase in density, is important for fracture prevention by reducing the likelihood of trabecular perforation. The Multiple Outcomes of Raloxifene Evaluation (MORE) trial was a multicenter randomized placebo-controlled study involving 7705 women with osteoporosis as diagnosed by BMD criteria. The data after 3 years have shown that raloxifene reduces the incidence of new vertebral fractures, particularly in women with pre-existing established osteoporosis. However, no significant reduction was seen in
367 nonvertebral fractures (250). A large randomized clinical trial of over 10,000 women has now shown conclusively that raloxifene does not prevent hip fractures (250a). Raloxifene appears to be well tolerated. Compared with placebo, women taking raloxifene experience more hot flushes and muscle cramps. The incidence of venous thromboembolism is increased twofold to threefold, similar to HRT-treated patients, but importantly it has no effect on the endometrium, acting as an antiestrogen on this tissue. The development of raloxifene and other SERMs has been a major recent advance in the prevention of postmenopausal osteoporosis and holds much promise for the future.
A. Phytoestrogens and Isoflavones Isoflavones are naturally occurring plant chemicals that are currently receiving attention as potential alternative therapies for a range of hormone-dependent conditions including cancer, coronary heart disease, osteoporosis, and the menopause. Isoflavones are present in legumes, but red clover and soya products contain the highest concentrations among foods thus far analyzed. There are plausible mechanisms to explain the potential health benefits of isoflavones. Because dietary isoflavones are structurally similar to estradiol but have 100 to 1000 times weaker estrogen activity, they have the capacity to bind to the ER and exert partial estrogen agonist/antagonist effects, which may be tissue specific. At certain concentrations, which may depend on many factors including receptor numbers, extent of protein mimics, and initiating estrogen-like actions, isoflavones may antagonize and inhibit estrogen action. This tissue-selective estrogen action is currently driving pharmacologists to develop SERMs. If dietary isoflavones can be shown to exert tissue-selective effects, these compounds would therefore have the potential to act as estrogen agonists, which may prove beneficial in postmenopausal women with respect to risk factors for heart disease, menopausal symptoms, and osteoporosis. However, under some conditions, these compounds may act as antiestrogens, which may assist in preventing the development of breast cancer. The main isoflavone compounds are biochanin, formononetin, daidzein, genistein, and equol. When ingested, these compounds are metabolized and absorbed and, like estrogens, undergo first-pass metabolism in the enterohepatic circulation. Phytoestrogens are effective in decreasing the severity of hot flushes in postmenopausal women. Dietary supplementation with 45 grams (g) of soya or wheat flower for 12 weeks significantly reduced the number and the severity of hot flushes, the reduction being more pronounced in the soya group (251). A larger randomized controlled trial investigated the effect of 60 g of isolated soy protein or 60 g
368 of placebo (casein) daily in 104 postmenopausal women (252). The 60 g of isolated soy protein contained 40 g of proteins and 76 mg of isoflavones. The 60 g of casein contained 40 g of protein but no isoflavones. At the end of the 12-week study period, there was a 45% reduction in the number of hot flushes in the soy group and a 30% reduction in the placebo group, which reached statistical significance. Unfortunately, there was a high withdrawal rate, largely due to gastrointestinal side effects such as constipation, bloating, nausea, and vomiting. Soya flour supplementation (45 g daily) or linseed (25 g daily) but not red clover sprout (10 g dry seed daily) given for 6 weeks to postmenopausal women led to vaginal cell maturation with an increase in the maturation index (253). These studies do suggest that phytoestrogens may be beneficial to the health of postmenopausal women. Ipriflavone, a synthetic isoflavone, has been shown to inhibit bone resorption both in vitro and in vivo. It is possible that the effect of ipriflavone on the bone may be due to the inhibition of recruitment and/or differentiation of preosteoclasts (254). Clinical randomized controlled studies evaluating the effect of ipriflavone on bone have been reported. Ipriflavone given to early postmenopausal women aged between 40 and 49 years led to a 1.1% improvement in the bone density of the lumbar spine after 12 months of treatment and a 1.2% improvement after 24 months. There was a 1.8% bone loss in the placebo group after 12 months and a 3.7% loss after 24 months (255). No changes in hot flush frequency, vaginal cytology, or osteocalcin level were observed, and there was a reduction in urinary hydroxyproline/creatinine ratio in the group receiving ipriflavone. A larger study invoMng 193 postmenopausal women between 50 and 65 years of age, with lumbar BMD less than 1 SD below the mean for agedmatched healthy young women (Z-score), was conducted and women were randomized to 600 mg ipriflavone/day or placebo and followed for 24 months. Lumbar BMD, serum skeletal alkaline phosphatase, serum osteocalcin, urinary excretion of calcium/creatinine, and hydroxyproline/creatinine ratio were measured every 6 months. BMD was measured with DEXA scanning with a coefficient of variation of 0.5% using phantom and <1% in vivo. There was a 1% increase in the lumbar BMD in the ipriflavone group and 0.7% decrease in the placebo group. The differences were statistically significant but should be interpreted with caution, bearing in mind the coefficient of variation of the equipment (256). Ipriflavone has been found to be effective in the treatment of established postmenopausal osteoporosis. Given in the usual dosage of 600 mg/day for 2 years, it led to improvement in the radial BMD by 4% (257). In all the studies, ipriflavone was well tolerated. The most common side effects were gastrointestinal, followed by skin reactions and central nervous system symptoms such as headache, depression, and drowsiness. There is currently much research interest in phytoestrogens, probably prompted by
KEATING AND STEVENSON
consumer interest in more "natural" alternatives to conventional menopausal treatments.
B. S t r o n t i u m R a n e l a t e It follows from the preceding pages that, with the exception of PTH, most current therapies for the treatment of postmenopausal osteoporosis involve inhibition of bone resorption as the chief mechanism of action. For patients with previous osteoporotic fractures, treatment with an anabolic agent that enhances bone formation would clearly be a preferable option, and this is now possible since the introduction of 1-34 recombinant (PTH), as discussed earlier in this chapter. However, the use of P T H is limited by its high cost and administration route (subcutaneous self-injection). Recently, strontium ranelate has been introduced for the treatment of osteoporosis and although not yet fully understood, it seems to have a dual effect with both inhibition of bone resorption and stimulation of bone formation (258). Strontium ranelate is composed of two atoms of stable strontium combined with organic ranelic acid; the latter acts as a carrier that makes the treatment palatable, and the former is the active component with regard to the skeleton. Trace amounts of strontium are present in bone, with about 100 gg in every gram of bone, and the absorption of strontium in the gut, its incorporation in bone, and elimination from the body through the kidneys is similar to that of calcium (259). Initially, strontium is adsorbed onto the surface of hydroxyapatite crystals, but in the longer term, some will exchange with calcium in the bone mineral and thus remain bound in the skeleton for years, as confirmed by radioactive strontium studies showing stable strontium still present in the skeleton 1 decade after 3 years of treatment (259). Strontium not incorporated into bone is excreted through the kidneys and feces. In an early 2-year, double-blind, placebo-controlled trial of strontium ranelate in postmenopausal women with vertebral osteoporotic fractures, strontium ranelate (500 rag, I g/day or 2 g/day) or placebo was given to 353 Caucasian women of mean age 66 years. All patients were also given a daily supplement of calcium (500 mg) and vitamin D2 (800 IU). A statistically significant mean annual increase in lumbar-adjusted BMD of 2.97% was observed in the 2-g strontium ranelate group, compared with placebo, and in the second year of treatment, the 2-g dose was associated with a 44% reduction in the number of patients experiencing a new vertebral deformity. Bone histomorphometry showed no mineralization defects, and there was no difference in withdrawal due to adverse effects compared with the placebo group (260). Further multinational, randomized, double-blind and placebo-controlled trials using the 2-g strontium dose derived from the previous study went on to assess the antifracture efficacy of strontium in the treatment of severe
369
CHAPTER 28 Interventions for Osteoporosis postmenopausal osteoporosis. These large studies were designed to assess the effect of strontium on risk of vertebral fracture (Spinal Osteoporosis Therapeutic Intervention [SOTI]) and peripheral (nonspinal) fracture (Treatment of Peripheral Osteoporosis [TROPOS]). The study duration was 5 years, with main statistical analysis planned after 3 years. All patients included in these two studies had previously participated in a normalization of calcium and vitamin D study called F I R S T ( F r a c t u r e International Run-in Strontium Ranelate Trials) and therefore throughout the studies, the patients received calcium/vitamin D supplements that were individually adapted according to their deficiencies (500 mg or 1000 mg of calcium, and 400 IU or 800 IU of vitamin D3) (261). The mean age of participants in the SOTI and T R O P O S trials was 69 and 77 years, respectively. The primary analysis of the SOTI study, evaluating the effect of 2 g of strontium ranelate on vertebral fracture rates, revealed a 41% reduction in RR of experiencing a first new vertebral fracture compared with placebo (RR 0.59; 95% confidence interval [CI], 0.48-0.73) in the intent-to-treat population. This antifracture efficacy of strontium ranelate was demonstrated from the first year, with a 49% reduction in RR of experiencing a first new fracture with strontium ranelate compared with placebo (RR 0.51, 95% CI, 0.360.74). The lumbar B M D increased by 14.4% in the treated group compared with the placebo group at 3 years. Strontium ranelate was well tolerated, without any specific adverse event, and no deleterious effects were observed on rates of nonvertebral fractures (262). The findings of the T R O P O S study have been equally impressive and in the entire sample, RR was reduced by 16% for all nonvertebral fractures (P < 0.04) and by 19% for major fragility fractures (hip, wrist, pelvis and sacrum, ribs and sternum, clavicle, humerus) (P < 0.031) in strontium ranelate-treated patients in comparison with the placebo group. Among women at high risk of hip fracture (age > 7 4 years and femoral neck B M D T-score < - 3 ) , the RR reduction for hip fracture was 36% (P < 0.046). Strontium ranelate increased B M D throughout the study, reaching at 3 years (P < 0.001) + 8.2% (femoral neck) and +9.8% (total hip). Incidence of adverse events was similar in both groups. It was concluded that strontium ranelate significantly reduced the risk of all nonvertebral and, in a high-risk subgroup, hip fractures over a 3-year period (263). Certainly the B M D improvements in vertebral and nonvertebral sites with strontium ranelate were impressive. However, caution is required in interpreting these data because much of the observed effect can be explained by the higher atomic number of strontium compared with calcium. In an editorial commentary in the British Medical Journal (264), it was explained that when B M D is measured by bone densitometry, atom for atom strontium attenuates x-rays more strongly than calcium. As a result, a 1% molar fraction of strontium causes a 10% overestimation of BMD. A sub-
sequent subset analysis of patients in the SOTI trial was performed using bone biopsy to assess bone strontium content, and the data for B M D was corrected on the basis of the correlation between the bone strontium content measured in the biopsy samples and concentrations of strontium in the blood. After the adjustment for bone strontium, the measured effect of treatment on B M D in the spine of 14.4% was corrected to 8.1%. Clearly, care is needed when predicting fracture efficacy solely with changes in BMD, although monitoring of B M D may be a useful clinical tool to gauge and encourage compliance with treatment with strontium ranelate due to the impressive B M D increases it typically produces.
References 1. van StaaTP, Dennison EM, Leufkens HG, Cooper C. Epidemiology of fractures in England and Wales. Bone 2001;29:517-522. 2. Keene GS, Parker MJ, Pryor GA. Mortality and morbidity after hip fractures. BMJ 1993;307:1248-1250. 3. Incidence of vertebral fracture in Europe: results from the European Prospective Osteoporosis Study (EPOS). J Bone Miner Res 2002;17: 716-724. 4. Holt G, Khaw KT, Reid DM, et al. Prevalence of osteoporotic bone mineral densityat the hip in Britain differs substantiallyfrom the US over 50 years of age: implicationsfor clinical densitometry.BrJRadiol 2002;75:736-742. 5. Cummings SR, Melton LJ. Epidemiologyand outcomes of osteoporotic fractures. Lancet 2002;359:1761-1767. 6. Todd CJ, Freeman CJ, Camilleri-Ferrante C, et al. Differences in mortality after fracture of hip: the east Anglian audit. BMJ 1995;310: 904-908. 7. Melton LJI. Epidemiology of fractures. In: Riggs BL, Melton LJI, eds. Osteoporosis."etiology, diagnosisand management. New York: Raven Press, 1988:133-154. 8. Burge RT. The cost of osteoporotic fractures in the UK. Projections for 2000-2020.JMed Econ 2001;4:51-62. 9. Consensus Development Conference. Prophylaxis and treatment of osteoporosis. Am J Med 1991;90:107-110. 10. World Health Organization. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. WHO Technical Report Series. Geneva:WHO, 1994. 11. Hui SL, Slemenda CW, Johnston CC. Age and bone mass as predictors of fractures in a prospective study. J Clin Invest 1988;81: 1804-1809. 12. Melton LJ, Atkinson EJ, O'Fallon WM, Wahner HW, Riggs BL. Long-term fracture prediction by bone mineral assessed at different skeletal sites.J Bone Miner Res 1993;8:1227-1233. 13. KanisJA, Melton LJ III, Christiansen C, Johnston CC, KhaltaevN. Perspective. The diagnosis of osteoporosis. J Bone Miner Res 1994;9:1137-1141. 14. Melton LJ III. Epidemiologyof fractures. In: Riggs BL, Melton LJ III. Osteoporosis."Etiology, diagnosis and management, ed 2. Philadelphia: Lippincott-Raven, 1995:225-247. 15. Chrischilles EA, Butler CD, Davis CS, Wallace RB. A model of lifetime osteoporosis impact.Arch Intern Med 1991;151:2026-2032. 16. SingerBR, McLauchlan GJ, Robinson CM, ChristieJ. Epidemiology of fractures in 15,000 adults: the influence of age and gender.J Bone Joint Surg Br 1998;80B:243-248.
370 17. Kanis JA, Johnell O, Oden A, De Laet C, Malstrom D. Epidemiology of osteoporosis and fracture in men. CalcifTissue Int 2004;75:90-99. 18. Smith RW, Risek J. Epidemiological studies of osteoporosis in women of Puerto Rico and southwestern Michigan. Clin Orthop 1962;45:31-48. 19. Nordin BEC, Crilly RG, Smith DA. Osteoporosis. In: Nordin B, ed. Metabolic bone and stone disease. Edinburgh: Churchill Livingstone, 1984:1-70. 20. Alffram PS, Bauer G. Epidemiology of fractures of the forearm. J Bone Joint Surg Am 1962;44A:105-114. 21. Holbrook TL, Grazier K, Kelsey JL, Stauffer RN. The frequency of occurrence, impact, and cost of selected musculoskeletal conditions in the United States. Chicago: American Academy of Orthopedic Surgeons, 1984. 22. Ralston SH. Osteoporosis (clinical review). Br Med J 1997;315: 469-472. 23. Slemenda CW, Christian JC, Williams CJ, Johnston CC Jr. The changing relative importance of genetics and environment in adult women. In: Cohn DV, Glorieux FH, Martin TJ, eds. Calcium regulation and bone metabolism. Amsterdam: Elsevier Scientific, 1990:491-496. 24. Chesnut CH III. Osteoporosis: a worldwide problem? In: Christiansen C, Overgaard K, Eds. Osteoporosis1990, vol 1. Copenhagen: Osteopress ApS, 1990:33-35. 25. Melton LJ III. Epidemiology of fractures in North America. In: Christiansen C, Overgaard K, Eds. Osteoporosis1990, vol 1. Copenhagen: Osteopress ApS, 1990:36-41. 26. Solomon L. Bone density in aging Caucasian and African populations. Lancet 1979;3:1326-1329. 27. Lo CW, Jarjou LM, Poppitt S, et al. Delayed development of bone mass in West African adolescents. In: Christiansen C, Overgaard K, eds. Osteoporosis 1990, vol 1. Copenhagen: Osteopress ApS, 1990:73-77. 28. Stevenson JC, Lees B, Devenport M, Cust MP, Ganger KF. Determinants of bone density in normal women: risk factors for future osteoporosis? Br MedJ 1989;298:924-928. 29. Compston J. The pathogenesis of osteoporosis. In: Arden NK, Spector TD, eds. Osteoporosisillustrated. London: Current Medical Literature, 1997:21. 30. Mundy GR. Bone remodelling and mechanisms of bone loss in osteoporosis. In: Meunier PJ, ed. Osteoporosis diagnosis and management. London: Martin Dunitz, 1998:17-35. 31. Mbalaviele G, Chen H, Boyce BF et al. The role of cadherin in the generation of multi-nucleated osteoclasts from mononuclear precursors in murine marrow.J Clin Invest 1995;95:2757-2765. 32. Mbalaviele G, Nishimura R, Feng J, et al. Molecular cloning of a novel cadherin in human osteoclasts (abstract). J Bone Miner Res 1995;10:10. 33. Horton MA, Davies J. Perspectives: adhesion receptors in bone. J Bone Miner Res 1989;4:803-807. 34. Parfitt AM. Bone remodelling in the pathogenesis of osteoporosis. Med Times 1981;109:80-92. 35. Mundy GR, Rodan SB, Majeska RJ, et al. Unidirectional migration of osteosarcoma cells with osteoblastic characteristics in response to products of bone resorption. CalcifTissue Int 1982;34:542-546. 36. Mundy GR, Poser JW. Chemotactic activity of the gamma-carboxyglutamic acid containing protein in bone. CakifTissue Int 1983;35: 164-168. 37. Pfeilschifter J, Wolf O, Naumann A, et al. Chemotactic response of osteoblast like cells to transforming growth factor B. J Bone Miner Res 1990;5:825-830. 38. Manolagas SC, Jilka RL. Bone marrow, cytokines, and bone remodelling: emerging insights into the pathophysiology of osteoporosis. New EnglJ Med 1995;332:305-310. 39. Kurihara N, Civin C, Roodman GD. Osteotropic factor responsiveness of highly purified populations of early and late precursors for human multinucleated cells expressing the osteoclast phenotype. J Bone Miner Res 1991;6:257-261.
KEATING AND STEVENSON
40. Ohsaki u Takahashi S, Scarcez T, et al. Evidence for an autocrine/ paracrine role for interleukin-6 in bone resorption by giant cells from giant cell tumours of bone. Endocrinology 1992;131:2229-2234. 41. Albright F, Smith PH, Richardson AM. Postmenopausal osteoporosis.JAm Med_dssoc 1941;116:2465-2474. 42. Albright F, Bloomberg E, Smith PH. Postmenopausal osteoporosis. Trans AssocAm Physicians 1940;55:305. 43. Heaney RP. A unified concept of osteoporosis Am J Med 1965; 39:877. 44. Lindsay R. Estrogen deficiency. In: Riggs BL, ed. Osteoporosis:etiology, diagnosis and management. New York: Raven, 1995:133-160. 45. Lindsay R. Estrogen and osteoporosis. In: Stevenson JC, Lindsay R, eds. Osteoporosis. London: Chapman and Hall, 1998:161-171. 46. Parfitt AM. Bone forming cells in clinical conditions. In: Hall BK, ed. The osteoblast and osteocyte. Bone, vol 1. Boca Raton, FL.: Telford/ CRC, 1990:351-429. 47. Heaney RP. Menopausal effects on calcium homeostasis and skeletal metabolism. In: Ryan KJ, Gibson DC, eds. Menopause and aging. Bethesda, MD: Government Printing Office, DHEW Publication (NIH) 73-319, 1973. 48. Riggs BL, Jowsey j, Kelly PJ, Jones JD, Maher FT. Effects of sex hormones on bone in primary osteoporosis.J Clin Invest 1969;48:1065. 49. Recker RR, Saville PD, Heaney RP. Effects of estrogens and calcium carbonate on bone loss in postmenopausal women. Ann Intern Med 1977;87:649. 50. Heaney RP, Recker RR, Saville PD. Menopausal changes in bone remodelling. J Lab Clin Med 1978;92:964-970. 51. Eriksen EF, Colvard DS, Berg NJ, et al. Evidence of estrogen receptors in normal human osteoblast-like cells. Science 1988;241:84-86. 52. Komm BS, Terpening CM, Benz DJ, et al. Estrogen binding, receptor mRNA, and biologic response in osteoblast-like osteosarcoma cells. Science 1988;241:81-84. 53. Rizzoli R, Bonjour J-P. Hormones and bones. Lancet 1997;349: sI20-sI23. 54. Girasole G, Jilka RL, Passeri G, et al. 17B estradiol inhibits interleukin-6 production by bone marrow-derived stromal cells and osteoblasts in vitro: a potential mechanism for the antiresorptive effect of estrogens. J Clin Invest 1992;89:883-891. 55. Felson DT, Zhang u Hannan MT, et al. The effect of postmenopausal estrogen therapy on bone density in elderly women. N EnglJ Med 1993;329:1141-1146. 56. Kiel DP, Felson DT, Anderson JJ, et al. Hip fracture and the use of estrogens in postmenopausal women. The Framingham Study. NEngl JMed 1987;317:1169-1174. 57. Lafferty FW, Fiske ME. Postmenopausal estrogen replacement: a long-term cohort study. Am J Med 1994;97:66-77. 58. Weiss NS, Ure CL, Ballard JH, et al. Decreased risk of fractures of the hip and lower forearm with postmenopausal use of estrogen. NEnglJ Med 1980;303:1195-1198. 59. CauleyJA, Seeley DG, Ensrud K, et al. Estrogen replacement therapy and fractures in older women. Ann Intern Med 1995;122:9-16. 60. Paganini-Hill A, Ross RK, Gerkins VR, et al. Menopausal estrogen therapy and hip fractures. Ann Intern Med 1982;95:28-31. 61. Ettinger B, Genant HK, Cann CE. Long-term estrogen therapy prevents bone loss and fractures. Ann Intern Med 1985;102:319-324. 62. Hutchinson TA, Polansky SM, Feinstein AR. Postmenopausal oestrogens protect against fractures of hip and distal radius. Lancet 1979;ii:705-709. 63. Hillard TC, Whitcroft SJ, Marsh MS, et al. Long-term effects of transdermal and oral hormone replacement therapy on postmenopausal bone loss. OsteoporosInt 1994;4:341-348. 64. Lindsay R, Tohme JF. Estrogen treatment of patients with established postmenopausal osteoporosis. Obstet Gyneco11990;76:290-295. 65. Prince RL, Smith M, Dick IM, et al. Prevention of postmenopausal osteoporosis: a comparative study of exercise, calcium supplementation, and hormone replacement therapy. N EnglJ Med 1991;325: 1189-1195.
CHAPTER 28 Interventions for Osteoporosis 66. Lindsay R, Hart DM, Aitken JM, et al. Long-term prevention of postmenopausal osteoporosis by oestrogen: evidence for increased bone mass after onset of oestrogen treatment. Lancet 1976;i:1038-1041. 67. Nachtigall LE, Nachtigall RH, Nachtigall RD, et al. Estrogen replacement therapy. I. A 10-year prospective study in relationship to osteoporosis. Obstet Gyneco11979;53:277-281. 68. Lufkin EG, Wahner HW, O'Fallon WM, et al. Treatment of postmenopausal osteoporosis with transdermal estrogen. Ann Intern IVied 1992;117:1-9. 69. M-Azzawi F, Hart DM, Lindsay R. Long-term effects of oestrogen replacement therapy on bone mass as measured by dual photon absorptiometry. Br MedJ 1987;294:1261-1262. 70. Lees B, Stevenson JC. The prevention of osteoporosis using sequential low-dose hormone replacement therapy with estradiol-1713 and dydrogesterone. OsteoporosInt 2001;12:251-258. 71. Lindsay R, Gallagher JC, Kleerekoper M, Pickar JH. Effect of lower doses of conjugated equine estrogens with and without medroxyprogesterone acetate on bone in early postmenopausal women. JAm IVied Assoc 2002;287:2668-2676. 72. Ettinger B, Ensrud KE, Wallace R, et al. Effects of ultralow-dose transdermal estradiol on bone mineral density: a randomized clinical trial. Obstet Gyneco12004;104:443-451. 73. CauleyJA, Robbins J, Chen Z, et al. Effects of estrogen plus progestin on risk of fracture and bone mineral density. The Women's Health Initiative randomized trial. JAm Med dssoc 2003;290:1729-1738. 74. Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy. JAm Meddssoc 2004;291:1701-1712. 75. Lamy O, Kreig MA, Burckhardt B, Wasserfallen JB. An economic analysis of hormone replacement therapy for the prevention of fracture in young postmenopausal women. Expert ODin Pbarmacother 2003;4:1479-1488. 76. Holland EFN, Leather AT, Studd JWW. The effect of a new sequential oestradiol valerate and levonorgestrel preparation on the bone density of postmenopausal women. Br J Obstet Gynaecol 1993;100: 966-967. 77. Lees B, Pugh M, Siddle N, et al. Changes in bone density in women starting hormone replacement therapy compared to those in women already established on hormone replacement therapy. Osteoporos Int 1995;5:344-348. 78. Riis BJ, Thomsen K, Strum V, et al. The effects of percutaneous estradiol and natural progesterone on postmenopausal bone loss. Am J Obstet Gyneco11987;156:61-65. 79. Stevenson JC, Cust MP, Ganger KF, et al. Effects of transdermal versus oral hormone replacement therapy on bone density in spine and proximal femur in postmenopausal women. Lancet 1990;335: 265-269. 80. Studd J w w , Holland EFN, Leather AT, et al. The dose-response of percutaneous oestradiol implants on the skeletons of postmenopausal women. BrJ Obstet Gynaeco11994;101:787-791. 81. Ryde SJS, Bowen-Simpkins K, Bowen-Simpkins P, et al. The effect of oestradiol implants on regional and total body mass: a three-year longitudinal study. Clin Endocrino11994;40:33-38. 82. Naess6n T, Persson I, Thor L, et el. Maintained bone density at advanced ages after long term treatment with low dose oestradiol implants. BrJ Obstet Gynaeco11993;100:454-459. 83. Persson I, Adam HO, Bergkvist L, et al. Risk of endometrial cancer after treatment with oestrogens alone or in adjunction with progestogens: results of a prospective study. Br MedJ 1989;28:147-151. 84. Christiansen C, Riis BJ. 17-[3 oestradiol and continuous norethisterone: an unique treatment for established osteoporosis in elderly women. J Clin Endocrinol Metab 1990;71:836-841. 85. Nielsen SP, B~irenholdt O, Hermansen F, et al. Magnitude and pattern of skeletal response to long term continuous and cyclical sequential oestrogen/progestin treatment. Br J Obstet Gynaecol 1994;101: 319-324.
371 86. Lindsay R, Hart DM, Purdie P, et al. Comparative effects of oestrogen and a progestogen on bone loss in postmenopausal women. Clin Sci Mol IVied 1978;54:193-195. 87. Bagger YZ, Tank6 LB, Alexandersen P, et al. Two to three years of hormone replacement treatment in healthy women have long-term preventive effects on bone mass and osteoporotic fractures: the PERF study. Bone 2004;34:728-735. 88. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. JAm MedAssoc 2002;288:321-333. 89. Million Women Study Collaborators. Breast cancer and hormonereplacement therapy in the Million Women Study. Lancet 2003;362: 419-427. 90. Stevenson JC on behalf of the International Consensus Group on HRT and Regulatory Issues. HRT, osteoporosis and regulatory authorities. O~is custodiet ipsos custodes? Human Reprod 2006;21:1668-1671. 91. Kanis JA, Johnell O, Oden A, et al. Ten year probabilities of osteoporotic fractures according to BMD and diagnostic thresholds. Osteoporos Int 2001;12:989-995. 92. Francis MD, Fogelman I. 99mTc diphosphonate mechanism on bone. In: Fogelman I, ed. Bone uptake scanning in clinicalpractice. London: Springer-Verlag, 1987:7-17. 93. Francis MD. The inhibition of calcium hydroxyapatite crystal growth by polyphosphonates and polyphosphates. Calcif Tissue Res 1969; 3:151-162. 94. Reynolds JJ, Minkin C, Morgan DB, Spycher D, Fleisch H. The effect of two diphosphonates on the resorption of mouse calvaria in vitro. Calcif Tissue Res 1972;10:302-313. 95. Fleisch H. Bisphosphonates: pharmacology and use in treatment of tumour-induced hypercalcaemic and metastatic bone disease. Drugs 1991;42:919-944. 96. Adami S, Zamberlan N, Mian M, et al. Duration of the effects of intravenous alendronate in postmenopausal women and in patients with primary hyperparathyroidism. Bone Miner 1994;5:75-82. 97. Nixon GA, Buehler EV, Newmann EA. Preliminary safety assessment of disodium etidronate as an additive to experimental oral hygiene products. Toxicol/lppl Pbarmaco11972;22:661-671. 98. Nolen GA, Buehler EV. The efffect of disodium etidronate on the reproductive functions and embryogeny of albino rats and New Zealand rabbits. ToxicolAppl Pbarmaco11971;18:548-561. 99. Gralpel P, Bentley P, Fritz H, et al. Reproduction toxicity studies with pamidronate. #Irzneimittel Forschung 1992;42:654-667. 100. Minsken DH, Manson JM, Chennekatu PE Effects of the biphosphonate, alendronate, on parturition in rats. Toxicol_/lppl Pharmacol 1993;121:217-233. 101. Masarachia P, Weinreb M, Balena R, Rodan GA. Comparison of the distribution of 3H-alendronate and 3H-etidronate in rat and mouse bones. Bone 1996;19:281-290. 102. Flanagan AM, Chambers TJ. Dichloromethilene bisphosphonate (C12MBP) inhibits bone resorption through injury to osteoclasts that resorb C12MBP-coated bone. Bone Miner 1989;6:33-43. 103. Hughes DE, Wright KR, Vy HL, et al. Bisphosphonates promote apoptosis in murine osteoclasts in vitro and in vivo.J Bone Miner Res 1995; 10:1478-1487. 104. Sat M, Grasse W, Endo M, et al. Bisphosphonate action. Alendronate localization in rat bone and effects on osteoclast ultrastructure. J Clin Invest 1991;88:2095-2105. 105. Hughes DE, MacDonald BR, Rissell RGG, Gonan M. Inhibition of osteoclast-like cell formation by bisphosphonates in long-term cultures of human bone marrow.J Clin Invest 1989;83:1930-1935. 106. Murakami H, Takahashi N, Sasald T, et al. A possible mechanism of the specific action of bisphosphonates on osteoclasts: tiludronate preferentially affects polarised osteoclasts having ruffled borders. Bone 1995;17:137-144.
372 107. Skovey IA, Ly HD, Kelly J, et al. How does alendronate inhibit protein-tyrosine phosphotases? J Bid Chem 1997;272:2472-2480. 108. Sahni M, Guenther H, Fleisch H, et al. Bisphosphonates act on rat bone resorption through the mediation of osteoblasts. J Clin Invest 1993;91:2004-2011. 109. Vitte C, Fleisch H, Guenther H. Bisphosphonates induce osteoblasts to secrete an inhibitor of osteoclast-mediated resorption. Endocrinology 1996;137:2324-2333. 110. Pouillies JM, Tremollieres F, R oux C, et al. Cyclical etiodronate for the prevention of early postmenopausal bone loss. Osteopor Int 1996;6:242-246. 111. PouilliesJM, Tremollieres F, Roux C, et al. Effects of cyclical etidronate therapy on bone loss in early postmenopausal women who are undergoing hormonal replacement therapy. OsteoporosInt 1997;213-218. 112. Herd RJM, Balena R, Blake GM, Ryan PJ, Fogelman I. The prevention of early postmenopausal bone loss by cyclical etidronate therapy: a 2-year, double-blind, placebo-controlled study. Am J Med 1997;103: 92-99. 113. Meunier PJ, Confravreux E, Tupinon I, et al. Prevention of early postmenopausal bone loss with cyclincal etidronate therapy (a doubleblind, placebo-controlled study and one-year follow-up).J Clin Endocrinol Metab 1997;82:2784-2791. 114. Livermay VA, Weiss SR, Broil J, et al. Effect of oral alendronate on bone mineral density and the incidence of fracture in postmenopausal osteoporosis. N EnglJ Med 1995;333:1437-1443. 115. Black DM, Cummings SR, Karpf DB, et al. Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet 1996;348:1535-1541. 116. Karpf DB, Shapiro DR, Seeman E, et al. Prevention of non-vertebral fractures by alendronate. A meta-analysis. JAMA 1997;277:1154-1164. 117. Hosking D, Chilvers C, Christiansen C, et al. Prevention of bone loss with Alendronate in postmenopausal women under 60 years of age. N EnglJ Med 1998;338:485-492. 118. McClung MR, Geusens P, Miller PD, et al. Hip intervention program study group: effect of risedronate on the risk of hip fracture in elderly women. N EnglJ Med 2001;344:333-340. 119. Reginster JY, Minne HW, Sorensen OH, et al. Randomized trial of the effects of risedronate on vertebral fractures in women with established postmenopausal osteoporosis. Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. OsteoporosInt 2000;11:83-91. 120. Woo S-B, Hellstein JW, Kalmar JR. Systematic review: bisphosphonates and osteonecrosis of the jaws. Ann Int Med 2006;144:753-761. 121. Chestnut CH III, Skag A, Christiansen C, et al. Effects of oral ibandronate administered daily or intermittently on fracture risk in postmenopausal osteoporosis. J Bone Miner Res 2004;19:1241-1249. 122. Kanis JA, Passmore R. Calcium supplement of the diet: 1. Br MedJ 1989;298:137-140. 123. Nordin CBE, Heaney RR Calcium supplementation in the diet: justified by present evidence. Br MedJ 1990;300:1056-1060. 124. Cadogan J, Eastell R, Jones N, Barker ME. Milk intake and bone mineral acquisition in adolescent girls: randomised, controlled intervention trial. Br MedJ 1997;315:1555-1560. 125. Nordin BEC, Peacock M. Calcium and bone metabolism. In: Horler AR, Foster JB, eds. Progress in clinical medicine. Edinburgh: Churchill Livingstone, 1983:287-315. 126. Hagenfeldt Y, Eriksson HA. The oestrogen receptor in the rat kidney: ontogeny, properties and effects of gonadoectomy on its concentration. J Steroid Biochem 1988;31:49-56. 127. Lau E, Donnan S, Barker DJP, Cooper S. Physical activity and calcium intake in fracture of the proximal femur in Hong Kong. Br Med J 1988;297:1441-1443. 128. Cooper S, Barker DJP, Wickham C. Physical activity, muscle strength and calcium intake in fracture of the proximal femur in Britain. Br MedJ 1998;297:1443-1446.
KEATING AND STEVENSON
129. Kanis JA, Johnell O, Gullberg B, et al. Evidence of efficacy of drugs affecting bone metabolism in preventing hip fracture. Br Med J 1992;305:1124-1128. 130. Dawson-Hughes B, Dallal GE, Kxall EA, et al. A controlled trial of the effect of calcium supplementation on bone density in postmenopausal women. NEnglJ Med 1990;323:878-883. 131. Prince RL, Smith M, Dick IM, et al. Prevention of postmenopausal osteoporosis. NEnglJMed 1991;325:1189-1195. 132. Reid IR, Ames RW, Evans MC, et al. Effect of calcium supplementation on bone loss in postmenopausal women. N Engl J Med 1993;328:460-464. 133. Riis B, Thomsen K, Christiansen C. Does calcium supplement prevent postmenopausal bone loss? NEnglJMed 1987;316:173-177. 134. Nicar MJ, Pak CYC. Calcium bioavailability from calcium carbonate and calcium citrate. J Clin Endocrinol Metab 1985;61:391-393. 135. Harvey JA, Zobitz MM, Pak CYC. Dose dependency of calcium absorption: a comparison of calcium carbonate and calcium citrate. J Bone Mineral Res 1988;3:253-258. 136. Reginster JY, Denis D, Bartsch V. Acute biochemical variations induced by four different calcium salts in healthy male volunteers. Osteopor Int 1993;3:271-275. 137. Hunt JN, Johnson C. Relationship between gastric secretion of acid and urinary excretion of calcium after oral supplements of calcium. Digest Dis Scien 1983;28:417-421. 138. Recker R. Calcium absorption and achlorhydria. N Engl J Med 1985;313:70-73. 139. Reichnel H, Koeffler HP, Norman AW. The role of the vitamin D endocrine system in health and diseases. NEnglJMed 1989;320:981. 140. Pike JW. Vitamin D3 receptors. Structure and function in transcription. Annu Rev Nutr 1991;11:189. 141. Morrison NA, Q[JC, Toldta A, et al. Prediction of bone density from vitamin D receptors alleles. Nature 1994;367:284-287. 142. Howard G, Nguyen T, Morrison N, et al. Genetic influence on bone density: physiological correlates of vitamin D receptor gene alleles in premenopausal women. J Clin Endocrinol Metabol 1995;80: 2800-2805. 143. Jorgensen HL, Scholler J, Sand JC, et al. Relation of common alMic variation at vitamin D receptor locus to bone mineral density and postmenopausal bone loss: cross-sectional and longitudinal population study. Br MedJ 1996;313:586-590. 144. Looney J, Yoon HK, Fischer M, et al. Lack of high prevalence of Bb vitamin D receptor genotype in severely osteoporotic women. J Clin Endocrinol Metabol 1995;80:2158-2162. 145. Holick MF, Shao Q M Liu WW, Chen TC. The vitamin D content of fortified milk and infant formula. N Engl J Med 1992;326: 1178-1181. 146. Lips P, Grafmans WC, Ooms ME, Bezemer PD, Bouter LM. Vitamin D supplementation and fracture incidence in elderly persons. A randomised, placebo-controlled clinical trial. Ann Intern Med 1996; 124:400-406. 147. Smith H, Anderson F, Raphael H, Crozier S, Cooper C. Effect of annual intramuscular vitamin D supplementation on fracture risk: population-based, randomised, double-blind, placebo-controlled trial. Osteoporos Int 2004:15:$8. 148. Grant AM, Avenell A, Campbell MK, et al. RECORD Trial Group. Oral vitamin D3 and calcium for secondary prevention of low-trauma fractures in elderly people (Randomised Evaluation of Calcium or Vitamin D, RECORD): a randomised placebo-controlled trial. Lancet 2005;365:1621-1628. 149. Trivedi DP, Doll R, Khaw KT. Effect of four monthly oral vitamin D3 (cholecalciferol) supplementation on fractures and mortality in men and women living in the community: randomised double-blind controlled trial. BMJ 2003;326:469-572.
CHAPTER 28 Interventions for Osteoporosis 150. Dawson-Hughes B, Harris SS, Krall EA, Dalla GE. Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. N E n g l J M e d 1997;337:670-676. 151. Chapuy MC, Arlot ME, Duboeuf F, et al. Vitamin D and calcium to prevent hip fractures in elderly women. N Eng! J Med 1992;327: 1637-1642. 152. Chapuy MC, Pamphile R, Paris E, et al. Combined calcium and vitamin D3 supplementation in elderly women: confirmation of reversal of secondary hyperparathyroidism and hip fracture risk: the Decalyos H study. OsteoparosInt 2002;13:257-264. 153. Larsen ER, Mosekilde L, Foldsprang A. Vitamin D and calcium supplementation prevents osteoporotic fractures in elderly communitydwelling residents: a pragmatic population-based 3-year intervention study. J Bone Miner Res 2004;19:370-378. 154. Porterhouse J, Cockayne S, King C, et al. Randomised controlled trial of supplementation with calcium and cholecalciferol (vitamin D3) for prevention of fractures in primary care. BMJ 2005;330:1003-1006. 155. Jackson RD, LaCroix AZ, Gass M, et al. Women's Health Initiative Investigators. Calcium plus vitamin D supplementation and the risk of fractures. N EnglJ Med 2006;354:669-683. 156. Sogaard CH, Mosekilde L, Thomsen JS, Richard A, McOsker JE. A comparison of the effects of two anabolic agents (fluoride and PTH) on ash mineral density and bone strength assessed in an osteopenic rat model. Bone 1997;20:439-449. 157. Ejersted C, Andreassen TT, Nilsson MHL, et al. Human parathyroid hormone (1-34) increases bone formation and strength of cortical bone in aged rats. EurJEndocrino! 1994;130:210-217. 158. Reeve J, Meunier pJ, Parsons JA, et al. Anabolic effect of human parathyroid hormone fragment on trabecular bone in involutional osteoporosis: a multicentre trial. Br MedJ 1990;I:1340-1344. 159. Finkelstein JS, Klibanski A, Schaefer EH, et al. Parathyroid hormone for the prevention of bone loss induced by oestrogen deficiency. N E n g l J M e d 1994;331:1618-1623. 160. Lindsay R, Nieves J, Formica C, et al. Randomised controlled study of effect of parathyroid hormone on vertebral bone mass and fracture incidence among postmenopausal women on oestrogen with osteoporosis. Lancet 1997;350:550-555. 161. Neer RM, Arnaud CD, Zanchetta JR, et al. Effect of parathyroid hormone (1-34) on fractures and bone mineral density in postmenopausal women with osteoporosis. N EnglJ Med 2001;344:1434-1441. 162. Tashjian AH Jr, Chabner BA. Commentary on clinical safety of recombinant human parathyroid hormone 1-34 in the treatment of osteoporosis in men and postmenopausal women. J Bone Miner Res 2002;17:1151-1161. 163. Marcus R, Wang O, Satterwhite J, Mitlak B. The skeletal response to teriparatide is largely independent of age, initial bone mineral density, and prevalent vertebral fractures in postmenopausal women with osteoporosis. J Bone Miner Res 2003;18:18-23. 163a. Jiang Y, Zhao JJ, Mitlak BH, et al. Recombinant human parathyroid hormone (1-34) [teriparatide] improves both cortical and cancellous bone structure.JBone Miner Res 2003;18:1932-1941. 164. Rich C, Enskinck J. Effect of sodium fluoride on calcium metabolism of human beings. Nature 1961;191:184-185. 165. Kleerecoper M. The role of fluoride in the prevention of osteoporosis. Endocrinol Metab Clin N d m 1998;27;441-452. 166. Riggs BL, Hogson SF, O'Fallam MW, et al. Effect of fluoride treatment on the fracture rate in postmenopausal women with osteoporosis. N E n g l J M e d 1990;352:802-809. 167. Pak CYC, Sakhaee K, Adams-Huet B, et al. Treatment ofpostmenopausal osteoporosis with slow-release sodium fluoride. Ann Intern ivied 1995;123:401-408. 168. Heaney RE Fluoride and osteoporosis. Ann Intern ivied 1994;120: 689-690. 169. Kleerekoper M. Osteoporosis and the primary care physician: time to bone-up. Ann Intern Med 1995;123:466-467.
373 170. Eastell R, Boyle IT, Compston J, et al. Management of male osteoporosis: report of the UK Consensus Group. QJM 1998;91:71-92. 171. Chambers TJ, Moore A. The sensitivity of isolated osteoclasts to morphological transformation by calcitonin.d C/in EndocrinolMetabol 1983;57:819-825. 172. Nicholson GC, Horton MA, Sexton PM, et al. Calcitonin receptors of human osteoclastoma. Harm Metab Res 1987;19:585-589. 173. Azria M, Copp DH, Zanelli JM. 25 years of salmon calcitonin: from synthesis to therapeutic use. CalcifTissue Int 1995;57:405-408. 174. Reginster JY, Gaspar S, Deroisy R, et al. Prevention of osteoporosis with nasal salmon calcitonin-effect of anti-salmon calcitonin antibody formation. OsteoporInt 1998;3:261-264. 175. Stevenson JC, Hillyard CJ, MacIntyre I, Cooper H, Whitehead MI. A physiological role for calcitonin: protection of the maternal skeleton. Lancet 1979; ii:769-770. 176. Hillyard CJ, Stevenson JC, MacIntyre I. Relative deficiency of plasma-calcitonin in normal women. Lancet 1978;i:961-962. 177. Taggart HM, Chesnut CH, Ivey JL, et al. Deficient calcitonin response to calcium stimulation in post-menopausal osteoporosis. Lancet 1982;1:475-478. 178. Stevenson JC, Allen PR, Abeyasekera G, Hill PA. Osteoporosis with hip fracture: changes in calcium regulating hormones. EurJ Clin Invest 1986;16:357-360. 179. Tiegs RD, Body JJ, Wahner HW, et al. Calcitonin secretion in postmenopausal osteoporosis. N EnglJ Med 1985;312:1097-1000. 180. MacIntyre I, Stevenson J, Whitehead M, et al. Calcitonin for prevention of postmenopausal bone loss. Lancet 1988;1:900-902. 181. Overgaard K, Riis BJ, Christiansen C, et al. Effect of calcitonin given intranasally on early postmenopausal bone loss. Br MedJ 1989;299: 477-479. 182. Ellerington MC, Hillard TC, Whitcroft SIJ, et al. Intranasal salmon calcitonin for the prevention and treatment of postmenopausal osteoporosis. Calcif Tissue Int 1996;59:6-11. 183. Overgaard K, Hansen MA, Jensen SB. Effects of salcatonin given intranasally on bone mass and fracture rates in established osteoporosis: a dose-response study. Br MedJ 1992;305:556-561. 184. Rico H, Hernandez ER, Revilla M, Gomez-Castresana E Salmon calcitonin reduces vertebral fracture rates in postmenopausal crush fracture syndrome. Bone Miner 1992;16:131-138. 185. Cardona JM, Pastor E. Calcitonin versus etidronate for the treatment of postmenopausal osteoporosis: a meta-analysis of published clinical trials. OsteoparosInt 1997;7:165-174. 186. Arnala I, Saastamonen J, Alhava EM. Salmon calcitonin in the prevention of bone loss in the perimenopause. Bone 1996;4:629-632. 187. Chestnut CH, Silverman S, Adriano K, et al. A randomized trial of nasal spray salmon calcitonin in postmenopausal women with established osteoporosis: the Prevent Recurrences of Osteoporotic Fractures study. Am J Med 2000;109:267-276. 188. Benedek-Jaszman LJ. Long-term placebo-controlled efficacy and safety study of Org OD 14 in climacteric women. Maturitas 1987;1: $25-$33. 189. Lippuner K, Haenggi W, Birkhaeuser MH et al. Prevention of postmenopausal bone loss using tibolone or conventional per oral or transdermal hormone replacement therapy with 17 beta estradiol and dydrogesterone. J Bone Miner Res 1997;12:806-812. 190. Lindsay R, Hart DM. Prospective double-blind trial of a synthetic steroid (Org OD14) for prevention of postmenopausal osteoporosis. Br MedJ 1980;1:1207-1209. 191. Lyritis GP, Karpathios S, Basdecis K, et al. Prevention ofpost-oophorectomy bone loss with Tibolone. Maturitas 1995;22:247-253. 192. Geusens P, Dequeker J, Gielen J, Schot LPC. Non-linear increase in vertebral density induced by a synthetic steroid (Org OD 14) in women with established osteoporosis. Maturitas 1991;13:155-162. 193. Kenemans P, SperoffL, Birkhauser M, et al. Tibolone: clinical recommendations and practical guidelines. A report of the International Tibolone Consensus Group. Maturitas 2005;51:21-28.
374 194. Cummings SR. LIFT study is discontinued. BMJ 2006;332:667. 195. Genazzani AR, Benedek-Jaszman LJ, Hart DM, et al. Org OD 14 and the endometrium. Maturitas 1991;13:243-251. 196. Von Dadelszen P, Gillmer MDG, Gray MD, et al. Endometrial hyperplasia and adenocarcinoma during Tibolone (Livial) therapy. BrJ Obstet Gynaeco11994;101:158-161. 197. Milner MH, Sinnott MM, Cooke TM, et al. A two-year study of lipid and lipoprotein changes in postmenopausal women with tibolone and estrogen-progestin. Obstet Gyneco11996;87:593-599. 198. Rymer J, Crook D, Sidhu M, Chapman M, Stevenson JC. Effects of tibolone on serum concentrations of lipoprotein (a) in postmenopausal women. Acta Endocrino11993;128:259-262. 199. Bjarnason NH, Bjarnason K, Haarbo J, et al. Tibolone: influence on markers of cardiovascular disease. J Clin Endocrinol Metab 1997;82: 1752-1756. 200. Zumoff B, Strain GW, Miller LK, Rosner W. Twenty-four-hour mean plasma testosterone concentration declines with age in normal premenopausal women. J Clin Endocrinol Metab 1995;80:1429-1430. 201. Meldrum DR, Davidson BJ, Tataryn IV, Judd HL. Changes in circulating steroids with aging in postmenopausal women. Obstet Gynaecol 1981;57:624-628. 202. Judd HL, Lucas WE, Yen SSC. Effect of oophorectomy on circulating testosterone and androstenedione levels in patients with endometrial cancer. Am J Obstet Gyneco11974;118:793-798. 203. Colvard CS, Ericsen EF, Keeting PE, et al. Identification of androgen receptors in normal human osteoblastic-like cells. Proc NatlAcad Sci 1989;86:854-857. 204. Kasperk CH, Wergedal JE, Farley JR, et al. Androgens directly stimulate proliferation of bone cells in vitro. Endocrinology 1989;124: 1576-1578. 205. Thorn MH, Studd JWW. Hormonal profile in postmenopausal women after therapy with subcutaneous implant. BrJ Obstet Gynaecol 1981;88:426-433. 206. Devogelaer JP, Crabbe J, De Deuxchaisnes CN. Bone mineral density in Addison's disease: evidence for an effect of adrenal androgens on bone mass. Br MedJ 1987;294:798-800. 207. Watts NB, Notelovitz M, Timmons MC, et al. Comparison of oral oestrogen and oestrogens plus androgen on bone mineral density, menopausal symptoms and lipid-lipoprotein profiles in surgical menopause. Obstet Gynaeco11995;85:529-537. 208. Hickok LR, Toomey C, Speroff L. A comparison of oesterified estrogens with and without methyltestosterone: effects on endometrial histology and serum lipoproteins in postmenopausal women. Obstet Gynaed 1993;82:919-924. 209. Savvas M, Studd JWW, Normah S, et al. Increase in bone mass after one year of percutaneous oestradiol and testosterone implants in postmenopausal women who have previously received long-term oral oestrogens. BrJ Obstet Gynaeco11992;99:757-760. 210. Davis SR, Burger HG. Androgens and the postmenopausal women. J Clin Endocrinol Metab 1996;81:2759-2763. 211. Need GA, Horowitz M, Bridges A, et al. Effects of nandrolone decanoate and anti-resorptive therapy on vertebral density in osteoporotic women. Arch Intern Med 1989;149:57-60. 212. Geusens P, Dequekere J. Long-term effects of nandrolone decanoate, 1 alpha-hydroxyvitamin D3 or intermittent calcium infusion therapy on bone mineral content, bone remodelling and fracture rate in symptomatic osteoporosis: a double-blind controlled study. Bone Miner 1986;1:347-357. 213. Passeri M, Pedrazzoni M, Pioli G, et al. Effects of nandrolone decanoate on bone mass in established osteoporosis. Maturitas 1993;17: 211-219. 214. Sherwin BB, Gelfand MM. Differential symptom response to parenteral oestrogen and/or androgen administration in the surgical menopause. Am J Obstet Gynaecol 1985; 151:153-160.
KEATING AND STEVENSON
215. Burger H, Hailes J, Nelson J, Menelaus M. Effects of combined implants on oestradiol and testosterone and libido in postmenopausal women. Br MedJ 1987;294:936-937. 216. Myers LS, Dixen J, Morrissette D, et al. Effects of estrogen, androgen, and progestin on sexual psychophysiology and behavior in postmenopausal women. J Clin Endocrin Metabol 1900;79:1124-1131. 217. Stevenson JC, Banks LM, Spinks TJ, et al. Regional and total skeletal measurements in the early postmenopause. J Clin Invest 1987;80: 258-262. 218. Recker RR, Davies MK, Hinders SM, et al. Bone gain in young adult women. JAMA 1992;268:2403-2408. 219. Rubin CT, Lanyon LE. Osteoregulatory nature of mechanical stimuli: function as a determinant for adaptive remodelling in bone. J Orthop Res 1987;5:300-310. 220. Taaffe DR, Snow-Carter C, Connolly DA, et al. Differential effect of swimming versus weight bearing activity on bone mineral status of amenorrhoeac athletes. J Bone Miner Res 1995;10:586-589. 221. Taaffe DR, Robinson TL, Snow CM, Markus R. High-impact exercise promotes bone gain in well-trained female athletes. J Bone Miner Res 1997;12:255-260. 222. Wolman RL. Osteoporosis and exercise. Br Med J 1994;309: 400-403. 223. Etherington J, Harris PA, Nandra D, et al. The effect of weight-bearing exercise on bone mineral density: a study of female ex-elite athletes and the general population. J Bone Miner Res 1996;11:1333-1338. 224. Skerry TM. Mechanical loading and bone: what sort of exercise is beneficial to the skeleton. Bone 1997;2:179-181. 225. Szoller SM, Martin EM, Parthermore JG, Sartoris DJ, Defros LJ. Demineralisation in tetraplegic and paraplegic man over time. Spinal Cord 1997;35:223-228. 226. Goemaere S, Van Laere M, De Neve P, Kaufman JM. Bone mineral status in paraplegic patients who do or do not perform standing. Osteopor Int 1994;4:138-143. 227. Rambaut PC, Goode AW. Skeletal changes during space flight. Lancet 1985;ii:1050-1053. 228. Hainonen A, Kannus P, Oja P, et al. Randomised controlled trial of effect of high-impact exercise on selected risk factors for osteoporotic fractures. Lancet 1996;348:1343-1347. 229. Nelson ME, Fiatarone MA, Morganti CM, et al. Effect of high-intensity strength training on multiple risk factors for osteoporotic fractures. JAMA 1994;272:1909-1914. 230. Paganini-Hill A, Chao A, Ross RK, Henderson BE. Exercise and other factors in the prevention of hip fracture: the leisure world study. Epidemiology 1991;2:16-25. 231. Tinetti ME, Baker DI, McAvay G, et al. A multifactorial intervention to reduce the risk of falling among elderly people living in the community. N EnglJ Med 1994;331:8221-8227. 232. Province MA, Hadley EC, Hornbrook MC, et al. The effect of exercise on falls in elderly patients. JAMA 1995;273:1341-1347. 233. Sato M, Glasebrook AL, Bryant HU. Raloxifene: a selective estrogen receptor modulator. J Bone Miner Metabol 1994;12:519-520. 234. Yang NN, Venugopalan M, Hardikar S, Glesebrook A. Identification of an oestrogen response element activated by metabolites of 17-beta estradiol and raloxifene. Science 1996;273:1222-1225. 235. Cole MP, Jones CTA, Todd IDH. A new antioestrogenic agent in late breast cancer: an early clinical appraisal of ICI 46 474. Br J Cancer 1971;25:270-275. 236. Ward HWC. Antioestrogen therapy for breast cancer: a trial of tamoxifen at two-dose levels. Br MedJ 1973;1:13-14. 237. Wright CDP, Garrahan NJ, Stanton M, et al. Effect of long-term tamoxifen therapy on cancellous bone remodelling and structure in women with breast cancer.J Bone Miner Res 1994;9:153-159. 238. Grey AB, Stapleton JP, Evans MC, et al. The effect of the antioestrogen tamoxifen on bone mineral density in normal late postmenopausal women. Am J Med 1995;99:636-641.
CHAPTER 28 Interventions for Osteoporosis 239. Love RR, Mazess RB, Barden HS, et al. Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. NEngl J Med 1992;326:852-856. 240. Powels TJ, Hickish T, Kanis JA, Tidy A, Ashby S. Effect of tamoxifen on bone mineral density measured by dual-energy x-ray absorptiometry in healthy premenopausal and postmenopausal women. J Clin Onco11996;14;78-84. 241. Love PR, Barden HS, Mazess RB, Epstein S, Chappell RJ. Effects of tamoxifen on lumbar spine bone mineral density in postmenopausal women after 5 years. Arch Int Med 1994;154:2585-2588. 242. Fornander T, Rutqvist LE. Sjoberg HE, et al. Long-term adjuvant tamoxifen in early breast cancer: effect on bone mineral density in postmenopausal women. J Clin Onco11990;8:1019-1024. 243. Andersen HT, Kristensen B, Ejlertsen B, Andersen KW, Lauritsen JB. Fractures in postmenopausal breast cancer patients treated with adjuvant tamoxifen. Breast 1995;4:245-246. 244. Clements JA, Bennett DR, Black LJ, Jones CD. Effects of a new antioestrogen, keoxifene (LY 1567758) on carcinogen-induced mammary turnouts and on LH and prolactin levels. Life Sci 1983;32: 2869-2875. 245. Wakeling AE, Valcaccia B, Newboult E, Green LR. Non-steroidal antioestrogens: receptor binding and biological response in rat uterus, rat mammary carcinoma and human breast cancer cells. J Steroid Bio Chem 1984;20:111-120. 246. Buzdar AU, Marcus C, Holmes F, Hug V, Hortobagyi G. Phase II evaluation of LY 156758 in metastatic breast cancer. Oncology 1988;45:344-345. 247. Sato M, Rippy MK, Bryant HU. Raloxifene, tamoxifen, nafoxidine or estrogen effects on reproductive and non-reproductive tissues in ovariectomized rats. FASEB J 1996;10:1. 248. Heany RP, Draper MW. Raloxifene and estrogen: comparative bone remodeling kinetics. J Clin EndocrinolMetab 1997;82:3425-3429. 249. Delmas PD, Bjarnason NH, Mitlak BH, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentration, and uterine endometrium in postmenopausal women. N Engl J Med 1997;337: 1641-1647. 250. Ettinger B, Black D, Mitlak BH, et al. Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene: results from a 3-year randomized clinical trial. JAMA 1999;282: 637-645. 250a. Barrett-Connor E, Mosca L, Collins P, et al. Effects of raloxifene on cardiovascular events and breast cancer in postmenopausal women. N EnglJ IVied 2006;355:125-137.
375 251. Murkies AL, Lombard C, Strauss BJG, et al. Dietary flour supplementation decreases postmenopausal hot flushes: effect of soy and wheat. Maturitas 1995;21:189-195. 252. Albertazzi P, Pansini F, Bonaccorsi G, et al. The effect of dietary soy supplementation on hot flushes. Obstet Gyneco11998;91:6-11. 253. Wilcox G, Wahlqvist ML, Burger HG, Medley G. Oestrogenic effects of plant foods in postmenopausal women. Br MedJ 1990;301: 905-906. 254. Reginster JY. Ipriflavone: pharmacological properties and usefulness in postmenopausal osteoporosis. Bone Miner 1993;23:223-232. 255. Gambacciani M, Ciaponi M, Cappagli B, Piaggesi L, Genazzani AR. Effects of combined low dose of the isoflavone derivative ipriflavone and estrogen replacement on bone mineral density and metabolism in postmenopausal women. Maturitas 1997;28:75-81. 256. Agnusdei D, Crepaldi G, Isaia G, et al. A double blind, placebocontrolled trial of ipriflavone for prevention of postmenopausal spinal bone loss. CalcifTissue Int 1997;61:142-147. 257. Agnusdei D, Bufalino L. Efficacy of ipriflavone in established osteoporosis and long term safety. CalcifTissue Int 1997;61;$23-$27. 258. Meunier PJ, Roux C, Seeman E, et al. The effects of strontium ranelate on the risk of vertebral fracture in women with postmenopausal osteoporosis. NEnglJMed 2004;350:459-468. 259. Marshall JH, Lloyd EL, Rundo J, et al. Alkaline earth metabolism in adult man. A report prepared by a task group of Committee 2 of the International Commission on Radiological Protection (ICRP publication 20). Health Phys 1973;24:129-221. 260. Meunier PJ, Slosman D, Delmas P, et al. Strontium ranelate: dose dependent effects in established postmenopausal vertebral osteoporosis: a 2-year randomized placebo-controlled trial. J Clin Endocrinol Metab 2002;87:2060-2066. 261. Reginster JY, Spector T, Badurski J, et al. A short-term run-in study can significantly contribute to increasing the quality of long-term osteoporosis trials. The Strontium Ranelate Phase III Program. Osteoporos lnt 2002;13:$30. 262. Meunier PJ, Roux C, Seeman E, et al. The effects of strontium ranelate on the risk of vertebral fracture in women with postmenopausal osteoporosis. NEnglJMed 2004;350:459-468. 263. Reginster JY, Seeman E, De Vernejoul MC, et al. Stontium ranelate reduces the risk of nonvertebral fractures in postmenopausal women with osteoporosis: Treatment of Peripheral Osteoporosis (TROPOS) Study. J Clin EndocrinolMetab 2005;90:2816-2822. 264. Fogelman I, Blake GM. Strontium ranelate for the treatment of osteoporosis. BMJ 2005;330:1400-1401.
This Page Intentionally Left Blank
..HAPTER 2.
Treatment of Osteoporosis LASZL6 B. TANK6 CLAUS CHRISTIANSEN
Center for Clinical and Basic Research, 2750 Ballerup, Denmark Center for Clinical and Basic Research, 2750 Ballerup, Denmark
I. I N T R O D U C T I O N Osteoporosis is a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture (1). Although both men and women experience some degree of bone loss as a natural part of the aging process, bone loss progresses rapidly in postmenopausal women, and the disorder has become a major health problem in Western countries (2), where increased life expectancy has placed new emphasis on aging-related disorders. Bone mass increases rapidly in growing children and adolescents, reaching a peak in adults between the second and third decades. After the mid-late thirties, bone mass begins to decline. Men lose bone mass at approximately the same rate over their lifetime; in women, however, the rate of bone loss increases dramatically after the natural menopause or surgical removal of the ovaries (3). It is also important to note that bone mass in women after the age of 50 is only two-thirds of that found in men (4). These two factors (i.e., lower initial adult bone mass and a more rapid rate of bone loss) together pose an inclination to develop osteoporosis in old age. Indeed, the prevalence of vertebral fractures increases steadily with age, ranging between 20% for 50-year-old postmenopausal women to 64.5% for older women (5). Significant morbidity and mortality are attributed to osteoporosisrelated fractures (6), and new therapeutic and preventive modalities must be evaluated and applied in high-risk populations. In postmenopausal women with low bone mass or osteoporosis, there are currently several options for therapy and preventive measures that may be implemented with TREATMENT OF THE POSTMENOPAUSAL WOMAN
377
positive results. Treatment of established osteoporosis in quite elderly women (>80 years) who have already suffered multiple nontraumatic fractures (e.g., vertebral crush fractures) is difficult, but in the past few years drugs have been developed that may also help this population. In adult women before the onset of menopause, rates of bone formation and bone resorption are approximately equal; calcium balance is maintained, and no loss of bone mass occurs (3). After the menopause, both bone formation and bone resorption rates increase, but the rate of bone resorption increases more rapidly, resulting in a negative calcium balance and consequent bone loss. Therefore, the first goal of therapy for osteoporosis should be the restoration of bone resorption and bone formation to premenopausal levels. Optimally, we would like to inhibit bone resorption while maintaining bone formation at higher levels, thereby producing a positive calcium balance and countering bone loss. The ultimate aim, however, is the prevention of fractures, both vertebral and nonvertebral (e.g., hip). There are currently numerous agents that provide sufficient antifracture efficacy. These antiresorptive agents reduce the risk of single vertebral fractures by about 40% to 50%. The benefits are usually seen within the first 6 to 18 months of the treatment. Regarding nonvertebral fractures with a focus on hip fractures, hormone replacement therapy (HRT), alendronate, risedronate, and strontium ranelate seem to provide effective protection against this serious complication of osteoporosis (7). The following sections include more detailed descriptions of the individual treatment modalities, with focus on their efficacy and safety for prevention and treatment of osteoporosis. Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
378
TANK6 AND CHRISTIANSEN
II. ESTROGEN AND ESTROGENPROGESTOGEN REPLACEMENT THERAPY Estrogen deficiency accompanying the menopause is now a generally accepted contributor to the pathogenesis of postmenopausal osteoporosis. Accordingly, the most plausible therapy to prevent consequences of estrogen deficiency is to administer HRT. Replacement therapy with estrogen alone (without a progestogen) is often called unopposedestrogen therapy and is abbreviated as ERT, whereas combined estrogen-progestogen therapy is abbreviated as HRT. If estrogen is given alone to women with an intact uterus, the risk of developing endometrial hyperplasia and cancer is considerably increased. Therefore, estrogen is often given in combination with a progestogen, either in a cyclic regimen or in a so-called continuous-combined regimen. In the cyclic regimen, a progestogen is added for a portion of the cycle, perhaps 10 or 14 days. This mimics the natural premenopausal cycle and thus provides monthly menstrual-like bleeding. In continuous combined therapy, the progestogen is given every day together with the estrogen. In this way, the endometrium is kept atrophic continuously, and bleeding is avoided in many, but not all, users (8). In the so-called cyclophasicregimen, the lowest possible dose of progestogen is given in an alternating dosing of 3 days with estrogen alone followed by 3 days with combined estrogen-progestogen (9).
A. Estrogen" Definition Estrogens can be defined as compounds that are able to produce vaginal cornification or uterotropic effects in oophorectomized rats or mice (10). Estrogens can be divided into four main groups, as follows: 1. Synthetic estrogen analogs without a steroid skeleton, often stilbestrol derivates 2. Synthetic estrogen analogs with a steroid skeleton 3. Nonhuman estrogens, produced from an equine source (conjugated estrogens) 4. Native human estrogens or compounds that are transformed to native estrogens in the body The potencies of different estrogens vary. The various bioassays used to determine potencies, however, do not give the same potency ratios. In addition to the differences in potency, estrogens have qualitative differences (10,11). Studies have indicated that nonsynthetic estrogens induce fewer metabolic side effects than do synthetic estrogens (12). Synthetic estrogens without a steroid structure are now obsolete. Synthetic estrogens with a steroid structure (such as ethinyl estradiol) are still the most used estrogens in oral contraceptives.
For postmenopausal therapy, conjugated equine estrogens (CEEs) are the most commonly prescribed estrogens in the United States, whereas in Europe there has been a tradition of using native human estrogen (i.e., 17~-estradiol or estradiol valerate).
B. Progestogen" Definition Substances described as progestogens differ greatly in their additional properties, but they have at least one property in common: they are all able to cause secretory transformation of an estrogen-primed endometrium (13,14). In addition, to this effect, progestogens may have estrogenic, antiestrogenic, androgenic, antiandrogenic, antigonadotropic, glucocorticoidlike, and adrenocorticotropic hormone (ACTH)-stimulating activities. The classes of progestogens may be divided into three subgroups, as follows: 1. 19-Norethisterone (norethindrone) derivates, with a relatively high androgenic effect 2. 17-Hydroxyprogesterone derivatives, with a less androgenic effect 3. Natural human progesterone, which is available in a micronized form for oral therapy
C. Estrogen Effects on Bone Remodeling Calcium reaches the extracellular and intracellular volumes through intestinal absorption, which is mainly regulated by the active metabolite of vitamin D, 1,25(OH)2 D. Furthermore, calcium is transported from the skeleton to the extracellular volume (bone resorption) and in the opposite direction (bone formation) (15,16). The metabolism of calcium changes dramatically when estrogen production declines in women. The main characteristic is increased bone remodeling, which is reflected in high serum concentration of biochemical indicators of bone resorption and bone formation. This increased bone turnover is maintained throughout life and is also seen in women with symptomatic postmenopausal osteoporosis (17). All conditions of estrogen deprivation result in loss of bone. This includes natural or surgical menopause and drugs that inhibit estrogen production or its effects (luteinizing hormone-releasing hormone (LH-RH) agonists, antagonists, or antiestrogens). Strenuous exercise or other conditions that provoke anovulation also may result in accelerated bone loss. Withdrawal of estrogen primarily affects bone resorption, as shown by rapid increase in the biochemical indicators of bone resorption. Because of the coupling of bone resorption and formation, bone formation will show a secondary increase, reflected by a delayed elevation in the biochemical estimates of bone formation (Fig. 29.1) (18).
CHAPTER 29 Treatment of Osteoporosis
379
D. Progestogen Effects on Bone Remodeling
FIGURE 29.1 The effect of hormone replacement therapy (HRT) on bone remodeling. Estrogen deficiency (menopause) primarily affects bone resorption, as shown by a rapid increase in the biochemical markers of bone resorption. Because of the coupling of bone resorption and formation, bone formation will show a secondary increase, reflected by a delayed elevation in the biochemical estimates of bone formation. The opposite sequence of events is seen when H R T is prescribed to an estrogen-deficient woman. Bone resorption is primarily normalized or at least decreased toward a premenopausal level, and only secondarily, bone formation is depressed. This phenomenon results in an increase in bone mass that may last several years but is transient by nature and will level off when the rates of bone resorption and formation approach each other.
Bone loss is the result of an imbalance between bone resorption and bone formation, which implies, in the case of postmenopausal bone loss, that the increase in bone resorption exceeds the parallel increase in bone formation. The difference between the two processes seems to be greatest in the first years after menopause, whereas in the long term it moves toward a new steady state at a higher level. Postmenopausal estrogen therapy rapidly normalizes both bone resorption and formation, leading to a reestablishment of homeostatic balance (19,20). Estrogen replacement therapy primarily decreases bone resorption and only secondarily decreases bone formation, which results in a positive calcium balance for some time. During this period, the resorption lacunae are filled, and bone mineral density (BMD) will increase. However, due to parallel decreases in bone formation (coupling), the increase in bone density will plateau after 2 to 3 years of therapy. Further treatment will result in a stabilization of BMD, but no further increase can be expected (see Fig. 29.1).
Because the progestogens have such different qualities, the effects on bone and calcium metabolism of the various progestogens and progesterone itself need to be studied separately. Only a few studies have been published on the individual effects of progestogens on postmenopausal calcium metabolism. Apparently, however, gestronol hexanoate and norethisterone acetate (NETA) may prevent the loss of bone (21,22), whereas medroxyprogesterone does not have this effect. Studies have suggested that NETA may have bone formation-stimulating effects. A short-term pathophysiologic investigation (23) studied the separate effects of E2 alone and E2 plus NETA on biochemical markers of bone turnover. The markers of bone resorption decreased relatively more than the markers of bone formation during E2-NETA therapy compared with therapy with E2 alone. The next study of the effect of NETA on BMD included a group of premenopausal women with endometriosis (24). To alleviate endometriosis symptoms, women were treated with the L H - R H agonist Nafarelin. This compound results in decreased estrogen production and thereby relieves pain associated with menstruation. However, the treatment has a severe adverse effect in that bone density declines relatively rapidly, accompanied by significant increases in biochemical markers of bone turnover. When this endometriosis treatment was combined with a small dose of NETA, the bone loss was prevented. The biochemical markers of bone turnover also displayed a different response than that of the group using Nafarelinalone, compatible with a relative high ratio of bone formation to resorption. In a recently published study (25) alendronate was compared to a combination of E2 and NETA in early postmenopausal women. The study included a total of 1460 women aged 45 to 59 years, and of those, 110 received HRT. The optimum dose of alendronate appeared to be 5 mg. After 2 years of treatment, BMD in the spine had increased by 3.34% in the alendronate group and by 5.14% in the hormonal group (p < 0.01). The same numbers in the hip were 1.60% versus 3.21% (p < 0.001); in the forearm, 1.14% versus 0.54%; and in the total body, 0.64% versus 2.59% (p <0.001), respectively. Thus, this hormonal therapy tended to build more bone independent of the skeletal site when compared with the bisphosphonate.
E. Combined Effects on Bone Remodeling A number of studies have been carried out with combined hormone therapy (19-21,26-28). Progestogens do not counteract the estrogenic effect on calcium metabolic parameters and may in some instances even add to the effect of the estrogen (29).
380
TANK6 AND CHRISTIANSEN
F. Effects on Bone at D i f f e r e n t Stages o f the M e n o p a u s e Both ERT and HRT prevent osteoporosis in postmenopausal women (26-28). Bone loss caused by estrogen deficiency is prevented by estrogen and estrogen-progestogen therapy, independent of chronological and menopausal age (29-32). 1. EARLY MENOPAUSE
In early postmenopausal women, numerous studies have demonstrated that oral estrogen and estrogen-progestogen regimens prevented bone loss. This is true for use of both synthetic and nonsynthetic estrogens, although nonsynthetic estrogens should be preferred for postmenopausal therapy. The estrogenic effect on bone is dose dependent; if a sufficient serum concentration of estrogen is not obtained, bone loss will not be arrested completely. For oral estrogen and estrogen-progestogen therapy, studies have demonstrated that doses of 0.625 mg of conjugated estrogens and 2 mg of 17f3-estradiol prevent early postmenopausal bone loss (Fig. 29.2) (33-35). However, other studies (9,36-39) suggest that lower doses may be sufficient to prevent bone loss, which is important because adverse effects are dose dependent. The postmenopausal bone loss is a generalized phenomenon affecting all parts of the skeleton; that is, areas with mainly trabecular bone (e.g., the spine) as well as areas with mainly cortical bone (e.g., the forearm and the hip) (27,28). Postmenopausal estrogen therapy prevents bone loss from all skeletal areas (Fig. 29.3). Studies of total body bone mineral density have demonstrated that estrogen therapy prevents bone loss in the head, chest, arms, pelvis, and legs (40,41). Other studies, where bone mass in the spine, hip and forearm have been examined directly, have revealed that HRT definitely arrests bone loss in all skeletal areas (27,28).
FIGURE 29.2 The effect of different doses of estrogens on loss of bone mineral content (BMC). The optimum bone-sparing dose of estradiol lies between 1 and 2 mg daily and is 0.625 mg daily for conjugated estrogens (Left from ref. 34 and right from res 35, with permission.)
In addition to oral HRT, other delivery systems are available. Percutaneous 17~-estradiol, (i.e., estradiol given in a gel applied every day on the skin) prevents skeletal bone loss as effectively as oral HRT (19). The same is the case with estradiol given in a patch that is changed two to three times per week, where studies have demonstrated that this delivery system also prevents bone loss (42). These alternative routes of administration may be particularly useful when prescribing HRT for smokers, in whom enhanced first-pass effects in the liver may hamper bioavailability of orally dosed estrogens (43).
2. LATE MENOPAUSE
The increased bone remodeling seems to be most pronounced in the early postmenopausal years. With increasing age postmenopausally, there may be a tendency towards a slowing down in bone turnover and thereby in the rate of bone loss. The greatest benefit from HRT therefore is obtained if it is instituted shortly after menopause. Indeed, it has been debated whether HRT in elderly postmenopausal women has any effect at all. However, the literature contains clear evidence that HRT prevents bone loss at all stages of postmenopausal lift (29-33,43). Lindsay et al. treated osteoporotic patients with estrogen and calcium or calcium alone (30). The mean number of years since menopause was 12, with a range of 4 to 35 years. At the end of 24 months, mean bone mass in the spine and hip was significantly higher in the estrogen-treated group than in the calcium-treated group. At both skeletal sites, bone loss was stopped, and bone density even increased moderately. Christiansen et al. treated 65-year-old women who had sustained their first osteoporotic fracture with a continuous combination of E2 and NETA (29). After 12 months of treatment, bone density in skeletal areas with a high content of trabecular bone (spine and ultradistal forearm) had increased by 8% to 10% when compared with placebo. Bone
FIGURE 29.3 Percent changes of bone mineral content (BMC) or bone mineral density (BMD) at different skeletal sites during therapy with HRT (0) or placebo (O). Note that the bone loss is prevented in all skeletal areas. (From ref. 27, with permission.)
CHAPTER29 Treatment of Osteoporosis BMCprox
381
BMCdist
110 100
O
~
~
90 I
120
I
I
I
I
I
I
TBBM
I
I
I
i
I
BMCspine
11o
100
~------------____~.~ 12
~ 24
6
' ;
' 12 '
24 months
FIGURE 29.4 Changesof bone mass at different skeletalsites in 65-yearold women taking either HRT (O) or placebo (O). (From ref. 29, with permission.)
mineral density of the total skeleton and the distal forearm, which contain approximately 80% cortical bone, both increased on the average by 3% to 5% (Fig. 29.4). Lufldn et al. treated 66 osteoporotic patients with transdermal estrogen, 0.1 mg, 25 days per month plus medroxyprogesterone acetate (MPA), 10 mg/day for 10 days per month, or placebo (43). All subjects received 800 mg of dietary calcium per day. Estrogen-treated patients showed a sharp decrease in histologic features of osteoporotic bone resorption and bone formation, an increase in spinal bone density, and a decrease in the fracture rate when compared with placebo-treated patients. These anatomic changes were accompanied by significant decreases in the serum markers of bone turnover in treated patients (serum osteocalcin "--20%, serum bone specific alkaline phosphatase ---40%, and urinary hydroxyproline ---40%).
G. Effects on Fractures Large epidemiologic studies demonstrated that use of estrogens is associated with decreased fracture risk in postmenopausal women (44,45). In randomized clinical trials, women receMng HRT had a 50% decreased risk for vertebral fractures compared with those treated with placebo (46). Meta-analysis of clinical trials reported between 1997 and 2000 suggested that HRT may decrease the risk of not only vertebral but also nonvertebral fractures (47,48). A pooled analysis indicated a significant 27% reduction in the risk of nonvertebral fractures. Stratification of women into
groups of younger or older than 60 years revealed that the reduction in risk is more pronounced in women <60 years old compared with women >60 years old (33% versus 12%). When focusing on hip and wrist fractures, the reduction in risk was even more pronounced (40%), particularly in women <60 years (55%). These findings were confirmed by the recently terminated Women's Health Initiative (WHI) study (49), which was the first large-scale randomized study to assess the effects of HRT in a primary prevention setting. In this study, which followed patients over a 5.2-year period, HRT reduced the risk of hip fracture and symptomatic fractures of the spine (both by 34%).
H. Adverse Effects In the W H I trial, long-term use of estrogen plus progestogen therapy was associated with an increased risk of adverse events compared with placebo (49). Absolute excess risks per 10,000 person-years attributable to estrogen plus progestin were 7 more C H D events, 8 more strokes, 8 more pulmonary embolisms, and 8 more invasive breast cancers, whereas absolute risk reductions per 10,000 person-years were 6 fewer colorectal cancers and 5 fewer hip fractures. The absolute excess risk of events included in the global index was 19 per 10,000 person-years. In the estrogen-only arm of the trial (50), there was an absolute excess risk of 12 additional strokes per 10,000 person-years and an absolute risk reduction of 6 fewer hip fractures per 10,000 personyears. The estimated excess risk for all monitored events in the global index was a nonsignificant 2 events per 10,000 person-years. Interestingly, estimated hazard ratios for conjugated estrogen versus placebo for the major clinical outcomes were below 1.0 for both coronary events (0.91, 0.75-1.12) and breast cancer (0.77, 0.59-1.01) (51). These findings shed light on a potential direct contribution of progestogens to cardiovascular and breast cancer risk. Regrettably, we still do not have definitive answers as to how the global index would look in a trial undertaken in early postmenopausal women. Only 250 women between ages 50 and 54 received active treatment in the trial including more than 16,000 participants.
I. Low- and Ultra-Low-Dose Therapy The adverse effect of estrogen therapy is dose dependent. Therefore, there is a genuine interest in reducing the dose of estrogen to decrease the risk of complications while maintaining effective fracture prevention. Currently ultra-lowdose estrogens are being explored, and preliminary results are promising, indicating significant effects on bone turnover and bone mass (51). However, it remains to be clarified
382
TANKd AND CHRISTIANSEN
whether they offer effective fracture protection and at the same time a better safety profile compared with the traditional oral treatments. The fact that ultra-low-dose estrogen patches do not require continuous progestogen supplementation may be advantageous for the overall safety profile.
J. When to Use Estrogen or Estrogen Plus Progestogen Therapy Numerous investigators have argued for the notion that ERT/HRT, which if administered in the early phase of the menopause (where bone loss is most accelerated) and not for more than 5 years, is likely to be beneficial for women. The benefits are not confined to the inhibition of menopausal symptoms only, but also the prevention of late complications of the menopause. We recently demonstrated that early postmenopausal women who received 2- to 3year HRT in randomized settings have significantly higher (>5%) BMD compared with untreated women, even many years after cessation of the treatment. Furthermore, the preservation of bone mass in the early postmenopausal years and consequent delay in bone loss led to a 52% decrease in the relative risk for osteoporotic fractures later in life (52). Thus, limited administration of HRT in the early postmenopausal years seems to offer long-lasting benefits for the prevention of postmenopausal bone loss and osteoporotic fractures.
III. ESTROGEN ANALOGS
A. Selective Estrogen Receptor Modulators (SERMs) In 1973, McGuire and Chamness (53) summarized their work on the estrogen receptor in animal and human breast tumors and described a target for therapeutic intervention. At that time, there were no clinically useful antiestrogens, but the subsequent development of tamoxifen for breast cancer therapy has revolutionized the therapeutic approach. In addition to breast cancer therapy with antiestrogens, a new strategy has been developed to exploit the target site for the beneficial actions of estrogen (i.e., prevention of bone loss and cardiovascular disease) (54). To date, only raloxifene has been approved by the U.S. Food and Drug Administration (FDA) for the prevention and treatment of osteoporosis. The two other SERMs approaching the final stage of clinical development are bazedoxifene and lasofoxifene. Both of these SERMs are claimed to be more potent than raloxifene and are being developed in lower doses for daily use. Furthermore, neither lasofoxifene nor bazedoxifene have demonstrated negative effects on vasomotor symptoms.
TABLE 29.1 Efficacy of Raloxifene* in Increasing Bone Mass and Decreasing Fracture Risk** Efficacy parameter CTX I Bone-specific alkaline phosphatase Osteocalcin Spine BMD Hip BMD Femoral neck BMD Total body BMD Vertebral fracture risk
Raloxifene 60 mg versus placebo - 34% -23.1% - 15% 2.4% 2.4% 2.5% 2.0% - 30% (OR 0.7, 95% CI 0.5-0.8)
*60 mg daily dose. **Data from the MORE Study Trial. (From ref. 55.) BMD, bone mineral density; CI, confidence interval; OR, odds ratio.
1. EFFECTS OF RALOXIFENE ON SKELETAL HOMEOSTASIS
The largest study evaluating the safety and efficacy of two doses of raloxifene was the Multiple Outcomes Raloxifene Evaluation (MORE) study, which included 7700 women with osteoporosis. In this trial, raloxifene was generally well tolerated, able to normalize bone turnover, counter postmenopausal bone loss effectively, and reduce the risk of incident vertebral fractures. However, there was no suggestion of reduced risk of any other osteoporotic fractures (i.e., nonvertebral fractures), although the study was neither designed nor powered for the demonstration of such effects (55,56). Effects on bone mass and fracture risk are summarized in Table 29.1. 2. OTHER Evvv.CTS ov RALOXIFENE
In addition to the beneficial effects on bone mass, 4-year treatment with raloxifene also decreased the risk of estrogen receptor-positive invasive breast cancer occurrence by 65% compared with placebo-treated women (57). Raloxifene has also been shown to reduce aortic atherosclerosis significantly in cholesterol-fed rabbits (58). In one recent publication in which cardiovascular events were evaluated in a post hoc analysis of the M O R E trial, raloxifene had no apparent effect on the risk of cardiovascular events in the whole study population. However, when focusing on women with an increased risk for cardiovascular events at baseline, a 40% decrease in the risk of incident cardiovascular events was achieved compared with placebo-treated women (59). This is being more systematically evaluated in the Raloxifene Use for the Heart (RUTH) trial, which includes more than 10,000 women. Collectively, the multiple benefits of raloxifene treatments will likely keep this medication an attractive option for preventive purposes in postmenopausal women.
CHAPTER29 Treatment of Osteoporosis 3. ADVERSE EFFECTS OF RALOXIFENE
Generally, SERMs are well tolerated. In animal models, raloxifene acts as an estrogen receptor antagonist on breast and endometrial tissue, whereas it acts as an estrogen agonist in the skeletal and cardiovascular systems. Long-term use of raloxifene is not associated with an increase in the risk of endometrial cancer and has no adverse effects on cognitive function, mood, or memory function. However, raloxifene (similar to estrogen) increases the risk of venous thromboembolism by about threefold. The most annoying side effect may be an increased number of hot flushes, which, seems to disappear after a few months' treatment (55).
B. Other Tissue- Selective Estrogenic Compounds Tibolone is a synthetic steroid. Tibolone is rapidly metabolized in the circulation to metabolites; 3 alpha- and 3 beta-hydroxy-tibolone exerting estrogenic effects, whereas the delta(4)-isomer possesses androgenic and progestogenic properties (60). The effects on bone, brain, and vagina can be accurately explained by the estrogenic activity of tibolone, but estrogenic activity is not expressed in the endometrium. Tibolone behaves differently from estrogen plus progestogen combinations on the breast (61), and it has antiatherogenic effects according to animal experiments (62). It is approved in many countries for treatment of menopausal symptoms such as hot flushes. This compound also prevents bone loss (38) and reduces the bone turnover to normal levels (63); with half of the recommended dose eliciting comparable responses (38). However, no information is available regarding the potentials of this therapeutic option in terms of fracture prevention. The Long term Intervention on Fractures and other endpoints Trial (LIFT) study, which was initiated to answer this question, was terminated after the release of concerning results from the W H I trial.
IV. BISPHOSPHONATES Bisphosphonates represent a class of drugs developed in the past 2 decades for use in various diseases of the skeleton, teeth, and calcium metabolism (64). Bisphosphonates are compounds characterized by two carbon-phosphorus bonds, and they are analogous to the physiologically occurring inorganic pyrophosphate, in which an oxygen atom has been replaced with a carbon atom. The structure of bisphosphonares allows for a great number of variations. Each bisphosphonate has its own physicochemical and biologic characteristics. One should thus not speak generally about the effect of "the bisphosphonates" but rather consider each bisphosphonate on its own.
383 It is known that pyrophosphate impairs the crystallization as well as the dissolution of calcium phosphate crystals. This effect is apparently related to the strong surface adsorption of pyrophosphate on solid-phase calcium phosphate. Bisphosphonates act in a similar way and thus have the ability to both block the production of apatite crystals and delay the dissolution of the crystals. When given both parenterally and orally, bisphosphonates have been found to inhibit experimentally induced calcification of soft tissues such as the arteries, kidneys, skin, and heart. When administered in sufficiently high doses, certain bisphosphonates can also inhibit the normal calcification that occurs in bone, cartilage, and teeth. Bisphosphonates are primarily resorption inhibitors, with a secondary impact on bone formation (coupling). The approved bisphosphonates are alendronate, risedronate, and ibandronate. The former two are available as oncedaily or once-weekly treatments, whereas the latter has just received approval as once-daily and once-monthly treatments. An upcoming new member is zoledronate, which is currently being tested for its safety and efficacy as a onceyearly treatment of osteoporosis (65).
A. Alendronate For prevention of bone loss, a dose of 5 mg/day of alendronate has been approved (66). Monitoring of the changes in BMD during 10-year therapy with 5 mg alendronate indicated progressive increases at all skeletal sites, reaching 9.3% in the lumbar spine, 2.9% in the total hip, 2.8% in the femoral neck, and 1% in the total body (67). Biochemical markers of bone turnover indicated sustained inhibitions of bone resorption and formation throughout the study period. The study also illustrated the rise of bone turnover early (within I year) after treatment withdrawal, resulting in slow but progressive propagation of bone loss. These results are illustrated in Fig. 29.5. For treatment of women who have established osteoporosis, a dose of 10 mg/day of alendronate is approved. In osteoporotic patients, alendronate markedly decreased bone turnover and increased BMD up to 8% in the spine and up to 6% in the hip compared with the effects of placebo over a 3-year study period. The Fracture Intervention Trial (HT), which included 2000 women with low BMD and at least one prevalent fracture, showed that alendronate was able to reduce the risk of a new single vertebral fracture by 50%, multiple vertebral fractures by as much as 90%, and hip fractures by as much as 50% (68). A similar study in more than 4000 women with low BMD but no fracture suggested that alendronate decreased vertebral and hip fractures by 50%. A later study in almost 2000 women with low BMD showed a 47% decrease in the incidence of nonvertebral fractures (69). Once-weekly dosing of 70 mg alendronate is
384
TANK6 AND CHRISTIANSEN
FIGURE 29.5 Mean percent changes in BMD (A, lumbar spine; B, femoral neck; C, trochanter; D, total hip; and E, total body) and biomarkers of bone turnover (F, urinary N-telopeptides of type I collagen values; and G, serum bone-specific alkaline phosphatase) during different treatment scenarios using either 5 or 10 mg daily dose of alendronate or placebo. Women in the discontinuation group received 20 mg of alendronate daily for 2 years and 5 mg daily in years 3, 4, and 5, followed by 5 years of placebo. (From ref. 67, with permission.)
therapeutically equivalent to the daily dosing of 10 mg and offers an option to improve patient compliance.
B. Risedronate Risedronate is another potent aminobisphosphonate with similar marked effects on bone turnover and bone mass. The Vertebral Efficacy with Risedronate Therapy (VERT) studies, two studies of identical design in more than 3600 women with established osteoporosis, indicate that risedronate reduced the risk of vertebral fractures by 40% to 50% (70,71). These studies also revealed a 33% to 40% reduction in the risk of nonvertebral fractures. The Hip Intervention Program (HIP) study, which included 9300 women, was specifically designed to assess the potentials of risedronate for the prevention of nonvertebral fractures (72). One arm included women in their 70s with proven osteoporosis, whereas a second arm included women in their 80s or older with risk factors for fracture but no proven osteoporosis. Results from the former group showed a 40% reduced risk for hip fracture in pooled risedronate groups (2.5 mg and 5 mg daily. In the second arm, a non-
significant 20% decrease in the relative risk of hip fractures to the treatment was apparent. More recent development generated two once-weekly dosing regimes for alendronate and risedronate, which both were shown to have equivalent effects compared with daily dosing regimes. In the Fosamax Actonel Comparison Trial including 1053 postmenopausal women who received either once-weeNy alendronate (70 mg) or risedronate (35 mg) for 12 months, alendronate induced greater increases in BMD and decreases in biomarkers of bone turnover than did risedronate. The differences between BMD gains at the lumbar spine, total hip, and femoral neck were 1.2%, 1.0%, and 0.7%, respectively (p < 0.05%). No differences in safety profile were seen (73).
C. I b a n d r o n a t e In a 1-year randomized, double-blind, placebo-controlled study investigating effects of different daily doses of ibandronate (0.25 mg to 5 mg) on BMD, Ravn et al. established 2.5 mg as the optimal dose for daily treatment (74). The FDA approval of Boniva for the prevention of osteo-
CHAPTER 29 Treatment of Osteoporosis
385
TABLE 29.2 Effects of Marketed Bisphosphonates in Major Clinical Trials
Alendronate Daily (10 rag)
Trial
Treatment period
FIT
3 years
Spine BMD 6.2%
Hip BMD 4.7%
Vertebral fracture risk
Nonvertebral fracturerisk
-47% -
Risedronate Daily (5 mg) Ibandronate Daily (2.5 mg) Monthly (150 mg) versus daily (2.5 mg)
-51% hip 48% wrist
VERT
3 years
4.3%
2.8%
-41%
-39%
BONE MOBILE
3 years I year
6.5% 4.9% vs. 3.9%
3.4%
-62%
-69%*
3.5% vs. 2.5%
--
-
*In high-riskpopulation (hip BMD T-score <-3.0). The BMD changes refer to placebo corrected changes. BMD, bone mineral density. (From refs. 68, 71, 76, and 78.)
porosis was based on a randomized, double-blind, placebocontrolled, 2-year trial of 653 postmenopausal women aged 41 to 82 years of age without osteoporosis. Subjects were given ibandronate at different dose (0.5 mg, 1.0 mg, and 2.5 mg), or placebo. Results showed that treatment with 2.5 mg daily dose achieved a mean increase in lumbar spine BMD of 3.1% compared with placebo after 2 years of treatment, accompanied by increases in hip, femoral neck, and trochanter BMD of 1.8%, 2.0%, and 2.1%, respectively (75). FDA approval of this dose for the treatment of osteoporosis was based on a 3-year, Phase III trial (BONE study) (76,77). The study enrolled 2946 postmenopausal women between age 55 and 80 years with osteoporosis. Subjects were treated with either placebo or one of two oral ibandronate schedules: daily (2.5 mg) or intermittent (20 mg) taken every other day for 24 days followed by a between-dose interval of greater than 2 months. Results showed that 2.5 mg daily of ibandronate reduced the risk of new vertebral fractures by 62% compared with placebo. In addition, data demonstrated that an intermittent (20 mg) dose of oral ibandronate taken every other day for 24 days reduced the risk of new vertebral fractures by 50% compared with placebo. Approval of the once-monthly dosing formulation was based on a 12-month, double-blind, placebo-controlled noninferiority study that compared the safety and efficacy of the 150 mg once-monthly dose of Boniva with the approved 2.5 mg once-daily dose (78). The trial enrolled 1602 postmenopausal women with low BMD due to confirmed osteoporosis, who were randomized to receive one of the two regimens. Primary efficacy data indicated that the oncemonthly dose produced a mean increase in BMD of 4.8% versus an increase of 3.9% for the daily dose. This relative increase exceeded the noninferiority endpoint, establishing the once-monthly regimen's statistical superiority in improving BMD.
D. Adverse Effects In the Phase III trials, all three amino-bisphosphonates were well tolerated, with no increase relative to placebo in the incidence of overall adverse experiences. Abdominal pain and dysphagia were the only individual adverse experiences that were significantly increased, but none of the events were serious or resulted in discontinuation. The risk of these adverse gastrointestinal tract events can be decreased by following the dosing instructions (e.g., avoid lying down for 30 minutes after taking the drug and take the drug with a full glass of water) and may be decreased with once-weekly or once-monthly dosing regimes. Ibandronate and zoledronate are being investigated as intravenous treatments, which offer an alternative route for those with limited tolerability of oral treatment.
V. C A L C I T O N I N Calcitonin is a long-chain polypeptide hormone that is produced by the parafollicular cells of the thyroid gland. Abundant calcitonin receptors are present on the surface of osteoclasts (79). Long-term treatment with calcitonin not only inhibits osteoclast activity, but also decreases the number of osteoclasts. Recent experimental studies indicate that human osteoblasts may also respond to calcitonin peptides
(80,81). To date, calcitonin has been isolated from humans and more than 15 other species, but currently, only human, eel, porcine, and salmon calcitonins are in medical use. Because of its protein-like nature, calcitonin undergoes enzymatic degradation during its gastrointestinal passage. Therefore, for many years, it was administered as a subcutaneous injection, which is not practical for long-term treatment. Later, it was developed as a nasal spray, which is a more convenient
386
TANKd AND CHRISTIANSEN
form for patients. Calcitonin has been approved by the FDA for the treatment of Paget's disease of bone, hypercalcemia, and postmenopausal osteoporosis. Since the discovery of calcitonin in 1961, it has also been used effectively to reverse the bone loss observed in immobilized paraplegic patients, Sudeck's atrophy of bone, patients treated with glucocorticoid medications, and in malignant diseases such as multiple myeloma.
A. C a l c i t o n i n in O s t e o p o r o s i s Because calcitonin acts primarily to suppress osteoclastic activity, one could anticipate that individuals with highturnover forms of the osteoporotic process could respond much more rapidly to calcitonin therapy than those with less-active disturbances in bone remodeling. In fact, this hypothesis has been substantiated with reports of larger effects in patients with high bone turnover (81). In 1987, Reginster et al. (82) provided early evidence that calcitonin therapy effectively counteracted the propensity for vertebral bone loss in women who had been menopausal for no more than 36 months and, as such, were most likely to have active or high-turnover bone remodeling. These findings, which demonstrated a protective effect of calcitonin after ovarian failure, when active bone turnover dominates, were further substantiated by Overgaard et al. in 1989 (83). A double-blind study (84) demonstrated that 200 IU daily of nasal salmon calcitonin (plus 500 mg calcium) prevents further bone loss in postmenopausal women with low bone mass. The nasal administration form produced no metabolic, vascular, or gastrointestinal side effects, nor was there any local irritation. From this study it was concluded that nasal calcitonin might be a potential treatment for women with low bone mass 10 to 15 years after the menopause. Twelve months after withdrawal of therapy, a subgroup of the women resumed treatment with calcitonin 200 IU plus calcium 500 mg daily in an open design for an additional 1-year period (85). A control group of 19 agematched women (no forearm fracture) did not receive any treatment. At the end of the 3 years, the control group had lost significantly more bone in the forearm and spine that had the group treated with intranasal calcitonin for 2 years. The group receiving calcitonin for 1 year had intermediate bone mass values. In the year after withdrawal, the rate of bone loss was similar in the women who had received calcitonin and those who had received placebo (Fig. 29.6). Calcitonin was especially effective in women with initially high bone turnover and low bone mass. The bone response in the spine could thus be estimated by the changes in bone turnover. The data suggest that discontinuous treatment with intranasal calcitonin affects bone and calcium metabolism in established osteoporosis. In women with high-turnover osteoporosis, the therapy results in a net gain of bone in the peripheral and axial skeleton.
FIGURE 29.6 Spinal bone mass in elderly osteoporotic women receiving either intranasal calcitonin or placebo for the first 12 months, no treatment in the second year (all women); and intranasal calcitonin in the third year (all women). (From ref. 85, with permission.)
B. C a l c i t o n i n a n d Fractures A Mediterranean osteoporosis study (86) designed to examine the incidence of hip fracture in more than 5000 men and women 50 years or older from six countries in southern Europe demonstrated that calcitonin therapy was associated with a 31% decrease in the relative risk of osteoporotic hip fractures. In a double-blind study of 400 women with osteoporosis treated with either 100, 200, or 400 IU/ day nasal salmon calcitonin or placebo for 2 years, the vertebral fracture rate was significantly reduced (87). These data later received confirmation by a 5-year, double-blind, randomized, placebo-controlled study (The Prevent Recurrence of Osteoporotic Fracture [PROOF] study), which included 1255 osteoporotic women (88). In this study, treatment with 200 IU salmon calcitonin reduced the risk of vertebral fractures by 33%. Calcitonin has analgesic properties, which make it a useful therapeutic option for elderly women with symptomatic vertebral fractures (89). Interestingly, in both studies, the increases in bone density were relatively modest (1% to 1.5%). The explanation of the significantly reduced fracture risk, despite these modest changes in BMD, might rest in improved bone quality in the trabecular compartment (90). In response to decreases in bone turnover during calcitonin treatment, the vertical trabeculae may strengthen and achieve increased resistance against buckling and perforation, thereby decreasing the risk of crush fractures.
387
CHAPTER 29 T r e a t m e n t o f Osteoporosis
C. Adverse Effects
formulation could ease some of the inconveniences associated with the subcutaneous or nasal administration routes and could further improve compfiance to long-term treatment.
Approximately 8% to 10% of patients treated with calcitonin have nausea and mild gastric discomfort. Only rarely is vomit or diarrhea encountered. These symptoms characteristically occur at the initiation of therapy but either decrease in severity or disappear completely during treatment. Facial flushing and dermatologic hypersensitMty may occur in 2% to 5% of patients, and local pruritic reactions are felt at injection sites in approximately 10% of patients. The dermatologic symptoms also disappear with continued administration of the drug. The incidence of side effects substantially decreased when the drug was introduced as a nasal spray.
VI. ANABOLIC STEROIDS A N D A N D RO G EN S A. Anabolic Steroids Anabolic steroids are chemically related to natural androgens. They are distinguished from the latter by a powerful protein anabolic effect in doses that produce little androgen effect. Anabolic steroids have been used for many years for the treatment of osteoporosis. The use of anabolics was based more on tradition than on scientific results. Recent clinical trials, however, have indicated that patients with established osteoporosis obtain a positive calcium balance during treatment with nandrolone decanoate (93) (Fig. 29.7) resulting in increases in BMD and even reduced vertebral fracture risk (94).
D. Recent Advances Innovative research of the past years led to the development of several oral formulations of calcitonin, of which an Efigen technology-based formulation shows the most promising results. The carrier molecule has been shown to effectively protect the peptide-hormone during its gastrointestinal passage and facilitate its rapid absorption to the circulation (91). A randomized, double-blind, placebo-controlled, Phase II clinical trial including 277 healthy elderly women indicated that this oral formulation induces pronounced dose-dependent decreases in bone resorption without a significant effect on bone formation. In the 0.15 to 2.5 mg dose range, acute responses had a nadir at -81%, whereas the mean decreases from pre-dose values after treatment for 3 months was -20.5% (92). Much expectation surrounds the upcoming Phase III evaluation of this drug candidate because this new
Proximal
B. Estrogen Plus Androgen Some interesting data in the literature indicate that combination of HRT with androgens (e.g., methyltestosterone) increases the benefits, particularly at the hip (95-98). The beneficial effects involve: (1) prevention of increases in sex hormone-binding globulin induced by oral estrogens, which in turn maintains the high bioavailability of estrogen for tissue exposure; and (2) the direct action of androgens on os-
Forearm
%
BMC 2
BMC 1 I09
,o6.
BMD 2
103 100 97-
,
0
,
1
;
I
"
1'2
w.--
I
l
months
Distal Forearm %
BMC 2
109 106 103
I00 97
" o
0 ~
h-
,
6
12
~
,
0
,
,
6
,
,
12 months
FIGURE 29.7 Forearm bone mass during treatment with nandrolone decanoate (O) or placebo (@). BMC1, uncorrected bone mineral content; BMC2, fat-corrected bone mineral density. Proximal and distal refer to two different measurement sites (see text). Values are given as percentage of initial mass, mean + 1 SEM. (From res 93, with permission.)
388
TANKd AND CHRISTIANSEN
teoblasts, stimulating bone formation. These effects were elegantly illustrated by a study published by Raisz et al. (99). The additive benefits of NETA when combined with estradiol also are likely attributable to androgenic properties of this progestogen (23,100). However, it remains to be investigated whether low-dose androgen supplementation could enhance antifracture efficacy.
VII. PARATHYROID H O R M O N E Parathyroid hormone (PTH) is one of the calciumregulating hormones. The PTH secretion increases in response to decreased serum levels of calcium. The PTH then stimulates bone resorption and the production of 1,25 dihydroxyvitamin D, which in turn increases the intestinal absorption of calcium. In this way, the serum concentration of calcium tends to rise. The physiologic effect of PTH is therefore to stimulate bone resorption. However, when dosed intermittently, injections of low-dose synthetic PTH may stimulate bone formation (101).
A. PTH (1-34) The first PTH analog that was launched in 2002 was PTH (1-34), also known as teriparatide. To date, the largest study assessing safety and efficacy is that of Neer et al. (102). The investigators allocated 1637 postmenopausal women, who received either 20 or 40 lag of parathyroid hormone (1-34) or placebo daily for 21 months. The medication was administered as subcutaneous injection. The results are summarized in Table 29.3. Treatment with teriparatide significantly reduced the severity and frequency of back pain. In general, teriparatide treatment is safe and well tolerated. A potential adverse effect worth mentioning is mild hypercalcemia, which usually arises in the initial phase of the treatment. Symptoms reported by treated women include nausea, headache, leg cramps, and dizziness. There is TABLE 29.3 Efficacy of Subcutaneous Teriparatide Therapy in Osteoporotic Women PTH (1-34) 20 pg Lumbar spine BMD Total hip BMD Femoral neck BMD Total body BMD Vertebral fractures Nonvertebral fractures Back pain
PTH (1-34) 40 pg
8.6% 3.6% 3.5% 3.1% -65% - 53% -9%
From ref. 102. BMD, bone mineral density; PTH, parathyroid hormone.
12.6% 5.0% 5.8% 4.5% -69% - 54% - 10%
no indication of an increased incidence of bone tumors in humans during teriparatide treatment.
B. PTH (1-84) PTH (1-84) is the flail-length form of parathyroid hormone, which is now available as a recombinant analog. It is yet unknown whether the flail-length form offers significant advantages compared with the already approved teriparatide. The Treatment of Osteoporosis with PTH (TOP) study investigated the efficacy and safety of daily subcutaneous doses of PTH (1-84) 100 lag/day versus placebo in an 18-month randomized, double-blind, placebo-controlled trial that included 2600 osteoporotic women older than 55 years. The primary end point was fracture reduction with increased BMD. The study was completed in September 2003, but 75% of patients entered an open-label extension study, increasing the total treatment time to 24 months (98). Another study assessing the potentials of PTH (1-84) is the PTH for Osteoporotic Women on Estrogen Replacement (POWER) study. In this trial, participants received daily subcutaneous injections of 100 lag/day PTH (1-84) or placebo in addition to their ongoing HRT (103). Results have not yet been published.
VIII. STRONTIUM RANELATE A. Basic Research Strontium ranelate (SR) is composed of an organic moiety (ranelic acid) and two atoms of stable nonradioactive strontium. Experimental findings obtained with the use of this novel medication were recently reviewed by Marie (104). SR decreases bone resorption and increases bone formation, promoting a lasting positive calcium and consequent increases in bone mass. In ovariectomized rats, SR prevented the reduction of BMC and the decrease of trabecular bone volume induced by estrogen deficiency. In this model, SR inhibited bone resorption while maintaining bone formation at a high level, as documented by plasma biochemical markers and histomorphometric indices of bone formation. In the model of osteopenia induced by hind-limb immobilization in rats, SR reduced histomorphometric parameters of bone resorption and diminished long bone loss, quantified by measurements of BMC, bone volume, and biochemical indices of bone resorption. In intact mice, SR increased bone formation and vertebral bone mass. Furthermore, SR induced an improvement of the mechanical properties of the humerus and/or the lumbar vertebra associated with a commensurate increase in bone dimension, shaft, and volume. In intact growing rats, SR increased the bone trabecular volume without alteration of mineralization. In vitro studies further corroborated the unique mode of action of SR. In rat calvaria culture systems and rat osteoblastic cell cultures, SR
CHAPTER 29 Treatment of Osteoporosis enhanced preosteoblastic cell replication and increased collagen synthesis by osteoblasts. Moreover, SR decreased bone resorption in organ cultures and decreased the resorbing activity of isolated mouse osteoclasts.
B. Clinical Research The effect of SR in postmenopausal women with established osteoporosis was assessed in the STRontium Administration for Treatment of OSteoporosis (STRATOS) study, a multinational, prospective, double-blind, randomized, placebo-controlled trial (105). In this study, 353 women with osteoporosis (i.e., at least one prevalent vertebral fracture and a lumbar T score < - 2 . 4 ) were treated with either SR (500, 1000, or 2000 rag/day) or placebo for 2 years. Those receMng treatment with 2000 mg SR showed significant increases in bone formation and significant decreases in bone resorption, suggested by the changes in biomarkers. The positive calcium balance resulted in a 7.3% increase in lumbar BMD. During the second year of treatment, there was a 44% reduction in incidence of vertebral deformities associated with the treatment with 2000 mg SR. Bone histomorphometry showed no mineralization defects. In the Phase III evaluation (106), 1649 postmenopausal women were treated with either placebo or 2000 mg SR. The findings of this major trial are summarized in Table 29.4. In the TROPOS study (103) including 5091 postmenopausal women, treatment with 2000 mg SR was associated with a 16% reduction in the risk of nonvertebral fractures during the 3-year study period. Among 1977 women at high risk of hip fracture (age >74 years and femoral neck BMD T-score < - 3 , corresponding to - 2 . 4 according to National Health and Nutrition Examination Survey (NHANES) reference), the relative risk reduction for hip fracture was 36%. In this study, relative risk of vertebral fractures was reduced by 39% in the 3640 patients with spinal x-rays and by 45% in the subgroup without prevalent vertebral fracture. The increases in total hip and femoral neck BMD from baseline at the end of the study were 9.8% and 8.2%, respectively. In all trials, SR was safe and well tolerated.
389 TABLE 29.4 Effects of Strontium Ranelate on Skeletal Tissue Efficacy parameter CTX I Osteocalcin Spine BMD Total hip BMD Femoral neck BMD Vertebral fracture risk (1 year) Vertebral fracture risk (3 year) Back pain Nonvertebral fracture risk Major fragility fractures Hip fracture in at-risk patients Histomorphometry
Strontium ranelate 2 mg versus placebo -
12.2% 8.1% 14.4% 9.8% 8.3% -49% -41% -3.6% -16% -19% -36%
No signs of osteomalacia or primary mineralization defect No increase in osteoid thickness and mineralization lag time
BMD, bone mineral density. The data presented is from refs. 106 and 107.
IX. GENERAL GUIDELINES
A. Whom Should We Treat? The greatest efficacy of antiresorptive agents for prevention of future fractures has been observed in patients with prevalent vertebral fractures at the time of the diagnosis, regardless of their BMD. These patients have a strong inclination for further fractures and therefore are in definite need of treatment. Women with a BMD T-score < - 2 . 5 are also at increased risk for future fractures and are also candidates for preventive treatment. Results with alendronate, raloxifene, and risedronate support the antifracture efficacy of treatment initiated in these patients. Whether these treatments provide true antifracture benefits for women with osteopenia is not fully established yet.
B. How Long Should We Treat? C. Adverse Effects During the first 3 years of the treatment, diarrhea was more frequently reported by strontium-treated women compared with placebo-treated women (6.1% versus 3.6%, p = 0.02). Serum calcium concentrations were lower, whereas serum phosphate concentrations were higher in patients at month 3, with a plateau thereafter. Serum creatine kinase concentration in the skeletal muscle was more frequently increased in strontium-treated women compared with placebo-treated women during the study (3.4% versus 1.8%), but without any clinical symptoms whatsoever.
Limited information is available regarding whether antifracture efficacy persists beyond 3 to 5 years. Most antiresorptive agents-HRT, raloxifene, bisphosphonates, and calcitonin-do not provide further benefits in terms of increasing BMD beyond 2 to 3 years. In the study by Bagger et al., we showed that by administering 2 to 3 years of HRT in the critical early years of the menopause, when bone loss in placebo-treated women is most accelerated, significant antifracture efficacy can be achieved, lasting up to 15 years (52). In another study on alendronate, we showed that the long-term benefits of the treatment in regard to preservation of BMD are greatly dependent on the
390
TANKd AND CHRISTIANSEN
overall dose loaded to the skeleton (67,108), which can be achieved by high doses administered within a shorter period or lower doses administered over a longer period. In contrast to hormones and hormone-like substances, which have shorter half-lives, bisphosphonates possess a strong affinity to calcified tissue and may exert persisting strong inhibitory effects on bone turnover (64). Animal studies indicate that persisting super-low bone turnover accompanying and following bisphosphonate treatment may have implications for bone quality (i.e., brittleness due to increased mineral content) (109). The human relevance of these findings is yet unknown but should not be ignored.
C. W h i c h D r u g Should W e U s e and W h e n ? Head-to-head comparisons are sparse in the literature for justifying major differences in antifracture efficacy. Accordingly, the recommendations should rather be based on individual considerations. The use of HRT should clearly be confined to the early menopausal years, especially if the patient complains about menopausal symptoms suggestive of serious estrogen deficiency. Raloxifene and bisphosphonates can also be recommended for women in the later postmenopausal years. The advantage of choosing raloxifene is evident when treating women with a familial predisposition for breast cancer. Alendronate and risedronate can be first choices for the elderly (women > 75 years) when the goal of the treatment is to prevent hip fractures. PTH should be considered in patients with severe and progressive osteoporosis when building bone is vital to achieve protection against further fractures and related morbidity. Calcium and vitamin D supplementation are always recommended for elderly women over 65 years.
References 1. Consensus Development Conference. Diagnosis, prophylaxis and treatment of osteoporosis. Am J Med 1993;94:646-650. 2. Melton LJ. Epidemiology of fractures. In: Riggs BL, Melton LJ, eds. Osteoporosis: etiology, diagnosis and management. New York: Raven Press, 1988:133-154. 3. Nilas L, Christiansen C. Bone mass and its relationship to age and the menopause. J Clin EndocrinolMetab 1987;65:697-702. 4. Thomsen K, Gotfredsen A, Christiansen C. Is postmenopausal bone loss an age related phenomenon? CakifTissue Int 1986;39:123-127. 5. Haczynski J, Jakimiuk A. Vertebral fractures: a hidden problem of osteoporosis. Med Sci Monit 2001;7:1108-1117. 6. Cummings SR, Melton LJ. Epidemiology and outcomes of osteoporotic fractures. Lancet 2002;359:1761-1767. 7. Seeman E, Eisman JA. Treatment of osteoporosis: why, whom, when and how to treat. The single most important consideration is the individual's absolute risk of fracture. MedJAust 2004;180:298-303. 8. Christiansen C, Riis BJ. Five years with continuous-combined oestrogen-progestogen treatment. Effects on calcium metabolism, lipoproteins, and bleeding pattern. BrJ Obstet Gynecd 1990;97:1087-1091.
9. Alexandersen P, Hassager C, Sandholdt I, Riis BJ, Christiansen C. The effect of cyclophasic hormone replacement therapy on postmenopausal bone mass and bone turnover. J Bone Miner Res 1997;12: $499. 10. Hammond CB, Maxson WS. Current status of estrogen therapy for the menopause. Fertil Steri11982;37:5-25. 11. Mashchak CA, Lobo RA, Dozono-Takono R, et al. Comparison of pharmacodymanic properties of various estrogen formulations. Am J Gyneco11982;144:511-518. 12. Hammond CB, Maxson WS. Estrogen replacement therapy. Clin Obstet Gynecol 1968;29:407-430. 13. Wiegratz I, Kuhl H. Progestogen therapies: differences in clinical effects? TrendsEndocrinolMetab 2004;15:277-285. 14. Neumann E The physiological action of progesterone and the pharmacological effects of progestogen. A short review. Postgrad Med 1978;54:11-24. 15. Nordin BBC. CalciumJFoodNutr 1986;42:67-82. 16. Parfitt AM. Q.uantum concept of bone remodeling and turnover: implications for the pathogenesis of osteoporosis. Calcif Tissue Int 1979;28, 1-5. 17. Riggs BL, Mann KG. Assessment of bone turnover in osteoporosis using biochemical marker. In: Christiansen C, ed. Osteoporosis1987, vol2. Copenhagen: Osteopress ApS, 1987:67-76. 18. Christiansen C, Rodbro P, Tjellesen L. Serum alkaline phosphatase during hormone treatment in early postmenopausal women. Acta Med Scand 1984;216:11-17. 19. Christiansen C, Christensen MS, NcNair P, et al. Prevention of early postmenopausal bone loss. Controlled 2-year study in 315 normal females. EurJ Clin Invest 1980;10:273-279. 20. Riis BJ, Thomsen K, Strom V, Christiansen C. The effect on percutaneous estradiol and natural progesterone on postmenopausal bone loss. Am J Obstet Gyneco11987;156:61-65. 21. Lindsay R, Hart DM, Purdie D, et al. Comparative effects of estrogen and a progestogen on bone loss in postmenopausal women. Clin Sci 1978;54:193-195. 22. Abdalla HI, Hart DM, Lindsay R, Leggate I, Hooke A. Prevention of bone mineral loss in postmenopausal women by norethisterone. Obstet Gyneco11985;66:789-792. 23. Christiansen C, Nilas L, Riis BJ, Rodbro P, Deftos L. Uncoupling of bone formation and resorption by combined oestrogen and progestogen therapy in postmenopausal osteoporosis. Lancet 1985;2:800-801. 24. Riis BJ, Christiansen C, Johansen JS, Jacobsen J. Is it possible to prevent the bone loss in young women treated with LH-RH agonist? J Clin EndocrinolMetab 1990;70:920-924. 25. Hosldng D, Chilvers C, Ravn P, et al. Prevention of bone loss with oral alendronate in postmenopausal women under age 60. N EnglJ Med 1998;338:485-492. 26. Christiansen C, Christensen MS, Transbol I. Bone mass in postmenopausal women after withdrawal of estrogen/gestagen replacement therapy. Lancet 1981;1:459-461. 27. Riis BJ, Jensen j, Christiansen C. Cyproteroneacetate, an alternative gestagen in postmenopausal oestrogen/gestagen therapy? Clin Endocrino11987;26:327-334. 28. Riis BJ, Thomsen K, Strom V, Christiansen C. The effect of percutaneous estradiol and natural progesterone on bone loss. Am J Obstet Gyneco11987;156:61-65. 29. Christiansen C, Riis BJ. 17(-Estradiol and continuous norethisterone: a unique treatment of established osteoporosis in elderly women. J Clin EndocrinolMetab 1990;71:836-841. 30. Lindsay R, Aitken JM, Andersen JB, et al. Long-term prevention of postmenopausal osteoporosis by oestrogen. Lancet 1976;1:1038-1040. 31. Jensen GF, Christiansen C, Transbol I. Treatment of postmenopausal osteoporosis. A controlled therapeutic trial comparing oestrogen/ gestagen, 1, 25-dihydroxyvitamin D3 and calcium. Clin Endocrinol 1982;16:515-524.
CHAPTER 29 Treatment of Osteoporosis 32. Q.uigley ME, Martin PL, Burnier AM, Brooks E Estrogen therapy arrests bone loss in elderly women. Am J Obstet Gynecol 1987;156: 1516-1523. 33. Horsman A, Jones M, Francis R, Nordin BC. The effect of estrogen dose on postmenopausal bone loss. N EnglJ Med 1983;309:1405-1407. 34. Christensen MS, Hagen C, Christiansen C, Transbol I. Doseresponse evaluation of cyclic estrogen/gestagen in postmenopausal women. Placebo-controlled trial of its gynecologic and metabolic actions, d m J Obstet Gyneco11982;144:873-879. 35. Lindsay R, Hart CM, Clark DM. The minimum effective dose of estrogen for prevention of postmenopausal bone loss. Obstet Gynecol 1984;63:759-763. 36. Bjarnason NH, Hassager C, Christiansen C. 1713-Estradiol 1 mg and 2 mg in combinations with a new gestagen, gestodene are equally preventive on bone loss in early postmenopausal women. Bone 1997;20:93S. 37. Bjarnason NH, Hassager C, Christiansen C. Profile of a new substitution principle: low dose 1713 estradiol and gestodene./icta Obstet Gynecol &and 1997;76:$56. 38. Bjarnason NH, Bjarnason K, Haarbo J, Rosenquist C, Christiansen C. Tibolone: prevention of bone loss in late postmenopausal women. J Clin Endocrinol Metab 1996;81:2419-2422. 39. Stadberg E, Mattson L-C, Uvebrant M. Low dose 17-beta-estradiol and norethisterone acetate as continuous combined hormone replacement therapy in postmenopausal women: lipid metabolic effects. Menopause 1996;3:90-96. 40. Gotfredsen A, Nilas L, Riis BJ, Thomsen K, Christiansen C. Bone changes occurring spontaneously and caused by oestrogen in early postmenopausal women; a local or generalized phenomenon? Br Med J 1986;292:1098-1100. 41. Riis BJ, Christiansen C. Measurement of spinal or peripheral bone mass to estimate early postmenopausal bone loss? Am J Med 1988;84:646-653. 42. Ribot C, Tremollierds F, Pouillds JM. Cyclic Estraderm TTS 50 plus oral progestogen in the prevention of postmenopausal bone loss over 24 months. In: Christiansen C, Overgaard K, eds. Osteoporosis 1990, vol2. Copenhagen: Osteopress ApS, 1984:1979-1984. 43. Lufkin EG, Wahner HW, O'Fallon WM, et al. Treatment of postmenopausal osteoporosis with transdermal estrogen. ~Inn Intern Med 1992:117;1-9. 44. Weiss NS, Ure CL, Ballard JH, Williams AR, Dalin JR. Decreased risk of fractures of the hip and lower forearm with postmenopausal use of estrogen. NEnglJMed 1980;303:1195-1198. 45. Kiel DR Felson DT, Andersen JJ, Wilson PWF, Moskowitz MA. Hip fracture and the use of estrogens in postmenopausal women. The Framingham study. N EnglJ Med 1987;317:1169-1174. 46. Lufldn EG, Wahner HW, O'Fallon WM, et al. Treatment of postmenopausal osteoporosis with transdermal estrogen. ~Inn Intern ivied 1992;117:1-9. 47. Wells G, Tugwell P, Shea B, et al. Meta-analyses of therapies for postmenopausal osteoporosis. V. Meta-analysis of the efficacy of hormone replacement therapy in treating and preventing osteoporosis in postmenopausal women. Endocr Rev 2002;23:529-539. 48. Torgerson DJ, Bell-Syer SE. Hormone replacement therapy and prevention of nonvertebral fractures: a meta-analysis of randomized trials. J/IMA 2001;285:2891-2897. 49. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. J/IMA 2002;288:321-333. 50. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. J/IMA 2004;291:1701-1712.
391 51. Ettinger B, Ensrud KE, Wallace R, et al. Effects of ultralow-dose transdermal estradiol on bone mineral density: a randomized clinical trial. Obstet Gyneco12004;104:443-451. 52. Bagger YZ, Tanko LB, Alexandersen R et al. Two to three years of hormone replacement treatment in healthy women have long-term preventive effects on bone mass and osteoporotic fractures: the PERF study. Bone 2004;34:728-735. . . . . McGuire WL, Chamness GC. In: O'Malley BW, Means AR, eds. Receptorsfor reproductive hormones. New York: Plenum Press, 1973. 54. Jordan VC. Third annual William L. McGuire memorial lecture. Studies on the estrogen receptor in breast cancer-20 years as a target for the treatment and prevention of cancer. Breast Cancer Res Treat 1995;36:367-385. 55. Delmas PD, Bjarnason NH, Mitlak BH, et al. The effects ofraloxifene on bone mineral density, serum cholesterol, and uterine endometrium in postmenopausal women. N EnglJ Med 1997;337:1641-1647. 56. Lufldn EG, Whitaker R, Argueta R, et al. Raloxifene treatment of postmenopausal osteoporosis. J Bone Miner Res 1997;12:S150. 57. Cauley JA, Norton L, Lippman ME, et al. Continued breast cancer risk reduction in postmenopausal women treated with raloxifene: 4year results from the MORE trial (Multiple Outcomes of Raloxifene Evaluation). Breast Cancer Res Treat 2001;65:125-134. 58. Bjarnason NH, Haarbo J, Byrjalsen I, Kauffman RF, Christiansen C. Raloxifene inhibits aortic accumulation of cholesterol in ovariectomized, cholesterol-fed rabbits. Circulation 1997;96:1964-1969. 59. Barrett-Connor E, Grady D, Sashegyi A, et al. Raloxifene and cardiovascular events in osteoporotic postmenopausal women: four-year results from the MORE (Multiple Outcomes of Raloxifene Evaluation) randomized trial. JAMA 2002;287:847-857. 60. Kloosterboer HJ. Tissue-selectivity: the mechanism of action of tibolone. Maturitas 2004;48:$30-40. 61. Kenemans P, Speroff L, International Tibolone Consensus Group. Tibolone: clinical recommendations and practical guidelines. A report of the International Tibolone Consensus Group. Maturitas 2005; 51:21-28. 62. Bjarnason NH, Bjarnason K, Hassager C, Christiansen C. The response in spinal bone mass to tibolone treatment is related to bone turnover in elderly women. Bone 1997;2:151-155. 63. Clarkson TB, Anthony MS, Cline JM, Lees CJ, Ederveen AG. Multisystem evaluations of the long-term effects of tibolone on postmenopausal monkeys. Maturitas 2004;48:$24-29. 64. Fleisch H. Mechanisms of action of the bisphosphonates. Medicina (t? Aires) 1997;57:65-75. 65. Reid IR, Brown JP, Burckhardt P, et al. Intravenous zoledronic acid in postmenopausal women with low bone mineral density. NEnglJ Med 2002;346:653-661. 66. Hosking D, Chilvers CED, Christiansen C, et al. Prevention of bone loss with alendronate in postmenopausal women under age 60 years of age. N EnglJ ivied 1998;338:485-492. 67. Bone HG, Hosking D, Devogelaer JP, et al. Ten years' experience with alendronate for osteoporosis in postmenopausal women. NEnglJMed 2004;350:1189-1199. 68. Black DM, Cummings SR, Karpf DB, et al. Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fracture. Fracture intervention trial research group. Lancet 1996;348:1535-1541. 69. Cummings SR, Black DM, Thompson DE, FIT Research Group. Alendronate reduces the risk of vertebral fractures in women without pre-existing vertebral fractures. Results of the Fracture Intervention (FIT) Trial. J Bone Miner Res 1997;12:$149. 70. Reginster J, Minne HW, Sorensen OH, et al. Randomized trial of the effects of risedronate on vertebral fractures in women with established postmenopausal osteoporosis. Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. OsteoporosInt 2000;11:83-91.
392 71. Harris ST, Watts NB, Genant HK, et al. Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis: a randomized controlled trial. Vertebral Efficacy With Risedronate Therapy (VERT) Study Group. JAMA 1999;282:1344-1352. 72. McClung MR, Geusens P, Miller PD, et al. Effect of risedronate on the risk of hip fracture in elderly women. Hip Intervention Program Study Group. N E n g l J M e d 2001;344:333-340. 73. Rosen CJ, Hochberg MC, Bonnick SL, et al. Treatment with onceweekly alendronate 70 mg compared with once-weekly risedronate 35 mg in women with postmenopausal osteoporosis: a randomized double-blind study. J Bone Miner Res 2005;20:141-151. 74. Ravn P, Clemmesen B, Riis BJ, Christiansen C. The effect on bone mass and bone markers of different doses of ibandronate: a new bisphosphonate for prevention and treatment of postmenopausal osteoporosis: a 1-year, randomized, double-blind, placebo-controlled dose-finding study. Bone 1996;19:527-533. 75. McClung MR, Wasnich RD, Recker R, et al. Oral daily ibandronate prevents bone loss in early postmenopausal women without osteoporosis. J Bone Miner Res 2004;19:11-18. 76. Delmas PD, Recker RR, Chesnut CHIII, et al. Daily and intermittent oral ibandronate normalize bone turnover and provide significant reduction in vertebral fracture risk: results from the BONE study. Osteoporos Int 2004;15:792-798. 77. Chesnut CH III, Skag A, Christiansen C, et al. Effects of oral ibandronate administered daily or intermittently on fracture risk in postmenopausal osteoporosis. J Bone Miner Res 2004;19:1241-1249. 78. Miller PD, McClung MR, Macovei L, et al. Monthly oral ibandronate therapy in postmenopausal osteoporosis: 1-year results from the MOBILE study. J Bone Miner Res 2005;20:1315-1322. 79. Nicholson GC, Moseley JM, Sexton PM, Mendelsohn FA, Martin TJ. Abundant calcitonin receptors in isolated rat osteoclasts. Biochemical and autoradiographic characterization.J Clin Invest 1986;78: 355-360. 80. Villa I, Dal Flume C, Maestroni A, et al. Human osteoblast-like cell proliferation induced by calcitonin-related peptides involves PKC activity. Am J Physiol Endocrinol Metab 2003;284:E627-E633. 81. Civitelli R, Gonnelli S, Zacchei F, et al. Bone turnover in postmenopausal osteoporosis. J Clin Invest 1988;82:1268. 82. Reginster JY, Denis D, Albert A, et al. One-year controlled randomized trial of prevention of early postmenopausal bone loss by intranasal calcitonin. Lancet 1987;2:1481-1483. 83. Overgaard K, Riis BJ, Christiansen C, Hansen MA. Effect of calcitonin given intranasally on early postmenopausal bone loss. Br MedJ 1989;299:477-479. 84. Overgaard K, Riis BJ, Christiansen C, Podenphant J, Johansen JS. Nasal calcitonin for treatment of established osteoporosis. Clin Endocrino11989;30:435-442. 85. Overgaard K, Hansen MA, Nielsen VH, Riis BJ, Christiansen C. Discontinuous calcitonin treatment of established osteoporosiseffects on withdrawal of treatment. Am J Med 1990;89:1-6. 86. Kanis JA, Johnell O, Gullberg B, et al. Evidence for efficacy of drugs affecting bone metabolism in preventing hip fracture. Br Med J 1992;305:1124-1129. 87. Overgaard K, Hansen MA, Birk-Jensen J, et al. Effect of salcatonin given intranasally on bone mass and fracture rates in established osteoporosis: A dose-response study. Br MedJ 1992;305:556-560. 88. Chesnut CHIII, Silverman S, Andriano K, et al. A randomized trial of nasal spray salmon calcitonin in postmenopausal women with established osteoporosis: the prevent recurrence of osteoporotic fractures study. PROOF Study Group. Am JMed 2000;109:267-276. 89. Silverman SL, Azria M. The analgesic role of calcitonin following osteoporotic fracture. OsteoporosInt 2002;13:858-867. 90. Jiang Y, Zhao J, Geusens P, et al. Femoral neck trabecular microstructure in ovariectomized ewes treated with calcitonin: MRI microscopic evaluation. J Bone Miner Res 2005;20:125-130.
TANK6 AND CHRISTIANSEN
91. Buclin T, Cosma Rochat M, Burckhardt P, Azria M, Attinger M. Bioavailability and biological efficacy of a new oral formulation of salmon calcitonin in healthy volunteers. J Bone Miner Res 2002;17:1478-1485. 92. Tanko LB, Bagger YZ, Mexandersen P, et al. Safety and efficacy of a novel salmon calcitonin (sCT) technology-based oral formulation in healthy postmenopausal women: acute and 3-month effects on biomarkers of bone turnover. J Bone Miner Res 2004;19:1531-1538. 93. Johansen JS, Hassager C, Podenphant J, et al. Treatment of postmenopausal osteoporosis: Is the anabolic steroid nandrolone decanoate a candidate? Bone Miner 1989;6:77-86. 94. Frisoli A Jr, Chaves PH, Pinheiro MM, Szejnfeld VL. The effect of nandrolone decanoate on bone mineral density, muscle mass, and hemoglobin levels in elderly women with osteoporosis: a doubleblind, randomized, placebo-controlled clinical trial. J Gerontol~l Biol Sci Med Sci 2005;60:648-653. 95. Davis SR, McCloud P, Strauss BJ, Burger H. Testosterone enhances estradiol's effects on postmenopausal bone density and sexuality. Maturitas 1995;21:227-236. 96. Garnett T, Studd J, Watson N, Savvas M, Leather A. The effects of plasma estradiol levels on increases in vertebral and femoral bone density following therapy with estradiol and estradiol with testosterone implants. Obstet Gyneco11992;79:968-972. 97. Watts NB, Notelovitz M, Timmons MC, et al. Comparison of oral estrogens and estrogens plus androgen on bone mineral density, menopausal symptoms, and lipid-lipoprotein profiles in surgical menopause. Obstet Gyneco11995;85:529-537. 98. Barrett-Connor E, Young R, Notelovitz M, et al. A two-year, doubleblind comparison of estrogen-androgen and conjugated estrogens in surgically menopausal women. Effects on bone mineral density, symptoms and lipid profiles. J Reprod Med 1999;44:1012-1020. 99. Raisz LG, Wiita B, Artis A, et al. Comparison of the effects of estrogen alone and estrogen plus androgen on biochemical markers of bone formation and resorption in postmenopausal women. J Clin Endocrinol Metab 1996;81:37-43. 100. Abdalla HI, Hart DM, Lindsay R, Leggate I, Hooke A. Prevention of bone mineral loss in postmenopausal women by norethisterone. Obstet Gyneco11985;66:789-792. 101. Seeman E, Delmas PD. Reconstructing the skeleton with intermittent parathyroid hormone. Trends Endocrinol Metab 2001;12:281-283. 102. Neer RM, Arnaud CD, Zanchetta JR, et al. Effect of parathyroid hormone (1-34) on fractures and bone mineral density in postmenopausal women with osteoporosis. N EnglJ Med 2001;344:1434-1441. 103. Adis International Ltd. ALX 111: ALXl-11, parathyroid hormone (1-84)-NPS Allelix, PREOS, PTH, recombinant human parathyroid hormone, rhPTH (1-84). Drugs R D 2003;4:231-235. 104. Marie PJ. Strontium ranelate: a novel mode of action optimizing bone formation and resorption. OsteoDorosInt 2005;16:$7-S10. 105. Meunier PJ, Slosman DO, Delmas PD, et al. Strontium ranelate: dose-dependent effects in established postmenopausal vertebral osteoporosis-a 2-year randomized placebo controlled trial. J Clin Endocrinol Metab 2002;87:2060-2066. 106. Meunier PJ, Roux C, Seeman E, et al. The effects of strontium ranelate on the risk of vertebral fracture in women with postmenopausal osteoporosis. N E n g l J Med 2004;350:459-468. 107. Reginster JY, Seeman E, De Vernejoul MC, et al. Strontium ranelate reduces the risk of nonvertebral fractures in postmenopausal women with osteoporosis: Treatment of Peripheral Osteoporosis (TROPOS) study. J Clin Endocrinol Metab 2005;90:2816-2822. 108. Bagger YZ, Tanko LB, Alexandersen P, Ravn P, Christiansen C. Alendronate has a residual effect on bone mass in postmenopausal Danish women up to 7 years after treatment withdrawal. Bone 2003;33:301-307. 109. Burr DB, Miller L, Grynpas M, et al. Tissue mineralization is increased following 1-year treatment with high doses of bisphosphonates in dogs. Bone 2003;33:960-969.
;HAPTER
3(
Potentials of Estrogens in the Prevention of OsteoarthrltlS: What Do We Know and What 0 uestions are Still Pending? Q
LASZL6 B. TANK6 CLAUS CHRISTIANSEN
MORTEN
A.
KARSDAL
9
9
Center for Clinical and Basic Research, 2750 Ballerup, Denmark Center for Clinical and Basic Research, 2750 Ballerup, Denmark Pharmos Bioscience, 2730 Herlev, Denmark
I. D E F I N I T I O N OF O S T E O A R T H R I T I S
II. P R E V A L E N C E A N D I N C I D E N C E OF O S T E O A R T H R I T I S
Osteoarthritis (OA) is a heterogeneous group of conditions that lead to joint symptoms and signs, which are associated with defective integrity of articular cartilage in addition to related changes in the underlying bone at the joint margins (1). The hallmark of OA is a progressive degradation of articular cartilage and overall joint structure associated with pain, joint stiffness, disability, and altogether impaired quality of life. In the United States, OA is second only to ischemic heart disease as a cause of work disability in people over 50 years of age.
OA is the most common joint disorder in the world. The joints most commonly affected are the hip, knee, hand, and spine. The prevalence of radiographic OA rises steeply with age at all joints sites. In a recent large population-based observational study (2), which included 6585 subjects, a total of 75% of women aged 60 to 70 years had OA in their distal interphalangeal (DIP) joints, and even by 40 years of age, 10% to 20% of subjects had evidence of severe radiographic disease of their hands or feet.
TREATMENT OF THE POSTMENOPAUSAL WOMAN
393
Copyright 9 2007by Elsevier,Inc. All rightsof reproductionin any formreserved.
394 The most recent data approximating the incidence of symptomatic hand, knee, and hip OA were obtained from the Fallon Community Health Plan in the United States (3). In this study, the age- and sex-standardized incidence rate of hand OA was 100 per 100,000 person-years; for hip OA, 88 per 100,000 person-years; and for knee OA, 240 per 100,000 person-years. The incidence of OA is higher in men than women before age 50 (4), but later in life, women have higher incidence rates than men, which also points to a likely contribution of the menopause (3). The U.S. population is growing older. The >65 age group represented only 4% of the population in 1900, accounted for 12.4% in 1988, and is projected to account for 22% by the year 2030 (5). These trends have ultimate implications for the prevalence of OA and the magnitude of its socioeconomic burden. In 2000 the number of patients with OA was 43 million (1 in 6), with most in the age group > 45 years. By the year 2020, 59.4 million persons in the United States will be affected by arthritis. The annual cost to society in medical care and lost wages is more than $100 billion, which is expected to increase by more than 25% by 2020 (5). Accordingly, physicians who provide care for the increased number of patients with OA must be aware of therapeutic modalities that carry the potential of providing help in terms of reducing arthritis, preserving articular cartilage, and thereby preventing the long-term complications of this chronic disease. This is currently a challenging task when one realizes that although a wide array of medications is available to relieve pain and joint stiffness, few to none have true chondroprotective effects.
llI. POTENTIALS OF ESTROGEN IN C H O N D R O P R O T E C T I O N Interestingly, there are notable differences in the prevalence and severity of OA between men and women. After the age of 50 years, women are more likely than men to develop generalized OA, and women have a predilection for painful and deforming form of OA in the hand ("menopausal arthritis") (6). Knee OA also appears to be more common in women than in men, although the female-tomale ratio varies among studies, from 1.4 to 4.0 (7,8). The prevalence rates of the less common hip OA are comparable for the two genders (9). The marked increase in the prevalence and incidence of OA in women after the menopause raises the hypothesis that ovarian estrogens have a role in the maintenance of the structural integrity of articular cartilage. In this chapter, we summarize the currently available data generated by in vitro, preclinical, and clinical studies investigating the effects ofestradiol on articular cartilage and the development of OA. In addition to reviewing the existing literature, we attempt to explain the
TANKd ET AL.
diversity of findings and draw attention to questions that need to be answered before we can draw conclusions, based on good evidence, regarding whether hormone therapy has a role in the prevention of OA in postmenopausal women.
A. Cellular Research 1. PRESENCE OF ESTROGEN RECEPTORS IN CHONDROCYTES
In general, 1713-estradiol acts on target tissues via two different pathways: (1) it may bind to dimerized cytosolic estrogen receptors (ERs) and act as transcriptional factors for estrogen response elements (EREs) in the nucleus (genomic effects) or (2) it may act via membrane-associated mechanisms involving G protein-coupled phospholipase C and connected signal transduction pathways (nongenomic effects) (10,11). These pathways are schematically illustrated in Fig. 30.1. Classical nuclear ERs (both cx and 13) have been demonstrated in articular chondrocytes from rats (12), monkeys (13), rabbits (14), and cows and pigs (15), as well as from humans (16). Binding capacity for estradiol was found to be significantly greater in chondrocytes from females than in chondrocytes from males (17), pointing to gender-related differences in the relative importance of estradiol action. Importantly, estradiol may upregulate the nuclear expression of ER in chondrocytes (18), thereby enhancing its own effects on chondrocytes (13). Collectively, these observations provide evidence for the existence of cellular structural elements, which are the sine qua non of hormonal action on cartilage tissue.
FIGURE 30.1 Schematicillustrationof the effectsof 1713-estradiolvia its
nuclear and membrane-boundestrogen receptors (ERs): the genomic and nongenomicpathways.ERE, estrogenresponsiveelements;ERK, extracellular signal-regulated kinase; Gp: G-protein; IP3: inositol triphosphates; mRNA, messengerRNA; PKC, protein kinase C; PLC, phospholipase C; Shc and Src, tyrosine kinases;TF, transcriptionfactor.
CHAPTER30 Potentials of Estrogens in the Prevention of Osteoarthritis: What Do We Know and What Questions are Still Pending? 395 2. PROVOSV.DMEDIATORS OF ESTROgeN ACTION
B. Preclinical R e s e a r c h
ON ARTICULAR CARTILAGE
The study by Cheng et al. (19) on articular chondrocytes elegantly demonstrated the direct effect of 17[3-estradiol on cell growth and the synthesis of DNA and proteoglycans. Their findings indicated a duality of estradiol action depending on the concentration of sex steroids that cells are exposed to. Estradiol was shown to increase cell number and 3H-thymidine incorporation in a dose-dependent manner, peaking at 10 nmol/L and 1 nmol/L, respectively. However, further increase in the concentration of the hormone causes inhibition of both functions, and this reaches significance at l lxmol/L. The effect of estradiol on proteoglycan synthesis is similar; the maximum stimulating effect can be seen at 10 nmol/L, whereas a significant inhibition can be detected at I lxmol/L. Factors intimately involved in cartilage metabolism that are influenced by the actions of estradiol include the following: 1. Growth factors (transforming growth factor [TGF]-[3, insulin-like growth factor [IGF] system) (20-26) 2. Proinflammatory cytokines (interleukin [IL]-I and -6, TNF-c~) (27-30) 3. Matrix metalloproteinases (MMP-1, MMP-3, MMP13, 53-58 gelatinase) (31-33) 4. Reactive oxygen species (34-37) 5. Inducible nitric oxide synthase (38,39) Numerous experimental observations shed light on the dual action of estradiol depending on the concentration of estradiol the cells are exposed to (Table 30.1) (19,23,29,31,33). Whereas low or physiologic concentrations promote favorable effects, high concentrations exert deleterious effects on cartilage matrix and chondrocytes. Figure 30.2 illustrates the proposed mechanisms by which adequately dosed estrogen is believed to exert chondroprotective effects via modulation of catabolic and anabolic pathways of cartilage turnover.
TABLE 30.1
Examples of the Dual Actions of Estradiol in Vitro
Effect IL-113 induced decrease in PG synthesis 53-58 kD gelatinolytic activity
MMP-1, -3, and -13 mRNA levels Chondrocyte cell number 3H-thymidine incorporation PG synthesis Expression of TGF-[3 Ratio of MMP- 1/TIMP- 1 and MMP-3/TIMP- 1 MMP-1 expression
Our research group has recently reported that surgical removal of the ovaries evoked pronounced increases in the urinary excretion of the C-terminal telopeptide of collagen type II (CTX-II) compared with sham-operated animals (40). This acceleration of collagen type II degradation indicated a more rapid development of surface erosions in the articular cartilage of the knee. The severity score of erosions was proportional to the magnitude of the early increase in CTX-II. Independent investigators reported similar responses to ovariectomy (41). Restoration of estradiol levels or estrogenic stimuli by estradiol replacement therapy or therapy using a selective estrogen receptor modulators (SERMs) was able to normalize cartilage turnover and reduce the propagation of cartilage surface erosions (42). In rabbits, the pioneering studies of Rosner et al. in the late 1970s described a favorable impact of estradiol valerate on the catabolism of proteoglycans, but without any apparent impact on erosive OA pathology (43). Studies by Tsai et al. [44] drew attention to the importance of the dose and route of administration of estradiol were exposed to the chondrocytes. When treating rabbits with intra-articular injection of high doses of estradiol, an apparent deleterious effect on articular cartilage (loss of condyle surface congruity, thinning, fissuring, and fibrillation of the articular cartilage) was observed 9 weeks after operation. At week 12, cartilage erosion extended to the calcified layer, exposing the subchondral bone. However, when estradiol was injected in low doses, no structural harm was induced. When assessed in larger animals such as sheep, ovariectomy evoked detrimental effects on the intrinsic material properties of the articular cartilage of the knee, even though there were no apparent changes in cartilage thickness compared with sham-operated animals 12 months after surgery (45).
Low dose Countering Decrease Decrease Increase Increase Increase Increase Decrease Neutral m i l i e u Inhibition
High dose Enhancing Increase Increase Decrease Decrease Decrease Decrease Slight increase Pro-inflammatory milieu* No effect
Reference 30 19 23 33 31
*Denotes the presence of TNF-c~ or IL-113exceeding10 nmol/L. Similardual action of estradiol has also been described in studies on CD4+ T-lymphocytes (88). PG, proteoglycan.
396
TANK6ET AL.
Anabolic pathway
y
17p-estradiol
Catabolic pathway
Plasma membrane stabilization Improved redox status
1]"TGF-p 1]' IGF
1]" Chrondrocyte proliferation 1]"Chrondrocyte maturation
Protection against reactive oxygen species
1]' Proteoglycan synthesis 1]"Collagen synthesis
,U,IL-1 ,U.IL-6
MMP- 1,3 and 13 1]' TIMP-1 |
J~ Proteoglycan degradation _- J~ Collagen degradation
-.....
I.
,i
Decreased collagenolytic activity Increased cartilage matrix synthesis
l
Chond roprotection? FmURE 30.2 Schematic diagram outlining the proposed mechanisms of action by which estradiol exerts effects on the anabolic and catabolic pathways of cartilage turnover and thereby chondroprotective effects.
Extensive research characterized the long-term effects of estradiol in nonhuman primates, which offers a particularly useful model to approximate the disease of humans. Cynomolgus macaques spontaneously develop OA, and their histologic lesions in the tibial plateau and femoral condyles are very similar to those in humans (46). In female macaques that underwent bilateral ovariectomy, 3-year treatment with conjugated equine estrogens (CEE) significantly reduced the severity of OA compared with untreated animals. The treatment effects at the end of the treatment period were analyzed at four levels. Level 1 described differences in OA severity (articular cartilage structure, number of chondrocyte clones, loss of toluidine blue staining); level 2 described differences in subchondral bone thickness; level 3 described the variability in articular cartilage thickness and area; and level 4 described the calcified cartilage areas and thickness. The investigators found significantly lower level 1 scores and higher level 2 scores (however, the number of osteophytes was significantly lower) in estrogentreated animals compared with controls. Assessment at levels 3 and 4 did not indicate significant treatment-related
benefits. Taking into account the large number of monkeys included in the study, the long duration of the HRT, and the state-of-the-art methodology, this study raises particularly noteworthy arguments for the chondroprotective effects of the sex steroid (47). An important observation recently reported by the same group is that bilateral ovariectomy evokes pronounced increases in dynamic histomorphometric indices of bone turnover in the subchondral but not in the epyphyseal/ metaphyseal cancellous bone compartment (48). The investigators suggested intimate links between the acceleration of subchondral turnover and cartilage homeostasis, with direct implications for the pathogenesis of OA. Thus, part of the favorable impact of estrogen might be mediated by its well-established antiresorptive effect. This notion is also supported by the observation that other antiresorptive agents, such as bisphosphonates, have also been proposed as OA-modifying agents (49). However, the effects of estrogens are likely conveyed in a more physiological manner (via receptors on osteoclasts), and the hormone has not only indirect but also direct effects on chondrocytes (9).
CHAPTER 30 Potentials of Estrogens in the Prevention of Osteoarthritis: What Do We Know and What Q.uestions are Still Pending? 397 C. E p i d e m i o l o g i c a n d C l i n i c a l R e s e a r c h 1. ENDOGENOUS ESTROGEN AND OSTEOARTHRITIS
Numerous small and larger studies correlated sex hormone levels to self-reported or radiologically verified OA findings with no evidence for a significant association in elderly women (50,51). Spector et al. (52) and Sowers et al. (53) reported increased levels of bioavailable estradiol (and androgens) in OA patients. In a small study including 21 patients, Tsai et al. (54) found an apparent strong association of excessive synovial estradiol and estrogen receptor binding with OA of the knee. These observations are in line with the message depicted in Table 30.1: that excessive estradiol exposure of chondrocytes may confer deleterious effects on the structural integrity of articular cartilage. However, these aforementioned associations also can be confounded by numerous pathogenic factors. For example, we have recently shown that the highest levels ofbioavailable estradiol are measured in elderly women with upper-body obesity, a phenotype that is prone to insulin resistance, dyslipidemia, low-grade inflammation, diabetes, and accelerated atherogenesis (55). Thus, we cannot exclude the possibility that the apparent lack of beneficial effects of endogenous estradiol on cartilage homeostasis actually reflects the presence of an array of pathogenic factors exerting adverse effects
TABLE 30.2
2. HORMONE THERAPY AND OSTEOARTHRITIS
In epidemiologic and clinical trials assessing the potential implications of hormone therapy for OA, it must be noted that there is a certain degree of variation in the definition of OA. In the present review, we have defined two major categories: (1) self-reported or symptom-based definition of OA (Table 30.2) and (2) image-based definition of OA (Table 30.3). The overall picture of different studies in which OA was self-reported or defined by joint symptoms suggests little to no benefits of estrogen or hormone replacement therapy (ERT/HRT) (see Table 30.2). Some studies even suggest tendencies toward a greater prevalence of OA among those who report hormone use. These results may be interpreted in several ways: (1) H R T per se offers little benefit in terms of managing joint symptoms and/or (2) H R T may not be beneficial in the elderly in whom cartilage loss is already substantial and cannot be reversed by HRT. The latter no-
Impact of Hormone Therapy on Osteoarthritis as Assessed by Symptom-Based Diagnostic Diagnosis of OA
Negative effects Von Muhlen et al. (56)
on articular cartilage metabolism. However, it is also possible that interactions of endogenous estradiol with visceral fat mass may have direct implications for the generation of proinflammatory cytokines that, by promoting systemic lowgrade inflammation, also increase the susceptibility to OA.
Participants
HRT
Pain history and clinical examination by a study nurse
1001 PMW (mean age 72 years)
14 + 10 year use (min 1 year)
Sandmark et al. (57)
Previous prosthetic surgery for knee OA
Ever-use versus no use
Sahyoun et al. (58)
Physician-diagnosed OA
380 cases versus 370 controls (>50 years) 2416 PMW
Outcome Minimal increase in the incidence of hip (4.1 versus 1.1%) and arm (15.8 versus 13.1%) OA 1.8-fold increase in risk
0-1 year use 1-4 year use 4-10 year use > 10 year use
1.3-fold increase in risk 1.3-fold increase in risk 1.96-fold increase in risk 2.0-fold increase in risk
4-year use of HRT (prospective study) Ever-use versus no
No apparent effects
Ever-use versus no use
No effects
Nevitt et al. (59)
Knee pain and WOMAC scores
Maheu et al. (60)
Various measures of disease activity
PMW with prevalent CVD followed for 4 years 711 PMW 50-75 years 238 "painful" OA 240 "quiescent " "" OA 233 controls
Physician diagnosed or requiring treatment with pills or surgery
42,464 PMW (mean age 53 years) 12,521 with OA
No apparent effects
use
Beneficial effects
Parazzini et al. (61)
27% decrease in risk
CVD, cardiovasculardisease; HRT, hormone replacement therapy; OA, osteoarthritis; PMW, postmenopausalwomen.
398
TANKO ET AL.
TABLE 30.3
Impact of Hormone Therapy on Osteoarthritis as Assessed by Imaging Technique-Based Diagnostic Diagnosis of OA
Beneficial effects Dennison et al. (63)
Radiography (hip)
Erb et al. ( 6 4 )
Radiography (hip or knee)
Nevitt et al. ( 6 5 )
Radiography (hip)
Participants PMW >45 y e a r s old 413 cases listed for hip replacement 413 controls 475 PMW (mean age 66.1 years) who underwent hip or knee replacement 4366 PMW (>65 years old) 539 with mild to moderate OA 214 with moderate to severe OA
Spector et al. (66)
HRT
Outcome
Bilateral oophorectomy Long-term use (>5 years) Short-term use 11.6% used hormones for a median of 5.4 years
1.9-fold increase in risk 40% decreased risk (NS) 70% increase in risk No apparent effects
Cross-sectional study (Low number of hormone users)
Current users
38% decreased risk
Largest crosssectional study
Current users who used HRT for less than 10 years Current users who had used HRT for at least 10 years Current users
25% decreased risk
Radiography of hand and knee (K-L index 2 +) Individual features of osteophytes and joint space narrowing Radiography (knee)
606 women (45-63 years old)
831 women (63-93 years)
Hormone use for at least 4 years
Wluka et al. (69)
Radiography of knee MRI of the knee
551 women (63-93 years) 42 users and 39 controls
Hormone use for 8 years Hormone use for at least 5 y e a r s
Cicuttini et al. (70)
MRI of patella
42 users and 39 controls
Hormone use for at least 5 years
Hannan et al. (67) Zhang et al. ( 6 8 )
Note
Cross-sectional study
43% decreased risk
69% decreased risk
Cross-sectional study
+ less joint space narrowing in the DIP joints 29% decreased risk for all OA 34% decreased risk for knee OA 60% decreased risk for knee OA Hormone use was associated with more articular cartilage No differences in patella cartilage
Prospective study
Prospective study Cross-sectional study Cross-sectional study
DIP, distal interphalangeal; HRT, hormone replacement therapy; MRI, magnetic resonance imaging;NS, non-significant; OA, osteoarthritis; PMW: postmenopausal women.
tion seems to be supported by the findings of an Italia study reported by Parazzini et al. (61), who found apparent benefits of H R T in a large population of early postmenopausal women who were an average of 53 years old at the time of assessment. A prospective analysis from the Chingford study (62) provides important clues as to why estrogen deficiency is an important risk factor for progressive OA and why hormone therapy is likely to provide beneficial effects in the early menopausal years. From the original study population of 1003 women, those women who had not received any bonemodifying medication, those who had radiographic informa-
tion on the 4-year progression of knee OA, and those who had baseline information on bone mineral density (BMD) were identified and separated into four groups as follows: controls (n = 50), progressive OA (n = 71), nonprogressive OA (n = 36), and osteoporosis (n = 59). The results demonstrated that bone resorption was increased in patients with progressive knee OA and was essentially unaltered in those with nonprogressive knee OA. The increase in bone resorption seen in patients with progressive knee OA was similar to that observed in patients with osteoporosis. These observations thus suggest that (1) the well-established effects of estrogen deficiency on bone turnover and related bone loss may
CHAPTER 30 Potentials of Estrogens in the Prevention of Osteoarthritis: What Do We Know and What Q.uestions are Still Pending? 399 also have implications for cartilage homeostasis and (2) the beneficial effects of hormone therapy in terms of inhibiting osteoclastic bone resorption might also provide benefits in terms of maintaining cartilage integrity if initiated in the early years of the menopause, when the impact of estrogen deficiency on bone loss is most pronounced. Table 30.3 lists major studies using imaging techniques as a basis for the diagnosis of joint lesions and OA. Interestingly, in this case most studies suggest decreased risk of OA associated with the use of HRT. Of these studies, some particular comments should be made. The Study of Osteoporotic Fractures published by Nevitt et al. (65) is a large cross-sectional analysis with validated radiographic data on hip OA. In this study, 539 mild-tomoderate cases and 214 moderate-to-severe cases of OA could be identified. The number of hormone users was also sufficiently high for proper statistical evaluation. Current users who used HRT for less than 10 years had a 25% decreased risk, whereas current users who used this therapy for more than 10 years had a 43% decreased risk for hip OA. The Framingham investigators published two reports regarding the effects of 4 and 8 years of HRT on the risk of OA; the former including 831 subjects and the latter including 551 subjects (67,68). Both analyses indicated moderate statistically nonsignificant protective effect against worsening of radiographic knee OA among elderly women. These findings corroborate the cross-sectional observations and point further to a potential benefit of female hormones in OA. Recently, a more advanced imaging technique, magnetic resonance imaging (MRI), has been introduced for a more direct visualization and morphometric assessment of articular cartilage. Preliminary studies using this technique indicated more articular cartilage in long-term users of HRT (>5 years) compared with never-users (69). Interestingly, the apparent benefits of HRT were seen in weight-bearing cartilage only, but not at other sites such as the patella (70).
IV. CONCLUSIONS AND PERSPECTIVES Currently hormone therapy is not recommended for primary prevention. The decision is based on findings of the Women's Health Initiative (WHI) study, which found risks exceeding benefits when the therapy was given as an estrogen plus progestin combination, and a balanced risk-benefit profile when the therapy was given as estrogen only (71,72). However, it is important to point out that the potential benefits of estrogen replacement for the prevention of OA have not been considered in the equation of the global index of risk and benefits. Current research on estrogen and OA appears to be conflicting, yet there is enough experimental
and clinical evidence to support further investigation into the therapeutic effects. Moreover, it remains to be clarified whether the conclusions of the W H I study would have been different had the majority of women in the study been in their early menopausal years. Hormone users in observational studies are women who initiate hormone therapy due to a need to control symptoms of estrogen deficiency arising in the early phase of the menopause. This period might actually open a therapeutic window for maximizing the multiple benefits of the therapy for effective long-term prevention of menopause-related diseases. In support of this line of thinking, we have recently shown that when applied in this critical period, even 2 to 3 years of H R T can bring lasting benefits for postmenopausal women after treatment withdrawal, in terms of prevention of osteoporotic fractures and cognitive decline (73,74). Another unanswered question is whether the safety profile of CEE plus medroxyprogesterone (MPA) is equally applicable to other combinations. In light of the W H I findings, the progestin component may carry greater importance for the coronary events and incidence of breast cancer during long-term therapy. Thus, estradiol with more novel progestins might confer different effects. Evaluation of progestins for their effects on insulin sensitivity and breast density is a crucial element of establishing safer combinations. Parenteral combination therapies, in particular ultralow-dose transdermal estradiol delivery systems carry great potential in this regard because they do not require continuous progestin supplementation (75). Finally, there are some methodologic issues worth mentioning with regard to the clinical assessment of the chondroprotective potential. The review illustrated that diversity of methods and selected end points had a considerable impact on the overall impression as to whether HRT carries chondroprotective effects. Therefore, there is an unmet need for standardizing and enhancing the accuracy and precision of methodology and thereby the reliability of our methodologic approach to the quantification of changes in articular cartilage. MRI allows direct three-dimensional visualization and quantification of articular cartilage and will likely improve the assessment of true chondroprotective effects of candidate drugs (76). In addition, much expectation regards the upcoming biomarkers. Because of their ability to respond rapidly to interventions, biologic markers will likely come to play an important role in the clinical development of novel structure-modifying agents (77). Biomarkers might also provide useful assistance in facilitating Phase I and II trials, which are supposed to define the optimal dose range of candidate drugs for large-scale assessments in Phase III trials. Randomized, double-blind, placebo-controlled trials using a methodologic approach that combines biomarkers
400
TANK6 ET AL.
with MRI will likely bring us closer to providing definitive answers as to whether hormone therapy administered in the early postmenopausal years carries the potential for decreasing the epidemiologic burden of OA.
References 1. Altman R, Asch E, Bloch D, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the knee. Arthritis Rheum 1986;29:1039-1049. 2. van Saase JL, van Romunde LK, Cats A, Vandenbroucke JP, Valkenburg HA. Epidemiology of osteoarthritis: Zoetermeer survey. Comparison of radiological osteoarthritis in a Dutch population with that in 10 other populations. Ann Rheum Dis 1989;48:271-280. 3. Oliveria SA, Felson DT, Reed JI, Cirillo PA, Walker AM. Incidence of symptomatic hand, hip, and knee osteoarthritis among patients in a health maintenance organization. Arthritis Rheum 1995;38:1134-1141. 4. Felson DT, Lawrence RC, Dieppe PA, et al. Osteoarthritis: new insights. Part 1: the disease and its risk factors. Ann Intern Med 2000;133:635-646. 5. Weinstein JN, Barmier JD. The Dartmouth atlas of musculoskeletal health care. Chicago: American Hospital Association Press, 2000. 6. Nelson DT, Nevitt MC. The effects of estrogen on osteoarthritis. Curt Opin Rheumato11998;10:269-272. 7. Felson DT, Zhang Y, Hannan MT, et al. The incidence and natural history of knee osteoarthritis in the elderly. The Framingham Osteoarthritis Study. Arthritis Rheum 1995;38:1500-1505. 8. Meachim G, Pedley RB. Implications of a sex difference in osteoarthrosis. Ann Rheum Dis 1980;39:199. 9. Srikanth VK, Fryer JL, Zhai G, et al. A meta-analysis of sex differences prevalence, incidence and severity of osteoarthritis. Osteoarthritis Cartilage 2005;13:769-781. 10. Sylvia VL, Walton J, Lopez D, et al. 17-beta-Estradiol-BSA conjugates and 17-beta-estradiol regulate growth plate chondrocytes by common membrane associated mechanisms involving PKC dependent and independent signal transduction. J Cell Biochem 2001;81:413-429. 11. Evinger AJ III, Levin ER. Requirements for estrogen receptor alpha membrane localization and function. Steroids 2005;70:361-363. 12. Pelletier G. Localization of androgen and estrogen receptors in rat and primate tissues. Histol Histopatho12000; 15:1261-1270. 13. Richmond RS, Carlson CS, Register TC, Shanker G, Loeser RF. Functional estrogen receptors in adult articular cartilage: estrogen replacement therapy increases chondrocyte synthesis of proteoglycans and insulin-like growth factor binding protein 2. Arthritis Rheum 2000; 43:2081-2090. 14. Dayani N, Corvol MT, Robel P, et al. Estrogen receptors in cultured rabbit articular chondrocytes: influence of age. J Steroid Biochem 1988;31:351-356. 15. Classen H, Hassenpflug J, Schunke M, et al. Immunohistochemical detection of estrogen receptor alpha in articular chondrocytes from cows, pigs and humans: in situ and in vitro results. Ann Anat 2001;183:223-227. 16. Ushiyama T, Ueyama H, Inoue K, Ohkubo I, Hukuda S. Expression of genes for estrogen receptors alpha and bets in human articular chondrocytes. Osteoarthrits Cartilage 1999;7:560-566. 17. Nasatzky E, Schwartz Z, Soskolne WA, et al. Evidence for receptors specific for 17 beta-estradiol and testosterone in chondrocyte cultures. Connect Tissue Res 1994;30:277-294. 18. Richmond RS, Carlson CS, Register TC, Shanker G, Loeser RF. Estrogen replacement therapy increases chondrocyte synthesis of proteoglycans and insulin-like growth factor binding protein 2. Arthritis Rheum 2000;43:2081-2090.
19. Cheng P, Ma X, Xue Y, Li S, and Zhang Z. Effects of estradiol on proliferation and metabolism of rabbit mandibular condylar cartilage cells in vitro. Chin MedJ (Engl) 2003;116:1413-1417. 20. _Arevalo-Silva CA, Cao Y, Weng Y, et al. The effect of fibroblast growth factor and transforming growth factor-beta on porcine chondrocytes and tissue-engineered autologous elastic cartilage. Tissue Eng 2001;7:81-88. 21. Yang X, Chen L, Xu X, Li C, Huang C, Deng CX. TGF-beta/Smad3 signals repress chondrocyte hypertrophic differentiation and are required for maintaining articular cartilage. J Cell Bio12001;153:35-46. 22. Goldaale JA, Frenkel SR, Dicesare PE. Estrogen and osteoarthritis.Am J Orthof 2004;33:71-80. 23. Saggese G, Federico G, Cinquianta L. In vitro effects of growth hormone and other hormones on chondrocytes and osteoblast-like cells. Acta Paediatr Supp11993;82:54-60. 24. Yaeger PC, Masi TL, De Ortiz JL, et al. Synergistic action of transforming growth factor-beta and insulin-like growth factor-I induces expression of type II collagen and aggrecan genes in adult human articular chondrocytes. Exp Cell Res 1997;237:318-325. 25. Kondo S, Cha SH, Xie WF, Sandell LJ. Cytoldne regulation of cartilage-derived retinoic acid-sensitive protein (CD-RAP) in primary articular chondrocytes: suppression by IL-1, bFGF, TGF beta and stimulation by IGF-1.J Orthop Res 2001;19:712-719. 26. Fernihough JK, Richmond RS, Carlson CS, et al. Estrogen replacement therapy modulation of the insulin-like growth factor system in monkey knee joints. Arthritis Rheum 1999;42:2103-2111. 27. Chambers MG, Bayliss MT, Mason RM. Chondrocyte cytokine and growth factor expression in murine osteoarthritis. Osteoarthritis Cartilage 1997;5:301-308. 28. Guerne PA, Carson D, Lotz M. IL-6 production by human chondrocytes: modulation of its synthesis by cytokines, growth factors and hormones in vitro. J Immuno11990;144:494-505. 29. Badger AM, Blake SM, Dodds RA, et al. Idoxifene, a novel selective estrogen receptor modulator, is effective in a rat model of adjuvantinduced arthritis. J Pharmacol Exp Ther 1999;291:1380-1386. 30. Richette P, Dumontier MF, Francois M, et al. Dual effects of 17 betaoestradiol on interleukin 1beta-induced proteoglycan degradation in chondrocytes. Ann Rheum Dis 2004;63:191-199. 31. Lee YJ, Lee EB, Kwon YE, et al. Effect of estrogen on the expression of matrix metalloproteinase (MMP)-I, MMP-3, and MMP-13 and tissue inhibitor of metalloproternase-1 in osteoarthritis chondrocytes. Rheumatol Int 2003;23:282-288. 32. Hui W, Rowan AD, Richards CD, Cawston TE. Oncostatin M in combination with tumor necrosis factor alpha induces cartilage damage and matrix metalloproteinase expression in vitro and in vivo. Arthritis Rheum 2003;48:3404-3418. 33. Song YJ, Wu ZH, Lin SQ~Weng XS, Qiu GX. [The effect of estrogen and progestin on the expression of matrix metalloproteinases, tissue inhibitor of metalloproteinase and interleukin-1 beta mRNA in synovia of OA rabbit model]. Zhonghua Yi Xue Za Zhi 2003;83:498-503. 34. Schwartz Z, Gates PA, Nasatzky E, et al. Effect of 17beta-estradiol on chondrocyte membrane fluidity and phospholipid metabolism is membrane specific, sex-specific, and cell maturation-dependent. Biochem Biophys Acta 1996;1282:1-10. 35. Liehr JG, Roy D. Pro-oxidant and antioxidant effects of estrogens. In: Armstrong D, ed. Free radical and antioxidant protocols, pp. 425-435, Totowa, NJ: Humana, 1998:425-435. 36. Ayres S, Tang M, Subbiah MT. Estradiol-17 beta as an antioxidant: some distinct features when compared with common fat-soluble antioxidants. J Lab Clin Med 1996;128:367-375. 37. Claassen H, Schtinke M, Kurz B. Estradiol protects cultured articular chondrocytes from oxygen-radical-induced damage. Cell Tissue Res 2005;319:439-445.
CHAPTER 30 Potentials of Estrogens in the Prevention of Osteoarthritis: W h a t Do We Know and W h a t Questions are Still Pending? 38. Josefsson E, Tarkowski A. Suppression of type II collagen-induced arthritis by the endogenous estrogen metabolite 2-methoxyestradiol. Arthritis Rheum 1997;40:154-163. 39. Palmer RM, Hickery MS, Charles IG, Moncada S, Bayliss MT. Induction of nitric oxide synthase in human chondrocytes. Biochem Biophys Res Commun 1993;193:398-405. 40. Hoegh-Andersen P, Tanko LB, Andersen TL, et al. Ovariectomized rats as a model of postmenopausal osteoarthritis: validation and application. Arthritis Res Ther 2004;6:R169-R180. 41. Ren YX, Deng YZ. [An experimental study on effect of estrogen on osteoarthritis in female rats]. Zhongguo Xiu Fu ChongJian Wai Ke Za Zhi 2003;17:212-214. 42. Christgau S, Tanko L.B, Cloos PA, et al. Suppression of elevated cartilage turnover in postmenopausal women and in ovariectomized rats by estrogen and a selective estrogen-receptor modulator (SERM). Menopause 2004;11:508-518. 43. Rosner IA, Goldberg VM, Getzy L, Moskowitz RW. Effects of estrogen on cartilage and experimentally induced osteoarthritis. Arthritis Rheum 1979;22:52-58. 44. Tsai CL, Liu TK. Estradiol-induced knee osteoarthrosis in ovariectomized rabbits. Clin Orthop Relat Res 1993;291:295-302. 45. Turner AS, Athanasiou KA, Zhu CF, Alvis MR, Bryant HU. Biochemical effects of estrogen on articular cartilage in ovariectomized sheep Osteoarthritis Cartilage 1997;5:63-69. 46. Carlson CS, Loeser RF, Jayo MJ, et al. Osteoarthritis in cynomolgus macaques: a primate model of naturally occurring disease. J Orthop Res 1994;12:331-339. 47. Ham KD, Loeser RF, Lindgren BR, Carlson CS. Effects of long-term estrogen replacement therapy on osteoarthritis severity in cynomolgus monkeys. Arthritis Rheum 2002;46:1956-1964. 48. Ham KD, Carlson CS. Effects of estrogen replacement therapy on bone turnover in subchondral bone and epiphyseal metaphyseal cancellous bone of ovariectomized cynomolgus monkeys. J Bone Miner Res 2004;19:823-829. 49. Spector TD. Bisphosphonates: potential therapeutic agents for disease modification in osteoarthritis. Aging Clin Exp Res 2003;15:413-418. 50. Samanta A, Jones A, Regan M, Wilson S, Doherty M. Is osteoarthritis in women affected by hormonal changes or smoking? Br J Rheumatol 1993;32:366-370. 51. Cauley JA, Kwoh CK, Egeland G, et al. Serum sex hormones and severity of osteoarthritis of the hand. J Rheumato11993;20:1170-1175. 52. Spector TD, Perry LA, Jubb RW. Endogenous sex steroid levels in women with generalised osteoarthritis. Clin Rheumatol 1991;10: 316-319. 53. Sowers MF, Hochberg M, Crabbe JP, et al. Association of bone mineral density and sex hormone levels with osteoarthritis of the hand and knee in premenopausal women. Am J Epidemio11996;143:38-47. 54. Tsai CL, Liu TK, Chen TJ. Estrogen and osteoarthritis: a study of synovial estradiol and estradiol receptor binding in human osteoarthritic knees. Biochem Biophys Res Commun 1992;183:1287-1291. 55. Tanko LB, Bruun JM, Alexandersen P, et al. Novel associations between bioavailable estradiol and adipokines in elderly women with different phenotypes of obesity: implications for atherogenesis. Circulation 2004;110:2246-2252. 56. Von Muhlen D, Morton D, Von Muhlen CA, Barrett-Connor E. Postmenopausal estrogen and increased risk of clinical osteoarthritis at the hip, hand, and knee in older women. J l/VomensHealth Gend Based Med 2002;11:511-518. 57. Sandmark H, Hogstedt C, Lewold S, Vingard E. Osteoarthrosis of the knee in men and women in association with overweight, smoking, and hormone therapy. Ann Rheum Dis 1999;58:151-155. 58. Sahyoun NR, Brett KM, Hochberg MC, Pamuk ER. Estrogen replacement therapy and incidence of self-reported physician-diagnosed arthritis. Prev Med 1999;28:458-464.
401
59. Nevitt MC, Felson DT, Williams EN, Grady D. The effect of estrogen plus progestin on knee symptoms and related disability in postmenopausal women: The Heart and Estrogen/Progestin Replacement Study, a randomized, double-blind, placebo-controlled trial Arthritis Rheum 2001;44:811-818. 60. Maheu E, Dreiser RL, Guillou GB, Dewailly J. Hand osteoarthritis patients characteristics according to the existence of a hormone replacement therapy. Osteoarthritis Cartilage 2000;8:S33-$37. 61. Parazzini F. Menopausal status, hormone replacement therapy use and risk of self-reported physician-diagnosed osteoarthritis in women attending menopause clinics in Italy. Maturitas 2003;46:207-212. 62. Bettica P, Cline G, Hart DJ, Meyer J, Spector TD. Evidence for increased bone resorption in patients with progressive knee osteoarthritis: longitudinal results from the Chingford study. Arthritis Rheum 2002;46:3178-3184. 63. Dennison EM, Arden NK, Kellingray S, et al. Hormone replacement therapy, other reproductive variables and symptomatic hip osteoarthritis in elderly white women: a case-control study. Br J Rheumatol 1998;37:1198-1202. 64. Erb A, Brenner H, Gunther KP, Sturmer T. Hormone replacement therapy and patterns of osteoarthritis: baseline data from the Ulm Osteoarthritis Study. Ann Rheum Dis 2000;59:105-109. 65. Nevitt MC, Cummings SR, Lane NE, et al. Association of estrogen replacement therapy with the risk of osteoarthritis of the hip in elderly white women. Study of Osteoporotic Fractures Research Group. Arch Intern Med 1996;156:2073-2080. 66. Spector TD, Nandra D, Hart DJ, Doyle DV. Is hormone replacement therapy protective for hand and knee osteoarthritis in women? The Chingford Study. Ann Rheum Dis 1997;56:432-434. 67. Hannan MT, Felson DT, Anderson JJ, Naimark A, Kannel WB. Estrogen use and radiographic osteoarthritis of the knee in women. The Framingham Osteoarthritis Study. Arthritis Rheum 1990;33:525-532. 68. Zhang Y, McAlindon TE, Hannan MT, et al. Estrogen replacement therapy and worsening of radiographic knee osteoarthritis: the Framingham Study. Arthritis Rheum 1998;41:1867-1873. 69. Wluka AE, Davis SR, Bailey M, Stuckey SL, Cicuttini FM. Users of oestrogen replacement therapy have more knee cartilage than nonusers. Ann Rheum Dis 2001;60:332-336. 70. Cicuttini FM, Wluka AE, Wang Y, Stuckey SL, Davis SR. Effect of estrogen replacement therapy on patella cartilage in healthy women. Clin Exp Rheumato12003;21:79-82. 71. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321-333. 72. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMA 2004 ;291:1701-1712. 73. Bagger YZ, Tanko LB, Alexandersen P, Qin G, Christiansen C. Early postmenopausal hormone therapy may prevent cognitive impairment later in life. Menopause 2005;12:12-17. 74. Bagger YZ, Tanko LB, Alexandersen P, et al. Two to three years of hormone replacement treatment in healthy women have long-term preventive effects on bone mass and osteoporotic fractures: the PERF study. Bone 2004;34:728-735. 75. Ettinger B, Ensrud KE, Wallace R, et al. Effects of ultralow-dose transdermal estradiol on bone mineral density: a randomized clinical trial. Obstet Gyneco12004;104:443-451. 76. Eckstein F, Reiser M, Englmeier KH, Putz R. In vivo morphometry and functional analysis of human articular cartilage with quantitative magnetic resonance imaging-from image to data, from data to theory. Anat Embryol (Berl) 2001;203:147-173. 77. Garnero R Delmas PD. Biomarkers in osteoarthritis. Curr Opin Rheumato12003;15:641-646.
This Page Intentionally Left Blank
SECTION VII
Cardiovascular Perhaps no other area has changed or advanced as much since the previous edition of this book as the subject of cardiovascular disease, in particular the effects of hormones on the cardiovascular system. This section updates our current understanding and provides the most recent data and hypotheses regarding the effects of hormones on the cardiovascular system. Although an attempt has been made to be comprehensive in covering this topic, it is not possible to cover everything in depth. Some of these potential deficiencies will be addressed here. For a great number of years, we have been reassured by data showing consistently that there was a coronary protective effect of estrogen. This led to many studies attempting to elucidate the various mechanisms of cardio-protection. Many protective mechanisms were found that tended to validate the observational data pointing to the cardio-protective effects of estrogen. Thus it was a reaction of confusion and disbelief when the findings of various secondary prevention trials and the estrogen plus progesterone arm of the Women's Health Initiative (WHI) were released, which showed no protection and an effect of "early harm" (increased number of coronary events) in the first year in older women who were prescribed hormones. It is now quite clear that there is no beneficial effect of estrogen in women with established coronary disease, as shown in the various secondary prevention studies, as well as a real concern for the effect of early harm if standard dose therapy is used and the woman is not on a statin. As will be reviewed in this section, the mechanism of this early harm is likely to be due to plaque instability and rupture leading to coronary thrombosis. Nevertheless, there are compelling data to suggest that initiating hormones at the onset of menopause in young healthy women may be protective and does not lead to the early harm phenomenon witnessed in older women. These findings parallel the observational literature on which we based the belief of the cardio-protective effect of estrogen. Among the many publications on this particular point and the excellent reviews the reader will find in this section, the reader might focus on several recent publications showing the benefit of estrogen in younger women in both W H I and the Nurses' Health Study (1,2) and the lack of evidence for early harm in a young healthy population (3). This section begins with an in-depth epidemiologic review of cardiovascular disease after menopause by George I. Gorodeski. Next, Richard H. Karas briefly reviews the multiple mechanisms whereby estrogen may be cardio-protective. These mechanisms are all valid but may not be effective in older women with established disease and/or certain risk factors. Next, Ronald Krauss revisits the changes in lipids after menopause and the effects of hormones. Note again that in trials such as The Heart and Estrogen/ Progestin Replacement Study (HERS) and W H I , whereas beneficial lipid changes were observed with hormones, there was no reduction in coronary events in older women with atherosclerosis. The next three chapters deal with inflammatory markers and coagulation factors in postmenopausal women without and with hormonal therapy. In general, estrogen decreases most inflammation markers, with the marked exception of CRP, which may be due to first-passage hepatic effects of oral therapy. The other important difference is in the matrix mellanoproteinases (MMPs), which are all important in explaining plaque instability and rupture. MMPs may also be influenced, at least in part, by hepatic metabolism. Next,
404
SECTION VII CARDIOVASCULAR
Morris Notelovitz reviews the coagulation cascade and the influence of hormones, and Carolyn Westhoff summarizes the data regarding the influence of hormones on thrombosis and pulmonary embolism. There is no question that oral estrogen and the selective estrogen receptor modulators (SERMs), at least in standard doses, increase thrombosis risk. Nevertheless, in young healthy women this risk is relatively rare, given the low prevalence of these disorders. The risk may be further reduced with using small doses and/ or non-oral therapy. The next chapter, by Sven O. Skouby, reviews the topic of carbohydrate metabolism and the influence of hormones. The importance of this topic is immense given the increasing prevalence of obesity, metabolic syndrome, and diabetes. The data on the effects of hormones are confusing indeed and need to be interpreted with the knowledge of the type of hormonal preparation, dose, route of administration, and the characteristics of the woman being treated. We have shown recently that there are major differences in the effects of estrogen on a "normal" postmenopausal woman compared with a woman who has metabolic syndrome (4). The final two chapters in this section summarize our current understanding of the effects of hormones on coronary disease. First, Thomas B. Clarkson reviews data from the monkey model and very convincingly points to the importance of the timing of initiation of hormonal therapy, as discussed previously. Finally, Howard N. Hodis reviews the clinical trial data and comes to similar conclusions as Tom Clarkson; in essence stressing that data as presented in the past (WHI, for example) may not be taken at face value but should be interpreted in the context of the population studied and the totality of the literature available to us. I would be remiss in not mentioning two other points. Although discussed in several areas in this section, there are not specific chapters on hypertension and stroke. In previous editions, there was a dedicated chapter on hypertension. While not wanting to ignore the enormous epidemiologic importance of hypertension in women as reviewed in this section, the effects of hormones is relatively minor. In general, apart from a known idiosyncratic hypertensive response with standard doses of oral estrogen in a small percent of the population, most women receiving hormones have no change or even a reduction in blood pressure. A recent editorial makes these and other points (5). In terms of the risk of stroke, it appears that there is likely to be a small increase (20% to 30%) in the risk ofischemic (not hemorrhagic) stroke in older women receiving standard oral hormones. This increase is of borderline significance and appears to be confined to the older population, not to young women. Also, observational data have shown that lower than standard-dose therapy does not increase the risk of stroke (6).
References 1. Hsia J, Langer RD, Manson JE, et al. Conjugated equine estrogens and coronary heart disease: the Women's Health Initiative. Arch Intern Med 2006;166:357-365. 2. Grodstein F, Manson JE, Stampfer MJ. Hormone therapy and coronary heart disease: the role of time since menopause and age at hormone initiation. J Womens Health (Larchmt) 2006;15:35-44. 3. Lobo R. Evaluation of cardiovascular event rates with hormone therapy in healthy, early postmenopausal women: results from 2 large clinical trials. Arch Intern Med 2004;164:482-484. 4. Chu MC, Cosper P, Nakhuda G, Lobo R. A comparison of oral and transdermal short-term estrogen therapy in postmenopausal women with metabolic syndrome. Fertil Steri12006;86:1669-1675. 5. Lobo R. What is the effect of estrogen on blood pressure after menopause? Menopause 2006;13:331-333. 6. Grodstein F, Manson JE, Colditz GA, et al. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease.Ann Intern Med 2000;133:933-941.
~ H A P T E R 31
Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women EIRAN Z E V GORODESKI
GEORGE I. GORODES KI
Departmentof Cardiovascular Medicine, Cleveland Clinic, Cleveland, OH 44195 Department of Reproductive Biology, CASE (Case Western Reserve) University, Cleveland, OH 44106
I. B A C K G R O U N D
women undergo cardiovascular diagnostic and surgical interventions. The situation is further complicated by the lack of clear understanding of the impact of menopause and hypoestrogenism on the cardiovascular system. The hypothesis that estrogens confer cardiovascular protection remains unchallenged, but its translation into clinical practice in postmenopausal women has not been successful. Recent randomized controlled studies researching the possible cardiovascular protective role of estrogen in postmenopausal women have concluded that estrogens should not be considered for that indication alone. However, these conclusions have been challenged, and many questions remain unanswered. The largest growing segment of the U.S. population is women over the age of 75, and more than one-fourth of the female population is in their postreproductive years. Although certain forms of CVD can be treated medically or surgically, the most effective means to decrease the impact of CVD on women's health is by modifying the contribution of specific factors that increase CVD risks. The major contribution to decreasing morbidity and mortality of CVD in men was achieved by relatively simple measures such as changing lifestyle habits, eliminating cigarette smoking, and
Cardiovascular disease (CVD) is the leading cause of morbidity and mortality in women, outweighing other causes such as cancer, cerebrovascular disease, lung disease, accidents, diabetes mellitus, and infectious diseases. Understanding the epidemiology, pathophysiology, and risk factors of the disease is important for developing an evidencebased program to improve clinical outcome of the disease in women. CVD in women remains a focus of multidisciplinary research interest. Recent studies have established genderrelated differences in the prevalence of cardiovascular risk factors and clinical management of CVD. The observations suggest that CVD differs among women and men, thereby confirming the paradigm that information obtained from studies involving men should not be directly applied to women. Only recently have large-scale studies begun to include women in equal numbers to men. However, many current clinical decisions continue to be made based on past studies that targeted primarily men. This can probably explain the fact that although more women than men ultimately die of heart disease, a greater proportion of men than TREATMENT OF THE POSTMENOPAUSAL W O M A N
405
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
406
GORODESKIAND GORODESKI
lowering plasma lipid levels. Similar effects were recently reported in women, suggesting that early detection of risk factors and an active approach to prevent or lower the impact of risk factors could positively affect the outcome of CVD in women. This chapter describes the epidemiology and pathogenesis of CVD in women and the risk factors involved, with special emphasis on gender-related differences and the relation between estrogen deficiency and heart disease.
II. EPIDEMIOLOGY OF CVD IN W O M E N In 2002, a total of 1,244,123 deaths were recorded among U.S. women, 40% of which were due to CVD, significantly outweighing other causes of death (Fig. 31.1). Accordingly, a woman in the United States has a 29% lifetime chance of dying of CVD compared with lung cancer, 5%; breast cancer, 3%, and colorectal cancers, 2% (1-5). Age-adjusted mortality rates show that CVD is the leading cause of death in women after the age of 60 (Fig. 31.2). However, about one-third of deaths from CVD in women were premature, in women younger than 65 years, accounting for more than 100,000 cases of death per year (1). More than 55% of CVD
All causes
deaths were due to ischemic heart disease, which is commonly referred to as coronary heart disease (CHD) (Fig. 31.3), and is the single leading cause of CVD deaths in women and in men (1). In 2002 the incidence of C H D began to increase after age 45 and continued to increase thereafter (Fig. 31.4). Heart disease is the third leading chronic morbid condition in women, following arthritis and back and neck conditions, affecting an estimated 2.2 million women (2). Heart disease is also the leading cause of physician office visits among women, about 5.6 million annual visits, and the leading cause for hospitalizations of women, about 1.1 million annual hospitalizations (3). Modeling the Framingham Heart Study cohort epidemiologic data into a life-expectancy analysis suggests that 50-year-old females live 1.2 years after myocardial infarction (MI), and having ever suffered acute MI reduces life expectancy by 13 years (6). Therefore, C H D morbidity and mortality in women carry a worse clinical prognosis than cancer. However, because women are relatively immune from C H D at an early age, and because cancers precede CVD and C H D by about 1 decade, the fact remains that neither women nor their physicians fully appreciate the magnitude of CVD and C H D in women. In women 25 years and older, only 8% cite CVD as their greatest health
Death R a t e - Cancer and Heart Disease per 100,000 Population
l
0000
Cardiovascular
o
Cancer Lung * Injuries "* Diabetes Alzheimer's Dis Pneumonia "'" Kidney Dis Septicemia
HIV
ill i i i i i i
5000
_
9
v -"
M e n - Cancer Women - Cancer M e n - Heart Disease WomenHeart Disease
10 000
/ 5000
Women Men
I
,
0
,
I
i
i
I
200
400
600
800
!
1000
1200
Death Rates (per 100,000 Population)
FIGURE 31.1 Major causes of deaths in the United States. *Chronic lower respiratory diseases; **includesnonintentional and intentional accidents and suicide; ***pneumoniaand influenza. Modified from ref. 1, with permission.
< 20
20-39
40-59
60-79
1
> 80
Ages
FmURE 31.2 Age and sex adjusted mortalityrates from cancerand heart disease. Modified from ref. 1, with permission.
407
CHAPTER31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women
Deaths
Per 100,000
Population
2000
All causes
Stroke
o
-
9 AA Men CMen - 9 AAWomen -[]C Women
1500
Isch Heart Dis
-o-
1000 /
Heart Failure
500
/o
HT Heart Dis
35-44
Women Men
Hypertension
k
1
I
100
200
300
55-64
65-74
75-84
Age
Atherosclerosis
0
45-54
c~
FIGURE 31.4 Age and sex adjusted mortality rates from CHD in African Americans (AA) and Caucasians (C). Modified from ref. 1, with permission. 400
Death Rates (per 100,000 Population)
FIGURE 31.3 Major cardiovascular deaths. HT, hypertensive; Isch, ischemic. Modified from ref. 1, with permission.
concern and less than 30% report that their physician has discussed cardiovascular risk with them (7). Even when women were identified as having risk factors for CVD, there was lower utilization of standard accepted therapy by physicians as compared with men (8-10).
III. PATHOGENESIS OF CARDIAC ISCHEMIA Proper function of the myocardium depends on supply of oxygen and nutrients. Decreased supply is usually caused by decreased arterial blood flow due to coronary artery occlusion. Understanding the processes of coronary artery occlusion is the key in identifying risk factors that cause, sustain, and augment cardiac ischemia.
A. Injury and Inflammation The coronary arterial tree is susceptible to flow injuries due to its special branching pattern. Disturbances in blood flow occur at bending points near branching sites of large and medium-sized arteries in areas with high hemodynamic shear stress, and they cause a perpetual process of intimal insult, attraction of inflammatory cells, inflammation, and endothelial damage (11). The earliest atherosclerotic lesions
can be observed in utero as deposits of cholesterol esters (fatty streaks) and typically consist of macrophages and T-cells embedded in a thin layer of lipids in the intima of large muscular arteries (12). Fatty streaks already can be found in the coronary arteries in infancy (13-15). Intimal accumulation of low-density lipoprotein cholesterol (LDL-C) and its oxidation augment expression of surface molecules, such as vascular-cell adhesion molecule-I, and stimulate adhesion and migration of monocytes and lymphocytes into the subendothelial space (16,17). Monocytes that enter the newly formed plaque differentiate into macrophages; they scavenge oxidized LDL-C and dead-cell remnants and trigger an inflammatory cascade. Activated macrophages release an array of cytokines and growth factors that further attract monocytes, lymphocytes, and platelets, thereby adding to the pool of effector molecules that expand and perpetuate the inflammatory response. As this cycle is repeated, the plaque develops a fatty core covered by fibrous matrix that stabilizes the structure (18). Inflammation characterizes all phases of atherothrombosis and provides a critical pathophysiologic link between early lesion formation and plaque rupture leading to occlusion and infarction. In early stages of atheromatous lesion development, proinflammatory cytokines potentiate the expression of adhesion molecules on vascular endothelial cells; they stimulate recruitment of leucocytes to the arterial wall and the migration of monocytes into the subendothelial space. Mononuclear ceils in concert with intrinsic vascular cells release growth factors and cytokines that stimulate the proliferation of smooth muscle cells. Cytokines produced by the macrophages are also released systemically and induce
408
GORODESKIAND GORODESKI
the release of interleukin (IL)-6, which stimulates the production of acute-phase reactants by the liver and adipose tissue, including C-reactive protein (CRP), serum amyloid A, and fibrinogen. The four key cells that interact to accentuate formation of the atheroma are the monocytes-macrophages, platelets, endothelial cells, and smooth muscle cells. The main perpetuating factor of atherosclerosis is LDL-C, which is modified by oxidation, glycation, and enhanced local inflammation. After becoming trapped in the arterial wall, LDL-C particles become progressively more oxidized, form lipid peroxides, and facilitate accumulation of cholesterol esters. Modified LDL-C is chemotactic for circulating monocytes and stimulates proliferation of macrophages already in the lesion, as well as formation of fibrointimal thickening and of a capsule over an existing lesion (18,19). This cycle of events stimulates growth and thickening of the plaque, whereby the arterial wall responds by gradually dilating to maintain the diameter of the vessel lumen ("remodeling").
B. The Coronary Syndromes One consequence of this process is formation of a lipidrich plaque, narrowing of the lumen of the coronary artery, and stiffness and dysfunction of the vessel, resulting in stenosis. Coronary artery stenosis is a chronic process associated with maladaptation to exercise, resulting in negative oxygen balance and the classic clinical presentation of stable angina pectoris. Another scenario is the release by macrophages of tissue metalloproteinases that act in concert with other extracellular proinflammatory factors to degrade the fibrous cap and render the plaque vulnerable to rupture (20). In contrast to chronic stenosis, rupture of the atherosclerotic plaque resulting in coronary thrombosis (21) produces acute myocardial ischemia, collectively referred to as acute coronary syndromes. Coronary thrombosis usually results from destabilization and rupture of previously silent atherosclerotic plaque. Vulnerable or unstable plaque is the product of local inflammatory processes within the artery. Activated macrophages,T-cells, and mast cells produce cytokines, proteases, coagulation factors, free radicals, and vasoactive molecules, which destabilize the fibrous cap of the atherosclerotic plaque. Activation of matrix metalloproteinases and cysteine proteases further promote plaque rupture (22). Upon rupture, the thrombogenic lipid core of the plaque becomes exposed to the blood, thereby stimulating adhesion and activation of platelets and release of adenosine diphosphate, thromboxane A2, and serotonin, and triggering the recruitment and activation of surrounding platelets. Platelets exteriorize the glycoprotein IIb/IIIa receptor that binds fibrinogen and leads to the linking of adjacent platelets, thereby leading to a formation of hemostatic platelet
plug (23). In some patients, thrombosis in coronary arteries can result from superficial erosion of the intima without frank plaque rupture. This occurs as a result of apoptosis of endothelial cells and proteinase production in response to inflammatory mediators. Vasospasm of coronary arterioles is an additional mechanism of acute ischemia. Atherosclerosis impairs endothelial integrity and function and can result in diminished production and release of vasodilators such as nitric oxide, thereby increasing the vascular resistance of the coronary tree. Changes in the rheologic properties of the blood, and in particular increased viscosity, can contribute to increased coronary resistance and diminished blood flow in the coronary vessels (24,25). The clinical presentation of patients with acute coronary syndromes varies and is categorized according to electrocardiographic (ECG) and biochemical criteria. The spectrum of disease ranges from partial to complete thrombotic occlusion of the coronary artery. Patients with partial occlusion may present with unstable angina (UA) or non-ST elevation MI (NSTEMI). Unstable angina is ischemia without myocardial necrosis. Patients present with ischemic symptoms but without ECG changes or positive myocardial markers on blood testing. Patients with partial coronary artery occlusion who progress to myocardial necrosis usually present with N STEMI, defined as lack of ST elevation but with elevated myocardial markers. Complete arterial occlusion usually results in ST elevation MI (STEMI), defined by ST-elevations and positive myocardial markers (26). A majority of patients with STEMI ultimately develop Qzwaves (i.e., Q_w-MI) on ECG, implying transmural infarction, whereas in some no Qzwaves are evident (i.e., non-Qw-MI), implying partial transmurality (Fig. 31.5). Although non-Qw-MI may have infarct that is smaller than Q w-MI, it is often associated with higher incidence of late cardiac events as compared with QMI, including recurrent ischemia angina or MI. Long-term survival is similar for both types of MI (27).
Acute Coronary Syndrome
No ST Elevation
UnstableAngina
ST Elevation
No-Qw-MI ...........................
Qw-MI T
FIGURE 31.5 Classificationof acute coronarysyndromes.MI, myocardial infarction.Modifiedfrom res 26.
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women
C. Gender Differences Fatty streak lesions of the coronary arteries are found in about half of infants less than I year of age, and their incidence in males and females is the same (28). Most lesions of early infancy reverse, but those that persist into childhood do not reverse (15,29). A few reports described coronary artery intimal layers that are thicker in male infants than in females, suggesting an inborn sex difference in the structure of the coronary artery that renders males more susceptible to ischemic heart disease (30). Other reports, however, have found no such sex difference in fetuses and children up to age 19 (31,32). In certain ethnic groups with unusually high prevalence of CAD, coronary artery intimal layers are thicker in males than in females (30). At all ages after puberty the incidence of fibrointimal fatty streak lesions is lower in women than in men (33), and the clinicopathologic patterns of C H D also differ between genders. Women are more likely to have nonobstructive CHD and single-vessel disease, whereas men are more likely to have obstructive multivessel or left main C H D (34). These observations suggest that female-specific atherogenesis-protective processes operate from puberty onward. In contrast, women have higher incidence of acute coronary ischemia due to vasoconstriction of the coronary arterial tree than do men (35). Chronic coronary ischemia due to severely stenotic vessels is frequently associated with formation of collaterals, which is a compensatory mechanism of arterial blood supply to the affected cardiac muscle (36). Relatively little is known about gender differences in the formation of cardiac collaterals. More is known about gender differences related to cardiac protection from the ischemia-reperfusion syndrome. Reperfusion of previously anoxemic tissues can lead to irreversible cell damage and cardiac death (37). Cells become necrotic and attract and activate polymorphonuclears and platelets, which could lead to impaired blood flow in the microvasculature, recurrent thrombosis, increased vascular resistance, accelerated necrosis, and cell death (38,39). Studies in women (40,41) and in animal models (42-47) have shown that estrogens protect the myocardium from the detrimental consequences of reperfusion, regardless of the status of the coronary circulation. Thus, it is apparent that gender-related differences in the pathogenesis of coronary atherosclerosis could also lead to gender differences in the incidence, prevalence, and clinical manifestations of CHD.
IV. RISK FACTORS FOR CARDIAC ISCHEMIA IN W O M E N Cardiac ischemia is the end result of genetic, constitutive, behavioral, and environmental factors that act in concert throughout life and result in progressive cardiac damage. Known risk factors for C H D in women include obesity,
409
abnormal plasma lipids, hypertension, diabetes mellitus, cigarettes smoking, sedentary lifestyle, adverse genetic background, increased blood viscosity, stress, and autonomic imbalance, and estrogen deficiency. Usually more than one risk factor can be identified, but some risk factors have a stronger impact on the prevalence of the disease and are therefore considered "independent" or even "causative." Better understanding of how risk factors cause or affect C H D may direct caregivers in formulating more effective intervention protocols.
A. Obesity Obesity is a worldwide epidemic (48,49). In the United States, more than 60% of adults are either overweight or obese (50). Current clinical definitions of obesity in women are body mass index (BMI, body weight [in kg] divided by height [in m] squared), whereby a BMI -> 25 kg/m 2 is defined as overweight, and >- 30 kg/m 2 as obese; fat mass >35% of total body weight; and waist circumference -> 88 cm (37 inches) (51-53). Obese premenopausal women tend to have gluteal (hip or thigh) type obesity, also termed gynecoid obesity. After menopause the fat distribution changes to centralized (waist, androgenic, or truncal) obesity (53,54). The waist-to-hip ratio (WHR) or waist-to-thigh ratio (WTR) correlates with CHD. Increased W H R remains positively correlated with C H D even after controlling for hypertension, glucose intolerance, blood lipids, smoking, and BMI (55). Moreover, the gender-associated risk disappears after controlling for
WHR (53). Abdominal fat is located in two major compartments, subcutaneous and intraperitoneal (visceral). The subcutaneous adipose tissue is a much larger compartment than the intraperitoneal fat, and it is composed of truncal and gluteofemoral adipose tissues. The intraperitoneal adipose tissue consists of omental and mesenteric fat. The subcutaneous truncal fat and the intraperitoneal omental and mesenteric fat are related to metabolic risk (56). The mechanism by which centralized obesity confers an increased C H D risk is not completely understood. It could be the result of differential response of upper body versus lower body adipocytes to free fatty acid mobilization (52,57). Obesity develops as a result of positive energy balance, intake being higher than expenditure. Females have a greater prevalence of obesity as compared with males. The differences could be accounted for by a higher 24-hour energy expenditure in men compared with women. Men have higher basal and sleeping metabolic rates, whereby the basal rate accounts for 60% to 70% of total energy expenditure (58). Obesity is both an independent and an intermediary risk factor for cardiovascular mortality. Mortality begins to increase with BMI levels greater than 25 kg/m 2, and it
410 increases most dramatically as BMI levels surpass 30 kg/m 2 (relative risk [RR] 1.5) (59). Obesity is also a modifier of other risk factors including hypercholesterolemia, hypertension, Type 2 diabetes, hypertension, obstructive lung disease, sleep apnea, stroke, asthma, back and lower extremity weight-bearing degenerative problems, osteoarthritis, several forms of cancer, and depression (60). The mechanisms include modulation of atherogenic dyslipidemia, insulin resistance, proinflammatory state, and prothrombotic state. The age and smoking-adjusted RR of CHD is 1.8 in mild to moderately obese women and 3.3 among severely obese women (55). Overall, 70% of CHD in obese women and 40% of CHD in all women could be attributed to being overweight (55). The majority of obese persons who develop CVD have a clustering of risk factors described as the metabolic syndrome (61). This syndrome consists of atherogenic dyslipidemia (elevated serum triglycerides, apolipoprotein B, and LDL-C, plus low high-density lipoprotein cholesterol ([HDL-C]); hypertension; hyperglycemia and insulin resistance; and prothrombotic and proinflammatory states. Metabolic syndrome is defined if a person has three of the following five features (62): 1. Increased waist circumference (abdominal obesity >-102 cm in men and >-88 cm in women) 2. Elevated triglycerides (>- 150 mg/dL) 3. Reduced HDL-C (-<40 mg/dL in men and -<50 mg/dL in women) 4. Elevated blood pressure (>-130/85 mm Hg, or on treatment for hypertension) 5. Elevated fasting glucose (>- 100 mg/dL) Patients with diabetes (fasting glucose >- 126 mg/dL) are said to have the metabolic syndrome if two other features are present. Diagnosis of the metabolic syndrome is a useful
GORODESKI AND GORODESKI
clinical tool in identifying patients at risk for development of diabetes, CVD, and CHD (Fig. 31.6). 1. PREVALENCE OF OBESITY AND THE METABOLIC SYNDROME
The prevalence of obesity in the United States is escalating (Fig. 31.7). Two-thirds of individuals living in the United States are overweight, and 5% are obese (63); about 57% of Caucasian females and 77% of African-American females are overweight (64). Obesity is identifiable already in childhood: between 1976-1980 and 1999-2002 the prevalence of obesity in U.S. children doubled (65). In 2002 the prevalence of overweight Caucasian females aged 6 to 19 was 13%; of African-American and American-Mexican females, 19%. These levels increased to 60% in adults (Fig. 31.8) (66), and the age-adjusted prevalence of overweight and obesity were 65% and 30%, respectively (67) (see Fig. 31.8). AfricanAmerican females tend to be more overweight and obese than Caucasian or Mexican-American females (see Fig. 31.8). Equally disturbing are the age-related trends of the metabolic syndrome. In the United States, about 2% of 12to 19-year-old adolescent women can be characterized as having the metabolic syndrome (68); the prevalence rises to
Prevalence of Obesity Percent of Population 35
30
25
20
_- / /J/
15
10
FIGURE 31.6 CHD, CVD, and total mortality stratified by history of CVD, diabetes, and the metabolic syndrome. Modified from ref. 5.
.........
1 ..............
1960-62
I
I
I
1971-74
1976-80 Years
t988-94
. . . . . . . . . . .
!
1999-02
FIGURE 31.7 Prevalence of obesity and the metabolic syndrome in the United States. Modified from ref. 5.
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women
411
TABLE 31.1 Interpretations of Plasma Levels of LDL Cholesterol, Total Cholesterol, and Triglycerides Relative to C H D Risk*
Percent of Women 60 Caucasians African-Americans Mexican-Americans
LDL cholesterol
Total cholesterol
Triglycerides
40
Desirable (optimal) Near optimal Borderline h i g h High Very high
3)
0 Age:
6-11
12-19
_> 20
Overweight
_> 20
<100 100-129 130-159 160-189 ->190
<200 200-230 ->240
<150 150-199 200-499 ->500
*All values in mg/dL. CHD, coronary heart disease; LDL, low-density lipoprotein. Data from ref. 62.
Obese
FIGURE 31.8 Age and ethnicity/race-dependent prevalence of overweight and obesity in women. Modified from ref. 5.
7% among women ages 20 to 29, and it escalates to 42% for ages 60 and older (69,70).
B. Abnormal Plasma Lipids Unfavorable plasma lipids are an independent causative factor for the development of CAD (62). LDL-C comprises 70% of the total serum cholesterol. It contains a single apolipoprotein (apo B) and is the major atherogenic lipoprotein. LDL-C is the primary target of cholesterol-lowering therapy, and most clinical trials show the efficacy of LDLlowering therapy for reducing risk for CHD. Because LDLC levels < 100 mg/dL are associated with a very low risk for CHD, they are referred to as optimal. Near-optimal LDL-C levels of 100-129 mg/dL are associated with some degree of atherogenesis, whereas higher levels of 130-159 mg/dL are associated with significant atherogenesis and are referred to as borderline. At high (160-189 mg/dL) and very high levels (-> 190 mg/dL), atherogenesis is markedly accelerated. Similar definitions for total cholesterol and triglycerides are shown in Table 31.1. With regard to HDL-C, studies have shown a continuous increase in C H D risk as H D L - C levels decline, but without a deflection point in the C H D versus H D L - C curve. Therefore, suggested low (i.e., undesirable) H D L cholesterol levels are defined as <40-50 mg/dL and high (i.e., desirable) levels as >60 mg/dL (62). An increase of 1% in plasma total cholesterol or LDLC increases the risk of C H D by 2%, and a decrease of 1% in H D L - C increases the risk by 2% to 4.7% (62). Elevated plasma triglycerides are also associated with increased risk
of C H D but are not an independent risk factor. Plasma H D L - C is an important predictive C H D risk factor in women, and elevated H D L - C levels above 55 mg/dL can modify the adverse effects of high total cholesterol and LDL-C levels (62). Unfavorable profiles of plasma lipids are found in persons with congenital and acquired hyperlipidemias and dyslipidemias, and these individuals are at an increased risk to die of C H D at a young age. Unfavorable plasma lipids can be also found in otherwise healthy asymptomatic children, adolescents, and young adults. In the Pathobiological Determinants of Atherosclerosis in Youth study (PDAY), investigators looked at coronary lesions of persons aged 15 to 34 years who died of accidental injury, homicide, or suicide (71). The extent of both fatty streaks and raised lesions (advanced lesions and fibrous plaques) in the right coronary artery was associated positively with non-HDL-C lipids levels, hypertension, impaired glucose tolerance, and obesity and associated negatively with HDL-C. After adjusting for race, sex, and smoking, unfavorable lipid profile was positively associated with both fatty streaks and raised lesions in the right coronary artery at ages 15 through 34, and lesions more than tripled through these ages (Fig. 31.9). Unfavorable plasma lipids are common in the adult population and may worsen with age. In the Framingham Study, levels of LDL-C in women increased after the age of 50 to 55. In contrast, levels of H D L - C in women remained essentially unchanged after the fifth decade (72). Other studies have shown that postmenopausal women have greater increases in LDL-C and greater decreases in H D L - C than age-adjusted premenopausal controls (73). In addition, acute changes after the menopause (within 6 months after cessation of menstrual periods) were observed in total cholesterol levels (6% increase), triglycerides (11% increase), and LDLC (10% increase). H D L - C levels have decreased gradually by 6% 2 years after menopause (74). One-third of postmenopausal women in the United States are at moderate
412
GORODESKI AND GORODESKI
A: Fatty Streaks
B: Raised Lesions
!--7 favorablelipids
I---] favorable!ipids
unfavomblelipids
P e r c e n t of W o m e n 6O
Caucasians African-Americas Mexican-Americans
4O
91~ unfavorableIlpids
2O
0
o
o
o T-
15-19
2 0 - 2 4 25-29
Age
30-34
15-19
2 0 - 2 4 25-29
Age
O P-
30-34
FIGURE 31.9 The meanextent(+ SE) of fattystreaksand raisedlesionsin the right coronaryarteryby 5-yearage groups in personswith a favorablelipoprotein profile (non-HDL-C < 108 mg/dL; HDL-C > 60 mg/dL) comparedwith an unfavorablelipoprotein profile (non-HDL-C > 150 mg/ dL, HDL-C < 43 mg/dL).Valueswere adjustedfor race,sex,and smoking. Modified from res 71.
risk and one-fourth at high risk of C H D as indicated by H D L - C levels lower than 45 mg/dL (62). The mechanism by which aging adversely affects lipid profile is unclear. LDL-C levels are likely increased as a result of age-related reduced capacity for its removal, mediated via reduced hepatic LDL receptor (75). Menopause-related changes in lipids could be the result of estrogen deficiency (see later discussion). Women with chronic anovulation, including those with polycystic ovarian syndrome (PCOS), have increased levels of LDL-C and decreased levels of HDL-C. Some of these patients also have insulin resistance and hyperandrogenemia (76,77).
1. P R E V A L E N C E OF H Y P E R L I P I D E M I A
The prevalence of high total serum cholesterol (->240 mg/dL) in women in the United States has decreased over the last 4 decades, but more than 18% still present with markedly elevated levels (78). Unfavorable plasma lipids can be detected already among children and adolescents ages 4 to 19 years. Caucasian women, in particular, have significantly higher average total cholesterol and LDL-C than do men and African-American women (Fig. 31.10). In 2002, 53% of Caucasian women had plasma total cholesterol ->200 mg/dL and 20% ->240 mg/dL; 43% had LDL-C ---130 mg/dL, and 14% H D L - C -< 40 mg/dL (see Fig. 31.10). These levels expose women to moderate to high degree of C H D risk (62).
O I-
AI
0
,
rl --I
o vI
(.~ --I I::I :I:
FIGURE 31.10 Prevalenceofhyperlipidemia in women. Modified from
ref. 5.
C. Hypertension Hypertension is a risk factor for CHD, stroke, and congestive heart failure, with the strongest impact in women (5,62). Hypertension is defined as systolic blood pressure ->140 mm Hg or diastolic blood pressure ->90 mm Hg. Both elevated systolic and diastolic blood pressure levels are independent risk factors, directly and positively associated with acute coronary events as well as with mortality from all cardiovascular causes (5,62). In 2002, hypertension was the direct cause of death of 4.7% of Caucasian and 22.8% of African-American women (1). African Americans develop hypertension earlier in life, and their average blood pressures are much higher. As a result, compared with Caucasians, African Americans have a 1.5 times greater rate of heart disease death (1,5,62). Yet 30% of persons with hypertension are unaware of having the disease or of its consequences (5,62). Hypertension is commonly diagnosed in patients who have other C H D risk factors such as obesity, unfavorable plasma lipid profile, and diabetes mellitus. The combination of hypertension and diabetes mellitus is of particular importance in women because diabetic patients frequently (50%) have hypertension and the prevalence of diabetes mellitus in hypertensive patients is also high (15% to 18%) (79).These women have also higher incidence of other atherogenic risk factors including dyslipidemia, hyperuricemia, and elevated fibrinogen levels (5,62). The association between obesity and hypertension could be the result of sympathetic nervous and renin-angiotensinaldosterone system activation because sodium and water
413
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women retention and blood pressure elevations are observed in obese individuals. Visceral obesity and the ectopic deposition of adipose tissue may be important in the activation of these systems, and in the damage of target organs that ensues (80). Due to its high prevalence, the attributable risk of hypertension for C H D is the highest of the known risk factors, reaching 55% as compared with 5% of diabetes mellitus and 2.8% of cigarettes smoking (79,81). Controlling hypertension reduces significantly the risk of CVD, but the reductions in acute coronary risks are less than expected, indicating that hypertension has had a chronic interrelated effect with other factors (5,62). The degree of risk imposed by hypertension is now recognized as a continuum and is associated with age. Beginning at a blood pressure of 115/75 mm Hg, CVD risk doubles with each increment of 20/10 mm Hg. Individuals who are normotensive at 55 years of age have a 90% lifetime risk for developing hypertension, and individuals with a systolic blood pressure of 120-139 mm Hg or a diastolic blood pressure of 80-89 mm Hg should be considered pre-hypertensive and require healthpromoting lifestyle modifications to prevent CVD (82).
1. PREVALENCE OF HYPERTENSION
Before the age of 55, the age-adjusted prevalence of prehypertension and hypertension is greater in men than in women, but after that age a much higher percentage of women have hypertension than men, and the prevalence of the disease increases further with age. Forty percent of persons in the United States are normotensive, 31% are prehypertensive, and 29% are hypertensive. Stated differently, two-thirds of persons are either pre-hypertensive or hypertensive (5). Hypertension is two to three times more common in women taking oral contraceptives, especially in obese and older women (5). In 1999 to 2000, the prevalence of hypertension in the United States at ages 20 to 74 was 42% in African-American women compared with 28% in Caucasian and Latino females (78).
mortality, and the effects were aggravated by cigarettes smoking, hypertension, and obesity (84). Diabetes increases the incidence of MI (85), 28-day and 1-year mortality after MI (86), and the 2-year mortality rate in patients admitted with unstable angina or non-Qcwave MI (87). Persons with diabetes often have other associated CVD risk factors, including hypertension, dyslipidemia and obesity (88). In women, diabetes mellitus accelerates atherosclerosis and increases the risk of acute coronary ischemia (79). Impaired glucose tolerance and insulin resistance often occur jointly with hypertension obesity and unfavorable plasma lipid profile (79). This association has suggested that there are pre-existing genetic traits or metabolic factors in the causal pathway common to these conditions (89). Impaired handling of blood glucose, either as pre-diabetes (fasting glucose 100-125 mg/dL) or overt diabetes (fasting glucose - 126 mg/dL) can be associated with coronary atherosclerosis already at a young age. In the PDAY study (90), investigators found that an elevated glycosylated hemoglobin concentration (corresponding to an average blood glucose concentration of-> 150 mg/dL in the previous 2 or 3 months) was associated with increased fatty streaks and raised lesions of the right coronary artery. Most of the increases in coronary atherosclerosis occurred after age 25 (Fig. 31.11). A number of mechanisms were proposed by which hyperglycemia induces atherosclerosis. Transient hyperglycemia is associated with overproduction of superoxide by the mitochondrial electron transport chain (91). It activates signaling cascades culminating with decreased bioavailabil-
A: Fatty Streaks
B: Raised Lesions
50r normal g-Hb
BB elevatedg-Hb
normal g-Hb
BB elevated g-Hb
D. D i a b e t e s Mellitus Impaired handling of glucose, including insulin resistance insulin, insulin deficiency, and overt diabetes mellitus, is both an independent risk factor for CVD and a causative factor of premature mortality from C H D (5,62). Blood glucose levels are a graded independent risk factor for CVD even within a relatively normal blood sugar range (79). Women with diabetes mellitus lose the female-related advantage of CVD over men (83). The age-adjusted RR for C H D in women with diabetes mellitus is 3-7 (62). In the observational Nurses Health Study, diabetes was associated with a markedly increased risk of fatal and nonfatal MI, stroke, and all-cause
15-19
20-24 25-29 30-34 Age
15-19
20-24 25-29 30-34
Age
FIGURE 31.11 The mean extent (+ SE) of fatty streaks and raisedlesions in the right coronaryartery by 5-year age groups and normal glycosylated hemoglobin concentration (normal g-Hb < 8%) comparedwith elevated glycosylatedhemoglobin concentration (elevatedg-Hb >- 8%).Valueswere adjusted for race and sex.Modified from res 90, with permission.
414
GORODESKI AND GORODESKI
ity of nitric oxide and prostacyclin, increased synthesis of vasoconstrictor prostanoids and endothelin, and increased vascular permeability that leads to endothelial damage and atherosclerotic plaque formation. Persistent hyperglycemia stimulates resident macrophage activation and upregulation of CD36 protein with increased uptake of oxidized lowdensity lipoprotein, increased macrophage-induced matrix metalloproteinase (MMP) expression, adventitial inflammation, and vasa vasorum neovascularization (92,93). The ensuing inflammatory microangiopathic process could lead to plaque rupture and coronary thrombosis. In this regard hyperglycemic endothelial damage mirrors the process of classical atherosclerosis (94,95). Hyperglycemia can also adversely affect platelet function, resulting in an added risk for thrombosis. In addition, hyperglycemia can affect directly cardiac myocytes and cause cardiac autonomic neuropathy and cardiomyopathy (5,62). 1. PREVALENCEOF DIABETES MELLITUS
The prevalence of diagnosed type 2 diabetes mellitus, which accounts for more than 90% of all cases of diabetes, has increased during the last 3 decades. By 2025 it is projected to involve more than 5% of the world population (96). The increase in prevalence of diabetes parallels those of obesity and the Metabolic Syndrome (97). In 2002, among Caucasian, African-American, and MexicanAmerican women, pre-diabetes was diagnosed in 4.6%, 5.9%, and 7.2%, respectively. In those groups, diabetes was diagnosed in 4.7%, 12.6%, and 11.3%, respectively (5). Currently it is estimated that 40% of U.S. females ages 40 to 74 have pre-diabetes (5).
E. Cigarette S m o k i n g Cigarette smoking is the single most important preventable risk factor of CVD (99,100). In women younger than 55, cigarette smoking is directly responsible for 21% of all CVD mortality and for 50% of all acute coronary events (62,101). Cigarette smoking is an independent predictor of sudden cardiac death in patients with CHD (102), leading to a twofold to threefold increased risk of dying from CHD (103). Cigarette smoking is also directly related to and can potentiate other CHD risk factors (5,62). In women younger than age 44, cigarette smoking is dose related to MI, with an RR of 2.5 for those smoking I to 5 cigarettes per day, compared with nonsmokers, and rising to 74.6 for those smoking more than 40 cigarettes per day (104). In women, smoking "low-yield nicotine" cigarettes (less than 0.4 mg nicotine per cigarette) did not affect significantly CHD risk compared with smoking the "regular" high-yield nicotine cigarettes (more than 1.3 mg nicotine
per cigarette) (RRs of 4.2 and 4.7, respectively) (101,105). Exposure to passive, environmental, cigarette smoke also predisposes to cardiovascular events (106-110). Observational studies in women estimated that exposure to passive smoking is associated with a 15% increase in the risk of dying from heart disease compared with nonsmokers not exposed to passive smoking (110). Even smokeless tobacco has been shown to increase CVD risk compared with nonsmokers (108). Cessation of smoking is associated with an estimated 50% to 70% reduction in CHD risk. One year after quitting, the risk of CHD decreases by 50%, and within 2 to 3 years it approaches that of a nonsmoker (5,62,111). Fifteen years after quitting, a woman's RR of dying from CHD equals that of a lifetime nonsmoker (111). The adverse coronary effects of cigarette smoking are exerted via several known mechanisms. Nicotinic alkaloids stimulate the sympathetic nervous system and increase plasma free fatty acids and LDL-C levels (113,114). Nicotine increases platelet aggregability (115) and fibrinogen plasma levels (116,117), thereby increasing the risk of coronary thrombosis. Cigarette smoking increases inflammation, thrombosis, and oxidation of LDL-C, thereby increasing tissue oxidative stress (118). Cigarette smokers have a higher incidence of insulin resistance compared with nonsmokers (119), and their plasma lipids demonstrate an unfavorable profile: increased levels of total cholesterol and triglycerides, increased LDL-C VLDL-C and VLDL-triglycerides, and decreased HDL-C (120,121). In women, cigarette smoking provokes a low-estrogenic, high-androgenic milieu. Premenopausal women who smoke cigarettes are usually deficient in estrogen, and cigarette smoking eliminates the estrogen-protective effect on the cardiovascular system (122,123). Constituents of cigarette smoke adversely modify production and metabolism of estrogens, and cigarette smoking attenuates estrogen effects on target tissues (124). Premenopausal cigarette smokers have a lower incidence of gluteal (gynecoid) adipose tissue distribution and a higher incidence of central (abdominal) adipose tissue distribution (122). Cigarette smoking positively correlates with a risk of early natural menopause, greater prevalence of hirsutism, oligomenorrhea, and infertility in premenopausal women (125). Premenopausal and postmenopausal smokers have a lower incidence of estrogendependent cancers such as endometrial cancer (126,127), and smoking eliminates the protective effect of oral estrogens on hip fracture in postmenopausal women (123). At the cellular level, nicotinic alkaloids directly inhibit human granulosa cell aromatase activity and the conversion of androgens to estrogens (128). Cigarette smoking induces hepatic microsomal mixed-function oxidase systems that metabolize sex hormones. This effect enhances 2-hydroxylation, resulting in synthesis of non-agonistic estrogen metabolites (123,129).
CHAPTER31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women
1. PREVALENCE OF CIGARETTE SMOKING
During the past 4 decades, the prevalence of smoking among women has declined. In the United States, the prevalence of smoking decreased by 0.3% per year in the past decade, but the rate of smoking initiation increased by 1% per year (5,62). In 2003, 24.6% of female students (grades 9 to 12) reported current tobacco use (130), and the most common age of initiation was 14 to 15 (131). This has resulted in a relatively stable prevalence of smoking among women of about 26% (5). About 60% of people in the United States have biologic evidence of exposure to passive cigarette smoke (132).
E Sedentary Lifestyle Data from studies in men and secondary analyses in women suggest that sedentary lifestyle increases CHD risk in women and that physical activity counteracts these effects. The RR of CHD associated with physical inactMty ranges from 1.5 to 2.4 and is comparable to that observed for high blood cholesterol, high blood pressure, or cigarette smoking (133). Sedentary people have about twice the risk of developing or dying from CHD compared with active people, and 37% of deaths from CHD can be attributed to physical inactivity, second only to raised blood cholesterol (134). Regular moderate intensity physical activity, which is attainable for the majority of men and women, gives considerable protection against CHD (135). Cross-sectional studies report that middle-aged women with higher levels of physical actMty had lower weight, lower systolic and diastolic blood pressure levels, favorable lipid profile, and lower fasting glucose (136). Physical activity can moderate risks of obesity in men (137) and women (138). In women, moderate actMty decreases total cholesterol, LDL-C, and triglycerides and increases HDL-C levels (139,140). In the Nurses Health Study, vigorous to moderate activity such as walking was associated with a 46% decrease in the risk of developing diabetes compared with least physically active women (141). Studies in men showed that physical actMty also improves the prevention and management of type 2 diabetes in obese people (142). Potential mechanisms by which physical exercise decreases the risk of CHD are direct cardiac effects such as development of collateral coronary circulation and improvement of the functional myocardial work capacity; systemic effects such as lowered heart rate and blood pressure; modulation of CHD risk factors such as weight reduction, improvement in lipid profile, decreased platelet adhesiveness and aggregability, enhanced fibrinolysis, and decreased adrenergic response to stress; and decreased incidence of depression, which may further contribute to a sedentary lifestyle (143,144).
415
1. PREVALENCE OF SEDENTARY LIFESTYLE
In 1999 to 2001 in the United States, 38% of Caucasian females, 55% of African-American females, and 57% of Hispanic or Latino females reported no leisure time physical activity (5).
G. Adverse Genetic Background Similar to other multifactorial polygenic diseases, most cases of CVD, CHD, and stroke involve interplay of various molecular and biochemical pathways with environmental factors. Among known genetic risk factors for CHD, family history is the most significant independent risk factor (145-147). Epidemiologic studies identified the heritability potential for most CHD risk factors, including total cholesterol (40% to 60%), HDL-C (45% to 75%), total triglycerides (40% to 80%), BMI (25% to 60%), systolic blood pressure (50% to 70%), diastolic blood pressure (50% to 65%), Lp(a) levels (90%), homocysteine levels (45%), type 2 diabetes (40% to 80%), and fibrinogen (20% to 50%) (147-150). Mendelian disorders associated with atherosclerosis, such as familial hypercholesterolemia (FH), explain only a small percentage of disease susceptibility. However, genetic studies of FH have contributed to the understanding of pathways for cholesterol homeostasis and to the development of lipidlowering drugs (151). Whole-genome scans of families with high prevalence of CHD revealed several loci linked to CHD (2q21-22, xq23-26 [152], and 16p13-pter [152,153]) or MI (14qter [154] and lp34-36 [155]), but thus far none of the underlying genes have been identified. Large-scale genome scans for diabetes, obesity, and hypertension have also failed to reveal specific loci accounting for the population variations, which suggests that many genes in concert determine the susceptibility to CHD (156,157). Contributing to this complexity are the confounding roles of age, ethnicity, and the environment (158). Gene polymorphism studies focused on relatively rare diseases and have presented evidence about point mutations that could be associated with CHD risk. For instance, CYPllB2, C-344T, and AT1R Al166C polymorphisms affect the autonomic modulation of heart rate that depends on sodium excretion (159). Similarly, the lipid response to diet could be modified by polymorphisms within the genes for apoE, apoB, apoCIII, lipoprotein lipase, hepatic lipase, endothelial lipase, the liver fatty acid-binding protein, the [33adrenergic receptor, adipsin, and the peroxisome proliferatoractivated receptor y (160). A large number of protein enzymes and receptors are involved in HDL metabolism, which regulates transport of cholesterol from peripheral cells to the liver. Mutations in the genes encoding these factors have been found to associ-
416
GORODESKI AND GORODESKI
ate with marked alterations in plasma HDL-C levels (161). Apolipoprotein A-I (ApoA-I) is the major protein constituent of the HDL particle (162), and ApoA-I and HDL have been shown to protect against atherosclerosis and CHD. The gene of ApoA-I is expressed in the liver and small intestine, and upon secretion into the plasma apoA-I is lapidated through ATP-binding cassette A1 (ABCA1)mediated efflux of free cholesterol and phospholipids from peripheral cells. This results in the formation of nascent disc-shaped HDL particles. Through the esterification of free cholesterol into cholesteryl ester by lecithin/cholesterol acyltransferase (LCAT), the nascent HDL particle can mature into larger and spherical HDL. The HDL-cholesteryl ester is transferred in part to apoB-containing lipoproteins in exchange for triglycerides by the action of cholesteryl ester transfer protein (CETP). Through this activity, tissuederived cholesterol can find its way to the liver via receptormediated uptake of LDL particles for secretion into bile. Recently, it was shown that hypo-o>lipoproteinemia due to mutations in apoA-I, ABCA1, and LCAT is associated with increased progression of atherosclerosis. In contrast, hyper-(x-lipoproteinemia as a result of loss of CETP function is associated with unaltered atherosclerosis progression (163). Gene polymorphisms of thrombosis functions can explain some of the estrogen-associated increased risk of thrombotic events observed in the The Heart and Estrogen/ Progestin Replacement Study (HERS) and Women's Health Initiative (WHI) trials (164). Suggested candidate genes are Factor V (AS06~G, Leiden), prothrombin (G20210-*A), Factor VII (A353-+G), plasminogen activator inhibitor-1 (4G/SG), t3-fibrinogen (G/A-455), and glycoprotein IIb/IIIa (pIA1/A2). Polymorphism of estrogen receptor (ER)-ot and ER-f3 can also explain differing responses to estrogen therapy. Naturally occurring splice variants of ER-oL and ER-[3 have been described, but only few point mutations were identified, and their biologic role is unclear (165,166). From the above it is apparent that CVD-CHD genomics is at its infancy, and future studies could provide important information about the genetics of these diseases. Understanding the genetics of CHD could improve the development of genotype-based tests and pharmacologic interventions based on genetic susceptibility.
H. Increased Blood Viscosity Increases in blood viscosity are associated with increased CHD risk. Plasma viscosity is higher in patients undergoing acute cardiovascular events such as stroke, MI, or sudden cardiac death (17,25,167,168). Increased blood viscosity can be caused by an increase in red cell mass or increased red cell deformity, increased plasma levels of fibrinogen and coagulation factors, and dehydration. CHD risk correlates with
increases in the activity of thrombogenic factors such as fibrinogen, procoagulants, and coagulation factors VII and VIII; with decreased activity of thrombolytic and fibrinolytic factors such as antithrombin III and plasminogen activator (25); and with conditions that promote platelets aggregation and adherence (17).
I. Depression Epidemiologic and observational studies have demonstrated an association between depression and cardiovascular disease (169-172). In the W H I Observational Study, depression was an independent predictor of CVD death (RR 1.5) after adjustment for multiple risk factors (172). Depression is also associated with the development of atherosclerosis (173), CHD (174), and a 2.5-fold increased risk of impaired cardiovascular outcome after MI (175). The mechanisms linking depression to CVD and CHD are unknown. Proposed mechanisms include sympathetic hyperactivity, increased platelet aggregation, lowered threshold for ventricular arrhythmias, and increased inflammation (176). Only a few studies have examined whether treatment of depression lowers CV-D risk, but no definitive conclusions could be drawn. A case-control study including 30- to 65year-old smokers (mostly premenopausal women) found an odds ratio (OR) of 0.35 for development of first MI among users of selective serotonin reuptake inhibitors (SSRIs) (177). However, a follow-up study showed that this association held only for SSRIs but not for other antidepressant medications, raising the possibility that SSRIs decrease MI risk through alternative mechanisms unassociated with depression (such as depletion of serotonin stores in platelets) (178). The Enhancing Recovery in Coronary Heart Disease Patients (ENRICHED) trial, a randomized controlled clinical trial of depression treatment through cognitive behavior therapy and SSRI use, showed no increase in post-Ml event-free survival despite improved depression outcomes (179). As this study included mostly men, it is uncertain whether the results could be extrapolated to women.
J. Stress and Autonomic Imbalance Persons under stress are at an increased risk for CVD. In men, the risk of CHD in type A personalities (competitive, achievement oriented, time urgent, and hostile) is twofold higher than in type B personalities (those who lack type A characteristics) (180). The onset of acute coronary syndromes is more prevalent in awakening hours, 6 AM to 9 AM, compared with 6 PM to 12 l'M. (181--183), and these effects correlate with peak diurnal sympathetic activity (183). Little is known about the effects of stress on CVD and CHD in women.
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women
The mechanisms by which stress modulates the risk of C H D are unclear. Stress can modify sympathetic neuroendocrine activity and the responses of the coronary vasculature to stimuli (180). Sympathetic hormones and glucocorticoids can promote the development of arteriosclerosis (17). In female animals, stress is associated with hypoestrogenism (185). In women, psychosocial stress exacerbates decreases in estrogen during the menstrual cycle and may lead to menstrual problems and infertility. Stressful life events are predictors of the premenstrual syndrome (PMS), possibly as a result of defective folliculogenesis and relative hypoestrogenism (185-187). Cardiac responses to [3-adrenergic sympathetic stimulation decline with age, as does the heart rate response to parasympathetic withdrawal (188). In postmenopausal women, the mean atrial effective refractory period is longer than in premenopausal women and agematched men. Experimentally, the atrial effective refractory period can be shortened during atrial pacing and during simultaneous atrioventricular pacing, but premenopausal women are relatively resistant to these effects compared with postmenopausal women and age-matched men (189). Arrhythmias are not considered a direct risk factor for CHD, but they could worsen CVD risk by weakening a failing heart. Most common among arrhythmias is atrial fibrillation (5). The incidence of atrial fibrillation is lower in women than in men (5). The lifetime risk for development of atrial fibrillation is 23% for women 40 years of age and older, and the mean age of developing atrial fibrillation in women is about 74 (5). In contrast to atrial fibrillation, women are at a greater risk for developing symptomatic ventricular arrhythmias and sudden death than men (190). The female gender is considered an independent risk factor for syncope and sudden death in the congenital long Q_T syndrome. Prolongation of the Q T interval is associated with increased risk of arrhythmia C H D and mortality. The higher propensity toward ventricular arrhythmia in females is associated with differences in repolarization and longer rate-corrected Q T intervals (191,192).
417
L. Gender-Related Differences in CHD Age-adjusted death rates increase exponentially after age 50, and significantly more so in men than in women (Fig. 31.12). In both sexes the leading causes of death are CVD (see Fig. 31.1), specifically C H D (Fig. 31.13), followed by cancer deaths (see Fig. 31.1). Prior to age 60, cancer deaths contribute relatively little to the gender-mortality gap (see Fig. 31.2), which suggests that the gender-related difference in mortality rates is mainly due to CHD. Although death rates from CVD and C H D are lower in women than in men (see Fig. 31.13), more women than men die of CVD because they have a longer life expectancy. In 2002, 488,946 women and 429,682 men died of major cardiovascular diseases in the United States (1). In childhood, CVD is rare and is mostly attributed to congenital heart disease (1). C H D mortality rates begin to increase by the second decade of life, at which age the maleto-female risk ratio increases and remains high throughout life (see Fig. 31.12). The gender gap narrows at more advanced ages, but even after age 85 the age-related gender mortality gap is substantial (see Fig. 31.12). These observations suggest that women are relatively protected from developing C H D at early adulthood and that the effect continues into advanced age.
Death Rates- All Causes per 100,000 Population 18000
15000 Men Women 12000
9000
K. Compounded Risk Factors 6000
In both women and men, risk factors for CVD and CHD are usually clustered: among persons with two risk factors, the most common combination is hypertension and hypercholesterolemia; among those with three risk factors, hypertension, hypercholesterolemia, and obesity; and among those with four risk factors, hypertension, hypercholesterolemia, obesity, and cigarette smoking or diabetes (5). However, risk profiles for acute ischemic syndromes differ among the genders: women tend to be older and less likely to have ever smoked; they have higher rates of associated diabetes, hypertension, and hypercholesterolemia; and they have lower incidence of prior MI, angioplasty, and coronary artery bypass surgery (193).
3000
03
03
",r i
~
~
!
U3
(,0 i
(0 i
I'~
I"-.
o3
Age FIGURE 31.12 Age- and sex-adjusted mortality rates in the United States. Modified from ref. 1, with permission.
GORODESKI AND GORODESKI
418
All
toms, illness, short and long-term disability, and poorer health than men (194). Women use preventive and curative medical services more frequently, including visits to caregivers, over-the-counter and prescription medications, hospitalizations (other than childbirth), psychiatric treatment, diagnostic tests, and surgical procedures (5,195). Recent reports indicate that although the age-related prevalence of CVD morbidity appears similar in women and men (Fig. 31.14, A), the age-related prevalence of hypertension (Fig. 31.14, B) and angina pectoris (Fig. 31.14, C) are higher in women than in men. In contrast, the prevalence of diagnosed cardiac conditions such as C H D (Fig. 31.15, A), acute MI (Fig. 31.15, B), and heart failure (Fig. 31.15, C) are lower in women than in men. The reason for these trends appears to be reporting bias, namely over-reporting by women and under-reporting by men, indicating greater health awareness by women. In addition to differences in CVD morbidity, genderrelated differences were also described relative to the clinical presentation of patients with ischemic heart disease, use of coronary diagnostic invasive studies, coronary findings in women, clinical and surgical management, outcome, and prognosis. These differences are discussed in greater detail in the following sections.
causes
Isch Heart Dis
Stroke
Heart Failure
HT Heart Dis
Women Men
Hypertensi on
Atherosclerosis
0
I
I
I
100
200
300
400
Death Rates (per 100,000 Population)
1. CLINICAL PRESENTATION
FIGURE. 31.13 Gender-dependent total cause and CVD-related mortality rates in the United States. Modified from res 1, with permission.
In the United States, the incidence of CHD in women lags behind men by 10 years; for MI and sudden death, by 20 years (196-197). More women than men have their initial manifestation of C H D as angina pectoris (65% versus 35%, respectively) (198), while fewer women than men suffer an acute MI as their first manifestation (29% versus 43%, respectively) (5). Among hospitalized patients with acute coronary syndromes, MI develops in a smaller percentage of
Gender-related differences are also observed relative to CVD morbidity. These phenomena are usually described in terms of disease prevalence based on self-reporting, visits to physicians, and hospital discharge coding. Past reports suggested that from early childhood women report more symp-
100
A. C V I )
-0-
t00
Men
-.J/. 0
'
-o- -8.
80
-0- ~ l
C. A n g i n a P e c t o r i s
13. H y p e r t e n s i on
- t - Wofmn
~
jo ~
....'
'
~
.......
Ages
FIGURE 31.14
0
"
z
~
20-34 35-44 45-54 55-64 65-74l 75+
/
6
......
*. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
" ........
21)-34 35.44 45.54 55-64 65-74 75§ Ages
0
-
~
.........
'
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
25-34 35-44 45-54 ~-64
65-7-4
7S +
A~s
Gender-related prevalence of CVD (A), hypertension (B), and angina pectoris (C). Modified from ref. 5.
419
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women
/
/
/
. 6
8
FIGURE 31.15
Gender-related prevalence o f C H D (A), acute MI (B), and congestive heart failure (C). Modified from ref. 5.
women than men (198-200). However, 36% of all women with C H D present with sudden cardiac death or fatal MI (201). Approximately 64% of women and 50% of men who died suddenly of C H D had no previous symptoms of the disease (5). In the United States, the average age of first heart attack is 65.8 for men and 70.4 for women (5), and the lifetime risk of developing C H D after age 40 is 49% for men and 32% for women. Analysis of annual rates of new MI or C H D death during 1989-2000 show similar racial trends but different gender trends (Fig. 31.16). At presentation, women usually have milder symptomatology than men, despite usually more severe disease (202). Chest pain is the most common symptom of MI in both men and women (203,204), but it is less predictive in women than in men and tends to subside more quickly (205). Women tend to complain of more back and jaw pain and are more likely to experience atypical symptoms such as shortness of breath, loss of appetite, indigestion, nausea and/or vomiting, palpitations, dizziness, fatigue, and syncope (205-208). Among MI patients younger than 65 years of age, fewer women than men present with ST elevation and fewer develop Qzwave MI, whereas in patients ---65 years old, there is no significant sex difference (209). In contrast to women, men are more likely to have a Qzwave infarction (210). Women younger than 65 years of age are more likely than men of the same age to be discharged with a diagnosis of unstable angina (211). These gender-related differences could be attributed to the greater lack of recognition of acute MI in women than in men (5,62,202).
2. USE OF DIAGNOSTIC INVASIVE STUDIES
Women are referred less often than men for coronary artery evaluation (Fig. 31.17), even when they have a positive exercise test (5,8). They are referred at a more advanced
Annual Rate (per 1000 persons) 60 -
40 -
0
l i b
-C]-O-II-0-
C Women CMen AA Women AA Men
~
65-74
/
,,,
/
I ..........................
75-84
Age
/./ I
/ ~ /f/01"'
I
> 85
FIGURE 31.16 Gender-related annual rates of new MI or CHD death in Caucasians (C) and African Americans. Modified from ref. 5.
stage of disease, are less likely to be diagnosed initially of having an acute ischemic heart condition, and take longer to be admitted for observation and treatment (5). The reasons for the decreased rate of referral for coronary testing are numerous and include the following: 9 Women more often present with an atypical pattern of chest pain. ~ Historically, women have been known to have nonobstructive C H D and single-vessel disease (34). ~ Women have less common causes of ischemia such as vasospastic and microvascular angina, as well as syndromes of nonischemic chest pain, such as mitral valve prolapse (212). ~ Women are more risk averse in their medical decisionmaking (213).
420
GORODESKI AND GORODESKI
TABLE 31.2 Gender Differences in the Prevalence of Nonobstructive ("Normal") Coronary Angiogram Diagnostic Cardiac Cath eterizations
Acute coronary syndrome Unstable angina No ST-elevation MI ST-elevation MI
Angioplasty
Women
Men
18% 28% 9% 10%
9% 11% 4% 7%
MI, myocardial infarction. Modified from ref. 214, with permission.
Open Heart Surgery
Cardiac Bypass Revascularization
Pacemakers and Implantable Defibrillators
Men Women 0
200
400
600
800
I000
Estimated No of Procedures, 2002
FIGURE 31.17 Gender differences in invasive diagnostic coronary procedures and cardiac surgical interventions. Modified from res 5.
9 Many physicians believe that angina pectoris in women is a benign disease and know that stress testing is less accurate for the diagnosis of C H D (5). 9 The fact that women with acute MI are on average 7 to 8 years older than men seems to contribute to the reluctance to perform invasive procedures (5).
3. CORONARY FINDINGS IN WOMEN
Three-vessel and left main C H D are more common in men than in women (211), and women are less likely than age-matched men to have obstructive C H D (5). Normal coronary angiograms ("nonobstructive atherosclerotic coronary disease") are present in 9% to 28% of women presenting with acute coronary syndromes compared with 4% to 11% in men (Table 31.2) (214). In some women with evidence of myocardial ischemia or infarction and normal coronary arteries, symptoms may be indistinguishable from those with obstructive coronary artery disease. In these women, pathologically important atherosclerotic coronary disease may be present even in the absence of angiographically observed stenosis (214). Advanced coronary perfusion imaging techniques will usually demonstrate vascular dysfunction in 50% to 60%, and 20% of those will have athero-
sclerosis-related endothelial dysfunction, as is evident by abnormal acetylcholine perfusion test (214). Most women with acute coronary syndromes and nonobstructive coronary angiograms will have positive exerciserelated ECG changes suggestive of ischemia. Although most have a good prognosis with regard to MI and cardiovascular mortality (215), some will have limitations in daily activities because of pain (216) and will be erroneously diagnosed as having other vascular disorders such as hot flushes, Raynaud's phenomena, and migraine (217,218). Of equal concern is the fact that unstable angina with "normal" coronary angiogram carries a 2% risk of death or MI at 30 days of follow-up (219,220).
4. CLINICAL AND SURGICAL MANAGEMENT Following the demonstration of normal or near-normal coronary arteries, women with angina are often told that they have no significant heart disease and are offered no specific treatment beyond reassurance. Even when women are identified as having risk factors for CVD, physicians less often use standard accepted therapy in women than in men. Compared with men with similar cardiovascular risk profiles, women are less likely to undergo additional coronary evaluation (38% versus 62%) or coronary revascularization (2% versus 5%). Less aggressive lipid-modifying strategies are used when treating women compared with men with similar risk profiles (8-10). The complacency and misunderstanding of the gravity of C H D outcome in women is best evident from a report that women took longer to arrive at the hospital after infarction than did men (median time 2.6 hr for women versus 2.0 hr for men) (221). Studies from a decade ago (late 1990s) reported gender differences in procedural outcomes after elective percutaneous transluminal coronary angioplasty. These differences could have been the result of the presence of more comorbidities and worse clinical characteristics in women such as older age, unstable angina, congestive heart failure, diabetes mellitus, and hypertension than in men. Women have a smaller vessel diameter, more coronary tortuosity, and different plaque composition compared with men, which could
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women lead to a higher dissection rate and a greater number of procedural complications. Increased awareness to these conditions has resulted in inclusion of more women in surgical coronary procedures. With the introduction of more refined surgical techniques and the more aggressive use of antithrombotic drugs, current procedures are almost equally safe and provide similar outcomes in women and men, despite worse baseline characteristics in women (222). Nevertheless, recent reports still describe underutilization of some procedures and facilities for women. In 2002 an estimated 6.8 million inpatient cardiovascular operations and procedures were performed in the United States; 4.0 million were performed on men but only 2.8 million on women. Studies also reported underutilization of reperfusion therapy (men 68.8%; women 49.7%), administration of beta blockers (men 76.0%; women 66.0%) (221), and cardiac rehabilitation after acute M1 (see Fig. 31.17)(5).
5. OUTCOME AND PROGNOSIS
Data about the immediate and long-term outcome of acute coronary syndromes show that women fare less well than men and that, once C H D is diagnosed, the case fatality rate for women exceeds that for men (5). Although these data should be interpreted with caution because of the more advanced age with which women present with new cardiac event, some studies reported that the age-adjusted hospital mortality in cases of acute MI in women equals that in men. Those studies also reported that women have greater risk for congestive heart failure, stroke, and reinfarction than men, even after age adjustment (200,223,224). During hospitalization for the acute MI, the mortality rate in women is more than doubled compared with men (18.6% versus 8.4%, respectively) (221). The prognosis of women with unstable angina and nonobstructive atherosclerotic coronary artery disease includes a 2% risk of death or MI at 30 days of follow-up (214). Approximately 38% to 39% of women who have an acute MI die within 1 year, compared with 25% to 31% of men. Twenty percent of women versus 15% of men will have repeat acute M1 during the first 4 to 6 years after the first episode. Six percent of women and 7% of men will experience sudden death; about 11% of women and 8% of men will have a stroke, and 46% of women and 22% of men will be disabled with heart failure (5). Women are less likely than men to return to work during the first 2 years after the acute MI. Attributed reasons to this phenomenon are a more advanced age of women at their index event and gender differences in behavioral response to illness (225). Women use cardiac rehabilitation services less frequently than men and when they enroll, women present differently, being older and having greater medical comorbidities (226). Women under age 55 have worse prognosis than men, with greater recurrence of MI and higher mortality. Women under age 55 have less favorable near-term
421
outcomes after myocardial revascularization procedures than men and are twice as likely to die after an acute MI than age-matched men (5). One of the consequences of ischemic heart disease is congestive heart failure. The incidence of congestive heart failure is 10 per 1000 population after age 65. The overall prevalence rates of congestive heart failure are similar in men and women, but the incidence is higher for men in age groups 54 to 74 (see Fig. 31.15, C). The etiology and risk factors for congestive heart failure also differ between men and women. Ischemic heart disease is the most common cause of heart failure in men of all ages. Heart failure in younger women is more often due to nonischemic heart disease, but older women are more likely than men to develop heart failure after MI and interventional procedures. Hypertension, diastolic dysfunction, diabetes, obesity, and inactivity are risk factors for congestive heart failure in women, whereas alcohol abuse, cigarette smoking, cardiomyopathy, ischemic heart disease, and systolic dysfunction are more important risk factors in men (227). Women tend to be older when diagnosed with heart failure and have better ventricular function, less ischemic heart disease, atrial fibrillation, and ventricular arrhythmias. Compared with men, women benefit less from established treatments and experience a lower overall quality of life (228). However, women have better long-term survival (229).
6. GENDER DIFFERENCES: BY WHAT MECHANISMS?
Studies by the PDAY group showed that coronary artery fatty streaks and raised atherotic lesions can be found as early as childhood and progress in extent through ages 15 to 34 in both men and women. However, women have about one-half the extent of raised lesions at all ages thereafter (Fig. 31.18) (230). These results are consistent with other studies (231,232), which indicate that the coronary arteries of women and men have an equal extent of fatty streaks but that women have less extensive raised lesions. A possible explanation for the delayed progression of raised coronary lesions in women is the H D L mechanism because, regardless of race and ethnicity, the single major C H D risk factor with a significantly higher prevalence in men versus women is low H D L - C (<40 mg/dl) (Fig. 31.19, A, B). Several well-documented functions of H D L could explain the ability of these lipoproteins to protect against atherosclerosis. The best recognized is the ability of H D L to promote efflux of cholesterol from ceils and minimize the accumulation of foam cells in the artery wall. However, H D L has additional antiatherogenic properties. H D L and its major proteins apo A-I, apo A-II, and paraoxonase are effective antioxidants. H D L possesses anti-inflammatory properties, and by virtue of its ability to inhibit the expression of adhesion molecules in endothelial ceils, it can reduce the recruitment of blood monocytes into the artery wall. In
422
GORODESKI AND GORODESKI
A: Fatty Streaks
B: Raised Lesions
Women
!---I Men
15-19
20-24
25-29
Age
30-34
15-19
20-24
25-29
Age
30.34
. . . .
FIGURE 31.18 The mean extent (2SE) of fatty streaks and raised lesions by 5-year age groups of men and women in the right coronary artery. The values were adjusted for race, non-HDL-C concentration, HDL-C concentration, and smoking. Modified from ref. 71.
experimental models, HDL therapy decreased plaque contents including macrophages and smooth muscle cells (233). These beneficial effects could explain the coronary plaque stabilization in patients treated with recombinant ApoA-I Milano/phospholipid complex (234). The mechanism by which the female gender affects H D L and ApoA-I is probably linked to estrogen. Plasma levels of ApoA-I decrease in postpubertal boys and in women after menopause but remain stable in postpubertal girls until menopause (235). Moreover, treatment with estrogen reverses menopause-related decrease in ApoA-I (236-241). Therefore, it is possible that postpubertal increases in plasma activity of estrogen in women stimulate liver cells to produce ApoA-I and increase HDL; these effects are maintained during the woman's reproductive years but are sufficient to reduce the impact of cholesterol on coronary plaque progression throughout a woman's life. One of the consequences of acute ischemia is an increased sensitivity of the myocardium to influx of plasma and gases upon coronary reperfusion. The effect is of importance due to the increased use of treatments and invasive techniques intended to unblock coronary arteries. A study looking at the consequences of primary percutaneous coronary intervention in patients with acute MI found that although the amount of myocardium at risk or initial perfusion defect did not differ between women and men, the final infarct size, measured in a follow-up scintigraphy, was smaller in women than in men whereas the myocardium salvage index was greater in women than in men (242). These data suggest that female gender is an independent predictor of greater myocardial salvage after percutaneous coronary intervention.
FIGURE 31.19 Gender-related prevalence of major CHD risk factors in Caucasians (A) and African Americans (B). Cig, cigarette smoking; DM, diabetes mellitus; HBP, high blood pressure; OW, overweight; TC, total cholesterol. Elliptical areas: p < 0.02-0.05. Modified from ref. 5.
The mechanism of myocardium protection in females is at present unclear. One possible explanation is the role of the anti-apoptotic factor Bcl-2, which controls cell apoptosis and provides greater tolerance to hypoxia. Myocardial cells of females express higher basal levels of Bcl-2 than do myocardial cells of males (243). Another explanation is intracellular Ca 2+ metabolism; in cardiac myocytes, estrogens block Ca 2+ influx and moderate changes in cytosolic calcium (244). Augmented Ca 2+ influx and abrupt increases in cytosolic calcium are an important mechanism of the ischemia-reperfusion syndrome. Female myocardial cells express receptors for estrogen, and even low blood
423
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women levels of estrogen may confer protection against C a 2+ influx induced by hypoxia-reoxygenation (245,246). In addition to controlling C a 2+ influx, estrogens exert direct myocardial protective effects (40,41,47). In female ER-[3 null mice, body weight, fluid retention, atrial natriuretic peptide (ANP) serum levels, ventricular pro-ANP expression, elevated phospholamban expression, and rates of mortality were increased compared with wild-type animals (247). These markers and effects were unrelated to C a 2+ handling, indicating that deletion of the ER-[3 aggravates clinical and biochemical markers of heart failure by a mechanism that is unrelated to cytosolic calcium. Additional potential mechanisms by which estrogens confer cardiac protection are discussed in the following section.
V. ESTROGEN AND CHD
Premenopausal
CVD
~
Postmenopausal
CHD
O
n, ~=
2
m o
a
0
<40
40-44 45-49 50-54 <40 40-44 45-49 50-54 Age (years)
FIGURE 31.20 Age-adjusted death rates in women: effect of menopause. Modified from ref. 248, with permission.
A. Epidemiologic and Observational Studies Epidemiologic data indicate that menopause is associated with increased prevalence of CVD and CHD. The Framingham Study showed that the age-adjusted risk of CVD was higher in postmenopausal women than in premenopausal women, regardless of the age of menopause. Premature menopause (i.e., earlier than age 40) was associated with increased CHD risk (Fig. 31.20) (31,248). In the observational Nurses Health Study, after adjusting for age, smoking status, and other cardiovascular risk factors, the RRs across categories of age at natural menopause (<40, 40-44, 45-49, 50-54, and -> 55 years) were 1.53, 1.42, 1.10, 1.00 (reference), and 0.95, respectively (249). Earlier epidemiologic studies have suggested that women with late menarche have an increased risk for CHD (250,251) and that the length of exposure to estrogens, rather than the age of menopause per se, is the detrimental cardioprotective factor. Collectively, the epidemiologic and the PDAY data (see Figs. 31.9, 31.11, and 31.18) suggest that female hormonal changes are the causative factor for the gender-related differences in CVD and that the candidate to account for those changes is estrogen. The theory that estrogen treatment could lower cardiovascular mortality and improve morbidity has stirred more controversy during the past 3 decades than any other women's health-related issue. A large volume of experimental laboratory data, human mechanistic studies, and observational longitudinal studies in women has concluded that estrogen is likely the main cardiovascular protective factor in females. By the 1990s, observational data from epidemiologic studies were remarkably consistent in showing that postmenopausal hormone users tended to be at lower cardiovascular disease risk than non-users. The number of epidemiologic studies of postmenopausal hormone use and its effects on CHD is sufficient to justify a quantitative assess-
ment of the evidence. A number of large-scale, randomized controlled trials were initiated to specifically address the questions of estrogen modulation of coronary atherosclerosis and CHD mortality. In the United States, these studies included the following: 9 Postmenopausal Estrogen/Progestin Interventions (PEPI) trial (252) 9 Estrogen in the Prevention ofAtherosclerosis (EPAT) randomized, double-blind, placebo-controlled trial (253) 9 Heart and Estrogen/Progestin Replacement Study (HERS) (254) 9 Estrogen Replacement in Atherosclerosis (ERA) trial (255) 9 Women's Health Initiative (WHI) study, which had two main components, the Estrogen plus Progestin arm (256) and the Estrogen-only arm (257). Some of the initial results and conclusions drawn from these studies differed from conclusions of the previous experimental and observational studies. This resulted in reversal of previous recommendations regarding the pharmacologic value of estrogen replacement for the prevention of CVD in women. Moreover, because of the reported increased rates of vascular thromboembolic phenomena and CHD events associated with de novo estrogen treatment in older postmenopausal women, patients and caregivers were cautioned against the use of these medications in all postmenopausal women. Pendulum-like swings in medical opinions are not uncommon, but the conclusions of the HERS, ERA, and W H I trials stand in sharp contrast to the voluminous epidemiologic and observational data mentioned previously. This section reviews the different aspects of estrogen actions on the cardiovascular system in females; the main results and conclusions of the PEPI, EPAT, HERS, ERA, and W H I
424
GORODESKI AND GORODESKI
trials (as they pertain to CVD and CHD); and the takehome messages after critical review of the data. B. E s t r o g e n Actions on the Cardiovascular System in Females 1. MOLECULARAND CELLULAR MECHANISMS
Most eukaryotic cells, including those of the cardiovascular tree, are targets for the actions of estrogens. At the cellular level some actions, such as antioxidant effects, are mediated independent of ERs and are due to the chemical structure of estrogens, which contains a hydroxyphenolic (antioxidant) ring. More commonly, however, estrogen effects are mediated by the nuclear ERs (258). There are two types of known ERs: ER-ci and ER-[3. In endothelial cells, ER-ci mediates estradiol-induced increase in endothelial nitric oxide synthase (eNOS); it facilitates angiogenesis in vivo and mediates the anti-atherosclerotic effects of estradiol. Both ER-ot and ER-[3 mediate the beneficial effects of estradiol on vascular smooth muscle proliferation after vessel injury (259). In cardiac myocytes, ER-o~ predominates and the role of ER-[3 is unclear: some reports suggest that ER-[3 mediates estradiol-dependent transcriptional activation of eNOS and inducible nitric oxide synthase (iNOS) (260), whereas others maintain that ER-[3 has no specific role in the heart (261). The nuclear ERs act by regulating transcriptional processes. Following estrogen binding to the receptors, the receptors dimerize and bind to specific estrogen-response elements that are located in the promoters of target genes. ERs can also regulate gene expression without directly binding to DNA, through protein-protein interactions with other DNA-binding transcription factors in the nucleus. In addition, membrane-associated ERs can mediate nongenomic actions of estrogens and lead both to altered functions of proteins in the cytoplasm and regulation of gene expression. The latter two mechanisms enable a broader range of genes to be regulated than those regulated by the classical mechanism of ER action alone (262,263). The recently discovered estrogen-related receptors (ERRs) provide an additional type of estrogen mechanism of action. The orphan nuclear receptor ERRs could be activated by the transcriptional coactivator PPAR-~/, a regulator of cellular energy metabolism. Through this pathway ERRcx is involved in mitochondrial energy-producing pathways in cardiac and skeletal muscle cells (264). One of the difficulties in translating the experimental laboratory data into the clinical arena is that acute effects of the hormone on target cells may not sustain into chronic effects. This situation is the result of a number of known mechanisms: (1) cells can develop habituation to estrogen effect; (2) activation of one signaling pathway in vivo can
activate parallel systems that may offset or delay the initial effect; and (3) aging can alter cells' responsiveness to estrogen (265). Therefore, the effects of estrogen on the cardiovascular system should be discussed in terms of acute versus chronic regulation. Failure to distinguish between these two groups of data has created confusion and misunderstanding of estrogen effects in vivo. The list of estrogen acute effects has grown exponentially within the past decade and are reviewed elsewhere (266). More pertinent to this chapter is the topic of the chronic effects of estrogen, including modulation of plasma lipids, insulin and glucose metabolism, vasculo-protective effects, autonomic tone and cardiac rhythm, and myocardium protective effects. 2. PLASMA LIPIDS EFFECTS
Estrogens' effects on plasma lipids are well documented and were confirmed by WHI data (266a). Estrogens elevate plasma HDL-C, Apo-AI, and Apo-AII and decrease plasma total cholesterol, LDL-C, and lipoprotein(a). These favorable effects are considered protective against atherosclerosis. Estrogens increase plasma triglycerides via increases in plasma VLDL-C, which have only mild effects on atherosclerosis. The intrahepatic lipid-related mechanisms of estrogen actions involve enhanced catabolism and clearance of LDL by increasing the number of LDL (Apo B/E) receptors in hepatocytes; decreases in hepatic HDL receptors and attenuation of HDL catabolism; reduced actMty of hepatic lipase, thereby raising levels of HDL-C; inhibition of regulatory oxysterols so that the increased uptake of LDL is not associated with reduced cholesterol synthesis; inhibition of catabolism of cholesterol to bile acids, thereby diminishing a major disposal path for hepatic free cholesterol and enhancement ofbiliary secretion of cholesterol and cholesterol esterification; and enhanced clearance of chylomicron remnants. The overall effect would tend to reduce cholesterol accumulation in peripheral tissues and to increase its biliary secretion (266). 3. GLUCOSE EFFECTS
Estrogens' effects on glucose metabolism in postmenopausal women are only partially understood. Post hoc analysis of the HERS data showed that among postmenopausal women with coronary heart disease who were followed for an average of 4.1 years, hormone replacement treatment (0.625 mg of conjugated estrogen plus 2.5 mg of medroxyprogesterone acetate [MPA] daily, orally) reduced the incidence of diabetes by 35% (267). W H I data confirmed a decline in patients' reported treated diabetes both in the Estrogen plus Progestin arm (down 21%) and in the Estrogen-only arm (down 12%)(266a).
CHAPTER31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women 4. VASCULO-PROTECTIVEEFFECTS
The anti-atherogenic vasculo-protective effects of estrogens have been described elsewhere (266). Estrogens modulate functions of endothelial and vascular smooth muscle cells. Estrogens promote angiogenesis by modulation of endothelial cell migration, inhibition of vascular smooth muscle cell migration, and suppression of collagen secretion and assembly. The mechanisms are: inhibition of influx of esterified cholesterol into the vessel wall; inhibition of lipid accumulation; attenuated oxidative conversion of LDL in coronary arteries, even before affecting plasma lipids, and enhanced efflux of cholesterol from the vessel wall; suppression of inflammatory cell adhesion to vascular endothelial cells; inhibition of NFKB mediated upregulation of vascularcell adhesion molecule-1 expression; inhibition of endothelial cell and smooth muscle cell hyperplasia; and inhibition of collagen and elastin deposition (268). The vasodilator effects of estrogen are also well documented (266). Estrogens promote dilation of large proximal epicardial arteries, including partially atherosclerosed vessels. Estrogens also dilate coronary arterioles, which determine the coronary vascular resistance. The later effect is of importance in women because they are more likely than men to have acute coronary syndromes due to diminished coronary blood flow in the absence of obstruction of proximal vessels. The latter is the result of coronary vasospasm of the coronary arterioles (5). The mechanisms by which estrogens promote vasodilation include upregulation of eNOS, iNOS, and prostacyclines and downregulation of endothelins and their receptors. Additionally, estrogens modulate vasoreactivity by regulating the sensitivity to adrenergic and cholinergic stimuli and by direct inhibition of smooth muscle cell contraction. The latter are mediated by inhibition of calcium influx, attenuation of platelet-dependent release of vasoconstrictors (serotonin and thromboxanes), and modulation of release of endothelial secretagogues and mediators of inflammatory reaction (266). 5. REGULATION OF AUTONOMIC TONE AND CARDIAC RHYTHM
Gonadal steroids have been shown to regulate human cardiac rhythm and to control arrhythmias (269). In postmenopausal women, estrogens modulate beneficially autonomic tone and control blood pressure regulation (270). Estrogen treatment in postmenopausal women lowers systolic blood pressure and mean blood pressure by decreasing the systemic vascular resistance (271). In most normotensive or hypertensive postmenopausal women, natural estrogens lower blood pressure; however, 5% to 7% of women may have an idiosyncratic reaction of elevated blood pressure, which is reversible upon stopping the treatment (272). Estrogens confer protective effect on the myocardium by sta-
425
bilizing the conductive atrial and AV systems, thereby lowering the incidence of atrial fibrillation (189). However, unopposed estrogens in menopausal women were shown to mildly prolong myocardial repolarization (273), and in patients with long Q T syndrome, estrogens can induce symptomatic ventricular arrhythmias and sudden death (190). 6. MYOCARDIUMPROTECTIVEEFFECTS
Cardiac ischemia may lead to arrhythmia, infarction, and death. In cases of reperfusion to previously ischemized areas, cardiac tissues may undergo additional injuries, leading to accelerated necrosis. Estrogens can favorably affect the outcome of cardiac ischemia in females and salvage myocardial tissues from the consequences of reperfusion injury. The cellular and molecular mechanisms by which estrogens protect cardiac myocytes from the consequences of ischemia and reperfusion are unclear. Estrogens can modulate endothelial permeability by increasing the resistance of the tight junctions. This would tend to control the flow of solutes from the blood through the endothelium and into the cardiac parenchyma. Estrogen's antioxidant properties could explain some of the effect. Direct cardioprotective effects of estrogens in females include modulation of lysosomal activity and metabolic activity of cardiac myocytes; upregulation of eNOS, which plays an important role in signal transduction; an increase in cardiac content of epinephrine; augmented release of f3-adrenergic transmitters from cardiac nerve endings; and modulation of cardiac muscarinic receptors (266). Estrogens inhibit the rise in cytosolic calcium and intracellular sodium during metabolic stress (274). Estrogens upregulate NFKB-mediated HSP72 activity in cardiac myocytes, an effect that provides cytoprotection during hypoxia independent of heat shock protein (HSP) induction (275). The hormone reduces cardiomyocytes apoptosis by ER- and phospho-inositide-3 kinase-Akt-dependent pathways (276), and estradiol-mediated induction of atrial natriuretic factor in cardiac hypertrophy could contribute to its modulatory effects in left ventricular hypertrophy (277). These data indicate that cardiac myocytes are targets for estrogen actions: estrogens enhance inotropy and decrease chronotropy, lower the cardiorespiratory and energy demands, and protect the cells from hypoxic and reperfusion insults. The data also suggest that hypoestrogenism may not be advantageous to cardiac function in females and that estrogen replacement can reverse, to a degree, the insults of low estrogen and restore functions to near desirable levels.
C. Randomized Clinical Trials The experimental randomized clinical trials in postmenopausal women have provided novel information about the effects of estrogen and progestins on atherosclerosis and
426 coronary arteriosclerosis. Many clinicians consider these well-designed and executed studies as definitive in their respective research arenas, but controversies exist with regard to conclusions drawn from the data. This section describes briefly the five major randomized clinical trials in the United States and provides critical evaluation of the conclusions relative to C H D in women. Issues related to other aspects of those studies, such as breast cancer, osteoporosis, and stroke, are not discussed in this chapter. The PEPI trial (252) was a randomized, double-blind, placebo-controlled trial to assess differences among unopposed estrogen, each of three estrogen/progestin regimens, and placebo on selected heart disease risk factors in healthy postmenopausal women. A total of 875 healthy postmenopausal women ages 45 to 64 years were randomly assigned to the following groups: (1) conjugated equine estrogen (CEE), 0.625 mg daily; (2) CEE 0.625 mg daily plus cyclic MPA, 10 mg daily for 12 days per month; (3) CEE 0.625 mg daily plus consecutive MPA, 2.5 mg daily; (4) CEE, 0.625 mg daily plus cyclic micronized progesterone, 200 mg daily for 12 days per month; and (5) placebo. All medications were given orally. End points related to CVD risk were systolic blood pressure, plasma HDL-C, insulin, and fibrinogen levels. The results showed that oral estrogen, alone or in combination with an oral progestin, improves plasma H D L - C and lowers plasma fibrinogen levels without detectable effects on post-challenge insulin or blood pressure. The EPAT randomized, double-blind, placebo-controlled trial studied effects of unopposed estrogen on the progression of subclinical atherosclerosis in healthy postmenopausal women without pre-existing cardiovascular disease (253). In this trial, 222 postmenopausal women (mean age 61 years) were randomized to receive oral micronized 17[3-estradiol 1 mg daily, or placebo, for 2 years. Women received lipid-lowering drug (LLD) if their LDLC level exceeded 160 mg/dL. Baseline plasma lipid levels in the no-LLD category were (in mg/dL) total cholesterol, 224; triglycerides, 144; HDL-C, 55; LDL-C, 142; and fasting glucose, 87. The respective levels in the LLD category were 262, 167, 53, 177, and 90. Changes in intima-media thickness of the right distal common carotid artery were assessed by ultrasonograms. The results showed that: (1) in postmenopausal women with borderline high-risk (noLLD) lipid profile and without pre-existing cardiovascular disease, treatment with oral unopposed estrogen is associated with less progression of subclinical atherosclerosis than that observed in women who received placebo; (2) estrogen had no significant effect in women with high-risk lipid profile; (3) lipid-lowering medication had no added benefit to unopposed estrogen in reducing the progression of subclinical atherosclerosis. The authors concluded that administration of oral unopposed estrogen to women entering menopause may slow the progression of atherosclerosis and its clinical sequelae.
GORODESKI AND GORODESKI
H E R S was the first randomized trial of hormone treatment for secondary prevention of C H D in postmenopausal women (254,278). In this trial, 2763 women (mean age 66 years and 10 years after menopause) were randomized to receive either 0.625 mg of CEE plus 2.5 mg of MPA in 1 tablet daily (E/P) or a placebo (P). The primary outcome was the occurrence of nonfatal MI or C H D death. Women were followed for an average of 4.1 years. At baseline, 18% of women in both groups had diabetes; 56% had a BMI > 27 kg/m2; 24% reported poor of fair health; 9% had congestive heart failure; 17% had a history of Qzw-MI; 45% had percutaneous coronary revascularization, and 41% had coronary artery bypass graft surgery. Mean LDL-C was 145 mg/dL, 46% were on LLDs, 55% were on calcium channels blockers, 78% were on aspirin, 33% were on [3-blockers, 17% were on ACE inhibitors, and 28% were on diuretics. The results showed that: (1) there were no significant differences between groups in the primary outcome; (2) within the overall null effect, there were more C H D events in the hormone group than in the placebo group in year 1 and fewer in years 4 and 5; (3) more women in the hormone group than in the placebo group experienced venous thromboembolic events (relative hazard 2.89); (4) LDL-C decreased by 11% and H D L - C increased by 10% in the E/P group compared with the P group. The authors concluded that in postmenopausal women with established coronary disease, treatment with oral CEE plus MPA for 4.1 years did not reduce the overall rate of C H D events, but the treatment increased the rate of thromboembolic events. The authors recommended not to start hormone replacement therapy for the purpose of secondary prevention of CHD. However, given the favorable pattern of C H D events after several years of therapy and the favorable effect on plasma lipids, women already receiving E/P replacement may continue with the treatment. Post hoc analyses of the HERS data showed that hormone therapy reduced the incidence of diabetes by 35% (267). In addition, among women assigned to hormone therapy, C H D events were independently predicted by baseline LDL-C and H D L - C levels but not by triglyceride levels, suggesting that changes in lipid levels with hormone therapy had no effect on C H D outcomes (279). The ERA randomized, double-blind, placebo-controlled trial (255) studied effects of oral estrogen plus progestin, or estrogen alone, on the progression of coronary atherosclerosis in postmenopausal women with established CHD. The ERA trial was designed as a follow-up to the HERS findings. In the latter, estrogens were found to have no beneficial role as secondary prevention management in older postmenopausal women with high-risk C H D profiles. One of the criticisms of the HERS trial was lack of an estrogenonly arm. The ERA trial included estrogen-progestin and estrogen-only arms in addition to the placebo group. In this trial, 309 women with angiographically verified coronary
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women disease (mean age 66 years) with >30% coronary stenosis and mean cholesterol 216 mg/dL (36% of whom were on lipid-lowering drugs) were randomized to receive orally 0.625 mg of conjugated estrogen per day (E), 0.625 mg of conjugated estrogen plus 2.5 mg of MPA daily (E/P), or placebo. Women were followed for 3.2 years, and baseline and follow-up coronary angiograms were analyzed quantitatively. The results showed that (1) E and E/P decreased LDL-C and increased HDL-C; (2) neither treatment altered the progression of coronary atherosclerosis; and (3) the rates of clinical cardiovascular events were similar among the treatment groups. The authors concluded that (1) neither estrogen alone, nor E/P affected the progression of coronary atherosclerosis in women with established disease and (2) older postmenopausal women with established C H D should not use estrogen replacement with an expectation of cardiovascular benefit. The IYHIwas a randomized, controlled, primary prevention trial aimed at assessing the major health benefits and risks of the most commonly used combined hormone preparation in the United States (256). In the W H I trial, 16,608 postmenopausal women (mean age 63, about 10 years after menopause) with an intact uterus at baseline were recruited for a planned 8.5-year study duration. Participants received oral CEEs (0.625 mg daily) plus MPA (2.5 mg daily) in 1 tablet (E/P), or placebo. The primary outcome was any myocardial infarction, both fatal and non-fatal. At baseline, 35% of women in both arms of the study were overweight and an additional 34% were obese; more than 35% were treated for hypertension. More than 12% had elevated cholesterol; 7% were on statins, and 19% were on aspirin. On May 31, 2002, after a mean of 5.2 years of follow-up, the data and safety monitoring board recommended stopping the trial because the test statistic for invasive breast cancer exceeded the stopping boundary for this adverse effect and the global index statistic supported risks exceeding benefits. The statistically significant major clinical outcomes in the E/P versus placebo arms (pertinent to CVD) were as follows (estimated relative hazards): CHD, 1.29; pulmonary embolism, 2.13; and death due to other causes, 0.92. Relative hazards for composite outcomes were 1.22 for total CVD and 0.98 for total mortality. Absolute excess risks per 10,000 person-years attributable to estrogen plus progestin were 7 more C H D events and 8 more pulmonary embolisms. The authors concluded that: (1) overall health risks exceeded benefits from use of combined estrogen plus progestin for an average 5.2-year follow-up among healthy postmenopausal U.S. women; (2) all-cause mortality was not affected during the trial; (3) the risk-benefit profile was not consistent with the requirements for a viable intervention for primary prevention of chronic diseases; and (4) this regimen should not be initiated or continued for primary prevention of CHD. The estrogen-alone component of the W H I trial (Estrogenalone I/VHI) (257) was a randomized, controlled, primary
427
prevention trial aimed at assessing the effects of the most commonly used postmenopausal hormone therapy in the United States on major disease incidence rates. In this trial, 10,739 postmenopausal women (mean age 63 years) with a history of prior hysterectomy were randomly assigned to receive either 0.625 mg daily of CEE or placebo orally. The primary outcome was any myocardial infarction, both fatal and non-fatal. At baseline, 34% of women in both study arms were overweight and an additional 44% were obese; more than 47% were treated for hypertension. More than 14% had elevated cholesterol; 7% were on statins, and 19% were on aspirin. In February 2004, the National Institutes of Health (NIH) decided to end the intervention phase of the trial after 6.8 years as opposed to the planned 8.5 years, claiming that the burden of incident disease events was equivalent in the CEE and placebo groups, indicating no overall benefit. Estimated relative hazards for CEE versus placebo (pertaining to CVD) were CHD, 0.91; pulmonary embolism, 1.34. Corresponding results for composite outcomes were total CVD, 1.12; total mortality, 1.04. The authors concluded that (1) the use of CEE has no effect on C H D incidence in postmenopausal women with prior hysterectomy over an average of 6.8 years and (2) CEE should not be recommended for chronic disease prevention in postmenopausal women. A 2006 update report, however, clearly demonstrated a lowered trend of MI or coronary death in CEE recipients compared with the placebo group among women aged 50 to 59 years at baseline (hazard ratio of 0.63, nominal 95% confidence interval, 0.36-1.08). In that age group, coronary revascularization was less common among women assigned to receive CEE (hazard ratio, 0.55; nominal 95% confidence interval, 0.35-0.86), as were several composite outcomes, which included the primary outcome and coronary revascularization (hazard ratio, 0.66; nominal 95% confidence interval, 0.44-0.97). In addition, there was a clear trend of increasing differences between the CEE and placebo groups relative to those parameters over the 7 years of follow-up (279a).
D. Critical E v a l u a t i o n o f Trial C o n c l u s i o n s The PEPI, EPAT, ERA, HERS and W H I provided important experimental clinical data and are considered milestones in menopausal medicine.
1. STRENGTHS OF THE RANDOMIZED CLINICAL TRIALS
The PEPI study (252) was the first randomized trial of hormones use in postmenopausal women. It showed for the first time in an experimental randomized trial the long-term beneficial effects of estrogen on plasma lipids, and it highlighted the clinical outcomes with the use of different addon progestins.
428 The EPAT (253) is considered a definitive study of estrogen effect on the progression of subclinical atherosclerosis in healthy postmenopausal women with borderline high-risk lipid profile but without pre-existing cardiovascular disease. The results support previous experimental data in humans, animal models, and ex vivo organ, tissue, cell culture, and broken cell systems of the anti-atherosclerotic potential of estrogen. The results of the EPAT study also showed that the effect of estrogen was lost in women with high-risk lipid profile. The major conclusion of the study was that benefits of estrogens, administered orally, require healthy target tissues and particularly healthy endothelium. The HERS (254,267,278,279) and ERA trials (255) have demonstrated that in women with high-risk C H D profile (HERS) or established C H D (ERA), oral unopposed estrogen or combined estrogen-progestin are ineffective in reversing or slowing the progression of coronary atherosclerosis or in reducing the incidence of C H D morbidity and mortality. The explanation provided by the authors, which is supported by circumstantial human and animal data, is that estrogens and progestins can sustain and augment pro-inflammatory reaction at the region of the coronary plaque. This action offsets the atherogenic-protective effect of estrogen and can reverse the estrogen effect from arterio-protection to deleterious atherogenic process. This hypothesis is supported by two groups of experimental data. First, it was shown that among markers of inflammation, CRP is the only one that increases after oral (but not transdermal) estrogen treatment. However, although baseline CRP is a good predictor of C H D risk, levels after estrogen treatment have no predictive role (280). Second, in postmenopausal women hormone treatment increases MMP-9 plasma levels and gelatinolytic activity (281). This extracellular proteinase, which can be downregulated by statins (282), plays an important role in modulation of coronary plaques and can render them unstable. Collectively, these observations could explain the phenomena noted in postmenopausal women in which statins reduced the incidence of plaque rupture (i.e., acute coronary syndromes), whereas estrogen with or without progestins augmented the incidence of plaque rupture. Additional observations in the HERS and ERA studies were that older postmenopausal women who have not been treated with estrogen/progestin or have been offhormones for a number of years are at risk for developing vascular thromboembolic events upon de novo treatment with estrogen. The exact mechanism of this phenomenon is unclear, but it could be related to the combination of the thrombogenic properties of estrogen and the increased susceptibility of the older postmenopausal woman to increased blood viscosity. The W H I studies (256,257) were the largest randomized clinical trials in women and are the only randomized trials that provide clinical end points for primary prevention of CVD. The W H I studies confirmed previous results by the HERS and ERA studies of increased vascular thromboem-
GORODESKI AND GORODESKI
boric effects upon de novo treatment with oral estrogen, as well as the association of estrogen-progestin treatment with mildly increased risk of breast cancer. Similar to the HERS and ERA studies, the W H I studies did not find evidence for estrogen cardioprotection. The W H I estrogen-progestin study reported increased C H D events without affecting overall mortality, suggesting that the attributed mortality to de novo C H D events is offset by decreased mortality due to other causes such as osteoporotic fractures and colon cancer.
2. WEAKNESSES OF THE RANDOMIZED CLINICAL TRIALS
Despite the major contributions to the understanding of the therapeutic role of hormone replacement to postmenopausal women, some of the conclusions of the HERS and ERA studies, and in particular of the W H I studies, may have been inappropriately generalized to types of populations who were not studied in those trials. The follow-up of the original HERS study, the HERS-II, was the continued treatment of the HERS cohort for a period of up to 6.8 years. The rationale was to determine whether the improving C H D trends in the estrogen-progestin group versus the placebo group materialized at subsequent years. The authors found that lower rates of C H D events among women in the hormone group in the final years of HERS did not persist during additional years of follow-up. The authors concluded that postmenopausal hormone therapy should not be used to reduce risk for C H D events in women with C H D (283). However, the high drop-out and low adherence in that study may have rendered this interpretation invalid due to lack of statistical power to ascertain such conclusions. Post hoc analysis of the HERS results suggested that the use of LLDs was associated with lower rates of cardiovascular and venous thromboembolic events (284). Because all the cardiovascular mortality in the estrogen-progestin arm occurred in women not on LLDs, it is possible that the increased cardiovascular mortality in the estrogen-progestin group was biased by the use of LLDs. A similar concern was raised regarding the W H I data (285). Of the randomized trials, the W H I authors' conclusions were subject to the most criticisms. The W H I studies concluded that estrogen, alone or in combination with progestin, should not be initiated or continued for primary prevention of CHD. The latter conclusions are debated by critics who argue that the W H I cannot be considered a primary prevention trial. Some of the criticisms are summarized as follows: 9 The W H I participants cannot be considered genuinely "healthy postmenopausal women" from the cardiovascular perspective, given their advanced age relative to the age of natural menopause (63 versus 51 years); the fact that most were at least 10 years after menopause;
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women and the prevalence of overweight, obesity, hypertension, and hyperlipidemia. Furthermore, most cases of vascular thromboembolic events occurred in years 1 and 2, mostly in older (ages 70-79) and obese women. It is possible that some of the W H I patients already had established coronary atherosclerosis at study entry. Therefore, the hormone intervention may have not been purely primary prevention, but rather secondary prevention similar to the HERS and ERA trials. 9 Sensitivity analysis of the published W H I data showed that 44.4% of the women on active treatment versus 6.8% of the placebo recipients had their treatments unblinded, mainly because of vaginal bleeding. Among those, detection bias could not be excluded. For the three cardiovascular outcomes, bias was likely as the women were twice cautioned about possible increased risks observed in the interim data among estrogen plus progestin recipients. Therefore, it may not be possible to discriminate between causation and detection bias as alternative explanations for the findings (286). 9 There was an excessive drop-out rate in the W H I studies, especially in study years 4 and 5. In such a setting, the pre-specified study objectives could be missed because the allowances made for drop-outs during the planning stage of the trial were too small to compensate for excessive loss of patients that may emerge during the trial (287). 9 Only 10% of the cardiovascular events were adjudicated, and after adjudication, C H D and nonfatal MI rates did not differ among the estrogen-progestin and placebo arms, indicating that the differences between the arms were so small that a shift of only few cases could result in loss of significance. Furthermore, after adjudication, the increases in C H D events in the estrogen-progesfin and placebo arms were significant only after year 1 (285). 9 In both W H I trials there was a trend towards decreased C H D risks with increasing duration of treatment and with earlier age of start of treatment (ages 50-59) (285). In the estrogen-progestin W H I trial, among women more than 10 years postmenopause, the RR of C H D with hormone therapy versus placebo was 0.89; for 10-19 years, 1.22; and for >20 years, 1.71. This trend is statistically significant (p = 0.036) (256). Similar trends were reported in the estrogen-only W H I trial (257,279a). Interestingly, in the W H I observational arm (288), estrogen-progestin treatment was associated with reduced C H D risks (RR 0.71), despite increased risks of vascular thromboembolic events (RR 1.06-1.17) and similar to the results obtained in the observational Nurses Health Study (289,289a). The most likely explanation is the younger age of participants and earlier start of hormone treatment in the women enrolled in the W H I observational arm compared with the W H I randomized trial.
429
This conclusion is supported by a number of previous studies. For instance, in rabbits, estrogens had either a beneficial or adverse effect depending on the state of the arteries. Animals with healthy arteries showed a protective effect of estrogen, but in those with damaged arteries the effect was adverse (290). Monkeys randomized to postmenopausal hormones at the time of oophorectomy had substantially reduced coronary atherosclerosis compared with those given placebo. In contrast, when hormones were begun 2 years after surgical menopause (equivalent to about 6 human years), no such protection was observed (291-295). In women, the effect of estrogen on acetylcholine-induced vasodilatation differs markedly according to the presence or absence of coronary atherosclerosis (296). Randomized trials of women with C H D in the HERS and ERA trials did not show beneficial effects of estrogen, but in younger postmenopausal women there was a reduction in progression of atherosclerosis in the treated group compared with those given placebo (296). 9 The decision to terminate the estrogen-progestin W H I study was based on the nominal, unadjusted RRs of breast cancer and cardiovascular events. When the number of statistical tests is taken into account (i.e., adjusted for multiple testing), all the "significant" test results other than those for deep vascular thromboembolic events were no longer significant, indicating that the minor, non-significant increase in RRs in many conditions could have been explained by chance alone (287).
E. Differences among the Conclusions of the Observational and Randomized Trials The disparity among findings from the observational studies and the randomized trials on the effects of hormone replacement on CVD and C H D have created considerable debate among caregivers, researchers, and patients. Pooled estimates from observational studies inferred a relative CVD reduction of about 50% with ever-use of hormone replacement, supporting a cardioprotective role for estrogens that is unlikely to be explained by confounding factors (289). Similar results were recorded in the W H I observational arm, as described previously (288). By contrast, the randomized W H I studies in supposedly healthy women found hormone replacement to either have null effect (257) or slightly increased risk of C H D (256). One possible explanation for this apparent disparity is the potential for confounding in the observational studies. For instance, in the Nurses Health Study, estrogen users tended to be healthier than those not on hormone treatment. The difficulty with this explanation is that no disparity was found between the Nurses Health Study and the
430
GORODESKI AND GORODESKI
W H I trials with regard to risk estimates for breast cancer, stroke, and others (297). Another possible explanation is that observational studies are limited in capturing acute effects. The increase in C H D events in W H I was concentrated in the initial year after starting hormones. This short-term effect would tend to be missed in observational studies, even with fairly frequent updating of exposure. However, re-analyses of observational data should be able to test this speculation. Indeed, reanalysis of the Nurses Health Study showed that there was a higher rate of recurrent coronary events in women with a prior history of cardiovascular disease, fitting with the W H I data (298). Moreover, it was shown that the risk of C H D depended on start of hormone replacement from the time of menopause: the RR of C H D was 0.72 if treatment with estrogen plus progestin started 1 to 4 years after menopause versus 0.9 if started more than 10 years after menopause. For estrogen-only treatment the RRs were 0.66 and 0.67, respectively (299). In addition, a meta-analysis of observational studies focusing on older postmenopausal women was consistent with recent randomized trials, and it demonstrated an almost identical increase in CVD RR, being 1.28 in the secondary prevention of CVD events (300). The likely explanation for the disparity between the observational studies and the randomized trials is therefore that women who participated in the trials were different from those who participated in the observational studies. Past observational studies reflected common clinical practice whereby women were typically initiated on postmenopausal hormones at a relatively young age and earlier after the beginning of menopause. In contrast, two-thirds of the participants in the W H I began the trial at an age older than 60, and about 10 years after the start of menopause. These women most likely had already an established atherosclerosis upon entry into the trial and were more prone to the prothrombotic effects of the hormone treatment (301). In the case of the W H I trials, a large number of the participants were not healthy postmenopausal women, and the W H I trials cannot answer the question of the primary C H D preventive role of estrogen in postmenopausal women (301).
F. Current Stand on Hormone Treatment for Postmenopausal Women Based on the previous discussion, the following conclusions are advanced: 9 Estrogen is a vasculo- and cardio-protective hormone in females, and the menopause transition renders a woman at an added increased risk for CHD. 9 Estrogen should not be used as first-choice treatment in women with established C H D or at increased C H D risk.
9 Oral, but not transdermal, estrogen replacement therapy is associated with risk of venous thromboembolism in postmenopausal women (302). 9 The W H I trials did not effectively address the issue of primary prevention of C H D with estrogen, and the current understanding in this matter still depends on the results of the observational studies such as the Nurses Health Study. One of the conclusions from these studies is that estrogen replacement therapy beginning early after menopause reduces overall risks of CHD. 9 When taking into consideration how early after menopause to start hormone replacement therapy in otherwise healthy woman without clinically apparent CHD, it is important to understand that coronary disease begins premenopausally and that menopause may be preceded by years of gradual declining estrogen activity. Studies in monkeys revealed the importance of premenopausal estrogen treatment in reducing the risk of atherosclerosis (303). 9 Current experimental, clinical, and observational data do not provide evidence that progestins attenuate estrogen's cardiovascular benefits (304). 9 Considerations whether and when to start hormone replacement therapy should be individualized to the patient's needs, emphasizing her global versus cardiovascular risk profile.
VI. CHD RISK ASSESSMENT Understanding the epidemiologic profile of patients at risk can predict development of C H D and its long-term outcome in 50% to 75% of cases (305-308). However, most models, including the Framingham Risk Score, fall short of predicting near-furore events in individual patients, and these models do not effectively predict occurrence of acute coronary syndromes and sudden death. Acute coronary events are the culmination of processes aggravated by thrombogenic background that promotes thrombus formation at the site of the coronary plaque and by myocardial electrical instability and susceptible autonomic tone that could lead to fatal arrhythmia. Accordingly, the modern approach to C H D risk assessment advocates early identification of the Vulnerable Patient along with the Vulnerable Blood and the Vulnerable Myocardium (309,310). Risk assessment of the Vulnerable Woman is based on the 10-year Framingham Risk Score Calculator (311). It classifies the risk of heart disease as high, intermediate, or low. Tabulation is based on scoring points for the parameters of age, smoking, plasma total- and HDL-cholesterol, and systolic blood pressure (Tables 31.3 to 31.7). The added points are transformed to a 10-year risk as a percentage, so that a risk ->20% in 10 years is considered high risk, intermediate
431
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women TABLE 31.6 The Framingham C H D Risk Score Calculator: H D L Cholesterol as a Risk Factor
TABLE 31.3 The Framingham C H D Risk Score Calculator: Age as a Risk Factor Age
Points
HDL (mg/dL)
Points
20-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79
-7 -3 0 3 6 8 10 12 14 16
>-60 50-59 40-49 <40
-1 0 1 2
CHD, coronary heart disease; HDL, high-density lipoprotein. Modified from ref. 311, with permission.
TABLE 31.7 The Framingham C H D Risk Score Calculator: Systolic Blood Pressure as a Risk Factor
CHD, coronary heart disease. Modified from res 311, with permission.
Systolic blood pressure (mm Hg) TABLE 31.4 The Framingham C H D Risk Score Calculator: Smoking as a Risk Factor Points per age range Smoking Nonsmoker Smoker
Age 20-39
Age 40-49
Age 50-59
Age 60-69
Age 70-79
0 9
0 7
0 4
0 2
0 1
<120 120-129 130-139 140-149 >-160
Untreated
Treated
0 1 2 3 4
0 3 4 5 6
CHD, coronary heart disease. Modified from res 311, with permission.
TABLE 31.8 The Framingham C H D Risk Score Calculator
CHD, coronary heart disease. Modified from ref. 311, with permission.
Point total TABLE 31.5 The Framingham C H D Risk Score Calculator: Total Cholesterol as a Risk Factor Points per age range Total cholesterol (mg/dL)
Age 20-39
Age 40-49
Age 50-59
Age 60-69
Age 70-79
<160 160-199 200-239 240-279 >-280
0 4 8 11 13
0 3 6 8 10
0 2 4 5 7
0 1 2 3 4
0 1 1 2 2
CHD, coronary heart disease. Modified from ref. 311, with permission.
<0 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 >--25
10-year risk % <1% 1% 1% 1% 1% 2% 2% 3% 4% 5% 6% 8% 11% 14% 17% 22% 27% >-30%
CHD, coronary heart disease. Modified from ref. 311, with permission. risk is 10% to 20%, and low risk is <10% (Table 31.8). Women with established C H D (history of acute MI, silent MI, myocardial ischemia, history of unstable angina and stable angina pectoris, and history of coronary procedures), women with other clinical atherosclerotic diseases (periph-
eral arterial disease, abdominal aortic aneurysm, and carotid artery disease), and women with diabetes meUitus are considered to be at high risk (62). Parameters of Blood Vulnerability include markers of inflammation (e.g., high-sensitivity CRP), hypercoagulability
432 (e.g., fibrinogen), D-dimer, and platelet activation and aggregation function tests. Parameters of Myocardial Vulnerability in patients with atherosclerosis-related myocardial ischemia include ECG at rest or during stress test, perfusion and viability disorder (e.g., positron emission [PET] scan), and wall motion evaluation (e.g., echocardiography). In patients without atherosclerosisrelated myocardial ischemia, the following findings will identify a vulnerable myocardium: sympathetic hyperactivity, left ventricular hypertrophy, cardiomyopathy and myocarditis, valvular disease, and electrophysiologic disorders including long-Q_T and Wolff-Parkinson-White syndromes and sinus and atrioventricular conduction disturbances. It should be realized that this strategy is still in its developing stages for women (311) and that commonly used terminology for chest pain, such as typical or atypical angina or non-anginal chest pain, are definitions derived from male populations. Therefore, the clinical assessment and diagnosis of C H D in women presents challenges not seen when testing men. Women at intermediate or high risk, or symptomatic women, are candidates for cardiovascular risk screening including imaging of subclinical atherosclerotic disease. However, the rationale for imaging has been established in men, and proposed advantages for women using standard tests carry lesser predictive value for guiding management decisions and therapeutic risk-reduction strategies. Treadmill testing with exercise ECG is frequently used in women, but results are of lower accuracy than in men because of more frequent resting ST-T-wave changes and lower ECG voltage. Sensitivity and specificity of exercise ECG studies are 61% and 70% as compared with 72% and 77% in men (312). In high-risk symptomatic women, exercise ECG testing can predict obstructive (->75%) disease in 89% of cases, but this number decreases to 19% and 35% in low- and moderate-risk women (313). The diagnostic and prognostic accuracy of the exercise ECG stress test in symptomatic high-risk women can be enhanced by the inclusion of additional parameters such as functional capacity and treadmill scores to the interpretation of the ST-segment response to exercise. Exercise ecbocardiograpby (stress test) is used for diagnosis and risk assessment of symptomatic women with intermediate to high likelihood of CHD. It can provide information about the presence of left ventricular systolic or diastolic dysfunction, valvular heart disease, the extent of infarction, and stress-induced ischemia. Because C H D in women occurs at a more advanced age than in men, women referred for stress testing are usually older and less likely to reach an adequate heart rate with exercise testing. Exercise echocardiography has demonstrated good diagnostic accuracy for detecting or excluding significant C H D in women, with sensitivity and specificity of 81% to 89% and 86%, respectively. It offers prognostic data beyond that provided by clinical evaluation or exercise treadmill variables alone (34).
GORODESKI AND GORODESKI
Gated myocardialperfusion single-photon emission computed tomography (SPECT) is commonly used for the diagnosis and risk assessment of symptomatic women with suspected CHD. The technique assesses perfusion defects, global and regional left ventricular function, and left ventricular volume. Although widely used, myocardial perfusion imaging may result in falsepositive results in women due to breast-related artifacts and small left ventricular chamber size. Overall, the accuracy of the method is similar in women and men: sensitivity of 87% and 88% and specificity of 91% and 96%, respectively (34). Studies in women have shown good predictive value for MI, cardiac death, or the need for coronary revascularization (reviewed in [34]). Furthermore, stress myocardial perfusion imaging provides incremental value over clinical variables and the exercise ECG in women with suspected or known ischemic heart disease, and it is recommended for the symptomatic woman with an intermediate to high likelihood of CHD. Three relatively new and rapidly developing imaging methods have included women in preliminary studies: computed tomography (CT), magnetic resonance imaging (MRI), and carotid intima-media thickness (IMT). However, more data are needed to define their role in the routine clinical evaluation of women with suspected CHD.
VII. CHD RISK R E D U C T I O N Guidelines for reducing the health impact of C H D in women and men focus on the following three lines of action: modifications of lifestyle; early diagnosis and treatment of associated medical conditions; and lipid-lowering treatment.
A. Modifications of Lifestyle Behavior modification remains the primary and most important step in CHD risk reduction. This intervention should be routinely provided and applied to all women regardless of level of risk. Measures should include the following: 9 Cessation of cigarette smoking and avoidance of second-hand smoke 9 Maintenance of BMI at 19-25 kg/m 2 by balancing intake and expenditure of calories Heart-healthy diet (fruits, vegetables, grains, low-fat or nonfat dairy products, fish, legumes and sources of protein low in saturated fat; along with limited intake of transfatty acids) 9 Exercise program 9 Cardiac rehabilitation program in symptomatic patients or patients with established heart disease
9
Of the above recommendations, the most studied is the subject of weight loss and maintenance of BMI at 19-25 kg/m 2. Prospective studies of weight loss by obese persons
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women have not demonstrated improvements in long-term morbidity and mortality. However, weight reductions correlate with reductions in other risk factors for C H D (314). Modest weight loss of about 10% body weight improves glycemic control, reduces blood pressure and cholesterol levels (315), and has beneficial effects on the psychologic status (316), quality of life (317), and relief of symptoms of dyspnea and chest pain (318). The combination of diet and physical activity results in greater weight loss compared with diet alone (60). Equally important is that weight loss requires conscientious effort of diet control physical activity and programmed exercise for maintaining the lower body weight (319). "Diet pills" are a controversial topic. Currently accepted pharmacotherapy treatment for weight loss includes two classes of drugs: appetite suppressors (e.g., serotoninnorepinephrine re-uptake inhibitors) and gastrointestinal lipase inhibitors (60). Neither has been shown to have a long-term lasting effect, and assessment of long-term safety remains to be determined. Bariatric or weight reduction surgery is considered for people with morbid obesity who are refractory to other weight-reduction interventions. Surgery is considered for persons with BMI -> 40 kg/m 2, or BMI _> 35 kg/m 2 with obesity-related disease when less invasive measures have failed. Bariatric surgical procedures usually fall into two categories: gastric bypass and gastroplasty. Mean weight loss is 60 to 100 lbs, but the procedures carry significant postoperative morbidity and mortality risks (320,321). Review of 136 bariatric surgery studies revealed that more than 72% of patients undergoing the procedure were women, with a mean age of 39 years and baseline mean BMI of 46.9 kg/m 2 (322). The surgery resolved or improved diabetes, hypertension, hyperlipidemia, and obstructive sleep apnea in more than 70% of the patients (322). The concept of dietary modifications, in contrast to the more drastic measures of dietary restrictions and active interventions for weight loss, was tested in the W H I Dietary Modification Trial, which was a randomized intervention trial of more than 48,000 postmenopausal women. About 40% were randomized to an intervention arm that included group and individual sessions to promote a decrease in fat intake and increases in vegetable, fruit, and grain consumption but did not include weight loss or caloric restriction goals. About 60% of the women comprised the control group and received diet-related education materials. Compared with the control group, women in the intervention group lost 1.9 kg at 1 year and 0.4 kg at 7.5 years (p < 0.01-0.1). No tendency toward weight gain was observed in intervention group women overall or when stratified by age, ethnicity, or BMI. Weight loss was greatest among women in either group who decreased their percentage of energy from fat. A similar but lesser trend was observed with increases in vegetable and fruit servings, and a non-significant trend toward weight loss occurred with increasing intake of fiber (322a).
433
Over a mean of 8.1 years, the dietary intervention that reduced total fat intake and increased intakes of vegetables, fruits, and grains did not significantly reduce the risk of CHD, stroke, or CVD in postmenopausal women. However, it achieved mild effects on CVD risk factors: LDL-C levels, diastolic blood pressure, and factor VlIc levels were significantly reduced by 3.55 mg/dL, 0.31 mm Hg, and 4.29%, respectively. In contrast, levels of H D L - C triglycerides, glucose, and insulin did not significantly differ in the intervention versus comparison groups (322b). At present it is difficult to draw definitive conclusions from those studies and their overall negative results. Despite their size and careful execution, criticisms presented above relative to the W H I Hormone Trials about participants' age and health condition apply to the W H I Dietary Modification Trial as well. Exercise, defined as formal training-type activity that can also include recreational sport, should not be confused with physical activity, which is activity at work and leisure, including sport. Moderate-intensity physical activity is defined as expending 5 to 7.5 kcal/min or exercising at 60% to 70% of maximum heart rate or at 60% of VO2max (e.g., brisk walking, swimming, or cycling). Regular exercise increases myocardial contractility and improves electrical stability; decreases heart rate at rest and at submaximal cardiac output; improves endothelial function and flow-mediated dilatation; and stimulates formation of collaterals. At the cellular level, platelet aggregation is reduced and fibrinolytic activity is increased. Inflammatory factors such as CRP are reduced; and lipid oxidation during activity and in postexercise recovery is increased, as are lipoprotein lipase activity, tissue sensitivity to insulin, and disposal of glucose (323,324). Studies conducted mostly in men showed the benefits of moderate exercise for cardiovascular health (325-327). The health risks of obesity in men could be controlled if a person is physically active and fit (137). Moderate exercise decreases total cholesterol LDL-C and triglycerides and increases H D L - C (139,140). Moderate exercise also improves hypertension (80,328) and type 2 diabetes (142). Physical activity halves the risk of type 2 diabetes, CHD, and stroke (134,135) and reduces the risks of hip and vertebral fractures, colon cancer, anxiety, depression, and cognitive decline with age (134). Two observational studies analyzed effects of exercise on C H D and overall mortality in women. In the Nurses Health Study, there was a graded inverse association between physical activity and C H D risk. Walking and vigorous exercise were associated with substantial reductions in C H D risk: women who walked 3 hours per week or exercised vigorously for 1.5 hour per week had 30% to 40% reduced RR of C H D compared with sedentary women, and women who became active in middle or late adulthood had decreased risk of C H D when compared with long-term sedentary counterparts (329). In the Nurses Health Study, vigorous to
434
GORODESKI AND GORODESKI
Diet Only . . . . .
~................
~ ~
..........
Diet + Walking ~
~ ~ 1 ...........- -
Diet + Aerobic Dance ~ / A .........- -
~ ~ ..........
g FmtJRE 31.21 Exercise effects, alone or in combination with diet on body weight parameters. BMI, body mass index; IFA, intraabdominal fat area; SFA, subcutaneous fat area. Modified from ref. 138, with permission.
'2 L
0.80 i~ 0.60 ~. 0.40 0.20 0.00
~,
6
9 eo :~
la e. o z
Time Spent Walking per wk
, .......................
The popular fear that vigorous exercise can trigger cardiac arrest or sudden death has been disproportionally exaggerated. Data obtained predominantly in men reported an absolute risk of sudden death during, and up to 30 minutes after, vigorous exercise of I case per 1.51 million episodes of exertion (332). The increased risk was only seen in men with hypertension (treated or untreated) (333), and it could be significantly reduced by a habitual exercise program. Contraindications to exercise include unstable angina, uncontrolled diabetes, uncontrolled hypertension, exercise-induced arrhythmias, severe stenotic or regurgitant valvular disease, and hypertrophic cardiomyopathy (134).
1.00
|
~
Usual Walking Pace (km/h)
FIGURE 31.22 Exercise effects on C H D risk. Modified from ref. 331, with permission.
moderate exercise was associated with 46% decreased RR of developing diabetes (141). In the Iowa Women's Health Study, which focused on postmenopausal women ages 55 to 69, women who engaged in moderate activities ->4 times a week had 47% reduced RR of cardiovascular mortality than those who did so rarely or not at all. Participants in vigorous activities ->4 times a week had 80% lower RR (330). A study of Japanese women showed that exercise augments diet effects on weight reduction and loss of intra-abdominal and subcutaneous fat (Fig. 31.21) (138). Vigorous intensity activity confers maximal cardiovascular benefit (Fig. 31.22) (331). However, regular moderate intensity physical activity, which is attainable for the majority of women and men, gives considerable protection against C H D (135). Therefore, walking is suggested as the most practical exercise from which the population can achieve such improvements in health (134).
B. Early Diagnosis and Treatment of Associated Medical Conditions A number of known medical conditions can increase the risk of C H D if not treated. Women with type 2 diabetes mellitus should be monitored for HbA1c goal <7%. The Diabetes Prevention Program (DPP) demonstrated that both medications and lifestyle interventions can delay or prevent progression from impaired glucose tolerance (IGT) to type 2 diabetes. Compared with the placebo intervention, intensive lifestyle intervention reduced the incidence of type 2 diabetes by 58% over 2.8 years compared with 31% by metformin (334). The DPP study conclusions were confirmed by the Finnish Diabetes Prevention Study (335). The results of both trials are specifically applicable to women because 67% to 68% of the participants in both studies were perimenopausal or postmenopausal women at mean ages of 50 to 55. Hypertension should be managed aggressively. Blood pressure is considered optimal at <120/80 mm Hg. If blood
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women pressure ->140/90 mm Hg (or ->130/80 mm Hg if the patient has diabetes), drugs should be used to control the hypertension. Current recommendations (82) do not clearly address gender-related issues other than pregnancy and birth control pills. However, as widely adopted also for women (5,62), they are as follows: 1. Thiazide-type diuretics should be used in drug treatment for most patients with uncomplicated hypertension, either alone or combined with drugs from other classes. Certain high-risk conditions are compelling indications for the initial use of other antihypertensive drug classes (angiotensin-converting enzyme inhibitors, angiotensinreceptor blockers, [3-blockers, calcium channel blockers). 2. Most patients with hypertension will require two or more antihypertensive medications to achieve goal blood pressure. 3. If blood pressure is more than 20/10 mm Hg above goal blood pressure, consideration should be given to initiating therapy with two agents, one of which should be a thiazide-type diuretic. Women post-MI or those with chronic angina or chest pain should be treated with ~3-blockers. Women with a history of heart failure should be treated with angiotensin converting enzyme (ACE) inhibitor therapy or angiotensin receptor blocker (ARB); atrial fibrillation should be controlled and warfarin therapy (or 325 mg of aspirin daily) is indicated; and high-risk women with depression should be aggressively treated (7,311). Antioxidant supplements, such as vitamin E and beta carotene, should not be used to prevent heart disease. Severn clinical trials have shown no benefit, and some have shown an unexpected increase in hemorrhagic strokes (7,311). The use of low-dose aspirin is recommended for secondary prevention in women with pre-existing CHD. However, the use of aspirin for primary prevention of C H D in healthy women remains a controversial topic. The Women's Health Study, a randomized placebo controlled trial of low-dose aspirin (100 mg every other day) in 39,876 perimenopausal, healthy health professionals at an average age of 54 showed no statistically significant benefit relative to the combined end point of MI, stroke, and mortality (336). The only derived benefit was a 34% reduction in RR of the secondary end point of ischemic stroke. These results contradict prior findings in men in whom daily use of aspirin reduced the risk of MI but not of stroke (337). However, questions have been raised about whether the dose of aspirin used in the Women's Health Study was too low and whether the results could be generalized to healthy women with greater C H D risk factors. Aspirin for low-risk patients is not recommended because the risks of gastrointestinal bleeding outweigh the potential benefits.
435
C. L i p i d L o w e r i n g T r e a t m e n t The association between cardiovascular events and LDLC is well established, and LDL-C is the primary target of therapy both in men and women (338). LDL-C comprises 60% to 70% of the total serum cholesterol and is the major atherogenic lipoprotein (62). Because LDL-C levels <100 mg/dL are associated with a very low risk for CHD, they are referred to as optimal. Near-optimal LDL-C levels of 100-129 mg/dL are associated with some degree of atherogenesis, and higher levels of 130-159 mg/dL are associated with significant atherogenesis and are referred to as borderline. At high (160-189 mg/dL)and very high levels (->190 mg/dL) atherogenesis is markedly accelerated (62). Similar definitions for total cholesterol and triglycerides are shown in Table 31.9. HDL-C is also an important predictor for coronary events (339), and it has a greater predictive potential in women than in men (340-342). Because there is a continuous rise in risk for C H D as H D L - C levels decline and no inflection point in the C H D versus H D L - C curve (62), low H D L - C levels are defined as <40 mg/dL, and high (i.e., desirable) levels as >60 mg/dL (62,311). Lipoprotein(a), which is unaffected by diet, exercise, and most lipid-modifying medications, is an independent predictor of the risk of recurrent C H D in postmenopausal women (343), but it is not used routinely for treatment evaluation. The intensity of LDL lowering therapy should be adjusted to the individual's absolute risk for CHD. For that purpose, risk assessment of major independent risk factors is based on the 10-year risk assessment Framingham scoring shown in Table 31.8, with some differences. Three categories of C H D risk are identified (62) as established CHD and CHD risk equivalents, multiple (2+) risk factors, and 0-1 risk factor. The major risk factors include age (->55 years in women), cigarette smoking, low H D L - C (<40 mg/dL), hypertension (blood pressure -> 140/90 mm Hg or on anti-
TABLE 31.9 Interpretations of Plasma Levels of LDL Cholesterol, Total Cholesterol, and Triglycerides in Terms of C H D Risk* LDL cholesterol Desirable (optimal) Near optimal Borderline h i g h High Very high
Total cholesterol
Triglycerides
<100 100-129 130-159 160-189 _>190
<200 200-230 _>240
<150 150.199 200-499 _>500
*Allvalues in mg/dL. CHD, coronaryheart disease; LDL, low-densitylipoprotein. Modified from ref. 62.
436
GORODESKIAND GORODV.SKI
TABLE 31.10 Goals of Lipid-Lowering Therapy for Primary Prevention of CHD Based on LDL Cholesterol Levels LDL Cholesterol Risk category Multiple (2 +) risk factors 0-1 risk factor
10-year risk for CHD
Level at which to consider drug therapy
Primary goal of therapy
>20% (includes CHD risk equivalents) 10-20% <10% < 10%
> 100 mg/dL ->130 mg/dL >-160 mg/dL >-190 mg/dL
<1000 mg/dL <130 mg/dL <130 mg/dL < 160 mg/dL
CHD, coronaryheart disease;LDL, low-densitylipoprotein. Modified from ref. 62.
hypertensive medication), family history of premature CHD (CHD in male first-degree relative <55 years; CHD in female first-degree relative <65 years), and type 2 diabetes mellitus. The goals of targeted lowering LDL-C for the primary prevention of CHD are according to the scheme shown in Table 31.10. Treatment options for dyslipidemia in women are based mainly on past clinical trials done predominantly in men with post hoc analyses for women. However, more recent trials also involved women directly and revealed similar beneficial effects in women (62,344-348). The common options for lowering LDL-C in the context of primary prevention of CHD are listed in the following discussion and summarized in Table 31.11. Statins (hydroxymethylglutaryl-coenzyme A reductase inhibitors) have emerged as an efficient, low-risk, cost-effective treatment modality for lowering plasma LDL-C. Statins interfere with the rate-limiting step of cholesterol synthesis; they decrease cholesterol production and upregulate expression and activity of LDL receptors. These changes stimulate the removal of circulating LDL-C, intermediate-density lipoproteins, and VLDL remnants, leading to lower levels of the lipoproteins. Statins are considered first-line therapy to achieve LDL-C goals. In women, reductions in the RR of cardiovascular events ranged from 11% to 46% (344-348), with an overall estimated RR reduction in major coronary events of 29% (349). The Heart Protection Study, which evaluated 5082 women, found that overall reductions in major vascular events with statin therapy was 20% in women and was unaffected by age (348). Post hoc analysis of HERS results showed that statin use was associated with lower rates of cardiovascular events: nonfatal MI or C H D death (hazard ratio [HR] 0.79), total mortality (HR 0.67), and venous thromboembolic events (HR 0.45) (284). Overall, statins reduce coronary mortality and total mortality by 22% to 42% and 9% to 30%, respectively (62). There is little if any protective effect of statins on the development of type 2 diabetes (350). Statins are relatively safe and well tolerated in the majority of patients. Their most common serious
adverse effects are hepatotoxicity and myopathy. They occur at very low rates, and the risk of myopathy in women increases with advanced age (see Table 31.11) (351). Niacin reduces the hepatic synthesis of VLDL and triglycerides by inhibiting mobilization of free fatty acids from peripheral adipose tissue to the liver, thereby decreasing LDL-C. Niacin also decreases HDL-C clearance by blocking hepatic uptake of ApoA-I, thereby increasing the amount of H D L available for reverse cholesterol transport. Clinically, niacin's greatest effect is on HDL-C, but it also effectively decreases triglycerides, LDL-C, and lipoprotein (a) (62). In women, niacin is recommended for the treatment of low HDL-C and elevated triglyceride levels once the LDLC goal has been achieved (338). Niacin use has been limited by serious side effects including flushing and liver toxicity (see Table 31.11) (352). The combination of statin plus niacin for treating dyslipidemia in women adds the favorable effects of niacin on atherogenic dyslipidemia to the LDL-C-lowering action of statins. In clinical studies in women this treatment showed improvements in the plasma lipid profile (353-355). Fibrates are used in patients with atherogenic dyslipidemia to lower triglycerides and raise HDL-C once the LDL-C goal has been achieved (338). Fibrates are believed to increase lipoprotein lipase activity and downregulate ApoC-III, thereby increasing the catabolism of triglyceriderich lipoproteins (62). Reductions in cardiovascular events associated with fibrate therapy have been documented mainly in men, but post hoc analysis referring to women showed similar effects (356-358). Fibrates are generally well tolerated and are considered relatively safe. The main concern is an increased risk of myopathy when fibrates are used in combination with statins, particularly in patients with renal disease (see Table 31.11) (359). The statin/niacin/fibrates combination could be a potential treatment for patients who require extremely aggressive risk reduction and those with lipid abnormalities that persist despite current combination therapy (360). Bile acid sequestrants (BASs) reduce bile acid enterohepatic recirculation, thereby increasing the conversion of
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women
437
TABLE 31.11 Lipid-Lowering Medications HMG CoA reductase inhibitors
Nicotinic acid
Available drugs
Lovastatin, pravastatin, simvastatin, fluvastatin, atorvastatin
Lipid-lipoprotein
LDL 18-55% $
Crystalline, sustained, timed-release, or extended-release nicotinic acid LDL 5-25% ,1,
HDL 5-15% 1"
HDL 15-35% 1"
Triglycerides 7-30% ,l, Lower LDL
Triglycerides 20-50% ,], Most lipid/lipoprotein abnormalities
Active or chronic liver disease
Chronic fiver disease, severe gout Hyperuricemia; high doses in type 2 diabetes
Safety
Co-treatment with cyclosporine, macrolides, anti-fungals, cytochrome P-450 inhibitors CHD and stroke risk reduction Minimal side effects
Major side/adverse effects
Myopathy, increased fiver transaminases
effects
Major use
Contraindications Absolute
Relative
Efficacy
CHD risk reduction Possible serious hepatoxicity with sustainedrelease form Flushing, hyperglycemia, hyperuricemia or gout, upper GI distress, hepatotoxicity for the sustainedrelease form
Fibric acid derivatives
Bile acid sequestrants
Gemfibrozil, fenofibrate, clofibrate
Cholestyramine, colestipol, colesevelam
LDL 5-20% ,], in nonhypertriglyceridemic persons HDL 10-35% 4: in hypertriglyceridemia Triglycerides 20-50% ,], Hypertriglyceridemia, atherogenic dyslipidemia
LDL 15-30% ,], HDL 3-5% $ Triglycerides: no effect Lower LDL
Severe hepatic or renal insufficiency
Familial dys-13-lipoproteinemia Triglycerides >400 mg/dL Triglycerides >200 mg/dL
Moderate CHD risk reduction Increase in non-CHD mortality (?)
CHD risk reduction
Dyspepsia, upper GI complaints, cholesterol gallstones, myopathy
Minimal systemic toxicity; GI side effects common Upper and lower GI complaints; decrease absorption of other drugs
CHD, coronaryheart disease; GI, gastrointestinal; HDL, high-densitylipoprotein; LDL, low-densitylipoprotein. Modified from ref. 62.
cholesterol into bile acids. BASs lower plasma LDL-C; however, they can potentially raise triglycerides, and their use is contraindicated in persons with hypertriglyceridemia or familial dys-13-1ipoproteinemia (359). Lowering LDL-C is considered the first and most important step in primary prevention of CHD, but there may be limits to the degree of benefit that can be achieved by lowering LDL-C alone. This has led to an increased interest in targeting other lipid-related risk factors for CVD, such as HDL-C. Nuclear receptors of the RXR heterodimer family have been shown to regulate key genes involved in H D L metabolism and reverse cholesterol transport. These include the peroxisome proliferator activated receptors (PPARs), the liver X receptor (LXR), and the farnesoid X receptor (FXR). The synthesis of specific and potent ligands for these recep-
tors has aided in ascertaining the physiologic role of these receptors as lipid sensors and the potential therapeutic utility of modulators of these receptors in dyslipidemias and cardiovascular disease (reviewed in [361]). Three subfamilies of the PPARs, including PPAR-ci,-8, and -y, have been identified with crucial roles in lipid metabolism, plaque inflammation, expression of adhesion molecules and cytokines, and regulation of matrix metalloproteinases. PPAR-y regulates several metabolic pathways including lipid biosynthesis and glucose metabolism. Thiazolidinediones (i.e., rosiglitazone, pioglitazone), which are synthetic PPAR-y agonists, act as insulin sensitizers and are used in the treatment of type 2 diabetes. More recently it was found that PPAR-y ligands also regulate cellular inflammatory and ischemic responses. PPAR-y is upregulated in
438
GORODESKI AND GORODESKI
intimal vascular smooth muscle cells after mechanical injury to the vessel wall, and PPAR-y ligands have been shown to inhibit neointima formation in both normal and insulinresistant vasculature. The suppression of intimal hyperplasia results from the potential of PPAR-y to inhibit vascular smooth muscle cells growth and promote apoptosis (362). Thiazolidinediones could attenuate coronary atherosclerosis and stabilize plaque formation and protect the heart and other organs against the tissue injury caused by ischemia/ reperfusion injury and shock (361). The role of thiazolidinediones in the primary prevention of C H D in women remains to be determined.
D. Alcohol The interest in the possible cardioprotective effects of alcohol continues despite conflicting reports (363). A review of 974 studies from 1966 to 2003 found that compared with no alcohol use, moderate consumption (1-3 drinks per day a few days per week) was associated with a 33% to 56% lower incidence of diabetes and a 34% to 55% lower incidence of diabetes-related C H D (364). For obese women (BMI -> 25 kg/m 2) there was an inverse association between moderate alcohol consumption and insulin activity (365). This conclusion confirms earlier reports that light to moderate drinking reduces the incidence of type 2 diabetes in men (366) and young women (367). The diabetes-related benefits seem to derive from alcohol's effects on insulin secretion, resistance, and sensitivity (368). More recent studies found a bimodal relationship between alcohol consumption and C H D morbidity and mortality. Light drinking (1-2 drinks per day a few days per week) reduced the risk, while ---5 drinks increased the risk
Relative Risk of CHD 1.50 [
Women
Me n
1.25
1.00
0.75
0.50 I ..... 0
l
i
i
25
50
75
~
i
100
t ..............
125
i
150
Alcohol Consumption (g/day)
FIGURE 31.23 Mcohol effects on CHD risk. Modified from ref. 369, with permission.
(369-373) (Fig. 31.23). The trends for beneficial C H D effects appeared when daily drinking exceeded 1 and 1.5 drinks/day for women and men, respectively, and was independent of age, smoking habits, and BMI (370,372,374,375). The protective effect was also a function of frequency of consumption because small amounts consumed several times per week reduced risk to a greater extent than the same amount consumed over fewer occasions (376). The mechanism of the cardioprotective effect of light alcohol consumption is not entirely understood. Effects could be secondary to regulation of glucose metabolism, as indicated earlier. Additionally, effects could be secondary to alcohol-induced increases in H D L - C (377), decreased LDL oxidation (378), or reduced blood clotting and platelet aggregation characteristics (reviewed in [363]). Whichever mechanism is responsible for the alcohol effects, the data in Fig. 31.23 suggest involvement of at least two mechanisms operating in different direction and that women are more sensitive than men to both mechanisms. As most of the data about alcohol consumption and CVD/ C H D risks came from observational studies, it would be premature to suggest the use of moderate alcohol consumption as a clinical method for primary prevention of CHD.
E. Achieving the Goal of Low CHD-Risk Increased awareness and better implementation of prevention principles has been shown to decrease the prevalence of C H D risk factors. Active awareness could be defined as lifestyle modifications and participation in health-related programs that will lead to reduction of C H D risk factors, such as cessation of cigarette smoking, decreases in blood pressure to less than 120/80 mm Hg, and decreases in plasma total cholesterol to less than 200 mg/dL. The prevalence of active awareness in the United States rose from 6% in 1971-1975 to 17% in 1999-2000 and was about twice as high in women as in men throughout that period (5,62). The prevalence of active awareness was initially higher in Caucasians than in African Americans (7% versus 3% in 1971-1975), and it increased more with time in the latter (17% versus 15% in 1999-2000). Prevalence of active awareness in 1999-2000 was lowest in persons ages 65-74 (3%) and was progressively greater at younger ages (29% at ages 25-34). The greatest changes in the components of risk from 1971 to 2000 were decreases in diastolic blood pressure (38% to 71%) and systolic blood pressure (32% to 47%), cessation of smoking (60% to 79%), and decreases in plasma cholesterol (33% to 46%) (5,62). In the Nurses Health Study, the incidence of coronary disease declined by 31% from the 2-year period 1980-1982 to the 2-year period 1992-1994. From 1980 to 1992 the proportion of participants currently smoking declined by 41%, but the prevalence of overweight increased by 38%. Taken individually, cessa-
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women tion of smoking could explain a 13% decline in the incidence of coronary disease, as well as the increase in BMI and 8% increase in the incidence of C H D (379).
F. Adherence, or Lack Of?. Despite accumulating evidence for the benefits of primary prevention of C H D and the methods to achieve this goal, participation in prevention programs, initiation of treatment, and long-term adherence to therapy remain far from optimal. The W H I Observational Study (380) has suggested that postmenopausal women are not treated aggressively enough to control their blood pressure to recommended levels. Treatment options for dyslipidemia are currently underused in women at significant risk for future coronary events (359), and adherence to lipid management is less than desirable, as reflected in the following findings: Less than half of persons who qualify for any kind of lipidmodifying treatment for C H D risk reduction are receiving it; less than half of the highest-risk persons, including those with symptomatic CHD, are receiving it; only one-third of treated persons are achieving their LDL-C goal; less than 20% of C H D patients are at their LDL-C goal; only half of the persons who are prescribed lipid-lowering drugs are still taking it 6 months later; after 12 months, this number falls to 30% of persons; and lack of adherence is not related to gender, age, ethnicity, or socioeconomic status (5,62). Patients are not the only ones to carry the burden of blame. Caregivers and health providers have little training in behavioral modification techniques and do not naturally apply behavioral change principles to improve clinical responsibilitytaking behavior. Many primary care providers and other health professionals spend little time in their practices to provide interventions to encourage adherence with therapy. There are too few incentives built into the health delivery system to encourage and support health professionals to address poor adherence among patients (5,62). Interventions to improve primary prevention of C H D should use evidence-based approaches for improving adherence and must include the patient, the caregiver, and the health delivery system. For instance, interventions focused on the patient should simplify medication regimens; provide explicit instructions and good counseling techniques to teach the patient how to follow the prescribed treatment; use systems to reinforce and reward adherence and maintain contact with the patient; and involve patients in their care through self-monitoring. Similarly, interventions focused on the physician and medical office should teach physicians to implement risk assessment and management guidelines; develop a standardized treatment plan to structure care; remind patients of appointments and follow-up on missed appointments; and use feedback from past performance to foster change in future care (5,62).
439
VIII. AN EYE INTO T H E FUTURE Life expectancy statistics are commonly used to forecast longevity from previously known mortality rates. Life expectancy at birth represents the average number of years that an infant is expected to live based on the observed death rates of the population at its birth. The trend in life expectancy during the past millennium has been characterized by a slow steady increase, a pattern frequently punctuated by volatility in death rates caused by epidemics and pandemic infectious diseases, famines, and war (381). In 2002, life expectancy at birth in the United States reached a record high of 77.3 years: 79.9 years for females and 74.5 years for males. This represented a 0.1-year increase for both females and males compared with 2001 (1). Trend analysis shows that since 1930, life expectancy in the United States has steadily risen both for women and men, Caucasians, and African-Americans (Fig. 31.24). Similar patterns were also reported for persons of other races and ethnic groups (1). However, during the past 30 years the rise in life expectancy in the United States slowed relative to that historical pattern, and gains in life expectancy became much smaller than they were in previous decades (382).
FIGURE 31.24 Trendsin life expectancyin the United States for Caucasians (C) and African Americans (AA). Modified from ref. 1, with permission.
440
GORODESKI AND GORODESKI
FIGURE 31.25 Trends in life expectancy in the United States for Caucasians. Arrows designate "turning points" in life expectancy (see Table 31.12). Modified from res 1, with permission.
FIGURE 31.26 Trends in life expectancy in the United States for African Americans. Modified from ref. 1, with permission.
Predictions of life expectancy can be made from extrapolation of data (see Fig. 31.24). Such an analysis is presented below as the lead hypothesis of life expectancy in women in the United States and is shown graphically in Figs. 31.25 and 31.26 and in Table 31.1. However, results of extrapolation models may fail to consider the health status of people currently alive, and they explicitly assume that the past can predict the future. This consideration is the basis for the alternative hypothesis of life expectancy in women.
A. Lead Hypothesis of Life Expectancy in Women in the United States The lead hypothesis predicts a continued increase in life expectancy for both women and men in the United States. When the data of Fig. 31.24 are redrawn (see Figs. 31.25
and 31.26), it is apparent that there was a biphasic trend for Caucasians: an early rise in life expectancy, shown as empty circles, and a late rise, shown as filled circles. The data of the early and the late increases in life expectancy in Caucasians and the data of life expectancy for African Americans could be fitted into exponential curves (see Figs. 31.25 and 31.26; r 2 > 0.97, p < 0.01-0.04). By extrapolation, the following parameters were calculated: (1) points of intersection between the early and late increases of life expectancy in Caucasians (designated as "turning points") (see Figs. 31.25 and 31.26); (2) "maximal" life expectancies; and (3) years to reach the "maximal" life expectancy. These parameters are shown in Table 31.12, suggesting significant differences in these parameters between women and men, Caucasians and African Americans. To better understand those trends, rates of the 10 leading causes of deaths were compared among Caucasian women
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women
441
TABLE 31.12 Life Expectancy Parameters* "Turningpoint" in life expectancy Caucasian women Caucasian men African-American women African-American men
Maximal life expectancy
kkkkkkkkkklkkiP ~ a '
Year to reach maximal life expectancy
1970 1970
82.5 83.0 77.0
2010 2065 2030
N/A
78.0
2100
I
!
Caucasian Wmnm
i
Caucasian
L
Men
0 ~'~
1975
*Calculated from Figs. 31.25 and 31.26.
~:.oa2
~
and men for the years 1960 (the pre-plateau of the early increase in life expectancy) and 2002 (latest available data) (Fig. 31.27). The greater life expectancy in Caucasian women compared with Caucasian men could be explained by the lower death rates from heart disease in women. In both groups, death rates from heart disease decreased significantly from 1960 to 2002. During that period, cancer mortality rates increased in both groups, but the increases were smaller in women than in men (see Fig. 31.27). Therefore, it is possible that (1) the steep increases in life expectancy after 1970 in Caucasians were due to decreases in deaths from heart disease and (2) decreases in heart disease mortality in Caucasian men were offset by increases in cancer mortality, thereby tempering the increase in life expectancy compared with Caucasian women. The predictions presented in Figs. 31.25 and 31.26 and in Table 31.12 are similar to those recently published by the Social Security Administration (382), which predicted that life expectancy in the United States will continue its steady increases, reaching the mid-80s later in the 21st century. The predicted linear increases in life expectancy of Caucasians suggest that if present health and mortality trends continue at their current pace, Caucasian women will reach the maximal predicted life expectancy sooner than Caucasian men (about 83 years of age for both sexes). Caucasian women will reach this target around the year 2010 and Caucasian men around the year 2065 (see Table 31.12).
B. An Alternative Hypothesis of Life Expectancy in Women in the United States The alternative and more ominous hypothesis is that both women and men have already reached their peak life expectancy and that in the future, life expectancy will either remain at its current level or will decrease if current trends of morbidity continue. Furthermore, decreases in life expectancy will be greater in women than in men (381). This hypothesis is supported by two groups of data. First, mortal-
19rio 1975
J
kip
i
!
i
s F
Af[ic~-American Women
L.,
African. Amer ica. Men
F 1975 2002 9
2002
|
10o 200 300 tOO Death Ra~e~fperlO0.Dl~ P~pul~on)
5o0 0
100 Death ~
~
300 800 ~er tOO,0~ Pepulation)
~1o
FIGURE 31.27 Major causes of death in the United States in years 1960, 1975, and 2002. Modified from ref. 1, with permission.
ity rates of colon/rectum cancers in women decreased through the 1950s and breast cancer mortality decreased in the past decade. However, mortality rates of lung/bronchus cancers, which are the leading cancers in women, have increased since the 1960s (Fig. 31.28). In men, mortality rates of all three leading cancers (lung/bronchus, prostate, and colon/rectum) have been declining since the past decade (see Fig. 31.28). If these trends continue into the next decade, rates of cancer mortality will increase more in women than men. The second argument is that rates of C H D risk factors (including obesity, hypercholesterolemia, hypertension, and physical inactivity) are escalating at a greater pace in women than in men (5). The prevalence of these risk factors already has increased in teenager females as compared with males and remains higher in women than in agematched men (5,381). Therefore, if increases in C H D risk factors such as obesity remain constant, the overall negative effect on life expectancy in the United States will be a reduction in life expectancy of one-third to three-fourths
442
GORODESKI AND GORODESKI
FIGURE 31.28 Cancer causes of deaths in the United States. Modified from res 1, with permission.
FIGURE 31.29 Age-related deaths in the United States in years 1968, 1978, and 2002. Modified from ref. 1, with permission.
of a year (381). This prediction applies to both women and men, and in particular to African Americans in whom escalating rates of C H D risk factors and cancer mortality determine the lower predicted life expectancy (ages 77 and 78 in African-American women and men, respectively; see Figs. 31.24 to 31.26, Table 31.12, and [1]).
9 The clinical management of women should be directed to maintain the achievements in health care that were gained in the past 4 decades in order to avoid reversal of the life expectancy trends. Women, more so than men, are at a greater perceived risk of developing heart disease and therefore for dying at an earlier age than in previous years. 9 Physicians will be faced with an increasing dilemma of having to screen for and diagnose heart disease in women at earlier phases of life, despite the low accuracy of current cardiovascular screening/diagnostic methods in women. 9 Statins are considered a breakthrough in the management of lipidemia, and their use since the mid-1980s has been associated with decreases in C H D risks and mortality (385). The accelerated increase in life expectancy in women around the 1970s occurred before statins were used, and their effect on life expectancy in women is unclear. It can be argued that the use of statins in women has not been popularized as in men. On the other hand, a period of 10-15 years should have produced an impact on cardiovascular mortality and on the estimates of life expectancy. Therefore, whether statins will increase life expectancy in women or will only maintain the current level of longevity remains to be seen.
C. Conclusions One of the questions raised in life expectancy analyses is their relevance to life span and longevity. Life span is defined as the recordable longest observed age at death of a human individual (383), which at present is 122 years (384). Usually, records of life span have little relevance because genetic, environmental, and lifestyle diversity guarantees that an overwhelming majority of the population will die long before attaining the age of the longest-lived individual. More pertinent to life expectancy analyses is the parameter of the amount of life gained (i.e., longevity). As shown in Fig. 31.29, in the United States between the years 1978 and 2002, there was an average gain of about 5 years of life, thereby approaching the projected life expectancy of about 83 years. In summary, even if current trends of morbidity continue, there will be only small additional net gain in longevity. From the above discussion, the following conclusions can be drawn: 9 Current mortality data support the hypothesis that in the 21st-century Caucasian women and men will reach life expectancy of about 83 years of age, which appears to be the maximal life expectancy. An alternative hypothesis is that if current trends of morbidity continue, life expectancy will decline in future years.
In summary, C H D in women is predicted to become the pandemic of the 21st century due to escalating increases in C H D risk factors from childhood. The consequences to women could be an earlier onset of CHD, increased prevalence of cardiovascular morbidity and diminished quality of life, decreased life expectancy, and premature death. Active prevention of C H D in women should aim at attenuating the
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal W o m e n
impact of cardiovascular risk factors, beginning as early as childhood. Efforts should include control of eating habits and diet; early detection of plasma lipidemia, diabetes, and hypertension; avoidance of primary and secondary cigarette smoking; and providing education and support for the proper active lifestyle. The role of estrogen replacement in perimenopausal and postmenopausal women for the primary prevention of CVD remains to be determined.
References 1. Kochanek KD, Murphy SL, Anderson RN, Scott C. Deaths: final data for 2002. National Vital Statistics Report CDC 2004;53:1-115. 2. National Heart, Lung, and Blood Institute. Tabulations of the National Health Interview Survey. National Center for Health Statistics. Accessed June, 15, 2005, from www.cdc.gov/nchs/datawh/ftpserv/ ftpdata/ftpdata.htm. 3. National Ambulatory Medical Care Survey, 2001-2002 Summary, CDC/NCHS. 4. Watkins LO. Perspectives on coronary heart disease in African Americans. Rev Cardiovasc Med 2004;5:$3-$13. 5. Heart Disease and Stroke Statistics: 2005 update. Accessed June, 15, 2005, from www.americanheart.org. 6. Peeters A, Mamun AA, Willekens F, Bonneux L. A cardiovascular life history: a life course analysis of the original Framingham Heart Study cohort. Eur HeartJ 2002;23:458-466. 7. Mosca L, Jones WK, King KB, et al. Awareness, perception, and knowledge of heart disease risk and prevention among women in the United States. Arch Fam Med 2000;9:506-515. 8. Shaw LJ, Miller DD, Romeis JC, et al. Gender differences in the noninvasive evaluation and management of patients with suspected coronary artery disease. Ann Intern Med 1994;120:559-566. 9. Schrott HG, Bittner V, Vittinghoff E, et al. Adherence to National Cholesterol Education Program Treatment goals in postmenopausal women with heart disease: the Heart and Estrogen/Progestin Replacement Study (HERS). JAm MedAssoc 1997;277:1281-1286. 10. Miller M, Byington R, Hunninghake D, et al. Sex bias and underutilization of lipid-lowering therapy in patients with coronary artery disease at academic medical centers in the United States and Canada. Arch Intern Med 2000;160:343-347. 11. Malek AM, Alper SL, Izumo S. Hemodynamic shear stress and its role in atherosclerosis.JAm MedAssoc 1999;282:2035-2042. 12. Napoli C, D'Armiento FR Mancini FP, et al. Fatty streak formation occurs in human fetal aortas and is greatly enhanced by maternal hypercholesterolemia. Intimal accumulation of low density lipoprotein and its oxidation precede monocyte recruitment into early atherosclerotic lesions. J Clin Invest 1997;100:2680-2690. 13. McGill HC Jr. Fatty streaks in the coronary arteries and aorta. Lab Invest 1968;18:560-564. 14. Strong JR McGill HC Jr. The natural history of coronary atherosclerosis. Am J Patho11962;40:37-49. 15. Berenson GS, Wattigney WA, Tracy RE, et al. Atherosclerosis of the aorta and coronary arteries and cardiovascular risk factors in persons aged 6 to 30 years and studied at necropsy: The Bogalusa Heart Study. Am J Cardio11992;70:851-858. 16. Ross R. The pathogenesis of atherosclerosis. N Engl J Med 1986;314:488-500. 17. Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of coronary artery disease and the acute coronary syndromes. N EnglJ Med 1992;326:242-250.
443
18. Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med 1999;340:115-226. 19. Steinberg D, Lewis A. Oxidative modification of LDL and atherogenesis. Circulation 1997;95:1062-1071. 20. Sch6nbeck U, Mach F, Sukhova GK, et al. Regulation of matrix metalloproteinase expression in human vascular smooth muscle cells by T lymphocytes: a role for CD40 signaling in plaque rupture? Circ Res 1997;81:448-454. 21. Libby R Current concepts of the pathogenesis of the acute coronary syndromes. Circulation 2001;104:365-372. 22. Hansson G K. Inflammation, atherosclerosis, and coronary artery disease. NEnglJMed 2005;352:1685-1695. 23. Lefkovits J, Plow EF, Topol EJ. Platelet Glycoprotein IIb/IIIa receptors in cardiovascular medicine. N EnglJ Med 1995;332:1553-1559. 24. Ernst E, Matrai A, Marshall M. Blood rheology in patients with ischemic attacks. Stroke 1988;19:634-636. 25. Resch KL, Ernst E, Matrai A, Paulson HE Fibrinogen and viscosity as risk factors for subsequent cardiovascular events in stroke survivors. Ann Int Med 1992;117:371-375. 26. Braunwald E, Antman EM, Beasley JW, et al. Committee on the management of patients with unstable angina. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction-summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines. J Am Coll Cardiol 2002; 40:1366-1374. 27. Yang H, Pu M, Rodriguez D, et al. Ischemic and viable myocardium in patients with non-Qcwave or Qzwave myocardial infarction and left ventricular dysfunction.JAm Coll Cardio12004;43:592-598. 28. Stary HC. Evolution and progression of atherosclerotic lesions in coronary arteries of children and young adults.Arteriosclerosis 1989;99: I19-I32. 29. Solberg LA, Strong JR Risk factors and atherosclerosis lesions: a review of autopsy studies. Arteriosclerosis 1983;3:187-198. 30. Vlodaver Z, Kahn HA, Neufeld HN. The coronary arteries in early life in three different ethnic groups. Circulation 1979;39:541-550. 31. Kalin MF, Zumoff B. Sex hormones and coronary disease: a review of the clinical studies. Steroids 1990;55:330-352. 32. Neufeld HN, Wagenvoort CA, Edwards JE. Coronary arteries in fetuses, infants, juveniles and young adults. Lab Invest 1962;11: 837-844. 33. Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of coronary artery disease and the acute coronary syndromes. N EnglJ Med 1992;326:310-318. 34. Mieres JH, Shaw LJ, Arai A, et al. Role of noninvasive testing in the clinical evaluation of women with suspected coronary artery disease: Consensus Statement from the Cardiac Imaging Committee, Council on Clinical Cardiology, and the Cardiovascular Imaging and Intervention Committee, Council on Cardiovascular Radiology and Intervention, American Heart Association. Circulation 2005;111:682-696. 35. Kessler KM. Syndrome X: the epicardial view. J Am Coll Cardiol 1992;19:32-33. 36. Billinger M, Kloos P, Eberli FR, et al. Physiologically assessed coronary collateral flow and adverse cardiac ischemic events: a follow-up study in 403 patients with coronary artery disease. JAm Coll Cardiol 2002;40:1545-1550. 37. Kloner RA, Rezkalla SH. Cardiac protection during acute myocardial infarction: where do we stand in 2004? JAm CoN Cardio12004;44:276286. 38. Lefer DJ. Myocardial protective actions of nitric oxide donors after myocardial ischemia and reperfusion. New Horizons 1995 ;3:105-112. 39. Young JJ, Kereiakes DJ. Pharmacologic reperfusion strategies for the treatment of ST-segment elevation myocardial infarction. Rev Cardiovasc Med 2003;4:216-227.
444 40. Aldrighi JM, Alecrin IN, Caldas MA, et al. Effects of estradiol on myocardial global performance index in hypertensive postmenopausal women. GynecolEndocrino12004;19:282-292. 41. Alecrin IN, Aldrighi JM, Caldas MA, et al. Acute and chronic effects of oestradiol on left ventricular diastolic function in hypertensive postmenopausal women with left ventricular diastolic dysfunction. Heart 2004;90:777-781. 42. McHugh NA, Cook SM, Schairer JL, Bidgoli MM, Merrill GE Ischemia- and reperfusion-induced ventricular arrhythmias in dogs: effects of estrogen. Am J Physio11995;268:H2569-H2573. 43. Delyani JA, Murohara T, Nossuli TO, Lefer AM. Protection from myocardial reperfusion injury by acute administraiton of 1713-estradiol. J Mol Cell Cardio11996;28:1001-1008. 44. Hale SL, Birnbaum Y, Kloner RA. [3-estradiol, but not ot-estradiol, reduces myocardial necrosis in rabbits after ischemia and reperfusion. Am HeartJ 1996;132:258-262. 45. Kim YD, Lees DE. Estrogen and ischemic heart disease. ChonnamJ Med Sci 1996;9:62-70. 46. Kim YD, Chen B, Beauregard J, et al. 17-beta-estradiol prevents dysfunction of canine coronary endothelium and myocardium and reperfusion arrhythmias after brief ischemia/reperfusion. Circulation 1996;94:2901-2908. 47. Gorodeski GI, Levy MN, Yang T, Goldfarb J, Utian WH. Beneficial effects of estrogen on cardiac stunning in female rabbits. J Soc Gynecol Invest 1999;6:80-87. 48. Report of a W H O consultation: obesity, preventing and managing the global epidemic. World Health Organ Tech ReD Ser 2002;894:i-xii, 1-253. 49. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999-2000. J Am Med Assoc 2002;288:1723-1727. 50. National Center for Health Statistics. NHANES 1999 prevalence of overweight and obesity among adults: US, 1999. Hyattsville, MD: US Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, 1999. 51. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. National Institutes of Health. ObesRes 1998;6:51S-209S. 52. Martin ML, Jensen MD. Effects of body fat distribution on regional lipolysis in obesity. J Clin Invest 1991;88:609-613. 53. Larsson B, Bengtsson C, Bjorntorp P, et al. Is abdominal body fat distribution a major explanation for the sex difference in the incidence of myocardial infarction? Am JEpidemio11992;135:266-273. 54. Armellini F, Micciolo R, Ferrari P, et al. Blood pressure, metabolic variables and adipose tissue distribution in pre and postmenopausal women. A cta Obstet GynecolScand 1990;69:627-633. 55. Manson JE, Colditz GA, Stampfer MJ, et al. A prospective study of obesity and risk of coronary heart disease in women. N EnglJ Med 1990;322:882-889. 56. Pi-Sunyer FX. The epidemiology of central fat distribution in relation to disease. Nutr Rev 2004;62:$120-$126. 57. Jensen DM, Haymond MW, Rizza RA, Cryer PE, Miles JM. Influence of body fat distribution on free fatty acid metabolism in obesity. J Clin Invest 1989;83:1168-1173. 58. Ferraro R, Lillioja S, Fontvieille AM, et al. Lower sedentary metaboric rate in women compared with men. J Clin Invest 1992;90: 780-784. 59. Manson JE, Wilier WC, Stampfer MJ, et al. Body weight and mortality among women. NEnglJMed 1995;333:677-685. 60. Orzano AJ, Scott JG. Diagnosis and treatment of obesity in adults: an applied evidence-based review. J Am Board Fam Pract 2004;17: 359-369. 61. Grundy SM. Obesity, metabolic syndrome, and cardiovascular disease. J Clin EndocrinolMetabolism 2004;89:2595-2600.
GORODESKI AND GORODESKI
62. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Circulation 2002;106:3143-3421. 63. National Center for Health Statistics NHANES IV Report. Accessed June, 15, 2005, from www.cdc.gov/nchs/product/pubs/pubd/hestats/ obes/obese99.htm2002. 64. Health, United States, 2003. HyattsviUe, MD: National Center for Health Statistics, 2003. Accessed June 15, 2005, from www.cdc.gov/ nchs/hus.htm. 65. Dietz WH, Robinson TN. Overweight children and adolescents. N EnglJ Med 2005;352:2100-2109. 66. CDC/NCHS, NHANES, 1999-2002. Accessed June 15, 2005, from www.americanheart.org. 67. Hedley AA, Ogden CL, Johnson CL, et al. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999-2002. JAm MedAssoc 2004;291:2847-2850. 68. Cook S, Weitzman M, Auinger P, Nguyen M, Dietz WH. Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 19881994. Arch PediatrAdd Med 2003;157:821-827. 69. CDC/NCHS, NHANES III (1988-1994). Accessed June 15, 2005, from http://www.americanheart.org. 70. Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults. Findings fom the Third National Health and Nutrition Examination Survey.JAm MedAssoc 2002;287:356-359. 71. McGill HC Jr, McMahan CA, Herderick EE, et al. Origin of atherosclerosis in childhood and adolescence.AmJ Clin Nutr 2000;72:1307S1315S. 72. Kannel WB. Nutrition and the occurrence and prevention of cardiovascular disease in the elderly. Nutr Rev 1988;46:68-78. 73. Matthews KA, Meilahn E, Kuller LH, et al. Menopause and risk factors for coronary heart disease. N EnglJMed 1989;321:641-646. 74. Jensen J, Nilas L, Christiansen C. Influence of menopause on serum lipids and lipoproteins. Maturitas 1990;12:321-331. 75. Ericsson S, Ericsson M, Vitols S, et al. Influence of age on the metabolism of plasma low density lipoproteins in healthy males. J Clin Invest 1991;87:591-596. 76. Wild RA, Painter PC, Coulson PB, Carruth KB, Ranney GB. Lipoprotein lipid concentrations and cardiovascular risk in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1985;61: 946-951. 77. Wild RA, Alanpovic P, Givens JR, Parker IJ. Lipoprotein abnormalities in hirsute women. Am J Obstet Gyneco11992;167:1813-1818. 78. Health, United States, 2003. Hyattsville, MD: National Center for Health Statistics, 2003. Accessed June, 15, 2005, from http://www. cdc.gov/nchs/hus.htm. 79. Kannel WB, Wilson PWF, Zhang TJ. The epidemiology of impaired glucose tolerance and hypertension Am HeartJ 1991;121:1268-1273. 80. Davy KP, Hall JE. Obesity and hypertension: two epidemics or one? Am J Physiol Regul Integr Comp Physio12004;286:R803-R813. 81. Isles CG, Hole DJ, Hawthorne VM, Lever AF. Relation between coronary risk and coronary mortality in women of the Renfrew and Paisley Survey: comparison with men. Lancet 1992;339:702-706. 82. Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. J Am Med Assoc 2003;289: 2560-2571. 83. Dittrich H, Gilpin E, Nicol P, et al. Acute myocardial infarction in women: influence of gender on mortality and prognostic variables.Am J Cardio11988;62:1-7. 84. Manson JE, Colditz GA, Stampfer MJ, et al. A prospective study of maturity-onset diabetes mellitus and risk of coronary heart disease and stroke in women.Arch Intern Med 1991;151:1141-1147.
CHAVTE~ 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal Women 85. Haffner SM, Lehto S, Ronnemaa T, Pyorala K, Laakso M. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without myocardial infarction. N Engl J IVied 1998;339:229-234. 86. Miettinen H, Lehto S, Salomaa V, et al. Impact of diabetes on mortality after the first myocardial infarction. The FINMONICA Myocardial Infarction Register Study Group. Diabetes Care 1998;21:69-75. 87. Malmberg K, Yusuf S, Gerstein HC, et al. Impact of diabetes on longterm prognosis in patients with unstable angina and non-Qzwave myocardial infarction: results of the OASIS (Organization to Assess Strategies for Ischemic Syndromes) Registry. Circulation 2000;102: 1014-1019. 88. Harris MI. Summary. In: National Diabetes Data Group. Diabetes in America, ed 2. NIH publication number 95-1468. Washington, DC: Government Printing Office, 1995:1-14. 89. Jarrett RJ, McCartney P, Keen H. The Bedford Study: ten year mortality rates in newly diagnosed diabetics, borderline diabetics and normoglycaemic controls and risk indices for coronary heart disease in borderline diabetics. Diabetologia 1982;22:79-84. 90. McGill HC Jr, Strong JP, Tracy RE, McMahan CA, Oalmann MC. Pathobiological determinants of atherosclerosis in Youth (PDAY) Research Group. Relation of a postmortem renal index of hypertension to atherosclerosis in youth. Arterioscler Thromb VascBio11995;15: 2222-2228. 91. Ceriello A, Hanefeld M, Leiter L, et al. Postprandial glucose regulation and diabetic complications. Arch Intern Med 2004;164:20902095. 92. Schmidt A, Stern D. Hyperinsulinemia and vascular dysfunction: the role of nuclear factor-KB, yet again. Circ Res 2000;87:722-724. 93. Verma S, Wang CH, Weisel RD, et al. Hyperglycemia potentiates the proatherogenic effects of C-reactive protein: reversal with rosiglitazone. J Mol Cell Cardio12003 ;35:417-419. 94. Moreno PR, Fuster V. New aspects in the pathogenesis of diabetic atherothrombosis. J A m Coll Cardio12004;44:2293-2300. 95. Dickinson S, Brand-Miller J. Glycemic index, postprandial glycemia and cardiovascular disease. Curr Opinion Lipido12005;16:69-75. 96. Boyle JP, Honeycutt AA, Narayan KM, et al. Projection of diabetes burden through 2050: impact of changing demography and disease prevalence in the US. Diabetes Care 2001;24:1936-1940. 97. Winer N, Sowers JR. Epidemiology of diabetes. J Clin Pharmacol 2004;44:397-405. 98. Bush TL, Comstock CW. Smoking and cardiovascular mortality in women. Am J Epidemio11983 ;118:480-488. 99. Rosenberg L, Kaufman DW, Helmrich SP, et at. Myocardial infarction and cigarette smoking in women younger than 50 years of age. J A m MedAssoc 1985;253:2965-2969. 100. LaCroix AZ, Lang J, Scherr P, et al. Smoking and mortality among older men and women in three communities. N Engl J Med 1991;324:1619-1625. 101. Willett WC, Green A, Stampfer MJ, et al. Relative and absolute excess risks of coronary heart disease among women who smoke cigarettes. N EnglJ Med 1987;317:1303-1309. 102. Goldenberg I, Jonas M, Tenenbaum A, et at. Current smoking, smoking cessation, and the risk of sudden cardiac death in patients with coronary artery disease. Arch Intern Med 2003;163:2301-2305. 103. Tobacco-relatedmortality. Fact sheet, http://www.cdc.gov/tobacco. Feb, 2004. 104. The health consequences of smoking, 2004. Report of the Surgeon General (CDC/NCHS). 105. Palmer JR, Rosenberg L, Shapiro S. "Low yield" cigarettes and the risk of nonfatal myocardial infarction in women. N Engl J Med 1989;320:1569-1573. 106. Law MR, Morris JK, Wald NJ. Environmental tobacco smoke exposure and ischemic heart disease: an evaluation of the evidence. Brit MedJ 1997;315:973-980.
445
107. Gupta R, Gurm H, Bartholomew JR. Smokeless tobacco and cardiovascular risk. Arch Intern Med 2004;164:1845-1849. 108. Bolinder G, Alfredsson L, Englund A, de Faire U. Smokeless tobacco use and increased cardiovascular mortality among Swedish construction workers. Am J Public Health 1994;84:399-404. 109. Barnoya J, Glantz SA. Cardiovascular effects of secondhand smoke. Circulation 2005;111:2684-2698. 110. Kaur S, Cohen A, Dolor R, Coffman CJ, Bastian LA. The impact of environmental tobacco smoke on women's risk of dying from heart disease: a meta-analysis. J Womens Health (LarchmO 2004;13:888897. 111. Rosenberg L, Palmer JR, Shapiro S. Decline in the risk of myocardial infarction among women who stop smoking. N Engl J Med 1990; 322:213-217. 112. Accessed June, 15, 2005, from www.who.ch/ntday/ntdayO8. 113. Mjos OD. Lipid effects of smoking. Am HeartJ 1988;115:272-275. 114. Schoenberger JC. Smoking change in relation to changes in blood pressure, weight and cholesterol. Prev Med 1982;11:441-453. 115. Renaud S, Blache D, Dumont E, Therenon C, Wissendanger T. Platelet function after cigarette smoking in relation to nicotine and carbon monoxide. Clin Pharmacol Ther 1984;36:389-395. 116. Davis JW, Davis RF. Acute effects of tobacco cigarette smoking on the platelet aggregate ratio. Am J Med Sci 1979;278:139-143. 117. Markowe HLJ, Marmot MG, Shipley MJ, et at. Fibrinogen: a possible link between social class and coronary heart disease. Br M e d J 1985;291:1312-1314. 118. Ambrose JA, Barua RS. The pathophysiology of cigarette smoking and cardiovascular disease: an update.JAm Coil Cardio12004;43:17311737. 119. Facchini FS, Hollenbeck CB, Jeppesen J, Chen YDI, Reaven GM. Insulin resistance and cigarette smoking. Lancet 1992;339:11281130. 120. Craig WY, Palomaki GE, Haddow JE. Cigarette smoking and serum lipid and lipoprotein concentration: an analysis of published data. Br MedJ 1989;298:784-788. 121. Muscat JE, Harris RE, Haley NJ, Wynder EL. Cigarette smoking and plasma cholesterol. Am Heart J 1991;121:141-147. 122. Daniel M, Martin AD, Faiman C. Sex hormones and adipose tissue distribution in premenopausal cigarette smokers. Int J Obesity 1992;16:245-254. 123. Keil DP, Baron JA, Anderson JJ, Hannan MT, Felson DT. Smoking eliminates the protective effect of oral estrogens on the risk for hip fracture among women. Ann Int Med 1992;116:716-721. 124. Tansavatdi K, McClain B, Herrington DM. The effects of smoking on estradiol metabolism. Minerva Gyneco12004;56:105-114. 125. Hartz AJ, Kelber S, Borkowf H, et al. The association of smoking with clinical indicators of altered sex steroids-a study of 50,145 women. Public Health Rep 1987;102:254-259. 126. Lesko SM, Rosenberg L, Kaufman DW, et al. Cigarette smoking and the risk of endometrial cancer. N EnglJMed 1985;313:593-596. 127. Baron JA, La Vecchia C, Levi F. The antiestrogenic effect of cigarette smoking in women. Am J Obstet Gynecol 1990;162:502-514. 128. Barbieri RL, McShane RM, Ryan KJ. Constituents of cigarette smoke inhibit human granulosa cell aromatase. Fertil Sterff 1986; 46:232-236. 129. Michnovicz JJ, Hershcopf RJ, Naganuma H, Bradlow HL, Fishman J. Increased 2-hydroxylation of estradiol as a possible mechanism for the antiestrogenic effect of cigarette smoking. N EnglJ Med 1986; 315:1305-1309. 130. Youth Risk Behavior Surveillance [YRBS], United States, 2003. M M W R May 21, 2004;53(SS-2), CDC/NCHS. 131. Cigarette Smoking Among High School Students-11 States, 19911997. M M W R Aug 1999;48(31), CDC/NCHS. 132. Second national report on human exposure to environmental chemicals: tobacco smoke. CDC/NCHS 2003:80.
446 133. Pate RR, Pratt M, Blair SN, et al. Physical activity and public health. A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. J Am Med Assoc 1995;273:402-407. 134. Britton A, McPherson K. Monitoring the progress of the 2010 target for coronary heart disease mortality: estimated consequences on CHD incidence and mortality from changing prevalence of risk factors. National Heart Forum: London, 2002. Accessed June 15, 2005, from www.americanheart.org. 135. Oguma Y, Shinoda-Tagawa T. Physical activity decreases cardiovascular disease risk in women: review and meta-analysis. Am J Prey Med 2004;26:407-418. 136. SallisJF, Patterson TL, Buono MJ, Nader PR. Relation of cardiovascular fitness and physical activity to cardiovascular risk factors in children and adults. Am J Epidemio11988;127:933-941. 137. Welk GJ, Blair SN. President's Council on Physical Fitness, Sports. Physical activity protects against the health risks of obesity. Research Digest 2000;3:1-7. 138. Okura T, Nakata Y, Tanaka K. Effects of exercise intensity on physical fitness and risk factors for coronary heart disease. Obesity Res 2003;11:1131-1139. 139. Stefanick ML, Wood PD. Physical activity, lipid and lipoprotein metabolism, and lipid transport. In: Bouchard C, Shephard RJ, Stephens T, eds. Physicalactivity,fitness and health: internationalproceedings and consensus statement. Champaign, IL: Human Kinetics, 1994:417-431. 140. Kelley GA, Kelley KS, Tran ZV. Aerobic exercise and lipids and lipoproteins in women: a meta-analysis of randomized controlled trials. J Womens Health (Larchmt) 2004;13:1148-1164. 141. Manson JE, Rimm EB, Stampfer MJ, et al. Physical activity and incidence of non-insulin-dependent diabetes mellitus in women. Lancet 1991;338:774-778. 142. Manson JE, Nathan DM, Krolewski AS, et al. A prospective study of exercise and incidence of diabetes among US male physicians. JAm MedAssoc 1992;268:63-67. 143. Owens JF, Matthews KA, Wing RR, Kuller LH. Can physical activity mitigate the effects of aging in middle aged women? Circulation 1992;85:1265-1270. 144. Considine RV. Acute and chronic effects of exercise on leptin levels in humans. JAppl Physio11997;83:3-4. 145. Colditz GA, Stampfer MJ, Willett WC, et al. A prospective study of parental history of myocardial infarction and coronary heart disease in women. Am J Epidemio11986;123:48-58. 146. Schildkraut JM, Myers RH, Cupples LA, Kiely DK, Kannel WB. Coronary risk associated with age and sex of parental heart disease in the Framingham Study. AmJ Cardio11989;64:555-559. 147. Lusis AJ, Weinreb A, Drake TA, Allayee H. Genetics of atherosclerosis. In: Topol E, ed. Textbook of cardiovascular medicine, ed 2. Philadelphia: Lippincott, Williams & Wilkins, 2002. 148. Jee SH, Song KS, Shim WH, Kim HK, Sub I, Park JY, et al. Major gene evidence after MTHFR-segregation analysis of serum homocysteine in families of patients undergoing coronary arteriography. Hum Genet 111:128-135, 2002. 149. King RA, Rotter JI, Motulsky AG. The genetic basis ofcommon diseases. New York: Oxford University Press, 2002. 150. Tournier-Lasserve E. New players in the genetics of stroke. NEnglJ Med 2002;347:1711-1712. 151. Goldstein JL, Hobbs HH, Brown MS. Familial hypercholesterolemia. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolicbasis of inherited disease. New York: McGraw-Hill, 1995:1267-1282. 152. Pajukanta P, Cargill M, Viitanen L, et al. Two loci on chromosomes 2 and X for premature coronary heart disease identified in early- and late-settlement populations of Finland. Am J Hum Genet 2000;67: 1481-1493.
GORODESKI AND GORODESKI
153. Francke S, Manraj M, Lacquemant C, et al. A genome-wide scan for coronary heart disease suggests in Indo-Mauritians a susceptibility locus on chromosome 16p13 and replicates linkage with the metabolic syndrome on 3q27. Hum Mol Genet 2001;10:2751-2765. 154. Broeckel U, Hengstenberg C, Mayer B, et al. A comprehensive linkage analysis for myocardial infarction and its related risk factors. Nat Genet 2002;30:210-214. 155. Wang Q~ Rao S, Shen GQ~ et al. Premature myocardial infarction novel susceptibility locus on chromosome lp34-36 identified by genome-wide linkage analysis. Am J Hum Genet 2004;74:262-271. 156. Lusis AJ. Atherosclerosis. Nature 2000;407:233-241. 157. Wang X, Paigen B. Q.uantitative trait loci and candidate genes regulating HDL cholesterol: a murine chromosome map. Arterioscler Thromb VascBid 2002;22:1390-1401. 158. Lusis AJ, Mar R, Pajukanta E Genetics of atherosclerosis. Annu Rev Genomics Hum Genet 2004;5:189-218. 159. Stolarz K, Staessen JA, Kawecka-Jaszcz K, et al. Genetic variation in CYPllB2 and AT1R influences heart rate variability conditional on sodium excretion. Hypertension 2004;44:156-162. 160. Massona LF, McNeill BG. The effect of genetic variation on the lipid response to dietary change. Curr Opin Lipido12005;16:61-67. 161. Miller M, Rhyne J, Hamlette S, Birnbaum J, Rodriguez A. Genetics of HDL regulation in humans. Curr Opin Lipido12003;14:273-279. 162. Glomset JA. The metabolic role of lecithin cholesterol acyltransferase: perspectives from pathology. Adv Lipid Res 1973;11:1-65. 163. Hovingha GK, de Groota E, van-der Steega W, et al. Inherited disorders of HDL metabolism and atherosclerosis. Curt Opinion Lipidol 2005;16:139-145. 164. Herrington D, Potvin-Klein K. Pharmacogenetics of estrogen replacement therapy. JAppl Physio12001;91:2776-2784. 165. Henrynk MH, Fuqua SAW. Estrogen receptor mutations in human disease. Endocrine Rev 2004;25:869-898. 166. Herrington DM, Howard TD. ER-alpha variants and the cardiovascular effects of hormone replacement therapy. Pharmacogenomics 2003;4:269-277. 167. Fraser GE, Strahan TM, Sabate J, Beeson L, Kissinger D. Effects of traditional coronary risk factors on rates of incident coronary events in a low risk population. Circulation 1992;86:406-413. 168. Hunt BJ. The relation between abnormal hemostatic function and the progression of coronary disease. Curt Opin Cardio11990;5:758-765. 169. Sesso HD, Kawachi I, Vokonas PS, Sparrow D. Depression and the risk of coronary heart disease in the Normative Aging Study. Am J Cardio11998;82:851-856. 170. Barth J, Schumacher M, Herrmann-Lingen C. Depression as a risk factor for mortality in patients with coronary heart disease: a metaanalysis. PsychosomMed 2004;66:802-813. 171. Vinkers DJ, Stek ML, Gusseldoo J, Van Der Mast RC, Westendorp RG. Does depression in old age increase only cardiovascular mortality? The Leiden 85-plus Study. IntJ Geriatr Psychiatry 2004;19:852-857. 172. Wassertheil-Smoller S, Shumaker S, Ockene J, et al. Depression and cardiovascular sequelae in postmenopausal women. The Women's Health Initiative (WHI). Arch Intern Med 2004;164:289-298. 173. Wolff B, Grabe HJ, Volzke H, et al. Relation between psychological strain and carotid atherosclerosis in a general population. Heart 2005;91:460-464. 174. Lett HS, Blumenthal JA, Babyak MA, et al. Depression as a risk factor for coronary artery disease: evidence, mechanisms, and treatment. Psychosom Med 2004;66:305-315. 175. Van Melle JR de Jonge P, Spijkerman TA, et al. Prognostic association of depression following myocardial infarction with mortality and cardiovascular events: a meta-analysis. PsychosomMed 2004;66:814-822. 176. Joynt KE, Whellan DJ, O'Connor CM. Depression and cardiovascular disease: mechanisms of interaction. Biol Psychiatry 2003;54: 248-261.
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal W o m e n 177. Sauer WH, Berlin JA, Kimmel SE. Selective serotonin reuptake inhibitors and myocardial infarction. Circulation 2001;104:1894-1898. 178. Sauer WH, Berlin JA, Kimmel SE. Effect of antidepressants and their relative affinity for the serotonin transporter on the risk of myocardial infarction. Circulation 2003;108:32-36. 179. Berkman LF, Blumenthal J, Burg M, et al. Effects of treating depression and low perceived social support on clinical events after myocardial infarction: the Enhancing Recovery in Coronary Heart Disease Patients (ENRICHD) Randomized Trial.JAm MedAssoc 2003;289: 3106-3116. 180. Herd JA. Cardiovascular response to stress. PhysiolRev 1991;71:305330. 181. Toiler GH, Brezinski D, Schafer AI, et al. Concurrent morning increase in platelet aggregability and the risk of myocardial infarction and sudden cardiac death. N E n g l J M e d 1987;316:1514-1518. 182. Willich SN, Toiler GH, Brezinski DA, et al. Platelet o~2 adrenoceptor characteristics during the morning increase in platelet aggregability. Eur HeartJ 1992;13:550-555. 183. Panza JA, Epstein SE, O.u)Tumi AA. Circadian variation in vascular tone and its relation to sympathetic vasoconstrictor activity. N EnglJ Med 1991;325:986-990. 184. Manfredini R, Gallerani M, Portaluppi F, Fersini C. Relationships of the circadian rhythms of thrombotic, ischemic, hemorrhagic, and arrhythmic events to blood pressure rhythms. Ann N Y Acad Sci 1996;783:141-158. 185. Ballinger SE, Walker WL. Not THE change of life." breaking the menopause taboo. Ringwood, Victoria, Australia: Penguin, 1987. 186. Ballinger S. Stress as a factor in lowered estrogen levels in the early postmenopause. Ann N YAcad Sci 1990;592:95-113. 187. Woods NF, Derby GK, Most A. Stressful life events and perimenstrual symptoms. J Hum Stress 1982;8:23-31. 188. Stratton JR, Levy WC, CaldwellJH, et al. Effects of aging on cardiovascular responses to parasympathetic withdrawal. J A m Coll Cardiol 2003;41:2077-2083. 189. Tse HF, Oral H, Pelosi F, et al. Effect of gender on atrial electrophysiologic changes induced by rapid atrial pacing and elevation of atrial pressure. J Cardiovasc Electrophvsio12001;12:986-989. 190. Malloy KJ, Bahinski A. Cardiovascular disease and arrhythmias: unique risks in women.J Gend Spec~Med 1999;2:37-44. 191. Pham TV, Rosen MR. Sex, hormones, and repolarization. Cardiovasc Res 2002;53:740-751. 192. Carnethon MR, Anthony MS, Cascio WE, et al. A prospective evaluation of the risk of Q_T prolongation with hormone replacement therapy: the atherosclerosis risk in communities study. Ann Epidemiol 2003;13:530-536. 193. Hsia J, Aragaki A, Bloch M, et al. Predictors of angina pectoris versus myocardial infarction from the Women's Health Initiative Observational Study. Am J Cardio12004;93:673-678. 194. National Institutes of Health: Chartbook on cardiovascular, lung and blood diseases. Bethesda, MD: US Department ofHHealth and Human Services, May 1992. 195. Weinblatt E, Shapiro S, Frank CW. Prognosis of women with newly diagnosed coronary heart disease: a comparison with course of disease among men. Am J Public Health 1973;63:577-593. 196. Murabito JM, Evans JC, Larson MG, Levy D. Prognosis after the onset of coronary heart disease: an investigation of differences in outcome between the sexes according to initial coronary disease presentation. Circulation 1993;88:2548-2555. 197. Hurst W. The heart, arteries and veins, ed 10. New York: McGrawHill, 2002. 198. Cunningham MA, Lee TH, Cook EF, et al. The effect of gender on the probability of myocardial infarction among emergency department patients with acute chest pain: a report from the Multicenter Chest Pain Study Group.J Gen Intern Med 1989;4:392-398.
447
199. Robinson K, Conroy RM, Mulcahy R, Hickey N. Risk factors and in hospital course of first episode of myocardial infarction or acute coronary insufficiency in women. J A m Coll Cardio11988;11:932-936. 200. Maynard C, Litwin PE, Martin JS, Weaver WD. Gender differences in the treatment and outcome of acute myocardial infarction: results from the Myocardial Infarction Triage and Intervention Registry. Arch Intern Med 1992;152:972-976. 201. Lerner DJ, Kannel WB. Patterns of coronary heart disease morbidity and mortality in the sexes. A 26 year follow-up of the Framingham population. Am HeartJ 1986;111:383-390. 202. Basaria S, Dobs AS. Treatment of hyperlipidemia in women. Int J Fertil Womens Med 2000;45:22-33. 203. Kudenchuk PJ, Maynard C, Martin JS, Wirkus M, Weaver WD. Comparison of presentation, treatment, and outcome of acute myocardial infarction in men versus women (the Myocardial Infarction Triage and Intervention Registry). Am J Cardio11996;78:9-14. 204. Goldberg RJ, O'Donnell C, Yarzebski J, et al. Sex differences in symptom presentation associated with acute myocardial infarction: a population-based perspective. Am HeartJ 1998;136:189-195. 205. Alexander KP, Peterson ED. Medical and surgical management of coronary artery disease in women. Am JManag Care 2001;7:951-956. 206. Kyker KA, Limacher MC. Gender differences in the presentation and symptoms of coronary artery disease. Curr Womens Health Rep 2002;2:115-119. 207. DeVon HA, Zerwic JJ. The symptoms of unstable angina: do women and men differ? Nurs Res 2003;52:108-118. 208. Patel H, Rosengren A, Ekman I. Symptoms in acute coronary syndromes: does sex make a difference? Am HeartJ 2004;148:27-33. 209. Toiler GH, Stone PH, Muller JE, et al. Effects of gender and race on prognosis after myocardial infarction: adverse prognosis for women, particularly black women.JAm Coll Cardio11987;9:473-482. 210. Tunstall-Pedoe H, Morrison C, Woodward M, Fitzpatrick B, Watt G. Sex differences in myocardial infarction and coronary deaths in the Scottish MONICA population of Glasgow 1985-1991: presentation, diagnosis, treatment, and 28-day case fatality of 3991 events in men and 1551 events in women. Circulation 1996;93:1981-1992. 211. Rosengren A, Wallentin L, Gitt AK, et al. Sex, age, and clinical presentation of acute coronary syndromes. Eur Heart J 2004;25:663670. 212. Douglas PS, Ginsburg GS. The evaluation of chest pain in women. N E n g l J M e d 1996;334:1311-1315. 213. Rosen AB, Tsai JS, Downs SM. Variations in risk attitude across race, gender, and education. Med Decis Making 2003;23:511-517. 214. Bugiardini R, Bairey-Merz CN. Angina with "normal" coronary arteries: a changing philosophy. J A m MedAssoc 2005;293:477-484. 215. Hutchinson SJ, Poole-Wilson PA, Henderson AH. Angina with normal coronary arteries: a review. Q J M e d 1989;72:677-688. 216. Ockene IS, Shay MJ, Alpert JS, Weiner BH, Dalen JE. Unexplained chest pain in patients with normal coronary arteriograms. N EnglJ Med 1980;303:1249-1252. 217. Leppert J, Aberg H, Ringqvist F, Sorensson S. Raynaud's phenomenon in a female population: prevalence and association with other conditions. Angiology 1987;38:871-877. 218. Kronenberg E Hot flashes: epidemiology and physiology. Ann N Y Acad Sci 1990;592:52-86. 219. Diver DJ, Bier JD, Ferreira PE, et al. Clinical and arteriographic characterization of patients with unstable angina without critical coronary arterial narrowing (TIMI-IIIA Trial). Am J Cardiol 1994; 74:531-537. 220. Johnson BD, Shaw LJ, Buchthal SD, et al. Prognosis in women with myocardial ischemia in the absence of obstructive coronary disease: results from the National Institutes of Health-National Heart, Lung, and Blood Institute-sponsored Women's Ischemia Syndrome Evaluation (WISE). Circulation 2004;109:2993-2999.
448 221. Theres H, Maier B, Gothe RM, et al. Influence of gender on treatment and short-term mortality of patients with acute myocardial infarction in Berlin. Z Kardio12004;93:954-963. 222. Tamis-Holland JE, Palazzo A, Stebbins AL, et al. Benefits of direct angioplasty for women and men with acute myocardial infarction: results of the Global Use of Strategies to Open Occluded Arteries in Acute Coronary Syndromes Angioplasty (GUSTO II-B) Angioplasty Substudy. Am HeartJ 2004;147:133-139. 223. Ayanian JZ, Epstein AM. Differences in the use of procedures between women and men hospitalized for coronary heart disease. N EnglJ Med 1991;325:222-225. 224. Steingart RM, Packer M, Hamm P, et al. Sex differences in the management of coronary artery disease. N Engl J Med 1991;325:226230. 225. Chirikos TN, Nickel JL. Work disability from coronary heart disease in women. Women Health 1984;9:55-74. 226. Todaro JF, Shen BJ, Niaura R, Tilkemeier PL, Roberts BH. Do men and women achieve similar benefits from cardiac rehabilitation? J Cardiopulm Rehabi12004;24:45-51. 227. Regitz-Zagrosek V, Lehmkuhl E, Lehmkuhl HB, Hetzer R. Gender aspects in heart failure: pathophysiology and medical therapy. Arch Mal Coeur Vaiss2004;97:899-908. 228. Stromberg A, Martensson J. Gender differences in patients with heart failure. Eur J CardiovascularNursing 2003;2:7-18. 229. Lund LH, Mancini D. Heart failure in women. Med Clin North Am 2004;88:1321-1345. 230. McGill HC Jr, McMahan CA, Malcom GT, et al. Effects of serum lipoproteins and smoking on atherosclerosis in young men and women. Arterioscler Thromb VascBio11997;17:95-106. 231. Holman RL, McGill HC Jr, Strong JP, Geer JC. The natural history of atherosclerosis. The early aortic lesions as seen in New Orleans in the middle of the 20th century. Am J Patbo11958;2:209-235. 232. Tejada C, Strong JP, Montenegro MR, Restrepo C, Solberg LA. Distribution of coronary and aortic atherosclerosis by geographic location, race, and sex. Lab Invest 1968;18:509-526. 233. Barter PJ, Nicholls S, Rye KA, Anantharamaiah GM, Navab M, Fogelman AM. Antiinflammatory properties of HDL. Circ Res 2004; 95:764-772. 234. Nissen SE, Tsunoda T, Tuzcu EM, et al. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes: a randomized controlled trial. J Am Med Assoc 2003 ;290:2292-2300. 235. Ohta T, Hattori S, Murakami M, Nishiyama S, Matsuda I. Age and sex related differences in lipoproteins containing apoprotein A-I. Arteriosclerosis 1989;9:90-95. 236. Bush TL. The epidemiology of cardiovascular disease in postmenopausal women.Ann NYAcad Sci 1990;592:263-271. 237. Walsh BW, Schiff I, Rosner B, Greenberg L, Ravinkar V, Sacks FM. Effects of postmenopausal estrogen replacement on the concentrations and metabolism of plasma lipoproteins. N EnglJ Med 1991; 325:1196-1204. 238. Sacks FM, McPherson R, Walsh BW. Effect of postmenopausal estrogen replacement on plasma Lp(a) lipoprotein concentrations. Arch Int ivied 1994;154:1106-1110. 239. Bruckert E, Turpin G. Estrogens and progestins in postmenopausal women, influence on lipid parameters and cardiovascular risk. Horm Res 1995;43:100-103. 240. Haines CJ, Chung TK, Masarei JR, Tomlinson B, Lau JT. An examination of the effect of combined cyclical hormone replacement therapy on lipoprotein(a) and other lipoproteins. Atherosclerosis 1996;119: 215-222. 241. Kim CJ, Min YK, Ryu WS, Kwak JW, Ryoo UH. Effect of hormone replacement therapy on lipoprotein(a) and lipid levels in postmenopausal women. Influence of various progestogens and duration of therapy. Arch Int Med 1996;156:1693-1700.
GORODESKI AND GORODESKI
242. Mehilli J, Ndrepepa G, Kastrati A, et al. Gender and myocardial salvage after reperfusion treatment in acute myocardial infarction. J A m Coll Cardio12005;45:828-831. 243. Zhao X, Eghbali-Webb M. Gender-related differences in basal and hypoxia-induced activation of signal transduction pathways controlling cell cycle progression and apoptosis, in cardiac fibroblasts. Endocrine 2002;18:137-145. 244. Curl CL, Wen& IR, Canny BJ, Kotsanas G. Effects of ovariectomy and 17 beta-oestradiol replacement on [Ca21]i in female rat cardiac myocytes. Clin Exp Pharmacol Physio12003;30:489-494. 245. Jovanovic S, Jovanovic A, Shen WK, Terzic A. Low concentrations of 17-beta-estradiol protect single cardiac cells against metabolic stressinduced Ca2i loading. J A m Coll Cardio12000;36:948-952. 246. Anderson SE, Kirkland DM, Beyschau A, Cala PM. Acute effects of 1713-estradiol on myocardial pH, Nal, and Ca21 and ischemiareperfusion injury. Am J Physiol Cell Physio12005;288:C57-C64. 247. Pelzer T, Loza PA, Hu K, et al. Increased mortality and aggravation of heart failure in estrogen receptor-beta knockout mice after myocardial infarction. Circulation 2005;111:1492-1498. 248. Kannel WB, Hjortland MC, McNamara PM, Gordon T. Menopause and risk of cardiovascular disease. The Framingham Study. Ann Intern Med 1976;85:447-452. 249. Hu FB, Grodstein F, Hennekens CH, et al. Age at natural menopause and risk of cardiovascular disease. Arch Intern Med 1999;159:10611066. 250. Colditz GA, Willett WC, Stampfer MJ, et al. A prospective study of age at menarche, parity, age at first birth and coronary heart disease in women. Am J Epidemio11987;126:861-870. 251. LaVecchia C, Decarli A, Franceschi S, et al. Menstrual and reproductive factors and risk of myocardial infarction in women under 55 years of age. Am J Obstet Gyneco11987;157:1108-1112. 252. The Writing Group for the PEPI Trial. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial.JAm MedAssoc 1995;273:199-208. 253. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis trial research group. Estrogen in the prevention of atherosclerosis. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 2001;135:939-953. 254. Grady D, Wenger NK, Herrington D, et al. Postmenopausal hormone therapy increases risk for venous thromboembolic disease. The Heart and Estrogen/Progestin Replacement Study. Ann Intern Med 2000; 132:689-696. 255. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N E n g l J M e d 2000;343:522-529. 256. RossouwJE, Anderson GL, Prentice RL, et al. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. J A m MedAssoc 2002;288:321-333. 257. Anderson GL, Limacher M, Assaf AR, et al. Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy. The Women's Health Initiative randomized controlled trial. J Am Med Assoc 2004;291: 1701-1712. 258. McDonnell DR The molecular pharmacology of estrogen receptor modulators: implications for the treatment of breast cancer. Clin Cancer Res 2005;1 l:871s-877s. 259. Rubanyi GM, Kauser K, Johns A. Role of estrogen receptors in the vascular system. VasculPharmaco12002;38:81-88. 260. Nuedling S, Karas RH, Mendelsohn ME, et al. Activation of estrogen receptor beta is a prerequisite for estrogen-dependent upregulation of nitric oxide synthases in neonatal rat cardiac myocytes. FEBS Lett 2001;502:103-108.
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal W o m e n 261. Forster C, Kietz S, Hultenby K, Warner M, Gustafsson JA. Characterization of the ER beta mouse heart. Proc Natl Acad Sci USA 2004;101:14234-14239. 262. Bjornstrom L, Sjoberg M. Mechanisms of estrogen receptor signaling: convergence ofgenomic and nongenomic actions on target genes. Mol Endocrino12005;19:833-842. 263. Hisamoto K, Bender JR. Vascular cell signaling by membrane estrogen receptors. Steroids 2005;70:382-387. 264. Huss JM, Torra IP, Staels B, Giguere V, Kelly DP. Estrogen-related receptor alpha directs peroxisome proliferator-activated receptor alpha signaling in the transcriptional control of energy metabolism in cardiac and skeletal muscle. Mol Cell Bio12004;24:9079-9091. 265. Gorodeski GI. Aging and estrogen effects on transcervical-transvaginal epithefial permeability. J C/in Endocrinol Metab 2005;90:345-351. 266. Gorodeski GI. Epidemiology and risk factors for cardiovascular disease in postmenopausal women. Baillieres Best Pract Res Clin Obstet Gyneco12002;16:329-357. 266a. National Institues of Health: The Will legacy to future generations of women." update on scientific contributions. Bethesda, MD, NIH, Feb 28-March 1, 2006. 267. Kanaya AM, Herrington D, Vittinghoff E, et al. Glycemic effects of postmenopausal hormone therapy: the Heart and Estrogen/Progestin Replacement Study: a randomized, double-blind, placebo-controlled trial. Ann Intern Med 2003;138:1-9. 268. Moria M, Tsukaharab F, Yoshiokac T, Irieb K, Ohtaa H. Suppression by 1713-estradiol of monocyte adhesion to vascular endothelial cells is mediated by estrogen receptors. Life Sciences 2004;75:599-609. 269. Bailey MS, Curtis AB. The effects of hormones on arrhythmias in women. Curt Womens Health ReD 2002;2:83-88. 270. Christ M, Seyffart K, Tillmann HC, Wehling M. Hormone replacement in postmenopausal women: impact of progestogens on autonomic tone and blood pressure regulation. Menopause 2002;9:127136. 271. Ganger KF, Vyas S, Whitehead M, et al. Pulsatility index in internal carotid artery in relation to transdermal oestradiol and time since menopause. Lancet 1991;338:839-842. 272. Lobo RA. Cardiovascular implications of estrogen replacement therapy. Obstet Gyneco11990;75:18S-25S. 273. Kadish AH, Greenland P, Limacher MC, et al. Estrogen and progestin use and the Q..T interval in postmenopausal women. Ann Noninvasive Electrocardio12004;9:366-374. 274. Sugishita K, Li F, Su Z, Barry WH. Anti-oxidant effects of estrogen reduce [Ca21]i during metabolic inhibition. J Mol Cell Cardiol 2003;35:331-336. 275. Hamilton KL, Gupta S, Knowlton AA. Estrogen and regulation of heat shock protein expression in female cardiomyocytes: cross-talk with NF kappa B signaling. J Mol Cell Cardio12004;36:577-584. 276. Patten RD, Pourati I, Aronovitz MJ, et al. 1713-estradiol reduces cardiomyocyte apoptosis in vivo and in vitro via activation of phosphoinositide-3 kinase/Akt signaling. Circ Res 2004;95:692-699. 277. Babiker FA, De Windt LJ, van Eickels M, et al. 1713-estradiol antagonizes cardiomyocyte hypertrophy by autocrine/paracrine stimulation of a guanylyl cyclase A receptor-cyclic guanosine monophosphatedependent protein kinase pathway. Circulation 2004;109:269-276. 278. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/Progestin Replacement Study (HERS) Research Group.JAm MedAssoc 1998;280:605-613. 279. Shlipak MG, Chaput LA, Vittinghoff E, et al. Heart and Estrogen/ Progestin Replacement Study Investigators. Lipid changes on hormone therapy and coronary heart disease events in the Heart and Estrogen/Progestin Replacement Study (HERS). Am Heart J 2003;146:870-875.
449
279a. Hsia J, Langer RD, Manson JE, et al. Conjugated equine estrogens and coronary heart disease: The Women's Health Initiative. Arch Intern Med 2006;166:357-365. 280. Pradhan AD, Manson JE, Rossouw JE, et al. Inflammatory biomarkers, hormone replacement therapy, and incident coronary heart disease: prospective analysis from the Women's Health Initiative observational study. J A m MedAssoc 2002;288:980-987. 281. Zanger D, Yang BK, Ardans J, et al. Divergent effects of hormone therapy on serum markers of inflammation in postmenopausal women with coronary artery disease on appropriate medical management. J A m Coll Cardio12000;36:1797-1802. 282. Koh KK, Son JW, Ahn JY, et al. Non-lipid effects of statin on hypercholesterolemic patients established to have coronary artery disease who remained hypercholesterolemic while eating a step-II diet. Coron Artery Dis 2001;12:305-311. 283. Grady D, Herrington D, Bittner V, et al. HERS Research Group. Cardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/Progestin Replacement Study follow-up (HERS II).JAm MedAssoc 2002;288:49-57. 284. Herrington DM, Vittinghoff E, Lin F, et al. The HERS Study Group. Statin therapy, cardiovascular events, and total mortality in the Heart and Estrogen/Progestin Replacement Study (HERS). Circulation 2002;105:2962-2967. 285. Manson JE, Hsia J, Johnson KC, et al. Women's Health Initiative Investigators. Estrogen plus progestin and the risk of coronary heart disease. N EnglJ Med 2003 ;349:523-534. 286. Shapiro S. Risks of estrogen plus progestin therapy: a sensitivity analysis of findings in the Women's Health Initiative randomized controlled trial. Climacteric 2003;6:302-310. 287. Machens K, Schmidt-Gollwitzer K. Issues to debate on the Women's Health Initiative (WHI) study. Hormone replacement therapy: an epidemiological dilemma? Human Reproduction 2003;18:1992-1999. 288. Howard BV, Kuller L, Langer R, et al. Risk of cardiovascular disease by hysterectomy status, with and without oophorectomy: the Women's Health Initiative Observational Study. Circulation 2005;111: 1462-1470. 289. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47-63. 289a. Grodstein F, Manson JE, Stampfer MJ. Hormone therapy and coronary heart disease: the role of time since menopause and age at hormone initiation.J Womens Health 2006;15:35-44. 290. Holm P, Andersen HL, Andersen MR, Erhardtsen E, Stender S. The direct antiatherogenic effect of estrogen is present, absent, or reversed, depending on the state of the arterial endothelium. A time course study in cholesterol-clamped rabbits. Circulation 1999;100:17271733. 291. Williams JK, Anthony MS, Honore EK, et al. Regression of atherosclerosis in female monkeys. Arterioscler Thromb Vasc Biol 1995;15: 827-836. 292. Adams MR, Register TC, Golden DL, Wagner JD, Williams JK. Medroxyprogesterone acetate antagonizes inhibitory effects of conjugated equine estrogens on coronary artery atherosclerosis. Arteriosder Thromb VascBio11997;17:217-221. 293. Clarkson TB, Anthony MS, Jerome CE Lack of effect of raloxifene on coronary artery atherosclerosis of postmenopausal monkeys. J Clin Endocrinol Metab 1998;83:721-726. 294. Clarkson TB, Anthony MS, Morgan TM. Inhibition of postmenopausal atherosclerosis progression: a comparison of the effects of conjugated equine estrogens and soy phytoestrogens. J Clin Endocrinol Metab 2001;86:41-47. 295. Mikkola TS, Clarkson TB. Estrogen replacement therapy, atherosclerosis, and vascular function. Cardiovasc Res 2002;53:605-619.
450 296. Herrington DM, Espeland MA, Crouse JR III, et al. Estrogen replacement and brachial artery flow-mediated vasodilation in older women. Arterioscler Thromb VascBio12001;21:1955-1961. 297. Stampfer M. Hormones and heart disease: do trials and observational studies address different questions? IntJEpidemio12004;33:454-455. 298. Grodstein F, Manson JE, Stampfer MJ. Postmenopausal hormone use and secondary prevention of coronary events in the Nurses' Health Study. A prospective, observational study. Ann Intern Med 2001;135:1-8. 299. Stampfer MJ. Postmenopausal hormones and CHD." update from the Nurses' Health Study. The 15th Annual meeting of the North American Menopause Society, Washington, DC, October 6-9, 2004. 300. Humphrey LL, Chan BK, Sox HC. Postmenopausal hormone replacement therapy and the primary prevention of cardiovascular disease. Ann Intern Med 2002;137:273-284. 301. Michels KB. Hormone replacement therapy in epidemiologic studies and randomized clinical trials-are we checkmate? Epidemiology 2003;14:3-5. 302. Scarabin PY, Oger E, Plu-Bureau G, Estrogen and Thrombo-Embolism Risk Study Group. Differential association of oral and transdermal oestrogen-replacement therapy with venous thromboembolism risk. Lancet 2003;362:428-432. 303. Kaplan JR, Manuck SB, Anthony MS, Clarkson TB. Premenopausal social status and hormone exposure predict postmenopausal atherosclerosis in female monkeys. Obstet Gyneco12002;99:381-388. 304. Clarkson TB, Appt SE. MPA and postmenopausal coronary artery atherosclerosis revisited. Steroids2003;68:941-951. 305. Nieto FJ. Cardiovascular disease and risk factor epidemiology: a look back at the epidemic of the 20th century. Am J Public Health 1999; 89:292-294. 306. Lefkowitz RJ, Willerson JT. Prospects for cardiovascular research. JAm MedAssoc 2001;285:581-587. 307. Magnus P, Beaglehole R. The real contribution of the major risk factors to the coronary epidemics: time to end the "only-50%" myth.Arch Intern Med 2001;161:2657-2660. 308. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002; 347:1557-1565. 309. Naghavi M, Libby P, Falk E, et al. From vulnerable plaque to vulnerable patient. A call for new definitions and risk assessment strategies: Part I. Circulation 2003;108:1664-1672. 310. Naghavi M, Libby P, Falk E, et al. From vulnerable plaque to vulnerable patient. A call for new definitions and risk assessment strategies: Part II. Circulation 2003;108:1772-1778. 311. Mosca L. Heart disease prevention in women. Circulation 2004;109: e158-e160. 312. Kwok Y, Kim C, Grady D, Segal M, Redberg R. Meta-analysis of exercise testing to detect coronary artery disease in women. Am J Cardio11999;83:660-666. 313. Alexander KP, Shaw LJ, Shaw LK, et al. Value of exercise treadmill testing in women.JAm Coll Cardio11998;32:1657-1664. 314. McTigue KM, Harris R, Hemphill B, et al. Screening and interventions for obesity in adults: summary of the evidence for the US Preventive Services Task Force. Ann Intern Med 2003;139:933-949. 315. Goldstein DJ. Beneficial health effects of modest weight loss. Int J Obes Relat Metab Disord 1992;16:397-415. 316. Garrow J. Christmas factor and snacking. Lancet 2000;355:8. 317. Karlsson J, Sjostrom L, Sullivan M. Swedish obese subjects (SOS)-an intervention study of obesity. Two-year follow-up of health-related quality of life (HRQ.L) and eating behavior after gastric surgery for severe obesity. IntJ Obes Relat Metab Disord 1998;22:113-126. 318. Karason K, Lindroos AK, Stenlof K, Sjostrom L. Relief of cardiorespiratory symptoms and increased physical activity after surgically induced weight loss: results from the Swedish Obese Subjects study. Arch Intern Med 2000;160:1797-1802.
GORODESKI AND GORODESKI
319. Skender ML, Goodrick GK, Del Junco DJ, et al. Comparison of 2-year weight loss trends in behavioral treatments of obesity: diet, exercise, and combination interventions. J Am Diet Assoc 1996;96: 342-346. 320. Hall JC, Watts JM, O'Brien PE, et al. Gastric surgery for morbid obesity. The Adelaide Study. Ann Surg 1990;211:419-427. 321. Ashley S, Bird DL, Sugden G, Royston CM. Vertical banded gastroplasty for the treatment of morbid obesity. BrJ Surg 1993;80:14211423. 322. Buchwald H, Avidor Y, Braunwald E, et al. Bariatric surgery: a systematic review and meta-analysis. J A m Med Assoc 2004;292:17241737. 322a. Howard BV, Manson JE, Stefanick ML, et al. Low-fat dietary pattern and weight change over 7 years: the Women's Health Initiative Dietary Modification Trial.JAm MedAssoc 2006;295:39-49. 322b. Howard BV, Van Horn L, Hsia J, et al. Low-fat dietary pattern and risk of cardiovascular disease: The Women's Health Initiative Randomized Controlled Dietary Modification Trial. J Am Med Assoc 2006;295:655-666. 323. Bouchard C. Physical activity and prevention of cardiovascular diseases: potential mechanisms. In: Leon AS, ed. Physical activity and cardiovascular health. Champaign, IL: Human Kinetics, 1997:48-56. 324. Press V, Freestone 1, George CE Physical activity: the evidence of benefit in the prevention of coronary heart disease. Q J Med 2003;96:245-251. 325. Morris JN, Everitt MG, Pollard R, Chave SPW, Semmence AM. Vigorous exercise in leisure-time: protection against coronary heart disease. Lancet 1980;2:1207-1210. 326. Lee IM, Hsieh CC, Paffenbarger RS. Exercise intensity and longevity in men: Harvard Alumni Health Study. J A m Med Assoc 1995;273: 1179-1184. 327. American College of Sports Medicine. Guidelinesfor exercise testing and prescription. Philadelphia, Lippincott, Williams & Wilkins, 2000. 328. Hagberg JM, Park JJ, Brown MD. The role of exercise training in the treatment of hypertension. Sports Med 2000;30:193-206. 329. Manson JE, Hu FB, Rich-Edwards JW, et al. A prospective study of walking as compared with vigorous exercise in the prevention of coronary heart disease in women. N EnglJMed 1999;341:650-658. 330. Kushi LH, Fee RM, Folsom AR, et al. Physical activity and mortality in postmenopausal women. JAm MedAssoc 1997;277:1287-1292. 331. Lee IM, Rexrode KM, Cook NR, Manson JE, Buring JE. Physical activity and coronary heart disease in women: is "no pain, no gain" passfi?JAm MedAssoc 2001;285:1447-1454. 332. Albert CM, Mittleman MA, Chae CU, et al. Triggering of sudden death from cardiac causes by vigorous exertion. N Engl J Med 2000;343:1355-1361. 333. Shaper AG, Wannamethee G, Walker M. Physical activity, hypertension and risk of heart attack in men without incidence of ischaemic heart disease. J Hum Hypertens 1994;8:3-10. 334. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N EnglJ Med 2002;346:393-403. 335. Lindstrom J, Louheranta A, Mannelin M, et al. Lifestyle intervention and 3-year results on diet and physical activity. Diabetes Care 2003;26:3230-3236. 336. Ridker PM, Cook NR, Lee IM, et al. A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women. N EnglJ Med 2005;352:1293-1304. 337. Eidelman RS, Hebert PR, Weisman SM, Hennekens CH. An update on aspirin in the primary prevention of cardiovascular disease. Arch Intern IVied2003;163:2006-2010. 338. Mosca L, Appel LJ, Benjamin EJ, et al. Evidence-based guidelines for cardiovascular disease prevention in women. Circulation2004;109:672693.
CHAPTER 31 Epidemiology and Risk Factors of Cardiovascular Disease in Postmenopausal W o m e n 339. Castelli WP, Garrison RJ, Wilson PWF, et al. Incidence of coronary heart disease and lipoprotein cholesterol levels. The Framingham Study. JAm MedAssoc 1986;256:2835-2838. 340. Castelli WP, Anderson K, Wilson PWF, Levy D. Lipids and risk of coronary heart disease: the Framingham Study. Ann Epidemiol 1992;2:23-28. 341. Bass KM, Newschaffer CJ, Klag MJ, Bush TL. Plasma lipoprotein levels as predictors of cardiovascular death in women. Arch Intern Med 1993;153:2209-2216. 342. Hokanson JE, Austin MA. Plasma triglyceride level is a risk factor for cardiovascular disease independent of high-density lipoprotein cholesterol level: a metaanalysis of population-based prospective studies. J Cardiovasc Risk 1996;3:213-219. 343. Shlipak MG, Simon JA, Vittinghoff E, et al. Estrogen and progestin, lipoprotein(a), and the risk of recurrent coronary heart disease events after menopause.JAm MedAssoc 2000;283:1845-1852. 344. Miettinen TA, Pyorala K, Olsson AG, et al. For the Scandinavian Simvastatin Study Group. Cholesterol-lowering therapy in women and elderly patients with myocardial infarction or angina pectoris: findings from the Scandinavian Simvastatin Survival Study (4S). Circulation 1997;96:4211-4218. 345. Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N EnglJ Med 1998;339:1349-1357. 346. Lewis SJ, Sacks FM, Mitchell JS, et al. For the CARE Investigators. Effect ofpravastatin on cardiovascular events in women after myocardial infarction: the cholesterol and recurrent events (CARE) trial. JAm Coil Cardio11998;32:140-146. 347. Clearfield M, Downs JR, Weis S, et al. Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS): efficacy and tolerability of long-term treatment with lovastatin in women. J Womens Health Gend Based Med 2001;10:971-981. 348. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:7-22. 349. LaRosa JC, He J, Vupputuri S. Effect of statins on risk of coronary disease: a meta-analysis of randomized controlled trials. J Am Med Assoc 1999;282:2340-2346. 350. Jick SS, Bradbury BD. Statins and newly diagnosed diabetes. BrJ Clin Pharmaco12004;58:303-309. 351. Pasternak RC, Smith SC Jr, Bairey-Merz CN, et al. ACC/AHA/ NHLBI clinical advisory on the use and safety of statins. J A m Coll Cardio12002; 40:567-5 72. 352. McKenney JM, Proctor JD, Harris S, Chinchili VM. A comparison of the efficacy and toxic effects of sustained Vs immediate-release niacin in hypercholesterolemic patients. JAm MedAssoc 1994;271:672-677. 353. Davignon J, Roederer G, Montigny M, et al. Comparative efficacy and safety of pravastatin, nicotinic acid and the two combined in patients with hypercholesterolemia. Am J Cardio11994;73:339-345. 354. Jacobson TA, Jokubaitis LA, Amorosa LE Fluvastatin and niacin in hypercholesterolemia: a preliminary report on gender differences in efficacy. Am J Med 1994;96:64S-68S. 355. Brown BG, Zhao XQ~ Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N EnglJ Med 2001;345:1583-1592. 356. Carlson LA, Rosenhamer G. Reduction of mortality in the Stockholm Ischaemic Heart Disease Secondary Prevention Study by combined treatment with clofibrate and nicotinic acid. Acta Med Scand 1998;223:405-418. 357. BIP Study Group. Secondary prevention by raising HDL cholesterol and reducing triglycerides in patients with coronary artery disease: the Bezafibrate Infarction Prevention (BIP) study. Circulation 2000;102: 21-27.
451
358. Diabetes Atherosclerosis Intervention Study Investigators. Effect of fenofibrate on progression of coronary-artery disease in type 2 diabetes: the Diabetes Atherosclerosis Intervention Study, a randomized study. Lancet 2001;357:905-910. 359. Meagher EA. Addressing cardiovascular disease in women: focus on Dyslipidemia. JAm Board Fam Pract 2004;17:424-437. 360. Brown BG, Zambon A, Poulin D, et al. Use of niacin, statins, and resins in patients with combined hyperlipidemia. Am J Cardiol 1998;81:52B-59B. 361. Pelton PD, Patel M, Demarest KT. Nuclear receptors as potential targets for modulating reverse cholesterol transport. Curr Top Med Chem 2005;5:265-282. 362. Bruemmer D, Blaschke F, Law RE. New targets for PPARgamma in the vessel wall: implications for restenosis. Int J Obes Relat Metab Disord 2005;29:$26-$30. 363. Gunzerath L, Faden V, Zakhari S, Warren K. National Institute on Alcohol Abuse and Alcoholism report on moderate drinking. Alcohol Clin Exp Res 2004;28:829-847. 364. Howard APt, Arnsten JH, Gourevitch MN. Effect of alcohol consumption on diabetes mellitus: a systematic review. Ann Intern Med 2004;140:211-219. 365. Kroenke CH, Chu N-F, Rifai N, et al. A cross-sectional study of alcohol consumption patterns and biologic markers of glycemic control among 459 women. Diabetes Care 2003;26:1971-1978. 366. Rimm EB, Chan J, Stampfler MJ, Colditz GA, Willett WC. Prospective study of cigarette smoking, alcohol use and the risk of diabetes in men. BMJ 1995;310:555-559. 367. Wannamethee SG, Camargo CA Jr, Manson JE, Willett WC, Rimm EB. Alcohol drinking patterns and risk of type 2 diabetes mellitus among younger women. Arch Intern Med 2003;163:1329-1336. 368. Davies MJ, Baer DJ, Judd JT, et al. Effects of moderate alcohol intake on fasting insulin and glucose concentrations and insulin sensitivity in postmenopausal women.JAm MedAssoc 2002;287:2559-2562. 369. Corrao G, Rubbiai L, Bagnardi V, Zambon A, Poikolainen K. Alcohol and coronary heart disease: a meta-analysis. Addiction 2000; 95:1505-1523. 370. Hines LM, Rimm EB. Moderate alcohol consumption and coronary heart disease: a review. PostgradMedJ 2001;77:747-752. 371. Rotondo S, DiCastelnuovo A, de Gaetano G. The relationship between wine consumption and cardiovascular risk: from epidemiological evidence to biological plausibility. Ital HeartJ 2001;2:1-8. 372. Murray R_P, Connett JE, Tyas SL, et al. Alcohol volume, drinking pattern, and cardiovascular disease morbidity and mortality: is there a U-shaped function? Am J Epidemio12002;155:242-248. 373. Rehm J, Sempos CT, Trevisan M. Average volume of alcohol consumption, patterns of drinking and risk of coronary heart disease a review. J Cardiovasc Risk 2003;10:15-20. 374. Sillanaukee P, Koivula T, Jokela H, Pitkajarvi T, Seppa K. Alcohol consumption and its relation to lipid-based cardiovascular risk factors among middle-aged women. Atherosclerosis 2000;152:503-510. 375. Baer DJ, Judd JT, Clevidence BA, et al. Moderate alcohol consumption lowers risk factors for cardiovascular disease in postmenopausal women fed a controlled diet. Am J Clin Nutr 2002;75:593-599. 376. Mukamal KJ, Conigrave KM, Mittleman MA, et al. Roles of drinking pattern and type of alcohol consumed in coronary heart disease in men. N EnglJ Med 2003;348:109-118. 377. De Oliveira e Silva ER, Foster D, McGee HM, et al. Alcohol consumption raises HDL cholesterol by increasing the transport rate of apolipoproteins AI and AII. Circulation 2000;102:2347-2352. 378. Serafini M, Laranjinha JA, Almeida LM, Maiani G. Inhibition of human LDL lipid peroxidation by phenol-rich beverages and their impact on plasma total antioxidant capacity in humans. J Nutr Biochem 2000;11:585-590.
452 379. Hu FB, Stampfer MJ, Manson JE, et al. Trends in the incidence of coronary heart disease and changes in diet and lifestyle in women. N EnglJ Med 2000;343:530-537. 380. Wassertheil-Smoller S, Psaty B, Greenland P, et al. Association between cardiovascular outcomes and antihypertensive drug treatment in older women. J A m Med~tssoc 2004;292:2849-2859. 381. Olshansky SJ, Passaro DJ, Hershow RC, et al. A potential decline in life expectancy in the United States in the 21st century. N E n g l J M e d 2005;352:1138-1145. 382. Cheng AW, Miller ML, Morris M, et al. A stochastic model of the long range financial status of the OASDI program. Actuarial study no. 117. Baltimore: Social Security Administration, publication no. 11-11543, 2004.
GORODESKI AND GORODESKI
383. Olshansky SJ, Hayflick L, Carnes BA. Position statement on human aging. J Gerontol Biol Sci 2002;57A:B292-B297. 384. Allard M, Lebre V, Robine JM, Calment J. Jeanne Calment.'from van Gogh's time to ours; I22 extraordinary years. New York: W H Freeman, 1998. 385. Tobert JA. Lovastatin and beyond: the history of the HMG-CoA reductase inhibitors. Nat Rev Drug Discov 2003;2:517-526.
2 H A P T E R 3~
Mechanisms of Action of Estrogen on the Cardiovascular System RICHARD H.
KARAS
Molecular Cardiology Research Center, Tufts-New England Medical Center, Tufts University School of Medicine, Boston, MA 02111
I. I N T R O D U C T I O N There is abundant evidence that estrogen regulates important aspects of cardiovascular function and affects the risk of developing cardiovascular disease (CVD). Populationbased studies have demonstrated differences between women and men in the pattern of onset of CVD, particularly as it relates to hormonal status and age-specific rates of disease, strongly supporting that endogenously produced estrogens reduce the risk of CVD (1,2). Based on such observations, a hypothesis emerged that postmenopausal treatment with exogenous estrogen will also reduce CVD risk. This hypothesis was first tested in observational and epidemiologic studies that in aggregate supported the idea that postmenopausal estrogen treatment did indeed protect against CVD (3). This then led to a number of randomized controlled trials of postmenopausal hormone therapies, largely with estrogen and a progestin combined, that did not demonstrate the expected benefits (4,5), as is discussed in more detail in Chapter 39. Although the discordant results between the observational studies and randomized controlled trials have led to considerable controversy in this field (6-9), several points are clear. First, it is quite clear that the original hypothesis, that postmenopausal hormone therapy will exert broad and extensive cardioprotective effects, in a manner analogous to the cholesterol-lowering statins, is not true. A second aspect supported by these studies is that a variety of factors, including patient-specific factors such as age and the underlying status of the cardiovascular system, TREATMENT OF THE POSTMENOPAUSAL WOMAN
453
as well as agent-specific factors such as what actual compound, alone or in combination, is being given and at what dose and by what mode of administration, all have an important impact on the cardiovascular effects of hormone therapy. Finally, a third point that is clear is that the cardiovascular effects of estrogen are complex and include a mixture of both potentially beneficial and potentially harmful effects. Thus, it is imperative that we learn more about the mechanisms of action of estrogen on the cardiovascular system so that we can learn to better use postmenopausal hormone therapy in an appropriate manner to reduce CVD risk when possible. CVD remains the leading cause of death for both women and men in the United States. The majority of the morbidity and mortality associated with CVD results from either diseases of the heart itself, including arrhythmias, cardiomyopathic disorders, and valvular diseases, or from diseases of the vasculature, of which atherosclerosis and the myocardial infarctions and strokes that it causes are the most prominent and prevalent. Venous diseases, particularly venous thrombosis and thromboembolism, are less prevalent than atherosclerosis, but their consideration is also relevant to hormone therapy. Currently, we have considerably more information regarding the effects of estrogens on vascular diseases and atherosclerosis than we do regarding the effects of estrogen on the heart itself. Atherosclerosis is a complicated disease, in part because it results from an interplay of many factors, including inflammation, thrombosis, hemodynamics, and oxidative state, all unfolding on a background of underlying Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
454 variability in genetic susceptibility to this disease. As a result, in the sections that follow, I will review the effects of estrogen on each of these etiologic aspects of the atherosclerotic process.
II. SYSTEMIC AND DIRECT EFFECTS OF ESTROGEN ON THE CARD I OVAS CULAR SYSTEM The mechanisms of action of estrogen on the cardiovascular system can be categorized in a variety of ways, but one useful method is to consider effects on systemic or circulating factors separately from direct effects on the heart and vascular beds (reviewed in [10]). The fact that estrogen regulates a variety of systemic factors relevant to CVD has been known for many years. The ability of estrogen to directly regulate the structure and function of cardiovascular cells and tissues has been appreciated only for about the past 10 to 15 years. Each of these aspects will now be discussed in turn, beginning with the direct effects of estrogen on the cardiovascular system.
A. Direct Cardiovascular Effects of Estrogen The predominant cells in the cardiovascular system are cardiomyocytes, vascular smooth muscle cells (VSMCs), and vascular endothelial cells (ECs), and all have been shown to express estrogen receptors (reviewed in [11]). There are extensive data from cell culture models that estrogen regulates critically important processes in each of these cell types. For example, estrogen inhibits VSMC proliferation and migration. Interestingly, estrogen has the opposite effect on ECs in that it promotes proliferation and migration in these cells. This is yet another important example of the ability of estrogen to exert opposite effects depending on the specific cell type examined. Estrogen has also been shown to regulate apoptosis in vascular (12) and cardiac cells (13). Animal models have also contributed extensively to our understanding of the direct cardiovascular effects of estrogen. Indeed, animal models ranging from mice to monkeys, including hyperlipidemic models, vascular injury models, and genetically altered animals, have been used to demonstrate the potential for atheroprotective effects of estrogen (see Chapter 38 or [14]). Animal models have also contributed importantly to the hypothesis that the timing of initiation of hormone therapy may have a significant impact on its potential for cardiovascular benefit, or lack thereof, and this is discussed in detail in Chapter 38. These studies have also contributed to our understanding of the multiple mechanisms by which estrogen regulates the cardiovascular system. For example, as discussed in more detail later, estrogen
RICHARD H. KARAS
has been shown to accelerate re-endothelialization following direct vascular injury in the mouse (15), in part by increasing the contribution of circulating progenitor cells (16), and this process limits the response to injury and subsequent VSMC hyperproliferation. Interestingly, atheroprotective effects of estrogen are no longer evident in mice devoid of monocyte chemoattractant protein-1 (MCP-1), suggesting an important link to the inflammatory system. The importance of estrogen-mediated effects on the inflammatory status of the vessel itself is also supported by studies demonstrating that in women, estrogen reduces levels of both vascular cell adhesion molecule (VCAM) and inflammatory cell adhesion molecule (ICAM) (16). These proteins are expressed on the surface of ECs and play an important role in allowing circulating inflammatory cells to adhere to the luminal surface of the artery, a prerequisite for their subsequent vessel wall invasion. Estrogen also upregulares the expression of matrix metalloproteinases (MMPs), including specifically MMP9 (17). MMPs are proteins that degrade extracellular matrix and thus have an important role in vessel wall remodeling. Estrogen's effects on MMP9 are also interesting because of their potential importance for the differential effects of early versus late administration of hormone therapy to postmenopausal women, discussed in Chapter 38. In women with relatively normal arteries, estrogeninduced increases in MMP9 expression could be beneficial in that this may enhance "positive remodeling" of the artery, a process by which the vessel expands in diameter and preserves lumen size. In contrast, in women with more advanced atherosclerotic plaques, estrogen-induced increases in MMP9 expression might contribute to plaque erosion and thus subsequent acute coronary syndromes. In addition to these effects on vascular structure, estrogen also alters vascular function in important ways. The beststudied example of this is the acute vasodilatory response to estrogen administration. In both monkeys and humans, a single dose of estrogen induces relaxation of the coronary arteries within a few minutes of exposure (18,19). This effect has been studied in detail because it is interesting for many reasons. First, the acute vasodilatory effects of estrogen result from estrogen-induced increases in nitric oxide bioavailability in the artery, and nitric oxide is one of the most potent anti-atherogenic, endogenously produced compounds known. In addition, other interventions such as statins, that also increase arterial vasodilation via this pathway, have been shown to decrease CV-D risk. Interestingly, the ability of estrogen to induce vasodilation is impaired in individuals with diabetes mellitus or with other risk factors for coronary disease or established atherosclerosis (20), lending further support to the hypothesis that the vascular effects of estrogen depend in part on the state of health of the vasculature at the time it is being administered. The signaling mechanisms that mediate the acute vasodilatory effects of estrogen are discussed in the following section.
CHAPTER 32 Mechanisms of Action of Estrogen on the Cardiovascular System 1. ESTROGEN-MEDIATED SIGNALING IN THE CARDIOVASCULAR SYSTEM
In general, estrogen regulates cell function by activating estrogen receptors (ERs), of which two are known, ER-oL and ER-[3. Estrogen receptors are expressed in the heart and the vasculature in cardiomyocytes, ECs, and VSMCs, strongly supporting the idea that the cardiovascular system is an end organ target for estrogen action (reviewed in [10]). Both ER-e~ and ER-[~ have been implicated in mediating processes important for cardiovascular function. For example, estrogen-mediated reductions in the response to vascular injury are not observed in mice devoid of ER-ot (21), but they are in wild-type mice. A role for ER-[~ in vascular function is supported by the observation that mice devoid of ER-[3 have abnormal regulation of vasomotor tone and are hypertensive (22). Classically, ERs are thought to transduce estrogen's effects by acting as ligand-activated transcription factors that regulate gene expression in response to hormone binding. These effects are often referred to as "genomic" effects of estrogen as they require alterations in gene expression. It is now clear, however, that estrogen-dependent signaling and ER function are much more complex than this (Fig. 32.1) (see color insert). For example, as shown in Fig. 32.1, point 1 (see color insert), ERs interact with a variety of other proteins that exert important effects on the manner in which the ligand-bound receptor alters gene expression (23). In addition, it has also been shown that vascular ERs can be transcriptionally activated in the absence of hormone binding by direct phosphorylation of the receptor by a variety of kinases (see Fig. 32.1, point 3) (see color insert), a process referred to as ligandindependent activation of the receptor (24). This adds another
455
layer of complexity to the genomic effects of these receptors as there can be cross-talk between the ligand-dependent and ligand-independent activation pathways. Further complexity in ER-dependent signaling results from the fact that ERs also regulate their own transcription, as well as that of other steroid hormone receptors and other nonsteroid hormone receptor transcription factors (see Fig. 32.1, points 4 and 6) (see color insert). Finally, there are genetic polymorphisms in the ER genes that may also alter receptor-mediated downstream responses (see Fig. 32.1, point 5) (see color insert). As noted earlier, estrogen can also exert vasodilatory and other vascular cell effects within a matter of minutes after exposure. These effects occur too rapidly to be mediated by changes in gene expression, and they continue to occur in cells treated with inhibitors of gene transcription. Thus, these rapid effects are often referred to as "nongenomic" signaling pathways (reviewed in [25]). More recently discovered than the genomic pathways, the best-studied nongenomic pathway in the vasculature involves the rapid activation of the enzyme that produces nitric oxide in endothelial cells, eNOS (reviewed in [25]). Interestingly, these nongenomic effects of estrogen are also mediated by the same receptors, ER-ot and ER-[3, as are the genomic effects. However, the receptors that mediate nongenomic eNOS activation are localized to the cell membrane in specialized signaling modules called caveolae, rather than in the nucleus, and it is in the caveolae that the receptors initiate these signaling cascades by activating a variety of specific kinases (see Fig. 32.1, point 2) (see color insert). The cell membrane-associated ERs form signaling complexes with other proteins by binding to a scaffold protein called striatin, which allows the receptor, by still currently unknown mechanisms, to activate specific kinases that FIOURE 32.1 Molecularmechanisms that mediate cardiovasculareffects of estrogen and estrogen receptors (ERs). ERs regulate cardiovascularcell function by regulatinggene expression.They do so by interacting with a variety of proteins important for these transcriptional regulatorypathways (point 1). The transcriptional activityof the receptor is also regulated by direct phosphorylation(point 3). ERs also regulate the expression of other transcription factors, including other sex steroid hormone receptors and non-sex hormone receptor transcription factors (points 4 and 6). Genetic polymorphismsin the receptors also contribute to interindividual responses to hormone treatment (point 5). Finally,in addition to these regulatorypathways that are all related to genomic effects, cell membrane-associated ERs can also regulate cellular functions by nongenomic pathways that involve the rapid activation of various kinase cascades (point 2). CoA, coactivator; CoR, corepressor; GFR, growth factor receptor; NHR, non-SSHR nuclear receptors; SSHR, sex steroid hormone receptor; SSHRE, SSH response element; TFs, transcription factors. From ref. 11, with permission.
456
RICHARD H. Ka~s
in turn phosphorylate eNOS and activate its enzymatic activity (26). The potential to ultimately identify novel pharmacologic compounds that can selectively activate these nongenomic vasodilatory pathways, without activating the classical genomic signaling pathways, holds great promise as novel therapeutic strategies for diseases such as atherosclerosis, hypertension, and heart failure.
B. Systemic Effects of Estrogen on the Cardiovascular System As noted earlier, in addition to the direct effects of estrogen on the cardiovascular system, estrogen regulates a variety of systemic or circulating factors that can also affect the cardiovascular system. These systemic effects include those on circulating factors including lipids, coagulation and fibrinolytic factors, inflammatory factors, vasoactive hormones, oxidative state, and metabolic state. These will now be reviewed.
1. EFFECTS OF ESTROGEN ON CIRCULATING LIPIDS
The lipid effects of estrogen are a good example of the complex mixture of both potentially beneficial and potentially harmful effects that estrogen can exert. Oral estrogen has long been known to lower circulating levels of lowdensity lipoprotein (LDL) cholesterol and to increase levels of high-density lipoprotein (HDL) cholesterol (27). Both effects of course would be expected to reduce CVD risk. Similarly, oral estrogen also lowers the atherogenic Lp(a) particle (28). In contrast, however, oral estrogen can raise triglycerides, a potentially harmful effect. It is important to note that these effects on circulating lipids result primarily from changes in hepatic production of lipid particles, and thus they occur with oral estrogen due to a first-pass hepatic effect, but not with other forms of administration such as transdermal patches. In the randomized, controlled clinical trials mentioned previously, the expected changes in lipid profiles were indeed observed, but in the populations studied, these lipid changes were not accompanied by the anticipated decrease in CVD risk (4,5,29). The reasons for this discordance are unclear. One suggestion is that the changes in LDL and H D L cholesterol, for example, do not occur in the specific subgroups of cholesterol particles that would be associated with reductions in CVD risk (30), but this hypothesis requires further testing.
2. EFFECTS OF ESTROGEN ON CIRCULATING COAGULATION AND FIBRINOLYTIC FACTORS
The effects of estrogen on coagulation and fibrinolytic factors are also largely the result of first-pass hepatic effects, and thus they occur primarily with oral administration of hormone (reviewed in [10]). Overall, hormone therapy is
generally considered to create a mildly hypercoagulable state. However, as one might expect, it also presents complexities that result from a mixture of opposing effects. First, although some procoagulant and antifibrinolytic factors do increase in abundance with estrogen therapy, in some cases these effects are offset by opposing effects on anticoagulant or profibrinolytic proteins. For example, while plasminogen activator inhibitor I (PAI-I) levels are decreased by estrogen, so are levels of the opposing tissue plasminogen activator (tPA) (31,32). Recent clinical trial data have supported the idea that the overall effect of hormone therapy is to produce a tendency toward coagulation as the risks of venous thrombosis and venous thromboembolism are increased (4,33). However, these studies used combination therapy including both estrogen and a progestin. Interestingly, the very recently reported results from the estrogen-alone arm of the Women's Health Initiative (WHI) demonstrated considerably and significantly smaller increases in risk for venous thrombosis and thromboembolism than that observed in the combined hormone therapy arm of the W H I (34), suggesting that part of the increased risk attributed to hormone therapy may be due to the progestin component. It is also important to differentiate effects of estrogen on coagulation in the venous circulation versus those in the arterial circulation. The thrombotic events of clinical relevance on the venous side are primarily due to activation of the coagulation cascade, and they are treated with anticoagulants such as coumadin. In contrast, however, the clinically relevant thrombotic events in the arterial vasculature are primarily driven by platelet activation, and thus they are primarily treated with antiplatelet agents such as aspirin. Little is known regarding the effects of estrogen on platelet function, but the existing data suggest that estrogen may inhibit platelet activation (35). Thus, care must be used when interpreting the effects of estrogen on the coagulation and fibrinolytic systems in terms of the impact this has on arterial atherothrombotic events. 3. EFFECTS OF ESTROGEN ON CIRCULATING INFLAMMATORY MARKERS Over the past several years, our understanding of the importance of inflammation in the pathophysiology of CVD has increased tremendously. Thus, there is interest in the effects of estrogen on the relevant aspects of inflammation as well. Much of the attention in this field has focused on the potential role of increases in circulating levels of C-reactive protein (CRP). This results in part because CRP is the inflammatory marker that has the most demonstrated relevance to clinical medicine (36). Although there is considerable evidence that CRP levels are predictive, and independently so, of CVD risk (37), it remains debated whether this is because CRP is a marker for other etiologic factors or because it directly contributes to the atherosclerotic process. In either
CHAPTER 32 Mechanisms of Action of Estrogen on the Cardiovascular System case, evidence suggests that oral estrogen increases circulating levels of CRP, whereas transdermal estrogen does not (38), again highlighting the potential importance of route of administration on the effects of estrogen. It is interesting to note, however, that an analysis of the observational Cardiovascular Health Study showed that regardless of levels of CRP, women who were taking hormone therapy had a lower risk of CVD than those who were not taking it (39). Because of the focus on CRP, many people refer to estrogen as "pro-inflammatory." However, the situation is not quite that clear, and this term is likely not very useful. For example, many other circulating levels of inflammatory markers are reduced by estrogen. This includes interleukin (IL)-6, an important regulator of hepatic CRP production (16). In addition, IL-1 and MCP-1 levels also are reduced by estrogen (16). An additional example of the difficulty of labels such as "pro-inflammatory" can be found by looking at the effects of estrogen on cyclooxygenase 2 (Cox2), an enzyme important for production of specific prostaglandins. In animal models, estrogen increases Cox2 levels, and the atheroprotective effects of estrogen in a hyperlipidemic mouse model are lost in mice devoid of Cox2, supporting the idea that activation of Cox2 plays an important role in the atheroprotective effects of estrogen (40). Thus, it is important to consider the complexity of effects of estrogen on the inflammatory system and to integrate the effects of estrogen on specific components of this system when assessing the contribution of inflammatory system changes to the effects of estrogen on the cardiovascular system.
4. EFFECTS OF ESTROGEN ON CIRCULATING VASOACTIVE HORMONES
As discussed further in a later section, the state of contraction or relaxation of the arterial system has important consequences for CVD risk. Although it is an admitted oversimplification, it is useful to realize that, in a broad sense, factors that promote vasoconstriction are in general associated with increased CVD risk, whereas factors that promote vasorelaxation are associate with reduced CVD risk. Overall, estrogen appears to reduce levels of vasoconstrictors and increase circulating levels ofvasodilators (reviewed in [10]). For example, estrogen has been shown to decrease circulating levels of the vasoconstrictors angiotensin II, renin, and endothelin; on the other hand, estrogen has been shown to increase levels of the vasodilator prostacyclin (produced by Cox2) and to increase the bioavailability of nitric oxide, a centrally important vasodilator discussed later.
5. EFFECTS OF ESTROGEN ON OXIDATIVE STATE
Increased levels of oxidation have been associated with a pro-atherosclerotic effect and an increase in CVD risk. Increased levels of oxidized LDL cholesterol, a particularly
457
atherogenic form of LDL, likely contributes to this. In addition, a large body of data demonstrates that alterations in the overall oxidation/reduction state of ceils have very important influences on cell signaling and cell function. Certain forms of estrogen have been shown to exert antioxidant effects using in vitro assays (41). The extent to which antioxidant effects of estrogen contribute to effects on the cardiovascular system are currently unclear.
6. EFFECTS OF ESTROGEN ON METABOLIC STATE
The importance of a constellation of metabolic changes in contributing to CVD risk is increasingly being appreciated and is currently an area of intensive research. The "metabolic syndrome," consisting of central adiposity, hypertension, a procoagulant state, and glucose intolerance, is emerging as a major contributor to CVD risk, as, of course, is diabetes mellitus. Previous studies have suggested that estrogen can favorably affect glucose metabolism by enhancing insulin sensitivity (42), and these findings have been confirmed and extended by recent analyses reporting a reduction in self-reported incidence of diabetes mellitus in recent combined hormone therapy trials (43,44). This is an intriguing finding, but, as is the case for lipids, it remains unclear why a reduction in the risk of diabetes observed in this study was not accompanied by the expected reduction in cardiovascular events. Recent data from both human and animal studies also suggest that estrogen has favorable effects on body weight (45,46). 7. EFFECTS OF ESTROGEN ON PROGENITOR CELLS
Another rapidly emerging and evolving area of research in CVD has identified circulating progenitor cells, particularly EC progenitor cells, as playing a potentially important role in protecting against atherosclerosis and in recovery from cardiovascular insults or injury after they have occurred. There is increasing evidence, primarily from animal studies, that estrogen increases the quantity of circulating progenitor cells by increasing their mobilization from the bone marrow and that estrogen also enhances the ability of these cells to repair vascular injury (47). This is a very new area of inquiry that will bear close follow-up as the field moves forward.
III. SUMMARY There is abundant evidence that estrogen regulates many processes central to the physiology and pathophysiology of the cardiovascular system. To date, clinical trials have provided confusing and conflicting data as to how postmenopausal hormone therapy alters cardiovascular disease risk. It is clear from these studies and others that the cardiovascular
458
RICHARD H. KARAS
FIGURE 32.2 The importance of timing of initiation of hormone therapy on its cardiovascular effects. The "timing hypothesis" states that the effects of hormone therapy on cardiovascular disease differs between early and later stages of atherosclerosis. Atherosclerosis is characterized by the gradual loss of vascular protective mechanisms and the emergence of advanced, unstable lesions. Estrogen's effects on vascular cells differ depending on the state or progression of atherosclerosis in the exposed vessels. CAMs, cell adhesion molecules; Cox2, cyclooxygenase 2; LDL, low-density lipoprotein; MCP-1, monocyte chemoattractant protein 1; MMR matrix metalloproteinase; TNF-ot, tumor necrosis factor-el; VSMC, vascular smooth muscle cell. From ref. 11, with permission.
effects of estrogen are complex and include both potentially beneficial and potentially harmful effects (Fig. 32.2) (see color insert). Furthermore, increasing evidence supports the idea that one of the factors having an important impact on the relative balance of beneficial versus harmful cardiovascular effects of estrogen is the timing of initiation of treatment as it relates to the extent of underlying atherosclerosis, as depicted in Fig. 32.2 (see color insert). Estrogen regulates both systemic (or circulating) factors relevant to cardiovascular function, but it also has direct effects on the heart and vasculature. The systemic effects include effects on circulating lipids, inflammatory markers, coagulation/fibrinolysis, oxidative state, vasoactive hormones, metabolic state, and progenitor cells. The direct effects of estrogen on cardiovascular cells include effects on cell growth, migration, and survival. The molecular mechanisms that mediate the cardiovascular effects of estrogen include genomic effects resulting from changes in gene expression, as well as nongenomic effects activating specific signaling pathways within cells. It is hoped that through a better understanding of the mechanisms that mediate estrogen's actions on the cardiovascular system, we can develop novel therapeutic strategies that manipulate this pathways in such a way as to allow us to reduce the risk of this leading cause of death in women.
References 1. Barrett-Connor E. Sex differences in coronary heart disease. Why are women so superior? The 1995 Ancel Keys Lecture. Circulation 1997;95:252-264. 2. Barrett-Conor E, Bush TL. Estrogen and coronary heart disease in women.JAm Meddssoc 1991; 265:1861-1867. 3. Grady D, Rubin SM, Petitti DB, et al. Hormone therapy to prevent disease and prolong life in postmenopausal women. Ann Intern Meat 1992;117:1016-1037. 4. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAm MedAssoc 1998;280:605-613. 5. Manson JE, Hsia J, Johnson KC, et al. Estrogen plus progestin and the risk of coronary heart disease. N EnglJ Med 2003;349:523-534. 6. Clarkson TB. The new conundrum: do estrogens have any cardiovascular benefits? IntJ Fertil WomensMed 2002;47:61-68. 7. Grodstein F, Clarkson TB, Manson JE. Understanding the divergent data on postmenopausal hormone therapy. N EnglJ Med 2003;348: 645-650. 8. Karas RH, Clarkson TB. Considerations in interpreting the cardiovascular effects of hormone replacement therapy observed in the WHI: timing is everything. MenopausalMed 2003;10:8-12. 9. Stampfer MJ, Colditz GA, Willett WC, et al. Postmenopausal estrogen therapy and cardiovascular disease. N Engl J Med 1991;325: 756-762. 10. Mendelsohn ME, Karas RH. Mechanisms of disease: the protective effects of estrogen on the cardiovascular system. N Engl J Med 1999;340:1801-1811. 11. Mendelsohn ME, Karas RH. Molecular and cellular basis of cardiovascular gender differences. Science2005;308:1.
CHAPTER 32 Mechanisms of Action of Estrogen on the Cardiovascular System 12. Spyridopoulos I, Sullivan AB, Kearney M, Isner JM, Losordo DW. Estrogen-receptor-mediated inhibition of human endothelial cell apoptosis-estradiol as a survival factor. Circulation 1997;95:1505-1514. 13. Van Eickels M, Patten RD, Aronovitz M, et al. 17 Beta-estradiol decreases infarct size but increases cardiac remodeling and mortality in mice with myocardial infarction. JAm Coll Cardio12003;41:2084-2094. 14. Karas RH. Animal models of the cardiovascular effects of exogenous hormones. Am J Cardio12002;90:22F-25F. 15. Krasinski K, Spyridopoulos I, Asahara T, et al. Estradiol accelerates functional endothelial recovery after arterial injury. Circulation 1997; 95:1768-1772. 16. Koh KK, Shin M-S, Sakuma I, et al. Effects of conventional or lower doses of hormone replacement therapy in postmenopausal women. Artheroscler Thromb VascBio12004;24:1516-1521. 17. Zanger D, Yang B K, Ardans J, et al. Divergent effects of hormone therapy on serum markers of inflammation in postmenopausal women with coronary artery disease on appropriate medical management.JAm Coll Cardio12000;36:1797-1802. 18. Reis SE, Gloth ST, Blumenthal RS, et al. Ethinyl estradiol acutely attenuates abnormal coronary vasomotor responses to acetylcholine in postmenopausal women. Circulation 1994;89:52-60. 19. Williams JK, Adams MR, Herrington DM, Clarkson TB. Short-term administration of estrogen and vascular responses of atherosclerotic coronary arteries.JAm Coll Cardio11992;20:452-457. 20. Herrington DM, Espleland MA, Crouse JR III, et al. Estrogen replacement and brachial artery flow-mediated vasodilation in older women. Arterioscler Thromb VascBid 2001;21:1955-1961. 21. Pare G, Krust A, Karas RH, et al. Estrogen receptor-alpha mediates the protective effects of estrogen against vascular injury. Circ Res 2002;90:1087-1092. 22. Zhu Y, Bian Z, Lu P, et al. Abnormal vascular function and hypertension in mice deficient in estrogen receptor-b. Science 2002;295: 505-508. 23. Tsai MJ, O'Malley BW. Molecular mechanisms of action of steroid/ thyroid receptor superfamily members. Annu Rev Biochem 1994: 451-486. 24. Karas RH, Gauer EA, Bieber HE, Baur WE, Mendelsohn ME. Growth factor activation of the estrogen receptor in vascular cells occurs via a MAP kinase-independent pathway. J Clin Invest 1998;101: 2851-2861. 25. Mendelsohn ME. Genomic and nongenomic effects of estrogen in the vasculture. Am J Cardio12002;90:3 F-6 E 26. Lu Q~ Pallas DC, Surks HK, et al. Striatin assembles a membrane signaling complex necessary for rapid, nongenomic activation of endothelial N O synthase by estrogen receptor-alpha. PNAS 2004; 101:1712617131. 27. Writing Group for the PEPI Trial. Effects of estrogen or estrogen/ progestin regimens on heart disease risk factors in postmenopausal women: The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial.JAm MedAssoc 1995;273:199-208. 28. Shlipak MG, Simon JA, Vittinghoff E, et al. Estrogen and progestin, lipoprotein(a), and the risk of recurrent coronary heart disease events after menopause.JAm MedAssoc 2000;283:1845-1852. 29. Shlipak MG, Chaput LA, Vittinghoff E, et al. Lipid changes on hormone therapy and coronary heart disease events in the Heart and Estrogen/Progestin Replacement Study (HERS). Am Heart J 2003; 146:870-875.
459
30. Wakatsuki A, Ikenoue N, Izumiya C, Okatani Y, Sagara Y. Effect of estrogen and simvastatin on low-density lipoprotein subclasses in hypercholesterolemic postmenopausal women. Obstet Gynecol 1998; 92:367-372. 31. Koh KK, Mincemoyer R, Bui MN, et al. Effects of hormone-replacement therapy on fibrinolysis in postmenopausal women. N EnglJ Med 1997;336:683-690. 32. Gebara OC, Mittleman MA, Sutherland P, et al. Association between increased estrogen status and increased fibrinolytic potential in the Framingham Offspring Study. Circulation 1995;91:1952-1958. 33. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women.JAm MedAssoc 2002;288:321-333. 34. Curb JD, Prentiss RL, Bray PF, et al. Venous thrombosis and conjugated equine estrogen in women without a uterus. Arch Intern Med 2006;166:772-780. 35. Bar J, Tepper R, Fuchs J, et al. The effect of estrogen replacement therapy on platelet aggregation and adenosine triphosphate release in postmenopausal women. Obstet Gyneco11993;81:261-264. 36. Verma S, Szmitko PE, Ridker PM. C-reactive protein comes of age. Nat C/in Pract Cardiovasc Med 2005;2:29-36. 37. Ridker PM, Hennekens C, Buring J, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N EnglJ Med 2000;342:836-843. 38. Decensi A, Omodei U, Robertson C, et al. Effect of transdermal estradiol and oral conjugated estrogen of C-reactive protein in retinoidplacebo trial in healthy women. Circulation 2003;106:1224-1228. 39. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N E n g l J M e d 2002;347:15571565. 40. Egan KM, Lawson JA, Fries S, et al. COX-2-derived prostacyclin confers atheroprotection on female mice. Science 2004;306:1954-1957. 41. Shwaery GT, Vita JA, Keaney JF Jr. Antioxidant protection of LDL by physiological concentrations of 1713-estradiol-requirement for estradiol modification. Circulation 1997;95:1378-1385. 42. Nabulsi AA, Folsom AR, White A, et al. Association of hormonereplacement therapy with various cardiovascular risk factors in postmenopausal women. N EnglJ Med 1993;328:1069-1075. 43. Kanaya AM, Herrington D, Vittinghoff E, et al. Glycemic effects of postmenopausal hormone therapy: the Heart and Estrogen/Progestin Replacement Study. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 2003;138:1-9. 44. Margolis KL, Bonds DE, Rodabough RJ, et al. Effect ofoestrogen plus progestin on the incidence of diabetes in postmenopausal women: results from the Women's Health Initiative Hormone Trial. Diabetologia 2004;47:1175-1187. 45. Simpson ER, McInnes KJ. Sex and fat--can one factor handle both? Cell Metab 2005;2:346-347. 46. D'Eon TM, Souza SC, Aronovitz M, et al. Estrogen regulation of adiposity and fuel partitioning. Evidence of genomic and non-genomic regulation of lipogenic and oxidative. J Biol Chem 2005;280:3598335991. 47. Strehlow K, Werner N, Bohm M, et al. Estrogen increases bone marrow-derived endothelial progenitor cell production and diminishes neointima formation. Circulation 2003;107:3059-3065.
This Page Intentionally Left Blank
Llplds and Llpoprotelns and Effects of Hormone Therapy RONALD M. KRAUSS Children'sHospital & Research Center Oakland, Oakland, CA 94609
I. MAJOR LIPOPROTEINS AND METABOLIC PATHWAYS
Levels of plasma lipids and lipoproteins are strong predictors for the development of atherosclerotic cardiovascular disease in postmenopausal women (1). In women, as in men, numerous factors contribute to variations in plasma lipoproteins that may affect cardiovascular disease risk. These include age, dietary components, adiposity, genetic traits, and hormonal changes. Each of these factors may operate to varying degrees in determining changes in plasma lipoprotein profiles accompanying menopause. Cross-sectional (2,3) and longitudinal studies (3-5) have suggested increases in levels of cholesterol, low-density lipoproteins (LDLs), and triglyceride-rich lipoproteins associated with menopause. Levels of high-density lipoproteins (HDLs), which are higher in women than men and are thought to contribute to relative protection of premenopausal women from cardiovascular disease, remain relatively constant in the years following menopause (6), although small and perhaps transient reductions in the HDL-2 subfraction have been reported in relation to reduced estradiol level following menopause (4,5). Despite these associations, it has been difficult to determine the role of endogenous hormones in influencing the plasma lipoproteins of postmenopausal women. In principle, the effects of hormone replacement should act to reverse any alterations in lipoprotein metabolism that are due to postmenopausal hormone changes. However, although there may be beneficial effects on lipoproteins, hormone treatment does not restore a premenopausal lipoprotein profile. Furthermore, there is now compelling evidence that although premenopausal hormonal status contributes to protection from cardiovascular disease, exogenous hormoneinduced lipoprotein changes do not result in reduced incidence of cardiovascular disease, and can in fact increase disease risk. T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
A. A t h e r o g e n i c Lipoproteins Endogenously synthesized triglycerides are secreted, along with cholesterol and phospholipids, in the form of very-lowdensity lipoproteins (VLDLs, density < 1.006 g/ml). VLDLs comprise an array of particles, each of which contains one molecule of a high-molecular-weight structural protein, apoB100, and varying amounts of lipids and smaller proteins, apoEs and apoCs. The triglycerides in VLDLs are hydrolyzed in peripheral tissues by the action of lipoprotein lipase, releasing VLDL surface lipids, apoEs and apoCs, which may be transferred to HDLs. Triglyceride-depleted VLDL remnants are returned to the liver and taken up by receptor-mediated processes, which are incompletely understood. There is evidence that LDL receptors, LDL receptorlike proteins, and perhaps other uptake mechanisms are involved (Fig. 33.1). Similar pathways operate in the metabolism of exogenously derived fat, which is transported in plasma in intestinally derived chylomicron particles (not shown in Fig. 33.1). Chylomicrons have a structure similar to that of VLDLs but are generally much larger and contain a smaller molecular weight form of apoB, designated apoB48. Larger VLDL and chylomicron remnants are generally cleared from plasma over the course of several hours. However, smaller VLDL particles may be metabolized more slowly and accumulate in plasma as cholesteryl ester-rich remnants. Prolonged residence of these particles in plasma may result in further cholesterol enrichment by transfer of cholesteryl ester from HDLs through the action of cholesteryl ester transfer 461
Copyright 9 2007 by Elsevier, Inc. All rights of reproduction in any form reserved.
462
FIGURE 33.1 pathways.
RONALDM. KRAUSS
Major lipoproteins and metabolic
protein. This process, in conjunction with progressive loss of triglyceride and smaller protein constituents from VLDL remnants, results in the formation of intermediate-density lipoproteins (IDLs, density 1.006-1.019 g/mL) and ultimately in the production of LDLs (density 1.019-1.063 g/mL). LDL particles retain apoB-100 and cholesteryl esters as the principal protein and lipid components, respectively, and are responsible for the bulk of cholesterol transport in plasma. As is the case for VLDLs, both IDLs and LDLs are heterogeneous, comprising multiple subpopulations differing in size, density, and composition. The metabolic and possible pathologic features of certain subclasses, particularly small, dense LDLs, have been reviewed elsewhere (7,8). The major fraction of both IDLs and LDLs are cleared from plasma by hepatic LDL receptors, which recognize both apoE and apoB-100 as ligands. This process is much slower than that for remnant clearance, resulting in LDL plasma residence times of several days. Alterations in composition of LDLs (and perhaps that oflDLs and remnants) may occur, likely related in part to prolongation of their circulation in plasma and to interaction with tissues. Transport of these lipoproteins into the subendothelial space of arteries, and their retention by arterial proteoglycans, can lead to oxidative changes and other modifications (9). Oxidized lipoprotein lipids can trigger release of cytokines from arterial macrophages. These inflammatory agents have a wide range of effects including cellular proliferation and recruitment of more macrophages. Unregulated uptake of modified lipoproteins by macrophages can in turn lead to cholesterol engorgement of these cells and the development of an early atherosclerotic plaque, designated a fatty streak. Excess arterial lipoprotein influx and the disruption of
cholesterol-enriched cells produces extracellular cholesterol deposition in the form of a lipid "core" that is covered by a raised fibrous cap. An increase in size of the lipid core, together with a weakening of the fibrous cap and perhaps vasoconstriction due to altered endothelial function, can lead to rupture of the atherosclerotic lesion and further inflammatory and procoagulant changes with resultant thrombus formation and vessel occlusion (10). Recently another type of lipoprotein particle, Lp(a), has been recognized as a factor of potential importance in the development of atherosclerosis (11). Lp(a) contains apoB100 complexed with another large protein, apo(a), which is highly homologous to plasminogen. Plasma concentrations of Lp(a) vary over a wide range in the general population and appear to be under strong genetic influence. Lp(a) levels have been highly correlated with coronary disease risk in epidemiologic studies, and apo(a) has been demonstrated in arterial lesions by immunochemical techniques. Although the mechanisms for the involvement of Lp(a) in atherosclerosis are not known, the interactions of apo(a) with the arterial endothelium, and perhaps also with the fibrinolytic system, may be of importance.
B. High-Density Lipoproteins HDLs comprise a complex array of particles with differing lipid and protein composition and metabolic behavior (12). Most HDL particles contain the protein apoAI, derived from both liver and intestine, and certain subpopulations also contain varying amounts of apoAII as well as apoCs and apoEs. The bulk of HDLs are thought to arise
CHAVTE~33 Lipids and Lipoproteins and Effects of Hormone Therapy from lipid-poor precursor particles secreted by both the liver and intestine, which acquire additional lipids and undergo a variety of transformations during their circulation in the blood. HDL lipids are derived both from the metabolism of apoB-containing lipoproteins, as described earlier, and from the direct uptake of cellular lipids, in particular, unesterified cholesterol. The enzyme lecithin:cholesterol acyltransferase (LCAT), activated by apoAI, converts unesterified to esterifled cholesterol and participates in the transformation of nascent HDLs in a stepwise manner to a series of progressively larger particles with increasing cholesterol and apoprotein content. The smaller and denser HDL subclasses are designated as HDL-3, while the larger and more buoyant are designated as HDL-2. Within both HDL-3 and HDL2, there are multiple discrete subspecies with differing content of apoAI and apoAII as well as other constituents. Of particular note is the largest major HDL subspecies, HDL2b, which contains four molecules of apoAI without apoAII. Levels of HDL-2b are higher in women than in men and in both sexes correlate with plasma concentrations of HDL cholesterol. HDL components leave the plasma by a variety of pathways. As described earlier, cholesteryl ester transfer protein may mediate cholesterol movement from HDLs to apoBcontaining lipoproteins, particularly lipolytic remnants, in exchange for triglycerides. The acceptor particles may then be taken up by LDL or remnant receptors. Subpopulations of HDLs containing apoE also may be cleared directly by such receptors. Alternately, HDLs may deliver cholesterol directly to tissues, either by selective uptake of HDL lipids or removal of intact HDL particles. A receptor, SR-B1, which is highly expressed in the liver as well as in adrenal and gonadal tissue, has been found to mediate tissue uptake of HDL cholesterol (13). Another key protein involved in HDL metabolism is ABCA1 (14). ABCA1 is a membrane transporter that facilitates the uptake of cholesterol and phospholipids from tissues by ApoA1. It plays a critical role in the hepatic biogenesis of HDL particles and is also a mediator of cholesterol efflux from tissues, including arterial macrophages. Recently a related transporter, ABCG1, has been found to have even greater capacity for promoting cholesterol uptake from tissues to large HDL-2 particles (15). Another mechanism implicated in regulating plasma HDL levels is the activity of hepatic lipase, an enzyme that catalyzes the hydrolysis of HDL phospholipids and triglycerides and potentiates hepatic HDL lipid uptake. Variations of hepatic lipase activity measured in postheparin plasma have been strongly inversely correlated with levels of HDL cholesterol. It has also been proposed that triglycerideenriched HDLs resulting from accelerated cholesterol/ triglyceride exchange are catabolized and perhaps more rapidly cleared as a result of hepatic lipase activity. The potential ability of HDLs to mediate delivery of cholesterol from peripheral tissues to the liver has led to
463
the hypothesis that HDLs have an anti-atherogenic effect by promoting "reverse cholesterol transport." However, a number of other factors may underlie the inverse relationship of HDL cholesterol to risk of atherosclerotic cardiovascular disease (16). For example, levels of HDL cholesterol are determined in part by the efficacy of clearance of triglyceride-rich lipoproteins and their remnants. Thus, high HDL levels may reflect lower levels or shorter plasma residence time of potentially atherogenic remnant particles. In addition, recent evidence suggests that HDLs may reduce the accumulation of lipid peroxides in LDLs. A direct effect of increased apoAI production in atherosclerosis prevention has been demonstrated in transgenic mice overexpressing the human apoAI gene (17). This suggests that, whatever the mechanism, genetic and other factors regulating apoAI synthesis may have important antiatherogenic effects.
II. EFFECTS OF ESTROGENS ON PLASMA LIPOPROTEIN METABOLISM Estrogen treatment has been shown to increase both the rate of hepatic triglyceride production and of VLDL apoB100 secretion (Fig. 33.2). Walsh et al. (18) have reported that oral micronized estradiol at a dose of 2 mg/day increased the production of apoB in large VLDL particles (of Svedberg flotation rate or Sf 60 to 400) to a much greater extent than smaller VLDL particles (Sf 20 to 60). Most of the additional large VLDLs were cleared directly from plasma, with a smaller portion converted to smaller VLDLs and LDLs. In addition, there was an increased fractional catabolic rate of LDLs. The latter finding is consistent with evidence from animal studies that estrogens increase the number of hepatic LDL receptors. The net influence of estrogens on apoB-containing lipoproteins thus depends on the balance between increased production of large, triglyceride-rich VLDLs and increased clearance of these particles as well as cholesterolrich remnants and LDLs. There is little information as to the metabolic mechanisms by which estrogens influence HDL levels. In nonhuman primates, estrogen treatment has been found to increase both apoA-1 and apoA-2 production rates (19). High-dose ethinyl estradiol treatment of premenopausal women was shown to increase production rates of HDL apoAI in particular (20). On the other hand, estrogen treatment results in significant suppression of hepatic lipase activity (21,22), and reduced HDL fractional clearance was reported in a study in postmenopausal women. Either increased apoAI production or reduced hepatic lipase activity might be expected to induce a preferential increase in HDL-2 subclasses. Estrogen has been shown to induce dramatic suppression of the SR-B1 (HDL) receptor in the liver, an effect that may also contribute to accumulation of cholesterol-rich HDL particles in plasma (23).
464
RONALD M. KRAUSS
FIC;URE 33.2 Effects of estrogens on plasma lipoprotein metabolism.
III. PROGESTIN EFFECTS ON LIPOPROTEIN METABOLISM The practice of using progestins to counteract estrogeninduced endometrial hyperplasia raises the question of what effects this may have on lipoprotein metabolism. Natural progesterone, although not extensively studied, has not been found to cause significant changes in levels of plasma lipoproreins (24-27). On the other hand, synthetic progestins, particularly those with evidence of androgenic activity (e.g., norethindrone, levonorgestrel), may have significant metabolic effects. The best documented effect of androgenic progestins on lipoprotein metabolism is increased hepatic lipase activity, which in turn has been strongly correlated with reduced levels of HDL cholesterol, particularly in the HDL-2 subclasses (21). Androgenic progestins are also known to reduce triglyceride levels (22), although the mechanism of this effect is not understood. Some effects of these agents may also be mediated by insulin resistance, which has been associated with a dyslipidemia characterized by increased levels of VLDLs and reduced levels of HDL cholesterol (27).
sources of variation, including study designs, size and characteristics of study populations, hormone formulations, treatment regimens, duration of use, and laboratory methodology. The Postmenopausal Estrogen/Progestin Intervention (PEPI) Trial has provided the most definitive information to date as to the effects of unopposed estrogen and various combination hormone treatment regimens on plasma lipoproteins and other risk factors for coronary heart disease (30). In this 3-year, double-blind study, 875 healthy postmenopausal women aged 45 to 64 were randomly assigned to placebo or treatment with conjugated equine estrogens (CEEs) 0.625 mg/day, alone or in combination with cyclic medroxyprogesterone acetate (MPA) 10 mg/day for 12 days/month, continuous MPA 2.5 mg/day, or cyclic micronized progesterone 200 mg/day for 12 days/month. The major findings from this and earlier studies may be summarized as follows.
A. Estrogen Therapy 1. Plasma triglyceride and VLDL levels increase in a dosedependent manner. In the Lipid Research Clinics of North
IV. EFFECTS OF POSTMENOPAUSAL HORMONE TREATMENT ON PLASMA LIPID AND LIPOPROTEIN CHOLESTEROL LEVELS A large number of cross-sectional and longitudinal studies have assessed the influence of postmenopausal hormone treatment on plasma lipoprotein concentrations (reviewed in [28,29]). The findings of these studies reflect a number of
America Study (LRC), a comparison of 370 nonmenstruating, 45- to 64-year-old women not using estrogens versus 239 women using equine estrogens at varying doses demonstrated a 26% higher median triglyceride level in the estrogen users (31). A summary analysis of 10 randomized and crossover studies employing conjugated estrogens, adjusted for variables including sample size (ranging from 6 to 265 women) and duration of treatment (ranging from 1 to 12 months), revealed an overall increase in triglycerides of 20%, with levels increasing as a function of estrogen dose (32). A similar mean increase
CHAPTER 33 Lipids and Lipoproteins and Effects of Hormone Therapy (24%) was observed by Walsh et al. (18) in a doubleblind, placebo-controlled crossover study of 31 postmenopausal women treated with CEE for 3 months at a dose of 0.625 mg/day, while the mean increase at 1.25 mg/day was substantially greater (38%). These findings are consistent with those of the PEPI trial, in which CEE 0.625 mg/day resulted in a 25% increase in plasma triglyceride among adherent subjects (33). It should be noted that these studies were carried out in normolipidemic women. Women with primary hypertriglyceridemia may develop severe hyperlipemia on estrogen replacement therapy (34), and in such patients, estrogens should be used with caution and only after therapeutic reduction of triglyceride levels to less than 500 mg/dL. 2. Plasma total and LDL cholesterol levels decrease. In the LRC study, median LDL cholesterol was 11% lower in CEE users versus nonusers (31). Three other large crosssectional studies revealed similar reductions in LDL cholesterol (35-37), with dose-response effects between -<0.625 rag/day and ->0.9 to 1.25 rag/day reported in two of the studies (35,37). In the study by Walsh et al. (18), the reductions were somewhat greater: 15% at the lower dose and 19% at the higher dose, with a nonsignificant dosage effect. LDL cholesterol reduction with CEE 0.625 rag/day in the PEPI trial was 10% (31), with a somewhat greater effect when the analyses were restricted to adherent subjects (33).
3. Plasma HDL cholesterol levels increase, principally HDL-2. Median H D L cholesterol was found to be 10% higher in users of conjugated estrogens in the LRC study (31), with similar increases observed in three other crosssectional studies (35-37), two of which detected an apparent effect of dose (35,37). In the 10 prospective trials summarized by Bush and Miller (32), the average adjusted increase was also 10%, without an apparent dose effect. The PEPI trial observed an H D L cholesterol increase of 9% with CEE 0.625 mg/dL (30) and a 13% increase among adherent subjects (32). In the studies of Sherwin and Gelfand (38) and Walsh et al. (18), increases in H D L cholesterol were 14% and 16% at CEE 0.625 rag/day and 19% and 18% at 1.25 rag/day, suggestive of a small but not significant dose dependence. As in a number of other studies (39,40), the increase was restricted to the HDL-2 subclasses. Thus, as is the case with LDL cholesterol, there may be saturation of the effects of CEE at 0.625 rag/day, although there is clearly much interindividual variation in response. 4. The lipid and lipoprotein effects of estrogens are greater with oral than with systemic administration. A number of studies have indicated that systemic estrogen therapy, administered as percutaneous estradiol patches or creams, have little or no impact on plasma lipid and lipoprotein cholesterol levels (41-47). Other studies, however, have detected moderate increases in levels of HDLs (40,48-50)
465 and reductions in LDLs (26,50,51). Although there are a number of differences in experimental methodology among these studies, it has been pointed out that some differences in the findings may be related to the relatively greater plasma levels of systematically administered estrogens required to achieve intrahepatic hormone levels comparable with those achieved via the portal circulation after oral administration (18).
B. Combination Hormone Therapy The effects ofestrogen/progestin combinations on plasma lipid and lipoprotein cholesterol levels in postmenopausal women have been examined in a large number of studies employing various hormonal replacement regimens, as reviewed elsewhere (28,29). The following represents general observations that may be drawn from an overview of these and other studies, including the PEPI trial (30,31):
1. The metabolic effects of addedprogestin are related to the dose and relative androgenic potency of the hormone preparation and the concomitant dose of estrogen. On a weight basis, the C21-hydroxyprogesterone derivatives (e.g., medroxyprogesterone and medrogestone) are less metabolically active than the 19-nortestosterone derivatives (e.g., norethindrone) and levonorgestrel (23,52,53). Progesterone appears to be relatively inert in its effects on lipoprotein metabolism (23). 2. Addition of progestin has small and variable effects on levels ofplasma triglyceride and VLDL levels compared with those produced by estrogen alone. In the PEPI trial, among adherent subjects, triglycerides increased 15% to 20% in each opposed CEE arm versus 25% with CEE alone, although this difference was not statistically significant (33). Estrogen-induced increases in triglyceride level tend to be blunted to a greater extent by the more androgenic progestins (54). This may be related to the finding that a combination of estradiol and (d,1)-norgestrel results in increased fractional VLDL apoB clearance without an increase in VLDL apoB production rate (54). Recently it has been reported that levels of remnant-like lipoproteins, a subgroup of triglyceride-rich lipoprotein particles with potentially atherogenic properties, are reduced by conjugated estrogens with or without the addition of MPA, but the clinical significance of this observation is not known (55). 3. Estrogen-induced reductions in LDL cholesterol are affected minimally, if at all, by addition of conventional doses ofprogestins. A number of studies have shown similar reductions of LDL cholesterol with or without the concomitant administration ofprogestins (28,30,51,54). Although this would appear to suggest minimal interference of progestogens with the beneficial effects of estrogen on LDL metabolism, it has been reported that an oral
466 estradiol-norgestrel combination decreased LDL apoB production rate without affecting fractional LDL apoB clearance (56). This finding stands in contrast to the report that estrogen therapy alone reduces LDL levels primarily by increasing fractional LDL apoB catabolic rate (18). Thus, estrogens and estrogen/progestin combinations might alter LDL levels by different mechanisms, and this could conceivably have consequences for the prediction of effects on coronary disease risk. 4. Mos@rogestins attenuate estrogen-induced increases in H D L cholesterol (principally HDL-2). Although this effect has been reported to be more pronounced with high doses of 19-nortestosterone derivatives or levonorgestrel than with 21-hydroxyprogesterone derivatives, substantial reversal of estrogen-induced increases in HDL and HDL2 cholesterol have been observed with endometrialsuppressive doses of cyclic medroxyprogesterone (10 rag), norethindrone (1 mg), and (d,1)-norgestrel (0.15 mg) (25,50). In PEPI, HDL cholesterol increased by only 2% with either the cyclic or continuous CEE/MPA regimen (30) and by 7% with the CEE/micronized progesterone combination. Among adherent subjects, these increases were significantly less than with unopposed CEE (33). 5. Observational studies of lipid and lipoprotein levels in older users of combination hormone replacement therapy have not demonstrated the effects on H D L levels observed in shortterm clinical trials. A cross-sectional study in a retirement community in California showed no differences in plasma triglyceride, L D L cholesterol, or HDL cholesterol levels in users of estrogen-only versus combination hormone replacement therapy (36). Similar results have been observed recently in studies carried out in a second California retirement community (37). In the latter study, there were also no detectable differences in lipids and lipoproteins with doses of medroxyprogesterone ranging from 2.5 to 10 mg/day, administered from 5 to 25 days/month in combination with estrogen. In both studies, differences in lipid and lipoprotein levels between nonusers of hormones and estrogen-only users were comparable to those reviewed earlier. Although multiple explanations might be offered for these results, ranging from subject selection to poor progestin compliance, it should also be noted that longer-term prospective treatment trials (e.g., >3 years) have not been performed in older women, and it is possible that progestin-induced alterations of lipoproteins (notably HDLs) become attenuated over time.
C. Selective Estrogen Receptor Modulators Effects on lipoproteins and other cardiovascular risk markers have been described for selective estrogen receptor modulators (SERMs) that have estrogen-agonistic effects on bone and estrogen-antagonistic effects on breast and
RONALD M. KRAUSS
uterus (57). At two doses of raloxifene, LDL cholesterol was reduced by 12%, similar to the effects of estrogen and hormone treatment. In contrast to hormone treatment, however, there were no significant increases in triglyceride or HDL cholesterol, although HDL-2 was increased by 11%. Similar lipoprotein effects have been observed in cholesterol-fed nonhuman primates, in which raloxifene, unlike CEE, had no benefit on atherosclerosis severity (58).
V. EFFECTS OF HORMONE REPLACEMENT THERAPY ON OTHER LIPOPROTEIN PARAMETERS THAT MAY AFFE CT CARD I OVAS CULAR DISEASE RISK Several recent studies have found reductions in levels of Lp(a) with hormone replacement therapy (59,60). In PEPI, these ranged from 17% to 23% in comparison with placebo and did not differ significantly among treatment arms (59). A post hoc analysis of the Heart and Estrogen/progestin Replacement Study also showed that combination hormone treatment lowered Lp(a) levels and, moreover, that coronary heart disease events were more favorably affected in women with higher versus lower Lp(a) levels (61). Raloxifene has also been reported to lower Lp(a), but to a significantly lesser extent (57). Thus, although there is no definitive evidence as to the benefits of Lp(a) reduction on cardiovascular disease risk, hormone treatment is one of the few therapies capable of lowering elevated Lp(a) levels. Changes in composition and metabolism of atherogenic lipoproteins may have an impact on cardiovascular risk that is not reflected in standard measurements of their plasma levels. Estrogens have been shown to increase triglyceride content of LDLs as well as other lipoprotein classes (31,40), but these changes have not been directly related to altered cardiovascular disease risk. Reductions in levels of plasma apoB-100 have been reported with estrogen replacement (62) and with combined estrogen/progestin therapy (63), but these reductions are not as great as for LDL cholesterol (40,63). This is due to preferential reduction in levels of large, buoyant LDL subspecies with a higher ratio of cholesterol to apoB LDL than is found in smaller denser LDL (64,65). Metabolic studies have indicated that estrogen induces increased turnover of both large and small LDL and that the preferential reduction in plasma levels of large LDL is due to a reduced rate of production of these particles (65). Although smaller LDL particles signify increased risk for coronary heart disease (8,66,67), the clinical implications of the differing effects of estrogen on metabolism of larger and smaller LDL particles are not known. Estrogens have also been found to increase plasma clearance of potentially ath-
CHAPTER 33 Lipids and Lipoproteins and Effects of Hormone Therapy erogenic remnants formed during the course of postprandial triglyceride-rich lipoprotein metabolism (68), an effect consistent with the earlier demonstration of an increase in fractional catabolic rate of smaller V L D L particles (18). Finally, although estrogens have been found to reduce susceptibility of L D L s (69-71) and H D L (71) to oxidative modification, it has not been established to what extent this may contribute to protection from L D L oxidation and atherosclerosis in vivo.
VI. RELATION OF H O R M O N E TREATMENT EFFECTS ON PLASMA LIPOPROTEINS TO CARDIOVASCULAR DISEASE RISK Despite observational data suggesting that higher H D L cholesterol could contribute to reduced cardiovascular disease risk in hormone users (71) and the potential further reduction that could be expected from L D L lowering, such benefits have not been demonstrated in clinical trials of estrogen treatment, either alone or in combination with progestin (72-74). Moreover, in the Women's Health Initiative ( W H I ) , a small but significant increase in risk of myocardial infarction and stroke was observed with combination hormone treatment (74). A number of possible explanations have been suggested for these results. As noted previously, estrogen-induced reductions in L D L cholesterol are due primarily to lower levels of larger L D L particles, which are less strongly related to C V D risk than smaller L D L (8,66). It has been shown recently that postmenopausal women with coronary atherosclerosis manifested by coronary artery calcification had higher levels of large V L D L and small L D L particles, higher L D L particle concentration, and smaller mean L D L size compared with women with no detectable coronary artery calcification. It was suggested that the finding of a lack of benefit of hormonal therapy on coronary artery calcium in this cohort could be attributed in part to the finding that despite having lower levels of L D L cholesterol, hormone users had no reduction in concentrations of total L D L particles or small L D L and also had higher levels of V L D L particles (75). It is also conceivable that the nature of the estrogeninduced changes in H D L metabolism do not confer the same anti-atherogenic benefit as that associated with other causes of higher H D L . Another possibility, discussed in Chapter 34, is that pro-inflammatory and/or prothrombotic effects of hormonal treatment counteract anti-atherogenic lipoprotein changes, although there is as yet no direct evidence to support this explanation. In addition, as suggested recently (76), any potential cardioprotection afforded by estrogen in the postmenopause may require earlier hormone
467
exposure than was the case in the trials conducted to date. Finally, there may be subsets of women who are predisposed to either protective or adverse metabolic effects of hormonal treatment, possibly on a genetic basis, and the impact of hormones on C V D risk in these groups are not discriminated in the data from the overall population.
References 1. Kannel WB. Metabolic risk factors for coronary artery disease in women: perspective from the Framingham study. Am Heart J 1987; 114:413-419. 2. Hjortland MC, McNamara PM, Kannel WB. Some atherogenic concomitants of menopause: the Framingham study.Am J Epidemio11976; 103:304-311. 3. Jensen J, Nilas L, Christiansen C. Influence of menopause on serum lipids and lipoproteins. Maturitas 1990;12:321-331. 4. Matthews KA, Meilahn E, Kuller LH, et al. Menopause and risk factors for coronary heart disease. N EnglJ Med 1989;321:641-646. 5. Kuller LH, Gutai JP, Meilahn E, et al. Relationship of endogenous sex steroid hormones to lipids and apoproteins in postmenopausal women. Arteriosclerosis 1990;10:1058-1066. 6. Rifldnd BM, Tamir I, Heiss G, et al. Distribution of high density and other lipoproteins in selected LRC Prevalence Study populations. Lipids 1979;14:105-112. 7. Krauss RM. Relationship of intermediate and low-density lipoprotein subspecies to risk of coronary artery disease.Am HeartJ 1987;113:578582. 8. Berneis, K, Krauss RM. Metabolic origins and clinical significance of LDL heterogeneity.J Lipid Res 2002;43:1363-1379. 9. Witztum JL. The role of oxidized LDL in atherosclerosis. Adv Exp Med Bio11991;285:353-365.
10. Libby P, Theroux R Pathophysiology of coronary artery disease. Circulation 2005;111:3481-3488. 11. Scanu AM, Scandiani L. Lipoprotein(a): structure, biology, and clinical relevance. Adv Intern Med 1991;36:249-270. 12. Krauss RM. Regulation of high density lipoprotein levels. Med Clin North Am 1982;66:403-430. 13. Acton S, Rigotti A, Landschulz KT, et al. Identification of scavenger receptor SR-BI as a high density lipoprotein receptor. Science 1996; 271:518-520. 14. Oram JF, Heinecke JW. ATP-binding cassette transporter AI: a cell cholesterol exporter that protects against cardiovascular disease. Physiol Rev 2005;85:1343-1372. 15. Wang N, Lan D, Chen W, Matsuura F, Tall AR. ATP-binding cassette transporters G1 and G4 mediate cellular cholesterol efflux to highdensity lipoproteins. Proc NatlAcad Sci U S A 2004;101:9774-9779. 16. Rubin EM, Krauss RM, Spangler EA, et al. Inhibition of early atherogenesis in transgenic mice by human apolipoprotein AI. Nature 1991;353:265-267. 17. Tribble DL, Krauss RM. HDL and coronary artery disease.Adv Intern Med 1993;38:1-29. 18. Walsh BW, Schiff I, Rosner B, et al. Effects of postmenopausal estrogen replacement on the concentrations and metabolism of plasma lipoproteins. N EnglJ Med 1991;325:1196-1204. 19. Kushwaha RS, Foster DM, Murthy VN, et al. Metabolic regulation of apoproteins of high-density lipoproteins by estrogen and progesterone in the baboon (Papio sp). Metabolism 1990;39:544-552. 20. Schaefer EJ, Foster DM, Zech LA, et al. The effects of estrogen administration on plasma lipoprotein metabolism in premenopausal females.J Clin Endocrinol Metab 1983;57:262-267.
468 21. Tikkanen MJ, Nikkila EA, Kuusi T, et al. High density lipoprotein-2 and hepatic lipase: reciprocal changes produced by estrogen and Norgestrel. J Clin EndocrinolMetab 1982;54:1113-1117. 22. Landschulz KT, Pathak RK, Rigotti A, et al. Regulation of scavenger receptor, class B, type I, a high density lipoprotein receptor, in liver and steroidogenic tissues of the rat.J Clin Invest 1996;98:984-995. 23. Krauss RM. Effects of progestational agents on serum lipids and lipoproteins. J Reprod Med 1982;27:503-510. 24. Fahraeus L, Larsson CU, Wallentin L. L-norgestrel and progesterone have different influences on plasma lipoproteins. Eur J Clin Invest 1983;13:447-453. 25. Ottosson UB, Johansson BG, Von Schoultz B. Subfractions of highdensity lipoprotein cholesterol during estrogen replacement therapy: a comparison between progestogens and natural progesterone. Am J Obstet Gyneco11985;151:746-750. 26. Jensen J, Riis BJ, Strom V, et al. Long-term effects of percutaneous estrogens and oral progesterone on serum lipoproteins in postmenopausal women. Am J Obstet Gyneco11987;156:66-71. 27. Laws A, Reaven GM. Evidence for an independent relationship between insulin resistance and fasting plasma HDL-cholesterol, triglyceride and insulin concentrations. J Intern Med 1992;231:25-30. 28. Rijpkema AHM, Van der Sanden AA, Ruijs AHC. Effects of postmenopausal oestrogen-progestogen replacement therapy on serum lipids and lipoproteins: a review. Maturitas 1990;12:259-285. 29. Sacks FM, Walsh BW. The effects of reproductive hormones on serum lipoproteins: unresolved issues in biology and clinical practice. Ann N YAcad Sci 1990;592:272-285. 30. The Writing Group for the PEPI Trial. Effects of estrogen or estrogen/ progestin regimens on heart disease risk factors in postmenopausal women: the Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. JAm MedAssoc 1995;273:199-208. 31. Wahl P, Walden C, Knopp R, et al. Effect of estrogen/progestin potency on lipid/lipoprotein cholesterol. N Engl J Med 1983;308: 862-867. 32. Bush TL, Miller VT. Effects of pharmacologic agents used during menopause: impact on lipids and lipoproteins. In: Mishell DR, ed. Menopause: physiology and pharmacology. Chicago: Year Book Medical Publishers, 1987:187-208. 33. Barrett-Connor E, Slone S, Greendale G, et al. The Postmenopausal Estrogen/Progestin Interventions Study: primary outcomes in adherent women. Maturitas 1997;27:261-274. 34. Glueck CJ, Scheel D, Fishback J, et al. Estrogen-induced pancreatitis in patients with previously covert familial type V hyperlipoproteinemia. Metabolism 1972;21:657-666. 35. Krauss RM, Perlman JA, Ray R, et al. Effects of estrogen dose and smoking on lipid and lipoprotein levels in postmenopausal women.Am J Obstet Gyneco11988;158:1606-1611. 36. Barrett-Connor CE, Wingard DL, Criqui MH. Postmenopausal estrogen use and heart disease risk factors in the 1980s. Rancho Bernardo, Calif, revisited.JAm MedAssoc 1989;261:2095-2100. 37. Paganini-Hill A, Dworsky R, Krauss RM. Hormone replacement therapy, hormone levels, and lipoprotein cholesterol concentrations in elderly women. Am J Obstet Gyneco11996;174:897-902. 38. Sherwin BB, Gelfand MM. A prospective one-year study of estrogen and progestin in postmenopausal women: effects on clinical symptoms and lipoprotein lipids. Obstet Gyneco11989;73:759-766. 39. Krauss RM, Lindgren FT, Wingerd J, et al. Effects of estrogens and progestins on high density lipoproteins. Lipids 1979;14:113-118. 40. Moorjani S, Dupont A, Labrie F, et al. Changes in plasma lipoprotein and apolipoprotein composition in relation to oral versus percutaneous administration of estrogen alone or in cyclic association with utrogestan in menopausal women.J Clin EndocrinolMetab 1991;73:373-379. 41. Fahraeus L, Larsson-Cohn U, Wallentin L. Lipoprotein during oral and cutaneous administration of estradiol-17b to menopausal women. Acta Endocrinol Copenh 1982;101:597-602.
RONALD M. KRAUSS 42. Elkik F, Gompel A, Mercier-Bodard C. Effects of percutaneous estradiol and conjugated estrogens on the level of plasma proteins and triglycerides in postmenopausal women. Am J Obstet Gynecol1982;143: 888-892. 43. Basdevant A, De Lignieres B, Guy-Grand B. Differential lipemic and hormonal responses to oral and parenteral 17 beta-estradiol in postmenopausal women. Am J Obstet Gyneco11983;147:77-81. 44. Mandel FP, Geola FL, Meldrum DR. Biological effects of various doses of vaginally administered conjugated equine estrogens in postmenopausal women. J Clin EndocrinolMetab 1983;57:133-139. 45. Farish ECD, Fletcher DM, Hart F, et al. The effects of hormone implants on serum lipoproteins and steroid hormones in bilaterally oophorectomised women. Acta Endocrinol (Copenh) 1984;106:116-120. 46. Chetkowski RJ, Meldrum DR, Steingold KA, et al. Biologic effects of transdermal estradiol. N EnglJ Med 1986;314:1615-1620. 47. De Lignieres B, Basdevant A, Thomas G, et al. Biological effects of estradiol-17 beta in postmenopausal women: oral versus percutaneous administration. J Clin EndocrinolMetab 1986;62:536-541. 48. Lobo RA, March CM, Goebelsmann U, et al. Subdermal estradiol pellets following hysterectomy and oophorectomy. Effect upon serum estrone, estradiol, luteinizing hormone, follicle-stimulating hormone, corticosteroid binding globulin-binding capacity, testosterone-estradiol binding globulin-binding capacity, lipids, and hot flushes. Am J Obstet Gyneco11980;138:714-719. 49. Stanczyk FZ, Shoupe D, Nunez V, et al. A randomized comparison of nonoral estradiol delivery in postmenopausal women. Am J Obstet Gyneco11988;159:1540-1546. 50. Sharf M, Oettinger M, Lanir A, et al. Lipid and lipoprotein levels following pure estradiol implantation in postmenopausal women. Gynecol Obstet Invest 1985;19:207-212. 51. Crook D, Cust MP, Gangar KF, et al. Comparison of transdermal and oral estrogen-progestin replacement therapy: effects on serum lipids and lipoproteins. Am J Obstet Gyneco11992;166:950-955. 52. Dorflinger LJ. Relative potency of progestins used in oral contraceptives. Contraception 1985;31:557-570. 53. Sonnendecker EW, Polakow ES, Benade AJ, Simchowitz E. Serum lipoprotein effects of conjugated estrogen and a sequential conjugated estrogen-medrogestone regimen in hysterectomized postmenopausal women. A m J Obstet Gyneco11989;1:1128-1134. 54. Miller VT, Muesing RA, LaRosa JC, et al. Effects of conjugated equine estrogen with and without three different progestogens on lipoproteins, high-density lipoprotein sub-fractions, and apolipoprotein A-I. Obstet Gyneco11991;77:235-240. 55. Lamon-Fava S, Posfai B, Asztalos BF, et al. Effects of estrogen and medroxyprogesterone acetate on subpopulations of triglyceride-rich lipoproteins and high-density lipoproteins. Metabolism 2003;52:13301336. 56. Wolfe BM, Huff MW. Effects of combined estrogen and progestin administration on plasma lipoprotein metabolism in postmenopausal women.J Clin Invest 1989;83:40-45. 57. Walsh BW, Kuller LH, Wild RA, et al. Effects of raloxifene on serum lipids and coagulation factors in healthy postmenopausal women. JAm MedAssoc 1998;279:1445-1451. 58. Clarkson TB, Anthony MS, Jerome CP. Lack of effect of raloxifene on coronary artery atherosclerosis of postmenopausal monkeys. J Clin Endocrinol Metab 1998;83:721-726. 59. Soma M, Fumagalli R, Paoletti R, et al. Plasma Lp(a) concentration after oestrogen and progestogen in postmenopausal women. Lancet 1991;337:612. 60. Espeland MA, Marcovina SM, Miller V, et al. Effect of postmenopausal hormone therapy on lipoprotein(a) concentration. Circulation 1998;97:979-986. 61. Shlipak MG, Simon JA, Vittinghoff E, et al. Estrogen and progestin, lipoprotein(a), and the risk of recurrent coronary heart disease events after menopause.JAm MedAssoc 2000;283:1845-1852.
CHAPTER 33 Lipids and Lipoproteins and Effects of Hormone Therapy 62. Applebaum-Bowden D, McLean R Steinmetz A, et al. Lipoprotein, apolipoprotein, and lipolytic enzyme changes following estrogen administration in postmenopausal women. J Lipid Res 1989;30:18951906. 63. Haarbo J, Hassager C, Jensen SB, et al. Serum lipids, lipoproteins, and apolipoproteins during postmenopausal estrogen replacement therapy combined with either 19-nortestosterone derivatives or 17-hydroxyprogesterone derivatives. Am J Med 1991;90:584-589. 64. Campos H, Sacks FM, Walsh BW, et al. Differential effects of estrogen on low-density lipoprotein subclasses in healthy postmenopausal women. Metabolism 1993;42:1153-1158. 65. Campos H, Walsh BW, Judge H, Sacks FM. Effect of estrogen on very low density lipoprotein and low density lipoprotein subclass metabolism in postmenopausal women. J C/in Endocrinol Metab 1997;82: 3955-3963. 66. Gardner CD, Fortmann SR Krauss RM. Association of small lowdensity lipoprotein particles with the incidence of coronary artery disease in men and women.JAm Meddssoc 1996;276:875-881. 67. St-Pierre AC, Cantin B, Dagenais GR, et al. Low-density lipoprotein subfractions and the long-term risk of ischemic heart disease in men: 13-year follow-up data from the Quebec Cardiovascular Study. Arterioscler Tbromb VascBid 2005;25:553-559. 68. Westerveld HT, Kock LA, van Rijn HJ, et al. 17 beta-Estradiol improves postprandial lipid metabolism in postmenopausal women.J Clin Endocrinol Metab 1995;80:249-253. 69. Maziagere C, Auclair M, Ronveaux MF, et al. Estrogens inhibit copper and cell-mediated modification of low density lipoprotein.Atherosclerosis 1991;89:175-182.
469
70. Sack MN, Rader DJ, Cannon RO III. Oestrogen and inhibition of oxidation of low-density lipoproteins in postmenopausal women. Lancet 1994;343:269-270. 71. Wakatsuki A, Ikenoue N, Sagara Y. Effects of estrogen on susceptibility to oxidation of low-density and high-density lipoprotein in postmenopausal women. Maturitas 1998;28:229-234. 72. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. J A m MedAssoc 2004;291:1701-1712. 73. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. J A m MedAssoc 2002;288:321-333. 74. Shlipak MG, Chaput LA, Vittinghoff E, et al. Lipid changes on hormone therapy and coronary heart disease events in the Heart and Estrogen/progestin Replacement Study (HERS). Am Heart J 2003; 146:870-875. 75. Mackey RH, Kuller LH, Sutton-Tyrrell K, et al. Hormone therapy, lipoprotein subclasses, and coronary calcification: the Healthy Women Study. Arch Intern Med 2005;165:510-515. 76. Grodstein F, Manson JE, Stampfer MJ. Hormone therapy and coronary heart disease: the role of time since menopause and age at hormone initiation.J Womens Health (Larcbmt) 2006;15:35-44.
This Page Intentionally Left Blank
The Effects of Hormone Therapy on Inflammatory, 9 ly"tic Hemostatlc,9 and Fibrlno Markers in Postmenopausal
Women KWANG K O N K O H
BYUNG-Koo Y O O N C. NOEL BAIREY M E R Z
ICHIRO SAKUMA
VascularMedicine and Atherosclerosis Unit, Division of Cardiology, Gil Medical Center, Gachon Medical School, Namdong-Gu, Incheon, Korea 405-760 Obstetrics and Gynecology, Samsung Medical Center, Sungkyunkwan University, School of Medicine, Seoul, Korea Division of Cardiology, Department of Medicine, Cedars-Sinai Research Institute, CedarsSinai Medical Center; Department of Medicine, University of California School of Medicine, Los Angeles, CA 90048 CardiovascularMedicine, Hokko Memorial Hospital, Sapporo, Japan
ROBERT W . REBAR American Society for Reproductive Medicine, Birmingham, AL 35216
I. I N T R O D U C T I O N
hinges on thrombogenic as well as inflammatory cellular and molecular pathways. Hormone therapy (HT) has both anti-inflammatory and pro-inflammatory effects and HT activates coagulation and improves fibrinolysis (1,4). These effects depend on the route of administration, doses of estrogen, the age of the woman, and presence of coronary artery disease or the coexistence of other risk factors for hypercoagulability. In this review, we discuss the effects of HT on inflammation, hemostasis, and fibrinolysis markers and briefly discuss the
Estrogen deficiency is associated with endothelial dysfunction accompanied by inflammation in the vessel wall, increased lipoprotein oxidation, smooth muscle cell proliferation, extracellular matrix deposition, accumulation of lipid-rich material, activation of platelets, and thrombus formation (1). These pathogenic features contribute to the development of atherosclerosis and coronary heart disease (1-3). The clinical manifestation of atherosclerotic disease TREATMENT OF THE POSTMENOPAUSAL WOMAN
471
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any formreserved.
472
KOH E T
effects of lower doses of H T and tibolone in postmenopausal women.
II. EFFECTS OF HORMONE THERAPY ON INFLAMMATION MARKERS The acute and chronic manifestations of atherosclerosis are increasingly being considered as a consequence of a chronic inflammatory process (2). Once activated by injury, endothelial and smooth muscle cells of large arteries become transcriptionally active and synthesize proinflammatory proteins. The gene transcription of many adhesion molecules, including vascular cell adhesion molecule-1 (VCAM-1) and intercellular CAM-1 (ICAM-1), is regulated by a nuclear transcription factor, NF-KB, normally maintained in an inactive state. In addition to VCAM-1, ICAM-1, and E-selectin, NF-KB also activates transcription of genes encoding chemoattractant factors, such as monocyte chemotactic peptide and macrophage stimulatory factor, which attract monocytes into the vessel wall (Fig. 34.1) (1). NF-KB additionally increases the synthesis and release of cytokines such as interleukin (IL)-I, IL-2, and IL-6, which activate inflammatory cells, enhancing their attachment to the vessel wall (1). 17]3-estradiol (E2) bound to the estrogen receptor (ER)-cx has shown to antagonize NF-KB activity in human hepatoma HepG2 cells (5).
A. C-Reactive Protein Obesity as well as age are positively correlated with C-reactive protein (CRP) concentrations (6,7). Messenger cytokines including IL-1 and TNF-ci, released by adipose tissue, contribute to the increase in levels of CRP. This
FIGURE34.1 Manystimuli initiate transcription of genes in endothelium that encode protein mediators of inflammation and hemostasis.Estrogens may modulate this processby inhibiting activationof nuclear transcription factor,NF-KB. (Modifiedfrom ref. 1, with permission.)
AL.
cross-talk between inflammation and metabolism might have a greater impact on coronary heart disease risk in women than in men (8). Mthough IL-6 levels increase following natural or surgical menopause (9,10), CRP concentrations are not influenced by menopause per se (11). Nested case-control analyses of the Women's Health Study (12) and the Nurses' Health Study (13) reported the predictive value of CRP in determining the risk of future cardiovascular events in women as well as men. Orally administered estrogens increase blood CRP levels (14-20) and maintain this sustained increase up to 3 years (15). The CRP increase by oral estrogen might be higher in heavier women (21). In contrast, transdermal estradiol significantly lowers CRP levels (22,23) or does not change levels in healthy postmenopausal women (16,21). In premenopausal women, CRP correlates inversely with blood estradiol concentrations during the menstrual cycle (24). In 389 postmenopausal women with increased cardiovascular risk, CRP levels significantly increased after 6 months of H T when compared with baseline. On the contrary, other inflammatory markers such as plasma levels of E-selectin, ICAM-1, VCAM-1, IL-6, and s-thrombomodulin are significantly decreased after H T (20). These observations strongly suggest that the CRP increase after oral H T might be related to metabolic hepatic activation rather than a systemic proinflammatory response. Co-administered medroxyprogesterone acetate (MPA) for 3 months may attenuate the action of conjugated equine estrogen (CEE) in a dose-dependent manner in healthy postmenopausal women (mean age 52 years, mean body mass index [BMI] 21.8 kg/m 2) (25). Compared with CEE alone, the addition of MPA at 2.5 mg increased CRP significantly, but to a less degree, and 5 mg of MPA was found to abolish the C RP increase. Other studies have confirmed the impact of MPA at 2.5 mg (15,26). The
CHAPTER34 The Effects of Hormone Therapy on Inflammatory, Hemostatic, and Fibrinolytic Markers in Postmenopausal Women 473 in humans has been suggested by its localization in atherosclerotic plaques. Transendothelial migration and differentiation of monocytes into macrophages in the subendothelial space is also controlled by monocyte chemoattractant protein-1 (MCP-1). In vitro and animal studies have demonstrated that estrogens suppress gene expressions of CAMs (33,34), MCP-1 (34-36), T N F - a (37), and IL-6 (34,38). Blood levels of the soluble CAMs, chemokines, and cytokines are potential candidates as markers of inflammation. At present, however, these assays remain research tools and are not appropriate for routine use in the clinical laboratory (39). Soluble CAMs (40-42) and MCP-1 (43,44) might reflect a local arterial inflammatory reaction. Koh et al. (45) first showed that transdermal estradiol alone or combined with MPA decreases ICAM-1 levels. CEE 0.625 mg also decreases E-selectin (15,25,45-47), and both ICAM-1 and VCAM-1 levels are reduced (26,46). C E E + M P A also consistently resulted in a decrease of soluble CAMS as well as E-selectin (Fig. 34.2) (15,25,26,47). C E E + M P showed a reduction comparable to that of C E E + M P A (15,47). A significant correlation has been found between the mean maximum carotid intimal medial thickness and MCP-1 levels at baseline in postmenopausal women. In this study, E2 significantly reduced MCP-1 levels (48). MCP-1 concentrations were also lowered by both progestogens in combination with CEE by a similar degree (see Fig. 34.2) (47,49). TNF-c~ is a multifunctional circulating cytokine derived from endothelial and smooth muscle cells as well as macrophages associated with coronary atheroma. Furthermore, TNF-e~ enhances the rate of monocyte recruitment into developing atherosclerotic lesions. CEE with MP or MPA significantly reduced TNF-R levels from baseline in hypertensive or overweight postmenopausal women (50). This observation appears to be consistent with the findings of
effects of micronized progesterone (MP) also have been studied. Koh et al. (27) demonstrated that MP 100 mg with CEE 0.625 mg daily for 2 months led to significant CRP increases in healthy postmenopausal women (mean age 57 years, mean BMI 24.8 kg/m2). Of note, MP 100 mg with CEE 0.3 mg daily did not increase CRP, which suggests a dose-dependency of oral estrogen similar to that of MPA. According to the Postmenopausal Estrogen/Progestin Interventions (PEPI) trial, there are no long-term differences in CRP levels between C E E + M P A and C E E + M P (15). However, the controversy regarding the significance of the pro-inflammatory effects of estrogen persists (28-30). The Women's Health Initiative (WHI) observational study confirmed that the increase in CRP levels with oral HT is not linked to the increase in the risk of coronary heart disease (31). This result is consistent with the report that there is no relationship between CRP and the Framingham Coronary Heart Disease Risk Score among women taking HT (32). Nonetheless, because CRP has several important atherogenic properties in addition to being an inflammation biomarker (28), the oral estrogen-induced CRP increase over several years may have the potential to contribute to atherosclerosis progression. B. Cell A d h e s i o n M o l e c u l e s , C h e m o k i n e s , and C y t o k i n e s The initial step of recruitment of circulating leukocytes to a specific locus of inflammation is the contact of leukocytes with activated endothelium, which is mediated by the selectin family, including E-selectin and P-selectin. Thereafter, ICAM-1 and VCAM-1 facilitate firm attachment (1,8). The pathophysiologic relevance of cell adhesion molecule (CAM)
6O 50oo > o3 u
1000
P<0.05
40
P
-
P
iii
&',
20
-~
lO-
800
>
30-
6oo -
2;
400
-
_o
200
-
8
8 I
0
I Base line
I
I CEE+ MP
I
0
P<0.05 -1-' '
400-
"7 2;
i
150
P=0.048
by A N O V A
200 -
_o 2
100
>
90
oL (D 2;
60
_5
-
P<0.05 [
. P<0 05
i
i CEE+ MPA P<0.005 by A N O V
12o
300
< o
i CEE+ MP
180 P<0.05
oJ
i Base line
CEE+ MPA
500
> o~ u
P=0.016 by A N OVA
-
_5 < o > __e
-
P
-r~
03
<-
3O
8 0
I
I Base line
I
I CEE+ MP
I
, CEE+ MPA
Base line
CEE+ MP
_+ CE MPA
FIGURE 34.2 Soluble E-selectin, vascular cell adhesion molecule (VCAM-1), intercellular adhesion molecule (ICAM-1) levels, and monocyte chemoattractant protein (MCP)-I levels before therapy (baseline) and following micronized progesterone (CEE +MP) or medroxyprogesterone acetate (CEE +MPA) combined with conjugated equine estrogen (CEE). Both therapies significantly decreased E-selectin, VCAM-1, ICAM-1, and MCP-1 levels from baseline values (p < 0.001, p = 0.016, p = 0.048, and p < 0.005 by A N O V A , respectively) (From ref. 47, with permission.)
474
KOH E T
TABLE 34.1 Effects of Estrogen on Inflammation
Effects C-reactive protein (CRP) Increases CRP in healthy women Decreases CRP in women Does not change CRP in healthy women Cell adhesion molecules (CAMs) Inhibits IL-l-induced expression of CAMS Lowers soluble CAMS in healthy women Monocyte chemoattractant protein-1 (MCP-1) Inhibits MCP-1 expression Lowers soluble MCP-1 in healthy women Tumor necrosis factor (TNF)-ot Inhibits TNF-o~ gene expression Lowers soluble TNF-cx in women Interleukin (IL)-6 Inhibits IL-6 expression Increases soluble IL-6 in healthy women Does not change IL-6 in healthy women
References 14-20 22",23" 16",21"
33,34 15-17, 25,26,45*,46,47
33,35,36 47-49
37 19,50 34,38 17,18 14,19
*Transdermal estrogen. Walsh et al. (19). The effects of estrogen therapy (ET) or H T on soluble IL-6 levels in postmenopausal women are inconsistent. Some studies have observed an increase oflL-6 levels (17,18), whereas our study and others observed no significant changes (14,19). In summary, the effect of H T on inflammation in postmenopausal women is variable. Oral H T increases the levels of CRP and, on the other hand, decreases levels of the soluble CAMs, MCP-1, and TNF-ot that may contribute to the risk of cardiovascular disease (Table 34.1). Although likely a first-pass effect of orally administered estrogen on the hepatic synthesis of CRP, elevated CRP could have deleterious effects on vascular inflammation, as discussed previously, and may contribute to the unexpected increase in myocardial infarction risk within the first year of treatment in the randomized and observational clinical studies.
III. EFFECTS OF HORMONE THERAPY ON HEMOSTASIS AND FIBRINOLYSIS MARKERS The coagulation and fibrinolytic systems are two separate, but interlinked, enzyme cascades that regulate the production and breakdown of fibrin. Overall pictures of the
AL.
coagulation and fibrinolysis schemes are depicted in Figs. 34.3 and 34.4, respectively (51). The release of tissue factor initiates the extrinsic pathway by activating factor VII. Antithrombin, proteins C and S, and the tissue factor pathway inhibitor are the most important inhibitors. There are a variety of laboratory assays to assess coagulatory activity and its major regulating factors. Prothrombin fragment 1+2 (FI+ 2) and the thrombin/antithrombin complex (TAT) are determined to evaluate the thrombin generation rate. Tissue factor antigen, factor VII activity, and F1 + 2 reflect ongoing coagulatory activity. Activity of the inhibitors of thrombin generation and thrombin activity, antithrombin III, protein C, and protein S, may also be measured. Fibrinolysis is achieved by the activation of plasminogen to plasmin by tissue-type plasminogen activator (t-PA), which is also balanced by a specific inhibitor, plasminogen activator inhibitor type I (PAI-1). Several assays are used to assess fibrinolytic activity and its major regulating factors. The plasmin-antiplasmin complex (P-AP) appears to be an indicator of the plasmin generation rate. t-PA and PAI-1 are measured by enzyme immunoassay. D-dimer, a degradation product of cross-linked fibrin by plasmin, is used to assess the net activity of the hemostatic system (51). Increased levels of fibrinogen (52), factor VII (53), or PAI-1 antigen (54) are associated with an increased risk of coronary heart disease. Following menopause, the levels of several coagulation factors increase, including factors VII and VIII in addition to fibrinogen, although the effects of aging, independent of estrogen status, also likely contribute to these changes (55,56). Furthermore, levels of a critical inhibitor of fibrinolysis, PAI-1 antigen, are higher in postmenopausal women than in premenopausal women, approaching PAI-1 levels observed in men of any age (57). However, no menopause-induced effect associated with the reaction products of thrombin activity has been observed in a longitudinal study (58). According to the PEPI trial, orally administered CEE 0.625 mg results in a small (<2%) but significant reduction in plasma fibrinogen levels in healthy postmenopausal women (59). Co-administration of MPA cyclic 5.0 mg or continuous 2.5 mg or MP cyclic 200 mg does not influence the CEE effect on fibrinogen. Small-sized clinical trials have reported less consistent results of transdermal estradiol 50 mg therapy; that is, either no change (60) or a decrease (61) in fibrinogen levels. Koh et al. (62) demonstrated that oral CEE 0.625 mg therapy for 1 month decreases PAI-1 antigen levels by 50% from pretreatment values, with the greatest reduction occurring in women with the highest baseline PAI-1 levels. In addition, MPA 2.5 mg combined with CEE showed a similar decrease in PAI-1 levels (Fig. 34.5). These findings are surprising because MPA stimulates PAI-1 release from bovine aortic (63) and human umbilical endothelial cells (64). In addition, the percent change in D-dimer levels exhibited a significant inverse
CHAPTER 34 The Effects of Hormone Therapy on Inflammatory, Hemostatic, and Fibrinolytic Markers in PostmenopausalWomen 475
FIGURE 34.3 A schematic diagram of some of the clinicallyimportant coagulationreactions. See explanations in text. (From res 51, with permission.)
FIGURE 34.4 A schematic diagram of the fibrinolyticpathway. See explanations in text. (From ref. 51, with permission.)
correlation with the percent change in PAI-1 levels, suggesting an enhanced potential for fibrinolysis (Fig. 34.6). Six months of HT with oral cyclic E2 combined with MP also increased global fibrinolytic capacity by 63% versus baseline and reduced both PAI-1 antigen and PAI activity in 45 healthy postmenopausal women (65). When administered orally with CEE, MP (cyclic 200 mg) reduced PM-1 levels to
a degree similar to that with MPA (cyclic 10 mg) (47). The effect of transdermal estradiol may differ as a function of duration of therapy. No change in PM-1 was observed after 1 month of a transdermal E2 100-mg patch (62), whereas a reduction was induced after 1 year with a 50-1xg patch (60). Factor VII levels increase with CEE, and MPA appears to attenuate this CEE effect (66). Transdermal estradiol therapy, however, does not affect factor VII (16,60). Differences in fibrinogen, PM-1, and factor VII responses by the route of estrogen administration strongly suggest first-pass effects of oral estrogen on hepatic synthesis of the thrombosis markers. There is evidence of the activation of coagulation system even with low estrogen doses used orally in HT. Caine et al. (67) showed that the administration of CEE 0.625 mg or 1.25 mg daily increased indices of thrombin generation (Fl12 and fibrinopeptide A) in a dose-dependent way, and CEE decreases levels of inhibitors of thrombin generation (antithrombin III and protein S). In this regard, observational studies have reported an increased risk of venous thromboembolism, or pulmonary embolism, in postmenopausal women treated with H T when compared with nonusers (68-70). Interestingly, an increasing daily dose is associated with increased risk of venous thromboembolism (69,71) or stroke (72). These observations raise the question as to whether potentiation of fibrinolysis could be a consequence of activation of the coagulation pathway as a primary response to estrogen administration. WinNer speculated that H T at conventional dosages might affect fibrinolytic activity to a greater extent than the coagulation activity, mainly by reducing PAI-1 antigen levels, whereas the converse may hold at higher estrogen doses (73). Koh and coworkers (74) observed that the increase in fibrinolytic po-
476
KOH E T
AL.
FIGURE 34.5 Changesin the plasma levels ofplasminogen activator inhibitor (PAI-1) before and after therapywith oral conjugated equine estrogen (CEE) 0.625 mg dailyfor 1 month, or a combination of CEE 0.625 mg with medroxyprogesterone acetate (MPA) 2.5 mg dailyfor 1 month by 30 healthy postmenopausalwomen. Mean values are identified by open circles. (From res 62, with permission.)
FIGURE34.6 Scatterplots with the predicted regression line showing the correlation between the percent change in PAI-1 antigen levels and the percent change in o-dimer levels after therapywith CEE alone and after CEE combined with MPA, each for I month in 30 healthypostmenopausal women. (From ref. 62, with permission.)
tential is independent of any effect on coagulation of CEE at conventional dosages. Other groups (75,76) also reported no correlation between fibrinolytic potential and coagulation activation using H T regimens. Cushman and colleagues (18) found that hemostasis markers and evidence of procoagulation are not correlated and that fibrinolytic potential was increased. These studies support the concept that the increase in the fibrinolytic potential with H T might be independent of the increase in coagulation with H T at conventional dosages in healthy postmenopausal women. However, activation of coagulation following ET or H T may not be balanced by activation of fibrinolysis in some postmenopausal women (4,62,77). Thus, ET or H T should not be initiated in women with coronary artery disease or the coexistence of other risk factors for hypercoagulability. Thrombogenic events are considered more likely in patients with certain heritable conditions, such as platelet antigen-2 polymorphisms (78). Homozygosity for factor V Leiden leads to enhanced arterial thrombosis and atherosclerosis in mice (79), and factor V Leiden mutation increases the risk of primary and recurrent venous thromboembolic events by
threefold to sixfold (80) as well as increases the risk of myocardial infarction (81). Indeed, E T and H T may decrease or increase atherothrombosis risk depending on the presence or absence of factor V Leiden mutation (82,83). Koh et al. (77) reported that CEE 0.625 mg combined with either M P 100 mg or MPA 2.5 mg both enhanced activation of coagulation, measured as tissue factor activity, F1 +2, and reduced antithrombin III, while not decreasing PAI-1 antigen levels in hypertensive and/or overweight postmenopausal women. H T may increase the thrombin formation by an enhanced activation of coagulation in these women. Koh et al. (27) also reported that compared with the conventional dose of CEE 0.625 mg combined with M P 100 mg, CEE 0.3 mg combined with M P 100 mg did not increase F1 +2 (Fig. 34.7) and decreased antithrombin III to a lesser extent, whereas PAI-1 was decreased by a similar degree. This observation is also in line with the concept of independent activation of fibrinolysis by CEE. Epidemiologic evidence also suggests a reduced thrombosis risk with lower oral doses of ET, although the data are not definitive.
CHAPTER 34 The Effects of Hormone Therapy on Inflammatory, Hemostatic, and Fibrinolytic Markers in Postmenopausal Women
FIGURE 34.7 Plasmaprothrombin fragment 1+2 (F1 +2) levels before treatment (baseline) and following micronizedprogesterone combinedwith conjugated equine estrogen 0.625 mg (C-HT) or 0.3 mg (L-HT) treatment. C-HT significantly increased F1+2 from baseline values (p < 0.001); however,L-HT did not significantlychange F1 +2 (p = 0.558). Of interest, the effects of C-HT and L-HT on Fl12 were significantlydifferent (p < 0.001). Median values are identified by open circles. (From ref. 27, with permission.)
IV. TIBOLONE Tibolone, a synthetic steroid with estrogenic, androgenic, and progestogenic properties, has long been used in Europe to relieve climacteric symptoms and prevent postmenopausal bone loss. The impact of tibolone on cardiovascular disease risk, however, remains unclear. Although it lowers high-density lipoprotein (HDL) cholesterol, tibolone improves the overall lipid profile and the flow-mediated brachial artery dilator response to hyperemia. The lowering of H D L cholesterol in postmenopausal women by tibolone, however, is not associated with changes in cholesterol elflux capacity or paraoxonase activity. In this study, tibolone significantly decreased circulating H D L cholesterol levels and apolipoprotein A - l ; however, these changes were not associated with changes in the activity of plasma to release 3H-cholesterol from radiolabeled fibroblasts or in the serum activity of the antioxidative enzyme paraoxonase/ arylesterase (84). In other words, the anti-atherogenic activities of H D L appears to remain unchanged despite decreased H D L cholesterol. Koh et al. (85) investigated the effects of tibolone on markers of inflammation and thrombosis. Compared with C E E 0.625 mg combined with M P 100 mg, tibolone showed different effects on CRP, antithrombin III, and F1+2. CRP and antithrombin III did not change (Fig. 34.8), and F1 +2 increased to a smaller degree. Meanwhile, tibolone did not affect fibrinogen as occurs with HT, although PAI-1 antigen decreased to a similar degree. In another study, when compared with C E E 0.3 mg combined
477
FIGURE 34.8 High sensitivityC-reactive protein (hsCRP) levels before treatment (baseline) and following HT (CEE+MP) and tibolone. HT significantly increased hsCRP from baseline values (p < 0.001); however, tibolone did not significantly change hsCRP (p = 0.425). Of interest, the effects of HT and tibolone on hsCRP were significantly different (p = 0.015). Median values are identified by open circles. (From ref. 84, with permission.)
with M P 100 mg [lower-dose (L)-HT], tibolone significantly reduced total cholesterol, triglyceride, H D L cholesterol levels, and triglyceride/HDL cholesterol ratios but not the total cholesterol/HDL cholesterol ratios (86). Tibolone also improved the flow-mediated response to hyperemia from baseline values by a magnitude similar as L-HT. Neither L - H T nor tibolone increased CRE L - H T reduced antithrombin III from baseline values, while tibolone showed no changes. In contrast, tibolone increased F 1 + 2 from baseline values, while L - H T showed no changes. Tibolone significantly reduced PAI-1 antigen levels from baseline values, compared with L-HT, which showed no changes. Tibolone also decreased factor VII activity (87). An in vitro study also reported that tibolone suppressed expression of VCAM-1 and E-selectin in human vascular endothelial cells (88). Taken together, these findings suggest a small activation of coagulation, independent activation of fibrinolysis, and an anti-inflammatory effect with tibolone.
V. CLINICAL IMPLICATIONS AND FUTURE PROSPECTS The fact that ET and H T do not confer cardio-protective effects in the recent randomized controlled trials can be assimilated readily according to the "healthy endothelium" concept (1,89-91). In short, the favorable vascular effects of estrogen on atherosclerosis, inflammation, hemostasis, and coronary flow reserve are dependent on the integrity of both the endothelium and estrogen receptor populations in endo-
478 thelial cells and vascular smooth muscle cells (92,93), and these conditions were probably not met by most women in these trials because of their advanced age, multiple risk factors, and existent coronary atherosclerosis. Optimization of estrogen's cardio-protective properties may depend on the presence of a healthy endothelium. Miller et al. (34) demonstrated that E2 inhibits expression of adhesion molecules (P-selectin, ICAM-1, VACM-1), MCP-1, and proinflammatory cytokines (IL-1, IL-6) in rat carotid arteries after balloon injury, thus attenuating the stimulus for leukocyte entry and negatively modulating the injury response. Q.uite interestingly, in the same model, this group observed that aged ovariectomized rats lose the vasculo-protective and anti-inflammatory responses to exogenous E2 seen in injured arteries of younger ovariectomized rats (94). In healthy postmenopausal women, CEE 0.625 mg reduced E-selectin, ICAM-1, and VACM-1 levels (46). However, CEE 0.625 mg did not reduce E-selectin, ICAM-1, and VACM-1 levels in type 2 diabetic postmenopausal women (95). L-HT might prevent coronary heart disease without increasing stroke or thromboembolism risk (69,71,72,96). Beneficial effects of CEE 0.3 mg combined with MP 100 mg daily on vasomotor function, inflammation, and hemostasis may pave the way for a new clinical trial (27). Indeed, the Kronos Early Estrogen Prevention Study (KEEPS) was launched in 2005 (97). KEEPS is a 5-year clinical trial that will evaluate the effectiveness of CEE 0.45 mg, transdermal estradiol 50 lag (both in combination with cyclic MP 200 mg), and placebo in preventing progression of carotid atherosclerosis and coronary artery calcification in early postmenopausal women. The Early versus Late Intervention Trial with Estradiol (ELITE) study will compare the effects of estradiol in both early and late postmenopausal women. These studies will answer the question as to whether lower doses of HT affect the healthy endothelium.
In conclusion, HT has both anti-inflammatory and proinflammatory effects and both activates coagulation and improves fibrinolysis. These effects depend on the route of administration, doses of estrogen, the age of the woman, and the presence or absence of coronary artery disease or the coexistence of other risk factors for hypercoagulability. Accordingly, the most effective approach may be to use HT immediately postmenopause (i.e., initiating HT at early menopause). As alternatives, lower doses of HT or tibolone might prevent coronary heart disease without increasing stroke or thromboembolism risk. However, clinical recommendations regarding the effects of lower doses of HT or tibolone on cardiovascular outcomes must await the performance of additional studies with clinical end points.
KOH ET AL.
References 1. Koh KK. Effects of estrogen on the vascular wall: vasomotor function and inflammation. Cardiovasc Res 2002;55:714-726. 2. Ross R. Atherosclerosis-an inflammatory disease. N Engl J Med 1999;340:115-126. 3. Koh KK. Effects of statins on vascular wall: vasomotor function, inflammation, and plaque stability. CardiovascRes 2000;47:648-657. 4. Koh KK, Home MKI, Cannon RO III. Effects of hormone replacement therapy on coagulation, fibrinolysis, and thrombosis risk in postmenopausal women. Thromb Haemost 1999;82:626-633. 5. Harnish DC, Scicchitano MS, Adelman SJ, Lyttle CR, Karathanasis SK. The role of CBP in estrogen receptor cross-talk with nuclear factorkB in HepG2 cells. Endocrinology 2000;141:3403-3411. 6. Visser M, Bouter LM, McQ.uillan GM, Wener MH, Harris TB. Elevated C-reactive protein levels in overweight and obese adults. J A m MedAssoc 1999;282:2131-2135. 7. Woloshin S, Schwartz LM. Distribution of C-reactive protein values in the United States. NEnglJMed2005;352:1611-1613. 8. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N EnglJ Med 2005;352:1685-1695. 9. Cioffi M, Esposito K, Vietri M, et al. Cytoldne pattern in postmenopause. Maturitas 2002;41:187-192. 10. Cantatore F, Loverro G, Ingrosso A, et al. Effect of oestrogen replacement on bone metabolism and cytokines in surgical menopause. Clin Rheumato11995;14:157-160. 11. Sites CK, Toth MJ, Cushman M, et al. Menopause-related differences in inflammation markers and their relationship to body fat distribution and insulin-stimulated glucose disposal. Fertil Steri12002;77:128-135. 12. Ridker PM, Hennekens CH, Buring JE, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N EnglJ Med 2002;342:836-843. 13. Pal JK, Pischon T, Ma J, et al. Inflammatory markers and the risk of coronary heart disease in men and women. N EnglJ Med 2004;351: 2599-2610. 14. Koh KK, Schenke WH, Waclawiw M, Csako G, Cannon RO III. Statin attenuates increase in C-reactive protein during estrogen replacement therapy in postmenopausal women. Circulation 2002;105: 1531-1533. 15. Cushman M, Legault C, Barrett-Connor E, et al. Effect of postmenopausal hormones on inflammation-sensitive proteins. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Study. Circulation 1999;100:717-722. 16. Vehkavaara S, Silveira A, Hakala-Ala-Pietila T, et al. Effects of oral and transdermal estrogen replacement therapy on markers of coagulation, fibrinolysis, inflammation and serum lipids and lipoproteins in postmenopausal women. Thromb Haemost 2001;85:619-625. 17. Herrington DM, Brosnihan KB, Pusser BE, et al. Differential effects of estrogen and droloxifene on C-reactive protein and other markers of inflammation in healthy postmenopausal women. J Clin Endocrinol Metab 2001;86:4216-4222. 18. Cushman M, Meilahn EN, Psaty BM, et al. Hormone replacement therapy, inflammation, and hemostasis in elderly women. Arterioscler Thromb VascBio11999;19:893-899. 19. Walsh BW, Cox DA, Sashegyi A, et al. Role of tumor necrosis factorcx and interleukin-6 in the effects of hormone replacement therapy and raloxifene on C-reactive protein in postmenopausal women. Am J Cardio12001;88:825-828. 20. Silvestri A, Gebara O, Vitale C, et al. Increased levels of C-reactive protein after oral hormone replacement therapy may not be related to an increased inflammatory response. Circulation 2003;107:3165-3169. 21. Decensi A, Omodei U, Robertson C, et al. Effect of transdermal estradiol and oral conjugated estrogen on C-reactive protein in retinoidplacebo trial in healthy women. Circulation 2002;106:1224-1228.
CHAPTER 34 The Effects of Hormone Therapy on Inflammatory, Hemostatic, and Fibrinolytic Markers in Postmenopausal Women 22. Sattar N, Perera M, Small M, Lumsden MA. Hormone replacement therapy and sensitive C-reactive protein concentrations in women with type-2 diabetes. Lancet 1999;35;4:487-488. 23. Modena MG, Bursi F, Fantini G, et al. Effects of hormone replacement therapy on C-reactive protein levels in healthy postmenopausal women: comparison between oral and transdermal administration of estrogen. Am J Med 2002;113:331-314. 24. Blum CA, Mfiller B, Huber P, et al. Low-grade inflammation and estimates of insulin resistance during the menstrual cycle in lean and overweight women. J Clin Endocrinol Metab 2005;90:3230-3235. 25. Wakatsuki A, Okatani Y, Ikenoue N, Fukaya T. Effect of medroxyprogesterone acetate on vascular inflammatory markers in postmenopausal women receiving estrogen. Circulation 2002;105:1436-1439. 26. Lamon-Fava S, Posfai B, Schaefer EJ. Effect of hormonal replacement therapy on C-reactive protein and cell-adhesion molecules in postmenopausal women. Am J Cardio12003;91:252-254. 27. Koh KK, Shin MS, Sakuma I, et al. Effects of conventional or lower doses of hormone replacement therapy in postmenopausal women. Arterioscler Thromb VascBid 2004;24:1516-1521. 28. Szmitko PE, Wang CH, Weisel RD, et al. Biomarkers of vascular disease linking inflammation to endothelial activation: Part II. Circulation 2003;108:2041-2048. 29. Esmon CT. Crosstalk between inflammation and thrombosis. Maturitas 2004;47:305-314. 30. Taylor KE, Giddings JC, van den Berg CW. C-reactive proteininduced in vitro endothelial cell activation is an artefact caused by azide and lipopolysaccharide. Arterioscler Thromb Vasc Bid 2005;25:12251230. 31. Pradhan AD, Manson JE, Roussouw JE, et al. Inflammatory biomarkers, hormone replacement therapy, and incident coronary heart disease. Prospective analysis from the Women's Health Initiative observational study. J A m Med Assoc 2002;288:980-987. 32. Albert MA, Glynn RJ, Ridker PM. Plasma concentration of C-reactive protein and the calculated Framingham Coronary Heart Disease Risk Score. Circulation 2003;108:161-165. 33. Caulin-Glaser T, Watson CA, Pardi R, Bender JR. Effects of 1713estradiol on cytokine-induced endothelial cell adhesion molecule expression.J Clin Invest 1996;98:36-42. 34. Miller AP, Feng W, Xing D, et al. Estrogen modulates inflammatory mediator expression and neutrophil chemotaxis in injured arteries. Circulation 2004; 110:1664-1669. 35. Frazier-Jessen MR, Kovacs EJ. Estrogen modulation of JE/monocyte chemoattractant protein-1 mRNA expression in murine macrophage. J Immuno11995;154:1838-1845.
36. Pervin S, Singh R, Rosenfeld ME, et al. Estradiol suppresses MCP-1 expression in vivo: implications for atherosclerosis..drterioscler Thromb Vasc Bid 1998;18:1575-1582. 37. Srivastava S, Weitzmann MN, Cenci S, et al. Estrogen decreases TNF gene expression by blocking JNK activity and the resulting production of c-Jun and JunD. J Clin Invest 1999;104:503-513. 38. Sukovich DA, Kauser K, Shirley FD, et al. Expression of interleukin-6 in atherosclerotic lesions of male apoE-knockout mice. Arterioscler Thromb VascBid 1998;18:1498-1505. 39. Myers GL, Rifai N, Tracy RP, et al. CDC/AHA workshop on markers of inflammation and cardiovascular disease. Application to clinical and public health practice report from the laboratory science discussion group. Circulation 2004;110:e545-e549. 40. Nakai K, Itoh C, Kawazoe K, et al. Concentration of soluble vascular cell adhesion molecule-1 (VCAM-1) correlated with expression of VCAM-1 mRNA in the human atherosclerotic aorta. CoronArtery Dis 1995;6:497-502. 41. Hwang SJ, Ballantyne CM, Sharrett AR, et al. Circulating adhesion molecules VCAM-1, ICAM-1, and E-selectin in carotid atherosclerosis and incident coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) study. Circulation 1997;96:4219-4225.
479
42. Ridker PM, Hennekens CH, Roitman-Johnson B, Stampfer M, Allen J. Plasma concentration of soluble intercellular adhesion molecule 1 and risks of future myocardial infarction in apparently healthy men. Lancet 1998;351:88-92. 43. Cipollone F, Marini M, Fazia M, et al. Elevated circulating levels of monocyte chemoattractant protein-1 in patients with restenosis after coronary angioplasty. Arterioscler Thromb VascBid 2001;21:327-334. 44. Aukrust P, Berge RK, Ueland T, et al. Interaction between chemokines and oxidative stress: possible pathogenic role in acute coronary syndromes. J A m Coll Cardio12001;37:485-491. 45. Koh KK, Bui MN, Mincemoyer R, Cannon RO III. Effects of hormone therapy on inflammatory cell adhesion molecules in postmenopausal healthy women. Am J Cardio11997;80:1505-1507. 46. Koh KK, Blum A, Hathaway L, et al. Vascular effects of estrogen and vitamin E therapies in postmenopausal women. Circulation 1999;100: 1851-1857. 47. Koh KK, Jin DK, Yang SH, et al. Vascular effects of synthetic or natural progestagen combined with conjugated equine estrogen in healthy postmenopausal women. Circulation 2001;103:1961-1966. 48. Stork S, Baumann K, yon Schacky C, Angerer P. The effect of 17 betaestradiol on MCP-1 serum levels in postmenopausal women. Cardiovasc Res 2002;53:642-649. 49. Koh KK, Son JW, Jin DK, et al. Effect of hormone replacement therapy on nitric oxide bioactivity and monocyte chemoattractant protein1 levels. I n t J Cardio12001;81:43-50. 50. Koh KK, Ahn JY, Jin DK, et al. Effects of continuous combined hormone replacement therapy on inflammation in hypertensive and/or overweight postmenopausal women. Arteriosder Thromb Vase Biol 2002;22:1459-1464. 51. Koh KK. Effects ofHMG-CoA reductase inhibitor on hemostasis. Int J Cardio12000;76:23-32. 52. Kannel WB, Wolf PA, Castelli WP, D'Agostino RB. Fibrinogen and risk of cardiovascular disease. The Framingham Study.J A m MedAssoc
1987;258:1183-1186. 53. Heinrich J, Balleisen L, Schulte H, Assmann G, van de Loo J. Fibrinogen and factor VII in the prediction of coronary risk: results from the PROCAM study in healthy men. drterioscler Thromb 1994;14: 54-59. 54. Hamsten A, de Faire U, Walldius G, et al. Plasminogen activator inhibitor in plasma; risk factor for recurrent myocardial infarction. Lancet 1987;2:3-9. 55. Meade TW, Imerson JD, Haines AP, Stirring Y, Thompson SG. Menopausal status and haemostatic variable. Lancet 1983;1:22-24. 56. Folsom AR, Wu KK, Davis CE, et al. Population correlates of plasma fibrinogen and factor VII, putative cardiovascular risk factors. Atherosclerosis 1991;91:191-205. 57. Gebara OCE, Mittleman MA, Sutherland P, et al. Association between increased estrogen status and increased fibrinolytic potential in the Framingham Offspring Study. Circulation 1995;91:1952-1958. 58. Pinto S, Rostagno C, Coppo M, et al. No signs of increased thrombin generation in menopause. Thromb Res 1990;58:645-651. 59. The writing group for the PEPI trial. Effects of estrogen or estrogen/ progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) trial. J A m MedAssoc 1995;273:199-208. 60. Post MS, van der Mooren MJ, van Baal WM, et al. Effects of low-dose and transdermal estrogen replacement therapy on hemostatic factors in healthy postmenopausal women: a randomized placebo-controlled study. Am J Obstet Gyneco12003;189:1221-1227. 61. Zegura B, Keber I, Sebestjen M, Koenig W. Double blind randomized study of estradiol replacement therapy on markers of inflammation, coagulation and fibrinolysis. Atherosderosis 2003;168:123-129. 62. Koh KK, Mincemoyer R, Bui MN, et al. Effects of hormone-replacement therapy on fibrinolysis in postmenopausal women. N EnglJ Med 1997; 336:683-690.
480 63. Blei F, Wilson EL, Mignatti P, Rifldn DB. Mechanism of action of angiostatic steroids: suppression of plasminogen activator activity via stimulation of plasminogen activator inhibitor synthesis. J Cell Physiol 1993;155:568-578. 64. Takada A, Takada Y, Urano T. The physiological aspects of fibrinolysis. Thromb Res 1994;76:1-31. 65. Scarabin P-Y, Mhenc-Gelas M, Plu-Bureau G, et al. Effects of oral and transdermal estrogen/progesterone regimens on blood coagulation and fibrinolysis in postmenopausal women. A randomized controlled trial. Arterioscler Tbromb VascBiol 1997;17:3071-3078. 66. Lobo RA, Pickar JH, Wild RA, Walsh B, Hirvonen E. Metabolic impact of adding medroxyprogesterone acetate to conjugated estrogen therapy in postmenopausal women. Obstet Gyneco11994;84:987-995. 67. Caine YG, Bauer KA, Barzegar S, et al. Coagulation activation following estrogen administration to postmenopausal women. Thromb Haemost 1992;68:392-395. 68. Daly E, Vessey MP, Hawkins MM, et al. Risk of venous thromboembolism in users of hormone replacement therapy. Lancet 1996;348: 977-980. 69. Jick H, Derby LE, Myers MW, Vasilakis C, Newton ICM. Risk of hospital admission for idiopathic venous thromboembolism among users of postmenopausal oestrogens. Lancet 1996;348:981-983. 70. Grodstein F, Stampfer MJ, Goldhaber SZ, et al. Prospective study of exogenous hormones and risk of pulmonary embolism in women. Lancet 1996;348:983-987. 71. Smith NL, Heckbert SR, Lemaitre RN, et al. Esterified estrogens and conjugated equine estrogens and the risk of venous thrombosis. J A m MedAssoc 2004;292:1581-1587. 72. Grodstein F, Manson JE, Colditz GA, et al. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease.Ann Intern IVied 2000;133:933-941. 73. Winkler UH. Menopause, hormone replacement therapy and cardiovascular disease: a review of haemostaseological findings. Fibrinolysis 1992;6($3):5-10. 74. Koh KK, Home MK III, Csako G, Waclawiw M, Cannon RO III. Relation of fibrinolytic potentiation by estrogen to coagulation pathway activation in postmenopausal women. Am J Cardio11999;83:466-469. 75. Scarabin P-Y, Alhenc-Gelas M, Plu-Bureau G, et al. Effects of oral and transdermal estrogen/progesterone regimens on blood coagulation and fibrinolysis in postmenopausal women; a randomized controlled trial. Arterioscler Tbromb VascBio11997;17:3071-3078. 76. Teede HJ, McGrath BP, Smolich JJ, et al. Postmenopausal hormone replacement therapy increases coagulation activity and fibrinolysis. Arterioscler Thromb VascBio12000;20:1404-1409. 77. Koh KK, Ahn JY, Kim DS, et al. Effect of hormone replacement therapy on tissue factor activity, C-reactive protein, and the tissue factor pathway inhibitor. Am J Cardio12003;91:371-373. 78. Weiss EJ, Bray PF, Tayback M, et al. A polymorphism of a platelet glycoprotein receptor as an inherited risk factor for coronary thrombosis. N E n g l J M e d 1996;334:1090-1094. 79. Eitzman DT, Westrick RJ, Shen Y, et al. Homozygosity for factor V Leiden leads to enhanced thrombosis and atherosclerosis in mice. Circulation 2005;111:1822-1825. 80. Price DT, Ridker PM. Factor V Leiden mutation and the risks for thromboembolic disease: a clinical perspective. Ann Intern Med 1997;127:895-903.
KoI-I E~" hi..
81. Rosendaal FR, Siscovick DS, Schwartz SM, et al. Factor V Leiden (resistance to activated protein C) increases the risk of myocardial infarction in young women. Blood 1997;89:2817-2821. 82. Glueck CJ, Wang P, Fontaine RN, et al. Effect of exogenous estrogen on atherothrombotic vascular disease risk related to the presence or absence of the Factor V Leiden mutation (resistance to activated protein C ) . A m J Cardio11999;84:549-554. 83. Herrington DM, Vittinghoff E, Howard TD, et al. Factor V Leiden, hormone replacement therapy, and risk of venous thromboembolic events in women with coronary disease. Arterioscler Thromb Vasc Biol 2002;22:1012-1017. 84. von Eckardstein A, Schmiddem K, Hovels A, et al. Lowering of H D L cholesterol in post-menopausal women by tibolone is not associated with changes in cholesterol efflux capacity or paraoxonase activity. Atberosclerosis 2001;159:433-439. 85. Koh KK, Ahn JY, Jin DK, et al. Significant differential effects of hormone therapy or tibolone on markers of cardiovascular disease in postmenopausal women. Arterioscler Thromb VascBio12003 ;23:1889-1894. 86. Koh KK, Han SH, Shin M-S, et al. Significant differential effects of lower doses of hormone therapy or tibolone on markers of cardiovascular disease in postmenopausal women: a randomized, double-blind, crossover study. Eur HeartJ 2005;26:1362-1368. 87. Winkler UH, Altkemper R, Kwee B, Helmond FA, Coelingh Bennink HJT. Effects of tibolone and continuous combined hormone replacement therapy on parameters in the clotting cascade: a multicenter, double-blind, randomized study. Fertil Steri12000;74:10-19. 88. Simoncini T, Genazzani AR. Tibolone inhibits leukocyte adhesion molecule expression in human endothelial cells. Mol Cell Endocrinol 2000;162:87-94. 89. Koh KK. Can a healthy endothelium influence the cardiovascular effects of hormone replacement therapy? IntJ Cardio12003;87:1-8. 90. Mendelsohn ME, Karas RH. The time has come to stop letting the HERS tale wag the dogma. Circulation 2001;104:2256-2259. 91. Koh KK, Sakuma I. Should progestins be blamed for the failure of hormone replacement therapy to reduce cardiovascular events in randomized controlled trials? Arterioscler Thromb VascBio12004;24 :1171-1179. 92. Mendelsohn ME, Karas RH. The protective effects of estrogen on the cardiovascular system. N EnglJ Med 1999;340:1801-1811. 93. Mendelsohn ME, Karas RH. Molecular and cellular basis of cardiovascular gender differences. Science 2005;308:1583-1587. 94. Miller AP, Xing D, Feng W, et al. Aged rats lose estrogen-induced vasoprotective and anti-inflammatory effects in injured arteries. Circulation 2004;110:1664-1669. 95. Koh KK, Kang MH, Jin DK, et al. Vascular effects of estrogen in type II diabetic postmenopausal women. J Am Coil Cardiol 2001;38: 1409-1415. 96. Ferrara A, Q.gesenberry CP, Karter AJ, et al. Current use of unopposed estrogen and estrogen plus progestin and the risk of acute myocardial infarction among women with diabetes: the Northern California Kaiser Permanente Diabetes Registry, 1995-1998. Circulation 2003;107:43-48. 97. Harman SM, Brinton EA, Cedars M, et al. KEEPS: The Kronos Early Estrogen Prevention Study. Climacteric 2005;8:3-12.
~ H A P T E R 3~
Hormone Therapy and Hemostasis MORRIS
NOTELOVITZ
Adult Women's Health & Medicine, Boca Raton, FL 33496
I. H E M O S T A S I S REVISITED
The effect of hormones on the coagulation system has been a matter of controversy for many years. Until 1996, the conventional wisdom was that hormone therapy (HT) after menopause was not associated with a significant risk of thrombosis. To a large degree, this was presumed to be because the doses used in H T are many times lower than those used in oral contraceptives (OCs) (1), where it was known that there is a twofold to threefold increase in the risk of thrombosis. It appears clear now, that with standard-dose oral HT, there is a similar increased risk (approximately 2-3 fold), as reviewed in Chapter 36. This increased risk was first consistently reported in a series of different epidemiologic studies in 1996 (2-4) and then reported in a randomized trial of H T and placebo in postmenopausal women with coronary disease (Heart and Estrogen/ Progestin Replacement Study [HERS]) (5). It is probable that with aging and the lifelong increase in the extent of atherosclerosis and vascular damage, smaller hormonal influences (HT rather than OC use) can lead to a greater probability of thrombosis. Data from the Women's Health Initiative (WHI) provide the largest database on the risks of HT and estrogen therapy (ET) (6,7), at least for the doses and regimens studied, and these data will be reviewed here. This chapter provides the basic information involved in coagulation and fibrinolysis, which was also discussed again in Chapters 34 and 36. It is clear that age, the type of ET and HT, and specifically the dose and the route of administration all influence the balance of factors reviewed later, and accordingly influence the level of risk. Inherited, albeit subtle thrombophilias, previously unrecognized, also substantially modify this risk. Finally, it must be accepted that there is not a good correlation between various changes in circulatory markers of the coagulation and fibrinolytic systems and the absolute risk of thrombosis. TREATMENT OF THE POSTMENOPAUSAL W O M A N
A review of hemostasis (8), while endorsing previous in vitro studies, notes variations in the sequential reactions that lead to blood clot formation in vivo. The process is very complex and has numerous coagulatory and fibrinolytic checks and balances. Fig. 35.1 depicts a simplified summary and will serve as a guide to some potential points of clinical relevance when prescribing HT. Although the basic process is the same, arterial thrombosis differs from venous thrombosis: the former results in plateletrich ("white") thrombi, whereas the low-blood-flow venous thrombosis is fibrin and red cell-rich ("red") and, as a consequence, is more liable to embolism (9). More importantly, thrombus formation is a key event in the origin and progression of atherosclerosis and, as will be discussed, may help to define the clinician's approach to cardiovascular health care in women.
II. VESSEL WALL The endothelium, which is a semipermeable barrier between the blood and the deeper layers of the vessel wall, serves two main functions in hemostasis: it helps to regulate vascular tone, and it acts as an anticoagulant. The endothelium synthesizes vasodilators such as prostacyclin (PGI2) and endothelium-derived relaxing factor (nitric oxide [NO]), as well as vasoconstrictors such as endothelin and plateletactivating factor (PAF) (see Fig. 35.1) (10). Prostacyclin and NO act synergistically to prevent platelet activation. Both factors are difficult to assay because they are either only locally active (NO) or are rapidly metabolized (PGI2). Nevertheless, two studies have shown that nonoral estrogen in481
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
482
MORRISNOTELOVITZ Endothelial damage Exposure of subendotheliurn,Vascular Tone
-- Platelet Factor
1"
Intact Endothellum
Proatacyclln 12--
]
TissUe '
Adherenceto
] ~X/~X/~ I ~ - -. / IX~.bl~ ~,r~lrem
F~;?r I ICl~ Vlla ,---t ..... T............ ~ Anti.thrombln IIII I- i [ ] r......... I ProteinC ]
I Vllt-e'Wlla .... :
[
,.o;o:x: ..... _ .
I Phosphollpld, + Xa + Va
Endothelln -~, -
.1
.J
Plate.t fmctor 4
............
Anticoagulation
E ,r!n.i,
Plitele: Intlmal defect
etmr
Coagulation Cascade----------~
i1,,,1 .1o
:
.o+--.Vasospasm t
"-J ....
>A ..
"">AaTrence
/
l ]
,,.,o,o,,,u,
| F,bn..,..
..........
,Inhibitory;
i I .
.
.
.
.
I i
I
~
"...............
[
Fl~rlnp~
I Protein S
I
,--~--"
.......
ot'el C:
.
I Clot dissolving:
[
:
Flblnoly$1s
I
...................
]
Activators ...... :! PAl-1
..... Inhib•ors I
FIDrlrr~ ........... k .... ]_~-Plasmlr~--ii FDP
.
.
'=Flb~lnmTomerl][l___-_
[Veno,,.. th~.ombosll-_j- I
[Arterial thrombosis
.
I Thrombln + Thrombomodulln I : ~ ~1 I~ ~ I i Protrombln-~Thrombln ............... t ..... t--"Plasmlnogen
I
~ l
:
I .
.
i
:--O;-antiplasmln i-.a,-antitrypsln (~' -macroglobulln
:--Endothelial flblnolyslns
,stimulatory
FIGURE 35.1 The integrated relationship between four systems (vascular tone, platelet activity, coagulation, and fibrinolysis) is depicted in separate panels showing their interrelationships, both stimulatory (solid lines) and inhibitory (dashed lines). ADP, adenosine diphosphate; EDRF, endothelium derived relaxing factor; FDP, fibrinogen-degradation products; PAI-1, plasminogen activator inhibitor-1.
creases plasma levels of both nitric oxide and prostacyclin (11,12). The serum estradiol level ranged between 80 and 120 pg/mL. A further study involving oral estradiol and norethindrone reported a 21% increased production in prostacyclin but no change in the potent vasoconstrictor, endothelial release (13). The endothelium also inactivates other known vasoactive substances such as serotonin and bradykinin (14). Blood flow--a major determinant ofintimal damage--depends on the balance among these various vasoactive substances, many of which are influenced by HT. The healthy endothelium prevents thrombosis in at least three ways. It synthesizes thrombomodulin, a protein that is an endothelial receptor for thrombin. The resulting complex activates protein C (see later), the primary inhibitor of factor Va and VIIIa (15); the intact endothelium is a surface anticoagulant. The cells consist of glycoproteins and proteoglycans such as heparin sulfate, the main cofactor to antithrombin III (the most potent endogenous anticoagulant). The glycoproteins emit a negative charge that propels platelets and prevents initiation of the intrinsic arm of the coagulation cascade (16); tissue plasminogen activator (t-PA) is continuously synthesized and secreted by endothelial cells and as such regulates the fibrinolytic activity of blood (17). This function may be a major contributor to the prevention of coronary artery disease (CAD). The activity of t-PA is enhanced by progestins (see later discussion).
III. PLATELET FACTOR Platelets circulate in blood in an inactive resting form. When they are exposed to the subendothelial layers, platelets develop pseudopodia, attach to the injured surface, and become highly adherent (Fig. 35.2) (8). The binding is determined by various surface receptors that complex with specific ligands such as yon WiUebrand factor. The bound platelets degranulate and expose receptors for factor VIII, factor V, and P-selectin (18). Neutrophils and monocytes bind to the P-selectin receptor and participate in the inflammatory response; in addition, the monocytes secrete tissue factor, the initiator of the extrinsic coagulation cascade (see Fig. 35.1). Platelet aggregation is enhanced by modifying and sensitizing their surface receptors to adhesive proteins, such as fibrinogen. The platelet membrane also releases arachidonic acid. Arachidonic acid is metabolized to thromboxane A2, which in turn promotes further platelet aggregation and local vasoconstriction. The net result is a platelet plug with fibrin formation, the first necessary step to permanent hemostasis and wound healing (9). Aspirin inhibits cyclo-oxygenase activity in both platelets (thereby reducing thromboxane activity), as well as endothelial prostacyclin synthesis. However, the process is more prolonged in platelets (+2 days) and is short rived in the endothelium (+_6 hours). It is for this reason that low-dose intermittent aspirin use optimizes its anti-
CHAPTER 35 Hormone Therapy and Hemostasis
483
FIGURE 35.2 Depictionof the initiation of blood coagulationand formation of the platelet plug within a blood vessel,involvingboth thrombogenesis
and inflammation. thrombotic potential (19). Two studies have confirmed that H T involving both combination oral treatment (conjugated equine estrogen [CEE] with medroxyprogesterone acetate [MPA]) and 1713-estradiol and norethindrone acetate and transdermal estrogen (1713-estradiol plus MPA) decreased platelet aggregation (20) and the cellular reactivity of platelets and monocytes (21). A positive effect of ET on platelet activity was the reported decrease in thromboxane metabolism, subsequent to both acutely administered intravenous estradiol and long-term estrogen replacement therapy (ET) with estradiol pellets (12). Although several reports, as noted previously, have shown a decrease in platelet aggregation with ET/HT, there has been the suggestion that the reverse is true with more platelet activation/aggregation. This may relate in part to the populations studied, with women with cardiovascular illnesses having altered platelet activities. Indeed even here the net long-term effect appears to be no overall change in platelet activation/aggregation (21a).
IV. COAGULATIONANTICOAGULATION The coagulation system has built-in checks and balances that preclude (in healthy individuals) inappropriate thrombosis and conserve the ability to help breach disruptions in the normal vasculature. This is achieved by three mechanisms. First, the procoagulant factors circulate in an inactive form; activation, and hence coagulation, only occurs at the site(s) of trauma. The process is thus localized to a relatively small area of intimal damage; second, the activated coagulants are inhibited and their activity modulated by naturally occurring anticoagulants. Third, both coagulants and anticoagulants are present in concentrations far in excess of the physiologic need. As a consequence, when statistically sig-
COAGULATION
I ANTI-COAGULATION [ RegulatoryProteins
Pro-Enzymes Factor VII* FactorIX FactorXl FactorX
Protein C *+ Protein S
Prothrombin 9 Pro.Cofactors Factor VIII FactorV TissueFactor*
Anti-ThrombinII1" ~
Pro-Cofactors Thrombomodulin Heparin
FIGURE35.3 Balancebetweenfactors involvedin coagulationand anticoagulation. *, systemic tests of possible clinical value; 9 best prognostic indicator (fragment 1.2); prothrombin2021A; +, factorV Leiden.
nificant alterations in plasma levels of activated zymogens occur, for example, in response to HT, they usually have little or no clinical impact (22,23). Factors involved in blood coagulation are summarized in Fig. 35.1. Based on in vitro studies (8), activation of factor X to Xa links the intrinsic and extrinsic pathways of the coagulation cascade and leads to the conversion of prothrombin (a vitamin K-dependent hepatic protein) to thrombin, which is the catalyst for the fibrinogen-fibrin reaction (see Fig. 35.1). The partial thromboplastin time (PTT) is a dynamic test that assesses the interrelated response of the intrinsic system; the prothrombin time (PT) is a measure of the activity of the extrinsic arm of the coagulation cascade. It has been noted that patients with hereditary factor XII deficiency, for example, have increased P T T but no bleeding diathesis, indicating that this factor is not important for the formation of a blood clot. Thus, the P T T is a test of dubious value (8). Tissue factor (a normal constituent of the surface of nonvascular cells) and stimulated monocytes activate coagulation. The physiologic
484
FIGURE 35.4 Pathway of coagulation involving factors from tissue and the cascade toward fibrin formation.
pathway of blood coagulation appears to bypass the first steps of the intrinsic pathway and is summarized in Fig. 35.4. Coagulation hinges on the formation of a tissue factoractivated VII-a complex that in turn converts factor IX to IXa and factor X to Xa (24). Activated factors VIIIa and IXa further catalyze the X-to-Xa conversion. Activated factor Xa, together with activated factor Va, mediates the conversion of prothrombin to thrombin in the presence of phospholipids and calcium (8). Thrombin, when free in solution, promotes clotting by converting fibrinogen to fibrin, by activating platelets, and, by positive feedback, stimulating the activation of factors XI, VIII, and V (see Fig. 35.4). By contrast, thrombin has a local anticoagulant effect (10,25). When thrombin forms a complex with thrombomodulin on the endothelial surface, it inhibits blood clotting by activating protein C (15). The latter is a zymogen formed in the liver and is vitamin K dependent. Protein S, another vitamin K-dependent protein, is a cofactor for protein C (26). Protein C also initiates fibrinolysis by activating plasminogen and neutralizing plasminogen activator inhibitor (9) (PAI-1) (see Fig. 35.1). Resistance to activated protein C has been discovered as a hereditary trait in individuals who develop venous thromboembolism without apparent cause. This is referred to as factor VLeiden and is said to occur in approximately 20% of patients with venous thromboembolism. In one study, the risk of recurrent venous thromboembolism in carriers of this mutation was
MORRIS NOTELOVITZ
increased with a hazard ratio of 2.4 compared with patients without this condition (27). Antithrombins are inhibitors of thrombin and other coagulation proteases (28). Antithrombin III, which is the most important, probably accounts for at least 50% of the natural fluidity of blood. The action of antithrombin III is catalyzed by heparin and, in its complex form, neutralizes thrombin instantly (29). Antithrombin III also inhibits other enzymes in the coagulation cascade, such as activated factors Xa, XII, XI, and X (15). However, its main efficacy as a natural anticoagulant is thought by some to be the prevention of factor Xa generation and the activation of prothrombin. A deficiency of antithrombin III is associated with an increased liability of thrombosis (30). The level at which antithrombin III deficiency would cause thrombosis is not known, but it would probably have to be reduced to approximately 50% of its normal activity. This estimate is based on the plasma levels of antithrombin III in families who have congenital antithrombin III deficiency. Not all individuals with antithrombin III deficiency experience thrombosis. As with protein C deficiency, thrombosis usually occurs in conjunction with other events such as trauma and surgery. Reliable assays are now available for the measurements of antithrombin III and protein C levels.
V. FIBRIN FORMATION A N D FIBRINOLYSIS The introduction of t-PA therapy into clinical practice emphasizes the importance of fibrinolysis in maintaining intimal health and vascular patency. Fibrinogen is a dimer with three polypeptide chains: ~x, 13, and ~/. Thrombin acts on the ci and 13 chains and produces two molecules of fibrinopeptide A and two molecules of fibrinopeptide B. This leaves a large residue molecule known as fibrin monomer, which spontaneously polymerizes into non-stable fibrin polymer. These are cross-linked in the presence of calcium and activated factor XIII to form the stable insoluble clot, fibrin (9). This process is vital to survival and is nature's way of preserving the integrity of the vascular tree subsequent to injury so that healing and normal function of the area concerned can be restored and maintained. Fibrin formation is regulated by the process of fibrinolysis, which involves the enzymatic degradation of fibrin and fibrinogen by plasmin (9). Plasmin is formed from plasminogen, a J3-globulin synthesized by the liver (25). The biologic activity of plasmin, and hence fibrinolysis, is determined by activators and inactivators of plasminogen and plasmin inhibitors (see Fig. 35.1). Fibrinolysis is initiated by either factor XIIa or urokinase plasminogen activator (intrinsic pathway) or tissue-type plasminogen activator (t-PA), the extrinsic pathway (31). Tissue plasminogen activator, which is produced by the endothelial cells and released into the circula-
CHAPTER 35 Hormone Therapy and Hemostasis
485
FIGURE 35.5 Diagrammaticdepiction of how fibrin polymerforms.Asterisksdepict potential markers for thrombosis.
tion, in the presence of fibrin binds together with plasminogen to generate plasmin, and thus fibrinolysis (17). Fibrin acts both as a substrate and a cofactor to plasminogen activity. Tissue plasminogen activator is inhibited by plasminogen activator inhibitor type 1 (PM-1) (32). PM-1 is produced by hepatocytes and endothelial cells. The balance between t-PA and PM-1 is said to be the major determinant of the spontaneous fibrinolytic activity of blood (32). Both fibrinogen and PM-1 are significantly reduced by combination transdermal estradiol plus MPA, and the mean plasma estradiol level was found to be 138 pg/mL (33). In another study, conjugated estrogen (alone and in combination with MPA) significantly reduced plasma PM-1 antigen levels (34). In this study, transdermal estrogen had very little effect on PM-1. As noted, plasminogen has to bind to fibrin in order to be activated to plasmin. The binding takes place at the socalled "lysine-binding" sites on the plasminogen molecule. Histidine-rich glycoprotein has an affinity for these sites and can, therefore, control the amount of biologically available free plasminogen (35). Estrogen decreases plasma histidinerich glycoprotein and may, therefore, enhance fibrinolysis (36). Exogenous progestins and estrogen (37) are associated with increased plasminogen activity (5). Three main plasmin inhibitors exist: c~2-plasmin inhibitor; ci2-macroglobulin, which reacts quickly and as a competitive plasmin inhibitor; and ci2-antitrypsin, which reacts more slowly but more firmly. The latter two proteases also inhibit thrombin and as such have conflicting functions: they prevent clot formation by antagonizing thrombin, and they encourage fibrin and fibrinogen integrity by inhibiting plasmin. Their overall effect
on thrombogenesis is not known. Antiplasmin inhibitor (ci2-plasmin inhibitor) is said to inhibit 35% of the plasmin generated from plasminogen (38). It acts in two ways: direct inactivation of plasmin and blockage of plasminogen binding to fibrin. Pharmacologic lowering of a2-plasmin inhibitor levels can be viewed as a positive side effect because the net effect will result in increased fibrinolysis. Cleavage of fibrin and fibrinogen produces a variety of fragments known as fibrin orfibrinogen degradationproducts (FDPs). FDPs have potent anticoagulant properties and may interfere with platelet activity as well.
VI. CLINICAL MARKERS OF THROMBOSIS Given the complexity of hemostasis, it is not surprising that relatively few tests can predict individuals at risk for thrombosis (23,39). Only 15% to 20% of patients with hypercoagulability have an identifiable biochemical abnormality of their plasma proteins (8). Thrombosis is a localized event. Tests based on systemic venous blood samples are insensitive markers of changes occurring on the damaged vascular intima~especially when the latter involves arteries~ and tests are performed on venous blood. Also, apart from the physiologic checks and balances of hemostasis, some factors can serve as a systemic procoagulant and a local vessel wall anticoagulant. For example, thrombin promotes coagulation by catalyzing activation of some zymogens (factors XI, VIII, and V) in the coagulation cascade (8), but it
486 stimulates the intact vascular endothelium to synthesize prostacyclin and NO and also promotes functions of t-PA, which together tend to inhibit platelet aggregation and vascular spasm while enhancing fibrinolysis (10,15,25). Various studies have identified plasma fibrinogen, factor VIIc (which can be indirectly measured by the prothrombin time), and PAI-1, as markers of arterial thrombosis (39,40). More recently, elevated levels of fragment 1.2, which is derived from the conversion of PT to thrombin, has been correlated with a thrombotic tendency (41). This test is clinically available and may identify patients at risk for thrombosis. As noted, the P T T is probably of little clinical value, whereas the role of the PT as a measure of anticoagulation is being reevaluated. Native prothrombin antigen has been identified as the sole type of prothrombin in normal blood and can readily be distinguished from abnormal prothrombin, which may be synthesized, for example, in response to treatment with anticoagulants such as warfarin (8). Studies have shown that the adequacy of anticoagulation is better monitored if native prothrombin antigen levels are assayed instead of the PT (42). The effect of exogenous sex steroids on native prothrombin antigen is unknown. Fibrinopeptides A and B, as well as various FDPs, do change in response to H T and may be an indirect sign of accelerated coagulation. In this context, the assay of the FDP, D-dimer, may be important. D-dimer, which is an FDP peptide with epitopes that cross-link with those found on activated fibrin, has been used as a marker of ongoing intervascular coagulation; for example, in patients with suspected pulmonary embolism (43). Although measurement of sensitive D-dimer has a clear place in the diagnosis of thromboembolic states, the sensitivity is not as high as once envisioned (only in the 80% range) (43a). Reliable assays are available for antithrombin III and proteins C and S. Deficiencies of protein C, protein S, or antithrombin III are associated with thromboembolic disease and are used for screening tests in women with histories of previous thrombotic disease. These protein assays also have to be distinguished from the measurements of activity. Antithrombin III activity must be measured in plasma. Serum antithrombin III only quantifies the amount of antithrombin III left after clotting, not that consumed during clotting; therefore, it has no clinical relevance to the risk of inappropriate thrombus formation (22). Although activated protein C resistance is said to be present in at least 20% (or more) of patients with venous thrombosis and has been associated with an increased risk of thrombosis in women taking OCs (see later), routine testing for this factor prior to treating women with hormone replacement therapy (HRT) is both impractical and costly. Thus, if the prevalence of factor V Leiden is 2% and the positive predictive value of the assay is 44%, the cost of preventing one thromboembolic death is more than $44 million (44). Currently the testing for genetic aspects of thrombophilia (i.e., evaluating polymorphisms and various mutations) has
MorRis NOTELOVlTZ become extremely commonplace, particularly in women who have recurrent miscarriages. Whether these tests are truly meaningful for the risk of thrombosis in a postmenopausal women in questionable, particularly without a family history or a history of prior thrombosis. In addition to mutations in Factor V resulting in activated protein C resistance (APC), mutation searches may also include those invoMng the prothrombin gene (G20210A) and in methylene-tetrahydrofolate reductase (MTHFR) Cb77T. Current clinical assays are able to detect both homozygous and heterozygous states. However, these assays are of dubious value, particularly without a strong family or personal history of thrombosis, and are not recommended for screening women before considering ET or HT.
VII. COAGULATION A N D CLINICAL SYNDROMES The association between accelerated coagulation, venous thrombosis (both superficial and deep), and the potential for pulmonary embolism is well recognized and will not be discussed further. Less well-recognized is the key role of thrombosis and fibrin formation in the origin and progression of atherosclerosis, as well as the pathogenesis of acute CAD and associated syndromes. This has been reviewed in the past (6). In brief, subsequent to injury of the coronary artery endothelium, macrophages and lipids accumulate at the site of epithelial denudation. The macrophages initiate changes that involve the intima and lead to platelet adhesion and the release of various growth factors. This results in smooth muscle cell proliferation, hypertrophy and migration, and the formation of foam cells, characteristic of the mature atherosclerotic plaque. These lesions appear in most children by the age of puberty (45); however, many regress and it is only in the third decade of life that the lipid-rich plaques are covered by a fibrotic cap with focal areas of microthrombi formation (46). Fibrin and fibrin-related products have been found in the intimal section of these lesions and are typical of a mature atherosclerotic plaque (47). Angiographic studies have shown that the intimal lesions responsible for subsequent myocardial infarction often have mild (50%) to moderate (70%) stenosis (48). Thus, recurrent mural thrombi-associated with fissuring and disruption of established plaque-rather than sudden occlusive events may be responsible for much of the unstable angina and eventual myocardial infarctions seen in clinical practice (49). The clinical impact of mural thrombosis is self-evident. The efficacy of thrombolytic therapy when treating patients with acute myocardial infarction is well established. The inhibition or prevention of atherosclerosis by prophylactic antithrombin or fibrinolytic therapy is less clear. Two unrelated (and speculative) observations suggest a potential role for preventive subclinical anticoagulation: pigs with homozygous von Willebrand disease are resistant to thrombosis and
CHAPTER
35 Hormone Therapy and Hemostasis
spontaneous atherosclerosis (50). Lower-dose aspirin prophylaxis is very effective in preventing coronary thrombosis (51). By inhibiting platelet adhesion at sites of vascular damage, aspirin may prevent microthrombosis and the subsequent evolution of atherosclerosis. The role of sex steroids in the pathogenesis of the intimal thrombosis and atherogenesis has not been established, but a potential relationship is seen. Apolipoprotein (a), a glycoprotein that is part of the Lp(a) molecule, is structurally similar to plasminogen (52). In vitro studies have demonstrated competition between apolipoprotein (a) and plasminogen for the latter's endothelial binding sites (53). By interfering with plasmin generation and clot lysis, endothelial thrombosis results. Lp(a) is also a potent and independent atherogenic moiety. It is absorbed into the arterial intima, where it accumulates in macrophages and is eventually degraded into foam cells, the precursor of arterial plaque (54). Lp(a) also binds to and immobilizes intimal fibrin and promotes plaque formation by the previously discussed thrombin hypothesis. The effect of exogenous estrogen on plasma levels of Lp(a) is variable (55), but some studies have reported a 50% reduction in plasma levels (56). Inconsistency among various studies may be caused by the initial plasma Lp(a) value. Estrogen decreases elevated Lp(a) but has little apparent effect on normal plasma Lp(a) levels. Progestins and, to a lesser extent, exogenous estrogen increase plasma plasminogen antigen and plasminogen activity (5,31). The net potential may be intimal protection via estrogen, because of inhibition of Lp(a) absorption and foam-cell formation, and reduced levels of plasma apolipoprotein (a). This could then free endothelial plasminogen receptors to respond to progestin-stimulated endothelial fibrinolysis if progestins are administered. A recently published study adds some credence to this hypothesis. After 1 year of combination HRT-either estradiol valerate plus MPA or transdermal estrogen plus MPA-significant reductions were noted in plasma levels (compared with baseline) of PAI-1 activity, Lp(a), with corresponding increases in tPA antigen and the percent of plasminogen (57). As reviewed here and elsewhere, hormonal therapy causes an increase in fibrinolytic activity, for example an increase in plasminogen, while at the same time changes also occur in markers of procoagulation.
487 nificant in some studies, is invariably well within the physiologic range. It is highly unlikely that these changes have any biologic impact in vivo. Transdermal estrogen has a negligible influence on the hepatic zymogens (63) and is the preferable route of administering estrogen when procoagulant activity is increased; for example, immediately postsurgery. Of importance, recent data by Scarabin et al. (64) confirmed a thrombosis risk with oral estrogen but not with transdermal preparations. Consistent with the more recent data (2-4,64) standard H T is associated with a twofold to threefold increase in thrombosis risk. This has been extensively reviewed in Chapter 36. Of note, there is suggestive evidence that lower doses of ET or H T carries a lower risk (65), although there are no prospective randomized trials on this issue. The existing data from the largest randomized trials shown an increased relative risk at the lower range of that in observational trials (i.e., closer to 2). This overall relative risk of venous thromboembolism (VTE) in W H I included women who were older (up to age 75 at study entry), who were obese, and who had other co-morbidities (one-third were hypertensive). Also at the beginning of the W H I H T trial, women with prior thrombosis were allowed to be enrolled. Even if that had not been the case, it is known that VTE risk with H T is observed earlier after initiation of H T and then decreases somewhat thereafter (Fig. 35.6). The overall rate in the estrogen-alone (E-alone) trial of W H I was lower, with all the end points such as pulmonary embolism not being statistically increased (Fig. 35.7). This is noteworthy in that this cohort of hysterectomized women was more obese. However, the fact that 50% of this group had used ET or H T in the past may suggest that the exposed group were less susceptible to thrombosis, having been exposed earlier to this risk without having a thrombotic episode. Finally, it is also possible that there is a real difference in the risk of thrombosis with H T versus ET, with progestogens imparting an additional risk.
HR :2.11 ] 0.7
0.6
F, o.s Ill
VIII. HRT AND ARTERIAL THROMBOSIS. CLINICAL CONSIDERATIONS
9 Hazard Year Ratio
95% aCI = 1.26-3.55 -a-- CEE/MPA a -4)- Placebo
0.4
o o .o L
91
3.60
92
2.26
93
1.67
94
1.84
0.2
5
2.49
0.1
9 6+
0.90
*. 0.3 c
0.0
An extensive literature exists on H T and its effect on hemostasis (58-62). Briefly stated, oral estrogens will induce an increase in the hepatic zymogens factors VII, X, IX, and II. However, most of the assays measure the inactive proenzyme, and the percent increase, although statistically sig-
95% n C I : 1.58-2.82
I 1
I 2
I 3
Year
I 4
I 5
I 6§
P < .05, significant for decreasing risk over time.
FIGURE 35.6 Annualized percentage of V T E events by year for women receiving C E E / M P A or placebo showing more events in the first year. (From ref. 6.)
488
MORRIS NOTELOVITZ
Kaplan-Meier Estimate
0.. i I.R=i.34 I I
-1-
0.04 i i 0.03 'i
95% nCI = 0.87-2.06 95% aCI = 0.70-2.55
_m " 0.02 'i E = 0.01 ..~ o
CEE ~ ~ - ~ " - - ~ - - -
0.00
0
1
2
3
4 Time (year)
5
6
Placebo 7
8
FIGURE 35.7 Cumulative hazard for pulmonary embolism by time, for women receiving CEE alone versus placebo. (From res 7.)
IX. C O N C L U S I O N Although the exact mechanisms of VTE risk with ET/ H T are not known, in the absence of a known thrombophilia, it is clear that the VTE risk is twofold increased with standard-dose oral ET/HT. Additional factors are necessary to explain this risk in that measurement of circulatory procoagulant and fibrinolytic factors, although potentially altered by ET/HT, do not correlate well with this risk. Indeed, many of the procoagulant changes observed are statistical changes within the normal range of most laboratories. Lower oral doses and transdermal estrogen probably reduce this risk. Finally, the increase in VTE does not appear to alter mortality, and the overall prevalence is still quite low. For example, the risk of pulmonary embolism (PE), if increased twofold with HT, results in a risk of approximately 30-40 per 100,000 women per year. This rate is still lower than the rate of PE in pregnancy, which is approximately 60 per 100,000 women per year.
References 1. Lobo RA. Estrogen and the risk of coagulopathy. Am J Med 1992;92:283-285. 2. Daly E, Vessey M, Hawkins M, et al. Risk of venous thromboembolism in users of hormone replacement therapy. Lancet 1996;348: 977-980. 3. Jick H, Derby LE, Myers MW, Vasilakis C, Newton KM. Risk of hospital admission for idiopathic venous thromboembolism among users of postmenopausal oestrogens. Lancet 1996;348:981-983. 4. Grodstein F, Stampfer MJ, Goldhaber SZ, et al. Prospective study of exogenous hormones and risk of pulmonary embolism in women. Lancet 1996;348:983-987. 5. Grady D, Wenger NK, Herrington D, et al. Postmenopausal hormone therapy increases risk for venous thromboembolic disease. The Heart and Estrogen/progestin Replacement Study. Ann Intern Mec12000; 132:689-696.
6. Rossouw JE, Anderson GL, Prentice RL, et al., Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAm MedAssoc 2002;288:321-333. 7. The Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative Randomized Controlled Trial. JAm MedAssoc 2004;291:1701-1712. 8. Furie B, Furie BC. Molecular and cellular biology of blood coagulation. NEnglJMed 1992;326:800-806. 9. Verstraete M, Verhaeghe R. The physiologic mechanism of blood coagulation and fibrinolysis.Adv Contracept 1991;(Suppl 3):244-258. 10. Vane JR, Anggard EE, Botting RM. Regulatory functions of the vascular endothelium. NEnglJMed 1990;323:27-36. 11. Cicinelli E, Ignarro LJ, Lograno M, et al. Acute effects of transdermal estradiol administration on plasma levels of nitric oxide in postmenopausal women. Fertil Steri11997;67:63-66. 12. Stanczynk F, Rosen G, Ditkoff E, et al. Influence of estrogen on prostacyclin and thromboxane balance in postmenopausal women. Men.JNorth Am Men Soc 1995;2:137-143. 13. Mikkola T, Ranta V, Orpana A, Viinikka L, Ylikorkala O. Hormone replacement therapy modifies the capacity of plasma and serum to regulate prostacyclin and endothelin-1 production in human vascular endothelial cells. Fertil Steri11996;66:389-393. 14. Golino P, Piscione F, Willerson JT, et al. Divergent effects of serotonin on coronary-artery dimensions and blood flow in patients with coronary atherosclerosis and control patients. NEnglJMed 1991;324: 641-648. 15. Esmon CT. The roles of protein C and thrombomodulin in the regulation of blood coagulation.J Biol Chem 1989;264:4743-4746. 16. Marcum JA, Rosenberg RD. Anticoagulantly active heparin-like molecules from vascular tissue. Biochemistry 1984;10;23:1730-1737. 17. Loscalzo J, Braunwald E. Tissue plasminogen activator. NEnglJMed 1988;319:925-931. 18. Larsen E, Celi A, Gilbert GE, et al. PADGEM protein: a receptor that mediates the interaction of activated platelets with neutrophils and monocytes. Cell 1989;59:305-312. 19. Heavey DJ, Barrow SE, Hickling NE, Ritter JM. Aspirin causes short-lived inhibition of bradykinin-stimulated prostacyclin production in man. Nature 1985;318:186-188. 20. Bar J, Tepper R, Fuchs J, et al. The effect of estrogen replacement therapy on platelet aggregation and adenosine triphosphate release in postmenopausal women. Obstet Gyneco11993;81:261-264. 21. Aune B, Oian P, Omsjo I, Osterud B. Hormone replacement therapy reduces the reactivity of monocytes and platelets in whole blood-a beneficial effect on atherogenesis and thrombus formation? Am J Obstet Gyneco11995;173:1816-1820. 21a. Williams MS, Vaidya D, Kickler T, Ouyang E Long-term hormone replacement therapy does not cause increased platelet activation. Am HeartJ 2005;150:434-438. 22. Mammen AE Oral contraceptives and blood coagulation: a critical review. Am J Obstet Gyneco11982;142:781-790. 23. Winkler UH. Hormone replacement therapy and hemostasis: principles of a complex interaction. Maturitas 1996;24:131-145. 24. Osterud B, Rapaport SI. Activation of factor IX by the reaction product of tissue factor and factor VII: additional pathway for initiating blood coagulation. ProcNatlAcad Sci U S A 1977;74:5260-5264. 25. Collen D. On the regulation and control of fibrinolysis. Edward Kowalski Memorial Lecture. Thromb Haemost 1980;43:77-89. 26. Dahlback B. Protein S and C4b-binding protein: components involved in the regulation of the protein C anticoagulant system. Thromb Haemost 1991;66:49-61.
CHAPTER 35 H o r m o n e Therapy and Hemostasis 27. Simioni P, Prandoni P, Lensing AW, et al. The risk of recurrent venous thromboembolism in patients with an Arg506-->Gln mutation in the gene for factor V (factor V Leiden). N EnglJ Med 1997;336: 399-403. 28. Rosenberg RD. Actions and interactions of antithrombin and hepatin. N EnglJ Med 1975;292:146-151. 29. Jordan RE, Fayreau LV, Braswell EH, Rosenberg RE). Heparin with two binding sites for anti-thrombin or platelet factor 4. J Biol Chem 1982;257:400-406. 30. Egeberg O. Inherited anti-thrombin deficiency causing thrombophilia. Thromb Diath Haemorrh 1965;13:516-530. 31. Jesperson J, Peterson KR, Skouby SO. Effects of newer oral contraceptives on the inhibition of coagulation and fibrinolysis in relation to dosage and type of steroid. A m J Obstet Gyneco11990;163:396-403. 32. Sprengers ED, Kluft C. Plasminogen-activated inhibitors. Blood 1987;69:381-387. 33. Lindoff C, Peterson F, Lecander I, Martinsson G, Astedt B. Transdermal estrogen replacement therapy: beneficial effects on hemostatic risk factors for cardiovascular disease. Maturitas 1996;24:43-50. 34. Koh K, Mincemoyer R, Bui M, et al. Effects of hormone-replacement therapy on fibrinolysis in postmenopausal women. N Engl J Med 1997;336:683-690. 35. Lijnen HB, Hoylaerts M, CoHen D. Isolation and characterization of a human plasma protein with affinity for the lysine binding sites in plasminogen. Role of the regulation of fibrinolysis and identification of histidine-rich glucoprotein. J Biol Chem 1980; 255:10214-10222. 36. Haukkamaa M, Morgan WT, Koskelo E Serum histamine-rich glycoprotein during pregnancy and hormone treatment. ScandJ Clin Lab Invest 1983;43:591-595. 37. Conard J, Gompel A, Pelissier C, Mirabel C, Basdevant A. Fibrinogen and plasminogen modifications during oral estradiol replacement therapy. Fertil Steri11997;68:449-453. 38. Aoki N, Saito H, Kamiya T, et al. Congenital deficiency of alpha 2-plasmin inhibitor associated with severe hemorrhagic tendency. J Clin Invest 1979;63:877-884. 39. Kluft C. Disorders of the hemostatic system and the risk of the development of thrombotic and cardiovascular disease: limitation of laboratory diagnosis. Am J Obstet Gyneco11990;163:305-312. 40. Kelleher CE. Clinical aspects of the relationship between oral contraceptives and abnormalities of the hemostatic system: relation to the development of cardiovascular disease. Am J Obstet Gynecol 1990; 163:392-395. 41. Conway EM, Bauer KA, Barzegar J, Rosenberg RD. Suppression of hemostatic system activation by oral anti-coagulants in the blood of patients with thrombotic diathesis. J Clin Invest 1987;80:1533-1544. 42. Hirsh J. Oral anticoagulant drugs. N Engl J Med 1991;324:18651875. 43. Bounameaux H, Cirafici P, de Moerloose P, et al. Measurement of d-dimer in plasma as diagnostic aid in suspected pulmonary embolism. Lancet 1990;1:96-200. 43a. Ray P, Bellick B, BiroUeau S, et al. Referent d-dimer enzyme-linked immunosorbent assay testing is of limited value in the exclusion of thromboembolic disease: result of a practical study in an ED. Am J E merg Med 2006;24:313-318. 44. Altes A, Souto J, Mateo J, Borrell M, Fontcuberta J. Activated protein C resistance assay when applied in the general population.AmJ Obstet Gyneco11997;176:358-359. 45. Stary HE. Evolution and progression of atherosclerotic lesions in coronary arteries of children and young adults. Arteriosclerosis 1989; 99:119-132. 46. Davies MJ, Woolf N, Rowles PM, Pepper J. Morphology of the endometrium over atherosclerotic plaques in human coronary arteries. Br HeartJ 1988;60:459-464.
489 47. Smith EB, Kean A, Grant A, Stirk C. Fate of fibrinogen in human arterial intima. Arteriosclerosis 1990;10:263-275. 48. Little WC. Angiographic assessment of the culprit coronary artery lesion before acute myocardial infarction. Am J Cardiol 1990;66: 44G-47G. 49. Falk E. Unstable angina with fatal outcome: dynamic coronary artery thrombosis leading to infarction and/or sudden death. Autopsy evidence of recurrent mural thrombosis with peripheral embolization culminating in total vascular occlusion. Circulation 1985;71:699-708. 50. Foster V, Griggs TR. Porcine yon WiUebrand disease: implications for the pathophysiology of atherosclerosis and thrombosis. Prog Hemost Thromb 1968;8:159-183. 51. Manson JE, Stampfer MJ, Colditz GA, et al. A prospective study of aspirin use and primary prevention of cardiovascular disease in women. JAm MedAssoc 1991;266:521-527. 52. Loscalzo J. A unique risk factor for athero-thrombotic disease. Arteriosclerosis 1990;10:672-679. 53. Hajjar KA, Gavish O, Breslow JL, Nachman RL. Lipoprotein (a) modulation of endothelial cell surface fibrinolysis and its potential role in atherosclerosis. Nature 1989;339:303-305. 54. Smith EB, Cochran S. Factors influencing the accumulation in fibrous plaques of lipid derived from low-density lipoprotein. II: Preferential immobilization of lipoprotein (a): LP (a). Arteriosclerosis 1990;84:173-181. 55. Lobo RA, Notelovitz M, Bernstein L, et al. Lipoprotein(a) relationship to cardiovascular disease risk factors, exercise and estrogen. Am J Obstet Gyneco11992;166:1182-1190. 56. Henricksen P, Angelin B, Berglund L. Hormonal effects of serum Lp(a) levels: marked reduction during estrogen treatment in males with prostatic cancer [Abstract]. Arterioscler Thromb VaseBiol 1991; 11:1423a. 57. Gilabert J, Estelles A, Cano A, et al. The effect of estrogen replacement therapy with or without progestogen on the fibrinolytic system and coagulation inhibitors in postmenopausal status. Am J Obstet Gyneco11995;173:1849-1854. 58. Bush TL, Barrett-Connor E. Noncontraceptive estrogen use and cardiovascular disease. Epidemiol Rev 1985;7:89-104. 59. Notelovitz M, Ware M. Coagulation risks with postmenopausal oestrogen therapy [Review Article]. In: Studd J, ed. Progress in obstetricsand gynecology, vol 2. Edinburgh: Churchill-Livingstone, 1983:228-240. 60. Young RL, Goepfert AR, Goldzieher HW. Estrogen replacement therapy is not conducive of venous thromboembolism. Maturitas 1991;13:189-192. 61. Kessler CM, Szymanski LM, Shamsipour Z, et al. Estrogen replacement therapy and coagulation: relationship to lipid and lipoprotein changes. Obstet Gyneco11997;89:326-331. 62. Lobo RA, Pickar JH, Wild RA, Walsh B, Hirvonen E. Metabolic impact of adding medroxyprogesterone acetate to conjugated estrogen therapy in postmenopausal women. The Menopause Study Group. Obstet Gyneco11994;84:987-995. 63. Chetkowski RJ, Meldrum DR, Steingold KA, et al. Biologic effects of transdermal estradiol. N EnglJ Med 1986;314:1615-1620. 64. Scarabin PY, Oger E, Bureau GP, ESTHER Study Group. Differential association of oral and transdermal oestrogen-replacement therapy with venous thromboembolism risk. Lancet 2003;362:428-432. 65. Lobo RA. The rationale for low-dose hormonal therapy. Endocrine 2004;24:217-221. [Review.] 66. Notelovitz M, Kitchens CS, Rappaport Y, et al. Menopausal status associated with increased inhibition of blood coagulation. Am J Obstet Gyneco11981;141:149-152. 67. Meade TW, Dyer S, Howarth OJ, et al. Anti-thrombin III and procoagulant activity: sex differences and effects of the menopause. Br J Haemato11990;74;77-81.
490 68. Bourne T, Hillard TC, Whitehead MI, et al. Oestrogens, arterial status, and postmenopausal women. Lancet 1990;325:1470-1471. 69. Daly E, Vessey M, Hawkins M, et al. Risk of venous thromboembolism in users of hormone replacement therapy. Lancet 1996;348: 977-980. 70. Jick H, Derby L, Myers M, Vasilakis C, Newton K. Risk of hospital admission for idiopathic venous thromboembolism among users of postmenopausal oestrogens. Lancet 1996;348:981-983. 71. Grodstein F, Stampfer M, Goldhaber S, et al. Prospective study of exogenous hormones and risk of pulmonary embolism in women. Lancet 1996;348:983-987. 72. Aylward M. Coagulation factors in opposed and unopposed oestrogen treatment at the climacteric. PostgradMedJ 1978;54:31-37. 73. Notelovitz M, Kitchens C, Coone R, et al. Low-dose oral contraceptive usage and coagulation. Am J Obstet Gyneco11981;141:71-76. 74. Belier FK, Ebelt E Effects of oral contraceptives on blood coagulation: a review. Obstet GynecolSurv 1985;44:425-436. 75. Notelovitz M, Kitchens C, Ware M, et al. Combination estrogen and progestogen replacement therapy does not adversely affect coagulation. Obstet Gyneco11983;62:596-600. 76. World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Venous thromboembolic disease and combined oral contraceptives: results of international multi centre case-control study. Lancet 1995;346:1575-1582. 77. World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Effect of different progestogens in low oestrogen oral contraceptives on venous thromboembolic disease. Lancet 1995;346:1582-1588. 78. Jick H, Jick S, Gurewich V, Myers M, Vasilakis C. Risk of idiopathic cardiovascular death and nonfatal venous thromboembolism in women using oral contraceptive with differing progestogen components. Lancet 1995;346:1589-1593. 79. Bloemenkamp K, Rosendaal F, Helmerhorst F, Buller H, Vandenbroucke J. Enhancement by factor V Leiden mutation of risk of deepvein thrombosis associated with oral contraceptives containing a third-generation progestagen. Lancet 1995;346;1593-1596. 80. Spitzer W, Lewis M, Heinemann L, Thorogood M, MacRae K. Third generation oral contraceptives and risk of venous thromboembolic disorders: an international case-control study. BMJ 1996;312: 83-88. 81. Lewis M, Spitzer W, Heinemann L, et al. Third generation oral contractive and risk of myocardial infarction: an international casecontrol study. BMJ 1996;312:88-90. 82. Notelovitz M, Ware M. Coagulation risks with postmenopausal oestrogen therapy [Review article]. In: Studd J, ed. Progress in obstetricsand gynecology, vol 2. Edinburgh: Churchill-Livingstone, 1983:228-240. 83. Young RL, Goepfert AR, Goldzieher HW: Estrogen replacement therapy is not conducive of venous thromboembolism. Maturitas 1991;13:189-192. 84. Kessler C, Szymanski L, Shamsipour Z, et al. Estrogen replacement therapy and coagulation: relationship to lipid and lipoprotein changes. Obstet Gyneco11997;89:326-331.
MORRIS NOTELOVITZ 85. Lobo R, Pickar J, Wild R, Walsh B, Hirvonen E. Metabolic impact of adding medroxyprogesterone acetate to conjugated estrogen therapy in postmenopausal women. Obstet Gyneco11994;84:987-995. 86. Chetkowski RJ, Meldrum DR, Steingold KA, et al. Biologic effects of transdermal estradiol. N EnglJ Med 1986;314:1615-1620. 87. Steingold KA, Matt DW, de Ziegler D, et al. Comparison of transdermal to oral estradiol administration on hormonal and hepatic parameters in women with premature ovarian failure. J Clin Endocrinol Metab 1991;73:275-280. 88. Boschetti C, Cortellaro M, Nencione T, et al. Short- and long-term effects of hormone replacement therapy (transdermal estradiol vs. oral conjugated equine estrogens, combined with medroxyprogesterone acetate) on blood coagulation factors in post-menopausal women. Thromb Res 1991;62:1-8. 89. Notelovitz M, Greig HBW. Natural estrogen and anti-thrombin III activity in postmenopausal women. J Reprod Med 1976;16:87-90. 90. Notelovitz M, Kitchens CS, Ware MD. Coagulation and fibrinolysis in estrogen treated surgically menopausal women. Obstet Gynecol 1984;63:621-625. 91. Lindberg UB, Crona N, Stigendal L, et al. A comparison between effects of estradiol valerate and low dose ethinyl estradiol on hemostasis parameters. Thromb Haemos 1989;61:65-69. 92. WinNer U. Effects of androgens on haemostasis. Maturitas 1996; 24:147-155. 93. Notelovitz M, Johnston M, Smith S, Kitchen C. Metabolic and hormonal effects of 25 mg and 50 mg 17~-estradiol implants in surgically menopausal women. Obstet Gyneco11987;70:749-754. 94. Clarke RJ, Mayo G, Price I, Fitzgerald GA. Suppression of thromboxane A, but not of systemic prostacyclin by controlled-release aspirin. NEnglJMed 1991;325:1137-1141. 95. Osterud B, Olsen JO, Wilsgard L. Effect of strenuous exercise on blood monocytes and their relation to coagulation. Med Sci Sports Exercise 1989;21:374-378. 96. Bartsch P, Haeberli A, Straub PW. Blood coagulation after longdistance running: anti-thrombin III prevents fibrin formation. Thromb Haemost 1990;6:430-434. 97. Notelovitz M, Zauner C, McKenzie L, et al. The effect of low-dose oral contraceptives on cardiorespiratory function, coagulation, and lipids in exercising young women: a preliminary report. Am J Obstet Gyneco11987;156:591-598. 98. Ornish D, Brown SE, Scherwitz LW, et al. Can lifestyle changes reverse coronary heart disease? Lancet 1990;336:129-133. 99. Kane JT, Malloy IM, Ports TA, et al. Regression of coronary atherosclerosis during treatment of familial hypercholesterolemia with combined drug regimens.JAm MedAssoc 1990;264:3007-3012. 100. Sullivan IM, Vanderzwaag R, Lemp G, et al. Estrogen replacement and coronary artery disease. Effect on survival in postmenopausal women. Ann Intern Med 1988;108:358-363.
Risk of Pulmonary Embolism/Venous Thrombosis CAROLYN WESTHOFF
Columbia University Medical Center, New York, NY 10032
I. CLINICAL ENTITIES
Clots in the leg may be silent, recognized only after a pulmonary embolism has occurred, if then, or they may come to clinical attention due to swelling and pain in the extremity. In clinically recognized cases of deep vein thrombosis of the leg, the main symptoms are swelling (88%), pain (56%), and tenderness (55%). Symptoms such as redness, palpable venous cord, or a positive Homans' sign are present in only a minority of confirmed cases (2). Pulmonary emboli present with dyspnea (80%), pleuritic chest pain (60%), and cough (41%) (3). On ausculation, 60% of angiogramconfirmed cases have audible crackles. There is no important difference between men and women in the frequency of these presenting complaints. Because of difficulties in diagnosing these conditions, it is likely that many milder cases may be unrecognized. The sequelae of venous thrombosis are immediate death due to pulmonary embolism, recurrent symptomatic or asymptomatic thrombosis with repeated risk of embolism, and the development of postphlebitic syndrome. The postphlebitic syndrome encompasses chronic swelling, skin changes including ulceration, and chronic pain interfering with ambulation (4). These chronic symptoms can also be associated with superficial phlebitis. The main goal of treatment of deep vein thrombophlebitis (DVT) is to prevent
Venous thrombosis mainly presents in the deep veins of the leg or in the lung but can also occur elsewhere, including the brain, retina, liver, mesentery, and upper extremities. Clots originating in the upper extremity are relatively rare and are usually sequelae of medical procedures such as catheter placement (1). Thrombophlebitis is a clinical syndrome of pain and swelling in the leg that originates as a valve pocket thrombus, most often in the soleal vein or posterior tibial or popliteal veins, during some period of stasis. Such a thrombus may propagate upward to the femoral and iliac veins, where pieces then break off into the venous circulation to become trapped in the lung as a pulmonary embolus. It is thought that distal leg clots do not embolize until after upward propagation to the thigh; however, clots may originate in the thigh or pelvis rather than in the leg. Further propagation in the lung or showers of emboli can be fatal. In most cases of pulmonary embolism it is possible to identify an apparent source clot in the lower extremity. About two-thirds of clinically recognized cases of venous thromboembolism (VTE) present with a venous thrombosis in the lower extremity only, and the remaining one-third presents as a pulmonary embolus with or without a symptomatic clot in the leg. TREATMENT OF THE POSTMENOPAUSAL WOMAN
491
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
492 pulmonary embolism (PE), prevent recurrence of thrombosis, and prevent the postphlebitic syndrome.
II. DIAGNOSIS A N D TREATMENT The method of diagnosis of clots depends primarily on the location of the suspected clot. Proximal thrombi (that is, above the knee) are diagnosed by real-time B-mode compression ultrasound with high sensitivity and high positive predictive value. This test is safe, noninvasive, and widely available. If the patient has had clots previously, compressability may be abnormal, and thus sonography will be less helpful for diagnosis of recurrent clots; it is also less usefid for distal clots. Contrast venography is still the standard for diagnosis of thrombi and is more accurate than sonogram when distal thrombi are suspected. Because venography is invasive and more difficult to perform, its use is limited to cases in which sonography is not helpful. Magnetic resonance imaging is proving to be useful for diagnosis but has not replaced other modalities, except perhaps in pregnant women (5). A suspected pulmonary embolus is first evaluated by a chest radiograph plus perfusion scan or by a ventilation-perfusion (WQ.) scan in which anatomic mismatches between air flow and blood flow in the lung may indicate the location of an embolus. The advantage of doing the perfusion scan as the first diagnostic test is that it is noninvasive and does not require highly specialized skills to carry out. A multicenter study, Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED), was carried out to assess the performance characteristics of WQ.scans used to assess 931 adults suspected of PE (3). A high-probability perfusion scan in women has a sensitivity of 41% and a positive predictive value for a PE of 86%. An intermediate- or high-probability WQ scan has a sensitivity of 84% and a positive predictive value of 44%. Thus, scans that are intermediate probability or nondiagnostic usually require proceeding to a pulmonary angiogram for a definitive diagnosis in order to avoid unnecessary treatment for those patients who will prove, by angiogram, to be negative for PE. Several useful diagnostic algorithms are available to guide patient assessment (6). Conventional treatment of DVT and PE has been intravenous unfractionated heparin during hospitalization followed by 3 to 6 months of oral anticoagulant therapy (OAT). Despite the ongoing risk of DVT recurrence, OAT is typically not continued past 6 months due to the risk of hemorrhage. Initial thrombolytic therapy in the acute phase is not widely used due to the risks of hemorrhage. Use ofthrombectomy or inferior vena cava filters is considered in selected patients with large thrombi or when adequate anticoagulation has not prevented recurrences. Also, treatment has been initiated with subcutaneous low-molecular-weight heparin (LMWH) used at home without laboratory dose adjustment. This appears to be at least similiar to unfractionated heparin in the prevention
CAROLYNWESTHOFF
of thrombus extension, death, hemorrhage, and recurrence of VTE events (7). An advantage of L M W H is that it can be used at home without laboratory monitoring; a disadvantage is its greater expense. The duration of subsequent therapy with OAT is extended, possibly for fife, in patients who have a continuing risk for VTE, such as those with cancer and those with an inherited predisposition.
III. PATHOPHYSIOLOGY The underlying causes of venous thrombosis have not been better described than by Virchow, who proposed in the nineteenth century that hypercoagulability, injury to the vessel wall, and circulatory stasis cause intravascular thrombosis. The disease was not described in antiquity, and Dexter proposes that the emergence of venous thrombosis as a major medical problem is due to the modern increase in chair sitting (8). Multiple factors are required for a venous clot to occur because each factor alone m t h a t is, stasis or injury or hypercoagulability--is only rarely associated with a clinically evident thrombus. Most of the recognized risk factors for VTE have an obvious connection to stasis, injury, or hypercoagulability. The coagulation system is a dynamic balance between procoagulant and anticoagulant factors. The coagulation system factors related to arterial thrombosis appear to be distinct from those related to venous thrombosis. The risk of venous clots has been linked to deficiencies of several anticoagulant factors. In particular, inherited deficiencies of protein C, protein S, and antithrombin III confer an increased risk of clinically apparent venous clots. Acquired or inherited resistance to activated protein C (APC resistance) also leads to increased risk of thromboses. APC resistance can be measured genotypically or phenotypicaUy. Genetic assays aim to identify in factor V an autosomal recessive point mutation called the Leiden mutation (9), which causes APC resistance. Phenotypic assays measure the resistance to cleavage of the factor V complex in the presence of activated protein C; this phenotype favors coagulation. The APCresistent phenotype is present in individuals with the Leiden factor V mutation (10), and also, in a brief report from the World Health Organization Monitored Investigation of Cardiac Disease (MONICA) survey, the APC-resistant phenotype has been associated with oral contraceptive (OC) users, hormone replacement therapy (HRT) users, and those with a body mass index (BMI) greater than 30 kg/m 2 (11). Screening of asymptomatic individuals for these inherited predispositions or for acquired APC resistance is not indicated on any routine basis. A genetically determined abnormality in prothrombin (a procoagulant) has been identified in thrombophilic families (12). This and other specific inherited or acquired abnormalities of the coagulation system are suspected to increase the risk of venous thrombosis. Additional genetic
CHAPTER 36 Risk of Pulmonary Embolism/Venous Thrombosis
493
studies will continue to pinpoint specific mutations in clotting factors that increase susceptibility to venous thrombosis, but these are likely to be rare and are currently not well established.
Age-specific incidence rates in women are generally similar to those obtained in the Worcester study; the higher rates observed for PE among the older women may reflect the inclusion of cases diagnosed post-mortem, a category that was missing in the other studies. A similar study evaluated DVT in Malmo, Sweden, a city of 230,000. Subjects included all patients referred for phlebography (venogram) for suspected DVT in 1987 (15). There was only one source of diagnostic testing for this population, and phlebography was the test of choice at the time of the study. There were 366 tests positive for DVT of 1009 referrals, including 189 women aged 40 years or older. Some of these cases may have been diagnosed and treated solely as outpatients, in contrast to most other studies that have considered only inpatient cases. Patients presenting with a primary pulmonary embolus were not included in this study, but PE was clinically suspected in 5% of the cases with DVT. As in the previous cohorts, the incidence rates increased with age and were similar in men and women. The incidence rates in Malmo may appear higher than those in Worcester because of the possible inclusion of outpatients and the inclusion of a few patients with PE in the DVT group and because new (76%) and recurrent (24%) cases were not separated. Fatality rates were not presented. Participants in the Nurses' Health Study ( N H S ) w e r e evaluated for the occurrence of PE between 1976 and 1992 (16). Medical records were examined for all of the 280 PEs that were reported by questionnaire or by death certificate. All but 36 of these cases occurred in women 40 years or older. Age-specific incidence rates are shown in Table 36.1; a rate for women older than 70 years is not calculated because very few members of the cohort had yet reached that age. Mortality rates were not presented. Based on the Tecumseh, Worcester, Malmo, and N H S cohorts, about one-quarter to one-third of all cases of PE and V T E among women over 40 years old are recurrences. The risk of a recurrence, even after 6 months of anticoagulant treatment, is much greater than the risk of a first event. In an 8-year study of 355 consecutive, new V T E cases in Padua, Italy, where the cases were diagnosed from 1986 to 1987, 78 patients experienced a recurrent venous thrombotic event during follow-up, including 9 fatal pulmonary emboli (4). The recurrences accrued gradually with a cumulative recurrence rate of 30% at 8 years. Those cases whose primary V T E was secondary to surgery or trauma had a decreased risk of recurrence. Whether sex or age influenced recurrence risk was not presented. Mortality at 8 years was 30% and was mainly related to cancer. After 8 years of follow-up, 29% of subjects developed post-thrombotic syndrome. In a similar study from Cleveland (17), 124 cases with venogramconfirmed V T E diagnosed in 1984 and 1985 were followed until recurrence or death; follow-up ended in 1992. During follow-up, 42% of the cases died; the main predictors of
IV. I N C I D E N C E A N D PROGNOSIS Several cohort studies have quantified the incidence of venous thrombosis and pulmonary embolism by identification of all new cases arising in a defined population. The first of these was the Tecumseh Community Health Study, whereby 9000 individuals of all ages in a city in Michigan were questioned and examined repeatedly in several cycles from 1957 to 1969 (13). Most of the 169 first thrombotic events and the 63 recurrent events were identified by report of the participants without further validation. The incidence of these events increased with age, and men and women over age 40 experienced similar rates of DVT and PE. Because there were only 45 cases reported by women over age 40, the incidence rates are imprecise, but nonetheless they are similar to those obtained subsequently in larger studies with more uniform and complete surveillance of the population. These investigators estimated there would be about 250,000 clinically recognized cases annually in the United States, of which nearly half would occur in women over age 40 years. The Worcester DVT study calculated incidence rates for V T E based on discharge diagnoses from all 16 hospitals covering the Worcester Standard Metropolitan Statistical Area (1985 population, 379,953) from July 1985 through December 1986. All medical records with an eligible discharge diagnosis were reviewed to identify cases living in the catchment area and to confirm the diagnoses (2). In total there were 615 cases included in the analysis--405 initial episodes and 210 recurrent episodes. W h e n extrapolated to the U.S. population, this study predicts about 170,000 initial and 99,000 recurrent cases annually, with nearly half of these in women over 40 years old. This study also identified an increase in risk with age and similar incidence rates for men and women. Age-specific incidence rates for first events in women over 40 are given in Table 36.1; the number of subjects used to calculate these rates was estimated from information in the paper. Overall, 12% of patients with a first episode died in the hospital, 5% after DVT, and 23% after PE. Case fatality was similar for men and women and increased with age. Immediate case fatality was only 2% below age 40 years but was 10% from ages 40 to 59 years and 11% from ages 60 to 79 years. All patients in the Worcester cohort who were discharged alive were followed for 2 to 3.5 years, and 30% of the patients died during the follow-up interval. Long-term survival depended on age but not on sex or whether the initial event was a PE or a DVT. Mayo Clinic investigators assessed the incidence of DVT from 1966 through 1990 in Olmsted County, M N (14).
494
CAROLYN WESTHOFF
TABLE 36.1 Age (Years) 40-49 50-59 60-69 70-79 Cases (N)
Incidence of Venous Thrombosis and Pulmonary Embolism in Women >40 Years Old a Worcester 1.0 4.2 9.9 21.1 129b
Venous thrombosis Mayo 4.9 5.6 9.8 16.2 282
Malmo
Worcester
9.7 10.3 21.7 42.9 189
0 1.1 6.2 6.9 54 b
Pulmonary embolism Mayo 4.1 6.6 13.7 26.9 365
NHS 1.3 1.8 3.2 m 344
aCases per 10,000 women per year. Rates from Worcester include only first events; rates from Malmo and the Nurses' Health Study (NHS) may include patients with recurrent thromboses or emboli. bThe numbers of cases from Worchester were calculated from data in the paper.
death were cancer, stroke, or age greater than 75 years at the initial diagnosis. Chronic symptoms of pain, swelling, or discoloration in the affected leg were reported by 42%, and a recurrence was diagnosed in 15% of patients. These follow-up studies indicate that chronic symptoms and recurrence are common after a first VTE even in patients who receive a full 6-month course of anticoagulation. Little else is known about the distribution of VTE in the general population. A study of all 23,000 VTE hospital discharges in California from 1991 through 1994 found that about 75% of cases are idiopathic and 25% are secondary to cancer, surgery, trauma, or another medical event (18). This study also revealed that the risk among African Americans was higher than among Caucasians for idiopathic events (relative risk [RR], 1.3; 95% confidence interval [CT], 1.1-1.5). The risk for Hispanics for idiopathic events was lower (RR, 0.6; 95% CI, 0.5-0.7) as was the risk for Asians (RR, 0.26; 95% CI, 0.2-0.3). The risk for secondary VTE was similar for Caucasians, African Americans, and Hispanics but was lower for Asians.
A. Risks o f V T E w i t h C o n g e n i t a l or A c q u i r e d Causes o f H y p e r c o a g u l a b i l i t y Familial clustering of VTE has allowed the investigation and recognition of several inherited disorders of coagulation that are associated with a dramatically increased risk of VTE. In population-based studies, however, the risk of thrombosis in the presence of specific genetically determined abnormalities is sometimes lower than the risk as estimated in studies of affected families; this suggests that for certain abnormalities, members of strongly affected families may have additional, unrecognized abnormalities that generate higher risks than seen in otherwise unselected individuals with the same defect (19). Bearing in mind these discrepancies, the frequency of each of the major genetic predispositions and estimates of the associated risk will be presented.
The prevalence of protein C deficiency has been estimated in large studies of blood donors to be about 0.2% to 0.4% (20). Among patients experiencing a first VTE episode, the prevalence is about 3% and is even higher among patients with recurrent VTE. The relative risk of VTE from family studies is about 8 and from a populationbased study is 6.5, which shows good agreement (21-23). DVT of the leg is the most common manifestation, and one-half of predisposed individuals will not experience a first clinical event until after age 40 years. The prevalence of protein S deficiency in a study of Scottish blood donors was 0.2% (24). Among patients with a first VTE episode, the prevalence of protein S deficiency is about 1% to 2% and seems to be higher among patients with recurrent thrombosis (23). There may be phenotypic variations in protein S deficiency, and the relative risks for VTE for individuals with this class of clotting abnormality have not been precisely quantified. Antithrombin III deficiencies are rare, occurring in only 0.02% of unselected blood donors (20), in whom the relative risk for VTE may be as high as 50 (23,25), and the first VTE event will often occur before age 25 years, which makes this the most serious of the inherited predispositions. The most common of the inherited predispositions may be APC resistance due to a point mutation in factor V (26). The population prevalence is between 3% and 6% and is greater among Caucasians than in other racial groups (27). The RR of VTE in individuals with this mutation is about 8 compared with the unaffected population, and the risk may be substantially higher for homozygotes (28). A first event in an affected individual may occur at any age. Altogether, about 30% of individuals with a first VTE event occurring in the absence of a predisposing clinical situation will have one of these underlying genetically determined abnormalities of the coagulation system. Another recently described mutation is a prothrombin variant from G to A at position 20210, which is present in 1% to 4% of European populations and leads to an increase in prothrombin levels (29). This variation may be associated with a threefold increase in VTE (12). The APC
CHaeTER 36 Risk of Pulmonary Embolism/Venous Thrombosis
495 TABLE 36.2 Risk Factors for V T E in Hospitalized Women Aged >40 Years Old
resistance due to the Leiden mutation is the most common of these genetic abnormalities identified thus far (9).
B. Risks of VTE with Clinically Identified Risk Factors The strongest risk factors for V T E have been readily identified clinically without the use of laboratory or epidemiologic studies to establish risk. Immobilization with or without injury is the hallmark of these risk factors. V T E associated with any of these antecedent factors is generally referred to as secondary VTE because the clot is presumed to have arisen as a result of a specific precipitating event. In most series of consecutive cases, from 40% to 80% of all subjects are considered to have a V T E that is secondary to a clinically recognized precipitating event. In studies to assess more subtle causes of VTE, those V T E patients with any of these major risk factors are excluded. Due to this approach, it is essentially unknown how cofactors might increase risk in the large proportion of V T E patients who have a precipitating factor. Studies of clinical risk factors have not presented separate analyses regarding menopausal women, but there is no a priori reason to suspect important differences in risk factors by age or sex. Overall, there is an increase in V T E risk after age 40 years, but much of this increase is probably due to the increased prevalence of the underlying medical risk factors with increasing age. Many V T E events occur after hospitalization for another problem and are not themselves the primary reason for the hospitalization. Risk of V T E occurring in patients who are already hospitalized has been well characterized. A risk classification for women over age 40 modified from the Thromboembolic Risk Factors ( T H R I F T ) Consensus Group (30) is presented in Table 36.2. Women who fall into the low-risk group do not warrant prophylactic anticoagulation, but for those in the moderate- or high-risk groups routine prophylaxis is advisable. A meta-analysis of the incidence of D V T following general surgery indicates that effective prophylactic measures include low-dose heparin, graduated elastic compression stockings, and intermittent pneumatic compression (31). In the P I O P E D study several risk factors were present among women who were referred for evaluation for possible PE (3), these included surgery within 3 months before the onset of symptoms (38%), immobilization (35%), malignancy (28%), a history of previous phlebitis (18%), stroke (7%), and trauma (7%). After evaluation, these factors were associated with a 50% or greater increased risk of a positive angiogram among the women in the study population. An increased risk of either deep vein thrombosis or pulmonary embolus has been identified in such patients in numerous studies (32).
Risk
Reason for hospitalization
<10%
Minor surgery (<30 min) Minor trauma or medical illness Minor trauma, surgery, or illness with past VTE history Surgery (>30 min), including urologic, gynecologic, general, cardiothoracic, or neurologic surgery Cancer, heart, or lung disease Major trauma or burns Major surgery, illness, or trauma with past history of VTE Fracture or orthopedic surgery of pelvis, hip, or leg Pelvic or abdominal surgery for cancer Lower limb paralysis (e.g., stroke) or amputation
10-40%
>40%
Adapted from THRIFT Consensus Group (30).
C. Other Risk Factors for Primary or Idiopathic VTE Few studies have evaluated causes of V T E after excluding cases with strong precipitating factors. The main exposures of interest have been weight, smoking, chronic medical conditions, and use of exogenous hormones. Use of hormones will be discussed separately later. Some studies have identified modestly increased risks for V T E in women with chronic medical conditions such as hypertension, diabetes mellitus, and gallbladder disease (33,34). Because evaluation of hormone use has often been a primary goal of the analyses, women with these conditions have often been excluded in order to control for confounding. Overall, evaluation of risk associated with the common chronic medical conditions has not been illuminating. Obesity, when defined as a BMI, has generally been found to be a risk factor for VTE. Obesity had no effect on V T E risk only in the Walnut Creek Contraceptive Drug Study, in which obesity was defined as weight 15% above the cohort mean (35), this study included 38 cases, of which 23 were 40 years or older. In the Nurses' Health Study there were 125 women who reported an idiopathic pulmonary embolism. Among these women, obesity (defined as a BMI > 29 kg/m 2) was associated with a relative risk of 2.9 (95% CI, 1.5-5.4). Most of these cases were older than 40 years, and the increased risk associated with obesity held for all ages (16). In a U.K. cohort that included 292 female V T E cases aged 50 to 79 years, a BMI of 26 kg/m 2 or greater had an RR for V T E of 2.0 (95% CI, 1.4-2.9) compared with
496
CAROLYN WESTHOFF
women with a BMI of 25 kg/m 2 or less (34). Large studies of OCs and thrombosis included younger study populations, but all have identified obesity as a risk factor for VTE (36-38). The evidence supports an increased risk of VTE in obese women in the menopausal age group as well as in younger women. There are no data concerning VTE and obesity in men. The data for smoking are less consistent. Studies in younger women find little or no increased risk of VTE in current smokers (36-38). The large U.K. cohort of women aged 50 to 79 years identified an RR of 1.2 (95% CI, 0.9-1.7) for V T E among current smokers compared with never smokers (34). Among the 125 idiopathic PE cases from the NHS, there was an RR of 1.9 (95% CI, 0.9-3.7) for women who smoked 25 to 34 cigarettes daily and an RR of 3.3 (95% CI, 1.7-6.5) for women who smoked 35 or more cigarettes daily compared with nonsmokers (16). In sum, cigarette smoking is not an important explanatory variable for V T E in women, regardless of age.
D. Hormonal Risk Factors for Venous Thromboembolism Case reports followed very soon by epidemiologic studies showed that use of the old, high-dose OCs were associated with an increased risk of venous thrombosis, including pulmonary embolus (39). As the dose of estrogen in the OC decreased, there was a concomitant decrease in VTE risk (40). This led to the conclusion that the risk of VTE in OC users was linked to the estrogen dose, a concept that was little challenged over the past 2 decades (41). Because the dose of estrogen in H R T is so much lower than the dose in even the lowest-dose OCs, it seemed unlikely that H R T would be likely to cause any problem with venous clots (42). The original studies (Table 36.3) that attempted to assess risk of V T E in H R T users supported this line of thought. The first report came from the Boston Collaborative Drug Surveillance Program (43) and reported on 18 cases of VTE. O f the cases, 14% were H R T users, and of the controls, 8% were users; if the study were larger these percentages would have indicated an increased risk. The authors correctly concluded "significant associations were not present"; however, the small number of cases and rare use of H R T meant that the study did not have statistical power to detect an association. The next report came from the Walnut Creek Contraceptive Drug Study (33,35) and included 17 cases, of whom 12% used HRT; 15% of the controls used HRT. This does not indicate any hint of increased risk, but again, due to the small number of cases and limited use of H R T the study lacked statistical power to detect an association. Both of these studies excluded the majority of cases to focus on idiopathic V T E only. In con-
TABLE 36.3 Current H R T Use and VTE Risk: The Early Negative Studies
Years
Number of cases
BCDSP (42) Nachtigall (44)
1972 1969-1978
18 30
Petitti ( 3 3 ) Devor ( 4 5 )
1969-1976 1980-1987
17 121
Study
Adjusted relativerisk (95% CI) 2.3 (0.6-8.0) ~ 0.85 (NS; CI not provided) 0.7 (0.2-2.5)* 0.6 (0.2-1.8)
*Relative risk and confidence interval (CI) not provided in original publication but subsequently estimated by Douketis (64). N S, not significant.
trast, NachtigaU (44) excluded none of the cases; in this randomized controlled trial of chronically ill, permanently hospitalized women, 84 women were randomized to an oral H R T regimen for 10 years, and 84 controls received placebo. These 168 women experienced 30 V T E events--a number that is at least 10 times greater than the largest expected number calculated from incidence rates in Table 36.1. The high rate of V T E in this population may have been due to limited mobility of the subjects during their chronic hospitalization. There are also cases of superficial thrombophlebitis included, and no requirement for confirmatory diagnostic tests. There was no indication of increased V T E risk among the women randomized to HRT; however, it is difficult to interpret this finding in this unusual population. Devor (45) also included all cases of V T E in a hospital-based case-control analysis. Based on 121 cases, there was no evidence of excess risk in the H R T users. The overall analysis showed no increase in risk associated with HRT, but when subsets of cases were excluded (such as women with a previous thrombosis) the RR increased. As is true for other consecutive, unselected series of V T E cases, most of the cases in this study were secondary to a known predisposing factor. These four early evaluations of the H R T - V T E association did not indicate an increased risk, but the first two studies were too small for a precise or sophisticated analysis, and the randomized trial included an unusual population. Finally, the larger case-control study included so many women with secondary V T E and other serious diseases that it is difficult to assess whether H R T would have been given to these women by the physicians who were caring for their medical illnesses. With the clarity of hindsight, it appears that none of these early studies had a design or number of cases appropriate to address the question.
CHAPTER 36 Risk of Pulmonary Embolism/Venous Thrombosis
E. Studies of the H R T - V T E Association Starting in 1996, a series of new large studies began to be published (Table 36.4), suggesting an increased risk of both deep vein thrombosis and pulmonary embolus among current users of HRT. These new studies include data from two large prospective cohort studies, the Nurses' Health Study (46) and the Oxford-Family Planning Association cohort (47), from an historical cohort study of the Group Health Incorporated (GHI) medical plan database (48), from an historical cohort study using record linkage with the U.K. General Practice Research Database (34), from a record linkage study using medical databases in Italy (49), from hospital-based case-control studies in the United Kingdom (50), in France (51), and a multicenter hospital-based casecontrol study (52), from a population-based case-control study in Washington State (53), and, finally, from two randomized controlled trials, the Heart and Estrogen/progestin Replacement Study (HERS) trial (54,55) and the Women's Health Initiative (56), both from the United States (54). All of these studies summarized in Table 36.4 consider just those cases with a first episode of VTE. All of the studies excluded cases and controls with risk factors for V T E (e.g., cancer, fracture, and stroke), as well as excluding cases with superficial phlebitis. All of the studies described specific diagnostic criteria that were required for inclusion as a case. TABLE 36.4
Risk of Primary V T E among Current H R T Users m Subsequent Studies Adjusted relative risk (95% CI)
Study Grodstein (58) Daly (letter)(47) Daly (47) Jick (48) Spitzer (38) Varas-Lorenzo (49) Hulley (54) Douketis (52) Cushman (56)
Number of cases 123 18 103 42
Current use
First-year use
2.1 (1.2-3.8)
Not calculated Not calculated 6.7 (2.1-21.3) 6.7 (1.5-30.8) 4.6 (2.5- 8.4) 2.9 (1.2-6.9)
2.2 (0.6-7.9)
171
3.5 3.6 2.1 2.3
46 95 147
2.9 (1.5-5.6) 1.9 (1.2-3.2) 2.1 (1.6-2.7)
Smith (53)
586
Scarabin (51)
155
1.7 (1.2-2.2) a 0.9 (0.7-1.2) b 3.5 (1.8-6.8) c 0.9 (0.5-1.6) d
292
aConjugated equine estrogen. bEsterified estrogen. COral estrogen. dTransdermal estrogen.
(1.8-7.0) (1.6- 7.8) (1.4-3.2) (1.0-5.3)
3.3 (1.1-10.1) Not calculated 4.0 (Not calculated) Not calculated 8.1 (0.9-7.4) 1.5 (0.3-6.9)
497 Mthough the methodology differed and the stringency of the diagnostic criteria varied somewhat between studies, the RRs shown in Table 36.4 are nonetheless extremely similar, and indicate a risk among current H R T users that is two to three times the risk among nonusers. Unlike previous smaller studies, these studies were able to assess risks in some subgroups of H R T users. The most consistent finding was that the increased risk of V T E was greatest in or limited to the first year of H R T use. Until recently, the usual approach to analysis was to look for increased risk as the time of exposure increased (e.g., the risk of lung cancer increases with increasing years of cigarette smoking). The notion of looking for an early, transient risk is more recent and an understanding of these consistent results is not yet established. Other studies indicate that the increased risk of V T E in OC users may also be early and transient (38,57). Other subgroup analyses have attempted to assess conventional markers of risk such as dose. In these analyses, there was an indication of an increased risk of V T E with higher estrogen doses in two studies (36,53), but no dose-response effect in the others that were able to assess dose (46,50,58). The studies did generally agree on little or no excess risk in past users (46,49,50,58). The populations that were studied often included only a few H R T users or a limited range of H R T regimens; therefore, the investigators had a limited ability to compare risks among regimens. In addition to the studies listed above, the Postmenopausal Estrogen/Progestin Intervention (PEPI) trial also evaluated phlebitis in women randomly assigned to five different H R T regimens; although there was a suggestion of an increased risk of V T E among the subjects receiving active treatment, this outcome was too rare to allow comparisons among the groups (59). The most recent studies have provided some evidence to distinguish the risks of opposed versus unopposed estrogen, of oral versus other mutes of administration, or between different formulations. Scarabin (51) found increased risk in users of oral estrogen but not in users of a transdermal formulation. In the Douketis study (52), few women used a transdermal hormone; among users of the oral preparations the excess risk was seen in users of estrogen plus progestin (RR 2.7, 95% CI 1.4-5.1) and not in users of estrogen alone (RR 1.2, 95% CI 0.6-2.6). Smith (53) also found an excess risk in combined rather than estrogen-only users (RR 1.6, 95% CI 1.1-2.6). In addition, Smith reported an excess risk limited to users of oral conjugated equine estrogen users (RR 1.65) and not in users of esterified estrogen (RR 0.92). The vast majority of V T E cases in women in the menopausal age group are secondary to other identifiable risk factors. The incidence of cases that might be attributed to H R T use can be calculated from the cohort studies. In the G H I cohort there were about two extra V T E cases per 10,000 H R T users per year (48). An estimate based on U.K. incidence data similarly suggested about two additional cases per
498 10,000 H R T users per year (50). In the HERS study the excess risk was about 7 cases per 1000 in year 1 and about 2 cases per 1000 in years 4 and 5 (54); however, women were selected for that randomized trial based on diagnosed cardiovascular disease, and they appear to have a much higher baseline risk for V T E than do women in the general population. The HERS results do suggest that women with an increased baseline risk of V T E may experience many more cases of V T E if using H R T than is seen among low-risk women. Moreover, Hoibraaten (60) carried out a randomized trial of H R T use in women with a history of VTE; the trial was stopped because of an excess risk of recurrent V T E among the women randomized to H R T (10.7% versus 2.3% among placebo users). This further substantiates the notion that women at high risk of V T E should avoid HRT. Data from the Women's Health Initiative ( W H I ) are comparable to the findings above, showing approximately a twofold increase in risk (see Chapters 45 and 46). O f interest, in W H I there appeared to be a greater risk among women treated with estrogen plus progestogen compared with hysterectomized women on estrogen alone, despite a higher risk status in the latter group. These data suggest but do not prove a greater risk with added progestogens. Some data are accumulating about the risk of V T E in women using the other, newer selective estrogen receptor modulators. Tamoxifen has the longest and widest use of these and has been associated with increased risk of V T E at a magnitude similar to the studies of estrogen (61). Tibolone, a nonestrogen treatment for hot flushes, was evaluated in some of the European studies with RRs for V T E slightly lower than those seen for estrogen; however, the number of users of tibolone in these studies was small and therefore the RR estimate thus far must be considered imprecise (46). Raloxifene use in placebo-controlled clinical trials has been associated with an RR of V T E of about 3.0 (62), and this excess is noted in the product labeling (63).
V. CLINICAL RECOMMENDATIONS Venous thromboembolism is a common problem among women in the menopausal age group and is most likely to occur among women with predisposing medical problems. In large part, V T E increases with age because the predisposing problems become more common with advancing age. Among healthy, low-risk women, even after age 40, the risk of V T E is probably about 1 to 2 new cases per 10,000 women per year. In this population, the risk of V T E for women using H R T may increase two- to threefold from this low level, yielding somewhere between 1 and 6 additional cases per 10,000 H R T users. This is comparable both in relative and absolute terms to the increased risk seen among younger women who use OCs. As with younger women and OCs, this risk of V T E in H R T users needs to be communicated to potential H R T
CAROLYNWESTHOFF users so that they can weigh this along with all other risks and benefits in making a decision to use HRT. The V T E risks may be smaller with the use of estrogen alone, with the use of esterified rather than conjugated estrogens, and with the use of transdermal rather than oral preparations; however, each of these associations needs further substantiation. Among menopausal women with chronic medical problems, with a previous DVT, with an inherited predisposition, or among women who will be having surgery or hospitalization for some other reasons, the baseline risk of a V T E is substantially higher. The results from the HERS trial (47) indicate that the women in that trial with known heart disease had a baseline risk of V T E of at least 20/10,000 women per year and that H R T use in these women increased that risk threefold. The proportional increase was the same as that seen in low-risk women, but because their baseline risk of V T E was higher, this increase in risk may translate into 20 to 40 additional cases per 10,000 users per year. Because most of the studies agree that the excess risk is concentrated in the first 1 to 2 years of H R T use, this information is most important for women who are considering whether to initiate HRT. Although there are no specific relevant data regarding the risk of V T E among hospitalized H R T users, it appears prudent to consider discontinuing oral H R T in women who are going to undergo major surgery or who develop a medical problem that is associated with immobilization or an other disease associated with a high risk of VTE. Among women who are currently being treated with anticoagulants, there are no data to indicate whether H R T use would modify their risk of a recurrent event. References 1. Ault M, Artal R. Upper extremityDVT: What is the risk?Arch Intern Med 1998;158:1950-1951. 2. AndersonFA Jr, Wheeler HB, Goldberg RJ, et al. A population-based perspective of the hospitalincidence and case-fatalityrates of deep vein thrombosis and pulmonaryembolism.The Worcester DT Study.Arch Intern Med 1991;151:933-938. 3. Q.uinnD, Thompson B, Terrin M, et al. A prospectiveinvestigation of pulmonary embolism in women and men.J A m MedAssoc 1992;268: 1689-1696. 4. PrandoniP, LensingAW, Cogo AJ, et al. The long-term clinicalcourse of acute deep venous thrombosis.Ann Intern Med 1996;125:1-7. 5. Baker W. Diagnosis of deep vein thrombosis and pulmonary embolism. Curr Concepts Thromb 1998;82:475-495. 6. Merli G. Diagnostic assessment of deep venous thrombosis and pulmonary embolism.Am J Med 2005;118:3S- 12S. 7. Haas S.Treatment of deep venousthrombosisand pulmonaryembolism. Current recommendations.Curr Concepts Thromb 1998;82:495-510. 8. DexterL. President'sAddress:The chair and venous thrombosis. Tram Am Clin ClimatolAssoc 1973;84:1-15. 9. Dahlback B, HiUarp A, Rosen S, Zoller B. Resistance to activated protein C, the FV.'Qs~ allele, and venous thrombosis. Ann Hematol 1996;72:166-176. 10. Bertina R, KoelemanB, KosterT, et al. Mutation in blood coagulation factorV associatedwith resistanceto activatedprotein C. Nature (London) 1994;369:64-67.
CrtheVE~ 36 Risk of Pulmonary Embolism/Venous Thrombosis 11. Lowe GDO, Rumley A, Woodward M, Reid E. End of the line for "third-generation" pill controversy? Lancet 1996;349:1113-1114. 12. Poort S, Rosendaal F, Reitsma P, Bertina R. A common genetic variation in the 3'-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996;88:3698-3703. 13. Coon W, Willis P, Keller J. Venous thromboembolism and other venous disease in the Tecumseh community health study. Circulation 1973;47:839-846. 14. Silverstein MD, Heit JA, Mohr DN, et al. Trends in the incidence of deep vein thrombosis and pulmonary embolism. Arch Intern IVied 1998;158:585-593. 15. Nordstrom M, Lindblad B, Bergqvist D, Kjellstrom T. A prospective study of the incidence of deep-vein thrombosis within a defined urban population. J Intern Med 1992;232:155-160. 16. Goldhaber S, Grodstein F, Stampfer M, et al. A prospective study of risk factors for pulmonary embolism in women. J Am Med Assoc 1997;277:642-645. 17. Beyth R, Cohen A, Landefield C. Long-term outcomes of deep-vein thrombosis. Arch Intern Med 1995;155:1031-1037. 18. White R, Zhou H, Romano P. Incidence of idiopathic deep venous thrombosis and secondary thromboembolism among ethnic groups in California. Ann Intern Med 1998;128:737-740. 19. Rosendaal E Risk factors for venous thrombosis: prevalence, risk, and interaction. Semin Hemato11997;34:171 - 187. 20. Tait RC, Walker ID, Reitsma PH, et al. Prevalence of protein C deficiency in the healthy population. Thromb Haemostasis 1995;73:87-93. 21. Allaart C, Briet E. Familial venous thrombophilia. In: Bloom A, Forbes C, Thomas D, et al, eds. Hemostasis and thrombosis. New York: Churchill-Livingstone, 1994:1349-1360. 22. Bovill EG, Bauer KA, Dickermann JD, Callas P, West B. The clinical spectrum ofheterozygous protein C deficiency in a large New England kindred. Blood 1989;73:712-717. 23. Koster T, Rosendaal FR, Briet E, et al. Protein C deficiency in a controlled series of unselected outpatients: an infrequent but clear risk factor for venous thrombosis (Leiden thrombophilia study). Blood 1995;85:2756- 2761. 24. Beauchamp NJ, Dykes AC, Parikh N, Tait RC, Daly ME. The prevalence of, and molecular defects underlying, inherited protein S deficiency in the general population. BrJ Haemotology 2004;125:647-654. 25. Hiejboer H, Brandjes DP, Buller HR, Sturk A, Ten Cate JW. Deficiencies of coagulation-inhibiting and fibrinolytic proteins in outpatients with deep-vein thrombosis. N EnglJ Med 1990;323:1512-1516. 26. Dahlback B, Carlsson M, Svensson R Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: prediction of a cofactor to activated protein C. Proc NatlAcad Sci U S i11993;90:1004-1008. 27. Ridker PM, Hennekens CH, Lindpainter K, et al. Mutation in the gene coding for coagulation factor V and the risk of myocardial infarction, stroke, and venous thrombosis in apparently healthy men. NEngl JMed 1995;332:912-917. 28. Vandenbroucke J, Helmerhorst E Risk of venous thrombosis with hormone-replacement therapy. Lancet 1996;348:972. 29. Rosendaal FR, Doggen CJM, Zivelin A, et al. Geographic distribution of the 20210 G to A prothrombin variant. Thromb Haemost 1998;79: 706- 708. 30. THRIFT Consensus Group. Risk of and prophylaxis for venous thromboembolism in hospital patients. Br MedJ 1992;305:567-574. 31. Clagett G, Reisch J. Prevention of venous thromboembolism in general surgical patients. Results of recta-analysis. Ann Surg 1988;208: 227-240. 32. Salzman E, Hirsch J. The epidemiology, pathogenesis, and natural history of venous thrombosis. In: Coleman R, Hirsch J, Marder V, Salzman E, eds. Hemostasis and thrombosis: bas# principles and clinicalpractice. Philadelphia: Lippincott, 1994:1275-1296.
499 33. Petitti D, Wingerd J, Pellegrin F, Ramcharan S. Risk of vascular disease in women: smoking, oral contraceptives, noncontraceptive estrogens, and other factors. J Am Med Assoc 1979;242:1150-1154. 34. Gutthann SP, Garcia Rodriguez LA, Castellsague J, Oliart AD. Hormone replacement therapy and risk of venous thromboembolism: population based case-control study. Br MedJ 1997;314:796-800. 35. Petitti D, Wingerd J, Pellegrin F, Ramcharan S. Oral contraceptives, smoking, and other factors in relation to risk of venous thromboembolic disease. Am J Efidemio11978;108:480-485. 36. Jick H, Jick S, Gurewich V, Myers MW, Vasilakis C. Risk of idiopathic cardiovascular death and nonfatal venous thromboembolism in women using oral contraceptives with differing progestogen components. Lancet 1995;346:1589-1592. 37. World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease. Lancet 1995;346:1582-1588. 38. Spitzer WO, Lewis MA, Heinemann LA, Thorogood M, MacRae KD. Third generation oral contraceptives and risk of venous thromboembolic disorders: an international case-control study. Br MedJ 1996; 312:83-88. 39. Inman WH, Vessey MP, Westerholm B, Engelund A. Thromboembolism disease and the steroidal content of oral contraceptives: a report to the Committee on Safety of Drugs. Br MedJ 1970;2:203-209. 40. Gerstman BB, Piper JM, Tomita DK, et al. Oral contraceptive estrogen dose and the risk of deep venous thromboembolic disease. Am JEfidemio11991;133:32-37. 41. Realini J, Goldzieher J. Oral contraceptives and cardiovascular disease: a critique of the epidemiological studies. _//mJ Obstet Gyneco11985;152: 729-798. 42. Lobo R. Estrogen and the risk of coagulopathy. Am J Med 1992; 92:283-285. 43. Boston Collaborative Drug Surveillance Program. Surgically confirmed gallbladder disease, venous thromboembolism, and breast tumors in relation to postmenopausal estrogen therapy. N EnglJ Med 1974;290:15-19. 44. Nachtigall L, Nachtigall R, Nachtigall R, Beckman E. Estrogen replacement therapy II: a prospective study in the relationship to carcinoma and cardiovascular and metabolic problems. Obstet Gyneco11979;54:74-79. 45. Devor M, Barred-Connor E, Renvall M, Feigal D Jr, Ramsdell J. Estrogen replacement therapy and the risk of venous thrombosis. Am J Med 1992;92:275-282. 46. Grodstein F, Stampfer M, Manson I, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N EnglJ Med 1996;335:453-461. 47. Daly E, Vessey MP, Painter R, Hawkins M. Case-control study of venous thromboembolism risk in users of hormone replacement therapy. Lancet 1996;348:1027. 48. Jick H, Derby L, Myers M, Vasilakis C, Newton K. Risk of hospital admission for idiopathic venous thromboembolism among users of postmenopausal oestrogens. Lancet 1996;348:981-983. 49. Varas-Lorenzo C, Garcia-Rodriguez L, Cattaruzzi C, et al. Hormone replacement therapy and the risk of hospitalization for venous thromboembolism: a population-based study in Southern Europe. Am J E f i demio11998;147:387-390. 50. Daly E, Vessey MP, Hawkins MM, et al. Risk of venous thromboembolism in users of hormone replacement therapy. Lancet 1996;348: 977-980. 51. Scarabin P, Oger E, Plu-Bureau G, on behalf of the EStrogen and THromboEmbolism Risk (ESTHER) Study Group. Differential association of oral and transdermal oestrogen-replacement therapy with venous thromboembolism risk. Lancet 2003;362:428-432. 52. Douketis JD, Julian JA, Kearon C, et al. Does the type of hormone replacement therapy influence the risk of deep vein thrombosis? A prospective case-control study. J Thromb Haemostasis 2005;3:943-948.
500 53. Smith NL, Heckbert SR, Lemaitre RN, et al. Esterified estrogens and conjugated equine estrogens and the risk of venous thrombosis. J A m MedAssoc 2004;292:1581-1587. 54. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. J A m MedAssoc 1998;280:605-613. 55. Grady D, Wenger NK, Herrington D, et al., for the Heart and Estrogen/ Progestin Replacement Study Research Group. Postmenopausal hormone therapy increases risk for venous thromboembolic disease. The Heart and Estrogen/Progestin Replacement Study. Ann Intern Med 2000;vl 32:689 - 696. 56. Cushman M, Kuller LH, Prentice R, et al., for the Women's Health Initiative Investigators. Estrogen plus progestin and risk of venous thrombosis. J A m MedAssoc 2004;292:1573-1580. 57. Lidegaard O. Thrombotic diseases in young women and the influence of oral contraceptives. Am J Obset Gyneco11998;179:$62- $67. 58. Grodstein F, Stampfer M, Goldhaber S, et al. Prospective study of exogenous hormones and risk of pulmonary embolism in women. Lancet 1996;348:983-987.
CAROLYNWESTHOFF 59. Writing Group for the PEPI Trial. Effects of estrogen or estrogen/ progestin regimens on heart disease risk factors in postmenopausal women.JAm MedAssoc 1995;273:199-208. 60. Hoibraaten E, Qyigstad E, Arnesen H, et al. Increased risk of recurrent venous thromboembolism during hormone replacement therapy. Thromb Haemostasis 2000;84:961 - 967. 61. Fisher B, Costantino J, Redmond C, et al. A randomized clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have estrogen-receptor-positive tumors. N Engl J Med 1989;320:479-484. 62. Delmas P, Mitlak B, Christiaansen C. Effects of raloxifene in post menopausal women (letter). N EnglJ Med 1998;338:1313 - 1314. 63. Eli Lilly and Company. Evista product information. Indianapolis, IN: Eli Lilly and Company, 1998. 64. Douketis JD, GinsbergJS, Holbrook A, et al. A reevaluation of the risk for venous thromboembolism with the use of oral contraceptives and hormone replacement therapy. Arch Intern Med 1997;157:1522-1530.
;HAPTER 3,
Glucose Metabolism After Menopause SVEN O . S KOUBY Department of Obstetrics and Gynecology, Frederiksberg Hospital, University of Copenhagen, Denmark STEN MADSBAD
Department of Endocrinology, Hvidovre Hospital, University of Copenhagen, Denmark
I. I N T R O D U C T I O N
This chapter discusses the impact of endogenous and exogenous sex steroids on the age-dependent decrease in glucose tolerance in normal women and the glycemic and metabolic control of women with type 2 diabetes.
Changes in the environment and in human behavior and lifestyle, in conjunction with genetic susceptibility, have resulted in a dramatic increase in the incidence and prevalence of diabetes worldwide. It has been estimated that there will be a 42% increase in cases of diabetes in developed countries, from 51 million to 72 million; and a 170% increase in developing countries, from 84 million to 228 million (1). The prevalence of the metabolic syndrome in nondiabetic adults in Europe has been reported to be as high as 15%. The rapid escalation of the number of people with type 2 diabetes and diabetes-related cardiovascular disease demands urgent action, focused on prevention. An important rule of any prevention strategy is the identification of individuals who are at high risk. In women, increased body weight and central body fat distribution are often observed throughout the climacteric period, and the prevalence of metabolic syndrome increases dramatically after the age of 40 (2). It has been reported that the decline in insulin sensitivity following the menopause relates largely to age rather to menopause itself (3). A large communitybased population survey indicated that in postmenopausal women, the years since menopause confers a negative influence on glucose tolerance, and the risk of impaired glucose tolerance increases by 6% for each year after menopause (4). TREATMENT OF THE POSTMENOPAUSAL WOMAN
II. T H E METABOLIC SYNDROME The metabolic syndrome is also known as the insulin resistance syndrome (5) and as syndrome X (6). The cluster of metabolic abnormalities includes glucose intolerance, insulin resistance, central obesity, dyslipidemia, and hypertension, all well-documented risk factors for cardiovascular disease (7). There is wide variation in the prevalence in men and women. In global studies, the prevalence varies in urban populations from 8% to 24% in men and from 7% to 43% in women. A very consistent finding is that the prevalence of the metabolic syndrome is highly age dependent. The prevalence of the metabolic syndrome in the United States (National Health and Nutrition Examination Survey [NHANES III]) increased from 7% in participants aged 20-29 years to 44% and 42% for those aged 60-69 years and those at least 70 years old, respectively (8). The metabolic syndrome is associated with an increased risk of both diabetes (9) and cardiovascular disease (10) (Fig. 37.1). The most accepted and unifying hypothesis to describe the pathophysiology of the 501
Copyright 9 2007by Elsevier,Inc. All rightsof reproductionin anyformreserved.
502
SKOUBY AND MADSBAD
FICURE 37.1 permission.)
The metabolic syndrome increases cardiovascular risk. (From ref. 10, with
metabolic syndrome is the existence of insulin resistance. A major contributor to the development of insulin resistance is an overabundance of circulating fatty acids. The association of the metabolic syndrome with inflammation also is well documented (11). The increases in pro-inflammatory cytokines including interleukin (IL)-6, resistin, tumour necrosis factor-el (TNF-ci) and C-reactive protein reflect overproduction by the expanded adipose tissue mass, particularly the increase in visceral fat. Adiponectin is an anti-inflammatory cytoldne that is produced exclusively by adipocytes. Adiponectin both enhances insulin sensitivity and inhibits many steps in the inflammatory process (12). An alternative concept to explain the metabolic syndrome is leptin resistance (13). Abnormalities of both low adiponectin and high leptin have been documented in obesity and diabetes as well as in postmenopausal women with metabolic syndrome (13a). Several investigative techniques have been devised for the in vivo assessment of resistance to insulin-stimulated glucose uptake. All have limitations, and none are suitable for routine clinical use. The euglycemic clamp technique, which involves simultaneous infusions of insulin and glucose, has been used extensively and is considered an important standard. However, the method is complex, somewhat invasive, and resource demanding. A computer (minimal model) modeling of an intravenous glucose tolerance test (IVGTT) is easier to carry out, although the differential equations necessary to assess simultaneous changes in glucose and insulin concentrations are complex. Many consider this to be the gold standard technique. Alternative methods involve comparison of glucose and insulin changes after a standard oral glucose tolerance test (13b), the intravenous insulin tolerance test (13c), and the use of fasting blood alone to assess the glucose/insulin ratio or calculations of Homeostatic Model Assessment (HOMA) (13d) and The Q.uantative Insulin Check Index (QUICKI) (13e). All these techniques have been found to correlate well with each other in the absence of overt diabetes (14).
A. Sex Steroids and Carbohydrate Metabolism Several conditions suggest important relationships between insulin sensitivity and sex steroids. Examples include the increased insulin resistance during the high estrogen and progesterone state of pregnancy, during the luteal phase of the menstrual cycle, and in women with high androgen production (e.g., polycystic ovarian syndrome [PCOS], stromal hyperthecosis). In the context of sex steroids, it is also of interest that during the reproductive years, insulin sensitivity to glucose is greater in women than in men. Levels of Sex Hormone Binding Globulin (SHBG) reflect in part endogenous estrogen status, with higher levels of SHBG correlating with higher estrogen status. The San Antonio Heart Study demonstrated that SHBG is negatively correlated with fasting insulin and insulin and glucose responses to oral glucose (15). Thus, the metabolic effects of endogenous estrogen suggest that estrogen replacement therapy may have the potential to reverse the cluster of metabolic changes included in the metabolic syndrome. Some discrepancies, however, exist because of the difference between the physiologic and pharmacologic actions of endogenous and exogenously administered sex steroids. Exogenous sex steroids with estrogen and progestogen represent a wide spectrum of hormones with different molecular structures and also differences in their pharmacokinetic properties and pharmacodynamic effects. Estrogens may be natural or equine from genuine sources or artificial with or without a steroid structure. The progestogens represent an even wider array of hormonal compounds with progesterone-like activity and some having partial androgen, antiandrogen, mineralocorticoid, or antimineralocorticoid activities (16). Moreover, the sex steroids are available in various formulations such as patches, implants, and various vaginal preparations, which have no first-pass hepatic effect. Such differences in pharmacokinetic properties will also influence the metabolic impact of hormonal treatment. In animal studies it has been shown that 17[3-estradiol (E2)
CHAPTER 37 Glucose Metabolism After Menopause treatment enhances insulin secretion in islet cells and thereby augments the insulin response to glucose and reduces insulin resistance (17). Also, natural progesterone may result in an increased insulin response to glucose, probably due to decreased insulin sensitivity in target tissues resulting in increased beta cell responsiveness. Much of our knowledge about the effect of the exogenous sex steroids on glucose metabolism in women is derived from studies using oral contraceptives (OCs). When administering OCs, often there is a decrease in glucose tolerance, concomitant with a hyperinsulinemic response to the glucose load. The impact on glucose metabolism has been shown to be clearly dose dependent, and the low-dose types of combined OCs have far less of an impact on glucose tolerance; most often there is no deterioration in the glucose curve whereas hyperinsulinemia and decreased insulin sensitivity can still be observed (18). Several investigations on hormonal administration and insulin sensitivity and glucose tolerance have been reported, but the conclusions are not uniform. Some of these discrepancies may be due to differences in the routes of administration. Following oral administration, hormones pass directly from the gut to the liver via the portal vein, giving a high local concentration that markedly affects hepatic metabolism. In contrast, transdermal administration avoids first-pass metabolism, and it is therefore not surprising to find different effects with oral compared with parenteral administration. Barret-Connor and Laakso have published results from a comparative study on post-challenge glucose and insulin levels in women using no hormonal therapy (HT), unopposed conjugated estrogen (CEE), and CEE plus medroxyprogesterone acetate (MPA) (19). The glucose and insulin values were adjusted for age and body mass index (BMI) before comparisons were made. Fasting glucose levels were significantly lower in the estrogen-treated group compared with the untreated group. No difference was seen between the CEE + MPA group compared with the group with unopposed CEE treatment, but fasting insulin values were lower in the estrogen-treated group compared with the group with no treatment. In contrast, Cagnacci et al. (20) found no significant effect on fasting insulin levels or beta cell sensitivity to oral glucose with CEE. This group compared the oral intake of CEE with transdermal administration of E2. Following the transdermal administration, fasting insulin levels decreased and the beta cell responses to glucose increased, indicating an increase in insulin sensitivity. The study is interesting because it evaluated the significance of the estrogen effect on the liver for glucose homeostasis, and in contrast to oral CEE, transdermal estrogen increased hepatic insulin clearance. Godsland et al. (21) have compared insulin sensitivity using the IVGTT, "minimal model" technique during the sequential intake of oral CEE plus levonorgestrel (LNG) and transdermal administration of E2 plus oral norethisterone (NET). Oral intake of
503 C E E + L N G caused a decrease in glucose tolerance and an overall increased plasma insulin response. Insulin resistance was greater during the combined phase compared with the estrogen-only phase. The transdermal regimen had relatively few effects on insulin metabolism, although the first-phase beta cell insulin secretion was enhanced. Both regimens increased hepatic insulin uptake. This study therefore suggests that the addition of a progestogen to estrogen therapy may markedly influence the effect on glucose metabolism in parallel to the observations made during administration of OCs. Lindheim et al. demonstrated the impact of dose, showing a bimodal effect of oral CEE estrogens on insulin sensitivity with an improvement occurring with the lower dose of 0.625 mg but with deterioration with the dose of 1.25 mg (22). The authors suggested that this effect may be related to the first-pass hepatic-portal effect in that transdermal E2 (0.1 mg), which may be equated more closely with the larger dose of oral estrogen (1.25 mg), improved insulin sensitivity. In the Estrogen in the Prevention of Atherosclerosis Trial (EPAT), in which oral micronized E2 (1 mg) was used alone for 2 years, there was statistical improvement in glucose, insulin, and hemoglobin (Hgb) Alc (22a). Progestogens, however, appeared to attenuate the beneficial effects of transdermal estrogen and may alter the clearance of insulin (23). This observation may be related to the dose and type of progestogen. Duncan et al. (24) found no beneficial effect of unopposed oral and transdermal E2 on insulin sensitivity and suggested that the addition of an oral progesterone (NET) confers no clinically important risk or benefit. Gaspard et al. (25) found that after 2 years of treatment with oral E2 combined sequentially with dydrogesterone (DG), presumably a metabolically more neutral progesterone deriviate, the oral glucose tolerance test (OGTT) was unaffected, but there was a decreased area under the insulin curve as well as fasting insulin values indicating an improvement in insulin sensitivity. Using transdermal E2 and the same progestogen, Cucinelli et al. (26) demonstrated with the euglycemic clamp technique an amelioration of insulin resistance after 12 weeks of treatment. Using both a high-dose (2 mg) and low-dose (1 mg) oral regimen of E2 combined with DG for 24 months, Godsland et al. (27), applying the minimal model, found favorable effects on insulin secretion and elimination in menopausal women. When comparing oral E2 (2 mg) plus DG with tibolone in obese women (BMI -> 27), Morin-Papunen et al. (28) found that following 12 months of tibolone, whole-body glucose uptake decreased, contrasted with the effect of oral E2 + DG using the euglycemic technique. Comparing also oral tibolone with oral CEE (0.625 mg) plus MPA (5 mg) Wiegratz et al. (29) found no significant changes with either treatment in the results obtained from OGTTs performed at baseline and after 6 months in nonobese women. We have examined the impact on glucose metabolism and insulin sensitivity from oral administration of 2 mg E2 in combination with the intrauterine delivery
504
SKOUBY AND MADSBAD
system o f L N G (Mirena) and compared the results with data obtained with a continuous combined treatment with 2 mg E2 plus 1 mg Norethindrone Acetate (NETA). After 6 months, no changes were observed in the OGTTs and insulin sensitivity using the minimal model technique. One of the largest randomized comparative studies on H T and risk markers for a deterioration of glucose metabolism was the Postmenopausal Estrogen/Progestin Intervention Study (PEPI), a 3-year trial to assess effects of four hormone regimens on cardiovascular risk factors (30). In 788 women, the impact of the hormone regimens was determined on insulin and glucose concentrations measured during a standard OGTT. When compared with women taking placebo, those taking CEE at 0.625 mg/day with or without a progestational agent had lower mean fasting insulin levels, lower mean fasting glucose levels, and higher mean 2-hour glucose levels. No significant differences were apparent between women taking CEE plus progestin regimens: MPA (2.5 mg daily), continuous CEE plus MPA, CEE and MPA at 10 mg on days 1-12, and micronized progesterone (MP) cyclically at 200 mg on days 1-12. The impact of H T on insulin and glucose depended on baseline levels of fasting insulin and 1-hour glucose. However, the treatment effects on carbohydrate metabolism appeared to be consistent across participant subgroups. The interpretation was that oral H T involving 0.625 mg/day of CEE may modestly decrease fasting levels of insulin and glucose. Post-challenge glucose concentrations are increased, which may indicate delayed glucose clearance. In WHI, there was a reduction in glucose and insulin in women receiving CEE 0.625 mg with MPA 2.5 mg (30a). H O M A was significantly decreased at 1 year (p < 0.05) but returned to baseline thereafter (Fig. 37.2). On the contrary, with this same regimen, sites showed no improvement and
a 17% reduction in insulin sensitivity at 6 months, which persisted for 2 years using the englycemic hyperinsulinemic clamp technique. In summary, oral estrogen (at low to moderate doses) seems to result in a mild improvement in insulin sensitivity in most but not all studies. Transdermal estrogen may be more favorable in this regard, particularly in some women. The addition of progestogens may or may not attenuate these responses based on the dose and type of progestogens. Finally, the age and metabolic status of the women studied may influence the findings.
III. HORMONAL THERAPY AND DIABETES Only a few studies have evaluated the effect of H T on diabetes incidence. In the Heart and Estrogen/Progestin Replacement Study (HERS), 2763 postmenopausal women with documented coronary heart disease (CHD)were randomly assigned to daily estrogen plus progestogen therapy or to placebo (31). The effect of H T on fasting glucose levels and incident diabetes over 4 years of follow-up was evaluated. A total of 2029 women without diabetes were included. Incident diabetes was defined by self-report of diabetes or disease complication, fasting glucose level of 6.9 mmol/L or greater (-> 126 mg/dL), or initiation of therapy with diabetes medication. Fasting glucose levels increased significantly among women assigned to placebo but did not change among women receiving HT. The incidence of diabetes was 6.2% in the H T group and 9.5% in the placebo group (hazard ratio [HR] 0.65; 95% confidence interval [CI], 0.48-0.89), corresponding to a 35% lower incidence among H T users. The number needed to treat for benefit to prevent
FIGURE 37.2 Women's Health Initiative CEE+MPA study: change in insulin resistance, calculated from fasting glucose and insulin according to the homeostasis model assessment of insulin resistance (HOMA-IR): insulin x (glucose/22.5); values are mean _+ SE; ap < 0.05 versus placebo. (From ref. 33, with permission.)
CHAPTER 37 Glucose Metabolism After Menopause one case of diabetes was 30. Changes in weight and waist circumference did not mediate this effect. Obviously, evaluation of the preventive effect of H T on the occurrence of type 2 diabetes was not the primary objective of HERS, and only one combination (0.625 mg of CEE plus 2.5 mg of MPA) was considered. Previous research examining the effect of H T on the development of diabetes has been inconclusive, and no long-term randomized clinical trial data have been available. The findings, however, are in accordance with the observational findings of the Nurses' Health Study, in which the influence of H T on the subsequent development of diabetes was examined in a prospective cohort of 21,028 postmenopausal U.S. women aged 30 to 55 years and free of diagnosed diabetes in 1976 (32). During 12 years of followup, it was confirmed that current H T users had a relative risk of diabetes of 0.80 (95% CI, 0.67-0.96) as compared with never users, after adjustment for age and BMI. In addition, the W H I also offers new insights into how diabetes may be prevented in women during H T (33). The study was designed to examine the effect of postmenopausal hormone replacement on a number of health outcomes, with a primary end point of coronary heart disease death or nonfatal myocardial infarction. In the estrogen/progestin arm, more than 15,000 women were randomized to receive either 0.625 mg of CEE plus 2.5 mg of MPA or placebo each day during 5.6 years of follow-up. The participants were women aged 50 to 79, and all had an intact uterus. Diabetes incidence was ascertained by self-report of treatment with insulin or oral hypoglycemic medications. Fasting glucose, insulin, and lipoproteins were measured in a random sample at baseline and at 1 and 3 years. The cumulative incidence of treated diabetes was 3.5% in the H T group and 4.2% in the placebo group (HR 0.79; 95% CI, 0.67-0.93) (see Fig. 37.2). There was little change in the hazard ratio after adjustment for changes in BMI and waist circumference. During the first year of follow-up, changes in fasting glucose and insulin indicated a significant fall in insulin resistance in actively treated women compared with the control subjects (Fig. 37.3). The interpretation of the data suggests that combined therapy with estrogen and progestin reduces the incidence of type 2 diabetes.
FIGURE 37.3 Women's Health Initiative CEE+MPA study: incidence of diabetes. (From ref. 33, with permission.)
505 Thus, the study confirms the observations made in the HERS and Nurses' Health Study trials. Although the results of these studies are consistent in showing that women on H T are less likely to progress to type 2 diabetes, it is unclear if these results should favor the prescription of H T to diabetic women. In a number of smaller studies, women with diabetes benefited from HT, either in terms of glucose excursions over time (33a), improved levels of Hgb AIC (33b), or an overall marginal improvement in the metabolic profile (33c,34). The search for the "optimal" and "low-dose" H T is still needed but not easy to define due to varying effects of the same compound in different target organs. The reported increased risk of stroke and "early harm" coronary risk in older women is of concern and points to the importance of adjusting underlying risk factors such as hypertension, hyperlipidemia, and elevated plasma glucose. These high-risk factors that accelerate atherosclerosis potentially place diabetic women at greater risk for early harm when using standard-dose oral therapy.
IV. CONCLUSION The association between deterioration of glucose metabolism and C H D has been observed in numerous studies. In the past few decades, research has demonstrated that the commonly used term insulin resistance syndrome promotes atherothrombosis and includes several components: insulin resistance with or without glucose intolerance, abdominal obesity, atherogenic dyslipidemia, raised blood pressure, a proinflammatory state, and a prothrombotic state. In the vessel wall, insulin stimulates proliferation and migration of smooth muscle cells as well as binding of low-density lipoproteins to receptors in monocytes, and it also influences the activity of major enzymes involved in lipogenesis. In any given woman, insulin resistance is largely genetically determined, with its clinical expression is linked to age, and is more frequently encountered in middle-aged women compared with men. Endogenous and exogenous sex steroids appear to have a modulating effect. In postmenopausal women, exogenous estrogens may decrease insulin resistance, although the effect may be blunted by the increased synthesis of growth factors and decreased hepatic insulin clearance following the first-pass effect of orally administered hormones. Moreover, administration of progestogens together with estrogens may counteract the estrogenic effects on insulin resistance. Although the data are inconclusive, low-dose oral estrogen as well as transdermal therapy may be preferable from a metabolic point of view. Presently the results from the HERS and W H I are encouraging in terms of prevention of type 2 diabetes with the postmenopausal use of HT, although more information is needed before clinical recommendations or guidelines can
506
SKOUBYAND MADSBAD
be rendered. In diabetic women, HT may have a bimodal effect, preventing the development of CHD in women without pre-existing coronary heart disease but accelerating the risk in those with established disease.
References 1. King H, Aubert RE, Herman WH. Global burden of diabetes, 19952025: prevalence, numerical estimates, and projections. Diabetes Care 1998:21;1414-1431. 2. King H, Rewers M. Global estimates for prevalence of diabetes mellitus and impaired glucose tolerance in adults. W H O Ad Hoc Diabetes Reporting Group. Diabetes Care 1993;16:157-177. 3. Walton C, Godsland I, Proudler A, Wynn W, Stevenson JC. The effects of the menopause on insulin sensitivity, secretion and elimination in nonobsese, healthy women. EurJ Clin Invest 1993;23:466-473. 4. Wu SI, Chou P, Tsai ST. The impact of years since menopause on the development of impaired glucose tolerance. Clin Epidemiol 2001;54: 117-120. 5. DeFronzo RA, Ferrannini E. Insulin resistance. A multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease. Diabetes Care 1991;14:173 - 194. 6. Reaven M. Banting lecture 1988. Role of insulin resistance in human disease. Diabetes 1988;37:1595-1607. 7. Isomaa B, Almgren P, Tuomi T, et al. Cardiovascular morbidity and mortality associated with the metabolic syndrome. Diabetes Care 2001; 24:683 -689. 8. Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults: findings from the third National Health and Nutrition Examination Survey.JAm MedAssoc 2002;287:356- 359. 9. Grundy SM, Hansen B, Smith SC Jr, Cleeman JI, Kahn RA. Clinical management of metabolic syndrome: report of the American Heart Association/National Heart, Lung, and Blood Institute/American Diabetes Association conference on scientific issues related to management. Circulation 2004;109:551-556. 10. Lakka HM, Laaksonen DE, Lakka TA, et al. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 2002;21:2709-2716. 11. Sutherland J, McKinnley B, Eckel RH. The metabolic syndrome and inflammation. Metabolic 8yndr Rel Disord 2004;2:82-104. 12. Trayhurn P, Wood IS. Adipokines: inflammation and the pleiotropic role of white adipose tissue. BrJNutr 2004;92:347- 355. 13. Nawrocki AR, Scherer PE. The delicate balance between fat and muscle: adipokines in metabolic disease and musculoskeletal inflammation. Curr Opin Pharmaco12004;4:281-289. 13a. Chu MC, Cosper P, Orio F, Carmina E, Lobo RA. Insulin resistance in postmenopausal women with metabolic syndrome and the measurements of adiponectin, leptin, resistin, and ghrelin. Am J Obstet Gynecol 2006;194:100-104. 13b. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999;22:1462-1470. 13c. Lindheim SR, Buchanan TA, Duffy DM, et al. Comparison of estimates of insulin sensitivity in pre- and postmenopausal women using the insulin tolerance test and the frequently sampled intravenous glucose tolerance test. J Soc GynecolInvestig 1994;1:150-154. 13d. Stumvoll M, Mitrakou A, Pimenta W, et al. Use of the oral glucose tolerance test to assess insulin release and insulin sensitivity. Diabetes Care 2000;23:295-301. 13e. Katz A, Nambi SS, Mather K, et al. Q.uantitative insulin sensitMty check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000;85:2402-2410.
14. Matthews DR, HoskerJP, Rudenski AS, et al. Homeostasis model assessment: insulin resistance and beta-ceU function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412-419. 15. Goodman-Gruen D, Barrett-Connor E. Sex hormone-binding globulin and glucose tolerance in postmenopausal women. The Rancho Bernardo Study. Diabetes Care 1997;20:645-649. 16. Sitruk-Ware R. Pharmacological profile of progestins. Maturitas 2004;47:277-283. 17. Kurmagai S, Hokmang A, Bjorntorp P. The effects of oestrogen and progesterone on insulin sensitivity in female rats. Acta Physiol &and 1993;49:91-97. 18. Skouby SO. Oral contraceptives: effect on glucose and lipid metabolism in previous gestational diabetic women and in insulin dependent diabetic women--a clinical and biochemical assessment. Dan Med Bull 1988;35:157-167. 19. Barret-Connor E, Laakso M. Ischemic heart disease risk in postmenopausal women. Effects of estrogen use on glucose and insulin levels. Arteriosclerosis 1990; 10:531 - 534. 20. Cagnacci A, Soldani R, Carriero PL, et al. Effects of low doses of transdermal 17-beta estradiol on carbohydrate metabolism in postmenopausal women. J Clin Endocrinol Metab 1992:74:1396-1400. 21. Godsland IF, Gangar K, Walton C, et al. Insulin resistance, secretion, and elimination in postmenopausal women receiving oral or transdermal hormone replacement therapy. Metabolism 1993;42:846-853. 22. Lindheim SR, Buchanan TA, Dully DM, et al. Comparison of estimates of insulin sensitivity in pre- and postmenopausal women using the insulin tolerance test and the frequently sampled intravenous glucose tolerance test.J Soc GynecolInvestig 1994;1:150-154. 22a. Hodis HN, Mack WJ, Lobo RA; Estrogen in the Prevention of Atherosclerosis Trial Research Group. Estrogen in the prevention of atherosclerosis. A randomized, double-blind, placebo-controlled trial.Ann Intern Med 2001;135:939-953. 23. Lindheim SR, Duffy DM, Kojima T, et al. The route of administration influences the effect of estrogen on insulin sensitivity in postmenopausal women. Fertil Steri11994;62:1176-1180. 24. Duncan AC, Lyall H, Roberts RN, et al. The effect of estradiol and a combined estradiol/progestagen preparation on insulin sensitivity in healthy postmenopausal women. J Clin Endocrinol Metab 1999;84: 2402-2407. 25. Gaspard UJ, Wery OJ, Scheen AJ, Jaminet C, Lefebvre PJ. Long-term effects of oral estradiol and dydrogesterone on carbohydrate metabolism in postmenopausal women. Climacteric 1999;2:93-100. 26. CucineUi F, Paparella P, Soranna L, et al. Differential effect of transdermal estrogen plus progestagen replacement therapy on insulin metabolism in postmenopausal women: relation to their insulinemic secretion. EurJ Endocrino11999;140:215-223. 27. Godsland IF, Manassiev NA, Felton CV, et al. Effects of low and high dose oestradiol and dydrogesterone therapy on insulin and lipoprotein metabolism in healthy postmenopausal women. Clin Endocrino12004; 60:541-549. 28. Morin-Papunen LC, Vauhkonen I, Ruokonen A, Tapanainen JS, Raudaskoski T. Effects of tibolone and cyclic hormone replacement therapy on glucose metabolism in non-diabetic obese postmenopausal women: a randomized study. EurJ Endocrino12004;150:705-714. 29. Wiegratz I, Starflinger F, TetzloffW, et al. Effect of tibolone compared with sequential hormone replacement therapy on carbohydrate metabolism in postmenopausal women. Maturitas 2002;41:133-141. 30. Espeland MA, Hogan PE, Fineberg SE, et al. Effect ofpostmenopausal hormone therapy on glucose and insulin concentrations. PEPI Investigators (Postmenopausal EstrogerdProgestin Interventions). Diabetes Care 1998;21:1589-1595. 30a. Margolis KL, Bonds DE, Rodabough RJ. Effect of oestrogen plus progestin on the incidence of diabetes in postmenopausal women: results from the Women's Health Initiative Hormone Trial. Diabetologica 2004; 47:1175-1187.
CHAPTER 37 Glucose Metabolism After Menopause 31. Kanaya AM, Herrington D, Vittinghoff E, et al. Glycemic effects of postmenopausal hormone therapy: the Heart and Estrogen/Progestin Replacement Study. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 2003;138:1-9. 32. Manson JE, Rimm EB, Colditz GA, et al. A prospective study of postmenopausal estrogen therapy and subsequent incidence of noninsulin-dependent diabetes mellitus. Ann Epidemio11992;2:665-673. 33. Margolis KL, Bonds DE, Rodabough RJ, Women's Health Initiative Investigators. Effect of oestrogen plus progestin on the incidence of diabetes in postmenopausal women: results from the Women's Health Initiative Hormone Trial. Diabetologia 2004;47:1175-1187. 33a. Friday KE, Dong C, Fontenot RU. Conjugated equine estrogen improves glycemic control and blood lipoproteins in postmenopausal women with type 2 diabetes.J Clin End Metab 2001;86:48-52.
507 33b. Ferrara A, Karter AJ, Ackerson LM, et al. Hormone replacement therapy is associated with better glycemic control in women with type 2 diabetes: The Northern California Kaiser Permanente Diabetes Registry. Diabetes Care 2001;24:1144-1150. 33c. Manning PJ, Allum A, Jones S, Sutherland WH, Williams SM. The effect of hormone replacement therapy on cardiovascular risk factors in type 2 diabetes: a randomized controlled trial. Arch Intern Med 2001; 161:1772-1776. 34. Anderson B, Mattsson L-A, Hahn L, et al. Estrogen replacement therapy decreases hyperandrogenicity and improves glucose homeostasis and plasma lipids in postmenopausal women with non-insulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1997;82:638-643.
This Page Intentionally Left Blank
2 H A P T E R 3t
Stage of Reproductive Life, Atherosclerosis Progression and Estrogen Effects on Coronary Artery
Atherosclerosis THOMAS
B.
CLARKSON
Wake Forest University School of Medicine, Comparative Medicine Clinical Research Center, Winston-Salem, NC 27157
JAY R. N A P L A N
Wake Forest University School of Medicine, Comparative Medicine Clinical Research Center, Winston-Salem, NC 27157
I. I N T R O D U C T I O N
following the perimenopause accounts for the large increase in CHD observed among women over 55 years of age. Not surprisingly, many investigators also concluded that postmenopausal estrogen therapy would be cardioprotective. Today, that belief has come to be questioned by some, perhaps most, health care professionals. Much of the doubt about the potential cardioprotective benefits of estrogen therapy (ET) comes from those who would extend to perimenopausal/early postmenopausal women the results of two randomized trials conducted with women 14 to 18 years postmenopausal (the Heart and Estrogen/Progestin Replacement Study [HERS] and the Women's Health Initiative [WHI]) (1,2). In both HERS and WHI, the initiation of hormone therapy (HT) was associated with increased CHD
Coronary heart disease (CHD) (and underlying coronary artery atherosclerosis) remains the largest cause of death among women over 55 years of age, with CHD mortality exceeding by several-fold the combined deaths due to cancers of the lung, breast, colon, and endometrium. A large body of experimental and epidemiologic evidence suggests that although normal levels of premenopausal estrogen inhibit the development of coronary artery atherosclerosis, estrogen deficiency (as occurs with premenopausal ovarian dysfunction or following surgical or natural menopause) is a major contributor to atherosclerosis progression. In this view, the relative decline in estrogen production during and TREATMENT OF THE POSTMENOPAUSAL WOMAN
509
Copyright 9 2007by Elsevier,Inc. All rights of reproductionin any formreserved.
510
CLARKSON AND KAPLAN
events during the first year of treatment. These findings appeared to contrast with experimental and observational data published in the 1990s. Numerous studies of monkeys provide evidence that ET begun immediately following surgical menopause inhibits the progression of coronary artery atherosclerosis by about 70% (3-5). Moreover, the results of large observational studies, such as that of 121,700 female nurses 30 to 55 years of age, suggest that H T initiated around the time of menopause reduces the risk for a major coronary event by about 50% (6,7). The foregoing contrast between the experimental/ observational studies and the clinical trials resulted in confusion about whether and to what extent estrogens had cardiovascular benefits at any life stage and, equally important, challenged the medical research community to reconcile the various studies into one consistent concept. Following considerable debate and reexamination of the relevant data, there is an emerging consensus that the disparate results of the clinical trials and observational studies reflect the fact that the effects of estrogen on arteries are contingent on the stage of reproductive life and the extent of underlying atherosclerosis. Specifically, estrogen appears to exert beneficial effects in premenopausal and perimenopausal women and women in whom atherosclerosis is minimal. Effects are diminished in postmenopausal women and women in whom atherosclerosis is extensive. The two-part conceptual framework for current thinking about cardiovascular effects of estrogens is illustrated in Fig. 38.1. Fig. 38-1, A, depicts the premenopausal period as important in determining the extensiveness of atherosclerosis at the beginning of the perimenopausal transition, which in turn determines the trajectory for plaque progression during the postmenopausal period. This figure implies that investigators should focus attention on the determinates of "high" and "low" risk during the premenopausal years. Fig. 38.1, B, illustrates the idea that estrogen's arterial benefits vary with menopausal status, which we have come to A
understand is largely inseparable from the degree of progression of subclinical atherosclerosis. Hence, benefits are considerable during premenopausal years and the perimenopausal transition when estrogen receptors (ERs) are intact and there is minimal atherosclerosis. However, there is little or no beneficial effect following 6 or more postmenopausal years, when receptor function is impaired and atherosclerosis is present. In this chapter, we summarize the evidence that the W H I was a seconaary intervention trial and elaborate the data supporting the conceptual framework described previously. After considering the results of the major observational studies, clinical trials, and experimental investigations, we conclude that primary cardiovascular benefits of estrogens are both rational and likely.
II. PROGRESSION OF CORONARY ARTERYATHEROSCLEROSIS A M O N G NORTH AMERICAN WOMEN Recognition that coronary artery atherosclerosis has its origin in childhood and progresses with each reproductive state is important in understanding the conceptual framework about estrogen's arterial effects (Fig. 38.2). Three sources of data provide information about the progression of subclinical coronary artery atherosclerosis of the average American woman. The muhicenter Pathologic Determinants of Atherosclerosis in Youth (PDAY) study provides information on the premenopausal precursors (fatty streaks) of postmenopausal coronary artery plaques (raised lesions) (8). The numbers shown in the lumen of the premenopausal women refer to the surface of coronary arteries with fatty streaks. Among both white and black women up to 35 years of age, the majority (about 70%) have fatty streaks. Only about onethird of these premenopausal women have the early stages of B §
Clinical Events Plaque ~ H i gRishk /
/
7 v
/
Fatty Streaks Normal ~ Artery
+
Hypothetical+ Arterial Benefits of + Estrogens +
Low Risk Pre Peri Post Menopausal Status
0
Pro
Peri Post Menopausal Status
FIGURE 38.1 Progressionof premenopausal atherosclerosis determines the coronary artery plaques extent at the time of perimenopausal transition and sets the trajectoryfor extent of postmenopausal atherosclerosis extent (A). Estrogen's arterial benefits varywith reproductive stage, being of key importance premenopausallyand perimenopausallybut losing benefits after the first 6 postmenopausalyears (B).
CHAPTER 38 Stage of Reproductive Life, Atherosclerosis Progression
FIGURE 38.2 The natural history of atherosclerosis among women in the United States is depicted schematically. ET, estrogen therapy; HT, hormone therapy.
atherosclerotic plaques, and the extent is small, covering only about 15% of the surface of the coronary artery. Data on adult and postmenopausal women have come largely from autopsies of women dying of accidental deaths. During the perimenopausal transition and the early years of the menopause (35-45 years), plaques become larger and fibrous caps begin to develop. By age 55, plaques are larger still, fibrous caps are more distinct, and atheronecrosis and calcification are often present (9). In vivo arterial imaging (via ultrasound) also reveals the presence of focal lesions late in the premenopausal phase and during the perimenopausal transition (10,11). Among women older than 65 years the amount of atheronecrosis increases, and inflammatory processes are often apparent within the lesion, particularly within the shoulders of the fibrous caps (12,13). Electron beam tomography (EBT) data support the pathologic studies. Coronary artery calcification, which is always associated with necrosis, begins at about age 55 and becomes distinct in most women by age 60 to 64. Specifically, it has been reported that about 60% of women reach the 90th percentile of coronary calcification by age 60 to 64 years (13). Generally, plaque instability, associated with the inflammatory process (primarily increases in metalloproteinase activities), coincides with increasing calcification.
III. PRIMARY VERSUS S E C O N D A R Y P R E V E N T I O N ~ C H D VERSUS CO RONARY ARTERY ATHEROSCLEROSIS There is much confusion concerning use of the terms
primary and secondary prevention in relation to estrogen's cardiovascular effects. Primary prevention of atherosclerosis at the vascular level refers to inhibiting the progression of coronary artery fatty streaks and small plaques (common in women at the time of perimenopausal transition) to advanced atherosclerotic plaques (common in women 5 to 15 years postmenopause). In contrast, primaryprevention of CHD generally refers to inhibiting the progression of complicated
511 atherosclerotic plaques to clinical C H D events. The results of the W H I suggest that H T is not associated with primary prevention of CHD. However, this trial did not test whether there was primary prevention of atherosclerosis progression because the participants were at an age that would already have had advanced (although subclinical) plaques when treatment was initiated. As mentioned previously, data from studies of monkeys as well as women indicate that the effect of ET and H T are robust when treatment is initiated at the life stage in which fatty streaks are progressing to plaques (primary prevention of atherosclerosis). However, the recent clinical trials (WHI and HERS) tell us that E T / H T initiated at the complicated plaque stage (beyond about 60 years of age) does not result in C H D benefit.
IV. P R E M E N O P A U S A L D E T E R M I N A N T S OF C O R O N A R Y ARTERY A T H E R O S C L E R O S I S PROGRESSION Conceptually, whether indMduals are at high or low risk for progression of coronary artery atherosclerosis during their premenopausal years determines the extent of intimal lesion present at the time of the perimenopausal transition, which in turn likely establishes the trajectory for plaque progression postmenopausally (see Fig. 38.1, A). Consequently, identifying the risk factors for atherosclerosis progression during the premenopausal period comprises a potentially fruitful focus for research. A close examination of the multicenter PDAY study provides quantitative data about the progression of coronary artery lesions during the important premenopausal years (age 15 to 34 years) (14). Fig. 38.3, A and B, depict the changes with increasing age. During the years from age 15 to 34, the surface area of women's coronary arteries having fatty streaks (the likely precursor of atherosclerotic plaques or raised lesions) increases from about 2% to about 8% with a rather large variation as indicated by the standard error bars (see Fig. 38.3, A). The most rapid increase in fatty streak extent is from ages 25-29 to 30-34, and the variation seen suggests that the population contained both high- and lowrisk individuals. By ages 30 to 34 years, a portion of the fatty streaks have presumably progressed to raised lesions that now occupy about 1% of the surface area of coronary arteries (see Fig. 38.3, B). The two risk factors that appear to be of the most importance in determining a high versus a low trajectory for premenopausal atherosclerosis progression are plasma lipid profiles and ovarian dysfunction (estradiol deficiency), which are not necessarily independent. The evidence supporting the role of these two risk factors is reviewed in the following paragraphs.
512
.-.
CLARKSON AND KAPLAN
lo I
B 4
g
10
8
O > r
6
>
Fatty Streaks
Raised Lesions
~s
7
O ;> c
5 4
<
~o
8
~: 2 0 15-19 20-24 25-29 30-,34
15-19 20-24 25-29 30-34
FIGURE 38.3 Mean extent (+ standard error) of fatty streaks (A) and raised lesions (B) in the right coronary artery of women by 5-year age groups. The data are adjusted for race, very-low-density lipoprotein + lowdensity lipoprotein cholesterol (VLDL + LDL-C), and high density lipoprotein cholesterol (HDL-C) concentrations and smoking. (From res 14, with permission.)
plasma concentrations of V L D L + L D L - C are clearly associated with increase extensiveness of fatty streaking. A similar association occurs in this age group with respect to raised lesions (plaques), although only for the highest tertile. Decreased plasma concentrations of HDL-C are more important than VLDL + LDL-C as a modulator of atherosclerosis progression in premenopausal women (Fig. 38.5. A and B). Hence, the well-defined inverse relationship between the HDL tertile and coronary artery fatty streaks among 25- to 34-year-old women is matched by a similar pattern in relation to raised lesions. Furthermore, plasma HDL-C concentrations also seem to affect atherogenesis among younger (ages 15 to 24) women, albeit only with respect to fatty streaks. B. Role o f O v a r i a n D y s f u n c t i o n in P r e m e n o p a u s a l Atherosclerosis Progression
A. Role o f P l a s m a Lipids in P r e m e n o p a u s a l Atherosclerosis Progression Again, the best evidence for the role of plasma lipids in the progression of premenopausal atherosclerosis progression comes from the PDAY study (14). McGill and his coworkers divided the women into tertiles (low, middle, or high) for very low-density lipoprotein plus low-density lipoprotein cholesterol (VLDL + LDL-C) and high-density lipoprotein cholesterol (HDL-C), and then related the plasma lipid tertile to the extent of coronary artery lesions observed at autopsy. Fig. 38.4, A and B, depict those relationships for VLDL + LDL-C. Between ages 15 and 24, there is no relationship between V L D L + L D L - C and either fatty streaks or raised lesions; however, between ages 25 and 34, increased
A
9
Fatty Streaks
7
0
B
,~
Raised Lesions
~'3
Low Med High
Low Med High
15-24
25-34
0
Although autopsy studies have revealed much about traditional risk factors and the natural history of atherosclerosis in premenopausal women, studies of socially-housed nonhuman primates first unmasked the influence of individual differences in estrogen status on premenopausal lesion development. A series of investigations demonstrated that normal ovarian functionmas evidenced by a normal estrogen surge at ovulation and adequate luteal phase progesterone concentrations--inhibits the progression of coronary artery atherosclerosis, even in the presence of a diet high in saturated fat and cholesterol. In contrast, impaired ovarian function is associated with a marked exacerbation of coronary artery atherosclerosis. Notably, such impairments are generally limited to monkeys of low social status and thus probably represent a response to the stress of being
A
B
10
Fatty Streaks
~
Raised Lesions
~a
Low Mecl High
Low Med High
15-24
25-34
FIGURE 38.4 Mean extent (+ standard error) of fatty streaks (A) and raised lesions (B) in the right coronary artery of women by 10-year age groups and thirds of very-low-density lipoprotein plus low-density-lipoprotein cholesterol (VLDL+LDL-C) concentrations (low, <108 mg/dL; medium, 108 to 150 mg/dL; high, > 150 mg/dL). The data are adjusted for race, highdensity lipoprotein cholesterol (HDL-C), and smoking. (From res 14, with permission.)
0
Low Med High
15-24
Low Med High
25-34
Low Med High
15-24
Low Med High
25-34
FIGURE 38.5 Mean extent (+ standard error) of fatty streaks (A) and raised lesions (B) in the right coronary artery of women by 10-year age groups and thirds of high-density-lipoprotein cholesterol (HDL-C) concentrations (low, <43 mg/dL; medium, 43 to 60 mg/dL; high, >60 mg/dL). The data are adjusted for race, very-low-density lipoprotein plus low-density lipoprotein cholesterol (VLDL + LDL-C) and smoking. (From res 14, with permission.)
CHAPTER 38 Stage of Reproductive Life, Atherosclerosis Progression subordinate in their social groups, a conclusion supported by the observation that such monkeys produce large amounts of the stress hormone cortisol (15-18). A summary of the results produced by the monkey model system is depicted in Fig. 38.6. The stress in subordinate animals occurs in response to the incessant harassment initiated by those monkeys occupying the dominant status positions in their social groups (19). The stressed or "high-risk" monkeys (approximately half of the animals in each social group) are characterized by hypothalamically mediated ovarian dysfunction, estrogen deficiency and luteal phase progesterone deficits, low plasma concentrations of plasma HDL-C, and premature acceleration of coronary artery atherosclerosis. These initial studies suggested the hypothesis that stressinduced estrogen deficiency was driving the premature exacerbation of atherosclerosis of the high-risk subordinate monkeys. We decided to test this hypothesis by conducting an experiment to determine whether providing exogenous estrogen, by way of an estrogen-containing OC, would protect the stressed (subordinate) high-risk monkeys from developing premature coronary artery atherosclerosis (20,21). This was a large study involving more than 200 monkeys for an experimental period (28 months) that would be equivalent to about 6 years for women. We refer to this study as a "periclinical" trial owing to its large size, randomized design, and the many similarities between monkeys and women in reproductive and pathobiologic characteristics. The major finding of this trial is summarized in Fig. 38.7. As predicted, the subordinate (i.e., high-risk, stressed) monkeys receiving no exogenous estrogen experienced a significant exacerbation of atherosclerosis compared
FIGURE 38.6 Differences in risk factors for coronary artery atherosclerosis in premenopausal cynomolgus monkeys subjected to stress (subordinate
status) or no stress (dominant status). Stressed animals had more anovulatory menstrual cycles (depicted as a failure to express an ovulatorysurge in estradiol at the end of the follicular phase), lower high-densitylipoprotein cholesterol (HDL-C) concentrations,and more coronaryartery atherosclerosis at necropsy[15,16]. (From ref. 99, with permission.)
513
p
p=O.O01 - -
0.45
I-'] Dominant BB Subordinate
0.40 0.35
E E 0.30
NS
0.25 <
0.20
E 0.15
_c 0.10 0.05 0.00
ill T
N:
41
46 Control
41
49
OC
FIGURE 38.7 Extent of coronary artery atherosclerosis of surgically postmenopausal cynomolgus monkeys presented by social status and oral contraceptive (OC) treatment. Animals habituallyand predictablywinning the majorityof their competitiveinteractionswere labeled "dominant."The others were subordinate. (From ref. 21, with permission. Reprinted with permission from ILAR Journal, Institute for LaboratoryAnimal Research, National Academy of Sciences, 500 Fifth Street NW, Washington DC 2001 [www.national-academies.org/ilar]). to untreated dominant (non-stressed, low-risk indMduals, p < 0.01). The bars on the right show that treatment of subordinate, high-risk monkeys with exogenous estrogen (OC) completely inhibited the excessive atherosclerosis observed in their untreated counterparts. This study suggests two conclusions" (1)"premenopausal protection" from atherosclerosis extends only to animals at low risk by virtue of their dominant status and (2) estrogen deficiency is responsible for the premature acceleration of atherosclerosis typically observed in subordinate (high-risk) monkeys. This trial was extended subsequently to include a (surgically induced) postmenopausal phase, providing the opportunity to determine whether premenopausal treatment of highrisk monkeys with exogenous estrogen (OC) would modify the trajectory of lesion progression and thus influence the final extent of atherosclerosis, long after termination of OC exposure (22). The results of that study are depicted in Fig. 38.8. The OC exposure not only resulted in significantly less premenopausal atherosclerosis, it also depressed the postmenopausal trajectory and resulted in significantly less extensive atherosclerosis at the end of the postmenopausal period (36 months, approximately equivalent to 9 years in women). This trial was notable in that it employed biopsies taken from the iliac arteries as a surrogate for the coronary arteries (atherosclerosis at these sites correlates >0.80). This allowed serial evaluation of atherosclerosis, once immediately following the premenopausal phase and once at the end of the surgically postmenopausal phase. As a result, the study was able to demonstrate clearly and for the first time that premenopausal hormonal and behavioral conditions significantly influence postmenopausal atherosclerosis outcomes. Although the results depicted here refer to the iliac arteries, the ultimate postmenopausal outcome was mirrored precisely in coronary artery atherosclerosis extent
514
CLARKSON AND KAPLAN
1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
A
E o :3
g
D
L
Control p
OC [ p
I |
Premenopause
Postmenopause
who showed that women with angiographically confirmed CAD had lower estradiol concentrations than those observed in a set of non-patient controls. Based on the substantial body of research using the monkey model and the strongly suggestive data from human studies, it could be hypothesized that premenopausal subjects with ovarian dysfunction/estrogen deficiency are on a high-risk trajectory for CHD and could be a target for intervention, perhaps with estrogen-containing OCs. In this context it could be noted that an additional set of data from the WISE study indicated that postmenopausal women who had taken OCs had less severe CAD than women without OC exposure (23).
FIGURE 38.8 Oral contraceptives (OCs) inhibit atherosclerosis progression in high-risk monkeys. (From res 22, with permission.)
(22). The further implications of this study are discussed in a subsequent section on the perimenopausal transition. Women cannot be subjected to the same kind of invasive, life course manipulation applied to monkeys. Nonetheless, there is now evidence from premenopausal women consistent with the outcomes observed in monkeys. Specifically, hypothalamically mediated ovarian dysfunction and estrogen deficiency of premenopausal women has been shown to be associated with premature coronary artery atherosclerosis in the NHLBI-sponsored Women's Ischemia Syndrome Evaluation (WISE) study (23). In that study, premenopausal women were examined by quantitative angiography. Individuals without evidence of CAD were found to have normal plasma concentrations of estradiol (E2) of about 80 pg/mL, whereas those with coronary artery disease had plasma E2 concentrations of less than 40 pg/mL (Fig. 38.9). A similar observation was reported by Hanke et al. (24), 100
p=0.003
80
E2 (pg/mL)
60 40 20
No CAD
CAD
CAD (>_ 70% stenosis of at least one epicardial coronary artery) FIGURE 38.9 Median estradiolconcentrations ofpremenopausal women with and without coronary artery disease (CAD). E2, estradiol. (From ref. 23, with permission.)
V. CORONARY ARTERY ATHEROSCLEROSIS AND THE PERIMENOPAUSAL TRANSITION As we have suggested in Fig. 38.1, A, and demonstrated in studies of monkeys, the extent of pre-atherosclerosis determines the extensiveness of the plaques during the postmenopausal years. Furthermore, autopsy and imaging studies indicate that atherosclerosis begins accelerating rapidly during the latter part of the premenopausal period. These observations emphasize the importance of investigating the effect of potential interventions (hormones, drugs, dietary supplements, lifestyle changes, etc.) on the progression of atherosclerosis during the 5- to 7-year period immediately prior to the menopause, now termed the perimenopausal transition. For example, there have been suggestions that interventions initiated coincident with menopause (in comparison to several years later) might effectively inhibit the progression of atherosclerosis (25-27). That even earlier intervention might also be efficacious is supported by the monkey studies and the observation that prior OC use by premenopausal women was an independent negative predictor of coronary artery severity in postmenopausal women (23). The possibility that interventions applied during follicle depletion could reduce the trajectory of early disease development and thereby inhibit the postmenopausal progression of coronary and carotid artery atherosclerosis is of extraordinary public health relevance. Investigators have been hampered in addressing the issue of perimenopausal and early postmenopausal interventions by the lack of a suitable animal model. The promise of such research is exemplified by studies evaluating the effects of estrogen supplementation in subordinate (psychosocially stressed and thereby estrogen-deficient) premenopausal monkeys described in the preceding sections. We would argue that these subordinate monkeys resemble perimenopausal women in their ovarian dysfunction (menstrual irregularity, luteal phase progesterone deficits, and anovulation), and in their demonstrably increased vulnerability for chronic disease. Hence, the treatment of socially subordinate monkeys with OCs could
CI~aVTER 38 Stage of Reproductive Life, Atherosclerosis Progression well mimic a perimenopausal hormonal intervention. In one of our trial cohorts, this manipulation was followed by ovariectomy and hormone replacement with conjugated equine estrogens (CEE). The complete design and postmenopausal coronary artery atherosclerosis results of that study are depicted in Fig. 38.10 (22). As we described previously, subordinate, highrisk monkeys ("perimenopausal-like syndrome") either did (+) or did not ( - ) receive an estrogen containing OC for 2 years prior to ovariectomy. Likewise, after ovariectomy, animals either did (+) or did not ( - ) receive ET (CEE) for 3 years. Coronary artery atherosclerosis measurements at the end of the study are depicted as percent reduction from untreated controls. Premenopausal OC use was associated with a marked (67%) reduction in coronary artery atherosclerosis (shown with respect to iliac artery atherosclerosis in Fig. 38.8). As expected, immediate postmenopausal ET also resulted in an inhibition of coronary artery atherosclerosis (64%). Perhaps of most clinical importance was the finding that premenopausal estrogen treatment, followed immediately by postmenopausal estrogen treatment, resulted in an 88% inhibition in coronary artery atherosclerosis compared with untreated controls. These findings imply that treatment of women perimenopausally and immediately postmenopausally could markedly inhibit atherosclerosis progression. Clearly, this intriguing and important hypothesis requires evaluation with a model that more directly mimics the ovarian changes in women.
A. Changes in Plasma Lipids and Lipoproteins During the Perimenopausal Transition The perimenopausal acceleration of atherosclerosis noted in autopsy and imaging studies and implied by the monkey trials is accompanied by adverse changes in plasma lipids and lipoproteins concomitant with the increasing ovarian dysfunction. Two studies (28-29) have shown that LDL-C
FIGURE 38.10 Percentreduction in coronary artery atherosclerosis in psychosocially stressed monkeys treated with either oral contraceptive (OC) (+) or not (-) prior to and after ovariectomy(OVX) [22]. CEE, conjugated equine estrogen.(Fromref. 99, with permission.)
515 concentrations increase by 10% to 20% during the menopausal transition (30). Associated with the change in LDL-C concentration are changes in LDL size and composition. The occurrence of small, dense LDL (associated with coronary heart disease risk) is low in premenopausal women (about 10%) and increases to 30% to 50% during the perimenopausal transition (31-33). Additionally, the changes in plasma lipids and lipoproteins during the menopausal transition may be related to the risk of developing the metabolic syndrome (34). In contrast to changes in total LDL-C, total HDL-C concentrations do not change significantly from premenopause to postmenopause in women. However, there are changes in HDL particle subclasses in an atherogenic direction (decreased HDL2b and HDL2a, as well as increased HDL3) (35-37). A recent report from the Framingham Study (38) suggests that during the perimenopausal transition (ages 30 to 45), there is a small decrease in large HDL particles, a large increase in intermediate HDL particles and an even larger increase in small H D L particles. Importantly, a significant association has been demonstrated between declining HDL-C and carotid artery intimal media thickness (CAIMT), a surrogate marker of atherosclerosis progression (30). Taken together, there seems little doubt that the hormonal and lipid changes occurring in the years leading up to menopause potentiate an acceleration of atherosclerosis and subsequent increase in C H D risk.
B. Endothelial Function and the Perimenopausal Transition The normal endothelium protects the intima of arteries from the initiation and progression of atherosclerosis. Much of the protective role of the endothelium is associated with its ability to produce nitric oxide (NO). For that reason, NOassociated, endothelial-mediated dilation (NO-EMD) is often considered as a critical marker of endothelial dysfunction associated with the initiation/progression of atherogenesis. It is known that N O - E M D changes across the menstrual cycle of women (39-41) and that the dilatory capacity of arteries is directly associated with the plasma estradiol concentration. Although we are not aware of any studies that have examined the relationship between hormonal changes and NO-EMD across the perimenopausal transition of women, it is plausible that the erratic and substantial changes in estradiol production at this time trigger alterations in N O - E M D similar to those observed in response to hormonal perturbations across the menstrual cycle. Using coronary artery infusions of acetylcholine and coronary angiography, it has been possible to evaluate the relationship between plasma estradiol concentrations of premenopausal cynomolgus monkeys and N O - E M D (42). The results of one such study is shown schematically in Fig. 38.11. This figure shows that premenopausal
516
CLARKSONAND
.z .17 _
% Change from Control -4.67 +3.97 +12.57
Left Circumflex Coronary Arteries
12-40
41-62 63-101 102-476 Qum'tiles of Plasma Estradiol (pg/ml)
FIGURE 38.11 Responsesof coronary arteries to ace@choline infusion
in vivo dividedby quartiles of plasmaestradiolconcentration. (Fromref.42, with permission.)
cynomolgus monkeys exhibit the range of plasma estradiol concentrations that bracket the range experienced by women during the perimenopausal transition. Notably, these data also indicate that there is a critical plasma estradiol concentration (approximately 60 to 70 pg/mL) above which E M D is apparently preserved and below which E M D is impaired.
C. T h e N e e d for a Suitable A n i m a l M o d e l for
Studies of the P e r i m e n o p a u s a l T r a n s i t i o n As suggested earlier, progress in understanding atherogenesis during the perimenopausal transition has been hampered by the lack of a suitable animal model. The challenge in modeling the perimenopausal transition and postmenopause relates to two factors: (1) the existence in women of a several-year period (perimenopause)
characterized by reduced numbers of primordial follicles, irregular menstrual cycles, and ultimately declining production of estradiol; and (2) the presence in naturally postmenopausal women of follicle-depleted residual ovarian tissue (stromal and hilar cells) capable of producing androstenedione and testosterone. Together, these features contribute to a hormonal profile distinct from that observed in ovariectomized models, which are characterized by an abrupt and complete cessation of ovarian hormone production. The absence of an adequate animal model of perimenopause/postmenopause has not only impeded progress in developing safe and effective strategies for improving postmenopausal health, it has prevented the invasive studies that could be used to better understand the hormonal and physiologic characteristics and pathophysiologic concomitants of the perimenopause as they may apply to women. We have addressed the foregoing deficiencies in menopausal models by extending to cynomolgus monkeys (Macaca fascicularis) a technique developed in mice and involving exposure to the occupational chemical 4vinylcyclohexene diepoxide (VCD), which selectively destroys ovarian primordial and primary follicles (Fig. 38.12). Because follicle loss is selective, the animal undergoes gradual ovarian failure, thereby modeling the perimenopausal transition as it occurs in women. Ultimately, this procedure results in a follicle-depleted animal that retains residual ovarian tissue. During the course of impending follicle depletion, the changing hormonal profile in these animals closely resembles that observed during the natural onset of menopause in women. We plan to establish a population of VCD-induced, follicle-depleted monkeys as a resource for scientists to conduct clinically relevant research on atherosclerosis progression and other health issues associated with the transition from the perimenopausal period into the postmenopausal state.
ose and lime of VCD "~ i exposure determinesnumberof] ] remahfiangprimordial oocytes ] land length ofpefimenopausal I (transition ) \. . . . . J~*19 18 months - - .... / / ...................... ....... 7 " [ ~Premenopause PerimenopausalTransition 9Erraticthendecliningestradiol concentrations . IncreasingFSH 9DecreasingInNbinB 9DecreasingAMH :
KAPLAN
(',,. f "3 ~ {lmeo ova | ( ctep,euon ) ~ ~ Postmenopause
I
9Residualovary"continuesto producetestosterone 9Plasmaestradiolsof 15-30 pg/mlvs. OVXof 1-2pg/ml 9Plasmatestosteronesof 30-40 ng/nflvs. OVXof,--9 ng/ml
FIGURE38.12 The expectedcharacteristics of the cynomolgus4-vinylcyclohexenediepoxide (VCD) model of perimenopause and postmenopause. FSH, follicle-stimulating hormone; AMH, anti-mfillerian hormone; OVX,ovariectomy.
517
CHAPTER 38 Stage of Reproductive Life, Atherosclerosis Progression
VI. ESTROGEN
EFFECTS
IN EARLY P O S T M E N O P A U S E
(EARLYINTERVENTION) In Fig. 38.2, we suggested that ET and H T have primary vascular benefits when administered during the perimenopausal transition and early postmenopausal years. That suggestion is based on research using monkeys, observational studies, clinical trial data, and some indications from the results of the W H I trial.
A. Monkey Studies The results of studies with the cynomolgus monkey model have been consistent in showing that premenopausal estrogen deficiency is atherogenic whereas premenopausal estrogen supplementation or early postmenopausal replacement is protective. The most compelling example of this effect is depicted in Fig. 38.13, which indicates that when estrogen replacement is initiated immediately after making cynomolgus monkeys surgically menopausal, there is an average inhibition of 70% in the progression of coronary artery atherosclerosis (3,43,44).
B. Observational Studies The majority of observational studies assessing the effect of H T on C H D (46) are based on evaluations of women who have chosen to initiate H T when entering menopause.
0.40
0.30
Coronary Artery
Plaque
0.20
The most comprehensive of these is the Nurses' Health Study that began in 1976 when 121,700 female nurses 30 to 55 years of age completed a mailed questionnaire about their postmenopausal hormone use and medical history (6). Approximately 80% of these women began to use hormones within 2 years after menopause. In the latest report, with 70,533 postmenopausal women followed for up to 20 years, 953 nonfatal myocardial infarctions and 305 coronary deaths were identified (7). Overall, current use of H T was associated with a relative risk (RR) for major coronary event of 0.61 (CI, 0.52-0.71) when adjusted for age and the common cardiovascular risk factors. Of particular relevance in the present context are new data from the Nurses' Health Study showing that among women 60 years and older at initiation of estrogen use, there was no beneficial effect on C H D risk (RR, 1.07; 95% CI, 0.31-1.38), whereas substantial risk reduction was found in women aged 50 to 59 years at estrogen initiation (RR, 0.51; 95% CI, 0.32-0.82) (47).
C. Clinical Trials E v a l u a t i n g E a r l y Initiation of H T Among the few trials evaluating the C H D effects of early initiation of H T (i.e., in women likely to have minimal subclinical atherosclerosis) is one by de Kleijn et al. (48), in which perimenopausal women between ages 40 to 60 were randomized to receive either estrogen or placebo and carotid artery intima media thickness was measured before and after 2 years of treatment. Estrogen treatment showed a trend toward inhibiting the progression of intima media thickness. In a more direct test of the potential cardioprotective effects of estradiol (because of screening to eliminate individuals with pre-existing CHD), the Estrogen in the Prevention of Atherosclerosis Trial (EPAT) evaluated progression of subclinical carotid atherosclerosis in women without clinical cardiovascular disease (49). These women were randomized to receive either oral micronized estradiol or placebo, and after 2 years, atherosclerosis progression was slower in estradiol-treated women (Fig. 38.14).
Size (mm 2) Placebo
0.10
0.004
r-p--o,045
0.003 Women J
Placebo
CEE
FIGURE 38.13 Coronary artery atherosclerosis ofcynomolgus monkeys was about 70% less than controls when estrogen replacement begins at onset of surgical menopause. Treatment was for 34 months (comparable to about 9 patient years) [45]. CEE, conjugated equine estrogen. (From ref. 99, with permission.)
61 yrs [ ~ {LDL-C I , > 130 I
0.002 ~ - - 2 yrs
Rate of
0.001 Atherosclerosis Progression 0 Micronized (CAIMT -0.001 in mrrdyr) 1 mg/day -0.002 n=lll Placebo
E2
FIGURE 38.14 Effect of estrogen therapy on "preclinical atherosclerosis" determined by carotid artery imaging [49]. CAIMT, carotid artery intima media thickness; E2, estradiol. (From ref. 49, with permission.)
518
CLARKSONAND KAPLAN
Additional evidence for beneficial effects of H T on coronary artery atherosclerosis was observed in a study that evaluated coronary artery calcium in 2213 women (50). The average age of the H T users was 59 years and the duration of use was 9 years; thus, the H T was started close to menopause. The duration of H T was significantly associated with a reduction in coronary artery calcification, suggesting coronary artery atherosclerosis protection by H T in these women. Finally, Lobo (51) reported on the cardiovascular events in two large clinical trials of women treated with several H T regiments similar to those used in the W H I . The mean age of the women was about 53 years. Unlike the H T arm of the W H I , no C H D events were observed during the first year of the trials. Thus, it seems that the data from monkey studies, observational studies, and from clinical trials are generally uniform in supporting the hypothesis that early (perimenopausal/immediately postmenopausal) use of estrogen provides significant C H D benefits.
D. Implications of the WHI Trial: A Second Look at the Data The results of the W H I trial also provide supporting evidence concerning potential cardiovascular benefits of earlier initiation of ET and HT. The effect of ET (by age) and H T (by years since menopause) on risk for C H D is depicted in Fig. 38.15. Among women treated with HT, the smallest hazard ratio (HR) for C H D was in the women <10 years postmenopausal (52). In the ET arm of the
Years Since Menopause
< 10 10-19
Hazard Ratio for CHD (HT Arm) 0.00 ,
0.5 ,
,
1.0 ,
J
1.5 ,
2.0
2.5
3.0
.,..........., ............!............,......,...........,............i
.0~ Hazard Ratio for CHD (ET Arm)
Age
0.00 0.5 1.0 1.5 , , , , , ~ ,
x .....
2.0 ,
,
2.5 ,
,
3.0 ,
50 - 59
60 - 69 70 - 79
o.~_o~,
..............~
..............
FIGURE 38.15 Effect of estrogen therapy (ET), by age, and hormone therapy (HT), by years since menopause,on coronaryheart disease (CHD) in postmenopausalwomen. HT arm data [52], ET arm data [53]. (From ref. 99, with permission.)
WHI, there was a strong indication of C H D protection in the youngest subset (50 to 59 years) (53).
VII. M E C H A N I S M S F O R A T H E R O P R O T E C T I O N OF ESTROGENS ADMINISTERED IN EARLY M E N O P A U S E
A. Plasma Lipids/Lipoproteins Clearly, the postmenopausal oral administration of either micronized estradiol or CEE increases plasma concentrations of H D L - C and lowers L D L - C (54-56). Generally, the increases in H D L - C are about 10% and the decreases in L D L - C are about 15%. The administration of a progestogen tends to attenuate these presumably beneficial effects whether the progestogen is given continuously or cyclically (57,58). The attenuating effect of medroxyprogesterone acetate (MPA) is somewhat more pronounced than that of oral micronized progesterone. It is not at all clear that these progestogen effects on estrogen-induced changes in lipid profiles diminish the beneficial changes on atherogenesis, primarily because the reductions in plasma concentrations of H D L - C may not be associated with reductions in the reverse cholesterol transport function of the HDLs (59). It could be suggested that too much emphasis is often placed on the relative importance of estrogen-induced changes in plasma lipid/lipoproteins in mediating their atheroprotective effects. In studies of surgically postmenopausal monkeys fed an atherogenic diet and treated with estradiol administered subcutaneously, it was noted that there was a 50% reduction in the progression of coronary artery atherosclerosis in the estradiol monkeys, whereas there was no significant changes in total plasma cholesterol or plasma concentrations of H D L - C (3). That observation, along with the results of numerous other monkey studies, lead Clarkson and Anthony (60) to suggest that only about 25% of the atheroprotective effects of estrogens related to their plasma lipid effects and that about 75% related to plasma lipidindependent effects, presumably direct effects on arteries. Recently, that estimate was confirmed by studies of early postmenopausal women (61). As shown in Fig. 38.14, the results of the EPAT trial showed that subclinical atherosclerosis was significantly reduced in postmenopausal women randomized to receive oral micronized estradiol. In the study reported by Karim et al. (61), those data were analyzed to determine the extent to which the estrogen-induced changes in plasma lipids explained the beneficial effects of the ET on the progression of CAIMT. The authors concluded that 30% of the estrogen-related influences on CAIMT progression was due
CHAPTER 38 Stage of Reproductive Life, Atherosclerosis Progression to beneficial alteration in the plasma concentrations of HDL-C and LDL-C. Both the monkey studies and the recent studies of the EPAT women are generally consistent with the estimates of 30% to 50% for the plasma lipidindependent effects of estrogens on cardiovascular disease obtained from observational studies (56,62).
B. Arterial Accumulation of LDL The accumulation of LDL in the intima of coronary arteries is directly involved in the initiation and progression of coronary artery atherosclerosis. Studies have been reported of ovariectomized monkeys fed a moderately atherogenic diet to determine the effects of estradiol and progesterone on this early event in atherogenesis. Estradiol and cyclic progesterone were administered via Silastic implants to one group, resulting in physiologic plasma concentrations of those hormones, whereas controls had low to undetectable concentrations of estradiol and progesterone. The hormone treatment resulted in significantly reduced accumulations of coronary artery LDL, 5.95 biL/gl hr for controls versus 1.75 gL/g/hr for the estradiol/ progesterone-treated monkeys (p = 0.04) (63). Although the exact mechanisms for these differences remain unexplained, they likely involve estradiol effects on LDL oxidation and accumulation by macrophages and binding affinities of the proteoglycans.
C. Estrogen and the Prevention of LDL Oxidation The oxidation of LDL particles has a prominent role in atherogenesis. As oxidized LDL particles enter the arterial intima or become oxidized within the intima, they are more readily taken up by macrophages and retained in the artery. Estrogens have been found to have relatively robust LDL antioxidant effects. Wilcox et al. (64) found significant inhibition in vivo of LDL oxidation by CEE in postmenopausal women. Similar antioxidant effects have been reported by Bhavnani et al. (65) and Subbiah et al. (66). The mechanisms by which estrogens protect LDL particles from oxidation are complicated. Critical to their ability to protect the LDL particles is esterification of the estradiol with a fatty acid to render it sufficiently lipophilic to be accumulated within the particle. The Finnish research group headed by Matti Tikkanen has pioneered in elucidating these mechanisms. They found that lecithin:cholesterol acytltransferase (LCAT) associated with H D L catalyzed the formation of estrogen esters, which could be transferred from H D L to LDL particles by a process enhanced by cholesteryl ester transfer protein (CETP) (67,68).
519
D. Sex Hormone Receptor Signaling The complex role of estradiol, progesterone, and testosterone in modulating cardiovascular health is just beginning to be understood. Mendelsohn and Karas (69) recently reviewed the emerging importance of interactions among these sex steroid hormones and their receptors, especially the receptor signaling properties. Sex steroid hormone receptors do not act alone but interact with numerous coregulatory proteins to signal transcription. Mendelsohn and Karas (69) point out that most current research focuses on the ERs, whereas progesterone and testosterone and their receptors in specific tissues have received far less attention. Furthermore, they indicate that a more complete understanding of each of the sex steroid hormones and their receptors in the cardiovascular system will be required to understand more fully the effects of postmenopausal hormone treatment on atherosclerosis.
E. Estrogens, Endothelium, and Vasodilation Estrogen deficiency and estrogen replacement have numerous effects on the endothelium. Mendelsohn and Karas (70) focused the attention of the research community on the role of estrogens as mediators of endothelial function. Dilator and constrictor substances are released by endothelial cells that mediate vascular responses to a wide variety of neurohumoral stimuli. Acetylcholine is often used to determine the functional capacity of the endothelium. As described in previous sections, studies of female monkeys reveal that intracoronary perfusion of acetylcholine induces vasodilation, but only in estrogen-replete individuals; in contrast, estrogen-deficient monkeys administered acetylcholine in the same way undergo vasoconstriction (71). Estrogens have been found to cause vasodilation through both rapid increases in nitric oxide production and induction of nitric oxide synthase genes (70,72,73). In addition to the monkey studies, vasodilation has been found to be affected by fluctuations in circulating concentrations of estrogen during the menstrual cycle, during pregnancy, and after estradiol treatment (73).
F. Inflammation and Inflammatory Markers Over the past 10 to 15 years, the view that lipids play a primary role in the pathogenesis of atherosclerosis and its complications has changed to encompass an understanding of how immune system activation and subsequent inflammatory responses play prominent, if not predominant, roles in atherogenesis and CHD events (74,75). Generally, the pathogenetic sequence begins with the endothelium producing an adhesion molecule--vascular cell adhesion
520
CLARKSON AND KAPLAN
molecule-1 (VCAM-1) ~ w h i c h is capable of binding monocytes and T-lymphocytes to the endothelial surface. The migration of the leukocytes through the endothelium occurs in response to a chemoattractant cytokine, monocyte chemoattractant protein-1 (MCP-1). The monocytes, then in the intimal space, take on modified lipoprotein particles by way of the scavenger receptor to become foam cells and elaborate cytokines and matrix metalloproteinases (MMPs). Estrogens, during the early progressing stages of atherosclerosis, modulate arterial inflammatory processes in beneficial ways (76). The downregulation of cell adhesion molecules such as intercellular cell adhesion molecule-1 (ICAM-1), VCAM-1, and E-selectin has been reported by a number of authors (77-88). As important as are such data, their significance is tempered by limitations to our current knowledge. As concluded in a recent review (89): "Based on the literature it is obvious that circulating markers of inflammation have a role as risk factors for unstable manifestations of atherosclerosis, but it is still unclear whether the different inflammatory markers merely are markers, or if they in an active way contribute to the development and progression of the atherosclerosis disease in their own." Because investigation of that question is not feasible for women, studies using monkey models of vascular inflammation offer a potential way to determine the causal chain of events underlying the progression of atherosclerosis from initial lesions to the unstable plaques that culminate in clinical events. A randomized trial with surgically postmenopausal cynomolgus monkeys was conducted to explore the relationship of inflammatory markers to atherosclerosis extent and plaque characteristics and the extent to which the inflammation markers were modified by treatment with CEE (90). Monkeys were fed a moderately atherogenic diet over a 3-year experimental period. The control group received no hormones, and the experimental monkeys were administered
TABLE 38.1
Coronary artery Carotid artery (common) Carotid artery (internal) Carotid artery (bifurcation)
CEE. The markers measured were MCP-1, sVCAM-1, sE-selectin, interleukin-6 (IL-6), and C-reactive protein (CRP). Serum concentrations of sVCAM-1 and MCP-1 were reduced by CEE treatment (both p < 0.0001). CEE treatment did not affect serum concentrations of lL-6 (p = 0.40), sE-selectin (p = 0.17), or CRP (p = 0.15). Serum MCP-1 and, to a lesser extent, IL-6 correlated significantly to atherosclerotic plaque size in coronary and carotid arteries (Table 38.1). Serum concentrations of MCP-1 were also strongly associated with plaque characteristics and with plaque M M P - 9 (Table 38.2). Another study using the cynomolgus monkey model explored the role of MCP-1 in atherosclerosis and plaque complications (91). Here, the investigators took advantage of a mutant MCP-1, 7ND, which has been shown to work as a dominant-negative inhibitor of MCP-1. Iliac artery plaque size was 0.8 _+ 0.2 mm 2 for M C P - l - b l o c k e d monkeys compared with placebo of 1.3 + 0.2 mm 2 (p < 0.05). Striking decreases in macrophage infiltration was achieved by 7ND gene transfer (p < 0.05). In studies of postmenopausal women, Koh and his colleagues (92,93) found that plasma concentrations of MCP-1 were increased compared with premenopausal women and decreased to premenopausal concentrations by treatment with CEE. Sumino and coworkers (94) reported on the effect of postmenopausal transdermal ET (continuous estradiol patch plus cyclic oral MPA) on plasma concentrations of MCP-1 and other inflammation markers and on endothelial function. The H T resulted in decreases in MCP-1, ICAM-1, VCAM-1, and E-selectin and improvements in flow-mediated dilation. The comparative evidence suggests that MCP-1 may be of particular importance in the early stages of atherosclerosis progression. The question of whether subsequent studies of women will support the important role of MCP-1 must await clinical studies.
Partial Correlations and p Values of Circulating Markers with Plaque Size in Key Arteries of Postmenopausal Monkeys M C P- 1
sVCAM- 1
sE- Selectin
IL-6
0.220 p - 0.009 0.219 p = 0.003 0.314 p = 0.001 0.137 p = 0.109
0.037 p = 0.667 0.008 p = 0.921 0.040 p = 0.640 0.119 p = 0.163
-0.053 p = 0.497 0.051 p = 0.504 -0.006 p = 0.941 0.013 p = 0.988
0.089 p = 0.302 0.228 p = 0.003 0.250 p = 0.003 0.068 p = 0.433
C RP -0.050 p = 0.559 0.007 p = 0.923 0.108 p = 0.208 -0.039 p = 0.654
Correlations were correctedfor effects of treatment. MCP-1, monocytechemoattractantprotein-l; sVCAM-1,solublevascularcell adhesion molecule-l; sE-selectin, soluble E-selectin;IL-6, interleukin-6;CRP, C-reactiveprotein. (From ref. 90, with permission.)
CHAPTER 38 Stage of Reproductive Life, Atherosclerosis Progression
TABLE 38.2
521
Partial Correlations and p Values of Circulating Markers with Coronary Artery Characteristics
Plaque characteristics
M C P- 1
sVCAM- 1
sE- Selectin
IL-6
C RP
Plaque area
0.272 p = 0.001 0.256 p = 0.002 0.362 p = 0.001 0.304 p = 0.001 0.352 p = 0.001 0.296 p = 0.001 0.361 p = 0.001
0.033 p - 0.697 -0.004 p = 0.963 0.077 p = 0.371 0.078 p = 0.366 0.055 p = 0.523 0.061 p = 0.478 0.031 p = 0.762
-0.064 p = 0.451 -0.110 p = 0.198 - 0.041 p = 0.632 0.020 p = 0.815 -0.076 p = 0.376 -0.064 p = 0.546 -0.028 p = 0.786
0.019 p = 0.828 0.100 p = 0.246 0.107 p = 0.213 0.066 p = 0.446 0.070 p = 0.415 0.032 p = 0.707 0.120 p = 0.240
-0.082 p = 0.337 0.008 p = 0.925 -0.046 p = 0.596 0.076 p = 0.375 -0.045 p = 0.601 -0.080 p = 0.351 -0.025 p = 0.813
Inflammation index Necrosis area Percent necrosis Fibrous cap thickness Fibrous cap thinness Plaque MMP-9
Correlations were corrected for effects of treatment. MCP-1, monocytechemoattractantprotein-I; sVCAM-1, solublevascularcell adhesion molecule-I; sE-selectin, soluble E-selectin; IL-6, interleukin-6; CRP, C-reactive protein. (From ref. 90, with permission.)
VIII. ESTROGEN EFFECTS IN LATE M EN O PAU S E (LATE INTE RVENTI O N) At this life stage, as we have indicated in Fig. 38.2, there likely are no cardiovascular benefits of ET/HT, which may in fact result in adverse effects soon after the initiation of treatment. Numerous experimental and observational studies (e.g., the Nurses' Health Study) gave rise to the general belief that E T and H T would be cardioprotective irrespective of the life stage when treatment was begun. Thus, it was unexpected when HERS (1), Estrogen Replacement Atherosclerosis Study (ERA) (95), and W H I (2) reported no cardioprotective benefits of estrogens when administered to older women (ages 63 to 67 years and 15 to 17 years postmenopausal). Much has been written dealing with possible/ probable reasons for the discordance between experimental and observational studies and the results of these three randomized trials (reviewed in [25,26,69,96-101]). A unifying hypothesis has evolved, often referred to as the timing hypothesis, taking the position that estrogens are athero-protective if administered during the fatty streak to uncomplicated plaque stage of atherosclerosis progression (perimenopausal-early postmenopausal reproductive stage) but have no or deleterious effects if administered after plaques have become "complicated" by necrosis and inflammation (the later postmenopausal years). Attempts to understand how estrogen's effects on arteries could be discordant depending on the stage of progression of the atherosclerotic lesions have necessitated reevaluation of a number of pathobiologic concepts. This has become a relatively new area of research focus, and the understanding
of loss of the athero-protective effects of estrogens when treatment is begun after atherosclerotic plaques have reached the "complicated" stage is not as well understood as are the mechanisms whereby estrogens inhibit progression of early lesions. Explanations of why H T is sometimes associated with cardiovascular events when treatment is initiated at the complicated plaque stage are even less well developed. Nevertheless, in this review we shall summarize the relevant findings.
A. Monkey Studies Because estrogen treatment effects cannot be compared to the stage of progression of atherosclerosis in women, monkey studies have been particularly helpful. The design of one study allowed us to assess the effects of E T (CEE) and H T (CEE + M P A ) on early versus late postmenopausal coronary artery atherosclerosis (45,102). In that study, cynomolgus monkeys were made surgically postmenopausal and then the initiation of E T and H T was delayed for 2 years (comparable to 6 postmenopausal years for women). During this period in which hormonal treatments were withheld, the coronary artery plaques progressed to a stage of atheronecrosis with inflammatory complications comparable to women's plaques at 60 to 65 years of age (see Fig. 38.2). In contrast to the robust atherosclerosis benefits of C E E administered to monkeys soon after ovariectomy (see Fig. 38.13), there was no inhibition of coronary artery atherosclerosis after either E T or H T treatment for a period comparable to 6 years for women (Fig. 38.16).
522
CLARKSON AND KAPLAN 0.40
-7
0.30 Coronary Artery Plaque 0.20 Size (mm 2) 0,10
Baseline
Placebo
CEE
CEE+MPA
FIGURE 38.16 Coronary artery atherosclerosis ofpostmenopausal cynomolgus monkeys was not reduced by estrogen therapy or hormone therapy
when delayed 2 years after ovariectomy (comparable to 6 patient years) [45]. CEE, conjugated equine estrogens; MPA, medroxyprogesterone acetate. (From ref. 99, with permission.)
To understand better whether the lack of effect of the CEE treatment was related to time since surgical menopause or the stage of progression of the atherosclerotic plaques, a reanalysis of another study was undertaken (25,103). In that study, prior to the administration of placebo or CEE, a portion of the common iliac artery was removed surgically for quantification of atherosclerosis extent. The animals were then randomized to receive either placebo or CEE. Pretreatment iliac artery atherosclerosis was used to divide the monkeys into tertiles and the effect of 3 years of CEE treatment was evaluated in the contralateral iliac artery samples at the end of the study when the animals were necropsied (Fig. 38.17).
FIGURE38.17 The effect of the extent of pre-existing atherosclerosison effectiveness/lack of effectiveness on postmenopausal conjugated equine estrogen (CEE) treatment of surgicallypostmenopausalcynomolgusmonkeys. Only the tertile with plaque extents comparable to women at the end of the perimenopausaltransition and earlypostmenopausalyears derived a significant atheroinhibitorybenefit of CEE treatment. (From ref. 25, with permission.)
Compared with animals treated with placebo, CEE was highly effective in inhibiting atherosclerosis progression, but only among those monkeys starting at the lowest tertile of atherosclerosis (/5 = 0.0001). We believe that the plaques among those animals were analogous to those of women during the perimenopausal transition and early postmenopausal years (i.e., relatively minimal). In contrast, those monkeys with plaques in the moderate and high tertiles began treatment when the stage of atherosclerosis progression was analogous to those of women several years postmenopausal; here CEE treatment was not associated with a significant inhibition of atherosclerosis progression in comparison to placebo.
B. Studies of Women Results from recent randomized trials are in general agreement with the monkey data. HERS tested whether daily treatment with 0.625 mg CEE and 2.5 mg MPA would reduce coronary heart disease events in older postmenopausal women (1). The women in HERS were generally beyond 65 years of age and had pre-existing coronary heart disease. No reduction in coronary events could be shown by the hormone treatment during an average of 4.1 years of follow-up. In fact, during the first year of the trial, there was an increase in events (RR, 1.52; CI 1.0-2.29), seen mostly in the first 4 months. In the ERA Trial, Herrington et al. (95) investigated the effect of H T on the progression of pre-existing coronary artery atherosclerosis as assessed by angiography. The average age of the women in the study was 65 years, and half had experienced a previous myocardial infarction. After 3.5 years of H T treatment (0.625 mg CEE alone, CEE combined with 2.5 mg MPA, or placebo), there were no differences in coronary artery atherosclerosis progression among any of the treatment groups that could be detected by repeated angiography. The C H D outcomes in the H T arm of the W H I were similar to those in HERS. Postmenopausal women aged 50 to 79 years were assigned randomly to receive either placebo or the same H T used in HERS (CEE +MPA). After 5.2 years of follow-up, a trend toward more C H D events among HTtreated women was observed. As in HERS, the largest difference in C H D events was noted in the first year following initiation of treatment. The C H D outcomes in the ET arm of the W H I were similar, although the HRs tended to be smaller and there was a strong indication of fewer C H D events among the ET-treated subjects in the youngest subset (50 to 59 years) (53). O f considerable interest, and consistent with predictions from the monkey data, there was a trend for increasing numbers of events among ET-treated subjects with increasing years since surgical menopause (see Fig. 38.15).
CHAPTER 38 Stage of Reproductive Life, Atherosclerosis Progression Additional evidence for a lack of effect of H T on older women with pre-existing coronary artery atherosclerosis comes from the results of a double-blind, placebo-controlled trial of postmenopausal women (mean age 63.5 years) who had at least one coronary artery lesion demonstrated angiographically (Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial [WELLHART]) (104). Unlike ERA, in which the women were given C E E + M P A , the women in W E L L - H A R T were given oral micronized estradiol alone or with sequentially administered MPA. W E L L - H A R T was undertaken to make possible a comparison of postmenopausal women with pre-existing cardiovascular disease with the previously described results of EPAT, in which the postmenopausal women were without such disease (see Fig. 38.14). In the EPAT trial, an athero-protective effect was demonstrated in the oral micronized estradiol-treated group, whereas among the older women with coronary artery atherosclerosis in WELL-HART, hormone treatments had no significant effects on progression of coronary artery atherosclerosis.
C. Mechanisms for Loss of Atheroprotection Explanations for the loss of atheroprotection following estrogen treatment of subjects with advanced atherosclerosis are not yet fully developed.
1. DIMINISHED ER EXPRESSION OR ACTIVITY
Availability of ERs in the artery wall is considered an important factor in how estrogens can modulate development of atherosclerosis. Losordo and colleagues (105) demonstrated loss of detectable ERs in atherosclerotic arteries. ER's were detectable in 71% of normal arteries but were found in only 32% of arteries with atherosclerosis (p = 0.01). Post and colleagues (106) reported a greater degree of ER-ot methylation in coronary atherectomy samples (10%) compared with macroscopically normal human aorta (4%,p < 0.01 versus atherectomy). Reduced expression or activity of the ERs in atherosclerotic lesions would contribute importantly to the lack of protection afforded by ET in patients with advanced atherosclerosis.
523 vasodilatory effect of H T only in the healthiest of these women (109). Our data in female monkeys with established atherosclerosis are consistent with these reports (102). It appears that the vasodilatory effects of H T are diminished as women develop a greater risk factor burden, and thus likely a greater degree of subclinical atherosclerosis.
D. P u t a t i v e M e c h a n i s m s for P l a q u e I n s t a b i l i t y A consistent finding in both HERS and W H I was that the adverse C H D events occurred soon after the initiation of treatment. In the H T arm of W H I , for example, the only statistically significant increase in C H D events was in the first year of treatment (52). That observation has led cardiovascular pathobiologist to conclude that the initiation of l i T in some individuals resulted in changes in complicated atherosclerotic plaques that had not been exposed to exogenous estrogens for as much as 15 to 18 years. That C H D events did not appear to be increased by treatment with CEE alone in the W H I trial is difficult to interpret. Some have concluded that the observation suggests some role for MPA (100,101); however, that may be an incorrect speculation given the differences in the patient population in the CEE + MPA and the CEE-only arms of the W H I trial. The patients in the CEE-only arm had prior hysterectomy. Howard et al. (110) reported that hysterectomy (regardless of oophorectomy status) was a significant predictor of cardiovascular disease (HR, 1.26; p < 0.001). A likely mechanism for the occurrence of adverse events following estrogen treatment in women with pre-existing coronary lesions relates to the induction of pro-inflammatory changes in complicated lesions (69,76,98,101). An illustration of these putative mechanisms is shown in Fig. 38.18. There is substantial interest in the hypothesis that estrogens can increase the production of MMPs. It seemed plausible to place a major emphasis on the MMPs because they have proteolytic capabilities and as a result can digest away the shoulder regions of the fibrous cap, thus increasing likelihood of the gruel coming into contact with the blood and potentiating the development of a mural thrombus. Supporting these ideas is the observation that macrophages in
2. VASODILATION
Vasodilatory effects of estrogen may be diminished in women with more advanced risk for vascular disease. Campisi et al. (107) showed that long-term H T did not modulate myocardial blood flow in postmenopausal women with risk factors for CHD. Koh et al. (108) reported that CEE did not affect the flow-mediated dilatory response to hyperemia in type 2 diabetic postmenopausal women. An analysis of elderly women in the Cardiovascular Health Study found a
FmURE 38.18 Adverse effects of initiating estrogen treatment at the complicatedplaque stage of atherosclerosis.MMP, matrixmetalloproteinase; PQ~plaque.
524
CLARKSON AND KAPLAN
the shoulder region of fibrous caps produce MMPs, which in turn can weaken the fibrous cap and lead to plaque rupture, mural thrombosis, and arterial occlusion (111-114). Additional plausible evidence for a role of the MMPs comes from the well-recognized association between premenopausal cyclical estrogen production and the upregulation of the production of MMPs in the endometrium, an important role in the remodeling of the endometrium with each menstrual cycle (115-117). It seems reasonable to assume that the same processes that occur in the endometrium are pertinent in the context of complicated atherosclerofic plaques when local concentrations of estradiol are markedly increased by estrogen treatment. Hu and coworkers reported on an extensive study of the effects of H T on markers of inflammation and endothelial function and on plasma concentrations of MMP-9 in postmenopausal women (118). Women assigned to CEE or CEE plus one of three progestin regimens had increases in CRP levels, but the increases were not associated with changes in IL-6. CEE with or without micronized progesterone decreased both ICAM and MMP-9. Although it is interesting that the HTs decreased plasma concentrations of MMP-9, it is unlikely that plasma concentrations reflect changes in M M P production within complicated plaques. In HERS (119) and in the H T arm of W H I (52), those patients receiving statins did not experience early increases in CHD events when treatment was initiated (Fig. 38.19). Although it has not yet been proven, it seems likely that the mechanism by which the statins inhibit early C H D events when administered to women with complicated coronary artery atherosclerotic plaques is via downregulation of the inflammatory process, particularly the production of MMPs. The statins have a powerful ability to inhibit the secretion of MMP-9 from vascular smooth muscle cells and foam cells from plaques. Fig. 38.20 shows the effect of statins on inhibiting MMP-9 secretion from human vascular smooth muscle cells and foam cells from rabbit atherosclerotic plaques (120). In a valuable recent experiment, Hwang and coworkers (121) found that the addition of estradiol to macrophages in vitro increased production of MMP-9; however, simvastatin
120
Human Vascular Smooth Muscle Cells F - - p<0.01
100
Rabbit Plaque Foam Cells !--- p<0.01
:~ 80 60
o
40
20
J
/
Control
Statin
Control
Statin
FIGURE 38.20 Statins inhibit secretion of matrix metaUoproteinase-9 from human vascular smooth muscle cells and rabbit plaque macrophage derived foam cells. (From ref. 120, with permission.)
inhibited, in a dose-dependent manner, the estradiolstimulated production of MMP-9 (Fig. 38.21). The inhibition of MMP-9 activity by simvastatin provides mechanistic support for the observation that statin users do not have an increased rate of CHD events within the first year of randomization of HT.
IX. C L I N I C A L OFTHE
TIMING
TRIALS HYPOTHESIS
This chapter has reviewed a large body of clinical and experimental evidence supporting what has been referred to as the "timing hypothesis": the notion that El" can inhibit the progression from early atherosclerofic lesions to complicated plaques but cannot prevent the progression of complicated plaques and may in fact initiate processes leading to the instability and rupture of such plaques. Importantly, two clinical trials have been initiated in the United States to test the timing
Hazard Ratlo for CHD Statin Use
0.00 l
I
0,5 ,
J
1,0 i
J
1,5 l
~
2.0 I
'
2.5 !
!
3.0 !
~
3,5 !
j
4.0 I
Yes
No
FIGURE 38.19 Coronary heart disease (CHD) events in statin-treated women in the Women's Health Initiative, years 1 to 5, p = 0.02 for trend overtime. (From ref. 52, with permission.)
FIGURE 38.21 The addition of estradiol (E2) (5 nano-molar) to macrophages in vitro increases the release of matrix metaUoproteinase-9 (MMP-9). Simvastatin, in a dose-dependent manner, inhibited the production of E2stimulated MMP-9 production. (From ref. 121, with permission.)
CHAPTER 38 Stage of Reproductive Life, Atherosclerosis Progression hypothesis. The Kronos Early Estrogen Protection Study (KEEPS, http://www.kronosinstitute.org/keeps.html) will include only women in early menopause who are treated with lower-dose C E E (0.459 mg/day) or with estradiol transdermal patches (50 pg), both with oral micronized progesterone. The Early versus Late Intervention Trial with Estradiol (ELITE, http://clinicaltrials.gov/show/NCT00114517) will contrast the effects of oral micronized estradiol (1 mg/day with intravaginal progesterone) given to women less than 6 years since menopause versus women 10 or more years postmenopausal. For both KEEPS and ELITE, the primary outcome measure will be change in C M M T . Recently, van der Schouw and Grobbee (122) presented evidence that different outcomes might prevail depending on whether H T is administered to women with or without menopausal symptoms. These investigators argue that major differences between the W H I and observational studies such as the Nurses' Health Study relate not only to age of the women but whether the women were experiencing menopausal symptoms at initiation of therapy. Almost without exception, the women participating in observational studies had menopausal symptoms, whereas such patients were excluded from the W H I . In Finland the S Y M P T O M trial (http://www.hus.fi) has been initiated to compare vascular and cardiac (unction in recently menopausal women with or without severe vasomotor symptoms, as well as to investigate vascular response to oral versus transdermal HT.
525 There is an urgent need to better understand the cellular and molecular events that occur when complicated plaques are exposed to exogenous estrogens. Although the current research focus is on the MMPs, their precise role in estrogenrelated C H D events has not been clarified. It is encouraging that trials are underway to evaluate the relation between time postmenopausal and estrogen effects on atherosclerosis. The trial end points, however, use a surrogate outcome, changes in CAIMT. If these trials result in positive outcomes, it will be a high priority to extend them to trials in which C H D events are the outcome. In this context it is clear that studies employing nonhuman primates will continue to extend our understanding of the factors affecting the cardiovascular health of postmenopausal women by allowing direct studies of coronary artery atherosclerosis extent and severity while varying hormone treatment and initial atherosclerosis status. Finally, monkeys are also well suited to disentangling the effects of age and disease stage in relation to HT. Perimenopausal women typically have little atherosclerosis, whereas women 10 or more years postmenopausal almost invariably have extensive atherosclerosis. Studies in monkeys will allow investigators to determine the relative contributions of arterial age and atherosclerosis stage to the efficacy of HT.
References X. CONCLUSIONS The results of current research provides evidence that estrogen effects on coronary arteries vary with reproductive life stage, time since menopause, and stage of progression of coronary artery atherosclerosis. The premenopausal life stage is a time when the trajectory for atherosclerosis progression may be set and likely determines, to a large extent, the burden ofpostmenopausal atherosclerosis. Although plasma concentrations of lipids and lipoproteins have a major influence on premenopausal atherosclerosis, experimental studies indicate that estrogen deficiency has a more significant role. The perimenopausal transition and early menopause is a life stage in which potentially there are robust cardiovascular benefits of estrogen treatment. About one-third of the benefits are mediated through improvements in plasma lipids, and two-thirds are plasma lipid independent, likely involving downregulation of the inflammatory process and maintenance of normal endothelial function. In the later postmenopausal life stage (beyond 5 or 6 postmenopausal years) there are no or possibly deleterious cardiovascular effects of estrogen treatment. The increase in C H D events during this life stage appears to be related to upregulation of the plaque inflammatory process with resulting plaque instability and may be downregulated by statin treatment.
1. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women.JAm MedAssoc 1998;280:605-613. 2. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women's Health Initiative randomized controlled trial.JAm MedAssoc 2002;288:321-333. 3. AdamsMR, KaplanJR, Manuck SB, et al. Inhibition of coronary artery atherosclerosis by 17-beta estradiol in ovariectomizedmonkeys.Lack of an effect of added progesterone.Arteriosclerosis 1990;10:1051- 1057. 4. Adams MR, Register TC, Golden DL, Wagner JD, Williams JK. Medroxyprogesterone acetate antagonizes inhibitory effects of conjugated equine estrogens on coronary artery atherosclerosis.Arterioscler Thromb VascBio11997;17:217-221.
5. ClarksonTB, AnthonyMS, WagnerJD. A comparison of tibolone and conjugated equine estrogen's effects on coronary artery atherosclerosis and bone densityof postmenopausal monkeys.J Clin EndocrinolMetab 2001;86:5396- 5404. 6. Stampfer MJ, Willett WC, Colditz GA, et al. A prospective study of postmenopausal estrogen therapy and coronary heart disease.NEnglJ Med 1985;313:1044-1049. 7. GrodsteinF, MansonJE, Colditz GA, et al. A prospective,observational study of postmenopausal hormone therapy and primary prevention of cardiovasculardisease.Ann Intern Med 200;133:933-941. 8. McGill HC, McMahan CA, Zieske AW, et al. Effects of nonlipid factors on atherosclerosis in youth with a favorable lipoprotein profile. Circulation 2001;103:1546-1550. 9. McGill HC Jr, Stern ME Sex and atherosclerosis. Atheroscler Rev 1979;4:157-242.
526 10. Sutton-Tyrrell K, Lassila HC, Meilahn E, et al. Carotid atherosclerosis in Premenopausal and postmenopausal women and its association with risk factors measured after menopause. Stroke 1998;29:1116-1121. 11. Tuzcu EM, Kapadia SR, Tutar E, et al. High prevalence of coronary atherosclerosis in asymptomatic teenagers and young adults: evidence from intravascular ultrasound. Circulation 2001;103:2705-2710. 12. Tejada C, Strong JP, Montenegro MR, Restrepo C, Solberg LA. Distribution of coronary and aortic atherosclerosis by geographic location, race, and sex. Lab Invest 1968;18:509-526. 13. Raggi P, Callister TO._, Cooil B, et al. Identification of patients at increased risk of first unheralded acute myocardial infarction by electronbeam computed tomography. Circulation 2000;101:850-855. 14. McGill HC Jr, McMahan CA, Malcom GT, Oalmann MC, StrongJE Effects of serum lipoproteins and smoking on atherosclerosis in young men and women. Arterioscler Thromb VascBio11997;17:95-106. 15. Hamm TE Jr, Kaplan JR, Clarkson TB, Bullock BC. Effects of gender and social behavior on the development of coronary artery atherosclerosis in cynomolgus macaques. Atherosclerosis 1983;48:221-233. 16. Adams MR, Kaplan JR, Koritnik DR. Psychosocial influences on ovarian endocrine and ovulatory function in Macaca fascicularis. Physiol Behav 1985;35:935-940. 17. Kaplan JR, Adams MR, Clarkson TB, Koritnik DR. Psychosocial influences on female "protection" among cynomolgus macaques. Atherosclerosis 1984;53:283-295. 18. Kaplan JR, Adams MR, Clarkson TB, Manuck SB, Shively CA. Psychosocial factors, sex differences, and atherosclerosis: lessons from animal models. Psychosom Med 1996;58:598-611. 19. Kaplan JR. Nonhuman primate and other animal models of women's health. ILAR 2004;45:83-211. 20. Kaplan JR, Adams MR, Anthony MS, et al. Dominant social status and contraceptive hormone treatment inhibit atherogenesis in premenopausal monkeys. Arterioscler Thromb VascBio11995;15:2094-2100. 21. Kaplan JR, Manuck SB. Ovarian dysfunction, stress, and disease: a primate continuum. II_~IRJ 2004;44:89-115. 22. Kaplan JR, Manuck SB, Anthony MS, Clarkson TB. Premenopausal social status and hormone exposure predict postmenopausal atherosclerosis in female monkeys. Obstet Gyneco12002;99:381- 388. 23. Merz CN, Johnson BD, SharafBL, et al. Hypoestrogenemia of hypothalamic origin and coronary artery disease in premenopausal women: a report from the NHLBI-sponsored WISE Study.J A m Co# Cardiol 2003;41:413-419. 24. Hanke H, Hanke S, Ickrath O, et al. Estradiol concentrations in premenopausal women with coronary heart disease. Coron Artery Dis 1997;8:511-515. 25. Karas RH, Clarkson TB. Considerations in interpreting the cardiovascular effects of hormone replacement therapy observed in the WHI: timing is everything. Menopaus Med 2003;10:8-12. 26. Mikkola TS, Clarkson TB, Notelovitz M. Postmenopausal hormone therapy before and after Women's Health Initiative study: what consequences? Ann Med 2004;36:1-12. 27. Clarkson TB, Appt SE. Coronary artery disease and postmenopausal hormone therapy: is there a time window for prevention? Gynaecol Forum 2004;9:11 - 14. 28. Poehlman ET, Toth MJ, Ades PA, Rosen CJ. Menopause-associated changes in plasma lipids, insulin-like growth factor I and blood pressure: a longitudinal study. EurJ Clin Invest 1997;27:322-326. 29. Matthews KA, Meilahn E, Kuller LH, et al. Menopause and risk factors for coronary heart disease. N EnglJMed 1989;321:641-646. 30. Matthews KA, Kuller LH, Sutton-Tyrrell K, Chang YF. Changes in cardiovascular risk factors during perimenopause and postmenopause and carotid artery atherosclerosis in healthy women. Stroke 2001;32: 1104-1111. 31. Campos H, McNamara JR, Wilson PW, Ordovas JM, Schaefer EJ. Differences in low density lipoprotein subfractions and apolipoproteins in
CLARKSON AND KAPLAN
premenopausal and postmenopausal women. J Clin Endocrinol Metab 1988;67:30-35. 32. Austin M, King M-C, Vranizan K, Newman B, Krauss R. Inheritance of low density lipoprotein subclass patterns: results of complex segregation analysis. Am J Hum Genet 1988;43:838- 846. 33. Carr MC, Kim KH, Zambon A, et al. Changes in LDL density across the menopausal transition. JInvest Med 2000;48:245-250. 34. Carr MC. The emergence of the metabolic syndrome with menopause. J Clin Endocrinol Metab 2003;88:2404-2411. 35. Matthews KA, Wing RR, Kuller LH, Meilahn EN, Plantinga E Influence of the perimenopause on cardiovascular risk factors and symptoms of middle-aged health women. Arch Intern IVied 1994;154:2349-2355. 36. Kuller LH, Meilahn EN, Lassila H, Matthews KA, Wing R. Cardiovascular risk factors during first five years postmenopause in nonhormone replacement therapy users. In: Forte T, ed. Proceedings of the American Heart Association scientific conferenceon hormonal, metabolic and cellular influences on cardiovascular disease in women. New York: Futura, 1997:273-287. 37. Williams PT, Vranizan KM, Austin MA, Krauss RM. Associations of age, adiposity, alcohol intake, menstrual status, and estrogen therapy with highdensity lipoprotein subclasses.Arteriosder Thromb 1993;13:1654-1661. 38. Freedman DS, Otvos JD, Jeyarajah EJ, et al. Sex and age differences in lipoprotein subclasses measured by nuclear magnetic resonance spectroscopy: the Framingham Study. Clin Chem 2004;50:1189-1200. 39. Bracamonte MP, Jayachandran M, Rud KS, Miller VM. Acute effects of 17 beta-estradiol on vemoral veins from adult gonadally intact and ovariectomized female pigs. Am J Physiol Heart Circ Physio12002;283: H2389-H2396. 40. Kalantaridou SN, Naka KK, Papanikolaou E, et al. Impaired endothelial function in young women with premature ovarian failure; normalization with hormone therapy.J Clin EndocrinolMetab 2004;89:3907-3913. 41. Kawano H, Motoyama T, Ohgushi M, et al. Menstrual cyclic variation of myocardial ischemia in premenopausal women with variant angina. Ann Intern Med 2001;135:1002-1004. 42. Williams JK, Shively CA, Clarkson TB. Determinants of coronary artery reactivity in premenopausal females cynomolgus monkeys with diet-induced atherosclerosis. Circulation 1994;90:983- 987. 43. Clarkson TB, Anthony MS, Klein KP. Hormone replacement therapy and coronary artery atherosclerosis: the monkey model. Br J Obstet Gynaeco11996;103:53-57. 44. Clarkson TB, Anthony MS, Morgan TM. Inhibition of postmenopausal atherosclerosis progression: a comparison of the effects of conjugated equine estrogens and soy phytoestrogens. J Clin Endocrinol Metab 2001;86:41-47. 45. Clarkson TB. The new conundrum: do estrogens have any cardiovascular benefits. IntJ Ferti12002;47:61-68. 46. Grady D, Rubin SM, Petitti DB, et al. Hormone therapy to prevent disease and prolong life in postmenopausal women. Ann Intern Med 1992;117:1016-1037. 47. Grodstein F, Manson JE, Stampfer MJ. Hormone therapy and coronary heart disease: the role of time since menopause and age at hormone initiation. J Womens Health 2006;15:35-44. 48. de Kleijn Mjj, Bots ML, Bak AAA, et al. Hormone replacement therapy in perimenopausal women and 2-year change of carotid intima-media thickness. Maturitas 1999;32:195 - 204. 49. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis: a randomized, double-blind, placebo-controlled trial. Ann Intern Med 2001;135:939-953. 50. Akhrass F, Evans AT, Wang Y, et al. Hormone replacement therapy is associated with less coronary atherosclerosis in postmenopausal women. J Clin Endocrinol Metab 2003;88:611-614. 51. Lobo RA. Evaluation of cardiovascular event rates with hormone therapy in healthy, early postmenopausal women: results from 2 large clinical trials.Arch Intern Med 2004;164:482-484.
CHAPTER 38 Stage of Reproductive Life, Atherosclerosis Progression 52. Manson JE, Hsia J, Johnson KC, et al. Estrogen plus progestin and the risk of coronary heart disease. N EnglJ Med 2003;349:523-534. 53. The Women's Health Initiative Steering Committee. Effects of conjugated equine estrogens in postmenopausal women with hysterectomy--the W H I randomized controlled trial. J Am Med Assoc 2004;291:1701-1712. 54. Walsh BW, SchiffI, Rosner B, et al. Effects of postmenopausal estrogen replacement on the concentrations and metabolism of plasma lipoproteins. N EnglJ Med 1991;325:1196-1204. 55. Kuhl H. Prevention of atherosclerosis by estrogen substitution? Geburtshilfe Frauenbeilkd 1988;48:747- 756. 56. Bush TL, Barrett-Connor E, Cowan LD, et al. Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the Lipid Research Clinics Program Follow-up Study. Circulation 1987; 75:1102-1109. 57. The Writing Group for the PEPI Trial. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial.JAm MedAssoc 1995;273:199-208. 58. Sitruk-Ware R. Progestins and cardiovascular risk markers. Steroids 2000;65:651-658. 59. Clarkson TB, Appt SE. MPA and postmenopausal coronary artery atherosclerosis revisited. Steroids 2003;68:941-951. 60. Clarkson TB, Anthony MS. Effects on the cardiovascular system-basic aspects. In: Dempster D, Lindsay R, Jordan VC, eds. Estrogens and anti-estrogens. Philadelphia: Lippincott-Raven, 1997:89-118. 61. Karim R, Mack WJ, Lobo RA, et al. Determinants of the effect of estrogen on the progression of subclinical atherosclerosis: Estrogen in the Prevention of Atherosclerosis Trial. Menopause 2005;12:366- 373. 62. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47-63. 63. Wagner JD, Clarkson TB, St Clair RW, et al. Estrogen and progesterone replacement therapy reduces low density lipoprotein accumulation in the coronary arteries of surgically postmenopausal cynomolgus monkeys. J Clin Invest 1991;88:1995-2002. 64. Wilcox JG, Hwang J, Hodis H, et al. Cardioprotective effects of individual conjugated equine estrogens through their possible modulation of insulin resistance and oxidation of low-density lipoprotein. Fertil Steri11997;67:57-62. 65. Bhavnani BR, Cecutti A, Gerulath A, Woolever AC, Berco M. Comparison of the antioxidant effects of equine estrogens, red women components, vitamin E, and probucol on low-density lipoprotein oxidation in postmenopausal women. Menopause 2001;8:408-419. 66. Subbiah MTR, Kessel B, Agrawal M, et al. Antioxidant potential of specific estrogens on lipid peroxidation. J Clin Endocrinol Metab 1993;77:1095 - 1097. 67. Helisten H, H6ckerstedt K, W~ihSl~i K, et al. Accumulation of highdensity lipoprotein-derived estradiol-1713 fatty acid esters in low-density lipoprotein particles. J Clin EndocrinolMetab 2001;86:1294-1300. 68. Tikkanen M, Vihma V, Jauhiainen M, et al. Lipoprotien-associated estrogens. CardiovascRes 2002;56:184-188. 69. Mendelsohn ME, Karas RH. Molecular and cellular basis of cardiovascular gender differences. Science2005;308:1583-1587. 70. Mendelsohn ME, Karas RH. The protective effects of estrogen on the cardiovascular system. N EnglJ Med 1999;340:1801 - 1811. 71. Williams JK, Honor4 EK, Adams MR. Contrasting effects of conjugated estrogens and tamoxifen on dilator responses of atherosclerotic epicardial coronary arteries in nonhuman primates. Circulation 1997; 96:1970-1975. 72. Edwards DR Regulation of signal transduction pathways by estrogen and progesterone. Annu Rev Physio12005;67:335-376. 73. Sader MA, Celermajer DS. Endothelial function, vascular reactivity and gender differences in the cardiovascular system. Cardiovasc Res 2002;15:597-604.
527 74. Libby P. Inflammation in atherosclerosis. Nature 2002;420:868-874. 75. Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation 2002; 105:1135 - 1143. 76. St6rk S, van der Schouw YT, Grobbee DE, Bots ML. Estrogen, inflammation and cardiovascular risk in women: a critical appraisal. Trends Endocrinol Metab 2004;15:66- 72. 77. St6rk S, yon Schacky C, Angerer R The effect of 1713-estradiol on endothelial and inflammatory markers in postmenopausal women: a randomized, controlled trial. Atherosclerosis 2002a;165:301 - 307. 78. St6rk S, Baumann K, yon Schacky C, Angerer P. The effect of 1713estradiol on MCP-1 serum levels in postmenopausal women. Cardiovasc Res 2002b;53:642-649. 79. Cushman M, Lagault C, Barrett-Connor E, et al. Effect of postmenopausal hormones on inflammation-sensitive proteins: The Postmenopausal Estrogen/Progestin Interventions (PEPI) study. Circulation 1999;100:717- 722. 80. Koh KK, Blum A, Hathaway L, et al. Vascular effects of estrogen and vitamin E therapies in postmenopausal women. Circulation 199%; 100:1851-1857. 81. Koh KK, Cardillo C, Bui MN, et al. Vascular effects of estrogen and cholesterol-lowering therapies in hypercholesterolemic postmenopausal women. Circulation 199%;99:354-360. 82. Scarabin PY, Alhenc-Gelas M, Oger E, Plu-Bureau G. Hormone replacement therapy and circulating ICAM-1 in postmenopausal women. A randomized controlled trial. Thromb Haemost 1999;81:673-675. 83. Blum A, Schenke WH, Hathaway L, et al. Effects of estrogen and the selective estrogen receptor modulator raloxifene on markers of inflammation in postmenopausal women. Am J Cardio12000;86:892- 895. 84. Sbarouni E, Kroupis C, Kyriakides ZS, Koniavitou K, Kremastinos DT. Cell adhesion molecules in relation to simvastatin and hormone replacement therapy in coronary artery disease. Eur HeartJ 2000;21:975-980. 85. Zanger D, Yang BK, Ardans J, et al. Divergent effects of hormone therapy on serum markers if inflammation in postmenopausal women with coronary artery disease on appropriate medical management.JAm Coll Cardio12000;36:1797-1802. 86. Seljeflot I, Arnesen H, Hofstad AE, Os I. Reduced expression of endothelial cell markers after long-term transdermal hormone replacement therapy in women with coronary artery disease. Thromb Haemost 2000;83:944-948. 87. Goudev A, Georgiev DB, Koycheva N, Manasiev N, Kyurkchiev S. Effects of low dose hormone replacement therapy on markers of inflammation in postmenopausal women. Maturitas 2002;43:49-53. 88. Koh KK, Schenke WH, Waclawiw MA, Csako G, Cannon RO. Statin attenuates increase in C-reactive protein during estrogen replacement therapy in postmenopausal women. Circulation 2002;105:1531-1533. 89. Lind L. Circulating markers of inflammation and atherosclerosis. Atherosclerosis 2003;169:203 - 214. 90. Register TC, Cann JA, Kaplan JR, et al. Effects of soy isoflavones and conjugated equine estrogens on inflammatory markers in atherosclerotic, ovariectomized monkeys. J Clin EndocrinolMetab 2005;90:1734-1740. 91. Kitamoto S, Nakano K, Hirouchi Y, et al. Cholesterol-lowering independent regression and stabilization of atherosclerotic lesions by pravastatin and by antimonocyte chemoattractant protein-1 therapy in nonhuman primates. Arterioscler Thromb VascBid 2004;24:1522-1528. 92. Koh KK, Jin DK, Yang SH, et al. Vascular effects of synthetic or natural progestagen combined with conjugated equine estrogen in healthy postmenopausal women. Circulation 2001a;103:1961-1966. 93. Koh KK, Son JW, Alan JY, et al. Effect of hormone replacement therapy on nitric oxide bioactivity and monocyte chemoattractant protein-1 levels. IntJ Cardio12001b;81:43-50. 94. Sumino H, Ichikawa S, Ohyama Y, et al. Effect of transdermal hormone replacement therapy on the monocyte chemoattractant protein-1 concentrations and other vascular inflammatory markers and on endothelial function in postmenopausal women. Am J Cardio12005;96:148-153.
528 95. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary artery atherosclerosis. N EnglJ Med 2000;343:522-529. 96. Grodstein F, Clarkson TB, Manson JE. Understanding the divergent data on postmenopausal hormone therapy. N EnglJ Med 2003;348: 645-650. 97. Harman SM, Brinton EA, Clarkson T, et al. Is the WHI relevant to HRT started in the perimenopause? Endocrine 2004;24:195-202. 98. Rosano GM, Vitale C, Lello S. Postmenopausal hormone therapy: lessons from observation and randomized studies. Endocrine 2004;24: 251-254. 99. Clarkson TB, Appt SE. Controversies about HRT--lessons from monkey models. Maturitas 2005;51:64- 74. 100. Klaiber EL, Vogel W, Rako S. A critique of the Women's Health Initiative hormone therapy study. Fertil Steri12005;84:1589-1601. 101. Phillips LS, Langer RD. Postmenopausal hormone therapy: critical reappraisal and a unified hypothesis. Fertil Steri12005;83:558-566. 102. Williams JK, Anthony MS, Honor~ EK, et al. Regression of atherosclerosis in female monkeys. Arteriosder Tbromb Vasc Bid 1995;15: 827-836. 103. Anthony MS, Clarkson TB. Does extent of pretreatment atherosclerosis influence the effects of conjugated equine estrogens on atherosclerosis progression? JAm Coll Cardio12002;39:248A. 104. Hodis HN, Mack WJ, Azen SP, et al. Hormone therapy and the progression of coronary-artery atherosclerosis in postmenopausal women. N EnglJ Med 2003 ;349:535 - 545. 105. Losordo DW, Kearney M, Kim EA, Jekanowski J, Isner JM. Variable expression of the estrogen receptor in normal and atherosclerotic coronary arteries of premenopausal women. Circulation 1994;89: 1501-1510. 106. Post WS, Goldschmidt-Clermont PJ, Wilhide CC, et al. Methylation of the estrogen receptor gene is associated with aging and atherosclerosis in the cardiovascular system. Cardiovasc Res 1999;43:985-991. 107. Campisi R, Nathan L, Pampaloni MH, et al. Noninvasive assessment of coronary microcirculatory function in postmenopausal women and effects of short-term and long-term estrogen administration. Circulation 2002;105:425-430. 108. Koh KK, Kang MH, Jin DK, et al. Vascular effects of estrogen in type II diabetic postmenopausal women. JAm Coll Cardio12001c;38: 1409-1415. 109. Herrington DM, Espeland MA, Crouse JR, et al. Estrogen replacement and brachial artery flow-mediated vasodilation in older women. Arteriosder Thromb VascBio12001;21:1955-1961. 110. Howard BV, Kuller L, Langer R, et al. Risk of cardiovascular disease by hysterectomy status, with or without oophorectomy. The Women's Health Initiative Observational Study. Circulation 2005;111: 1462-1470.
CLARKSON AND KAPLAN
111. van der Wal AC, Becker AE, van der Loos CM, Das PK. Site ofintimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation 1994;89:36-44. 112. Moreno PR, Falk E, Palacios IF, et al. Macrophage infiltration in acute coronary syndromes: implications for plaque rupture. Circulation 1994;90:775-778. 113. Matrisian LM. The matrix degrading metalloproteinases. Bioessays 1992;14:455-463. 114. Rennick RE, Ling KLE, Humphries SE, et al. Effect of acetyl-LDL on monocyte/macrophage expression of matrix metalloproteinases. Atberosclerosis 1994;109:192. 115. Goffin F, Manaut (2, Frankenne F, et al. Expression pattern of metalloproteinases and tissue inhibitors of matrix-metalloproteinases in cycling human endometrium. Bid Reprod 2003;69:976-984. 116. Hampton AL, Salamonsen LA. Expression of messenger ribonucleic acid encoding matrix metalloproteinases and their tissue inhibitors is related to menstruation. J Endocrino11994; 141:1 - 3. 117. Kokorine I, Marbaix E, Henriet P, et al. (1996). Focal cellular origin and regulation of interstitial collagenase (matrix metalloproteinase-1) are related to menstrual breakdown in the human endometrium. J Cell Sci 1996;109:2151-2160. 118. Hu P, Greendale GA, Palla SL, et al. The effects of hormone therapy on the markers of inflammation and endothelial function and plasma matrix metaUoproteinase-9 level in postmenopausal women: the postmenopausal estrogen progestin intervention (PEPI) trial.Atberosclerosis 2006;185:347-352. 119. Herrington DM, VittinghoffE, Lin F, et al. Statin therapy, cardiovascular events, and total mortality in the Heart and Estrogen/Progestin Replacement Study (HERS). Circulation 2002;105:2962-2967. 120. Luan Z, Chase AJ, Newby AC. Statins inhibit secretion of metalloproteinases-1, -2, -3, and -9 from vascular smooth muscle cells and macrophages. Arterioscler Tt~romb Vasc Biol 2003;23: 769-775. 121. Hwang J, Hodis HN, Hsiai TK, Asatryan L, Sevanian A. Role of annexin II in estrogen-induced macrophage matrix metalloproteinase-9 activity: the modulating effect of statins. Atberosclerosis 2006;189: 76-82. 122. van der Schouw YT, Grobbee DE. Menopausal complaints, oestrogens, and heart disease risk: an explanation for discrepant findings on the benefits of post-menopausal hormone therapy. Eur Heart J 2005; 26:1358-1361.
~HAPTER
3 (.
Randomized Controlled
Trials and the Effects of Postmenopausal Hormone Therapy onCardiovascular " D~sease: Facts,Hyp otheses,
and Clinical Perspective HOWARD N. HODIS WENDY
J. M A C K
Atherosclerosis Research Unit, Division of Cardiovascular Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033 Department of Preventive Medicine, Division ofBiostatistics, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
I. I N T R O D U C T I O N
cardiovascular disease mortality in men whereas in women, cardiovascular disease mortality has continued to increase (Fig. 39.1). Over their life span, women exhibit two distinct patterns of cardiovascular risk. During the premenopausal phase of life, relative to men, women are protected from the clinical complications of atherosclerosis, such as myocardial infarction (2). However, in the postmenopausal phase of life, the clinical manifestations of cardiovascular disease in women equal or exceed those of their male counterparts. Although
Mthough our understanding of the pathophysiology of cardiovascular disease has increased substantially over the past 50 years, atherosclerotic cardiovascular disease continues to be the number one cause of death in both women and men (1). In fact, 53% of all deaths in women are attributable to cardiovascular disease, exceeding the number of deaths in men as well as the next 7 causes of death in women combined (1). Since 1979, there has been a steady decline in TREATMENT OF THE POSTMENOPAUSAL WOMAN
529
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
530
HODIS AND MACK
with a reduction in the incidence of cardiovascular disease in postmenopausal women. In the 1990s, a series of randomized controUed trials were initiated to test this hypothesis. Contrary to the many consistent observational studies, the trial results have not been straightforward. After the initial sensationalism (13) of the first few trial results, a more tempered critical reappraisal of all the randomized controlled trials in relation to the totality of data has emerged (14). In this chapter, the randomized controlled trials designed to study the effects of hormone therapy on cardiovascular disease will be reviewed and placed into perspective with the totality of data currently available.
FIGURE 39.1 U.S. cardiovascular disease mortality trends for males and females, 1979-2002.
there is an age-associated increase in the incidence of cardiovascular disease for both postmenopausal and premenopausal women, the age-specific incidence of cardiovascular disease across the age range is two to six times greater for postmenopausal than premenopausal women (Fig. 39.2). These and other data indicate that loss of a cardiovascular protective factor (presumed to be estrogen), or loss of an array of factors, increases a woman's risk for cardiovascular disease. If these protective factors are endogenous sex steroid hormones, then the transition from premenopausal cardiovascular protection to postmenopausal cardiovascular risk provides the opportunity for reducing the incidence of cardiovascular disease to premenopausal levels through exogenous hormones as preventive therapy. This assumption is based on more than 40 observational studies in which exogenous estrogen (3,4) and estrogen plus progestogen (5-12) were associated
FIGURE 39.2 Age-specific incidence of cardiovascular disease according to menopausal status.
II. MORBIDITY/MORTALITY TRIALS OF POSTMENOPAUSAL H O R M O N E THERAPY As summarized in Table 39.1, five published randomized controlled trials of postmenopausal hormone therapy and coronary heart disease have used morbidity and mortality as the trial end point. Three of these trials are secondary prevention trials (women with symptomatic cardiovascular disease), and two of these trials are primary prevention trials (majority of women without clinical evidence of cardiovascular disease). Although no primary or secondary prevention trial has demonstrated a statistically significant overall difference between hormone therapy and placebo in the risk of coronary heart disease or all-cause mortality, significant trends to benefit are evident with longer duration of hormone therapy and timing of initiation of hormone therapy relative to menopause.
A. The Heart and Estrogen / Progestin Replacement Study The Heart and Estrogen/progestin Replacement Study (HERS) was the first randomized controlled trial to determine whether hormone therapy would reduce the incidence of coronary heart disease (15). Specifically, HERS was designed to test whether the combination of oral daily continuous combined conjugated equine estrogen (CEE) 0.625 mg plus medroxyprogesterone acetate (MPA) 2.5 mg (Prempro | Wyeth-Ayerst, Philadelphia, PA) (n = 1380) would reduce the combined incidence of nonfatal myocardial infarction and coronary heart disease death relative to women who received placebo treatment (n = 1383). The HERS cohort consisted of 2763 postmenopausal women who were on average ( + standard deviation) 66.7 + 6.7 years old, 18 + 8 years since menopause at baseline, and had pre-existing cardiovascular disease defined as a prior history of myocardial infarction, coronary
531
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy TABLE 39.1
Coronary Heart Disease and All-Cause Mortality with Postmenopausal Hormone Therapy: Randomized Controlled Trials Relative risk
Trial
Mean age at entry (years)
Time from menopause (years)1
Treatment
Mean duration (years)
Coronary heart disease
Overall mortality
0.99 (95% CI, 0.80-1.22) 1.29 (95% CI, 0.84-1.95) 0.99 (95% CI, 0.70-1.41)
1.08 (95% CI, 0.84-1.38) Not reported
1.24 (95% uCI, 1.00-1.54) (95% aCI, 0.97-1.60) 0.95 (95% uCI, 0.79-1.16) (95% aCI, 0.76-1.19)
0.98 (95% uCI, 0.82-1.18) (95% aCI, 0.70-1.37) 1.04 (95% uCI, 0.88-1.22) (95% aCI, 0.81-1.32)
Secondary Prevention HERS (15)
66.7
18.8
PHASE (18)
66.5
>102
ESPRIT (19)
62.6
16
WHI-EP (20,21)
63.3
13.4
WHI-E (22,23)
63.6
Oral CEE 0.625 mg + MPA 2.5 mg Transdermal 1713-E2 alone or sequential NE Oral E2-V 2 mg alone
4.1 2.6 2.0
0.79 (95% CI, 0.50-1.27)
Primary Prevention
>102
Oral CEE 0.625 mg + MPA 2.5 mg
5.2
Oral CEE 0.625 mg alone
6.8
1At randomization. 2Exact information has not been published. HERS, Heart and Estrogen/progestin Replacement Study; PHASE, Papworth HRT Atherosclerosis Study; ESPRIT, Estrogen in the Prevention of Reinfarction Trial; WHI-EP, Women's Health Initiative Estrogen+ Progestin Trial; WHI-E, Women's Health Initiative Estrogen Trial; CEE, conjugated equine estrogen; MPA, medroxyprogesteroneacetate; 17~-E2, 1713-estradiol;NE, norethisterone; E2-V, estradiol valerate; uCI, unadjusted confidence interval, describes the variabilityof the point estimate that would arise from a simple trial for a single outcome; aCI, adjusted confidence interval, describes the variabilityof the point estimate corrected for multiple outcome analyses.
artery revascularization, or angiographic evidence of coronary artery disease. After an average 4.1 years of randomized treatment, there was no significant difference between the C E E + M P A - t r e a t e d (172 women with events) and placebotreated (176 women with events) groups in the incidence of the primary trial end point (hazard ratio [HR], 0.99; 95% confidence interval [CI], 0.80-1.22). Although the overall treatment effect was null, there was a statistically significant time trend in coronary heart disease events. In the first year of randomized treatment, the incidence of coronary heart disease events was significantly greater in the C E E + M P A treated group relative to the placebo treated group (57 events versus 38 events). After the first year of randomized treatment, there was a statistically significant trend toward a reduction in the primary trial end point in years 4 and 5 in the C E E + M P A treated group compared with the placebo group (p -- 0.009 for trend) (Fig. 39.3). The greater incidence of coronary heart disease events in the first year of randomized treatment in the C E E + M P A - t r e a t e d versus placebo-treated group was only significant among the women not using statins at baseline (HR, 1.75; 95% CI, 1.02-3.03,p = 0.04) (Fig. 39.4). Among the women using statins at baseline, the H R was not
FIGURE 39.3 Percent(incidence), absolute number, relative risk, and 95% confidence intervalfor coronaryheart disease events by year in the Heart and Estrogen/Progestin Replacement Study (HERS). CEE+MPA = conjugated equine estrogen 0.625 mg plus medroxyprogesteroneacetate 2.5 mg daily. P = trend in relative risk over time.
532
HoDis AND MACK
B. Papworth Hormone Replacement Therapy Atherosclerosis Study
FIGURE 39.4
Effectof statin use on coronaryheart disease events in the Heart and Estrogen/Progestin Replacement Study (HERS). Among the women who did not use statins, the incidence of coronary heart disease in the first year of randomization was greaterin the conjugated equine estrogen 0.625 mg plus medroxyprogesterone acetate 2.5 mg daily (CEE+MPA) group relative to placebo,whereas in women who used statins there was no difference between treatment groups.
significantly different between the C E E + M P A - t r e a t e d and placebo-treated groups (HR, 1.34; 95% CI, 0.63-2.86, p = 0.45) (16). Statins most likely prevent atherosclerotic plaque rupture that may be induced under certain circumstances by CEE + MPA (see VIII. Evidence for a Duality of CEE + MPA Therapy on Coronary Heart Disease). Overall mortality did not differ between the C E E + M P A - t r e a t e d group (131 deaths) and the placebo-treated group (123 deaths) in HERS (HR, 1.08; CI, 0.84-1.38). There was no significant difference between hormone therapy and placebo in breast cancer incidence. In an unblinded, open-label, mean follow-up of 2.7 years of the HERS subjects in which all event data were selfreported, there was no difference in coronary heart disease events between women originally assigned to the C E E + MPA-treated group and the placebo-treated group (17). As an unblinded, open-label follow-up, the data require careful interpretation. For example, during the follow-up period, less than half of the women originally assigned to the C E E + MPA-treated group were actually using C E E + M P A , with only 45% of the cohort adherent to the hormone regimen, and 8% of the original placebo-treated group were taking hormone therapy. In addition, 36% of the C E E + M P A - t r e a t e d group and 39% of the placebo-treated group were using statins while 80% of all subjects were using aspirin during the follow-up period. Although the authors interpreted these data as indicating that CEE + M P A therapy had no long-term effect on coronary heart disease, an equally likely interpretation is that continued hormone therapy is required to maintain the beneficial effects of hormone therapy on the reduction of coronary heart disease events as seen in the latter years of the HERS randomized trial (15).
A total of 255 postmenopausal women older than 55 years with at least one coronary artery lesion of 50% diameter stenosis or greater who were admitted for coronary angiography were randomized to the Papworth Hormone Replacement Therapy Atherosclerosis Study (PHASE) (18). P H A S E was a prospective, open-label, blind end point study of transdermal hormone therapy versus placebo. Allocation was to transdermal estrogen (n = 58), sequential estrogen/progestogen therapy (n - 76), or placebo (n = 121). Women randomized to active treatment received transdermal 1713-estradiol 320 gg/day; those with a prior hysterectomy received 1713estradiol alone and those women with an intact uterus additionally received transdermal norethisterone 480 gg/day sequentially for 14 days of each 28-day cycle. Women were postmenopausal for at least 5 years; the average age of the women who received active treatment was 66.3 ___ 11.2 years, and for those in the placebo group, the average age was 67.0 ___ 11.8 years. A total of 53 subjects (40%) dropped out of the active treatment group, and 8 subjects (7%) dropped out of the placebo group. After an average 2.5 years of randomized treatment, there were 53 primary trial end point events (nonfatal myocardial infarction, coronary heart disease death, or hospitalization with unstable angina) in the active treatment group and 37 primary trial end point events in the placebo group. In the intention-to-treat analysis, there was no significant difference in the primary trial end point event rate (Event Rate Ratio, 1.29; 95% CI, 0.84-1.95; p = 0.24) between the active treatment group (15.4%) and the placebo group (11.9%). However, this study was limited by very low power to test event rate differences.
C. Estrogen in the Prevention of Reinfarction Trial The Estrogen in the Prevention of Reinfarction Trial (ESPRIT) was designed to test whether estrogen alone would reduce the primary trial end points of nonfatal myocardial infarction or cardiac death and all-cause mortality relative to women who received placebo treatment (19). Postmenopausal women ages 50 to 69 years who survived a first myocardial infarction were randomized to estradiol valerate 2 mg daily (n = 513) or placebo (n = 504). The ESPRIT cohort consisted of 1017 postmenopausal women who were on average 62.6 years old, were approximately 16 years since menopause at baseline, and had pre-existing cardiovascular disease defined as a prior history of myocardial infarction. After 2 years of randomized treatment, there were 62 nonfatal myocardial infarctions or cardiac deaths in the estradiol group and 61 such events in the placebo group,
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy TABLE 39.2
533
Cumulative Number and Relative Risk of Overall Mortality and Cardiac Death in the Estrogen in the Prevention of Reinfarction Trial (ESPRIT) Any death 1
Time to events (months)
Estradiol valerate (n = 513)
Placebo (n = 504)
Relative risk (95% confidence interval)
3 6 12 18 24
8 14 20 27 32
14 20 30 39 39
0.56 (0.23-1.33) 0.68 (0.34-1.35) 0.65 (0.37-1.14) 0.67 (0.41-1.10) 0.79 (0.50-1.27)
Cardiac death2
P-value
Estradiol valerate (n = 513)
Placebo (n = 504)
Relative risk (95% confidence interval)
P-value
0.19 0.27 0.13 0.11 0.34
4 9 14 19 21
12 18 25 30 30
0.33 (0.11-1.01) 0.49 (0.22-1.09) 0.54 (0.28-1.05) 0.61 (0.35-1.09) 0.68 (0.39-1.19)
0.052 0.079 0.059 0.096 0.170
171 total deaths. 251 cardiac deaths.
with 32 deaths from any cause in the estradiol group and 39 such events in the placebo group. There was no significant difference between the estradiol-treated and placebo-treated groups in any of the primary trial end points of nonfatal myocardial infarction or cardiac death (rate ratio [RR], 0.99; 95% CI, 0.70-1.41; p = 0.97) and all-cause mortality (RR, 0.79; 95% CI, 0.50-1.27; p - 0.34). Although not significantly different, the relative risks of any death and cardiac death were lower in the estradiol-treated group than the placebo-treated group by 21% to 44% and 32% to 67%, respectively, throughout the 2-year trial, with both end points being lowest after 3 months of randomization (Table 39.2).
D. W o m e n ' s H e a l t h I n i t i a t i v e The Women's Health Initiative (WHI) was a parallel, placebo-controlled trial classified as primary prevention (although 10.2% of the women had known cardiovascular disease at baseline) of 27,347 postmenopausal women ages 50 to 79 years who were randomized according to hysterectomy status. Women with an intact uterus were randomized to oral daily continuous combined CEE 0.625 mg plus MPA 2.5 mg (Prempro | Wyeth-Ayerst) (n = 8506) or placebo ( n - 8102) (20,21), and women who had a prior hysterectomy were randomized to oral daily CEE 0.625 mg (Premarin | Wyeth-Ayerst) (n = 5310) or placebo (n - 5429) (22,23). The primary trial end point was the combined incidence of nonfatal myocardial infarction and coronary heart disease death. The initially planned duration of the W H I trials was 8.5 years. On May 31, 2002, after an average follow-up of 5.2 years, the Data and Safety Monitoring Board stopped the C E E + M P A trial (WHI-EP), and on February 2, 2004 the National Institutes of Health (the funding source of the trial) unilaterally stopped the CEE trial (WHI-E) after an average follow-up of 6.8 years. The W H I - E P cohort consisted of 16,608 postmenopausal women (Fig. 39.5) who were on average 63.3 + 7.1 years old
I
wo~176
.... [
! 18,845 Reported No Hysterectomy
1
3,444 9 Unblinded
[-~
42% 9 Discontinued CEE+MPA
I "38% Discontinued Placebo
6% 9 Started HT on Own
[ 11% 9 Started HT on Own
Unblind~
FIGURE39.5 Studyflowchart of the conjugatedequineestrogen0.625 mg plus medroxyprogesteroneacetate 2.5 mg daily (CEE+MPA) arm of the Women's Health Initiative (WHI) trial. HT - hormonetherapy.
(two-thirds were 60 years and older) with an average body mass index (BMI) of 28.5 kg/m 2 (34.1% of the subjects had a BMI of 30 kg/m 2 or greater); 7.6% of the women had known cardiovascular disease at baseline (20). Women with menopausal symptoms such as hot flushes that would have precluded randomization to placebo were excluded. After an average 5.2 years of randomized treatment, there was no significant difference between the C E E + M P A - t r e a t e d (annualized percentage of events 0.39) and placebo-treated (annualized percentage of events 0.33) groups in the incidence of the primary trial end point of nonfatal myocardial infarction and coronary heart disease death (HR, 1.24; 95% unadjusted CI, 1.00-1.54; 95% adjusted CI, 0.97-1.60)(21). Although the overall treatment effect was null, there was a statistically significant time trend toward the reduction of coronary heart disease events (Fig. 39.6). In the first year of randomized treatment, the incidence of coronary heart disease events was significantly greater in the C E E + MPA-treated group relative to the placebo-treated group (42 events versus 23 events) as analyzed with the unadjusted
534
HODIS AND MACK
FIGURE 39.6 Percent (incidence), absolute number, relative risk, and 95% confidence interval for coronary heart disease events by year in the Women's Health Initiative (WHI) estrogen plus progestin (E+P) trial. C E E + M P A = conjugated equine estrogen 0.625 mg + medroxyprogesterone acetate 2.5 mg daily. P is for the trend in relative risk over time.
statistic (HR, 1.81; 95% unadjusted CI, 1.09-3.01). These data have not been reported using the adjusted analysis for multiple comparisons. After the first year of randomized treatment, there was a statistic@ significant trend toward a reduction in the primary trial end point in year 6 and beyond in the C E E + M P A - t r e a t e d group compared with the placebo-treated group (p = 0.02 for trend). Overall mortality did not differ between treatment groups in W H I - E R Further analysis of the W H I - E P results indicated a statistically significant trend in the treatment effect of C E E + M P A on coronary heart disease according to time since menopause (Fig. 39.7) (24). Women in the C E E + M P A - t r e a t e d group who were less than 10 years postmenopausal had an 11%
Years
Since
reduced risk for coronary heart disease compared with placebo, those who were 10 to 19 years postmenopausal had a 22% increased risk for coronary heart disease, and those who were 20 years or more postmenopausal had a 71% increased risk for coronary heart disease. As two-thirds of the W H I - E P cohort was 60 years old or older, the majority of women randomized to this trial were 10 years or more (average of 13.4 years) postmenopausal at baseline (25). In WHI-EP, 87% of the coronary heart disease events occurred among the women randomized to C E E + M P A 10 years or more after menopause (24). Data regarding the event rates of C E E + M P A versus placebo treatment by time since menopause are not available. Because the majority of coronary heart disease events occurred among women randomized to CEE + M P A 10 years or more after menopause, these data are important in understanding whether the early increase in coronary heart disease events in the CEE + M P A group relative to placebo group was dependent upon when a woman initiated C E E + M P A therapy relative to her time since menopause. Importantly, analysis of both W H I - E P and HERS indicates that CEE + MPA significantly reduces the incidence of diabetes mellims. In WHI-EP, C E E + M P A reduced new onset diabetes meUitus by 21% (HR, 0.79; 95% CI, 0.67-0.93) relative to placebo (Fig. 39.8) (26). In HERS, C E E + M P A reduced new-onset diabetes mellitus by 35% (HR, 0.65; 95% CI, 0.48-0.89) relative to placebo (27). These findings have profound importance as more than 75% of diabetics die from cardiovascular disease and cardiovascular morbidity in women is greater than that in men (28). Additionally, the lifetime residual risk of new-onset diabetes in women 50 years old is (30%) 3000 cases per 10,000 women; for Hispanic and African-American women, it is (40%) 4000 cases per 10,000 women (29). As such, W H I EP and HERS predict a lifetime residual risk reduction of 0.06 -
Menopause <10
I
10-19
'"
-
"!
0.89
1
~"
'
"
lI
l
-BIB -
J I
I
0
"
I
0.5
-.-
I
1.5
-
Placebo CEE+MPA
r'-
........................
r ~ .'2-.'=-
0.04
m =,,. ,,..
=
0.03
t - - /
,,-
0
,
I
i
......
2.0
i
Timing of initiation of hormone therapy (conjugated equine estrogen 0.625 mg plus medroxyprogesterone acetate 2.5 mg daily [CEE+MPA]) according to years since menopause at randomization and its effect on coronary heart disease events in the Women's Health Initiative (WHI) trial.
a,
d
f
0.01
0.00
....
2.5
Hazard Ratio (95% Confidence Interval)
/
I-"
22 o.02 E~ =i5
BII
'
1.0
-
m o~
1.71
~r
-
r
1.22
i
__~>20
0.05
i. . . . . . . . . . .
Years Placebo (n) 7627 CEE+MPA (n) 8014
7618 7894
7403 7785
7271 7675
7049 7461
5049 5418
2528 2842
922 1252
FIGURE 39.7
FIGURE 39.8 Incidence of new-onset diabetes mellitus in women randomized to conjugated equine estrogen 0.625 mg plus medroxyprogesterone acetate 2.5 mg daily (CEE +MPA) and placebo in the Women's Health Initiative (WHI) trial.
535
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy new-onset diabetes of 630-1050 cases per 10,000 women 50 years old. Currently, there are 40 million postmenopausal women in the United States who are 50 years old or older. With the growing epidemic of obesity and diabetes linked with the growing number of women entering menopause, these findings have immense public health implications. Although the mechanism for the reduction in newonset diabetes mellitus in W H I - E P and HERS is not known, W H I - E P results are consistent with a large body of data that show improvement in insulin resistance with hormone therapy (26). Reduction in insulin resistance has been shown to significantly reduce new-onset diabetes mellitus by 50% in insulin-resistant women (30). Furthermore, the improvement in insulin resistance results in a significant reduction in the progression of subclinical atherosclerosis (31). The W H I - E cohort consisted of 10,739 postmenopausal women (Fig. 39.9) who were on average 63.6 ___7.3 years old (two-thirds were 60 years and older) with an average BMI of 30.1 kg/m 2 (45% of the subjects had a BMI of 30 kg/m 2 or greater); 14.2% of the women had known cardiovascular disease at baseline (22). Women with menopausal symptoms such as hot flushes that would have precluded randomization to placebo were excluded. After an average 6.8 years of randomized treatment, there was no significant difference between the CEE-treated (annualized percentage of events 0.53) and placebo-treated (annualized percentage of events 0.56) groups in the incidence of the primary trial end point of nonfatal myocardial infarction and coronary heart disease death (HR, 0.95; 95% unadjusted CI, 0.79-1.16; 95% adjusted CI, 0.76-1.19) (23). Although initially reported as statistically significant in year 7 and beyond in the W H I - E cohort (p = 0.02 for trend) (22), the time trend toward a reduction in coronary heart disease in the CEE-treated
group compared with the placebo-treated group was not significant in the fully adjudicated data (p = 0.14 for trend) (23). Unlike C E E + M P A in the W H I - E P trial, CEE alone was not associated with an early increase in coronary heart disease events in the W H I - E trial (22,23). Overall mortality did not differ between treatment groups in W H I - E (22). Analysis of the fully adjudicated W H I - E data indicates that the treatment effect of CEE therapy on coronary heart disease is age dependent (p-value for interaction = 0.07) (Fig. 39.10) (Table 39.3) (23) and that the reduction of coronary heart disease is most relevant for women less than 60 years old (Fig. 39.11) (23). Women in the CEE group who were 50 to 59 years old had a 37% reduced risk for coronary heart disease (nonfatal myocardial infarction or coronary heart disease death) compared with placebo (HR, 0.63; 95% CI, 0.36-1.08), those who were 60 to 69 years old had a 6% reduced risk for coronary heart disease (HR, 0.94; 95% CI, 0.71-1.24), and those who were 70 to 79 years old had an 11% increased risk for coronary heart disease (HR, 1.11; 95% CI, 0.82-1.52) (p value for interaction = 0.07) (see Table 39.3) (23). Several composite coronary heart disease events were significantly reduced in the CEE-treated group relative to the placebo-treated group in the 50-59 year age range but not in the 60-69 or 70-79 year age ranges (see Table 39.3) (23). In the CEEtreated women in the 50-59 year age range, coronary artery bypass grafting or percutaneous coronary intervention was significantly reduced 45% (HR, 0.55; 95% CI, 0.35-0.86); combined nonfatal myocardial infarction, coronary heart disease death, coronary artery bypass grafting, and percutaneous coronary intervention was significantly reduced 34% (HR, 0.66; 95% CI, 0.44-0.97); and combined nonfatal myocardial infarction, coronary heart disease death, coronary artery bypass grafting, percutaneous coronary
A g e (Years) at Randomization 50-59
0.63
I
I
~-m
eo-~9
,
70-79
o~
I |
;
o'.5
0.%
t i
,,,
.... .,,,m
1.o
...................
:
1'.25
l'.S
l'.e
Hazard Ratio (95% Confidence Interval)
FIOURE 39.9 Study flow chart of the conjugated equine estrogen 0.625 mg daily (CEE) arm of the Women's Health Initiative (WHI) trial. HT = hormone therapy.
FIGURE 39.10 Initiation of hormone therapy (conjugated equine estrogen 0.625 mg daily [CEE]) and its effect on coronary heart disease events according to age at randomization in the Women's Health Initiative (WHI) trial.
CHD event
Number (Annualized Percentage) of Coronary Heart Disease Events with Conjugated Equine Estrogen or Placebo by Age at Baseline from the Women's Health Initiative Estrogen Trial
Hazard ratio (95% CI)
P-value 1
70-79 years old Placebo (n = 1291)
0.07 0.18 0.09 0.09
60-69 years old CEE (n = 1286)
1.11 (0.82-1.52) 1.12 (0.78-1.60) 1.04 (0.78-1.39) 1.08 (0.85-1.38)
0.11
50-59 years old
Hazard ratio (95% CI)
77 (0.86) 56 (0.63) 94 (1.06) 130 (1.46)
1.05 (0.84-1.33)
Placebo (n = 2465)
84 (0.96) 62 (0.71) 95 (1.08) 137 (1.56)
141 (1.58)
CEE (n = 2387)
0.94 (0.71-1.24) 1.03 (0.76-1.41) 0.99 (0.78-1.27) 1.02 (0.83-1.25)
148 (1.69)
Hazard ratio (95% CI)
105 80 130 177
0.98 (0.80-1.20)
Placebo (n = 1637) 96 80 129 177
194 (1.12)
(0.61) (0.46) (0.75) (1.02)
0.63 (0.36-1.08) 0.59 (0.34-1.02) 0.55 (0.35-0.86) 0.66 (0.44-0.97)
186 (1.11)
(0.57) (0.48) (0.77) (1.06)
34 35 52 65 0.66 (0.45-0.96)
(0.27) (0.28) (0.42) (0.52)
21 21 29 42 70 (0.56)
(0.17) (0.17) (0.24) (0.35)
46 (0.38)
CEE (n = 1637)
TABLE 39.3
CHD (MI or CD) Angina 2 CABG or PCI MI, CD, CABG, and PCI MI, CD, CABG, PCI, and Angina 2
1P-value for interaction. 2Angina pectoris confirmed with hospitalization for angina and confirmatory stress test or coronary artery disease by angiography, CHD, coronary heart disease; CEE, conjugated equine estrogen; CI, confidence interval; MI, nonfatal myocardial infarction; CD, coronary heart disease death; CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention.
o
O',
rao
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy
537
FIGURE 39.11 Percent (incidence) and absolute number for coronary heart disease events by year according to age in the Women's Health Initiative (WHI) estrogen trial.A: Age 50-59 years. B: Age 60-69 years. C: Age 70-79 years. CEE = conjugated equine estrogen 0.625 mg daily.
Body Mass Index (kg/m 2) <25
0.76 m am
II
! 0.87
25 to <30
I'
t
""
m
1.11
'l1
>3O
0'5
o:~5
baseline BMI less than 30 kg/m 2 had a reduction in coronary heart disease events, whereas women with a baseline BMI 30 kg/m 2 or more did not (Fig. 39.12) (23). As such, approximately 50% of the women randomized to the W H I E trial were in the weight range unlikely to show a reduction in coronary heart disease events with CEE therapy (see IX. Evidence for a Cardioprotective Effect of Postmenopausal Hormone Therapy According to Body Mass Index).
lib
.
.
1.o
.
.
...........
.
.
.
I
.
1'.2s
1'.5
Hazard Ratio (95% Confidence Interval)
FIGURE 39.12 Effect of hormone therapy (conjugated equine estrogen 0.625 mg daily [CEE]) on coronary heart disease events according to baseline body mass index in the Women's Health Initiative (WHI) trial.
intervention, and confirmed angina pectoris was significantly reduced 34% (HR, 0.66; 95% CI, 0.45-0.96). Confirmed angina pectoris in the CEE-treated women in the 5 0 - 5 9 year age range was reduced 41% (HR, 0.59; 95% CI, 0.34-1.02) relative to placebo. Because only one-third of the W H I - E cohort was less than 60 years old (only 10% of the women were less than 55 years old) at baseline, the majority of women randomized to the W H I - E trial were in the age range unlikely to show a cardiovascular benefit with estrogen therapy. An additional important factor in W H I - E was baseline weight. Women in W H I - E who were randomized to CEE alone and had a
III. A R T E R I A L I M A G I N G TRIALS OF POSTMENOPAUSAL HORMONE THERAPY Arterial imaging trials are important in that they provide direct imaging of atherosclerosis and measurement of the effects of an intervention on atherosclerosis progression, the pathophysiologic precursor to clinical cardiovascular events (32). Measurement of atherosclerosis progression by serial coronary angiography and noninvasive carotid artery intima media thickness (IMT) are predictive of clinical cardiovascular events and as such closely parallel morbidity/mortality trials of cardiovascular end points (32-35). There have been five published randomized controlled serial arterial imaging trials of hormone therapy in postmenopausal women: three studies using quantitative coronary angiography to measure arterial lesions and two studies using B-mode ultrasound to measure arterial wall thickness (carotid intima media thickness) (Table 39.4). These trials have been of even shorter duration (less than 1 year to 3 years) than the morbidity/mortality trials.
538
HODIS AND MACK TABLE 39.4
Coronary Angiographic and Carotid Artery Intima Media Thickness Trials of Postmenopausal Hormone Therapy Mean age at entry (years)
Time from menopause (years)1
ERA (36)
65.8
23
WAVE (37)
65
WELL-HART (38)
63.5
18.2
62.2
13.4
Trial
Mean duration (years)
Treatment
Outcome
P-value
Coronary Angiographic Trials
> 15
Oral CEE 0.625 mg alone or continuous MPA 2.5 mg Oral CEE 0.625 mg + continuous MPA 2.5 mg Oral 1713-E2 1 mg alone or sequential MPA 5 mg
Placebo CEE CEE + MPA Placebo CEE + MPA
-0.09 + 0.02 mm 0.38 -0.09 + 0.02 mm -0.12 + 0.02 mm -0.010 _+ 0.15 mm/y 0.17 -0.048 + 0.14 mm/y
3.3
Placebo 1713-E2 17~-E2 + MPA
1.89 + 0.78 %S 2.18 + 0.76 %S 1.24 + 0.80 %S
2.0
Overall Placebo 0.0036 mm/yr 1713-E2 -0.0017 mm/yr No lipid-lowering medication Placebo 0.0134 mm/yr 1713-E2 -0.0013 mm/yr Placebo 0.02 +_ 0.05 ram* 1713 + gest (hd) 0.03 _+ 0.05 mm* 1713 + gest (ld) 0.03 + 0.05 ram*
3.2
2.8
0.66
Carotid ArteryIMT Trials EPAT (39)
PHOREA (44)
NR
NR
Oral 1713-E2 1 mg alone
Oral 17~-E2 1 mg + gestodene 0.025 mg hd or ld
48 weeks
0.045
0.002
NS
1At randomization. IMT, intima-media thickness; NR, not reported; ERA, Estrogen Replacement AtherosclerosisTrial; WAVE, Women's Angiographic Vitamin and Estrogen Trial; WELL-HART, Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis RegressionTrial; EPAT, Estrogen in the Prevention of Atherosclerosis Trial; PHOREA, Postmenopausal Hormone Replacement against AtherosclerosisTrial; CEE, conjugated equine estrogen; MPA, medroxyprogesterone acetate; 1713-E2= 17[3-estradiol; hd, high dose, gestodene 0.025 mg days 17 to 28 every month; ld, low dose, gestodene 0.025 mg days 17 to 28 every 3 months; mm, millimeter change from baseline in minimum lumen diameter by quantitative coronary angiography (QCA); mm/y, millimeter per year change in minimum lumen diameter by Q_CA;%S, percent diameter stenosis change from baseline by QCA; mm/yr, millimeter per year change in common carotid artery intima-media thickness; mm*,change in carotid artery intima-media thickness over 48 weeks.
A. Estrogen Replacement Atherosclerosis Trial The Estrogen Replacement Atherosclerosis (ERA) Trial was a randomized, double-blind, placebo-controlled, serial coronary angiographic trial designed to test the effects of oral daily continuous combined C E E 0.625 mg plus MPA 2.5 mg (Prempro | and oral daily unopposed C E E 0.625 mg (Prematin | versus placebo on the progression of coronary artery disease (36). The ERA cohort consisted of 309 postmenopausal women with angiographically documented coronary artery disease, defined as at least one coronary artery lesion of 30% diameter stenosis. The women in ERA were on average 65.8 years old and 23 years postmenopausal at baseline. After a mean treatment period of 3.2 years, women randomized to C E E + M P A (n = 85), C E E alone (n = 79), and placebo (n = 84) all showed similar coronary artery disease progres-
sion rates measured as minimum lumen diameter by serial quantitative coronary angiography (see Table 39.4).
B. Women's Angiographic Vitamin and Estrogen Trial The Women's Angiographic Vitamin and Estrogen (WAVE) Trial was a randomized, double-blind, placebocontrolled, serial coronary angiographic trial of 403 postmenopausal women with angiographically documented coronary artery disease, defined as at least one coronary artery lesion of 15% to 75% diameter stenosis (37). Participants were randomized in a 2 x 2 factorial design to oral daily continuous combined C E E 0.625 mg plus M P A 2.5 mg (Prempro | (women with a hysterectomy received
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy oral daily unopposed CEE 0.625 mg [Premarin| or matching placebo and vitamin E 400 IU twice daily plus vitamin C 500 mg twice daily or matching placebo. The women in WAVE were on average 65 years old and more than 15 years postmenopausal at baseline. After a mean treatment period of 2.8 years, coronary artery disease progression measured as minimum lumen diameter was not significantly different between women receiving hormone therapy (n = 87) versus placebo (n = 77) (see Table 39.4). In addition, coronary artery disease progression was not significantly different among women receiving vitamin supplementation and placebo.
C. Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial The Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial (WELL-HART) was a randomized, double-blind, placebo-controlled, serial coronary angiographic trial designed to test the effects of oral daily micronized 1713-estradiol 1 mg and oral daily 17]3-estradiol 1 mg plus MPA 5 mg for 12 days of each month versus placebo on the progression of coronary artery disease (38). All participants had their low-density lipoprotein (LDL)-cholesterol level reduced to below 130 mg/dL with lipid-lowering therapy (predominantly statins) and lifestyle and dietary intervention. The WELL-HART cohort consisted of 226 postmenopausal women with angiographically documented coronary artery disease, defined as at least one coronary artery lesion of 20% diameter stenosis. The women in W E L L - H A R T were on average 63.5 _ 6.5 years old and were on average 18.2 years postmenopausal at baseline; 70% of the participants were members of a racial or ethnic minority group, and 51% had diabetes mellitus. After a median treatment period of 3.3 years, women randomized to 1713-estradiol alone (n - 54), sequentially opposed 17f3-estradiol (n - 53), and placebo (n = 59) all showed similar coronary artery disease progression rates measured as percent diameter stenosis by serial quantitative coronary angiography (see Table 39.4).
539
menopausal at baseline. Minorities comprised 42% of the cohort. The primary trial end point was the rate of change in the distal common carotid artery (CCA) far wall intima media thickness (IMT) in computer image-processed B-mode ultrasonograms obtained every 6 months (32,34,35,40,41). At the end of 2 years of treatment, there was a significant reduction in the progression of CCA IMT in those women randomized to 1713-estradiol versus those randomized to placebo (Fig. 39.13). As subjects with an LDL-cholesterol level greater than 160 mg/dL received lipid-lowering therapy, predominantly with a 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitor, preplanned subgroup analyses revealed two additional findings: (1) 1713-estradiol reduced the progression of subclinical atherosclerosis to the same degree as lipid-lowering therapy and (2) lipid-lowering therapy added to 17[3-estradiol did not appear to reduce the progression of subclinical any further than 1713-estradiol alone (see Fig. 39.13). Because EPAT was designed to specifically test the hypothesis that estrogen therapy reduces the progression of subclinical atherosclerosis in postmenopausal women, several trial design factors were taken into account. As 1713estradiol is the endogenously produced hormone, postmenopausal hormone replacement with a bio-identical estrogen preparation was chosen over the variety of other commercially available estrogen compounds. A regimen of 17]3-estradiol alone was chosen because data indicate that progestogens negate the beneficial cardiovascular effects of estrogen. Two exclusion criteria of particular importance, current smoking and previous use of estrogen therapy for more than 10 years were employed because of the possibility of reducing the detectability of a cardiovascular benefit of estrogen therapy under trial conditions. Active smoking increases the metabolism of estradiol and reduces serum estradiol levels and long-term use of estrogen therapy appears to be cardioprotective.
D. E s t r o g e n in the P r e v e n t i o n o f Atherosclerosis Trial The Estrogen in the Prevention of Atherosclerosis Trial (EPAT) was a randomized, double-blind, placebocontrolled, carotid artery ultrasound trial designed to test the hypothesis that estrogen therapy reduces the progression of subclinical atherosclerosis in healthy postmenopausal women without pre-existing cardiovascular disease (39). The 222 subjects randomized to EPAT received either oral micronized 17[3-estradiol 1 mg daily or placebo. The women in EPAT were on average 62.2 years old and 13.4 years post-
FIGURE 39.13 Rate of change in carotid artery intima media thickness over all subjects and those subjects taking and not taking lipid-lowering medications (meds) from the Estrogen in the Prevention of Atherosclerosis Trial (EPAT).
540 The primary outcome from EPAT is consistent with the observational studies that indicate that healthy postmenopausal women who use estrogen therapy (initiated at the time of menopause primarily for symptomatic relief of menopausal symptoms) have lower cardiovascular disease event rates than women who do not use estrogen therapy (3,4). Findings from EPAT are also consistent with the previous findings from the Asymptomatic Carotid Atherosclerosis Progression Study (ACAPS). ACAPS was a randomized, double-blind, placebo-controlled carotid artery IMT trial oflovastatin therapy in indMduals without pre-existing cardiovascular disease (42). Subgroup analyses from this trial indicated that the effects of estrogen therapy on the progression of subclinical atherosclerosis were similar to those of EPAT, whereby estrogen therapy (self-selected) reduced atherosclerosis progression in the women not receiving lipid-lowering therapy while not adding any additional effect on carotid artery IMT progression in women assigned to lovastatin (Fig. 39.14). In a study by de Kleijn, et al., 121 perimenopausal women with a mean age of 47 years were randomized to receive oral daily CEE 0.625 mg and norgestrel 0.15 mg on days 17-28, oral daily micronized 1713-estradiol 1.5 mg and desogestrel 0.15 mg on days 13-24 with 4 matched placebo pills taken on days 25-28, or placebo only (43). After 2 years of treatment, women randomized to hormone therapy showed a reducton in the progression of CCA IMT compared with placebo. However, a high 48% dropout rate prevented reliable conclusions from this trial. Although the sample size was limited, the direction of the observed effects was in agreement with other trials of hormone therapy in postmenopausal women, such as EPAT and the hormone therapy subgroup analysis from ACAPS. In addition, the magnitude of reduction in CCA IMT progression in the 1713-estradiol treatment arm in the study of de Kleijn et al. and EPAT were similar, -0.0014 mm/ year versus -0.0013 mm/year, respectively. The differences between the 1713-estradiol and placebo arms and the CEE and
FIGURE 39.14 Rate of change in carotid artery intima media thickness in subjects who were randomized to placebo or lovastatin in the Asymptomatic Carotid Atherosclerosis Progression Study (ACAPS) and who used or did not use estrogen therapy.
HODIS AND MACK
placebo arms from the de Kleijn study were of similar magnitude to the difference between 17[3-estradiol and placebo in EPAT,-0.0102 mm/year,-0.0159 mm/year, and -0.0147 mm/ year, respectively.
E. Postmenopausal Hormone Replacement against Atherosclerosis Trial The Postmenopausal Hormone Replacement against Atherosclerosis (PHOREA) trial was a randomized, single-blind, placebo-controlled trial of 321 postmenopausal women who were randomly assigned to oral daily 1713-estradiol 1 mg plus gestodene (a progestin) 0.025 mg on days 17-28 of each 4-week cycle, oral 17[3-estradiol 1 mg every day plus gestodene 0.025 mg on days 17-28 of each third cycle only (every 3 months), or placebo (44). The study participants were 40 to 70 years old, and although they did not have clinical evidence of cardiovascular disease, they were preselected for a carotid artery IMT greater than 1 mm. After a follow-up of 48 weeks (12 cycles), there was no difference in the progression of carotid artery IMT among the three treatment groups. Several important points must be noted when interpreting the results from PHOREA. First, although the women randomized to this trial did not have clinical evidence of cardiovascular disease, defining a cohort with a carotid artery IMT greater than 1 mm preselects individuals with advanced atherosclerosis (32,35). Second, there was no unopposed 17[3-estradiol comparison arm in this trial. Third, a treatment effect on the progression of atherosclerosis with 1 year or less of any intervention is unlikely. Although a statistically significant reduction in the progression of atherosclerosis after I year of intervention has been reported with aggressive lipid alteration in post hoc analyses, these findings are not consistent (32). Reduction in cardiovascular disease events with anti-atherosclerosis therapies consistently shows divergence between intervention and placebo after 2 years of therapy. Therefore, based on the results of many atherosclerosis and cardiovascular mortality/morbidity trials, a significant reduction in the progression of atherosclerosis with only 48 weeks of hormone therapy is highly unlikely. Analysis of the data from EPAT (39) also indicated no significant treatment group difference in the progression of carotid artery IMT between estrogen replacement and placebo after 1 year of intervention. On the other hand, consistent with other antiatherosclerosis interventions, physiologic parameters such as arterial blood flow, vasoreactMty, and arterial stiffness improve over short treatment intervals with hormone therapy. Although arterial wall thickness (an anatomic measure of atherosclerosis) was not significantly different between treatment groups after 48 weeks of hormone therapy in PHOREA, carotid arterial stiffness (a physiologic measure of atherosclerosis)significantly improved in the every-third-month, estrogen-opposed hormone therapy group relative to placebo (45). This is an
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy important finding because carotid arterial stiffness is related to an increased risk of future cardiovascular and all-cause mortality independently from established risk factors (46).
F. Summary In HERS and WHI-EP, randomization to oral daily continuous combined CEE + M P A resulted in an early increase in coronary heart disease events followed by a significant trend in the reduction of events over time. Although the pattern of treatment effect with CEE + M P A in HERS and W H I - E P was similar, the cohorts of the 2 trials were different: women with previous coronary heart disease in HERS and women without previous coronary heart disease in W H I - E E In both cohorts, with and without previous coronary heart disease, a trend toward reduction of coronary heart disease with CEE + M P A relative to placebo became evident after 4 to 6 years of use. The remainder of trials with randomization to estrogen alone, including the W H I - E trial, did not show an early increase in coronary heart disease events. W H I - E also showed an age-dependent reduction of coronary heart disease events particularly among women less than 60 years old, consistent with the significant reduction in atherosclerosis progression with estrogen alone in EPAT. As such, current data indicate that the pattern of treatment effect on coronary heart disease may differ between oral continuous combined CEE + M P A and estrogen alone. Duration of hormone therapy and timing of initiation of hormone therapy relative to menopause and according to age and BMI appear to be important factors in determining whether hormone therapy is effective in reducing risk of coronary heart disease. ERA, W E L L - H A R T and WAVE indicate that neither CEE alone, nor continuous combined C E E + M P A or estradiol alone, nor estradiol plus sequential MPA affect the progression of coronary atherosclerosis over a 2- to 3-year treatment period in older women with angiographically documented coronary artery disease who were on average more than 15 years postmenopausal. The coronary angiographic trial findings are concordant with those of the secondary prevention trials HERS and PHASE, in which no treatment effect of hormone therapy on coronary heart disease was demonstrated over a 2 - 3 year treatment period in older women who were also more than 10 years postmenopausal at the time of randomization. Women randomized to oral micronized 1713-estradiol alone had a significant reduction in the progression of subclinical atherosclerosis compared with women randomized to placebo in EPAT. The results from EPAT are consistent with the more than 40 observational studies indicating that postmenopausal women who use estrogen therapy have lower rates of coronary heart disease than postmenopausal women who do not use estrogen therapy. In addition, EPAT results are consistent with ESPRIT, which demonstrated a
541
nonsignificant reduction in overall and coronary heart disease mortality with estradiol alone. The cumulated data indicate that there is a time-dependent beneficial effect of hormone therapy on the reduction of coronary heart disease in postmenopausal women. As with all good scientific endeavors, the randomized controlled trials of hormone therapy and cardiovascular disease have generated questions and hypotheses. This has been particularly spawned by the fact that not all of the randomized controlled trial results are concordant with the body of observational and experimental literature within this field of investigation.
IV. DISCORDANCE BETWEEN RANDOMIZED CONTROLLED TRIALS AND OBSERVATIONAL STUDIES OF POSTMENOPAUSAL HORMONE THERAPY Of all randomized controlled trial outcomes of hormone therapy, coronary heart disease is the most discordant from observational studies. Whereas observational studies have demonstrated a consistent 30% to 50% reduction in coronary heart disease in users versus nonusers of hormone therapy (3,4), randomized controlled trials have shown an overall null effect on coronary heart disease outcomes in women randomized to hormone therapy versus placebo (see Table 39.1). Although on the surface this discordance appears inexplicable, close examination of the differences in cohort characteristics between randomized controlled trials and observational studies provides several legitimate and probable explanations for the discordance in outcomes between randomized controlled trials and observational studies (Table 39.5). Whether any one factor or a
TABLE 39.5 Differences between Randomized Controlled Trials and Observational Studies of Postmenopausal Hormone Therapy
Mean age or age range at enrollment (years) Duration of therapy (years) Time since menopause (years) Menopausal symptoms (flushing) Body mass index (mean)
Randomized controlled trials
Observational studies
>62
30-55
<7
> 10
>10
<6 (<2) 1
Excluded
Predominant
---29 kg/m2
~25 kg/m 2
aMore than 80% of the women initiated hormone therapy at the time of menopause.
542 combination of these factors accounts for the discordant outcomes between randomized controlled trials and observational studies is not known at the current time. However, it is clear that the women selected for randomized controlled trials were considerably different than the women studied in observational studies. Observational studies reflect the pattern of hormone use within the general population. As such, the women who used hormone therapy in these studies were relatively young (30 to 55 years old), recently postmenopausal (majority of the women initiated hormone therapy at the time of menopause), and were predominantly symptomatic, mainly with flushing and other menopausal symptoms as these were the primary reasons for seeking treatment. Additionally, the women studied who used hormone therapy did so for decades (10 to 40 years). The women who used hormone therapy in these studies were also relatively lean (approximate BMI of 25 kg/m2). On the other hand, women selected for randomized controlled trials were much older, with more than 90% of women older than 55 years (the upper age bound for women in observational studies) and on average more than 10 years beyond menopause when randomized (range 13-23 years). Because of the concern ofunblinding to treatment group, women with menopausal symptoms, predominantly flushing, were purposefully excluded from trial participation. Duration of therapy (1-6.8 years) in randomized controlled trials was also considerably less than that of women studied in observational studies (10-40 years). Additionally, the women randomized to controlled trials were on average overweight (approximate BMI of 29 kg/m2). It is clear from the comparison of the cumulated studies that the cohorts of women selected for randomized controlled trials were markedly different from the women studied from the general population in observational studies. As such, randomized controlled trials have yet to completely test the hormone cardioprotective hypothesis in the same population of women studied in observational studies from which the hypothesis was derived.
V. EVIDENCE FOR A CARDIOPROTECTIVE EFFECT ACCORDING TO DURATION OF POSTMENOPAUSAL H O R M O N E THERAPY The controlled trials with the longest randomized followup, HERS and WHI, provide evidence for a long-term beneficial effect of hormone therapy on coronary heart disease. In HERS, there was a statistically significant trend toward a reduction in coronary heart disease outcome in years 4 and 5 in the CEE+MPA-treated group compared with placebo (p = 0.009 for trend) (see Fig. 39.3). Compared with
HODIS AND MACK
placebo in the W H I - E P trial, there was a statistically significant trend toward a reduction in coronary heart disease outcome in year 6 and beyond in the CEE+MPA-treated group (p = 0.02 for trend) (see Fig. 39.6). Initially reported as statistically significant in year 7 and beyond for the overall W H I - E cohort (p = 0.02 for trend) (22), the time trend toward a reduction in coronary heart disease in the CEEtreated group compared with the placebo-treated group was not significant in the fully adjudicated data (p = 0.14 for trend) (23). The temporal trend in the reduction of coronary heart disease in the women less than 60 years old in W H I - E has not been reported (23). Data in regard to a beneficial effect of hormone therapy on cardiovascular disease according to duration of hormone therapy use is strongly supported by consistency between the W H I - E P randomized controlled trial and the estrogen plus progestin (E+P) data from the W H I observational study (12). As with other observational studies (3-11), the W H I observational study of 53,054 women showed a 50% decreased risk for coronary heart disease (RR, 0.50) and stroke (RR, 0.52) in E + P users relative to nonusers, which differed from the W H I - E P trial (Table 39.6) (12). However, in both the W H I - E P trial and the W H I observational study of E + P users (n = 17,503 women), the relative risks for coronary heart disease, stroke, and venous thromboembolic events (VTEs) associated with E + P therapy all consistently decreased with increasing duration of use (Table 39.7) (12). After 5 years of use of E + P, the relative risks for coronary heart disease, stroke, and VTE in both the W H I - E P clinical trial and the W H I observational study of E + P users were similar, and the risks for coronary heart disease and stroke for those women who used E + P were reduced relative to nonusers in both studies (Table 39.7) (12). Randomized controlled trial data are supported by other studies showing that longer duration of estrogen use is associated with a reduced risk for coronary heart disease. The Nurses' Health Study, the largest longitudinal observational study to date, was conducted in 121,700 female nurses ages 30 to 55 years (10). In a report of 70,533 postmenopausal women followed for up to 20 years in the Nurses' Health Study, risk of nonfatal myocardial infarction and coronary heart disease death after adjustment for cardiovascular risk factors and age was reduced 39% (RR, 0.61; 95% CI, 0.52-0.71) in current estrogen users relative to never users (47). Among women from the Nurses' Health Study with established coronary heart disease at enrollment (n = 2489), the trend for the reduction in recurrent myocardial infarction or coronary heart disease death with duration of hormone therapy use was significant (p for trend = 0.002) (48). Compared with never users, women who used hormone therapy for less than 1 year had a nonsignificant increased risk for recurrent myocardial infarction or coronary heart disease death of 25% (RR, 1.25; 95% CI, 0.78-2.00). On the other hand, in women who used hormone therapy for
543
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy TABLE 39.6
Incidence Rates of Cardiovascular Disease in the Women's Health Initiative Clinical Trial and Observational Study of Estrogen and Progestin Clinical trial Placebo
Number of women Coronary Heart Disease Number of events Annualized incidence (%) Stroke Number of events Annualized incidence (%) Venous Thromboembolism Number of events Annualized incidence (%)
Observational study
CEE + MPA
8102
Ratio
8506
Control I
Estrogen + progestin
35,551
17,503
Ratio
147 0.33
188 0.39
1.18
615 0.32
158 0.16
0.50
107 0.24
151 0.32
1.29
490 0.25
123 0.13
0.52
76 0.17
167 0.35
2.10
336 0.17
153 0.16
0.94
1Nonusers of estrogen plus progestin.
more than 2 years, the risk for recurrent myocardial infarction or coronary heart disease death was reduced 62% (RR, 0.38; 95% CI, 0.22-0.66) (48). In a case-control study of 559 women between ages 35 and 65 who suffered an acute myocardial infarction, compared to 1118 age-matched community controls from the same geographic area, the overall risk of acute myocardial infarction for ever-use of postmenopausal hormone therapy versus never-use was reduced 26% (adjusted odds ratio [OR], 0.74; 95% CI, 0.55-0.99) (49). The trend for TABLE 39.7
reduced risk of acute myocardial infarction with duration of hormone therapy was significant (p < 0.001) (Fig. 39.15). During the first 12 months of hormone therapy use, there was a nonsignificant increased risk of acute myocardial infarction of 14% (OR, 1.14; 95% CI, 0.72-1.80). Hormone therapy use for 1 3 - 6 0 months was associated with a small decreased risk of acute myocardial infarction of 15% (OR, 0.85; 95% CI, 0.55-1.29), and use for more than 60 months was associated with a large risk reduction for acute myocardial infarction of 58% (OR, 0.42; 95% CI, 0.24-0.73).
Hazard Ratios for Cardiovascular Disease in the Women's Health Initiative Clinical Trial and Observational Study According to Time Since Initiation of Estrogen + Progestin* Clinical trial
Duration of use (years)
Coronary Heart Disease <2 2-5 >5 Stroke <2 2-5 >5 Venous Thromboembolism <2 2-5 >5
Observational study
Number of CEE +MPA cases
Hazard ratio I
95% confidence interval
80 80 28
1.68 1.25 0.66
1.15-2.45 0.87-1.79 0.36-1.21
43 79 29
1.15 1.49 0.74
73 72 22
3.10 1.89 1.31
Number of estrogen + progestin cases
Hazard ratio 2
95% confidence interval
126
1.12 1.05 0.83
0.46-2.74 0.70-1.58 0.67-1.01
0.71-1.87 1.02-2.17 0.39-1.39
7 12 104
2.10 0.48 0.89
0.86-4.56 0.24-0.93 0.71 - 1.18
1.85-5.19 1.24-2.88 0.64-2.67
7 27 119
2.37 1.52 1.24
1.08-5.19 1.01-2.29 0.99-1.56
5 27
*Adjusted for age, race, body mass index, educational level, smoking, age at menopause, and physical activity. 1Relative to placebo. 2Relative to nonusers of hormone therapy.
544
HODIS AND MACK
Hormone Therapy Use (Months) Never Use <12
1.14
I
mm mm
0.85
13--60
I ....... -----
I
0.42
>60 f ............
0
I
0.5
......
1.0
|
!
1.5
2.0
Odds Ratio (95% Confidence Interval) FIGURE 39.15
Effect of hormone therapy on risk of nonfatal acute myocardial infarction according to duration of use.
Hormone therapy use was only associated with a significant reduction in coronary heart disease risk when used for more than 60 months. The pattern of reduction in risk of acute myocardial infarction with duration of use was similar both for estrogen alone and E + P, although the risk reduction was greater for combination hormone therapy (49). In the Puget Sound Group Health Cooperative casecontrol study of women aged 30 to 79 years, any use of postmenopausal estrogen was associated with a 28% (OR, TABLE 39.8
0.72; 95% CI, 0.59-0.88) decreased risk for myocardial infarction compared with never estrogen users (50). Among postmenopausal women currently using estrogen, a longer duration of use was associated with a reduced risk of myocardial infarction (Table 39.8). Hormone therapy use was only associated with a significant reduction in coronary heart disease risk when used for greater than 8.2 years. Among subjects who used estrogen in the past but not currently, there was no evidence of deceasing risk with increasing duration of use of estrogen. Examination of the past estrogen users also indicated that women who had discontinued the use of estrogen within 8 months (0.7 years) had a higher risk for myocardial infarction than those who discontinued estrogen within 0.7 to 2.7 years before their myocardial infarction (OR, 1.43; 95% CI, 0.67-3.06). These data support the interpretation that continued hormone therapy is required to maintain the beneficial effects of hormone therapy on the reduction of coronary heart disease events, as seen in the latter years of the HERS randomized trial but rapidly lost in the open-label follow-up, in which the majority of subjects originally randomized to hormone therapy discontinued therapy or had low compliance (17). Similar results were reported in a study of 24,420 postmenopausal diabetic women, in which the risk for myocardial infarction was unchanged (HR, 1.03; 95% CI, 0.74-1.44) among women whose current use of hormone therapy was less than 1 year, whereas among those women who used
Duration of Estrogen Use and Risk for Nonfatal Myocardial Infarction: Puget Sound Group Health Cooperative Case-Control Study ,,
Estrogen use Never Ever
Duration (years) >0-<1.8 1.8-<4.2 4.2-<8.2 ---8.2 Current
Duration (years) >0-<1.8 1.8-<4.2 4.2-<8.2 _>8.2 Past
Duration (years) >0-<1.8 1.8-<4.2 4.2-<8.2 ---8.2
No. control subjects
Odds ratio
95% confidence interval
621 229
1274 700
1.00 0.72
Reference 0.59-0.88
95 51 52 31 126
267 171 145 117 411
0.77
0.58-1.02 0.49-1.00 0.51-1.06 0.39-0.93 0.55-0.89
No. case subjects
P for trend
0.35 0.70 0.74 0.60 0.70
0.05 40 29 33 24 103
101 104 103 103 289
0.91 0.70 0.65 0.55 0.74
0.60-1.38 0.45-1.10 0.42-1.01 0.34-0.88 0.57-0.96
55 22 19 7
166 67 42 14
0.69 0.70 0.93 0.96
0.49-0.97 0.41-1.18 0.51-1.69 0.36-2.52
0.29
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy
hormone therapy for I year or more, the risk for myocardial infarction was reduced 19% (HR, 0.81; 95% CI, 0.66-1.00) (51). In addition, results similar to those of acute myocardial infarction have been reported for cardiac death (52). Compared with nonusers of estrogen, women who used estrogen for less than 3 years, 4 - 1 4 years, and 15 years or more had relative risks for cardiac death of 0.88, 0.78, and 0.53, respectively (p for trend <0.05). Consistent with observational and case-control studies are cross-sectional imaging studies indicating that longer duration of hormone therapy (9 to 10 years) relative to shorter-term use is associated with significantly less coronary artery plaque burden measured by coronary artery calcium (53,54). Duration of hormone therapy is a significant predictor of coronary artery calcium after adjustment for age and cardiovascular risk factors, with maximum benefit of hormone therapy on coronary artery calcium observed after 23 years of use (53). The reduction in risk of coronary heart disease with prolonged use of hormone therapy (more than 6 years) is consistent across the large randomized controlled trials (HERS and WHI-EP), observational studies, case-control studies, and arterial imaging studies. Comparison of these results from randomized controlled trials (HERS and WHI-EP), observational studies, and case-control studies indicate that selection bias and confounding do not explain the consistent evidence that hormone therapy is associated with a duration-dependent lowering of coronary heart disease risk. The W H I trial along with observational studies suggest that this durationdependent lowering may predominantly, if not only, manifest in women initiating hormone therapy in close proximity to menopause (within 6 years) or before 60 years of age.
VI. EVIDENCE FOR A CARDIOPROTECTIVE EFFECT ACCORDING TO TIMING OF INITIATION OF POSTMENOPAUSAL H O R M O N E THERAPY In randomized controlled trials, the effect of hormone therapy on coronary heart disease appears also to be related to the timing of initiation of therapy relative to menopause (55). This is evident in both arms of the largest randomized controlled trial, W H I (21-24). In WHI-EP, women who were randomized to CEE +MPA within 10 years of menopause had an 11% reduction in coronary heart disease events, those randomized between 10 and 20 years after menopause had a 22% increased risk of coronary heart disease, and those randomized more than 20 years after menopause had a 71% increased risk of coronary heart disease (Fig. 39.7) (p = 0.036 for trend) (24). Data from the Nurses' Health Study additionally indicate that the beneficial coronary heart disease effects
545
of hormone therapy depend on the time from menopause when hormone therapy is initiated (24). Women who initiated hormone therapy within 4 years of menopause had a statistically significant reduced risk for coronary heart disease (RR, 0.62; 95% CI, 0.52-0.76 for estrogen alone; RR, 0.71; 95% CI, 0.56-0.89 for E + P), whereas women who initiated hormone therapy 10 or more years after menopause had a nonsignificantly reduced risk for coronary heart disease (RR, 0.87; 0.69-1.10 for estrogen alone; RR, 0.90; 95% CI, 0.62-1.29 for E + P) (24). In W H I - E , hysterectomized women who were randomized to CEE therapy at ages 50-59 years had less frequent coronary heart disease events than those women randomized to placebo, whereas coronary heart disease events in women who were randomized at ages 60-69 years and 70-79 years did not differ between treatment groups (p = 0.07 for trend) (see Table 39.3) (23). In W H I - E , time from menopause cannot be determined because the cohort was comprised of hysterectomized women. However, the beneficial effect of CEE on coronary heart disease according to age at randomization in W H I - E in hysterectomized women is consistent with the beneficial effect of C E E + M P A on coronary heart disease according to time since menopause at randomization in W H I - E P in nonhysterectomized women. As sister studies, W E L L - H A R T (38) and EPAT (39) demonstrated the role of timing of hormone therapy in obtaining a beneficial effect of estrogen on atherosclerosis progression. Together, W E L L - H A R T and EPAT were designed to determine the effects of hormone therapy on the progression of atherosclerosis in postmenopausal women with and without pre-existing atherosclerotic vascular disease, respectively. The different outcome of no effect of hormone therapy on atherosclerosis progression in W E L L HART and a reduction in atherosclerosis progression in EPAT may be related to the timing of the intervention relative to the stage of atherosclerosis as reflected by the imaging methods used in these two trials. Q.uantitative coronary angiography used in W E L L - H A R T is used to measure symptomatic late-stage atherosclerosis, whereas carotid artery wall thickness used in EPAT is used to evaluate asymptomatic early subclinical atherosclerosis (32). Observational studies support the findings that timing of hormone therapy initiation influences the cardiovascular benefit of hormone therapy. Almost all women using hormones studied in observational studies initiated hormone therapy within 6 years of menopause (see Table 39.5); 80% of the hormone users who participated in the Nurses' Health Study initiated hormone therapy at the time of menopause (24,56). In addition, the concept of timing of initiation of hormone therapy and cardioprotection is supported by biologic effects in other tissues. A tissue-protective effect appears greater with early initiation of hormone therapy in the skeletal (57) and central nervous system (58,59) as it concerns osteoporosis and cognitive decline, respectively. Among
546 women with normal cognitive function at baseline, hormone therapy appears to reduce the risk of cognitive decline; the same effect is not apparent in women who have belowaverage cognitive function at baseline when hormone therapy is initiated (58,59). The concept of timing of initiation of hormone therapy and cardiovascular benefit is supported by other lines of evidence including data from animal studies. Experiments of estrogen intervention for surgical menopause using cynomolgus monkeys, a well-accepted non-human primate model for human atherosclerosis, eloquently illustrate the concept of vascular benefit dependent upon timing of hormone therapy. In the first experiment, monkeys had very little or no atherosclerosis prior to being made surgically menopausal. At the time of surgical menopause, the monkeys were fed an atherogenic diet and also given estrogen. Relative to a placebo-treated group of surgically menopausal monkeys also fed an atherogenic diet, there was a 70% reduction in the development of atherosclerosis in the estrogen-treated monkeys (60,61). In the second experiment, a moderate amount of atherosclerosis was initiated premenopausally, and then the monkeys were made surgically menopausal. At the time of surgical menopause, estrogen was given to the monkeys and the atherogenic diet was continued. Relative to a placebo-treated group of surgically menopausal monkeys also fed an atherogenic diet under the same conditions, there was a 50% reduction in development of atherosclerosis in the estrogen-treated monkeys (62). In the third experiment, monkeys were made surgically menopausal and immediately fed an atherogenic diet. However, in this experiment, estrogen therapy was delayed for 2 years (equivalent to approximately 6 years in humans) after the monkeys were made surgically menopausal before it was given to the monkeys. When the estrogen was given to the monkeys, the atherogenic diet was stopped. After 2 years of estrogen therapy, and even with the atherogenic diet halted, there was no significant difference in the amount of atherosclerosis in the estrogen-treated monkeys compared with the placebo-treated monkeys (63). The same concept of a differing effect of estrogen on new and established atherosclerotic lesions has also been demonstrated in apolipoprotein E deficient mice (64). Estrogen administration failed to reduce established atherosclerotic lesions, whereas it prevented the formation of new lesions when initiated at the time of atherogenesis. The clinical implications of these animal studies are profound because they indicate that once atherosclerosis is established, estrogen therapy initiated years after menopause has no effect in reducing plaque burden. On the other hand, if estrogen therapy is initiated simultaneously with menopause, the extent of atherosclerosis is markedly reduced. The randomized controlled trial results are consistent with these non-human primate data. The study designs and results from HERS, ESPRIT, PHASE, WELL-HART, ERA,
HODIS AND MACK
WAVE, and W H I are consistent with the non-human primate studies in which hormone therapy was initiated a decade or more after menopause, when atherosclerosis was already established and/or the arterial wall was less responsive to hormonal manipulation (see Tables 39.1 and 39.4). HERS, ESPRIT, PHASE, WELL-HART, ERA, and WAVE were conducted in older women with pre-existing cardiovascular disease, and the time from menopause to randomization to hormone therapy was on average 18 to 23 years. The age distribution in W H I was similar to these trials, with 90% of the cohort over 55 years old and twothirds of the women 10 years or more beyond menopause when randomized (25). Additionally, as reviewed earlier, the women closest to menopause when randomized to W H I showed a reduction in coronary heart disease events compared with women randomized a decade or more after menopause (see Figs. 39.7 and 39.10) (21,23,24). On the other hand, the design and results of EPAT (see Table 39.4) (39) and the study by de Kleijn et al. (43) are consistent with the non-human primate studies in which estrogen therapy was initiated earlier in the postmenopausal period when the atherosclerotic process was relatively young and the arterial wall still responsive to hormonal manipulation. The time from menopause to randomization was approximately 5 to 10 years shorter in EPAT relative to the other randomized controlled arterial imaging trials, and in the de Kleijn et al. study, hormone therapy was initiated in the perimenopausal period. It is important to note that observational studies are also consistent with the non-human primate studies in which estrogen therapy was initiated early postmenopausally when the atherosclerotic process was relatively early. The majority of observational studies were conducted in predominantly healthy women who initiated mainly unopposed estrogen therapy in the early postmenopausal period primarily for symptomatic relief of menopausal symptoms, a population in which observational studies have consistently shown a reduction in coronary heart disease events (3-12). Almost all (more than 95%) of women in observational studies initiated hormone therapy within 6 years of menopause. EPAT has been the closest test of the observational findings thus far, as it was a randomized controlled trial of unopposed estrogen therapy administered relatively early in postmenopausal healthy women. Animal data provide a strong pathobiologic explanation for the discordant data between observational studies and randomized controlled trials. The animal data clearly indicate that the timing of initiation of estrogen in relation to menopause and the extent of pre-existing atherosclerosis determine whether estrogen will be effective in preventing the development and progression of atherosclerosis. Data from randomized controlled trials, including W H I and EPAT, are consistent with animal studies indicating a beneficial effect of hormones on atherosclerosis and coronary
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy
547
heart disease with early postmenopausal hormone initiation. In total, animal and human data support the hypothesis that estrogen reduces the progression of atherosclerosis if initiated early relative to the time since menopause, when the vascular endothelium is in a relatively healthy state, versus later when the endothelium has lost its responsiveness to estrogen. Studies of other tissues such as skeletal tissue also support this hypothesis. The window of opportunity for the greatest benefit of estrogen on the arterial wall appears to be within 6 years of menopause.
VII. EVIDENCE THAT THE CARDIOPROTECTIVE EFFECT OF POSTMENOPAUSAL HORMONE THERAPY REQUIRES HEALTHY ENDOTHELIUM The findings reviewed previously support the healthy endothelium concept, whereby estrogen benefits are receptor mediated and depend on the integrity and functional status of the vascular endothelium. Advancing age and atherosclerosis lead to vessel wall injury with depletion of estrogen receptors and/or their function (perhaps through methylation of the estrogen receptor gene) and compromise of endothelial function (65). Reduction or lack of estrogen receptor gene expression with atherosclerosis may result in decreased vascular tissue responsivity to estrogen (66). The expression of estrogen receptors is directly influenced by the circulating level of estrogen. Estrogen receptors normally found in endothelial and smooth muscle cells disappear soon after menopause when the circulating level of estrogen is low. Preservation and function of estrogen receptors after menopause may require maintenance of premenopausal circulating levels of estrogen. In humans, vasodilation and increased blood flow are the normal responses of healthy intact vascular endothelium to estrogen administration. Studies have consistently shown that younger women and those women without cardiovascular disease and cardiovascular disease risk factors have a normal response to estrogen administration with increased brachial artery flow-mediated dilation (67), increased myocardial blood flow (68), and increased coronary artery dilation (69), whereas elderly women and those women with established cardiovascular disease and risk factors for cardiovascular disease do not have these normal vascular changes in response to estrogen administration (Fig. 39.16). In animal models and as demonstrated in EPAT and WELL-HART, as reviewed previously, the antiatherogenic effect of estrogen depends on the pre-existing pathologic state of the vessel wall. This concept has also been demonstrated in an endothelium-damaged, cholesterol-clamped rabbit model (70). At the same plasma cholesterol concentra-
FIGURE 39.16 Effect of hormone therapy on brachial artery vasodilation in women according to health status with or without risk factors. Risk factors = cardiovascular risk factors, diseases, and/or medications.
tion, cholesterol accumulation in undamaged rabbit aorta is almost completely inhibited by estrogen. In balloon-injured aorta, re-endothelialized areas accumulate similar amounts of cholesterol in estrogen and placebo-treated animals, whereas de-endothelialized areas accumulate significantly more cholesterol in estrogen-treated animals than in placebo-treated animals. These data clearly indicate that the antiatherosclerotic action of estrogen is dependent on a healthy intact endothelium. The inhibition of aortic cholesterol accumulation by estrogen is mediated independently of plasma cholesterol concentration, suggesting that estrogen in some way increases the resistance of healthy intact endothelium to atherosclerosis. However, the estrogen atheroprotective effect is reversed in the damaged arterial wall where the endothelium is denuded (70). These findings have also been reported in a non-human primate model, in which CEE inhibited progression of atherosclerosis in animals with intact endothelium but had no effect on intimal hyperplasia after balloon injury (71). Although there were relatively few menopausally symptomatic women with hot flushes randomized to WHI (less than 10%) as women with severe menopausal symptoms were excluded from the trial, the data derived from this small sample are consistent with the healthy endothelium concept. Data from WHI-EP (21) showed that the women randomized to C E E + M P A who were 50 to 59 years old and did not have hot flushes had an elevated relative risk of 1.98 for coronary heart disease, whereas women of the same age range who had hot flushes (n - 476) (25) did not have an elevated relative risk (Fig. 39.17). Women with menopausal symptoms may be those women who have the lowest estrogen tissue levels and/or healthier vascular tissue, and hence have the greatest tissue responsiveness and benefit from exogenously administered estrogen. A requirement of both flushing (healthy endothelium) and early postmenopausal initiation of hormones may account for the lack of a
548
HODIS
Hot Flashes
0.95i
Yes
I
i .....
No
i. . . . . . .
I
0.5
I
1.0
I
BB
I
mm
!1 0
1.98
....
I .............
1.5
!
2.0
"
i
2.5
Hazard Ratio (95% Confidence Interval) FIGURE 39.17 Effect of conjugated equine estrogen 0.625 mg plus medroxyprogesterone acetate 2.5 mg daily (CEE + MPA) on coronary heart disease in postmenopausal women ages 50-59 years who did or did not have hot flushes in the Women's Health Initiative (WHI) trial.
trend in the treatment effect when age alone was examined in the W H I - E P (21). In observational studies, women who had menopausal symptoms consisting predominantly of hot flushes were the most likely to initiate hormone therapy. Vasomotor instability manifested as hot flushes may be the critical sign of a healthy endothelium and vessel wall responsivity to hormone therapy. Early restoration or continued maintenance of estrogen receptors and function with circulating and/or tissue estrogen levels comparable to premenopausal levels through exogenous estrogen administration may be necessary for the cardioprotection afforded by estrogen therapy. Loss of cardioprotection with cessation of hormone therapy, as seen in the HERS follow-up and in observational studies, suggest that continued hormone therapy is necessary for maintenance of estrogen receptor function and cardioprotection. Both animal and human studies strongly suggest that estrogen therapy is more effective in maintaining vascular health than in treating established vascular disease. These data also suggest that a beneficial vascular effect of estrogen therapy is more likely expressed in healthy women with healthy endothelium who are close to menopause.
VIII. EVIDENCE FOR A DUALITY OF CEE + MPA T H E R A P Y O N C O R O N A R Y H E A R T DISEASE Atherosclerosis is an age-dependent progressive process with different stages of responsiveness to prevention and treatment. A clinical cardiovascular event is the final result of a chronic process of atherosclerosis progression throughout life. As such, although a group of individuals may not have clinically apparent cardiovascular disease, each individual within that group will have a certain degree of atherosclerosis,
ANDMACE
with the degree of involvement increasing with age. By 35 years of age, women typically have only fatty streaks and minimal atherosclerotic plaques (72). From 45 to 55 years of age, there is progression of atherosclerosis with development of atherosclerotic lesions (73). By 65 years of age, women begin to develop complicated atherosclerotic lesions that have the potential for plaque destabilization and rupture (74). Atherosclerotic lesions in younger women tend to consist of more cellular fibrous tissue and lipid-rich foam cells, found in earlier stages of development of atherosclerosis, and of lesser amounts of dense fibrous and calcified tissue found at later stages of development (75). Therefore, as demonstrated in several animal models including non-human primates, it is highly probable that the timing of intervention with hormones in women is important in whether they are effective in slowing the progression of atherosclerosis and averting a cardiovascular event, whether they have no effect, or whether they perhaps induce an event. Certain randomized controlled trials are consistent with the evolving concept of the duality of hormone therapy on atherothrombosis. Both HERS and W H I - E P suggest that daily continuous combined CEE +MPA has a dual effect on atherothrombosis, with an early increased risk of coronary heart disease events followed by a reduction in events (Figs. 39.3 and 39.6). In the latter stages, atherosclerosis consists of complex lesions comprised of a necrotic core of lipid gruel and other constituents, calcification, a fibrous cap, and infiltration of macrophages especially along the shoulder of the fibrous cap. Depending on the overall constituents of the plaque, size of the necrotic core, and extent of macrophage activation, these plaques can become vulnerable to fissuring, plaque rupture, and extrusion of the necrotic core into the vessel lumen, thereby activating the thrombotic pathway. Plaque rupture is the final event of atherosclerosis progression that leads to thrombosis and acute cessation of blood flow with deprivation of the end organ of blood leading to the cardiovascular event, myocardial infarction, or stroke. Plaque destabilization leading to rupture may be induced through the expression of certain enzymes with gelatinolytic activity. In particular, one of the more important of such enzymes belongs to the matrix metalloproteinase (MMP) family of gelatinolytic enzymes, MMP-9. MMP-9 expression and activity is upregulated by C E E + M P A therapy (Fig. 39.18) (76). The pattern of an early increased risk of coronary heart disease that was later offset by benefit in HERS and W H I - E P is also consistent with the findings from the Nurses' Health Study in the women with established coronary heart disease (n - 2489). Compared with never users, women who used hormone therapy for less than 1 year had a nonsignificant increased risk for recurrent myocardial infarction or coronary heart disease death of 25% (HR, 1.25; 95% CI, 0.78-2.00), whereas in women who used hormone therapy for more than 2 years, the risk was reduced 62% (RR, 0.38; 95% CI, 0.22-0.66)
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy
FIGURE 39.18 Expression of matrix metalloproteinase (MMP)-9 plasma level and gelatinolytic activity in postmenopausal women treated with conjugated equine estrogen 0.625 mg plus medroxyprogesterone acetate 2.5 mg daily (CEE + MPA).
(p for trend = 0.002) (48). Similar results were reported from a case-control study of 559 women with acute myocardial infarction matched to 1118 controls (49). During the first 12 months of hormone therapy use, there was a nonsignificant increased risk of acute myocardial infarction of 14% (OR, 1.14; 95% CI, 0.72-1.80), women who used hormone therapy for 13 to 60 months had a reduced risk of 15% (OR, 0.85; 95% CI, 0.55-1.29), and women who used hormone therapy for more than 60 months had a reduced risk of acute myocardial infarction of 58% (OR, 0.42; 95% CI, 0.24-0.73) (p for trend <0.001) (see Fig. 39.15)(49). Closer analysis of the HERS data in the first year of hormone therapy indicated that the risk for coronary heart disease events was greatest in the first 4 months of therapy, intermediate in the second 4 months of therapy, and then close to unity during the final 4 months of therapy (15). Occurrence of very early events in HERS strongly suggests precipitation of plaque rupture with C E E + M P A therapy. Further lines of evidence strongly suggest that plaque rupture most likely results from the induction of M M P activity. In vitro data indicate that hormones induce MMP-9 protein expression and activity from macrophages and endothelial cells, whereas statin therapy nullifies this effect of hormones (77). In HERS, the elevated risk of coronary heart disease (relative to placebo) that occurred in the first year of C E E + M P A therapy was evident in those women who only took C E E + M P A therapy but not in those women who concomitantly used statin therapy (see Fig. 39.4) (16). As such, the HERS randomized controlled trial data parallel in vitro mechanistic studies and highly suggest that induction of M M P activation through C E E + M P A therapy precipitates plaque rupture of vulnerable lesions while concomitant use of a statin nullifies the effect of
549
C E E + M P A on M M P activation and plaque rupture, preventing clinical coronary events. Once beyond the vulnerable period after rupture of susceptible lesions, those women in HERS who did not use a statin later expressed benefits in coronary heart disease event reduction. Other studies, such as the Northern California Kaiser Permanente Diabetes Registry Study of 24,420 postmenopausal diabetic women, provide other lines of evidence that hormone therapy may precipitate plaque rupture in susceptible lesions (51). In a subgroup of 580 women with a recent myocardial infarction (within 1 year of baseline), current users of hormone therapy had a significantly greater risk for a recurrent myocardial infarction (HR, 1.78; 95% CI, 1.06-2.98) than women not currently using hormone therapy. This is compared to a significantly reduced risk for myocardial infarction of 16% (HR, 0.84; 95% CI, 0.72-0.98) among current users without a recent myocardial infarction (Fig. 39.19). Acute myocardial infarction results from plaque rupture and typically indicates the presence of other plaques susceptible to potential rupture. After an initial myocardial infarction, the coronary artery system is "hot" and other coronary artery lesions remain susceptible to plaque rupture for up to 1 year or more. Cooling the coronary artery lesions down and reducing the risk for early recurrent myocardial infarctions is the premise behind using heparin, aspirin, and statins at the time and immediately following an acute myocardial infarction. Hormone therapy, through the upregulation of MMP-9 and other gelatinolytic enzymes, may have the potential for inducing plaque rupture of susceptible lesions immediately following an acute myocardial infarction while the coronary artery lesions remain "hot." On the other hand, 4065 young healthy postmenopausal women predominantly symptomatic with vasomotor symptoms, with a mean age of approximately 53.5 years and on average 4.9 years since menopause when randomized to
FIGURE 39.19 Association of current use of postmenopausal hormone therapy with risk of myocardial infarction in diabetic women with or without a recent myocardial infarction. Analyses adjusted for age, ethnicity, education, obesity, diabetes duration, hypoglycemic therapy, hemoglobin, hypertension, lipid-lowering medications, smoking, alcohol, and exercise.
550
hormone therapy, in two clinical trials did not exhibit increased cardiovascular event rates (78). Although MMP-9 may be necessary for plaque rupture, it is not sufficient because susceptible atheromatous substrate on which MMP-9 acts must be present for an event to occur (Fig. 39.20). Induction of MMP-9 in the absence of susceptible complicated plaques, as in younger healthy menopausally symptomatic women at the onset of menopause, is unlikely to result in a cardiovascular event. On the other hand, induction of MMP-9 in the presence of susceptible complicated plaques is likely to result in plaque rupture and a cardiovascular event. This is consistent with plaque destabilization in women with established atherosclerosis years after menopause, as seen in W H I - E P and HERS. Taken together, the cumulated data indicate that timing of initiation of hormone therapy relative to menopause and pathologic state of the vascular wall determines whether hormone therapy will be beneficial, have no effect, or will be detrimental. In perimenopausal or early postmenopausal (within 6 years) women in whom the vascular endothelium is healthy and responsive to estrogen, estrogen therapy may exert beneficial effects and prevent or delay atherosclerosis progression and coronary heart disease. As atherosclerosis develops, the vascular endothelium becomes diseased and unable to respond to estrogen therapy either beneficially, possibly due to reduced or nonfunctional estrogen receptors, or negatively as susceptible atherosclerotic plaques have not yet developed. Remote from menopause (approximately 10 years or more), complex plaques susceptible to rupture develop, and estrogen therapy may provoke plaque rupture. Time since menopause rather than natural or surgical menopause is the major factor associated with increased subclinical atherosclerosis (79), and a significant increase in the prevalence of plaque and intima media thickness occurs rapidly within 5 to 8 years after menopause (80). Because the progression of atherosclerosis appears to be rapid in the absence of estrogen and the cumulated data
FIGURE 39.20 Hypothetical schema of plaque rupture requiring both matrix metalloproteinase(MMP)-9 and susceptible atherosclerosissubstrate. HT, hormone therapy.
HODIS AND MACK
strongly indicate a window of opportunity for benefit of approximately 6 years, the timing of initiation of hormone therapy with respect to menopause may have important ramifications for the therapeutic efficacy of hormone therapy in delaying the progression of atherosclerosis and preventing coronary heart disease.
IX. EVIDENCE FOR A CARDIOPROTECTIVE EFFECT OF POSTMENOPAUSAL H O R M O N E THERAPYACCORDING TO BODY MASS INDEX Another important distinction between randomized controlled trials and observational studies that may have contributed to the discordance in coronary heart disease outcome is BMI (see Table 39.5). This observation is supported by the Cancer Prevention Study II (81). In this 12-year observational study of 290,823 women without a history of cancer and cardiovascular disease at enrollment, all-cause death rates were lower among users of hormone therapy than never users (RR, 0.82; 95% CI, 0.78-0.87). For specific causes of death, the lowest relative risk was found for coronary heart disease mortality (RR, 0.66; 95% CI, 0.58-0.77). Use of hormone therapy was also associated with a lower risk of stroke mortality (RR, 0.81; 95% CI, 0.67-1.00). Unlike coronary heart disease mortality, the association of hormone therapy with stroke mortality did not vary with BMI. For coronary heart disease mortality, the inverse association between hormone therapy and coronary heart disease mortality was strongest for thin women (BMI less than 22 kg/m 2) and was not apparent in obese women (BMI 30 kg/m 2 or greater) (p for interaction = 0.02). Among women with a BMI less than 22 kg/m 2, the risk of coronary heart disease mortality associated with hormone therapy was reduced 51% (RR, 0.49; 95% CI, 0.37-0.65), in women with a BMI between 22 kg/m 2 to less than 25 kg/m 2 the risk of coronary heart disease mortality associated with hormone therapy was reduced 28% (RR, 0.72; 95% CI, 0.57-0.91), and for women with a BMI between 25 kg/m 2 to less than 30 kg/m 2 the risk of coronary heart disease mortality associated with hormone therapy was reduced 23% (RR, 0.77; 95% CI, 0.59-1.01). These risk reductions contrast with the 45% increased risk of coronary heart disease mortality associated with hormone therapy in women with a BMI of 30 kg/m 2 or more (RR, 1.45; 95% CI, 1.00-2.11). Also of note was that risk of breast cancer mortality was nonsignificantly reduced with the use of hormone therapy (RR, 0.83; 95% CI, 0.69-1.01), even for women who used hormone therapy for 10 years or more. Body mass index did not modify the association between use of hormone therapy and breast cancer mortality (81). These data are consistent with another study in which women using estrogen alone
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy even for up to 25 years or longer had no increased risk of breast cancer; BMI did not modify the association between estrogen use and breast cancer (82). Assuming that results from the Cancer Prevention Study II extend to coronary heart disease incidence (including both fatal and nonfatal events), these results suggest another explanation for the differences in coronary heart disease outcome between randomized controlled trials and observational studies. Women in observational studies who used hormone therapy tended to be leaner than women who were randomized to controlled trials. For example, the women randomized to W H I - E P and W H I - E were predominantly overweight to obese (normal weight = 18.5-24.9 kg/m2; overweight = 25.0-29.9 kg/m2; obesity -> 30 kg/m2). The average BMI in W H I - E P and W H I - E was 28.5 kg/m 2 and 30.1 kg/m 2, respectively, and 34% of the W H I - E P cohort and 45% of the W H I - E cohort had a BMI greater than 30 kg/m 2 (20,22). In HERS, 56% of the women had a BMI greater than 27 kg/m 2 (15). Based on the average BMI of women randomized to controlled trials, data from the Cancer Prevention Study II would predict no overall significant treatment effect of hormone therapy on coronary heart disease outcome relative to placebo in these trials, in particular no treatment effect in women with a BMI greater than 25 kg/m 2 and perhaps an increased risk of coronary heart disease in those women with a BMI 30 kg/m 2 or greater (81). This contrasts to observational studies where women who used hormone therapy had reductions in coronary heart disease risk and were much leaner with an approximate average BMI of 25 kg/m 2 (3-12). It is interesting to note that in terms of weight, women randomized to controlled trials were similar to nonusers of hormone therapy in observational studies (3-12). The results from the W H I - E trial closely parallel those from the Cancer Prevention Study II (23). Women in the W H I - E trial randomized to CEE therapy with a BMI less than 25 kg/m 2, 25 kg/m 2 to less than 30 kg/m 2, and 30 kg/m 2 or greater had a 24% reduction, 13% reduction, and 11% increase in coronary heart disease events relative to placebo, respectively (see Fig. 39.12) (23). In the Cancer Prevention Study II, the risks for coronary heart disease mortality in users versus nonusers of hormone therapy were 28% reduced, 23% reduced, and 45% increased for the respective BMI categories (81). The results from the Cancer Prevention Study II and the W H I - E trial contrast with findings from EPAT, in which there was no significant difference in the estradiol treatment effect on the reduction of subclinical atherosclerosis progression rates between women with a BMI less than 30 kg/m 2 and those with a BMI 30 kg/m 2 or more (p = 0.52) (83). This may be the result of differing end points between these studies. However, it is of interest to note that in EPAT, the on-trial serum estradiol level and serum estradiol change from baseline due to exogenous estradiol treatment were significantly greater in those women with a BMI of 30 kg/m 2
551
or greater relative to women with a BMI less than 30 kg/m 2. The treatment group by BMI interaction was significant for both the on-trial serum estradiol level and serum estradiol change from baseline (p for interaction = 0.004). Whether this greater response in serum estradiol levels due to exogenous estrogen therapy in women with a BMI 30 kg/m 2 or greater relative to women with a BMI less than 30 kg/m 2 has clinical significance remains untested.
X. STROKE AND POSTMENOPAUSAL H O R M O N E THERAPY Although the Women's Estrogen for Stroke Trial (WEST) has been the only randomized controlled trial of hormone therapy designed with stroke as the primary trial outcome (84), HERS and W H I have also provided information concerning hormone therapy and stroke as an additional secondary trial outcome. W E S T was a randomized controlled trial designed to test whether oral 1713-estradiol I mg daily reduces the combined outcome of nonfatal stroke or all-cause mortality in 664 postmenopausal women who were on average 71 years old and approximately 20 years postmenopausal at baseline (84). The women had a documented non-disabling stroke or transient ischemic attack within 90 days of randomization into the trial. As such, W E S T was a trial that assessed the effect of 1713estradiol on the secondary prevention of stroke. Deaths or nonfatal stroke occurred in 99 participants in the estradiol group and in 93 participants in the placebo group (RR, 1.1; 95% CI, 0.8-1.4). There were 51 nonfatal strokes in the estradiol group and 52 nonfatal strokes in the placebo group (RR, 1.0; 95% CI, 0.7-1.4). There were 48 deaths in the estradiol group and 41 deaths in the placebo group (RR, 1.2; CI, 0.8-1.8). There were no increased rates of VTE or breast cancer in the estradiol compared with placebo group. The cumulated randomized controlled trial results concerning stroke and hormone therapy are mixed (Table 39.9), consistent with previous studies that show decreased risk (12,52,85-87), no effect (10,88-90), and increased risk, at least among smokers (91). W E S T showed that oral 17f~estradiol had a null effect on the secondary prevention of nonfatal and fatal stroke and both strokes combined (84). HERS showed that oral CEE + M P A had a null effect on the primary prevention of nonfatal and fatal stroke in postmenopausal women with established coronary heart disease (92). W H I - E P (20,93) and W H I - E (22,94) showed that nonfatal stroke was significantly increased with oral CEE + MPA and CEE alone when unadjusted statistical analysis was used, but was nonsignificant when the statistical analysis was adjusted for multiple trial outcome comparisons. Fatal stroke was not statistically different between the C E E + M P A and placebo treatment groups in W H I - E P (20,93) or between the CEE and placebo treatment groups in W H I - E (22,94) when analyzed using either the adjusted or unadjusted statistical
552
HODIS AND MACK
TABLE 39.9
Trial
Sample size
Postmenopausal Hormone Therapy and Stroke
Mean age (years)
Cohort
Duration, years
Combined stroke RR (95% CI)
Nonfatal stroke RR (95% CI)
Fatal stroke RR (95% CI)
1.23 (0.89-1.70) 1.10 (0.80-1.40) 1.31 (1.02-1.68) 1 (0.93-1.84) 2 1.39 (1.10-1.77) 1,3 (0.97-1.99) 2
1.18 (0.83-1.66) 1.00 (0.70-1.40) NR
1.61 (0.73-3.55) 2.90 (0.90-9.00) NR
1.39 (1.05-1.84) 1 (0.91-2.12) 2
1.13 (0.54-2.34) 1 (0.38-3.36) 2
HERS (92) WEST (84) WHI-EP (93)
2763 664 16,608
66.7 71 63.3
+CHD +CVA/TIA -CVD
4.1 2.8 5.6
WHI-E (22,94)
10,739
63.6
-CVD
6.8
1Unadjusted confidence interval, describes the variabilityof the point estimate that would arise from a simple trial for a single outcome. 2Adjusted confidence interval, describes the variabilityof the point estimate corrected for multiple outcome analyses. 3Fully adjudicated data (7.1 years follow-up) reported in reference 94 only for combined stroke and onlywith unadjusted confidence interval. HERS, Heart and Estrogen/progestin Replacement Study; WEST, Women's Estrogen for Stroke Trial; WHI-EP, Women's Health Initiative Estrogen+ Progestin Trial; WHI-E, Women's Health Initiative Estrogen Trial; +CHD, previous coronary heart disease; +CVA/TIA, previous cerebrovascular accident or transient ischemic accident; -CVD, no previous cardiovascular disease; RR (95% CI), relative risk (95% confidence interval) of hormone therapy compared with placebo; NR, fully adjudicated data not reported in reference 93, original data from reference 20: Combined Stroke: 1.4l (95% unadjusted CI, 1.07-1.85; 95% adjusted CI, 0.86-2.31). Nonfatal Stroke: 1.50 (95;%unadjusted CI, 1.08-2.08; 95% adjusted CI, 0.83-2.70). Fatal Stroke: 1.20 (95% unadjusted CI, 0.58-2.50; 95% adjusted CI, 0.32-4.49).
TABLE 39.10
Estrogen use Never 0.3 mg 0.625 mg ->1.5 mg
Estrogen Dosage and Stroke in the Nurses' Health Study
Person-years of s
No. of cases
313,661 19,964 116,150 39,026
290 9 124 46
Age-adjusted relative risk (95% confidence interval) 1.0 (reference) 0.43 (0.22-0.83) 1.11 (0.90-1.37) 1.58 (1.16-2.15)
Multivariate-adjusted relative risk (95% confidence interval) 1 0.54 (0.28-1.06) 1.35 (1.08-1.68) 1.63 (1.18-2.26)
1Adjusted for age, body mass index, history of diabetes, hypertension, hypercholesterolemia,age at menopause, cigarette smoking, and parental history of coronary heart disease.
analyses. Combined stroke followed the same pattern as nonfatal stroke in both W H I - E P (20,93) and W H I - E (22, 94). The fully adjudicated W H I data showed a lower C E E and C E E + M P A risk of combined stroke than that originally reported from W H I - E and W H I - E P (93,94). Although the cumulated data do not conclusively indicate that hormone therapy significantly increases stroke relative to placebo, the randomized controlled trials do indicate a consistent increase in the point estimate of risk. However, the time- and age-related stroke data from the W H I indicate that the risk of stroke from hormone therapy when initiated in close proximity to menopause or in young postmenopausal women is rare. In W H I - E P , the risk of stroke was 3 additional cases per 10,000 women per year of C E E + M P A therapy for women within 5 years of menopause and 6 additional cases per 10,000 women per year of C E E § therapy for women 5 years to less than 10 years since menopause (93). In W H I - E , the risk of stroke in women less than 60 years old was 1 additional case per 10,000 women per year of C E E therapy (94). In addition, the W H I - E P randomized
controlled trial data indicate that increasing duration of C E E + M P A therapy reduces the risk of stroke (12). The risk from stroke was reduced 26% in the women who used C E E + M P A for more than 5 years (see Table 39.7) (12). The reduction in stroke risk associated with hormone therapy was consistent between the W H I - E P trial and the W H I observational study of E + P (see Table 39.7) (12). Data from the Nurses' Health Study indicate that stroke risk associated with hormone therapy may be a function of estrogen dosage (Table 39.10) (47).
XI. VENOUS THROMBOEMBOLIC EVENTS AND POSTMENOPAUSAL HORMONE THERAPY Venous thromboembolism (VTE) is a disease process comprised of two different outcomes, deep vein thrombosis (DVT) and pulmonary embolism (PE). For DVT, there are
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy two distinct clinical entities classified according to location: calf and other. Calf DVTs are considered clinically benign and typically not treated because pathophysiologically they are rarely associated with PE. To date, the distinction between calf and other DVT locations associated with randomization in hormone therapy trials of cardiovascular disease have not been published. Although there was a twofold greater risk of VTE in the CEE + M P A group relative to the placebo group in WHI-EP, the increase in the absolute risk was small (Table 39.11) (95). The increased relative risk for VTE was greatest in the first year and significantly decreased but remained elevated over time: H R for VTE was 4.01 (CIs not reported) in year I of treatment, 1.97 in year 2, 1.74 in year 3, 1.70 in year 4, 2.90 in year 5, and 1.04 in year 6 and beyond (p for trend = 0.01) (95). The elevated absolute risk of VTE was lowest for women who were younger than 60 years old when randomized. In women younger than 60 years old, there were 11 additional cases of VTE per 10,000 women per year of CEE + M P A therapy versus 51 additional cases of VTE per 10,000 women per year of C E E + M P A therapy in women older than 60 years old (95). In HERS, C E E + M P A was associated with a 2.7-fold increased risk for VTE relative to placebo (96). Although DVT and PE were similarly increased in the C E E + M P A group, PE was not statistically different than placebo (see Table 39.11). The increase in the absolute risk of VTE from CEE + M P A was small. Similar to WHI-EP, over 6.8 years of follow-up, the H R for VTE declined from 2.66 during HERS (15,96) to 1.40 during the open label follow-up of 2.7 years (17) (test for time trend, p = 0.08). The results from W H I - E showed a nonsignificant RR of 1.32 with respect to the risk of VTE with CEE alone relative to placebo (97). The absolute risk for VTE was lowest for women who were less than 60 years old when randomized to W H I - E . For women younger than 60 years old, there were 4 additional cases of VTE per 10,000 women per year of CEE therapy versus 18 additional cases of VTE per 10,000 women per year of CEE therapy in women older than 60 years old (97). Analyzed separately, DVT with CEE alone was significantly increased relative to placebo, whereas the risk of PE was not (see Table 39.11). Comparison of the results from W H I - E P (HR, 2.1) and W H I - E (HR, 1.3) suggest that the VTE risk associated with CEE alone is less than the risk associated with C E E + M P A (62% increased risk associated with MPA relative to CEE alone). The results with estradiol alone from W E S T (84) are consistent with the W H I - E results with respect to the risk of VTE with CEE alone (97). In WEST, VTE and DVT were nonsignificantly lower in the estradiol group relative to placebo, whereas there was no effect on PE (see Table 39.11). The cumulated data suggest that the increased risk of VTE associated with C E E + M P A is not generalizable to CEE alone or to other forms of estrogen. In a case-control
553
TABLE 39.11 Relative and Absolute Risks ofVTE, DVT, and PE in Randomized Controlled Trials of Postmenopausal Hormone Therapy Absolute risk events per 10,000 women per year Trial and outcome WHI-EP (95) VTE DVT PE WHI-E (97) VTE DVT PE HERS (96) VTE DVT PE WEST (84)
Relative risk (95% CI)
Placebo
2.06 (1.57-2.70) 1.95 (1.43-2.67) 2.13 (1.45-3.11)
17 13 8
1.32 (0.99-1.75) 1.47 (1.06-2.06) 1.37 (0.90-2.07)
22 15 10
2.7 (1.4-5.0) 2.8 (1.3-6.0) 2.8 (0.9-8.7)
23 16 7
0.8 (0.2-3.4) 0.5 (0.0- 5.8) 1.0 (0.1-7.1)
44 22 22
Hormone therapy CEE + MPA 35 26 18 CEE alone 30 23 14 CEE +MPA 62 44 19 17[3-Estradiol alone
VTE DVT PE
32 11 21
VTE, venousthromboembolism;DVT,deep vein thrombosis;PE, pulmonaryembolus;WHI-EP, Women'sHealth Initiative Estrogen+ ProgestinTrial;WHI-E, Women'sHealth Initiative EstrogenTrial; HERS, Heart and Estrogen/progestinReplacement Study;WEST, Women's Estrogenfor StrokeTrial; CEE, conjugatedequine estrogen; MPA, medroxyprogesteroneacetate. study of 586 incident VTE cases and 2268 controls, compared with women not currently using hormone therapy, current users of esterified estrogen alone, esterified estrogen plus MPA, and CEE alone did not have an increased risk for VTE, whereas current users of CEE + MPA had a twofold statistically significant increased risk for VTE compared with nonusers of hormone therapy (Table 39.12) (98). In combination with MPA, esterified estrogen (OR, 1.08) and CEE (OR, 2.17) had increased relative risks for VTE; C E E + M P A was statistically significant (see Table 39.12). The increased risk for VTE associated with C E E + M P A (OR, 2.17) in this case-control study was similar to the relative risk reported from HERS (HR, 2.7) and W H I - E P (HR, 2.06). The relative risk for VTE associated with CEE alone (OR, 1.31) in this case-control study was similar to W H I - E (HR, 1.33). Esterified estrogen alone (OR, 0.78) was also similar to estradiol alone in W E S T (RR, 0.8). Most studies of VTE risk in hormone users have been conducted in women using orally administered exogenous hormones. Theoretically, non-oral routes of administration of hormone therapy should be associated with lower risk of VTE than
554 TABLE 39.12
HODIS AND MACK Risk of Venous Thrombembolism with Current Use of Different Hormone Therapies: Puget Sound Group Health Cooperative Case-Control Study
Participants, No. 1 Cases Controls Odds Ratio (95% CI)
None
EE alone
EE +MPA
CEE alone
372 1439
39 278
47 237
57 167
Reference
0.78 (0.53-1.15)
1.08 (0.75-1.56)
1.31 (0.91-1.88)
CEE +MPA 64 122 2.17 (1.49-3.14)
1Excludes 7 case and 25 control subjectswho were current users of estrogens other than EE and CEE. EE, esterified estrogens;CEE, conjugatedequine estrogen;MPA, medroxyprogesteroneacetate.
Overall mortality in HERS (HR, 1.08; 95% CI, 0.84-1.38), W H I - E P (HR, 0.98; 95% unadjusted CI, 0.82-1.18; 95% adjusted CI, 0.70-1.37) and W H I - E (HR, 1.04; 95% unadjusted CI, 0.88-1.22; 95% adjusted CI, 0.81 - 1.32) was unaffected by CEE + M P A and CEE therapies (Figures 39.21 to 39.23) (see Table 39.1). However, there was crossover in mortality risk between placebo and the C E E + M P A (see Fig. 39.22) and CEE (see Fig. 39.23) therapies after 5 to 6 years in W H I , suggesting a trend to decreasing mortality in the hormone group relative to placebo with longer duration of hormone therapy, as seen in previous observational studies (52). ESPRIT also showed a
nonsignificant reduction in cardiac and overall mortalities with estradiol alone (see Table 39.2) (19). To examine the effect of hormone therapy on overall mortality according to age, a meta-analysis of randomized controlled trials reporting at least 1 death in postmenopausal women comparing hormone therapy with placebo of at least 6 months duration was conducted (Table 39.13) (99). In this study, the effect of hormone therapy on overall mortality was null (OR, 0.98; 95% CI, 0.87-1.12). However, when the data were examined by the age of subjects, a statistically significant reduction in overall mortality was found for subjects less than 60 years old (mean age 54 years). The magnitude of the reduction in overall mortality of 39% (OR, 0.61; 95% CI, 0.39-0.95) for the women younger than 60 years old was similar to that of observational studies (3-12). The point estimate as well as the confidence interval for the reduction in overall mortality in the women less than 60 years old from this large meta-analysis are quite comparable to those of the largest observational study, the Nurses' Health Study (HR, 0.63; 95% CI, 0.56-0.70) (47). The age of initiation
FIGURE 39.21 All-causemortality (cumulative incidence) with conjugated equine estrogen0.625 mg plus medroxyprogesteroneacetate 2.5 mg daily (CEE+MPA) in the Heart and Estrogen/Progestin Replacement Study (HERS).
FIGURE 39.22 All-causemortality (cumulative incidence) with conjugated equine estrogen 0.625 mg plus medroxyprogesteroneacetate 2.5 mg daily (CEE+MPA) in the Women'sHealth Initiative (WHI) trial of estrogen plus progestin.
oral hormone therapy because of avoidance of the first-pass liver effects on procoagulation.
XII. MORTALITY AND POSTMENOPAUSAL H O R M O N E THERAPY
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy 0.05-
- -
....
,~
XIII. CLINICAL PERSPECTIVE OF POSTMENOPAUSAL H O R M O N E THERAPY
Placebo CEE
t
J
0.03-
-~'~'E . ~,. ~ , 0
=
0.02-
= 0
0.01-
0.0C,
0
,
1
,
2
i
3
--
~
4
,
5
,
6
~
7
,
8
Years
FIGURE 39.23 All-cause mortality (cumulative incidence) with conjugated equine estrogen 0.625 mg daily (CEE) in the Women's Health Initiative ( W H I ) trial of estrogen alone.
of hormone therapy of the women in the observational studies and the age of the younger women randomized to the clinical trials examined in the meta-analysis is similar. On the other hand, in this study, the effect of hormone therapy on overall mortality in women greater than 60 years old (OR, 1.03; 95% CI, 0.90-1.18) was similar to that reported by HERS (HR, 1.08; 95% CI, 0.84-1.38), W H I - E P (HR, 0.98; 95% unadjusted CI, 0.82-1.18; 95% adjusted CI, 0.70-1.37) and W H I - E (HR, 1.04; 95% unadjusted CI, 0.88-1.22; 95% adjusted CI, 0.81-1.32). Similar to the effect of hormone therapy on coronary heart disease events, both duration and time of initiation of hormone therapy appear to be important in reducing overall mortality. These results parallel observational studies (3-12,52). 39.13 Cohort Characteristics of Meta-Analysis of Randomized Controlled Trials of Hormone Therapy and Mortality in Younger and Older Women
TABLE
Number of trials Number of participants Patient-years Mean duration of follow-up (years) Range of follow-up (years) Mean study size (number of participants) Range of study size (number of participants) Mean age at baseline (hormone therapy), years _+ SD Mean age at baseline (placebo), years _+ SD Age range (years) Mean dropout rate (hormone therapy) Mean dropout rate (placebo) SD, standard deviation.
555
30 26,708 119,118 4.46 0.7-10 890 52-16,608 62.2 _+ 8.9 63.4 _+ 9.1 36-87 11.5% 10.6%
Because hormones exert systemic effects, weighing the benefits and side effects of hormone therapy is important. Although formal analyses to compare the benefits and side effects of hormone therapy have not been conducted, W H I - E P is interpreted by some as indicating that the benefits are outweighed by the side effects. This oversimplification of data from a single clinical trial that used one hormone therapy regimen and dosage when placed into perspective to the totality of data has been interpreted with an opposite conclusion~ that the benefits outweigh the side effects (14). Furthermore, conclusions comparing benefits and side effects have been based on the initial W H I - E P report that stated a significantly increased risk (in the unadjusted analysis) for coronary heart disease with CEE + MPA therapy (20) but later was shown not to be statistically significant when the fully adjudicated data were analyzed (12,21). Additionally, these conclusions have not considered the trends in reduction in coronary heart disease risk with longer duration of therapy or time of initiation of hormone therapy in relation to menopause (21,24), the age dependency of benefit of hormone therapy on cardiovascular disease (22,23), reduction in new-onset diabetes mellitus (26), or reduction in absolute risk of breast cancer, as shown in W H I - E (22,100). Any single randomized controlled trial does not provide complete information for determination of benefits and side effects for all specific hormones, dosages, or regimens. Based on the W H I - E P and W H I - E data as well as the cumulated studies, side effects of hormone therapy are clearly not a class effect. In addition, current randomized controlled trial data do not provide adequate information for the determination of the long-term benefits or side effects of hormone therapy. No randomized controlled trial has been conducted with sufficient follow-up to do so. Randomized controlled trials provide information for up to 6.8 years, the approximate point at which the beneficial effects of hormone therapy on coronary heart disease begin to manifest. Consequently, short-term use of hormone therapy may not provide for the full expression of the benefits of estrogen and may paradoxically give the appearance of greater side effects than benefits. The cumulated data indicate that continued exposure to exogenous estrogen is required for continued beneficial expression of estrogen on its target organs such as the cardiovascular and skeletal systems. Once hormone therapy is stopped, the beneficial effects on coronary heart disease and osteoporosis rapidly disappear (17,57,101,102). Exposure to hormone therapy reduces coronary heart disease, osteoporosis, new-onset diabetes mellitus, and overall mortality. Unfortunately, as only observational studies have
556
HODIS AND MACK
provided long-term information about hormone therapy, the long-term side effects essentially remain unknown and determination of long-term risk-benefits cannot be determined with currently available data. Considering the relatively short-term benefits and side effects from randomized controlled trials, the cumulated data indicate a balance between benefit and side effects (14). Table 39.14 provides the most recent information published from W H I - E P . Overall, the effect of oral daily continuous combined C E E + M P A on coronary heart disease risk was elevated but null (with a significant time trend to benefit) (21), increased the relative risk of breast cancer with borderline significance in one statistic (unadjusted) and nonsignificance in another (adjusted) (103), increased the relative risk of stroke with significance in one statistic (unadjusted) and nonsignificance in another (adjusted) (93), and increased the risk for V T E (95). Although of small magnitude, W H I EP is the only randomized controlled trial that has shown an increased relative risk of breast cancer with any hormone therapy. On the other hand, C E E + M P A resulted in significant reductions in colorectal cancer (104), osteoporotic bone fractures (105), and new-onset diabetes mellitus (26). Only reductions in colorectal cancer and bone fracture and increased V T E were statistically significant with both the unadjusted and adjusted analyses. In W H I - E , the overall effect of oral C E E alone on coronary heart disease was null
TABLE 39.14
(with a significant reduction in several composite coronary heart disease outcomes in women less than 60 years old) (22,23), as was the effect on breast cancer (100), colorectal cancer (22), and VTE (97) (Table 39.15). The relative risk of stroke was significantly increased in one statistic (unadjusted) and nonsignificant in another (adjusted) (22,94). On the other hand, C E E alone significantly reduced osteoporotic bone fractures (22). Only the reduction in bone fractures was statistically significant in both the unadjusted and adjusted analyses. Other randomized controlled trials add further to the mixed picture of benefits and side effects, with HERS showing no effect on stroke and breast cancer but increased VTE; W E S T showing no effect on stroke, breast cancer, or VTE; and E S P R I T showing nonsignificant risk reduction for overall and cardiac mortality. The mixed picture of benefits and side effects of hormone therapy may be driven by variation in the populations studied, sample size variation, varying follow-up, and differences in hormone therapies. In the final analysis, however, side effects with the hormone therapy regimens studied thus far are consistent with observational studies and gravitate toward the null. The one exception is VTE, where the risk with oral daily continuous combined C E E 0.625 mg plus MPA 2.5 mg appears to be elevated. It is important to understand absolute risks in relation to the acceptable standard for the practice of medicine and use
Women's Health Initiative Estrogen + Progestin Trial Outcomes ,
No. of subjects (annualized %)1 Outcome Coronary heart disease (21) Stroke (93) Venous thromboembolism (95) Breast cancer (invasive) (103) Colorectal cancer (104) Total fractures (105) New-onset diabetes (26)6
CEE + MPA 2 (N = 8506)
Placebo (N = 8102)
188 (0.39) 151 (0.31)
147 (0.33) 107 (0.24)
167 (0.35)
Adjusted
Increased absolute benefit per 10,000 women/year
95% CI 3
Unadjusted 95% CI 4
1.24 1.31
0.97-1.60 0.93-1.84
1.00-1.54 1.02-1.68
76 (0.17)
2.06
NR s
1.57-2.70
199 (0.42)
150 (0.33)
1.24
0.97-1.59
1.01-1.54
43 (0.09)
72 (0.16)
0.56
0.33-0.94
0.38-0.81
733 (1.52)
896 (1.99)
0.76
NR 5
0.69-0.83
47
277 (0.61)
324 (0.76)
0.79
NR
0.67-0.93
15
Hazard ratio
Increased absolute risk per 10,000 women/year
18
1Mean follow-up time = 5.6 years. 2Conjugated equine estrogen 0.625 mg + medroxyprogesteroneacetate 2.5 mg daily. 3Adjusted 95% CI, confidence interval describes the variabilityof the point estimate corrected for multiple outcome analyses. 4Unadjusted 95% CI, confidence interval describes the variabilityof the point estimate that would arise from a simple trial for a single outcome. SNot reported in updated report. Reported as significant in original manuscript from Women's Health Initiative Estrogen+ Progestin Trial (20). 6CEE+MPA = 8014 subjects; Placebo = 7627 subjects.
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy
TABLE 39.15
557
Women's Health Initiative Estrogen Trial Outcomes
No. of subjects (annualized %)1
Increased absolute benefit per 10,000 women/year
Increased absolute risk per 10,000 women/year
CEE 2 (N = 5310)
Placebo (N = 5429)
Hazard ratio
Adjusted 95% CI 3
Unadjusted 95% CI 4
Coronary heart disease (23)
201 (0.53)
217 (0.56)
0.95
0.76-1.19
0.79-1.16
Stroke (94)
168 (0.45)
127 (0.33)
1.37
NR 5
1.09-1.73
12
Venous thromboembolism (97)
111 (0.30)
86 (0.22)
1.32
NR 5
0.99-1.75
8
Breast cancer (invasive) (100)
104 (0.28)
133 (0.34)
0.806
NR s
0.62-1.04
61 (0.17)
58 (0.16)
1.08
0.63-1.86
0.75-1.55
503 (1.39)
724 (1.95)
0.70
0.59-0.83
0.63-0.79
Outcome
Colorectal cancer (22) Total fractures (22)
3
6 1 56
1Mean follow-up time = 6.8 years for colorectal cancer and total fractures; 7.1 years of follow-up time for other end points. 2Conjugated equine estrogen, 0.625 mg daily. 3Adjusted 95% CI, confidence interval describes the variabilityof the point estimate corrected for multiple outcome analyses. 4Unadjusted 95% CI, confidence interval describes the variabilityof the point estimate that would arise from a simple trial for a single outcome. 5Not reported in updated report. Reported as nonsignificant in original manuscipt from Women's Health Initiative Estrogen Trial (22). 6Total breast cancer incidence statisticallysignificantlyreduced in CEE versus placebo in adherent subjects (hazard ratio -- 0.67, 95% CI, 0.47-0.97). Statistically significant reduction of ductal (most common) breast cancer in subjects randomized to CEE versus placebo (hazard ratio = 0.71, 95% CI, 0.52-0.99) (100). of medications. As demonstrated by W H I , the side effects of C E E + M P A and C E E alone are of small magnitude (less than 1 additional case per 1,000 women) and, according to the World Health Organization Council for International Organization of Medical Sciences (CIOMS), are considered rare (Table 39.16) (106). The side effects are even rarer in women who are within 5 years of menopause or younger than 60 years old when initiating hormone therapy. The development of side effects with any long-term therapy should be kept in perspective, especially those used for the primary prevention of cardiovascular disease. As such, it is highly instructive to view the benefits and side
TABLE 39.16 World Health Organization Council for International Organizations of Medical Sciences frequency of Adverse Drug Reactions Category Very common Common (frequent) Uncommon (infrequent) Rare Very rare
Absolute frequency
Percentage frequency
-> 1/10 -> 1/100-<1/10
-> 10% -> 1%-<10%
>- 1/1000-<1/100 -> 1/10,000-<1/1000 <1/10,000
--- 0.1%-1% 0.01%-<0.1% <0.01%
effects of hormone therapy in perspective with other therapies used for coronary heart disease prevention. Prophylactic use of aspirin for primary prevention of cardiovascular disease is a common medical practice and recommended by major health organizations (107). However, the Women's Health Study of 39,876 healthy women randomized to aspirin 100 mg every other day or placebo for 10 years showed a null effect of aspirin on the primary trial end point of nonfatal myocardial infarction, nonfatal stroke, or cardiovascular death (RR, 0.91; 95% CI, 0.80-1.03) (108). Overall mortality and cardiovascular death from any cause were also unaffected by aspirin. Within the null finding was no significant effect on fatal or nonfatal myocardial infarction (RR, 1.02; 95% CI, 0.84-1.25) and a 17% (RR, 0.83; 95% CI, 0.69-0.99) reduction in stroke with aspirin versus placebo. Although ischemic stroke was reduced 24% (RR, 0.76, 95% CI, 0.63-0.93) with aspirin versus placebo, hemorrhagic stroke was increased 24% (RR, 1.24; 95% CI, 0.82-1.87) with aspirin versus placebo. In addition, bleeding diatheses were all significantly increased with aspirin versus placebo. Any gastrointestinal bleeding was increased 22% (RR, 1.22; 95% CI, 1.10-1.34) with aspirin versus placebo, and gastrointestinal bleeding requiring blood transfusion was increased 40% (RR, 1.40; 95% CI, 1.07-1.83) with aspirin versus placebo. The absolute increased risk for any gastrointestinal bleeding was 8 additional cases per 10,000 women per year of aspirin
558
use. The other randomized controlled trials of aspirin that included women were of smaller size and shorter duration than W H S but showed similar magnitudes of risk with no overall reduction in cardiovascular events: Hypertension Optimal Treatment (HOT) (n = 8883 women, 3.8 years) (109) and Primary Prevention Project (PPP) (n - 2583 women, 3.6 years) (108,110). Meta-analyses confirm the same or greater levels of risks for men, for whom the data are more substantial because the majority of study participants in the randomized controlled trials of aspirin have been men (111). Sudden death was nonsignificantly increased in men randomized to aspirin versus placebo. In the Physicians' Health Study (PHS) (n = 22,071), there were 22 sudden deaths in the men randomized to aspirin and 12 events in men randomized to placebo (OR, 1.96; 95% CI, 0.91-4.23) or an increased absolute risk of 5 sudden deaths per 10,000 men per year of aspirin use (112). Whether aspirin causes sudden death in women is unknown because these data have not been reported in WHS, HOT, or PPP. The magnitude of the side effects associated with aspirin resembles those from C E E + M P A and CEE from W H I (see Tables 39.14 and 39.15). Lipid-lowering medication, predominantly H M G - C o A reductase inhibitors (statins), are the mainstay for the primary prevention of cardiovascular disease in women, as they are for men (107,113). However, the cumulated data across six randomized trials and 11,435 women indicate that lipid lowering does not significantly reduce coronary heart disease events (RR, 0.87; 95% CI, 0.69-1.09), nonfatal myocardial infarction (RR, 0.61; 95% CI, 0.22-1.68), total mortality (RR, 0.95; 95% CI, 0.62-1.46), or coronary heart disease mortality (RR, 1.07; 95% CI, 0.47-2.40) when used for primary prevention of coronary heart disease in women (114). In contrast, CEE therapy under primary prevention conditions in the W H I - E trial significantly reduced several composite coronary heart disease outcomes by approximately 34% to 45% in 3310 postmenopausal women less than 60 years old (see Table 39.3) (23). In addition, a recent meta-analysis of 23 randomized controlled trials indicates a 32% (OR, 0.68; 95% CI, 0.48-0.96) reduction of coronary heart disease in women younger than 60 years old or less than 10 years since menopause when randomized to hormone therapy relative to placebo (115). On the other hand, the cumulated data across eight randomized controlled trials of lipid-lowering medication use in secondary prevention in 8272 women demonstrate significantly reduced coronary heart disease events (RR, 0.80; 95% CI, 0.71-0.91) and nonfatal myocardial infarction (RR, 0.73; 95% CI, 0.59-0.90) (114). In the final analysis, data supporting the use of lipid-lowering medication for the reduction of cardiovascular disease in women and upon which to base recommendations for primary prevention are small compared with that of men (114). Recommendations for using lipidlowering medications in the primary prevention of cardio-
HODIS AND MAcIr
vascular disease in women are predominantly extrapolated from data derived from men (107,113). On the other hand, certain risks of lipid-lowering medications resemble those of C E E + M P A reported from W H I EP (see Table 39.14). For example, in the Prospective Study of Pravastatin in the Elderly Risk (PROSPER), a 3.2-year randomized controlled trial of pravastatin versus placebo in men and women 70 to 82 years old, the diagnosis of new cancer was significantly increased by 25% (HR, 1.25; 95% CI, 1.04-1.51), or approximately 52 additional cancer cases per 10,000 persons per year of pravastatin use (116). The effect of pravastatin on cardiovascular events in 3000 women was null (HR, 0.96; 95% CI, 0.79-1.18). The risk for breast cancer was increased 65% (RR, 1.65; 95% CI, 0.78-3.49) in those women randomized to pravastatin (18 events) relative to placebo (11 events). This risk (annualized rate of approximately 0.38%) equates to a similar rate of breast cancer risk seen in W H I - E P (annualized rate of 0.42%), or 15 additional cases of breast cancer per/0,000 women per year of pravastatin use. The confidence interval for the risk of breast cancer with pravastatin in PROSPER was wider than that of W H I - E P because the sample size in PROSPER was 5.5 times smaller. The elevated risk of breast cancer associated with the use of pravastatin was reported in an additional randomized controlled trial of 576 women 59 _+ 9 years old treated for 5 years (117), in 997 postmenopausal women ages 55-73 years treated with lovastatin for 5.2 years (118) and in 827 women 60.5 + 6.4 years old followed for 10.4 years after randomized treatment to simvastatin for 5.4 years (119). In the Cholesterol and Recurrent Events (CARE) trial, there were 12 cases of breast cancer in women randomized to pravastatin and 1 woman randomized to placebo (p = 0.002), or approximately 77 additional cases of breast cancer per 10,000 women per year of pravastatin use (117). In the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS), there were 13 cases of breast cancer in women randomized to lovastatin and in 9 women randomized to placebo, or approximately 15 additional cases of breast cancer per 10,000 women per year oflovastatin use (118). In the 10-year follow-up of the Scandinavian Simvastatin Survival Study (4S), there were 7 cases of breast cancer in women randomized to simvastatin and in 5 women randomized to placebo, or 5 additional cases of breast cancer per 10,000 women per year of simvastatin use (114). As breast cancer events were not reported in the original 5.4-year intervention trial and the subsequent follow-up period was open-label, the incidence of breast cancer solely due to simvastatin exposure cannot be completely determined from the information provided from this study. However, the greater number of breast cancer events in the simvastatin-treated versus placebo-treated group in open-label follow-up suggests a possible residual breast cancer exposure risk to simvastatin (119). The 15 to 77 additional cases of breast cancer per 10,000 women per year of statin use exceeds the 9 additional cases of breast cancer per 10,000
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy women per year of CEE +MPA use in W H I - E P (see Table 39.14) (103). The increased absolute risk of breast cancer with statin therapy contrasts with CEE therapy. In WHI-E, the risk of breast cancer was decreased with CEE therapy relative to placebo by 18% (HR, 0.82; 95% CI, 0.65-1.04) or 6 fewer cases of breast cancer per 10,000 women per year of CEE therapy (100). In subgroup analyses, breast cancer risk was further reduced in women adherent to CEE therapy relative to placebo (HR, 0.67; 95% CI, 0.47-0.97) and in those women who developed ductal carcinoma (HR, 0.71; 95% CI, 0.52-0.99). The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) (120) and the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT) (121) showed no difference in breast cancer between pravastatin and placebo in the former study and two fewer cases of breast cancer per 10,000 women per year of pravastatin use in the latter trial. The Heart Protection Study (HPS) showed 10 fewer breast cancer cases per 10,000 women per year of simvastatin use (122). Unlike WHI-EP, in which colorectal cancer was significantly reduced with hormone therapy relative to placebo, gastrointestinal cancer was increased with pravastatin treatment relative to placebo in PROSPER (p -- 0.053). There was a 46% (RR, 1.46; 95% CI, 1.00-2.13) increased risk for gastrointestinal cancer in the pravastatin-treated group relative to the placebo group, or 22 additional cases of gastrointestinal cancer per 10,000 persons per year of pravastatin use (116). Although lipid-lowering trials other than PROSPER have not shown statistically significant increased risk for cancer overall, the overall absolute cancer risk in the primary prevention trials with statins that have reported cancer incidence is consistently elevated (118,121). The overall absolute cancer risk specifically determined in women in the primary prevention trial AFCAPS/TexCAPS was elevated, with approximately 15 additional cases of cancer per 10,000 women per year of lovastatin use (118). In the primary prevention trial ALLHAT-LLT, the overall absolute cancer risk was 5 additional cases of cancer and 7 additional cancer deaths per 10,000 persons per year of pravastatin use (121). Site-specific cancers, especially breast cancer incidence, have not been reported in all lipid-lowering randomized controlled trials, and thus the absolute risk remains unknown across all lipid-lowering medications (123,124). However, a recent meta-analysis reported data from five randomized controlled trials of statin therapy (PROSPER, CARE, 4S, LIPID, and ALLHAT-LLT) indicating a 33% (OR, 1.33; 95% CI, 0.79-2.26) increased risk of breast cancer associated with statin use. There were 81 cases of breast cancer among 16,875 women randomized to statin therapy versus 64 cases of breast cancer among 16,901 women randomized to placebo, or 10 additional breast cancer cases per 10,000 women who used statin therapy (123). A previous meta-analysis of 7 randomized controlled
559
trials of statin therapy showed a 4% (RR, 1.04; 95% CI, 0.81-1.33) increased risk of breast cancer associated with statin use (124). As commonly used drugs for the primary prevention of coronary heart disease, neither aspirin nor lipid-lowering therapy have been conclusively shown to reduce coronary heart disease events or mortality in women under primary prevention conditions. On the other hand, certain risks associated with these therapies are of similar magnitude to the risks reported from W H I and other randomized controlled trials of hormone therapy and cardiovascular disease. As in W H I , these risks on an absolute scale are small and considered rare (see Table 39.16) (106). Unlike hormone therapy, however, the risks associated with lipid-lowering therapy and aspirin in the primary prevention of coronary heart disease are considered acceptable. It should be appreciated that randomized controlled trials designed to study the effects of medications for the primary prevention of cardiovascular disease in women have not been as rigorously designed as W H I to monitor side effects and thus, most of the potential side effects carefully monitored in hormone therapy trials are essentially unknown for other therapies of cardiovascular disease prevention in women. However, data from trials such as PROSPER, other statin trials, and W H S provide reason for pause and careful analysis of all available data. In addition, most studies examining the effects of lipid-lowering medications for primary prevention of coronary heart disease under randomized controlled conditions have contained relatively few women compared with studies that have included up to 8363 men (125). For example, the largest randomized, double-blind, placebo-controlled trial of a statin in the primary prevention of coronary heart disease in women was ALLHAT-LLT, which contained 5074 women (121). These sample sizes are in general insufficient to detect side effects on the order of magnitude reported from WHI.
XIV. CONCLUSION The cumulated randomized controlled trial data indicate that after 4 to 6 years of use, hormone therapy reduces coronary heart disease risk relative to placebo in women with or without pre-existing coronary heart disease. In addition, these data strongly suggest that early initiation of hormone therapy relative to menopause is also associated with a reduction in coronary heart disease. The randomized controlled trial results are supported by approximately 40 observational studies also indicating that initiation of hormone therapy at the menopausal transition or early in the postmenopausal period and continued for a prolonged period of time results in significant reductions in coronary heart disease and overall mortality. Comparison of the results from randomized controlled trials, observational studies, and
560 case-control studies indicate that selection bias and confounding do not explain the consistent evidence that hormone therapy is associated with a duration- and timedependent lowering of coronary heart disease. The window of opportunity for maximal expression of the beneficial effects of hormone therapy on coronary heart disease appears to be initiation of hormone therapy within 6 years of menopause and/or by 60 years of age and continued for 10 years or more. As with the use of all medications, especially those used for primary prevention of disease, benefits must be carefully weighed against side effects. Randomized controlled trials of hormone therapy have confirmed that the side effects of hormones are rare and even rarer in women younger than 60 years old or within 5 years of menopause and of the same magnitude as other primary prevention therapies for coronary heart disease. Further analyses of the subgroups of women within randomized controlled trials that resemble the women from observational studies indicate a consistency between the two types of study designs, with similar benefit of hormone therapy on the reduction of coronary heart disease risk. Only two interventions, the reduction of elevated blood pressure and hormone therapy currently have data derived from studies conducted in sufficient numbers of women to support a potential role in the reduction of coronary heart disease (126). No other single preventive therapy offers systemicwide effects for women and as such, the use of hormone therapy in the primary prevention of disease must be considered differently than medications currently used in prevention. As exogenous estrogen therapy is a replacement therapy, its use can only be viewed as long term or permanent, much in the same way that hormones are used for endocrinopathies (e.g., hydrocortisone for Addison's disease and thyroxine for hypothyroidism). As an interesting and important parallelism with other forms of hormone replacement therapies, such as the examples with hydrocortisone and thyroxine, continued exposure to exogenous estrogen may be required for continued benefit. This has been well established for other tissues, such as bone, in which continued hormone therapy is required for the prevention of bone loss (57,101,102). Administration of exogenous estrogen after menopause should not be viewed as a therapy for any specific disease entity but as a replacement for a hormone that appears by the cumulated data to lessen the impact of aging on a multitude of organ systems such as the cardiovascular (127,128), skeletal (57,101,102), and potentially the central nervous (58,59) systems. Timing in the initiation of hormone replacement before tissue damage due to aging becomes too extensive appears to be the key for successful prevention and amelioration of any further damage. In the final analysis, the discordance in coronary heart disease outcome between randomized controlled trials and
HODIS AND MACK
observational studies is a function of the differences in study design and the characteristics of the populations studied and not the result of one study design being superior to another. In fact, as data from randomized controlled trials accumulate (99,115), it is important to realize how much the results look more and more like the more than 40 consistent observational studies that indicate that young postmenopausal women with menopausal symptoms who use hormone therapy for long periods of time have lower rates of coronary heart disease than comparable postmenopausal women who do not use hormone therapy. To date, no randomized controlled trial has properly studied the cardioprotective hypothesis of hormone therapy, as the cohorts studied in observational studies from which the hypothesis has been generated have not been selected for study under randomized controlled trial conditions. In addition, the duration of randomized controlled trials have not been of comparable duration to observational studies. However, even in cohorts of women not representative of observational studies, the side effects of hormone therapy are rare and the benefits are expressed. Further research to understand the effects of hormone therapy on coronary heart disease in women selected to replicate those women in observational studies is currently underway. One such study, the Early versus Late Intervention Trial with Estradiol (ELITE), is a randomized, double-blind, placebo-controlled, serial arterial imaging trial with a 2 • 2 factorial design. Funded by the National Institute on Aging, ELITE is specifically designed to study the hypothesis that the beneficial effects of estrogen therapy depend on timing of initiation of estrogen therapy relative to menopause. As such, healthy postmenopausal women without clinical evidence of cardiovascular disease and diabetes mellitus are randomized to oral 1713estradiol or placebo according to time since menopause, within 6 years of menopause and more than 10 years since menopause. The primary trial end point is progression of subclinical atherosclerosis measured as the rate of change of CCA IMT. Additional end points include cognition and other postmenopausal health issues. Interpretation of data in isolation, especially from one trial, can be misleading and result in conclusions that potentially could subvert an effective preventive therapy, especially those that require long-term administration. Putting study data into perspective can salvage important preventive therapies. Until results from trials such as ELITE become available, the consistent findings from randomized controlled trials and observational studies indicating a decreased risk of coronary heart disease with initiation of hormone therapy in close proximity to menopause and continued long term should be considered with other information concerning benefits and risks when women and their health care providers consider using hormone therapy.
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal Hormone Therapy
References 1. American Heart Association. Heart disease and stroke statistics--2005 update. Dallas, TX: American Heart Association, 2005. 2. Kannel WB, Hjortland MC, McNamara PM, Gordon T. Menopause and risk of cardiovascular disease: the Framingham study. Ann Intern Med 1976;85:447-452. 3. Grodstein F, Stampfer M. The epidemiology of coronary heart disease and estrogen replacement in postmenopausal women. Prog CardiolDis 1995;38:199-210. 4. Grodstein F, Stampfer M. Estrogen for women at varying risk of coronary disease. Maturitas 1998;30:19 - 26. 5. Nachtigall LE, Nachtigall RH, Nachtigall RD, Beckman M. Estrogen replacement therapy II: a prospective study in the relationship to carcinoma and cardiovascular and metabolic problems. Obstet Gynecol 1979;54:74-79. 6. Thompson SG, Meade TW, Greenberg G. The use of hormonal replacement therapy and the risk of stroke and myocardial infarction in women. J Epidemiol Community Health 1989;43:173-178. 7. Hunt K, Vessey M, McPherson K. Mortality in a cohort of long-term users of hormone replacement therapy: an updated analysis. BrJ Obstet Gynaeco11990;97:1080-1086. 8. Falkeborn M, Persson I, Adami HO, et al. The risk of acute myocardial infarction after oestrogen and oestrogen-progestogen replacement. BrJ Obstet Gynaeco11992;99:821- 828. 9. Psaty BM, Heckbert SR, Atkins D, et al. The risk of myocardial infarction associated with the combined use of estrogens and progestins in postmenopausal women. Arch Intern Med 1994;154:1333-1339. 10. Grodstein F, Stampfer MJ, Manson JE, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N EnglJ Med 1996;335:453-461. 11. Grodstein F, Stampfer MJ, Falkeborn M, Naessen T, Persson I. Postmenopausal hormone therapy and risk of cardiovascular disease and hip fracture in a cohort of Swedish women. Epidemiology 1999;5:476-480. 12. Prentice RL, Langer R, Stefanick ML, et al., for the Women's Health Initiative Investigators. Combined postmenopausal hormone therapy and cardiovascular disease: toward resolving the discrepancy between observational studies and the Women's Health Initiative clinical trial. Am J Epidemio12005 ;162:404- 414. 13. Mendelsohn ME, Karas RH. The time has come to stop letting the HERS tale wag the dogma. Circulation 2001;104;2256-2259. 14. Phillips LS, Langar RD. Postmenopausal hormone therapy: critical reappraisal and a unified hypothesis. Fertil Steri12005;83:558-566. 15. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAm MedAssoc 1998;280:605-613. 16. Herrington DM, VittinghoffE, Lin F, et al., for the HERS study group. Statin therapy, cardiovascular events, and total mortality in the Heart and Estrogen/Progestin Replacement Study (HERS). Circulation 2002; 105:2962-2967. 17. Grady D, Herrington D, Bittner V, et al., for the HERS Research Group. Cardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/Progestin Replacement Study follow-up (HERS II).JAm MedAssoc 2002;288:49-57. 18. Clarke SC, Kelleher J, Lloyd-Jones H, Slack M, Schofield PM. A study of hormone replacement therapy in postmenopausal women with ischemic heart disease: the Papworth HRT Atherosclerosis Study. BJOG 2002;109:1056-1062. 19. Cherry N, Gilmour K, Hannaford P, Heagerty A, Khan MA, Kitchener H, McNamee R, Elstein M, Kay C, Seif M, Bucldey H. Oestrogen therapy for the prevention of reinfarction in postmenopausal women: a randomized placebo controlled trial. Lancet 2002;360:2001-2008. 20. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy menopausal women:
21. 22.
23.
24.
25.
26.
27.
28. 29.
30.
31.
32.
33.
34.
35. 36.
37.
38.
39.
561
principal results from the Women's Health Initiative randomized controlled trial. JAm MedAssoc 2002;288:321- 323. Manson JE, Hsia J, Johnson KC, et al. Estrogen plus progestin and the risk of coronary heart disease. NEnglJ Med 2003;349:523-534. The Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAm Med Assoc 2004;291:1701-1712. Hsia J, Langer RD, Manson JE, et al., for the Women's Health Initiative Investigators. Conjugated equine estrogens and coronary heart disease: the Women's Health Initiative. Arch Intern Med 2006;166:357-365. Grodstein F, Manson JE, Stampfer MJ. Hormone therapy and coronary heart disease: the role of time since menopause and age at hormone initiation. J Women's Health 2006;15:35-44. Barnabei VM, Cochrane BB, Aragaki AK, et al., for the Women's Health Initiative investigators. Menopausal symptoms and treatmentrelated effects of estrogen and progestin in the Women's Health Initiative. Obstet Gyneco12005;105:1063-1073. Margolis KL, Bonds DE, Rodabough RJ, et al., for the Women's Health Initiative Investigators. Effect of oestrogen plus progestin on the incidence of diabetes in postmenopausal women: results from the Women's Health Initiative hormone trial. Diabetologia 2004;47:1175-1187. Kanaya AM, Herrington D, Vettinghoff E, et al. Glycemic effects of postmenopausal hormone therapy: the Heart and Estrogen/progestin Replacement Study. Ann Intern Med 2003;138:1-9. Kannel WB. Lipid, diabetes, and coronary heart disease: insights from the Framingham study. Am HeartJ 1985;110:1100-1107. Narayan KMV, Boyle JP, Thompson TJ, Sorensen SW, Williamson DE Lifetime risk of diabetes mellitus in the United States.JAm Med Assoc 2003;290:1884-1890. Buchanan TA, Xiang AH, Peters RK, et al. The TRIPOD Study: preservation of pancreatic B-cell function and prevention of type 2 diabetes by pharmacological treatment of insulin resistance in high risk Hispanic women. Diabetes 2002;51:2796-2803. Xiang AH, Peters RK, Kjos SL, et al. Effect of thiazolidinedione treatment on progression of subclinical atherosclerosis in premenopausal women at high risk for type 2 diabetes. J Clin Endocrinol Metabo12005. 90:1986-1991. Blankenhorn DH, Hodis HN. Duff Memorial Lecture: arterial imaging and atherosclerosis reversal. Arterioscl Thromb 1994;14: 177-192. Azen SP, Mack W, Cashin-Hemphill L, et al. Progression of coronary artery disease predicts clinical coronary events: long-term follow-up from the Cholesterol Lowering Atherosclerosis Study. Circulation 1996;93:34-41. Hodis HN, Mack WJ, LaBree L, et al. The role of carotid arterial intima-media thickness in predicting clinical coronary events. Ann Intern Med 1998;262-269. Hodis HN, Mack WJ. Carotid artery intima-media thickness and risk of cardiovascular events. Curt Pract Med 1999;2:171-174. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary artery atherosclerosis. NEngld Med 2000;343:522- 529. Waters DD, Alderman EL, Hsia J, et al. Effects of hormone replacement therapy and antioxidant vitamin supplements on coronary atherosclerosis in postmenopausal women. J Am Med Assoc 2002;288: 2432-2440. Hodis HN, Mack WJ, Azen SP, et al., for the WELL-HART Research Group. Hormone therapy and the progression of coronary artery atherosclerosis in postmenopausal women. N Engl J Med 2003;349: 535-545. Hodis HN, Mack WJ, Lobo RA, et al., for the Estrogen in the Prevention of Atherosclerosis Trial Research Group. Estrogen in the prevention of atherosclerosis: a randomized, double-blind, placebo-controlled trial. Ann Intern ivied 2001;135:939-953.
562 40. Selzer RH, Hodis HN, Kwong-Fu H, et al. Evaluation of computerized edge tracking for quantifying intima-media thickness of the common carotid artery from B-mode ultrasound images. Atherosclerosis 1994;111:1-11. 41. Selzer RH, Mack WJ, Lee PL, Kwong-Fu H, Hodis HN. Improved common carotid elasticity and intima-media thickness measurement from computer analysis of sequential ultrasound frames. Atherosclerosis 2001;154:185-193. 42. Espeland MA, Applegate WB, Furberg CD, et al., for the ACAPS Investigators. Estrogen replacement therapy and progression of intimalmedial thickness in the carotid arteries of postmenopausal women. Am J Epidemio11995;142:1011-1019. 43. de Kleijn Mjj, Bots ML, Bak AAA, et al. Hormone replacement therapy in perimenopausal women and 2-year change of carotid intimamedia thickness. Maturitas 1999;32:195-204. 44. Angerer P, Stork S, Kothny W, Schmitt P, yon Schacky C. Effect of oral postmenopausal hormone replacement on progression of atherosclerosis: a randomized controlled trial. Arterioscler Thromb Vasc Biol 2001;21:262-268. 45. Angerer P, Kothny W, Stork S, yon Schacky C. Hormone replacement therapy and distensibility of carotid arteries in postmenopausal women: a randomized controlled trial.JAm Coll Cardio12000;36:1789-1796. 46. Oliver JJ, Webb DJ. Noninvasive assessment of arterial stiffness and risk of atherosclerotic events. Arteriosder Thromb VascBio12003;23:554- 566. 47. Grodstein F, Manson JE, Colditz GA, et al. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med 2000;133:933-941. 48. Grodstein F, Manson JE, Stampfer MJ. Postmenopausal hormone use and secondary prevention of coronary events in the Nurses' Health Study. Ann Intern Med 2001;135:1- 8. 49. Chilvers CED, Knibb RC, Armstrong SJ, Woods KL, Logan RFA. Postmenopausal hormone replacement therapy and risk of acute myocardial infarction: a case control study of women in the East Midlands, UK. Eur HeartJ 2003;24:2197-2205. 50. Heckbert SR, Weiss NS, Koepsell TD, et al. Duration of estrogen replacement therapy in relation to the risk of incident myocardial infarction in postmenopausal women. Arch Intern Med 1997;157:1330-1336. 51. Ferrara A, Quesenberry CP, Karter AJ, et al. Current use of unopposed estrogen and estrogen plus progestin and the risk of acute myocardial infarction among women with diabetes: the Northern California Kaiser Permanente diabetes registry, 1995-1998. Circulation 2003;107:43-48. 52. Henderson BD, Paganini-Hill A, Ross RK. Decreased mortality in users of estrogen replacement therapy. Arch Intern Med 1991;151:75-78. 53. Akhrass F, Evans AT, Wang Y, et al. Hormone replacement therapy is associated with less coronary atherosclerosis in postmenopausal women. J Clin Endocrinol Metab 2003;88:5611-5614. 54. Barrett-Connor E, Laughlin GA. Hormone therapy and coronary artery calcification in asymptomatic postmenopausal women: the Rancho Bernardo Study. Menopause 2005;12:40-48. 55. Hodis HN, Mack WJ, Lobo R. What is the cardioprotective role of hormone replacement therapy? Current Atherosclerosis Reports 2003;5: 56-66. 56. Barr RG, Wentowski CC, Grodstein F, et al. Prospective study of postmenopausal hormone use and newly diagnosed asthma and chronic obstructive pulmonary disease. Arch Intern Med 2004;164: 379-386. 57. Cauley JA, Seeley DG, Ensrud K, et al., for the Study of Osteoporotic Fractures Research Group. Estrogen replacement therapy and fractures in older women.Ann Intern Med 1995;122:9-16. 58. Viscoli CM, Brass LM, Kernan WN, et al. Estrogen therapy and risk of cognitive decline: results from the Women's Estrogen for Stroke Trial (WEST). Am J Obstetr Gyneco12005;192:387- 393. 59. Bagger YZ, Tanko LB, Alexandersen P, Qin G, Christiansen C, for the PERF Study Group. Early postmenopausal hormone therapy may prevent cognitive impairment later in life. Menopause 2005;12:12-17.
HODIS AND MACK 60. Adams MR, Register TC, Golden DL, Wagner JD, Williams JK. Medroxyprogesterone acetate antagonizes inhibitory effects of conjugated equine estrogens on coronary artery atherosclerosis. Arterioscler Thromb VascBio11997;17:217- 221. 61. Clarkson TB, Anthony MS, Jerome CE Lack of effect of raloxifene on coronary artery atherosclerosis of postmenopausal monkeys. J Clin Endocrinol Metab 1988:83;721- 726. 62. Clarkson TB, Anthony MS, Morgan TM. Inhibition of postmenopausal atherosclerosis progression: a comparison of the effects of conjugated equine estrogens and soy phytoestrogens. J Clin Endocrinol Metab 2001:86;41-47. 63. Williams JK, Anthony MS, Honore EK, et al. Regression of atherosclerosis in female monkeys. Arterioscler Thromb Vasc Biol 1995;15: 827-836. 64. RosenfeldME, Kauser K, Martin-McNulty B, et al. Estrogen inhibits the initiation of fatty streaks throughout the vasculature but does not inhibit intra-plaque hemorrhage and the progression of established lesions in apolipoprotein E deficient mice.Atherosclerosis2002;164:251-259. 65. Post WS, Goldschmidt-Clemont PJ, Wilhide CC, et al. Methylation of the estrogen receptor gene is associated with aging and atherosclerosis in the cardiovascular system. CardiovascRes 1999;43:985-991. 66. Losordo DW, Kearney M, Kim EA, Jekanowski J, Isner JM. Variable expression of the estrogen receptor in normal and atherosclerotic coronary arteries of premenopausal women. Circulation 1994;89:1501-1510. 67. Herrington DM, Espeland MA, Crouse JR, et al. Estrogen replacement and brachial artery flow-mediated vasodilation in older women. Arteriosder Thromb VascBio12001;21:1955-1961. 68. Campisi R, Nathan L, Pampaloni MH, et al. Noninvasive assessment of coronary microcirculatory function in postmenopausal women and effects of short-term and long-term estrogen administration. Circulation 2002;105:425-430. 69. Reis SE, Gloth ST, Blumenthal RS, et al. Ethinyl estradiol acutely attenuates abnormal coronary vasomotor responses to acetylcholine in postmenopausal women. Circulation 1994;89:52-60. 70. Holm P, Andersen HL, Andersen MR, Erhardtsen E, Stender S. The direct antiatherogenic effect of estrogen is present, absent, or reversed, depending on state of the arterial endothelium; a time course study in cholesterol-clamped rabbits. Circulation 1999;100:1727-1733. 71. Geary RL, Adams MR, Benjamin ME, Williams JK. Conjugated equine estrogens inhibit progression of atherosclerosis but have no effect on intimal hyperplasia or arterial remodeling induced by balloon catheter injury in monkeys.JAm Coll Cardio11998;31:1158-1164. 72. Strong JP, Malcom GT, McMahan CA, et al. Prevalence and extent of atherosclerosis in adolescents and young adults: implications for prevention from the Pathobiological Determinants of Atherosclerosis in Youth Study.JAm MedAssoc 1999;281:727-735. 73. McGill HC Jr, Stern ME Sex and atherosclerosis. Atheroscler Rev 1979;4:157-242. 74. Tejada C, StongJP, Montenegro MR, Restrepo C, Solberg LA. Distribution of coronary and aortic atherosclerosis by geographic location, race and sex. Lab Invest 1968;18:509-526. 75. Dollar AL, Kragel AH, Fernicola DJ, Waclawiw MA, Roberts WC. Composition of atherosclerotic plaques in coronary arteries in women <40 years of age with fatal coronary artery disease and implications for plaque reversibility. Am J Cardio11991;67:1223-1227. 76. Zanger D, Yang BK, Ardans J, et al. Divergent effects of hormone therapy on serum markers of inflammation in postmenopausal women with coronary artery disease on appropriate medical management.JAm Coll Cardio12000;36:1797-1802. 77. HwangJ, Hodis HN, Hsiai TK, Asatryan L, Sevanian A. Role of annexin II in estrogen-induced macrophage matrix metalloproteinase activity: the modulating effect of statins. Atherosclerosis2006;189:76- 82. 78. Lobo RA. Evaluation of cardiovascular event rates with hormone therapy in healthy early postmenopausal women: results from 2 large clinical trials. Arch Intern Med 2004;164:482-484.
CHAPTER 39 Randomized Controlled Trials and the Effects of Postmenopausal H o r m o n e Therapy 79. Mack WJ, Slater CC, Xiang M, et al. Elevated subclinical atherosclerosis associated with oopherectomy is related to time since menopause rather than type of menopause. Fertil Steri! 2004;82:391-397. 80. Sutton-Tyrrell K, Lassila HC, Meilahn E, et al. Carotid atherosclerosis in premenopausal and postmenopausal women and its association with risk factors measured after menopause. Stroke 1998;29:1116-1121. 81. Rodriguez C, Calle EE, Patel AV, et al. Effect of body mass on the association between estrogen replacement therapy and mortality among elderly US women. Am J Epidemio12001;153:145-152. 82. Li CI, Malone KE, Porter PL, et al. Relationship between long durations and different regimens of hormone therapy and risk of breast cancer. JAm MedAssoc 2003;289:3254-3263. 83. Mack WJ, Hameed AB, Xiang M, et al. Does elevated body mass modify the influence of postmenopausal estrogen replacement on atherosclerosis progression: results from the estrogen in the prevention of atherosclerosis trial. Atherosclerosis 2003;168:91-98. 84. Viscoli CM, Brass LM, Kernan WN, et al. A clinical trial of estrogenreplacement therapy after ischemic stroke. N Engl J Med 2001;345:1243-1249. 85. Finucane FF, Madans JH, Bush TL, Wolf PF, Kleinman JC. Decreased risk of stroke among postmenopausal hormone users: results from a national cohort. Arch Intern Med 1993;153:73-79. 86. Falkeborn M, Persson I, Terent A, et al. Hormone replacement therapy and the risk of stroke: follow-up of a population-based cohort in Sweden. Arch Intern Med 1993;153:1201-1209. 87. Schairer C, Adami HO, Hoover R, Persson I. Cause-specific mortality in women receiving hormone replacement therapy. Epidemiology 1997;8: 59-65. 88. Lindenstrom E, Boysen G, Nyboe J. Lifestyle factors and risk of cerebrovascular disease in women: the Copenhagen City Heart Study. Stroke 1993;24:1468-1472. 89. Pedersen AT, Lidegaard O, Kreiner S, Ottesen B. Hormone replacement therapy and risk of non-fatal stroke. Lancet 1997;350: 1277-1283. 90. Petitti DB, Sidney S, Quesenberry CP, Bernstein A. Ischemic stroke and use of estrogen and estrogen/progestogen as hormone replacement therapy. Stroke 1998;29:23 - 28. 91. Wilson PWF, Garrison RJ, CasteUi WP. Postmenopausal estrogen use, cigarette smoking, and cardiovascular morbidity in women over 50: the Framingham study. N EnglJ Med 1985;313:1038-1043. 92. Simon JA, Hsia J, Cauley JA, et al., for the HERS Research Group. Postmenopausal hormone therapy and risk of stroke: the Heart and Estrogen-progestin replacement study (HERS). Circulation 2001;103: 638-642. 93. Wassertheil-Smoller S, Hendrix SL, Limacher M, et al., for the WHI Investigators. Effect of estrogen plus progestin on stroke in postmenopausal women: the Women's Health Initiative: a randomized trial.JAm MedAssoc 2003;289:2673-2684. 94. Hendrix SL, Wassertheil-Smoller S, Johnson KC, et al., for the WHI Investigators. Effects of conjugated equine estrogen on stroke in the Women's Health Initiative. Circulation 2006;113:2425-2434. 95. Cushman M, Kuller LH, Prentice R, et al., for the Women's Health Initiative investigators. Estrogen plus progestin and risk of venous thrombosis. JAm MedAssoc 2004;292:1573-1580. 96. Grady D, Wenger NK, Herrington D, etal., for the Heart and Estrogen/ progestin Replacement Study Research Group. Postmenopausal hormone therapy increases risk for venous thromboembolic disease: the Heart and Estrogen/progestin Replacement Study. Ann Intern Med 2000;132:689-696. 97. Curb JD, Prentice RL, Bray PF, etal. Venous thrombosis and conjugated equine estrogen in women without a uterus. Arch Intern Med 2006; 166:772-780. 98. Smith NL, Heckbert SR, Lemaitre RN, et al. Esterified estrogens and conjugated equine estrogens and the risk of venous thrombosis. J A m MedAssoc 2004;292:1581-1587.
563
99. Salpeter SR, Walsh JME, Greyber E, Ormiston TM, Salpeter EE. Mortality associated with hormone replacement therapy in younger and older women: a meta-analysis. J Gen Intern Med 2004;19:791-804. 100. Stefanick ML, Anderson GL, Margolis KL, et al., for the WHI Investigators. Effects of conjugated equine estrogens on breast cancer and mammography screening in postmenopausal women with hysterectomy. JAMA 2006;295:1647-1657. 101. Lindsay R, Hart DM, MacLean A, etal. Bone response to termination of oestrogen treatment. Lancet 1978;1:1325-1327. 102. Christiansen C, Christensen MS, Transbol I. Bone mass in postmenopausal women after withdrawal of oestrogen/gestagen replacement therapy. Lancet 1981;1:459-461. 103. Chlebowski RT, Hendrix SL, Langer RD, et al., for the WHI Investigators. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women: the Women's Health Initiative randomized trial. J Am Med Assoc 2003;289: 3243-3253. 104. Chlebowski RT, Wactawski-Wende J, Ritenbaugh C, etal., for the Women's Health Initiative investigators. Estrogen plus progestin and colorectal cancer in postmenopausal women. N Eng! J Med 2004; 350:991-1004. 105. Cauley JA, Robbins J, Chen Z, et al., for the Women's Health Initiative investigators. Effects of estrogen plus progestin on risk of fracture and bone mineral density: the Women's Health Initiative randomized trial. JAm MedAssoc 2003;290:1729-1738. 106. Utian WH, Archer DF, Bachmann GA, et al. Recommendations for estrogen and progestogen use in peri- and postmenopausal women: March 2007 position statement of the North American Menopause Society. Menopause 2007;14:1 - 17. 107. Mosca L, Appel LJ, Benjamin EJ, et al. Evidence-based guidelines for cardiovascular disease prevention in women. Circulation 2004;109: 672-693. 108. Ridker PM, Cook NR, Lee IM, etal. A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women. N EnglJ Med 2005 ;352:1293 - 1304. 109. Kjeldsen SE, Kolloch RE, Leonetti G, etal., for the HOT study group. Influence of gender and age on preventing cardiovascular disease by antihypertensive treatment and acetylsalicylic acid: the HOT study. J Hypertens 2000;18:629-642. 110. Collaborative Group of the Primary Prevention Project (PPP). Lowdose aspirin and vitamin E in people at cardiovascular risk: a randomized trial in general practice. Lancet 2001;357:89-95. 111. Hayden M, Pignone M, Phillips C, Mulrow C. Aspirin for the primary prevention of cardiovascular events: a summary of the evidence for the U.S. preventive services task force.Ann Intern Med 2002;136:161-172. 112. Steering Committee of the Physicians' Health Study Research Group. Final report on the aspirin component of the ongoing Physicians' Health Study. N EnglJ Med 1984;321:129-135. 113. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAm MedAssoc 2001;285:2486-2497. 114. Walsh JME, Pignone M. Drug treatment of hyperlipidemia in women. JAm MedAssoc 2004;291:2243-2252. 115. Salpeter SR, Walsh JME, Greyber E, Salpeter EE. Coronary heart disease events associated with hormone therapy in younger and older women: a meta-analysis. J Gen Intern Med 2006;21:363-366. 116. Shepherd J, Blauw GJ, Murphy MB, et al., on behalf of the PROSPER study group. Pravastatin in elderly individuals at risk of vascular disease (PROSPER): a randomized controlled trial. Lancet 2002;360: 1623-1630. 117. Sacks FM, Pfeffer MA, Moye LA, et al., for the Cholesterol and Recurrent Events Trial Investigators. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N EnglJ Med 1996;335:1001 - 1009.
564 118. Clearfield M, Downs JR, Weis S, et al. Air Force/Texas coronary atherosclerosis prevention study (AFCAPS/TexCAPS): efficacy and tolerability of long-term treatment with lovastatin in women. J Women's Health GendBasedMed 2001;10:971-981. 119. Strandberg TE, Pyorala K, Cook TJ, et al., for the 4S Group. Mortality and incidence of cancer during 10-year follow-up of the Scandinavian Simvastatin Survival Study (4S). Lancet 2004;364:771-777. 120. The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. NEnglJMed 1998;339:1349-1357. 121. ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. Major outcomes in moderately hypercholesterolemic, hypertensive patients randomized to pravastatin vs usual care: the antihypertensive and lipid-lowering treatment to prevent heart attack trial (ALLHAT-LLT). JAm MedAssoc 2002;288:2998-3007. 122. Heart Protection Study Collaboration Group. Effects of cholesterollowering with simvastatin on stroke and other major vascular events in 20,536 people with cerebrovascular disease or other high-risk conditions. Lancet 2004;363:757-767. 123. Dale KM, Coleman CI, Henyan NN, Kluger J, White CM. Statins and cancer risk: a meta-analysis. JAm MedAssoc 2006;295:74-80.
HoDis AND MACK 124. Bonovas S, Filiovssi K, Tsavaris N, Sitaras NM. Use of statins and breast cancer: a meta-analysis of seven randomized clinical trials and nine observational studies. J Clin Onco12005;23:8606-8612. 125. Sever PS, Dahlof B, Poulter NR, et al., for the ASCOT investigators. Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the Anglo-Scandinavian Cardiac Outcomes TrialmLipid Lowering Arm (ASCOT-LLA): a multicentre randomized controlled trial. Lancet 2003;361:1149-1158. 126. Blood Pressure Lowering Treatment Trialists' Collaboration. Effects of different blood-pressure-lowering regimens on major cardiovascular events: results of prospectively designed overviews of randomized controlled trials. Lancet 2003;362:1527-1535. 127. Takahashi K, Tanaka E, Murakami M, et al. Long-term hormone therapy delays the age related progression of carotid intima-media thickness in healthy postmenopausal women. Maturitas 2004;49:170-177. 128. McGrath BP, Liang YL, Teede H, et al. Age-related deterioration in arterial structure and function in postmenopausal women: impact of hormone replacement therapy. Arterioscler Tbromb VascBio11998;18: 1149-1156.
SECTION VIII
Neoplasia After menopause, aging explains the increased incidence of all cancers. Breast cancer is the major site of cancer mortality at young ages but by age 60 lung cancer mortality exceeds that of breast cancer (1) (Fig.l), while deaths from cardiovascular disease are by far the major cause of death at older ages. In surveys of postmenopausal women, it has been found consistently that the major fear of women contemplating hormonal therapy (HT) is the risk of cancer, primarily breast cancer (2) (Fig. 2). This Section focuses on the primary cancers of concern to postmenopausal women, and specifically the effect of HT. Included in this Section are discussions of surveillance for cancer (breast, uterus, and ovary). This important emphasis in this Section is key in my view for maintaining good health in postmenopausal women, leading to early detection and cure if a cancer develops. Not specifically discussed in the setting of surveillance is evaluation of the colon-rectum. Here, too, the detection and treatment of early lesions proves to be curative. A discussion of breast cancer is covered both by Malcolm Pike and colleagues and by Carlo La Vecchia. In the first chapter, Pike and colleagues stress the variable of body mass index in modifying the influence of hormone therapy on breast cancer risk. Also discussed are differences in the relative risks of breast cancer in the U.S. and European studies. This is followed by a chapter on breast surveillance by William H. Hindle. Next, endometrial cancer is discussed by John P. Curtin. Ovarian cancer is the most deadly of the gynecologic cancers because of difficulties in early detection and not having well-defined precursor lesions. Surveillance strategies, therefore, become the most relevant discussion point in that there is no clear evidence that estrogen therapy or hormone therapy increases the risk of ovarian cancer. The last chapter again reviews data of the risk of breast, colorectal, and lung cancer. An extremely important aspect of cancer surveillance in women is the use of ultrasound. Here surveillance of the uterus/endometrium is more valuable than ovarian ultrasound because ovarian cancer grows very rapidly and is difficult to diagnose at an early stage. However endometrial cancer gives many early clues and often has a precursor lesion: endometrial hyperplasia, reflected in endometrial thickness. The use of vaginal ultrasound was a separate chapter in the previous edition, and this chapter by Anna Parsons is still relevant and current in that the basic techniques and attributes of this technique were aptly described (3). After menopause the finding of an endometrial stripe of 4 mm or less is extremely reassuring, in the absence of postmenopausal bleeding. With postmenopausal bleeding, vaginal ultrasound is imperative to assess the endometrium. Here apart from cancer detection (suggested by thickening, irregularity and sometimes abnormal vascularity with color flow doppler) benign polyps and fibroids can be detected and are very prevalent. For this purpose saline sonohysterography (SHG) has become a mainstay to management, as described in detail previously (3). SHGs are carried out routinely now in the office setting and can guide treatment without any delay. Occasionally the need may arise for 3D-ultrasound to better isolate the location of various types of fibroids and structural abnormalities of the uterus; and magnetic resonance imaging (MRI) may be useful in the case of adenomyosis, as well as to assess cancer invasion. On this latter point as pointed out earlier,
566
SECTION VIII Neoplasia
FIGURE 1
Percentageof deaths from breast and lung cancer versus cardiovasculardisease.
color flow doppler studies, carried out simultaneously by vaginal ultrasound may also be helpful for cancer detection. Finally vaginal ultrasound of the uterus is helpful as a guide for hormonal therapy and is being used increasingly for this purpose. Since many practitioners feel that progestogen therapy should be minimized and should be administered only to protect the uterus for endometrial hyperplasia and developing endometrial cancer, endometrial thickness on ultrasound can be used as a guide to when progestogen therapy should be used. This approach is helpful primarily when assessed prospectively and longitudinally, and with the use of lower doses of estrogen, where proliferation is less likely to occur. That is, after a baseline
Q: What do you personal/y think is the effect of liT on the risks of suffering from the fo//owing complaints?
FIGURE 2
U.S.women'sopinions on effects of hormone therapy on selected risks.
567
SECTION VIII Neoplasia
assessment, the degree of proliferation (endometrial thickness) as assessed by ultrasound after several months of low dose estrogen can serve as a guide to when progestogen should be administered. This use of vaginal ultrasound has been advocated by me as well as others (4).
References 1. Phillips KA, Glendon G, Knight JA. Putting the risk of breast cancer in perspective. New EnglJMed 1999;340:141-144. 2. Strothmann A, Schneider HE Hormone therapy: the U.S. women's perspective. MenopauseMgt (in press). 3. Parsons AK, Londono JL. Detection and surveillance of endometrial hyperplasia and carcinoma, Chapter 47 in Treatment of the Postmenopausal Woman, second edition. Pages 513-538. 4. Goldstein SR. Attempt to reduce the progestogen component in hormonal therapy. Menopause2006;13:538.
This Page Intentionally Left Blank
;HAPTER
4(
Body Weight, Menopausal Hormone Therapy, and Risk of Breast Cancer ANNA H. WIJ
Department of Preventive Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
C . LEIGH PEARCE Department of Preventive Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
DARCY V. S P I C E R
Department of Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
SULGGI LEE
Department of Preventive Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
MALCOLM C . PIKE
Department of Preventive Medicine, Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
Much epidemiologic, endocrinologic, clinical, pathologic, and experimental data exist to provide overwhelming evidence that ovarian hormones play a crucial role at all stages of the development of breast cancer. Two observations hold the key to our understanding of the relationship of ovarian hormones to breast cancer risk as it applies to the treatment of postmenopausal women:
1. Earlier menopause, whether natural or through bilateral oophorectomy, reduces breast cancer risk. 2. The rate of breast (epithelial) cell proliferation in the luteal phase of the menstrual cycle is about double the rate in the follicular phase, and the rate in the follicular phase is considerably higher than in the postmenopausal period.
In this chapter we will use these two observations, together with the observations that later age at menarche, earlier age at first birth, and increasing parity decrease breast cancer risk, to build a basic mathematical model of breast cancer etiology.
This work was supported by grant CA14089 from the National Cancer Institute and by grant BC044808 from the Department of Defense Congressionally Directed Breast Cancer Research Program. TREATMENT OF THE POSTMENOPAUSAL WOMAN
569
All
Copyright 9 2007 by Elsevier,Inc. rights of reproductionin any form reserved.
570
Wu ET AL.
We will show how this model allows us to gain a deep understanding of the quantitative effect of obesity on risk. We will then summarize the epidemiologic data, including that found in the Women's Health Initiative (WHI) randomized trials, on the quantitative effects of menopausal estrogen therapy (ET) and menopausal estrogen-progestin therapy (EPT) on breast cancer risk; by integrating this with the model we obtain a comprehensive understanding of the effect of estrogen and progestin treatment on breast cancer risk.
I. KEY RIS K FACTORS
A. Age at Menopause Trichopoulos and colleagues (1) found in a large casecontrol study that women whose natural menopause occurred before age 45 had only one-half the breast cancer risk of women whose menopause occurred after age 55. These investigators also found that bilateral oophorectomy reduced breast cancer risk. The effect of oophorectomy was somewhat greater than that of natural menopause, and women with an oophorectomy before age 35 had only one-quarter the breast cancer risk of women with a natural menopause at age 50. Similar results were found in many other epidemiologic studies (2,3). The protective effect of early oophorectomy proves that the association is one of cause and effect. This indicates that the hormonal pattern ofpremenopausal women (cyclic production of relatively large amounts of estradiol [E2] and progesterone [P4]) causes a greater rate of increase in risk of breast cancer than the hormonal pattern of postmenopausal women (constant low E2 and very low P4). Factors other than E2 and P4 may be involved, but these factors alone provide a comprehensive explanation of the observed effect.
B. Age at Menarche just as earlier menopause decreases breast cancer risk, later menarche decreases breast cancer, and for a given age at menarche, the later regular menstruation is established, the lower the risk (2,4).
C. Parity Having a full-term pregnancy (a "birth") before about age 32 has a long-term protective effect against breast cancer, and the younger the mother when birth occurs, the greater the protection. Early epidemiologic studies suggested that this protective effect of births was confined to, or at least much greater with, a first birth (5), but it is now clear that the long-term protection from any (and all) births
under about age 32 provides about the same level of added protection as the first (6,7). Although a birth prior to the age of about 32 affords long-term protection, births at older ages increase risk, with the risk increasing with increasing age at the birth, as was seen in the early studies of MacMahon and colleagues (5). Even births at young ages are followed by short-term increases in incidence that are evident for the first 10 years after the birth. Pike and colleagues (8) showed how this short-term increase could be seen as the same phenomenon as the observation that a late age at first birth is associated with a long-term increase in risk (see later discussion).
II. MODELING BREAST CANCER RISK A. Age and Breast Cancer Incidence To understand the manner in which ages at menopause and menarche affect breast cancer risk, it is essential to understand the relationship of age and breast cancer incidence (i.e., the risk of breast cancer diagnosis during a 1-year period at different ages). The incidence of the common non-hormonedependent adult cancers (stomach, colon, etc.) rises continuously and increasingly rapidly with age. On a log-log scale the age-incidence curve of such a cancer is linear (Fig. 40.1) (9,10). The log-log plot is also most useful in providing an understanding of how various factors affect cancer incidence, as the log-incidence plot shows the rate of change in incidence with age. The incidence, fit), at age t of such a cancer rises as the kth power of age (10) and can be written as: I(t) = a • ? so that
log[fit)] = log(a) + k • log(t) Fig. 40.2 shows the age-specific incidence curve for breast cancer in U.S. Caucasian females in 1969-1971, before mammographic screening and widespread use of menopausal hormone therapy had distorted the picture (9). Breast cancer incidence also continues to increase with age, but there is a distinct slowing of the rate of increase around age 50, that is, around menopause. The important etiologic elements for breast cancer thus appear to be present in premenopausal women and to be sharply reduced following the menopause, as the epidemiologic studies of the effect of age at menopause showed.
B. Modeling Risk The age-specific incidence curve for breast cancer can be made to fit into the common linear log-log form if age, t, is replaced by breast-tissue age, b(t), so that the linear
CHAPTER 40 Body Weight, Menopausal Hormone Therapy, and Risk of Breast Cancer
571
menopause, age at first birth, and the age-incidence curve in Fig. 40.2 is shown in Fig. 40.3 (8). This definition assumes that breast-tissue aging starts at menarche and continues at a constant rate up to first birth, then at a reduced rate until the start of the perimenopausal period, declining gradually thereafter to a low postmenopausal rate. To account for a birth after around age 32 being associated with a breast cancer rate that is larger than that of nuUiparous women, the model includes a transient increase in breast-tissue aging during the pregnancy. Taking r(t) from menarche to the first flail-term pregnancy (FFTP) as 1.0, the rate increases to ---4.0 during the year of the F F T P and then falls to "--0.7 until the perimenopause, declining gradually thereafter to ---0.11 after menopause. The fitted exponent of time (k) is 4.5, similar to the best-fitting value for many epithelial cancers. The fit of the model is shown by the smooth curve in Fig. 40.2. This model provides an excellent description of the observed effects of ages at menarche, menopause, and first birth. The short-term increase in risk after a birth could be equated to the observation that a late age at first birth is associated with a long-term increase in risk; effectively the benefit of first birth in reducing the subsequent breast-tissue aging rate does not have long enough (before the postmenopausal drop in rate occurs) to compensate for the increased rate during the pregnancy. This model has been extended by Rosner and
FIGURE 40.1 Log-logplot of age-specific incidence rates for colorectal cancer in United States Caucasian women, 1969-1971 (9).
relationship is between the log of incidence, I(t), and the log of breast-tissue age:
log[I(t)] = log(a) + k •
log[a(t)]
and
I(t) = a[b(t)] k At a given age t, the rate of breast-tissue aging, r(t), is the rate at which the underlying carcinogenic process is taking place, and r(t) is summed from 0 to t to give b(t) (8). We assume that r(t) is roughly equal to mitotic rate, or perhaps cell death and resulting replacement in some stem cell compartment (11). Breast-tissue age represents at least approximately the cumulative number of stem cell divisions up to age t. The hypothesis is simply that hormones affect cancer incidence mainly through their effect on mitotic rates in the stem-cell compartment, both by increasing the probability of a DNAdamaging event becoming fixed as a mutation and through their promoting effect as seen in animal experiments. A definition of the rate of breast-tissue aging with age that accounts in quantitative terms for the effects of menarche,
FIGURE 40.2 Log-log plot of age-specific incidence rates for breast cancer in United States Caucasian women, 1969-1971 (9).
572
W u ET AL.
This information was forthcoming in the early 1980s in a series of studies showing that mitotic activity of breast epithelium varies markedly during the normal menstrual cycle, with peak activity occurring in the luteal phase (Fig. 40.4) (13,14). This shows that regular ovulatory cycling is positively correlated with mitotic activity and explains why a more rapid establishment of regular cycles after menarche will be associated with the greatest increase in breast cancer risk. The mitotic activity during the luteal phase is roughly double the level in the follicular phase. Based on the fit shown in Fig. 40.2, we calculate that the mitotic rate in the postmenopausal period is approximately one-fifth the mitotic rate during the follicular phase of the cycle: data on this are sparse, but this is completely in line with what has been found (15). FIGURE 40.3 Modelof rate of breast-tissue aging.
colleagues (3,12) to incorporate the effects of second and later births.
C. Breast-Cell Mitotic Activity Understanding the relationship of breast-cell mitotic activity to the phase of the menstrual cycle is essential to relating this model of breast cancer risk to the biology of the breast.
III. WEIGHT EFFECTS IN MODELTERMS Body weight has profound effects on breast cancer risk. By combining these weight effects with the insights gained from the model shown in Fig. 40.3 and the relationship of E2 and P4 to breast cell proliferation shown in Fig. 40.4, we are in a position to gain further quantitative insights into the relation of estrogen and progesterone to breast cancer risk. The relationship of weight to breast cancer risk is critically dependent on age. For older postmenopausal women, breast cancer risk increases with weight, whereas in premenopausal
FIGURE40.4 PlasmaE2 and P4, and breast-tissue mitotic rate by day of menstrual cycle.(From ref. 14, with permission.)
CHAPTER 40 Body Weight, Menopausal Hormone Therapy, and Risk of Breast Cancer
women, increasing weight is associated with a decrease in breast cancer risk (2). Premenopausal obesity is associated with increased anovulation with decreased serum hormone levels of both estrogen and progesterone (16). In contrast, postmenopausal obesity is associated with increased serum estrogen levels (17,18). When these differences in hormone levels are taken into consideration with the known effects on mitotic rate of both estrogen and progesterone (see Fig. 40.4), the observed changes in effect of weight on risk with age are precisely what are predicted. Table 40.1 shows, using the results of the pooled analysis of cohort studies (19), the relative risks for a woman with a body mass index (BMI) of 30 kg/m 2 compared with a woman with a BMI of 20 kg/m 2 at the end of her premenopausal period at age - 5 0 years and at a postmenopausal age of---62 years, some 12 years after her natural menopause at "--50 years. The relative risk of 0.75 at age 50 becomes a relative risk of 1.2 at age 62. The relative risks can be converted into the associated breast-tissue age with the above model, and these values are shown in the lower half of the table. The relative risk of 0.75 at age 50 implies a breasttissue age of---28.6 years for the 30-kg/m 2 woman compared with the breast-tissue age of "-30.5 years for the 20-kg/m 2 woman (assuming age at menarche = 12, age at first birth = 24, and parity = 1 for both women). If we estimate the breast-tissue aging rate in the - 2 0 - k g / m 2 woman to be ---0.1 or slightly less, that is, somewhat less than the value found in fitting the model shown in Fig. 40.3, then the relative risk of 1.2 at age 62 implies a breast-tissue age of "--33.0 years for the 30-kg/m 2 woman compared with a breast-tissue age of---31.7 years for the 20-kg/m 2 woman. Thus, the increase in breast-tissue age per year, r(t), in the postmenopausal period in the 30-kg/m 2 woman is the difference between her breast-tissue ages at age 62 and age 50, namely, (33.0 - 28.6) = 4.4 years, divided by 12 (the number of years between 50 and 62), so that, r(t) = 4.4/12 - 0.37.
TABLE 40.1 Relative Risks of Breast Cancer by BMI in Premenopausal and Postmenopausal Women, and Calculated Associated Breast-Tissue Ages to Derive These Relative Risks BMI
Premenopausal (age 50)
Relative risks: "-20 kg/m2 "--30 kg/m 2 Breast-tissue age: " 2 0 kg/m 2
"30 kg/m 2 *Age at menopause 50 years. BMI, body mass index.
Postmenopausal (age 62)*
1 0.75
1 1.20
30.5 yrs 28.6 yrs
31.7 yrs 33.0 yrs
573
Assuming that the r(t) values of 0.1 and 0.37 for the 20-kg/m 2 and 30-kg/m 2 woman, respectively, are highly correlated with the "effective estrogen" level and the estrogen effect to be simply proportional to level, the relative levels of effective estrogen will be 0.37/0.1 = 3.7. With no threshold, this 3.7 is simply the ratio of effective estrogen levels, but with a threshold value for effective estrogen, the ratio of effective estrogen will be lower and may be considerably lower. If the effective estrogen is equated to serum non-sex-hormone-binding globulin-bound E2 (nonSHBG-bound E2), that is, - 8 . 4 pg/mL and - 4 . 0 pg/mL for a woman with a BMI of 30 kg/m 2 compared to a woman with a BMI of 20 kg/m 2 (calculated with the mass action approach of S6dergard and colleagues [20] with the estimated association constants as given in Dunn et al. [21] that were used by the Endogenous Hormones Group [18] and using the E2 and SHBG values given by the Endogenous Hormones Group [17]), we find that the ratio of effective estrogen is (8.4/4.0) = 2.1. To convert this to a 3.7 ratio, one has only to assume a threshold value of n o n - S H B G - b o u n d E2 of 2.37 pg/mL, calculated so that (8.4- 2.37)/(4.0 - 2.37) = 3.7. Fig. 40.4 shows that approximately two-thirds of the breast cell proliferation in a premenopausal woman can be accounted for by an estrogen effect alone. The breast-tissue aging rate, r(t), of the parous premenopausal woman is approximately 0.7, so that we estimate the effect due solely to estrogen is 0.7 • 2/3 = 0.47, or the effect of the estrogen level of the 30 kg/m 2 woman is (0.37/0.47) = 79% that of the effect of the follicular phase estrogen. If we continue to equate effective estrogen with n o n - S H B G - b o u n d E2, the n o n - S H B G - b o u n d E2 level of the postmenopausal woman with a BMI of 30 kg/m 2 is - 8 . 4 pg/mL, so that the effective ceiling for n o n - S H B G - b o u n d E2 effect is unlikely to be greater than [(8.4 - 2.37)/0.79 + 2.37] = 10.0 pg/mL. We will show next how this profoundly important conclusion, namely that estrogen levels above a n o n - S H B G bound E2 value of 10.0 pg/mL are unlikely to increase risk further than is already achieved with a n o n - S H B G - b o u n d E2 value of 10.0 pg/mL, provides an explanation of the reduced effects of menopausal hormone therapy (HT) seen with increasing weight.
IV. MENOPAUSAL H O R M O N E THERAPY EFFECTS IN MODEL TERMS Menopausal ET at a conjugated equine estrogen (CEE) dose of 0.625 mg/day, the most commonly used dose in the United States until very recently, is roughly equivalent to a 50-~lg E2 transdermal patch. A 50-~lg E2 patch increases steady state E2 levels by - 3 0 pg/mL (22-24) and has little effect on SHBG (25,26), so that n o n - S H B G - b o u n d E2 will increase from - 4 . 0 pg/mL to - 1 8 . 4 pg/mL in a
574 20-kg/m 2 woman and from --~8.4 pg/mL to ---27.4 pg/mL in a 30-kg/m 2 woman (calculated as described above). The steady-state, n o n - S H B G - b o u n d E2 level of all women on a 50-lag E2 patch is thus well above the ceiling level. A woman on E T is thus predicted (in model terms) to have a breast-tissue aging rate of--~0.47 (two-thirds of 0.7) independent of her BMI, assuming she takes the E T every day (Fig. 40.5). The predicted consequences of this for E T use of 5 years' duration are shown in Table 40.2 and Fig. 40.6 for a woman aged 55 years whose age at menopause was 50 years. We note that the predicted effect of E T use in a woman of 20 kg/m 2 is to increase her breast cancer risk by 30% after 5 years of use. In sharp contrast, the predicted effect of E T use in a woman of 30 kg/m 2 is to increase her breast cancer risk by only 3% after 5 years of use. Following this same mode of argument, one can use the model to roughly predict what the effect of E P T will be. The area under the curve (AUC) of P4 during a normal menstrual cycle is approximately 97 ng/mL (see Fig. 40.4). The usual oral dose of progestin with CEE was 10 mg/day of medroxyprogesterone acetate (MPA) for 10-14 days per 28-day cycle. An oral M P A dose of 10 mg/day is estimated to have the same progestin effect as 500 rag/day of oral P4, and this dose of P4 is estimated to produce a serum P4 level equivalent to a steady state level of 12.2 ng/mL (27), leading to an AUC of between 122.0 ng/mL and 170.8 ng/mL for 10 to 14 days of use per 28-day cycle. On a straightforward linear model of P4 action, one would therefore predict a 1.26- to 1.50-fold increased progestin effect for 10 to 12 days MPA use compared with that seen during a menstrual cycle. The contribution to the breast-cell aging rate from the progesterone during a menstrual cycle is roughly one-third of the total effect, that is, 0.7/3 = 0.23, so that the MPA at 10 mg/day for 10 days is
Wo ET AL.
FIGURE 40.6 Relative risks of breast cancer at age 55 years (age at menopause 50 years) after 5 years of hormone therapy (HT) use by body mass index. EPT, estrogen-progestin therapy; ET, estrogen therapy; HT, ET or EPT.
estimated to add (0.23 • 1.26) = 0.29 to the rate associated with the estrogen component of EPT, namely, 0.47, giving a total breast-ceU aging rate of 0.76. The equivalent figure for use of MPA at this dose for 12 days per cycle is (0.23 • 1.50) + 0.47 = 0.82. The predicted breast-tissue aging rate of use for 12 days per 28-day cycle are shown in Fig. 40.5; the rates are again independentof BMI. The predicted effects of such a 12-day E P T regimen of 5 years' duration on breast cancer incidence are shown in the far-right column of Table 40.2 and in Fig. 40.6 for a woman aged 55 years whose age at menopause was 50 years. We note that the predicted effect of such E P T use in a woman of 20 kg/m 2 is to increase her breast cancer risk by 64% after 5 years of use. The predicted effect of
TABLE 40.2 Relative Risks of Breast Cancer at Age 55 Years (Age at Menopause 50 Years) after 5 Years of E T or E P T Use by BMI BMI 20 kg/m2 30 kg/m2
FIGURE40.5 Breast-tissueagingrate in postmenopausalwomen by body mass index and use of hormone therapy (HT). EPT, estrogen-progesfin therapy; ET, estrogen therapy; HT, ET or EPT.
Breast tissue age RR Breast tissue age RR RR
No. HT
ET
EPT
31.00 a 1.00 30.45 b 0.92 1.00
32.83 1.30 30.93 0.99 1.07
34.58 1.64 32.68 1.27 1.38
aThis figure is calculated as the sum of the 30.5 breast-tissue age shown in Table 40.1 at age 50 plus 5 years of the breast-tissue aging rate of 0.1 in the postmenopausalperiod. bThis figure is calculated as the sum of the 28.6 breast-tissue age shown in Table 40.1 at age 50 plus 5 years of the breast-tissue aging rate of 0.37 in the postmenopausalperiod. BMI, body mass index; HT, hormone therapy; ET, estrogen therapy; EPT, estrogen-progestin therapy; RR, relative risk.
CHAPTER 40 Body Weight, Menopausal Hormone Therapy, and Risk of Breast Cancer such E P T use in a woman of 30 kg/m 2 is to increase her breast cancer risk by 31%, roughly half of the relative risk seen in the 20-kg/m 2 woman but still a significant added risk. (Note: The relationship of BMI to breast cancer risk shown in Table 40.2 and Fig. 40.6 is very slight in contrast to the effect of BMI shown in Table 40.1. This is because the effect of BMI on risk becomes greater and greater with increasing time since menopause, and the results shown in Table 40.2 and Fig. 40.6 refer to --~5 years after menopause, whereas the results shown in Table 40.1 refer to --~12 years after menopause. Immediately after menopause there is still a negative relationship between weight and risk as a carry-over from the reduced risk with increased weight seen in premenopausal women. At older ages, the effect of increasing BMI will be greater than that noted in Table 40.1.) These predicted effects are robust but not precise. A significant number of assumptions have had to be m a d e - for example, that the relative risks associated with obesity refer to women aged 50 and 62, respectively--and we have had to use hormone level figures from different studies, whereas one actually needs the measurements to have been made at the same time using the same methods. However, the overall picture is quite clear, and we shall see in the next section that these predicted effects are precisely what are seen in epidemiologic studies. This gives one a considerable level of confidence in the general validity of the approach adopted here.
V. EPIDEMIOLOGIC STUDIES OF ET A N D EPT
A. Estrogen Therapy Epidemiologic studies show that menopausal ET causes an approximately 2% increase in breast cancer risk per year of use (28); that is, a 10% increase in risk at the 5-year point after 5 years of use or a cumulative increase in risk for 5 years of use of approximately half this amount. The increased risk is much more evident in slender women, and little if any effect is seen in obese women (28-31). The model, as further developed in the earlier discussion of weight, predicts precisely these results, and the few epidemiologic study results published are in good agreement with the predictions shown in Table 40.2 and Fig. 40.6. The previous discussion of ET is based on an overview of all epidemiologic studies of the effect of E T on breast cancer risk. However, the W H I randomized trial of ET in hysterectomized women found a decrease in risk (32). The lack of effect of E T in the W H I trial is difficult to reconcile with at least three other observations: (1) increased postmenopausal serum estrogen levels are associated with increased risk; risk increases twofold to threefold
575
between the extreme quintiles of E2, estrone (El), and E1 sulphate (E1S); (2) increasing postmenopausal weight increases breast cancer risk and is largely explained by the association of increased weight with increased serum estrogen levels (17,18); and (3) the risk of contralateral breast cancer is sharply reduced by treating the original cancer with an aromatase inhibitor that drastically reduces estrogen levels. The evidence considered as a whole shows that ET increases breast cancer risk and that the effect is greatest in slender women and difficult to see in women with a B MI over 30 kg/m 2.
B. Estrogen-Progestin Therapy The publication of the overview of H T studies by the Collaborative Group on Hormonal Factors in Breast Cancer (28) contained very little information on the effects of EPT. A large number of studies have been conducted on breast cancer risk from E P T since that time. We have carried out an overview of all these studies (33). Ten recent studies (29,30,34-41) provided information on the risk from E P T use in a quantitative assessment that adjusted for age at menopause, a critical factor in assessing H T use and breast cancer risk, and provided information on risk by duration of E P T use. The results from these studies and from the Collaborative Group (28) pooled analysis were used to obtain an overall assessment of E P T and breast cancer risk (33). Special measures were taken to permit the results from the randomized trial (40) and the cohort studies (29,30,37,41) to be expressed in the same relative risk terms as in the case-control studies. First, the W H I randomized trial (40) found an average relative risk of breast cancer of 1.24 for E P T use after an average of 5.6 years of follow-up. This 1.24 figure converts to a relative risk at 5 years after 5 years of use, RRs, of 1.47. Second, for cohort studies, the true duration of E P T use is underestimated in current hormone users. This is because E P T use is assessed at baseline but continues for an unknown proportion of individuals for at least some further period until last follow-up time, so that an additional duration of use should be added for current hormone users in the cohort studies considered. We made the conservative assumption that current users of E P T remained users during follow-up; this approach slightly underestimates the effect of E P T use because some women will have stopped use before the last follow-up date in the studies. The overall summary of the studies (Table 40.3) showed a weighted average relative risk at the end of 5 years of use, RR5, of 1.44 (95% confidence interval [CI], 1.40-1.48). The RR5 for the United States studies was 1.29 (95% CI, 1.19-1.39); for the European studies, 1.46 (95% CI, 1.42-1.50); and for the Scandinavian studies, 1.53 (95% CI. 1.37-1.72). The difference between the United States
576
Wu E T
TABLE 40.3 Relative Risks of Breast Cancer at 5 Years after 5 Years of Use (RRss) of Estrogen-Progestin Therapy Studies
RR5 (95% CI)
All studies (28 - 30,34- 41) United States studies (29,35-37,39,40) European studies (30,34,41) Scandinavian studies (34,41)
1.44 (1.40-1.48) 1.29 (1.19-1.39) 1.46 (1.42-1.50) 1.53 (1.37-1.72)
CI, confidenceinterval.
studies and these other studies was highly statistically significant. Five studies had adequate data by progestin schedule of EPT use (30,34,35,39,41). Overall, sequential EPT use was associated with a lower risk than continuous-combined EPT use (Table 40.4). The most obvious difference between the continuous-combined and sequential schedules was seen in the two Scandinavian studies (34,41). This difference was also supported by the results of the Scandinavian study reported by Jernstrom and colleagues (42), which we excluded from our summary relative risk calculations because the authors only presented results for continuous-combined EPT use; the authors reported their results in this manner because of the greater observed effect for continuous-combined than for sequential EPT use. In the United States, the most common form of sequential EPT provides 5 - 1 0 mg of MPA per day for 10 days per 28-day cycle, whereas subjects assigned to receive continuouscombined EPT are typically given 2.5 mg of MPA every day. The total doses for sequential and continuous-combined are therefore very close at approximately 75 mg and 70 mg, respectively, per cycle. In contrast, in Scandinavia the total dose of the progestin is much higher with continuouscombined than with sequential EPT, at least for two commonly prescribed regimens using norethisterone acetate
TABLE 40.4 Relative Risks of Breast Cancer at 5 Years after 5 Years of Use (RRss) of Estrogen-Progestin Therapy by Progestin Schedule Studies All studies (30,34,37,39,41) United States studies (37,39) European studies (30,34,41) Scandinavian studies (34,41) CI, confidenceinterval.
Sequential RR5 (95% CI)
Continuous RR5 (95% CI)
1.53 (1.47-1.60)
1.63 (1.55-1.72)
1.32 (1.11-1.56)
1.20 (1.01-1.44)
1.55 (1.48-1.62)
1.69 (1.60-1.78)
1.40 (1.19-1.64)
1.88 (1.61-2.21)
AL.
(NETA). In these regimens, the same daily dose of NETA, 1 mg, is used with both the sequential and the continuouscombined EPT, so that the total NETA dose per cycle is roughly 10 mg and 28 mg, respectively. The situation in the United Kingdom is more similar to that in the United States, with the continuous-combined regimens using lower daily progestin doses than that used in the sequential regimens, so that total progestin doses are not that different. This would be in agreement with the results found by the Million Women Study (30). Some of the greater overall risk found in the European, mainly Scandinavian, studies is due to the higher risks found with continuous-combined regimens in Scandinavia and is likely due to higher total doses of progestin (as described earlier). Although some of the remaining excess risk in the European studies may be due to the much greater use of NETA and norgestrel in Europe, most of the remaining excess is very likely due to a greater relative effect of EPT use on breast cancer risk among leaner women. Women in the United States studies are much heavier than the women in the European studies (31), consistent with overall population demographics. In this meta-analysis we did not assess differences in risk by weight, as only two studies evaluated risk with duration of EPT use by BMI (29,43); the others only gave results with duration of H T use (34) or with ever-use of H T or EPT (28,30,36,39). Summary risk estimates were higher for lobular than for ductal carcinoma, but the difference was not statistically significant. The results of these studies suggest that past use of EPT is associated with a lower risk than recent (or current) use, as has been consistently seen in studies of the effect of ET on breast cancer risk, even when proper account is taken of duration of use. This result was seen in the only study that directly compared risk for recent versus past EPT use (39). Magnusson and colleagues (34) compared recent and past H T use, as did the Million Women Study (30); they also found greater risk in recent users. However, the results from these two studies are difficult to interpret because H T use in past users includes proportionately more ET use. The observed lower risk with past use may be due at least in part to the fact that duration of hormone use is not measured the same in current as in past hormone users. The actual duration of use within a duration category will tend to be longer in current than in past users. Misclassification of duration of use is also likely to be higher with past use, leading to a greater underestimate of the risk associated with past use.
VI. C O N C L U S I O N S ET use is associated with a significant increased risk of breast cancer, a risk that is especially significant in slender women due to their much lower endogenous levels of estrogen and to the fact that the ceiling of estrogen effect
CHAPTER 40 Body Weight, Menopausal H o r m o n e Therapy, and Risk of Breast Cancer
on the breast is quite low. The ceiling level is not much higher than that found in a woman with a BMI of 30 kg/m 2, so that the effect of ET on breast cancer risk in such women or heavier women is very small. Reducing the estrogen dose in ET will reduce the excess risk in slender women, but calculations based on the model presented here suggest that reducing the dose by 50% (transdermal patch of 25-pg E2 or equivalent) in a 20-kg/m 2 woman will only reduce her excess risk by 20%. EPT use is associated with a much greater increased risk of breast cancer than ET use, and this increase, although greater in slender women, is still very evident in heavier women due to the fact that progestin appears to act independently of estrogen on the breast without the estrogen ceiling blunting the progestin effect. Reducing the progestin dose is likely to significantly reduce the risk; this is possible with an endometrial route of administration (44,45).
References 1. Trichopoulos D, MacMahon B, Cole E Menopause and breast cancer risk. J Natl Cancer Inst 1972;48:605-613. 2. Kelsey JL, Bernstein L. Epidemiology and prevention of breast cancer. Annu Rev Public Health 1996;17:47-67. 3. Colditz GA, Rosner B. Cumulative risk of breast cancer to age 70 years according to risk factor status: data from the Nurses' Health Study. Am J Epidemio12000;152:950- 964. 4. Henderson BE, Ross RK, Judd HL, Krailo MD, Pike MC. Do regular ovulatory cycles increase breast cancer risk? Cancer 1985;56:1206-1208. 5. MacMahon B, Cole P, Brown J. Etiology of human breast cancer: a review. J Natl Cancer Inst 1973;50:21-42. 6. Trichopoulos D, Hsieh CC, MacMahon B, et al. Age at any birth and breast cancer risk. IntJ Cancer 1983;31:701-704. 7. Wohlfahrt J, Melbye M. Age at any birth is associated with breast cancer risk. Epidemiology 2001;12:68-73. 8. Pike MC, Krailo MD, Henderson BE, Casagrande JT, Hod DG. "Hormonal" risk factors, "breast tissue age" and the age-incidence of breast cancer. Nature 1983;303:767- 770. 9. Cutler SY, Young JL. Third National Cancer Survey." incidence data, vol 4. Washington DC: National Cancer Institute Monograph, 1975. 10. Cook PJ, Doll R, Fellingham SA. A mathematical model for the age distribution of cancer in man. Cancer 1969;4:93-112. 11. Cairns J. Somatic stem cells and the kinetics of mutagenesis and carcinogenesis. Proc NatlAcad Sci USA 2002;99:10567-10570. 12. Rosner B, Colditz GA, Willett WC. Reproductive risk factors in a prospective study of breast cancer: The Nurses' Health Study. Am J Epidemio11994;139:819- 835. 13. Anderson TJ, Ferguson DJ, Raab GM. Cell turnover in the "resting" human breast: influence of parity, contraceptive pill, age and literality. BrJ Cancer 1982;46:376-382. 14. Pike MC, Spicer DV, Dahmoush L, Press ME Estrogens, progestogens, normal breast cell proliferation, and breast cancer risk. Epidemiol Rev 1993;15:17-35. 15. Meyer JS, Connor RE. Cell proliferation in fibrocystic disease and postmenopause breast ducts measured by thymidine labeling. Cancer 1982;50: 746- 750. 16. Hartz AJ, Barboriak TN, Wong A, Katama KP, Rim A_A.The association of obesity with infertility and related menstrual abnormalities in women. IntJ Obes 1979;3:57-73.
577
17. Endogenous Hormones and Breast Cancer Collaborative Group. Body mass index, serum sex hormones, and breast cancer risk in postmenopausal women. J Natl Cancer Inst 2003;95:1218-1226. 18. Endogenous Hormones and Breast Cancer Collaborative Group. Free estradiol and breast cancer risk in postmenopausal women: comparison of measured and calculated values. Cancer Epidemiol Biomark Prey 2003;12:1457-1461. 19. van den Brandt P, Spiegelman D, Yaun S-S, et al. Pooled analysis of prospective cohort studies on height, weight, and breast cancer risk. Am J Epidemio12000;152:514- 527. 20. S6dergard R, Backstrom T, Shanbhag V, Carstensen H. Calculation of free and bound fractions of testosterone and estradiol-17 beta to human plasma proteins at body temperature. J Steroid Biochem 1982;16: 801-810. 21. Dunn JF, Nisula BC, Rodbard D. Transport of steroid hormones: binding of 21 endogenous steroids to both testosterone-binding globulin and corticosteroid-binding globulin in human plasma. J Clin Endocrinol Metab 1981;53:58-68. 22. Schiff I, Sela HK, Cramer D, Tulchinsky D, Ryan KJ. Endometrial hyperplasia in women on cyclic or continuous estrogen regimens. Fertil Steril 1982;37:79- 82. 23. Powers MS, Schennkel L, Darley PE, et al. Pharmacokinetics and pharmacodynamics of transdermal dosage forms of 17-[3 estradiol: comparison with conventional oral estrogens used for hormone replacement. Am J Obstet Gyneco11985;152:1099-1106. 24. Selby PL, Peacock M. Dose dependent response of symptoms, pituitary, and bone to transdermal estrogen in postmenopausal women. BMJ 1986;293:1337-1339. 25. Nachtigall LE, Raju U, Banerjee S, Wan L, Levitz M. Serum estradiol profiles in postmenopausal women undergoing three common estrogen replacement therapies: associations with sex hormone-binding globulin, estradiol, and estrone levels. Menopause 2000;7:243-250. 26. Ropponen A, Aittomald K, Vihma V, Tikkanen MJ, Ylikorkala O. Effects of oral estradiol administration on levels of sex hormone-binding globulin in postmenopausal women with and without a history of intrahepatic cholestasis of pregnancy. J Clin Endocrinol Metab 2005; 90:3431-3434. 27. Stanczyk FZ. Pharmacokinetics and potency of progestins used for hormone replacement therapy and contraception. Rev Endocrin Metab Disord 2002;3:211-224. 28. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Lancet 1997;350:1047-1059. 29. Schairer C, Lubin J, Troisi R, et al. Menopausal estrogen and estrogenprogestin replacement therapy and breast cancer risk. J A m MedAssoc 2000;283:485-491. 30. Million Women Study Collaborators. Breast cancer and hormonereplacement therapy in the Million Women Study. Lancet 2003;362: 419-427. 31. Beral V, Reeves G, Banks E. Current evidence about the effect of hormone replacement therapy on the incidence of major conditions in postmenopausal women. BrJ Obstet Gynaeco12005;112:692-695. 32. Women's Health Initiative. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy. JAm MedAssoc 2004;291: 1701-1712. 33. Lee SA, Ross RK, Pike MC. An overview of menopausal oestrogenprogestin hormone therapy and breast cancer risk. Br J Cancer 2005 ;92:2049 - 2058. 34. Magnusson C, Baron JA, Correia N, et al. Breast-cancer risk following long-term oestrogen- and oestrogen-progestin-replacement therapy. IntJ Cancer 1999;81:339-344. 35. Ross RK, Paganini-Hill A, Wan PC, Pike MC. Effect of hormone replacement therapy on breast cancer risk: estrogen versus estrogen plus progestin. J Natl Cancer Inst 2000;92:328- 332.
5
7
8
W
36. Newcomb PA, Titus-Ernstoff L, Egan KM, et al. Postmenopausal estrogen and progestin use in relation to breast cancer risk. Cancer Epidemiol Biomarkers Prey 2002;11:593-600. 37. Porch JV, Lee IM, Cook NR, Rexrode KM, Burin JE. Estrogenprogestin replacement therapy and breast cancer risk: the Women's Health Study (United States). Cancer Causes Control 2002;13: 847-854. 38. Kirsh V, Kreiger N. Estrogen and estrogen-progestin replacement therapy and risk of postmenopausal breast cancer in Canada. Cancer Causes Control 2002;13:583-590. 39. Weiss LK, Burkman RT, Cushing-Haugen KL, et al. Hormone replacement therapy regimens and breast cancer risk. Obstet Gynecol 2002;100:1148 - 1158. 40. Chlebowski RT, Hendrix SL, Langer RD, et al. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women: the Women's Health Initiative Randomized Trial. JAm MedAssoc 2003;289:3243- 3253.
u
ET AL.
41. Stahlberg C, Pedersen AT, Lynge E, et al. Increased risk of breast cancer following different regimens of hormone replacement therapy frequently used in Europe. IntJ Cancer 2004;109:721- 727. 42. Jernstrom H, Bendahl PO, Lidfeldt J, et al. A prospective study of different types of hormone replacement therapy use and the risk of subsequent breast cancer: the women's health in the Lund area (WHILA) study (Sweden). Cancer Causes Control 2003;14:673-680. 43. Ursin G, Tseng CC, Pagini-Hill A, et al. Does menopausal hormone replacement therapy interact with known factors to increase risk of breast cancer? J Clin Onco12002;20:699- 706. 44. Shoupe D, Meme D, Mezrow G, Lobo RA. Prevention of endometrial hyperplasia in postmenopausal women with intrauterine progesterone. NEnglJMed 1991;325:1811-1812. 45. Varila E, Wahlstrom T, Rauramo I. A 5-year follow-up study on the use of a levonorgestrel intrauterine system in women receiving hormone replacement therapy. Fertil Sterff 2001;76:969-973.
~HAPTER
41
Postmenopausal Breast Surveillance WILLIAM H.
HINDLE
Department of Obstetrics and Gynecology, University of Southern California; The William Hindle, M.D. Breast Diagnostic Center, Los Angeles, CA 90033
Breast cancer is the most common systemic (nonskin) cancer of women. The incidence rate of invasive breast cancer continues to increase as women grow older. Breast cancer is the most common cause of cancer death in women ages 20 to 59 and the second most common cause of cancer death for women of all ages (1). For 2005 it is estimated that there were 211,240 new cases of invasive breast cancer diagnosed and that 40,410 breast cancer deaths occurred (1). For women born in the United States, the lifetime (birth to death) probability of developing invasive breast cancer is calculated at 13.39%, approaching 1 in 7 (1). However, there are encouraging trends in early diagnosis of breast cancer, minimally invasive diagnostic testing and surgical therapy, and more effective adjuvant therapies. Simultaneously, the incidence rate of breast cancer deaths has decreased about 2.5% annually for the past 10 years (1). The key role of primary care health care providers for women is rigorous breast cancer screening (2). Due to the frequent occurrence of medical malpractice suits brought for "failure to diagnose breast cancer," it is imperative that the recommendation, ordering, and results of the annual screening mammogram and clinical breast examination be clearly and accurately documented in the patient's medical record (3). At menopause, with its subsequent estrogen deficiency, the breasts begin to atrophy; the glandular tissue continues TREATMENT OF THE POSTMENOPAUSAL WOMAN
to diminish and be progressively replaced by fat; benign breast diseases, including neoplasms, subside; and tenderness and palpable nodularity clear. Estrogen replacement therapy (ERT) can reverse or delay these breast-aging changes in some cases. However, the incidence of cancer continues to rise with advancing age. Annual screening mammography is the most important tool in the surveillance of the breasts of postmenopausal women. To detect small cancers, which have an excellent prognosis when treated, physicians who care for postmenopausal women should obtain annual mammograms (bilateral craniocaudal and mediolateral oblique views) for and perform complete annual breast examinations on all their patients. Breast cancer is the most common cancer of postmenopausal women. The incidence of breast cancer is more than twice the combined incidence of cervical, endometrial, and ovarian cancer in this age group (1). The incidence rate of breast cancer continues to increase every year as a woman grows older (Fig. 41.1) (3,4). The cumulative lifetime risk of breast cancer is 1 in 8for women born in the United States (1). More than two-thirds (67%) of all invasive breast cancers occur after age 60 (1). The age-specific incidence rates (1,4) and the aging of the population combine to make breast cancer an everincreasing health care dilemma, particularly for postmenopausal women (5). 579
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
580
WILLIAM H. HINDLE 500 400 Menopause
Annual Rate per 100,000 women
300 200 -
100-
0 ~ 0
L J" 2O
I
I
4O
60
, I 8O
100
Age FIGURE 41.1 Linearscale U.S. age-specificincidence rates per 100,000 white women for breast cancer based on the Surveillance,Epidemiology, and End Results (SEER) program data, 1978-1984. (From ref. 41, with permission.)
Benign breast disease, including cysts, fibroadenomas, fibrocystic changes, and mastalgia, markedly decrease with menopause and its subsequent estrogen deficiency (6). Elicited milky nipple discharge subsides after menopause. However, spontaneous watery, serous, or bloody nipple discharge from a single duct must be investigated. These may be signs of underlying carcinoma, although such nipple discharges are most commonly related to intraductal papillomas. Ductography (galactography) is useful in defining the extent of the intraductal lesion and whether it is single or multiple. Multiple papilloma formation is more commonly associated with papillary carcinoma. Cytology of bloody nipple discharge can be useful but only if malignant cells are identified. Nipple discharge cytology is not cost effective, does not alter the clinical management, and does not circumvent the need for definitive surgery. Surgical excision should be limited to the involved collecting duct draining a single glandular lobe into a single opening (i.e., galactophore) on the nipple. However, all of the ductal system showing intraductal abnormalities should be excised. This can usually be performed through a limited circumareolar incision and microdochectomy. Only permanent section histologic evaluation of the excised duct can establish the definitive diagnosis. If ductography is not available, the oncologic surgeon can pass a fine probe down the single duct that produces the discharge (6). Mammary ectasia (i.e., periductal mastiffs), which is most common during the perimenopausal ages, usually presents as dark greenish discharge, often from multiple duct openings on the nipple. Treatment is usually not necessary for this self-limited, benign condition. Hemostix or similar dipsticks are useful to determine if dark nipple discharge contains hemoglobin. If so, it must be investigated.
Monthly breast self-examination (BSE) should be encouraged for postmenopausal women (2,7). Because the aging breast usually becomes smaller, less tender, less nodular, and less thickened, BSE is potentially more accurate in detecting palpable breast masses. Significant changes from prior BSE and new differences between the right and left breast, such as dimpling of the skin or inversion of the nipple, which can be secondary signs of underlying breast cancer, must be evaluated. If a new mass is identified, it must be definitely diagnosed. Contrary to the case for younger women, a high percentage of palpable dominant breast masses in postmenopausal women are cancers. Methodical bilateral clinical breast examination (CBE) should be performed at least once each year and include both axiUae and the supraclavicular and subareolar areas. The vertical strip method recommended by the American Cancer Society is the most effective technique for detecting palpable breast masses (2,8,9). Unfortunately, palpable breast cancers have usually already reached stage II or are even more advanced and carry an unfavorable prognosis. It is critical that CBE findings be recorded in the medical record, particularly any abnormal findings. A pictorial diagram is useful to record the exact location of any abnormality, especially a suspected or suspicious mass. A specific foUow-up plan should be recorded. All palpable dominant masses and suspicious mammographic lesions in postmenopausal women must be specificaUy diagnosed in a timely manner to definitively exclude cancer. Palpable masses can be evaluated by fine-needle aspiration (FNA), with cysts being evacuated and aspirates of solid masses sent for cytology. Because FNA may produce hematoma formation, which can mammographically mimic a suspicious lesion in the breast, mammography should be done before or 2 weeks after FNA (10). However, if the mammogram is taken shortly thereafter and the mammographer is aware that an FNA has been performed (including the exact location in the breast), clinical management is rarely altered (11). FNA is performed with a 22-gauge, 1-inch needle with a transparent plastic hub and a 10-mL syringe. Most physicians prefer a pistol-type syringe holder or a three-finger control syringe. The FNA technique is illustrated in Fig. 41.2. The dominant breast mass must be completely immobilized, preferably over a rib, with the fingers of the nonaspirating hand. The skin is cleaned with an alcohol wipe, and the needle tip is sharply thrust downward in jackhammer-like motions back and forth about 20 times within the mass while negative pressure is maintained in the syringe. At all times during the aspiration (negative pressure), the needle tip is kept within the mass. Local anesthesia usually is not necessary because the procedure is no more painful than a venipuncture and takes about the same amount of time to perform.
CHAPTER 41 Postmenopausal Breast Surveillance
FIGURE41.2 Five-steptechnique of fine-needle aspiration for cytology of a palpable dominantbreast mass. An alternate method for FNA is the needle-alone technique, which requires more thrusts (preferably 30 or more) as the cellular material is accumulated in the barrel of the needle by capillary action, not by negative pressure in the syringe. With this technique, usually less cellular material is obtained. However, the learning curve is shorter, less blood is obtained, and tactical sensation (and control) is increased. A cyst should be completely drained and followed to be certain there is no residual mass and that the cyst does not reform (8). Cytologic examination of cyst fluid is rarely rewarding and is indicated only for frankly bloody fluid. Routine cytologic examination of cyst fluid is not cost effective and rarely alters clinical management (12). Cellular material that collects in the bore of the needle during the FNA of a solid mass should be quickly ejected onto a properly identified slide, smeared, and fixed. The slide can then be stained at leisure, usually with Papanicolaou or hematoxylin and eosin stain. An alternate method of slide preparation is air drying and then staining, usually with some version of Wright-Giemsa (8,13). With an adequate cell sample, as many as 90% of solid breast masses can be specifically diagnosed by FNA cytology (8,13,14). If a specific cytologic diagnosis is not made by FNA, a definitive histologic diagnosis should be obtained by tissue core-needle biopsy or open surgical biopsy. Open surgical excision biopsies are usually done under local anesthesia in an outpatient setting. Evaluation of multiple permanent histologic sections is required for definitive diagnosis. Excision biopsies should be done following the National Surgical Adjuvant Breast Project (NSABP) guidelines for lumpectomy with clear margins (15). Nonpalpable masses can be evaluated by stereotactic mammography-guided FNA or tissue core-needle biopsy (16,17). However, if a specific diagnosis is not obtained, mammographic needle localization of the nonpalpable mass and lumpectomy with clear surgical margins are required. These procedures are usually performed simultaneously on an outpatient basis using local anesthesia. Specimen radiography before the operation is completed is essential to be
581
certain the nonpalpable mammographic lesion has been excised. If not, repeat excision is required. The weight and dimensions of the lumpectomy specimen should be recorded. Multiple histologic sections should be evaluated to establish a definitive diagnosis and that all the surgical margins are clear. The maximum diameter of the malignant lesion should be measured and recorded. Screening mammography is the only proven method of meaningfully decreasing the overall mortality from breast cancer (16,18,19). Clinically significant mammographic findings that should be specifically reported are a spiculated mass, a circumscribed mass, microcalcifications, an asymmetric density, and skin changes (i.e., thickening and retraction). The mammographic impression can be negative (i.e., normal or no abnormalities noted), benign finding, probably benign finding, suspicious abnormality, or highly suggestive of malignancy. Most mammographers add a qualifier about their level of confidence in the mammographic impression and add their recommendation. The mammogram report should follow the breast imaging reporting and data system (BIRADS) format and terminology, including the complete assessment category (categories 1 to 5) (16,20,21). Even complete mammographic evaluation and workup does not give a definitive clinical diagnosis, although experienced mammographers develop a high clinical correlation. Definite cytology by FNA or histology by tissue core-needle or open biopsy is necessary for the final specific definitive diagnosis of a breast lesion (8). A "negative" mammogram does not exclude cancer and gives no useful clinical information other than the fact that the mammographer does not perceive a significant abnormality on viewing the films. A negative mammogram does not ensure disease-free status. Even though the overall accuracy of mammography is about 90% (22-24) and the progressive fatty replacement of glandular tissue in the aging breast increases the accuracy of mammography (22,23), CBE must continue to be performed in the postmenopausal group. Typical changes in premenopausal and postmenopausal mammograms are illustrated in Figs. 41.3 and 41.4. Screening mammography is useful throughout a woman's life, and there is no upper age limit (2,25). However, after age 85, estimated life expectancy and quality of life issues may outweigh the benefits, particularly when viewed on a population-wide basis and with consideration of health care costs. Each elderly woman should be evaluated individually, her family consulted, and her informed consent obtained. Breast infections occur in postmenopausal women but are unusual. Diffuse mastiffs must be differentiated from inflammatory carcinoma. If the mastiffs does not respond to antibiotic therapy, a skin biopsy should be done to look for possible carcinoma involvement of the dermal lymphatics. Subareolar fistulas continue to occur in postmenopausal women. Excision often is necessary to eradicate these chronic or recurrent infections.
582
WILLIAM H. HINDLE
FIGURE 41.3 Typical normal mammogram (left craniocaudal and mediolateral oblique views) of a postmenopausal woman.
Before initially prescribing ERT for a patient, the physician should obtain a mammogram unless one has been performed within the preceding year. Annual mammograms should be obtained thereafter. CBE should also be done before ERT and every year thereafter. Monthly BSE should be encouraged. Breast cysts or fibroadenomas may become symptomatic, palpable, or enlarged when a women begins ERT. Such symptoms and palpable masses must be definitively diagnosed to exclude cancer. The diagnostic triad of
CBE, mammography, and FNA (Fig. 41.5) should give a specific diagnosis. However, if uncertainty remains in the mind of the patient or her physician, a definitive histologic diagnosis should be obtained by tissue core-needle biopsy or open surgical biopsy. Women who have been treated for breast cancer should be followed in the same manner as other postmenopausal women with annual mammography and CBE (26). They should have an annual chest film and blood obtained for alkaline phosphatase. Further studies such as bone, liver, or brain scans should be done to evaluate symptoms that suggest recurrent disease. Women on tamoxifen should be monitored for evidence of osteoporosis (by evaluation of bone mineral content) or cardiovascular disease (by measurements of blood pressure, serum cholesterol or lipid profile, and weight). However, osteoporosis and cardiovascular disease are not thought to be adversely affected by tamoxifen therapy. Studies on the long-term effects of tamoxifen indicate, however, that there is an increased incidence of endometrial carcinoma (27,28), warranting close monitoring of each woman with a uterus. Data suggest decreased osteoporosis (29-31) and unchanged or decreased cardiovascular disease with tamoxifen (32- 34). The National Surgical Adjuvant Breast Project (NSABP) clinical trials are evaluating tamoxifen therapy as a prevention strategy for invasive breast cancer in high-risk women (NSABP P-l), as adjuvant therapy after lumpectomy for ductal carcinoma in situ (DCIS), for prevention of ipsilateral or contralateral invasive cancer (NSABP B-24), and as adjuvant therapy after lumpectomy for occult invasive carcinoma with and without adjuvant radiation therapy (NSABP B-21).
FIGURE 41.4 Typicalnormal mammogram (left craniocaudal and mediolateral oblique views) of a postmenopausalwoman.
CHAPTER 41 Postmenopausal Breast Surveillance
FIGURE 41.5 The diagnostic triad of clinical breast examination (CBE), mammography, and fine-needle aspiration (FNA) for a persistent dominant breast mass. Congruence of all three techniques provides a definitive specific diagnosis.
Compliance with surveillance decreases as women become elderly (35). A personal physician's recommendations and advice are the strongest motivators for compliance (36). A careful follow-up system with automatic reminders assists the physician in maintaining compliance. Twelve major medical organizations recommend screening mammography for women every year after 50 years of age (37-39). Annual mammography and CBE are required for postmenopausal women to have the best chance for early diagnosis of breast cancer (2,22,39,40). Optimal treatment of women with invasive breast cancer continues to change due to advances in minimally invasive techniques and improving effectiveness of adjuvant therapies particular chemotherapy. Ideally every woman with diagnosed invasive breast cancer should benefit from pretreatment evaluation by a multispecialty breast center. After a precise comprehensive individualized treatment plan has been determined, the ongoing care of the patient is usually under the purview of a medical oncologist.
References 1. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005;55:10-30. 2. Smith RA, Cokkinides V, Eyre HJ. American cancer society guidelines for the early detection of cancer, 2005. CA CancerJ Clin 2005;55:31-44. 3. Hindle WH. Breast cancer: introduction. Clinical Obstet Gynecol 2002;45:738-745. 4. Boring MS, Squires T, Tong T. Cancer statistics, 1991. Cancer 1991;1:19-39. 5. National Cancer Institute. Division of Cancer Prevention and Control. 1987 annual cancer statistics review, including cancer trends." 1950-1985, pub no 88-2789. Bethesda, MD: National Institutes of Health, 1988: III-36. 6. Hindle WH, Arias RE). Nipple discharge. In: Hindle WH, ed. Breast care. New York: Springer-Verlag, 1999. 7. Saunders KJ, Pilgrim CA, Pennypacker HS. Increased proficiency of search in breast self-examination. Cancer 1986;58:2531-2537. 8. Hindle WH. Breast mass evaluation. Clinical Obstet Gynecol 2002;45: 750-757.
583 9. Miller BA, Feuer EJ, Hankey BE. The increasing incidence of breast cancer since 1982: relevance of early detection. Cancer Causes Control 1991;2:67-74. 10. Sickles EA, Klein DL, Goodson W H 3rd, et al. Mammography after fine-needle aspiration of palpable breast masses. Am J Surg 1983;145: 395-397. 11. Hindle, WH, Chen EC. Accuracy of mammographic appearance after breast fine-needle aspiration. Am J Obstet Gynecol 1997;176:1286-1292. 12. Hindle WH, Arias RD, Florentine B, et al. Lack of utility in clinical practice of cytologic examination of nonbloody cyst fluid from palpable breast cysts. Am J Obstet Gyneco12000;182:1300-1305. 13. Hindle WH, Arias RD, Felix J, Sue& A. Adaptation of FNA to office practice. Clinical Obstet Gyneco12002;45:761-766. 14. Kline TS, Kline IK. Breast: guides to clinical aspiration biopsy. New York: Igaku-Shoin Medical Publishers, 1989:1-19. 15. Margolese R, Poisson R, Shibata H, et al. The technique of segmental mastectomy (lumpectomy) and axillary dissection: a syllabus from the National Surgical Adjuvant Breast Project Workshops. Surgery 1987;102:828-834. 16. Hindle WH. Mammography--screening and diagnosis. Clinical Obstet Gyneco12002;45:746- 749. 17. Costanza ME. Breast cancer screening in older women: synopsis of a forum. Cancer 1992;69:1925-1931. 18. Duffy SW, Tabar L, Fagerberg G, et al. Breast screening, prognostic factors and survival: results from the Swedish two county study. BrJ Cancer 1991;64:1133-1138. 19. Tabar L, Fagerberg G, Duffy SW, Day NE. The Swedish two county trial of mammographic screening for breast cancer: recent results and calculation of benefit. J Epidemiol Commun Health 1998;43:107-114. 20. BIRADS American College of Radiology (ACR). Breast imaging reporting and data system (BI-tL4DS), 2nd ed. Reston, VA: American College of Radiology, 1995. 21. Hindle WH. Appendix B: Reporting system. In: Hindle WH, ed. Breast care. New York: Springer-Verlag, 1999. 22. Edeiken S. Mammography in the symptomatic woman. Cancer 1989; 63:1412-1414. 23. Bender HG, Schnurch HG, Beck L, et al. Breast cancer detection: agerelated significance of findings on physical exam and mammography. Gynecol Onco11988;31:166 -175. 24. Feig SA. Should breast self-examination be included in a mammographic screening program? Recent Results CancerRes 1990;119:151-164. 25. Mandelblatt JS, Wheat ME, Monane M, et al. Breast cancer screening for elderly women with and without comorbid conditions. Ann Intern Med 1992;116:722-730. 26. Hindle WH. Treatment of invasive carcinoma. Clinical Obstet Gynecol 2002;45:767-769. 27. Fornander T, Cedemark B, Mattsson A, et al. Adjuvant tamoxifen in early breast cancer: occurrence of new primary cancers. Lancet 1989;1: 117-120. 28. Killackey MA, Hakes TB, Pierce VK. Endometrial adenocarcinoma in breast cancer patients receiving antiestrogens. Cancer TreatRep 1985;69: 237-238. 29. Love RR, Mazess RE, Barden HS, et al. Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. N Engl J M e d 1992;326:852-856. 30. Fornander T, Rutquist LE, Sjoberg HE, et al. Long-term adjuvant tamoxifen in early breast cancer: effect on bone mineral density in postmenopausal women. J Clin Oncol 1990;8:1019 - 1024. 31. Love RR, Mazess RE, Tormey DC, et al. Bone mineral density in women with breast cancer treated with adjuvant tamoxifen for at least two years. Breast Cancer Res Treat 1988;12:297-302. 32. Love RR, Wiebe DA, Newcomb PA, et al. Effects of tamoxifen on cardiovascular risk factors in postmenopausal women. Ann Intern Med 1991;115:860-864.
584 33. Bagdade JD, Rayn G, Wolter JM. Effects of tamoxifen treatment on plasma lipids lipoprotein composition. J Clin Endocrinol Metab 1990; 70:1132-1135. 34. Love RR, Newcomb PAL,Wiebe DA, et al. Effects of tamoxifen therapy on lipid and lipoprotein levels in postmenopausal patients with node-negative breast cancer. J Natl Cancer Inst 1990;82:1327-1332. 35. Weinberger M, Saunders AF, Samsa GP, et al. Breast cancer screening in older women: practice and barriers reported by primary care physicians. JAm Geriatr Soc 1991;39:22-29. 36. Glockner SM, Holden MG, Hilton SVW, Norcross WA. Women's attitudes toward screening mammography. Am J Prey Med 1992;8:69-77.
WILLIAM H. HINDLE 37. American College of Obstetricians and Gynecologists. Breast cancer screening. ACOG Practice Bulletin No. 42. Obstet Gyneco12003;101: 821-832. 38. Medical News and Perspectives. Breast cancer screening guidelines agreed on by AMA, other medically related organizations. JAMA 1989;262:1155. 39. Council on Scientific Affairs. Mammographic screening in asymptomatic women aged 40 years and older. JAMA 1989;261:2535-2542. 40. Hindle WH. Breast cancer: prevention and surveillance. Clin Obstet Gyneco12002;45:778- 783.
t A P T E R 4s
Endometrial Cancer and Hormonal Replacement Therapy J O H N P. C U R T I N
Department of Obstetrics and Gynecology, New York University School of Medicine, New York, NY 10016
Endometrial carcinoma is the most common gynecologic malignancy in the United States (1). Worldwide this cancer is a disease predominately affecting women from Western, industrialized countries (2). The incidence of endometrial cancers may be up to 10 times higher in the industrialized nations, and for the majority of patients lifestyle factors are responsible for the differences in incidence rates (2,3). The majority of uterine cancer cases in the United States belongs to the typical endometrioid adenocarcinoma or type I endometrial cancers. These cancers are characterized by several features, including early stage at diagnosis, low-grade histology, association with atypical endometrial hyperplasia (Fig. 42.1) (see color insert), and overall good prognosis. A common feature is a history of either known exogenous estrogen exposure or clinical syndromes that suggest an excess of endogenous estrogen production. The single most important environmental factor influencing the incidence of endometrial cancer is diet and obesity. Exogenous estrogens, most commonly prescribed for management of postmenopausal symptoms, are also known to be associated with endometrial cancers. TREATMENT OF THE POSTMENOPAUSAL WOMAN
585
I. I N C I D E N C E In the United States the American Cancer Society estimates that more than 40,800 women will be diagnosed with and 7300 women will die of cancer of the uterus in 2006 (1). The overall lifetime risk of endometrial adenocarcinoma is 2.61 cases per 100 women. The disease is uncommon in young menstruating women, with fewer than 10% of all cases occurring in patients less than 50 years of age. The median age at the time of diagnosis is 63 years. The incidence of endometrial cancers begins to rise after age 50 years and peaks during the sixth through the eighth decades. There continues to be a significant annual incidence that does not begin to decline until after age 85 years. The age-adjusted incidence rate is 24.2 per 100,000 women, and the age-adjusted death rate is 4.1 per 100,000 women. In comparison, the age-adjusted incidence and death rate for ovarian cancer is 13.9 per 100,000 and 8.9 per 100,000 women per year in the United States, respectively (4). In addition to differences in the incidence by age of the patient, there are also significant differences in the incidence as well as deaths due to uterine cancer by ethnicity and race. Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
JOHN P. CURTIN
586
Overall the highest incidence of uterine cancer occurs among Caucasian women (see Fig. 42.1) (see color insert); for women age 50 years or older the age-adjusted incidence rate is approximately 80 cases per 100,000 Caucasian women, compared with approximately 60 cases per 100,000 AfricanAmerican women. Conversely the death rate is higher among African-American women as compared with Caucasian women with uterine cancers (4). In part this has been attributed to several factors, including a higher incidence of type II cancers as well as health care access issues related to lower socioeconomic status (Fig. 42.2).
II. RISK FACTORS
FIGURE 42.1 hyperplasia.
A typical photomicrograph of atypical endometrial
FIGURE 42.2
Risk factors for endometrial cancer can be viewed as related to an estrogen effect or another factor. For example, patients who have a longer lifetime history of estrogen exposure due to menses (i.e., early onset menarche, late menopause, or a combination of both) have a higher risk of endometrial cancer. Although endometrial cancer is uncommon before the age of 45 years, in the majority of cases there is clinical evidence of anovulation with resultant chronic unopposed estrogen exposure. In series of women diagnosed with endometrial cancer who are in the
SEER data on cancer incidence in the U.S. (From SEER 9 areas and NCHS public use date file.)
CHAPTER 42 Endometrial Cancer and Hormonal Replacement Therapy
reproductive age group, approximately two-thirds of patients have a clinical syndrome compatible with polycystic ovarian syndrome (PCOS) (5). Further study of the pathogenesis of PCOS may shed additional light on another commonly associated medical condition associated with endometrial cancers, namely diabetes. Diabetics are known to have an increased lifetime risk of endometrial cancer. Obesity, a medical condition of epidemic proportions, is another common cause of endogenous exposure to excessive estrogen. There is a direct and near linear correlation between obesity and estrogen levels (6,7). There is also a strong correlation between obesity and the overall risk of endometrial cancer. Another "natural experiment" occurs in the case of hormonaUy active tumors of the ovary such as granulosa cell tumors (GCT). GCTs are usually diagnosed at an early stage. Approximately 80% of these tumors secrete estrogen. In more than half of the cases there is evidence of either preneoplastic or invasive endometrial cancer at the time of diagnosis. The effect of the excess production of estrogen often is abnormal bleeding, which in turn may be the presenting symptom.
587
endogenous or exogenous estrogen stimulation of type II cancers (9). Because estrogen stimulation appears to be a critical factor in type I endometrial carcinogenesis, it is important to briefly review the effect of estrogen on the endometrium. Estrogen acts to increase the insulin growth factor-1 (IGF-1); this has a direct impact on the antiapoptotic pathways promoting endometrial hyperplasia. This hyperplasia may result in clinical symptoms of bleeding but does not result in carcinogenesis alone. Endometrial hyperplasia with atypia and endometrial cancer require multistep changes in genetic pathways, which in turn lead to invasive cancer. As an example, P T E N mutations are common in both atypical endometrial hyperplasia and endometrial cancer. PTEN, an essential lipid phosphatase, acts as a tumor suppressor gene that is a key component of many signal transduction pathways. P T E N negatively regulates signal pathways and acts as a suppressor factor. Given that P T E N mutations are found in both normal endometrium and endometrial hyperplasia, it has been postulated that endometrial carcinoma occurs when there is a clonal selection of cells without the suppressor function of PTEN, which under the stimulation of estrogen and in the absence of progesterone results in neoplastic and malignant transformation (10).
III. PATHOPHYSIOLOGY It has been widely accepted that endometrial neoplasms can be broadly divided into two categories (7). Type I cancers are the most common and account for approximately 80% of all endometrial cancers. Type I cancers are characterized by an association with estrogen. The type I cancers almost always have some estrogen receptors and are often associated with either endogenous or exogenous estrogen exposure. The presumed pathogenesis and progression begin with chronic estrogen hyperstimulation of the endometrial glands. This initially results in endometrial hyperplasia. Endometrial hyperplasia without nuclear atypia does not appear to be associated with a significant risk of progressing to carcinoma. However, once nuclear atypia develops, there is a significant risk of either developing subsequent adenocarcinoma or having an early invasive lesion associated with atypical endometrial hyperplasia. Type II cancers are associated with high-grade lesions, as well as papillary serous carcinomas of the endometrium. These lesions account for approximately 20% of all endometrial cancers. Type II cancers are characterized by advanced stage at diagnosis in many patients, as well as an overall poor prognosis. Histologically these cancers often develop in a background of atrophic endometrium. As compared with type I cancers, estrogen receptors are usually absent in type II cancers. Aberrant 1)53 gene (a tumor suppressor gene) is another common feature of type II cancers. There does not appear to be a significant role for either
IV. EFFECT OF H O R M O N A L THERAPY Although empiric observations had identified endometrial cancer as an estrogen-induced cancer, the introduction of estrogen replacement therapy (ERT) in the late 1960s provided the evidence of direct effect of exogenous hormones on the endometrium. Synthetic estrogens were available as early as the 1930s in the United States; however, the widespread prescribing of estrogen started in the 1960s. As the number of prescriptions for estrogen therapy increased, so did the incidence rate of endometrial cancers. The peak incidence of endometrial cancer cases occurred in the mid 1970s and coincided with the publication of several papers that associated the use of unopposed estrogen and the development of endometrial carcinoma (11-13). Following the publication of these epidemiologic studies, the number of new prescriptions for estrogen therapy fell by approximately 50% over a 5-year period. The decline in the incidence rate of endometrial cancer again mirrored the prescribing patterns. Within the same 5-year time frame, the incidence of endometrial cancer returned to the current level following the publication of these reports. It is clear that estrogen acts directly on the endometrium in a dose-dependent relationship. The risk of estrogen therapy alone (ERT) is dependant on route of administration as well as agent used and duration of use.
588
JOHN P. CURTIN
A recent population-based, case-control study from Sweden studied the impact of low dose versus high dose as well as duration of therapy for estrogen alone. In this study the odds ratio (OR) for women taking estrogens alone was 3.0; the relative risk increased by 17% per year so that after 10 years of estrogen alone the OR was greater than 8 times the baseline risk in the control population. The use of a "low" versus "high" dose was also associated with a doserelated increase in risk. After 5 years of estrogen therapy alone, low-dose users had a fourfold increase in risk as compared with the high-dose users, who had an eightfold increase. As an example of a low-dose regimen, patients were taking less than 0.625 mg/day of conjugated estrogen; 0.625 mg/day or greater was considered high dose (14). Addition of progesterone to ERT appears to counteract the carcinogenic potential of estrogen (15-17). The dose of progesterone is critical in the overall reduction of endometrial cancer; longer duration of treatment is directly related to the incidence of endometrial cancer. However, there have been conflicting reports in the literature. In the same population-based, case-control study from Sweden described earlier, estrogen plus cyclic progestin did not protect from endometrial cancer and in fact was associated with a small but statistically significant 30% (OR 1.3) increase in the incidence of endometrial cancer (15). A recent publication analyzed the data collected as part of the Women's Health Initiative (WHI) randomized trial of continuous combined hormonal replacement therapy (14). Patients randomized to the treatment arm were prescribed a formulation containing both a progestin and a conjugated estrogen in a single daily tablet. Among women who were enrolled in the study, the observed incidence of endometrial cancers was calculated to be 62 per 100,000 person-years, which was lower than the predicted estimate of 83 cases per 100,000 person-years based on SEER data (Table 42.1). One explanation for the lower than anticipated incidence is the study requirement that all patients who were
TABLE 42.1
Type of cancer Endometrial (annualized %) Adenocarcinoma/ endometrioid Tumor grade I
II III Undifferentiated Data from ref. 27.
WHI: C C H vs. Placebo ~ Incidence of Endometrial Cancer CCH (n = 8506)
Placebo (n = 8102) HR (95% CI)
27 (.0.06)
31 (.0.07)
22
26
9
6 15 7 3
10 5 3
0.81 (0.048-1.36)
enrolled must first had to have a baseline endometrial biopsy; the data reported did not state how many patients were excluded due to abnormal endometrial pathology. A second large study that clarified the risk of endometrial cancer was the Million Women Study conducted in the United Kingdom from 1996 to 2001. Although not a randomized study, the data derived was based on enrollment of large numbers ofwomen undergoing immunographic screening and cancer incidence was calculated from a National Health Service reporting system. In the Million Women Study, enrollees were categorized into never users versus users of hormonal replacement therapy. Hormonal replacement users were further assigned to four groups; continuous combined therapy, cyclic therapy, tibolone, and estrogen alone. In the analysis of the risk of endometrial cancer, only those women who were at risk (i.e., had not undergone a hysterectomy) were included. In addition to endometrial cancer, breast cancer cases were also recorded. Tibolone, a synthetic hormone, has both estrogenic and progestogenic effects. This agent has been utilized in the United Kingdom as an alternative to estrogen therapy but is not available in the United States. In the analysis, both tibolone and estrogen alone were associated with an increase risk of endometrial cancer. In women who took continuous estrogen combined with progestin, the relative risk of endometrial cancer was 0.71 (CI 0.58-0.90). Addition of a progestin to the estrogen regimen reduced the overall rate of endometrial cancer. There was a trend toward lower rates with longer duration of treatment with progestins. As an example, for patients taking progestins fewer than 10 days per month, the relative risk of endometrial cancer was 1.12 compared with a relative risk of endometrial cancer of 0.75 for women who took a progestin for 13 to 14 days per month. Although there were fewer cases of endometrial cancer among patients who were randomized to continuous combined hormonal therapy, the difference was not statistically significant. There were no differences in the histology, grade of the tumor, or stage of disease based on whether the patient received continuous combined therapy. The goal of reducing the overall risk of endometrial cancer in a patient under treatment with combination progestogen and estrogen must be balanced by the associated risk of adding progestogens to HRT. Recent studies have supported the hypothesis that progestogen may contribute to an increased risk of breast cancer. In the W H I study of combination estrogen and progestin therapy, there was an overall increase in the incidence of breast cancer. The data from the Million Women Study indicated an added risk of breast cancer when progestins were added to hormonal therapy. Although the Million Women Study did demonstrate that the addition of continuous or cyclic progestin either lowered or had no effect on the incidence of endometrial cancer in women on estrogens, there was an
CHAPTER 42 Endometrial Cancer and Hormonal Replacement Therapy
overall increase in the cancer rate due to the higher numbers of breast cancer cases. This theory is further supported by newer data that have shown that progestins can promote carcinogenesis in the breast. Conversely, in the W H I study of women receiving estrogen alone, the incidence of breast cancer may have been lower among users as compared with nonusers. Also, in the Million Women Study the increased incidence of breast cancer was greatest among women who took both estrogen and progestin as compared with either tibolone or estrogen (18).
V. HORMONAL REPLACEMENT, OBESITY, AND ENDOMETRIAL CANCER Obesity is the single most important nongenetic factor influencing the incidence of endometrial cancer. It could be assumed that the addition of an exogenous estrogen in a patient already at risk due to obesity could further raise the overall risk. One key finding of the Million Women Study suggests that this is not the case. The Million Women Study not only collected data regarding hormonal usage but also collected height and weight data, allowing for calculation of body mass index (BMI) for all registrants. This allowed the investigators to examine the relative contribution of the risk of developing endometrial cancer as related to BMI. Women were classified as either not being overweight (BMI < 25 kg/M2), overweight (BMI 2 5 - 2 9 kg/M2) or obese (BMI > 30 kg/M2). The results are shown in Fig. 42.2 (18). For users of hormonal replacement therapy who were not overweight, use of cyclic combined therapy, estrogen alone, and tibolone increased the incidence of endometrial cancer, whereas continuous combined therapy conferred no benefit on this group of women. However, in the obese subset of women (BMI > 30 kg/M2), both continuous and cyclic hormonal therapy significantly reduced the risk of endometrial cancer as compared with those women who never took hormone replacement therapy. Tibolone users had similar rates as compared with neverusers; there were too few women who were obese and taking estrogen alone to compare the risks.
VI. PROGNOSIS WITH AN ESTROGEN-ASSOCIATED ENDOMETRIAL CANCER Several studies have examined the impact of ERT and H R T on the prognosis of endometrial cancer (19-21). Uniformly these studies have shown that patients diagnosed with endometrial cancer while on or following exogenous
589
estrogen therapy have an improved prognosis compared with patients who do not have a history of prior estrogen therapy. One problem with many of these studies is that there is inadequate accounting for either the risk factors in the non-estrogen users or inclusions of known high-risk factors, including patients with type II cancers. Because type II cancers are known to be unassociated with estrogen and have a much poorer prognosis, a true comparison would exclude these patients. There are at least two possible explanations as to why other factors may account for the improved prognosis of endometrial cancers in patients on estrogens. There is also the issue of early diagnosis bias, which several authors have suggested is related to the improved prognosis of endometrial cancers among estrogen users. The first explanation has to do with the underlying background incidence of subclinical endometrial cancer in the population. As is true with men and prostate cancer, the subclinical undiagnosed incidence of endometrial cancer may be a multiple of the known incidence by a factor as great as threefold to fivefold. Several authors have postulated that one cause of the higher incidence of endometrial cancer in estrogen users can be explained by the induction of symptomatic bleeding by an otherwise asymptomatic occult cancer. This theory is supported by the reports of "estrogenassociated" cancers occurring within a short time of initiation of estrogen therapy. It is clear that a patient who is receiving estrogen replacement therapy, whether alone or in combination with a progestin, has access to medical care. It can be further assumed that patients under the care of a physician for menopausal symptoms will be advised to report any unusual bleeding or discharge; the further assumption is that any symptom of such a patient will be addressed more directly by the physician. The data have shown that patients who are diagnosed with an endometrial cancer either during or after receiving estrogen therapy, with or without progestogens, have predominately less aggressive tumors. There is some suggestion that conjugated estrogens are associated with higher grade and more deeply invasive tumors as compared with synthetic estrogens.
VII. ESTROGEN REPLACEMENT THERAPY AFTER TREATMENT FOR ENDOMETRIAL CANCER Given the epidemiologic data supporting the carcinogenic effect on the endometrium of both exogenous and endogenous estrogens, it has long been considered to be a contraindication to either continue or initiate estrogen replacement therapy for patients treated for endometrial cancer. The widely held assumption is that because the
590 majority of endometrial cancers have estrogen receptors, prescribing of exogenous estrogen replacement therapy could stimulate occult (possibly dormant) residual disease. Beginning in the 1980s, several authors challenged the conventional wisdom. Creasman reported on 221 patients with stage I endometrial cancer; 47 received estrogen replacement and 174 who did not (22). After controlling for known risk factors, the estimated distributions of time to recurrence for the two groups were significantly different, with the surprising finding that the estrogen-treated patients actually experienced a longer disease-free survival as compared with the women who did not receive estrogen replacement. Creasman and colleagues concluded that the history of endometrial cancer was not a contraindication to hormone replacement therapy (HRT) in patients with stage I disease (23). Chapman reported another retrospective review of estrogen replacement therapy after treatment for endometrial cancer; in this study, both stage I and II patients were included. Of the 123 patients reported, there were two recurrences among 62 estrogen-treated (3.2%) women compared with 6 of 61 (9.8%) patients who never received estrogen replacement therapy. Again the authors concluded that there was no evidence within the estrogen replacement therapy group to show that estrogen decreased the disease-flee interval or increased the risk for recurrence in early-stage endometrial cancer (24). The problems with these retrospective studies are well known and include the bias prescribing estrogen to patients with less aggressive lesions. The regimens prescribed were not standardized in either study. Further complicating the analysis is the long interval between surgical treatment of endometrial cancer and the initiation of estrogen replacement therapy. A prospective randomized, placebo-controlled study was initiated by the Gynecologic Oncology Group (GOG) in 1997 to attempt to answer whether estrogen replacement therapy after surgical treatment for stage I and II endometrial cancer would increase the risk of recurrence in this group of patients. All patients had to initiate ERT within 20 weeks of surgery, and the dose of therapy was standardized (0.625 mg of conjugated equine estrogen). The G O G had planned to accrue 2500 women; however, the study was closed in 2002 after enrollment of 1236 women. The trial closure was due to several concerns. First, there were a large number of enrollees with low-risk disease. Second, with the publication of the W H I study, which raised concerns regarding the safety of HRT, interest in the study dropped significantly. Despite these problems, this study remains the only prospective, randomized trial of estrogen therapy for this group of patients. Recurrence and survival data are similar and are presented in Table 42.2 (25). As the authors correctly note, the safety of exogenous estrogen use in patients treated for endometrial cancer may never be determined. This study
JOHN P. CURTIN
TABLE 42.2
Recurrence and Survival Treatment group Placebo ERT (n = 618) (n = 619)
Recurrence and survival Alive, NED Alive, with disease recurrence Total deaths Endometrial cancer MI/CHD Pulmonary embolism Other Unknown
No.
%
No.
%
591
95.6
583
94.3
8 19
1.3 3.1
9 26
1.5 4.2
4 4
0.6 0.6
5 3
0.8 0.5
0 6 5
0.0 1.0 0.8
2 9 7
0.3 1.5 1.1
NED, no evidence of disease; MI/CHD, myocardial infarction/coronary heart disease. Source: Res 25.
does provide the practitioner with additional information to individualize the treatment of women who may benefit from estrogen therapy.
VIII. CONCLUSIONS Estrogen replacement therapy alone is associated with an increase risk of endometrial cancer. The usage of estrogen alone as hormonal replacement therapy for women with a uterus is almost never indicated due to this association. These cancers are most commonly characterized as low-grade and low-stage tumors with an excellent prognosis. The greatest risk is for patients who do not have known risk factors, especially in regard to obesity; conversely, in an obese women the addition of estrogen alone does not appear to be an additive risk factor. The addition of progestogens, either as cycfic or as a continuous daily dose, reduces the risk of endometrial cancer; however, it appears to increase the overall rate of hormone-associated cancers due to the increase in breast cancer rates associated with progestin therapy. This places the patient and physician in the difficult position of balancing various risk factors. As the frequency of progestins increases, the risk of endometrial cancer decreases while the risk of breast cancer increases. The prescribing patterns of hormonal replacement therapy have shifted significantly since the publication of the W H I data, and new regimens are in the process of development (25). Despite the risks, there continues to be a substantial number of women who derive a benefit from hormonal replacement therap3~ which may outweigh any risk of endometrial cancer.
CHAPTER 42 Endometrial Cancer and Hormonal Replacement Therapy
References 1. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005;55:10-30. 2. Amant F, Moerman P, Neven P, et al. Endometrial cancer. Lancet 2005;366:491-505. 3. Pisani P, Parkin P, Tenet V, et al. Estimates of the worldwide mortality from the eighteen major cancers in 1985. Implications for prevention and projections of future burden. IntJ Cancer 1993;55:891-903. 4. Parkin DM, Bray F, Ferlay J, Pisani E Global cancer statistics, 2002. CA CancerJ Clin 2005;55:74-108. 5. National Cancer Institute. Surveillance, epidemiology and end point results (SEER). Stat database (1973-2002). http://www.seer.cancer.gov. 6. Hardiman P, Pillay OC, Atiomo W. Polycystic ovary syndrome and endometrial carcinoma. Lancet 2003;361:1810-1812. 7. Kaaks R, Lukanova A, Kurzer MS. Obesity, endogenous hormones and endometrial cancer risk: a synthetic review. Cancer Epidem Biomakers Prevent 2002;11:1531-1543. 8. Modesitt SC, van Nagell JR Jr. The impact of obesity on the incidence and treatment of gynecologic cancers: a review. Obstet Gynecol Surv 2005;60:683-692. 9. Deligdisch L, Holinka CF. Endometrial carcinoma: two diseases? Cancer Detect Prey 1987;76:2034-2040. 10. Lax SE Molecular genetic pathways in various types of endometrial carcinoma: from a phenotypical to a molecular-based classification. VirchowsArch 2004;444:213-223. 11. Resinger JI, Maxwell GL, Chandramouli GVR, et al. Microarray analysis reveals distinct gene expression profiles among different histologic types of endometrial cancer. CancerRes 2003;63:6-11. 12. Smith DC, Prentice R, Thompson DJ, et al. Association of exogenous estrogen and endometrial carcinoma. N Engl J Med 1975;293: 1164-1167. 13. Akhmedkhanov A, Zeleniuch-Jacquotte A, Toniolo E Role of exogenous and endogenous hormones in endometrial cancer: review of the evidence and research perspectives. Ann N Y Acad Sci 2001;943: 296-315. 14. Shields TS, Weiss NS, Voigt LF, Beresford SA. The additional risk of endometrial cancer associated with unopposed estrogen use in women with other risk factors. Epidemiology 1999;10:733- 738. 15. Weiderpass E, Baron JA, Adami HO, et al. Low-potency oestrogen and risk of endometrial cancer: a case-control study. Lancet 1999;353: 1824-1828.
591
16. Weiderpass E, Adami HO, Baron JA, et al. Risk of endometrial cancer following estrogen replacement with and without progestins. J Natl Cancer Inst 1999;91:1131-1137. 17. Grady D, Gebretsadik T, Kerlikowske K, et al. Hormonal replacement therapy and endometrial cancer: a meta-analysis. Obstet Gynecol 1995;85:304-324. 18. Reed SD, Voight LF, Beresford SAA, et al. Dose of progestin in postmenopausal-combined hormone therapy and risk of endometrial cancer. Am J Obstet Gyneco12004;191:1146-1151. 19. Beral V, Bull D, Reeves G. Million Women Study. Lancet 2005; 365:1543-1551. 20. Collins J, Donner A, Allen LH, Adams O. Oestrogen use and survival in endometrial cancer. Lancet 1980;2:961-964. 21. Chu J, Schweid AI, Weiss NS. Survival among women with endometrial cancer: a comparison of estrogen users and nonusers. Am J Obstet Gyneco11982;143:569-573. 22. Shapiro JA, Weiss NS, Beresford SAA, Voigt LE Menopausal hormone use and endometrial cancer by tumor grade and invasion. Epidemiology 1998;9:99-101. 23. Creasman WT, Henderson D, Hinshaw W, et al. Estrogen replacement therapy in the patient treated for endometrial cancer. Obstet Gynecol 1986;67:326-330. 24. Chapman JA, DiSaia PJ, Osann K, et al. Estrogen replacement in surgical stage I and II endometrial cancer survivors. Am J Obstet Gynecol 1996;175:1195- 2000. 25. Barakat RR, Bundy BN, Spirtos NM, Bell J, Mannel RS. Randomized double-blind trial of estrogen replacement therapy versus placebo in stage I or II endometrial cancer: a gynecologic oncology group study. J Clin Onco12006;24:587-592. 26. Hubacher D, Grimes DA. Noncontraceptive health benefits of intrauterine devices: a systematic review. Obstet Gynecol Surv 2002;57: 120-128. 27. Anderson GL, Judd HL, Kaunitz AM, et al. Effects of estrogen plus progestin on gynecologic cancers and associated diagnostic procedures. JAMA 2003:290:1739-1748. 28. Horn LC, Dietel M, Einenkel J. Hormone replacement therapy (HRT) and endometrial morphology under consideration of the different molecular pathways in endometrial carcinogenesis. Eur J Obstet GynecolReprod Bid 2005;122:4-12. 29. Pisani P, Parkin P, Tenet V, et al. Estimates of the worldwide mortality from the eighteen major cancers in 1985. Implications for prevention and projections of future burden. IntJ Cancer 1993;55:891-903.
This Page Intentionally Left Blank
- I A P T E R 4!
Ovarian Cancer and Its Detection R A D H I KA G O G O I
Section of Gynecologic Oncology, Department of Obstetrics and Gynecology, New York University School of Medicine, New York, NY 10016
B LANK
Section of Gynecologic Oncology, Department of Obstetrics and Gynecology, New York University School of Medicine, New York, NY 10016
S T E P H A N I E V.
DAVID A. FI S H M A N
Section of Gynecologic Oncology, New York University School of Medicine, New York, NY 10016
In the United States, ovarian cancer is responsible for more deaths than all other gynecologic malignancies combined (1). The lifetime risk of a woman developing ovarian cancer is approximately 1 in 70 (1). It is estimated that in 2005 there will be 22,220 new cases and 16,210 deaths from ovarian cancer (1). Unfortunately 70% to 75% of these new diagnoses continue to be stage III or IV epithelial ovarian cancer (EOC) with a 5-year survival rate of 28.5% for distant disease (1). In contrast, the 5-year survival rate for local disease approaches 94% (1). This suggests that strategies involving early detection of early-stage disease may significantly impact women's health care. The F I G O staging system for ovarian cancer is described in Table 43.1.
to have a protective effect (relative risk [RR] 0.6) against E O C (3-5). A further 10% to 15% decrease in risk is seen with each additional pregnancy (5). Oral contraceptive use has been shown to reduce risk by 50% to 70%, offering a protective effect after as little as 3 to 6 months of use (6,7), even for members within a recognized hereditary syndrome with ovarian cancer predisposition (7a). Their protective effect continues for up to 15 years after last use and is regardless of the formulation of the oral contraceptive (8-10). Depo-Provera does not provide the same benefits (11). Infertility (5,12,13) and ovulation stimulation medications (5,13) have been linked to an increased risk of ovarian cancer in some but not all studies. A hysterectomy with or without removal of the ovaries or a tubal ligation have been shown to decrease the risk of developing ovarian cancer, with a RR of 0.61 and 0.64, respectively (14-16). In contrast to the protective effects of the oral contraceptive, recent studies have shown an increase in ovarian cancer risk with the use of exogenous hormone replacement therapy (HRT). Lacey et al. (17) reported on menopausal H R T and ovarian cancer risk and found a significant increase in RR (1.8) after more than 10 years of use of estrogen-only replacement therapy (ERT). No increase in risk was observed
I. RISK A S S E S S M E N T A. Modifiable Risk Factors It is estimated that approximately 90% of EOCs occur sporadically with no prior family history of disease (2). Pregnancies regardless of outcome have consistently been shown TREATMENT OF THE POSTMENOPAUSAL W O M A N
593
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
594
GOGOI ET AL.
TABLE 43.1
H G O Staging System for Carcinoma of the Ovary
STAGE
III
Growth limited to the ovaries Growth limited to one ovary. No ascites containing malignant cells. No tumor on external surface. Growth limited to both ovaries. No ascites containing malignant cells. No tumor on external surface. Tumor either stage IA or IB but with tumor on the surface of the ovary or ascites containing malignant cells or positive peritoneal washings. Growth involving one or both ovaries with pelvic extension Extension to uterus and/or tubes Extension to other pelvic tissues Tumor either stage IIA or IIB but with tumor on the surface of the ovary or ascites containing malignant cells or positive peritoneal washings. Tumor involving one or both ovaries with peritoneal implants, extension to the small bowel or omentum and/or
IIIA IIIB IIIC IV
Tumor limited to the pelvis with microscopic seeding of the peritoneum. Lymph nodes negative. Tumor limited to the pelvis with peritoneal implants < 2 cm. Lymph nodes negative. Peritoneal implants > 2 cm and/or positive retroperitoneal or inguinal nodes. Growth involving one or both ovaries with distant metastasis. Pleural effusion containing malignant cells. Parenchymal
I
IA IB
IC II IIA IIB IIC
positive retroperitoneal or inguinal nodes
liver metastasis
after any duration of use of estrogen and progestin combination H R T (17). Similar results were reported by Rodriguez et al. (18), who additionally showed an increased risk in former users of ERT for more than 10 years (RR 1.55). Benign gynecologic diseases such as endometriosis (19,20), with an RR of 1.32, have also been linked to a higher risk of developing ovarian cancer, most commonly clear cell and endometrioid types (21). At present, a family history of ovarian cancer is the strongest risk factor for the development of EOC. The RR of E O C for a woman with one first-degree relative with ovarian cancer is 3.1; with two or more affected relatives, it is 12 (22).
B. H e r i t a b l e F a c t o r s About 10% of EOCs are hereditary, with the presently recognized BRCA1 or BRCA2 mutations accounting for 90% of cases and hereditary nonpolyposis colorectal syndrome, or Lynch II syndrome (HNPCC), accounting for the remaining 10% (2,23). The frequency of the BRCA mutation significantly varies among various ethnic groups, with approximately 2% of Ashkenazi Jewish women having one of three founder mutations (185delAG and 5382insC in BRCA1 and 6174delT in BRCA2) (24), as compared with 0.2% of the North American population (23). Of women who develop EOC, 30% to 40% of Ashkenazi Jewish women and 10% to 13% of the North American population will carry the BRCA mutation. BRCA1 and BRCA2 are located on chromosomes 17q21 and 13q12-13 respectively and code large tumor suppressor genes involved in repair of D N A damage, transcriptional regulation of gene expression,
and inhibition of cell proliferation (2). The risk of developing breast cancer with either BRCA1 or BRCA2 mutations ranges from 50% to 85%. The majority of these breast cancers have been shown to be E R / P R and Her 2 neu negative (24a,24b). Women with a BRCA1 mutation have a 30% to 40% lifetime risk of developing E O C and with BRCA2 a 15% to 25% lifetime risk. Mutations within the ovarian cancer cluster region of BRCA2 increase the risk of E O C to 20% and mutations outside this region to 11% (25). Hereditary EOCs are predominantly serous carcinomas, high grade, late stage, nonmucinous or borderline, but have longer overall survival and recurrence-free intervals after chemotherapy (26). The improved response to chemotherapy may be due to the increased sensitivity of BRCAdeficient cells to DNA-damaging agents such as platinum (26). In addition to the increased risk in breast cancer and EOC, BRCA1 mutation carriers have an increased risk of prostate and colon cancer and in those BRCA2 carriers an increased risk of prostate, pancreatic, gallbladder, and
TABLE 43.2. Lifetime Risk of Breast and Ovarian Cancer in BRC/I1 and BRCA2 Mutation Carriers
BRCA1 BRCA2
Breast cancer
Ovarian cancer
50- 85% 50- 85%
30-40% 15-25%
Other malignancies Prostate, colon Pancreatic, gallbladder, melanoma, gastric, male breast cancer
595
CHAPTER 43 Ovarian Cancer and Its Detection gastric cancer, malignant melanoma, and male breast cancer (2). EOC associated with a BRCA1 mutation generally presents at an average age of 51 years compared with BRCA2 mutations and sporadic EOC, which presents later at 57.5 years. With these findings, in 1995 a consensus panel of the National Institutes of Health stated, "The risk of ovarian cancer in women from families with hereditary ovarian cancer syndromes is sufficiently high to recommend prophylactic oophorectomy in these women at 35 years of age or after childbearing is completed" (27). In addition to an increased risk of breast and ovarian cancer, BRCA1 and BRCA2 carriers at time of bilateral salpingo-oophorectomy (BSO) were found to have a higher incidence of fallopian tube cancer (27a,27b). Thus, it is suggested that the entire ovary and fallopian tube be removed, serially sectioned, and pathologically examined (28). Additionally a thorough inspection of the abdomen and pelvis and peritoneal washings are advocated, and directed peritoneal and omental biopsies are our practice whenever any suspicious anomalies are visualized (29). BRCA1 and BRCA2 carriers undergoing a riskreducing BSO decrease their risk of subsequent ovarian and fallopian tube cancer by 99% and breast cancer by 50% to 68%. The incidence of primary peritoneal cancer may be unaffected and has been reported in up to 3% of women after BSO (30-32). Three hereditary syndromes have also been identified as having an increased risk of EOC. They are the Lynch syndrome II associated with HNPCC, hereditary site-specific ovarian cancer, and hereditary breast and ovarian cancer (33).
1. WHO SHOULD BE TESTED?
Individuals with a family history suggestive of an inherited malignancy syndrome or those with a living affected relative should be considered for testing. These include the following individuals: those with a diagnosis of breast or EOC before age 30, those with a significant family history of breast or EOC (one or more affected first-degree relatives), those with a blood relative with a known mutation in BRCA1 or BRCA2, and Ashkenazi women who have breast cancer or EOC or a family history of either. Because approximately 40% of Jewish women with EOC treated at our institutions had a BRCA1 or BRCA2 mutation, we now offer all affected Jewish women formal genetic counseling by board-certified genetic counselors and geneticists and testing when indicated by pedigree analysis. The benefits of genetic testing for the BRC,4 genes include the identification of those individuals at increased risk for the development of breast cancer or EOC, individualizing surveillance measures to enhance early detection of cancer, offering prophylactic surgery such as mastectomy or bilateral salpingooophorectomy (BSO), offering chemopreventatives (such as oral contraceptives or Tamoxifen), enrolling in clinical trials for those at increased risk, and knowledge of the potential
for passing the mutation to future generations (2,34). Genetic testing for BRCA1 and BRCA2 mutations may involve sequencing of the entire coding region of the two genes in families who have not had previous genetic testing or "multisite analysis" with sequencing of specific regions known to have mutated in an affected family member or the sequencing of three known founder mutations in the Ashkenazi Jewish population. Genetic testing for mutations in these genes also has potential risks: adverse psychologic effects, disruption of family dynamics, insurance or employer discrimination, and potential for revealing nonpaternity. Prior to genetic testing it is imperative to assess for whom such testing is appropriate, provide expert genetic counseling on the implications of genetic testing, and obtain individual consent.
2. WHAT TO DO WITH THE GENETIC TEST RESULTS?
The ethical, legal, and psychologic considerations of genetic testing for inherited cancer syndromes are substantial and thus justify the need for expert counseling prior to and appropriate care after the results are available. The American Society of Clinical Oncology (ASCO) and the Ethical, Legal, and Social Issues branch of the Human Genome Project emphasize the importance of expert counseling for genetic testing.
II. S C R E E N I N G FOR EOC To better understand the role of a screening test, a few basic concepts will be reviewed. Sensitivity is defined as the percentage of patients with the disease who are correctly identified by a positive test result. Specificity is the percentage of patients without the disease who are correctly identified by a negative test result. Thus, a test with 98% specificity would result in 50 procedures for every case of EOC detected on screening in postmenopausal women. Positive predictive value (PPV) is the percentage of patients with a positive test result who also have the disease. Negative predictive value is the percentage of patients with a negative test result who do not have the disease. An effective screening test for a relatively rare disease such as ovarian cancer requires a sensitivity greater than 75%, specificity greater than 99.6%, and positive predictive value of more than 10%, although lower specificity may be acceptable in those women appropriately identified to be within the high-risk population. Unfortunately there are no effective screening strategies for the detection of early-stage ovarian cancer. Commonly used modalities include the pelvic exam, transvaginal ultrasound (TVUS), serum marker CA-125, or a combination thereof. Although a pelvic exam has an important role in the gynecologic care of women, Grover et al. (35) found that the bimanual exam has little value as a screening tool for EOC.
596 The TVUS remains the best diagnostic imaging modality to evaluate the adnexa. Ultrasound characteristics associated with an increased risk of ovarian malignancy include the presence of solid components, papillary excrescences, increased ovarian volume incorporated as part of the morphology index (MI) (36), and vascular analysis using tools such as Doppler velocimetry. Although these characteristics may be helpful in determining risk of malignancy, numerous studies have found that TVUS is not effective as a sole screening tool for EOC in asymptomatic women. Van Nagell et al. (37) performed ultrasound on 14,469 asymptomatic women, of which 180 underwent surgery for a persistent ultrasound abnormalities. Seventeen patients were found to have EOC. Four patients developed cancer within 12 months of normal results on screening, and four patients developed cancer more than 12 months after a normal screening. As a screening tool, the positive predictive value of TVUS was 9.4%, with 11 benign tumors removed for every one malignancy detected. Excluding granulosa and borderline ovarian tumors, the PPV of TVUS as a routine screening tool falls below the cutoff of 10%. Fishman et al. (38) recently reported the results of TVUS screening in a cohort of 4526 high-risk women referred to the national ovarian cancer early detection program. High risk was defined by (1) one or more affected first-degree relatives with ovarian cancer; (2) a personal history of breast or ovarian cancer; (3) multiple first- and second-degree relatives with breast or ovarian cancer; (4) BRCA mutation carrier; or (5) membership in a recognized familial cancer syndrome. A total of 12,709 TVUS were performed, with 98 persistent adnexal masses identified. Forty-nine surgeries were performed, resulting in findings of 12 malignancies and 37 benign ovarian masses. The malignancies were classified as four primary peritoneal, four fallopian tube, two EOC, and two endometrial carcinomas. All EOC and fallopian tube malignancies were found at advanced stages (IIIA-IIIC), limiting the use of TVUS as a screening tool for the early detection of EOC. These studies confirm the results of multiple studies suggesting the limited utility of TVUS as a sole screening tool in asymptomatic high-risk women for the accurate detection of early-stage disease. Recent sonographic developments involving improved scanning with harmonics and pulse inversion and the use of contrast agents justify the hope that the sonographic depiction of tumor microvascularity will provide visualization of vascular changes associated with early-stage rather than advanced disease. A variety of tumors markers have been studied for EOC, the most common of which is the CA125. CA125 is a tumor antigen encoding MUC16 first found to be elevated in patients with advanced nonmucinous ovarian cancer (39). A serum level of less than 35 U/mL is often accepted as normal; however, values after hysterectomy and in postmenopausal women may be lower (2). Eighty-five percent of patients with nonmucinous EOC will have elevated levels of CA125. However, approximately 47% of women with stage
GOGOI ET AL.
1 EOC have elevated CA125 levels, compared with greater than 90% with advanced EOC. Complicating the picture is the fact that CA125 can be elevated in many nongynecologic conditions, including pancreatic, breast, colon, and lung cancers; pregnancy; and benign gynecologic conditions, including endometriosis, menstrual cycle, fibroids, salpingitis, and benign ovarian neoplasms. CA125 as a sole screening test in a low-risk population has a sensitivity of 100% and a specificity of 97%, but a positive predictive value of only 2.8%, thus limiting its utility as a single screening tool for EOC (2). However, a CA125 in combination with TVUS increased the PPV of detecting EOC to 26.8% (40). Jacobs et al. (41) further analyzed the value of serial CA125 screening and TVUS. Postmenopausal women were randomized to either three annual screens with CA125 and ultrasound or a control group. Women were followed for 7 years for development of EOC or fallopian tube cancers. The study identified 468 women with an elevated CA125, with 29 undergoing surgery. Six cancers were identified, with a positive predictive value of 20.7%. Thus, although CA125 alone seems to have little efficacy as a screening tool, the combination of CA125 and TVUS is being further investigated in larger randomized clinical trials. Currently the use of CA125 should be limited to surveillance of known EOC, fallopian tube or primary peritoneal malignancies, or as preoperative evaluation of a known pelvic mass. In addition, numerous other tumor markers, including CA19-9, CEA, M-CSF, and LPA, are being evaluated for their role along with CA125 as part of a panel of screening markers for EOC. Ongoing studies evaluating the role of combined screening approaches for EOC include the Prostate, Lung, Colorectal, and Ovarian cancer screening trial (PLCO) enrolling 74,000 women into a control or an intervention arm receiving baseline CA125 and TVUS followed by annual CA125 and every-3-year TVUS screening (41a). Preliminary findings report a PPV of 3.7% for an abnormal CA125, 1% for an abnormal TVUS, and 23.5% if both tests are abnormal. Another ongoing trial is the United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS), which is a randomized trial enrolling women in one of three arms: no screening, annual screen with CA125 followed by ultrasound, or annual ultrasound screening alone. The trial is due to close in 2010 with an estimated 200,000 women enrolling (2).
III. N E W SCREENING MODALITIES Better insights into the biologic basis of cancer and the tumor-host microenvironment have facilitated development of novel screening modalities for EOC. It has been shown that changes in the host proteome may be one of the earliest signs of tumor progression. Tumor cells interacting with the surrounding stroma, extracellular matrix, organ
597
CHAPTER 43 Ovarian Cancer and Its Detection
parenchyma, and vascular and immune cells create protein modifications that influence both protein function and response to treatment and may represent a biomarker record of the tumor-host interaction. Using the technique of surface-enhanced laser desorption ionization time of flight (SELDI-TOF), Petricoin et al. (42) analyzed the serum protein profiles of 50 healthy women and 50 women with ovarian cancer. A computer-generated algorithm was derived and subsequently tested on an independent group of 116 samples (50 ovarian cancer cases and 66 control subjects). The algorithm successfully identified all 50 cases of ovarian cancer and 63 of 66 control cases with a PPV of 94% compared with 35% for a CA125. Using similar proteomic technology, Zhang et al. (42a) identified three biomarkers (apolipoprotein A, transthyretin, and a cleavage fragment of inter-el-trypsin inhibitor) that as a panel had a higher sensitivity (74% versus 65%) and equal specificity to that of CA125 alone. Improvements in proteomic technology include the use of serum albumin as a "protein sink" to amplify low abundance and low-molecular-weight proteins, improving mass spectroscopy analysis of the serum proteome. Using sera from 110 patients (40 high-risk patients without cancer, 30 with stage 1, and 40 with stage III), more than 700 peptides were identified from proteins not previously reported to circulate in blood from patients with ovarian cancer (42b). It is hoped that these types of approaches will increase our knowledge of circulating proteins in ovarian cancer patients, allowing further validation of their use as potential biomarkers for EOC. Advances in EOC biology have redefined critical components of the metastatic process~specifically, how cellular adhesion, migration, extraceUular matrix degradation, invasion, proliferation, apoptosis, and neovascularization are all significantly influenced by biologically active lipids such as lysophosphatidic acid (LPA), lysophosphatidylinositol (LPI), and lysophosphatidylcholine (LPC). To study the role of LPA as a screening marker for EOC, Xu et al. (43) evaluated LPA levels in patients with ovarian cancer, other gynecologic cancers, benign gynecologic diseases, breast cancers, and leukemias and healthy control subjects. Using a cutoff of 1.31aM, Xu reported elevated plasma LPA levels in 9 of 10 patients with stage I; 24 of 24 patients with stage II, III, and IV ovarian cancer; and 14 of 14 patients with recurrent ovarian cancer. The majority of patients with other gynecologic cancers (33 of 36) also showed elevated LPA levels compared with control subjects. In particular, LPA levels were elevated in 6 of 6 patients with cervical cancer. In contrast, elevated plasma LPA levels were detected in only a minority of healthy control subjects (5 of 48) and patients with benign gynecologic diseases (4 of 18) and in none of the patients with breast cancer (0 of 11) or leukemia (0 of 5). CA125 data from the same group of ovarian cancer patients revealed a substantially lower rate of detection, compared with LPA, especially in stage I. These results suggested that LPA represents a promising marker for the early detection of
ovarian cancer. Other factors showing promise as biomarkers for EOC detection include human kallikrein 6 (hK6) (44,45) and epidermal growth factor receptor (EGFR) (46). Although the incidence of EOC continues to rise in the United States, considerable progress has been made in understanding both the biology and heredity of the disease, advancing screening tools and new methods of treatment. We believe that future progress in early EOC detection lies in understanding the impact of the tumor-host microenvironment and the role of the host in allowing tumor progression. The use and refinement of serum proteomic technology allows new insights into the subtle changes in the proteome that may occur at the onset of tumor progression. We hope that the new knowledge gained through these technologies has continued use for the accurate detection of early stage EOC, targeted therapies, and an overall improvement in women's health care.
References 1. American Cancer Society Facts and Figures 2005. 2. Hoskins WJ. Principlesandpractice ofgynecologic oncology, 4th ed. 2005, Philadelphia: Lippincott Williams & Wilkins, 2005:xix. 3. Gwinn ML, Lee NC, Rhodes PH, Layde PM, Rubin GL. Pregnancy, breast feeding, and oral contraceptives and the risk of epithelial ovarian cancer. J Clin Epidemio11990;43:559-568. 4. Risch HA, Marrett LD, Howe GR. Parity, contraception, infertility, and the risk of epithelial ovarian cancer. Am J Epidemiol 1994;140: 585-597. 5. Whittemore AS, Harris R, Itnyre J. Characteristics relating to ovarian cancer risk: collaborative analysis of 12 US case-control studies. II. Invasive epithelial ovarian cancers in white women. Collaborative Ovarian Cancer Group. Am J Epidemio11992;136:1184-1203. 6. The reduction in risk of ovarian cancer associated with oral-contraceptive use. The Cancer and Steroid Hormone Study of the Centers for Disease Control and the National Institute of Child Health and Human Development. N EnglJ Med 1987;316:650-655. 7. Epithelial ovarian cancer and combined oral contraceptives. The W H O Collaborative Study of Neoplasia and Steroid Contraceptives.
IntJ Epidemio11989;18:538-545. 7a. Whittemore AS, Balise RR, Pharoah PD, Dicioccio RA, et al. Oral contraceptive use and ovarian cancer risk among carriers of BRCA1 or BRCA2 mutations. BrJ Cancer2004;91:1911-1915. 8. Ness RB, Grisso JA, Klapper J, et al. Risk of ovarian cancer in relation to estrogen and progestin dose and use characteristics of oral contraceptives. SHARE Study Group. Steroid Hormones and Reproductions. Am J Epidemio12000;152:233 - 241. 9. Rosenblatt KA, Thomas DB, Noonan EA. High-dose and low-dose combined oral contraceptives: protection against epithelial ovarian cancer and the length of the protective effect. The W H O Collaborative Study of Neoplasia and Steroid Contraceptives. EurJ Cancer1992;28A: 1872-1876. 10. Royar J, Becher H, Chang-Claude J. Low-dose oral contraceptives: protective effect on ovarian cancer risk. IntJ Cancer2001;95:370-374. 11. Depot-medroxyprogesterone acetate (DMPA) and risk of epithelial ovarian cancer. The W H O Collaborative Study of Neoplasia and Steroid Contraceptives. IntJ Cancer 1991;49:191-195. 12. Whittemore AS, Wu ML, Paffenbarger RS, et al. Epithelial ovarian cancer and the ability to conceive. CancerRes 1989;49:4047-4052.
598 13. Rossing MA, Daling JR, Weiss NS, Moore DE, Self SG. Ovarian tumors in a cohort of infertile women. N Engl J Med 1994;331: 771-776. 14. Kreiger N, Sloan M, Cotterchio M, Parsons P. Surgical procedures associated with risk of ovarian cancer. IntJEpidemio11997;26:710- 715. 15. Hankinson SE, Hunter DJ, Colditz GA, et al., Tubal ligation, hysterectomy, and risk of ovarian cancer. A prospective study. JAMA 1993;270:2813-2818. 16. Green A, Purdie D, Bain C, et al. Tubal sterilisation, hysterectomy and decreased risk of ovarian cancer. Survey of Women's Health Study Group. IntJ Cancer 1997;71:948-951. 17. Lacey JV, Mink PJ, Lubin JH, et al. Menopausal hormone replacement therapy and risk of ovarian cancer. JAMA 2002;288:334-341. 18. Rodriguez C, Patel AV, Calle EE. Estrogen replacement therapy and ovarian cancer mortality in a large prospective study of US women. JAMA 2001;285:1460-1465. 19. Ness RB. Endometriosis and ovarian cancer: thoughts on shared pathophysiology. Am J Obstet Gyneco12003;189:280-294. 20. Modugno F, Ness RB, Allen GO, et al. Oral contraceptive use, reproductive history, and risk of epithelial ovarian cancer in women with and without endometriosis. Am J Obstet Gyneco12004;191:733-740. 21. Van Gorp T, Amant F, Neven P, Vergote I, Moerman P. Endometriosis and the development of malignant tumours of the pelvis. A review of literature. Best Pract Res Clin Obstet Gynaeco12004;18:349-371. 22. Stratton JF, Pharoah P, Smith SK, Easton D, Ponder BA. A systematic review and meta-analysis of family history and risk of ovarian cancer. BrJ Obstet Gynaeco11998;105:493-499. 23. Risch HA, McLaughlin JR, Cole DE, et al. Prevalence and penetrance of germline B RCA1 and B RCA2 mutations in a population series of 649 women with ovarian cancer. Am J Hum Genet 2001;68:700- 710. 24. Moslehi R, Chu W, Karlan B, et al. BRCA1 and BRCA2 mutation analysis of 208 Ashkenazi Jewish women with ovarian cancer.//m J Hum Genet 2000;66:1259-1272. 24a. Grushko TA, Blackwood MA, Schumm PL, et al. Molecularcytogenetic analysis ofHER-2/neu gene in BRCAl-associated breast cancers. CancerRes 2002;62:1481-1488 24b. Musolino A, Naldi N, Michiara M, Bella MA, et al. A breast cancer patient from Italy with germline mutations in both the BRCA1 and BRCA2 genes. Breast CancerRes Treat 2005;91:203-205. 25. Thompson D, Easton D. Variation in cancer risks, by mutation position, in BRCA2 mutation carriers. Am J Hum Genet2001 ;68:410- 419. 26. Boyd J, Sonoda Y, Federici MG, et al. Clinicopathologic features of BRCA-linked and sporadic ovarian cancer.JAM//2000;283:2260-2265. 27. NIH consensus conference. Ovarian cancer. Screening, treatment, and follow-up. NIH Consensus Development Panel on Ovarian Cancer. J./IM_/11995;273:491-497. 27a. Carcangiu ML, Peissel B, Pasini B, et al. Incidental carcinomas in prophylactic specimens in B RCA1 and B RCA2 germ-line mutation carriers, with emphasis on fallopian tube lesions: report of 6 cases and review of the literature. Am J Surg Patho12006;30:1222-1230. 27b. Finch A, Beiner M, Lubinski J, et al. Salpingo-oophorectomy and the risk of ovarian, fallopian tube, and peritoneal cancers in women with a BRCA1 or BRCA2 Mutation.JAMd 2006;296:185-192. 28. Paley PJ, Swisher EM, Garcia RL, et al. Occult cancer of the fallopian tube in BRCA-1 germline mutation carriers at prophylactic oophorectomy: a case for recommending hysterectomy at surgical prophylaxis. Gynecol Onco12001;80:176 -180. 29. Powell CB, Kenley E, Chen LM, et al. Risk-reducing salpingooophorectomy in B RCA mutation carriers: role of serial sectioning in the detection of occult malignancy. J Clin Oncol 2005;23: 127-132.
GOGOI ET AL.
30. Kauff ND, Satagopan JM, Robson ME, et al. Risk-reducing salpingooophorectomy in women with a BRCA1 or BRCA2 mutation. NEngl JMed 2002;346:1609-1615. 31. Rebbeck TR, Lynch HT, Neuhausen SL, et al. Prophylactic oophorectomy in carriers of B RCA1 or B RCA2 mutations. N Engl J Med 2002;346:1616-1622. 32. Rutter JL, Wacholder S, Chetrit A, et al. Gynecologic surgeries and risk of ovarian cancer in women with B RCA1 and B RCA2 Ashkenazi founder mutations: an Israeli population-based case-control study. J Natl CancerInst 2003;95:1072-1078. 33. Pharoah PD, Ponder BA. The genetics of ovarian cancer. Best PractRes Clin Obstet Gynaeco12002;16:449-468. 34. Fishman DA, Bozorgi K. The scientific basis of early detection of epithelial ovarian cancer: the National Ovarian Cancer Early Detection Program (NOCEDP). CancerTreat Res 2002;107:3-28. 35. Grover SR, Q.uinn MA. Is there any value in bimanual pelvic examination as a screening test. MedJAust 1995;162:408-410. 36. DePriest PD, Shenson D, Fried A, et al. A morphology index based on sonographic findings in ovarian cancer. GynecolOnco11993;51:7-11. 37. Van Nagell JR Jr, DePriest PD, Reedy MB, et al. The efficacy of transvaginal sonographic screening in asymptomatic women at risk for ovarian cancer. GynecolOnco12000;77:350-356. 38. Fishman DA, Cohen L, Blank SV, et al. The role of ultrasound evaluation in the detection of early-stage epithelial ovarian cancer. Am J Obstet Gyneco12005;192:1214-1221; discussion 1221-1222. 39. Bast RC Jr, Klug TL, St John E, et al. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N EnglJ Med 1983;309:883- 887. 40. Jacobs I, Davies AP, Bridges J, et al. Prevalence screening for ovarian cancer in postmenopausal women by CA 125 measurement and ultrasonography. Br MedJ 1993;306:1030-1034. 41. Jacobs IJ, Skates SJ, MacDonald N, et al. Screening for ovarian cancer: a pilot randomised controlled trial. Lancet 1999;353:1207-1210. 41a. Buys SS, Partridge E, Greene MH, et al. Ovarian cancer screening in the Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening trial: findings from the initial screen of a randomized trial. Am J Obstet Gyneco12005;193:1630-1639. 42. Petricoin EF, Ardekani AM, Hitt BA, et al. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 2002;359:572-577. 42a. Zhang Z, Bast RC, Yu Y, Li J, et al. Three biomarkers identified from serum proteomic analysis for the detection of early stage ovarian cancer. Cancer Res 2004;64:5882- 5890. 42b. Lowenthal MS, Mehta AI, Frogale K, et al. Analysis of albuminassociated peptides and proteins from ovarian cancer patients. Clin Chem 2005;51:1933-1945. 43. Xu Y, Shen Z, Wiper DW, et al. Lysophosphatidic acid as a potential biomarker for ovarian and other gynecologic cancers. JAMA 1998;280: 719-723. 44. Ghosh MC, Grass L, Soosaipillai A, Sotiropoulou G, Diamandis EP. Human kallikrein 6 degrades extracellular matrix proteins and may enhance the metastatic potential of tumour cells. TumourBid 2004;25: 193-199. 45. Rosen DG, Wang L, Atkinson JN, et al. Potential markers that complement expression of CA125 in epithelial ovarian cancer. Gynecol Onco12005;99:267- 277. 46. Baron AT, Cora EM, Lafliy JM, et al. Soluble epidermal growth factor receptor (sEGFR/sErbB1) as a potential risk, screening, and diagnostic serum biomarker of epithelial ovarian cancer. CancerEpidemiolBiomarkers Prey 2003;12:103-113.
~HAPTER 4 z
Hormone Replacement Therapy in Menopause and Breast, Colorectal, and Lung Cancer: An Update CARLO LA V E C C H I A
Istituto di Ricerche Farmacologiche "Mario Negri," 20157 Milano, Italy; Istituto di Statistica Medica e Biometria, Universit/l degli Studi di Milano, 20133 Milano, Italy
that reanalysis, menopause HRT increased (discussed earlier) breast cancer risk by about 2.3% per year of use. The effect of steroid hormones is, however, thought to act on the later stages of the process of carcinogenesis (i.e., they are promoters). Consequently, the increased breast cancer risk associated with HRT declines within a few years after stopping use (2-5). An additional question left open by the collaborative reanalysis was the impact on breast cancer risk of the combination of estrogens and progestogens, a therapy effective in reducing the excess endometrial cancer risk associated with estrogen use alone (2). In the American Nurses' Health Study cohort (6) the relative risks (RRs) were 1.3 for estrogens alone, and 1.4 (95% confidence interval [CI] 1.2-1.7) for estrogens plus progestins. Recent studies are reviewed here, including observational (cohort and case-control studies) and randomized controlled trials. A case-control study in Sweden, involving 3345 women with breast cancer, found a trend of increasing risk with duration of different types of combined estrogen-progestin
Hormone replacement therapy (HRT) in menopause has been related to the risk of cancers of the breast, the female genital tract, the large bowel, and the lung. Breast, colorectal, and lung cancers, however, are the key issues on an individual risk and public health level, because no material endometrial cancer risk is related to combined estrogen-progestin treatment, and the possible association between HRT, cervical, and ovarian cancer remains undefined (1,2). This chapter, therefore, will review available knowledge on HRT and breast, colorectal, and lung cancers. Another analysis of the risk of breast cancer can be found in Chapter 40.
I. BREAST CANCER Most data on menopause, HRT, and breast cancer risk available until the mid-1990s were included in a collaborative reanalysis of individual data from 51 epidemiologic studies, based on more than 52,000 women with breast cancer and 108,000 without breast cancer (3,4). According to TREATMENT o r THE POSTMENOPAUSAL W O M A N
599
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
600 use (RR = 2.4 for women treated for at least 10 years) (7). A report on 46,355 participants followed for a mean of 10.2 years in the Breast Cancer Detection and Demonstration Project (8) showed that women who had used combined estrogens and progestins had an RR of 1.4 (95% CI 1.1-1.8) of developing breast cancer compared with neverusers. The risk from combined therapy was greater than that observed with unopposed estrogens (RR = 1.2; 95% CI 1.0-1.4). In that study, too, women who had used H R T in the past but had stopped did not have an increased risk. Likewise, a population-based case-control study of 1897 postmenopausal cases and 1637 postmenopausal population controls from Los Angeles County (9) found an RR of 1.06 (95% CI 0.97-1.15) for each 5 years of estrogen therapy, but of 1.24 (95% CI 1.07-1.45) for combined estrogenprogestin treatment. The excess risk for combined therapy was around 5% per year. In a cohort study of 29,508 women for Sweden, followed up between 1990 and 2001, and including 556 breast cancer cases, the RR was significantly elevated in women who had used continuous combined HRT, and nonsignificant excess risks were also observed among longterm users of combined sequential H R T (RR = 2.2), progestogen only (RR = 3.7), and estriol only (RR = 1.9). No excess risk was observed after 5 years or longer since stopping use (10). Likewise, another cohort study for southern Sweden, including 6586 women and 101 breast cancer cases, found an excess risk (RR = 3.3) for users of combined continuos H R T (11). In a multicentric case-control study of 1870 cases and 1953 controls from the United States, continuous combined H R T was associated with increasing breast cancer risk among users for 3 years or more (RR = 1.5) (12). Mammographic density was also related to combined HRT, more than to estrogens alone, in the Postmenopausal Estrogen/ Progestin Interventions (PEPI) trial (13). Likewise, in the Women's Health Study (14), a prospective investigation of 17,835 women randomized to aspirin and vitamin E (but not to HRT), the RR was 1.0 for ever use of estrogens only but 1.4 for estrogen/progestogen HRT. The RR was 1.8 for continuous progestogen administration. In a case-control study of 975 cases at 1007 controls from Washington State (15), women who had used estrogen replacement therapy (ERT) had no appreciable excess risk of breast cancer, even for 25 years of use or longer. Users of combined HRT, however, had an overall 1.7-fold (95% CI 1.3-2.2) increased risk of breast cancer, including a 2.7-fold increased risk of invasive lobular carcinoma, a 1.5-fold increased risk of ductal carcinoma, and a 2.0-fold increased risk of E R + / P R + breast cancers. The risk increases with increasing duration of use (15). This reflects the observation of a substantial rise in the incidence of lobular cancers in population-based series (16,17).
CARLO LA VECCHIA
In the prospective observational Million Women Study (18), including 9364 incidents of invasive breast cancer, current but not past H R T users were at increased risk (RR = 1.66). The association was stronger for users of estrogen-progestogen preparations (RR = 2.0) but was observed also for estrogens only (RR = 1.3) and tibolone (RR = 1.45). The RRs were increased for oral, transdermal, and implanted formulations. Likewise, in the Danish Nurses cohort study (19), based on 244 breast cancer cases, the RR was 2.0 for estrogens only, 2.7 for estrogen-progestogen, 4.3 for tibolone, and 4.2 for H R T with androgen-like progestins. The RR estimates were not heterogeneous across types of preparations. With reference to intervention studies, relevant information on cancer risk in users of combined H R T derives from the Women's Health Initiative ( W H I ) (20), a randomized controlled primary prevention trial including 8506 women aged 50 to 70 treated with combined H R T and 8102 untreated women. The combined treatment was discontinued in July 2002 when the U.S. National Institutes of Health issued a press release stating that the combination hormone medication used in the trial posed more risks than benefits. This was based on increased risks of invasive breast cancer, cardiovascular disease, stroke, and blood clots. An additional estrogen-only compared with placebo treatment group of the same trial was stopped on March 2, 2004, on the basis of an increased risk of stroke in treated women (21). For the combined H R T treatment, no difference in risk was observed during the first 4 years after starting treatment, but an excess risk of breast cancer was evident thereafter. At a 7-year follow-up, 166 breast cancer cases were registered in the H R T group versus 124 in the placebo group, corresponding to an RR of 1.24 (95% CI 1.03-1.66). In the estrogen-only section of the W H I , based on 10,739 postmenopausal women ages 50 to 79 with prior hysterectomy, at 8-year follow-up, 94 invasive breast cancers were observed in the treated group versus 124 in the placebo group. The corresponding RR was 0.77 (95% CI 0.59-1.01) (22). Data from two other smaller randomized studies are available, one (Heart and Estrogen-Progestin Replacement Study [HERS]) with combined estrogen-progestogen therapy (23), and one (Women's Estrogen for Stroke Trial [WEST]) with estrogen only (24). In a combined analysis of the three randomized trials (25), 205 cases of breast cancer were observed in the treated groups, versus 154 in the placebo one, corresponding to a pooled RR of 1.27 (95% CI 1.03-1.56). Thus, the apparent excess breast cancer risk in H R T users in all four available randomized clinical trials (RCT) (RR = 1.12) is attributable to combined estrogen-progestogen therapy only. Likewise, with relevance to the global index score in the W H I , including vascular disease, breast and colon
CHAPTER 44 Hormone Replacement Therapy in Menopause and Breast, Colorectal, and Lung Cancer: An Update cancer, hip fractures, and death, the estrogen-alone component (RR = 1.01; 95% CI 0.91-1.12) was more favorable than estrogen-progestogen (RR = 1.15; 95% CI 1.03-1.26) (20,22,26). Although H R T has been associated with an increased incidence of breast cancer, its use appeared to lead to improved prognosis in some observational studies. Thus, in the American Cancer Society Cancer Prevention Study II (27), breast cancer mortality did not increase with estrogen use overall, and no excess risk was observed for thin or heavy women. This effect may be due to increased breast cancer surveillance among H R T users. However, in the W H I , invasive breast cancers diagnosed in the estrogen plus progestogen group were similar in histology and grade but were larger and at a more advanced stage. Thus, the results from the W H I did not confirm the suggestion that breast cancers in women using H R T have a more favorable prognosis. Further, after 1 year, the percentage of women with abnormal mammograms was greater in the H R T group (28). Likewise, in the Million Women study, there was no consistent suggestion that prognostic indicators were more favorable among women who had ever used H R T (18). In conclusion, the evidence from observational epidemiologic studies (cohort and case-control), as well as from randomized clinical trials, indicates that the risk of breast cancer (mainly lobular cancer) is elevated among women using (combined) HRT, increases with longer duration of use, is reduced after cessation of use, and levels off after stopping use. Prognosis does not appear to be more favorable in women who had used HRT.
II. COLORECTAL CANCER Colorectal cancer is the most common cancer in nonsmokers of both sexes combined and the second leading cause of cancer death for both sexes combined in western countries (29,30). A male predominance in colorectal cancer mortality was found in most countries of the world, except a few Latin American ones, such as Cuba or Colombia (31). During the 1980s and early 1990s, mortality trends from colorectal cancer have been more favorable in women than in men in Europe (32). This may be due to healthier dietary and lifestyle patterns in women, but also to a potential protective effect of exposure to exogenous female hormones (i.e., H R T and perhaps oral contraceptives) (33-35). However, data on reproductive and menstrual correlates of colorectal cancer risk are inconclusive. Moderate inverse associations with parity and oral contraceptive use have been reported, but a possible favorable role of later age at menopause is unclear (36-42). In the Framingham study (43), higher bone mass, which is likely related to greater
601
cumulative estrogen exposure, was associated to a reduced risk of colorectal cancer. Some reviews and meta-analyses, including data from several observational studies until 1998 (1,33,44-47), estimated an approximately 20% reduced risk of colorectal cancer among H R T users. Several biologic hypotheses supporting such an inverse association were postulated. Female hormones may confer a protection against colorectal cancer as a result of changes in bile synthesis, which may lead to reduced concentration of bile acids in the colon (48,49). Estrogens inhibit the growth of colon cancer cells in vitro, and estrogen receptors have been identified in normal and neoplastic colon epithelial cells. The estrogen receptor (ER) gene may play a tumor suppressor role, because the hypermethylation of the promotor region of the ER gene results in a reduced expression and deregulated growth in colonic mucosa, and has been associated with apoptosis (50). Estrogens have been associated with reduced risk of microsatellite instability positive colon cancer (51). Estrogens may also reduce serum levels of insulin-like growth factor-1 (IGF-1), a mitogen that may play an important role in the pathophysiology of colorectal and other cancers (46,52). Between 1980 and 2000, use of combined H R T had increased among postmenopausal women in western countries: An estimated 20 million women worldwide were using H R T in the early 2000s (4,25). With the publication of several trials of H R T (particularly the Women's Health Initiative) showing excess vascular mortality among users, the prevalence of H R T use has substantially declined over the last few years in the United States (53,54) and probably in Europe as well. It is therefore useful to update the results of observational studies and randomized clinical studies on H R T and colorectal cancer published over the last few years. At least nine cohort studies (Table 44.1) reported information on H R T use and colorectal cancer risk, including a total of more than 2500 cases (2). Most studies showed relative risks around or below unity. A significant inverse association was found in two cohort investigations, including the largest one, focusing on fatal colon cancers. Two studies also suggested that H R T use may improve shortterm survival after a diagnosis of colon cancer (55,56). Table 44.2 considers case-control investigations. O f 14 case-control studies for a total of more than 6000 cases, 5 reported 20% to 40% risk reductions among ever-users of HRT. Two additional investigations showed moderate, nonsignificant inverse associations. Studies showing an inverse association between H R T and colorectal cancer were among the largest ones. Differences in RRs by duration of H R T use and anatomic subsite were not consistent, but the apparent protection tended to be stronger among recent users. Therefore, available observational
USA, Nurses' Health Study
Iowa, USA
USA, Cancer Prevention Study II
Chute et al., 1991 (84) and Grodstein et al., 1998 (62)
Bostick et al., 1994 (85) and Folsom et al., 1995 (86) Calle et al., 1995 (63)
Finland
Pukkala et al., 2001 (90)
7,701 (14.5 years) 249 66
59,002 (14 years) 47O
Long cycle HRT Monthly cycle HRT
Unopp. HRT Opposed HRT Any HRT
Former users Current users
Current users Past users
Opposed HRT Any type
Estriol
HRT
0.81 (0.63-1.04)
0.99 (0.79-1.2)
1.0 (0.7-1.5)
0.8 (0.7-1.1)
(0.5 -0.8)
0.7
1.00 (ns)
Colon-rectum
0.85 (0.63-1.1) 0.67 (0.34-1.2)
1.1 (0.7-1.5) 1.4 (0.7-2.5) 1.1 (0.81-1.6) 0.70 + (0.45-1.09)
1.3 (0.9-1.9)
0.8 (0.6-1.1) 0.7 (0.5-1.1) 0.7 (0.6-0.8)
0.9 (0.7-1.1) 1.0 (0.8-1.3) 0.6 (0.4-1.0) 0.9 (0.7-1.2) 0.6 (0.5-0.9) 0.9 (0.7-1.1)
Colon
1.1 (0.59-1.9) 0.52 + (0.21-1.31)
1.2 (0.7-2.3)
0.6 (0.3-1.2)
0.9 (0.7-1.2) 0.8 (0.5-1.2) 0.8 (0.4-1.3) 0.9 (0.7-1.1) 0.7 (0.4-1.1) 0.8 (0.5-1.2)
Rectum
Relative risk, RR (95% confidence interval) (ever- vs. never-users)
RR = 0.75 for -> 5-year use
Significant trend (RR for > 11 year use = 0.5; 95% CI 0.4-0.8) RR = 0.7 (0.2-2.6) for -> 5-year use No effect
Inverse trend (RR = 0.31 for -<5-year use)
No effect (RR ->5-year use 0.7; 95% CI 0.5-1.0)
No effect (RR -> 8 year use = 1.02; 95% CI 0.6-1.8) No effect
Duration of use
RR for recent use = 0.66 (0.44-0.98)
RR for recent use = 0.78 (0.55,1.1)
Stronger effect among current users (RR = 0.5, 0.4-0.8) Not shown
No effect
0.8-~.2)
No risk reduction after 5 years of discontinuation (RR = 0.9;
Not shown
Not shown
Recency of use
Cohort Studies of H o r m o n e Replacement Therapy and Colorectal Cancer
BCDDP, Breast Cancer Detection Demonstration Project; BMI, body mass index; ns, not significant; W/H, waist/hip ratio; OC, oral contraceptives; +, recent users (-<1 year).
USA, Leisure World Cohort
Paganini-Hill, 1999 (89)
USA, BCDDP
Canada
422,373 (7 years) 897 deaths 32,973 (14 years) 230 33,779 (7.7 years) 313
Sweden
Adami et al., 1989 (74) and Persson et al., 1996 (58)
Risch and Howe, 1995 (87) Troisi et al., 1997 (88)
41,837 (6 years) 293
California, USA
Wu et al., 1987 (83)
7345 (4 years) 68 22,597 (13 years) 233 62 deaths
Country
Reference
Population, (follow-up), no. cancers
TABLE 44.1
age
Significant trend with recency of use
Age
Age (but unaltered by additional allowance for education, BMI, parity and OC use)
Record linkage study
Age
Age, BMI, parity, menopause, OC, diet, exercise
Age, BMI, W / H ratio, alcohol, exercise, and medical history
Age, BMI, OC use, cancer family history, diet, alcohol, smoking, and age at menopause
RR for colon mortality = 0.6; 0.4-0.9
Age
Age
Adjustment and comments
CHAPTER 44 Hormone Replacement Therapy in Menopause and Breast, Colorectal, and Lung Cancer: An Update studies support an inverse association between H R T and colorectal cancer incidence, but prevention and surveillance bias cannot be ruled out (33). Very few studies have enabled separate evaluation of unopposed estrogens from combined HRT, because most studies included few subjects exposed only to unopposed estrogens. Of one case-control and three cohort investigations, two (57,58) suggested an inverse association of unopposed estrogens with cancer of the colon, similar to what has been observed for H R T of any type. Differences in RRs by anatomic subsite were not consistent, but the data for rectal cancer are scantier than those for colon cancer. Route of administration was considered in a case-control study of 1197 cases of colorectal cancer from Saskatchewan, Canada. The RR for 3 years or more of use was 0.75 for oral estrogens and 0.33 for transdermal estrogens (59). The authors suggested that the reduction is colorectal cancer may be greater for transdermal estrogen cancers, but the results were not significantly heterogeneous. A case-control study, including 608 cases and 1922 control subjects for the Kaiser Permanente Medical Case Program of Northern California, suggested an interaction between 5,10-methylenetetrahydrofolate (MTHFR) reductase polymorphisms and H R T use on colorectal cancer risk (60). The RRs for H R T users, however, were below unity in various M T H F R genotypes, and the results were not heterogeneous. Previous H R T use was associated with a 41% reduced mortality in a cohort study of 699 women ages 50 to 69 from Seattle (56). This finding needs confirmation because of the small numbers of users (14.7%). With reference to reanalyses of observational data, a meta-analysis of 20 studies of colorectal cancer published up to the end of 1996 (45) found an overall RR for ever-users of H R T of 0.85 (95% CI 0.7-0.9). The protection was greater for current or recent users (RR 0.69, 95% CI 0.5-0.9) and for users of more than 5 years (RR 0.73, 95% CI 0.5-1.0). The inverse relation between H R T and colorectal cancer risk tends to emerge within a few years after first exposure (61,62) and seems to level off 5 to 10 years after cessation. The apparent protection increases with duration of use in some (61,63) but not all (57,62) studies. Such a pattern is compatible with a late-stage, promoting effect of H R T on colorectal carcinogenesis (64). O f the several studies on precursors for colorectal cancer (not systematically reviewed here), a large prospective investigation (62) found a decreased risk for large colorectal adenomas but no association for small adenomas. O f concern is the possibility that women may discontinue H R T when symptoms of disease develop (65), leaving selectively healthy women in the category of current users. However, no difference in risk was found between current
603
and recent users (i.e., those who had stopped H R T in the previous 5 years) (62). However, H R T users may differ from nonusers in ways that selectively influence colorectal cancer risk (the so-called healthy user effect) (66). Postmenopausal women treated with H R T tend in fact to be of higher social class and more educated (66-70). This selection may imply a healthier lifestyle (e.g., more frequent consumption of vegetables, higher levels of leisure time physical activity, and lower prevalence of overweight). In addition, long-term H R T users are, by definition, compliant, and this is, per se, a favorable health indicator (66). With reference to randomized clinical trials, in the W H I study, after 7-years follow-up, 45 cases of colorectal cancer were observed in the combined H R T group versus 67 in the placebo one, corresponding to an RR of 0.63 (95% CI 0.41-0.92) (20). A difference in risk between the H R T group and the placebo one became apparent 4 years after starting treatment. Such a time-risk relation gives support to the existence of a real association. A combined reanalysis of the W H I with the HERS data included 56 cases in the combined H R T treatment and 83 cases in the placebo group (pooled RR = 0.64; 95% CI 0.45-0.92) (25). In the HERS, however, the RR of colorectal cancer increased from 0.71 during the first 4.1 years of randomization (HERS-I) to 1.01 in the subsequent openlabel observational follow-up (23). Thus, with reference to H R T and colorectal cancer risk, the recent findings from randomized trials are in broad agreement with those of observational (cohort and case-control) studies and therefore provide conclusive evidence that H R T has a favorable effect on colorectal cancer incidence (71). However, additional data are needed to further quantify any potentially favorable effect of H R T not only on colorectal cancer incidence but also on mortality. In fact, a decrease in incidence from colorectal cancer among H R T users could appreciably affect the balance of risks and benefits associated with the use of HRT, but data on mortality are less convincing. The issue of H R T and colorectal cancer mortality appears less promising. In fact, at the 7-year follow-up of the W H I , 43 invasive colorectal cancer cases were observed in the combined H R T group versus 72 in the placebo one (RR = 0.56; 95% CI 0.38-0.81) (72). However, cancers diagnosed in the H R T group were more advanced and had a greater number of affected lymph nodes. It is unclear whether such an unfavorable clinical pattern is due to a real effect of H R T on colorectal cancer or to delayed diagnosis of colorectal cancer in postmenopausal women using HRT. Furthermore, the estrogen-only arm of the W H I in hysterectomized women did not show any difference at the 8-year follow-up between treated women (n = 61) and control subjects (n = 68) (22).
TABLE 44.2
Case-Control Studies of Hormone Replacement Therapy and Colorectal Cancer Relative risk (95%
Reference
Country
Cases:Controls (type of controls)
Weiss et al., 1981 (91)
Washington, USA
143:707 (population)
Potter and McMichael, 1983 (92) Davis et al., 1989 (93)
Adelaide, Australia
155:311 (population)
Canada
720:349 (cancer patients)
Furner et al., 1989 (94) Negri et al., 1989 (36); Fernandez et al., 1996 (95); Talamini et al., 1998 9(40); Fernandez et al., 1998(61) Peters et al., 1990 (96)
Chicago, USA Italy
90:208 (spouses) 1536:3110 (hospital)
Los Angeles, USA
327:327 (neighbors)
Colorectal --- 5-year use: 1.1 (0.7-1.9) -> 6-year use: 1.0 (0.6-1.6)
Current users: Former users:
0.5 (0.3-0.9) 0.6 (0.4-0.8)
<5 years 5-14 years
Wu-Williams et al., 1991 (97)
North America and China
189:494 (neighbors)
->15 years North America
206:618 (neighbors)
China
Gerhardsson de Verdier and London, 1992 (98)
Sweden
299:276 (population)
Jacobs et al., 1994 (38)
Seattle, USA
148:138 (population)
Newcomb and Storer, 1995 (57)
Wisconsin, USA
694:1622 (population)
1.5 (0.8-2.7) 1.1 (0.7-1.9)
1.3 (0.9-2.0) 1.1 (0.6-1.8) 1.1 (0.6-1.9)
Unopposed HRT Opposed HRT Any HRT (recent use)
Kampman et al., 1997 (99)
USA, KPMC
815:1019 (KPMC members)
Yood et al., 1998 (100)
Detroit, USA
60:143 (HMO members)
Jacobs et al., 1999 (101)
Seattle, USA
341:1679 HMO
Prihartono et al., 2000 (102)
Massachusetts, USA
404:404 population
Csizmadi et al., 2004 (59)
Saskatchewan, Canada
1197:4669 population
Current use Past use
0.3 (0.1-1.0) 0.4 (0.1-1.4) -1.0
Oral Transdermal
0.82 (0.70-0.97) 0.40 (0.39-0.9)
BMI, body mass index; HMO, health maintenance organization; KPMC, Kaiser Permanente Medical Care; OC, oral contraceptives.
confidence interval) (ever- vs. never-users) Colon
Rectum
0.8 (0.4-1.5)
--
0.6 (o.5-o.9)
Duration of use
Recency of use
Adjustment and comments
No trend
Not shown
Age
1.5 (0.8- 3.0)
0.2 (o.o-o.8)
0.5 (0.3-0.7)
Not shown
Not shown
No trend Significant (RR for -> 2-year use = 0.5; 0.3-0.8)
Not shown RR -> 10 years since last use: 0.5; 0.3-1.0
No effect
Not shown
Cancer family history, parity, menopause, exercise, fat, alcohol, and calcium intake
Unadjusted (but unaltered by exercise, saturated fat intake, and years in the U.S.) Artificial menopause was a risk factor in China Age Hormone use included both HRT and OC, but mostly HRT Age, vitamin intake and hysterectomy. Greater protection in multiparous women Age, alcohol, BMI, cancer family history, and sigmoidoscopy
2.1 p = 0.14 2.9 p = 0.01
0.5 p = 0.23 p = 0.56
Not shown Mostly shortduration use
Not shown
0.6 (0.4-1.0)
0.7 (0.4-1.3)
No trend
Not shown
Significant trend (RR -> 5-year use = 0.5; 0.2-0.9) Significant trend (p = 0.002)
RR in current users = 0.5; 0.3-1.0
0.6 (0.4-1.0)
0.5 (0.3-0.9) 0.5 (0.3-1.1) o.7 (0.6-0.9) (recent use)
0.90 (0.46-1.76) 1.1 (0.5-2.5) 1.2 (0.8-1.6) (recent use)
0.8 (0.7-1.0)
0.8 (0.5-1.2)
1.7 (0.8-3.6)
Reproductive variables (diet was uninfluent Age and parity. No distinction was possible between HRT and OC use Age, parity, hysterectomy Age, education, cancer family history, BMI, parity, menopause, OC, and energy intake
Lower RR for < 10 years since last use = 0.5; 0.4-0.8, for colon
No trend (RR -> 10 yr use = 0.86) Not shown
RR for recent use = 0.71, (0.56-0.89) Not shown
No trend in risk
No association with recent use RR = 0.6 of colon cancer among recent users For >- 3 years use RR 0.75 for oral, 0.33 for transdermal
Age, cancer family history, aspirin and energy intake, OC, and exercise Age, race, reproductive variables, dietary habits, and colonoscopy Age Fat, fruit, vegetable intake; physical exercise; colorectal cancer screening Age
606 In conclusion, available data from randomized clinical trials do not support the use of combined H R T nor of estrogen alone to reduce colorectal cancer mortality.
III. LUNG CANCER Lung cancer is the leading cause of cancer mortality among women in the United States (31), and mortality from lung cancer in women has been substantially increasing across Europe over the last few decades (73). Thus, any influence of H R T on lung cancer risk would be of major public health importance. Epidemiologic data on H R T and lung cancer risk have therefore been updated here (1,2). A cohort study in Sweden of 23,244 women followed for 6.7 years suggested a moderate excess risk of lung cancer associated with use of estrogens (RR = 1.3; 95% CI 0.9-1.7) (74). No information was available on duration of use or any other risk factors. Two case-control studies from the United States have examined the relation of H R T use to risk of adenocarcinoma of the lung. A hospital-based casecontrol study of 181 cases found a 70% excess risk associated with estrogen replacement therapy, with the RR increasing twofold for smokers of 25 or more years (75). However, residual confounding by smoking remains possible, and in another population-based case-control study from Los Angeles County, California, including 336 cases, no significant relation was found between H R T use and lung cancer risk (RR 1.3; 95% CI 0.7-2.2) (76). Likewise, in a multicentric U.S. case-control study including 662 women with lung cancer and 4621 control subjects, the RR was 1.0 (95% CI 0.7-1.4) for E R T use, and also 1.0 for users of conjugated estrogens only, in the absence of any consistent trend with duration or any other time factors (77). In a population-based case-control study of 811 lung cancer cases 912 and controls from Germany, the RR for ever H R T use was 0.69 (95% CI 0.57-0.92), but again no relation was evident with duration, age at first use, and calendar year of use (78). In a cohort study of 28,508 women ages 25 to 65 from Sweden followed-up between 1990 and 1999, a nonsignificant decreased risk of lung cancer (RR = 0.73) was observed, although the risk of all tobacco-related neoplasms was significantly reduced (RR = 0.24) (79). In a case-control study of 499 lung cancer cases and 519 population controls from Texas, the multivariate RR was 0.66 (95% CI 0.51-0.89) for ever H R T use. The risk was significantly reduced in current, moderate smokers only and in a subgroup of genetically susceptible women (80), although the definition of susceptibility remains open to discussion. In the W H I randomized clinical trial (20), combined estrogen H R T was unrelated to lung cancer risk (RR = 1.04; 95% CI 0.71-1.53).
CARLO LA VECCHIA
Thus, taken together, observational studies and the limited available evidence from randomized clinical trials do not show any consistent association between H R T and lung cancer risk, because there are a similar number of studies showing RRs about unity, no association, or RRs below unity. The absence of material association between H R T and lung cancer risk is plausible, because it is now clear that women are not more susceptible to lung cancer than men for similar level of smoking (81,82), thus indicating that female hormones do not have a relevant role in lung carcinogens. Indeed, given the prominent role of tobacco smoking in lung cancer, minor differences in quantity or duration of smoking may have caused differences in risk as those observed in several studies between H R T users and nonusers.
References 1. International Agency for Research on Cancer. LdRC Monograph on the Evaluation of CarcinogenicRisks to Humans, vo172. Hormonal contraception andpost-menopausal hormonal therapy. Lyon, France: IARC, 1999. 2. La Vecchia C, Brinton LA, McTiernan A. Menopause, hormone replacement therapy and cancer. Maturitas 2001;39:97-115. 3. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Lancet 1997;350: 1047-1059. 4. Beral V, Banks E, Reeves G, Appleby P. Use of HRT and the subsequent risk of cancer.J Epidemiol Biostatistics 1999;4:191-215. 5. La Vecchia C. Menopause, hormone therapy and breast cancer risk. EurJ CancerPrey 2003;12:437-438. 6. ColditzGA, Hankinson SE, Hunter DJ, et al. The use of estrogens and progestins and the risk of breast cancer in postmenopausal women. NEnglJMed 1995;332:1589-1593.
7. Magnusson C, BaronJA, Correia N, et al. Breast-cancer risk following long-term oestrogen- and oestrogen-progestin-replacement therapy. IntJ Cancer 1999;81:339-344. 8. SchairerC, Lubin J, Troisi R, et al. Menopausal estrogen and estrogenprogestin replacement therapy and breast cancer risk.J_//M.d2000;283: 485-491. 9. Ross RK, Paganini-Hill A, Wan PC, Pike MC. Effect of hormone replacement therapy on breast cancer risks: estrogen versus estrogen plus progestin.JNatl CancerInst 2000;92:328- 332. 10. Olsson HL, Ingvar C, Bladstrom A. Hormone replacement therapy containing progestins and given continuously increases breast carcinoma risk in Sweden. Cancer2003;97:1387-1392. 11. Jernstrom H, Bendahl P-O, Lidfeldt J, et al. A prospective study of different types of hormone replacement therapy use and the risk of subsequent breast cancer: the women's health in the Lund area (WHILA) study (Sweden). Cancer Causes Control2003;14:673-680. 12. Weiss LK, Burkman RT, Cushing-Haugen KL, et al. Hormone replacement therapy regimens and breast cancer risk. Obstet Gynecol 2002;100:1148-1158. 13. Greendale GA, Reboussin BA, Slone S, et al. Postmenopausal hormone therapy and change in mammographic density.J Nard Cancer Inst 2003;95:30- 37.
14. Porch JV, Lee I-M, Cook NR, Rexrode KM, Buring JE. Estrogenprogestin replacement therapy and breast cancer risk: the Women's Health Study (United States). Cancer Causes Control 2002;13: 847-854.
CHAPTER 44 H o r m o n e Replacement Therapy in Menopause and Breast, Colorectal, and Lung Cancer: An Update 15. Li C, Malone ICE, Porter PL, et al. Relationship between long durations and different regiments of hormone therapy and risk of breast cancer. JAMA 2003;289:3254- 3263. 16. Verkooijen HM, Fioretta G, Vlastos G, et al. Important increase of invasive lobular breast cancer incidence in Geneva, Switzerland. Int J Cancer 2003;104:778- 781. 17. Levi F, Te V-C, Randimbison L, La Vecchia C. Increase in lobular breast cancer incidence in Switzerland. Int J Cancer 2003;107: 164-165. 18. Million Women Study Collaborators. Breast cancer and hormonereplacement therapy in the Million Women Study. Lancet 2003;362: 419-427. 19. Stahlberg C, Pedersen AT, Lynge E, et al. Increased risk of breast cancer following different regimens of hormone replacement therapy frequently used in Europe. IntJ Cancer 2004;109:721-727. 20. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. JAMA 2002;288:321 - 333. 21. Powledge TM. NIH terminates W H I oestrogen-only study. Lancet 2004;363:870. 22. Women's Health Initiative Steering Committee. Effect of conjugated equine estrogen in postmenopausal women with hysterectomy. JAMA 2004;291:1701 - 1712. 23. Hulley S, Furberg C, Barrett-Connor E, et al. Noncardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/progestin Replacement Study follow-up (HERS II). JAMA 2002;288:58-66. 24. Viscoli CM, Brass LM, Kernan WN, et al. A clinical trial of estrogenreplacement therapy after ischemic stroke. N Engl J Med 2001;345: 1243 - 1249. 25. Beral V, Banks E, Reeves G. Evidence from randomised trials on the long-term effects of hormone replacement therapy. Lancet 2002;360: 942-944. 26. Hulley SB, Grady D. The W H I estrogen-alone trial. Do things look any better? JAMA 2004;291:1769-1771. 27. Rodriguez C, Calle EE, Patel AV, et al. Effect of body mass on the association between estrogen replacement therapy and mortality among elderly US women. Am J EDidemio12001;153:145-152. 28. Chlebowski RT, Hendrix SL, Langer RD, et al. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women. JAAdA 2003;289:3243-3253. 29. Parkin DM, Bray F, Ferlay J, Pisani R Estimating the world cancer burden: Globocan 2000. IntJ Cancer 2001;94:153-156. 30. Levi F, Lucchini F, Negri E, Boyle P, La Vecchia C. Cancer mortality in Europe, 1995-1999, and an overview of trends since 1960. Int J Cancer 2004;110:155-169. 31. Bosetti C, Malvezzi M, Chatenoud L, et al. Trends in cancer mortality in the Americas, 1970-2000. Ann Onco12005;16:489-511. 32. Fernandez E, Bosetti C, La Vecchia C, et al. Sex differences in colorectal cancer mortality in Europe, 1955-1996. EurJ Cancer Prev 2000a; 9:99 - 104. 33. Fernandez E, Franceschi S, La Vecchia C. Colorectal cancer and hormone replacement therapy: a review of epidemiological studies. J Br Menop Soc 2000;6:8-14. 34. Fernandez E, La Vecchia C, Balducci A, et al. Oral contraceptives and colorectal cancer risk: A meta-analysis. BrJ Cancer 2001;84:722-727. 35. Fernandez E, Gallus S, Bosetti C, et al. Hormone replacement therapy and cancer risk: a systematic analysis from a network of case-control studies. IntJ Cancer 2003;105:408-412. 36. Negri E, La Vecchia C, Parazzini F, et al. Reproductive and menstrual factors and risk of colorectal cancer. CancerRes 1989;49:7158-7161. 37. La Vecchia C, Franceschi S. Reproductive factors and colorectal cancer. Cancer Causes Control 1991;2:193-200.
607
38. Jacobs EJ, White E, Weiss NS. Exogenous hormones, reproductive history, and colon cancer (Seattle, Washington, USA). Cancer Causes Control 1994;5:359-366. 39. Martinez ME, Grodstein F, Giovannucci E, et al. A prospective study of reproductive factors, oral contraceptive use, and risk of colorectal cancer. CancerEDidemiol Biomarkers Prev 1997;6:1-5. 40. Talamini R, Franceschi S, Dal Maso L, et al. The influence of reproductive and hormonal factors on the risk of colon and rectal cancer in women. EurJ Cancer 1998;34:1070-1076. 41. Franceschi S, Gallus S, Talamini R, et al. Menopause and colorectal cancer. BrJ Cancer 2000;82:1860-1862. 42. Van Wayenburg C, van der Schouw YT, van Noord PA, Peetersw PHM. Age at menopause, body mass index, and the risk of colorectal cancer mortality in the Dutch Diagnostich Onderzoek Mammacarcinoom (DOM) cohort. Epidemiology 2000;11:304- 308. 43. Zhang Y, Felson DT, Ellison RC, et al. Bone mass and the risk of colon cancer among postmenopausal women. Am J Epidemiol 2001;153: 31-37. 44. Franceschi S, La Vecchia C. Colorectal cancer and hormone replacement therapy: an unexpected finding. EurJ Cancer 1998;7:427-438. 45. H6bert-Croteau N. A meta-analysis of hormone replacement therapy and colon cancer among women. Am J Epidemiol 1998;147(suppl): $87 (Abs 345). 46. Grodstein F, Newcomb PA, Stampfer MJ. Postmenopausal hormone therapy and the risk of colorectal cancer: a review and meta-analysis. AmJMed 1999;106:574-582. 47. Nanda K, Bastian LA, Haselblad V, Simel SL. Hormone replacement therapy and the risk of colorectal cancer: a meta-analysis. Obstet Gynecol 1999;93:880-888. 48. McMichael AJ, Potter JD. Host factors in carcinogenesis: certain bileacid metabolic profiles that selectively increase the risk of proximal colon cancer. J Natl CancerInst 1985;75:185-191. 49. Costarelli V, Key TJ, Appleby PN, et al. A prospective study of serum bile acid concentrations and colorectal cancer risk in post-menopausal women on the Island of Guernsey. BrJ Cancer 2002;86:1741-1744. 50. Carothers AM, Hughes SA, Ortega D, Bertagnolli MM. 2Methoxyestradiol induces p53-associated apoptosis of colorectal cancer cells. CancerLett 2002;187:77-86. 51. Slattery ML, Potter JD, Curtin K, et al. Estrogen reduce and withdrawal of estrogens increase risk of microsatellite instability-positive colon cancer. CancerRes 2001;61:126-130. 52. Manousos O, Souglakos J, Bosetti C, et al. IGF-I and IGF-II in relation to colorectal cancer. IntJ Cancer 1999;83:15-17. 53. Buist DS, Newton KM, Miglioretti DL, et al. Hormone therapy prescribing patterns in the United States. Obstet Gynecol 2004;104: 1042-1050. 54. Majumdar SR, Almasi EA, Stafford RS. Promotion and prescribing of hormone therapy after report of harm by the Women's Health Initiative, WHI. JAMA 2004;1983-1988. 55. Slattery ML, Anderson K, Samowitz W, et al. Hormone replacement therapy and improved survival among postmenopausal women diagnosed with colon cancer (USA). Cancer Causes Control 1999;10: 467-473. 56. Mandelson MT, Miglioretti D, Newcomb PA, Harrison R, Potter JD. Hormone replacement therapy in relation to survival in women diagnosed with colon cancer. Cancer Causes Control2003;14:979-984. 57. Newcomb PA, Storer BE. Postmenopausal hormone use and risk of large-bowel cancer. J Natl CancerInst 1995;87:1067-1071. 58. Persson I, Yuen J, Bergkvist L, et al. Cancer incidence and mortality in women receiving estrogen and estrogen-progestin replacement therapy-long-term follow-up of a Swedish cohort. IntJ Cancer 1996; 67:327-332.
608 59. Csizmadi I, Collet J-P, Benedetti A, Boivin J-F, HanleyJA. The effects of transdermal and oral oestrogen replacement therapy on colorectal cancer risk in postmenopausal women. BrJ Cancer 2004;90:76-81. 60. Curtin K, Bigler J, Slattery ML, et al. MTHFR C677T and A1298C polymorphisms: diet, estrogen, and risk of colon cancer. CancerEpidemiol Biomarkers Prev 2004;13:285-292. 61. Fernandez E, La Vecchia C, Braga C, et al. Hormone replacement therapy and risk of colon and rectal cancer. CancerEpidemiol Biomarkers Prev 1998;7:329-333. 62. Grodstein F, Martinez E, Platz EA, et al. Postmenopausal hormone use and risk for colorectal cancer and adenoma. Ann Intern Med 1998;128:705-712. 63. Calle EE, Miracle-McMahill HL, Thun MJ, et al. Estrogen replacement therapy and risk of fatal colon cancer in a prospective cohort of postmenopausal women. JNatl Cancer Inst 1995;87:517- 523. 64. Day NE. Epidemiology: the role of multi-stage models. Cancer Surv 1983;2:579-593. 65. Sturgeon SR, Schairer C, Brinton LA, et al. Evidence of a healthy estrogen user survivor effect. Epidemiology 1995;6:227-231. 66. Barrett-Connor E. Postmenopausal estrogen and prevention bias. _~Inn Intern Med 1991;115:455-456. 67. Hemminki E, Kennedy DL, Baum C, et al. Prescribing of noncontraceptive estrogens and progestins in the United States, 1974-86. Am J Public Health 1988;78:1479-1481. 68. Parazzini F, La Vecchia C, Negri E, et al. Determinants of estrogen replacement therapy use in northern Italy. Rev Epidem et Santg Publ 1993;41:53-58. 69. Chiaffarino F, Parazzini F, La Vecchia C, et al. Correlates of hormone replacement therapy use in Italian women, 1992-1996. Maturitas 1999;33:107-115. 70. Csizmadi I, Collett J-P, Boivin J-F, Hanley JA, Benedetti A. Healthrelated behavior and the use of hormone replacement therapy. Pharmacoepidemiol Drug Saf2004;13:65- 71. 71. La Vecchia C, Franceschi S. Hormone replacement therapy and cancer: an update. EurJ Cancer Prev 2003;12:3-4. 72. Chlebowski RT, Wactawski-Wende J, Ritenbaugh C, et al. Estrogen plus progestin and colorectal cancer in postmenopausal women. N EnglJ Med 2004;350:991 - 1004. 73. Borras JM, Fernandez E, Gonzalez JR, et al. Lung cancer mortality in European regions (1955-1997). Ann Onco12003;14:159-161. 74. Adami H-O, Persson I, Hoover R, et al. Risk of cancer in women receiving hormone replacement therapy. IntJ Cancer 1989;44:833-839. 75. Taioli E, Wynder EL. Re: Endocrine factors and adenocarcinoma of the lung in women.JNatl CancerInst 1994;86:869-870. 76. Wu AH, Yu MC, Thomas D e , et al. Personal and family history of lung disease as risk factors for adenocarcinoma of the lung. CancerRes 1988;48:7279- 7284. 77. Blackman JA, Coogan PF, Rosenberg L, et al. Estrogen replacement therapy and risk of lung cancer. Pharmacoepidemiol Drug Saf 2002; 11:561-567. 78. Kreuzer M, Gerken M, Heinrich J, Kreienbrock L, Wichmann HE. Hormonal factors and risk of lung cancer among women? IntJEpidemiol 2003;32:263-271. 79. Olsson H, Bladstrom A, Ingvar C. Are smoking-associated cancers prevented or postponed in women using hormone replacement therapy? Obstet Gyneco12003;102:565-570. 80. Schabath MB, Wu X, Vassilopoulou-Sellin R, Vaporciyan AA, Spitz MR. Hormone replacement therapy and lung cancer risk: a case-control analysis. Clin CancerRes 2004;10:113-123. 81. Bain C, Feskanich D, Speizer FE, et al. Lung cancer rates in men and women with comparable histories of smoking. J Natl Cancer Inst 2004;96:826- 834. 82. Blot WJ, McLaughlin JK. Are women more susceptible to lung cancer? J Natl Cancer Inst 2004;96:812-813.
CARLO LA VECCHIA 83. Wu AH, Paganini-Hill A, Ross RK, et al. Alcohol, physical activity and other risk factors for colorectal cancer: a prospective study. BrJ Cancer 1987;55:687-694. 84. Chute CG, Willett WC, Colditz GA, et al. A prospective study of reproductive history and exogenous estrogens on the risk of colorectal cancer in women. Epidemiology 1991;2:201-207. 85. Bostick RM, Potter JD, Kushi LH, et al. Sugar, meat, and fat intake, and non-dietary risk factors for colon cancer incidence in Iowa women (United States). Cancer Causes Control 1994;5:38-52. 86. Folsom AR, Mink PJ, Sellers TA, et al. Hormonal replacement therapy and morbidity and mortality in a prospective study of postmenopausal women. Am J Public Health 1995;85:1128 - 1132. 87. Risch HA, Howe GR. Menopausal hormone use and colorectal cancer in Saskatchewan: a record linkage cohort study. CancerEpidemiol Biomarkers Prev 1995;4:21-28. 88. Troisi R, Schairer C, Chow W-H, et al. A prospective study of menopausal hormones and risk of colorectal cancer (United States). Cancer Causes Control 1997;8:130-138. 89. Paganini-Hill A. Estrogen replacement therapy and colorectal cancer risk in elderly women. Dis Colon Rectum 1999;42:1300-1305. 90. Pukkala E, Tulenheimo-Silfvast A, Leminen A. Incidence of cancer among women using long versus monthly cycle hormonal replacement therapy, Finland 1994-1997. Cancer Causes Control 2001;12: 111-115. 91. Weiss NS, Daling JR, Chow WH. Incidence of cancer of the large bowel in women in relation to reproductive and hormonal factors. J Natl CancerInst 1981;67:57-60. 92. Potter JD, McMichael AJ. Large bowel cancer in women in relation to reproductive and hormonal factors: a case-control study. J Natl Cancer Inst 1983;71:703- 709. 93. Davis FG, Furner SE, Persky V, et al. The influence of parity and exogenous female hormones on the risk of colorectal cancer. Int J Cancer 1989;43:587-590. 94. Furner SE, Davis FG, Nelson RL, et al. A case-control study of large bowel cancer and hormone exposure in women. Cancer Res 1989;49: 4936-4940. 95. Fernandez E, La Vecchia C, D'Avanzo B, et al. Oral contraceptives, hormone replacement therapy and the risk of colorectal cancer. BrJ Cancer 1996;73:1431-1435. 96. Peters RK, Pike MC, Chang WWL, et al. Reproductive factors and colon cancers. BrJ Cancer 1990;61:741 - 778. 97. Wu-WiUiams AH, Lee M, Whittemore AS, et al. Reproductive factors and colorectal cancer risk among Chinese females. Cancer Res 1991;51:2307- 2311. 98. Gerhardsson de Verdier M, London S. Reproductive factors, exogenous female hormones, and colorectal cancer by subsite. Cancer Causes Control 1992;3:355-360. 99. Kampman E, Potter JD, Slattery ML, et al. Hormone replacement therapy, reproductive history, and colon cancer: a multicenter, casecontrol study in the United States. Cancer Causes Control 1997; 8:146-158. 100. Yood SM, Ulcickas Yood M, McCarthy B. A case-control study of hormone replacement therapy and colorectal cancer. Ann Epidemiol 1998;8:133. 101. Jacobs EJ, White E, Weiss NS, et al. Hormone replacement therapy and colon cancer among members of a health maintenance organization. Epidemiology 1999;10:445-451. 102. Prihartono N, Palmer JR, Louik C, Shapiro S, Rosenberg L. A casecontrol study of use of postmenopausal female hormone supplements in relation to the risk of large bowel cancer. Cancer Epid Biomarkers Prey 2000;9:443-447.
SECTION IX
Clinical Trials and Observational Data This section deals with differences between randomized clinical trial (RCT) data and those derived from observational studies such as the Nurse's Health Study (NHS), the Leisure World Cohort, and so on. In so doing a discussion emanates on the overall risk-benefit equation for the use of hormones after menopause. In previous editions of this book a specific chapter was written on the cost-benefit analysis of hormonal therapy. Because the field is still somewhat unsettled after the publication of the recent randomized controlled trials (RCTs), and there is no clear consensus regarding who should be treated with hormones after menopause, that chapter has been omitted from this section. As will be pointed out in this section, there is remarkable agreement between the results of the observational data and the various RCTs. Only in two areas are the data discrepant; namely, coronary disease and dementia. The latter differences may be explained by the ages of the patients when treated, as pointed out by many authors in this book. (See the Section VII, on cardiovascular disease, and Section V, on brain function.) Nevertheless, there is no denying that certain intrinsic biases occur in observational studies that may affect the results to a varying degree. It is important, however, to point out that a good study is a good study and there is no reason to "trash" all studies that are not RCTs. Indeed, it has been shown that the results of RCTs and observational trials are largely in full agreement (1,2). Further, RCTs themselves also have intrinsic biases and problems. These concerns include problems of blinding (particularly hormonal studies in women where bleeding may occur), unique characteristics of women willing to be randomized in long trials, the socioeconomic status of many trial participants seeking "free care," and an overrepresentation of obese subjects in such trials. Perhaps for me the greatest concern regarding RCTs is the issue of poor external validity. It is very difficult and potentially erroneous in my view to extrapolate the findings of any given RCT, in one population and with one treatment regimen, to the entire population of women and to other potential treatment regimens within the same class. It has been said many times before, but first by the late Trudy Bush, "No single study has a monopoly on the truth." In this section, Margery L.S. Gass, a W H I investigator, reviews the recent findings in the various W H I trials. Next, Karin B. Michels and JoAnn E. Manson, who was an investigator in both W H I and the NHS, point out differences between observational data and RCTs, and offer possible explanations for the discrepancy. Finally, Annlia Paganini-Hill, who represents the rich experience of the Leisure World cohort study, discusses this issue and expands the discussion to an overall risk-benefit assessment.
References 1. Benson K, Hartz AJ. A comparison of observational studies and randomized, controlled trials. N Engl J Med 2000;342: 1878-1886. 2. ConcatoJ, Shah N, Horwitz RI. Randomized,controlled trials, observationalstudies, and the hierarchy of research designs. N EnglJ Med 2000;342:1887-1892.
This Page Intentionally Left Blank
~ H A P T E R 4_
The Women's Health Data nlt atlve and Discussion 9
9
9
MARGERY L.S. GASS ClinicalObstetrics and Gynecology,University of Cincinnati College of Medicine, Cincinnati, OH 45267
I. B A C K G R O U N D
II. W H I H O R M O N E T H E R A P Y TRIAL
The Women's Health Initiative is a large, intricately designed multisite study sponsored by the National Heart, Lung, and Blood Institute of the National Institutes of Health. The overarching purpose of the Women's Health Initiative (WHI) was to evaluate the efficacy and safety of three commonly used preventive health interventions: postmenopausal hormone therapy, low-fat diet, and calcium plus vitamin D supplementation (Fig. 45.1). All three W H I trial interventions were potentially beneficial for the majority of postmenopausal women regardless of age. Further, because all major outcomes of the study are more common with aging, it was important to include the older, more susceptible population for greater power in the data analysis. Healthy women ages 50 to 79 were enrolled. Women thought likely to discontinue the study (e.g., poor health, severe vasomotor symptoms, or likely to move from the area) were excluded. A detailed explanation of the rationale for the study design has been published (1). T R E A T M E N T OF T H E P O S T M E N O P A U S A L W O M A N
The trend in hormone therapy (HT) use at the time the W H I was designed in 1991 was to recommend H T for an expanding list of potential health benefits. The United States Food and Drug Administration (FDA) approved many estrogen products for the treatment of vasomotor symptoms, vulvovaginal atrophy, and the prevention of osteoporosis. There were recommendations for long-term use of H T for bone protection (2). Well-designed clinical trials demonstrated efficacy for the treatment of vasomotor symptoms and vulvovaginal atrophy and for preservation of bone mineral density. In the absence of fracture data, however, the FDA withdrew the indication for treatment of osteoporosis. A variety of trials and studies suggested that H T might protect against cardiac disease and dementia, as well as improve bone mineral density in older women who already had osteoporosis (3-8). See other chapters in this book for the scientific data that accumulated on the use of H T for these health conditions after the W H I had been designed. 611
Copyright 9 2007 by Elsevier, Inc. All rights of reproduction in any form reserved.
6
1
2
M
A
R
G
161,809women age 50-79 ] [ H0rm0neTria127,347] f ~ w F a t D i e t ] Estrogen+Progestin 16,608 1 | Trial EstrogenAlone 10,739,J l, 48,836
Calcium Vitamin D Trial Enrolled at Years 1 and
," Observational l Trial l 93,676
E
R
Y
L.S. GASS
information to the medical community, the participants, and the public as promptly as possible (10,11). This rapid publication necessitated using the database before health outcomes from the last few months of the trial had been entered. Subsequent publications focused more narrowly on specific endpoints and took advantage of the completed database. The different databases account for slightly different figures in the later papers.
,
FIGURE 45.1 Overviewof the Women's Health Initiative.Total clinical trial (CT) included 68,133 subjects; 11.8% of the CT enrolled in both the Hormone Therapy (HT) and Dietary Modification (DM) Trials; 37% of the CT enrolled in the DM and Calcium Vitamin D (CAD)Trials; 23.6% of the CT enrolled in the HT and CaD Trials; 7.3% of the CT enrolled in all three arms of the CT.
To assess the role of hormone therapy as a preventive health measure, the most widely prescribed hormones in the United States were selected: conjugated equine estrogens 0.625 mg (CEE) and medroxyprogesterone acetate (MPA). A daily regimen of 2.5 mg of MPA was chosen over the higher cyclic dose in an effort to reduce the amount of uterine bleeding, the most common reason for discontinuing HT, especially in older women (9). Another consideration in selecting the regimen was the concern that the new onset of cyclic bleeding would unblind the participants to the fact that they had received the active study drug. The doubleblind aspect of the study would have been undermined with a cyclic regimen. The H T trial had two arms with different treatment regimens: estrogen therapy (ET) for women who had undergone hysterectomy and estrogen plus progestogen therapy (EPT) for those who had not undergone hysterectomy. The primary endpoints for the two H T arms were coronary heart disease and breast cancer. Secondary endpoints included hip and other osteoporotic fractures, other cardiovascular events, and other cancers. A data and safety monitoring board of independent experts who were not involved in the trial met every 6 months to review the accumulated data and to determine if either the benefits or the risks were of sufficient magnitude to stop the trial before the designated 8.5 years were completed. The E P T and the E T intervention phases were both stopped earlier than planned, albeit at different times and for different reasons. The E P T intervention was discontinued in July 2002 because an elevated occurrence of breast cancer reached a predetermined safety stopping point. The intervention phase of the E T arm was stopped in February 2004 because of an elevated risk of stroke along with evidence that there was no cardiac benefit. A summary of key findings for each trial was published at the time the trial was stopped in order to transmit the
A. Osteoporosis Results from the two H T arms were not congruent on all primary endpoints (Fig. 45.2). The one statistically significant beneficial finding in both arms of the Hormone Trial was the decreased risk of hip fracture. In the E P T arm, the hazard ratio (HR) was 0.67; 95% confidence interval (CI), 0.47-0.96 (12). In the E T arm, the H R was 0.61; 95% CI, 0.41-0.91 (11). The number of total fractures was also reduced in both arms. The H T trial was the first randomized, placebo-controlled hormone trial to demonstrate a statistically significant reduction in hip fractures and other osteoporotic fractures. The finding is notable because unlike
FIGURE 45.2 Annualized incidence of events per 10,000 women per year. EPT, estrogen-progestin therapy; ET, estrogen therapy; Hip Fx, hip fracture; CHD, coronary heart disease events; Breast CA, breast cancer; Colon CA, colon cancer; VTE, venous thrombosis events. See references 10, 11, 13, 14, 15, 19, 20.
CHAPTER 45 The Women's Health Initiative--Data and Discussion TABLE 45.1
Global Index of Risk Associated with Tertile of Fracture Risk.
Tertile of fracture risk
Global index 1.20 (95% CI 0.93-1.58) 1.23 (95% CI 1.04-1.46) 1.03 (95% CI 0.88-1.24)
Lowest Tertile Middle Tertile Highest Tertile See reference 12.
most osteoporosis drug trials, W H I participants were not required to have osteoporosis or a history of vertebral fracture to enroll in the study. Because of the growing number of elderly women in the population and the association of osteoporosis with aging, it is important to consider whether H T should be used on a widespread basis for prevention of osteoporosis. Cauley et al. addressed that question using the E P T data. Taking into consideration the documented side effects and stratifying study participants according to their risk of fracture, the authors found that even in the highest fracture risk category, the risk/benefit ratio of E P T was neutral, indicating that the fracture-prevention benefit was not outweighing the risks of E P T (Table 45.1)(12).
B. Cardiovascular Disease A key finding in the E P T and ET arms was the failure of either hormone regimen to demonstrate prevention of coronary heart disease, the major endpoint of the Hormone Trial (10,11,13,14). An increase in cardiac events (myocardial infarction [MI] or coronary death) occurred in the first year in the E P T arm (HR 1.81; 95% CI, 1.09-3.01) but was no longer statistically significant by the time the study was stopped (HR 1.24; 95% CI, 1.00-1.54). Mean follow-up was 5.6 years (13). Unlike EPT, ET did not produce a statistically significant increase in coronary events in the first year (HR 1.11; 95% CI, 0.64-1.94) (14). However, after a mean follow-up of 7.1 years, there was no significant benefit with regard to the primary coronary endpoints (HR 0.95; 95% CI, 0.79-1.16). Additional analyses of the effect of H T on cardiovascular disease (CVD) by subgroup failed to demonstrate prevention of MI or coronary death according to age decade with either E P T or ET. Among participants ages 50 to 59 at baseline, analysis by composite coronary outcomes suggested a benefit for those participants assigned to ET. The H R for a combination of MI, coronary death, revascularization procedures, and confirmed angina was 0.66 (95% CI, 0.45-0.96); however, the test for interaction of treatment effect by age was not significant (14). Secondary coronary events did not differ among the older ET and placebo participants in the ET arm. The expected effects
613 of oral ET on lipids were observed, and it was noted that the effect of ET did not vary by baseline lipoprotein levels. This finding was in contrast to findings in the E P T arm, where the risk of C H D increased with higher levels of L D L cholesterol. Elevated C-reactive protein at baseline was associated with an increased risk of C H D among those assigned to ET but not to EPT. On the other hand, women assigned to ET who had coronary disease or multiple risk factors for coronary disease at baseline were not found to be at increased risk for cardiac events on treatment. In totality, these data, while not proving a benefit of H T for CVD prevention, do suggest relative safety of H T in regard to C H D for the younger symptomatic postmenopausal woman interested in using HT.
C. Stroke and Venous Thrombosis Another area of similarity between the two H T arms was the finding of an increased risk of stroke. Both E P T and ET resulted in an increased risk of stroke by the second year that persisted over the course of the study (HR 1.31; 95% CI, 1.02-1.68, and H R 1.39; 95% CI, 1.10-1.77, respectively) (11,15). Both H T arms also demonstrated an increased risk of venous thrombosis (VT), confirming previous reports on hormone therapy (16-18). The risk of thrombosis was highest in the first year of the E P T trial (HR 4.01), declined thereafter, but remained above average throughout the study (HR 2.06, 95% CI, 1.57-2.70) (19). In the ET arm the hazard ratio of thrombotic events was less (HR 1.33; 95% CI, 0.99-1.79) (11). The rate of pulmonary emboli was statistically significant in the E P T arm (HR 2.13; 95% CI, 1.45-3.11), but not in the ET arm (HR 1.34; 95% CI, 0.87-2.06) (11,19). Other factors can increase the baseline risk of thrombosis and thereby increase the absolute number of events seen with HT. Both older age and greater body mass index (BMI) increased the risk of venous thrombosis in W H I . Compared with women ages 50 to 59 on placebo, women ages 70 to 79 on E P T had a hazard ratio of 7.46 (95% CI, 4.32-14.38). Women on E P T with a BMI greater than 30 had a hazard ratio of 5.61 (95% CI, 3.12-10.11) compared with women on placebo with a BMI less than 25. Among E P T users no protective effect was seen with statins or aspirin (19).
D. Cancer Two areas of disagreement between the two hormone arms were the main cancer outcomes, colorectal and breast. Invasive breast cancer was increased in the E P T arm
6
1
4
M
A
R
G
(HR 1.24; 95% CI,1.01-1.54), whereas there was an insignificant decrease in breast cancer in the ET arm (HR 0.80; 95% CI, 0.62-1.04) (20,21). In both trials the tumors in the active arm were larger and more likely to have spread to the lymph nodes, raising a possible role for diagnostic delay due to a masking effect of H T on breast density or consistency. Further analyses of the breast cancer data across the entire W H I cohort revealed that the incidence rates and types of breast cancer differed among racial and ethnic groups (22). Age-adjusted incidence of breast cancer was lowest in African Americans (HR 0.75; 95% CI, 0.61-0.92), but the mortality rate was higher compared with Caucasian women (HR 1.79; 95% CI, 1.05-3.05). African Americans were more likely to have tumors with an unfavorable prognosis, such as high-grade and estrogen-receptor-negative tumors (adjusted odds ratio [OR] 4.70; 95% CI, 3.12-7.09). Colorectal cancer was decreased among women randomized to EPT compared with placebo (HR 0.56; 95% CI, 0.38-0.81; p = 0.003) (23). The percentage of invasive tumors with regional or metastatic disease was higher in the EPT group compared with placebo (76.2% versus 48.5%; p = 0.004). There was no statistically significant difference in the incidence of colorectal cancer in the ET arm (HR 1.08; 95% CI, 0.75-1.55) (11).
E. C o g n i t i v e D e c l i n e The W H I Memory Study (WHIMS) is a W H I ancillary study that published key findings soon after each H T arm was stopped (24-27). Women in this ancillary study were required to be age 65 or older to participate because the younger age group was unlikely to be diagnosed with dementia during the time course of the study. Neither EPT nor ET prevented cognitive decline or dementia. The analysis of the pooled EPT and ET data, as was originally planned, suggested harm rather than benefit with regard to probable dementia (HR 1.76; 95% CI, 1.19-2.60; ? = 0.006)(26).
F. M e n o p a u s a l S y m p t o m s Vasomotor symptoms were not the focus of the WHI. Women with severe vasomotor symptoms (VMS) were discouraged from participating in the randomized controlled hormone trials because it was thought they would be likely to discontinue the study if randomized to the placebo arm. Nonetheless, 2046 women with moderate to severe VMS did enroll in the EPT trial, and W H I confirmed a statistically significant benefit for VMS as well as small benefits for sleep, physical functioning, and body pain at the 1-year visit. Health-related quality of life was reported in several papers (28-30). Vaginal dryness was significantly improved
E
R
Y
L.S. GASS
for hormone users, as was the occurrence of joint pain and general aches and pains. W H I is the first trial to document H T benefit for aches and pains. In a subgroup analysis over a 3-year period of women 50 to 59 who were asymptomatic at baseline, fewer women developed vaginal dryness (OR 0.22, 95% CI, 0.08-0.59) and fewer women developed joint aches and pains (OR 0.54; 95% CI, 0.29-0.98) (30). The benefits of H T for menopausal symptoms have been the mainstay of the list of reasons for using H T as far as the Food and Drug Administration is concerned. These benefits, however, are sometimes offset by the side effects. The serious side effects were incorporated as major endpoints of the study and are listed earlier in the key findings. Less serious side effects play a role through their negative effect on quality of life as well as their effect on cost of medical care, including the need for repeat mammograms and breast biopsies. Justification for the concern about bleeding in the choice of hormone regimens is evident in the reasons participants gave for discontinuing therapy. In the EPT trial, 234 women mentioned vaginal bleeding as one of the reasons they discontinued therapy, compared with 8 in the placebo group, a 29-fold difference (30). Throughout the course of the EPT trial, 66.7% of the hormone users reported bleeding at some point despite being on continuous HT, compared with only 13.8% of the placebo users. Of the 875 endometrial aspirations and 217 transvaginal ultrasounds performed for uterine bleeding in the first year of the EPT trial, 91.3% and 86.2%, respectively, were performed in the active hormone group. Dilation and curettage procedures, as well as hysterectomies, were more common in the hormone users (30). The effect of H T on the breast was manifested in several ways: (1) an effect on breast cancer incidence as described earlier; (2) changes in mammographic density and incidence of abnormal mammogram; (3) symptoms of breast tenderness. A statistically significant increase in breast tenderness was seen at year 1 in the EPT 50- to 59-year-old group who were asymptomatic at baseline (OR 4.26; 95% CI, 3.59- 5.04) (30). An increase in the incidence of abnormal mammograms was noted in the first year (EPT 9.4% versus placebo 5.4%; p < 0.001). Over the course of the study the incidence of abnormal mammograms in the EPT group was 31.5% versus 21.2% in the placebo group (p < 0.001) (20). The Mammogram Density Ancillary Study involved 17 of the 40 W H I clinical centers. An analysis of mammographic density and abnormal mammograms was performed centrally over 1- and 2-year intervals. At year 1, mammographic percent density increased by 6% in the EPT group and declined by 0.9% in the placebo group, for an absolute difference of 6.9% (11.6% in the 70- to-79year-old group). At year 2, the absolute difference was 5.7%. Findings were similar across racial and ethnic groups. The relative risk of an abnormal mammogram report was 3.9 (95% CI, 1.5-10.2) and was not explained by the
CHAPTER 45 The Women's Health Initiative m Data and Discussion increase in density. At year 2, the relative risk for an abnormal mammogram in the EPT group was 1.8 (95% CI, 0.68-4.9) (31). Pre-WHI information on the subject of urinary incontinence generally taught that hormone therapy had a beneficial effect on incontinence (32). A combined analysis of the two W H I hormone arms revealed that women who were continent at baseline were more likely to develop incontinence when randomized to hormone therapy compared with those who were randomized to placebo (RR 1.87 [95% CI, 1.61-2.18] and 2.15 [95% CI, 1.77-2.62] for EPT and ET respectively compared to placebo) (33). Among those women who reported incontinence at baseline, the amount and frequency of incontinence increased on HT. Estrogen alone increased the incidence of urge incontinence (RR 1.32; 95% CI, 1.10-1.58). These findings do not support prescribing ET or EPT for either prevention or treatment of incontinence. The report from the H T trials on incontinence does not address the issue of recurrent cystitis in older women, a condition that some have thought benefits from topical estrogen (34).
III. WHI DIETARY MODIFICATION TRIAL The W H I Low-Fat Dietary Modification (DM) study was designed to test the hypothesis that lowering total fat intake would reduce the risk of breast and colorectal cancer in postmenopausal women. A secondary endpoint was the effect of low-fat diet on cardiovascular disease. The goal of the dietary intervention was to decrease the total fat consumption to 20% of energy intake while increasing the intake of fruits and vegetables to 5 servings a day and grains to 6 servings per day. To accomplish the dietary modification, specially trained nutritionists met with the women in the active intervention group 18 times in the first year and quarterly thereafter. Numerous strategies, including self-monitoring programs and motivational techniques, were employed in the effort to achieve and maintain the desired outcome. Despite intense efforts to effect dietary change, only 31% of the participants achieved the 20% fat goal at year 1 and only 14% at year 6.
A. Breast and Colorectal Cancers The hazard ratio for breast cancer was 0.91 (95% CI, 0.83-1.01). It was noted, however, that the women in the intervention arm with the highest baseline fat intake had a greater reduction in breast cancer (P for trend = 0.04) (35). The low-fat diet group had a statistically significant increase in sex-hormone-binding globulin as well as lower levels of serum estradiol.
615 Although there was no statistically significant reduction in colorectal cancer among women on the low-fat diet, the active intervention women did report fewer colorectal polyps (HR 0.91; 95% CI, 0.87-0.95) (36). Whether this will result in fewer cases of colorectal cancer in the years to come is unknown. The 5-year extension of the W H I may contribute to a better understanding of the long-term effects of the lowfat dietary intervention.
B. Cardiovascular The effect of the low-fat intervention on cardiovascular disease was a modest reduction in some of the known cardiac risk factors that did not lead to a statistically significant reduction in cardiovascular events (37). The intervention was not intended to be a weight loss diet, although the intervention group did weigh a statistically significant 1.29 kg less than the control group at year 3, a difference that dwindled to 0.4 kg by the end of the study.
IV. WHI CALCIUM VITAMIN D TRIAL In the Calcium Vitamin D (CAD) Trial, participants were randomized to 1000 mg/day of elemental calcium given in divided doses as calcium carbonate, plus 400 IU vitamin D3 or placebo. At 3 of the 40 W H I research sites bone mineral density was measured at years 1 (baseline for this trial), 3, 6, and 9. The primary endpoint of the CaD Trial was the effect of calcium and vitamin D supplementation on hip and other fractures. The secondary endpoint was the effect of CaD supplements on colorectal cancer.
A. Osteoporosis There was a 12% decrease in hip fractures in the CaD group that was not statistically significant (38). Subgroup analyses determined that women who were adherent to the supplements (defined as taking 80% or more of the tablets) had a statistically significant 29% decrease in hip fracture (HR 0.71; 95% CI, 0.52-0.97). Those women 60 years and older experienced a 21% decrease in hip fracture (HR 0.79; 95% CI, 0.64-0.98). Bone mineral density (BMD) increased from the baseline BMD to one year later at the hip for both the intervention and control group and then declined thereafter. BMD in the intervention group revealed a statistically significant increase compared with the control group at all time points after baseline. Bone mineral density at the hip was 1.06% higher in the intervention group (P < 0.01), but the 12% reduction in hip fractures and the 10% reduction in clinical spine fractures were not statistically significant (38).
6
1
6
M
A
R
G
B. Colorectal Cancer The CaD arm of the W H I did not confirm the hypothesis of a reduced rate of colorectal cancer with calcium and vitamin D supplementation (39). Unlike the low-fat diet trial, there was no finding of a decrease in colorectal polyps. Women should not be advised to use CaD supplements for prevention of colorectal cancer. Bone mineral density at the hip was 1.06% higher in the intervention group (P < 0.01), but the 12% reduction in hip fractures and the 10% reduction in clinical spine fractures were not statistically significant. A statistically significant adverse event associated with the CaD Trial was the increased rate of renal calculi (HR 1.17; 95% CI, 1.02-1.34). The finding was unrelated to a high calcium intake at baseline (38). Confounding the study results from the CaD Trial is the fact that the control group was allowed to take calcium and vitamin D. It was considered unethical to proscribe these supplements in an older population of women at risk for osteoporosis. As a result very few of the women in the placebo arm were calcium deficient or insufficient. Mean intakes at baseline in the control group for calcium and vitamin D were 1154 mg/day and 368 IU/day, respectively (38).
V. WHI OBSERVATION TRIAL The W H I Observation Trial is comprised of 93,676 women. It is yet another extensive source of biospecimens for the study of disease markers (1).
VI. CHALLENGES Most H T products reached the market by demonstrating relatively short-term effective and safe treatment of vasomotor symptoms and vulvovaginal atrophy. Clinicians are now in a position of translating risks and benefits found in a chronic disease prevention trial conducted across a wide age range to their predominantly younger (i.e., 45 to 55 year old) patients seeking treatment for specific symptoms. Some have approached this challenge by using qualitative phrases such as "small increase" in risks. Others discuss the smaller absolute risk in the youngest age decade of WHI. While not designed to address this particular situation, W H I does provide more specific risk and benefit information than previously existed for the 50- to-59-year-old age group. Some of the unexpected outcomes from the W H I present scientific challenges because they defy simple explanations. The cardiac and cognitive findings from W H I and W H I M S were unexpected. Much effort has been spent trying to understand the discrepancy between the observational data and the randomized controlled W H I and
E
R
Y
L.S. GASS
W H I M S data. Potential explanations for the conflicting findings have included the following topics. 1. The hormone regimens used in the W H I differed from what most women were using in the earlier observational studies. H T users from earlier decades often used conjugated equine estrogens alone. When a progestogen was used, it was typically used in an intermittent, cyclic fashion. It has been variously proposed that in the W H I the progestogen offset the potential benefits of estrogen; that a cyclic rather than the continuous combined regimen of EPT might have resulted in more benefit; or that bioidenfical estradiol and progesterone should have been used. 2. The beneficial effect seen in observational studies may have been the result of selection bias; healthy user effect; and failure to capture the early adverse events such as thrombosis, pulmonary emboli, myocardial infarction, and stroke among users. 3. The W H I H T Trial may have failed to find cardiac and cognitive benefits because H T was not initiated in all participants at the time of menopause. This theory holds that in order to see benefit in these two domains, one must begin H T prior to the development of atherosclerosis in the coronary and cerebral vasculature as well as prior to neurologic aging in the brain. These hypotheses have been explored elsewhere (40). Although there is evidence of no harm in the 50- to-59year-old women in the ET arm and a suggestion of benefit in some combinations of various cardiac outcomes, there was no statistically significant difference in the hazard ratios across age groups (11,14). Another scientific challenge lies in the different results seen in the two H T arms (see Fig. 45.2). An obvious difference between the two arms is the progestogen that was given along with estrogen to women who still had a uterus. However, progestogen was not the only difference between the groups. All the ET participants had undergone hysterectomy, and in some of them one or both ovaries had been removed. Furthermore, there were a number of differences between the ET and EPT groups at baseline (Table 45.3) (41). It is clear from the baseline
TABLE 45.2 Event Rates in Placebo Group of the ET and EPT Arms of the W H I Hormone Trial Event Stroke CHD rate CHD death Death Global index From references 10 and 11.
ET (5429)
EPT (8102)
0.32 0.54 0.16 0.78 1.90
0.21 0.30 0.06 0.53 1.51
617
CHAPTER 45 The Women's Health Initiative m Data and Discussion
TABLE 45.3 Baseline Characteristics of Participants in the ET and EPT Arms of the W H I Hormone Trial Characteristic Mean age (yrs) BMI (kg/m 2) 1st birth at age < 20 (%) Black (%) W h i t e (%)
E T (5310)
E P T (8506)
63.6 30.1 27.3 15.1 75.3
63.3 28.5 16.4 6.8 84.0
From reference 41.
data that the ET women were less healthy at baseline. During the course of the study, ET placebo group participants had higher rates of clinical events compared to the EPT placebo group (Table 45.2). Further research and analysis may be needed to address the different outcomes regarding invasive breast cancer from the EPT and ET arms. Ongoing data collection and in-depth analyses may provide better understanding of the findings from all three trials. The high retention rate over the 10 years of the original study is a tribute to the commitment of the participants to advancing the science of women's health.
VII. F U T U R E W H I REPORTS The principal papers for each arm of the W H I (10, 11,35-39) presented a summary of the primary and secondary endpoints for each trial. Subsequent publications have elaborated those findings and presented additional information. Data published to date represent a small fraction of the health information stored in the W H I database. The extensive demographic and health outcomes data for various ethnic groups make W H I a rich source of information on women's health. The repository of biospecimens provides opportunity for genomic and proteomic research that could lead to earlier identification of disease and disease risk factors. The 5-year WHI Extension Study (2005-2010) enhances the original study by providing long-term follow-up of participants with a unique opportunity to track the duration of benefits and risks associated with each intervention once the intervention is discontinued. Ongoing and new ancillary studies further expand the dividends from this program. The WHI database can be accessed via the Internet at www.whiscience.org or www.nhlbi.nih.gov/whi/index.html. The database is open to other scientists to further the research in women's health.
References 1. The Women's Health Initiative Study Group. Design of the Women's Health Initiative Clinical Trial and Observational Study. Controlled Clin Trials 1998;19:61-109. 2. Lufldn EG, Carpenter PC, Ory SJ, Malkasian GD, Edmonson JH. Estrogen replacement therapy: current recommendations. Mayo Clin Proc 1988;63:453-460. 3. Bush TL, Barrett-Connor E, Cowan LD, et al. Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the lipid research clinics program follow-up study. Circulation 1987;75: 1102-1109. 4. Sullivan JM, Vander Zwaag R, Hughes JP, et al. Estrogen replacement and coronary artery disease. Effect on survival in postmenopausal women. Arch Intern Med 1990;150:2557-2562. 5. Stampfer MJ, Colditz GA, Willett WC, et al. Postmenopausal estrogen therapy and cardiovascular disease: ten-year follow-up from the Nurses' Health Study. NEnglJMed 1991;325:756-762. 6. Fillit H, Weinreb H, Cholst I, et al. (1986). Observations in a preliminary open trial of estradiol therapy for senile dementia-Alzheimer's type. Psychoneuroendocrinology 1986;11:337- 345. 7. Honjo H, Ogino Y, Naitoh K, et al. In vivo effects by estrone sulfate on the central nervous system-senile dementia (Alzheimer's type). J Steroid Biochem 1989;34:521-525. 8. Christiansen C, Riis BJ. 17 Beta-estradiol and continuous norethisterone: a unique treatment for established osteoporosis in elderly women. J Clin Endocrinol Metab 1990;71:836- 841. 9. Ettinger B, Pressman A, Silver E Effect of age on reasons for discontinuation of hormone replacement therapy. Menopause 1999;7: 282-289. 10. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;321- 333. 11. The Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative Randomized Controlled Trial. JAMA 2004;291:1701-1712. 12. Cauley JA, Robbins J, Chen Z, et al., for the Women's Health Initiative Investigators. Effects of estrogen plus progestin on the risk of fracture and bone mineral density: The Women's Health Initiative Randomized Trial. JAMA 2003;290:1729-1738. 13. Manson JE, Hsia J, Johnson KC, et al., for the Women's Health Initiative Investigators. Estrogen plus progestin and the risk of coronary heart disease. N EnglJ Med 2003;349:523-534. 14. Hsia J, Langer R, Manson JE, et al., for the Women's Health Initiative Investigators. Conjugated equine estrogens and coronary heart disease. The Women's Health Initiative. Arch Intern Med 2006;166:357-365. 15. Wassertheil-Smoller S, Hen&ix SL, Limacher M, et al., for the WHI Investigators. Effect of estrogen plus progestin on stroke in postmenopausal women. JACMA2003;289:2673-2684. 16. Hulley S, Grady D, Bush T, et al., for the Heart and Estrogen/progestin Replacement Study (HERS) Research Group. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAM,4 1998;280:605-613. 17. Daly E, Vessey MP, Hawkins MM, et al. Risk of venous thromboembolism in users of hormone replacement therapy. Lancet 1996;348: 977-980. 18. Jick H, Derby LE, Myers MW, Vasilakis C, Newton KM. Risk of hospital admission for idiopathic venous thromboembolism among users of postmenopausal oestrogens. Lancet 1996;348:981-983. 19. Cushman M, Kuller LH, Prentice R, et al., for the Women's Health Initiative Investigators. Estrogen plus progestin and risk of venous thrombosis. JAMA 2004;292:1573-1580.
6
1
8
M
A
R
G
20. Chlebowski RT, Hendrix SL, Langer RD, et al., for the Women's Health Initiative Investigators. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women. JAM,4 2003;289:3243-3253. 21. Stefanick ML, Anderson GL, Margolis KL, et al., for the WHI Investigators. Effects of conjugated equine estrogens on breast cancer and mammography screening in postmenopausal women with hysterectomy. JAMA 2006;295;1647-1657. 22. Chlebowski RT, Chen Z, Anderson GL, et al. Ethnicity and breast cancer: Factors influencing differences in incidence and outcome. J Natl CancerInst 2005;97:439-448. 23. Chlebowski R, Wactawski-Wende J, Ritenbaugh C, et al., for the Women's Health Initiative Investigators. Estrogen plus progestin and colorectal cancer in postmenopausal women. N Eng! J Med 2004; 350:991-1004. 24. Rapp SR, Espeland MA, Shumaker SA, et al., for the WHIMS Investigators. Effect of estrogen plus progestin treatment on global cognitive function in postmenopausal women: the Women's Health Initiative Memory Study. A randomized controlled trial. JAMA 2003;289:2663-2672. 25. Shumaker SA, Legault C, Rapp SR, et al., for the WHIMS Investigators. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: The Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003 ;289:2651 - 2662. 26. Shumaker SA, Legault C, Kuller L, et al., for the Women's Health Initiative Memory Study. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: WHIMS. JAMA 2004;291:2947-2958. 27. Espeland MA, Rapp SR, Shumaker SA, et al., for the Women's Health Initiative Memory Study. Conjugated equine estrogens and global cognitive function in postmenopausal women: The Women's Health Initiative Memory Study. JAMA 2004;291:2959-2968. 28. Hays J, Ockene JK, Brunner RL, for the Women's Health Initiative Investigators. Effects of estrogen plus progestin on health-related quality of life. N EnglJ Med 2003;348:1839-1854. 29. Brunner RL, Gass M, Aragaki A, et al., for the Women's Health Initiative Investigators. Effects of conjugated equine estrogen on healthrelated quality of life in postmenopausal women with hysterectomy: results from the Women's Health Initiative randomized clinical trial. Arch Intern Med 2005;165:1976-1986.
E
R
Y
L.S. GASS
30. Barnabei VM, Cochrane BB, Aragaki AK, for the Women's Health Initiative Investigators. Menopausal symptoms and treatment-related effects of estrogen and progestin in the Women's Health Initiative. Obstet Gyneco12005;105:1063-1073. 31. McTiernan A, Martin CF, Peck JD, et al., for the Women's Health Initiative Mammogram Density Study Investigators. Estrogen plus progestin influence on mammogram density in healthy postmenopausal women in the Women's Health Initiative. J Natl Cancer Inst 2005;97:1366-1376. 32. McCully KS, Jackson S. Hormone replacement therapy and the bladder. J BR Menopause Soc 2004;10:30-32. 33. Hen&ix SL, Cochrane BB, Nygaard IE, et al., Effects of estrogen with and without progestin on urinary incontinence. JAMA 2005;293: 935-948. 34. Raz R, Stamm WE. A controlled trial of intravaginal estriol in postmenopausal women with recurrent urinary tract infections. N EnglJ Med 1993;329:753-756. 35. Prentice RL, Caan B, Chlebowsld RT, et al., Low-fat dietary pattern and risk of invasive breast cancer: The Women's Health Initiative Randomized Controlled Dietary Modification Trial. JAMA 2006;295;629-642. 36. Beresford SA, Johnson KC, Ritenbaugh C, et al., Low-fat dietary pattern and risk of colorectal cancer: the Women's Health Initiative Randomized Controlled Dietary Modification Trial. JAMA 2006;295:643-654. 37. Howard BV, Van Horn L, Hsia J, et al., Low-fat dietary pattern and risk of cardiovascular disease: the Women's Health Initiative Randomized Controlled Dietary Modification Trial. JAMA 2006;295:655-666. 38. Jackson RE), LaCroix AZ, Gass M, et al., for the Women's Health Initiative Investigators. Calcium plus vitamin D supplementation and the risk of fractures. NEnglJ Med 2006;354:669-683. 39. Wactawski-Wende J, Kotchen JM, Anderson GL, et al., for the Women's Health Initiative Investigators. Calcium plus vitamin D supplementation and the risk of colorectal cancer. N EnglJMed 2006;354:684-696. 40. Prentice RL, Langer R, Stefanick ML, et al., for the Women's Health Initiative Investigators. Combined postmenopausal hormone therapy and cardiovascular disease: toward resolving the discrepancy between observational studies and the Women's Health Initiative clinical trial. Am J Epidemio12005;162:404- 414. 41. Stefanick ML, Cochrane BB, Hsia J, et al. The Women's Health Initiative Postmenopausal Hormone Trials: overview and baseline characteristics of participants. Ann Epidemio12003;13:$78- $86.
2HAPTER 4(
Postmenopausal Hormone Therapy in the 21st Century: Reconciling Findings from Observational Studies and Randomized Clinical Trials KARIN B. M I C H E L S
Obstetrics and Gynecology Epidemiology Center, Department of Obstetrics, Gynecology and Reproductive Biology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115
JOANN E.
Divisionof Preventive Medicine, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02215
MANSON
I. INTRODUCTION
disease (CVD), and breast cancer (1-3). In this chapter we discuss the current state of knowledge regarding effects of H T and try to elucidate apparent discrepancies in results from observational studies and randomized clinical trials (RCTs), taking into account differences in methodology between the studies as well as biologic differences in the populations and drug regimens studied.
Recent years have seen a wealth of new data on possible effects of postmenopausal hormone therapy (HT). Some of these data appear inconsistent and contradictory, which has rendered clinical decision making more difficult. For decades, H T was thought to be beneficial not only to treat menopausal symptoms and to counter osteoporosis, but potentially to prevent cardiovascular disease. After the release of results from the Women's Health Initiative (WHI), physicians and patients became more reluctant to use HT. The weight of evidence appeared to shift to a negative balance for HT, particularly for the combination of estrogen and progestin: an increase in the risk of thromboembolic events, cardiovascular TREATMENT OF THE POSTMENOPAUSAL WOMAN
II. EFFECTS OF HT H T was introduced in the mid-20th century as an estrogen-only formulation. The sales of exogenous oral estrogen to counter menopausal symptoms was spurred by 619
Copyright 9 2007by Elsevier,Inc. All rights of reproductionin any formreserved.
620
MICHELS AND MANSON
publication of the bestseller F e m i n i n e F o r e v e r in 1968, which advertised H T as a designer drug promising "extended youth, .... femininity," and beauty (4). Despite the undisputed effectiveness of H T in treating the symptoms of menopause, enthusiasm for the use of H T dampened when it was found to substantially increase the risk of endometrial cancer. Opposing estrogen with progesterone eliminated this unwanted effect. Estrogen receptors can be found in many tissues, in particular in the endometrium, breast, and ovary. Exogenous estrogen promotes cell proliferation and increases the risk of endometrial cancer, breast cancer, and probably ovarian cancer. Although progesterone counters the proliferative effect of estrogen in the endometrium, it may enhance proliferation in the breast. Given the higher absolute risk of breast cancer than endometrial cancer, there is a greater increase in the incidence of both cancers combined with the estrogen plus progesterone combination therapy than with estrogen alone among women with a uterus (5). Exogenous estrogen also has several vascular and metabolic effects, including reducing plasma low-density lipoprotein (LDL) cholesterol levels and raising high-density lipoprotein (HDL) levels, lowering lipoprotein (a) levels and reversing postmenopausal increases in fibrinogen and plasminogen-activator inhibitor type 1 (6). Estrogen also improves glucose metabolism (7), boosts endothelial vascular function (6), and reduces plasma homocysteine levels (8). These effects of estrogen would predict a decrease in the risk of coronary heart disease (CHD) associated with H T use. However, exogenous estrogen also has adverse effects on related biomarkers, including increasing triglyceride levels; promoting coagulation through increases in factor VII, prothrombin fragments 1 and 2, and fibrinopeptide A; raising levels of the inflammatory marker C-reactive protein (6); and promoting the production of metalloproteinases, degradative enzymes important in the destabilization and rupture of plaques (9). Thus, estrogen's effects on vascular function are diverse and complex.
III. THE EFFECT OF HT ON CARDIOVASCULAR DISEASE INCIDENCE A. Data from Observational Studies Results from dozens of observational studies conducted during the past three decades have fairly consistently suggested a substantially lower risk of coronary heart disease (CHD) events among women using H T compared with those who do not use HT. A meta-analysis of 40 epidemiologic studies suggested a 50% reduction in the risk of CHD associated with current estrogen use (10). Because combination therapy with
estrogen and progestin has become prevalent only in the past two decades, most data from long-term observational studies are based on unopposed estrogen use. Data on the effects of opposed estrogen use from observational studies are limited. The Nurses' Health Study, with a 30-year follow-up, is one of the largest prospective cohorts for which information on HT use was collected. Among 70,533 postmenopausal women with no history of CVD, the relative risk (RR) for CHD was 0.55 (95% CI 0.45-0.68) among women taking oral conjugated estrogen alone during 808,825 person-years of follow-up after adjusting for coronary risk factors and 0.64 (95% CI 0.52-0.71) among women using estrogen plus progestin compared with never users (11). Among 2489 postmenopausal women with prior CHD, the RR for a second coronary event was 1.25 (95% CI 0.78-2.00) during the first year of H T use and 0.65 (95% CI 0.45-0.95) during the total follow-up period of 20 years compared with never users of l i T (12).
B. Randomized Clinical Trials on Secondary Prevention Six randomized clinical trials have addressed the question whether HT use may confer CHD benefits among women with prior heart disease (Table 46.1). Although clinical events data are not available from all these secondary prevention trials, these studies have generally not confirmed the inverse association between H T use and CHD reported in observational studies. In the Heart and Estrogen/Progestin Replacement Study (HERS), including 2763 women with documented CHD, the RR for a secondary CHD event was 0.97 (95% CI 0.82-1.14) among women using 0.625 mg of oral conjugated equine estrogen (CEE) plus 2.5 mg of medroxyprogesterone (MPA) compared with placebo (13). Participants randomized to HT experienced a 50% higher incidence of secondary CHD events during the first year of the trial compared with women receiving placebo, but this difference was offset in later years (13,14). The Estrogen Replacement and Atherosclerosis Trial (ERA) and the Women's EstrogenProgestin Lipid-Lowering Hormone Atherosclerosis Regression Trial (WELL-HART) were three-arm trials comparing the effects of estrogen, estrogen plus MPA, and placebo on atherosclerotic progression. No differences were found between groups in the rates of angiographically determined progression ofstenoses (15,16). Similarly, the Papworth Hormone Replacement Therapy Atherosclerosis Study, which evaluated the effects of transdermal estradiol with and without norethindrone compared with placebo, found no significant difference in secondary CHD events in the three groups (17). In the Women's Angiographic Vitamin and Estrogen Trial (WAVE), which tested CEE, CEE plus MPA, and placebo, and in the Estrogen in the Prevention of Reinfarction Trial (ESPRIT), which tested estradiol valerate and
621
CHAPTER 46 Postmenopausal Hormone Therapy in the 21st Century TABLE 46.1
Results of Randomized Clinical Trials of Postmenopausal Hormone Therapy and Coronary Heart Disease
Trial
Study group
Age range at entry (mean) in years
Treatment
Mean duration in years
Results
Secondary Prevention Heart and Estrogen/ Progestin Replacement Study 13
2763 women with documented CHD
44-79 (66.7)
0.625 mg of oral CEE plus 2.5 mg of MPA or placebo
6.8
Primary CHD events: RR = 0.97 (95% CI 0.82-1.14)
Estrogen Replacement and Atherosclerosis Trial 15
309 women with CHD
42-80 (65.8)
3.2
No significant differences in rates of angiographic stenoses or clinical events
Women's EstrogenProgestin LipidLowering Hormone Atherosclerosis Regression Trial 16
226 women with at least one coronary-artery lesion
<75 (63.5)
3.3
No significant difference in progression of angiographically assessed coronaryartery atherosclerosis
Papworth HRT Atherosclerosis Study a7
255 women with CHD
55 + (66.5)
2.6
CHD events: RR = 1.49 (95% CI 0.93-2.36)
Women's Angiographic Vitamin and Estrogen Trial TM
423 women with CHD
Mean: 65
0.625 mg of oral CEE, 0.625 mg of oral CEE plus 2.5 mg of MPA, or placebo. I mg of oral micronized 1713-estradiol alone, or 1713-estradiol plus sequentially administered medroxyprogesterone acetate or placebo Transdermal 1713estradiol alone or with cyclic norethindrone, or placebo 0.625 mg of oral CEE plus 2.5 mg of MPA, or placebo
2.8
Estrogen in the Prevention of Reinfarction Trial 19
1017 women with CHD
50-69 (62.6)
2 mg estradiol valerate, or placebo
2
Slightly greater angiographic progression with HT than placebo; CVD events: HR = 1.9 (95% CI 0.97-3.6) CHD events: RR - 0.99 (95% CI 0.70-1.41) All-cause mortality: RR = 0.79 (95% CI 0.50-1.27)
4124 women enrolled in RCTs of HRT 16,608 women at usual risk of CHD
Not provided
Several formulations of HRT or placebo or other therapy 0.625 mg of oral CEE plus 2.5 mg of MPA, or placebo
<3
P r i m a r y Prevention Overview of 22 HRT Trials 2~ Women's Health Initiative Estrogenplus-Progestin Trial21 Women's Health Initiative Estrogen-only Trial23
10,739 women at usual risk of CHD
50-79
50-79
0.625 mg of oral CEE, or placebo
5.2
6.8
CVD events: RR = 1.39 (95% CI 0.48-3.95) CHD events: HR = 1.29 (95% CI 1.02-1.63) CHD events: HR = 0.91 (95% CI 0.75-1.12)
CEE, conjugated equine estrogen; MPA, medroxyprogesterone acetate.
placebo, no evidence of cardiovascular benefit was found for any of the treatment regimens (18,19) among women with prior myocardial infarction (MI). The follow-up in these secondary prevention trials was 6.8 years or less.
C. Randomized Clinical Trials on Primary Prevention Randomized data on primary prevention of C H D are available only from the W H I and a meta-analysis of 22 small, short-term RCTs of H T including a total of
4124 women (see Table 46.1). The meta-analysis included RCTs testing several formulations of H T and provided an overall RR for C V D events of 1.39 (95% CI 0.48-3.95) (20). The W H I included generally healthy postmenopausal women ages 50 to 79 years old: 16,608 women with intact uterus were randomized to 0.625 mg C E E plus 2.5 mg of M P A or placebo, and 10,739 women who had undergone a hysterectomy were randomized to estrogen only or placebo. In June 2002 the estrogen-plus-progestin arm of the W H I R C T was halted prematurely after 5.2 years of follow-up due to an overall unfavorable risl~enefit ratio and a significant increase in the risk of breast cancer. Participants randomized
622
MICHELS AND MANSON
to estrogen plus progestin were advised to discontinue their medication. Women receiving estrogen plus progesfin were 29% more likely to develop C H D than women on placebo (RR = 1.29; 95% CI 1.02-1.63) (21,22) (Table 46.2). The increase in C H D incidence associated with H T was most pronounced during the first years of the trial (year 1: H R 1.81; 95% CI 1.09-3.01) and diminished with longer follow-up (6+ years: H R - 0.70; 95% CI 0.42-1.14) (22). In early 2004, the estrogen-only arm of the W H I was also stopped prematurely due to an increased risk of stroke and a lack of overall benefit of estrogen on a combined endpoint of CHD, stroke, pulmonary embolism, breast cancer, endometrial cancer, colorectal cancer, hip fracture, and death due to other causes (23). During 6.8 years of follow-up, women randomized to estrogen did not experience a significant reduction of their C H D risk (RR = 0.91; 95% CI 0.75-1.12) (see Table 46.2). In both trial arms, there was an apparent reduction in C H D risk among women followed for the longest duration (6 or more years in the estrogen plus progestin group and 7 or more years in the estrogen-only arm).
IV. OTHER DISEASE END POINTS Observational studies and the estrogen plus progestin arm of the W H I have had generally consistent findings for n o n - C H D clinical outcomes, while some results differed for the W H I estrogen only arm. Observational studies had indicated an increase in the incidence of ischemic stroke associated with H T use (RR = 1.26; 95% CI 1.00-1.61) (11), which was similar in the W H I (for estrogen plus progestin [E +P]: H R = 1.44; 95% CI 1.09-1.90; for estrogen only: HR=1.55; 95% CI 1.19-2.01) (24,25). The risk of hemorrhagic stroke was not affected by H T in either obser-
TABLE 46.2
vational studies (RR = 0.93; 95% CI 0.64-1.34) (11), nor the W H I (for E + P : H R = 0.82; 95% CI 0.43-1.56; for estrogen only: H R = 0.64; 95% CI 0.35-1.18) (24,25). The risk for venous thrombosis was approximately doubled with estrogen plus progestin in both observational studies (26) and the W H I (27), whereas the association was insignificantly elevated among estrogen-only users (23,26). Hip fractures were uniformly reduced by 30% to 40% in observational studies and the W H I (13,21,23). Breast cancer incidence was significantly increased in observational studies such as the Nurses' Health Study (NHS) among women who used estrogen alone (RR = 1.32; 95% CI 1.14-1.54) compared with women who never used H T (28) and the Breast Cancer Detection Demonstration Project (BCDDP) among estrogen users (RR = 1.2; 95% CI 1.0-1.4) (29), especially for long duration of use. H T use for less than 5 years did not seem to increase breast cancer risk (28,29). Progestin added to estrogen increased this risk further (RR = 1.41; 95% CI 1.15-1.74 in NHS and RR = 1.4; 95% CI 1.1-1.8 in the BCDDP) (28,29). In the W H I , a risk increase was confirmed in the estrogen plus progestin arm (HR = 1.26; 95% CI 1.00-1.59) after 5.2 years of use (21), but an unexpected borderline reduction in risk was observed in the estrogen-only arm with an H R of 0.77 (95% CI 0.59-1.01) during 6.8 years of follow-up (23). Colorectal cancer risk was reduced in a meta-analysis of 18 observational studies among women on H T (RR = 0.80; 95% CI 0.74-0.86) (30) and similarly among participants of the W H I randomized to estrogen plus progestin after relatively short-term use (HR = 0.63; 95% CI 0.43-0.92) (21), but not among women randomized to estrogen alone (HR = 1.08; 95% CI 0.75-1.55) (23). Some of the differences may be due to the shorter follow-up in the W H I , which may have impacted more on the estrogen-only arm.
Women's Health Initiative: Summary of the Hormone Trials
Estrogen + progestin (N = 16,608; ages 50- 79; mean duration 5.2 years)
Estrogen alone (N = 10,739; ages 50-79; mean duration 6.8 years)
Endpoint
RR
(95% CI) a
RR
95% CI) a
Coronary heart disease Stroke Pulmonary embolism Breast cancer Colorectal cancer Hip fracture Total mortality Global indexb
1.29 1.41 2.13 1.26 0.63 0.66 0.98 1.15
(1.02-1.63) (1.07-1.85) (1.39- 3.25) (1.00-1.59) (0.43-0.92) (0.45-0.98) (0.82 - 1.18) (1.03-1.28)
0.91 1.39 1.34 0.77 1.08 0.61 1.04 1.01
(0.75-1.12) (1.10-1.77) (0.87-2.06) (0.59-1.02) (0.75-1.55) (0.41-0.91) (0.88 - 1.22) (0.91-1.12)
"95% CIs are nominal. bFirst occurrence of CHD, stroke, pulmonaryembolism,breast cancer,colorectal cancer,endometrial cancer,hip fracture, death from other causes. Data from: Writing Group for the WHI Investigators.JAMA 2002;288:321-333; The Women's Health Initiative Steering Committee.JAMA 2004;291:1701-1712.
CHAPTER 46 Postmenopausal Hormone Therapy in the 21st Century
V. POSSIBLE EXPLANATIONS FOR DISCORDANT FINDINGS BETWEEN OBSERVATIONAL STUDIES AND RCTS Both methodologic and biologic factors may have contributed to the divergent findings on H T and C H D in observational studies and RCTs. The relevant issues are briefly reviewed in the following sections.
A. Methodologic Explanations 1. CONFOUNDING/SELECTIONBIAS Confounding is one of the most serious threats to the validity of observational studies. In epidemiologic studies, women who choose to use H T have been found to be different in many characteristics from women who choose not to use HT. H T users generally pursue a healthier lifestyle; are leaner, more physically active, and less likely to smoke; have better access to medical care; and have a higher socioeconomic status than postmenopausal women who do not use H T (31). Self-selection of healthier women into the H T user group (or prescription of H T to women with health-seeking behaviors by physicians) could generate confounding and lead to a "healthy-user" bias. Although some confounding factors can be controlled for in statistical analyses, residual confounding is likely to remain. Furthermore, because many traits of a healthy lifestyle cannot be readily ascertained, unmeasured confounding is probable. In the Nurses' Health Study, the age-adjusted and the covariate-adjusted HR for Coronary heart disease Observational studies11 WHt2~ Stroke Observational studies11 Writ
623 C H D among current users of unopposed estrogen differed slightly, indicating the presence of confounding, but a strong inverse association remained after covariate adjustment (ageadjusted HR 0.54; 95% CI 0.46-0.62; covariate-adjusted HR 0.61; 95% CI 0.52-0.71) (11). Furthermore, indicators of improved access to medical care and compliance with other medical treatments were not measured and therefore not controlled. Conversely, the concordance of results between observational studies and the W H I for most end points other than CHD, including stroke and venous thromboembolism, which have several risk factors in common with CHD, suggests that major confounding is unlikely to be the primary explanation for the discrepant findings for C H D (Fig. 46.1).
2. LACK o r COMPLIANCE
At the time the estrogen plus progestin component of the W H I was stopped, 42% of the women randomized to estrogen plus progestin had discontinued their medication, whereas 10.7% of women randomized to placebo initiated hormone therapy use during the study period (21). In the estrogen-only arm, the discontinuation rate at the time of study termination was 53.8% among the women randomized to estrogen, and 9.1% of women randomized to placebo had initiated estrogen use by year 6 of follow-up (23). The intention-to-treat analysis is, thus, diluted by noncompliance or only short-term compliance with treatment. Longer exposure to H T may be required to confer C H D protection. In most observational studies, women using H T tend to use it for a longer time than in the WHI.
41k v
,P
Pulmonary embolism Observational studies" WIll Hip fractures Observational studies "" WHI Breast cancer Observational studies29 WIll
Colorectal cancer
=-.-
Observational studies 3~ WIll
0
0.5
1
1.5
2
2.5
3
3.5
4
Relative Risk
FIGURE 46.1 Comparison of results from observational studies of postmenopausal hormone therapy and the Women's Health Initiative Estrogen Plus Progesfin Trial. *Grodstein F, Stampfer MJ, Goldhaber SZ, et al. Prospective study of exogenous hormones and risk of pulmonary embolism in women. Lancet 1996;348:983-987. **Grady D, Rubin SM, Petitti DB, et al. Hormone therapy to prevent disease and prolong life in postmenopausal women. Ann Intern Med 1992;117:1016-1037. (From Michels KB, Manson JE. Postmenopausal hormone therapy: a reversal of fortune. Circulation 2003;107:1830-1833)
624
MICHELS AND MANSON
3. FAILURE TO CAPTURE EARLY EVENTS
In prospective cohort studies, participants may change their exposure status during follow-up. Some prospective cohorts assess H T only at baseline and do not update actual use during follow-up. In the NHS, H T user status and type of hormone use is assessed every 2 years. If a clinical event occurs soon after H T initiation (in NHS within the 2-year time interval before change of H T exposure status is registered), it could be incorrectly classified as occurring during a "nonuser" interval. This misclassification may be more problematic for C H D than for other events, if H T increases early C H D events as seen in HERS and WHI, whereas delayed or persistently elevated risks were observed for stroke, venous thromboembolism, and breast cancer. However, in the NHS, only a tiny fraction of C H D events occurred during these intervals when H T exposure status changed, so this methodologic limitation is unlikely to explain the risk reduction observed.
B. Biologic Explanations 1. STUDY POPULATION
A number of fundamental characteristics differ among women included in observational studies and clinical trials. Participants in an observational study make the decision to take H T together with their physician based on a variety of factors, the most important of which may be the severity of their menopausal symptoms. Women with severe menopausal symptoms may also have the lowest endogenous estrogen levels. The women recruited to take part in the W H I had to be largely free of menopausal symptoms and neutral about H T use as they had to agree to be assigned by chance to take it or not to take it. This randomization scheme does not reflect clinical practice. Also, differences in age and time since menopause may be important. W H I participants were 50 to 79 years old when they were randomized to HT, and many of them had never used H T before. Because 67% of women in the W H I were age 60 years or older at the start of the trial, most participants started H T many years after the onset of their menopause. In contrast, participants of the NHS were 30 to 55 years old at baseline and 80% of H T users initiated H T use within 2 years of menopause. In most observational studies, the majority of H T users begin treatment close to the time of menopause. Thus, there is little overlap in the study populations of observational studies and RCTs with respect to age, menopausal symptoms, and time since menopause when H T is started (Fig. 46.2). It is likely that RCT participants who initiated H T a decade or so after their menses ceased had more advanced atherosclerosis than the younger participants. It is possible that exogenous estrogen has beneficial cardiovascular effects in
FIGURE 46.2 Time since menopause and effect of starting hormone therapy. (From Manson JE, Bassuk SS, Harman SM, et al. Postmenopausal hormone therapy: new questions and the case for new clinical trials. Menopause 2006;13:139-147.)
women who have relatively intact endothelial function and minimal atherosclerosis at the onset of menopause. In contrast, exogenous estrogen may exert prothrombotic and proinflammatory effects with ensuing plaque rupture and thromboocclusion among women with more advanced atherosclerotic plaques (see Fig. 46.2). Nonhuman primate data lend further support to this idea. Although exogenous estrogens administered to cynomolgus monkeys immediately after oophorectomy reduced atherosclerotic plaques by 50%, no effect was seen when hormones were first administered 2 years after oophorectomy (equivalent of 6 years past menopause in humans) (32). Similarly, in the Estrogen in the Prevention of Atherosclerosis Trial (EPAT), which randomized treatment with oral micronized 1713-estradiol versus placebo among postmenopausal women 45 years or older without preexisting cardiovascular disease, progression of subclinical atherosclerosis of the carotid artery determined by ultrasonograms was slower among women taking 1713-estradiol (33). Whether differences in age and time since menopause at initiation of H T may explain the apparent differences in resuits from observational studies and RCTs can be directly explored in subgroup analyses. In the WHI, the RR of C H D among women on estrogen only compared with women on placebo increased with the age at which H T was initiated: Hsia and colleagues report for participants aged 50 to 59 years, 60 to 69 years, and 70 to 79 years at baseline hazard ratios for C H D of 0.63 (95% CI 0.36-1.08), 0.94 (95% CI 0.71-1.24), and 1.11 (95% CI 0.82-1.52), respectively, and for a combined end point of myocardial infarction, coronary death, coronary artery bypass graft, or percutaneous coronary intervention of 0.66 (95% CI 0.44-0.97), 1.02 (95% CI 0.83-1.25), and 1.08 (95% CI 0.85-1.38), respectively (34). Although similar trends with age were not observed in the estrogen-plus-progestin arm of the W H I (35,36), the RRs for
CHAPTER 46 Postmenopausal Hormone Therapy in the 21st Century C H D increased with time since menopause: Women who initiated H T within 10 years of onset of menopause had an RR of 0.89 (95% CI 0.55-1.46); within 10 to 19 years, 1.22 (95% CI 0.85-1.77); and after 20 years or more, 1.71 (95% CI 1.18-2.49) (22). Moreover, an analysis of the NHS data on H T use according to time since menopause revealed a reduced risk of C H D among women who initiated H T near menopause for unopposed estrogen (RR = 0.66; 95% CI 0.54-0.80) and for estrogen and progestin (RR = 0.72; 95% CI 0.56-0.92), whereas no significant association was found for women who initiated H T 10 or more years after menopause (RR = 0.87; 95% CI 0.69-1.10 for estrogen alone; RR = 0.90; 95% CI 0.62-1.29 for estrogen and progestin) (37). Hence timing of initiation of H T may be an important modifier of the effect of H T on C H D risk: H T used at the beginning of menopause may reduce atherosclerotic progression and reduce C H D risk, whereas initiation many years into menopause may precipitate plaque rupture or thrombosis in vessels with advanced atherosclerotic lesions that had formed during previous years. The difference in timing of H T initiation in observational studies, which reflect clinical practice with women opting to use H T soon after menopausal symptoms first arise, and the W H I , which assigned H T many years after onset of menopause, may have contributed to the divergent findings from the two types of studies. A recent meta-analysis of 23 RCTs of H T confirmed a reduced incidence of C H D events in younger women (within 10 years of onset of menopause or less than age 60 years) (odds ratio [OR] = 0.68; 95% CI 0.48-0.96), but not in older women (OR = 1.03; 95% CI 0.91-1.16) (38). Two trials are under way to address the importance of timing of l i T initiation. The Early versus Late Intervention Trial with Estradiol (ELITE) is designed to test the hypothesis of timing of initiation of estrogen therapy in the reduction of subclinical atherosclerosis progression and cognitive decline in women less than 6 years versus greater than 10 years postmenopausal (www.usc.edu/medicine/aru). E L I T E is a 2 • 2 randomized double-blind placebocontrolled single-center serial arterial imaging trial of healthy postmenopausal women without CVD or diabetes mellitus conducted by the University of Southern California. The Kronos Early Estrogen Prevention Study (KEEPS) is a multicenter, 5-year, randomized, double-blind, placebocontrolled clinical trial of oral estrogen plus progesterone and transdermal estradiol plus oral progesterone (39). Primary end point of the KEEPS trial is the prevention of atherosclerosis progression.
VI. CONCLUSIONS Recent randomized clinical trials have painted a less favorable picture of the impact of H T on overall health than observational studies had suggested. Although many
625 associations between H T and diseases end points were consistent in the RCTs and observational studies, results for C H D differed substantially. These differences might be explained at least in part by methodologic differences and by biologic differences in the study populations. Although observational studies are susceptible to selection and confounding biases, it is likely that differences in characteristics of the study populations of RCTs and observational studies are of even greater importance. The treatment regimen laid out in the W H I differed from clinical practice in randomizing women with no or moderate menopausal symptoms to first use of l i T a decade or more after menopause. Time since menopause at initiation of H T and severity of menopausal symptoms may be important modulators of the effect of H T on C H D . Women a decade or more into their menopause may be unlikely to derive cardiovascular benefit from H T and may even be at an increased risk of a thromboembolic event precipitated by treatment. Current decision making about H T use should be guided by the severity of menopausal symptoms at the onset of menopause (Table 46.3). Such therapy, however, should involve the lowest effective dose and the shortest duration of treatment necessary to relieve symptoms. Although H T remains the most effective treatment to counter the symptoms of menopause, current guidelines discourage long-term use for chronic disease prevention.
TABLE 46.3 Women's Health Initiative: Attributable Risk Summary for Estrogen Plus Progestin and Estrogen Alone Per 10,000 women per year on estrogen + progestin
Per 10,000 women per year on estrogen alone
Risks: 7 more women with CHD 8 more women with stroke 8 more women with pulmonary embolism 8 more women with breast cancer
Risks: 12 more women with stroke Benefits: 6 fewer women with hip fractures
Benefits: 6 fewer women with colorectal cancer 5 fewer women with hip fractures
Null (not statistically significant): CHD, pulmonary embolism, breast cancer, colorectal cancer
Net Effect: 19-20 extra adverse events
Net Effect: 2 extra adverse events
Data from: Writing Group for the WHI Investigators.JdMd 2002;288:321-333; The Women's Health Initiative Steering Committee. JdMd 2004; 291:1701-1712.
626
MICHELS AND MANSON
References 1. Michels KB, Manson JE. Postmenopausal hormone therapy: a reversal of fortune. Circulation 2003;107:1830-1833. 2. Michels KB. Hormone replacement therapy in epidemiologic studies and randomized clinical trialsmare we checkmate? Epidemiology 2003; 14:3-5. 3. Grodstein F, Clarkson TB, Manson JE. Understanding the divergent data on postmenopausal hormone therapy. N EnglJ Med 2003;348: 645 -650. 4. Wilson R. Feminineforever. New York: J.B.Lippincott Company, 1966. 5. Beral V, Bull D, Reeves G. Endometrial cancer and hormonereplacement therapy in the million women study. Lancet 2005;365: 1543-1551. 6. Chae C, Manson J. Postmenopausal hormone therapy. Philadelphia: Saunders/Elsevier, 2004:349- 363. 7. Kanaya AM, Herrington D, Vittinghoff E, et al. Glycemic effects of postmenopausal hormone therapy: the heart and estrogen/progestin replacement study. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 2003;138:1-9. 8. Chiantera V, Sarti CD, Fornaro F, et al. Long-term effects of oral and transdermal hormone replacement therapy on plasma homocysteine levels. Menopause 2003;10:286-291. 9. Wakatsuki A, Ikenoue N, Shinohara K, Watanabe K, Fukaya T. Different effects of oral and transdermal estrogen replacement therapy on matrix metalloproteinase and their inhibitor in postmenopausal women. Arterioscler Thromb VascBio12003;23:1948-1949. 10. Grodstein F, Stampfer MJ. The epidemiology ofpostmenopausal hormone therapy and cardiovascular disease. New York: Marcel Dekker, 2002: 67-78. 11. Grodstein F, Manson JE, Colditz, GA, et al. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med 2000;133:933-941. 12. Grodstein F, Manson JE, Stampfer MJ. Postmenopausal hormone use and secondary prevention of coronary events in the nurses' health study. A prospective, observational study. Ann Intern Med 2001; 135:1-8. 13. Grady D, Herrington D, Bittner V, et al. Cardiovascular disease outcomes during 6.8 years of hormone therapy: heart and estrogen/ progestin replacement study follow-up (HERS II). JAMA 2002;288: 49-57. 14. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and estrogen/progestin replacement study (HERS) research group. JAMA 1998;280:605-613. 15. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N EnglJ Med 2000;343:522-529. 16. Hodis HN, Mack WJ, Azen SP, et al. Hormone therapy and the progression of coronary-artery atherosclerosis in postmenopausal women. N EnglJ Med 2003;349:535- 545. 17. Clarke SC, Kelleher J, Lloyd-Jones H, Slack M, Schofiel PM. a study of hormone replacement therapy in postmenopausal women with ischaemic heart disease: the Papworth HRT atherosclerosis study. Br J Obstet Gynaeco12002;109:1056-1062. 18. Waters DD, Alderman EL, Hsia J, et al. Effects of hormone replacement therapy and antioxidant vitamin supplements on coronary atherosclerosis in postmenopausal women: a randomized controlled trial. JAMA 2002;288:2432-2440.
19. Cherry N, Gilmour K, Hannaford P, et al. Oestrogen therapy for prevention of reinfarction in postmenopausal women: a randomised placebo controlled trial. Lancet 2002;360:2001-2008. 20. Hemminld E, McPherson K. Impact of postmenopausal hormone therapy on cardiovascular events and cancer: pooled data from clinical trials. BMJ 1997;15:149-153. 21. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the women's health initiative randomized controlled trial. JAMA 2002;288:321 - 333. 22. Manson JE, Hsia J, Johnson KC, et al. Estrogen plus progestin and the risk of coronary heart disease. NEnglJ Med 2003;349:523-534. 23. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the women's health initiative randomized controlled trial.JAMA 2004;291: 1701-1712. 24. Wassertheil-Smoller S, Hendrix SL, Limacher M, et al. Effect of estrogen plus progestin on stroke in postmenopausal women: the women's health initiative. A randomized trial. JAMA 2003;289:2673-2684. 25. Hendrix SL, Wassertheil-Smoller S, Johnson KC, et al. Effects of conjugated equine estrogen on stroke in the Women's Health Initiative. Circulation 2006;113:2425- 2434. 26. Smith NL, Heckbert SR, Lemaitre RN, et al. Esterified estrogens and conjugated equine estrogens and the risk of venous thrombosis. JAMA 2004;292:1581-1587. 27. Cushman M, Kuller LH, Prentice R, et al. Estrogen plus progestin and risk of venous thrombosis. JAMA 2004;292:1573-1580. 28. Colditz GA, Hankinson SE, Hunter DJ, et al.The use of estrogens and progestins and the risk of breast cancer in postmenopausal women. NEnglJMed 1995;332:1589-1593. 29. Schairer C, Lubin J, Troisi R, et al. Menopausal estrogen and estrogenprogestin replacement therapy and breast cancer risk. JAMA 2000;283: 485-491. 30. Grodstein F, Newcomb PA, Stampfer MJ. Postmenopausal hormone therapy and the risk of colorectal cancer: a review and meta-analysis. AmJMed 1999;106:574-582. 31. Matthews KA, Kuller LH, Wing RR, et al. Prior to use of estrogen replacement therapy, are users healthier than nonusers? Am J Epidemiol 1996;143:971-978. 32. Mikkola TS, Clarkson TB. Estrogen replacement therapy, atherosclerosis, and vascular function. Cardiovasc Res 2002;53:605-619. 33. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 2001;135:939-953. 34. Hsia J, Langer RE), Manson JE, et al. Conjugated equine estrogens and coronary heart disease. The Women's Health Initiative. Arch Intern Med 2006;166:357-365. 35. Speizer FE. Risks of postmenopausal hormone replacement. JAMA 2002;288:2819-2820; author reply 2823-2814. 36. Radford N, Church T. Risks of postmenopausal hormone replacement. JAMA 2002;288:2819; author reply 2823-2814. 37. Grodstein F, Manson JE, Stampfer MJ. Hormone therapy and coronary heart disease: the role of time since menopause and age at hormone initiation. J Women'sHealth 2006;15:35-44. 38. Salpeter SR, Walsh JME, Greyber E, et al. Coronary heart disease events associated with hormone therapy in younger and older women. J Gen Intern Med 2006;21:363-366. 39. Harman SM, Brinton EA, Cedars M, et al. KEEPS: The Kronos Early Estrogen Prevention Study. Climacteric 2005;8:3-12.
~ H A P T E R 4,
Morbidity and Mortality Changes with Hormone Therapy ANNLIA PAGANINI-HILL
Departmentof Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
Lower estrogen levels after menopause can produce effects on many organs of the body. Over the years, the idea that postmenopausal women might enhance their later years and prevent or delay some chronic diseases of aging by use of estrogen therapy gained acceptance. This was supported by accumulating evidence from observational studies, clinical trials with intermediate outcomes, and experimental studies in animals and isolated cells. Physiologic mechanisms provided biologic plausibility to the theory. Recent findings of the Women's Health Initiative (WHI) clinical trials that hormone therapy (HT) (1,2) increased risk of cardiovascular disease have stimulated much discussion among researchers, medical practitioners, the news media, and perimenopausal and postmenopausal women about the risks and benefits of such therapy. Although the WHI confirmed many of the associations of H T reported by previous observational studies, the cardiovascular finding was unexpected. Randomized, double-blind, placebo-controlled trials, such as the W H I trials, are generally considered the gold standard for testing treatment effects. Nonetheless, each type of study and each individual study has its own strengths and weaknesses. It is important to consider the entire body of evidence when evaluating the effects of HT. This paper summarizes the effects of estrogen plus progestin (combined hormone therapy [CHT]) and unopposed estrogen therapy (ET) on chronic diseases, comparing the findings of the WHI trials (1-14) with those of other clinical trials and observational studies. Specific reference is made to two well-known and reported prospective cohort studies ~ the Leisure World Cohort Study (15-25) and the Nurses' Health T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
Study (26-35). The results for the major clinical outcomes from these three studies are given in Table 47.1.
I. T H E S T U D I E S
A. The Women's Health Initiative With its hormone trials of C H T (0.625 mg conjugated equine estrogen [CEE] plus 2.5 mg medroxyprogesterone acetate [MPA] daily) (1) and ET (0.625 mg CEE daily) (2), WHI sought to examine the effects of H T in healthy postmenopausal women. Between 1993 and 1998, W H I recruited 27,347 women, ages 50 to 79 years; 16,608 with intact uterus were randomly assigned to receive C H T or placebo, and 10,739 hysterectomized women were randomized to receive ET or placebo. The C H T trial was stopped early (after a mean of 5.2 years of follow-up) because of adverse effects in cardiovascular outcomes and breast cancer. Likewise, the ET trial was stopped (after a mean of 6.8 years) because of increased risk of stroke and the likelihood that neither cardioprotection nor breast cancer risk would be demonstrated in the remaining intervention period.
B. The Leisure World Cohort Study The Leisure World Cohort Study (LWCS) is an ongoing prospective study designed, in part, to evaluate the risks and benefits of ET in a community setting. In 1981 627
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
628
ANNLIA PAGANINI-HILL
TABLE 47.1 Relative Risk (95% Confidence Interval) of Disease Outcomes for Women's Health Initiative (WHI) Clinical Trials, Leisure World Cohort Study (LWCS), Nurses' Health Study (NHS), and Meta-Analyses WHI-CHT
WHI-ET
Cancer Endometrial
0.81 (0.48-1.36) (3)
Breast
1.24 (1.01-1.54) (4)
0.77 (0.59-1.01)(2)
Colorectal
0.56 (0.38-0.81)(5)
1.08 (0.75-1.55)(2)
Fractures Hip
0.67 (0.47-0.96)(6)
0.61 (0.41-0.91)(2)
Vertebral
0.66 (0.44-0.98) (1)
0.62 (0.42-0.93)(2)
Cardiovascular disease CHD
1.24 (1.00-1.54) (8)
0.91 (0.75-1.12) (2)
Stroke
1.31 (1.02-1.68) (9)
1.39 (1.10-1.77) (2)
Venous thromboembolic disease
1.33 (0.99-1.79)(2)
Dementia
2.06 (1.57-2.70)(7) 4.01 first year 2.05 (1.21-3.48)(10)
1.49 (0.83-2.66)(12)
Death
0.98 (0.82-1.18)(1)
1.04 (0.88-1.22) (2)
Wrist
aMyocardialinfarction. bpulmonaryembolism.
to 1985, health surveys, including detailed questions on use of estrogen during and after menopause, were mailed to the residents of Leisure World Laguna Hills, a California retirement community. Residents are fairly homogeneous, being predominantly white, moderately affluent, and well educated. Cohort members are followed by periodic
resurvey, review of local hospital discharge records, and determination of vital status. Data from 8850 women (median age at baseline = 73 years) who provided information on estrogen use have been analyzed for the effects of ET on morbidity and mortality from several chronic diseases of the elderly.
CHAPTER 47 Morbidity and Mortality Changes with Hormone Therapy
LWCS-ET
4.98 (p < 0.0001) ever (22) 11.4 (p < 0.0001) --> 15 years duration 3.31 (p < 0.001) -> 15 years since last use 10 (p < 0.0001) ever (17) 20 (7.2-54) -> 15 years duration 5.8 (2.0-17) --- 15 years since last use 0.93 (NS) recent (22)
0.66 (0.44-0.98) recent (21)
1.02 0.80 1.08 1.02 0.95 0.82 0.95 0.60
1.05 (NS) ever (23) 0.95 (NS) recent
0.65 0.44 0.80 0.64 0.90 0.84 0.84
NHS
(0.49-0.88 ever (20) (0.26-0.75) 15+ years (0.70-0.87) ever (18) (0.52-0.78) recent (0.85-0.94) ever (25) (0.78-0.92) recent (0.77-0.91) 20+ years
Meta-analyses
2.3 (2.1-2.5) ever (37) 9.5 (7.4-12.3) > 10 years duration 2.3 (1.8-3.1) >- 5 years since last use
1.32 1.41 1.05 1.37 1.49 0.65
(1.14-1.54) (1.15-1.74) (0.92-1.20) (1.18-1.59) (1.29-1.74) (0.50-0.83)
current ET (27) current C H T past HT (35) current < 5 years HT current --- 5 years HT current H T (31)
(0.81-1.27) ever (19) (0.53-1.21) recent (NS) ever (24) (NS) recent (NS) ever (24) (NS) recent (NS) ever (24) (0.42-0.84) recent
0.72 (0.53-0.97) ever (15) a
629
1.21 (2p = 0.00002) current H T (42) 1.023 (1.011 - 1.036) increase per year
0.67 (0.59-0.77) recent H T (50) 0.66 (0.59-0.74) current H T (51) 0.75 (0.68-084) ever (55)
0.66 (0.41-1.07) current (54)
0.60 (0.43-0.83) current ET (28) 0.39 (0.19-0.78) current C H T 0.55 (0.45-0.68) current ET (33) 0.64 (0.49-0.85) current C H T 1.27 (0.95-1.69) current ET (28) 1.09 (0.66-1.80) current C H T 1.18 (0.95-1.45) current ET (33) 1.45 (1.10-1.92) current C H T 2.1 (1.2- 3.8) current HT (34) b 1.3 (0.7-2.4) past HT
0.64 0.50 0.70 0.66 1.12
(0.59-0.68) ever ET (63) (0.45-0.59) current ET (0.65-0.75) ever ET (64) (0.53-0.84) ever C H T (1.01-123) ever HT (69)
2.14 (1.64-2.81) current H T (72) 3.49 (2.33-5.59) first year H T 0.66 (0.53-0.82) (76)
0.63 (0.56-0.70) current (32) 1.03 (0.94-1.12) past 0.80 (0.67-0.96) 10+ years
C. The Nurses' Health Study T h e Nurses' Health Study ( N H S ) began in 1976 when questionnaires were mailed to all female, married, registered nurses, ages 30 to 55 years, who were living in one of 11 large U.S. cities. A total of 121,700 women
completed the survey, which included information on menopause and use of postmenopausal hormones. Being registered nurses, participants are quite homogenous with respect to occupation, education, and health awareness. Biennial follow-up questionnaires update the information originally obtained and inquire about the development of
630
ANNLIA PAGANINI-HILL
new illnesses and death. This cohort has been analyzed for the effect of C H T and ET on morbidity and mortality in postmenopausal women.
II. T H E O U T C O M E S A. Endometrial Cancer Endometrial cancer is the fourth most common cancer in U.S. women, with an annual incidence of 40,880 cases and 7310 deaths each year (36). More than three dozen observational studies have confirmed the strong association between endometrial cancer and ET (17,22,37-40). Women who have used unopposed estrogen are at greater risk of endometrial cancer than neverusers. Risk increases with higher doses and longer durations of ET and persists for several years after discontinuation of estrogen. Women using ET for more than 10 years have nearly a tenfold increased risk. Five years or more after stopping estrogen, risk is still increased more than twofold. The substantial increase in risk of endometrial cancer with unopposed estrogen led physicians to add progestin to the treatment regimen for women with an intact uterus. More recent studies have shown that combined estrogen-progestin regimens (with progestin given sequentially for 10 or more days or continuously) do not increase the risk of endometrial cancer compared with nonusers (39-41). The W H I clinical trial confirmed the finding of no increased risk of endometrial cancer in women receiving continuous C H T therapy (hazard ratio [HR] = 0.81; 95% confidence interval [CI] 0.48-1.36) (3). However, it also found an increased frequency of uterine bleeding leading to more frequent biopsies, ultrasounds, and hysterectomies in treated women.
type nor dose markedly affects the results. In contrast to the preponderance of observational study results, including those of recent studies (43-46), the ET trial of W H I found no increased risk of breast cancer among treated women (2). Although current ET may increase breast cancer risk slightly, observational studies found that adding progestin to estrogen therapy adds substantially to the risk (relative risk [RR] = 1.24-2.00) (27,43-47). Similarly, W H I found C H T increased risk (RR = 1.24; 95% CI 1.01-1.54) (4). Also, more women in the estrogen plus progestin group than in the placebo group had abnormal mammograms requiring additional medical evaluation. C H T decreases both the sensitivity and specificity ofmammography because of increased radiographic breast density (48,49).
C. Colorectal Cancer Colorectal cancer is the third most common cancer in incidence among U.S. women. Each year 73,470 women are diagnosed with colorectal cancer and 27,750 die of the disease (36). More than two dozen observational studies, including LWCS (21) and NHS (31), have explored the relationship of H T and colorectal cancer. Meta-analyses indicate a 10% to 20% reduction among ever-users compared with never-users and a 35% reduction among current users (50,51). The relation appears to be similar for cancers of the colon and rectum. The W H I clinical trial of C H T reported similar outcomes (5). The risks were 0.56 (95% CI 0.38-0.81) for colorectal, 0.54 (95% CI 0.36-0.82) for colon, and 0.66 (95% CI 0.26-1.64) for rectal cancers, though the number of rectal cancers (n = 19) was small. The W H I trial of ET found no difference in rates of colorectal cancer between treatment and placebo groups.
D. Osteoporosis and Fracture B. Breast Cancer As the most common cancer in U.S. women, breast cancer will affect one in seven women in their lifetime. In 2005, 211,240 new cases will develop, and 40,410 women will die of breast cancer (36). The magnitude of the risk of breast associated with H T has been clarified in recent years. A reanalysis of 90% of the worldwide epidemiologic evidence of primarily unopposed estrogen included 51 studies in 21 countries (42). The two main findings are based on more than 53,000 postmenopausal women with known age at menopause, of whom 33% had used HT. First, while women are using H T and in the 5 years after they cease use, breast cancer risk increases 2.3% per year of use. Among women who use estrogen for 5 years or more, the relative risk is 1.35. Second, 5 years or more after ceasing use, no excess risk persists. Neither estrogen
Osteoporosis is common among elderly women and predisposes to fractures of the hip, vertebrae, distal forearm, and other less common sites (52). Nearly half of the 40 million women in the United States over age 50 have or will develop osteoporosis, and 1.5 million per year endure fractures (53). In addition to the enormous economic costs, these fractures often cause permanent disabili~, dependency, and premature death. Hip fractures, the most serious consequence of osteoporosis, occur in 300,000 each year and kill 20% in the first year. A wealth of data from randomized trials indicates that estrogen use prevents or greatly retards bone loss at the menopause and thereafter (53,54). Findings are similar for opposed and unopposed regimens, oral and transdermal routes of administration, and types of progestins. Bone density increases at lumbar spine, femoral neck, and forearm sites are greater with higher estrogen doses. Rapid bone loss follows estrogen
631
CHAPTER 47 Morbidity and Mortality Changes with Hormone Therapy discontinuation, so that within a few years, bone mass in treated patients approaches that in untreated controls. Likewise, evidence from prospective cohort studies (19,55-58) and randomized clinical trials (including WHI) (1,2,6,53,54,59) support the antifracture efficacy of estrogen therapy in postmenopausal women. Randomized clinical trials show a decrease in hip, vertebral, and wrist fracture frequency of about 25% to 35%. In WHI, ET decreased fractures by 30% (RR = 0.70; 95% CI 0.63-0.79) (2) and C H T by nearly a quarter (RR = 0.76; 95% CI 0.69-0.83) (6). The degree of protection against fractures afforded by estrogen is greater among current or recent users, with previous use appearing to give little protection (19,56-58). In addition, some studies suggest that elderly women who use estrogen are not protected against fracture to the extent of younger women (59-61).
E. Coronary Heart Disease Coronary heart disease (CHD) is the most common and deadly disease of women. An American woman's lifetime risk of developing C H D is 32%, and it kills 242,000 each year (62). Women are more likely than men to die within one year after a heart attack (38% versus 25%) and have higher rates of disability. Early evidence from observational studies suggested a 30% to 50% lower risk of C H D among estrogen users (15,18,26,28,33,55,63,64), consistent with randomized trials of H T on intermediate markers (lipids, lipoproteins, fibrinogen, carotid intimal medial thickness) that showed beneficial changes (65,66). However, NHS showed an increased C H D risk in the first year among hormone users with a previous myocardial infarction or atherosclerosis (34). The randomized clinical trials of W H I provide no support for cardioprotection and some suggestion of harm by H T (1,2,8). Unopposed estrogen appeared not to affect incidence of C H D (RR = 0.91; 95% CI 0.75-1.12) (2). Among women assigned to CHT, the risk for C H D (myocardial infarction and coronary death) was increased (RR = 1.24; 95% CI 1.00-1.54), but not for coronary revascularization, angina, acute coronary syndrome, or congestive heart failure (8). Risk of C H D was substantially increased in the C H T group in year 1 (RR = 1.81; 95% CI 1.09-3.01) but tended to decrease over time, with no elevation past year 5.
E Stroke Stroke ranks as the third leading cause of death (behind heart disease and cancer) in U.S. women (62). Not only does stroke kill, it is the major cause of serious, long-term disability (62) and is the second leading cause of dementia (after Alzheimer's disease) (67). Stroke is a heterogeneous
group of diseases that can be classified by the pathology as ischemic (70-88%), hemorrhagic (12-27%), and other etiologies (62,68). Over the past 25 years, epidemiologic studies have produced no conclusive evidence of a beneficial effect of H T on stroke (55,68). However, a recent meta-analysis of nine selected observational studies found increased risk of incident stroke among ever-users (RR = 1.12; 95% CI 1.01-1.23) with no difference between ever-, current, or past users (69). Risk was increased for ischemic stroke (RR = 1.20; 95% CI 1.01-1.40) but not subarachnoid hemorrhage (RR = 0.80; 95% CI 0.57-1.04) or intracerebral hemorrhage (RR = 0.71; 95% CI 0.25-1.29). The NHS reported a significant dose-response relationship with risks of 0.54, 1.35, and 1.63 for doses of 0.3, 0.625, and 1.25+ mg, respectively (33), as did a recent casecontrol study in a health maintenance organization (70). The NHS also found that risk among C H T users was 45% higher compared with never-users of HT, whereas the risk among ET users was increased 18% but not statistically significant. Some observational data indicate that estrogen users have a moderately reduced risk of fatal stroke (16,68,69). A meta-analysis of nine studies estimated a 20% lower risk for stroke mortality (RR = 0.81; 95% CI 0.71-0.92) (69). Both ET and C H T increased stroke risk in W H I (1,2,9). In the ET trial, risk of stroke was 1.39 (95% CI 1.10-1.77) (2). Similar results were found in the C H T trial: The risk was 1.31 (95% CI 1.02-1.68) for all stroke subtypes. Most of the elevation occurred for ischemic stroke (RR = 1.44; 95% CI 1.09-1.90) versus hemorrhagic stroke (RR = 0.82; 95% CI 0.43-1.56) and for nonfatal events (9).
G. Venous Thromboembolism Venous thromboembolic events (deep vein thrombosis and pulmonary embolism) contribute to 200,000 deaths per year (71 ). Case-control, cohort (29), and randomized clinical trials have consistently found risk of venous thromboembolic events greater than 1.0 among H T users. A meta-analysis of 12 studies found a twofold increase in risk among current users, with risk highest (more than a threefold increase) in the first years of use (72). The W H I also reported significant doubling in risk among C H T users (RR = 2.06; 95% CI 1.57-2.70) (7) and a 30% higher risk in ET users (RR = 1.33; 95% CI 0.99-1.79) compared with placebo (2). However, a recent case-control study found that CEE, but not esterified estrogen, was associated with venous thrombolic risk (73). Among CEE users, risk increased with increasing dosage. Risk was also higher with current use of estrogen plus progestin compared with use of estrogen alone.
632
ANNLIA PAGANINI-HILL
H. Cognition and Dementia Although nearly all cognitive functions decline, on average, with age, the variability ranges from successful aging to dementia. More than one-third of women aged 65 and older will develop dementia, of which three-quarters will be Alzheimer's disease (AD) (74). Today AD affects 4.5 million Americans, and it is expected to claim 13 million by 2050 (75). Although observational studies provide little support for the hypothesis that H T preserves overall cognitive function, a few found improvement in selected aspects of cognition such as verbal fluency, verbal memory, and reasoning (32,76-78). A review of 15 double-blind, randomized controlled trials of H T on cognitive function over a treatment period of 2 weeks to 9 months in healthy postmenopausal women also found little evidence of an effect of estrogen on overall cognitive function (78). The Women's Health Initiative Memory Study (WHIMS) was an ancillary study to W H I designed to assess, in women ages 65 and older, whether H T can decrease risk of dementia. W H I M S involved 7510 women in the CHT/placebo trial and 4532 in the ET/placebo trial (10-13). With an average of 4 to 5 years of follow-up, both ET or C H T were found to have a small adverse effect on a measure of global cognitive function, the Modified Mini-Mental State Exam (3MMSE) (11,13). Several observational studies with dementia outcomes (generally AD) have suggested that H T is associated with a 29% to 44% decreased risk for dementia (20,76-78). In a meta-analysis of cohort and prospective case-control studies, H T was associated with a 34% reduction in risk of dementia (RR = 0.66; 95% 0.53-0.82) (76). In WHIMS, the only randomized controlled trial to examine the effects of l i T on dementia risk, women assigned to C H T had a twofold increased risk of developing dementia compared with the placebo group (RR -- 2.05; 95% CI 1.21-3.48) (10). Although the increased risk of dementia was not statistically significant in the ET trial, estrogen alone increased risk of mild cognitive impairment or dementia combined (RR = 1.38; 95% CI 1.01-1.89) (12).
I. Mortality Prospective studies have consistently shown a 20% to 50% decrease in mortality among users of estrogens (18,25,30,55). Grady et al. (55) calculated that the life expectancy for a 50-year old woman choosing long-term ET was nearly a year greater than nonusers. The association with reduced mortality was related to both duration and recency of estrogen use (25,30). In WHI, H T was unrelated to mortality. In the C H T trial, the numbers of overall deaths in the estrogen/progestin
and placebo groups were statistically and clinically similar in this short-duration (5-year) study (RR = 0.98; 95% CI 0.82-1.18) (1). Likewise, in the ET trial, overall death rates did not differ between the estrogen and placebo arms (RR = 1.04; 95% CI 0.88-1.22) (2).
III. WHY RESULTS DIFFER Overall, the results of the W H I study are consistent with the growing body of literature, including observational studies, on the effects of liT. The increasing risk of breast cancer with duration of use and the reductions in risk of colon cancer and fracture are in the expected direction and magnitude. As with observational studies, risk of venous thromboembolism was greatest in year 1 but continued throughout the 5 years of therapy. W H I also found greater risk of stroke as suggested by recent observational studies. In contrast with observational studies, in W H I the risk of C H D was elevated, especially in the first years, and in older women, risk of dementia was increased. Commentaries addressing the differences in findings between observational studies, which found a beneficial effect on C H D and dementia, and the randomized trials, which found none, continue. Some suggest that the divergent findings may be partially due to differences in the clinical characteristics of the study populations, including differences in age, years since menopause, and risk factors for CHD. Others have pointed out the methodologic limitations of both types of studies.
A. Confounding A major concern in observational studies is the possibility that unrecognized confounders or bias account for the observed results. In general populations, women taking H T may have other health-promoting habits and may differ from nonusers in unmeasured ways that influence longevity--the healthy user effect (79-81). However, the Leisure World cohort is a relatively homogeneous group of mostly white, highly educated, upper-middle class women with access to health care. Likewise, all participants in NHS are registered nurses, with similar education and health care knowledge. Thus, both studies controlled for the potential confounders of education and social class by focusing on populations with uniform education, income, and/or occupation. In addition, differences in lifestyle practices and cardiovascular risk factors between estrogen users and nonusers in these studies were not great and adjustment was made for these potentially confounding factors. Nonetheless, uncontrolled confounding cannot be ruled out in any observational study. Observational studies (including LWCS and NHS) have yielded relative risks for fracture, breast cancer, colon cancer, venous thromboembolism, and stroke that were similar to
CHAPTER47 Morbidity and Mortality Changes with Hormone Therapy those reported by WHI. Because confounding would be expected to alter the findings for these disease endpoints as well, perhaps other factors explain the different findings for C H D and dementia between observational studies and WHI.
B. A g e at S t a r t i n g T r e a t m e n t a n d P r e - e x i s t i n g Disease A major difference between participants in the observational studies and those in W H I trials is the age at initiation of HT. In WHI, most of the women were at ages that are, on average, substantially older than those at which most women in the general population initiate HT. In the observational studies, most hormone use began at menopause. In both LWCS and NHS, 80% of women initiated H T within 2 years of menopause. In contrast, W H I women were on average age 63 years, two-thirds were over age 60, and most had been postmenopausal for more than 10 years at enrollment (1,2). This is especially true in WHIMS, where all women took H T after age 65 years. This difference in age at starting treatment raises concern that W H I trial results may not apply to treatment begun early in menopause. The W H I found a trend toward lower risk of C H D with decreasing time since menopause; the relative risk was 0.89, 1.22, and 1.71 among women whose time since menopause was less than 10, 10 to 19, and 20 or more years, respectively (8). In addition, the W H I ET trial found that women in the youngest age group (ages 50 to 59 years) responded more favorably than older postmenopausal women for many outcomes (2). Even though the W H I subjects were designated healthy, many had pre-existing disease or other cardiovascular risk factors (older, obese, smoking, hypertensive, diabetic) and the process of atherosclerosis may have been active. In these women, the trial may have been more similar to a secondary (than a primary) prevention trial of cardiovascular disease. In the first year of WHI, more cardiac events occurred in the C H T group consistent with the findings of HERS (82), other secondary prevention trials, and NHS (34). Some researchers have speculated that estrogen's cardioprotective properties may depend on maintenance of a healthy endotheliummthe healthy endothelium concept (83,84). Here the favorable effects of estrogen on atherosclerosis, inflammation, hemostatis, and coronary flow reserve depend on the integrity of the endothelium and the estrogen receptor populations in endothelial and vascular smooth muscle cells. These would likely be seen only in younger women before the development of substantial atherosclerotic plaques. In older women with pre-existing disease, H T may not inhibit progression from later stages of atherosclerosis to coronary events but may cause prothombotic plaque &stabilization and result in an early harmful effect. This is consistent with studies in rabbits showing the estrogen benefits were reversed in the presence of endothelial dysfunction (85) and in
633
monkeys showing that a 2-year (equivalent to a 6-year in women) delay in H T is enough to reduce the protective actions of H T (86). Likewise, H T may not have been protective against dementia in W H I M S because it was not taken during the critical early menopausal years or for long duration (87,88). In addition, only 50% of the dementia cases in W H I M S were classified as AD, less than the 60% to 70% in the general population (74). This could suggest that H T increases the risk of vascular dementia more than the risk of AD and is consistent with the twofold increased risk of stroke in hormone-treated women in WHI.
C. S h o r t - t e r m a n d L o n g - t e r m Effects o f H T By design, most observational studies have sparse numbers concerning the first years of H T and therefore limited ability to capture acute short-term effects, including the increase in C H D events observed in the initial years of the W H I trial. This effect probably led to a modest overestimate of the apparent benefit of hormones for C H D and an underestimate of the apparent risk for stroke and venous thromboembolism in the observational studies. However, several observational studies provided hints of early CHD, stroke, venous thromboembolism, and mortality risk elevations with H T (25,29,34,89,90). Although the risk estimates from observational study are heavily weighted by longer-term use, clinical trial data are comparably sparse after 5 years' duration and their risk estimates reflect shorter-term use. Long-term treatment and follow-up in the context of a randomized clinical trial like W H I is unlikely. Thus, LWCS, NHS, and other long-term observational studies will continue to contribute valuable data about long-term use of l i T that complement those of clinical trials. Comparing the W H I observational study with the W H I clinical trial, Prentice and colleagues found that the relative risks for CHD, stroke, and venous thromboembolism tended to decrease with increasing time from initiation of C H T and that following control for time from initiation the estimates were rather similar for the trial participants and observational cohort (91).
D. Specific R e g i m e n The W H I evaluated only two drug regimens (0.625 mg CEE with or without 2.5 mg MPA daily) and therefore cannot answer questions regarding risks associated with other hormones, doses, schedules, or routes of administration. Whether other regimens carry similar risks to those studied in W H I is unclear. However, one review of clinical studies that investigated different types and regimens of estrogens in combination with progestins concluded that the risk/ benefit analysis reported by the W H I can be generalized to
634
ANNLIA PAGANINI-HILL
other estrogens (e.g., oral ethinyl estradiol, 17[3-estradiol, estradiol valerate, piperazine estrone sulfate, estriol, or transdermal 17[3-estradiol) and other progestins (e.g., oral NETA; levonorgestrel, norgestrel, or transdermal norethisterone; and dydrogesterone) (92). Although recent findings with lower doses of H T have suggested that some benefits (osteoporosis, fracture, colorectal cancer) are maintained, whether lower doses will minimize the risks seen with standard doses remains to be determined.
E. D r o p o u t a n d D r o p - i n Rates Another limitation in W H I is the relatively high rate of discontinuation of hormone therapy. Over the course of the trial, 42% of the women in the C H T group and 38% in the placebo groups stopped taking their medications and 6.2% and 10.7%, respectively, initiated HT. Likewise, more than 50% of the women in the ET trial had stopped therapy by the time the study was terminated. With this substantial crossover, the intention-to-treat analysis may have underestimated the true effects of hormones. Also, if duration of treatment is important, as with breast cancer risk, and if compliance decreases over time, 5-year results may underestimate longer-term treatment effects.
IV. THE RISKS AND BENEFITS OF HORMONE
THERAPY
The purpose of healthy women taking longer-term H T is to preserve health and prevent disease. The results of W H I studies provide strong evidence of significant risks with such therapy. Although population and biologic differences offer plausible explanations for the divergent findings between observational studies and WHI, clinically it may not matter. Risk of breast cancer increases with longer duration of HT, especially CHT. Even if l i T were cardioprotective in younger menopausal women, it does not balance the excess risk of breast cancer. Safe options are available for reducing C H D risk as well as for preventing osteoporosis and fractures. Like all drug therapy, H T has benefits and risks. 9 H T is the most effective treatment for moderate to severe menopausal symptoms (hot flashes, night sweats, vaginal dryness, urogenital atrophy) (93,94). 9 ET increases risk of endometrial cancer in women with an intact uterus. 9 C H T increases frequency of uterine bleeding, leading to more biopsies, ultrasounds, and hysterectomies. 9 H T increases risk of breast cancer during use and several years after. The risk disappears several years after discontinuing therapy. C H T increases risk over that seen with ET.
9 H T (current or recent use) reduces risk of colorectal cancer. 9 H T reduces risk of osteoporotic fractures of the hip, vertebrae, and other sites. H T is not warranted solely for fracture prevention. Alternative therapies are available. 9 H T is not indicated for the prevention of cardiovascular disease. Alternative health strategies (proper diet, exercise, refrain from smoking) and, if needed, pharmaceutical agents (for maintenance of low LDL cholesterol levels and normal blood pressure) with established value should be used. 9 H T increases risk of ischemic stroke. 9 H T increases risk of venous thromboembolism especially in the first years of use. 9 H T in older women (more than 65 years) does not enhance cognition or prevent the development of dementia. H T is appropriate for the relief of menopausal symptoms, as long as a woman has weighed the risks and benefits with her health care provider (95). The lower absolute risk for many adverse outcomes at ages 50 to 59 and the possibility of more favorable findings in this age group both suggest that use of H T to treat menopausal symptoms for a limited duration early in menopause is reasonable. If chosen, a woman should take the lowest dose possible for the shortest time possible and without a progestin if feasible. She should review the decision to take H T annually. However, H T should not be initiated or continued for the prevention of cardiovascular disease.
References 1. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321- 333. 2. The Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy. The Women's Health Initiative randomized controlled trial. JAMA 2004;291:1701 - 1712. 3. Anderson GL, Judd HL, Kaunitz AM, et al. Effects of estrogen plus progestin on gynecologic cancers and associated diagnostic procedures. The Women's Health Initiative randomized trial. JAMA 2003;290: 1739-1748. 4. Chlebowski RT, Hendrix SL, Langer RD, et al. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women. The Women's Health Initiative randomized trial. JAM,q 2003;289:3243- 3253. 5. Chlebowski RT, Wactawski-Wende J, Ritenbaugh C, et al. Estrogen plus progestin and colorectal cancer in postmenopausal women. NEngl J Med 2004;350:991-1004. 6. Cauley JA, Robbins J, Chen Z, et al. Effects of estrogen plus progestin on risk of fracture and bone mineral density. The Women's Health Initiative randomized trial. JAMA 2003;290:1729-1738. 7. Cushman M, KuUer LH, Prentice R, et al. Estrogen plus progestin and risk of venous thrombosis. JAMA 2004;292:1573-1580.
CHAPTER 47 Morbidity and Mortality Changes with H o r m o n e Therapy 8. Manson JE, Hsia J, Johnson KC, et al. Estrogen plus progestin and the risk of coronary heart disease. N EnglJ Med 2003;349:523- 534. 9. Wassertheil-Smoller S, Hendrix SL, Limacher M, et al. Effect of estrogen plus progestin on stroke in postmenopausal women. The Women's Health Initiative: a randomized trial. JAMA 2003;289: 2673-2684. 10. Shumaker SA, Legault C, Rapp SR, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women. The Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651 - 2662. 11. Rapp SR, Espeland MA, Shumaker SA, et al. Effect of estrogen plus progestin on global cognitive function in postmenopausal women. The Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2663-2672. 12. Shumaker SA, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women. Women's Health Initiative Memory Study. JAMA 2004;291:2947-2958. 13. Espeland MA, Rapp SR, Shumaker SA, et al. Conjugated equine estrogens and global cognitive function in postmenopausal women. Women's Health Initiative Memory Study. JAMA 2004;291:2959-2968. 14. Barnabei VM, Cochrane BB, Aragaki AK, et al. Menopausal symptoms and treatment-related effects of estrogen and progestin in the Women's Health Initiative. Obstet Gyneco12005;105:1063-1073. 15. Henderson BE, Paganini-Hill A, Ross RK. Estrogen replacement therapy and protection from acute myocardial infarction. Am J Obstet Gyneco11988;159:312-317. 16. Paganini-Hill A, Ross RK, Henderson BE. Postmenopausal oestrogen treatment and stroke: a prospective study. Brit Med J 1988;297: 519-522. 17. Paganini-Hill A, Ross RK, Henderson BE. Endometrial cancer and patterns of use of oestrogen replacement therapy: a cohort study. BrJ Cancer 1989;59:445-447. 18. Henderson BE, Paganini-Hill A, Ross RK. Decreased mortality in users of estrogen replacement therapy. Arch Intern Med 1991;151: 75-78. 19. Paganini-Hill A, Chao A, Ross RK, Henderson BE. Exercise and other factors in the prevention of hip fracture: the Leisure World Study. Epiderniology 1991;2:16-25. 20. Paganini-Hill A, Henderson VW. Estrogen replacement therapy and risk of Alzheimer's disease. Arch Intern Med 1996;156:2213-2217. 21. Paganini-Hill A. Estrogen replacement therapy and colorectal cancer risk in elderly women. Dis Colon Rectum 1999;42:1300-1305. 22. Paganini-Hill A. Morbidity and morality changes with estrogen replacement therapy. In: Lobo RA, ed. In: The treatment of the postmenopausal woman." basic and clinical aspects, 2nd ed. New York: Lippincott Williams &Wilkins, 1999:549-555. 23. Paganini-Hill A, Perez Barreto M. Stroke risk in older men and women: aspirin, estrogen, vitamins, and other factors.J Gend SpecifMed 2001;4:18-28. 24. Paganini-Hill A, Atchison KA, Gornbein JA, et al. Menstrual and reproductive factors and fracture risk: the Leisure World Cohort Study. J Women's Health 2005;14:773-784. 25. Paganini-Hill A, Corrada MM, Kawas CH. Increased longevity in older users of postmenopausal estrogen therapy: the Leisure World Cohort Study. Menopause 2006;13:12-18. 26. Stampfer MJ, Colditz GA, Willett WC, et al. Postmenopausal estrogen therapy and cardiovascular disease. Ten-year follow-up from the Nurses' Health Study. N EnglJ Med 1991;325:756- 762. 27. Colditz GA, Hankinson SE, Hunter DJ, et al. The use of estrogens and progestins and the risk of breast cancer in postmenopausal women. NEnglJMed 1995;332:1589-1593. 28. Grodstein F, Stampfer MJ, Manson JE, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N EnglJ ivied 1996;335:453-461; erratum, 1406.
635
29. Grodstein F, Stampfer MJ, Goldhaber SZ, et al. Prospective study of exogenous hormones and risk of pulmonary embolism in women. Lancet 1996;348:983 - 987. 30. Grodstein F, Stampfer MJ, Colditz GA, et al. Postmenopausal hormone therapy and mortality. N EnglJ Med 1997;336:1769-1775. 31. Grodstein F, Martinez ME, Platz EA, et al. Postmenopausal hormone use and risk for colorectal cancer and adenoma. Ann Int Med 1998;128:705-712. 32. Grodstein F, Chen J, Pollen DA, et al. Postmenopausal hormone therapy and cognitive function in healthy older women. J A m Geriatr Soc 2000;48:746-752. 33. Grodstein F, Manson JE, Colditz GA, et al. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med 2000;133:933-941. 34. Grodstein F, Manson JE, Stampfer MJ. Postmenopausal hormone use and secondary prevention of coronary events in the Nurses' Health Study. Arch Intern Med 2001;135:1-8. 35. Chen WY, Coldtiz GA, Rosner B, et al. Use of postmenopausal hormones, alcohol, and risk of invasive breast cancer. Ann Intern Med 2002;137:798- 804. 36. American Cancer Society. Cancer facts and figures 2005. Atlanta: American Cancer Society, 2005. 37. Grady D, Gebretsadik T, Kerlikowske K, et al. Hormone replacement therapy and endometrial cancer risk: a meta-analysis. Obstet Gyneco!1995; 85:304-313. 38. Cushing KL, Weiss NS, Voigt LF, et al. Risk of endometrial cancer in relation to low-dose, unopposed estrogens. Obstet Gyneco! 1998; 91:35-39. 39. Pike MC, Peters RK, Cozen W, et al. Estrogen-progestin replacement therapy and endometrial cancer. J Nat/Cancer Inst 1997;89: 1110-1116. 40. Weiderpass E, Adami HO, Baron JA, et al. Risk of endometrial cancer following estrogen replacement with and without progestins. J Nat/ Cancer Inst 1999;91:1131-1137. 41. Hill DA, Weiss NS, Beresford SA, et al. Continuous combined hormone replacement therapy and risk of endometrial cancer. Am J Obstet Gyneco! 2000;183:1456-1461. 42. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Lancet 1997;350:1047-1059. 43. Ross RK, Paganini-Hill A, Wan PC, Pike MC. Effect of hormone replacement therapy on breast cancer risk: estrogen versus estrogen plus progestin. J Nat/Cancer Inst 2000;92:328- 332. 44. Million Women Study Collaborators. Breast cancer and hormonereplacement therapy in the Million Women Study. Lancet 2003;362: 419-427. 45. Schairer C, Lubin J, Troisi R, et al. Menopausal estrogen and estrogenprogestin replacement therapy and breast cancer risk. JAMA 2000;283: 485-491. 46. Chen C-L, Weiss NS, Newcomb P, et al. Hormone replacement therapy in relation to breast cancer. JAMA 2002;287:734-741. 47. Li CI, Malone KE, Porter PL, et al. Relationship between long durations and different regimens of hormone therapy and risk of breast cancer. JAMA 2003;289:3254- 3263. 48. Carney PA, Miglioretti DL, Yankaskas BC, et al. Individual and combined effects of age, breast density, and hormone replacement therapy use on the accuracy of screening mammography. Ann Intern Med 2003;138:168-175. 49. Greendale GA, Reboussin BA, Slone S, et al. Postmenopausal hormone therapy and change in mammographic density. J Nat/ CancerInst 2003;95:30-37. 50. Nanda K, Bastian LA, Hasselband V, Simel DL. Hormone replacement therapy and the risk of colorectal cancer: a meta-analysis. Obstet Gyneco11999;93:880- 888.
636 51. Grodstein F, Newcomb PA, Stampfer MJ. Postmenopausal hormone therapy and the risk of colorectal cancer: a review and meta-analysis. Am J Med 1999;106:574- 582. 52. Cummings SR, Melton LJ III. Epidemiology and outcomes of osteoporotic fractures. Lancet 2002;359:1761 - 1767. 53. American College of Obstetricians and Gynecologists. Osteoporosis. Obstet Gyneco12004;104(suppl):66S- 76S. 54. Wells G, Tugwell P, Shea B, et al. Meta-analysis of therapies for postmenopausal osteoporosis. V. Meta-analysis of the efficacy of hormone replacement therapy in treating and preventing osteoporosis in postmenopausal women. Endocr Rev 2002;23:529-539. 55. Grady D, Rubin SM, Petitti DB, et al. Hormone therapy to prevent disease and prolong life in postmenopausal women. Ann Internal Med 1992;117:1016-1037. 56. Cauley JA, Seeley DG, Ensrud K, et al., for the Study of Osteoporotic Fractures Research Group. Estrogen replacement therapy and fractures in older women. Ann Intern Med 1995;122:9-16. 57. Yates J, Barrett-Connor E, Barlas S, et al. Rapid loss of hip fracture protection after estrogen cessation: evidence from the National Osteoporosis Risk Assessment. Obstet Gyneco12004;103:440-446. 58. Barrett-Connor E, Wehren LE, Siris ES, et al. Recency and duration of postmenopausal hormone therapy: effects on bone mineral density and fracture risk in the National Osteoporosis Assessment (NORA) study. Menopause 2003;10:412-419. 59. Torgerson DJ, Bell-Syer SEM. Hormone replacement therapy and prevention of nonvertebral fractures. A meta-analysis of randomized trials. JAMA 2001;285:2891 - 2897. 60. Naesstn T, Persson I, Adami H-O, Bergstrtm R, Bergksvist L. Hormone replacement therapy and the risk for first hip fracture. A prospective, population-based cohort study. Ann Intern Med 1990;11:95-103. 61. Kanis JA, Johnell O, Gullberg B, et al. Evidence for efficacy of drugs affecting bone metabolism in preventing hip fracture BMJ 1992;305: 1124-1128. 62. American Heart Association. Heart disease and stroke statistics--2005 update. Dallas: American Heart Association, 2005. 63. Grodstein F, Stampfer M. The epidemiology of coronary heart disease and estrogen replacement in postmenopausal women. Prog Cardiovascular Diseases 1995;38:199-210. 64. Barrett-Connor E, Grady D. Hormone replacement therapy; heart disease, and other considerations. Annu Rev Public Health 1998;19:55- 72. 65. Writing Group for the PEPI Trial. Effects of estrogen or estrogen/ progestin regimens on heart disease risk factors in postmenopausal women. JAMA 1995;273:199-208. 66. Mendelson ME, Karas RH. The protective effects of estrogen on the cardiovascular system. N EnglJ Med 1999;340:1801 - 1811. 67. Skoog I, Nilsson L, Palmertz B, Andreasson L-A, Svanborg A. A population-based study of dementia in 85 year-olds. N EnglJ Med 1993;328:153-158. 68. Paganini-Hill A. Hormone replacement therapy and stroke: risk, protection or no effect? Maturitas 2001;38:243-261. 69. Nelson HD, Humphrey LL, Nygren P, et al. Postmenopausal hormone replacement therapy. Scientific review.JAMe12002;288:872- 881. 70. Lemaitre RN, Heckbert SR, Psaty BM, et al. Hormone replacement therapy and associated risk of stroke in postmenopausal women. Arch Intern Med 2002;162:1954-1960. 71. American College of Obstetricians and Gynecologists. Venous thromboembolic disease. Obstet Gyneco12004;104(suppl):118S- 127S. 72. Miller J, Chan B KS, Nelson HD. Postmenopausal estrogen replacement and risk for venous thromboembolism: a systematic review and meta-analysis for the U.S. Preventive Services Task Force. Ann Intern Med 2002;136:680-690. 73. Smith NL, Heckbert SR, Lemaitre RN, et al. Esterified estrogens and conjugated equine estrogens and the risk of venous thrombosis. JAMel 2004;292:1581-1587.
ANNLIA PAGANINI-HILL 74. Ott A, Breteler MMB, van Harskamp F, et al. Incidence and risk of dementia. The Rotterdam Study. Am J Epidemiol 1998;147: 574-580. 75. Hebert LE, Scherr PA, Bienias JL, et al. Alzheimer disease in the US population: prevalence estimates using the 2000 census. Arch Neurol 2003 ;60:1119 - 1122. 76. LeBlanc ES, Janowsky J, Chan BKS, Nelson HD. Hormone replacement therapy and cognition. Systematic review and meta-analysis.JAMA 2001 ;285:1489-1499. 77. American College of Obstetricians and Gynecologists. Cognition and dementia. Obstet Gyneco12004;104(suppl):25 S - 40S. 78. Hogervorst E, Yaffe K, Richards M, Huppert F. Hormone replacement therapy for cognitive function in postmenopausal women (review). Cochrane System Rev 2002;2:CD003122. 79. Humphrey LL, Chan BKS, Sox HC (2002). Postmenopausal hormone replacement therapy and the primary prevention of cardiovascular disease. Ann Intern ivied 2002;137:273-284. 80. MacMahon S, Collins R. Reliable assessment of the effects of treatment on mortality and major morbidity. II. Observation studies. Lancet 2001;357:455-462. 81. Petitti DB. Hormone replacement therapy and heart disease prevention. Experimentation trumps observation. JAMA 1998;280:650-652. 82. Hulley S, Grady D, Bush T, et al., for the Heart and Estrogen/Progestin Replacement Study (HERS) Research Group. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMd 1998;280:605-613. 83. Koh KK. Can a healthy endothelium influence the cardiovascular effects of hormone replacement therapy? IntlJ Cardio12003;87:1- 8. 84. Phillips LS, Langer RD. Postmenopausal hormone therapy: critical reappraisal and a unified hypothesis. Fertility and Sterility 2005;83:558-566. 85. Holm P, Andersen HL, Andersen MR, et al. The direct antiatherogenic effect of estrogen is present, absent, or reversed, depending on the state of the arterial endothelium. A time course study in cholesterolclamped rabbits. Circulation 1999;100:1727-1733. 86. Mikkola TS, Clarkson TB, Notelovitz M. Postmenopausal hormone therapy before and after the women's health initiative study: what consequences? Ann Med 2004;36:402-413. 87. Zandi PP, Carlson MC, Plassman BL, et al. (2002). Hormone replacement therapy and incidence of Alzheimer disease in older women. The Cache County Study. JAMA 2002;288:2123-2129. 88. Resnick SM, Henderson VW. Hormone therapy and risk of Alzheimer disease. A critical time. JAMA 2002;288:2170- 2172. 89. Heckbert SR, Kaplan RC, Weiss NS, et al. Risk of recurrent coronary events in relation to use and recent initiation of postmenopausal hormone therapy. Arch Intern Med 2001;161:1709-1713. 90. Alexander KP, Newby LK, Hellkamp AS, et al. Initiation of hormone replacement therapy after acute myocardial infarction is associated with more cardiac events during follow-up. JAm Coll Cardio12001;38:1- 7. 91. Prentice RL, Langer R, Stefanick ML, et al. Combined postmenopausal hormone therapy and cardiovascular disease: toward resolving the discrepancy between observational studies and the Women's Health Initiative Clinical Trial. Am J Epidemio12005;162: 404-414. 92. Warren MP. A comparative review of the risks and benefits of hormone replacement therapy regimens. Am J Obstet Gyneco12004;190: 1141-1167. 93. American College of Obstetricians and Gynecologists. Vasomotor symptoms. Obstet Gyneco12004;104(suppl):106S- 117S. 94. Cardozo L, Bachmann G, McClish D, et al. Meta-analysis of estrogen therapy in the management of urogenital atrophy in postmenpausal women: second report of the Hormones and Urogenital Therapy Committee. Obstet Gyneco11998;92:722-727. 95. Peterson HB, Thacker SB, Corso PS, et al. Hormone therapy. Making decisions in the face of uncertainty. Arch Intern Med 2004;164: 2308-2312.
SECTION X
Life Cycle and Q OL Only in more recent years has there been a real focus aimed at understanding the full impact of a given treatment on the life of the patient. It is often said in the cancer field that improved quality-of-life years is more important than total longevity. Part of the problem here is that it has been difficult to accurately assess quality of life. The validity of various instruments has come into question in the past. This section deals with lifestyle and quality-of-life issues in postmenopausal women. Included here is a chapter by Adriane Fugh-Berman on complementary medicine, including the use of herbs and phytoestrogens, but also discussing choices such as the use of acupuncture. A valid argument may be made that this chapter belongs in Section XIII, Some Alternative Medical Therapies. Whether taking herbs and nutraceuticals should be considered a treatment or is part of a lifestyle management choice is a matter of debate. This section begins with an in-depth discussion of quality of life and the various instruments used to assess this. When evaluating studies on this point, it is important to know the validity of the instrument used and also the characteristics of the patient population studied. For example, in my view, regardless of the instrument used, Q.OL is improved across the board in postmenopausal women receiving hormones for symptoms. However, this may not be the case in asymptomatic and older women. Next, Michelle Warren discusses the important lifestyle variables of exercise and nutrition. As noted earlier, the final chapter in this section reviews the most current information we have on the effect of certain complementary approaches on various menopausal problems.
This Page Intentionally Left Blank
2 H A P T E R 41
Issues Relating to O..uality of Life in Postmenopausal Women and Their Measurement HERMANN
P. G . S C H N E I D E R
Department of Obstetrics & Gynecology, University of Muenster, 48149 Muenster, Germany
I. I N T R O D U C T I O N
Those years of life in which a woman passes through a transition from the reproductive stage of life to the menopausal years form a period marked by waning ovarian function, best referred to as the climacteric. According to the Massachusetts Women's Health Study, which provided data from 2570 women, the median age for the onset of menopause is 51.3 years and the range of menopause is approximately 48 to 55 (3). The perimenopausal transition, for most women, lasts for approximately 4 years. Thinner women experience a slightly earlier menopause (4), obviously related to the contribution of body fat to estrogen production. Factors of little influence are race, income, geography, parity, or height (3,4). The cessation of menses by most women is perceived to have almost no impact on subsequent physical and mental health (3). The Massachusetts Women's Health Study has provided information that women would express either positive or neutral feelings about menopause with the exception of those who experience surgical menopause. By that token, the majority of women feel healthy and happy and do not seek contact with physicians. Medical intervention at this point of life should rather be regarded as an
Health-related quality of @ refers to the effects of an individual's physical state on all aspects of psychosocial functioning. In a more broad sense, quality of life may also be defined as "the extent to which our hopes and ambitions are matched by experience" (1). Health status and quality of life are not linearly related. In recent years, there has been a growing awareness of the aspects of quality of life and aging. Q.uality of life is a subjective parameter, and direct questioning is therefore a simple and appropriate way of accruing information about how patients feel and function. Accordingly, measures of quality of life (Q_OL) attempt to gauge the effect ill health has across a number of physical, psychologic, and social parameters. Menopause is a transition in life and therefore differs from illness. Moreover, menopause challenges self-identity, which may be determined by age-related changes, illness cognitions, and symptom attributions. These need to be taken into account when assessing changes that affect quality of life during the menopausal transition (2). TREATMENT OF THE POSTMENOPAUSAL WOMAN
639
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any formreserved.
640 opportunity to provide and reinforce a program of preventive health care. These issues of preventive health care for women include family planning, cessation of smoking, control of body weight and alcohol consumption, prevention of heart disease and osteoporosis, maintenance of mental wellbeing (including sexuality), cancer screening, and treatment of neurologic problems. Menopause has also been looked at as a wonderful signal occurring at the right time of life when preventive health care is especially critical (5). This way, menopause is considered a normal stage in development, incorporating biology, psychology, society, and culture. Fries summoned up three eras in health and disease (6). The first era, which extended into the early 1900s, was characterized by acute infectious diseases. The second era, highlighted by cardiovascular disease and cancer, is now beginning to move into the third era, marked by problems of frailty (fading eyesight and hearing, impaired memory and cognitive function, decreased strength and reserve). Chronic disease in an aging population is incremental in nature. The best health strategy would be to change the rate at which illness develops and thus postpone the clinical illness; in the end, if it is postponed long enough, it might be prevented effectively. This postponement of illness has been termed compression of morbidity (7,8). The target is to lead a relatively healthy life and compress illness into a short period just before death. Thus, disease is something not necessarily best treated by medication or surgery, but by prevention or, more accurately, by postponement. Improvement of quality of life is a primary purpose of health promotion. This can be achieved by preventive health programs, with their greater impact on morbidity rather than mortality (8). The aim is maximal vigor in life rather than accepting linear senescence. Some linear decline is unavoidable, but the slope can be changed by effort and practice.
II. H O W T O ASSESS MENOPAUSAL SYMPTOMS AND QUALITY OF LIFE The first widely accepted attempt to document the severity of menopausal complaints in women was the Kupperman Index (9,10), a listing of the relevant symptomatology in this age span. This questionnaire focused primarily on symptomatic relief, assessed on the basis of the physician'ssummary of the severity of the climacteric complaints, and assisted by a weighting index, rather than letting women assess their perceived symptoms. This is an example for a simple symptoms inventory without attempts to standardize it or to apply psychometric methodology. Some decades later, time had come for the development of more specific symptom fists or other questionnaires as instruments to measure changes and to validate them in a scientific manner. Psychometric methods
HERMANN P. G. SCHNEIDER
were more frequently applied in the 1950s and 1960s; this knowledge, however, was greatly restricted to psychology and social science and not yet common in medicine. Test theory and test construction developed rapidly in the 1960s also, spreading to the medical field. It was the period when social scientists had to acknowledge the differences between "objective hard data" and "subjective soft data," as different degrees of proos In particular, increased awareness emanated of subjectively perceived quality of life to best serve the description of treatment benefits. Attention focused on instruments with applied psychometrics to develop a scale and evaluate their basic properties such as dimensions (domains) of a scale to measure complex human characteristics. This would, for example, require to analyze the structure of a construct such as menopausal complaints. By analyzing the possible intercorrelations of all symptom combinations, it was found that symptoms would cluster into "factors," which allow to assess variation. This is the simple basis of factor analysis. The factors represent the "domains or subscales" of a complex construct such as menopause. There is a recent trend of increasing recognition among clinicians and researchers of the role of patient-reported data as outcome measure for clinical and drug research. Health authorities are in support of this growing interest. As a result, multiple attempts have been undertaken for a state-of-the-art development of health-related qualityof-life scales applicable to women in their menopausal transition.
A. Methodologic Considerations The standard method used for collecting information on the prevalence and severity of menopausal symptoms has been a checklist of symptoms. Symptoms are defined as "an indication of a disease or a disorder noticed by the patient himself. A presenting symptom is one that leaves a patient to consult a doctor" (11). This way, symptoms represent a subjective expression or manifestation of some underlying physical, psychologic, or social dysfunction. Symptoms are, in effect, evidence of disease (12). Why do women seek medical assistance? Most of the time, for any of the following symptoms: 9 Vasomotor instability, resulting in the leading symptoms of menopause--the hot flushes and sweating 9 Atrophic conditions of estrogen-sensitive tissues-among these, atrophy of the vaginal epithelium; atrophic urethritis with a formation of urethral carbuncles; and dyspareunia and pruritus due to vulvar, introital, and vaginal atrophy 9 Urinary difficulties, such as stress incontinence, urgency, and abacterial urethritis and cystitis.
CHAPTER48 Issues Relating to Q.uality of Life in Postmenopausal Women and Their Measurement In addition, psychologic symptoms, such as anxiety, depressive mood, irritability, insomnia, and decreased libido are extremely common. Although the menopause is blamed for many of these problems, it is associated with a multiplicity of symptoms, some of which are similar to psychiatric disturbances. A deleterious effect of the menopause per se on mental health is not supported by the psychiatric literature. In addition, there is no proof for a specific psychiatric disorder such as involutional melancholia. Even depression is less common, not more common, among middle-aged women (13). Menopause, however, may increase minor so-caUed psychologic complaints. Becoming forgetful in the advanced years is a concern of many, and differentiation of the aging process and pathologic memory loss has a major clinical impact. Research on gender differences in cognition during the aging process and in the course of dementia is in its early stages. Many women become anxious or otherwise disturbed by aspects of a normal aging process in conjunction with a lessening of energy level compared with younger years, changes in sleeping patterns, and changes in the frequency and intensity of sexual desire. Coronary risk factors are highly prevalent among older women. About one-third of middle-aged women have hypertension. More than one-fourth of these are cigarette smokers; more than one-fourth are also overweight. Coronary risk factors tend to predominate in populations of lower socioeconomic class as well as lower educational levels. Relative coronary risk is imparted by gender. The relative risk (RR) of hypertension is comparable for women and men, about 1.5; the gender risk for hypercholesterolemia is 1.1 and 1.4, respectively; for diabetes 2.4 and 1.9, respectively; for overweight 1.4 and 1.3, respectively; and for smoking 1.8 and 1.6, respectively (14). Other long-standing metabolic consequences of the climacteric include osteoporosis and osteoporotic fractures, skin changes, weight gain, and obesity, as well as degenerative disease of the central nervous system (CNS). There is still much to be discovered about the action of estrogen and other sex steroids on the vascular system, immunity, CNS function, and musculoskeletal disease, with special emphasis on the cellular level. A knowledge of symptoms in particular, however, and their effect on the daily lives of women, will assist the caregiver in providing competent care and providing women with long-standing professional assistance during the aging process. In fact, it will be very helpful to provide objective information about an individual's climacteric symptoms that may affect her quality of life.
1. THE CONSTRUCTION OF SCALES AND SUBScALES
As indicated earlier, reliable and valid measures of multisymptom conditions generally come in the form of scales and subscales, developed on the basis of principles of test construction and scaling (15). In the field of psychology, the
641
techniques developed to construct such measures became known as psychometrics. The first experimental psychology laboratory was founded by Wilhelm Wundt at the University of Leipzig in 1870. The interest was to establish the general principles of psychologic expression. However, it was very quickly realized that there is a wide variation in individual expression. The assessment of differences between individuals required the construction of measures sensitive enough to distinguish between subjects and the various items under investigation. This, in turn, led to attempts to construct "scales." Scales, by definition, are instruments that measure phenomena on a continuum using ordinal scaling (12). Scales measuring more complex human characteristics, such as intelligence or personality traits, invariably consist of a number of items that are summated to give an overall score for each person. They usually consist of a number of symptoms, yielding a total score that reflects the degree of severity of a condition along a graded continuum for each individual. Moreover, each symptom is usually rated in terms of its frequency of occurrence or severity. Scales for measuring a complex phenomenon or multifaceted syndromes are generally made up of a number of subscales, each measuring a different facet of the syndrome. Very often, summating symptoms from apparently different domains is meaningless. Greene, in his methodologic evaluation, has compared this to adding a person's height and waist measurement to give an overall measure of "size." Such a measure would fail to distinguish tall, thin people from small, obese people, because both would tend to have a similar overall "size" score (12). Similarly, the common practice of reporting symptoms indMdually is bound to fail because such a measure will not assess a condition comprehensively. Scales for measuring human characteristics or conditions can be categorized into two types, general scales and conditionspecific scales. Thus, the types of questionnaire developed focus on either generic or disease- and treatment-specific aspects. The contents of the different generic scales show many similarities, assessing the ability of patients to cope with their condition physically, emotionally, and socially, as well as their general performance at work and in daily life (16). Among the more commonly used instruments are the Sickness Impact Profile (17), Nottingham Health Profile (18), Q.gality of Well-Being Scale (19), and Short Form-36 (SF-36) Health Survey (20). The generic measures cover the multidimensional aspects of quality of life over a wide range of health problems. The scales may be less responsive to treatment-induced changes and could be considered lengthy and time consuming. The disease-specific measures are more likely to be responsive and make sense to clinicians as well as to patients. Their specific measures relate to concepts and domains in patient populations, diagnostic groups or diseases. One of the very first was the Women's Health Q.uestionnaire (WH Q) (21), a menopause-specific instrument. The W H Q_ consists of 37 items, including nine scales, and it assesses, in
642
addition to vasomotor symptoms, important areas such as other somatic symptoms, mood, sleeping problems, cognitive difficulties, and sexual functioning. Other test systems and questionnaires refer to psychiatric problems, such as the Beck Depression Index (22), which was designed to assess clinical depression in psychiatric patients. This index proved to be much less sensitive to change than were other nonpsychiatric measures. The depressed mood experienced by climacteric women obviously may not be less severe than that of psychiatric patients, but it has different origins and may have a different context from that of psychiatric depression. Other test systems include pain scores, sleep disturbances, and the assessment of sexual dysfunction, mental and cognitive function. The description of the merits and shortcomings of the different tools is presented elsewhere in this chapter (23). 2. MERITS OF FACTORANALYSIS
Factor analysis is a multivariant mathematical technique traditionally used in psychometrics to construct measures of psychologic and behavioral characteristics, such as intellectual abilities or personality traits (12). Theoretically, it addresses the problem of how to analyze the structure of the interrelationship (correlations) among a large number of variables (test scores, questionnaire responses, behavior, symptoms) by identifying a set of underlying dimensions known as factors. The overall objective of factor analysis is data summarization and data reduction. A central aim of factor analysis is the orderly simplification of a number of interrelated measures. Factor analysis describes the data using many fewer dimensions than original variables. It aims to order and give structure to observed variables and, by virtue of that, allows for the construction of instruments in the form of scales and subscales. The relationship between a symptom and a factor is measured by a correlation coefficient known as a factor loading. This wa~4 an instrument can be constructed that consists of several separate subscales and will measure different aspects of the symptom picture, based on the way symptoms cluster together within factors and on the size of the factor loadings. This results in a scale that yields a symptom profile for each subject (12). In the end, by identifying symptoms that cluster together or form groups of factors, one may be able to delineate facets of the symptom picture and identify those symptoms that are an essential part of a syndrome and those that are not.
B. Currently Employed Menopause-Specific Quality-of-Life Scales There has been a lack of standardized menopausespecific instruments for measuring the symptoms of climacteric women. Currently there are at least seven standardized
HERMANN P. G. SCHNEIDER
menopause-specific scales that satisfy the following four criteria formulated by Gerald Greene (12): 1. They have been constructed on the basis of a factor analysis. 2. They consist of several subscales, each measuring a different aspect of climacteric symptomatology. 3. The scales possess sound psychometric properties. 4. They have been standardized using populations of climacteric women. Of the instruments which fulfill these criteria and include self-report by the patients, the following eight are currently dominating. Although some of them do not necessarily need the criterion to be considered health-related quality-of-life (HRQoL) instruments, they are preferred because of their extensive clinical usage and the large amount of statistical information collected. A definition of HRQoL was developed by the PRO-Harmonization Group (24): "HRQoL represents the patients' evaluation of the impact of health condition and its treatment on daily life." According to their chronologic order of construction, the scales are: 9 Greene Climacteric Scale 9 Women's Health Q.uestionnaire (WHO.) 9 Q.ualifemme 9 Menopause-Specific Q OL Q.uestionnaire (MENQ_OL) 9 Menopausal Symptom List (MSL) 9 Menopause Rating Scale (MRS) 9 Menopausal Quality of Life Scale (MQOL) 9 Utian Menopause Q~ality of Life Scale (UQ_OL)
1. GREENE CLIMACTERIC SCALE
The first properly analyzed climacteric symptom scale was the Greene Climacteric Scale. In 1976, J. G. Greene developed his original 30-item self-administered scale (25). In essence, it was derived from an earlier study by Neugarten and Kraines (26). Gerald Greene recognized endocrine and emotional factors underlying the etiology and dynamics of menopause. Based on this theory, he investigated the relationship between menopausal symptoms. Climacteric symptoms factor analysis was used to establish independent domains such as vasomotor and physical. Originally, the symptoms of 50 women ages 40 to 55 years were scored on a four-point Likert scale (0 to 3). The results were intercorrelated using product-moment coefficients with a resulting matrix being submitted to principal component analysis. The final scale yielded three independent symptom groups or factors, equivalent to subscales. These were psychologic, somatic, and vasomotor symptoms. Items with factor loadings greater than +0.40 on one factor and less than 0.30 on the other two factors were included in the questionnaire. The resulting 21 items from
CHAPTER 48 Issues Relating to Q.uality of Life in Postmenopausal Women and Their Measurement an initial list of 30 were included in the scale. Those items with factor loading above +0.50 were given a weighting factor of 2. Gerald Greene's tool represents a pioneering piece of work. The original Greene Climacteric Scale, however, was never designed to be a genuine HRQpL instrument as it is defined today. It first applied quantitative techniques to questionnaire construction and marked the beginning of the use of factor analysis in clinical studies with "patientreported" outcomes as the endpoint in the field of women's health. Factor analysis has since been applied to samples for various size and characteristics in order to generate new menopausal scales. In order to reconcile the findings of seven such factor analytic studies and meet the demand for a "communal and comprehensive measure" of climacteric symptoms, Greene based his new tool on a sample of 200 rather than 50 women. His 1998 report (27) looked at the optimal number of factors or domains to be established with resultant "communal" scales of psychologic, somatic, and vasomotor symptoms. By only selecting symptoms found to have a factor loading of more than 0.35 in three or more studies, he also determined which symptoms should be included. Greene's new study replaced four items from the original 1976 scale with four new ones. Four other symptoms underwent a change in the wording. An additional item on loss of sexual interest was added, and the psychologic symptoms domain was broken down into an anxiety and a depressed mood subscale. The result is a 21-item, four-level questionnaire. This "standardized" Greene Climacteric Scale of 1998 was employed in a trial of Kliogest, a continuous-combined regimen with 2 mg 1713-estradiol and 1 mg NETA (28). In his detailed analysis of HRQgL instruments, ZSllner (29) classified the Greene Climacteric Scale as a valid instrument for the use in clinical trials. The impact of adverse effects of menopausal hormone therapy, however, is not adequately addressed, resulting in the need for a more comprehensive instrument. Table 48.1 presents an overview of currently employed menopause-specific quality-of-life scales. 2. WOMEN'S HEALTH QUESTIONNAIRE
The Women's Health Q.uestionnaire ( W H Q ) was developed by Myra Hunter in 1986 to assess a wide range of physical and emotional symptoms and to study possible health changes in women ages 45 to 65 years (21). It is composed of 36 items, which cover mood states, physical sensation and experience, sexual behavior, and vasomotor symptoms. A clinical sample of 682 women attending a routine ovarian screening program in London hospitals served as the basis. Self-reported symptoms were scored on a fivepoint Likert scale (values 0 to 4), which was reduced to a binary scale for factor analysis. None of the original 36 items
643
were excluded after factor analysis, but rather grouped into nine subscales or factors (see Table 48.1). Psychometric properties with respect to test-retest reliability, internal consistency, and validity are listed in Table 48.2. Subscale 9 (attractiveness) accounted for a small proportion of the variance and may be omitted, depending on the needs of research. Some interdependence between subscales cannot be denied. The WHQ_was used both in epidemiologic and intervention studies. It was employed in the Adelphi Women's Health Program in 1998, with subsequent publications (30-32). A double-blind, randomized, placebo-controlled multicenter clinical study was performed in 1995 (33). This trial examined 223 volunteering Swedish postmenopausal women with mild to severe climacteric symptoms at baseline in terms of their HRQgL response to transdermal estradiol or placebo patches, respectively. The W H Q . i s a valid and well-documented instrument for use in clinical trials. According to ZSllner (29), potentially MHT-driven adverse effects are difficult to delineate. Therefore, a more comprehensive instrument for HRQgL assessment of postmenopausal hormone therapy is required. 3.
QUALIFEMME
The Qualifemme instrument was developed in France to measure the impact of menopausal hormone deficiency on a woman's quality of life. The first version consisted of 32 items delineated from several other validated and accepted H R Q g L questionnaires. These items were translated and linguistically validated for use in France (34). The Qualifemme instrument is scored using a visual analog scale. Item weighting was achieved by a group of menopausal experts contributing their clinical experience. Initially, a subject pool of 351 women ages 51 to 68 was investigated. A principal component analysis resulted in the identification of five domains with 32 items: general (9), psychologic (12), vasomotor (2), urogenital (6), and a final domain covering pain and problems with hair and skin (3). Cronbach's c~ coefficient of 0.87 demonstrated internal consistency. The original instrument underwent a reduction process and 17 items were removed, resulting in the current 15-item instrument. The reduction was achieved while maintaining the instrument's quality psychometric standards (35). For details of test-retest reliability, internal consistency, and validity, see Table 48.2. The Qualifemme was applied in a multicenter trial in France during the years 1996 and 1998 with 36 centers comparing HRQoL before and after sequential versus continuous application of 1713-estradiol percutaneous gel and nomegestrol acetate in a total of 141 postmenopausal women. The Qualifemme proved sensitive by demonstrating a 44.6% increase in the global quality-of-life score for those on sequential treatment and 38.3% in the continuous-combined
644
HERMANNP. G. SCHNEIDER TABLE 48.1
Instrument
Currently Employed Menopause-Specific Q.uality-of-Life Scales Author
Greene Climacteric Scale
Greene (1976, 1998)
Women's Health Q.uestionnaire (WHO.)
Hunter (1992)
Qualifemme
Floch (1994)
The Menopause-Specific Q.OL Q.uestionnaire (MENQ_OL)
Hilditch (1996)
Menopause Rating Scale (MRS)
Schneider (1996, 2000)
Menopausal Symptom List (MSL)
Perz (1997)
Menopause Q.uality of Life Scale (MQOL)
Jacobs (2000)
Utian Q.uality of Life Scale (UQOL)
Ufian WH (2002)
treatment group (36). Thus, Q.ualifemme appears to be a valid instrument and attempts to include side effects of menopausal hormone therapy such as androgenic skin effects.
4. MENOPAUSE-SPECIFICQUALITY OF LIFE QUESTIONNAIRE The Menopause-Specific Q.uality of Life Q.uestionnaire (MENQ.OL) was developed by a group of researchers from Canada during the mid-1990s (37). A list of symptoms or problems that could be experienced by postmenopausal women was created using the menopause and quality-of-life literature and existing menopause and quality-of-life questionnaires plus the clinical experience of the investigators. The final item-reduction questionnaire had 106 items. From the original five domains, physical, vasomotor, psychosocial and sexual, and working
Domains covered 9 Psychologic 9 Somatic 9 Vasomotor 9 Depressed mood 9 Somatic symptoms 9 Vasomotor 9 Anxiety/fears 9 Sexual behavior 9 Sleep problems 9 Menstrual symptoms 9 Memory/concentration 9 Attractiveness 9 Climacteric 9 Psychosocial 9 Somatic 9 Urogenital 9 Vasomotor 9 Psychosocial 9 Physical 9 Sexual 9 Psychologic symptoms 9 Somatovegetative symptoms 9 Urogenital symptoms 9 Psychologic 9 Vasosomatic 9 General somatic 9 Physical 9 Vasomotor 9 Psychosocial 9 Sexual 9 Occupational quality of life 9 Health-related quality of life 9 Emotional quality of life 9 Sexual quality of life
Items List of symptoms
Fully phrased statements (symptoms and feelings)
List of symptoms
List of symptoms, signs, feelings
List of symptoms
List of symptoms
Fully phrased statements (symptoms and feelings)
Fully phrased statements (symptoms and feelings)
life, upon completion of the study, the domain "working life" was finally omitted. A 32-item menopause-specific H R Q p L instrument resulted. It encompasses four subscales (physical, vasomotor, psychosocial, and sexual) plus one overall HRQ.gL item (see Table 48.2). Each domain is scored separately within a possible range from 1 (not experiencing a problem) to 8 (extremely bothered). The mean of the subscale is taken as the overall subscale score. In analogy to the W H Q ~ no overall score can be obtained from this questionnaire, as the relative contribution of each domain to an overall score is unknown. Reliability, responsiveness, and construct validity of the M E N Q O L instrument (see Table 48.2) were determined within a randomized, parallel-group design trial of conjugated versus transdermal estrogen, both supplemented with medroxyprogesterone acetate (MPA) in a sequential fashion.
645
CHAPTER 48 Issues Relating to Quality of Life in Postmenopausal Women and Their Measurement TABLE 48.2 Scale Greene Climacteric Scale
WHO..
Qualifemme
Structure Psychologic (P) scale 11 items: 1 to 11 (anxiety I to 6; depression: 7 to 11): 1. Tachycardia 2. Nervousness 3. Insomnia 4. Being excitable 5. Panic attacks 6. Difficulty concentrating 7. Tired, lack of energy 8. Lost interest in most things 9. Unhappy or depressed 10. Crying spells 11. Irritability Somatic (S) scale 7 items: 12 to 18: 12. Dizzy or faint feelings 13. Pressure, tightness in head, body 14. Paresthesia 15. Headaches 16. Arthralgia, myalgia 17. Loss of feeling in hands, feet 18. Breathing difficulties Vasomotor (V) scale 2 items: 19 and 20: 19. Hot flushes 20. Sweating at night "Probe" for sexual dysfunction: 21. Loss of interest in sex 9 Depressed mood 9 Somatic symptoms 9 Vasomotor 9 Anxiety/fears 9 Sexual behavior 9 Sleep problems 9 Menstrual symptoms 9 Memory/concentration 9 Attractiveness NB: Symptoms are not listed in blockwise manner to complete one domain and proceed to the next; rather, items from all domains are shuffled across the questionnaire
15-item questionnaire 9 Climacteric (2) 9 Psychosocial (5) 9 Somatic (4) 9 Urogenital (4) 9 Item weighting was determined by a group of menopausal experts adding clinical experience to the instrument
Characteristics of the Scales Scoring 4-point scale Patient asked to indicate degree to which she is bothered "at the moment" by listed symptoms 0: not at all 1: a little 2: quite a bit 3: extremely
4-point scale Patient asked to indicate her agreement with statements as listed: Yes, definitely Yes, sometimes No, not much No, not at all NB: Scoring had to be reversed for some items, as these were phrased positively rather than negatively
Visual analog scale 100 mm
Psychometric properties
Reliability Reliability coefficients: P scale: 0.87 S scale: 0.84 V scale: 0.83 Validity Content validity: Only symptoms with statistically significant factor loading (confirmed by other factorial studies in the latest version) have been included Construct validity: Has been demonstrated in relation to 9 Life stress 9 Bereavement 9 Psychologic 9 Treatment 9 HRT
Reliability Test-retest (2 weeks): Correlation between factor scores (scores of same factor at two different times) ranged between 0.69 and 0.96 across factors Internal consistency: Assessment not deemed necessary as underlying factor analysis provides sufficient information about item to scale relationship Validity Concurrent validity of psychologic scales was estimated by comparison with the 30-item GHQ_, which correlated 0.81 with the depression scale and 0.46 with the anxiety scale
Reliability Test-retest (4 weeks): Pearson's correlation coefficient ranged from 0.84-0.98 Internal consistency: Cronbach's o~ coefficient of 0.73 Validity PCA resulted in five distinct factors explaining 71.8% of variance
Continued
646
HERMANN R G. SCHNEIDER TABLE 48.2
Scale MENQ_OL
MSL
MRS
MQOL
Characteristics of the Scales - - cont'd
Structure Physical (16 items) Vasomotor (3 items) Psychosocial (10 items) Sexual (3 items) Plus general Q_OL question, which was not included in analysis
Total of 25 items covering three domains 9 Psychologic symptoms 9 Vasosomatic symptoms 9 General-somatic symptoms Total of 11 items covering three domains 9 Psychologic symptoms 9 Somatovegetative symptoms 9 Urogenital symptoms
9 Energy (10 items) 9 Sleep (3 items) 9 Appetite (1 item) 9 Cognition (5 items) 9 Feelings (12 items) 9 Interactions (7 items) 9 Symptoms impact (10 items)
Scoring 7-point scale Patient asked to indicate how strongly she is bothered by symptoms listed: 0: not at all 1
2 3 4 5 6: extremely Intermediate levels do not carry any description NB: Scoring had to be reversed for some items, as these were phrased positively rather than negatively
6-point scale to indicate severity and frequency
5-point scale Patient asked to indicate the severity of symptoms 0: no symptoms 1: mild 2: moderate 3: marked 4: severe
6-point scale Patient asked to rate the degree to which the following statements apply to you 1: I am never like this 2 3 4 5 6: I am always like this In addition, a global item included, in which patients are asked to score their overall quality of life on a 100 point scale
Psychometric properties Reliability Test-retest (1 month): Correlation between factor scores (scores of same factor at two different points in time) ranged between 0.55 and 0.85 across factors Internal consistency: Cronbach's ot ranged between 0.81 and 0.89, indicative of fairly high internal consistency
Validity Construct validity: Correlation coefficients are computed for both evaluative and discriminative validity and oscillate between 0.40 and 0.65 (discriminative), or 0.28 and 0.60 (evaluative) (global Q OL item excluded) Test-retest reliability with satisfactory coefficients Validity of limited experience
Reliability Test-retest (1.5 years): Correlation between factor scores (scores of same factor at two different points in time) for severity of scores: K = 0.25 P = 0.000 Somatic symptoms: K = 0.25 P - 0.000 Psychological symptoms: K - 0.30 P - 0.000 Urogenital symptoms: K = 0.19 P = 0.000 Comparison with SF-36 Reliability Internal consistency: Cronbach's oLranged from 0.92 for the total MQ.OL scale, and from 0.91 to 0.69 for each of the menopausal Q OL domains demonstrating a good level of internal consistency
CHAPTER 48 Issues Relating to Quality of Life in Postmenopausal Women and Their Measurement TABLE 48.2 Scale
UQOL
Characteristics of the Scalesmcont'd
Structure
9 Occupational (7 items) 9 Health (7 items) 9 Emotional (6 items) 9 Sexual (3 items)
647
Scoring Named end points: "Perfect quality of life," "Might as well be dead," as well as seven additional quantifiers along the scale 5-point scale Patient asked to rate the degree to which the following statements, apply to you 1: Not true for me 2 3: Moderately true 4: 5: Very true of me Negative item: Scores reversed Missing values: Single missing value scored at the mean of the remaining domain values, second missing score resulted in missing domain score
Improvement was observed in all domains; however, there were significant differences between groups (38). Poor discriminative power was observed in the vasomotor domain, but it was good in other domains. Evaluative performance was fair in the vasomotor and libido, poor in the global subscale. In addition to such identified shortcomings, the lack of factor analysis is another disadvantage, as it withholds correlation patterns of data variants. M E N Q O L also does not address the full picture of potential side effects of menopausal hormone therapy (29). 5. MENOPAUSALSYMPTOM LIST The Menopausal Symptom List (MSL) was developed in 1997 to measure the severity of symptoms commonly associated with menopause. The theoretical symptom checklist was sent to 40 women ages 45 to 55 years living in Australia. Following completion of two principal component analyses, 25 significant items emerged in three domains, labeled psychologic, vasosomatic, and general somatic (39). The latter combines the anxiety and depression subscales of the Greene Climacteric Scale and the Women's Health Questionnaire. The vasomotor subscale, besides two vasomotor symptoms, also includes other somatic symptoms for reasons not quite apparent. The items are scored on a six-point Likert scale of both frequency and severity. The MSL is a symptom inventory in terms of the selection, wording of items, and its scoring. Test-retest reliability coefficients are satisfactory. Validation experience is limited.
Psychometric properties
Reliability Internal consistency: Cronbach's oLvalue of 0.83 indicative of fairly high internal consistency
Validity
Comparison with SF-36 resulting in Pearson correlations demonstrating scale validity
Standard hormone replacement therapy will significantly reduce both frequency and severity of classical menopausal symptoms. In 2002, Hogervorst et al. (40) systematically reviewed the effect of menopausal hormone therapy on cognitive function. Their study included 15 publications with a total of 566 postmenopausal women. This meta-analysis did not report any favorable effect of M H T on cognitive functions (verbal measures, spatial measures, speed of reading, or memory). In the W H I Study, no favorable effect on cognitive function associated with the use of M H T was found (41). Overall, randomized data systematically report that H R T improves quality of life when it is hampered by the presence of climacteric symptoms. W h e n symptoms are not present, M H T does not improve quality of life and would not do so in elderly women. Therefore, clinical practice would foster the application of menopause-specific questionnaires and their validated association with quality of life in women of transient menopausal age. A globally validated and short symptom scale would optimally serve such practical purposes. The Menopause Rating Scale (MRS) is the prototype (42).
6. MENOPAUSERATING SCALE The first version of the Menopause Rating Scale has been used since 1992 (43,44). It was initially developed to provide the physician with a tool to document specific climacteric symptoms and their changes during treatment and was seen as an improvement over the commonly applied Kupperman Index.
648 A critical assessment of this new scale, however, disclosed methodologic deficiencies, which both in theory and practice limited its use. This resulted in proposals to improve the physician-based scale: 9 Application of the scale in a representative sample of women after questionnaire revision. 9 Revision of the questionnaire such that women will complete it themselves; first of all because self-assessment is more sensitive, and second, a self-administered questionnaire would not limit future application. 9 Modification of the wording of items to a simple, laymen-appropriate form. 9 Proper psychometric evaluation of the revised scale based on a representative sample and development of simple-to-use standardized items with clear dimensions. 9 Classification of the severity of complaints based on a normal population sample. 9 Provision of normative data, representative for the climacteric age in the female population. In early 1996, the MRS questionnaire was standardized using a representative random sample of 689 German women ages 40 to 60 years in order to obtain a new and standardized Menopause Rating Scale (45). The first revision of the questionnaire mainly concerned the layout, including some adjustments regarding the number, structure, and wording of items; these were made to support applicability as self-administered questionnaire. The MRS was formally standardized following up-to-date psychometric rules. The revised scale consists of a list of 11 items to be answered: The respondent has a choice among five categories: no symptom, mild, moderate, marked, or severe complaints/symptoms (see Table 48.2). Factor analysis of the standardized l 1-item version encompassed three domains: psychologic, somatovegetative, and urogenital dimension. Scoring is based on a five-point Likert scale ranging from no symptoms to severe symptoms. A follow-up investigation was performed from August to October 1997 on 306 women from the original study. The retest reliability of scores between the two points was evaluated using Pearson's correlation coefficient. The results of the follow-up survey demonstrate stability of the individual scores. The total score and scores of the three defined factors have significant agreement as demonstrated using K statistics (46). The validity of the MRS to measure HRQoL in postmenopausal women was determined by comparing the instrument to both the Kupperman Index (1,2) and the SF-36 (47). In the Kupperman Index, weighting factors are based on the intuition of the developers. The criterion of assignment of a weighting factor based on "prevalence and consequence" is not explicit and merges distinct concepts into one coefficient. No quantitative research or psychometric validation has ever been given to this simple symptom questionnaire of the late 1950s. The MRS proved to be a much more
HERMANN E G. SCHNEIDER
sound and accurate instrument than the Kupperman Index; the differences between scores could easily be explained by the domains resulting from factor analysis. Kendall's T-b coefficient and Pearson's correlation coefficient both demonstrated a high degree of association between the results of both instruments (47). Truly more significant were the results of comparing the MRS to SF-36. When comparing the psychologic and somatovegetative MRS subscales to the eight domains of the SF-36, they did not correlate equally well across all domains. However, the pattern of correlation was understandable, as the highest degree of correlation occurred in the domains of the SF-36 that are most relevant to women during the menopausal transition (48). Thus, the MRS is a reliable, well-defined instrument for measuring the impact of climacteric symptoms on quality of life (49,50). It should be regarded as a brief and compact instrument, easy to complete and to score, and suitable for routine controls. It covers the key complaints of women during and after menopause. This type of scale is not tailored to detail specific therapies to the needs of each individual woman. The MRS had an excellent international response and acceptance. The first translation was into English (51). Other translations followed (52), this way applying international methodologic recommendations. Currently, the following versions are available: Brazilian, Bulgarian, Belgium-French, Belgium-Dutch, Chilean, Chinese, Croatian, English, French, German, Greek, Indonesian, Mexican/Argentinean, Polish, Spanish, Swedish, Rumanian, Russian, South African Engfish, South African Afrikaans, Turkish, Ukrainian (Russia), Ukrainian (Ukraine). Some of these versions are available in a published form (52), and all--including the unpublished-can be downloaded in PDF format from the internet (see ref. 48 and www.menopause-rating-scale.info). Details of compatibility of the MRS scores among countries are given elsewhere (53). 7. MENOPAUSALQUALITY OF LIFE SCALE
The Menopausal Q.uality of Life Scale (MQOL) was developed in 2000 (54). It was intended to create a condition-specific questionnaire that examines the effects of menopause on HRQoL, as well as the impact of employment, age, and medical history; in addition, cross-sectional information on differences in HRO_gL was obtained in a community-based sample of women consequent to a self-rated change in menopausal status. Hnally, the effects of M H T in the early postmenopause were investigated. Subject pools of 32 and later another 29 women were interviewed, and the results of such preliminary information served to develop a pilot questionnaire containing 63 items divided into seven domains. These were energ~ sleep, appetite, cognition, feelings, interactions, and symptoms impact. Each of these items is reported using a six-point Likert scale. Of the total of 191 distributed questionnaires, 99 were returned. The following psychometric analysis resulted in a
CHAPTER 48 Issues Relating to Q.uality of Life in Postmenopausal Women and Their Measurement 48-item questionnaire, as well as a global HRQoL question to rate the overall quality of life. A second analysis using oblimin rotation was performed resulting in a seven-factor hierarchical structure, which accounted for 57% of the data variance. This structure proved unstable across subsamples. As a result, the MQ_OL questionnaire was scored by an overall score instead of seven subscale scores for each of the seven domains. Reliability data are presented in Table 48.2. Strong correlations of interdependence between domains could be shown. Consequently, the global quality-of-life index was disregarded as a single factor, indicating that all the items were evaluated with the same importance and were added in a total score. The attempts to circumvent or mask the psychometric shortcomings of this instrument both undermine the empirical foundation of this questionnaire construction. 8. UTIAN QUALITYOF LIFE SCORE
The Utian Q.uality of Life Score is a modification of the original Utian questionnaire from the 1970s. The UQ_OL was systematically developed from the old questionnaire designed to assess the sense of well-being of participants in a treatment study comparing estrogen to placebo (55). The UQ_OL is focused on general quality of life rather than Q_OL in menopausal women. It was developed with twostage application of factor analysis. The 23-item instrument with a five-point rating scar for each item has four subscales (occupational, health, emotional, and sexual). A field study was conducted on 327 women recruited from 11 separate communities throughout the East and Midwest of the United States. Subjects ranged from 46 to 65 years of age. The resulting 23-item instrument was subsequently administered to a second sample of 270 menopausal women and then readministered to determine test-retest validity. The SF-36 was concurrently administered to determine scale validity. This scale can measure severity of Q.OL burden. Only limited data on reliability and validity are as yet available. The relatively low number of menopausal symptom-specific items may require a parallel application of another more menopausal symptom-related scale for the most widely practiced application of such scales, which is during the menopausal transition. Reliability and validity properties are documented in Table 48.2.
C. Health-related Quality of Life in Postmenopausal Women In order to document the practice of assessing quality of life in postmenopausal women, a few pertinent studies shall be referred to.
649
1. THE BERLIN STUDY
As a basic instrument for quantification of menopausal symptoms, the Menopause Rating Scale was applied in a Berlin Study in 1994 with 230 women with psychosocial determinants of menopausal symptoms. The latter included lack of social support, deficit in self-esteem, and stressful reorientation. For personality identification in this Berlin sample, the Freiburg Personality Inventory (FPI), widely acknowledged and evaluated in Germany, and a projective sentence-accomplishing technique, were applied. The returns were analyzed as material for the evaluations of wellbeing in menopausal women. Attitudes of the women toward menopause could be transformed into items and scales of well-defined diagnostic quality. Scales of this Berlin Menopause Q.uestionnaire have been factor-analyzed and evaluated on a large (n = 603) nationwide German representative sample (56). A questionnaire applied in the Berlin Study was based on results of a pilot investigation and contained 90 items within 13 psychosocial domains. Eight scales were characterized as a result of a factor analysis (Table 48.3). The first four scales relate to complaints such as depressive mood, sleeping disorders, irritability, and exhaustion. Furthermore, the instrument records women's self-confidence, quality of partner relationship, reorientation initiated by menopause, and the absence of menopause-related problems. Based on this information, a validated test instrument consisting of 32 items was developed creating individual profiles of coping and quality of life in menopause conditions. According to their own appraisals, menopause is inconspicuous in 80% of women, who experience no apparent loss in quality of fife. Two out of five women emphasize the physical relief of the menopause, with a resultant general improvement in well-being (Fig. 48.1). Instead of focusing on detailed climacteric complaints, the test asks about the joy of riving and quality of fife. A total of 62% of our probands reported positive attributions to the menopause itself. Altogether, 78% of women took the interview experience as a means to further organize fife in a more conscious manner. TABLE 48.3 Subscales to the Berlin Menopause Q.uestionnaire Depressive moods: "I live in constant worry." Reorientation: "A new life period starts for me." Sleeping disorders: "At night, I lay awake." Irritability: "When I get frustrated, I cannot control myself." Problem-free: "I have no problems with menopause." Self-esteem: "I am happy with myself." Exhaustion: "I have no energy." Q.uality of relationship: "My partner relationship is trustful." Reproduced from res 49.
650
HERMANN E G. SCHNEIDER
TABLE 48.4
Age-Related Perception of Personal Image Percentage of postmenopausal women ages 50-70 years
Feature
FIGURE 48.1 Individualperception of positive effects of menopausal
age. (Reproducedfrom res 49.)
In public opinion, menopausal transition relates to biologic changes co-occurring with social and psychologic alterations during midlife. Very often, psychosocial relief is not encountered and does not contribute positively to perimenopausal age. The fact that children often leave home around mother's menopause is predominantly associated with a loss ("empty nest syndrome"). The psychoanalytic literature on menopause stresses the "empty nest syndrome," mental depression, and the offense of not being in command of reproduction any longer. In contrast to this general attitude, women in the Berlin Study pointed to the advantages of greater personal independence. Only 20% of all participants complained of empty nest symptoms, pointing to different attitudes in the younger generation; they also feel strong relief from menstrual problems, premenstrual complaints, contraceptional obligations, and pregnancy complications of older age (see Fig. 48.1). With increasing age, the quality of sexual life is of growing importance. Tender loving care is a dominant issue in 75% of the interrogated women. In a cluster analysis of this Berlin Study, three types of menopausal coping styles were identified. The first cluster identifies pragmatic women, a more or less problem-free group of 37% with discrete menopausal complaints but with a good level of self-esteem, regarding themselves as attractive. They deny being seriously affected by menopausal changes. Possibly, this pragmatic group has a repressor coping strategy and shields the occurring symptoms with self-discipline. The majority of women (75%) stated no loss in their attractiveness. Women who judged themselves attractive showed fewer menopausal symptoms. With increasing age this individual positive body image seems to modify. Individual features, as seen from our study in 1997 (56) with 1000 postmenopausal women ages 50 to 70, are depicted in Table 48.4 (49). Women with low self-esteem score much higher in the MRS in all our studies (Fig. 48.2). A very important finding is the high correlation between personal professional activity and a quantified low degree of menopausal complaints. The positive feedback of health-promoting behavior needs to be emphasized. Regular exercise correlated signifi-
Drop in efficiency Figure changes Gain in weight Skin slackness Get wrinkles Decrease in attraction
(~ = 1038)
49 39 35 35 30 13
Reproduced from ref. 49. cantly with high self-confidence and with fewer menopausal complaints. Another factor analyzed was reorientation in life. The Berlin Study revealed that 50% of women experienced a reorientation process in their life that was initiated by menopause and that had the consequence of rearranging their lifestyle. A trend toward a creative form of reorientation dominates. Those subjects presenting with a high level of positive reorientational motives in the questionnaire were the ones with a low level of psychologic complaints in the MRS. This group of women look forward to new perspectives in their individual fives. Another group considers themselves as being forced into a form of reorientation that is not at all dominated by the women's own intentions. Here we found a correlation with high levels of psychologic complaints (Fig. 48.3). A further predictor for a rather problem-free menopausal coping is a satisfactory partner relationship. The impact of satisfactory interpersonal relationships and of a secure social net is clearly evident. The personal relationship and its quality grow in importance during midlife. Women without partners score both low and high levels of menopausal complaints in the MRS. Looking at women with partners, when asked about the quality of their partnership, those who were dissatisfied were those who suffered more complaints (Fig. 48.4). What is the overall message of this Berlin Study? Important sequelae for the understanding of well-being in menopausal women are women's self-confidence, the quality of their
FIGURE 48.2 Self-confidenceand well-being in menopausal women.
(Reproduced from ref. 49.)
CHAPTER 48 Issues Relating to Q.uality of Life in Postmenopausal Women and Their Measurement
FIGURE 48.3 Reorientation and well-being in menopausal women. (Reproduced from ref. 49.)
partner relationship, and the reorientation process initiated by menopause or by their psychosocial condition. Good selfconfidence is a predictor for successful coping. A satisfying relationship and social network improves quality of life. Employment is confirmed as a protective factor. Furthermore, several types of physical and psychosocial relief have to be considered in the assessment of well-being in menopausal women. Introducing these variables into the interaction between a woman and her counseling doctor will allow for better cooperation and a higher degree of compliance with treatment.
651
The absolute improvement of the symptoms during treatment was 9.3 points of the MRS total score on average. An important aspect is whether or not a scale like the MRS is good enough to detect even treatment-related changes in women with only few or mild symptoms. This question seems even more important when referring to a metaanalysis of HRT and cognitive function (40). One of Hogervorst's conclusions from this study was: "When symptoms are not present, H R T does not impress Q OL and would not do so in elderly women." Our MRS experience is documented in Fig. 48.5. The relative improvement of complaints or quality of life increases with the degree of severity of symptoms at baseline, which is consistent with the general expectation. It is, however, important to underscore that the MRS scale apparently detects a positive treatment effect also in women with few complaints (57). The MRS-assisted assessments of menopausal hormone therapy and the meta-analysis of Eva Hogervorst both would explain why the W H I investigationmwith menopausal complaints as exclusion criterionmdid not produce major benefits in terms of quality-of-life outcomes except improved vasomotor symptoms and a small benefit in terms of sleep disturbance (40,58).
3. THE PAN-ASIA MENOPAUSE (PAM) STUDY 2. THE MENOPAUSE RATING SCALE AS OUTCOME MEASURE FOR HORMONE TREATMENT
An open, uncontrolled postmarketing study with more than 9000 women with pretreatment and posttreatment data of the MRS scale was organized to evaluate the capacity of the scale to measure the health-related effects of hormone treatment independent from the severity of complaints at baseline. Hormone therapy consisted of a combination of 2 mg estradiol valerate continuously and 1 mg cyproterone acetate in a sequential addition (Climen). The mean age was 49.8 years (SD 6.4); about half the women participating were still perimenopausal (51.9%), the others already in the postmenopausal period (48.1%). The mean body mass index was 24.7 (SD 3.7).
FIGURE 48.4 Relationshipwith partner and well-being in menopausal women. (Reproduced from ref. 49.)
A group of Asian colleagues has organized a prospective, randomized, double-blind multinational clinical trial in 1028 healthy postmenopausal women of nine ethnic groups from eleven Asian countries or regions (59). Following 2 weeks of basal observation, the women received one of three conjugated estrogens and medroxyprogesterone acetate doses. At baseline and at the end of 1, 3, and 6 months following the start of therapy, the study participants were asked to record on a M E N Q O L questionnaire 29 menopausal symptoms, as experienced during the preceding month. The symptoms were categorized into four domains: vasomotor,
FIGURE 48.5 HRT: relative change of the MRS. Mean values (SD) in four categories of severity at baseline. (From ref. 49.)
652
HERMANN R G. SCHNEIDER
psychosocial, physical, and sexual. The results are shown in Table 48.5. Overall, Vietnamese and Pakistani women had the highest baseline scores--that is, were most afflicted by each set of symptoms in a given d o m a i n ~ a n d Indonesian, Malay, Taiwanese, and Thai women were least afflicted. In the overall population, intervention resulted in statistically significant decreases in the scores of all four domains within 1 month of intervention. It is of special interest to observe that the prevalence of four domains of menopausal symptoms, representative of quality of life as recorded on a MENQ_OL questionnaire, vary considerably among ethnic groups of Asian women. Consequently, multinational studies across ethnic groups need to consider such variation.
performed in North America, Europe, and Australia. The PFSF is reliable and valid for use in women with hypoactive sexual disorder and was demonstrated to have robust psychometric properties across numerous geographies (61). Sexuality certainly has great bearing on quality of life. Recent European questionnaires in four to six different countries agree to reduced sex drive concerning a good third of women ages 50 to 60 years. Of these, only small proportions of up to 10% ever seek treatment. In a scale of 1 (extremely unhappy) to 10 (extremely happy), sexual satisfaction of the average European woman was rated around 6, vaginal pain being the leading symptom. These women also admit that an "increased sex life would make me feel more feminine and helps confidence and selfesteem" (62). A detailed report is in preparation (63).
4. ASPECTS OF SEXUALITY All these mentioned questionnaires and inventories more or less consider sexual behavior a separate domain or item. More specific treatment options as well as considerable variation across populations urged clinical scientists to develop broader profiles of female sexual function and appropriate ways of assessment. Hypoactive sexual desire disorder (HSDD) is defined as a persistent deficiency or absence of sexual fantasies and desire for sexual activity that causes marked distress or interpersonal difficulty (60). Women who have undergone menopause, whether natural or surgical, may experience a significant reduction in sexual desire. The Profile of Female Sexual Function (PFSF) is a new, self-reported multinational instrument designed to measure sexual desire and associated symptoms in women with H S D D following menopause. Specifically, this instrument, which contains 37 items in seven separate domains and an overall sexual satisfaction question, has been developed to reflect the clinical phenomenology of this disorder and its effects on the patients' thoughts, feelings, behaviors, and emotions. The initial development of the PFSF was
TABLE 48.5
Ethnic origin Chinese Filipino Indonesian Korean Malay Pakistani Taiwanese Thai Vietnamese SD, standard deviation. Data from ref. 60.
No. of women 249 199 60 97 24 60 81 150 100
III. CONCLUSION In recent years, there has been a growing awareness among clinicians of the importance of learning all about how patients cope with symptoms of the climacteric. Healthrelated quality of life is a subjective parameter commonly used to assess the views of the patients in terms of the physical, social, and emotional aspects of living with their condition. Direct questioning is a simple and appropriate way of accruing information about how patients feel and function. Using standardized questionnaires ensures welldocumented psychometric properties. For routine application in clinical practice or in clinical trials, it is essential that the instruments employed are simple and comparatively short. The majority of patients or probands welcome the opportunity to report how symptoms and their subsequent treatment affect daily life. Psychometrically evaluated questionnaires allow uniform administration and unbiased quantification of data because the response options are predetermined and thus equal for all respondents. A core set of
PAM Study: Baseline Domain Scores by Ethnic Group Vasomotor
MENQ_OL (29) domain (mean _+ SD) Psychosocial Physical
3.13 (1.67) 3.17 (1.60) 2.28 (0.87) 2.21 (1.40) 3.02 (1.56) 4.96 (2.41) 2.29 (1.39) 2.87 (1.61) 5.71 (1.59)
2.84 (1.37) 3.33 (1.41) 2.40 (0.68) 3.06 (1.46) 2.78 (1.11) 4.24 (1.64) 2.37 (1.32) 3.10 (1.22) 5.96 (1.48)
3.21 (1.15) 3.20 (1.23) 2.66 (0.63) 3.29 (1.24) 2.93 (1.08) 4.84 (1.61) 2.84 (1.23) 3.28 (1.08) 5.39 (1.20)
Sexual 4.04 (2.20) 3.03 (2.03) 2.63 (1.18) 3.55 (2.29) 3.14 (1.78) 2.90 (1.70) 2.11 (1.32) 2.89 (1.90) 6.55 (1.67)
CHAPTER 48 Issues Relating to Q.uality of Life in Postmenopausal Women and Their Measurement questionnaires would allow the comparison of study results in patient populations. The growing awareness of an interest in the subjective aspects of quality-of-life outcomes is evident by the increasing number of publications in this area. A growing emphasis has been on self-administered questionnaires. Unless conventional variables are supplemented with self-assessment measures, a limited picture of the impact of symptoms and the effect of treatment is obtained. Certain difficulties, however, introduce bias into the interpretation of data. These include the experiences of some interviewed individuals, particularly of older age who might have difficulty with reading or writing, or who have been exposed to less experienced interviewers, or simply the expenses involved in gathering quality-of-life data. Standardization, compatibility, eradication of possible bias, and economy are therefore important variables for the validity of any type of quality-of-life assessment. The application of healthrelated quality-of-life instruments requires the same scrutiny and intention as the measurement of physiologic outcomes. Random and representative samples of the population should be investigated in sufficient numbers and over prolonged periods. In terms of statistics, quality of life is, by definition, an assessment of multiple variables. The use of many measures and multiple statistical tests reduces the statistical power of the analysis. Health-related quality of life certainly is a multidimensional concept; there is a continuing debate as to whether or not the aggregation of several dimensions into a summary index is appropriate. A summary score may falsely suggest improvement in one vital area and conceal deterioration in another. Indices, however, are practical and are a convenient method of information transfer. Q O L measures are increasingly used for measuring health outcomes in evaluative research. There is evidence of a lack of consistency in the selection of measures for clinical trials which hinders comparison among studies. Concurrent evaluation and professional consensus will assist to determine the most suitable measure for a particular application. There may be a need for combining general with disease- or populationspecific measures of Q OL, as they complement each other. In a larger representative Berlin Study, important sequelae for the understanding of well-being in menopausal women were found to be women's self-confidence, the quality of their partner relationship, and the reorientation process initiated by menopause and their psychosocial condition. Employment is considered to be a protective factor. The experience of relief from several physical and psychosocial conditions has to be considered in the assessment of well-being in menopausal women. Other examples of application document the prevalence of individual menopausal symptoms to differ among ethnic groups of Asian women. Within each ethnic group, the percentage of women reporting each symptom varies substantially. A hypoactive sexual desire disorder causes marked distress or interpersonal difficulty with severe impact on quality of life. In addition to the menopause-related questionnaires and inventories,
653
which more or less consider sexual behavior a separate domain, a more specific evaluation has emanated. H u m a n beings are social individuals. If one changes the health status or quality of life of an aging person, the partner might also be affected, sometimes strongly and with positive or negative interaction. This is rarely considered in the development of tools to measure treatment. A well-defined menopausal complaint rating scale may serve as a less troublesome, less time-consuming, and therefore rather practical instrument to address the impact of treatment on various aspects of quality of life and at the same time avoid a wide-ranging battery of questionnaires with their practical drawbacks.
References 1. Calman KC. Quality of life in cancer patients--an hypothesis.J ivied Ethics 1984;10:124-127. 2. Alder B. How to assess quality of life--aspects of methodology. In: Schneider HPG, ed. Hormone replacement therapy and quality of life. Carnforth, New York: Parthenon Publishing, 2002:11- 22. 3. McKinlay SM, Brambilla DJ, Posner JG. The normal menopause transition. Maturitas 1992;14:103-115. 4. MacMahon B, WorcesterJ. Age at menopause.United States 1960-62. Vital Health Stat 1966;11:1-20. 5. Speroff L. A signal for the future. In: Lobo R, ed. Treatment of the postmenopausal woman. New York: Raven Press, 1994:1-8. 6. FriesJE Aging, illness, and health policy: implications of the compression of morbidity. Perspect Biol Med 1988;31:407-428. 7. Fries JE Aging, natural death and the compression of morbidity. N E n g l J M e d 1980;303:130-135. 8. FriesJF, Green LW, Levine S. Health promotion and the compression of morbidity. Lancet 1989;1:481-483. 9. KuppermanHS, Blatt MHG, Wiesbader H, et al. Comparative clinical evaluation of estrogen preparations by the menopausal and amenorrhoea indices.J Clin Endocrino11953;13:688-703. 10. Kupperman HS, Wetchler BB, Blatt MHG. Contemporary therapy of the menopausal syndrome.JdMd 1959;171:1627-1637. 11. Martin EA. The Oxford medical dictionary. Oxford, UK: Oxford University Press, 1994. 12. Greene JG. Generic score assessment of quality of life and climacteric subscales. In: Schneider HPG, ed. Hormone replacement therapy and quality of life. Carnforth, NY: Parthenon Publishing, 2002:35-44. 13. Hiillstr6m T, Samuelsson S. Mental health in the climacteric. The longitudinal study of women in Gothenburg.Acta Obstet Gynecol &and 1985;130(suppl):13 - 18. 14. Anda RF, Waller MN, Wooten KG, et al. Behavioral risk factor surveillance, 1988. M M W R CDC Surveill Summ 1991;39:1-22. 15. Peck D, Shapiro C. Measuring human problems." a practical guide. Chichester, UK: Wiley, 1990. 16. Fitzpatrick R, Fletcher A, Gose S, et al. Q.uality of life measures in health care. I: Applications and issues in assessment. Br MedJ 1992; 305:1074-1077. 17. BergnerM. Development,use and testing of the sickness impact profile. In: Walker S, Rosser M, eds. Quality of life assessment."key issues in the 1990s. Dordrecht, Netherlands: KluwerAcademic Press, 1993:201-209. 18. Hunt SM, McKenna SP, McEwen J, et al. The Nottingham Health Profile: subjective health and medical consultations. Soc Sc Med 1981; 15A:221-229. 19. Kaplan RM, Anderson JP, Ganiats T. The Quality of Wellbeing Scale: rationale for a single quality of life index. In: Walker S, Rosser M, eds: Quality of life assessment." key issues in the 1990s. Dordrecht, Netherlands: KluwerAcademic Press, 1993:65ff.
654 20. McHorney CA, Ware JE, Raczek AE. The MOS 36-item short-form health status survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 1993;31:247-263. 21. Hunter M. The Women's Health Q.uestionnaire (WHO.J: a measure of mid-aged women's perceptions of their emotional and physical health. Psychol Health 1992;7:45-54. 22. Beck AT, Ward CH, Mendelson M, et al. An inventory for measuring depression.Arch Gen Psychiatry 1962;4:561-574. 23. Wiklund I, Dimen~is E, Wahl M. Factors of importance when evaluating quality of life in clinical trials. ControlClin Trials 1990;11:169-179. 24. Acquadro C, Berzon R, Dubois D, et al. Incorporating the patient's perspective into drug development and communication: an ad hoc task force report of the patient-reported outcomes (PRO) harmonization group meeting at the Food and Drug Administration, February 16, 2001. ValHealth 2003;5:522-531. 25. Greene JG. A factor analytic study of climacteric symptoms.JPsychosom Res 1976;20:425-430. 26. Neugarten BL, Kraines RJ. Menopausal symptoms in women of various ages. Psychom Med 1965;27:266-273. 27. Greene JG. Constructing a standard climacteric scale. Maturitas 1998;29:25-31. 28. Ulrich LG, Barlow DH, Sturdee DW, et al., for the UK continuous combined HRT study investigators. Q.ualityof life and patient preference for sequential versus continuous combined HRT: the UK Kliofem multicenter study experience, lntJ GynaecolObstet 1997;59(suppl 1):11-17. 29. Z611nerYF, Acquadro C, Schaefer M. Literature review of instruments to assess health-related quality of life during and after menopause. Qual Life Res 2005;14:309-327. 30. Z6llner Y, Piercy J, Alt J. Mental health aspects of peri- and postmenopausal women, attitudes, quality of life, and the role of HRT (poster). Arch Women Mental Health 2001;3(suppl 2):68. 31. Z6llner Y, Kay S, Abetz L, et al. La qualit6 de vie sexuelle des europtennes. Gyn Info 2001;51:9-11. 32. Piercy J, Z6llner Y, Kay S, et al. Q.uality of life in postmenopausal women in five European countries (poster). ValHealth 2001;4:168. 33. Karlberg J, Mattsson LA, Wiklund I. A quality of life perspective on who benefits from estradiol replacement therapy. Acta Obstet Gynecol Scand 1995;74:367-372. 34. Le Floch JP, Colau JCI, Zartarian M. Validation d'une mtthode d'dvaluation de la qualit6 de vie en mtnopause. Refs en Gyn~colObst~tr 1994;2:179-188. 35. Le Floch JP, Colau JCI, Zartarian M, Gelas B. Rtduction d'un questionnaire d'tvaluation de la qualit6 de vie en mtnopause. Contracept Fertil Sex 1996; 24:238-245. 36. Le Floch JP, Chevalier T, Gelas B, et al. Q.uality of life improvement and hormonal replacement therapy: comparison of sequential versus continuous combined schedules with 1713-estradiol percutaneous gel and nomegestrol acetate. Menopause Rev 1999;4:87-96. 37. Hilditch JR, Lewis J, Peter A, et al. A menopause-specific quality of life questionnaire: development and psychometric properties. Maturitas 1996;24:161-175. 38. Hilditch JR, Lewis JE, Ross AH, et al. A comparison of the effects of oral conjugated equine estrogen and transdermal estradiol-17[3 combined with an oral progestin on the quality of life in postmenopausal women. Maturitas 1996;24:177-184. 39. Perz JM. Development of the menopause symptom fist: a factor analytic study of menopause associated symptoms. WomensHealth 1997;25:53-69. 40. Hogervorst E, Yaffe K, Richards M, et al. Hormone replacement therapy for cognitive function in postmenopausal women. Cochrane Database Syst Rev 2002;CD003122. 41. Shumaker SA, Legault C, Rapp SR, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651-2662.
HERMANN P. G. SCHNEIDER 42. Schneider HPG, Behre HM. Contemporary evaluation of climacteric complaints: its impact on quality of life. In: Schneider HPG, ed: Hormone replacementtherapy and quality of life. New Yorlc Parthenon, 2002:45-61. 43. Hauser GA, Huber IC, Keller PJ, et al. Evaluation der klimakterischen Beschwerden (Menopause Rating Scale [MRS]). Zentralbl Gynakol 1994;116:16-23. 44. Schneider HPG, Hauser GA. The Menopause Rating Scale (MRS II) m clusters of menopausal symptoms.Maturitas 1996;27(suppl. 1):201. 45. Potthoff P, Heinemann LAJ, Schneider HPG, et al. MenopauseRating-Skala (MRS II): Methodische Standardisierung in der deutschen Bevoelkerung. Zentralbl Gynako12000;122:280-286. 46. Schneider HPG, Heinemann LAJ, Rosemeier HP, et al. The Menopause Rating Scale (MRS): reliability of scores of menopausal complaints. Climacteric2000;3:59- 64. 47. Schneider HPG, Heinemann LAJ, Rosemeier HP, et al. The Menopause Rating Scale (MRS): comparison with Kupperman index and quality-of-life scale SF-36. Climacteric2000;3:50- 58. 48. Greene JG. Measuring the symptoms dimension of quality of life: general and menopause specific scales and their subscale structure. In: Schneider HPG, ed. Hormone replacement therapy and quality of life. Carnforth, NY: Parthenon 2002:35-45. 49. Schneider HPG, Schultz-Zehden B, Rosemeier HP, et al. Assessing well-being in menopausal women. In: Studd J, ed. The management of the menopausemthe millennium review 2000. New York: Parthenon Publishing, 2000:11 - 19. 50. Wiklund I. Methods of assessing the impact of climacteric complaints on quality of life. Maturitas 1998;29:41-50. 51. Schneider HPG, Heinemann LAJ, Thiele K. The Menopause Rating Scale (MRS): cultural and linguistic validation into English. Life Med Sc Online 2002;3: DOI:10.1072/LO0305326. 52. Heinemann LAJ, PotthoffP, Schneider HPG. International versions of the Menopause Rating Scale (MRS). Health Qual Life Outcomes 2003;1:28. 53. Heinemann LAJ, Schneider HPG. Q.uality of life assessment in the menopause. In: Eskin B, ed. The menopause." endocrinologic basis and management options, 5 ed. Oxon: Informa Healthcare, 2007:79-85. 54. Jacobs P, Hyland ME, Ley A. Self rated menopausal status and quality of life in women aged 40- 63 years. BrJ Health Psych2000;5:395- 411. 55. Utian WH. The mental tonic effect of oestrogens administered to oophorectomised females. S Afr MedJ 1972;46:1079-1082. 56. Schultz-Zehden B. FrauenGesundheit in und nach den Wechseljahren.Die 1000 Frauenstudie. Gladenbach: Verlag Kempkes, 1998. 57. Heinemann LAJ, DoMinh T, Strelow F, et al. The Menopause Rating Scale (MRS) as outcome measure for hormone treatment? A validation study. Health Qual Life Outcomes2004;2:67. 58. Hays J, Ockene JK, Brunner RL, et al. Effects of estrogen plus progestin on health-related quality of life. NEnglJMed 2002;348:1839-1854. 59. Limpaphayom KK, Darmasetiawan MS, Hussain RI, et al. Differential prevalence of quality of life categories (domains) in Asian women and changes after therapy with three doses of conjugated estrogens/ medroxyprogesterone acetate: the Pan-Asia Menopause (PAM) study. Climacteric 2006;9:204- 214. 60. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4 ed. Washington, DC: APA, 1994. 61. Derogatis L, Rust J, Golombok S, et al. Validation of the Profile of Female Sexual Function (PFSF) in surgically and naturally menopausal women. J Sex Marital Ther 2004;30:25-36. 62. Strothmann A, Schneider HPG. Hormone replacement therapy: the European women's perspective. Climacteric2003;6:337- 346. 63. Genazzani AR, Schneider HPG, Nijland E. The European Menopause Survey 2005. What do women think right now about menopause and HRT? Climacteric2005;8 (suppl 2):96.
( ; H A P T E R 4~
Role of Exercise and Nutrition MICHELLE P. WARREN
Departmentof Obstetrics/Gynecology,Columbia University,New York, NY 10032
CECILIA ARTACHO
Departmentof Obstetrics/Gynecology,Columbia University,New York, NY 10032
ALLISON R.
HAGEY Departmentof Obstetrics/Gynecology,Columbia University,New York, NY 10032
I. ROLE OF EXERCISE AND NUTRITION The effects of menopause and the aging process itself cause many physiologic changes, which explain the increased prevalence of chronic diseases observed in postmenopausal women. Exercise and nutrition play important roles in the prevention and treatment of cardiovascular disease, cancer, obesity, diabetes, osteoporosis, and depression, which are some of the major health problems seen in postmenopausal women. Clinicians caring for older women should stress the relevance of these two lifestyle factors to overall health and advise their patients about adequate exercise prescriptions and nutrition.
II. CARDIOVASCULAR DISEASE A N D T H E ROLE OF N U T R I T I O N Cardiovascular disease (CVD) incidence increases with advancing age in women, with a notable increase after menopause, possibly as a result of estrogen deficiency. Coronary heart disease (CHD) is the leading cause of death among postmenopausal women (1,2). It is believed that the T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
655
atherosclerotic process is attenuated in women until perimenopause because circulating estrogen prevents the incorporation of low-density lipoprotein cholesterol (LDLc) into atherosclerotic plaques (3). By the year 2015, one-half of all women in the United States will be more than 45 years old (3), and primary prevention of CVD in this group is becoming increasingly important. Smoking, hypertension, hypercholesterolemia, obesity, diabetes mellitus, a sedentary lifestyle, and estrogen deficiency all increase the risk of atherosclerotic plaque formation (4,5). Changes that take place during menopause, including atherogenic changes in serum cholesterol profiles and weight gain, contribute to the increased risk of CHD after menopause (7). CVD causes other diseases, besides heart disease and stroke, that result in significant morbidity in women. As many as 25% of women ages 55 to 74 years suffer from lower extremity atherosclerosis (8). Although these CVD and CHD risk factors become more pronounced during menopause, they, in turn, are affected by nutrition and physical activity. Cardiovascular disease risk factors, such as lipid changes, become more pronounced during menopause. In a large cohort of premenopausal and postmenopausal French women, menopause was associated with higher levels of serum cholesterol, triglycerides, apolipoprotein (apo) B, and Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
656 apo A-I, and also with elevated diastolic blood pressure (9). Lipoprotein profiles change with age, and LDLc increases in women after age 50 (10). Postmenopausal women not only have higher total LDLc plasma levels, but also the L D L particle itself becomes more dense (11). This smaller, denser form of L D L is linked to C H D risk (12). The highdensity lipoprotein cholesterol (HDLc) plasma level declines with the onset of menopause, and the H D L particle itself is altered. The relative proportions of its two subfractions change, and the more dense H D L 3 C increases and the less dense H D L 2 C decreases (13). The HDL3b subclass is usually present in individuals with C H D (14). Serum ferritin, total cholesterol, and L D L cholesterol all increase in postmenopausal women, and their parallel rise may be partly responsible for the increased risk of C H D in this population (15). The Framingham Offspring/Spouse Study, which examined the relationship between diet and plasma total and L D L cholesterol levels in a large sample set of premenopausal and postmenopausal women, showed that cholesterol levels were directly related to consumption of saturated fat and inversely related to total caloric intake, but that dietary cholesterol was not a predictor of cholesterol levels (16). Another study of the same cohort showed that plasma triglycerides were inversely related to protein, fiber, and polyunsaturated fat and directly related to saturated fat and oleic acid (17). High triglyceride levels are better predictors of C H D risk in women than in men, and triglycerides are strongly related to HDLc, which is an important predictor of C H D risk in women (18,19). High triglyceride levels are an important risk factor for C H D in women, but the effect of high triglycerides is most pronounced when they are present in conjunction with low H D L c levels. Patients who have both high triglyceride levels and low HDLc levels have an increased incidence of CVD (20). A low HDLc appears to be the strongest predictor of C H D in women (20,21). An association of H D L c with cardiovascular disease in women was also found in the 20-year follow-up of the Donolo-Tel Aviv cohort (21). The importance of HDLc and triglycerides as risk factors for CVD in women was also apparent in the Lipid Research Clinics Follow-up Study (20), which found that H D L c levels of less than 1.3 mmol/L (50 mg/ dL) and triglyceride levels of 2.25 to 4.49 mmol/L (200-399 mg/dL) were independent predictors of death from CVD. A low H D L c level places women at a greater risk for C H D than a high LDLc level (20). The Framingham, Donolo-Tel Aviv, and Lipid Research Clinics studies all indicate that low H D L c levels in women are stronger predictors of cardiovascular disease mortality than total cholesterol levels (20-22). C H D in women is also more strongly related to the total cholesterol H D L c ratio than to either total cholesterol or to LDLc (23). An age-related increase exists in the ratio of total to HDLc, and when the ratio is greater than 7.5, women have the same C H D risk as men (24).
WARREN ET AL.
The entire lipid profile should be assessed yearly, and dietary and exercise interventions should be prescribed to women who are not within the range of optimal cholesterol and lipoprotein profile (Table 49.1). Only one-half of the women in the Framingham study met the 1996 National Cholesterol Education Program (NCEP) guidelines for a desirable blood level of LDLc (130 mg/dL) (5,25). In 2001 the N C E P updated these cholesterol designations to more rigorous standards and now suggests that people at low or
TABLE 49.1 Adult Treatment Plan (ATP) III Classification of LDL, Total, and H D L Cholesterol (mg/dL) L D L cholesterol < 100 100-129 130-159 160-189 -> 190
Optimal Near optimal/above optimal Borderline high High Very high
Total cholesterol < 200 200-239 >- 240
Desirable Borderline high High
H D L cholesterol < 40 -> 60
Low High
Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol and Adults (AdultTreatment Plan III). Executive summary.AccessedMay 2001, from http://www.nhlbi.nih.gov/guidelines/ cholesterol/index.htm.
TABLE 49.1A Major Risk Factors (Exclusive of L D L Cholesterol) that Modify L D L Goals a Cigarette smoking Hypertension (blood pressure -> 140/90 mm Hg or an antihypertensive medication) Low HDL cholesterol (< 40 mg/dL) b Family history of premature CHD (CHD in male first-degree relative < 55 years; CHD in female first-degree relative < 65 years) Age (men -> 45 years; women -> 55 years)a aln ATP Ill, diabetes is regarded as a CHD risk equivalent. bHDL cholesterol -> 60 mg/dL counts as a "negative"risk factor; its presence removesone risk factor from the total count. From Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol and Adults (AdultTreatment Plan III). Executive summary.AccessedMay 2001, from http://www.nhlbi.nih.gov/guidelines/ cholesterol/index.htm.
CHAPTER 49 Role of Exercise and Nutrition TABLE 49.1B
657
Three Categories of Risk that Modify L D L Cholesterol Goals LDL level at which to initiate therapeutic lifestyle changes (TLC)
Risk category
LDL goal (mg/dL)
CHD and CHD risk equivalents Multiple (2 +) risk factors a
< 100 mg/dL
-> 100 mg/dL
< 130 mg/dL
-> 130 mg/dL
< 160 mg/dL
-> 160 mg/dL
Zero to one risk factor
LDL level at which to consider drug therapy -> 130 mg/dL (100-129 mg/dL: drug optional) b 10-year risk 10-20%: -> 130 mg/dL 10-year risk < 10%: -> 160 mg/dL -> 190 mg/dL (160-189 mg/dL: LDL-lowering drug optional)
~Risk factors that modify the LDL goal are listed in Table 49.1a. bSome authorities recommend use of LDL-lowering drugs in this categoryif an LDL cholesterol < 100 mg/dL cannot be achievedby therapeutic lifestyle changes. Others prefer use of drugs that primarily modify triglycerides and HDL, such as nicotinic acid or fibrate. Clinicaljudgment also may call for deferring drug therapy in this subcategory. From Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol and Adults (Adult Treatment Plan III). Executive summary.Accessed May 2001, from http:llwww.nhlbi.nih.gov/guidelineslcholesterol/ index.htm.
moderate risk of coronary heart disease maintain LDLc levels less than 100 mg/dL (26). Although women in the Framingham cohort were adhering to a moderate cholesterol intake of no more than 300 mg/day, their average fat intake constituted 38% of calories (22). Only one-fifth of the women adhered to the recommended upper limit for fat intake of no more than 30% of calories (22). In light of these findings, it is important for postmenopausal women to receive adequate counseling about ways to reduce their saturated fat, total fat, oleic acid from animal sources, and total caloric intake and to increase their intake of fiber and polyunsaturated fat in order to improve their lipoprotein profile and reduce their risk for CVD. Dietary change resulting in lowered blood cholesterol correlates with a reduction in C H D rates among women in the United States (3). For each 1% decrease in serum cholesterol, a corresponding 2% to 3% reduction is seen in CVD risk in U.S. adults, (27) and dietary intervention can lower cholesterol concentrations by approximately 10% (28). One of the important steps in the primary prevention of C H D should be a nutritional intervention aimed at the many women who still consume diets that place them at increased risk for CHD. Women with hypercholesterolemia clearly benefit from a reduction in dietary fat and cholesterol (3). Healthy women with only slightly elevated cholesterol levels may or may not benefit from a dietary intervention because such a lipid-lowering intervention can also reduce H D L c (3). Because a low H D L c level places women at increased risk of C H D , dietary interventions should be geared toward raising H D L and lowering L D L (3). The American Heart Association (AHA) recommends that total fat intake be
reduced to 30% of calories, that saturated fat intake be decreased below 10% of calories, and that dietary cholesterol be lowered to less than 300 mg/day for healthy people and less than 200 mg/day for at-risk individuals. Other L D L lowering enhancers include plant stanols/sterols (2 g/day), increased viscous (soluble) fiber (10-25 g/day), protein approximating 15% of total calories, and moderate physical activity (29). This A H A Phase I diet (30) should be followed by all nonobese, normocholesterolemic women (31). Women who have a higher CVD risk profile should follow the A H A Phase II or the A H A Phase II1 diets. Both the A H A step I diet and the A H A step II diet have proven to be equally effective in reducing total serum cholesterol and LDLc in men and women (Fig. 49.1) (32). Reducing fat intake may have additional protective effects. Data from the Iowa Women's health study also indicated that a high fat intake is associated with decreased survival of postmenopausal women with breast cancer (33). A 5-year randomized trial designed to determine whether increases in LDLc and body weight during menopause can be prevented through changes in diet and physical activity showed that a dietary and behavioral intervention was successful during the first 6 months in a cohort of premenopausal women (7). The risk of CAD can be reduced through dietary intervention, which can improve the lipid profile and lower blood pressure (34). The A H A encourages reaching an optimal blood pressure of less than 120/80 mm H G through diet and exercise modifications. For individuals with blood pressure ranges greater than 140/90 mm H G with related target-organ damage or diabetes, the A H A recommends pharmacology such as thiazide diuretics (25). Diet and other nonpharmacologic interventions should,
658 0
WARREN ~.v aT..
Per 9
-2 -4 -6 -8 -10
-12 -14 -16
I
1
Low
Mid Men
~
High Women
FIGURE 49.1 Low-density lipoprotein cholesterol responses in women and men following the American Heart Association step I diet for an 8-week period. Subjects are grouped by tertile based on initial serum cholesterol level. Women had the following serum cholesterol responses: highest titer, initial values 295 (7.6 mmol/L) _+ 6 mg/dL,-12.8% (p < 0.0001); middle titer, initial values 253 (6.5 mmol/L) _+ 2 mg/dL,-9.0% (p < 0.0001); lowest titer, initial values 224 (5.8 mmol/L) _+ 3 mg/dL,-1.9% (p = 0.34). (From ref. 32, with permission.)
however, be the first step in management of both hypercholesterolemia and hypertension (35). Lower caloric intake, decreased total, saturated fat, and cholesterol intake, weight loss, sodium restriction, and abstinence from alcohol are effective dietary interventions used to lower blood pressure (34,35). In terms of dietary profiles that can affect blood pressure, increased potassium consumption has been shown to significantly lower it (36), and hypertensive patients may benefit from potassium supplements. Complex carbohydrates and soluble fiber have also been shown to lower blood pressure (37). Thus, women should actively attempt to improve unfavorable lipid profiles by restricting caloric intake, losing excess weight, consuming the recommended quantities and types oflipids, consuming sufficient complex carbohydrates and soluble fiber, and exercising. Evidence shows that 100 mg of aspirin taken every other day has little or no effect on all-cause mortality, but it does reduce the risk for stroke in healthy middle-aged and older women (38). Evidence of aspirin's effect on heart disease remains unclear (38). Although aspirin use may benefit women older than 65 with a higher baseline risk for cardiovascular disease, the A H A suggests that routine use of aspirin in lower-risk women under the age of 65 be delayed pending the results of ongoing trials (38,39). Aspirin recommendations are currently challenging because most data from primary prevention trials lack female patients and data on men may not necessarily be extrapolated to women. Further, aspirin therapy may increase the risk of hemorrhagic stroke in women with uncontrolled hypertension and may also cause gastrointestinal bleeding. Nonsteroidal anti-inflammatory medications should not be substituted for aspirin for CVD prevention. ~3-blockers are recommended for women who have
had a myocardial infarction or who have chronic ischemic syndromes, and angiotensin-converting enzyme (ACE) inhibitors should be prescribed for high-risk women. Angiotensin II receptor blockers (ARBs) should be used for high-risk women with clinical evidence of heart failure or an ejection fraction, of which 40% are intolerant to ACE inhibitors (25). Oxidative stress, which is defined as an imbalance between prooxidative factors and the antioxidative defense mechanisms, has been linked with the development of early stages of arteriosclerosis and cancer (40-43). Epidemiologic studies have shown that diets rich in antioxidants from fruit, vegetable, and vegetable oil sources reduce the relative risk of premature death from CVD and cancer (44). It is believed that an optimal antioxidant status is necessary for optimal health. Threshold antioxidant plasma levels associated with minimal relative risk of premature death by CVD and cancermin other words, desirable optimal levels--can tentatively be established from available consistent data (41-43,45). At these optimal antioxidant plasma levels, relative risks seem to disappear (44). Most middle-aged persons can obtain optimal plasma levels by consuming antioxidants at levels close to or only slightly above the current recommended daily allowances (RDAs) (46): 75 mg vitamin C in nonsmokers, 125 to 130 mg in smokers; 2 to 3 mg ~3-carotene in nonsmokers, and 15 mg or more for females over the age of 50 (44,47). The risk for CVD and cancer is twice as high at levels 25% to 35% below these levels. Suboptimal levels of a single antioxidant may increase relative risk, and suboptimal levels of several antioxidants further increase relative risk (44). The First U.S. National Health and Nutrition Examination Survey showed that habitual consumption of a vitamin C-containing multivitamin by the general middle-aged U.S. population reduced the mortality rate from CVD by approximately 42% in men and by about 25% in women (48). The effects of vitamin C consumed as a multivitamin did not have a significant effect in terms of lower cancer mortality (48). The U.S. Health Professionals Study and the U.S. Nurses' Health Study showed a protective effect against coronary risk for f3-carotene supplements in smokers but not in nonsmokers, probably because nonsmokers had above optimal levels of ~3-carotene even without the use of supplements, but smokers had low levels and they have an increased requirement (44). A low intake of vitamin A may increase the relative risk of breast cancer; thus, any benefit of vitamin A supplements may be limited to women with diets already low in vitamin A (49). Randomized antioxidant intervention trials conducted in China and Finland in middle-aged and elderly subjects over a 5- to 6-year period were successful in preventing the earlier stages of CV-D and cancer by rectifying previously poor antioxidant levels (50-52). However, antioxidant supplementation did not have an effect on irreversible precancerous lesions, clinically established common cancers, and vascular lesions in chronic smokers (44). It is therefore important for
CHAPTER 49 Role of Exercise and Nutrition women to start consuming a diet rich in antioxidants as a preventive measure before menopause, when the risk of CVD and cancer begins to escalate with advancing age. Some studies report no association between intake of antioxidant vitamins and relative risk of breast cancer (53), but because of the protective effects of some antioxidants on CVD and possible protective effects against breast cancer and other types of cancers, increasing dietary antioxidant intake seems like a prudent preventive measure. Studies have shown that antioxidant supplementation aimed at preventing or correcting previously poor levels may protect from both CVD and cancer. No evidence indicates that megadoses of antioxidants have an additional protective effect. Practitioners should discourage self-prescribed overconsumption of antioxidant supplements, especially because overdoses of the fat-soluble vitamin A can be extremely toxic. Antioxidants are most effective when overall nutritional status is optimized. The AHA guidelines recently stated that antioxidant vitamin supplements should not be used to prevent CVD pending the results of ongoing trials. In fact, the AHA recognizes that while the use of antioxidant supplements may have some benefits, currently there is evidence that certain levels may be harmful. Additional large-scale randomized trials of antioxidants in the primary and secondary prevention of CVD should be conducted to prove their effectiveness in reducing CVD risk before advising patients to take antioxidant supplements (25,54). Oxidative modification of LDLc may accelerate atherosclerotic plaque formation, and dietary antioxidant vitamins may play an important role in preventing CHD. However, intake of the antioxidant vitamins, vitamins A and C, was not correlated with decreased risk of death from C H D (25). Antioxidant vitamins exert an antioxidant effect on LDL and may also preserve the endogenous antioxidants of LDL (55). Preliminary data indicate that H R T can preserve the LDL particle content of two antioxidants, c~-tocopherol and [3-carotene, and keep the LDL in a reduced antioxidant state (55). However, the A H A states that estrogen plus progestin hormone therapy should not be initiated or continued to prevent CVD in postmenopausal women, and other forms of H R T such as unopposed estrogen, should not be initiated or continued pending ongoing trials. Overall, the A H A recommends a conservative approach in HRT use but suggests health care providers weigh the risks of therapy against the potential benefits for menopausal symptom control (25). Nutrition can play a role, not only in the primary prevention of CVD through the effects of the antioxidant vitamins, but also in the treatment of premature arteriosclerotic disease. Mild hyperhomocysteinemia, which is often present in patients with premature arteriosclerotic disease, can be corrected by treatment with compounds involved in folic acid metabolism, such as vitamin B6, folic acid, and betaine (25,56). The U.S. Food and Nutrition Board recommends
659 that women over the age of 50 consume 1.7 mg of vitamin B6 and 400 Ixg of folate (47). The A H A has concluded that the intake of omega-3 fatty acids in the form of fish has been associated with a reduced risk of CVD. However, the A H A encourages women of childbearing age and pregnant women to avoid shark, swordfish, king mackerel, and tilefish because the high mercury levels in these fish impair fetal neurologic development (25). Increased fruit and vegetable consumption lowers the risk of ischemic heart disease by almost 15% for those in the 90th percentile of fruit consumption compared with those in the 10th percentile of fruit consumption. Although the mechanism responsible for the protective effect on heart disease of fruit and vegetables is not well understood, the positive effect is commensurate with the estimated protective effects of the potassium and folate in these fruits and vegetables (57). The 2005 U.S. Department of Health and Human Services and the U.S. Department of Agriculture 2005 Dietary Guidelines recommend that people over the age of 50 meet their recommended dietary allowance (RDA) for vitamin B12 (2.4 Ixg/day) by eating foods fortified with vitamin 812 such as fortified cereals, or by taking the crystalline form of vitamin B12 supplements (58). Also of recent interest are studies indicating that moderate alcohol consumption may lower the risk factors for cardiovascular disease in postmenopausal women on a controlled diet (59). However, further studies should be conducted before moderate alcohol consumption is recommended to women. Because CVD is one of the main health problems of postmenopausal women, adequate dietary habits should be initiated as early as childhood or adolescence and maintained throughout life, especially during menopause and in the postmenopausal years.
III. CARDIOVASCULAR DISEASE: THE ROLE OF EXERCISE In recent years many studies have shown that increased physical activity plays an independent role in the primary prevention of CHD, the most serious and common form of CVD. Physically active individuals are at lower risk for C H D than those who have a sedentary lifestyle. Physical inactivity is a high risk factor for CVD, but it can also exert its effects by contributing to the physiologic changes associated with atherogenesis, such as hypertension, obesity, diabetes, and hypercholesterolemia. Exercise favorably affects plasma lipids and lipoproteins and thus may have protective effects against CVD and C H D mortality (60,61). Exercise increases the levels of the less dense HDL2 cholesterol subfraction, which normally decreases during menopause, by causing an increase in lipoprotein lipase, the enzyme that catabolizes triglyceride-rich lipoproteins (31). Higher concentrations of
660 lipoprotein lipase are found in the slow-twitch skeletal muscle fibers of endurance athletes (31). Exercise also lowers triglyceride levels. Population studies such as the Framingham Study have shown that age, total cholesterol, H D L cholesterol, systolic blood pressure, treatment for hypertension, cigarette smoking, and triglyceride levels (25,62,63) are the best predictors of C H D in women, and dietary or exercise interventions that can alter these lipids can decrease C H D risk. Thirty of forty studies conducted between 1985 and 1995 suggest that exercise can reduce cardiovascular risk between 10% and 50% (64). However, research on C H D risk and exercise has been conducted mostly in men. Few studies have examined the effects of exercise on CHD in women, and data from available studies are often inconsistent, possibly because the number of individuals in the high activity category is too small. Most cross-sectional studies examining the effects of exercise in women show a positive correlation between exercise and HDLc in both premenopausal and postmenopausal women (62). Longitudinal studies on the effects of exercise in women have given inconsistent results because of methodologic problems; nevertheless, about half the studies showed that exercise increases HDLc (62). HDLc can be expected to increase by approximately 10% in both men and women as a result of training (62). Large-scale prospective studies have consistently shown exercise to be effective in reducing heart disease (65-67). A meta-analysis of 27 cohort studies comparing sedentary and physically active individuals indicates an association between lack of physical actMty and increased risk of CHD (68). This study also found that the association is stronger when a high activity group is compared with a sedentary group, rather than when the active group only has a moderate activity level, indicating that a dose response relationship exists between physical activity and protection from C H D (68). Exercise was found to have a protective effect in terms of preventing major cardiovascular events in this study, but not in terms of reducing the severity of such events (68). Cardiovascular disease risk occurs less in women who exercise than it does in nonexercisers (69). One mechanism that may account for the decreased risk seen in active women is that these women may have more favorable blood pressure-related risk factors than do sedentary women (69). Physically active postmenopausal women in their mid-50s have more favorable systolic blood pressure-related CVD risk factors than do sedentary healthy women in their late 50s (69). The observed decrease in risk may result from the lower levels of abdominal adiposity found in the more active women. The study group was small (active women n = 18; less-active control subjects n = 34). Prospective studies need to be conducted with a larger group to establish more definitively the effects of exercise on blood pressure. In light of these findings, however, it would be prudent to advise patients to increase their level of physical activity because of its potential beneficial effects on blood pressure and consequent
WARREN ET AL.
reduction of CVD risk. Exercise therapy may be particularly appropriate for postmenopausal women, in whom LDLc levels rise sharply with increasing age, because antihypertensive medication, thiazide diuretics, and [3-adrenergic blockers all tend to increase blood cholesterol levels (3). Before prescribing antihypertensive medication, a patient's overall C H D risk profile should be evaluated and, in some patients, exercise and weight loss may be a safer alternative. Exercise improves cardiovascular fitness, which has been linked to decreased mortality in prospective studies. Longitudinal studies on cardiovascular disease indicate threshold levels of approximately 20 minutes per day of moderate physical activity or 2 to 3 hours per week (70). One study of 3120 women found that those with the lowest fitness level, as measured by maximal treadmill test, had a relative risk of death almost fivefold greater than those with the highest fitness level (60). Being in the lowest fitness level was associated with greater risk of death but slightly lower than the risk from having an elevated systolic blood pressure (> 140 mm Hg) or an elevated serum glucose (6.7 mmol/L). The risk associated with cigarette smoking was approximately equal to the risk associated with being in the lowest fitness category (60). A study in an Australian cohort showed that previously sedentary women who improved their fitness level over a 4-year period had significantly improved blood lipid profiles and systolic blood pressure (61). Similar benefits were not reported for the men in this study. Exercise appears to have many protective effects that can improve the C H D risk profile. In a cross-sectional study of premenopausal and postmenopausal runners and joggers, HDLc was higher in the women who exercised than in the inactive control subjects (71). The rise in HDLc was the same for premenopausal and postmenopausal women exercisers. In premenopausal women, exercise did not affect the HDLc/LDLc ratio, but in postmenopausal women exercise had a favorable effect on the ratio, which increased significantly from 0.57 mg/dL in the inactive women to 0.85 mg/dL in runners (71). Postmenopausal women may exhibit a greater response to exercise (72). In a crosssectional study of premenopausal and postmenopausal trained runners, the postmenopausal exercise group had a higher HDLc (74 mg/dL vs. 56 mg/dL), a lower LDLc (141 mgldL vs. 185 mgldL), and a higher HDLc/LDLc ratio (0.57 mg/dL vs. 0.32 mg/dL) than the sedentary postmenopausal control group (73). In the premenopausal group, only LDL changed significantly with exercise. In these two studies, exercise blunted the age-dependent increase in LDLc and prevented the age-dependent decrease in HDLc experienced by menopausal women. Both premenopausal and postmenopausal women who exercise can improve their lipoprotein profile and thereby reduce their risk of developing CHD. In prospective studies, sedentary women had a greater risk of developing hypertension
661
CHAPTER 49 Role of Exercise and Nutrition independent of body weight (74). In a large study of perimenopausal women, those who increased their exercise participation over the course of 3 years gained less weight and had a smaller decrease in H D L 2 c (75). The activity levels in this study were moderate and seem to indicate that even slight increases in exercise at the time of menopause can help prevent the atherogenic changes in lipid profiles and the weight gain experienced by menopausal women. Thus, moderate exercise seems to lead to improved weight and blood pressure and to more favorable lipid profiles in both premenopausal and postmenopausal women. Controversy exists regarding the frequency and intensity of exercise that must be maintained to observe the beneficial rise in HDLc. Few studies have been conducted in postmenopausal women. At least 4 months of fairly strenuous activity, such as running 16 to 24 km per week, may be needed to obtain a significant increase in H D L c (31). More moderate activities, such as walking 48 km per week, require 3 months to observe a significant rise in H D L c (31). Walking can lower CVD risk because of its effects on cardiorespiratory function and on body fat. Walking 3 or 5 days per week increased peak volume of oxygen utilization (VO2
peak) and decreased body fat, but it did not alter serum lipids of nonobese, normolipidemic postmenopausal women (76). A case-control study of postmenopausal women found that the risk of myocardial infarction in this population is decreased by 50% with energy expenditures corresponding to 30 to 45 minutes of walking three times per week (77). In a large cross-sectional study of male runners participating in the National Runners Health Study, substantial increases in H D L c were observed in women who exercised at levels exceeding current guidelines (78). Official guidelines from the Centers for Disease Control and Prevention (CDC) state that most health benefits from physical activity can be achieved by walking 2 miles (3.2 km) briskly most days, which is the energy equivalent of running 8 to 12 km per week (79). However, the study conducted in runners demonstrated that women obtain additional health benefits from exercise at levels higher than currently recommended, because plasma H D L c concentrations increased for every additional kilometer run per week (Fig. 49.2). A 2-year randomized trial testing the effects of different intensities and formats of exercise on participation rates, fitness and plasma H D L c levels in older men and in
FIGURE 49.2 Plasma high-density lipoprotein (HDL) cholesterol concentrations according to weekly distance run. Menstrual periods were reported by 1390 women; the absence of periods was reported by 447. A total of 236 women reported using oral contraceptives and 176 reported using postmenopausal estrogen-replacementtherapy.The p-values shown in the figure are for the regression slope of HDL cholesterol plotted against distance run, with adjustment for age, education, progesterone use, and intake of red meat, fish, fruit, and vitamins C and E. (From ref. 78, with permission.)
662 postmenopausal women showed that moderate-intensity exercise improved cardiorespiratory fitness levels and improved HDLc levels (Fig. 49.3) (80). Two years were required to observe a change in HDLc. Frequency of participation played an important role in influencing HDLc in this age group (80). This finding differs from previous reports that exercise intensity is the determining factor affecting H D L c levels. In summary, most studies have shown that exercise causes an increase in HDLc, which may be related to the amount, frequency, and intensity of exercise, and that as many as 2 years may be required to produce this increase in HDLc. The effects of exercise, exercise intensi~, and hormone replacement therapy (HRT) on lipid and lipoprotein profiles are controversial. Some studies in women on HRT indicate that exercisers have higher HDLc levels than nonexercisers (62). In a cross-sectional study of postmenopausal women, regular exercise was associated with significantly greater HDLc concentrations, especially in women using exogenous estrogens (81). In the group of women who were not taking estrogen, the most significant increase in HDLc was observed for the sedentary versus the fight activity categories. Increasing activity levels from fight to moderate or heavy did not produce an incremental change in HDLc in this study (81). However, other investigators studying the effects of HRT and exercise on fipid metabolism had conflicting results. In a study assessing the independent effects of a moderate exercise program, with and without oral estrogen replacement, on lipids and lipoproteins in a group of postmenopausal women, estrogen therapy alone had the greatest beneficial effect on the lipid and lipoprotein profile (82). Exercise alone also favorably altered lipid and lipoprotein levels, resulting in a significant reduction in
FmORE 49.3 Bar graph shows mean change (with standard error bars) in high-density lipoprotein cholesterol based on the average number of exercise sessions per week completed across 2 years by exercise training condition (higher-intensity, group-based exercise; higher-intensity, homebased exercise; and lower-intensity, home-based exercise). (Adapted from ref. 80, with permission.)
WARREN ET AL.
cholesterol, triglycerides, and LDLc and in an increase in the H D L / L D L ratio. The combination of oral estrogen and exercise did not produce additional improvements in lipid metabolism (82). In a controlled, prospective 2-month clinical trial in postmenopausal women who participated in a 2-month, lowintensity exercise regimen followed by a 9-month period of high-intensity exercise for 45 minutes per day, three or more days per week, those in the exercise group had lower total cholesterol and LDLc levels, but HDLc and triglycerides were unaffected by the exercise regimen (83). Those in the exercise plus HRT group had decreased total cholesterol and LDLc and increased HDLc levels. Exercise was found to be protective against the HRT-related increase in triglycerides observed in the HRT group (83). However, recent studies examining H R T therapy on postmenopausal women indicate further risks from hormones. In March 2004 the National Institutes of Health (NIH) instructed participants in the Women's Health Initiative (WHI) to stop taking the estrogen study pills because of an increased stroke risk (84). The W H I study was established to assess the effects of long-term use of hormone therapy in healthy postmenopausal women on the prevention of heart disease and hip fractures as well as breast cancer. In early 2004, the NIH concluded that estrogen alone does not affect heart disease or b r e a s t cancer and that estrogen alone appears to increase the risk of stroke and decrease the risk of hip fracture. The increased risk of stroke in the estrogen-alone study is similar to the findings in the estrogen plus progestin study, which was discontinued in July 2002. The NIH follows the FDA guidelines, which recommend that postmenopausal women considering using estrogen or estrogen with progestin discuss the risks and benefits with their physicians. Estrogen and estrogen with progestin are still approved therapies for relief from moderate to severe hot flashes and symptoms of vulvar and vaginal atrophy (84). Hormone therapy is also approved for treatment of osteoporosis; however, the NIH advises that hormone therapy should only be administered to women at significant risk of osteoporosis who cannot take nonestrogen medications (84). The W H I study found a 24% reduction in all fractures and a 33% reduction in hip fractures in women assigned to estrogen plus progestin (85). In addition, the estrogen plus progestin trial was discontinued after 5.6 years of follow-up because the increased risk of breast cancer, coronary heart disease, stroke, and blood clots outweighed the benefits on hip fracture and colorectal cancer (84). The increased risk of breast cancer due to estrogen plus progestin was 8 additional cases of breast cancer for every 10,000 women over 1 year, and there was a 24% overall increase in breast cancer due to the estrogen plus progestin therapy (86). The breast cancer tumors in the estrogen plus progestin group tended to be larger and with more local spread as compared with women taking a placebo (86). Even 1 year after the study was discontinued, quite a few women
663
CHAPTER 49 Role of Exercise and Nutrition had abnormal mammograms in the estrogen plus progestin group (9.4%) as compared with the placebo group (5.4%) (86). The relative risk of coronary heart disease was 1.24 (6 more heart attacks annually per 10,000 women using estrogen plus progestin) with most of the events occurring in the first year (87). The relative risk of stroke was 1.31 (8 more cases per 10,000 per year) (88). Most of the strokes were ischemic rather than hemorrhagic (88). The estrogen studied in the estrogen alone trial was conjugated equine estrogens (CEE; Premarin) at a dose of 0.625 mg/day. There was no effect on heart disease or breast cancer; however, stroke risk was increased by 12 excess cases per 10,000 women per year and tended to increase deep vein thrombosis by 6 per 10,000 (89). The latter was much lower than the estrogen plus progestin arm of the study, which reported an excess incidence of 18 per 10,000. The study did find that CEE had no effect on pulmonary embolisms and decreased hip fractures at a rate of 6 fewer cases per 10,000 women per year and total bone fractures by 30%. These numbers are similar to those found by the W H I combined estrogen-progestin study (89). Although HRT remains controversial, all women should be encouraged to engage in physical activity starting at a young age so that they can enter menopause with improved CVD risk factors. Exercise is especially important during menopause and in the postmenopausal years as a means of attenuating the development of CVD risk factors that become more pronounced at this time. Evidence suggests that exercise, initiated sufficiently early in a woman's life, also protects against breast cancer, cancers of the reproductive system, non-reproductive system cancers, diabetes, and obesity (90-94). In fact, one study examining 74,171 W H I participants found that physical activity plays a protective role in women who engage in strenuous activity at the ages of 35 and 50 years, with the greatest associations observed for the lightest-weight women. Moderately overweight women, however, did also benefit from increased total physical activity (95). The study found that increasing physical activity (5.1-10.0 metabolic equivalent [MET] hours per week, which is equivalent to 1.25-2.5 hours per week of brisk walking) reduces the risk of breast cancer by 18%. And increasing the intensity of the physical activity more than 40 M E T hours per week further reduces the risk of breast cancer by 22% as compared with sedentary women (95). Women who engaged in strenuous exercise at least three or more times per week at the age of 35 had a 14% reduction in breast cancer risk compared with women who did not exercise at this level (95). Likewise, women over the age of 50 who engaged in strenuous exercise at least three or more times per week had a slight, though not statistically significant reduction in risk in breast cancer (95). The study also found that physical activity reduces risk among women who are using hormone therapy; a group that is at increased risk for developing breast cancer. Physical activity also helps women maintain a health body mass index, which is a risk factor for breast cancer. A
study of 85,917 W H I participants found that generalized obesity is an important risk factor for postmenopausal breast cancer, but only among women who have never taken HRT (96). For women of average size or less who exercised regularly and are taking or have taken HRT, there is no association between anthropometrics and breast cancer risk (96). Women above average size (body mass index [BMI] >-- 28.4) did not increase their risk of breast cancer with HRT use; however, exercise did not decrease their risk (96). Among HRT nonusers, their weight at the time of enrollment was the strongest predictor of breast cancer risk (96). For the HRT nonusers, heavier women (BMI > 31.1) had a greater relative risk of postmenopausal breast cancer compared with slimmer women (BMI -< 22.6); however, this effect was more pronounced in younger women (50-69 years of age) compared with older postmenopausal women (96). Exercise is also important for maintaining cardiorespiratory fitness, which decreases as a result of the aging process. Sedentary individuals have a 1% loss of VO2 max per year with age, particularly after age 50 (31). Physical activity can slow the natural age-related decline in aerobic power. The age-associated decline in muscle mass, strength, and flexibility can largely be prevented by regular exercise participation. Paradoxically, lowering body fat can adversely affect lipids. Exercise at an intensity that results in reduction in body fat and in a consequent drop in endogenous estrogen can cause atherogenic changes in lipoprotein profiles (3). Heavier postmenopausal women tend to have higher endogenous levels of estrone, because of increased aromatization of androstenedione to estrone. Higher estrone levels are related to higher HDLc and lower LDLc levels in perimenopausal women. Both men and women, but especially women, experience an age-related loss of muscle strength. Before beginning an exercise regimen, all women would probably benefit from a strength training program, especially those women who have been inactive for extended periods. Few studies have examined whether exercise can effectively prevent CVD. Lack of compliance has been a problem in many studies. Randomized clinical trials should be conducted to prove that exercise reduces cardiovascular risk and to establish the level of intensity, frequency, and exercise duration that produces maximal protective effects in postmenopausal women. These studies are hard to do because of lack of blinding, long-term intervention, and compliance issues. The combined effects of exercise, diet, and HRT on CV-D risk remain controversial.
IV. OBESITY Obesity is characterized by excess body fat or an excess storage of triglycerides in adipose tissue that results from consumption of a high-fat, high-calorie diet and a sedentary lifestyle. Obesity is a condition that has become more prevalent in
664 the United States and one that is often resistant to treatment. Women make up a greater percentage of the obese population than do men (97). The levels of central and total adiposi~, which increase with age, contribute to the development of cardiovascular and metabolic disease. The incidence of obesity increases threefold in adult women until age 65 (98). Individuals who are above the 85th percentile of BMI, which is the weight in kilograms divided by the height in meters squared, are usually classified as obese. Obesity is a risk factor for diabetes, low HDLc level, hypertension, degenerative arthritis, lipid disorders, gallbladder disease, renal disease, cirrhosis of the fiver, and several forms of cancer (24). Obesity places people at increased risk for CVD, because it is associated with four major risk factors for atherosclerosis: hypertension, diabetes, hypercholesterolemia, and hypertriglyceridemia (24). The waist/hip ratio is used to estimate the relative proportions of upper body or android obesity and lower body or gynoid obesity. A waist/hip ratio greater than 0.85 is indicative of android obesity and a ratio of less than 0.75 is indicative of gynoid obesity. The waist/hip ratio is the index of obesity most strongly associated with C H D (24). Android obesity is associated with increased risk of CVD. Central body fat is metabolically active: It is sensitive to catecholamines and insensitive to insulin. Gynoid adiposity is a store of fat that does not exhibit significant fatty acid fluxes because, unlike android obesity, it is resistant to catecholamines and sensitive to insulin. Women with gynoid obesity are less likely to develop diabetes mellims and C H D than are women with android obesity. Body fat in women tends to be distributed superficially on the body frame, whereas men are prone to central obesity (97). A significant increase in waist/hip ratio, which is indicative of central obesity, is observed in menopausal women, and this increase places women at greater risk for coronary artery disease, whereas lower-body or appendicular adiposity appears less harmful and is more favorably associated with serum triglyceride and HDLc levels and markers of insulin resistance (97,99). It is more difficult for the obese woman to lose weight than it is for the obese man; therefore, it is important for women to avoid gaining excessive amounts of weight during middle age. Men are able to maintain HDLc levels and lose central obesity by diet alone, whereas women must reduce their caloric intake and exercise to obtain the same results (97). A body mass index greater than 27 or more in late middle age, or approximately 65 years old, is associated with increased risk of coronary heart disease in late life (100). Women with a BMI greater than 27 would probably benefit from initiating a dietary and exercise regimen, and those with a BMI of 28 or more should definitely be treated (24). Increased mortality is seen at a BMI of 30, which corresponds to roughly 30% excess body weight (24). The American Heart Association (AHA) recommends women maintain a BMI between 18.5 and 24.9 kg/m 2 and a waist circumference less than 35 inches (25).
WARREN ET AL.
Central obesity is associated with an androgenic hormonal stares, hypertension, and irregularities in lipid and carbohydrate metabolism in middle-aged women (63), and it places women at an increased risk of C H D (101). Abdominal fat is associated with an atherogenic lipid profile in women that is influenced by insulin and estrogen (102). A direct relationship exists between central adiposity and increased total cholesterol, triglycerides, and LDLc in women and an inverse relationship between central fat and HDLc (103). The waist/hip ratio is the variable that is most strongly and inversely associated with the level of the HDL2 cholesterol subfraction (104). The HDL2 cholesterol subfraction is strongly related to protection from CVD. Women with large waist/hip ratios, which are indicative of android obesity, will have low HDL2 cholesterol levels, and they will therefore be at increased risk of developing CV-D. The amount of weight gained during menopause is strongly correlated with the degree to which cardiovascular risk factors such as changes in the lipid profile, blood pressure, and insulin levels become more pronounced (105). Blood insulin levels are elevated in obesity because excess body fat affects insulin secretion and sensitivity, resulting in insulin resistance. In both men and women, insulin resistance is influenced by total adipose tissue content, daily caloric intake, carbohydrate content of the diet, and daily exercise participation (24). In obese individuals, the observed increase in insulin secretion causes downregulation of insulin receptors, which in turn leads to insulin resistance. Carbohydrate, fat, and protein metabolism are all adversely affected by insulin resistance, and inadequate insulin suppression of fat cells leads to increases in circulating free fatty acids. HDLc levels decrease and LDLc levels increase because of the reduced catabolism of triglycerides caused by the insulin resistance (24). The atherogenic changes in lipid profiles caused by insulin resistance result in increased risk of CVD. High insulin levels can also contribute to the development of hypertension. Although hyperinsulinemia in obese individuals significantly increases their CVD risk profile, it is responsive to diet and exercise. Hyperinsulinemia is reversible with weight loss. Exercise promotes a more sensitive insulin response (94), and long-term athletic training is associated with a lower risk of developing diabetes (93). The resting metabolic rate decreases about 2% per decade after age 18, resulting in progressive weight gain over the years if no change occurs in caloric intake or exercise level (24). A study examining the effects of age and gender on energy expenditure independent of differences in body composition found no gender effect and no linear decrease in energy expenditure with increasing age; however, the middle-aged subjects had a lower basal metabolic rate (BMR) than did the younger ones (106). This effect on BMR is independent of body size, body composition, and level of activity (106).
665
CHAPTER49 Role of Exercise and Nutrition Dietary and exercise interventions to prevent weight gain during menopause may also have a protective effect in terms of breast cancer risk (107). Several studies have shown that a history of weight gain in early adult life is associated with increased breast cancer risk in Western women (107). Obesity also seems to be associated with increased breast cancer risk in Asian women. In a prospective case-control study of 1086 Singaporean Chinese women, central obesity was associated with the highest risk for breast carcinoma (107). Excessive weight gain in early adult life can lead to the development of hyperinsulinemia in women who are genetically susceptible to insulin resistance (107). The metabolic relationship between weight gain and breast cancer risk in Western women has been supported by evidence of insulin resistance in these women (107). Some studies have found that hyperinsulinemia is related to overall obesity in postmenopausal women; however, in premenopausal women it is related to abdominal adiposity (107). This finding could be the reason why a high BMI is a risk factor for breast cancer in postmenopausal but not in premenopausal women (107). It is believed that overnutrition and insufficient exercise can promote the development of hyperinsulinemia and increase breast cancer risk in genetically susceptible women, but this hypothesis has not been tested in intervention studies (107). Overweight and obese postmenopausal women should follow a weight reduction program to reduce their risk for CVD, hypertension, diabetes, and cancer. At the time of menopause, the rise in LDLc, triglycerides, and insulin levels is most pronounced in women who put on the most weight (105). The Nurses' Health Study demonstrated the importance of weight reduction in preventing CVD. The study found that women with a BMI of 29 or greater had a threefold increase in risk for CVD (109). In this study, 40% of coronary events were caused by excess body weight. Women who are mildly overweight have a substantial increase in coronary risk, and women who are very overweight have an even higher risk. In the Nurses' Health Study, 70% of coronary events were the result of excess body weight. Physicians should prescribe a diet and actively oversee the management of a realistic weight loss program to which the patient will adhere. In the first month, patients should lose 4 to 5 pounds, and in the next 4 to 5 months they should lose 20 to 30 pounds (24). An appropriate rate of weight loss can be achieved when energy expenditure exceeds energy intake by 500 to 1000 calories (110). As patients lose weight they should decrease their energy intake, because their energy requirements will be lower than they were before. For patients wishing to maintain their current weight, women over the age of 51 who are sedentary should consume approximately 1600 calories a day, women who are moderately active should consume 1800 calories a day, and women who are active should consume between 2000 and 2200 calories a day (58). The optimal diet for weight loss should be composed of 50% carbohydrates, 15% to 20% protein, and less than 30% fat.
Less than 10% of calories should be from saturated fatty acids, and the consumption of cholesterol should be less than 300 mg/day (58). This diet will not compromise vitamin and mineral status, but further caloric restriction has an adverse effect on overall nutritional status. The majority of fats should be from polyunsaturated and monounsaturated fatty acids, such as fish, nuts, and vegetable oils. Trans fatty acid consumption should be as low as possible. Reducing fat intake is the most successful method of weight loss, because fat has twice as many calories per gram than either carbohydrate or protein. Successful weight loss programs set realistic goals that can be reached by gradual weight loss through diet and exercise, and they teach patients behavior modification strategies. Weight loss and increased physical activity have been shown to reduce LDLc levels and to increase HDLc (111). In overweight and obese women, exercise alone or exercise combined with a low-fat diet was found to independently raise HDLc (112,113). Strenuous or prolonged physical activity inhibits appetite for an extended period and increases the resting metabolic rate for 24 to 48 hours. Obese individuals must incorporate increased physical activity into their lifestyles, because exercise is necessary to increase caloric expenditure significantly and to lose weight. It is important for the patient and the physician to design a weight loss program that encourages compliance, because repeated dieting and recidivism have a negative impact on metabolism. With successive diets the body can become more calorically efficient, resulting in difficulty in achieving and maintaining weight loss.
V. DIABETES The metabolic syndrome in women is characterized by risk factors including abdominal obesity, atherogenic dyslipidemia, elevated blood pressure, insulin resistance or glucose intolerance, prothrombotic state, and proinflammatory state (Fig. 49.4) (114). People with metabolic syndrome are at increased risk of coronary heart disease and other disease relating to plaque buildup in artery walls, as well as type 2 diabetes (114). Physical inactivity is often associated with the syndrome. Studies show that adoption of regular exercise by older adults (over the age of 67) is associated with reduced development of metabolic risk factors for cardiovascular disease, fewer exercise-induced cardiac abnormalities, and reduced comorbidity (115). A recent study by researchers at Johns Hopkins determined that in people age 55 to 75, a moderate program of physical exercise can significantly offset the metabolic syndrome risk factors causing heart disease and diabetes (116). In the Johns Hopkins study, exercise improved overall fitness, but the 23% fewer cases were more strongly linked to reductions in total and abdominal body fat and increases in muscle leanness, rather than improved fitness.
666
WARREN ET AL.
Diabetes mellitus is a chronic metabolic syndrome characterized by glucose intolerance, hyperglycemia, absolute or relative insulin availability, and insulin resistance. The alterations in carbohydrate, protein, and lipid metabolism in diabetics produce atherosclerosis and microvascular complications. Type 2, or non-insulin dependent diabetes mellitus (NIDDM), occurs more frequently with advancing age, and it is one of the major chronic disorders of older women. It is characterized by insulin resistance. A family history of diabetes, obesity, and age are the major risk factors for diabetes (117). An increased prevalence of diabetes is seen in women with advancing age, because women have a high tendency toward weight gain. Diabetes is a more important risk factor for CVD in women than in men, possibly because of its effects on lipoprotein profiles (8). Exercise and weight reduction are effective in the primary prevention of diabetes meUitus in most older women. Type 2 diabetics are insulin resistant and thus have hyperinsulinemia. Insulin resistance appears to be caused by decreased insulin receptor numbers and to postreceptor abnormalities, both of which cause alterations in the insulin action system. Genetic predisposition, aging, physical inactivity, and weight gain, especially android obesity, affect the development of insulin resistance. Certain patterns of fat distribution in middle-aged adults may confer additional risks for metabolic syndromes such as diabetes mellitus. The Health, Aging and Body Composition Study found that visceral
....R i s k
facs
"~ominal
........... Obesity
Defining--level '.......W.a i s t
cirCumferenCe
Men
> 102 cm (> 40 in)
Women
> 88 cm (> 35 in)
...T....r...i....g...l ....y....c...e....r....i...d....e....s...........................................z........1....5....0... m g / ~
................
HDL cholesterol Men
< 40 mg/dL
Women
< 50 mg/
B l o o d pressure
.....F a s s
-~ 1301185 ~ g
.....
giueose ............. e ~ l0 m g / ~
FmURE49.4 Clinicalidentificationof the metabolicsyndrome.The ATP III paneldid not find adequateevidenceto recommendroutine measurement of insulin resistance (e.g. plasma insulin), proinflammatorystate (e.g. highsensitivity C-reactive protein), or prothrombotic state (e.g. fibrinogen or PAI-1) in the diagnosis of metabolic syndrome.FromThird Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol and Adults (Adult Treatment Plan III). Executivesummary.AccessedMay 2001, from
http://www.nhlbi.nih.gov/guidelines/cholesterol/index.htm.
abdominal adipose tissue (AT) and muscle-associated AT are related to insulin resistance in older subjects of normal weight, and that accumulation of these regional AT depots is characteristic of older people with type 2 diabetes and impaired glucose tolerance (118,119). Metabolic syndromes should therefore not be discounted on the basis of body composition, including body weight and BMI (118). Hyperinsulinemia in postmenopausal women is related to overall obesity, whereas in premenopausal women hyperinsulinemia is related to abdominal adiposity (107). Type 2 diabetics have a high incidence of complications caused by atherosclerosis, such as CHD, cerebrovascular disease, peripheral vascular disease, and premature and severe coronary artery disease (CAD). Type 2 diabetes also produces multiple lipid disorders, which place patients at increased risk of developing CHD. Elevated LDLc, elevated triglycerides, oxidized LDLc, and decreased HDLc are commonly seen in patients with poorly controlled type 2 diabetes. Both diet and exercise can prevent many of these complications. Improved diet and exercise are also protective against breast cancer. The incidence of breast cancer is high in societies with a sedentary lifestyle and with diets rich in fat, saturated fat, and refined carbohydrates (120). Exercise is considered an important intervention, because it increases insulin sensitivity and decreases triglycerides and total cholesterol. Exercise also improves glucose tolerance in both lean and obese type 2 diabetics under age 55; however, in those older than age 55, this effect is only seen in obese patients (121). Women who are obese, who have a family history of diabetes, or who have a history of gestational diabetes should periodically monitor their fasting plasma glucose levels, and they should initiate an exercise and weight reduction program for the primary prevention of diabetes mellitus (24). A sedentary individual who initiates an exercise program consisting of walking approximately 5 km per day or swimming, running, or biking 30 to 60 minutes per day is 50% less likely to develop diabetes (24). Long-term athletic training in premenopausal women is associated with a lower risk of developing diabetes (93). Preventing weight gain and possibly losing weight also decrease the likelihood of developing diabetes (93). A group of women ages 34 to 59 years who initiated a once a week exercise program had a 16% decrease in the relative risk of diabetes during an 8-year follow-up (122). Women in this cohort who lost weight and exercised showed a 33% decrease in their relative risk. Primary prevention of type 2 diabetes mellitus should be initiated at an early age in high-risk women. All women more than 30 years old who are overweight, hypertensive, or who have a family history of diabetes should exercise at least once a week to lose weight and thereby prevent type 2 diabetes. Therapy for type 2 diabetes consists of relieving symptoms and decreasing the risk of vascular complications, which are a common problem in these patients. Based on observations in
667
CHAPTER 49 Role of Exercise and Nutrition type I diabetics (123,124), it has been hypothesized that lowering glucose levels can reduce microvascular complications in type 2 diabetics; however, this has not been supported by data in type 2 diabetics (24). Type 2 diabetics should attempt to obtain good control of blood glucose through diet and exercise. The American Diabetes Association (ADA) and the World Health Organization diagnosis diabetes mellitus at random plasma glucose measurements greater than 200 mg/ dL with associated diabetes symptoms, at fasting plasma glucose levels greater than 126 mg/dL, or with 2-hour plasma glucose levels greater than 200 mg/dL (125). The ADA targets fasting glucose measurements less than 120 mg/dL (125). Patients without diabetes mellims may still demonstrate impaired glucose tolerance, with fasting plasma glucose levels less than 126 mg/dL and 2-hour postglucose load levels greater than 140 mg/dL (125). Dietary therapy for diabetes focuses on weight loss and reduced consumption of refined carbohydrates (126). Weight gain and overeating should be prevented because they increase insulin resistance (127). Simple carbohydrates such as sucrose should be avoided because they produce sharp increases in blood glucose. Complex carbohydrates such as starch do not cause drastic fluctuations in blood glucose. Women should restrict caloric intake to 34 kcal/kg of ideal body weight (24). Obese patients require a more severe caloric restriction to achieve weight loss. The diet should consist of 60% or less carbohydrates (mostly complex carbohydrates and dietary fiber); approximately 5% refined sugar; 12% to 20% protein; less than 30% total fat, 10% saturated fat, 10% unsaturated fat, and 10% monounsaturated fat; and 300 mg or less cholesterol per day (126). Dietary fiber has been shown to reduce serum cholesterol and to decrease the postprandial rise in blood glucose (128,129). The ADA initially recommended a total of 40 g of crude and soluble fiber intake per day (126) but subsequently found that such a high fiber intake interferes with the absorption of some vitamins and minerals. It now recommends an intake of 20 to 35 g of fiber per day for healthy adults (130). A low protein intake is essential to prevent the renal complications associated with high-protein diets. Type 2 diabetics will benefit from strict adherence to a fixed diet consisting of three meals a day with no snacks, because such a diet will improve plasma glucose levels when insulin levels are limited. Patients should be introduced to the Exchange Listsfor Meal Planning (126) to ensure proper nutritional balance. A nutrition assessment and an annual dietary history should be conducted in diabetic patients to determine compliance with the prescribed meal plan (131). Nutritional needs of diabetic patients require particular attention because complications associated with diabetes can affect these requirements (131). Vitamin and mineral requirements in diabetics may exceed the RDAs, and micronutrient imbalances or deficiencies can result (131). Vitamin and mineral deficiencies may play a role in decreased insulin secretion and
increased insulin resistance (131). Patients with poor control can lose large amounts of water-soluble vitamins and minerals (131). Careful dietary monitoring and supplementation may be useful in patients at risk of developing deficiencies; however, routine vitamin and mineral supplementation is not indicated in the management of most patients (131).
VI. OSTEOPOROSIS In the United States, approximately 15% of women age 50 or older have osteoporosis and 35% to 50% have low bone mass (135). Prevention of osteoporosis is extremely important, because compromised bone strength predisposes women to an increased risk of fracture. Bone strength is reflected in bone mass, bone density, bone architecture, bone size, and bone mineral quality (135). More than 40% of women in the United States over the age of 50 will suffer an osteoporotic fracture, typically in the vertebrae, hip, pelvis, ribs, distal forearm, and other limb bones (135). Further, there is a high association of hip fractures with increased mortality and decreased independence, with women suffering from a hip fracture showing 20% mortality in the first year; 50% of these patients will have some long-term decrease in mobility and independence, and 25% of these women will require long-term care (135). Osteoporosis is a multifaceted disease in which genetics, endocrine function, exercise, and nutrition play important roles. Once the disease becomes established, treatment options are not highly effective. Therefore, preventative strategies aim at preventing fractures by slowing or preventing bone loss, maintaining bone strength, and minimizing or eliminating factors that may contribute to falls. Preventive strategies attempt to achieve and maintain peak bone mass through diet, exercise, HRT, and avoiding behaviors such as smoking that have adverse effects on bone (132). Parathyroid hormones used as a therapeutic agent may be available for some patients wishing to stimulate new bone formation (135). Risk factors for osteoporosis include menopause, smoking, poor vitamin D and calcium nutrition, lack of weight-bearing exercise, high alcohol intake, and a family history of the disease (132,133). The contribution of dietary factors, most notably calcium intake, and exercise to peak bone development has recently been under increased scrutiny. Improved diet and exercise during childhood are essential to reach adulthood with optimal bone mass. Bone mass accumulation until the age of 20 is determined largely by hereditary factors (134). Heredity accounts for 50% to 70% of the accumulated bone mass, but the remaining 30% to 50% is likely determined by dietary and other lifestyle factors until early adulthood (134). After age 20, lifestyle factors such as diet and physical activity become increasingly important in promoting an increase in bone mass. Attainment of peak
668 bone mass (PBM) and peak bone density (PBD) during the adolescent and early adult years and reducing bone loss after menopause are factors believed to be most important in preventing osteopenia and osteoporotic fractures during the postmenopausal years. Adequate calcium along with vitamin D intake prevents bone loss and reduces the risk of spine, hip, and other fractures in perimenopausal and postmenopausal women (135,136,138). Proper calcium nutrition increases bone mineral density (BMD) during periods of skeletal growth and prevents loss of bone and osteoporotic fractures in the elderly (137). During menopause, calcium requirements increase because bone resorption rate increases and bone mass declines (135). This increase in calcium needs results from the fall in ovarian estrogen production and the resulting decrease in efficiency of utilization of dietary calcium (135). At menopause, calcium absorption efficiency is typically 50% below that of adolescent peak absorption and is believed to be associated with a lack of vitamin D resulting from age-related declines in ingestion, dermal synthesis (139), renal enzymatic activity (140), and intestinal responsiveness (135,141). Total BMD decrease in the spine is approximately 15%, beginning 1.5 years after the last menstrual period and continuing at a rate of 3% for about 5 years (135). Hip BMD declines at a rate of approximately 0.5% per year before and after menopause and loses approximately 5% to 7% across the menopause transition period (135,142). Calcium supplementation slows the rate of postmenopausal bone loss by 30% to 50% (143). A review of more than 20 comprehensive studies indicates a bone loss of 0.014% per year in postmenopausal women receiving calcium supplementation compared with a bone loss of 1.0% per year in untreated women (144). The smaller decrease in bone loss was maintained for up to 4 years in longer-term trials (145-147). A 3-year study of older women (65 years and older) given 500 mg/day supplemental calcium (with vitamin D) demonstrated significant reduction in the loss of total BMD compared with a placebo after 1 and 3 years of therapy (148). The North American Menopause Society recommends at least 1200 mg/day of calcium for most women, with levels not exceeding 2500 mg/day (135). In postmenopausal women, loss of bone density in the forearm is significantly attenuated by daily calcium supplementation with 1000 to 2000 mg (149-151). Several studies indicate that calcium supplementation does not have a beneficial effect on spinal bone loss in early postmenopausal women (149,152,153). One study in immediately postmenopausal women reports a beneficial effect on cortical bone from calcium supplements (154). Calcium supplementation in women 3 to 6 years postmenopausal significantly reduced the rate of total body and femoral neck bone loss (155). In late postmenopausal women with low dietary calcium intakes (400 mg/day), a 500 mg/day
WARREN ET AL.
calcium supplement resulted in improvements in bone density of the spine, proximal femur, and radius as compared with the control group; however, the supplementation did not benefit women with higher initial calcium intakes (153). A 4-year randomized, placebo-controlled trial of 1 g/day calcium supplements in late postmenopausal women found significant reductions in the rate of bone loss from the total body, lumbar spine, and proximal femur (156,157). In the second half of the study, the group receiving calcium supplementation showed statistically significantly less bone loss in the total body. Based on evidence from this study, it can be concluded that the long-term use of calcium supplements produce small but significant cumulative benefits at baseline calcium intakes of 750 mg, which are typical of postmenopausal women in Western countries (143,157). A further study found that vitamin K1 supplementation in conjunction with calcium, magnesium and vitamin D over a 3-year period reduced bone loss of the femoral neck (158). Calcium appears to augment the effect of exercise on improving BMD in postmenopausal women; however, the benefits of exercise plus calcium were only observed with doses of calcium above 1000 mg (159). A 2-year randomized, placebo-controlled study examining the effects of calcium supplementation (1 g/day) and weight-bearing exercise in women who were more than 10 years postmenopausal found that calcium supplementation resulted in cessation of bone loss at the hip and in a significant reduction in the rate of bone loss at the tibia (160). Exercise plus calcium supplementation resulted in less bone loss at the femoral neck than calcium supplementation alone. Bone loss in elderly women may be attenuated by calcium in conjunction with vitamin D supplementation (133). In elderly women who are vitamin D deficient, rectification of the deficiency has been associated with a significant decline in hip fractures (161,162). In fact, a large 18-year prospective study found that adequate vitamin D intake is associated with a lower risk of osteoporotic hip fractures in postmenopausal women, whereas milk and high-calcium diets do not appear to reduce the risk (163). A large trial with elderly women with low vitamin D levels and low calcium intake who received supplemental vitamin D (800 IU/day) and calcium (1200 mg/day) for 18 months had significantly fewer nonvertebral fractures (32%) and hip fractures (43%) than placebo recipients (164). A recent study of daily calcium and vitamin D supplementation in men and women more than 65 years old found that 500 mg of calcium and 700 IU of vitamin D3 moderately reduced bone loss in the femoral neck, spine, and total body during the 3-year study period (165). The supplementation program also reduced the incidence of nonvertebral fractures. Optimal calcium absorption is dependent on proper vitamin D nutrition (166). A fall in intestinal calcium absorption causes an increase in parathyroid hormone secretion leading to mobilization of calcium from the skeleton and bone loss. This process is reversed by vitamin D therapy, which
669
CHAPTER 49 Role of Exercise and Nutrition improves calcium absorption from the bowel by 20% of the level of insufficiency (167). Elderly patients are at increased risk of vitamin D deficiency (levels less than 10 n g / m L of 25-hydroxyvitamin D [25OH]D]), and their vitamin D status should consequently be monitored. T h e level of insufficiency is the level of 2 5 ( O H ) D at which there appears to be an inadequate absorption of calcium to maintain necessary physiologic levels of circulating calcium levels. A parathyroid hormone increase is triggered, which occurs at the expense o f b o n e - - a level at about 30 ng/mL. One study of 78 hospitalized patients with
TABLE 49.2
M e a n intake of vitamin D (Ixg) from food and food plus dietary supplements. CSFI a
Sex/age category Both sexes, 1-3 years Both sexes, 4 - 8 years Males, 9-13 years Males, 14-18 years Males, 19-30 years Males, 31-50 years Males, 51-70 years Males, -> 71 years Females, 9-13 years Females, 14-18 years Females, 19-30 years Females, 31-50 years Females, 51-70 years Females, - 71 years
Sample size
osteoporotic fractures (76 were hip fractures) reported that 97% had vitamin D levels less than 30 ng/ml (168). T h e National Health and Nutrition Examination Survey ( N H A N E S ) III report showed that 70% of women between 50 to 70 had inadequate vitamin intake as well as 90% of women over 70 (Table 49.2) (169). Another study recently showed that more than 50% of women over 50 in North America have low levels of vitamin D, and the proportion increases with age (170). Daily supplementation with 600 to 800 IU of vitamin D in institutionalized elderly patients improves calcium balance and reduces
Vitamin D from food (mg/d 6 IU/d) c
NHANES III b
Sample size
Vitamin D from food (mg/d 6 IU/d) d
Vitamin D from food and supplements (mg/d 6 IU/d) d
Percent using vitamin D-containing supplements
3777
6.0 _+ 0.05 (242)
3309
5.7 + 0.09 (228)
9.8 _+ 0.32 (393)
39.3
3769
5.9 + 0.08 (235)
3448
6.0 _+ 0.10 (240)
9.4 _+ 0.22 (376)
36.0
569
6.6 _ 0.19 (265)
1219
6.6 _+ 0.27 (264)
8.7 _+ 0.33 (346)
22.0
446
6.9 _ 0.28 (274)
909
7.0 _+ 0.40 (280)
7.7 +_ 0.40 (310)
11.0
854
4.0 _+ 0.18 (190)
1902
5.5 _+ 0.21 (220)
8.0 _ 0.53 (320)
20.4
1684
5.2 _+ 0.11 (210)
2533
5.7 _+ 0.25 (230)
7.9 _ 0.35 (315)
23.7
1606
5.2 _+ 0.10 (207)
1942
5.7 _+ 0.18 (227)
8.0 _ 0.27 (320)
25.4
674
5.5 _+ 0.15 (220)
1255
5.9 +_ 0.16 (236)
8.3 __ 0.24 (332)
26.0
580
5.2 _+ 0.15 (206)
1238
5.2 _+ 0.24 (210)
7.3 __ 0.31 (291)
24.1
436
3.9 ___0.14 (154)
974
4.3 _+ 0.20 (172)
5.3 +_ 0.23 (210)
14.9
760
3.5 _+ 0.15 (140)
1919
3.9 _ 0.18 (156)
6.1 _+ 0.23 (244)
25.2
1614
3.7 _+ 0.09 (148)
2953
3.9 _+ 0.13 (158)
7.1 _+ 0.35 (283)
30.4
1539
4.0 _+ 0.14 (160)
2076
4.5 _+ 0.16 (178)
7.8 _+ 0.29 (312)
31.0
623
4.3 _ 0.12 (172)
1368
4.5 _ 0.12 (180)
8.1 _ 0.56 (325)
29.7
Two-day average estimates based on data from the U.S. Department of Agriculture, Agricultural Research Service, Continuing Survey of Food Intakes by Individuals (CSFIII), 1994-1996, 1998, and ENVIRON Health Sciences Institute Vitamin D database. Error terms for the CSFII data were calculated with WesVar Complex Samples 3.0 software (Westat, 1998). bDay 1 estimates based on data from U.S. Department of Health and Human Services, National Center for Health Statistics, Third National Health and Nutrition Examination Survey (NHANES III), 1988-1994. Standard errors for the NHANES data were estimated by the Taylor linearization method utilizing Software for Statistical Analysis of Correlated Data (SUDAAN User's Manual, Release 7.5, Research Triangle Institute, 1997). cl Ixg vitamin D = 40 International Unites (IU). dContinuing Survey of Food intakes by individuals (CSFII) and the Third National Health and Nutrition Examination Survey (NHANES III). Excludes breastfeeding infants and pregnant and lactating females. Reprinted with permission from Moore C, Murphy MM, Keast DR, Holick ME Vitamin D intake in the United States.JAm Diet Assoc 2004;104:980-983. a
670 fracture risk (171). The supplementation program also reduced the incidence of nonvertebral fractures. In studies of calcium supplementation without concomitant vitamin D, fracture rates were reduced in postmenopausal women with prior spine fractures and low calcium intakes; however, there was no reduction in fracture rates in women with a history of fractures (172). A recent study reports that vitamin D levels are a better predictor of calcium absorption than calcium intake (173-176). Vitamin D also appears to have other beneficial actions, including antiproliferative effects, which may decrease cancer risk, including colon breast and prostate, and effects on muscle and the cardiovascular system and enhance immunity. Vitamin D can be acquired via sunlight alone or by diet. Sunlight is the most effective source of vitamin D, with a slight sunburn yielding the equivalent of 10,000 to 25,000 IU of oral vitamin D (177). A randomized trial found that a group of stroke patients exposed to sunlight over a 1-year period had a higher mean level (25[OH]D), higher bone mineral density, and a lower incidence of hip fracture than a group of patients not exposed to sunlight (177,178). The provision of vitamin D by sunlight can be hampered by sunscreen use, melanin in darkly pigmented people, winter months in northern latitudes, and increased age. Dietary sources of vitamin D include oily fish and fish liver oil; it is found commonly in fortified milk, cereals, and bread products. However, because milk also contains vitamin A, which does not support bone health, it is recommended that women obtain vitamin D through supplements or consumption of dark fish (163). In addition, recent studies indicate that increased milk and calcium consumption are not associated with a lower risk of osteoporotic hip fracture (55,177). Based on current studies, the Food and Nutrition Board recently set the adequate daily intake of vitamin D at 10 ~g for women 51 to 70 years of age and 15 ~tg for women greater than 70 years of age (163,179). A recent study of 2310 adults found that as long as vitamin D intake is adequate, calcium intake levels of more than 800 mg/day may be unnecessary in maintaining calcium metabolism (180). Studies examining the effect of calcium supplementation on fracture number in postmenopausal women indicate that calcium supplements may be effective in preventing fractures despite only a modest 1% to 4% difference in BMD between the treated and placebo groups (157,181). A 4-year follow-up study of elderly women found that calcium supplementation is associated with a 45% reduction in the incidence of vertebral fractures in women with a pre-existing fracture but had no effect on women who were fracture free at the beginning of the study (182). The difference in BMD in the two groups in this study was less than 2%. Small effects on BMD may thus result in significant protective effects on fracture risk, especially in high-risk women who have already suffered a fracture. Calcium supplements cause decreases in bone turnover rates, which result in the preservation of trabecular connectivity
WARREN ET AL.
(143). Supplements improve neuromuscular function and therefore may prevent falls, ultimately resulting in fewer fractures. Long-term studies are needed to definitively assess whether the small benefit on BMD in postmenopausal women is protective against fractures. In most studies, calcium supplementation produced small but significant benefits on axial bone density in women more than 5 years after the menopause. Calcium supplementation is probably most effective when given in several doses, because divided doses will prevent saturation of the calcium-active transport system and will result in its increased absorption (143). Calcium-active transport is also stimulated by the major vitamin D metabolite, 1,25dihydroxyvitamin D. Consuming supplements with meals allows food acids to contribute to the dissolution of insoluble calcium salts and lengthens its transit time in the small bowel, also resulting in improved absorption (143). It is important to have calcium available at night, when bone resorption rates are greatest. Ideally, patients should consume small amounts of calcium with meals and at bedtime. Consuming 500 mg of calcium carbonate after breakfast and after dinner or 1000 mg of calcium lactate-gluconate at night is probably easier for most patients (143). All women older than 51 years should consume 1200 mg/day of calcium (47). Calcium-rich foods such as dairy products should constitute the primary source of daily calcium intake, but supplements can be used by patients who cannot consume adequate amounts of calcium through diet alone. In addition to proper calcium and vitamin D intake, a recent study demonstrated that elderly women with higher protein intakes, approximately 71 g/day as a percentage of energy, had higher BMDs when daily calcium intake exceeded 408 mg/day (183). Bone mass density was significantly higher in the spine (7%), midradius (6%) and total body (5%), but no significant difference was observed in hip BMD (184). The study did not observe any significant difference between dairy and nondairy protein consumption. On the other hand, studies of young adult women show that low-protein diets depresses intestinal calcium absorption and were often accompanied by secondary hyperparathyroidism (184,185). Although the exact repercussions of a low-protein diet are not fully understood, there definitely exists a potential to affect bone health (184,185). The type, duration, and frequency of physical activity needed to result in observable beneficial effects on bone remains controversial. Regular exercise including resistance training and high-impact activity contributes to the development of high peak bone mass and may reduce the risk of falls in older women (138). In a study of female college athletes, activities that involve high skeletal impacts, such as gymnastics, were found to be especially osteotropic for young women (186). Bone mineral density at the lumbar spine and femoral neck was found to respond dramatically to the mechanical loading exercises typical of gymnastics training, whereas running and swimming did not have pronounced effects on
671
CHAPTER 49 Role of Exercise and Nutrition BMD. Muscle strengthening activity did not seem to have a significant effect on BMD in gymnasts (186). A study of mature female athletes had similar findings, namely that women who regularly engage in high-impact physical activity in the premenopausal years have higher BMD than nonathletic control subjects (187). Amenorrhea in premenopausal women, particularly in competitive athletes, can adversely affect bone mass (188). Increasing physical activity can slow bone loss in postmenopausal women, even if exercise produces no significant increase in BMD (189). Studies indicate that moderate levels of physical activity provide protection against later hip fracture, whereas decline in physical activity level over time is an important risk factor for hip fracture (70,190). Longitudinal studies examining the effects of walking interventions show that walking, which is frequently prescribed to postmenopausal women, does not prevent bone loss (191-194). Higher intensity exercises may be required to attenuate menopausal bone loss. A randomized, controlled trial of high-intensity strength training exercises two days per week
conducted in postmenopausal women ages 50 to 70 years showed that the exercises preserved bone density and improved muscle mass, strength, and balance (195). No evidence indicates that exercise alone can replace bone loss during menopause (189). A 12-month study investigating the effects of aerobic training conducted three times a week at 70% to 85% of maximal heart rate (Fig. 49.5) for 30 to 45 minutes and calcium supplementation did not find significant increases in forearm or lumbar BMD, but the training did attenuate lumbar BMD loss in early postmenopausal women (approximately 6 years of the onset of menopause) (196). The exercise program produced significant gain in aerobic power. Prospective studies of strength-training, muscle loading, and aerobic exercise programs in postmenopausal women are difficult to compare because they differ in terms of exercise prescription, length of follow-up, subject age hormonal status, and method and site of BMD measurement. Nevertheless, the literature seems to indicate that exercise increases forces on bone and may thereby attenuate loss of bone mass (197).
FIGURE 49.5 Aerobictraining guide. Exercise performed at 70% to 85% of the maximal heart rate was found to be protective againstlumbar bone loss in earlypostmenopausalwomen and to produce significantgain in aerobicpower.(From ref. 24, with permission.)
672 It is important to understand the influence of behavioral and genetic factors on BMD in premenopausal and postmenopausal women to assess adequately the effectiveness of interventions aimed at increasing or maintaining BMD. A study conducted on 25 elderly women, whose mean age was 72 years, and on their premenopausal daughters, whose mean age was 41 years, investigated the associations between lifetime milk consumption, calcium intake from supplements, lifetime weight-bearing exercise, premenopausal and postmenopausal hormone use, and BMD to determine the relative importance of lifestyle factors, such as calcium intake and physical activity, and genetic contributions (198). In the older women, multiple regression analyses showed that total and peripheral BMD were positively related to calcium intake from supplements after age 60, body weight, current ERT, and past oral contraceptive (OC) use. Axial BMD in this group was positively associated with body weight and past OC use. By contrast, in the premenopausal women, total and peripheral BMD were found to be determined by lifetime weight-bearing exercise, and axial BMD was found to be determined by total lean body mass. Mothers and daughters had similar lifetime milk consumption patterns (198). The results of this study indicate that supplemental calcium intake and exogenous estrogen have a positive effect on bone mass in postmenopausal women and that physical actMty is effective in preventing osteoporosis. Behavioral and hormonal factors had a stronger effect on BMD than did familial similarity in the premenopausal and postmenopausal women studied. Women, therefore, are capable of increasing their genetically determined bone mass by engaging in weight-bearing physical activity, following a postmenopausal ERT regimen, and consuming an adequate amount of calcium (198). Calcium alone does not prevent menopause associated bone loss as effectively as when combined with estrogen replacement therapy (ERT)/HRT, selective estrogen receptor modulators (SERMs), or bisphosphonates (135,199). Although calcium plus vitamin D can reduce the risk of fracture, it is no substitute for an antiresorptive agent in early postmenopausal women. The Multiple Outcomes of Raloxifene Evaluation (MORE) study of 7705 postmenopausal women with osteoporosis showed a reduced risk of vertebral fractures (200). In addition, the bisphosphonates alendronate and risedronate have also shown a reduction in vertebral and hip fractures in women with low bone density, although their effectiveness was dependent on the women's fracture status at the entry into the clinical trial (201-204). Vitamin D deficiency results in secondary hyperparathyroidism and increased rates of bone catabolism. The effects of the potent vitamin D metabolites such as calcitriol and alphacalcidol on postmenopausal bone density remain controversial (143). HRT or the potent bisphosphonates produce greater increases in BMD than the vitamin D metabolites (143). HRT
WARREN ET AL.
is routinely used to prevent and treat osteoporosis. In fact, the 2004 Position Statement issued by The North American Menopause Society recognizes that ET and combined estrogen-progestogen therapy (EPT) are effective in reducing the risk for postmenopausal osteoporosis fractures; however, it is suggested that ET/EPT use must be balanced against the risks (204). Studies of early postmenopausal women indicate that calcium alone was not as effective as ERT/HRT alone in reducing estrogen-withdrawal bone loss, but supplemental calcium did improve the efficacy of ERT/HRT (206). Several studies investigating whether calcium supplementation is beneficial to women on HRT seem to indicate that it is, but more study is needed in this area (152,207,208). Estrogen, when used as an antiresorptive agent, can be prescribed alone as estrogen replacement therapy or as hormone replacement therapy (135). The effects of HRT and exercise on bone have been found to be synergistic in many studies; however, some studies have not found a complementary effect. A placebocontrolled, 2-year prospective trial of two estrogen-progestin regimens in healthy postmenopausal women investigating the effects of HRT and exercise on bone density, muscle strength, and lipid metabolism found that exercise exerted a positive effect on BMD in the placebo group, however, in the HRT group no synergistic effect of exercise and estrogen on BMD was observed (209). Exercise cannot substitute for HRT, which clearly has the most significant impact on BMD. Estrogen therapy is effective in the prevention of osteoporosis and in treatment of established osteoporosis (210). A doubleblind, placebo-controlled randomized study conducted in 120 postmenopausal women (mean age 56 years) with low forearm bone density examined the effects of an exercise regimen, exercise combined with 1000 mg of supplemental calcium, and exercise plus estrogen and progesterone (Fig. 49.6) (150). The control group, which consisted of women with normal bone density, had a 2.7% decrease in distal forearm bone density per year. The exercise group showed a similar 2.6% decrease. The exercise-calcium group had significantly reduced bone loss (-0.5% of the baseline value per year). Bone density actually increased by 2.7% over the baseline value per year in the exercise-estrogen group. Similar patterns of bone loss and accrual were observed in the median forearm. In postmenopausal women with low BMD, bone loss can be slowed or prevented by a combined regimen of HRT and exercise or by calcium supplementation plus exercise (150). An exerciseestrogen regimen was more effective in increasing bone mass than calcium plus exercise. The exercise regimen used in this study, a 1-hour low-impact aerobics session per week with 30% of the time devoted to arm exercises, and two 30-minute brisk walks per week, was fairly moderate. A more significant exercise effect might have been observed at a higher training intensity. Nevertheless, the results of this study seem to indicate that both exercise and nutrition play important roles in the prevention and treatment of osteoporosis. In addition, a study administering estrogen and high amounts of calcium
673
CHAPTER 49 Role of Exercise and Nutrition
for the proper development and maintenance of the skeleton, and sedentary women can slightly increase their bone mass and prevent further bone loss by becoming more active (189). Exercise programs for older women should aim to improve muscle tone, strength, flexibility, and coordination, which may reduce falls in older women and prevent osteoporotic fractures (133,189). Many postmenopausal women are inactive. Clinicians should help their patients design an indMdualized moderate exercise program that is both safe and encourages compliance. The osteogenic effects of such a program may be modest, but the long-term effects on cardiorespiratory fitness, strength, and agility will improve the overall health and quality of life of postmenopausal women well into old age.
VII. TRACE ELEMENTS
FIGURE 49.6 How various treatments affect bone density. Effects of exercise only, calcium with exercise, and estrogen with exercise in postmenopausal women. (Adapted from ref. 150,with permission.)
(1183 mg/day through diet and supplements) demonstrated a significant increase in BMD at various skeletal sites in comparison to estrogen therapy combined with low calcium intakes (563 mg/day)(206). Body weight also contributes to BMD. Obesity and N I D D M are associated with increased bone mineral density. In a study of 559 women, fasting insulin levels were significantly and positively associated with bone density in the radius and in the spine (211). This study indicates that hyperinsulinemia may partly explain the association of diabetes and obesity with BMD in women (211). Weight loss in postmenopausal women can lead to decreased BMD. The number of pregnancies and a BMI of less than 20 kg/m 2 also had an adverse effect on BMD in a group of Mexican women (212). Increasing age and lack of exercise were found to be significant predictors of bone demineralization in this group. In summary, women with established osteoporosis can be treated with estrogen, or with estrogen and progestin, calcium, and vitamin D (133). Although HRT has the most significant impact on bone, its use remains controversial because of the side effects. Neither calcium supplementation nor exercise can substitute for HRT at the time of menopause, but their importance should not be overlooked. Estrogen-related bone loss usually takes place 3 to 6 years postmenopause, but loss caused by calcium deficiency results in continued bone loss until the nutritional inadequacy is corrected (213). Many postmenopausal women have low calcium intakes. Absorption efficiency decreases with age, exacerbating the existing nutritional inadequacy. Supplementation is therefore important in this population. Weight-bearing exercise is essential
Although the role of trace minerals in osteoporosis is still controversial, increasing evidence indicates that trace minerals are important for adequate bone formation and maintenance. Calcium, vitamin D, fluoride, magnesium, and trace elements such as copper, manganese, and zinc are essential for optimal bone matrix development and the maintenance of bone mineral density (214). These trace elements are necessary because they serve as cofactors for certain enzymes. A study investigating the role of copper, manganese, and zinc in bone metabolism in a rat model found that female rats placed on diets low in manganese or in copper and manganese for a period of 12 to 24 months had significantly lower BMD than those placed on a trace mineral-sufficient diet (215). The trace element-deficient groups had higher serum calcium concentrations. Serum and femur calcium content were inversely correlated; examination of isolated femurs showed an increase in femur porosity. Based on this observation, the authors concluded that trace mineral deficiency can result in changes in bone crystal composition and alteration of calcium control at the level of the bone (manifested as decreased mineralization or increased bone resorption), intestine, or kidney (215). Another group investigated the effects of dietary copper or manganese restriction on serum levels of calcium, copper, and manganese and on body mass density of the right femoral shaft in male rats kept on one of four diets for 12 weeks (216). By 8 weeks, BMD in the femoral shaft was significantly lower in the rats that were copper and manganese deficient than in the controls; rats placed on diets deficient in only one of the two trace minerals had intermediate BMD (216). Subcutaneous implants of demineralized bone powder (DBP) and bone powder (BP) were used to determine the effects of long-term dietary deficiencies in manganese and copper on the cellular activity of bone formation and bone resorption in vivo (217). This study found that osteoblast activity is compromised more than osteoclast activity,
674 resulting in increased bone resorption. Both bone formation and resorption are impaired by long-term dietary manganese and copper deficiency (215). Manganese deficiency in rats also resulted in lowered proteoglycan content in the organic matrix of bone (214). Evidence from animal studies for the importance of trace elements in bone formation and maintenance led investigators to study the role of trace elements in humans. A retrospective study in normal and osteoporotic women showed that bone integrity correlates with serum manganese levels (218). The osteoporotic women had low trabecular bone volume, low bone mineral content, low BMD, and significantly lower serum manganese levels than the control group. Cross-sectional studies in a group ofpostmenopausal women showed a strong correlation between low BMD and low dietary calcium intake and serum copper levels (219). A 2-year prospective, double-blind, placebo-controlled clinical trial evaluating the effect of supplementary calcium with and without a combination of copper, manganese, and zinc found a significant association between BMD and supplementation with calcium and trace minerals (220,221). Dietary calcium, copper, manganese, and zinc supplements could significantly decrease the loss of BMD in postmenopausal women (220,221). A recent study of postmenopausal women found that iron was associated with greater BMD at all sites (222). The study uncovered a complex calcium-iron relationship in which women who consumed 800 to 1200 mg of calcium had significantly higher BMD with increasing levels of iron intake. However, women with lower or higher intakes of calcium did not demonstrate the same association of iron and BMD. Although the exact relationship of iron to BMD is not fully understood, it is known that iron is essential for the synthesis of collagen structure upon which bone mineralization occurs (223). Iron is also involved in the conversion ofhydroxyvitamin D to 1,25-dihydroxyvitamin D, the active form of vitamin D (222,224). Animal studies have confirmed the association of iron to BMD. In one study on iron deficient rats it was found that the iron deficient rats had decreased bone mechanical strength in their femurs as compared to normal rats with similar BMD, BMC and dietary levels of calcium (222,225). Other studies found that iron deficiency resulted in low bone mass and bone volume (222,226,227). However, it appears that limitations to iron supplementation exist and iron overload has been associated with low bone density (222,228,239). Further studies of iron intake and the biologic mechanisms of iron and bone mass density are warranted. These studies suggest that trace elements are needed for optimal bone development and for maintaining bone density. Postmenopausal women who consume a balanced diet can significantly reduce the incidence ofosteoporosis. The amount of calcium used in the clinical trial was 1000 mg/day and the amount of zinc used was 15 mg. The current U.S. RDA for
W A R R E N ET AL.
calcium is 1200 mg/day for women over the age of 51 and 8 mg/day of zinc and 1.8 mg/day of manganese (46). The amount of copper used was 5 mg and 2.5 mg. The U.S. RDA of copper is 900 p~g/day (46,220,221). Many multivitamin supplements with minerals contain similar amounts of copper, zinc, and manganese. It therefore might be prudent for postmenopausal women to take a daily multivitamin with minerals in addition to consuming at least 1200 mg of calcium through the diet or by the use of supplements.
VIII. DEPRESSION The incidence of psychiatric disorders increases with each decade of life (230), and depression is more prevalent in women than in men. Many women deny they are suffering from depression and often refuse to see psychiatrists. Women have a 20% to 25% lifetime risk of suffering from a major depression (231). Many women and some clinicians share the misconception that women are likely to become depressed or irritable during menopause. Cross-sectional and longitudinal analyses of 2565 women ages 45 to 55 years participating in the Massachusetts Women's Health Study (MWHS) found no association between menopause status or change in menopause status and depression (231,232). Prior depression was found to be highly predictive of increased depression. A lengthy perimenopausal period was associated with increased depression (232). Depression typically manifests itself as depressed mood; however, older women more often have other symptoms such as loss of appetite, weight loss, decreased energy levels, decreased motivation, and sleep disturbances (24). Physicians should prescribe a suitable course of antidepressant medication and psychotherapy for treatment of depression, but the potentially adverse impact of a depressive disorder on a woman's nutritional status should be carefully monitored. No hard data exist on the role of nutrition in the management of depression. Nevertheless, the use of nutritional supplements, which provide a good balance of essential nutrients, might prove helpful in the depressed patient who may not have the appetite or the motivation to consume regular meals. Health professionals and regular exercisers share the belief that exercise produces psychologic benefits (233). The effects of exercise on anxiety, depression, personality, cognition, fatigue, socialization, and work performance have been extensively investigated, but most of the studies were not carefully designed or controlled (233,234). One study reviewed randomized, controlled experiments conducted mostly in male populations of all age groups on the psychologic effects of habitual aerobic exercise on mood, personality, and cognition (233). The study found evidence that exercise improves self-concept; however, it did not find significant evidence to substantiate claims that exercise
CHAPTER 49 Role of Exercise and Nutrition improves anxiety, depression, body image, personality, or cognition (233). An earlier review of the literature found suggestion that physical fitness results in improved mood, self-concept, and work behavior, and that improvements in physical fitness have an effect on self-concept but not on other personality traits in male and female cohorts of various age groups (234). A popular belief that does not appear to be supported by the scientific literature is the concept of the runner's high, a feeling of euphoria believed to result from release of [3-endorphin. A study examined the effect of running on plasma [3-endorphin in 6 female and 20 male trained long distance runners (235). Although running did produce increases in plasma [3-endorphin that were more pronounced at higher training intensities, the authors caution that the decrease in anxiety reported after running, the other mood changes, and the runner's high cannot be caused by the small changes that occurred in peripheral plasma [3-endorphin concentration (235). Most studies of the effects of exercise on psychologic health have been conducted in men. A clear need exists for well-designed studies in women, especially because preliminary evidence indicates that postmenopausal women will derive benefits from exercise in terms of fitness, psychologic well-being, and overall health (236). A randomized, controlled trial studying the impact of a 12-month exercise program on the physical and psychologic health in 124 postmenopausal, osteopenic women ages 50 to 70 years found that exercise produced significant increases in functional fitness, psychologic well-being, and self-perceived health (236). The exercise program consisted of weight-bearing exercises, aerobic dancing, and flexibility exercises performed for 60 minutes three times per week. This program was tailored for osteopenic women, who also benefited from decreased back pain and a stabilization of spinal BMD, but similar programs suited to all postmenopausal women should be studied in controlled, randomized trials. All women should be encouraged to exercise because of the potential physical and psychologic benefits they can derive from exercise. Even if future studies do not find a scientific basis for a psychologic benefit from exercise, the placebo effect can be beneficial and possibly more effective in motivating women to exercise.
IX. PHYTOESTROGENS Natural estrogen-like substances called phytogens found in plants are attracting increasing attention, particularly with reference to their role in menopause. Epidemiologic data from Asian countries, where the diet naturally contains large amounts of these products, suggest that they may behave like estrogen by modifying symptoms such as hot flashes and thus attenuating menopausal symptoms. This diet may offer an intriguing alternative treatment in the menopause
675 and may explain why some countries appear to have a lower incidence of menopausal symptoms. Phytoestrogens are naturally occurring compounds that are structurally or functionally similar to estradiol (E2). They consist of a number of classes, including isoflavones, lignans, coumestrans, and resorcyclic acid lactones. They are biologically active, and this biologic activity has been shown in animals (237). In general the human diet provides precursors for mammalian lignans and isoflavones. Even more intriguing is that the metabolism of the precursors to the phytoestrogens may be highly variable, suggesting that the production of the phytoestrogens can vary from individual to individual. The major components of isoflavones, genistein and daidzein, are found in chickpeas, lentils, soybeans, bluegrass, and red clover. Most of the metabolism occurs as a result of fermentation by gut flora, but the liver may also be involved (238). In humans, 30% to 70% of dietary isoflavones are converted into various metabolites (239,240). Four isoflavones m formononetin, biochanin, daidzein, and genisteinmappear to have a wide range of biologic effects on human cells, including activating the steroid receptor. In countries where soy and legumes are a major part of the diet, they can provide 30 to 100 mg of the four isoflavones daily. Unfortunately, current studies are inconclusive as to the efficacy of phytoestrogens in treating menopausal symptoms (241). A meta-analysis evaluating the effect of phytoestrogens from soy found an average of 9.3% decrease in total cholesterol, a 12.9% decrease in LDL cholesterol, a 10.5% decrease in triglycerides, and a 2.4% increase in HDL; however, these changes were not significant (242). One study did find a significantly lower Lp(a) levels with higher habitual isoflavone intake (242). It is believed that Lp(a) is a strong risk factor for atherosclerotic heart disease and has been associated with premature cardiovascular disease (242). In addition to their effect on the menopause, some studies show that the excretion of phytoestrogen is associated with a substantial reduction in breast cancer risk. Thus, the action of these phytoestrogens may be different in some respects from estradiol, which has been associated with a slight increase in breast cancer risk (243). This suggests that the phytoestrogens may act selectively on some estrogen receptors and not on others, in particular on the estrogen receptor beta. This receptor is expressed more prominently in the brain, prostrate, and urinary tract and only weakly in the breast cells, where the classic estrogen receptor alpha is prominent.
X. SUMMARYAND CONCLUSIONS Physicians should encourage their patients to initiate or continue a program of regular physical activity and improved nutrition. A better diet and regular exercise can help prevent
676
W A R R E N ET AL.
the development of the chronic diseases of older women. In patients with established disease, exercise and nutrition may delay its progression, help reverse the disease, and improve prognosis. Aerobic exercise has beneficial effects on cardiovascular disease, cancer, obesity, diabetes, and, possibly, on depression; it specifically improves lipid profiles and insulin resistance and lowers body fat. Strength training and flexibility exercises can reduce bone loss and prevent osteoporotic fractures, respectively. Trace minerals may be important in the management of osteoporosis. Phytoestrogens may improve menopausal symptoms and reduce breast cancer risk. Healthy women can safely initiate an exercise program or change their diet with adequate counseling. Exercise and changes in diet, however, may pose serious health risks in women with certain conditions. These patients should be carefully evaluated to determine the safety of a dietary and exercise regimen, and they should be monitored on a regular basis. Exercise frequency and intensity should be prescribed according to the ability and motivation of each patient. Frequent, relatively high intensity exercise might be more beneficial, but such a regimen is less likely to encourage compliance, may result in increased injury rates, and thus may not have lasting effects on overall health. Exercise performed at 70% to 75% of the maximal heart rate at least three times per week for 50 to 60 minutes is protective against cardiovascular disease and osteoporosis (24). A brief period of warm-up and cool-down is important. Jumping, bouncing, and quick bending of the spine should be avoided. Compelling evidence suggests that improved nutritional and activity profiles can significantly affect the development of chronic diseases that accelerate after menopause. Diet composition should be altered in specific conditions such as CHD and diabetes and appropriate exercise programs prescribed. At-risk patients should also be aware of nutritional and exercise interventions that can significantly attenuate the development of such conditions as osteoporosis.
References 1. Sharp PC, Konen JC. Women's cardiovascular health. Prim Care 1997; 24:1-14. 2. Wenger NK, Speroff L, Packard B. Cardiovascular health and disease in women. NEnglJMed 1993;329:247-256. 3. Meilahn EN, Becket RC, Corrao JM. Primary prevention of coronary heart disease in women. Cardiology 1995;86:286-298. 4. Sullivan JM. Practical aspects of preventing and managing atherosclerotic disease in postmenopausal women. Eur Heart J 1996;17(suppl D):32-37. 5. National Cholesterol Education Program, National Heart Lab. Summary of the Second Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA 1993;269:3015- 3023. 6. Eaker ED, Packard B, Thorn TJ. Epidemiology and risk factors for coronary heart disease in women. Cardiovasc Clin 1989;19:129-145.
7. Simkin-Silverman L, Wing RR, Hansen DH, et al. Prevention of cardiovascular risk factor elevations in healthy premenopausal women. Prev Med 1995;24:509-517. 8. Gerhard M, Baum P, Raby KE. Peripheral arterial-vascular disease in women: prevalence, prognosis and treatment. Cardiology 1995;86: 349-355. 9. Dallongeville J, Marecaux N, Isorez D, et al. Multiple coronary heart disease risk factors are associated with menopause and influenced by substitutive hormonal therapy in a cohort of French women.Atherosclerosis 1995;118:123-133. 10. Heiss G, Tamir I, Davis CE, et al. Lipoprotein-cholesterol distributions in selected North American populations: the Lipid Research Clinics Prevalence Study. Circulation 1980;61:302. 11. Campos H, McNamara JR, Wilson PWF, Ordovas JM, Schaefer EJ. Differences in low density lipoprotein sub fractions and apolipoproteins in premenopausal and postmenopausal women. J Clin Endocrinol Metab 1988;67:30-35. 12. Campos H, Genest JJ, Blijlevens E, et al. Low-density lipoprotein particle size and coronary artery disease. Arterioscler Thromb VaseBid 1992;12:187-195. 13. Mathews KA, Meilahn E, Kuller LH, et al. Menopause and risk factors for coronary heart disease. NEnglJMed 1989;321:641-646. 14. Cheung MC, Brown BG, Wolf AC, Albers JJ. Altered particle size distribution of apolipoprotein AI-containing lipoproteins in subjects with coronary artery disease. J Lipid Res 1991;32:383 - 394. 15. Berge LN, Bonaa KH, Nordoy A. Serum ferritin, sex hormones, and cardiovascular risk factors in healthy women. Arteriosd Thromb 1994; 14: 857-861. 16. Mullen BE, Franz MM, Q.uatromoni PA, et al. Diet and plasma lipids in women. Macronutrients and plasma total low-density lipoprotein cholesterol in women: the Framingham nutrition studies. J Clin Epidemid 1996;49:657-663. 17. Sonnenberg LM, Q.uatromi PA, Gagnon DR, et al. Diet and plasma lipids in women. II. Macronutrients and plasma triglycerides, highdensity lipoprotein, and the ratio of total to high-density lipoprotein cholesterol in women: the Framingham nutrition studies. J Clin Epidemio11996;49:665-672. 18. Stensvold I, Tverdal A, Urdal P, Graff-Lissner L. Non-fasting serum triglyceride concentration and mortality from coronary heart disease and any cause in middle aged Norwegian women. Br Med J 1993; 307:1318-1322. 19. Jacobs DRJ, Mebane IL, Bangdiawala SI, Criqui MH, Tyroler HA. High density lipoprotein cholesterol as a predictor of cardiovascular disease mortality in men and women: the follow-up study of the Lipid Research Clinic Prevalence Study. Am J Epidemio11990;131:32-47. 20. Bass KM, Newschaffer CJ, Klaq MJ, Bush TL. Plasma lipoprotein levels as predictors of cardiovascular death in women. Arch Intern Med 1993; 153:2209 - 2216. 21. Brunner D, Weisbort J, Meshulam N, et al. Relation of serum total cholesterol and high-density lipoprotein cholesterol percentage to the incidence of definite coronary events: twenty-year follow-up of the Donolo-Tel Aviv Prospective Coronary Artery Disease Study. Am J Cardio11987;59:1271-1276. 22. Posner BM, Cupples AD, Gagnon D, et al. The rationale and potential efficacy of preventive nutrition in heart disease: the Framingham Offspring-Spouse Study. Arch Intern Med 1993;153:1549-1556. 23. Hong MK, Romm PA, Reagan K, Green CE, Rackley CE. Usefulness of total cholesterol to high-density lipoprotein cholesterol ratio in predicting angiographic coronary artery disease in women. Am J Cardiol 1991;68:1646-1650. 24. Byyny RL, SperoffL.A clinicalguidefor the careof older women:primary and preventive care, 2nd ed. Baltimore: Williams & Wilkins, 1996. 25. Expert Panel/Writing Group, Mosca L. Evidence-based guidelines for cardiovascular disease prevention in women. Arterioscler Thromb Vasc Bio12004;24:e29-e50.
CHAva'v.~ 49 Role of Exercise and Nutrition 26. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol and Adults (Adult Treatment Plan III). Executive summary. Accessed May 2001, from http://www.nhlbi.nih.gov/guidelines/ cholesterol/index.htm. 27. National Heart Lab. Declining serum cholesterol levels among U.S. adults. JAMA 1993;269:23. 28. Wilson PWF, Christiansen JC, Anderson KM, Kannel WB. Impact of national guidelines for cholesterol risk factor screening: the Framingham Offspring Study.JAMA 1989;262:41-44. 29. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol and Adults (Adult Treatment Plan III). Executive summary. Accessed May 2001, from http://www.nhlbi.nih.gov/guidelines/ cholesterol/index.htm. 30. National Cholesterol Education Program, National Heart La. Report of the National Cholesterol Education Program Expert Panel on detection, evaluation and treatment of high blood cholesterol in adults. Arch Intern IVied 1988;148:36-39. 31. Notelovitz M. Nutrition and the coagulation effects of estrogen replacement on cardiovascular health. Obst Gynecol Clin North Am 1987; 14:121-141. 32. Anderson JW. Diet, lipids, and cardiovascular disease in women. J A m Coli Nutr 1993;12:433-437. 33. Zhang S, Folsom AR, Sellers TA, Kushi LH, Potter Jo. Better breast cancer survival for postmenopausal women who are less overweight and eat less fat. The Iowa Women's Health Study. Cancer 1995;76:275-283. 34. Preuss HG. Nutrition and disorders of women: cardiovascular disorders.JAm Coli Nutr 1993;12:417-425. 35. Wild RA, Taylor LE, Knehans A. The gynecologist and the prevention of cardiovascular disease. Am J Obstet Gyneco11995;172:1 - 13. 36. Siani A, Stazzullo P, Russo L, et al. Controlled trial of long term oral potassium supplements in patients with mild hypertension. Br MedJ 1987;294:1453-1456. 37. Rytlig K. Treatment of mild to moderate hypertension with dietary fibre. Lancet 1987;1:622-623. 38. Mulrow C. An editorial update: Should she take aspirin? Ann Intern Med 2005;142:942-943. 39. American Heart Association Docket Branch. Re." The American Heart Association's Comments on File Code 205P-0048." PROFESSIONAL LABELING FOR ASPIRIN DOSING I N ORDER FOR SECONDARY CARDIOVASCULAR PREVENTION. Accessed October 13, 2005, from www.fda.gov/ohrms/dockets/dockets/05p0048/05p-0048-c00000501-vol3.pds 40. Ames BN, Shigenaga MK. Antioxidants are a major contributor to cancer and ageing. In: Halliwell B, Arumona OR, eds. DNA and free radicals. Chichester: Horwood, 1993:1-15. 41. Gey KF. Ten-year retrospective study on the antioxidant hypothesis of arteriosclerosis: threshold plasma, levels of antioxidant micronutrients related to minimum cardiovascular risk.JNutr Biochem 1995;6:206-236. 42. Gey KE Optimum plasma levels of antioxidant micronutrients. Ten years of antioxidant hypothesis on arteriosclerosis. Bibl Nutr Dieta 1994;51:84-99. 43. Gey KE Prospects for the development of free radical disease, regarding cancer and cardiovascular disease. Br Med Bull 1993;49:679-699. 44. Gey KE Cardiovascular disease and vitamins. Concurrent correction of 'suboptimal' plasma antioxidant levels may, as important part of'optimal' nutrition help to prevent early stages of cardiovascular disease and cancer, respectively. Bibl Nutr Dieta 1995;52:75- 91. 45. Gey KE Vitamin E and other essential antioxidants regarding coronary heart disease: risk assessment studies. Epidemiological basis of the antioxidant hypothesis of cardiovascular disease. In: Packer L, Fuchs J, eds. Vitamin E." biochemistry and clinical applications. New York: Dekker, 1993:589-633.
677 46. Food and Nutrition Board, National Research Council. Recommended dietary allowances. Washington, DC: National Academy Press, 1989. 47. Food and Nutrition Board, National Research Council. Dietary reference intakes: vitamins. Washington, DC: National Academy Press, 2002. http://www.nal.usda.gov/fnic/etext/OOOlOS.html. 48. Enstrom JE, Kanim LE, Klein MA. Vitamin C intake and mortality among a sample of the United States population. Epidemiology 1992;3: 194-202. 49. Hunter oJ, Manson JE, Colditz GA, et al. A prospective study of the intake of vitamins C, E, and A and the risk of breast cancer. N EnglJ Med 1993;329:234-240. 50. Blot WJ, Li JY, Taylor PR, et al. Nutrition intervention trials in Linxian, China: supplementation with specific vitamin/mineral combinations, cancer incidence and disease-specific mortality in the general population. J Natl Cancer Inst 1993;85:1483-1492. 51. Blot wJ, Li JY, Taylor RR, et al. The Linxian trials: mortality rates by vitamin-mineral intervention group. Am J Clin Nutr 1995;62: 1424S- 1426S. 52. Heinonen OP, Albanes D. The effect of vitamin E and beta-carotene on the incidence of lung cancer and other cancers in male smokers. N EnglJ IVied 1994;330:1029-1035. 53. Kushi LH, Fee RM, Sellers TA, Zheng W, Folsom AR. Intake of vitamins A, C, and E and postmenopausal breast cancer. The Iowa Women's Health Study. Am J EDidemiol 1996;144:165-174. 54. Manson JE, Gaziano JM, Jonas MA, Hennekens CH. Antioxidants and cardiovascular disease: a review.JAm Coli Nutr 1993;12:426-432. 55. Clemente C, Caruso MG, Berloco P, et al. Alpha-tocopherol and betacarotene serum levels in women treated with transdermal estradiol and oral medroxyprogesterone acetate. Horm Metab Res 1996;28:558-561. 56. Franken DG, Boers GH, Blom HJ, Trijbels FJ, Kloppenborg PW. Treatment of mild hyperhomocysteinemia in vascular disease patients. Arteriosd Thromb 1994;14:465-470. 57. Law MR, Morris JK. By how much does fruit and vegetable consumption reduce the risk of ischaemic heart disease? EurJ Clin Nutr 1998; 52:549-556. 58. Weisweiler E Plasma lipoproteins and lipase and lecithin: cholesterol acyl transferase activities in obese subjects before and after weight reduction. J Clin Endocrinol Metab 1987;65:969-973. 59. Baer DF, Judd JT, Clevidence BA, et al. Moderate alcohol consumption lowers risk factors for cardiovascular disease in postmenopausal women fed a controlled diet. Am Soc Clin Nutr 2002;75:593-599. 60. Blair SN, Kohl HW, Paffenberger RS, et al. Physical fitness and allcause mortality. A prospective study of healthy men and women.JAMA 1989;262:2395-2401. 61. Sedgwick AW, Thomas DW, Davies M. Relationships between changes in aerobic fitness and changes in blood pressure and plasma lipids in men and women: the Adelaide 1000 4-year follow-up. J Clin EDidemio1199 3 ;461:141-151. 62. Krummel O, Etherton TO, Peterson S, Kris-Etherton PM. Effects of exercise on plasma lipids and lipoproteins of women. Proc Soc Exp Biol Med 1993;204:123-137. 63. Lapidus L, Bengtsson C, Lindquist O, Sigurdsson J, Rybo E. Triglycerides main lipid risk factor for cardiovascular disease in women. Acta Med &and 1985;217:481-489. 64. Powell KE, Thompson PO, Caspersen CJ, Kendrick JS. Physical activity and the incidence of coronary heart disease. Annu Rev Pub Health 1987;8:253-287. 65. Slattery ML, Jacobs DRJ, Nichaman MZ. Leisure time physical activity and coronary heart disease death: the US railroad study. Circulation 1989;79:304-311. 66. Leon AS, Connet J, Jacobs DR, Raumaraa R. Leisure-time physical activity levels and risk of coronary heart disease and death: the Multiple Risk Factor Intervention Trial. JAMA 1987;258:2388-2395.
678 67. Paffenbarger RSJ, Hyde RT, Wing AI, Hsieh CC. Physical activity, all-cause mortality and longevity of college alumanganesei. N EnglJ Med 1986;314:605-613. 68. Berlin JA, Colditz GA. A meta-analysis of physical activity in the prevention of coronary heart disease. Am J Epidemiol 1990;132: 612-628. 69. Stevenson ET, Davy KP, Jones PP, Desouza CA, Seals DR. Blood pressure risk factors in healthy postmenopausal women: physical activity and hormone replacement. JAppl Phsio11997;82:652-660. 70. Schroll M. Physical activity in ageing population. ScandJ Med Sci Sports 2003;13:63-69. 71. Hartung GH, Moore CE, Mitchell R, Kappus CM. Relationship of menopausal status and exercise level to HDL cholesterol in women. Exp Aging Res 1984;10:13-18. 72. Douglas PS, Clarkson TB, Flowers NC, et al. Exercise and atherosclerotic heart disease in women. Med Sci Sports Exerc 1992;14: $266-$276. 73. Rainville S, Vaccaro R The effects of menopause and training on serum lipids. IntJ Sports Med 1984;5:137-141. 74. Blair SN, Goodyear NN, Gibbons LW, et al. Physical fitness and incidence of hypertension in healthy normotensive men and women. JAMA 1984;252:487-490. 75. Owens IF, Matthews KA, Wing RR, Kuller LH. Can physical activity mitigate the effects of aging in middle-aged women? Circulation 1992;85:1265-1270. 76. Ready AE, Naimark B, Ducas J, et al. Influence of walking volume on health benefits in women post-menopause. Med Sci Sports Exerc 1996;28:1097-1105. 77. Lemaitre RN, He&bert SR, Psaty BM, Siscovik OS. Leisure-time physical activity and the risk of nonfatal myocardial infarction in postmenopausal women. Arch Intern Med 1995;155:2302-2308. 78. Williams PT. High-density lipoprotein cholesterol and other risk factors for coronary heart disease in female runners. NEnglJMed 1996; 334:1325-1327. 79. Pate RR, Pratt M, Blair SN, et al. Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA 1995;273: 402-407. 80. King AC, Haskell WL, Young DR, Oka RK, Stefanick ML. Longterm effects of varying intensities and formats of physical activity on participation rates, fitness, and lipoproteins in men and women aged 50 to 65 years. Circulation 1995;91:2596-2604. 81. Reaven PP, McPhillips JB, Barret-Connor EL, Criqui MH. Leisure time exercise an lipid and lipoprotein levels in an older population. JAm Geriatr Soc 1990;38:847-854. 82. Lindheim SR, Notelovitz M, Feldman EB, et al. The independent effects of exercise and estrogen on lipids and lipoproteins in postmenopausal women. Obstet Gyneco11994;83:167-172. 83. Binder EF, Birge SJ, Kohrt WM. Effects of endurance exercise and hormone replacement therapy on serum lipids in older women. JAm Geriatr Soc 1996;44:231-236. 84. NIH News. NIH asks participants in Women's Health Initiative Estrogen-Alone Study to stop study pills, begin follow-up phase. Accessed March 2, 2004, from www.nhlbi.nih.gov/new/press/04-03-02.htm. 85. Women's Health Initiative Participant Web site. Estrogen plus progestin effects on bone density and the risk of fractures. Accessed February 28, 2007, from http://www.whi.org/findings/ht/eplusp_bone.php. 86. Women's Health Initiative Participant Web site. Estrogen plus progestin effects on breast cancer and mammograms. Accessed February 28, 2007, from http://www.whi.org/findings/ht/eplusp_bc.php. 87. Women's Health Initiative Participant Web site. Estrogen plus progestin and risk of coronary heart disease. Accessed February 28, 2007, from http://www.whi.org/findings/ht/eplusp_chd.php.
WARREN ET AL.
88. Women's Health Initiative Participant Web site. Effects of estrogen plus progestin on stroke in the Women's Health Initiative. Accessed February 28, 2007, from http://www.whi.org/findings/ht/ eplusp_stroke.php. 89. Women's Health Initiative Participant Web site. WHI letter for healthcare providers. Accessed February 28, 2007, from http://www. whi.org/findings/ht/ealone_provider_letter.php. 90. Frisch RE, Wyshak G, Albright NL, et al. Former athletes have a lower lifetime occurrence of breast cancer and cancers of the reproductive system. Adv Exp Med Bid 1992;322:29-39. 91. Frisch RE, Wyshak G, Witschi J, et al. Lower lifetime occurrence of breast cancer and cancers of the reproductive system among former college athletes. IntJ Ferti11987;32:217-225. 92. Frisch RE, Wyshak G, Albright NL, Albright TE, Schiff I. Lower prevalence of non-reproductive system cancers among female former college athletes. Med Sci Sports Exerc 1989;21:250-253. 93. Frisch RE, Wyshak G, Albright TE, Schiff I. Lower prevalence of diabetes in female former college athletes compared with nonathletes. Diabetes 1986;35:1101-1105. 94. Frisch RE, Snow RC, Johnson LA, et al. Insulin response of women athletes in relation to body fat quantified by magnetic resonance imaging. GynecolObstet lnvest 1995;40:195-199. 95. McTiernan A, Kooperberg C, White E, et al. Recreational physical activity and the risk of breast cancer in postmenopausal women: the Women's Health Initiative Cohort StudyyAMA 2003;290:1331-1336. 96. Morimoto LM, White E, Chen Z, et al. Obesity, body size, and risk of postmenopausal breast cancer: the Women's Health Initiative (United States). Cancer Causes Control 2002;13:741-751. 97. Legato MJ.Gender-specific aspects of obesity. IntJFertil WomensMed 1997;42:184-197. 98. Barret-Connor E. Obesity, atherosclerosis, and coronary artery disease. Ann Intern Med 1995;103:1010. 99. Van Pelt RE, Jankowski CM, Gozansky WS, Schwartz RS, Kohrt WM. Lower-body adiposity and metabolic protection in postmenopausal women. J EndocrinolMetab 2005;90:4573-4578. 100. Harris TB, Launer LJ, Madans J, Feldman JJ. Cohort study of effect of being overweight and change in weight on risk of coronary heart disease in old age. BMJ 1997;314:1791. 101. Wing RR, Matthews KA, Kuller LH, Meilahn EN, Platinga R Waist-to-hip ratio in middle-aged women. Associations with behavior and psychosocial factors and with changes in cardiovascular risk factors. Arterioscler Thromb VaseBio11992;11:1250-125 7. 102. Soler JT, Folsom AR, Kaye SA, Prineas RL. Associations of abdominal adiposity, fasting insulin, sex hormone binding globulin and estrogen with lipids and lipoproteins in postmenopausal women. Atherosclerosis 1989;79:21-27. 103. Haarbo J, Hassager C, Riis BJ, Christiansen C. Relation of body fat distribution to serum lipids and lipoproteins in elderly women. Atherosclerosis 1989;80:57-62. 104. Ostlund REJ, Staten M, Kohrt W, Schultz J, Malley M. The ratio of waist-to-hip circumference, plasma insulin level, and glucose intolerance as independent predictors of the HDL2 cholesterol level in older adults. N EnglJ IVIed 1990;322:229-234. 105. Wing RR, Mathews KA, Kuller LH, Meilahn EN, Plantinga PL. Weight gain at the time of menopause. Arch Intern Med 1991;151: 97-102. 106. Klausen B, Toubro S, Astrup A. Age and sex effects on energy expenditure. Am J Clin Nutr 1997;65:895-907. 107. Stoll BA. Timing of weight gain in relation to breast cancer risk.Ann Onco11995;6:245-248. 108. Ng EH, Gao F, Ji CY, Ho GH, Soo KC. Risk factors for breast carcinoma in Singaporean Chinese women: the role of central obesity. Cancer 1997;80:725-731.
CHAPTER 49 Role of Exercise and Nutrition 109. Manson JE, Colditz GA, Stampfer MJ, et al. A prospective study of obesity and risk of coronary heart disease in women. N EnglJ Med 1990;322:882-889. 110. Garrow JS. Treatment of obesity. Lancet 1992;340:409-413. 112. Nieman DC, Haig JL, Fairchild KS, et al. Reducing-diet and exercisetraining effects on serum lipids and lipoproteins in mildly obese women. Am J Clin Nutr 1990;52:640-645. 113. Wood PD, Stefanick ML, Williams PT, Haskell WL. The effects on plasma lipoproteins of a prudent weight-reducing diet, with or without exercise, in overweight men and women. N E n g l J M e d 1991;325: 461-466. 114. American Heart Association. What is metabolic syndrome? Accessed November 7, 2005, from www.americanheart.org/presenter.jhtml? identifier= 4756-4 lk. 115. Petrella RJ, Lattanzio CN, Demeray A, Varallo V, Blore R. Can adoption of regular exercise later in life prevent metabolic risk for cardiovascular disease? Diabetes Care 2005;28:694-701. 116. Stewart KJ, Bacher AC, Turner K, et al. Exercise and risk factors associated with metabolic syndrome in older adults. Am J Prev Med 2005;28:9-18. 117. Shimokata H, Muller DC, Fleg JL, et al. Age as independent determinant of glucose tolerance. Diabetes 1991;40:44-51. 118. Goodpaster BH, Krishnaswami S, Harris T, et al. Obesity, regional body fat distribution and the metabolic syndrome in older men and women. Arch Intern Med 2005;165:777-783. 119. Goodpaster BH, Krishnaswami S, Resnick H. Association between regional adipose tissue distribution and both type 2 diabetes and impaired glucose tolerance in elderly men and women. Diabetes Care 2003;26:372-379. 120. Kaaks R. Nutrition, hormones and breast cancer: is insulin the missing link? Cancer Causes Control 1996;7:605-625. 121. Zierath JR, Wallberg-Henriksson H. Exercise training in obese diabetic patients. Special considerations. Sports Med 1992;14:171-189. 122. Manson ILL, Rimm EB, Stamp Fer MF, et al. Physical activity and incidence of non-insulin-dependent diabetes mellitus. Lancet 1991; 338:774-778. 123. The Diabetes Control and Complications Trial Research Group (DCCT). The effect of intensified treatment of diabetes on the development and progression of long-term complications in insulindependent diabetes mellitus. N EnglJ Med 1993;329:977. 124. Reichard P, Britz A, Carlsson P, et al. Metabolic control and complications over 3 years in patients with insulin dependent diabetes (IDDM): the Stockholm Diabetes Intervention Study. J Intern Med 1990;228:511 - 517. 125. Zimmerman B R. Postprandial hyperglycemia: implications for practice. dm J Cardio12001;88:32H-6H. [Review.] 126. American Diabetes Association. Nutritional recommendations and principles for individuals with diabetes mellitus: 1986. Diabetes Care 1987;10:126-132. 127. ~'tamura S. Diet therapy food exchange lists for diabetic patients. Diabetes Res Clin Pract 1994;24:$233-$240. 128. Anderson JW, Ward K. Long-term effects of high-carbohydrate, high-fiber diets on glucose and lipid metabolism: a preliminary report on patients with diabetes. Diabetes Care 1978;1:77-82. 129. Simpson HCR, Simpson RW, Lousley S, et al. A high carbohydrate leguminous fiber diet improves all aspects of diabetic control. Lancet 1981;1:1-5. 130. American Dietetic Association. Position of the American Dietetic Association: health implications of dietary fiber-technical support paper. J A m Dietet# dssoc 1988;88:217-221. 131. Mooradian AD, Failla M, Hoogwerf B, Marynuik M, Wylie-Rosett I. Selected vitamins and minerals in diabetes. Diabetes Care 1994;17: 464-479.
679 132. Bellantoni ME. Osteoporosis prevention and treatment. Am Faro Physician 1996;54:986- 992. 133. WarkJD. Osteoporotic fractures: background and prevention strategies. Maturitas 1996;23:193-207. 134. Anderson JB, Metz JA. Contributions of dietary calcium and physical activity to primary prevention of osteoporosis in females. J A m Call Nutr 1993;12:378-383. 135. North American Menopause Society. The role of calcium in peri- and postmenopausal women: consensus opinion of The North American Menopause Society. Menopause 2001;8:84-95. 136. Chevalley T, Rizzoli R, Nydeggar V, et al. Effects of calcium supplements on femoral bone mineral density and vertebral facture rate in vitamin-D-replete elderly patients. OsteoporosInt 1994;14:$181. 137. Murray TM. Prevention and management of osteoporosis: consensus statements from the Scientific Advisory Board of the Osteoporosis Society of Canada. 4. Calcium nutrition and osteoporosis. Can Med AssocJ 1996;155:935-939. 138. NIH Consensus Statement. Osteoporosis prevention, diagnosis and therapy. 2000;17:1-45. 139. Holick MF, Matsuoka LY, Worstman J. Age, vitamin D, and solar ultraviolet. Lancet 1989;2:1104-1105. [Letter. ] 140. Slovick DM, Rosenthal DI, Doppelt SH, et al. Restoration of spinal bone in osteoporotic men by treatment with human parathyroid hormone (1-34) and 1,25 dihydroxyvitamin D. J Bone Miner Res 1986;1: 377-381. 141. Gallagher JC, Riggs BL, Eisman J, et al. Intestinal calcium absorption and serum vitamin D metabolites in normal subjects and osteoporotic patients: effect of age and dietary calcium. J Clin Invest 1979;64: 729-736. 142. Recker RR, Lappe J, Davies K, Heaney R. Characterization of perimenopausal bone loss: a prospective study.J Bone Miner Res 2000;15: 1965-1973. 143. Reid IR. Therapy of osteoporosis: calcium, vitamin D and exercise. A m d M e d Soc 1996;312:278-286. 144. Nordin BE. Calcium and osteoporosis. Nutrition 1997;13:664-686. 145. Riggs BL, O'Fallon WM, Muhs J, et al. Long-term effects of calcium supplementation on serum parathyroid hormone level, bone turnover, and bone loss in elderly women. J Bone Miner Res 1998; 13:168 - 174. 146. Elders PJ, Lips P, Netelenbos JC, et al. Long-term effect of calcium supplementation on bone loss in perimenopausal women. J Bone Miner Res 1994;9:963 - 970. 147. Reid JR, Ames RW, Evans MC, Gamble GD, Sharpe SJ. Long-term effects of calcium supplementation on bone loss and fractures in postmenopausal women: a randomized controlled trial.AmJMed 1995;98: 331-335. 148. Dawson-Huges B, Harris SS, Krall EA, Dallal GE. Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age and older. N E n g l J M e d 1997;337;670-676. 149. Riis B; Thomsen SK, Christiansen C. Does calcium supplementation prevent postmenopausal bone loss? N EnglJ Med 1987;316:173 - 177. 150. Prince RL, Smith M, Dick IA, et al. Prevention of postmenopausal osteoporosis. A comparative study of exercise, calcium supplementation, and hormone replacement therapy. N Engl J IVied 1991;325: 1189-1195. 151. Smith EL, Gilligan C, Smith PE, Sempos CT. Calcium supplementation and bone loss in middle-aged women. Am J Clin Nutr 1989;50: 833-842. 152. Ettinger B, Genant HK, Cann CE. Postmenopausal bone loss is prevented by treatment with low-dosage estrogen with calcium. Ann Intern Med 1987;IQ6:40-45. 153. Dawson Hughes B, Dallal GE, Krall EA, et al. A controlled trial of the effect of calcium supplementation on bone density in postmenopausal women. N EnglJ Med 1990;323:878- 883.
680 154. Elders PJ, Netelenbos JC, Lips P, et al. Calcium supplementation reduces vertebral bone loss in perimenopausal women: a controlled trial in 248 women between 46 and 55 years of age. J Clin Endocrinol Metab 1991;73:533-540. 155. Aloia IF, Vaswani A, Yeh JK, et al. Calcium supplementation with and without hormone replacement therapy to prevent postmenopausal bone loss.Ann Intern Med 1994;120:97-103. 156. Reid IR, Ames RW, Eyans MC, Gamble GD, Sharpe SI. Effect of calcium supplementation on bone loss in postmenopausal women. N EnglJ Med 1993 ;328:460 - 464. 157. Reid IR, Ames RW, Evans MC, Gamble GD, Sharpe SI. Long-term effects of calcium supplementation on bone loss in postmenopausal women-- a randomized controlled trial. Am J Med 1995;98:331- 335. 158. Braam LA, Knapen MH, Geusens P, et al. Vitamin K1 supplementation retards bone loss in postmenopausal women between 50 and 60 years of age. CalcifTissue Int 2003;73:21-26. 159. Specker BL. Evidence for an interaction between calcium intake and physical activity on changes in bone mineral density. JBone Miner Res 1996;11:1539-1544. 160. Prince R, Devine A, Dick I, et al. The effects of calcium supplementation (milk powder or tablets) and exercise on bone density in postmenopausal women. J Bone Miner Res 1995;10:1068-1075. 161. Chapuy MC, Arlot ME, Delmas PD, Meunier PI. Effect of calcium and cholecalciferol treatment for three years on hip fractures in elderly women. BMJ 1994;308:1081-1082. 162. Heikinheimo RJ, Inkovaara JA, Harju EJ, et al. Annual injection of vitamin D and fractures of aged bones. Calcif Tissue Int 1992;51: 105-110. 163. Feskanich D, Willett WC, Colditz GA. Calcium, vitamin D, milk consumption, and hip fractures: a prospective study among postmenopausal women. Am J Clin Nutr 2003;77:504- 511. 164. Chapuy MC, Arlot ME, Duboeuf F, et al. Vitamin D3 and calcium to prevent hip fractures in elderly women. N EnglJ Med 1992;327: 1637-1642. 165. Dawson-Hughes B, Harris S, Krall EA, Dallal GE. Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age and older. NEnglJMed 1997;337:670-676. 166. Holick M. Vitamin D: the underappreciated D-rightful hormone that is important for skeletal and cellular health. Curr Opin Endocrinol Diabetes 2002;9:87- 98. 167. Heaney RP, Dowell S, Hale CA, Bendich A. Calcium absorption varies within the reference range for serum 25-hydroxyvitamin D. JAm Coll Nutr 2003;22:142-146. 168. Simonelli C, Weiss TW, Morancey J, Swanson L, Chen Y. A high prevalence of vitamin D inadequacy in a minimal trauma fracture population. Curr Med Res Opin 2005:21:1069-1074. 169. Moore C, Murphy MM, Keast DR, Holick ME Vitamin D intake in the United States. JAm Diet Assoc 2004;104:980-983. 170. Holick MF, Siris ES, Binkley N, et al. Prevalence of vitamin D inadequacy among postmenopausal North American women receiving osteoporosis therapy. J Clin Endocrinol Metab 2005;90: 3215-3224. 171. Anonymous. Optimal calcium intake: NIB Consensus Statement 1994 June 6-8. Nutrition 1995;12:1-31. 172. Recker RR, Hinders S, Davides KM, et al. Correcting calcium nutritional deficiency prevents spine fractures in elderly women. J Bone Miner Res 1996;11:1961-1966. 173. Munger KL, Zhang SM, O'Reilly, et al. Vitamin D intake and incidence of multiple sclerosis. Neurology 2004;62:60-65. 174. Plotnikoff GA, Qgigley JM. Prevalence of severe hypovitaminosis D in patients with persistent, nonspecific musculoskeletal pain. Mayo Clin Proc 2003;78:1463-1470. 175. Holick ME Vitamin D: importance in the prevention of cancers, type 1 diabetes, heart disease, and osteoporosis. Am J Clin Nutr 2004;79: 362-371.
WARREN ET AL.
176. Dietrich T, Joshipura KJ, Dawson-Hughes B, Bischoff-Ferrari HA. Association between serum concentrations of 25-hydroxyvitamin D3 and periodontal disease in the US population. Am J Clin Nutr 2004;80:108-113. 177. Hickey L, Gordon CM. Vitamin D: the underappreciated D-lightful hormone that is important for skeletal and cellular health. Curr Opin Endocrinol Diabetes 2002;9:87- 98. 178. Sato Y, Metoki N, Iwamoto J, et al. Amelioration of osteoporosis and hypovitaminosis D by sunlight exposure in stroke patients. Neurology 2003;61:338- 342. 179. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes, Food and Nutrition Board, Institute of Medicine, National Academy of Sciences. Dietary referenceintakesfor calcium,phosphorus, magnesium, vitamin D, andfluoride. Washington, DC: National Academy Press, 2000. 180. Steingrimsdottir L, Gunnarsson O, Indridason OS, Franzson L, Sigurdsson G. Relationship between serum parathyroid hormone levels, vitamin D sufficiency, and calcium intake. JAMA 2005;294: 2336-2341. 181. Chevalley T, Rizzoli R, Nydegger V, et al. Effects of calcium supplements on femoral bone mineral density and vertebral fracture rate in vitamin-D-replete elderly patients. OsteoporosInt 1994;4:245-252. 182. Recker RR, Kimmel DB, Hinders S, Davies KM. Anti-fracture efficacy of calcium in elderly women. J Bone Miner Res 1994;9:S154. 183. Rapuri PB, Gallagher JC, Haynatzka V. Protein intake: effects on bone mineral density and the rate of bone loss in elderly women. Am J Clin Nutr 2003;77:1517-1525. 184. Kerstetter JE, O'Brien KO, Insogna KL. Dietary protein affects intestinal calcium absorption. Am Soc Clin Nutr 1998;68:859-865. 185. Kerstetter JE, Looker AC, Insogna KL. Low dietary protein and low bone density. CalcifTissue Int 2000;66:313. 186. Taaffe DR, Robinson TL, Snow CM, Marcus R. High-impact exercise promotes bone gain in well-trained female athletes. J Bone Miner Res 1997;12:255-260. 187. Dook JE, James C, Henderson NK, Price RI. Exercise and bone mineral density in mature female athletes. Med Sci Sports Exerc 1997;29:291-296. 188. Constantini NW, Warren ME Physical activity, fitness, and reproductive health in women: clinical observations. In: Bouchard C, Shephard RJ, Stephens T, eds. Physical activity, fitness, and health: international proceedings and consensus statement. Champaign: Human Kinetics, 1994:955-966. 189. Anonymous. American College of Sports Medicine position stand on osteoporosis and exercise. IViedSci Sports Exerc 1995;27:i-vii. 190. Hoidrup S, Sorensen TIA, Stroger U, et al. Leisure-time physical activity levels and changes in relation to risk of hip fracture in men and women. Am J Epidemio12001;154:60-68. 191. Cavanaugh DJ, Cann CE. Brisk walking does not stop bone loss in postmenopausal women. Bone 1988;9:201-204. 192. Nelson ME, Fisher EC, Dilmanian FA, Dallal GE, Evans WI. A 1-year walking program and increased calcium in postmenopausal women: effects on bone.AmJ Clin Nutr 1991;53:1304-1311. 193. Sandler RB, Cauley JA, Horn DL, Sashin D, Kriska AM. The effects of walking on cross-sectional dimensions of the radius in postmenopausal women. CalcifTissue Int 19F;41:65-69. 194. White MK, Martin RE, Yeater RA, Butcher RL, Radin EL. The effects of exercise on the bones of postmenopausal women. Int Orthop 1984;7:209-214. 195. Nelson ME, Fiatarone MA, Morganti CM, Trice I, Greenberg RA, Evans WI. Effects of high-intensity strength training on multiple risk factors for osteoporotic fractures. A randomized controlled trial. JAMA 1994;272:1909-1914.
CHAPTER 49 Role of Exercise and Nutrition 196. Martin P, Notelovitz M. Effects of aerobic training on bone mineral density of postmenopausal women. J Bone Miner Res 1993;8:931-936. 197. Snow CM, Shaw JM, Matkin CC. Physical activity and risk for osteoporosis. In: Marcus R, Feldman D, Kelsey J, eds. Osteoporosis. New York: Academic Press, 1996:511-528. 198. Ulrich CM, Georgiou CC, Snow-Harter CM, Gillis DE. Bone mineral density in mother-daughter pairs: relations to lifetime exercise, lifetime milk consumption, and calcium supplements. Am J Clin Nutr 1996;63:72-79. 199. Heaney RE Calcium, dairy products and osteoporosis.JAm CollNutr 2000;19:83S-99S. 200. Ettinger B, Black DM, Mitlak BH, et al., for the Multiple Outcomes of Raloxifene Evaluation (MORE) Investigators. Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene: results from a 3-year randomized clinical trial. JAMA 1999;282:637-645. 201. Black DM, Cummings SR, Karpf DB, et al. Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet 1996;282:637-645. 202. Harris ST, Watts NB, Genant HK, et al., for the Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. Effects of rise&onate treatment on vertebral and nonvertebral factures in women with postmenopausal osteoporosis: a randomized controlled trial. JAMA 1999;282:1344-1352. 203. Liberman UA, Weiss SR, Broll J, et al., for the Alendronate Phase III Osteoporosis Treatment Study Group. Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. N EnglJ Med 1995;333:1437-1443. 204. Reginster J, Minne HW, Sorensen OH, et al. Randomized trial of the effects of risedronate on vertebral fractures in women with established postmenopausal osteoporosis. OsteoporosInt 2000;11:83 - 91. 205. The North American Menopause Society. Recommendations for estrogen and progestogen use in peri- and postmenopausal women: October 2004 position statement of the North American Menopause Society. Menopause 11;6:589-600. 206. Nieves JW, Komar L, Cosman F, Lindsay R. Calcium potentiates the effect of estrogen and calcitonin on bone mass: review and analysis. A m J Clin Nutr 1998;67:18-24. 207. Haines CJ, Chung TKH, Leung PC, Hsu SY, Leung DHY. Calcium supplementation and bone mineral density in postmenopausal women using estrogen replacement therapy. Bone 1995;16:529-531. 208. Davis JW, Ross PD, Johnson NE, Wasnich RD. Estrogen and calcium supplement use among Japanese-American women: effects on bone loss when used singly and in combination. Bone 1995;17:369-373. 209. Heikkinen J, Kyllonen E, Kurttila-Matero E, et al. HRT and exercise: effects on bone density, muscle strength and lipid metabolism. A placebo controlled 2-year prospective trial on two estrogen-progestin regimens in healthy postmenopausal women. Maturitas 1997;26:139-149. 210. Lindsay R. Hormone replacement therapy for prevention and treatment of osteoporosis. Am J Med 1993;95:37S- 39S. 211. Barret-Connor E, Kritz-Silverstein D. Does hyperinsulinemia preserve bone? Diabetes Care 1996;19:1388-1392. 212. Parra-Cabrera S, Hernandez-Avila M, Tamayo-y-Orozco J, LopezCarrillo L, Meneses-Gonzalez E. Exercise and reproductive factors as predictors of bone density among osteoporotic women in Mexico City. Calcif Tissue Int 1996;59:89-94. 213. Heaney RE Nutrition and risk for osteoporosis. In: Marcus R, Feldman D, Kelsey J, eds. Osteoporosis. New York: Academic Press, 1996:483-509. 214. Saltman PD, Strause LG. The role of trace minerals in osteoporosis. JAm Call Nutr 1993;12:384-389. 215. Strause L, Saltman P. The role of manganese in bone metabolism. Washington, DC: American Chemical Society, 1987:45-55.
681 216. Andon M, Luhrsen K, Kanerva R, Chatzidakis C. Effects of dietary copper and manganese restriction on serum mineral concentrations and femoral shaft bone density in rats.JAm CollNutr 1992;11:600. 217. Strause L, Saltman P, Glowacki I. The effect of deficiencies of manganese and copper on resorption of bone particles in rats. ClacifTissue Int 1987;41:145-150. 218. Reginster JY, Strause LG, Saltman P, Franchimont E Trace elements and postmenopausal osteoporosis: a preliminary study of decreased serum manganese. Med Sci Res 1988;16:337-338. 219. Howard G, Andon M, Bracker M, Saltman P, Strause L. Serum trace mineral concentrations, dietary calcium intake and spine bone mineral density in postmenopausal women. J Trace Elem Med Bid 1992;5:23-31. 220. Strause L, Saltman P, Smith K, Andon M. Calcium, copper, manganese arid zinc supplementation sustains bone density in postmenopausal women. In: Burckhardt P, Heaney RP, eds. Nutritional aspects ofosteoporosis. New York: Raven Press, 1991:223-232. 221. Strause L, Saltman P, Smith KT, Bracker M, Andon ME. Spinal bone loss in postmenopausal women supplemented with calcium and trace minerals. J Nutr 1994;124:1060-1064. 222. Harris MM, Houtkooper LB, Stanford VA, et al. Dietary iron is associated with bone mineral density in healthy postmenopausal women. J Nutr 2003;133:3598-602. 223. Propckop DJ. Role of iron in the synthesis of collagen in connective tissue. Fed Proc 1971;30:984-990. 224. Deluca HE Metabolism of vitamin D: current status. Am J Clin Nutr 1976;29:1258-1270. 225. Medeiros DM, Ilich J, Ireton J, et al. Femurs from rats fed diets deficient in copper or iron have decreased mechanical strength and altered mineral composition. J Trace Elem Exp Med 1997;10:197-203. 226. Kipp D, Pinero D, Beard JL. Low bone mass and volume in irondeficient rats. FASEB J 1998;12:A508. [Abstract.] 227. Kipp D, Beard J, Lees C. Mild iron deficiency results in altered bone mass and histomorphometry in growing female rats. FASEBJ2002;16: A273. [Abstract.] 228. Schnitzer CM, Macphail AP, Shires R, et al. Osteoporosis in African hemosiderosis: role of alcohol and iron. J Bone Miner Res 1994;9: 1865-1873. 229. Diamond T, Stiel D, Posen S. Osteoporosis in hemochromatosis: iron excess, gonadal deficiency or other factors? Ann Intern Med 1989;110: 430-436. 230. Buffer RN. The geriatric patient. In: Usdin G, Lewis JM, eds. Psychiatry in generalmedicalpractice. New York: McGraw-Hill, 1979. 231. McKinlay JB, McKinlay SM, Brambilla D. The relative contributions of endocrine changes and social circumstances to depression in middleaged women. J Health Soc Behav 1987;28:345- 363. 232. Avis NE, Brambilla D, McKinlay SM, Vass K. A longitudinal analysis of the association between menopause and depression. Results from the Massachusetts Women's Health Study. Ann Epidemiol 1994;4: 214-220. 233. Hughes JR. Psychological effects of habitual aerobic exercise: a critical review. Prev Med 1984;13:66-78. 234. Folkins CH, Sime WE. Physical fitness training and mental health. Am Psycho11981;36:373-389. 235. Colt EWD, Wardlaw SL, Frantz AG. The effect of running on plasma beta-endorphin. Life Sci 1981;28:1637-1640. 236. Bravo G, Gauthier P, Roy PM, et al. Impact of a 12-month exercise program on the physical and psychological health of osteopenic women.JAm Geriatr Soc 1996;44:756-762. 237. Kaldas R, Hughes CL Jr. Reproductive and general metabolic effects of soy extracts in mammals. Reprod Toxico11989;25:1917-1925. 238. Nilsson A. Demethylation of the plant oestrogen biochanin A in the rat. Nature 1961;192:358.
682 239. Kelly GE, Nelson C, Waring MA, et al. Metabolites of dietary (soya) isoflavones in human urine. Clinica Chimica 1993;223:2-22. 240. Joannou GE, Kelly GE, Reeder AY, Waring M, Nelson C. A urinary profile study of dietary phytoestrogens. The identification and mode of metabolism of new sioflavonoids. J Steroid Biochem Mol Bioi 1995;54:167-184. 241. CornweUT, Cohick W, Raskin H. Dietary phytoestrogens and health. Phytochemistry 2004;65:995-1016. [Review.]
WARREN ET AL.
242. Kreijkamp-Kaspers S, Kok L, Bots M, Grobbee D, van der Schouw Y. Dietary phytoestrogens and plasma lipids in Dutch postmenopausal women: a cross-sectional study.Atherosclerosis 2005;178: 95-100. 243. Ingram D, Sanders K, Kolyabba M, Lopez D. Case control study of phytoestrogens and breast cancer. Lancet 1997;350:990-994.
-IAPTER 5(
Herbs, Phytoestrogens, and Other CAM Therapies ADRIANE FUGH-BERMAN
Department of Physiology and Biophysics, Georgetown University Medical Center, Washington, DC 20057
I. P H Y T O E S T R O G E N S
Complementary and alternative medicine (CAM), sometimes called unconventional, natural, integrative, traditional, or non-traditional medicine, is quite popular in the United States. An analysis of 2002 National Health Interview Survey (NHIS) data found that 62% of American adults used CAM in that year (1). In the Study of Women's Health Across the Nation (SWAN), almost half of midlife women had used CAM in the past year (2). An analysis of the 1999 NHIS found that one-third of American women used CAM in the previous year, with spiritual healing/prayer and herbal medicine being the most popular (3). A national survey that oversampled minority women found that 43% of 812 African-American women reported using religion/ spirituality in the past 12 months for health reasons (4). (It bears noting that estimates of use are higher when prayer is included as a CAM therapy; in the Barnes study, excluding prayer specifically for health dropped the percentage of CAM users from 62% to 36%.) The well-delineated risks of menopausal hormone therapy may increase the utilization of CAM therapies by consumers. It is important for physicians to familiarize themselves with complementary therapies in order to counsel patients effectively.
Many food plants contain phytoestrogens, or plant estrogens, primarily phenolic compounds that include isoflavones and lignans. Phytoestrogens can occupy estrogen receptors but are less than 1% as potent as endogenous estrogens (5). Soybeans (Glycine max) are rich in the conjugated isoflavones genistein and daidzin, but many other beans, including Anasasi, brown, black, navy, pinto, and turtle beans (all Phaseolus vulgaris), have as much or more genistin as soybeans (6). Gut bacteria convert conjugated isoflavones to the unconjugated (aglycone) active isoflavones, primarily genistein, daidzein, and equol. It is unclear whether aglycone or glucoside forms of isoflavones are more bioavailable. The isoflavones in cooked soybeans, textured vegetable protein (TVP), and soy milk contain 95% glycosides, whereas isoflavones in tofu (soybean curd) and tempeh (fermented soybean curd) contain 20% to 40% aglycones. It has been suggested that differences in gut bacteria could affect bioavailability of isoflavones. One study compared the bioavailability of daidzein and genistein in American women and found no difference in bioavailability (7). Lignan precursors are found in whole grains, seeds, fruits, and vegetables, especially flaxseed (linseed), rye, millet, and legumes. Gut bacteria convert plant lignans to mammalian lignans (enterolactone and enterodiol).
The author acknowledges Jenna Bythrow for contributing to this update. TREATMENT OF THE POSTMENOPAUSAL WOMAN
683
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
684
ADRIANE FUGH-BERMAN
A. Hot Flushes Evidence is mixed on whether phytoestrogen supplementation can help hot flushes. Only two (8,9) of nine controlled trials found a clear benefit of dietary soy for hot flushes; six studies were negative (10-15) and one was mixed (16). Phytoestrogen supplements have also been tested in eight trials, which have been evenly split between positive (17-20) and negative (21-24) results. A crossover study of pure genistein 90 mg/day in 100 postmenopausal women for 6 weeks found that the treatment significantly reduced hot flushes compared with placebo but that the effect was mild (25). Both dietary and supplemental phytoestrogens appear to be safe; however, long-term studies have not been performed with concentrated phytoestrogen supplements or purified isoflavones.
B. Vaginal Epithelium It is not clear how changes in vaginal epithelium correlate with discomfort during sex. Of four studies that looked at phytoestrogen supplemention and vaginal epithelium, two studies found an estrogenic effect and two did not. A study of 45 g soy flour daily in 25 women found significant improvements in the vaginal maturation index (26). The Brzezinski study (9) also found improvements in vaginal cytology. Negative studies include the Murkies study (12) and another study of soy foods 165 mg isoflavones/day in 97 women (91 completed)(27).
C. Breast Cancer Risk Until recently, dietary intake of phytoestrogens has been primarily determined by cultural differences in diet. Although breast cancer patients excrete lower amounts of phytoestrogens than other women do, and breast cancer risk is low in areas of the world where excretion of phytoestrogens is high, observational studies are insufficient evidence for benefit because other differences between populations exist. For example, Japanese women who eat traditional foods have lower breast cancer rates than those who eat a Western diet (28); however, phytoestrogens are not the only difference in these two diets. Among other factors, a traditional diet is lower in meat and fat and higher in fiber and vegetable intake, including seaweed. Although breast cancer rates in Asia are lower at all ages than breast cancer rates in the West, it is unclear to what extent phytoestrogens contribute to this disparity. No randomized controlled trials have tested the effect of phytoestrogens on breast cancer risk. The following are observational studies, which can never prove benefit.
A case control study among Chinese, Japanese, and Filipino women in Los Angeles County (501 breast cancer patients and 594 controls) found that women who reported soy intake at least once weekly during adolescence had a reduced risk of breast cancer. There was also a significant trend of decreasing risk with increasing soy intake during adult life (29). Phytoestrogen intake among non-Asian women had no effect on breast cancer risk in a case-control study of almost 3000 African-American, Latina, and Caucasian women aged 35 to 79 years (30). A case-control study of 117 postmenopausal women in Shanghai found that an inverse association between urinary phytoestrogen excretion and breast cancer risk was more evident among women with a high body mass index or waist-to-hip ratio; the inverse association was also more pronounced among women with high blood concentrations of estradiol or low levels of estrone sulfate or sex hormone binding globulin (SHBG) (31). An earlier study of Chinese women in Singapore found that soy product intake was associated with lower rates of breast cancer in premenopausal, but not postmenopausal, women (32). There is no evidence that merely adding phytoestrogens to a Western diet would decrease breast cancer risk. Dosage, formulation, duration, and timing may all have an effect. It has been theorized, for example, that phytoestrogens may affect terminal end bud differentiation and that the most important period for phytoestrogen intake is around puberty. This effect would not be possible to extract from epidemiologic data on lifelong consumption of soy foods. On the other hand, there is no evidence that soy foods increase risk of breast cancer or any other disease. Soy foods have been consumed for thousands of years and are not dangerous. However, concentrated phytoestrogen products and purified isoflavones are now readily available, and there are no long-term safety data on these food-free phytoestrogens.
D. Endometrial Cancer Soy foods do not increase endometrial cancer rates. A case-control study of 332 endometrial cancer cases and 511 controls among a multiethnic population in Hawaii found that high soy intake appears to protect against endometrial cancer in both premenopausal and postmenopausal women (33). A study of soy cereal containing 82 mg isoflavones/day for 6 months caused no changes in endometrial biopsies (34). Although isoflavones may have a selective estrogen receptor modulator (SERM) effect, it is not enough to counteract estrogen; soy protein isolate (containing 120 mg aglycone isoflavones) over 6 months did not prevent estradiol-induced endometrial hyperplasia (35).
685
CHAPTER 50 Herbs, Phytoestrogens, and Other CAM Therapies
E. Cardiovascular Disease Risk Factors Soybeans may have a beneficial effect on cardiovascular disease risk factors. In a meta-analysis of 38 controlled clinical trials on the effect of soy intake on serum lipids, ingestion of an average of 47 grams of soy protein daily was associated with reduced cholesterol (23.2 mg/dL, or 9.3%), reduced low-density-lipoprotein (LDL) cholesterol (21.7 mg/dL, or 12.9%), and reduced triglycerides (13.3 mg/dL, or 10.5%) (36). High-density-lipoprotein (HDL) cholesterol was unaffected. In most studies, intake of fat, saturated fat, and cholesterol was similar between the control group and soy group, thus lipid changes cannot be attributed to the lower fat content in soy products compared with animal products.
E Osteoporosis In a double-blind trial of 66 postmenopausal, hypercholesterolemic women aged 49 to 73 (37), subjects were randomly assigned to 40 g protein per day from one of three sources: milk (nonfat dried milk and casein), soy protein with medium isoflavone content (equivalent to 55.6 mg isoflavones daily), or isolated soy protein with high isoflavone content (equivalent to 90 mg isoflavones daily). All women also followed a low-fat, low-cholesterol diet. Dualenergy x-ray absorptiometry (DXA) bone density studies of the lumbar spine, proximal femur, and total body were done at the beginning of the study (after a 2-week lead-in) and at the end of the 6-month study. No differences were seen among the three groups in bone density studies of the hip or total body, but subjects receiving the high-isoflavone preparation experienced a significant increase in lumbar bone density and mineral content (2%) compared with the milk protein group.
G. Risks Although consumption of subterranean clover (Trifolium reDens) and other phytoestrogen-rich forage or feed has caused lower conception rates in several animal species, the implicated plant species are not consumed in the human diet. Soy foods, on the other hand, are consumed in large quantities in China, Japan, and Korea and have not been associated with fertility problems or other adverse effects. In North America, soy infant formula has been used for more than 30 years; formula contains primarily the glycosides genistin and daidzin. A retrospective cohort study of adults aged 20 to 34 fed soy milk or cow milk as infants in a controlled feeding study found no significant differences between groups in pubertal maturation, reproductive history,
menstrual history, hormonal disorders, height, weight, or current health. Women fed soy as infants reported slightly longer duration of menstrual bleeding and greater discomfort with menstruation (38). Soybeans are quite high in oxalates, containing 0.67 to 3.5 grams oxalates per 100 grams of dry weight. Commercial soy foods contain between 16 and 638 mg oxalates per serving, comparable to peanut butter, reffied beans, and lentils (39). The oxalate content of soy foods may be of concern in patients who are avoiding oxalates because of nephrolithiasis (most kidney stones are composed of calcium oxalate). Some women with vulvodynia also choose to minimize dietary oxalates, which may aggravate symptoms. There is much misinformation on the Internet about the putative adverse effects of soy on the thyroid. Genestin does inactivate thyroid peroxidase, and soy consumption may be linked to thyroid dysfunction if iodine is deficient. Iodine deficiency is a serious problem in many developing countries, but it is rare in North America due to iodine fortification of table salt. Soy supplementation does not appear to affect thyroid function in most iodine-replete women (40). It is possible, however, that soy interferes with the absorption of thyroid medication. In one case, a 45-year-old woman with hypothyroidism who required high doses of levothyroxine was able to reduce her dose by separating ingestion of a soy drink from ingestion of levothyroxine (41). Although consumption of soy or other beans is probably benign, the safety of concentrated extracts or high doses of purified isoflavones has not been established. This is an issue because pure genistein or isoflavone mixtures are available at heath food stores.
II. HERBS Ginseng (Panax ginseng), chaste-tree berry (Vitex agnus castus), Dong quai (Angelica sinensis), black cohosh (Actaea racemosa), and licorice (Glycyrrbiza glabra) are a few of the herbs commonly used to treat hot flushes and other symptoms associated with menopause. However, there are few clinical studies on these therapies, and little information is available on the long-term effects of medicinal herbs.
A. Black Cohosh
(Actaea racemosa)
Black cohosh is the most studied herb for hot flashes; at least nine controlled studies have been performed. The most recent, methodologically sound placebo-controlled studies have been largely negative. Studies performed in the United States have been largely negative, while those performed in
686
ADRIANE FUGH-BERMAN
Germany have been more positive. An NIH-funded, yearlong, placebo- and treatment-controlled trial found no benefit of black cohosh (160 mg/day 2.5% triterpene glycosides, 70% ethanol extract) or a multibotanical preparation containing 200 mg black cohosh over placebo; conjugated equine estrogens 0.625 mg/day was effective (42). A crossover study in 132 women found no benefit of Remifemin (a standardized product that has changed formulation over time) 20 mg twice daily over placebo for four weeks (43). Another U.S. study found that Remifemin (40 mg/day 3< 2 months) had no effect over placebo for hot flashes in 85 breast cancer survivors (59 taking tamoxifen) (44). A German study in 304 women found a benefit of black cohosh over placebo on several subscores, including hot flashes, of the Menopause Rating Scale 1 (45). Recent treatment-controlled trials comparing black cohosh treatments to estrogens have been negative. An Italian study found no difference between Remifemin 40 mg extract/day and a transdermal estradiol patch (25 mcg/week plus dihydrogesterone 10 mg/day for 12 days over the three-month trial) (46). A randomized controlled trial found no significant effect of the standardized black cohosh preparation BNO1055 (Klimadynon/Menofem) over conjugated estrogens or placebo on hot flashes in 62 women, although a menopausal symptom score improved in women who completed the trial, and both conjugated estrogens and black cohosh improved vaginal epithelium (47). Older studies have found previous formulations of Remifemin equivalent to estrogens in reducing symptoms (48-50). The effects of long-term use of black cohosh are unknown. No clinical trials of black cohosh have been longer than six months in duration. Recently, several cases of hepatitis have been linked to black cohosh (51-53).
B. D o n g Q v a i
(Angelica sinensis)
Dong quai was tested in a randomized, double-blind, placebo-controlled clinical trial in 71 women. 4.5 grams of dong quai root/day 3< 6 months was not superior to placebo for hot flashes (54). Traditionally, dong quai is almost always used as part of a mixture. Dong quai can cause bleeding when administered concurrently with warfarin (55); the furocoumarins it contains can cause photosensitization.
C. E v e n i n g Primrose
(Oenethera biennis)
Evening primrose oil, a good source of linoleic and gamma linolenic acid, has been evaluated in a double-blind controlled trial of 56 women and found to be no more effective than placebo for hot flashes (56).
D. Kava
(Piper metbysticum)
Widely used in Polynesia, rhizomes of the kava shrub, a psychoactive member of the pepper family, are used medicinaUy for anxiety and insomnia. Active constituents in kava are the kavapyrones (also called kavalactones).Two doubleblind studies of kava for climacteric symptoms have been performed by the same investigator. A study in 40 women using doses of 30 to 60 mg/day for 56 to 84 days found significant improvements in the Kupperman index (57). A trial utilizing a higher dose (equivalent to 210 kavapyrones/ day) in 40 menopausal women also found improvements in menopausal symptoms (Kupperman index), and mood (assessed with the Hamilton Anxiety Scale [HAMA] and the Depression Status Inventory) (58). Extremely heavy, chronic recreational use of kava results in yellowing of the skin and an ichthyosiform eruption known as kava dermopathy,sometimes accompanied by eye irritation (59). Kava should not be recommended because of numerous case reports of hepatotoxicity (60,61). E. G i n s e n g
quinquefolius,
(Panax ginseng, Panax or Panax notoginseng)
The most common types of ginseng are Chinese ginseng (Panaxginseng), American ginseng (Panax quinquefolius), and Siberian "ginseng" (Eleutherococcussenticosus), which is not ginseng at all, although it is in the same family. Ginseng contains terpenoids, especially a group of compounds called ginsenosides. One randomized, doubleblind, placebo-controlled clinical trial in 384 menopausal women found no effect of ginseng (Panaxginseng; in this case Ginsana, containing 100 mg standardized extract Gl15 x 14 weeks) on hot flushes, endometrial thickness, or the vaginal maturation index (62). Although case reports have associated ginseng with postmenopausal bleeding (63,64), implicated products were not examined for hormone adulterants.
E Licorice
(Glycyrrhiza glabra)
Licorice is the most common herb in Chinese medicine products. In Chinese medicine, licorice is always used as part of a mixture, and the synergistic effects of mixtures as well as perhaps dose limitations may prevent problems. It bears noting that not all of the reported cases of licorice-induced problems were from licorice-containing candies, gum, laxatives, or chewing tobacco, and not from the use of licorice as herbal medicine (most "licorice" candies manufactured in the United States are actually flavored with anise; imported candies usually contain real licorice). However, licorice tinctures and extracts, capsules, lozenges, and such are available
CHAPTER50 Herbs, Phytoestrogens, and Other CAM Therapies in the United States, so it is useful to know about possible side effects. Glycyrrhizinic acid and its derivatives affect the metabolism of cortisol, apparently by inhibiting the 11 betahydroxysteroid dehydrogenase system that converts cortisol to cortisone (65). Large chronic doses may result in a pseudoprimary aldosteronism with symptoms that may include edema, hypertension, and hypokalemia (66,67). Cardiac arrhythmias and cardiac arrest, including two deaths, have occurred in users of licorice products. Cardiomyopathy has been reported (68) and pulmonary edema (69).
G. Sage
(Salvia officinalis)
Sage is reputed to help hot flushes and night sweats but should not be recommended because it contains thujone, a neurotoxin. Long-term use can cause seizures or other neurologic symptoms (70). Although thujone is inactivated by heat, sage products may not be heat-treated.
H. Vitex or Chaste-Tree Berry (Vitex agnus-castus)
687
j. Chinese Herbs A randomized, double-blind, placebo-controlled clinical trial in 78 menopausal women found no benefit of a Chinese herb mixture for 3 months (75). Chinese herbs are usually individualized; in this study all subjects received the same mixture.
K. Yam and Progesterone Creams A double-blind, placebo-controlled, 3-month crossover trial in 23 menopausal women with hot flushes found no benefit of wild yam (Dioscorea villosa) topical cream over placebo (76) on hot flushes or night sweats. Wild yam contains diosgenin, a precursor to progesterone; however, this conversion is not known to occur endogenously. Topical progesterone cream (20 mg progesterone/day) reduced hot flushes in a 1-year study of 102 healthy postmenopausal women with a primary endpoint of bone mineral density (77). Another study, however, found no effect (78).
L. Bioidentical Hormones Vitex contains flavonoids and an alkaloid called viticin. There have been no clinical studies on vitex for menopausal symptoms, but it is commonly used for irregular or heavy menstrual bleeding. Vitex can cause an acneiform rash but has not been linked to serious adverse effects.
I. Red Clover
(Trifolium pratense)
Red Clover contains the isoflavones formononetin, biochanin A, daidzein, and genistein. Red clover is currently being marketed as a long-term phytoestrogen source and natural hormone therapy. The largest randomized controlled trial to date, the Isoflavone Clover Extract study, found no advantage of Promensil or another red clover preparation, Rimostil, on hot flashes in 252 symptomatic women over three months (71). Two 3-month randomized, double-blind, placebocontrolled clinical trials of Promensil (containing 40 mg total isoflavones) conducted in Australia (one with 37 postmenopausal women, one with 51 postmenopausal women) reported no significant benefit of red clover extract for hot flushes (72,73). A randomized, double-blind, placebocontrolled trial in 30 menopausal women reported that Promensil 80 mg/day • 12 weeks reduced hot flushes and improved Greene scores, based on the difference in proportion of patients above and below the mean at 8 and 12 weeks (74). Only median percentage changes in hot flushes were provided.
Bioidentical or "natural" hormones, including estriol, estradiol, estrone, and progesterone, are being promoted to consumers as safe alternatives to conventional menopausal hormone therapy. No data support the claim that bioidentical hormones are safer than other hormones (79). Commercial or compounded estrogen preparations may be effective for treating hot flushes or vaginal dryness. If estriol is used to treat menopausal symptoms, it should be opposed with an oral progestin. Although some progesterone creams may provide some endometrial protection in some women, studies have been inconsistent and not reassuring. Serum levels of progesterone after application of transdermal creams are insufficient to prevent estrogenic stimulation of the endometrium, and it is inappropriate to prescribe topical progestin cream as the progestin component of hormone therapy. Any menopausal hormone therapy should be reserved for women with bothersome symptoms and used in the lowest effective dose for as brief a period as possible.
M. Vitamin E Several poorly controlled trials of vitamin E as a menopause treatment were done as early as the 1940s. In 1953, a double-blind, 3-year study compared vitamin E (50 to 100 mg/day) with two estrogen preparations, phenobarbital, and a placebo in 658 women (80). On an 11-symptom menopause index (hot flushes were not analyzed separately,
688
ADRIANE FUGH-BERMAN
vitamin E was no more effective than placebo. A recent randomized, placebo-controlled crossover study tested 400 IU vitamin E succinate twice daily on 125 breast cancer survivors with hot flushes. After 4 weeks, vitamin E was statisticaUy superior to placebo, but the difference between phases (one hot flush a day) was not considered clinically significant; subjects did not prefer vitamin E to placebo. Vitamin E is benign. It is a common misperception among clinicians that vitamin E may cause bleeding, but studies designed to assess bleeding risk have found no effect of vitamin E (81), even in those on warfarin (82).
N. A c u p u n c t u r e A study of acupuncture for hot flushes randomized 24 menopausal women either to electroacupuncture (electrical stimulation of acupuncture needles) at standardized points or to control (shallow acupuncture needle insertion) at the same points (83). Both groups improved, with no difference between groups. Shallow needle insertion at correct acupuncture points should have some effect, so the chosen control may have been suboptimal. Acupuncture occasionally causes tissue trauma and rarely, serious complications, including pneumothorax and cardiac tamponade. The most common complication has been hepatitis or other infections from inadequately sterilized needles (84). Disposable needles, the standard of care in the United States, obviate this
be other differences between exercisers and the general population. Some women report that exercise triggers hot flushes.
III. SUMMARY In summary, most CAM therapies for menopausal symptoms are benign. Because CAM therapies are widely used, studies should be done to delineate both benefits and risks. The current evidence on the use of phytoestrogens and herbs for hot flushes and other menopausal symptoms is unimpressive; most studies of herbs found them no better than placebo. Hot flushes are notoriously placebo-responsive. Among herbs tested for hot flushes, limited evidence supports a beneficial effect only for black cohosh (and long-term safety questions remain). Red clover has been shown ineffective in two of three trials. Single trials show no effect of dong quai, wild yam, evening primrose oil, ginseng, and a Chinese herb mixture. Although some trials of herbs were quite small and may have been underpowered, it is clear that the herbs tested to date lack a dramatic effect. Behavioral techniques, including relaxation and biofeedback, may be helpful for some women. Because of the wellestablished risks of hormone therapy, interest in CAM therapies for menopausal symptoms may increase. A variety of complementary therapies deserve further research.
concern.
References O. Paced Respiration and Relaxation A 4-month study compared paced respiration (slow, deep breathing) with progressive muscle relaxation or alpha electroencephalogram (EEG) biofeedback (control) in 33 postmenopausal women with hot flushes (85). Only paced respiration training significantly reduced hot flush frequency. In another trial by the same investigators, only paced respiration decreased hot flushes significantly in 24 menopausal women randomized to either paced respiration or control biofeedback (86). A randomized, controlled, 7-week study in 45 women with hot flushes compared 20 minutes daily of trained relaxation, reading, or symptom-charting (87). Hot flush frequency did not improve in any group; hot flush intensity decreased significantly only in the relaxation group. No adverse effects of paced respiration or relaxation have been reported.
E Exercise Although women who exercise regularly appear to have fewer hot flushes than women in the general population (88), this cannot be considered proof of benefit, as there may
1. Barnes PM, Powell-Griner E, McFann K, Nahin RL. Complementary and alternative medicine use among adults. United States, 2002. Adv Data 2004;1-19. 2. Bair YA, Gold EB, Azari RA, et al. Use of conventional and complementary health care during the transition to menopause: longitudinal results from the Study of Women's Health Across the Nation (SWAN). Menopause 2005;12:31- 39. 3. Upchurch DM, Chyu L. Use of complementary and alternative medicine among American women. WomensHealth Issues 2005;15:5-13. 4. Dessio W, Wade C, Chao M, et al. Religion, spirituality, and healthcare choices of African-American women: results of a national survey. Ethn Dis 2004;14:189-197. 5. Ziegler J. Soybeans show promise in cancer prevention. JNCI 1994; 86:1666-1667. 6. Kaufman PB, Duke JA, Brielmann H, Boik J, Hoyt JE. A comparative survey of leguminous plants as sources of the isoflavones genistein and daidzein: implications for human nutrition and health. JAlt Complement Med 1997;31:7-12. 7. Zubik L, Meydani M. Bioavailability of soybean isoflavones from aglycone and glucoside forms in American women. Am J Clin Nutr 2003;77:1459-1465. 8. Albertazzi P, Pansini F, Bonaccorsi G, et al. The effect of dietary soy supplementation on hot flushes. Ob Gyn 1998;91:6-11. 9. Brzezinski A, Adlercreutz H, Shaoul R, et al. Short-term effects of phytoestrogen-rich diet on postmenopausal women. Menopause 1997;4: 89-94.
CHAPTER 50 Herbs, Phytoestrogens, and Other C A M Therapies 10. Burke GL, Legault C, Anthony M, et al. Soy protein and isoflavone effects on vasomotor symptoms in peri- and postmenopausal women: the Soy Estrogen Alternative Study. Menopause 2003;10:147-153. 11. Van Patten CL, Olivotto IA, Chambers GK, et al. Effect of soy phytoestrogens on hot flashes in postmenopausal women with breast cancer: a randomized, controlled clinical trial. J Clin Onco12002;20:1449-1455. 12. Murkies AL, Lombard C, Strauss BJG. Dietary flour supplementation decreases postmenopausal hot flushes: effect of soy and wheat. Maturitas 1995;21:189-195. 13. St Germain A, Peterson CT, Robinson JG, Alekel DL. Isoflavone-rich or isoflavone-poor soy protein does not reduce menopausal symptoms during 24 weeks of treatment. Menopause 2001;8:17-26. 14. Kotsopoulos D, Dalais FS, Liang YL, McGrath BP, Teede HJ. The effects of soy protein containing phytoestrogens on menopausal symptoms in postmenopausal women. Climacteric2000;3:161 - 167. 15. Dalais FS, Rice GE, Wahlquist ML, et al. Effects of dietary phytoestrogens in postmenopausal women. Climacteric 1998;1:124-129. 16. Washburn S, Burke GL, Morgan T, Anthony M. Effect of soy protein supplementation on serum lipoproteins, blood pressure, and menopausal symptoms in perimenopausal women. Menopause 1999;6: 7-13. 17. Faure ED, Chantre P, Mares E Effects of a standardized soy extract on hot flushes: a multicenter, double-blind, randomized, placebo-controlled study. Menopause 2002;9:329-334. 18. Han KK, Soares JM Jr, Haidar MA, de Lima GR, Baracat EC. Benefits of soy isoflavone therapeutic regimen on menopausal symptoms. Obstet Gyneco12002;99:389-394. 19. Scambia G, Mango D, Signorile PG, et al. Clinical effects of a standardized soy extract in postmenopausal women: a pilot study. Menopause 2000;7:105-111. 20. Upmalis DH, Lobo R, Bradley L, et al. Vasomotor symptom relief by soy isoflavone extract tablets in postmenopausal women: a multicenter, double-blind, randomized, placebo-controlled study. Menopause2000; 7: 236-242. 21. Q.uella SK, Loprinzi CL, Barton DL, et al. Evaluation of soy phytoestrogens for the treatment of hot fashes in breast cancer survivors: a North Central Cancer Treatment Group Trial. J Clin Oncol 2000; 18:1068-1074. 22. Penotti M, Fabio E, Modena AB, et al. Effect of soy-derived isoflavones on hot flushes, endometrial thickness, and the pulsatility index of the uterine and cerebral arteries. Fertil Steril 2003;79: 1112-1117. 23. Nikander E, Killddnen A, Metsa-Heikkila M, et al. A randomized placebo-controlled crossover trial with phytoestrogens in treatment of menopause in breast cancer patients. Obstet Gynecol 2003;101: 1213-1220. 24. Secreto G, Chiechi LM, Amadori A, et al. Soy isoflavones and melatonin for the relief of climacteric symptoms: a multicenter, doubleblind, randomized study. Maturitas 2004;47:11-20. 25. Albertazzi P, Steel SA, Bottazzi M. Effect of pure genistein on bone markers and hot flushes. Climacteric2005;8:371-379. 26. Wilcox G, Wahlqvist ML, Burger HG, Medley G. Oestrogenic effects of plant foods in postmenopausal women. BMJ 1990;301:905-906. 27. Baird DD, Umbach DM, Lansdell L, et al. Dietary intervention study to assess estrogenicity of dietary soy among postmenopausal women. J Clin Endocrinol Metab 1995;80:685-690. 28. Adlercreutz H, Honjo H, Higashi A, et al. Urinary excretion of lignans and isoflavonoid phytoestrogens in Japanese men and women consuming a traditional Japanese diet. Am J Clin Nutr 1991 ;54:1093-1100. 29. Wu AH, Wan P, Hankin J, et al. Adolescent and adult soy intake and risk of breast cancer in Asian-Americans. Carcinogenesis2002;23:1491 - 1496. 30. Horn-Ross PL, John EM, Lee M, et al. Phytoestrogen consumption and breast cancer risk in a multiethnic population: the Bay Area Breast Cancer Study. dmJEpidemio12001;154:434-441.
689 31. Dai Q_, Franke AA, Yu H, et al. Urinary phytoestrogen excretion and breast cancer risk: evaluating potential effect modifiers endogenous estrogens and anthropometrics. Cancer Epidemiol Biomarkers Prev 2003;12:497-502. 32. Lee HP, Gourley L, Duffy SW, et al. Dietary effects on breast cancer risk in Singapore. Lancet 1991;337:1197-1200. 33. Goodman MT, Wilkems LR, Hankin JH, et al. Association of soy and fiber consumption with the risk of endometrial cancer, d m J E p i 1997; 146:294-306. 34. Balk JL, Whiteside DA, Naus G, DeFerrari E, Roberts JM. A pilot study of the effects of phytoestrogen supplementation on postmenopausal endometrium. J Soc GynecolInvestig 2002;9:238-242. 35. Murray MJ, Meyer WR, Lessey BA, et al. Soy protein isolate with isoflavones does not prevent estradiol-induced endometrial hyperplasia in postmenopausal women: a pilot trial. Menopause 2003;10:456-464. 36. Anderson JW, Johnstone BM, Cook-Newell ME. Meta-analysis of the effects of soy protein intake on serum lipids. N EnglJ Med 1995; 33:276-282. 37. Potter SM, Baum JA, Teng H, et al. Soy protein and isoflavones: their effects on blood lipids and bone density in postmenopausal women.Am J Clin Nutr 1998;68:1375S- 1379S. 38. Strom BL, Schinnar R, Ziegler EE, et al. Exposure to soy-based formula in infancy and endocrinological and reproductive outcomes in young adulthood. JAm MedAssoc 2001;286:2402-2403. 39. Massey LK, Palmer RG, Homer HTJ. Oxalate content of soybean seeds (Glycine max: Leguminosae), soyfoods, and other edible legumes. Agric Food Chem 2001;49:4262-4266. 40. Messina M, Redmond G. Effects of soy protein and soybean isoflavones on thyroid function in healthy adults and hypothyroid patients: a review of the relevant literature. Thyroid 2006;16:249-258. 41. Bell DS, Ovalle E Use of soy protein supplement and resultant need for increased dose of levothyroxine. Endocr Pract 2001;7:193-194. 42. Newton KM, Reed SD, LaCroix AZ, et al. Treatment of vasomotor symptoms of menopause with black cohosh, multibotanicals, soy, hormone therapy, or placebo: a randomized trial. Ann Intern IVied 2006;145:869-879. 43. Pockaj BA, Gallagher JG, Loprinzi CL, et al. Phase III double-blind, randomized, placebo-controlled crossover trial of black cohosh in the management of hot flashes: NCCTG Trial N01CC1. J Clin Oncol 2006 ;24:2836 - 2841. 44. Jacobson JS, Troxel AB, Evans J, et al. Randomized trial of black cohosh for the treatment of hot flashes among women with a history of breast cancer. J Clin Onco12001;19:2739-2745. 45. Osmers R, Friede M, Liske E, et al. Efficacy and safety of isopropanolic black cohosh extract for climacteric symptoms. Obstet Gynecol 2005;105:1074-1083. 46. Nappi R, Malavasi B, Brundu B, Facchinetti E Efficacy of Cimicifuga racemosa on climacteric complaints: a randomized study versus low dose transdermal estradiol. GynecolEndocrino 2005;20:30-35. 47. Wuttke W, Sedlova-Wuttke D, Gorkow C. The Cimicifuga preparation BNO 1055 vs. conjugated estrogens in a double-blind placebocontrolled study: effects on menopause symptoms and bone markers. Maturitas 2003;$67- $77. 48. Warnecke G. Beeinflussung ldimakterischer Beschwerden durch ein Phytotherapeutikum: Erfolgreiche therapie mit Cim#ifuga-Monoextrakt (Influence of phytotherapy on menopausal syndrome: successful treatments with monoextract of cimicifuga). Medizinische Welt 1985;36: 871-874. 49. Stoll W. Phytotherapeutikum beeinflusst atrophisches Vaginalepithel: Doppelblindversuch Cimicifuga vs. Ostrogenpr~iparat (Phytotherapy influences atrophic vaginal epithelium--double-blind studymCimicifuga vs. estrogenic substances). Therapeutikon 1987;1:23 - 31.
690 50. Lehmann-Willenbrock E, Riedel H. Klinische und endokrinologische Untersuchengen zur Therapie ovarieller Ausfallserscheinungen nach Hysterektomie unter Belassung der Adnexe (Clinical and endocrinological examinations concerning therapy of climacteric symptoms following hysterectomy with remaining ovaries). Zentralblatt fur Gynakologie 1988;110:611-618. 51. Lynch CR, Folkers ME, Hutson WR. Fulminant hepatic failure associated with the use of black cohosh: a case report. Liver Transpl 2006;12:989- 992. 52. Cohen S, O'Connor AM, Hart J, Merel NH, Te Hs. Autoimmune hepatitis associated with the use of black cohosh: a case study. Menopause 2004;11:575-577. 53. Whiting P, Clouston A, Kerlin P. Black Cohosh and other herbal remedies associated with acute hepatitis. MJA 2002;177:440-443. 54. Hirata JD, Swiersz LM, Zell B, Small R, Ettinger B. Does dong quai have estrogenic effects in postmenopausal women? A double-blind, placebo-controlled trial. Fertil Steri11997;68:981-986. 55. Fugh-Berman A. Herb-drug interactions. Lancet 2000;355:134-138. 56. Chenoy R, Hussain S, Tayob Y, et al. Effect of oral gamolenic acid from evening primrose oil on menopausal flushing. BMJ 1994;308:501-503. 57. Warnecke G, Pfaender H, Gerster G, Gracza E. Wirksamkeit yon Kava- Kava- Extrakt beim klimakterischen Syndrom. Z Phytother 1990; 11:81-86. 58. Warnecke G. Psychosomatic dysfunctions in the female climacteric. Clinical effectiveness and tolerance of Kava Extract WS 1490]. Fortschr Med 1991;109(4):119-122. 59. Norton SA, Ruze P. Kava dermopathy. J Am Acad Derm 1994;31: 89-97. 60. Tickel F, Baumuller HM, Seitz K, et al. Hepatitis induced by Kava (Piper methysticum rhizoma). J Hepato12003;39:62-67. 61. Humberston CL, Akhtar J, Krenzelok EP. Acute hepatitis induced by kava kava. J Toxicol Clin Toxico12003;41:109 - 113. 62. Wiklund IK, Mattsson L-A, Lindgren R, Limoni C. Effects of a standardized ginseng extract on quality of life and physiological parameters in symptomatic postmenopausal women: a double-blind, placebocontrolled trial. IntJ Clin Pharm Res 1999;XIX:89-99. 63. Greenspan EM. Ginseng and vaginal bleeding. J Am Med Assoc 1983;249:2018. 64. Hopkins MO, Androff L, Benninghoff AS. Ginseng face cream and unexplained vaginal bleeding. Am J Ob Gyn 1988;159:1121-1122. 65. Farese RV, Biglieri EG, Shackleton CHL, et al. Licorice-induced hypermineralocorticoidism. N EnglJ Med 1991;325:1223-1227. 66. Epstein MT, Espiner EA, Donald RA, Hughes H. Effect of eating liquorice on the renin-angiotensin aldosterone axis in normal subjects. BMJ 1977;1:488-490. 67. Chandler RF. Glycyrrhiza glabra. In De Smet PAGM, Keller K, Hansel R, Chandler RF. Adverse effectsof herbaldrugs, vol 1. Berlin:Springer-Verlag, 1992. 68. Shintani S, Murase H, Tsukagoshi H, Shiigai T. Glycyrrhizin (Licorice)induced hypokalemic myopathy. Eur Neuro11992;32:44-51. 69. Chamberlain JJ, Abolnik IZ. Pulmonary edema following a licorice binge. WesternJ Med 1997;167:184-185. 70. Wichtl M. Herbal drugs andphytopharmaceuticals. Stuttgart: Medpharm Scientific Publishers, 1994.
ADRIANE FUGH-BERMAN 71. Tice JA, Ettinger B, Ensrud K, et al. Phytoestrogen supplements for the treatment of hot flashes: the Isoflavone Clover Extract (ICE) Study: a randomized controlled trial. JANM 2003;290:207-214. 72. Barber RJ, Templeman C, Morton T, Kelly GE, West L. Randomized placebo-controlled trial of an isoflavone supplement and menopausal symptoms in women. Climacteric 1999;2:85- 92. 73. Knight DC, Howes JB, Eden JA. The effect of Promensil, an isoflavone extract, on menopausal symptoms. Climacteric 1999;2:79-84. 74. van de Weijer PH, Barentsen R. lsoflavones from red clover (Promensil) significantly reduce menopausal hot flush symptoms compared with placebo. Maturitas 2002;42:187-193. 75. Davis SR, Briganti EM, Chen RQ~ et al. The effects of Chinese medicinal herbs on postmenopausal vasomotor symptoms of Australian women. A randomised controlled trial. MedJAust 2001;174:68-71. 76. Komesaroff PA, Black CV, Cable V, Sudhir K. Effects of wild yam extract on menopausal symptoms, lipids and sex hormones in healthy menopausal women. Climacteric2001;4:144-150. 77. Leonetti HB, Longo S, Anasti JN. Transdermal progesterone cream for vasomotor symptoms and postmenopausal bone loss. Obstet Gynecol 1999;94:225-228. 78. Wren BG, Champion SM, Willetts K, Manga RZ, Eden JA. Transdermal progesterone and its effect on vasomotor symptoms, blood lipid levels, bone metabolic markers, moods, and quality of life for postmenopausal women. Menopause 2003;10:13-18. 79. Fugh-Berman A, Bythrow J. Bioidentical hormones for menopausal therapy: variation on a theme. J Gen Int Med 2007; published online March 7, 2007. http://www.springerlink.com/content/ q4772464371784g2/. 80. Blatt MHG, Wiesbader H, Kupperman HS. Vitamin E and climacteric syndrome. Arch Intern Med 1953;91:792-796. 81. Meydani SN, Meydani M, Blumberg JB, et al. Assessment of the safety of supplementation with different amounts of vitamin E in healthy older adults. Am J Clin Nutr 1998;68:311 - 318. 82. Kim JM, White RH. Effect of vitamin E on the anticoagulant response to warfarin. Am J Cardio11996;77:545-546. 83. Wyon Y, Lindgren R, Lundeberg T, Hammar M. Effects of acupuncture on climacteric vasomotor symptoms, quality of fife, and urinary excretion of neuropeptides among postmenopausal women. Menopause 1995;2:3-12. 84. Ernst E, White A. Life-threatening adverse reactions after acupuncture? A systematic review. Pain 1997;71:123-126. 85. Freedman RR, Woodward S. Behavioral treatment of menopausal hot flashes:evaluation by ambulatory monitoring. Am J Obstet Gynecol 1992;167:436-439. 86. Freedman RR, Woodward S, Brown B, Javaid JI, Pandey GN. Biochemical and thermoregulatory effects of behavioral treatment for menopausal hot flashes. Menopause 1995;2:211-218. 87. Irvin JH, Domar AD, Clark C, Zuttermeister PC, Friedman R. The effects of relaxation response training on menopausal symptoms. J Psychosom Obstet Gynaeco11996;17:202-207. 88. Hammar M, Berg G, Lindgren R. Does physical exercise influence the frequency of postmenopausal hot flushes? Acta Obstet Gynecol Scand 1990;69:409-412.
SECTION XI
Urinary Symptoms and Pelvic Support The total population of older women in the world is increasing. It has been estimated that there will be approximately 38 million women over the age of 55 in the United States by 2010. With aging, and particularly after menopause, a larger percentage of women will experience problems of pelvic support. These symptoms include a variety of urinary complaints, uterine and vaginal prolapse (including cystocele, rectocele, and enterocele), and rectal and anal problems. A recent survey of more than 4000 women in Southern California reported that the prevalence of urinary incontinence and overactive bladder was 28%, and 7% of women had uterine prolapse (1). Thirty-seven percent of women had various combinations of pelvic support problems. In an Australian survey the prevalence of urinary problems was 35.3% (2). These numbers are similar to a survey of women attending a gynecology clinic in the United Kingdom, where the prevalence of urinary problems was estimated to be 26.8%, and 8.4% of women had mixed urinary and anal incontinence (3). Thus, many millions of women worldwide suffer from pelvic support problems. In order to address some of the needs of this growing problem, a subspecialty of obstetrics and gynecology was developed to establish an advanced training fellowship program directed by the American Board of Obstetrics and Gynecology. The American Board of Urology has a similar program and, indeed, these two specialties may have a joint fellowship program. This section, therefore, attempts to provide the reader with up-to-date information on pelvic floor problems in postmenopausal women. There is a little overlap in the chapter of urinary incontinence in this section and an earlier chapter by Goran Samsioe (Chapter 18) where the intent was to bring this issue to light as a major change that occurs after menopause and may adversely affect quality of life. Both chapters have been written by Mat H. Ho and Narender N. Bhatia, who are involved in the advanced training of this subspecialty area. This new section has been added to this third edition of the book because of its immense importance in the care of postmenopausal women.
References 1. Lukacz ES, Lawrence JM, Contreras R, Nager CW, Luber KM. Parity, mode of delivery, and pelvic floor disorders. Obstet
Gyneco12006;107:1253-1260. 2. MacLennan AH, Taylor AW, Wilson DH, Wilson D. The prevalence of pelvic floor disorders and their relationship to gender, age, parity and mode of delivery. BJOG 2000;107:1460-1470. 3. Griffiths AN, Makam A, Edwards GJ. Should we actively screen for urinary and anal incontinence in the general gynaecology outpatients setting?--A prospective observational study.J Obstet Gynaeco12006;26:442-444.
This Page Intentionally Left Blank
-IAPTER 5
Lower Urinary Tract Disorders in Postmenopausal Women MAT H. H o
Divisionof Female Pelvic Medicine and Reconstructive Surgery, Department of Obstetrics and Gynecology, Harbor-UCLA Medical Center, David Geffen School of Medicine, University of California at Los Angeles, Torrance, CA 90509
NARENDER N. B HATIA Division of Female Pelvic Medicine and Reconstructive Surgery,Department of Obstetrics and Gynecology, Harbor-UCLA Medical Center, David Geffen School of Medicine, University of California at Los Angeles, Torrance, CA 90509
I. INTRODUCTION
This chapter focuses on the pathophysiology, diagnosis, and management of urinary incontinence, interstitial cystitis, and urinary tract infections in postmenopausal women.
During the past century, life expectancy has been increased, and this trend will continue, particularly in the industrialized countries. Women will, therefore, spend a significant part of their life in the postmenopausal years. In the United States, postmenopausal women now comprise more than 15% of the population, with a growth rate of 1.5% predicted until the year 2020 (1). The incidence of lower urinary tract disorders increases in these women, which is probably due to estrogen deficiency, hormonal alterations, muscular and neuronal damages secondary to previous childbirths, and other age-related physiologic changes (2,3). Disturbances in the lower urinary tract produce a wide variety of symptoms in postmenopausal women and account for a significant number of medical visits. Although not life threatening, lower urinary tract disorders can have a severe impact on quality of life. Patients with these problems may suffer from a loss of self-esteem, loss of independence, decrease in sexual actMty, social isolation, and depression (4). In addition to the medical impact and psychologic burden, disorders of the lower urinary tract are also associated with significant medical costs (5). T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
II. ANATOMY O F T H E LOWER URINARY TRACT In order to understand the evaluation and treatment of lower urinary tract disorders in postmenopausal women, the anatomy of these structures is briefly discussed.
A. Anatomy of the Pelvic Floor and Lower Urinary Tract 1. THE PELVIC FLOOR
The pelvic floor, which consists of levator ani muscles and their associated connective tissue attachments, prevents pelvic organ displacement and help to maintain urinary continence. The levator ani is a broad term that consists of two parts: the diaphragmatic part (coccygeus and 693
Copyright 9 2007 by Elsevier, Inc. All rights of reproduction in any form reserved.
694
Ho
iliococcygeus muscles) and the pubovisceral part (pubococcygeus and puborectalis muscles) (6,7). These are bilaterally paired muscles (Fig. 51.1). The coccygeus muscles attach anteriorly to the ischial spines, then fan out medially to attach to the lateral surface of coccyx. The iliococcygeus muscles arise from the lateral wall of the pelvis, run over the obturator internus to attach to the arcus tendineus, and then insert into a midline raphe behind the rectum. The pubococcygeus fibers run from the pubis and the fascia covering the obturator internus and meet in the midline behind the rectum to form the levator plate. Because the iliococcygeus merges together with the pubococcygeus, the distinction between these two muscles is arbitrary. The puborectalis provides muscular sling that pulls the rectum toward the pubic bones when the muscle contracts. The pelvic floor is a dynamic support system, with constant muscle tones and tension adjustments in response to pressure changes, rather than a rigid structure. The levator ani contains both type I (slow-twitch) fibers, which maintain muscle tone over a long period, and type II (fast-twitch) fibers, which increase muscle tone quickly to compensate for any increased abdominal pressure. Reflex contractions of both types of muscle fibers are important in supporting the pelvic contents and keeping urinary and fecal continence. When pubococcygeus and puborectalis muscles contract, they pull the rectum, vagina, and urethra anteriorly toward the pubic bone and constrict the openings of these organs. The anterior movement of the vagina and the contraction of the levator ani also produce strongly occlusive forces on the urethra. These actions help to maintain urinary continence. The pelvic floor is discussed in more detail in Chapter 52.
AND BHATIA
2. THE LOWERURINARYTRACT a. The Bladder The bladder's role is to store and empty urine under voluntary control. The bladder lining consists of transitional epithelium and has a rugose appearance formed by mucosal folds. The bladder neck has three layers of muscles: inner longitudinal, middle circular, and outer longitudinal (8). The remainder of the bladder musculature consists of smooth muscles that run in many directions and can expand to accommodate increasing volume without an appreciable rise in pressure. The trigone is a triangular area in the bladder base, and its three corners are formed by the paired ureteral orifices and the internal urethral opening. The trigone has superficial and deep muscle layers. The superficial layer is continuous with muscle fibers of the ureters and the urethra, and the deep layer fuses with the detrusor fibers. The deep muscle innervation is identical to that of the detrusor, with predominant cholinergic nerves and sparse noradrenergic fibers (6). The superficial layer has greater numbers of noradrenergic nerves and only few cholinergic activities. b. The Urethra In the adult female, the urethra is a muscular tube 3 to 4 cm in length and about 6 mm in diameter (6,8). Throughout its length, the urethra is embedded in the adventitia of the anterior vaginal wall. It is lined proximally with transitional epithelium that becomes striated squamous epithelium distally. The urethra is surrounded by smooth muscle that is primarily composed of oblique and longitudinal fibers and has extensive cholinergic innervations with few
FIGURE51.1 Superiorviewof the pelvicfloor structures. (Reproducedfrom RetzkySS, Rogers RM Jr, RichardsonAC. Anatomyof femalesupport.In: BrubakerLT, SaclaridesTJ, eds. Thefemale pelvicfloor." disordersoffunction and support. Philadelphia:FA Davis Company,1996:11.
695
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women noradrenergic nerves (6). The striated muscular urethral sphincter, which surrounds the proximal two-thirds of the urethra, contributes about 50% of the total urethral resistance and serves as a defense mechanism against incontinence. It is also responsible for the interruption of urinary flow at the end of micturition. The two posterior pubourethral ligaments provide a strong suspensory mechanism for the urethra and serve to hold it forward and in close proximity to the pubic bone under conditions of stress (Fig. 51.2). They extend from the lower part of the pubis to the urethra at the junction of its middle and distal third. The anterior vaginal
wall also provides a slinglike support to the proximal urethra and bladder base through its attachment to the levator ani muscles and the arcus tendineus fascia pelvis. The urethral functions are maintained by intrinsic and extrinsic factors. The intrinsic factors include the tones, contractions, and elasticity of the urethral muscles and the coaptation of the urethral mucosa. The tones and contractions are mediated by alpha-adrenergic receptors of the sympathetic nervous system. The extrinsic factors include the levator ani muscles, the endopelvic fascia, and their attachments. These structures form a hammock beneath the urethra to provide a backboard for its compression when intra-abdominal pressures are increased. Hypermobility of the urethra and bladder neck secondary to the loss of their support can occur with the muscular weakening and endopelvic fascia detachment, resulting in stress urinary incontinence.
B. Innervation of the Lower Urinary Tract 1. SYMPATHETIC AND PARASYMPATHETIC INNERVATION S
FIGURE 51.2 The components of bladder and urethral supports and periurethral structures. A. Interrelationships of these structures. B. Sphincter urethrae, urethrovaginal sphincter, and compressor urethrae are all parts of striated urogenital sphincter muscle. LA: levator ani muscles; VLA: vaginal levator attachment; D: detrusor muscle; AT: arcus tendineus fasciae pelvis; PUL: pubourethral ligament; US: urethral sphincter (or sphincter urethrae); CU: compressor urethrae; UVS: urethrovaginal sphincter; IC: ischiocavernosus muscle; BC: bulbocavernosus muscle. (A, From Mishell DR Jr, Stenchever MA, Droegemueller W, Herbst AL. Comprehensive gynecology. St. Louis, Mosby, 1997:576; B, From DeLancey JOL, Richardson AC. Anatomy of genital support. In: Hurt WG, ed. Urogynecologicsurgery. New York, Raven Press, 1992,:27.)
The lower urinary tract is under the control of both sympathetic and parasympathetic nerves (9). Although the parasympathetic system seems to control primarily bladder emptying while the sympathetic system controls bladder storage, the mechanism of micturition involves the interplay of both these systems (Fig. 51.3). Furthermore, a variety of nonadrenergic and noncholinergic neurotransmitters as well as neuropeptides are also involved in the modulation of these activities at the spinal cord and higher levels of the central nervous system. The sympathetic nerves originate from thoracolumbar segments (T10 through L2) of the spinal cord. These nerves have alpha- and beta-adrenergic components. The alpha fibers terminate primarily in the urethra and bladder neck, whereas the beta fibers terminate primarily in the detrusor muscle. Alpha-adrenergic stimulation contracts the bladder neck and urethra, and beta-adrenergic stimulation relaxes the detrusor muscle. The parasympathetic fibers originate in the sacral spinal cord segments $2 through $4. The main neurotransmitter is acetylcholine, which acts on muscarinic receptors. Stimulation of the pelvic parasympathetic nerves causes the detrusor muscle to contract, while administration of anticholinergic drugs controls detrusor overactivity, reduces the vesicle pressure, and increases the bladder capacity.
2. SOMATIC MOTOR CONTROL
The pudendal nerve ($2 through $4) provides motor innervation to the striated urethral sphincter. At the perineal membrane, this nerve divides into the inferior rectal nerve, the dorsal nerve to the clitoris, and the perineal nerve. Although there is considerable variation with the
696
Ho ANDBHATIA sacral reflex arc. Later, connections to the higher centers become established, and by training and conditioning, this spinal reflex becomes socially influenced so that voiding can be voluntarily accomplished.
III. U R I N A R Y I N C O N T I N E N C E A. Classification and Definition Urinary incontinence was recently redefined by the International Continence Society (ICS) as the complaint of any involuntary leakage of urine, and it can be classified into six different types: stress urinary incontinence, urge urinary incontinence, mixed urinary incontinence, nocturnal enuresis, continuous urinary incontinence, and other types of urinary incontinence (10). This new definition and classification reflect the recent advances in the field. Although this classification system is not perfect, it assists the clinicians in the evaluation and treatment of urinary incontinence.
1. STRESS URINARY INCONTINENCE
FIGURE 51.3 Innervations of the female lower urinary tract and mechanism of micturition. A. Innervations of the lower urinary tract. B. Actions of the autonomic and somatic nervous systems during bladder filling/storage and voiding. (Reproduction from Benson JT, Waiters MD. Neurophysiology of the lower urinary tract. In: Waiters MD, Karram MM, eds. Urogynecology and reconstructive pelvic surgery, ed. 2. St. Louis, Mosby, 1999:16-17.)
branching, the perineal nerve divides into a superficial branch that goes to the labia and a deep branch that innervates the periurethral striated muscles. The involved neurotransmitters are acetylcholine, and receptors are nicotinic type. The somatic efferent branches of the pelvic nerve may also innervate the proximal striated sphincter muscle (9). 3. SENSORY INNERVATIONS
Sensory afferents can be found in the pelvic, hypogastric, and pudendal nerves. Afferent impulses from the bladder, trigone, and proximal urethra pass to the spinal cord. Urethral sensation is carried out primarily by the pudendal nerve. It has been postulated that the loss of the sensory receptors at the trigone, which are different from the stretch receptors in the rest of the bladder, may lead to urge incontinence (9). In infancy, the storage and expulsion of urine is automatic and controlled at the level of the
Stress urinary incontinence (SUI) is the complaint of involuntary leakage of urine on effort or exertion, or with sneezing or coughing (10). SUI can also occur with lifting, running, or any activities that increase the intra-abdominal pressure. The urine leakage can range from a few drops to large volumes. The detrusor contraction should be absent. SUI is caused by ineffective urethral closure, which is related to the deficiencies in extrinsic urethral support or intrinsic urethral integrity, during times of increased intra-abdominal pressure. SUI therefore can be classified further into bladder neck hypermobility and intrinsic sphincter deficiency (ISD). Although bladder neck hypermobility is the major cause of SUI, many women with this condition do not have urinary incontinence if their urethral sphincters are strong and well functioning. Bladder neck hypermobility and ISD can occur together, and the classification in this case is based on the predominated mechanism of incontinence. This decision will guide the type of treatment, which will be discussed later. Genuine stress incontinence (GSI) is used for patients who have undergone urodynamic testing and met the diagnostic criteria. These criteria include the demonstrable urinary leakage when intravesical pressure exceeds the maximal urethral closure pressure and the absence of detrusor contractions. The term GSI is referred to as urodynamic stress incontinence in the new ICS terminology (10).
2. URGE URINARY INCONTINENCE
Urge urinary incontinence (UUI) is defined as the complaint of involuntary leakage of urine accompanied or immediately preceded by urgency (10). The symptom of
697
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women
urgency is defined as the complaint of a sudden compelling desire to pass urine that is difficult to defer (10). Several other terminologies need to be clarified. Detrusor overactivity (DO) is a urodynamic observation characterized by involuntary detrusor contractions during the filling phase (10). These detrusor contractions may be spontaneous or provoked. The older terms "detrusor instability" and "unstable bladder" are now replaced by DO. Detrusor overactivity incontinence is a new term to describe urinary incontinence due to DO. The difference between UUI and detrusor overactivity incontinence is that the former is characterized by the symptom of urgency and the latter is characterized by the sign of DO. DO can be divided into neurogenic detrusor overactivity (NDO) and idiopathic detrusor overactivity (IDO). N D O is used when there is a relevant neurologic condition that causes the detrusor overactivity. N D O replaced the older term "detrusor hyperreflexia." However, in the majority of patients the cause of involuntary bladder contractions is unknown, and this condition is referred to as IDO. This type of DO can be triggered by specific events similar to UUI or can occur in the absence of any identifiable event and without any feeling of urgency. Overactive bladder (OAB), or overactive bladder syndrome, is the condition whereby a patient has symptoms of urgency, with or without urge incontinence, in the absence of infection or other proven etiologies (10). These patients may also have urinary frequency and nocturia. OAB is a symptomatic diagnosis and therefore does not require urodynamic testing for confirmation. 3. MIXED URINARY INCONTINENCE
Mixed urinary incontinence (MUI) is defined as the complaint of involuntary leakage of urine associated with urgency and also with exertion, effort, sneezing, or coughing (10). MUI is, therefore, a coexistence of SUI and UUI. However, the degree to which each component contributes to the patient's symptoms varies, and one type of incontinence often predominates the other. Patients with MUI tend to have more incontinent episodes and leak larger volume of urine than those with either SUI or UUI alone (11). 4. NOCTURNALENURESIS
Nocturnal enuresk is the complaint of urinary loss occurring during sleep (10). It should be distinguished from the complaint of waking up at night one or more times to void, which is nocturia, and from waking with urgency and then leaking before arriving at the toilet, which is UUI. Nocturnal enuresis can be primary or secondary. Primary nocturnal enuresis starts in childhood and can persist into adulthood, and secondary nocturnal enuresis starts in adulthood.
5. CONTINUOUS URINARY INCONTINENCE
Continuous urinary incontinence is the complaint of continuous leakage of urine. Nonfunctioning urethra caused by partial urethral resection secondary to vulvar cancer or by scarring and fibrosis from previous surgeries can be the etiologies. Urethral sphincter paralysis secondary to lower motor neuron diseases can also cause continuous incontinence. Other causes include urogenital fistulas, noncompliance bladder secondary to pelvic radiation and fibrosis, and congenital malformations of the urogenital tract, such as bladder exstrophy and ectopic ureters. 6. OTHER TYPES OF URINARY INCONTINENCE
a. Overflow Urinary Incontinence
Ove~flozo incontinence
can occur with underactive detrusor and/or overactive urethra. This type of incontinence is commonly related to neurogenic bladder and outlet obstruction (12). Several medical conditions such as diabetes mellitus, multiple sclerosis, meningomyelocele, lumbosacral tumors, high spinal cord injuries, and prolapsed vertebral disks can result in bladder neuropathy and overflow incontinence. When detrusor muscle becomes acontractile, bladder emptying is not effective and incontinence can occur as an overflow of urine. In the early stage of diabetic bladder neuropathy, the detrusot functioning may wax and wane between hypoactivity and overactivity.
b. Transient Urinary Incontinence Transient incontinence (or functional incontinence) is caused by reasons other than neurologic, anatomic, or other urinary tract dysfunctions. The detrusor and urethra are normal. The common causing factors are described in the mnemonic DL/IPERS (delirium and psychiatric disorders, infection of the urinary tract, atrophic urethritis or vaginitis, pharmacologic agents, excessive urine production, reduced mobility, and stool impaction). Delirium and psychiatric disorders may cause disorientation that result in incontinence. Urinary tract infection may lead to bladder irritation and DO. Atrophic urethritis and vaginitis in postmenopausal women can cause urinary urgency, frequency, and incontinence. Several pharmacologic agents (as described in the section on incontinence evaluation) can induce urinary leakage. Excessive urine production from diabetes, hypercalcemia, or excessive fluid intake as well as restricted mobility and stool impaction can also lead to urinary incontinence. Bowel disorders are usually associated with lower urinary tract disorders because these structures share the innervation. Because these conditions are reversible, thorough medical history and evaluations are important components of the work-up of patients with urinary incontinence, particularly for older women.
698
Ho
c. Potential Urinary Incontinence Potential incontinence (or masked incontinence) refers to the SUI that is revealed only after reduction of the pelvic organ prolapse. The prolapse may kink the urethra and mask the presence of SUI. Typically, the patient has a history of SUI with some improvement as the pelvic organ prolapse progressed, and finally the SUI resolved coincidently with worsening of the prolapse. The diagnosis can be made with the stress test or urodynamics after reduction of the prolapse with a pessary in the vagina. In the evaluation of patients with pelvic organ prolapse for surgical treatment, the potential incontinence must always be ruled out in order to avoid the new-onset incontinence that follows the surgical correction of the prolapse. In the patients with both pelvic organ prolapse and potential incontinence, anti-incontinence surgery should be performed concomitantly with the antiprolapse procedure. Other types of incontinence such as coital incontinence, which occurs during sexual intercourse, or giggle incontinence, which occurs when a women laughs, are usually reversible and poorly reproducible.
B. Epidemiology Urinary incontinence in postmenopausal women may be under-reported because only one-quarter of incontinent women in the United States and one-third in the European countries seek medical attention for this problem (13). Many postmenopausal women may be embarrassed about their symptoms, think that urinary incontinence is normal as they get older, and delay the seeking of treatment until the condition is severe (4). The precise prevalence of urinary incontinence in these women is difficult to estimate because most studies are questionnaire based and only few are comprehensive, standardized epidemiologic investigations (13,14). Furthermore, many clinicians have misconceptions about urinary incontinence. An earlier study in 1989 reported that 70% of primary care physicians and 80% of nursing home physicians failed to acknowledge this problem in their patients (15). Although this number has improved in recent years, a study published in 2001 still showed that 46% of patients seeking help for urinary incontinence were untreated (16). When compared with men, women have twice the prevalence and are more likely to suffer moderate to severe urinary incontinence (4,13,17). The prevalence of urinary incontinence increases with age, and this increase was demonstrated in a review of 21 epidemiologic studies in which the pooled mean prevalence of urinary incontinence among women at age of 50 or older was 34% versus 25% among middle-aged and younger women (14). Urinary incontinence has been reported to affect more than 50% of nursing home residents (4). Genitourinary atrophy due to hypoestrogenism, weakening of connective tissue, increasing in
AND BHATIA
medical comorbidity, decreasing in cognitive functions, impairments in mobility, and increasing in nocturnal diuresis are contributing factors to this age-related increase in urinary incontinence. The prevalence of postmenopausal incontinence in the community is thought to be between 16% and 29% (18-20). Of the group of perimenopausal and postmenopausal women who attended a menopause clinic, however, 20% complained of severe urinary urgency and nearly 50% complained of stress incontinence (18). Recent data on the distribution of urinary incontinence by type reported that 33.8%, 31.8%, and 34.4% of women with incontinence disorders have SUI, UUI, and MUI, respectively (21,22). SUI tends to be more common among younger women, whereas older women have more UUI and MUI (14). The data on prevalence of urinary incontinence in racial and ethnic groups are somewhat limited. One study reported that the prevalence rates among Caucasian women and African-American women are similar (13), whereas another study found higher prevalence among Caucasian women (23). In a study of older women, urinary incontinence was found in 23% of Caucasian women and in 16% of African-American women (22). Recently, Thorn and coworkers (24) reported that the risk of SUI is lower, while the risk of UUI was similar, in African-American and Asian-American women when compared with Caucasian women.
C. Pathophysiology Urinary incontinence is a complex, multifactorial disorder, and the etiologies are diverse and, in many cases, incompletely understood. It usually involves more than one anatomic structure. Contributing factors include structural and functional disorders of the urethra, bladder, ureters, and surrounding supporting elements. At the urethral level, urinary incontinence can be caused by intrinsic etiologies, extrinsic factors, or obstruction. Intrinsic urethral etiologies can be anatomic, such as urethral fistula and diverticula, or functional, such as ISD and urethral instability. Extrinsic factors refer to the loss of anatomic support to the proximal urethra and urethrovesical ]unction causing urinary incontinence. At the bladder level, urinary incontinence can be caused by idiopathic or neurogenic detrusor overactivities as well as other conditions such as urinary tract infection, bladder calculi, foreign bodies, bladder neoplasms, radiation exposure, bladder overdistention, and outlet obstruction. Damage to the bladder anatomic integrity, including vesicovaginal fistula, vesicocutaneous fistula, and untreated exstrophy, can also result in incontinence. At the ureteral level, the causes of incontinence are very rare and include congenital abnormalities such as ectopic ureter or anatomic injuries such as ureterovaginal fistula.
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women It should also be noted that urinary incontinence can have a spontaneous remission. In a 5-year follow-up of 90 women who were incontinent at baseline, 28% of these women became continent, which corresponds to an annual remission rate of 5.6% (25). Women in the remission group tended to be younger, and the highest remission rate (10.6%) was in the 20- to 29-year-old group. Women with SUI at baseline had a higher annual remission rate than those with UUI or MUI. In order to understand the pathophysiology of different types of urinary incontinence, brief discussions of the mechanism of micturition and continence are warranted.
1. MECHANISM OF MICTURITION AND CONTINENCE
During the filling phase, the detrusor muscle relaxes to allow gradual expansion of the bladder while the urethral sphincter closes to prevent urine leaking. Normally, there is little or no increase in intravesical pressure despite a large increase in urine volume. This process is coordinated by the parasympathetic and sympathetic nervous systems, pelvic nerve, spinal cord, and brain (9). The bladder is controlled by neural impulses from both afferent and efferent pathways. As the bladder volume increases during the filling phase, tension-stretch receptors on the bladder wall become activated, leading to a desire to void. However, normal women are able to suppress this micturition reflex until an appropriate time. These activities seem to be mediated primarily by the sympathetic nervous system. In the bladder-emptying phase, cholinergic signal from the pelvic nerves coordinates the detrusor contraction and urethral sphincter relaxation (9). The voiding occurs with voluntary relaxation of the urethral sphincter and the pelvic floor as well as the contraction of the detrusor muscle. The bladder-emptying phase is mediated by the parasympathetic nervous system. In order for these mechanisms to work properly, thus maintaining continence, the anatomic and functional properties of the involved structures and their innervations should be intact.
2. STRESS URINARY INCONTINENCE
a. Bladder Neck Hypermobility SUI occurs when the rise in the intravesical pressure, which is caused by an increase in intra-abdominal pressure, overcomes the urethral resistance, resulting in urinary leakage. For continent women, the bladder and proximal urethra are located retropubically within the sphere of intra-abdominal pressure. The urethral sphincter and urethra have a higher pressure than the bladder, and this pressure gradient is preserved during period of physical stress because the changes in intra-abdominal pressures are transmitted equally to the bladder and proximal urethra. The commonly accepted theory for the pathogenesis of SUI is
699
that the bladder neck and proximal urethra drop below the pelvic floor because of the loss of anatomic support. This condition is often referred as bladder neck hypermobility (17,26). The increase in intra-abdominal pressures during strenuous activities or coughing is, therefore, not transmitted equally to the bladder and proximal urethra. As a result, intraurethral pressure falls below bladder pressure and urinary leakage occurs. The anterior vaginal wall and the endopelvic fascia that attaches bilaterally to the arcus tendineus and levator ani muscles provide support to the bladder neck and urethra. The attachment of suburethral endopelvic fascia to the pubococcygeus further limits bladder neck descent during physical stress (6,8). Damages to these structures and to the nerves, muscles, and connective tissues of the pelvic floor are thought to be the main culprits. Trauma from childbirth has been shown to predispose women to SUI (17). b. Suburethral Hammock Support In continent women, the urethra is firmly supported by the anterior vaginal wall and endopelvic fascia that are attached laterally to the arcus tendineus and levator ani muscles. These tissues provide a "hammock" of supporting structure, not only for the urethra but also for the urethrovesical junction (6,8,27). Any increase in intra-abdominal pressure will transmit a downward force to the anterior urethral wall and compress it against the firm, supported posterior urethral wall, thus occluding the urinary flow. If the anterior vaginal wall is damaged or weakened, or if the endopelvic connective tissue is detached from its normal lateral fixation, the optimal urethral compression does not occur, and urinary leakage results. With the success of tension-free suburethral sling procedures, it is now recognized that both of the hypermobility and loss of support to the bladder neck and urethra are important mechanisms of SUI. It has also been postulated that the integrity of the suburethral hammock is more important for continence than the location of the urethra relative to the pelvic floor (27). c. Intrinsic Sphincter Deficiency ISD, in which the urethral sphincter fails to close appropriately during period of stress, is another cause of SUI (26). Several factors such as childbearing and delivery, hypoestrogenism, periurethral vascular damage, neurologic diseases, radiation exposure, scarring secondary to previous surgery, or certain medications (i.e., alpha-adrenergic antagonists) can contribute to the development of ISD. Mucosal atrophy and hypovascularity of the urethral wall as well as neuromuscular damage to the urethral sphincter can also decrease urethral tone and cause ISD. This condition may result from myelomeningocele or epispadias or may be acquired after trauma or a sacral cord lesion. ISD is also common in women who had multiple anti-incontinence surgeries. Urine leakage in this
700 condition can occur with little exertion or with minimal activity such as throat-clearing or lifting a light weight (17). SUI secondary to ISD is less common but more challenging to diagnose and treat. It has been shown that suburethral sling procedures can effectively restore long-term continence in 76% to 96% of lSD patients (28). 3. URGE URINARY INCONTINENCE
UUI can be quite debilitating, and its pathophysiology is not completely understood. This type of incontinence can be triggered by some events such as water running, handwashing, orgasm, changes in posture or position, changes in temperature, or opening the front door. Neurologic diseases, outflow obstruction, local bladder and urethral irritants, and medications must be considered as etiologies. Neurologic disorders such as multiple sclerosis (MS), cerebrovascular diseases, Parkinson's disease, spinal cord injuries, and neoplasms of the central nervous system or spinal cord can cause N D O and UUI (17,26). Approximately 90% of patients with MS have lower urinary tract dysfunction during the course of the disease. Any hemorrhage, infarction, or vascular diseases in the areas of the cerebral cortex, internal capsule, brainstem, and cerebellum can result in N D O and UUI. Spinal cord injuries interrupt the sacral reflex and other pathways that are responsible for the voluntary and involuntary inhibition of detrusor contractions. Although the bladder is areflexic during the initial phase of injury that results in overflow incontinence, detrusor muscle eventually becomes hyperreflexic in the later phase. Bladder outlet obstruction due to advanced pelvic organ prolapse, such as severe cystocele or vaginal vault prolapse, is another cause of UUI (17,26). Local bladder and urethral irritants such as foreign bodies, permanent sutures, stones, and neoplasms as well as cystitis and medications such as parasympathomimetics must also be considered as etiologies in the work-up of UUI. However, in the majority of patients the cause of UUI is unknown. Theories regarding the intrinsic bladder abnormalities in UUI include disorders of bladder ganglia, disorders of pacemaker cells, increased activity of sensory nerves, and deficiency in prostaglandin production (29). The common final pathway of these mechanisms is probably the myogenic dysfunction of the detrusor muscle with increased capacity for spontaneous contractions. Damage to the detrusor muscle caused by trauma, aging, atrophy, or loss of innervation as well as disruption in the neurochemical pathways are responsible for this condition. UUI and OAB can also be caused by changes at the cellular level of the detrusor muscle (29). With an increased ratio of abnormal-to-normal cell junctions, the detrusor can spread these uninhibitory contractions. In a small subgroup of patients with UUI, the leakage of urine into a funneled and incompetent proximal urethra may trigger the micturition reflex, causing involuntary bladder contraction (30).
Ho AND BHATIA
UUI or OAB that occurs after anti-incontinence procedures in patients who preoperatively had only GSI is referred as de novo U U I or de novo OAB. It was hypothesized that elevation of the bladder neck in anti-incontinence procedures could cause urethral compression and outflow obstruction resulting in de novo U U I or OAB. However, this theory has been challenged (31). It was also proposed that repeat surgeries at the bladder neck may interfere with the autonomic nerve supply of the bladder and result in de novo 13151 or OAB. The incidence of de novo O A B ranges from 5% to 18% in two series (32,33) and from 5% to 27% in another series (34). 4. MIXED URINARY INCONTINENCE
MUI is a coexistence of both SUI and UUI, and the degree to which each component contributes to the patient's incontinence varies. It has been shown that 33% of women with SUI may have coexistent DO (35). These patients represent a therapeutic challenge, and whether coexistence of DO decreases the surgical cure rate of SUI is controversial. Currently, the exact etiology of MUI is not clear, and this condition often responds poorly to treatment, either pharmacologic or surgical. A better understanding of its pathophysiology is needed to target the treatment more effectively.
D. E v a l u a t i o n Initial evaluation of urinary incontinence in postmenopausal women requires a systemic approach that includes a detailed history, the bladder diaries, the quality of life questionnaires, the physical examinations, a cotton swab Qztip test, a stress test, a urinalysis (with urine culture and cytology if indicated), and a postvoid residual volume. For at least 90% of patients with urinary incontinence, these steps in combination with a simple cystometry are adequate investigations (36,37). They can be carried out in the office, and noninvasive empirical therapy such as pelvic floor muscle exercises, behavioral modification, or pharmacologic therapy can be initiated following these evaluations. Additional multichannel urodynamic, uroflowmetric, cystourethroscopic, electromyographic, electrophysiologic, or imaging studies may be necessary in patients with a history of multiple previous surgeries for urinary incontinence and for patients with associated neurologic diseases. They are also indicated for patients who failed the empiric therapy and are being considered for surgical intervention. 1. MEDICAL HISTORY
A thorough medical history should include the presenting symptoms, the degree to which these symptoms are bothersome, the onset and frequency of incontinence, and the amount of urine loss. Triggering events causing urine
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women loss, character of urine loss as constant or .intermittent, associated symptoms of dysuria or bladder pain, and other concomitant symptoms are also essential. Medical history of chronic cough, chronic obstructive pulmonary disease (which can lead to SUI due to chronic coughing), diabetes mellitus (which can produce osmotic diuresis if glucose control is poor), vascular insufficiency (which can mobilize fluid from peripheral edema into the vascular system at night, resulting in diuresis), cancer treatment (adverse effects of radiotherapy or chemotherapy on the urinary tract), and renal disorders should be obtained. Neurologic conditions that especially affect the bladder and urethral sphincter functions, such as MS, stroke, Parkinson's disease, spinal cord injury or neoplasm, and myelodysplasia, should be included. Symptoms suggesting neuropathy, such as muscle weakness, paralysis or poor coordination, gait disturbance, tremor, tingling sensation, numbness, and double vision, are also important in the medical history. Surgical histories of the pelvic, urologic, and central nervous systems are essential. Radical hysterectomy for uterine or cervical cancer may cause detrusor denervation, and surgeries involving the spine and brain can affect bladder function. Obstetric histories, such as the number of pregnancies, spontaneous deliveries, cesarean deliveries, and instrumental deliveries, are important because they may have significant effects on the pelvic support and lower urinary tract structures. Modifiable risk factors such as diet, smoking, alcohol and caffeine intake, mobility problems, and other transient causes of urinary incontinence (DIAPERS causes, as described earlier) should be obtained. All medications that may affect the lower urinary tract should be carefully reviewed. Antihypertensive medications, such as alpha blockers and calcium channel blockers, may decrease detrusor contractility or relax the bladder neck and urethra, leading to incontinence and voiding problems. The increased urine volume produced by diuretic therapy can be a challenge to older patients because of their decreased bladder capacity. Anticholinergic drugs, such as antihistamines, antidepressants, antipsychotics, antispasmodics, opiates, and anti-Parkinson's disease medications, may impair detrusor contractility and cause voiding difficulty or overflow incontinence. Over-the-counter cold medications that contain sympathomimetics or antihistamines with significant anticholinergic effects can adversely affect the detrusor and urethral sphincter functions. Tricyclic antidepressants, neuroleptics, narcotic analgesics, sedatives (e.g., benzodiazepines), and alcohol can be detrimental to cognition and, therefore, incontinence, especially in older patients. The presentation of urinary incontinence can range from a simple complaint to a combination of symptoms. The medical history may not accurately identify the type of urinary incontinence, and attempting to diagnose this disorder with medical history alone frequently leads to mistakes. A recent meta-analysis reported that the symptoms of urine
701
loss only with coughing, sneezing, laughing, running, or Valsalva maneuvers have a false-positive rate of 23% and a positive predictive value of 56% for the diagnosis of GSI (38). For the UUI, the reported symptom (i.e., urine loss only following a strong desire to void) has a sensitivity of 77% and a specificity of 34% (38). In patients with MUI, the described symptoms of both stress and urge incontinences have a false-positive rate of 36% and a positive predictive value of 79% (38). Although the patient's presenting symptoms may not be reliable in the determination of the type of incontinence, her perceptions of the severity of these symptoms are important in the evaluation and treatment. The decision on surgical procedures should not be based on medical history alone because these procedures are more likely to fail if the diagnosis is incorrect or incomplete. 2. VOIDINGDIARY
Although a voiding diary can be a valuable aid in the evaluation of patients with urinary incontinence, its use is often neglected. There is no standardized format for the voiding diary; however, any of the published forms (37,39) can be employed because they are similar and provide adequate information. Because studies have demonstrated that a 3-day diary is as reliable as a 7-day diary, recording for 3 continuous days is a common approach. The volumes of each void can be obtained with a plastic measuring "hat" that fits over the patient's toilet bowl. The voiding diary should record the time and volume of each void, the type and volume of fluid intake, the frequency and volume of incontinence, and the associated activities and symptoms with incontinence. Many types of fluid intake, such as caffeinated or alcoholic drinks, carbonated beverages, and certain food ingredients, can act as bladder irritants. Initial management in these patients is to eliminate or minimize these irritants. The voiding diary also helps the clinician to confirm complaints and to determine whether the incontinence is in part due to an abnormally high or low urinary output. For example, in diabetic patients with urinary incontinence secondary to osmotic diuresis (high urine output), the first step is to control blood sugar in order to decrease the urine output, rather than pharmacologic or surgical interventions. On the other hand, patients who limit the fluid intake dramatically in an attempt to control their incontinence will make highly concentrated urine (low urine output), which can irritate the bladder and cause UUI. Increase fluid intake will help these patients. The voiding diary can also help to determine the initial interval in the timed voiding technique for the treatment of UUI. Generally, a functional bladder capacity of 400 to 600 mL, an average voided volume of approximately 250 mL, a voiding frequency of six to eight times per day, and a daily urine output of 1500 to 2500 mL are considered normal (39,40). Although this recorded information can provide a
702 reliable voiding pattern and severity of the incontinence over a given period, it cannot differentiate stress from urge or mixed incontinences. Voiding diaries are shown to be reproducible in the setting of stress incontinence; however, data regarding its reproducibility in urge and mixed incontinences are lacking (3 9). 3. QUALITY OF LIFE QUESTIONNAIRES Validated instruments such as the Urogenital Distress Inventory (UDI) and Incontinence Impact Q.uestionnaire (IIQ] were developed to assess the impact of incontinence on quality of life (41,42). The UDI and IIQ instruments provide a multiple-component quantification of the severity of incontinence and its effects on the daily activities. These questionnaires allow the inclusion of the patient's perception of incontinence into the overall assessment. They can be easily administered in the clinic. 4. PHYSICALAND PELVIC EXAMINATIONS
General, focused neurologic, and detailed pelvic examinations should be parts of the evaluation. When the patient arrived with a full bladder, the voided volume is measured and the bladder is then catheterized to obtain the postvoid residual volume as well as samples for urinalysis and urine culture. If urodynamics are a part of the evaluation, these volumes can be obtained more conveniently at that time. a. GeneralExamination The generalexamination should include the vital signs, height, and weight as well as the specific cardiac, pulmonary, abdominal, and extremity evaluations. The body mass index should be calculated because obesity is a contributing factor to SUI. The examination should also include detection of medical conditions, such as cardiovascular insufficiency, pulmonary disease, and neurologic disorders, that may affect the lower urinary tract. The abdomen should be examined for surgical scars, hernia, masses, and bladder distention. The presence of hernias may indicate inherent connective tissue weakness, a possible contributor to the incontinence. b. Neurologic Examination Neurologic disorders can cause or exacerbate urinary incontinence (43). The patient's mental status and gait should be observed, and any abnormalities should prompt more in depth investigations such as examination of the cranial nerves, extremity function, and reflexes. Neurologic examination should also focus on the motor, sensory, and proprioceptive functions of the lumbosacral nerve roots by evaluating the strength, sensation to dull and sharp stimuli, and deep tendon reflexes of the lower extremities (43,44). Two reflexes may help in the evaluation of
Ho AND BHATIA sacral integrity. The anal wink reflex, or sensation of the perianal areas, can be detected by stroking laterally to the anal canal with a cotton swab and observing the reflex contraction of the external anal sphincter muscle. The bulbocavernosus reflex, which reflects the integrity of the $2 to $4 levels of the sacral cord, is elicited by gently tapping the clitoris with a cotton swab and observing the contraction of the bulbocavernosus and ischiocavernosus muscles. Although the presence of these reflexes may rule out any significant sacral cord problem, their absence does not indicate neuropathy. Up to 30% of neurologically intact women do not demonstrate these reflexes (43). c. Pelvic Examination For the pelvic examination, the external genitalia, urethral meatus, and vagina should be inspected for evidences of atrophy and hypoestrogenism, such as pallor and thinness of tissues, loss of rugae, fleshy lesion, and caruncle. Because the distal urethra is estrogen dependent, the patient with atrophic vaginitis may also have atrophic urethritis. The urethra should be palpated for suburethral masses, which raise the suspicion for a urethral diverticulum. If tenderness and watery or purulent discharge occur with compression of the urethra, diverticulum or urethritis may be present. Tenderness with the urethral and trigonal palpations may also indicate urethral syndrome or interstitial cystitis. Skin irritation suggests the presence of prolonged incontinence. Pelvic floor relaxation usually coexists with incontinence and should be carefully evaluated because surgical treatments for both of these conditions can be carried out simultaneously. Pelvic examination should be performed systematicaUy for the apical, anterior, and posterior vaginal wall defects. The presence of uterine or apical prolapse, cystocele, urethrocele, rectocele, enterocele, and perineal laxity should be noted. Details of these examinations are described in Chapter 52. In the digital examination, the uterus and adnexa are palpated for masses or tenderness. The urethra should be carefully palpate against the pubic symphysis to rule out suburethral masses and diverticula. The bladder should also be palpated for tenderness that may represent urinary tract infection, interstitial cystitis, or bladder stone. The levator ani muscles are palpated and their strength is appreciated while the patient squeezes the examining fingers. The perineal body should also be carefully inspected for its length and tissue thickness. A short perineal body with mostly skin and little or no underlying muscle indicates a damaged or compromised structure. The thickness of the rectovaginal septum is determined by a rectovaginal examination. Internal and external rectal sphincter tones should also be noted. With fingers in the vagina and rectum, any sliding down of tissue while the patient is in maximal straining may suggest enterocele.
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women
5. STRESSTEST
The stress test can be carried out during the pelvic examination or urodynamic study. Because the test must be done with full bladder, many clinicians prefer to perform it following the urodynamic procedure when the bladder is filled with fluid. While the physician observes the urethral meatus, the patient is asked to perform provocative maneuvers such as coughing or straining in the supine position. SUI is suggested if short spurts of urine escape simultaneously with each cough or straining. Generally, the patient with SUI displays immediate urine loss of relatively short duration. A delayed leakage, or loss of large volumes of urine, suggests uninhibited bladder contractions. If loss of urine is not demonstrated in the supine position, the test should be repeated with the patient in a standing position. If more than mild pelvic organ prolapse is present, the prolapse should be reduced (with a pessary or a single-blade speculum) during the stress test because the prolapse may mask the urine leakage. Care must be taken not to compress the urethra while reducing the prolapse. If the stress test is positive with a relatively empty bladder, in supine position, and with minimally increased intra-abdominal pressure, ISD is suspected and urodynamic testing is indicated. The stress test must be done with a full bladder, the urine loss should be visualized with each coughing or straining maneuver, and the urine loss should cease between the tests for it to be reliable (36,37). If the history strongly suggests SUI but stress test findings are negative and DI is ruled out, the pyridium pad test (45) can be employed. 6. THE Q2TIP TEST The Qztip test determines the mobility and descent of the urethrovesical junction on straining (36). With the patient in the lithotomy position, a sterile and lubricated cotton swab Qztip is inserted into the urethra to the level of the urethrovesical junction, and the angle between the Qztip and the horizontal is measured. Another simple way is to place the Qztip all the way into the bladder and then pull back until increased resistance is met, which indicates that the cotton tip is at the urethrovesical junction. The patient then asked to strain maximally, which produces a descent of the urethrovesical junction. Along with this descent, the Q:tip moves, producing a new angle with the horizontal. The resting and straining angles are measured with a simple goniometer. The normal change in these angles is up to 30 degrees. Hypermobility of the bladder neck is defined as a deflection angle with straining of greater than 30 degrees from the resting angle or greater than 30 degrees from the horizontal. In patients with pelvic relaxation and SUI, the change in Qztip angles can be in the range of 50 to 60 degrees or more.
703
The Qztip test does not establish the diagnosis of SUI and is not essential for patients in whom noninvasive treatment is intended (36). However, it may be valuable in differentiating SUI caused by bladder neck hypermobility from that caused by ISD and in predicting the failure of incontinence surgery. This test is useful if surgical intervention is planned because those with SUI and a well-supported bladder neck and urethra (the Qztip test is negative) may have ISD and should be treated with periurethral bulking agents or suburethral sling procedure rather than with bladder neck suspension. 7. POSTVOIDRESIDUALVOLUME
Although the Agency for Health Care Policy and Research (AHCPR) guidelines recommend that postvoid residual volume should be included in the initial evaluation (36), its value is limited except in patients suspected of having obstructional or neuropathic causes for their incontinence, and in postoperative patients. As compared with men, obstructive voiding is rare in women. Furthermore, the postvoid residual volumes have a poor test/retest reliability (44). The postvoid residual volume less than 50 mL is considered normal, and the volumes greater than 100 mL is generally considered abnormal (17,44). Volumes of greater than 100 mL but less than 200 mL should be repeated and correlated clinically. Volumes of greater than 200 mL should be referred for evaluation by a specialist. Patients with a large postvoid residual volume may have a diminished functional bladder or outlet obstruction. Incontinence can occur in these patients as a result of irritation from urinary tract infection (UTI) or overflowing from an overdistended bladder. Bladder overdistention may also provoke uninhibited contractions in the detrusor muscle, leading to incontinence. In these patients, bladder emptying with intermittent self-catheterization can eliminate recurrent UTI and correct the incontinence. 8. URINALYSIS
Urinary tract infection (UTI) can cause urinary urgency, frequency, and incontinence. A simple urine dip or analysis in the office can diagnose the absence of infection with a specificity of 97% to 99% (36). There is no need for urine culture and sensitivity in the patients with negative urinalysis. However, in those patients with high clinical suspicion, urine culture and sensitivity should be obtained regardless of the urinalysis results. Although urinalysis is excellent at ruling out UTI, its positive predictive value for diagnosing the presence of infection is poor. Patients with positive urinalysis should, therefore, have urine culture and sensitivity results before diagnosing infection as a cause of incontinence.
704
Ho AND BHATIA
Although urine cytology is not obtained routinely, it is useful as a screening test for urinary tract malignancy. Microscopic or macroscopic hematuria should be followed with cytology and cystoscopy. 9. PAD TEST
Pad test can be used to quantify the amount of urine loss; however, it is not always required for the diagnosis of incontinence, except in the evaluation of the efficacy of therapy. Patients are given pre-weighted pads and plastic bags and instructed to wear these pads for 24 hours, change them whenever necessary, and immediately place the wet pads into the zip-top bag. These pads and bags are returned to the office after completion of the 24-hour period for reweighing. A gain of up to 8 grams is within the normal range (44,45). 10. URODYNAMICS
In general, urodynamic studies involve a variety of procedures that together evaluate the structure and function of the lower urinary tract. In this sense, obtaining voiding diary and measurements of voided and postvoid residual volumes are parts of the urodynamic studies. The urodynamic methods range from simple cystometry (or singlechannel urodynamics) to complex multichannel urodynamics and videoradiographic studies (40,46-50). The findings in urodynamics should correlate with clinical observations and reproduce the patient's symptoms; if not, further evaluation is necessary. For example, if the stress test during urodynamic study demonstrates only scant urine loss after several repeated deep coughs while the patient complaints of continuous urine loss in large amounts with any provocative maneuver, the finding is not correlated with the patient's symptoms. Further evaluation of this patient, either to validate the complaint or to look
for other causes, is necessary. For most primary care physicians, the ability to employ and interpret simple cystometry and stress test is important in the management of urinary incontinence. Patients with complicated incontinence or who require multichannel urodynamics and/or cystourethroscopy can be referred to specialists.
a. Single Channel Cystometry Single channel cystometry involves distending the bladder with known volumes of sterile water or normal saline and observing the changes in the bladder function during the filling (40,46). A simple ofrice cystometry consists of a bladder catheter, a large syringe with the plunger removed, and a 500-mL bag of normal saline or sterile water (Fig. 51.4). After the patient voided, the residual volume can be measured by draining it through a 10 or 12 French rubber catheter. The bladder catheterization should be performed in a sterile fashion. The patient's bladder is then gradually refilled with the fluid at a rate of 60-80 mL/min until her functional bladder capacity is reached. During bladder filling, the patient is asked to report her first bladder sensation, initial urge to void, and maximal bladder capacity. Detrusor contraction is indicated by a rise in the meniscus of the fluid in the syringe. The most important observation is the presence of a detrusor reflex and the patient's ability to control or inhibit this reflex. For the normal female bladder, the first sensation of bladder filling occurs at volumes of 150 to 200 mL, and the strong desire to void occurs at 400 to 600 mL (40,46). During filling, an initial rise in detrusor pressure between 2 and 8 cm H20 usually occurs. The maximal cystometric capacity (usually 400 to 600 mL) is the volume that the bladder musculature can tolerate before the patient experiences a strong, uncontrollable desire to urinate. At this point, if the patient is asked to void, a terminal contraction may appear and is seen as a sudden rise in intravesical pressure. At the peak of the contraction, the patient is instructed to inhibit this reflex. A normal person should be able to inhibit this detrusor
FIGURE 51.4 Single channel cystometry. Pves = bladder pressure. (Reproduction from Karram MM. Urodynamics: cystometry. In: Walters MD, Karram MM, eds. Urogynecology and reconstructivepelvic surgery, ed. 2. St. Louis, Mosby, 1999:57.)
CHAPTER51 Lower Urinary Tract Disorders in Postmenopausal Women reflex and thereby bring down intravesical pressure. In urologically or neurologically abnormal patients, the detrusor reflex may appear without the specific instruction to void, and the patients cannot inhibit it; this observation is referred to as an uninhibited detrusor contraction or detrusor overactivity (40,46). These cystometric procedures allow differentiation between patients who are incontinent as a result of uninhibited detrusor contraction and those who have SUI. Conversely, the hypotonic bladder accommodates excessive amounts of fluid medium with little increase in intravesical pressure, and the terminal detrusor contraction is absent when the patient is asked to void. Single channel cystometry is not sensitive, with a poor negative predictive value, in the diagnosis of DO as compared with multichannel urodynamics (47). However, simple cystometry can be combined with a stress test to provide useful tools for the determination of incontinence type in the typical uncomplicated patients (48). If the patient experiences urgency with accompanying urine loss and detrusor contraction during bladder filling (positive cystometrogram) while the stress test is negative, she is likely to have UUI due to DO. If the patient has urine leakage with each time she coughs or strains (positive stress test) and there is no associated detrusor contraction, she has SUI. If both cystometric and stress tests are positive, the patient may have either MUI or UUI. In patients with UUI, a cough can induce detrusor contraction that will lead to urinary leakage with the subsequent cough, mimicking a positive stress test. Thus, in a patient who has strong detrusor contractions followed by a positive cough stress test, further studies are warranted prior to the treatment because this patient may only have UUI rather than MUI. Patients who have complaints that cannot be verified, who exhibit prolonged urine leakage with coughing that suggests a cough-induced detrusor contraction, and who exhibit both stress and urge incontinence require further evaluation with multichannel urodynamics. Patients in whom the cystometric findings do not reproduce their symptoms and in whom cystometric and stress tests do not provide a clear diagnosis also need multichannel urodynamic studies.
b. Multichannd Urodynamics In the single channel cystometry, only intravesical pressure is measured. Changes in the intravesical pressure can be caused by a detrusor contraction, an increase in the abdominal pressure, or both. In order to obtain the bladder pressure exerted only by detrusor activities (Pdet), the intra-abdominal pressure (Pabd) is measured and then subtracted from the total intravesical pressure (Pves) in the multichannel urodynamics (Fig. 51.5). The pressures of the bladder, urethra, and abdomen are recorded simultaneously by using microtransducers, which are incorporated in a small catheter (usually 8 French or less). The intravesical and urethral pressures are measured directly with a sensor-equipped catheter placed in both the
705
bladder and urethra. The intra-abdominal pressure is approximated by measuring either vaginal or rectal pressure. The detrusor pressure (Pdet = Pves - Pabd) is obtained by subtracting the intra-abdominal pressure from the intravesical pressure using electronic equipment. This procedure sometimes described as subtracting cystometry. Urodynamic studies are helpful if the results are interpreted with reference to the findings in patient's history and physical examination. The investigation includes assessment of filling and emptying, urethral pressure profile, leak point pressure, and uroflowmetry (40,46). Assessment of Filling and Emptying: Usually sterile water or normal saline is used to fill the bladder at a rate of 50-100 mL/min. The study can be carried out in the supine, modified lithotomy, sitting, or standing position; however, standing position is preferable because most patients experience incontinence when they are erect. During the filling phase, patients are asked to report the first sensation, strong desire to void, urgency, and maximal bladder capacity as in the single channel cystometry. The stress test is also performed. Provocative maneuvers, such as straining, coughing, heel-bouncing, fast filling of fluid, or listening to running water, are performed in an attempt to produce any urine leakage or uninhibited detrusor contractions. In normally continent patients, the detrusor does not contract even though the patient may have an urge to urinate. The bladder compliance is obtained by dividing the change bladder volume (in mL) by the change in detrusor pressure (in cm H20) (40). Urethral Pressure Profile and Leak Point Pressure: Because continence requires the pressure in the bladder to be lower than the pressure in the urethra, measuring the differences between these two pressures may provide useful clinical information. The urethral pressure profile is obtained by slowly pulling a pressure-sensitive catheter through the urethra, from the urethrovesical ]unction to the urethral meatus. Although the pulling can be performed manually, urodynamic instruments can carry out this procedure automatically with high precision. The urethral pressure profile is a graphic record of pressure along the length of the urethra (Fig. 51.6). From this profile the functional length and anatomic length of the urethra can be calculated (40,49). The maximum urethral closure pressure is obtained by subtracting the intravesical pressure (Pves) from the maximum urethral pressure (Pure). A low urethral closure pressure (Pucp = Pure - Pves) may be found in patients with SUI, whereas an abnormally high urethral closure pressure may be associated with voiding difficulties, hesitancy, and urinary retention. The urethral closure pressure normally varies between 50 and 100 cm H20 (40,49). The leak point pressure is the intravesical pressure, or sometimes intra-abdominal pressure, at the moment when stress incontinence occurs (40,49). The patient is asked to strain in order to increase intravesical pressure gradually, and the lowest pressure at which urine leakage occurs is the
706
Ho AND BHATIA
FIGURE 51.5 Multichannel urodynamics. A. Instrumentation. Abdominal pressure (Pabd), bladder pressure (Pves) and urethral pressure (Pure) are measured directly. Detrusor pressure (Pdet) and urethral closure pressure (Pucp) are electronically derived. Electromyography (EMG) and urine flow studies are also performed. B. Cystometrogram of a 70 year old multiparous woman with the complaint of stress urinary incontinence. She develops urgency at a volume of 70 ml and demonstrates leakage due to detrusor overactivity after several coughs. This patient has no urinary leakage without triggered detrusor spasms. (A, Reproduction from Karram MM. Urodynamics: cystometry. In: Waiters MD, Karram MM, eds. Urogynecologyand reconstructivepelvic surgery, ed. 2. St. Louis: Mosby, 1999:58; B, Reproduction from Theofrastous JP, Swift S. Urodynamic testing. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecology and pelvicfloor dysfunction, ed. 5. Philadelphia: Lippincott Williams & Wilkins, 2003:122.)
Valsalva leakpoint pressure. Patients may cough, with gradually increasing intensity, until urine leakage occurs and the intravesical pressure, or intra-abdominal pressure, at this point is the cough leak point pressure. If urine leakage does not occur at the highest pressure obtainable, this pressure is a reasonable estimation of the urethral sphincteric strength.
An abdominal leak point pressure of less than 60 cm H20 or urethral closure pressure of less than 20 cm H20 is suggestive of the diagnosis of ISD (40,49). However, the clinical applications of urethral closure pressure and abdominal leak point pressure are controversial and should not be the sole determinant for important management issues such as surgical intervention.
707
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women
~-
0
Mid-urethra
I00--
-r-
E 80-o
Q.
g
60-n
40--
,-
20--
Bladde
External meatus
A !
I
I
tract anatomy and function (53). The bladder neck mobility, bladder neck funneling, and urethra or bladder diverticula can be observed simultaneously with the urodynamic profiles. Although this technique is useful in the investigation of obstructive voiding and can show the anatomic level of the obstruction, it adds little information to the uncomplicated patient. The cost and radiation exposure associated with voiding cystourethrogram and videoradiographic urodynamics also obviate their need in the incontinent evaluation of the typical female patient. These techniques are reserved for complicated cases of urinary incontinence.
I
Urethral Length (cm) FIGURE 51.6 Urethral pressure profile. As the pressure sensor is withdrawn through the urethrovesical junction, urethral pressure rises to a maximum urethral closure pressure (MUCP) and then decreases to zero as it exits the urethra. FUL is the functional urethral length. (From Theofrastous JR Swift S. Urodynamic testing. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecology and pelvic jToor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:124.)
Uroflowmetry: Uroflowmetry records the rates of urine flow through the urethra when the patient is asked to void spontaneously. It is a simple test that can be used to document voiding dysfunction objectively. The urine flow rate and profile can be measured alone or concurrently with the detrusor pressure and urethral closure pressure to provide a better assessment of voiding function (50). In this way, poor bladder contraction that results in failure to empty can be identified. These pressure-flow studies help to differentiate voiding dysfunction secondary to obstruction from that secondary to an underactive detrusor. The high detrusor pressure and low flow rate of voiding indicate outlet obstruction as the cause of abnormal voiding. On the other hand, if a patient voids with a low flow rate and minimal or no rise in detrusor pressure, her voiding dysfunction is probably secondary to an acontractile or underactive detrusor.
11. VOIDING CYSTOURETHROGRAM AND VIDEORADIOGRAPHIC URODYNAMICS
In the voiding cystourethrogram, fluoroscopy is used to observe bladder filling, mobility of the urethra and bladder base, and anatomic changes during voiding (51,52). The procedure provides valuable information regarding bladder size and the competence of the bladder neck during increased intra-abdominal pressure with coughing or Valsalva maneuver. It also helps to determine the presence of bladder trabeculation and vesicoureteral reflux as well as the integrity of the urethral sphincter. Videoradiographic urodynamics incorporate fluoroscopy with simultaneous measurement of detrusor and urethral pressures to provide a comprehensive analysis of lower urinary
12. CYSTOURETHROSCOPY
Cystourethroscopy, which involves insertion of an endoscopic instrument into the urethra and bladder, allows the physician to examine inside the urethra, urethrovesical junction, bladder walls, and ureteral orifices (54). This procedure can detect bladder stones, tumors, diverticula, or suture from prior surgeries. Due to the discomfort associated with this procedure, cystourethroscopy is usually reserved for patients with confounding factors such as microscopic or macroscopic hematuria, bladder pain on palpation, suburethral mass, or persistent UTI despite adequate therapy or for patients with urge incontinence who failed the first-line therapy. Some clinicians also advocate cystourethroscopy and multichannel urodynamics in all patients with persistent incontinence prior to surgical treatment.
13. ELECTROPHYSIOLOGIC STUDIES
When the history, physical examination, and urodynamic findings suggest the possibility of nervous system impairment as the cause of incontinence, the adjunctive use of electrophysiologic testing is very helpful. Pelvic floor electrophysiologic testings include surface and needle electromyography, nerve conduction, terminal latency, evoked potential, and reflex response studies. It is beyond the scope of this chapter to discuss these techniques in details; their descriptions can be found elsewhere (55). 14. ULTRASONOGRAPHY
Employing real-time or sector ultrasonography, information can be obtained about the mobility and inclination of the urethra as well as the mobility and funneling of the urethrovesical junction, both at rest and with Valsalva maneuver (52,56). In addition to imaging these structures, bladder and urethral diverticula, periurethral cysts, and flatness of the bladder base can be identified. Ultrasound is also useful in the evaluation of postvoid residual volumes. Recently, the three-dimensional (3D) ultrasound, with its improved resolution, has provided another modality to capture the events related to the bladder, urethrovesical
708
Ho AND BHATIA
junction, and urethra. The simultaneous combination of real-time ultrasonography and urodynamics is under investigation. Ultrasound presents the advantages of low cost and no radiation exposure or reaction to contrast medium. 15. MAGNETIC RESONANCE IMAGING
Magnetic resonance imaging (MRI) is a relatively new and emerging technique for the study of pelvic floor dysfunction and urinary incontinence. It is useful for evaluating abnormalities not readily imaged with other modalities. Whereas fluoroscopic studies provide little information on the anatomic details of the bladder neck, urethra, pelvic floor, and surrounding tissues, MRI produces unsurpassed images of these structures (52). MRI can be useful in assessing urethrovesical junction and urethral abnormalities because the anatomy of these structures is well visualized. This imaging technique can also be employed in the assessment of the pelvic floor dysfunction and organ prolapse. Because of the cost associated with this imaging technique and because the information can also be obtained from clinical evaluation and multichannel urodynamics, the role of MRI in the diagnosis and management of typical female patients with urinary incontinence remains investigative. E. T r e a t m e n t 1. STRESS URINARY INCONTINENCE
Stress urinary incontinence is almost always treatable with nonsurgical or surgical interventions. Nonsurgical therapy can be initiated without urodynamic testing, and primary care physicians should be able to treat a large number of patients. Referrals to specialists should be reserved for patient with severe incontinence, complicated medical or surgical history, unclear diagnosis, or treatment failure and those contemplating surgery. a. Conservative Management
Treatment of Medical Comorbidities and Transient Causes: Conservative management and behavioral intervention are generally recommended as a first step in the treatment of urinary incontinence. Medical conditions such as chronic cough, chronic obstructive pulmonary disease (COPD), and diabetes mellitus should be controlled. All medications that may affect the lower urinary tract functions are carefully reviewed and modified if possible. These medications were discussed in the medical history section. The transient causes of urinary incontinence, as described earlier in the mnemonic DIAPERS, should be identified and treated. These issues are important in the postmenopausal female and elderly populations. Fluid management and dietary modification are also important. Generally, the fluid intake of 6 to 8 glasses, or 2 to 3 liters, per 24-hour period is
adequate. Bladder irritants such as caffeine, tea, carbonated drink, alcohol, chocolate, and certain food ingredients should be minimized or avoided. Absorbent Products: Absorbent pads are the most commonly used product in the elderly population for treatment of urinary incontinence. Although they do not cure or improve the incontinence, these pads are simple and noninvasive to use. Improved products featuring better absorbency, smaller size, greater comfort, and reusable undergarments are now used by younger women as an alternative approach to deal with urinary incontinence (57). Prosthetic Devices: There are two types of prosthetic devices that can be used for anti-incontinence: intravaginal devices and urethral occlusion devices (57). For intravaginal devices, large diaphragms, tampons, and various types of pessaries have been used to elevate and support the bladder neck and urethra. Although ring-type pessaries are traditionally employed for this purpose, the bladder neck support prosthesis (Introl) is very useful. This specially designed pessary consists of two arms at the end that help to support and elevate the bladder neck to the pubic bone in order to prevent urine leakage. Urethral occlusive devices can be inserted into the urethral meatus to achieve mechanical occlusion of the urethra, thus preventing urinary leakage. Placement of these urethral plugs may also stimulate the pelvic floor muscles, which contributes to the therapeutic effects of the devices (57). Other devices such as adhesive patch (Miniguard), urethral cup (Fern-Assist), and urethral catheter (Reliance) have been marketed with suboptimal outcomes. Mthough the prosthetic devices are not very effective, they may provide an acceptable alternative for patients who are unfit for surgery, for elderly women who have difficulty with behavioral and medical therapies, or for younger patients who are awaiting surgery. b. Behavioral Intervention The behavioral therapy includes bladder training, pelvic floor exercises, and biofeedback techniques. Bladder training, which will be discussed in detail later in this chapter, has been shown to be effective in patients with SUI without the presence oflSD (57). Pelvic Floor Exercises: Pelvic floor exercises (Kegel exercises) are known to improve or cure mild forms of SUI. These exercises supposedly increase the strength of periurethral and perivaginal muscles, thus improving anatomic support to the bladder neck and urethra (58,59). Strengthening the striated urogenital sphincter also enhances its ability to compress the urethra. The aim of this physical therapy is to train women to identify their pelvic floor muscles and to contract these muscles in anticipation of increased abdominal pressure during coughing, sneezing, lifting, or physical activities. For the pelvic floor exercises to be effective, they must be supervised, performed regularly, and aided by some form of
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women feedback so that the progress can be followed. The majority of patients are unlikely to benefit from written instruction alone. During the pelvic examination, the patient can learn to correctly isolate her levator ani muscles by contracting around the examiner's fingers. Patient can also insert her own finger into the vagina to detect these muscle contractions. Other adjunct approaches such as electromyography or electrical stimulation are useful in identifying the corrected muscles. After identifying the pelvic floor muscles, patients are instructed to "pull your pelvic in and up" or "squeeze as if you were trying to stop your urine." These contractions should be hold as long as possible and gradually increased up to 10 seconds. Typically, the exercises are performed three to five times per day in sets of 15 to 20 contractions each time. These numbers can be further increased each week. In addition to these exercises, which will strengthen slow-twitch muscle fibers, patients should be instructed to perform a similar exercise with rapid contractions to strengthen the fast-twitch fibers. For those who are unmotivated or have poor strength of contractions, the help of a physical therapist is beneficial. Improvement in incontinence typically takes at least 4 to 6 weeks and sometimes as long as 6 months of continuous exercises (58-60). Pelvic floor muscle exercises are recommended for all women with mild to moderate incontinence, for women who choose not to pursue surgical options, for young women who have not completed childbearing, and for elderly patients who are unfit for surgery. These exercises should also be incorporated into routine health maintenance for postmenopausal women because they are noninvasive, are virtually without side effects, and may prevent the development of pelvic organ prolapse. In general, exercises alone can result in a 56% to 95% reduction in incontinence episodes (60). In an intensive program of pelvic floor exercises for 3 months, Benvenuti and coworkers (61) reported that 32% of patients with G SI were cured and 68% had marked improvement in symptoms. Improvements of bladder neck support and contractility of the pubococcygeus muscle were observed with urodynamic and radiologic evaluations. This study also showed that at 12 to 36 months post-treatment follow-up, 77% of patients still maintained the functional level they had attained at the end of treatment. Pelvic floor exercises before and after delivery may help patients with postpartum urinary incontinence. Pelvic floor exercises require diligence and willingness to practice at home and at work, particularly in incontinenceprovoking situations and daily activities. They require a high level of motivation and compliance. The success of pelvic floor exercises can be enhanced by the use of adjunctive techniques such as vaginal cones, biofeedback, or electrical stimulation. Vaginal Cones: Vaginal cones are a useful adjunction to the pelvic floor exercises. These tampon-shaped devices can be inserted into the vagina for 15 to 30 minutes while the
709
patient is going about her normal activities. The patients are instructed to contract her pelvic floor muscles to keep the cones from falling out. After the lightest cone can be retained, the weights are then increased progressively from 20 to 100 grams. The feeling that the cone is slipping out of the vagina forms a sensory biofeedback to increase contraction of the pelvic floor muscles. Vaginal cones are reported to improve the symptoms of stress incontinence in 60% to 90% of patients (57,59). In a study of 30 postmenopausal women with GSI, Peattie and coworkers (62) reported that 70% of patients were either cured or had significant improvement of symptoms and only 37% had subsequently undergone surgical intervention. Biofeedback Techniques: Other biofeedback techniques can also be used in combination with the pelvic floor exercises. The two most commonly employed devices are vaginal perineometer and electromyographic units (57-59). The perineometer is a probe that can be inserted into the vagina to measure the pressure changes during pelvic muscle contractions. The electromyography helps patients to identify and contract appropriate pelvic muscles in their pelvic floor exercises in response to auditory or visual feedback (57,58). The data on using biofeedback in conjunction with pelvic floor exercises for SUI are conflicting. One meta-analysis showed that combining pelvic floor exercise with biofeedback was more effective than exercise alone (63), whereas another review of randomized trials indicated that the combination was no more effective than pelvic floor exercise alone (64).
c. Electrical Stimulation Therapy Electrical stimulation therapy is another adjunction to pelvic floor exercises. This technique involves the use of vaginal or rectal probes to deriver a low-level electrical current to the muscles and nerves of the pelvic floor (59). Stimulation of the afferent fibers of the pudendal nerve can produce contractions of the pelvic floor and periurethral skeletal muscles, which augment their tone in women with SUI. Simultaneously, the electrical current also causes a reflex inhibition of detrusor activity and improves UUI. Electrical stimulation can be used to increase pelvic floor muscle strength or to identify these muscles for voluntary control in Kegel exercises. Improvement rates of 60% to 90% and cure rates of 10% to 30% were reported in patients with stress, urge, or mixed incontinence (57). Although this mode of therapy remains an option for the treatment of urinary incontinence, patient acceptance of the technique is often poor. Furthermore, the effectiveness of electrical stimulation therapy in SUI has not been formally evaluated. d. Pharmacologic Therapy Estrogen Treatment: Although estrogen can improve vaginal epithelium thickness and vascularity, and possibly urethral function, in the postmenopausal women with urogenital atrophy, the role of estrogen in the treatment of SUI is
710 controversial. Some earlier studies reported promising results; however, they were poorly designed and were not randomized, controlled, and blinded. A meta-analysis of several studies of estrogen therapy showed significant subjective improvements for all patients, including those with SUI (65). These subjective improvements were probably due to the effects of estrogen on the feeling ofweU-being by the subjects. Another study also reported that estrogen was not effective in the treatment of SUI but might be useful for associated symptoms of urgency and frequency (66). Other studies of estrogen for the treatment of GUI in hypoestrogenic and postmenopausal women reported no significant changes in quality of life or objective outcomes (19,67). Recently, the effects of hormone therapy on the symptoms of SUI, UUI, and MUI in healthy postmenopausal women were investigated in a large study (68). The results showed that conjugated equine estrogen, either alone or in combination with medroxyprogesterone, increases the risks of urinary incontinence among continent women and worsens the characteristics of incontinence among symptomatic women after 1 year. A1pha-adrenergics: Alpha-adrenergic activities from the sympathetic nervous system help to maintain continence by increasing the tone of the urethra and bladder neck. Alphaadrenergic stimulants such as ephedrine, pseudoephedrine, or phenylpropanolamine are useful for treatment of SUI. However, they are not uroselective and are associated with systemic side effects, particularly hypertension, a condition that affects many postmenopausal women. Ephedrine and its stereoisomer, pseudoephedrine, can produce anxiety, headaches, and insomnia. Although phenylpropanolamine causes less central nervous stimulation, it can produce hypertension, cardiac arrhythmias, and anxiety. A review of eight randomized controlled trials of phenylpropanolamine demonstrated a low cure rate (0% to 14%) and a wide range in improving SUI symptoms (19% to 60%) (57,69). These data indicate that the effectiveness of phenylpropanolamine in the treatment of SUI is questionable. Duloxefine: Duloxetine is a combined serotonin and norepinephrine reuptake inhibitor. This drug has been shown to enhance the external urethral sphincter contraction and is currently in clinical trials for treatment of SUI. Early results of duloxetine uses in SUI have been encouraging (70-72); however, long-term data are not yet available. Duloxetine also appears to be safe and tolerable; adverse effects are common but not serious (71,72). e. Surgical Treatment
General Considerations: Surgery is the most common type of treatment in SUI because it offers the potential for effective and long-term results. When behavioral and pharmacologic managements fail to improve symptoms of SUI, surgical treatment is the next step. However, in patients with severe SUI or those unable to comply with nonsurgical interventions, surgery may be considered as a first-line treat-
Ho
AND BHATIA
ment. The aim of all surgical procedures is to correct the pelvic relaxation as well as to stabilize and restore the normal position of the bladder neck and urethra and provide adequate support to these structures. Another goal of surgical treatment is to improve the patient's quality of life. The first surgical procedure for SUI treatment is critical because the cure rate may decline with subsequent operations. Every effort should be made for a proper preoperative evaluation before embarking on any kind of surgical procedure for incontinence. A thorough medical history, physical examination, and urodynamic investigation should be performed to determine the cause and type of incontinence. Based on these findings, the choice of surgical procedure should be formalized and discussed with patients. The potential benefits should be weighed against the potential risks of surgery, and the decision resides ultimately with the patient because SUI is not a life-threatening condition and the ability to cope with the incontinence is subjective. Surgical treatment for GSI is not contraindicated in patients with coexistent DO. However, these patients should be counseled that the surgery may cure, may not have any effect, or may worsen the urge incontinence. In patients with both pelvic organ prolapse and potential stress incontinence (masked incontinence), the antiincontinence procedure should be performed concomitantly with the anti-prolapse procedure. A prospective study demonstrated that potential incontinence occurred in 58% of patients with severe pelvic organ prolapse (73). For those patients with potential incontinence who underwent concomitant anti-prolapse and anti-incontinence procedures, 86% had no postoperative stress incontinence. For those patients without demonstrable potential incontinence who underwent only anti-prolapse procedure, no postoperative SUI was observed. The mean follow-up of this study was 44 to 47 months. These data demonstrated that the antiincontinence surgery may not be needed if potential incontinence has been ruled out. Surgical treatment for SUI can be performed vaginally, abdominally or laparoscopically. Vaginal Approach: Vaginal approaches are as follows: Anterior Colporrhaphy with Suburethral Plication: Anterior colporrhaphy (Kelly's plication) is an excellent procedure for correction of cystocele but is less effective for correction of SUI. Anterior colporrhaphy for SUI includes simple plication of the bladder neck, plication of the fascia under the urethra to elevate the bladder neck, or both (Fig. 51.7). This procedure does not hold up well over time, and most of the studies have shown long-term success rates of only 37% to 75% (74), which are quite low compared with other surgical procedures. One series on the revised technique of anterior colporrhaphy showed a success rate of 91% (75). Needle Suspension Procedures: A needle suspension procedure was first described by Pereyra in the 1950s, and since then it has been modified to better anchor the sutures
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women
FIGURE 51.7 Anterior colporrhaphy with suburethral plication. A. Plication of the urethrovesical junction with one to three sutures. B. The plication sutures at the urethrovesical junction are tied. C. A cystocele is reduced with pursestring sutures incorporating vaginal muscularis. D. Completion of the repair and vaginal epithelium is trimmed before closure. (From Weber AM. Surgical correction of anterior vaginal wall prolapse. In: Waiters MD, Karram MM, eds. Urogynecology and reconstructive pelvic surgery, ed. 2. St. Louis: Mosby, 1999:215.)
(74). These procedures involve passage of sutures through the periurethral tissue, near the bladder neck, and then to the suprapubic space and rectus fascia using a specially designed long needle carrier. After these steps are carried out on both sides of the urethra, the sutures are tied over the rectus fascia to suspend the proximal urethra and bladder neck. These procedures have fallen into disfavor due to their poor long-term success rates. In a review of the longterm outcomes (more than 48 months), the needle suspension procedure was found to have 67% success rate whereas retropubic suspension and the midurethral sling have 84% and 83% success rates, respectively (76). In several other studies, the reported 5-year success rates of needle suspension procedures were considerable lower than retropubic approaches or suburethral slings (74). This procedure is simple and takes less time to perform; however, complications such as perforation of the bladder or urethra by the long needle, infections, formation of granulation tissue around the sutures, and nerve entrapment syndrome have been reported (74,76). Suburethral Sting Procedures: The best approach for SUI patients with ISD dysfunction is the suburethral sling procedure. In these patients, the sling supports and compresses the urethra when abdominal pressure is increased,
711
thus compensating for a faulty urethral closure mechanism and preventing incontinence (77). The suburethral sling procedures are suitable for patients who failed retropubic suspension (MMK or Burch) or needle suspension, and, recently, for all types of SUI (78,79). Sling procedures are also recommended for patients with SUI who engage in heavy occupational lifting, who have COPD or chronic cough, and who have post-hysterectomy vaginal vault eversion, because these patients seem to have higher failure rates with other procedures (79). The traditional procedure involves placement of a sling under the bladder neck and urethra and fixation of its both ends to the anterior rectus fascia or other abdominal sites. The sling is adjusted for appropriate tension, under the direct visualization of a cystoscope, to cradle the urethra in a supporting hammock (Fig. 51.8). The sling materials can be autologous tissue (fascia lata or rectus fascia), allografts (cadaveric fascia or dermal graft), xenografts (porcine dermis or porcine intestine), or synthetic meshes (Mersilene, Marlex, Prolene, Gore-Tex, Silastic band) (78,79). Autografts can eliminate the risks of tissue reactions and erosions; however, tissue harvesting increases the operative time and morbidity. Tissue failure can also occur over time with autografts. Allografts and xenografts are associated with durability and tissue reaction problems. Synthetic materials offer better uniformity, consistency, and durability than biomaterials; however, they are more prone to infection and can have significant complications, such as erosion into the vagina, urethra, or bladder. The selection of sling materials remains controversial, and currently there is no ideal graft material. Long-term cure rates for suburethral sling procedures are reported to be 73% to 90% (74,79,80). These cure rates are comparable to the Burch procedure. Although obstructed voiding after the sling placement can occur, this problem usually resolves over time. However, some patients may require long-term bladder drainage with suprapubic catheter or clean intermittent self-catheterization. Tension-Free Mid-Urethral Sling Procedures: Recently, tension-free support of the mid-urethra has quickly gained popularity in the treatment of SUI (81-84). There are two different techniques for the mid-urethral sling procedures: the tension-free vaginal tape (TVT) and the tension-free transobturator tape (TOT). The TVT approach has been gaining wide acceptance as an effective and minimally invasive procedure for treatment of SUI (81,82). Following a small sagittal incision in the vaginal mucosa, a synthetic polypropylene tape is placed at the level of the mid-urethra and extended bilaterally, behind the pubic symphysis, into two separated small suprapubic incisions (Fig. 51.9). There should be a visible space between the tape and the urethra in order to ensure that the urethral support is free of tension. The data accumulated for the past 6 to 8 years showed that the TVT procedure has cure rates of 85% to 90% (81,82). The TVT procedure is less
712
H o AND BHATIA
FIGURE 51.8 Suburethral sling procedure. A. Placement of suburethral sling. B. Attachment of the mini sling to endopelvic fascia or pubic rami bilaterally. C. Attachment of the mini sling to anterior rectus fascia by permanent sutures. D. Attachment of the full length sling to anterior rectus fascia. (A, From Chaikin DC, Blaivas JG. Sling procedures-organic. In: Cardozo L, Staskin D. Textbook offemale urology and urogynaecology. London: Isis Medical Media, 2001:540; B, C, D, From Kohli N, Karram MM. Surgery for genuine stress incontinence. In: Waiters MD, Karram MM, eds. Urogynecology and reconstructive pelvic surgery, ed. 2. St. Louis, Mosby, 1999:183.)
invasive and requires less time than retropubic suspension; however, it can expose patients to a number of possible complications. The blind passage of the trocar and tape using the retropubic route has been associated with several intraoperative complications resulting from penetration into the bladder, urethra, bowels, nerves and vessels, although the rates of these complications are low (81,82). To minimize these complications, a TOT procedure using the prepubic approach has been developed (83,84). This technique takes the full advantages of the tension-free sling in the TVT approach but spares the retropubic space (Fig. 51.10). In general, a 1.5- to 2-cm vertical incision is made in the vaginal wall at the middle third of the urethra. A polypropylene tape is passed from the vaginal incision, through the obturator foramen, and toward the thigh fold to provide a tension-free sling under the mid-urethra. Although no long-term data are available, the effectiveness and safety of this approach during the past 5 years are promising, and the TOT procedure has now become a popular surgical treatment for female SUI. The continent rates obtained with the TOT procedure have been similar to those of the retropubic TVT at least on the shortterm follow-up (83,84). These data also suggest that complication rates can be decreased with the obturator approach.
Abdominal Approach: Abdominal approaches include the following:
Retropubic Suspension: The MarshaU-MarchettiKrantz Vesicourethral Suspension: Most of the abdominal retropubic procedures for suspension of the bladder neck are modifications of the Marshall-Marchetti-Krantz (MMK) vesicourethral suspension or Burch retropubic urethropexy (85,86). The MMK procedure, which was described in 1949, begins with an abdominal incision to gain access to the retropubic space (space of Retzius) and to identify the vesicourethral junction and proximal urethra (Fig. 51.11). Two pairs of nonabsorbable sutures are then placed through the endopelvic fascia on each side of the urethrovesical junction and the proximal urethra and attached to the fibrocartilage of the pubic symphysis. The sutures are adjusted to provide suspension of the bladder neck into their normal intra-abdominal position. Occasionally, the patient may develop osteitis pubis after the MMK procedure. The Burch Retropubic Urethropexy: The Burch retropubic urethropexy, which was described in 1961, is also performed extraperitoneally in the space of Retzius and is similar to the MMK procedure. The retropubic space is entered through an abdominal incision, and sutures are
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal W o m e n
713
FIGURE 51.9 Tension-free transvaginal tape (TVT) procedure. The polypropylene tape is placed under the mid-urethra and without any tension. The tape is then cut just under the skin line. (Courtesy of CR Bard, Inc., Covington, GA.)
FIGURE 51.11 MMK vescicourethral suspension and Burch urethropexy. A. In the MMK procedure, the sutures are placed through a full thickness of vaginal wall, excluding the epithelium, on each side of the bladder neck and then into the pubic symphysis. B. In the Burch procedure, the sutures are placed into the Cooper's (pectineal) ligament. Usually two sutures are placed on each side of the bladder neck. (A, From Mishell DR Jr, Stenchever MA, DroegemueUer W, Herbst AL. Comprehensivegynecology. St. Louis: Mosby, 1997:587; B, From Waiters MD. Retropubic operations for genuine stress incontinence. In: Waiters MD, Karram MM, eds. Urogynecologyand reconstructivepdvic surgery, ed. 2. St. Louis: Mosby, 1999:161.)
FIGURE 51.10 Tension-free transobturator tape (TOT) procedure. The polypropylene tape passed through the obsturator foramen and toward the thigh fold to provide a tension-free sling under the mid-urethra. (From Karram M, Blaivas J, Waiters M. Which sling for which patient? OBG Management 2005;17:68.)
placed in the fascia lateral to and on each side of the bladder neck and proximal urethra. The Burch procedure differs from the MMK procedure in that it involves attachment to the Cooper's ligament (iliopectineal ligament) rather than the pubic symphysis. The advantages of the Burch procedure include easier suture placement, eliminating the risk of pubic osteitis and allowing the correction of small cystourethroceles (85,86).
714
H o AND BHATIA
Abdominal retropubic urethropexy has a long-term success rate of 82% to 90% (85,86). The cure rates for the MMK procedure are similar to the Burch retropubic urethropexy. Both procedures are associated with complications including urethral and bladder injuries, intraoperative bleeding, and postoperative retropubic hematoma. Postoperative urinary tract infection and retention can occur. Although rare, long-term complications such as detrusor instability, de novo UUI, bladder pain, pelvic organ prolapse, and retropubic abscess can also occur. Laparoscopic Approaches: Recent advances in the technique combined with the popularity of operative laparoscopy for various gynecologic procedures have stimulated the use of laparoscopic approaches for retropubic suspension of the bladder neck. The Burch procedure and paravaginal repairs have been successfully performed by laparoscopy (86,87). The advantages of the laparoscopic approach include short hospitalization and minimal invasiveness. However, many surgeons are not skillfully trained for these procedures, and long-term success rates have been disappointing. On the other hand, some reports demonstrated that laparoscopic Burch urethropexy and paravaginal repairs are associated with less patient morbidity and a cure rate similar to traditional procedures (77,86,87). These results are usually from experienced laparoscopists.
f Periurethral Bulking Injections Conventional surgical procedures for SUI sometimes fail in patients with diagnosis of ISD. These patients may be treated by suburethral sling procedures, periurethral bulking injections, or placement of an artificial urinary sphincter. For patients with ISD and a hypermobile bladder neck, suburethral or mid-urethral slings are ideal choices. For ISD without bladder neck hypermobility, use of periurethral bulking injection or an artificial sphincter are recommended. Injection of bulking agent is an attractive alternative to invasive operations for elderly patients who are unsuitable
for surgery or for women who have undergone prior surgery that has failed (74,88). The aim of this procedure is to inject bulking agents into the periurethral space to provide urethral coaptation, thus preventing urinary incontinence under conditions of increased intra-abdominal pressure. Several bulking agents, such as polytetrafluoroethylene (Teflon), glutaraldehyde cross-linked bovine collagen (Contigen), and carbon-coated microbeads (Durasphere), have been employed (74,88). Teflon is hard to inject and has been shown to migrate to other parts of the body under some circumstances. Contigen is easier to inject but requires skin testing for possible allergic reactions before the injection. Recently, ethylene vinyl alcohol copolymer suspended in a dimethyl sulfoxide carrier (Tegress, Uryx) has been developed and has shown to be a promising bulking agent (88); however, clinical trials are currently in progress and no published data are yet available. There are two different techniques for the injection. The transurethral technique involves injecting a bulking agent into the submucosa of the proximal urethra, under the direct visualization of a cystoscope (Fig. 51.12). Usually, the injection is carried out at 3:00 and 9:00 positions. For the periurethral injection technique, the needle is advanced in the submucosa, parallel to the urethra, to the area near the bladder neck. Bulking agent is injected under direct visualization until occlusion of the urethrovesical junction is observed. The injection is usually carried out at 3:00 and 9:00, similar to the transurethral technique. Bulking agent injections are a reasonable alternative to a suburethral sling procedure in women with a well-supported but poorly functioning urethra. Patient satisfaction rates up to 80%, marked improvement rates of 50% to 60%, shortterm cure rates up to 70%, and long-term cure rates of 20% to 30% with collagen injections have been reported (74,77,88). These procedures may be associated with complications, such as injury to the bladder or urethral wall, bleeding from the injection site, urinary retention, voiding dysfunction, urge
FIGURE 51.12 Periurethral bulking injection. (From Kohli N, Karram MM. Surgery for genuine stress incontinence. In: Walters MD, Karram MM, eds. Urogynecology and reconstructive felvic surgery, ed. 2. St. Louis: Mosby, 1999:187.)
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women incontinence, and urinary tract infection. The dissolution of the collagen agent requires repeat applications, usually 2 to 3 months apart. It has been reported that after two injections, the effects can last from 3 months to a few years (77).
g. Artificial Urinary Sphincter Although rare, some patients with severe incontinence may have a scarred, fibrotic, and nonfunctional urethra that is very difficult to treat. These patients may benefit from periurethral bulking injection, placement of a deliberately obstructive sling, or implantation of an artificial urinary sphincter. Periurethral injections often fail in these cases, and an obstructive sling requires lifetime self-catheterization. Artificial urinary sphincter implantation is an attractive mode of therapy for these patients because the device allows urethral obstruction to prevent urinary leakage. This obstruction can be voluntarily relieved at the time of voiding. As reported in two series of 66 patients, the success rates for incontinence were 91% to 100% (74). Mechanical failure with the device can occur and requires surgical repair. 2. URGE URINARYINCONTINENCE
In treating urge incontinence, it is important to exclude significant outflow obstruction that may precipitate DO. Although the majority of women with UUI have IDO, treatable conditions such as bladder inflammation or irritation due to infection, calculi, or foreign bodies should be identified and managed properly. In addition to the attention to medical comorbidities and transient causes of urinary incontinence, as described earlier, the treatment of UUI includes behavioral modifications, pharmacologic therapy, electrical stimulation, intravesical therapy, and surgical intervention. Among these options, behavioral therapy and pharmacologic treatment are the main approaches.
a. Behavioral Therapy Behavioral therapy was recommended as a first-line treatment for UUI (57,60). The best approach is to incorporate behavioral therapy into the longterm management because most of the drugs used in treating UUI have unpleasant anticholinergic side effects and patients will likely discontinue these medications over time. Behavioral therapy also has good successful rates and can help to enhance the effects of pharmacologic treatments. The therapy includes fluid management and bladder training. Fluid management: Depending on the patient condition, fluid management may involve restriction or increase in fluid intake. It is generally recommended that the fluid intake should be about 6 to 8 glasses, but not exceeding 2 to 3 liters, in a 24-hour period (60). However, too much restriction in fluid intake can produce highly concentrated urine and constipation, which subsequently irritate the bladder and aggravate the symptoms of UUI. Bladder irritants such as caffeine, tea, carbonated drinks, and chocolate should be avoided.
715
Bladder Training: Bladder training involves deferred voiding, timed voiding, and desensitization programs. The aim of these techniques is to increase functional capacity of the bladder and control urgency. Bladder training represents a behavioral modification designed to repeat the process of toilet training. It is based on the assumption that DO is caused by the loss of cortical control over the detrusor muscle. In the deferred voiding technique, the patient delays urination for as long as possible when a feeling of urge occurs. This can be achieved with various distraction methods such as deep breathing, changing posture, or quick and repeated contractions of the pelvic floor muscles to inhibit detrusor activities (57,60). Timed voiding is another technique that can benefit patients with DO and urge symptoms. The essential aim is to increase bladder capacity week by week and to prolong the intervals between voids. Based on the voiding diaries, the shortest interval is selected and the patient is asked to void at this fixed interval, regardless of whether she has the urge feeling. For example, if the interval is 1 hour, then the patient has to void every hour even she does not feel the need to void. If the patient is desperate to urinate before the scheduled time, she must wait even if urinary leakage occurs. The objective of this bladder retraining is to make the patient's bladder do what she wants it to do. This fixed interval is gradually increased by 15 minutes every week to achieve a period of 3 hours between voids and with less urgency. At night, the patient is allowed to void only when she is awakened from sleep by the problem. Compliance is critical for the success of timed voiding method. N D O does not respond well to this behavioral therapy. The desensitization technique is useful for patients with urgency that is triggered by an event such as water running or rising from a seated position. The patient is asked to delay voiding while gradually exposed to these events. b. Pharmacologic Treatment Medications employed in the treatment of UUI involve those that decrease contraction of the detrusor muscle and those that act on both the detrusor and the urethral sphincter together. It is best to start with a low dose, particularly in older patients. It is also reasonable to try several drugs, one at a time, and/or to increase the dose up to the maximal tolerable level, in order to find the most effective drug and its optimal dose for each patient. In general, patients with N D O require more medication than those with IDO. Anticholinergic Drugs: Anticholinergic medications are useful in the treatment of UUI because ace@choline is the primary neurotransmitter involved in the contraction of the detrusor muscle. These medications act by inhibiting the cholinergically innervated detrusor to reduce its hyperactivities. The systemic side effects of these drugs, such as dry mouth, constipation, blurred vision, increased heart rate, and
716 drowsiness, may limit their uses. These medications are contraindicated in patients with untreated narrow-angle glaucoma. The most frequently employed agents are oxybutynin chloride (Ditropan, Ditropan XL) and tolterodine tartrate (Detrol, Detro LA). Recently, darifenacin (Enablex), solifenacin (Vesicare), and trospium chloride (Sanctura) have been approved for the treatment of UUI and OAB. Hyoscyamine sulfate (Levsin), propantheline (Pro-Banthine), and dicyclomine (Bentyl) are also used for the treatment of UUI. Oxybutynin reduces detrusor hyperactivity by its antispasmodic activities, and in randomized and placebo-controlled trials, the cure or reduction of UUI was reported in 9% to 56% of patients (69,89). Similar to other anticholinergic agents, its use is frequently associated with central nervous effects and dry mouth. Ditropan XL, a slow-release formulation for single dose daily, was reported to produce a better pharmacokinetic profile and fewer systemic side effects than the immediaterelease formulation (69,89). The transdermal delivery oxybutynin formulation is also available for twice-weekly applications. In a randomized and placebo-controlled study, this formulation was shown to be effective in the treatment of UUI or MUI but with fewer systemic side effects, except for local skin irritation (90). Tolterodine is another anticholinergic agent that is more selective for the bladder. In a meta-analysis of randomized trials, tolterodine was shown to be less effective but to have a better side effect profile than oxybutynin (91). Patients who received tolterodine were better tolerated, had less dry mouth, and were less likely to withdraw from the studies due to side effects than those given oxybutynin. However, oxybutynin produced a greater increase in voiding volume and a lesser number of incontinence episodes per 24 hours than tolterodine. The long-acting tolterodine (Detrol LA) is a single-dose daily formulation and provides better patient compliance. A comparative study of the efficacy and safety of Detrol LA and transdermal oxybutynin showed that both formulations are effective treatments for patients with UUI or MUI. However, transdermal oxybutynin formulation was associated with fewer systemic side effects (90). Solifenacin, darifenacin, and trospium chloride are three new anticholinergic agents for the treatment of UUI and AOB (89). Solifenacin and darifenacin seem to have high selectivity for the M3 receptors in the bladder, whereas trospium may involve multiple subtypes of muscarinic receptors, including M2 and M3. Although solifenacin is marginally more effective than tolterodine, it is well tolerated and has a better side effect profile. In clinical trials, solifenacin treatment has been associated with statistically significant reductions in all key symptoms of OAB, while adverse effects are few and usually mild in nature (89). Darifenacin is also effective for the treatment of OAB; however, its uroselective properties have not yet been confirmed in clinical studies. Trospium chloride, which is a quaternary amine, has the lowest ability to penetrate the blood-brain barrier
Ho AND BHATIA
among anticholinergic agents and, therefore, produces favorable CNS tolerability. In addition, trospium is not metabolized extensively by the liver but excreted largely as an active compound in the urine, thus minimizing the systemic effects. In clinical investigations, this medication significantly improved the symptoms of UUI and showed minimal adverse effects, except for dry mouth, constipation and headache, when compared with placebo. The reported side effects of hyoscyamine sulfate are conflicting, with some authors promoting its use as an initial agent, whereas others rarely use it because of adverse systemic effects (69). There are no studies that adequately compare its effects with placebo. Two studies reported that propantheline produces a 13% to 17% reduction of UUI symptoms over placebo (29). TricyclicAntidepressants:Tricyclic antidepressants, such as imipramine (Tofranil) and doxepin (Sinequan), relax the detrusor muscle by virtue of their anticholinergic action and increase bladder outlet resistance by their action on the alpha-adrenergic receptors. These drugs are, therefore, useful in patients with MUI. For treatment of UUI, the reported cure rate of these drugs was 31% and the improvement rates were 20% to 77% (69). Randomized and placebo-controlled studies also revealed the effectiveness of doxepin and imipramine in reducing nocturnal enuresis in patients with UUI (92,93). The side effects of these agents include fatigue, dizziness, blurred vision, nausea, and insomnia. Imipramine may cause orthostatic hypotension and cardiac arrhythmias. Other Pharmacologic Agents: Beta-sympathomimetic agonists have also been employed in the treatment of UUI. The detrusor-relaxing action of beta-sympathomimetics forms the basis for the use of drugs such as metaproterenol (Alupent). They also enhance the effect of propantheline (69). Musculotropic drugs, such as flavoxate (Urispas), act by causing direct relaxation of the detrusor muscle (29,69). Diazepam (Valium) acts by a combination of direct smooth muscle relaxation, anticholinergic effect, and central nervous system sedation (69). c. Sacral Neuromodulation and Electrical Stimulation Therapy Recently, implantable sacral neuromodulation
has been employed in refractory cases of UUI with good success (94). The reported cure rates range from 47% to 56%, although potential bias may exist in these studies. These cure rates seems to be low; however, it should be noted that the patients have failed other therapies and sacral neuromodulation is the only available option. This mode of therapy may cause detrusor relaxation through its activation of the afferent fibers, which inhibits the spinal and supraspinal signals. Sacral neuromodulation may also cause a reflex relaxation in the detrusor muscle by activation of the efferent fibers to the striated urethral sphincter. However, the exact mechanism of neuromodulation in the treatment of UUI is not known. Sacral neuromodulation
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women has been shown to be an effective mode of therapy for the patient with UUI who has a history of poor response to other therapies (94-96). Detailed description of the sacral modulation and implanting technique can be found elsewhere (94). Other techniques of electrical stimulation, such as transcutaneous electrical nerve stimulation (TENS) and percutaneous posterior tibial nerve stimulation (PTNS), offer additional alternatives to the treatment of UUI. They are less invasive and have been shown to improve the symptoms of UUI (29).
d. Intravesical Injection of Botulinum Toxin Botulinum toxin has been investigated for the treatment of DO. When injected into the bladder wall, botulinum toxin prevents the release of acetylcholine and, therefore, inhibits detrusor contractions. Botulinum toxins A and B have been used for N D O and IDO with varying success (97). Although this mode of therapy is not yet an approved treatment for UUI, available data suggest that intravesical injection ofbotulinum toxin can be a therapeutic option in patients with N D O and IDO who are refractory to anticholinergic medications. e. Surgical Treatment Generally, surgery is not a treatment option for UUI or OAB, except in the patients who are refractory to pharmaceutical, behavioral, or other nonsurgical management. In these patients, surgical procedures such as bladder denervation, bladder augmentation, and urinary diversion can be tried (29). These procedures are associated with significant morbidity, and their usefulness remains controversial. They should be considered the last resort in the treatment ofUUI or OAB.
717
treatment for MUI patients with predominant stress incontinence (22). Tolterodine, an anticholinergic agent, has been shown to significantly reduce urge incontinence, but not the stress component, in patients with MUI (22). The transdermal formulation of oxybutynin and the long-acting form of tolterodine (Detrol LA) are also effective in the treatment of MUI patients who have predominant urge symptoms (90).
b. Surgical Treatment In some patients, surgery to suspend the bladder neck cures not only stress but also the urge component of MUI (22). In other patients, surgical treatment for the stress component of MUI does not have any effect or may even worsen the urge component (22). Considerable debate also exists regarding the role of suburethral sling procedures in women with MUI (79). However, there is some evidence that surgical treatment may relieve symptoms of DO and provide a better prognosis if stress symptoms predominate and predate urge symptoms (98,99). For patients with pure DO, bladder neck suspension and sling placement are contraindicated. A comprehensive clinical and urodynamic evaluation is, therefore, essential before making the decision for the surgical intervention. Recently, tension-flee suburethral sling procedures, such as TVT, have been proposed for the treatment of MUI, and cure rates of 66% to 85% were reported (22,100). However, a retrospective study on the long-term results of the TVT procedure for MUI reported that the cure rates continued to maintain at 60% for 4 years postoperatively and then declined to 30% at 8 years after the surgery (101). The decline in cure rates was due to the increases of urgency symptoms.
IV. INTERSTITIAL CYSTITIS 3. MIXED URINARYINCONTINENCE
MUI is a coexistence with both SUI and UUI. The treatment of MUI is challenging, and controversy exists regarding the best approach. Some authors advocate the use of conservative therapy first because surgery for MUI is not as effective as for pure stress symptoms, whereas others recommend that treatment be based on the predominated symptoms.
a. Nonsurgical Therapy Because pelvic floor exercises are recommended for both SUI and UUI, these exercises can be employed in MUI. Some vaginal support devices designed to treat SUI have modest efficacy in the treatment of MUI (57). Electrical stimulation can also be used to treat MUI, and improvement rates of 60% to 90% and cure rates of 10% to 30% have been reported (57). Pharmacologic approaches to the treatment of MUI include serotoninnorepinephrine reuptake inhibitors and anticholinergic agents. Duloxetine, a combined serotonin-norepinephrine reuptake inhibitor, is currently in clinical trials as a potential
Interstitial cystitis (IC) is a poorly defined clinical syndrome that manifests as urinary frequency, urinary urgency, nocturia, and bladder pain without an identifiable etiology. Until recently, most of the women with these symptoms were presumed to have UTI, overactive bladder, or chronic pelvic pain and were often treated accordingly. IC is now gaining recognition, and the diagnostic methods as well as treatment modalities are emerging.
A. Epidemiology Despite recent advances, the epidemiology of IC remains incompletely understood. IC affects mainly women, with a female to male ratio of approximately 10:1 (102). Reports on the prevalence of IC are conflicted depending on the criteria used for the diagnosis and method of study. In a population-based study, the estimated prevalence o f l C was 66 per 100,000 adult American females (103). Using the validated questionnaires, other studies demonstrated
718
Ho AND BHATIA
that the estimated prevalence of IC may range from 0.23% to 12.6% (104,105). Parsons and coworkers suggested that the prevalence of IC may be as high as 20% (106). Using potassium sensitive test as a diagnostic method, IC prevalence rates of 81% to 85% in gynecologic patients with complaints of pelvic pain were also reported (107,108). Although prevalence estimates vary greatly, IC is obviously much more common than traditionally believed.
B. Pathophysiology Currently, there are no proven etiologies for IC, and the pathophysiology of this disease is not completely understood. The etiology of IC is probably multifactorial rather than a single entity because a variety of etiologies have been proposed, but none adequately explains the variable presentations, clinical courses, or responses to therapies. It is also possible that different patients may have different etiologies. Recently, several studies reported the associations between IC and other disorders, such as endometriosis, vulvodynia, chronic pelvic pain, irritable bowel syndrome, inflammatory bowel disease, fibromyalgia, chronic fatigue syndrome, systemic lupus erythematous, Sj6rgen's syndrome, and allergies (109). These associations suggest that IC may be a systemic syndrome or a genetically predisposed disease. The suggested genetic predisposition to I C is further supported by the observation that the concordance rate of IC among monozygotic twins was greater than among dizygotic twins (110). One of the earliest theories suggested that IC is a result of a "leaky" urothelium in the bladder caused by the deficiency of the glycosaminoglycan (GAG) layer. The presence of a GAG layer in the urothelium helps to protect the bladder against both bacteria and urinary toxins (111). When urothelium in the bladder becomes more permeable to these bacteria and toxins, IC can develop. This theory, which forms a basis for potassium chloride test and pentosan polysulfate therapy in IC, was derived from several studies in animal models and humans that used protamine to induce an increased permeability in urothelium by stripping the GAG layer (112). However, whether there is an increase in the permeability of the IC urothelium and whether the penetration of bacteria and urinary toxins can be prevented by the GAG layer are still debatable (113). Another proposed etiology is the augmentation of the sensory function in the urothelium of IC patients (114). This theory was based on the findings that the bladder urothelium may have a sensory function in addition to providing a protective barrier. Altering the production of peptide growth factors, such as antiproliferative factors (APF) and heparin-binding epidermal growth factor, in the urothelium was also proposed to explain the etiology of IC (115). These factors may inhibit the growth or the regeneration of normal bladder urothelium. Although much
more investigation is needed to elucidate the mechanism, these growth factors may provide the basis to develop a urinary test for IC. The degranulation of mast cells causing the release of neuroactive and vasoactive agents in the bladder is also believed to play an important role in IC (116). This proposed mechanism forms a basis for the use of antihistamines such as hydroxyzine in the treatment of IC. Other proposed theories include neurogenic hypersensitivity or inflammation mediated locally at the bladder or spinal cord level, release of substance P leading to inflammation and pain, and chronic bladder infection with a poorly characterized agent (116). C. Evaluation 1. CLINICAL PRESENTATION
The symptoms of IC, such as urinary frequency, urinary urgency, and bladder pain, overlap with the symptoms of overactive bladder or chronic pelvic pain. The main differences between these conditions are that overactive bladder does not have bladder pain and chronic pelvic pain may not have urinary frequency and urgency. The presentation of IC symptoms is highly variable from one patient to another and does not necessarily follow a set pattern. Generally, IC patients complain of urinary urgency, urinary frequency, nocturia, and pelvic pain that do not have identifiable etiologies. Other patients may complain of vague symptoms of pain and incomplete bladder emptying or a constant sensation to void. These symptoms can wax and wane during the course of the disease, and one symptomatic component may predominate over the others. IC is characterized by periods of exacerbation followed by variable periods of remission, and the spontaneous remissions, although temporarily, occur in as many as 50% of patients at a mean of 8 months (116). The pain component of IC is nonspecific because the bladder is autonomic innervated and patients often have difficulty in locating or describing the pain. The typical pain over the bladder (suprapubic) area that can be relieved by voiding is described only by a small number of patients, whereas the majority of the women with IC may complain of referral pains such as urethral pain, back pain, vulvar pain, rectal pain, dyspareunia, dysuria, constant burning sensation, or general pelvic pain. Nocturia and urinary frequency and urgency are other components of IC. These symptoms are difficult to separate from each other and from the pain complaint because urinary frequency can be a result of urgency or bladder pain. IC patients may also have a constantly strong urge to void, despite low bladder volume, which is often described as pain. The typical patient voids 16 times a day and 2 or more times at night; however, in severe cases, the patient may urinate as often as 60 times a day and every half hour at night (106). Voided volumes are usually small. Q.uality of life is severely
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women impaired in these women. The patient may be anxious, depressed, angry, and sleep deprived, which can exacerbate the pain and urinary symptoms. 2. DIAGNOSIS
The ability to diagnose IC definitely does not exist because the proven etiology, typical physical examination findings, and conclusive tests or markers are not available. Furthermore, the symptoms are highly variable and can wax and wane. The key to diagnosis is to have highly clinical suspicion, exclude other identifiable conditions, and quantify the symptoms as objectively as possible. Currently, the clinical diagnosis o f l C is made primarily by the combination of symptomatic features and exclusionary criteria. Although the dilemma still exists in the diagnosis of IC, the investigation described in this section can be employed together with the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) exclusionary criteria and questionnaire instruments to guide the evaluation and treatment while awaiting further studies on the pathophysiology of lC.
a. The N I D D K Criteria In 1988 a list of criteria was formulated by the N I D D K to standardize subject selections for the research purposes (117). These criteria are summarized in Table 51.1. Although these criteria were not designed as a diagnosis tool, they can provide the clinicians with a useful guideline. However, recent studies showed that 60% of IC cases were missed if the diagnosis relied only on these criteria (118). Furthermore, the early IC may not show a reduction in bladder capacity and glomerulation, thus excluding these cases from the N I D D K criteria. The challenge to clinicians is to diagnose and treat IC early because some authors have reported that the more severe and advanced stages of I C are more resistant to current therapies (109,116). The usefulness of the N I D D K criteria in clinical practice is, therefore, limited. b. Physical Examination and Laboratory Studies General and pelvic examinations are performed to rule out other diseases and pelvic pathologies. Typically, the pelvic examination is negative in the IC patients, except for suprapubic and/or trigonal tenderness. Sexually transmitted diseases, urinary tract infections, urethral diverticulum, and pelvic masses should be ruled out. During the pelvic examination, specimens are collected for the determination of sexually transmitted diseases if clinically indicated. Urinalysis and urine culture are warranted, although the results are usually negative. Urine cytology should be obtained if microscopic hematuria is present or if the patient has other risk factors such as a history of smoking and is more than 40 years old.
719 TABLE 51.1 NIDDK (National Institute of Diabetes and Digestive and Kidney Diseases) Diagnostic Criteria for Interstitial Cystitis
Inclusion criteria 9 Symptoms of urinary frequency/urgency or bladder pain 9 Presence of glomerulations (in at least 3 quadrants with at least 10 glomerulations per quadrant) and/or Hunner's ulcers on cystoscopy and hydrodistention Exclusion criteria 9 Maximal bladder capacity greater than 350 ml on cystometry while patient is awake 9Absence of an intense urge to void with the bladder filled to 150 ml of water during cystometry using a fill rate of 30 to 100 ml/min 9 Demonstration of phasic involuntary bladder contractions on cystometry using the fill rate of 30 to 100 ml/min 9 Duration of symptoms less than 9 months 9Absence of nocturia 9 Symptoms relieved by antimicrobials, urinary antiseptics, anticholinergics, or antispasmodics 9 Frequency of urination while awake of less than 8 times a day 9 Diagnosis of bacterial cystitis in the past 3 months 9 Bladder, ureteral or urethral calculi 9Active genital herpes 9 Uterine, cervical, vaginal, or urethral cancer 9 Urethral diverticulum 9 Cyclophosphamide or any type of chemical cystitis 9Tuberculous cystitis 9 Radiation cystitis 9 Benign or malignant bladder tumors 9Active vaginitis 9 Age less than 18 years
c. Voiding Diary and Symptoms Scale A voiding diary, as described earlier for urinary incontinence, is very helpful in establishing baseline frequency, urgency, and bladder capacity in IC patients. These patients should also keep a voiding diary before and after treatment, as well as during any flare-up, to document the improvements and identify the triggers. Several patient symptom scales have been developed and validated for IC symptom quantification. These instruments include the O'Leary IC Indices (Table 51.2), the University of Wisconsin IC Scale (UW-ICS), and the Pelvic Pain and Urinary Urgency and Frequency Scale (PUF). The O'Leary IC Indices have two components: the IC Symptom Index (ICSI), which quantifies the symptoms, and the IC Problem Index (ICPI), which quantifies the quality of life (120). Using the cystoscopic and hydrodistention findings, and NIDDK criteria as the objective diagnosis of IC, the sensitivity, specificity, positive predictive value, and negative predictive value of ICSI and ICPI were found to be 94%, 50%, 53%, and 93%, respectively (121). The [A/V-ICS (Table 51.3), which is based on how much the patients experienced the symptoms, has also been developed and validated for IC (122). Recently, PUF scale (Table 51.4) has become more
720
Ho TABLE 51.2 The O'Leary-Sant Symptom Index and Problem Index for Interstitial Cystitis
Interstitial Cystitis Symptom Index (ICSI) 1. During the past month, how often have you felt the strong need to urinate with little or no warning? 0. not at all 1. less than 1 time in 5 2. less than half the time 3. about half the time 4. more than half the time 5. almost always 2. During the past month, have you had to urinate less than 2 hours after you finished urinating? 0. not at all 1. less than I time in 5 2. less than half the time 3. about half the time 4. more than half the time 5. almost always 3. During the past month, how often did you most typically get up at night to urinate? 0. none 1. once 2. 2 times 3. 3 times 4. 4 times 5. 5 or more times 4. During the past month, have you experienced pain or burning in your bladder? 0. not at all 1. a few times 2. almost always 3. fairly often 4. usually
Interstitial Cystitis Problem Index (ICPI) During the past month, how much has each of the following been a problem for you: 1. Frequent urination during the day? 0. no problem 1. very small problem 2. small problem 3. medium problem 4. big problem 2. Getting up at night to urinate? 0. no problem 1. very small problem 2. small problem 3. medium problem 4. big problem 3. Need to urinate with little warning? 0. no problem 1. very small problem 2. small problem 3. medium problem 4. big problem 4. Burning, pain, discomfort, or pressure in your bladder? 0. no problem 1. very small problem 2. small problem 3. medium problem 4. big problem Adapted from res 120.
AND BHATIA
popular as a simple but reliable symptom index for I C screening (106). This scale has been validated, and the scores are strongly correlated with a positive PST. Consequently, a high PUT score (10 or above), in combination with a history and physical examination suggesting IC, is considered by many clinicians to be sufficient to make presumptive diagnosis (106). Any of these validated instruments can be used to quantify the patient symptoms and help in the evaluation and treatment of IC as well as in the following patient improvement as objectively as possible.
d. Cystoscopy and Hydrodistention Cystoscopy is generally recommended for ruling out bladder neoplasm and other pathologies. The bladder capacity under anesthesia can also be determined, and the reduction in this capacity suggests the presence of IC, although this criterion is not reliable because many IC patients may have normal anesthetic bladder capacity. The disadvantages of this procedure are its invasiveness and requirement of anesthesia. A number of investigators advocate bladder biopsies to confirm the presence of inflammation; however, the morbidity of the procedure outweighs any potential benefits because histologic findings are not specific for IC. Hydrodistention can be carried out at the time of cytoscopy under general or regional anesthesia. The bladder is fully examined first with cystoscopy to rule out any abnormal appearing or evidence of lesions. The bladder is then distended with sterile water or normal saline at a pressure of 80 to 100 cmH20 for 2 to 5 minutes. After that, the bladder is emptied, and bloody efflux of irrigant suggests the presence of IC. Repeat cystoscopy is performed to reexamine the bladder epithelium, and any glomerulation and/or Hunner's ulcers are noted. Bladder rupture can occur in these patients, so careful inspection during filling and avoiding overdistention are crucial. The presence of glomerulations and/or Hunner's ulcer in the bladder on hydrodistention is the specific N I D D K criteria for the diagnosis of IC. In the United States, Hunner's ulcers (ulcerative form) only occur in less than 10% o f l C patients, and some authors consider it to be more resistant to therapy (116). Although they are uncommon, the presence of Hunner's ulcers is thought to be a specific sign for IC. The appearance of glomerulation after hydrodistention is more common; however, its specificity for the diagnosis of IC is subject to debate because the incidences of glomerulation between IC patients and asymptomatic controls were not significantly different, as reported by one study (123). The sensitivity of glomerulation findings in the diagnosis of IC is also questionable (116). Hydrodistention may have a therapeutic effect on the IC symptoms, although a randomized trial to compare hydrodistention with a sham cystoscopy (without hydrodistention) has not been performed. Some patients may have a temporary worsening of symptoms after hydrodistention, before getting better. Chai and coworkers (124)
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women TABLE 51.3
721
University of Wisconsin Interstitial Cystitis Scale
Interstitial Cystitis Items How much have you experienced the following symptoms today? (0: not at all; 6: a lot)
1. 2. 3. 4. 5. 6. 7.
0
1
2
3
4
5
6
0
1
2
3
4
5
6
Bladder pain Bladder discomfort Getting up at night to go to the bathroom Going to the bathroom frequently in the day Urgency to urinate Difficulty sleeping because of bladder problems Burning sensation in the bladder
Reference Items How much have you experienced the following symptoms today? (0: not at all; 6: a lot)
8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.
Other pelvic discomfort Backache Abdominal cramps Dizziness Aches in joints Heart pounding Chest pain Headache Nausea Numbness, tingling Blind spots, blurry vision Sore throat Swollen ankles Nasal congestion Coughing Suffocation Ringing in ears Flu Adapted from res 122.
suggested that the bladder stretching during hydrodistention can normalize growth factors, which are abnormally expressed in the IC patients, resulting in improvement of symptoms. Hunner's ulcer may be treated at the time of cystoscopy by laser fulguration (125). e. Potassium Sensitive Test The potassium sensitive test (PST) is based on the theory that the bladder epithelium in IC patients is leaky because of the deficiency in the GAG layer (106,126). If potassium is present in the urine, it will cross the leaky urothelium to activate the sensory nerve endings in the suburothelium and cause pain. This test is performed while the patient is awake and without anesthesia. About 40 mL of sterile water is infused into the bladder at a rate of 15-20 mL/min, and 5 minutes after completion of the infusion, the patient rates her pain and urgency using a visual scale from 0 to 5, with 5 being the worst. The bladder is then emptied, and 40 mL of a solution of 0.4 M potassium chloride (KC1) is instilled into the bladder and kept for 5 minutes in the same fashion as with sterile water. The patient then rates her pain and urgency using the same visual scale before voiding. The test is considered positive if
the patient is asymptomatic with water instillation and has a score of ->2 in either pain or urgency with KC1 solution (106). This test is quite painful in IC patients, and the bladder should be emptied immediately after completion of the test. Subsequent irrigation with sterile water or rescue therapy may be necessary. The positive PST was obtained in 75% to 78% of patients with IC (based on N I D D K criteria) and in 4% of controls (106,126). However, a negative PST does not rule out IC because up to 46% of patients with a negative PST meet the N I D D K criteria for IC (127). In gynecologic patients with chronic pelvic pain, 81% to 85% had positive PST (107,108) whereas 38% had positive findings for IC with cystoscopy and hydrodistention (121). This observation suggests that PST is more sensitive for IC, and most gynecologic patients with chronic pelvic pain may have this condition. However, in another study on a population who had symptoms suggestive of IC, the positive predictive values of PST and cystoscopy with hydrodistention were not significantly different (56% and 66%, respectively) (127). A false-positive result can be caused by infection or prior exposure to radiation or chemotherapy. The PST may help
722
Ho TABLE 51.4
AND BHATIA
Pelvic Pain and Urgency/Frequency (PUF) Scale for Interstitial Cystitis
Please circle the answer that best describes how you feel for each question
How many times do you go to the bathroom during the day? 2 a. How many times do you go to the bathroom at night? b. If you get up at night to go the bathroom, does it bother you? 3 Are you sexually active? Yes No 4 a. If you are sexually active, do you now or have you ever had pain or symptoms during or after sexual intercourse? b. If you have pain, does it make you avoid sexual activity? 5 Do you have pain associated with your bladder or in your pelvis (vagina, labia, lower abdomen, urethra, perineum, testes, or scrotum)? 6 a. If you have pain, is it usually b. Does your pain bother you? 7 Do you still have urgency after going to the bathroom? 8 a. If you have urgency, is it usually b. Does your urgency bother you? Symptom Score (1, 2a, 4a, 5, 6, 7a, 8a) = B o t h e r Score (2b, 4b, 7b, 8b) = Total Score (Symptom Score + Bother Score) =
Score 0
Score 1
Score 2
Score 3
Score 4
3-6
7-10
11-14
15-19
20+
0
1
2
3
Never
Occasionally
Usually
Always
Never
Occasionally
Usually
Always
Never
Occasionally
Usually
Always
Never
Occasionally Mild Occasionally
Usually Moderate Usually
Always Severe Always
Never
Occasionally Mild
Usually Moderate
Always Severe
Never
Occasionally
Usually
Always
1
Never
4+
PUF Symptom Scale. 9 2000 C. LowellParsons,MD (From ref. 106).
to predict the response to pentosan polysulfate therapy. Patients with a positive PST seem to have a better response than those who have a negative PST (128). f Urodynamics In general, urodynamic studies are not necessary in the evaluation of IC because voiding diaries can provide adequate information. However, some investigators believe that urodynamics will allow determination of detrusor instability and urethral dysfunction that can be treated differently. If urinary incontinence is present, urodynamic investigations are recommended. g. Urinary Markers An attractive approach is to find a sensitive and specific maker in the urine that might serve as a noninvasive diagnostic tool for IC. Among many urinary substances that have been investigated, A P F (115) and glycoprotein-51 (GP-51) (129) are potentially useful. Both of these markers were evaluated, and there was no overlapping in the urinary levels of A P F and GP-51 in those who
met the N I D D K diagnostic criteria and in controls. The alteration of these markers in patients who do not fulfill the N I D D K criteria remains unknown (130). Although attractive, the usefulness of these urinary markers in the diagnosis of lC requires much more investigations.
D. Treatment Because the etiology of IC is not completely understood, effective treatment for this disease does not exist. Currently, there is no cure for this condition, and clinicians should carefully counsel patients on the best form of therapy and the realistic expectation of the outcomes. The current treatments of IC are empirical and can only alleviate symptoms. Most of the employed modalities have not been studied in prospective, randomized, and placebo-controlled trials. Currently, some ongoing prospective studies are being carried out by the National Institutes of Health (NIH)-sponsored Interstitial
CHAPTER51 Lower Urinary Tract Disorders in Postmenopausal Women Cystitis Clinical Trial Group, and the results are pending. In addition to the therapeutic effects of hydrodistention, the current treatments of IC include supportive management, oral therapy, intravesical instillation, and surgical approach. NSMDs can be used adjunctively to reduce pain and inflammation. It should be noted that anticholinergic and antispasmodic agents are ineffective in women with IC. Anticholinergics block the motor pathway, but IC is primarily a hypersensory condition. 1. SUPPORTIVEMANAGEMENT
Supportive management consists of dietary modification, stress management, pelvic floor exercises, bladder training programs, and other behavioral measures. Although the link between dietary factors and IC symptoms has not been fully established, some foods and beverages may exacerbate the symptoms. It has been shown that about 53% of IC patients associate symptom aggravation with acidic foods and beverages (131). Another study demonstrated that urinary levels of tryptophan metabolites were elevated in women with hypersensitive bladders as compared with the controls (132). This increase in the tryptophan metabolites may disrupt the GAG layer of the urothelium and thus predispose to the development of IC or exacerbate its symptoms. However, more studies are needed to investigate the role of dietary factors. Stress management and behavioral modification may play a role in the treatment of IC. Distraction techniques can be employed to increase the time between voids. Contracting the pelvic floor muscles and overriding the first urge to void are very helpful for this purpose. Pelvic floor exercises and bladder training programs, as described earlier for urinary incontinence, can be employed in IC patients. These are good initial or adjunctive interventions and have been used with some success. 2. ORALTHERAPY
a. Sodium Pentosan Polysulfate (Elmiron) Sodium pentosan polysulfate is a weak heparinoid that supposedly reverses the GAG layer deficiency in the urothelium and thus minimizes the permeation of urinary bacteria or toxins. This medication is approved by the U.S. Food and Drug Administration (FDA) for oral use in the treatment of 1C. Hanno (133) reported that pain relief occurred in approximately 40% to 60% of patients after 3 months of treatment with 100 mg, three times per day. In other prospective, randomized, and placebo-controlled trials, the improvement of certain IC symptoms was significant with pentosan polysulfate (28% of treated subjects versus 13% of placebo controls), although the degree of improvement was not dramatically from a clinical standpoint (134). A recent study showed that the response rates of pentosan polysulfate at 32 weeks of treatment range
723
from 45% to 49% and these responses are not dose dependent, but the duration of therapy appears to be more important (135). Because pentosan polysulfate may take 3 to 6 months to achieve maximal benefit (133), adequate time should be allowed for the treatment to become effective. Long-term efficacy studies demonstrated that the benefit was maintained for I to 2 years in those who responded to the therapy (133). Although this medication is well tolerated, gastrointestinal side effects and reversible alopecia occur in 4% of patients.
b. dmitriptyline (Elavil) This medication has been used to decrease the chronic pain component of IC (136,137). In addition to its inhibiting properties in the noradrenaline and serotonin reuptakes, amitriptyline also has analgesic, sedative, anticholinergic, and antihistaminic effects. Amitriptyline is usually given at night to obtain additional benefit of improving sleep disturbance and decreasing nocturia. The starting dose is 10 to 25 mg at bedtime, gradually increased to 75 mg or until the side effects are intolerable in order to achieve maximal benefit. Weight gain, sedation, and anticholinergic side effects such as dry mouth and constipation occur in 20% to 80% of patients (136). When compared with placebo, amitriptyline significantly improved pain and urgency in patients with IC, and a long-term efficacy study (mean of 17 months) revealed a 64% improvement rate using the global assessment questionnaire (137). When used in conjunction with pentosan polysulfate, amitriptyline can be tapered off once remission is attained. Other tricyclic antidepressants have not been studied in the treatment of IC. c. Hydroxyzine (Atarax) Hydroxyzine, which is an antihistamine that prevents mast cell &granulation, has been employed in the treatment of IC. The degranulation of mast cells causing the release of neuroactive and vasoactive agents is believed to be responsible for the symptoms of IC. For IC patients who have a history of allergies, or in whom mast cells were confirmed on bladder biopsy, an antihistamine such as hydroxyzine is a good choice. In an open-label investigation with hydroxyzine, 40% reductions in IC symptoms were reported (138). This improvement increased to 55% in patients with history of allergies. Hydroxyzine can be taken alone, or given together with pentosan polysulfate, at a dose of 10 to 25 mg at bedtime for 1 week, then gradually increased to 50 to 75 mg. Hydroxyzine also has sedative and anxiolytic side effects, which are beneficial for nocturia and other IC symptoms. d. Gabapentin (Neurontin) Gabapentin is an antileptic medication that can hyperpolarize the neurons involved in pain transduction and thus increase their sensory thresholds. Because of this action, gabapentin has been tried for chronic pelvic pain and IC patients with variable success (139).
7
2
4
H
3. INTRAVESICAL INSTILLATION As compared with oral therapy, intravesical instillation is an attractive approach in the treatment of IC for three reasons. First, the side effects can be minimized in the intravesical therapy because of the lack of systemic absorption. Second, the therapy should be more effective because it targets directly on the urothelium where the abnormalities are believed to occur. Third, a "cocktail" mixture of multiple medications can be employed in intravesical instillation to provide the synergic effects. Several agents, either alone or as a mixture, have been used; however, the selection of these agents is empiric, and data on the prospective and randomized studies of these agents are limited. These medications are instilled through a urethral catheter and left in the bladder for 20 to 30 minutes or as long as the patient can tolerate. Although the schedule of treatment varies with each agent, the instillation is usually carried out once per week for 6 weeks. After this initial treatment period, some patients may need a maintenance schedule of instillations, usually biweekly or monthly. Intravesical instillation can also be used as an adjunct to oral therapy. A potential risk for UTI via catheterization and a transient chemical cystitis that exacerbates the symptoms can occur.
a. Dimethyl Sulfoxide Besides pentosan polysulfate, dimethyl sulfoxide (DMSO) is the only other drug approved by the FDA for the treatment of IC. DMSO is thought to improve the IC symptoms through its anti-inflammatory actions, mast cell inhibitions, muscle relaxant effects, and sensory neuropeptide depletions of the afferent nerves. About 50 mL of a 50% DMSO solution is instilled into the bladder and held for 20 to 30 minutes before voiding. Instillations are performed every 1 to 2 weeks for a total of 4 to 8 treatments. Some patients may need a maintenance schedule of instillations because of relapsed symptoms. A garlic-like odor may occur in the skin or breath because DMSO is secreted through the skin and lungs. DMSO can induce remission in 35% to 60% of patients for up to 24 months (140). If patients do not respond to DMSO initially, the combination of DMSO with hydrocortisone, heparin, and sodium bicarbonate in the instilled solution has been recommended (116). b. Sodium Pentosan Polysulfate (Elmiron) The action of sodium pentosan polysulfate is to replenish the defected GAG layer in the urothelium of the bladder. Because sodium pentosan polysulfate is poorly excreted in the urine, intravesical application of this medication is expected to be more effective than the oral route. However, this effect was not observed clinically. This agent is administered as a solution of 300 mg in 50 ml of normal saline, twice weekly for 12 weeks. It has been shown that intravesical instillation of pentosan polysulfate increases the bladder capacity and improves the symptoms of IC (141). These improvements are similar to the oral route, but symptoms were relieved quicker with the instillation.
o
AND BHATIA
c. Steroids Steroids such as hydrocortisone, methylprednisolone, or triamcinolone can be reconstituted in a small volume (10 to 15 mL) and instilled intravesically, either alone or in a mixture with other agents. Steroids are though to improve the IC symptoms through their immunosuppression and anti-inflammatory properties. d Heparin Heparin is thought to help replenish the GAG layer in the bladder's urothelium, thus improving the IC symptoms. In addition to this surface protective action, heparin is believed to have anti-inflammatory effects in the bladder. Heparin solution can be used alone or in a mixture with other agents. Because this medication has been shown to reduce relapses in patients who responded to DMSO, it is common for heparin and DMSO to be instilled together into the bladder. The instillation can be carried out 1 to 3 times per week, at a dose of 10,000-20,000 U each time. When administered 3 times per week for 3 months, remission of symptoms was reported in 56% of patients (142). The remission was maintained for up to 1 year in 80% to 90% of patients, if the treatment was continued at a maintenance level. Recently, bladder instillation of a mixture of 40,000 units heparin and 2% alkalinized lidocaine (lidocaine and bicarbonate) has been shown to produce immediate symptom relief in 94% of patients before heparin reaches its full effect (143). e. Local Anesthetics Local anesthetics, such as lidocaine (1%) or bupivacaine (Marcaine, 0.5%), can be used intravesically, either as a single solution or in a mixture with other agents (116,143). These agents are used in bicarbonate buffered solution. The responses of lidocaine and bupivacaine are significant; however, lidocaine requires frequent instillation due to its short duration of action. The longer-acting bupivacaine has been used in a variety of mixtures with other agents for bladder instillation. fi Hyaluronic Acid (Cystistat) Hyaluronic acid, which occurs namraUy in the human connective tissues, is another agent for intravesical therapy. When instiUed into the bladder, as a 40 mg in 40 ml normal saline solution, hyaluronic acid is believed to coat the lining of the bladder and protect it from irritating substances in the urine. Hyaluronic acid can be instilled weekly for 4 to 6 weeks, and then monthly for 6 months, or twice weekly and then monthly after improvement in symptoms occurs. In a small study involving 20 patients, weekly administration of hyaluronic acid decreased pain and nocturia in 30% and 40% of patients, respectively, while improving the IC symptoms in 65% of patients (144). g. Capsaicin/Resiniferatoxin Capsaicin is an extract of chili pepper that can reduce the activation of C-fiber. Symptoms oflC are thought to involve the C-fibers in the bladder (109,116). If the sensation of bladder pain mediated by these fibers can be blocked by capsaicin, then the symptoms of I C
725
CHAPTER51 Lower Urinary Tract Disorders in Postmenopausal Women may be improved. Resiniferatoxin (RTX) is a potent analog of capsaicin that acts similarly by desensitizing the sensory nerves of the bladder. This agent has been shown to be effective in some studies with small numbers of patients (116). RTX has an advantage over capsaicin because it does not cause a burning sensation with instillation. However, a recent prospective, randomized, and placebo-controlled clinical trial involving 163 patients demonstrated that RTX is not effective for IC (145).
h. Bacillus Calmette Gugrin Bacillus Calmette Gu6rin (BCG) is currently approved by the FDA for bladder cancer treatment. The mechanism of BCG action in the treatment of IC is unknown, although the modulation of the host immune response by this agent has been proposed. This agent has been employed intravesically in a prospective study involving 30 patients and showed a 60% improvement in IC symptoms versus a 27% for placebo (146). However, another study was unable to reproduce these results (147). Recently, a prospective, randomized, and placebo-controlled study involving 248 patients showed a response rates (in global assessment) of 21% for BCG and 12% for placebo (148). Although the safety profile of BCG is acceptable and the effectiveness of this agent is confirmed, the response rate is not dramatic from a clinical standpoint. i. Mixture of Multiple Agents Several "cocktail" regimens, which are mixtures of different agents, have been employed empirically for bladder instillations (116). These solutions involve lidocaine (or bupivacaine), bicarbonate, hydrocortisone (or triamcinolone), heparin, and DMSO, in a variety of mixtures, to provide immediate symptom relief as well as long-term effects. The common "cocktail" solutions are: (a) a mixture of 50 ml of DMSO, 10,000 units of heparin, 10 mg of triamcinolone, and 44 mEq of bicarbonate that can be administered weekly for 6 weeks (116), and (b) a mixture of 40,000 units of heparin, 8 ml of 2% lidocaine, 3 ml of 8.4% bicarbonate, and 5 ml of sterile water that can be administered 3 times per week for 2-3 weeks (143). Alternatively, an oral medication can be added to a patient who has already started with intravesical therapy to enhance the effects. No data on the formal evaluation of these "cocktail" regimens are available. 4. SURGICALTHERAPY
a. Laser Therapy Neodymium:YAG laser can be used to fulgurate the Hunner's ulcers during cystoscopy. This method is effective in improving the IC symptoms but carries a high relapse rate. About 50% of the 24 studied patients required one to four re-treatments (125). The application of laser therapy is limited because most of the IC patients do not have Hunner's ulcers (116).
b. Sacral Neuromodulation Sacral neuromodulation is a long-term electrical stimulation of the $3 nerve root that has been approved by the FDA for the treatment of IDO, frequency-urgency syndrome, and idiopathic urinary retention. The mechanism of action of sacral neuromodulation in the treatment of IC is unclear, although this modality has been applied to a small number of patients who failed other modes of treatment, with promising results (94,149). Thus far, it has significantly reduced urinary frequency and urgency symptoms as well as the pain associated with IC. Recently, a multicenter clinical trial also showed significant improvements in urinary frequency, pain symptoms, and voided volumes (150). The neuromodulation is achieved with an implanted lead that is placed through $3 foramen and an implanted electrical pulse generator as described in detail elsewhere (94). c. Cystectomy and Bladder Augmentation Surgical therapies, particularly major surgical interventions with denervation procedures, cystectomy, and bladder augmentation, are the last resort in the treatment oflC. Cystectomy and bladder augmentation has been described in the treatment of IC in patients with a severely contracted bladder (116). Although cure rates ranging from 50% to 80% were reported in some series, other studies demonstrated that IC symptoms may persist despite of total cystectomy and urinary diversion, and repeat surgery is often required (116,119). For this reason, patients must be carefully selected and fully counseled. 5. OTHER THERAPIES
Transcutaneous electrical nerve stimulation has been described for the treatment of IC with some benefits (151). This treatment modality supposedly stimulates the afferent nerves, thereby activating the inhibitory circuits and decreasing the sensation of pain. However, its exact mechanism remains unclear. Other approaches under investigation include intravesical injection of botulinum toxin and gene therapy (152). Botulinum toxin has been shown some benefits in the treatment of IC (152); however, a recent report demonstrated that intravesical injection of botulinum toxin A is not effective in 8 patients who have refractory IC (153).
V. LOWER URINARY TRACT INFECTION UTI can be classified as involving the upper and the lower urinary tract. The discussion in this chapter focuses only on lower UTI, which occur much more commonly than upper tract infection, and the term UTI is used to indicate lower urinary tract infection. Although the diagnosis and management of UTI in postmenopausal women are often simple, new challenges have been occurred recently. These challenges
726
Ho
include the recognition of additional pathogens such as Staphylococcus saprophyticus and Chlamydia trachomatis, the understanding that many cases of acute cystitis may have less than the standard 105 colony forming units (CFU)/mL in urine, the realization that many other conditions such as interstitial cystitis and urethral syndrome may present with UTI-like symptoms, and the awareness of recurrent and instrument-associated infections.
AND BHATIA
pathology. Detrusor dysfunctions, UTI, urethral defects, and other anatomic and functional defects of the urinary tract should be ruled out before attempting to use this term. Pydonephritis is a bacterial infection of the renal parenchyma and the renal pelvicaliceal system. The discussion ofpyelonephritis is beyond the scope of this chapter.
B. Epidemiology A. Terminology The commonly used terminology surrounding UTI requires some definitions because it can be confusing. Bacteriuria means the presence of bacteria in the urine, including both renal and bladder bacteria. Asymptomatic bacteriuria is generally accepted as a bacterial colony count of 105 or more per mL of urine in at least two properly collected "clean catch" specimens in the absence of clinical UTI symptoms. These patients may or may not have pyuria. Lower colony counts, even as low as 102 CFU/mL of urine, may be accepted as bacteriuria in symptomatic patients. Persistent bacteriuria is the presence of microorganisms that were isolated at the start of treatment and continue to be isolated while the patient is receiving therapy. Persistence may be caused by several factors, including the presence of resistant organisms, inadequate drug therapy, and poor patient compliance. Superinfection is the appearance of a different organism while a patient is still receiving therapy. The new organism may be a different strain or a different serologic type. Relapse of infection occurs with the recurrence of significant bacteriuria with the same species and serologic strain of organism. Relapse usually appears within 2 to 3 weeks of completion of therapy and most likely represents perineal colonization by the infecting organism. Recurrent infection (or reinfection) is an infection occurring after cessation of therapy with a different strain of microorganism or a different serologic type of the original infecting strain. Typically, reinfection occurs 2 to 12 weeks after a previous episode of infection and indicates recurrent bladder bacteriuria. Cystitis is an inflammation of the urinary bladder, which may have infectious or noninfectious etiology. Infectious cystitis or bacterialcystitis may be classified as simple, complicated, or recurrent infection. Recurrent UTI is diagnosed with 2 infections within 6 months or 3 or more infections during 1 year in which the initial episode is resolved and subsequently followed by another UTI. Trigonitis refers to inflammation or localized hyperemia of the bladder's trigone, commonly seen by cystoscopy. Urethritis indicates inflammation of the urethra and is usually caused by Chlamydia trachomatis, Neisseria gonorrhea, genital herpes, or other nonspecific conditions. Urethral syndrome is a poorly defined condition in patients with urinary frequency and urgency, dysuria, voiding difficulty, and suprapubic discomfort without any discernible
UTI occur much more frequently in females than in males, with a female to male ratio of 10:1. The incidence of UTI increases with age, and the infection is more common in women who are sexually active and in those who use diaphragms and spermicides for contraception (154). Approximately 20% to 50% of women will have at least one episode of UTI during their lifetime (155). Although UTI can occur in women of all ages, its prevalence is higher in the postmenopausal women. Impairment of the bladder emptying and alterations in the vaginal flora after menopause are thought to place these women at an increased risk of UTI. Furthermore, poor perineal hygiene, urinary incontinence, and fecal incontinence in these women can contribute to the problem. The elderly are usually sicker and at greater risk of dying with UTI than younger women. Because the pH of the vaginal secretion is increased and the number of lactobacilli is decreased in postmenopausal women, Gram-negative bacteria can colonized the vagina and serve as the uropathogens. The increase in residual volume of urine in the bladder of postmenopausal women also promotes bacterial growth. However, the analysis of midstream urine showed that the proportion of positive infections increases with age in both men and women, with no specific changes in the rates of infection at or after the menopause (156). The exact role of the menopause in the development of UTI is, therefore, not clear, and more investigations are needed.
C. Pathophysiology 1. Pathogenesis The most common pathogen in UTI is Escherichia co& which accounts for about 80% of the community-acquired infections (157). The second most common pathogen is Staphylococcus saprophyticus, which presents in 5% to 15% of young women with acute cystitis. Klebsiella species were found in about 5%, Enterobacter species in about 2%, and Proteus species in another 2% of community-acquired UTI. Approximately 1% to 2% of UTI are caused by other Gram-positive organisms such as group B and group D streptococci. Staphylococcus epidermidis is a frequent cause of nosocomial UTI in catheterized patients (158). This organism is also frequently resistant to
CHAPTER51 Lower Urinary Tract Disorders in Postmenopausal Women antibiotics. Serratia marcescens and Pseudomonas aeruginosa are primarily hospital acquired and are responsible for the infection in hospitalized patients with urethral catheterization or manipulation (157,159). These pathogens may gain entry to the urinary tract by three pathways: the ascending route, the hematogenous route, and the lymphatic route. The ascending route appears to be the most important mechanism for organisms to enter the urinary tract. Normally, the urinary tract in women is sterile above the level of the distal urethra, and organisms that gain access to the bladder tend to do so from neighboring sites such as the perineum, vaginal vestibule, lower urethra, paraurethral tissues, and bowel. Women are more susceptible to UTI due to their short urethra and the decreased urethral resistance after menopause. Urinary infection via the hematogenous route is uncommon but is seen occasionally in elderly, debilitated, or immunosuppressed patients with overwhelming infections or in whom kidney infection is only part of the multisystemic involvement. Renal seeding of staphylococcal organisms and tuberculosis are almost always acquired via the hematogenous route. Large bacterial inoculums and virulent organisms are also spread through hematogenous dissemination. Although very rare, lymphatic route of infection can occur. Sexual intercourse and use of diaphragm increase the risk of UTI (154). Other risk factors in postmenopausal women include diabetes mellitus, history of UTI, and urinary incontinence (160). Additional sources of infections include vulvovaginitis, urethra] diverticula, poor hygiene, and indiscriminate urethral catheterization. Infrequent and incomplete voiding resulting in large bladder volumes also increases the susceptibility to UTI.
727
urine may also block bacterial adherence. The deficiency of the GAG layer in the bladder may play a role in recurrent bacterial cystitis (162). Because voiding tends to dilute and wash away the inoculated organisms, bladder infection depends on the residual volume of urine, rate of urine flow, and frequency of voiding. Voiding not only displaces infected urine with freshly sterile urine, but also flushes out bacteria attached to the urothelium. Furthermore, urine inhibits bacterial growth with its low pH, high organic acid concentration, and very low or high osmolarity. Anaerobes are rarely seen in UTI, although they are abundant in the feces, because oxygen tension in the urine prevents their growth.
D. Evaluation 1. CLINICALPRESENTATION Acute cystitis generally has an abrupt onset of severe urinary tract symptoms such as dysuria, frequency and urgency that are associated with suprapubic or low back pain. Systemic symptoms such as fever and chills are usually absent in lower UTI. Occasionally, mild urinary incontinence may occur. Physical examination reveals suprapubic tenderness. Infection is characterized by a large number of leukocytes and organisms in the urine. Microscopic hematuria may sometimes occur in urine analysis. The symptoms of urethritis are usually milder and the onset is more gradual than acute cystitis. Patients may have lower abdominal pain and abnormal vaginal discharge or bleeding related to concurrent cervicitis. In female patients, symptoms of urethritis may be difficult to distinguish from those of cystitis. 2. DIAGNOSTICTESTS
2. HOST DEFENSE MECHANISM
Susceptibility to UTI depends on the status of the host defense mechanisms, the virulence of microorganism, and the inoculum size. The host defense mechanisms can be found in the vagina, urinary tract, and urine. Normally, the vagina and periurethral areas provide important defense mechanisms to prevent the progression of microorgansims from the rectum to the urethra. The acidity of vaginal secretions promotes the growth of normal flora such as lactobacillus and prevents the growth of uropathogens such as E. coli. In the premenopausal woman, the vaginal pH is around 4. When the pH becomes less acidic, enterobacteria can colonize vaginal introitus and periurethral areas, thus predisposing to UTI. The normal urinary tract in the female is remarkably resistant to infection. The urinary tract secretes proteins, such as Tamm-Horsfall protein, that inhibit bacterial adherence and allow the bacteria to be flushed away (161). The presence of GAG in the bladder urothelium and immunoglobin in the
Diagnosis of cystitis is based on the clinical manifestations, microscopic examinations, urinalysis, and culture results. The method of urinary collection must be carefully performed to avoid contamination. Collecting a cleancatch and midstream specimen is widely employed; however, urine sample can be obtained by bladder catheterization or suprapubic aspiration in special cases. Bladder catheterization and suprapubic aspiration should only be used when it is impossible to obtain uncontaminated urine samples or in symptomatic patients who have very low bacterial counts. a. Urine Analysis In addition to presenting symptoms, urinalysis can provide important evidences of UTI. The pH and specific gravity of urine are usually not helpful, except in the case of Proteus mirabilis infection (pH 8). The presence of distinctive crystals in the urine is not sensitive or specific for acute cystitis. Dipstick tests for proteinuria also have a poor predictive value in detecting UTI. However, dipstick
728 tests for esterase, hematuria, and nitrite are very useful in the diagnosis of UTI. Although less accurate than microscopic examination, these rapid dipstick tests in the office are reasonable substitutions (163). The nitrite test is based on the conversion of urinary nitrate to nitrite by bacterial action. The esterase test is based on a substrate color change that is caused by the esterase found in leukocytes. The nitrite and esterase tests, which indicate the presence of bacteriuria and pyuria, respectively, depend on the bacterial counts for their sensitivities. With the bacterial counts of 105 CFU/mL or more, the sensitivities of these tests are 60%; whereas infections with 104 to 105 CFU/mL, the sensitivities are only 22% (164). False-negative results can occur with a poor sampling technique, an enterococci infection (because they do not convert nitrate to nitrite), and the presence ofbilirubin, methylene blue, or phenazopyridine that may interfere with the test. Other rapid tests are also available for detecting the presence of bacteria and leukocytes in the urine simultaneously (165). These tests are based on the staining of the bacteria and leukocytes in a concentrated sample and then comparison of the resulting colors with a reference. Their sensitivities are 79% to 85% with the bacterial counts of 105 CFU/ mL or greater, but only 34% to 65% with infections of 104 to 105 CFU/mL. These tests, which are more sensitive but less specific than the nitrite and esterase tests, can be a good screening method for UTI. b. Microscopic Examination Microscopic examination of bacteria, leukocytes, and red blood cells in urine is an easy and valuable method of evaluating UTI. Bacteria can be detected with microscopy and the presence of several organisms per high power field usually correlates with a culture count of 105 CUF/ml urine. Gram stain may further help in determining the organism. Yeast infection, such as Candida albicans, can also be detected with microscopy. The presence of leukocytes in the urine indicates host injuries from infectious or noninfectious causes. In patients with bacteriuria, leukocyte counts help to differentiate colonization (bacteriuria without evidence of tissue invasion) from UTI (bacteriuria with evidence of host injury). A leukocyte count of more than 5 leukocytes/high power field in a centrifuged urine sample is usually considered significant pyuria; however, this method has been shown to have a poor reproducibility (166). Studies have suggested the nonpathologic limit for pyuria is less than 10 leukocytes/ml of uncentrifuged urine. Stamm (166) reported that 96% of patients with symptomatic bacteriuria have 10 leukocytes/ml of urine or more, while less than 1% of asymptomatic and a bacteriuric subjects have this level of leukocyte counts. Examination for microscopic hematuria may be helpful because it can be found in about 50% of women with acute
Ho
AND BHATIA
UTI while it is rarely present in patients who have dysuria from other causes (166,167). The presence of red blood cells in the urine frequently occurs when there is an infection in the mucosa of the bladder. However, microscopic hematuria, that does not resolve with UTI treatment or occurs in asymptomatic patients, requires further diagnostic workup to rule out other pathologies. c. Urine Culture and Sensitivity In the uncomplicated patient, the diagnosis can be presumed with presenting symptoms and positive screening tests for bacteriuria, pyuria, or hematuria, and starting antibiotics without urine culture is a reasonable approach. If the screening tests are inconclusive in a symptomatic patient, or if persistent or recurrent infection occurs in a patient who has been treated with antibiotics, urine culture and sensitivity should be performed. Urine culture and sensitivity should also be performed in patients with signs and symptoms that are consistent with upper urinary tract infection (i.e., pyelonephritis). Positive urine culture traditionally requires a growth of bacteria at least 105 CFU/ mL; however, the application of this criterion may have some limitations. It has been shown that 20% to 24% of women with symptomatic urinary infection have less than 105 C F U / m L in urine (168). A diagnostic criterion of 102 CFU/mL, rather than 105 CFU/mL, has been proposed for young symptomatic women. The other limitation is the contamination from other bacteria on the perineum, which can occur in 18% of specimens (169). Therefore, urine specimen collection must be carried out with considerable care. d. Cystourethroscopy Cystourethroscopy is not routinely performed in the evaluation of female patients with UTI. However, it should be considered in women with asymptomatic hematuria or with recurrent or persistent UTI. In hematuria, cystoscopy should be performed to rule out malignancy or other noninfectious pathologies, whereas recurrent or persistent UTI may result from a fistula or diverticulum (170). e. Radiologic Studies Radiographic study is not routinely necessary. However, if urethral diverticulum is thought to be the cause of recurrent infection in the lower urinary tract, a voiding cystourethrogram or a double-balloon catheter study should be performed. Intravenous pyelography (IV-P) may be indicated in UTI patients who have a history of urinary stones or obstruction, history of recurrent infection caused by urea-splitting organisms such as Proteus mirabilis, history of previous upper urinary tract infection, history of childhood UTI, infection associated with painless hematuria, and infection suggesting the presence of an enterovesical fistula.
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women
F. M a n a g e m e n t 1. GENERALMANAGEMENT
The bacteriuria may clear spontaneously without antibiotic therapy due to the host defense mechanisms. Two of the important protective factors are the washout effects cause by urination and the dilution of pathogenic organisms caused by the freshly sterile urine. Hydration is a simple way that may help to wash out and dilute the bacteria in the urinary tract. Cranberry juice may inhibit bacterial adherence and protect against the development of UTI (171). Pain and burning sensation with urination can be relieved with analgesic agents such as phenazopyridine hydrochloride (Pyridium). This agent is usually prescribed for 2 to 3 days along with the antibiotic treatment, and patients should be warned that their urine and contact lenses may turn an orange color. 2. ANTIBIOTICTREATMENT
Verifying the presence of complicated or recurrent infections and selecting the antimicrobial agent that is most likely to be effective are important tasks in the management of UTI. Two considerations should be kept in mind when prescribing antibiotic treatment for UTI. First, the treatment agent should ideally not alter the bacteria in the bowel because the fecal flora is the reservoir for most of the organisms causing infection. This alteration can occur if the drug has a high serum level or passes through the gastrointestinal tract without being absorbed. Second, the treatment agent should not disturb the vaginal flora in order to avoid yeast vaginitis. Disrupting the vaginal flora allows yeast to grow and causes vaginitis, which can subsequently lead to a vaginitis-cystitis cycle that may be difficult to treat. The spectra of antimicrobial activities against common pathogens in UTI are shown in Table 51.5. The combinations of trimethoprim and sulfamethoxazole (Bactrim, Septra) have become popular in the treatment of UTI. They provide a broad range of activity against uropathogens but have a low incidence of side effects and infrequent occurrence of bacterial resistance. These agents have moderate effects on the vaginal and bowel flora (157). Nitrofurantoin has excellent activity against E. coli but no significant changes in vaginal or bowel flora and low bacterial resistance. Fluoroquinolones such as ciprofloxacin, norfloxacin, ofloxacin, levofloxacin, and amifloxacin have also been introduced for the treatment of UTI. These agents offer excellent actMty against E. coli and other organisms such as Pseudomonas aeruginosa, staphylococci, and enterococci. Their adverse effects are infrequent; however, an increase in bacterial resistance, particularly by E. coli, has already occurred with ciprofloxacin due to its widespread use (157). These agents are also expensive and should be reserved for patients with persistent or recurrent infections, or as an
729
alternative to parenteral antibiotics in complicated infections. Amoxicillin or first-generation cephalosporins are generally avoided by many clinicians because of relatively high failure rates. Dosage and side effects of antibiotics commonly used in the treatment of UTI are shown in Table 51.6. 3. ASYMPTOMATICBACTERIURIA Asymptomatic bacteriuria can resolve spontaneously. It has been shown that 66% and 40% of elderly women with asymptomatic bacteriuria clear their bacteriuria when they are treated with antibiotics or placebo, respectively (172). Although the long-term effects of asymptomatic bacteriuria are not completely known, it is seldom associated with adverse outcomes. Except for pregnant women and for the preoperative evaluation before urologic or gynecologic surgeries, screening for asymptomatic bacteriuria in adults offers very little value. It has been shown that postoperative complications can be reduced by detecting and treating asymptomatic bacteriuria before urologic surgeries (173). There are no data to indicate that the routine treatment for asymptomatic bacteriuria is necessary in postmenopausal women.
4. SIMPLEINFECTION
In the uncomplicated patients with acute cystitis, treatment can be carried out without urine culture, and follow-up visit is unnecessary unless symptoms persist or recur. There are several regimens for the treatment of simple UTI. A single-dose therapy has been proposed; however, this approach had significantly higher treatment failure rate when compared with the 10-day course of trimethoprim (TMP)sulfamethoxazole (SMX) (174). However, a recent study demonstrated that a single dose of 500 mg of ciprofloxacin and a 3-day course of norfloxacin (400 mg twice a day) have the same efficacy (175). With most uncomplicated UTI, the 3-day regimens of TMP-SMX or nitrofurantoin are effective, but with less cost and fewer side effects than the 7-day regimens. Although a 3-day course of TMP-SMX is the current standard for empiric treatment of acute cystitis in uncomplicated patients, there has been a trend towards increasing resistance among uropathogens to this antimicrobial agent. The fluoroquinolones are highly effective and have low side effects, but they are more expensive. Prudent use of fluoroquinolones for the treatment of uncomplicated UTI is warranted because resistant organisms may develop with widespread use of these antibiotics. For patients with systemic diseases, such as diabetes meUitus, known structural abnormalities of the urinary tract, history of a treatment failure in the past 6 months, history of acute pyelonephritis, and history of childhood urinary tract infections, a 7- to 10-day course of therapy should be given. For patients whose UTI symptoms persist beyond the third day of
730
Ho AND BHATIA TABLE 51.5
Drug
Dosage and Adverse Reactions of C o m m o n Antibiotics Used in the Treatment of Lower Urinary Tract Infections Dosage
Common Adverse Reactions
TMP-SMX (160 mg/800 mg)
1 tablet PO q 12 h
Hypersensitivity, photosensitivity, skin reactions, GI upset, blood dyscrasia
Nitrofurantoin
100 mg PO q 6 h
GI upset, peripheral neuropathy, pulmonary hypersensitivity reactions, hemolysis in patients with G6PD deficiency
Nitrofurantoin, sustained release (Macrobid) Norfloxacin
100 mg PO q 12 h
Same as nitrofurantoin above
400 mg PO q 12 h
Nausea, vomiting, diarrhea, abdominal pain, skin rash, convulsions, psychoses, joint damage
Levofloxacin
500 mg PO q 24 h
Ciprofloxacin
500 mg PO q 12 h
Enoxacin
400 mg PO q 12 h
Lemofloxacin
400 mg PO q 24 h
Cephalexin
250-500 mg PO q 6 h
Ampicillin
500 mg PO/IV q 6 h
Gentamicin
1 mg/kg IM/IV q 8 h or 5-7 mg/kg IV q 24 h
Disturbance of blood glucose, allergic, photosensitivity, nausea, headache, allergic reactions, tendon rupture, pseudomembranous colitis Anosmia, taste loss, myalgia, dyspepsia, central nervous system effects, theophylline interactions, pseudomembranous colitis, anaphylactic reaction Mild GI effects, photosensitivity, CNS effects as other fluoroquinolones above Mild GI effects, photosensitivity, CNS effects as other fluoroquinolones above Allergic reactions (less than with penicillins), GI upset, pseudomembranous colitis Hypersensitivity, allergic reactions, diarrhea, GI upset, pseudomembranous colitis, decreased platelet aggregation, candidal overgrowth Ototoxicity, nephrotoxicity, neuromuscular blockade with high levels
Precautions Hematologic toxicity in AIDS patients, increased risk of hematologic effects in folateor G6PD-deficient patients, avoid in patients receiving warfarin Avoid concomitant use of magnesium or quinolones, which are antagonistic to nitrofurantoin, do not use in patients with low creatinine clearance (<40 ml/min), use with caution in patients with G6PD deficiency Same as nitrofurantoin above
Avoid in children and pregnancy due to arthropathic effects. Can significantly increase theophylline plasma levels, can lower seizure threshold, and can enhance warfarin effects. Need to monitor glucose levels in patients on antidiabetic agents. Concomitant use of antacid or iron or zinc or sucralfate dramatically decreases oral absorption Same as norfloxacin above
Same as norfloxacin above
Same as norfloxacin above Same as norfloxacin above
Hepatic dysfunctions
Increased risk of rash with concomitant viral disease, allopurinol therapy
Avoid in patients with pregnancy, impaired renal function, diabetes, or hepatic failure. Use with caution in myasthenia gravis and with other potentially ototoxic and nephrotoxic drugs
+ +
+ +
-
-
+
Proteus species
Enterobacter species
Enterococcus species
Staphylococcus species
Serratia marcescens
.
.
_
+
.
-
+
+ + .
Nitrofurantoin
Antimicrobial
.
.
.
.
+ +
+ +
.
+ +
-
+ +
Ampicillin
.
.
.
+
+ +
-
_
_
+ +
+
+ +
+ +
+ +
++
++
++
++ ++
+
++
++
++
++
++
++
Norfloxacin
++
++
++
-
++
++
++
++
Gentamicin
++
++
++
___
++
++
++
++
++
Levofloxacin
in L o w e r U r i n a r y T r a c t I n f e c t i o n s
Carbenicillin
against Pathogens
Cephalexin
Antibiotics
Tetracycline
Coverage of Common
++
++
-+-
++
++
++
++
++
Ciprofloxacin
Note: + +, excellent; +, good; _+, occasionally effective; -, resistant. (Reproduce from Karram MM, Kleeman SD. Lower urinary tract infection. In: Bent AE, Ostergard DR, Cundiff G W , Swift SE (eds). Ostergard's Urogynecology and Pelvic Floor Dysfunction, 5 th ed. Lippincott, Williams & Wilkins, Philadelphia, 2003, p. 273.)
+ +
.
+ +
TMP-SMX
Klebsiella species
Pseudomonas aeruginosa
Escherichia coli
Organism
TABLE 5 1 . 6
--.a
9
~,..~
9
v~
t.rt
a: >
732 therapy, a urine culture and sensitivity should be obtained to rule out the possibility of the diagnostic error and bacterial resistance. Post-treatment assessment with urine culture is not necessary in simple cases of UTI. Recently, the use of telephone triage and offering antibiotics based on symptoms without an office visit, urine analysis, or urine culture have been investigated, and the results showed that potential adverse outcomes were not significantly increased during the 60 days after the diagnosis and treatment (176).
5. COMPLICATED INFECTION
The wide variety of underlying conditions, such as anatomic, functional, and metabolic defects in the urinary tract, and the diverse spectrum of possible pathogens that are resistant to antibiotics can cause complicated UTI. In these complicated infections, the clinical presentation can range from mild cystitis to life-threatening urosepsis. Urine culture and sensitivity must be obtained in patients suspected of having complicated infection before starting the treatment. Empiric therapy can be carried out with fluoroquinolones, which provide a broad spectrum of antimicrobial activity and cover most of the expected pathogens including pseudomonas and enterococci. Mild or moderate cases can be treated as outpatients, whereas severe cases require hospitalization. The period of antibiotic treatment is usually 10 to 14 days; however, pseudomonas and enterococcal infections require longer therapy. Urine culture should be repeated 1 to 2 weeks after completion of the therapy. Any anatomic, functional, or metabolic defects should also be corrected to avoid recurrent infections.
6. RECURRENT INFECTION
The pathogens in recurrent infections are usually different from the initial strain. Even in cases of the same species, the organisms in the recurrent infection may have different colonial morphology and antimicrobial activities. Ascending route from the introital area is an important source of recurrent UTI, and sexual intercourse as well as occult urinary tract abnormalities can facilitate these infections. Urethral diverticulum, infected stone, foreign body, significant anterior vaginal wall relaxation, and duplicated or ectopic ureter can cause persistent bacteriuria and recurrent UTI. Cystourethroscopic and radiographic evaluations should be selectively performed to rule out these abnormalities. Urine culture should be obtained in patients with recurrent infection to rule out resistant microorganisms. Recurrent UTI should also be documented by urine culture at least once. After the urine has been sterilized by a flail course of antibiotic therap34 the management of recurrent UTI can be continuous prophylaxis, postcoital prophylaxis, or seN-initiated therapy by the patient. For women who have frequent reinfections,
Ho AND BHATIA
continuous prophylaxis is cost effective. Nightly doses of nitrofurantoin (100 rag), cephalexin (250 mg), TMP-SMX (80 mg TMP and 400 mg SMX), or cinoxacin (250 to 500 mg) can be used (157). Resistant strains are not likely to occur with these low doses, and most patients continue to maintain sterile urine. Breakthrough infections should be treated with the flail dose of a sensitive antimicrobial drug. In the coital prophylaxis, a single antimicrobial dose can be taken just before or after intercourse. Nitrofurantoin, nalidixic acid, and TMP-SMX are all effective (177). In the self-imflated therapy; the patient is instructed to collect a urine sample for culture when she has the symptoms consistent with a recurrent UTI and then start empirically a 3-day course of flail-dose antibiotic therapy. The antimicrobial drug is usually the previously used antibiotic. This approach has been shown to be safe, effective, and economical in women with recurrent UTI (178,179). Postmenopausal women may have frequently recurrent UTI due to the large residual volume of urine after voiding and the lack of estrogen. Large residual volume is often associated with pelvic organ prolapse. Hypoestrogenism can cause marked changes in the vaginal flora, including a decrease in lactobacilli and an increase in E. coli colonization. In these women, topical estrogen cream can be used as a preventive measure. Estrogen can reverse the microbiologic changes in the vaginal flora that occur after the menopause; however, there is conflicting evidence regarding the benefits of estrogen for prophylaxis against recurrent UTI in these women (180,181). The results from a multicenter and randomized clinical trial indicated that Estring (contains 2 mg estradiol and releases approximately 7.5 ~tg estradiol every 24 hours) may be of benefit when used in postmenopausal women with recurrent UTI (182). Further studies are required before conclusions can be made concerning the role of estrogen prophylaxis for these patients. 7. CATHETER OR INSTRUMENT-AssoCIATED INFECTION
Catheter-associated bacteriuria and UTI are common in hospitalized patients. The risk factors for catheter-associated infections are female sex, advanced age, and serious underlying illnesses. These infections occurred in 16% of newly catheterized patients at a university hospital; however, 90% of the infected patients had no symptoms as reported in one study (183). Although the routes of entry have not been clearly defined, the infections can occur as the results of the introduction of bacteria into the bladder at the time of catheterization, the colonization of bacteria along the catheter, and the ascension of bacteria within the catheter lumen. Prospective studies reported that 66% of the catheter-associated infections were from extraluminal and 34% from intraluminal contaminants (184), and the organisms causing these infections can be identified in the urethral or rectal flora 2 to 4 days before the onset of bacteriuria (185).
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal Women T h e use o f local antimicrobial o i n t m e n t s applied to the meatus and antimicrobial irrigants have no d e m o n strable efficacy (186). Prophylactic a d m i n i s t r a t i o n o f antibiotics c a n n o t be r e c o m m e n d e d because o f its cost and risk of" resistant development, a l t h o u g h antimicrobial agents may reduce the occurrence of bacteriuria in the first few days o f catheterization (187). Clinical trials with silver-hydrogel coated catheters have shown conflicting results and their uses remain undefined ( 1 8 8 - 1 9 0 ) . Preventive efforts should be focused on avoiding catheter manipulations, exchanging the catheter if infection is suspected or cessation of flow, replacing the catheter every 8 to 12 weeks, and keeping fluid intake at about 1.5 liters per day. Normally, asymptomatic bacteriuria in elderly patients with l o n g - t e r m catheterization (more than 3 months) does not require antibiotics. However, 10% of these patients may develop bacteremia and gram-negative septicemia, a serious condition with a 20% to 50% mortality rate (157). These patients must be diagnosed promptly, hospitalized i m m e d i ately, and started on systemic antibiotic treatment vigorously because complications of bacteremia, such as shock, adult respiratory distress syndrome, disseminated intravascular coagulation, and gastric hemorrhage, can occur. For urodynamic and cystourethroscopic procedures, the use of prophylactic antibiotics is controversial. O n e study demonstrated that the prevalences of bacteriuria before and after these procedures were similar and the rates of bacteriuria between the prophylactic and placebo groups were not significantly different, although the power of this study was limited (191). Currently, most clinicians do not give prophylactic antibiotics to low-risk patients who undergo urodynamics or cystourethroscopy.
References 1. Hill K. The demography of the menopause. Maturias 1996;23: 113-127. 2. Iosif CS, Bekassy Z. Prevalenceof genito-urinary symptoms in the late menopause. Acta Obstet Gynecol&and 1984;63:257-260. 3. Barlow DH, Cardozo LD, Francis RM, et al. Urogenital aging and its effect on sexual health in older British women. Br J Obstet Gynaecol 1997;104:87-91. 4. Roberts RO, Jacobsen sJ, Rhodes T, et al. Urinary incontinence in a community-based cohort: prevalence and healthcare-seeking. J Am Geriatr Soc 1998;46:467-472. 5. Wilson L, Brown JS, Shin GP, Luc KO, Subak LL. Annual direct cost of urinary incontinence. Obstet Gyneco12001;98:398-406. 6. Stepp KJ, Walters MD. Anatomy of the lower urinary tract, rectum, and pelvic floor. In: Walters MD, Karram MM, eds. Urogynecologyand reconstructivepelvic surgery, ed. 3. Philadelphia: Mosby, 2007:17- 30. 7. Rogers RM. Anatomy of pelvic support. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecology and pelvk floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003: 19-33.
733
8. DeLancey JOL. Anatomy of the female bladder and urethra. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard'surogynecology andpelvicfloor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:3-18. 9. Benson JT, Wakers MD. Neurophysiology and pharmacology of the lower urinary tract. In: Wakers MD, Karram MM, eds. Urogynecologyand reconstructivepelvic surgery, ed. 3. Philadelphia: Mosby, 2007:31-43. 10. Abrams P, Cardozo L, Fall M, et al., Standardisation Sub-committee of the International Continence Society.The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn 2002;21:167-178. 11. FantlJA, Bump RC, McClish DK. Mixed urinary incontinence. Urology 1990;36:21-24. 12. Germain MM. Urinary retention and overflow incontinence. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard'surogynecology andpdvicfloor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:285-291. 13. Bump RC, Norton PA. Epidemiology and natural history of pelvic floor dysfunction. Obstet Gynecol Clin North Am 1998;25:723-746. 14. Thom D. Variation in estimates of urinary incontinence prevalence in the community: effects of differences in definition, population characteristics, and study type.JAm Geriatr Soc 1998;46:473-480. 15. Pannill FC III. Practical management of urinary incontinence. Med Clin North Am 1989;73:1423-1439. 16. Milsom I, Abrams P, Cardozo L, Roberts RG, Thuroff J, Wein AJ. How widespread are the symptoms of an overactive bladder and how are they managed? A population-based prevalence study. BJU Int 2001;87:760-766. [Erratum in BJUInt 2001;88:807.] 17. Robinson D. Pathophysiology of female lower urinary tract dysfunction. Obstet Gynecol Clin North Am 1998;25:747-756, vi. 18. JolleysJV. Reported prevalence of urinary incontinence in women in a general practice. Br MedJ (Clin Res Ed) 1988;296:1300-1302. 19. Hextall A, Cardozo L. Menopause. In: Cardozo L, Staskin D: Textbook of female urology and urogynaecology. London: Isis Medical Media, 2001:994-1006. 20. Rekers H, Drogendijk AC, Valkenburg H, Riphagen E Urinary incontinence in women from 35 to 79 years of age: prevalence and consequences. EurJ Obstet GynecolReprod Bio11992;43:229-234. 21. Stewart WF, Van RooyenJB, Cundiff GW, et al. Prevalenceand burden of overactivebladder in the United States. WorldJ Uro12003;20:327-336. 22. Chaliha C, Khullar V. Mixed incontinence. Urology2004;63:51-57. 23. Fultz NH, Herzog AR, Raghunathan TE, Wallace RB, Diokno AC. Prevalence and severity of urinary incontinence in older African American and Caucasian women.J Gerontol/l Biol Sci Med Sci 1999;54: M299-M303. 24. Thom DH, van den Eeden SK, Ragins AI, et al. Differences in prevalence of urinary incontinence by race/ethnicity. J Urol 2006;175: 259-264. 25. Samuelsson EC, Victor FT, Svardsudd KE Five-year incidence and remission rates of female urinary incontinence in a Swedish population less than 65 years old. d m J Obstet Gyneco12000;183:568-574. 26. Bent AE. Pathophysiology.In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard'surogynecologyand pelvicfloor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:43-50. 27. DeLancey JO. Structural support of the urethra as it relates to stress urinary incontinence: the hammock hypothesis. Am J Obstet Gynecol 1994;170:1713-1720; discussion, 1720-1723. 28. Weinberger MW, Ostergard DR. Long-term clinical and urodynamic evaluation of the polytetrafluoroethylene suburethral sling for treatment of genuine stress incontinence. Obstet Gyneco11995;86:92-96. 29. Montella JM. Management of overactive bladder. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecology and pelvic floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:293-306.
734 30. Petros PE, Ulmsten U. Bladder instability in women: a premature activation of the micturition reflex. Neurourol Urodyn 1993;12:235-239. 31. Stanton SL, Williams LE, Ritchie D. The colposuspension operation for urinary incontinence. BrJ Obstet Gynaecd 1976;83:890-895. 32. Cardozo LD, Stanton SL, Williams JE. Detrusor instability following surgery for genuine stress incontinence. BrJ Uro11979;51:204-207. 33. Langer R, Ron-el R, Newman M, Herman A, Caspi E. Detrusor instability following colposuspension for urinary stress incontinence.
BrJ Obstet Gynaeco11988;95:607-610. 34. Vierhout ME, Mulder AE De novo detrusor instability after Burch colposuspension. Acta Obstet Gynecol&and 1992;71:414-416. 35. Karram MM, Bhatia NN. Management of coexistent stress and urge urinary incontinence. Obstet Gyneco11989;73:4-7. 36. Swift SE, Bent AE. Basic evaluation of the incontinent female patient. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecologyandpdvicfloor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:69-76. 37. Culligan PJ, Heit M. Urinary incontinence in women: evaluation and management. Am Faro Physician 2000;62:2433-2444,2447,2452. 38. Harvey MA, Versi E. Predictive value of clinical evaluation of stress urinary incontinence: a summary of the published literature. Int UrogynecolJ Pelvic FloorDysfunct 2001;12:31 - 37. 39. Harvey MA, Versi E, Weiss JR Voiding diary. In: Cardozo L, Staskin D. Textbook offemale urology and urogynaecology. London: Isis Medical Media, 2001:167-173. 40. Karram MM. Urodynamics: cystometry and urethral function tests. In: Waiters MD, Karram MM, eds. Urogynecologyand reconstructivepelvic surgery, ed. 3. Philadelphia: Mosby, 2007:78-101. 41. Shumaker SA, Wyman JF, Uebersax JS, McClish D, Fantl JA. Healthrelated quality of life measures for women with urinary incontinence: the Incontinence Impact Q.uestionnaire and the Urogenital Distress Inventory. Continence Program in Women (CPW) Research Group. Qual Life Res 1994;3:291-306. 42. Uebersax JS, Wyman JF, Shumaker SA, McClish DK, Fantl JA. Short forms to assess life quality and symptom distress for urinary incontinence in women: the Incontinence Impact Q.uestionnaire and the Urogenital Distress Inventory. Continence Program for Women Research Group. Neurourol Urodyn 1995;14:131-139. 43. Lind LR, Bhatia NN. Screening neurologic examination with clinical correlation in the evaluation of the incontinent patient. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard'surogynecology and pelvic floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:103-113. 44. Flisser AJ, Blaivas JG. Evaluating incontinence in women. Urol Clin North Am 2002;29:515- 526. 45. Harvey MA, Versi E. Pad Tests. In: Cardozo L, Staskin D. Textbook of female urology and urogynaecology. London: Isis Medical Media, 2001:175-182. 46. Theofrastous JP, Swift SE. Urodynamic testing. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecology and pelv# floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:115-139. 47. van Waalwijk van Doom ES, Returners A, Janknegt RA. Extramural ambulatory urodynamic monitoring during natural filling and normal daily activities: evaluation of 100 patients.J Uro11991;146:124-131. 48. Swift SE, Ostergard DR. Evaluation of current urodynamic testing methods in the diagnosis of genuine stress incontinence. Obstet Gynecol 1995;86:85-91. 49. Karram MM, Miklos JR. Urodynamics: urethral pressure profile and leak point pressures. In: Walters MD, Karram MM, eds. Urogynecology and reconstructivepelvic surgery, ed. 2. St. Louis: Mosby, 1999: 81-93. 50. Karram MM. Urodynamics: voiding studies. In: Walters MD, Karram MM, eds. Urogynecology and reconstructivepelvic surgery, ed. 3. Philadelphia: Mosby, 2007:102-113.
Ho AND BHATIA 51. Walters MD, Huang WC. Radiologic studies of the lower urinary tract. In: Walters MD, Karram MM, eds. Urogynecologyand reconstructivepelvic surgery, ed. 3. Philadelphia: Mosby, 2007:124-137. 52. Pannu H, Nihira MA, Genadry R. Imaging of the pelvic floor. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard'surogynecology andpdvicfloor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:185-204. 53. Herschorn S. Videourodynamics. In: Cardozo L, Staskin D. Textbook of female urology and urogynaecology. London: Isis Medical Media, 2001:263-274. 54. Cundiff GW, Bent AE. Cystourethroscopy. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecology andpelvic floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:141-153. 55. Barber MD. Neurophysiologic testing for pelvic floor disorders. In: Waiters MD, Karram MM, eds. Urogynecologyand reconstructivepelvic surgery, ed. 3. Philadelphia: Mosby, 2007:138-154. 56. Khullar V. Ultrasonography. In: Cardozo L, Staskin D. Textbook of female urology and urogynaecology. London: Isis Medical Media, 2001: 299-312. 57. Iselin CE, Webster GD. Office management of female urinary incontinence. Urol Clin North Am 1998;25:625-645, ix. 58. Visco AG, Figuers C. Nonsurgical management of pelvic floor dysfunction. Obstet GynecolClin North Am 1998;25:849-865, vii. 59. Mantle J. Physiotherapy for incontinence. In: Cardozo L, Staskin D. Textbook of female urology and urogynaecology. London: Isis Medical Media, 2001:352-358. 60. Gormley EA. Biofeedback and behavioral therapy for the management of female urinary incontinence. Urol Clin North Am 2002;29:551-557. 61. Benvenuti F, Caputo GM, Bandinelli S, et al. Reeducative treatment of female genuine stress incontinence. Am J Phys Med 1987;66:155-168. 62. Peattie AB, Plevnik S, Stanton SL. Vaginal cones: a conservative method of treating genuine stress incontinence. Br J Obstet Gynaecol 1988;95:1049-1053. 63. Weatherall M. Biofeedback or pelvic floor muscle exercises for female genuine stress incontinence: a meta-analysis of trials identified in a systematic review. BJU Int 1999;83:1015 - 1016. 64. Berghmans LC, Hendriks HJ, Bo K, et al. Conservative treatment of stress urinary incontinence in women: a systematic review of randomized clinical trials. BrJ Uro11998;82:181-191. 65. FanflJA, Cardozo L, McClish DK. Estrogen therapy in the management of urinary incontinence in postmenopausal women: a meta-analysis. First report of the Hormones and Urogenital Therapy Committee. Obstet
Gyneco11994;83:12-18. 66. Sultana CJ, Waiters MD. Estrogen and urinary incontinence in women. Maturitas 1994;20:129 - 138. 67. Jackson S, Shepherd A, Brookes S, Abrams R The effect of estrogen supplementation on post-menopausal urinary stress incontinence: a double-blind placebo-controlled trial. BrJ Obstet Gynaecol 1999;106: 711-718. 68. Hen&ix SL, Cochrane BB, Nygaard IE, et al. Effects of estrogen with and without progestin on urinary incontinence. J Am IVied/Issoc 2005;293:935-948. 69. Wein AJ, Rovner ES. Pharmacologic management of urinary incontinence in women. Urol Clin North Am 2002;29:537-550, viii. 70. Dmochowski RR, Miklos JR, Norton PA, et al., Duloxetine Urinary Incontinence Study Group. Duloxetine versus placebo for the treatment of North American women with stress urinary incontinence. J Uro12003;170:1259-1263. [Erratum in J Uro12004;171:360.] 71. Drutz H. Duloxetine in women awaiting surgery. BJOG 2006; 113:17-21. 72. Mariappan P, Alhasso A, Ballantyne Z, et al. Duloxetine, a serotonin and noradrenaline reuptake inhibitors (SNRI) for the treatment of stress urinary incontinence: a systemic review. Eur Uro12007;51:67-74.
735
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal W o m e n 73. Chaikin DC, Groutz A, Blaivas JG. Predicting the need for antiincontinence surgery in continent women undergoing repair of severe urogenital prolapse. J Uro12000;163:531 - 534. 74. Kohli N, Karram MM. Surgery for genuine stress incontinence: vaginal procedures, injections, and the artificial urinary sphincter. In: Waiters MD, Karram MM, eds. Urogynecologyand reconstructivepelvic surgery, ed. 2. St. Louis: Mosby, 1999:171-196. 75. Beck RP, McCormick S, Nordstrom L. A 25-year experience with 519 anterior colporrhaphy procedures. Obstet Gyneco11991;78:1011-1018. 76. Bodell DM, Leach GE. Needle suspension procedures for female incontinence. Uro!Clin North elm 2002;29:575- 584. 77. Bent AE, McLennan MT. Surgical management of urinary incontinence. Obstet Gynecol Clin North elm 1998;25:883-906. 78. Niknejad K, Plzak LS 3rd, Staskin DR, Loughlin KR. Autologous and synthetic urethral slings for female incontinence. Uro! Clin North elm 2002;29:597- 611. 79. Kubic K, Horbach NS. Suburethral sling procedures and treatment of complicated stress incontinence. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard'surogynecologyandpelvicftoor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:469-493. 80. Weber AM, Walters MD. Butch procedure compared with sling for stress urinary incontinence: a decision analysis. Obstet Gyneco12000;96: 867-873. 81. Klutke JJ, Klutke CG. The tension-free vaginal tape procedure: innovative surgery for incontinence. Curt Opin Obstet Gynecol 2001;13: 529-532. 82. Atherton MJ, Stanton SL. The tension-free vaginal tape reviewed: an evidence-based review from inception to current status. BJOG 2005;112:534-546. 83. Delorme E, Droupy S, de Tayrac R, Delmas V. Transobturator tape (Uratape): a new minimally-invasive procedure to treat female urinary incontinence. Eur Uro12004;45:203-207. 84. deTayrac R, Deffieux X, Droupy S, et al. A prospective randomized trial comparing tension-free vaginal tape and transobturator suburethral tape for surgical treatment of stress urinary incontinence, elm J
Obstet Gyneco12004;190:602-608. 85. Stoffel JT, Bresette JF, Smith JJ. Retropubic surgery for stress urinary incontinence. Urol Clin North elm 2002;29:585-596, viii-ix. 86. Blomquist J, Germain MM. Surgical correction of stress incontinence with hypermobility. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecology and pelvic floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:457-468. 87. Liu CY. Laparoscopic treatment of stress urinary incontinence. Obstet Gynecol Clin North elm 1999;26:149-167. 88. Gross M, Appel RA. Periurethral injections. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecologyandpelvic floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:495-502. 89. Staskin DR. Overactive bladder in the elderly: a guide to pharmacological management. DrugsAging 2005;22:1013-1028. 90. Dmochowski RR, Sand PK, Zinner NR, et al., Transdermal Oxybutynin Study Group. Comparative efficacy and safety of transdermal oxybutynin and oral tolterodine versus placebo in previously treated patients with urge and mixed urinary incontinence. Urology 2003; 62:237-242. 91. Harvey MA, Baker K, Wells GA. Tolterodine versus oxybutynin in the treatment of urge urinary incontinence: a meta-analysis, elm J Obstet
Gyneco12001;185:56-61. 92. Lose G, Jorgensen L, Thunedborg E Doxepin in the treatment of female detrusor overactivity: a randomized double-blind crossover study. J Uro11989;142:1024-1026. 93. Castleden CM, Duffin HM, Gulati RS. Double-blind study of imipfamine and placebo for incontinence due to bladder instability. Age Aging 1986;15:299-303.
94. McLennan MT. Sacral neuromodulation. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecologyandpelvic floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:325-338. 95. Janknegt RA, Hassouna MM, Siegel SW, et al. Long-term effectiveness of sacral nerve stimulation for refractory urge incontinence. Eur
Uro12001;39:101-106. 96. van Voskuilen AC, Oerlemans DJ, Weil EH, de Bie RA, van Kerrebroeck PE. Long term results of neuromodulation by sacral nerve stimulation for lower urinary tract symptoms: a retrospective single center study. Eur Uro12006;49:366-372. 97. Ho MH, Lin LL, Haessler AL, Bhatia NN. Intravesical injection of botulinum toxin for the treatment of overactive bladder. Curr Opin
Obstet Gyneco12005;17:512-518. 98. Langer R, Ron-E1 R, Bukovsky I, Caspi E. Colposuspension in patients with combined stress incontinence and detrusor instability.
Eur Uro11988;14:437-439. 99. Scotti RJ, Angell G, Flora R, Greston WM. Antecedent history as a predictor of surgical cure of urgency symptoms in mixed incontinence.
Obstet Gyneco11998;91:51-54. 100. Paick JS, Ku JH, Kim SW, et al. Tension-free vaginal tape procedure for the treatment of mixed urinary incontinence: significance of maximal urethral closure pressure. J Uro!2004;172:1001-1005. 101. Holmgren C, Nilsson S, Lanner L, Hellberg D. Long-term results with tension-free vaginal tape on mixed and stress urinary incontinence. Obstet Gyneco12005;106:38-43. 102. Simon LJ, Landis JR, Erickson DR, Nyberg LM. The Interstitial Cystitis Data Base Study: concepts and preliminary baseline descriptive statistics. Urology 1997;49:64- 75. 103. Curhan GC, Speizer FE, Hunter DJ, Curhan SG, Stampfer MJ. Epidemiology of interstitial cystitis: a population based study. J Uro! 1999;161:549-552. 104. Leppilahti M, Sairanen J, Tammela TL, et al., Finnish Interstitial Cystitis-Pelvic Pain Syndrome Study Group. Prevalence of clinically confirmed interstitial cystitis in women: a population based study in Finland. J Uro!2005;174:581- 583. 105. Rosenberg MT, Hazzard M. Prevalence of interstitial cystitis symptoms in women: a population based study in the primary care office.
J Uro12005;174:2231-2234. 106. Parsons CL. Diagnosing chronic pelvic pain of bladder origin. J Reprod Med 2004;49:235-242. 107. Parsons CL, Dell J, Stanford EJ, et al. The prevalence of interstitial cystitis in gynecologic patients with pelvic pain, as detected by intravesical potassium sensitivity.Am J Obstet Gyneco12002;187:1395-1400. 108. Parsons CL, Bullen M, Kahn BS, Stanford EJ, Willems JJ. Gynecologic presentation of interstitial cystitis as detected by intravesical potassium sensitivity. Obstet Gyneco12001;98:127-132. 109. Chai TC. Interstitial cystitis. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard'surogynecologyandpelvicfloor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:307-323. 110. Warren JW, Keay SK, Meyers D, Xu J. Concordance of interstitial cystitis in monozygotic and dizygotic twin pairs. Urology 2001;57: 22-25. 111. Parsons CL. A model for the function of glycosaminoglycans in the urinary tract. WorldJ Uro11994;12:38-42. 112. Lilly JD, Parsons CL. Bladder surface glycosaminoglycans is a human epithelial permeability barrier. Surg GynecolObstet 1990; 171:493 - 496. 113. Chelsky MJ, Rosen SI, Knight LC, et al. Bladder permeability in interstitial cystitis is similar to that of normal volunteers: direct measurement by transvesical absorption of 99mtechnetiumdiethylenetriaminepentaacetic acid. J Uro11994;151:346-349. 114. Sun Y, Keay S, De Deyne PG, Chai TC. Augmented stretch activated adenosine triphosphate release from bladder uroepithelial cells in patients with interstitial cystitis. J Uro12001;166:1951-1956.
736 115. Keay SK, Zhang CO, Shoenfelt J, et al. Sensitivity and specificity of antiproliferative factor, heparin-binding epidermal growth factor-like growth factor, and epidermal growth factor as urine markers for interstitial cystitis. Urology2001;57:9-14. 116. Hanno PM. Interstitial cystitis and related disorders. In: Campbell MF, Walsh PC, Alan B, Retik AB, eds. Campbell's urology, ed. 8. Philadelphia: W.B. Saunders, 2002:631-670. 117. Gillenwater JY, Wein AJ. Summary of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases Workshop on Interstitial Cystitis, National Institutes of Health, Bethesda, Maryland, August 28-29, 1987.J Uro11988;140:203- 206. 118. Hanno PM, Landis JR, Matthews-Cook Y, Kusek J, Nyberg L Jr. The diagnosis of interstitial cystitis revisited: lessons learned from the National Institutes of Health Interstitial Cystitis Database study. J Uro11999;161:553-557. 119. Nickel JC. Interstitial cystitis: a chronic pelvic pain syndrome. Med Clin North Am 2004;88:467-481, xii. 120. O'Leary MP, Sant GR, Fowler FJ Jr, Whitmore KE, Spolarich-Kroll J. The interstitial cystitis symptom index and problem index. Urology 1997;49:58-63. 121. Clemons JL, Arya LA, Myers DL. Diagnosing interstitial cystitis in women with chronic pelvic pain. Obstet Gyneco12002;100:337-341. 122. Goin JE, Olaleye D, Peters KM, et al. Psychometric analysis of the University of Wisconsin Interstitial Cystitis Scale: implications for use in randomized clinical trials.J Uro11998;159:1085-1090. 123. Waxman JA, Sulak PJ, Kuehl TJ. Cytoscopic findings consistent with interstitial cystitis in normal women undergoing tubal ligation. J Urol 1998;160:1663 - 1667. 124. Chai TC, Zhang CO, Shoenfelt JL, et al. Bladder stretch alters urinary heparin-binding epidermal growth factor and antiproliferative factor in patients with interstitial cystitis. J Uro12000;163:1440-1444. 125. Rofeim O, Hom D, Freid RM, Moldwin RM. Use of the neodymium:YAG laser for interstitial cystitis: a prospective study. J Urol 2000;166:134-136. 126. Parsons CL, Greenberger M, Gabal L, Bidair M, Barme G. The role of urinary potassium in the pathogenesis and diagnosis of interstitial cystitis. J Uro11998;159:1862-1866; discussion, 1866-1867. 127. Chambers GK, Fenster HN, Cripps S, Jens M, Taylor D. An assessment of the use of intravesical potassium in the diagnosis of interstitial cystitis. J Uro11999;162:699- 701. 128. Teichman JM, Nielsen-Omeis BJ. Potassium leak test predicts outcome in interstitial cystitis. J Uro11999;161:1791-1794; discussion, 1794-1796. [Erratum in J Uro11999;162:503.] 129. Byrne DS, Sedor JF, Estojak J, et al. The urinary glycoprotein GP651 as a clinical marker for interstitial cystitis. J Uro11999;161:1786-1790. 130. Erickson DR, Xie SX, Bhavanandan VP, et al. A comparison of multiple urine markers for interstitial cystitis.J Uro12002;167:2461-2469. 131. Whitmore KE. Serf-care regimens for patients with interstitial cystitis. Urol Clin North Am 1994;21:121-130. 132. Gillespie L. Metabolic appraisal of the effects of dietary modification on hypersensitive bladder symptoms. BrJ Uro11993;72:293-297. 133. Hanno PM. Analysis of long-term Elmiron therapy for interstitial cystitis. Urology 1997;49:93 - 99. 134. Mulholland SG, Hanno P, Parsons CL, Sant GR, Staskin DR. Pentosan polysulfate sodium for therapy of interstitial cystitis. A doubleblind placebo-controlled clinical study. Urology1990;35:552-558. 135. Nickel JC, Barkin J, Forrest J, et al., for Elmiron Study Group. Randomized, double-blind, dose-ranging study of pentosan polysulfate sodium for interstitial cystitis. Urology2005;65:654-658. 136. van Ophoven A, Pokupic S, Heinecke A, Hertle L. A prospective, randomized, placebo controlled, double-blind study of amitriptyline for the treatment of interstitial cystitis. J Uro12004;172:533 - 536. 137. Van Ophoven A, Hertle L. Long-term results of amitriptyline treatment for interstitial cystitis. J Uro12005;174:1837-1840.
Ho AND BHATIA 138. Theoharides TC, Sant GR. Hydroxyzine therapy for interstitial cystitis. Urology1997;49:108-110. 139. Sasaki K, Smith CP, Chuang YC, et al. Oral gabapentin (neurontin) treatment of refractory genitourinary tract pain. Tech Urol 2001;7: 47-49. 140. Perez-Marrero R, Emerson LE, Feltis JT. A controlled study of dimethyl sulfoxide in interstitial cystitis. J Uro11988;140:36- 39. 141. Bade JJ, Laseur M, Nieuwenburg A, van der Weele LT, Mensink HJ. A placebo-controlled study of intravesical pentosan polysulphate for the treatment of interstitial cystitis. BrJ Uro11997;79:168-171. 142. Parsons CL, Housley T, Schmidt JD, Lebow D. Treatment of interstitial cystitis with intravesical heparin. BrJ Uro11994;73:504-507. 143. Parsons CL. Successful downregulation of bladder sensory nerves with combination of heparin and alkalinized lidocaine in patients with interstitial cystitis. Urology2005;65:45-48. 144. Kallestrup EB, Jorgensen SS, Nordling J, Hald T. Treatment of interstitial cystitis with Cystistat: a hyaluronic acid product. ScandJ Urol Nephro12005;39:143 - 147. 145. Payne CK, Mosbaugh PG, Forrest JB, et al., ICOS RTX Study Group (Resiniferatoxin Treatment for Interstitial Cystitis). Intravesical resiniferatoxin for the treatment of interstitial cystitis: a randomized, doubleblind, placebo-controlled trial. J Uro12005;173:1590-1594. 146. Peters K, Diokno A, Steinert B, et al. The efficacy ofintravesical Tice strain bacillus Calmette-Guerin in the treatment of interstitial cystitis: a double-blind, prospective, placebo-controlled trial. J Uro11997;157: 2090-2094. 147. Peeker R, Haghsheno MA, Holmang S, Fall M. Intravesical bacillus Calmette-Guerin and dimethyl sulfoxide for treatment of classic and nonulcer interstitial cystitis: a prospective, randomized, double-blind study. J Uro12000;164:1912-1915; discussion, 1915-1916. 148. Mayer R, Propert KJ, Peters KM, et al., Interstitial Cystitis Clinical Trials Group. A randomized controlled trial of intravesical bacillus Calmette-Guerin for treatment-refractory interstitial cystitis. J Urol 2005;173:1186-1191. 149. Comiter CV. Sacral neuromodulation for the symptomatic treatment of refractory interstitial cystitis: a prospective study. J Uro12003;169: 1369-1373. 150. Whitmore KE, Payne CK, Diokno AC, Lukban JC. Sacral neuromodulation in patients with interstitial cystitis: a multicenter clinical trial. Int Urogynecol J Pelvic Floor Dysfunction 2003;14:305-308; discussion, 308-309. 151. Fall M, Lindstrom S. Transcutaneous electrical nerve stimulation in classic and nonulcer interstitial cystitis. Urol Clin North Am 1994;21: 131-139. 152. Chancellor MB, Yoshimura N. Treatment of interstitial cystitis. Urology 2004;63:85-92. 153. Kuo HC. Preliminary results of suburothelial injection of botulinum A toxin in the treatment of chronic interstitial cystitis. Urol Int 2005;75:170-174. 154. Strom BL, Collins M, West SL, KreisbergJ, Weller S. Sexual activity, contraceptive use, and other risk factors for symptomatic and asymptomatic bacteriuria. A case-control study. Ann Intern Med 1987;107: 816-823. 155. Asscher AW. Urinary tract infection. J Royal Coll Physicians Lond 1981;15:232-238. 156. Hextall A, Hooper R, Cardozo L, Stringer C, Workman R. Does the menopause influence the risk of bacteriuria? Int UrogynecolJ Pelvic Floor Dysfunction 2001;12:332- 336. 157. Karran MM, Kleeman SD. Lower urinary tract infection. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard'surogynecology and pelvicfloor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:261-284.
CHAPTER 51 Lower Urinary Tract Disorders in Postmenopausal W o m e n 158. Nicolle LE, Hoban SA, Harding GK. Characterization of coagulasenegative staphylococci from urinary tract specimens. J Clin Microbiol 1983;17:267-271. 159. Daifuku R, Stamm WE. Bacterial adherence to bladder uroepithelial cells in cathether-associated urinary tract infection. N Engl J Med 1986;314:1208-1213. 160. Hu KK, Boyko EJ, Scholes D, et al. Risk factors for urinary tract infections in postmenopausal women. Arch Intern Med 2004;164:989-993. 161. Saemann MD, Weichhart T, Horl WH, Zlabinger GJ. Tamm-Horsfall protein: a multilayered defense molecule against urinary tract infection. EurJ Clin Invest 2005;35:227-235. 162. Cengiz N, Baskin E, Anarat R, et al. Glycosaminoglycans in childhood urinary tract infections. Pediatr Nephro12005;20:937- 939. 163. Deville WL, Yzermans JC, van Duijn NP, et al. The urine dipstick test useful to rule out infections. A meta-analysis of the accuracy. BMC Uro12004;4:4. 164. Wu TC, Williams EC, Koo SY, MacLowry JD. Evaluation of three bacteriuria screening methods in a clinical research hospital. J Clin Microbio11985;21:796- 799. 165. Bixler-Forell E, Bertram MA, Bruckner DA. Clinical evaluation of three rapid methods for the detection of significant bacteriuria. J Clin Microbio11985;22:62-67. 166. Stamm WE. Measurement of pyuria and its relation to bacteriuria. AmJMed 1983;73:53-58. 167. Johnson JR, Stamm WE. Diagnosis and treatment of acute urinary tract infections. Infect Dis Clin North Am 1987;1:773-791. [Erratum in Infect Dis Clin North Am 1990;4, following xii.] 168. Stamm WE, Counts GW, Running KR, et al. Diagnosis of coliform infection in acutely dysuric women. N EnglJ Med 1982;307:463-468. 169. Valenstein P, Meier E Urine culture contamination: a College of American Pathologists QzProbes study of contaminated urine cultures in 906 institutions. Arch Pathol Lab Med 1998;122:123-129. 170. Fowler JE Jr, Pulaski ET. Excretory urography, cystography, and cystoscopy in the evaluation of women with urinary tract infection: a prospective study. N EnglJ Med 1981;304:462-465. 171. Sobatao AE. Inhibition of bacterial adherence by cranberry juice: potential use for the treatment of urinary tract infections. J Urol 1984;131:1013-1016. 172. Potts L, Cross S, MacLennan WJ, Watt B. A double-blind comparative study of norfloxacin versus placebo in hospitalized elderly patients with asymptomatic bacteriuria. Arch Gerontol Geriatr 1996;23:153 - 161. 173. Nicolle LE. Asymptomatic bacteriuria: when to screen and when to treat. Infect Dis Clin North Am 2003;17:367-394. 174. Fihn SD, Johnson C, Roberts PL, Running K, Stamm WE. Trimethoprim-sulfamethoxazole for actute dysuria in women: a single-dose or 10-day course. A double-blind, randomized trial. Ann Intern Med 1988;108:350-357. 175. Auquer F, Cordon F, Gorina E, et al., Urinary Tract Infection Study Group. Single-dose ciprofloxacin versus 3 days of norfloxacin in uncomplicated urinary tract infections in women. Clin Microbiol Infect 2002;8:50-54.
737
176. Saint S, Scholes D, Fihn SD, Farrell RG, Stamm WE. The effectiveness of a clinical practice guideline for the management of presumed uncomplicated urinary tract infection in women. Am J Med 1999; 106:636-641. 177. Pfau A, Sacks T, Engelstein D. Recurrent urinary tract infection in premenopausal women: prophylaxis based on an understanding of the pathogenesis. J Uro11983;129:1153-1157. 178. Gupta K, Hooton TM, Roberts PL, Stamm WE. Patient-initiated treatment of uncomplicated recurrent urinary tract infections in young women. Ann Intern Med 2001;135:9-16. 179. Schaeffer AJ, Stuppy BA. Efficacy and safety of self-start therapy in women with recurrent urinary tract infections. J Urol 1999;161: 207-211. 180. Raz R, Stamm WE. A controlled trial of intravaginal estriol in postmenopausal women with recurrent urinary tract infections. N EnglJ Med 1993;329:753-756. 181. Cardozo L, Benness C, Abbott D. Low dose oestrogen prophylaxis for recurrent urinary tract infections in elderly women. Br J Obset Gyneco11998;105:403-407. 182. Eriksen B. A randomized, open, parallel-group study on the preventive effect of an estradiol-releasing vaginal ring (Estring) on recurrent urinary tract infections in postmenopausal women. Am J Obset Gynecol 1999;180:1072-1079. 183. Tambyah PA, Maki DG. Cathether-associated urinary tract infection is rarely symptomatic: a prospective study of 1497 catheterized patients. Arch Intern Med 2000;160:678-782. 184. Tambyah PA, Halvorson KT, Maki DG. A prospective study of pathogenesis of catether-associated urinary tract infections. Mayo Clin Proc 1999;74:131-136. 185. Garibaldi RA, Burke JP, Britt MR, Miller MA, Smith CB. Meatal colonization and catheter-associated bacteriuria. N Engl J Med 1980;303:316-318. 186. Tambyah PA. Catheter-associated urinary tract infections: diagnosis and prophylaxis. IntJAntimicrob Agents 2004;24:$44- $48. 187. Warren JW. Cathether-associated urinary tract infections. Int J Antimicrob Agents 2001;17:299- 303. 188. Rupp ME, Fitzgerald T, Marion N, et al. Effect of silver-coated urinary catheters: efficacy, cost-effectiveness, and antimicrobial resistance. AmJ Infect Control 2004;32:445-450. 189. Lai KK, Fontecchio SA. Use of silver-hydrogel urinary catheters on the incidence of catheter-associated urinary tract infections in hospitalized patients. Am J Infect Control 2002;30:221-225. 190. Johnson JR, Kuskowski MA, Wilt TJ. Systematic review: antimicrobial urinary catheters to prevent catheter-associated urinary tract infection in hospitalized patients. Ann Intern Med 2006;144:116-126. 191. Cundiff GW, McLennan MT, Bent AE. Randomized trial of antibiotic prophylaxis for combined urodynamics and cystourethroscopy. Obstet Gyneco11999;93:749- 752.
This Page Intentionally Left Blank
; H A P T E R 5s
Pelvic Organ Prolapse in Postmenopausal Women MAT H. H o
Division of Female Pelvic Medicine and Reconstructive Surgery, Department of Obstetrics and Gynecology, Harbor-UCLA Medical Center, David Geffen School of Medicine, University of California at Los Angeles, Torrance, CA 90509
NARENDER N. B H A T I A
Division of Female Pelvic Medicine and Reconstructive Surgery, Department of Obstetrics and Gynecology, Harbor-UCLA Medical Center, David Geffen School of Medicine, University of California at Los Angeles, Torrance, CA 90509
I. I N T R O D U C T I O N
II. A N A T O M Y O F T H E PELVIC SUPPORT
In the United States and other developed countries, women are expected to spend a significant part of their life in the postmenopausal years, a period when pelvic organ prolapse (POP) is more likely to occur (1). POP is the abnormal protrusion of the pelvic organs from their normal locations into or out of the vaginal canal. These abnormalities, which result from defects of the pelvic support, include anterior compartment prolapse such as cystocele or urethrocele, posterior compartment prolapse such as rectocele or enterocele, and apical problems such as uterine or vaginal vault prolapse. Enterocele can also occur in conjunction with the defects of the anterior and apical compartments. These conditions often occur in various combinations and/or with the coexistence of urinary incontinence. Although POP seldom leads to serious medical illnesses, it may have a major effect on the patient's quality of life. Significant morbidity can occur with POP including alterations in bladder, bowel, or sexual function. These patients often suffer from urinary frequency and urgency urinary incontinence, fecal incontinence, constipation, and a decrease in sexual activity. Conditions of POP also are associated with loss of self-esteem, social isolation, depression, and other psychologic burdens. TREATMENT OF THE POSTMENOPAUSAL WOMAN
The pelvic floor is made up of muscular and fascial structures (2). The muscular component consists of levator ani muscles that provide dynamic support to the pelvis with their constantly adjusting tone. The fascial component, which involves parietal and visceral fasciae, provides passive support. The parietal fascia covers the pelvic muscles and attaches them to the bony pelvis. The visceral fascia is involved in supporting pelvic organs and is collectively known as the endopelvic fascia, a heterogeneous group of tissues consisting of collagen, elastin, smooth muscle, blood vessels, and nerves. The condensation of various parts of the visceral fascia form structures such as the uterosacral-cardinal ligament complex, the pubocervical fascia, and the rectovaginal fascia. Although many questions remain unresolved, our understanding of female pelvic support has improved greatly during the past decade. The integrity of the pelvic support network depends on the pelvic bones, muscles, ligaments, endopelvic fascia, and their innervations. Detailed knowledge of these structures facilitates the evaluation and management of POE 739
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
740
Ho
A. Bony Pelvis The borders of the bony peMs involve the pubic bone anteriorly, the ischial spines laterally, and the sacrum posteriorly. The ilium, ischium, and pubis bones are fused together to form the bony pelvis (Fig. 52.1). It articulates posteriorly with the sacrum at the sacroiliac joint and anteriorly with the other pubic bones at the pubic symphysis. In the normal standing position, the ischial spines and the posterior aspect of the pubic bones are almost in the horizontal plane. The sacrospinous ligaments and their overlying coccygeus muscles travel from the ischial spine to the lower sacrum and coccyx. The sacrotuberous ligaments connect the ischial tuberosity to the lateral and posterior aspects of the lower sacrum and form the posterior border of the pelvic outlet. The obturator foramen is formed by the pubis and ischium and covered by the obturator membrane.
B. Levator Ani Muscles The arcus tendineus levator ani (muscle white line), which is the thickening of the parietal fascia covering the obturator internus muscle, runs from the posterior and lateral aspects of the pubic bone to the ischial spine, and gives rise to the levator ani muscles. The levator ani muscles include the bilaterally paired pubococcygeus, puborectalis, and iliococcygeus (Fig. 52.2). Some authors also include the
AND BHATIA
coccygeus muscle in the levator ani complex. The pubococcygeus muscles originate from the superior ramus of the pubis and the fascia covering the obturator internus. It surrounds the lower vagina, meets at the midline behind the rectum to form the levator plate, and inserts onto the coccyx. The puborectalis muscle is the medial and inferior part of the pubococcygeus that fuses with the external anal sphincter to form a sling around the rectum at the anorectal junction, which upon contraction moves this junction toward the pubic symphysis to maintain fecal continence. The iliococcygeus muscles arise from the arcus tendineus levator ani, fuse together in the midline behind the rectum to form the levator plate, and insert onto the coccyx. The coccygeus muscles attach anteriorly to the ischial spines and fan out medially to the lateral surface of the coccyx. This triangular structure does not contribute to the active movement of the pelvic floor. The genital hiatus is the gap in the midline through which passes the urethra, vagina, and rectum. The levator ani muscles and the levator plate are dynamic structures that continually adjust their tension in response to changes in intra-abdominal and intra-pelvic pressures. Contractions of the pubococcygeus and puborectalis are responsible for the anterior movements of the vagina and anorectal junction toward the pubic symphysis. These anterior movements also produce occlusive forces on the urethra, thus contributing to urinary continence. In a normal woman in the standing position, the levator plate, which runs from the anorectal junction to the lower sacrum and coccyx, is in a
FIGURE52.1 Bonesand ligaments of the femalepelvisviewedfrom median (sagittal) section. (From Netter FH. Atlas of human anatomy. East Hanover, NJ: Novartis, 1997:plate 330. Copyright 9 Icon Learning Systems, LLC, a subsidiaryof MediMedia, USA, Inc.)
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal Women
741
FIGURE 52.2 Muscles of the female pelvis. (From Netter FH. Atlas of human anatomy. East Hanover, NJ: Novartis, 1997:plate 333. Copyright 9 Icon Learning Systems, LLC, a subsidiary of MediMedia, USA, Inc.)
horizontal plane. Its primary function is to prevent the uterine and vaginal wall from prolapsing by keeping the urogenital hiatus constricted and relatively closed. When its muscle tone is lost, the levator plate can sag down, thus widening the levator hiatus and predisposing to prolapse (see Fig. 52.2). The arcus tendineus fasciae pelvis (or fascial white line) is a thickening of the parietal fascia of levator ani muscles along the straight line from the pubic arch to the ischial spine bilaterally (see Fig. 52.2). The fascial white lines are the attachment sites for the pubocervical and rectovaginal fasciae. These fascial white lines should be distinguished from the arcus tendineus levator ani (muscle white line). The arcus tendineus fasciae rectovaginalis is a thickening of the parietal fascia of the iliococcygeus muscle along the line from the ischial spine to the perineal body. From the ischial spine to the level of the mid-vagina, the arcus tendineus fasciae rectovaginalis and the arcus tendineus fasciae pelvis share the same line. After that point, the arcus tendineus fasciae rectovaginalis turns toward the perineal body while the arcus tendineus fasciae pelvis continues to the pubic arch. The lower portion of the rectovaginal fascia inserts onto this fascial line and the apex of perineal body. Levator ani muscles receive innervations from both sacral efferent and pudendal nerves (2,3). The $2-$4 sacral nerves innervate the pelvic or superior surface of these muscles, while branches of the pudendal nerve innervate the perineal or inferior surface. The pudendal nerve, which is derived from the lumbosacral plexus, is a mixed nerve that carries both motor and sensory fibers.
C. Perineum and Perineal Body The perineum lies just inferior to the levator ani muscles and has a border with the pubic arch anteriorly and the coccyx tip posteriorly. The perineum can be divided into two triangular portions by the line between the ischial tuberosities. The anterior portion is called the urogenital triangle and contains the vaginal and urethral outlets. The posterior portion is known as the anal triangle and contains the anal canal. The perineal membrane, which travels between the two ischiopubic rami, provides the key support for the urogenital triangle and divides this triangle into superficial and deep compartments. The perineal body is a pyramidal structure that is located between the anus and vaginal outlet (Fig. 52.3). The bulbocavernosus muscles, the superficial and deep transverse perineal muscles, a portion of levator ani muscles, and the rectovaginal fascia are all inserted into the perineal body (2). The apex of this structure, which is at the level of the lower middle third of the vagina, is attached to the rectovaginal fascia and then to the uterosacral ligaments. This attachment helps to suspend and stabilize the perineal body.
D. Pelvic Support 1. LEVELS OF SUPPORT
The pelvic support can be divided into three levels from superior to inferior (4). The first level involves the cardinal-uterosacral ligament complexes that support the
742
Ho AND BHATIA
FIGURE 52.3 Female perineum and muscles. (Reprinted from Skandalalds JE, Gray SW, Mansberger AR Jr, et al. Hernia: surgical anatomy and technique. New York: McGraw-Hill, 1989:244-250.)
cervix and upper vagina bilaterally. The second level involves the pubocervical and rectovaginal fasciae that support the bladder, the proximal urethra, upper two-thirds of the vagina, and the rectum. The third level involves the levator ani muscles, the perineal body, and the perineum. At the first level, the cardinal-uterosacral ligament complexes support the cervix and upper vagina and, in posthysterectomy, the vaginal cuff. The cardinal ligament is a sheath of visceral fibers that follow the course of internal iliac vessels and fuse into the pericervical ring. The cardinal ligament also attaches to the lower uterine segment and upper vagina. Laterally and posteriorly, this ligament anchors firmly to the parietal fascia of the obturator internus, piriformis, and greater sciatic foramen. The uterosacral ligament is a condensation of the visceral connective tissues that attaches to the presacral fascia of $2-$4 vertebra. After curving around the rectum, this ligament inserts onto the posterior and lateral aspects of the pericervical ring. Detachment of the cardinal-uterosacral ligament complexes from the pericervical ring may lead to uterine prolapse, vaginal vault prolapse (post-hysterectomy), and apical enterocele. At the second level, the pubocervical and rectovaginal fasciae provide support and stabilization to the horizontal orientation of the bladder, the upper two-thirds of the vagina, and the rectum. It should be noted that pubocervical and rectovaginal fasciae are the clinical and surgical designations rather than the histologic terms. The pubocervical fascia is a thickening of the visceral fibromuscular coat of the anterior vaginal wall and is fused with the vaginal epithelium (2). Laterally, the pubocervical fascia attaches to the arcus tendineus fasciae pelvis on both sides, and posteriorly, it inserts onto the pericervical ring to provide a platform upon which the bladder and urethra rest. The pubocervical fascia also connects to the cardinal-uterosacral ligament complexes at the pericervical ring. Inferiorly, this fascia inserts onto the
perineal membrane. Intact pubocervical fascia and its attachments prevent anterior vaginal wall prolapse, such as cystocele or urethrocele, and anterior enterocele. On its superior surface, this fascia also reflects around the urethra and condenses to form the pubourethral ligament, which attaches to the pubic arch on both sides and provides stabilization for the mid-urethra. The pubocervical fascia also forms a hammock underneath the urethrovesical junction and proximal urethra, thus playing an important role in supporting bladder neck and maintaining incontinence. The rectovaginal fascia (also known as the rectovaginal septum) lies between the fascial coats of the vagina and the rectum. It is not a thickening of fascial coat, but rather a fusion of two peritoneal layers (2). Posteriorly, the rectovaginal fascia connects to the uterosacral ligaments, at the point of their insertions onto the pericervical ring, and to the cul-de-sac peritoneum. Laterally, its upper half attaches to the fascia of the iliococcygeus muscles, which then inserts into the fascial white line while the lower half inserts onto the arcus tendineus fasciae rectovaginalis. Anteriorly, it fuses with the apex of the perineal body. Intact rectovaginal fascia and its attachments prevent the development of posterior vaginal wall prolapse, such as rectocele and enterocele. Rectovaginal fascia also suspends the perineal body by its connections to this structure and to the uterosacral ligaments. The third level of support involves the distal urethra, lower third of the vagina, and anal canal that travel through the levator hiatus of the levator ani muscle complex and then through the perineum in both the urogenital and anal triangles. The supports at this level are provided by the levator ani muscles as well as the attachments and fusions of the visceral fascia, that envelops the distal urethra, lower third of the vagina, and anal canal, to the parietal fascia of the pubococcygeus and puborectalis muscles. The perineal body also supports the closure of vaginal introitus.
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal Women
2. SUPPORT OF THE CERVIX
The cervix is normally found supported above the level of the ischial spines and stabilized by the cardinaluterosacral ligament complexes bilaterally. The pubocervical fascia and the rectovaginal fascia fuse with the pericervical ring just underneath the vaginal epithelium to form the anterior fornix and posterior fornix, respectively. The posterior fornix is also bounded by the insertion of the uterosacral ligament into the posterolateral aspects of the pericervical ring. In women after hysterectomy, the pubocervical fascia should be attached to the rectovaginal fascia at the apex of the vagina.
3. SUPPORT OF THE VAGINA
The anterior vaginal wall consists of the pubocervical fascia and its covering epithelium. From the vaginal canal, the observed anterolateral suci represent attachments of the pubocervical fascia along each sidewall to the fascial white line. The posterior vaginal wall contains the epithelium and underlying visceral fascia, which is generally thinner than the pubocervical fascia. Between the visceral fascia of the vagina and perirectal fat is the fibroelastic sheath of the rectovaginal fascia. In standing position, the bladder, upper two-thirds of the vagina, and rectum are oriented in a more horizontal axis, whereas the urethra, distal third of the vagina, and anal canal are more vertical. In the supine position, the upper vagina is horizontal and superior to the levator plate (2,5). 4. SUPPORT OF THE ANORECTAL CANAL
The walls of the anorectal canal are supported by the levator plate posteriorly and the rectovaginal fascia anteriorly. The lateral walls are attached to the downward extension of the uterosacral ligaments. The anal canal is also surrounded anteriorly by the perineal body and laterally as well as posteriorly by the puborectalis muscles.
III. CLASSIFICATION AND QUANTIFICATION OF PROLAPSE A standardized description and classification of POP are important in the documentation and communication of the problem as well as in establishing the guidelines for treatment. Although the traditional descriptions of POP have been widely used, the compartmental and site-specific descriptions are now advocated. Methods for clinical grading or staging of POP, which include the halfway system, the modification system, and the Pelvic Organ Prolapse Q.uantification (POPQ) system, are also critical in the evaluation and treatment of these problems.
743
A. Classification and Description of Pelvic Organ Prolapse 1. TRADITIONAL CLASSIFICATION Traditionally, POP referred to the displacement of the bladder, urethra, uterus, vaginal vault, rectum, or bowel from their normal position into or out of the vaginal canal (Fig. 52.4). Cystocde is the protrusion of the bladder into or out of the vaginal canal that results from tearing or weakening of the supporting structure in the anterior vaginal wall. The bladder bulging should be beyond its normal limit, which is usually at the mid-portion of the vagina. Uretbrocde results from the protrusion of the urethra into the vaginal canal and is also associated with anterior vaginal wall defects. The term cystouretbrocde is used when it is difficult to discriminate between prolapse of the urethra or prolapse of the bladder. It is also used for a cystocele that includes the urethra as part of the prolapsing complex. Rectocele is the protrusion of the rectum into or out of the vaginal canal and is associated with posterior vaginal wall defects. Enterocde involves the protrusion of bowel into or out of the vagina, in which the peritoneum is in contact with vaginal mucosa and the normal intervening endopelvic fascia is absent. Upper posterior vaginal wall prolapse is nearly always associated with herniation of the pouch of Douglas containing loops of bowel, thus forming an enterocele. Defects in the upper anterior wall and apex of the vagina also predispose to enterocele. Uterine prolaDse is the protrusion of the cervix/uterus into or out of the vaginal canal and is associated with apical defects of the vagina. Because the uterus cannot descend without carrying some parts of the upper, anterior, or posterior vaginal walls with it, concomitant defects of these compartments can occur. Procidentia (or total uterine prolapse) represents failure of all the pelvic support, and the uterus protrudes completely out of the vaginal canal. Hypertrophy, elongation, congestion, and edema of the cervix may sometimes cause a large protrusion of tissue beyond the introitus, which may be mistaken for a procidentia. Vaginal vault prolapse occurs in post-hysterectomy patients and results from the protrusion of the vaginal apex into the vaginal canal or totally out of introitus. 2. COMPARTMENTALCLASSIFICATION
The traditional descriptive terms for POP are convenient but can be inaccurate and misleading because they imply an unrealistic certainty as to the prolapsing structures underneath the bulging vaginal wall. For example, it is often difficult to discriminate between a high rectocele and an enterocele or to distinguish between an anterior enterocele and a high cystocele on physical examinations. The traditional terms focus on the bladder, urethra, uterus, or rectum rather than on the
744
H o AND BHATIA
FIGURE 52.4 Diagrammatic representation of the various types of pelvic organ prolapse. A. Cystocele, B. Rectocele, C. Enterocele, D. Massive eversion of the vagina with cystocele, enterocele, and rectocele. In the majority of patients with massive eversion of a post-hysterectomy vagina, an enterocele is present. (Reproduced from Bhatia NN. Genitourinary dysfunction: pelvic organ prolapse, urinary incontinence, and infection. In: Hacker NF, Moore JG, Gambone, JC, eds. Essentials of obstetrics and gynecology, ed. 4. Philadelphia: Elsevier Saunders, 2004:310, and from Nichols DH. Massive eversion of the vagina. In: Nichols DH, Clarke-Pearson DL, eds. Gynecologic, obstetric, and related surgery, ed. 2. St. Louis: Mosby, 2000:476.)
site-specific defects responsible for the prolapse. Descriptions of POP based on compartmental defects are, therefore, being advocated in order to avoid erroneous assumptions regarding the prolapsing organs, unless the involved organs are identified by ancillary tests. In the compartmental classification, anterior vaginal wall (or anterior compartment) prolapse is used instead of cystocele, urethrocele, or anterior enterocele (Table 52.1). Likewise, posterior vaginal wall (or posterior compartment)prolapse is preferable to rectocele or enterocele, and apicalprolapse is used instead of vaginal vault prolapse or apical enterocele. According to the International Continence Society (ICS) standardization of terminology (6), anterior vaginal wall prolapse is defined as any descent of the anterior vagina so that the urethrovesical junction, or any midline point on the anterior vaginal wall proximal to this junction, is less than 3 cm above the plane of the hymen. Posterior vaginal wall prolapse is defined as any descent of the posterior vagina so that a reference point, which is a point in the midline of posterior vaginal wall and at 3 cm proximal to the level of the hymen, or any point proximal to this, is less than 3 cm above the plane of the hymen. Prolapse of the apical
segment of the vagina is defined as any descent of the cervix, or vaginal cuff scar (for post-hysterectomy), below a reference point. This reference point locates proximally from the hymen plane by a distance of 2 cm less than the total vaginal length. These definitions and the use of the TABLE 52.1
Different Types of Pelvic Organ Prolapse
Anterior vaginal wall prolapse 9 Cystocele 9 Urethrocele 9 Cystourethrocele 9 Anterior enterocele
Apical vaginal prolapse 9 Uterovaginal prolapse/uterine prolapse 9 Vaginalvault prolapse 9 Apical enterocele
Posterior vaginal wall prolapse 9 Rectocele 9 Enterocele
CHAPTER52 Pelvic Organ Prolapse in Postmenopausal Women POPQ system provide a more accurate way to describe the prolapse problems.
B. Q.uantification of Pelvic Organ Prolapse Grading or staging of the POP is important for documentation, communication, correlation of symptoms with severity of the prolapse, following of the patients longitudinally, and assessment of treatment outcomes. Although multiple grading schemes have been proposed, two widely used methods are the halfway system (7) and the modification system (8). Recently, the POPQ system was developed for staging the problem more accurately and reproducibly (8,9). 1. THE HALFWAY SYSTEM
The halfway system was developed by Baden and Walker in 1968 and revised in 1992 (7). This grading system has been widely employed in clinical practice to describe the severity of POP. The hymen is chosen as a reference, and the most dependent position of the pelvic organs during maximal straining is used for grading. Grade 0 represents the normal position, and grade 4 is the maximum possible descent of the prolapse (Table 52.2). The prolapse halfway to the hymen is considered to be grade 1; at the hymen, grade 2; and halfway past the hymen, grade 3. If there is any doubt, the patient should be regraded in a standing position, and when the decision has to be made between two grades, the greater grade should be used. If multiple sites of defects occur together, the worst grade of the worst site is employed for the quantification. This system is simple and useful in the grading of POP for clinical practice; however, it does not provide an exact measurement of the prolapse but rather an estimate of descent of the involved structures proximal or distal to the hymen.
745 considered to be first degree; between the introitus and complete eversion, second degree; and a complete eversion, third degree. In this system, the prolapse that is visible at the introitus by depressing the perineum should be considered first degree. The major drawback of this system is that the patient is not allowed to strain during the examination. All other grading systems require maximal straining or the standing position to reproduce the prolapse. The use of the introitus as a reference point is also not as reproducible as the hymen ring. In order to avoid these problems but retain the simplicity of the grading system, several modifications have been made over the years. One proposed system is a modification with grades 0 to 4 (Table 52.3). Another proposed system, which was modified from the Beecham and Baden-Walker methods, is simple to use and has reasonable interobserver variability (8). This system classifies cystocele, uterine or vaginal vault prolapse, and rectocele into first degree, second degree, and third degree, as explained in Table 52.3. The most dependent position of the pelvic organs during maximal straining or standing is used, and the hymen is employed as a reference point. First degree is when the prolapse descends halfway to the hymen, second degree is when it extends to the hymen, and third degree is when it protrudes beyond the hymen. Classification of rectocele is aided by performing a rectovaginal examination. For enterocele, the presence and depth of the enterocele sac, relative to the hymen, is described. The terms grade and degree are sometimes used interchangeably in this system. TABLE 52.3 Clinical Grading of Pelvic Organ Prolapse by the Modification System A. Modification with grades 0 to 4 Grade
No descent Descent between normal position and ischial spines Descent between ischial spines and hymen Descent within hymen Descent through hymen
2. THE MODIFICATION SYSTEM
The Beecham system for grading POP was introduced in 1980 (10). The reference for this system is the introitus, and the prolapse is expressed as first degree, second degree, and third degree. The prolapse above or to the introitus is TABLE 52.2 Clinical Grading of Pelvic Organ Prolapse by the Halfway System Grade
Description
No descent (normal position for each respective site) Descent halfway to the hymen Descent to the hymen Descent halfway past the hymen Maximum of possible descent for each site
Description
B. Modification with degrees 1 to 3 Cystocele, rectocele, uterine prolapse, or vaginal vault prolapse First degree: Descent halfway to the hymen Second degree: Descent to the hymen Third degree: Descent beyond the hymen Enterocele
The presence and depth of the enterocele sac, relative to the hymen, should be described anatomically,with the patient in the supine and standing positions during Valsalva maneuver
746
Ho AND BHATIA
3. THE PELVIC ORGAN PROLAPSE
QUANTIFICATION (POPQ) SYSTEM The POPQ system (8,9), which has been adopted by the ICS, the American Urogynecologic Society (AUGS), and the Society of Gynecologic Surgeons (SGS), provides a precise description of the prolapse; however, it is complicated for routine use. There are six points within the vaginal canal (two on the anterior wall, two on the apex, and two on the posterior wall) to be located and measured with reference to the plane of the hymen (Fig. 52.5). On the midline of the anterior vaginal wall, point Aa is 3 cm proximal to the external urethral meatus, and point Ba is the most distal (or most dependent) position of any part of the anterior vaginal wall segment between the anterior vaginal fornix (or vaginal cuff) and point Aa. On the midline of the posterior vaginal wall, point Ap is 3 cm proximal to the hymen, and point Bp represents the most distal (or most dependent) position of any part of the posterior vaginal wall segment between the posterior vaginal fornix (or vaginal cuff) and point Ap. By definition, the ranges of points Aa and Ap relative to the hymen are from -3 cm for normal support to +3 cm for complete eversion of the vagina. On the vaginal apex, point C is the most distal (or most dependent) edge of the cervix or the leading edge of the vaginal cuff (in patients who had a hysterectomy), and point D represents the location of the posterior vaginal fornix. Point D is omitted in the absence of the cervix. Genital hiatus (gh) is measured from the middle of the external urethral meatus to the posterior midline hymen. Perineal body (pb) is measured from the posterior margin of the vagina to the middle of the anal opening. The total vaginal length (tvl) is the greatest depth of the vagina when point C or D is reduced to its full normal position. All measurements of points Aa, Ba, C, D, Ap, and Bp are in centimeters and expressed as negative if above (proximal) and positive if below (distal) the hymen ring. The lengths of gh, pb, and tvl are also measured in centimeters. A diagram and grid are used to describe the result showing normal support or defects (see Fig. 52.4). These measurements can also be converted to a staging system from 0 to IV, as shown in Table 52.4. An easy way to remember this staging system is that stage 0 represents no descensus whereas stage IV is completely everted. The prolapse with a leading edge at or between 1 cm above and 1 cm below the hymen ring represents stage II. Grades I and III are in between. The good reproducibility of the POPQ system is reflected by a highly significant correlation between interobserver and intraobserver examinations (11). Although a valuable tool for quantification of POP in both research and patient care, the POPQ..system tends to be complicated and confusing for general use in clinical practice.
FIGURE 52.5 Diagrammatic representation of the Pelvic Organ Prolapse Quantification (POPQ) System. A. Six sites (Aa, Ba, C, D, Bp, and Ap), genital hiatus (gh), perineal body (pb), and total vaginal length (tvl) are used for pelvic organ prolapse quantification. B. Line and grid diagram of complete eversion of vagina. Most distal point of anterior wall (Ba), vaginal cuff scar (C), and most distal point of the posterior wall (Bp) are all at the same position (+ 8). C. Line and grid diagram of normal support. Points Aa and Ba and points Ap and Bp are all -3 because there is no anterior or posterior wall descent. D. Line and grid diagram of predominant anterior support defect. Leading point of prolapse is upper anterior vaginal wall, point Ba (+6). E. Line and grid diagram of predominant posterior support defect. Leading point of prolapse is upper posterior vaginal wall, point Bp (+ 5). (Reproduced from ref. 9.)
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal Women TABLE 52.4 Clinical Staging of Pelvic Organ Prolapse by the Pelvic Organ Prolapse Q.uantification (POPQ) System Stage
III IV
Description
No prolapse is demonstrated. Points Aa, Ap, Ba, and Bp are all at -3 cm and either point C or D is at no more than -(X-2) cm (a) The most distal portion (leading edge) of the prolapse is less than -1 cm (b) and the criteria for stage 0 are not met The most distal portion of the prolapse is at least -1 cm but no more than + 1 cm (c) The most distal portion of the prolapse is greater than + 1 cm but less than +(X-2) cm (a.d) The most distal portion of the prolapse is at least + (X-2) cm (a,~)
From ref. 9. Note: (a) X is the total vaginal length in centimeter; (b) The leading edge of the prolapse locates more than 1 cm above the level of the hymen; (c) The leading edge of the prolapse locates between the points of less than 1 cm proximal to 1 cm distal to the level of the hymen; (d) The leading edge of the prolapse locates at more than 1 cm below the plane of the hymen but protrudes no further than 2 cm less than the total vaginal length; (e) The leading edge of the prolapse protrudes to at least 2 cm less than the total vaginal length.
IV. EPIDEMIOLOGY The epidemiology of these conditions remains poorly understood secondary to the lack of definitions to distinguish between normal pelvic support and P O E Furthermore, the prevalence of P O P in women is difficult to determine because most studies are not comprehensive, and results of these epidemiologic investigations vary greatly. In a study of Swedish women between the ages of 20 and 59, the reported prevalence of prolapse that reached the vaginal introitus was 2% (12). Recently, another study using validated questionnaires demonstrated that 8.3% of urban Swedish women have symptomatic P O P and the prevalence increases with age but levels off after the age of 60 (13). Using the 1997 hospital discharge survey and the national census in the United States, the prevalence of surgical procedures to correct P O P in women was found to be 0.23% (14). The problem with this report is that women with P O P who decided not to have surgery were not included. Using the P O P Q s y s t e m to define the stages of P O P in perimenopausal women aged 45 to 55 years old, Bland and coworkers (15) found that stage 0 presents in 73% of these women, stage I in 23%, stage II in 4%, and no subjects have stage llI or IV. Another study on gynecologic patients between the ages of 18 and 86, who presented for routine annual examinations, reported that 6.4% of this population have stage 0, whereas 43.3% have stage I, 47.7% have stage II, 2.6% have stage III, and no subjects have stage IV (16). In a cross-sectional analysis of women who enrolled in the
747 Women's Health Initiative Hormone Replacement Therapy Clinical Trial ( W H I - H R T ) and had baseline examinations for POP, uterine prolapse was found in 14.2%, cystocele in 34.3%, and rectocele in 18.6% of the women with a uterus (17). For women who have undergone hysterectomy, the rate of cystocele was 32.9% and the rate of rectocele was 18.3%. Another group of women in the W H I - H R T study (age range 63-74) who underwent P O P Q examinations demonstrated 2.3% in stage 0, 33% in stage I, 62.9% in stage II, 1.9% in stage III, and no subjects with stage IV (18). Although the reported prevalence rates vary, it can be estimated that 2% to 3% of women in the United States may have symptomatic POP, defined as P O P Q stages III or higher. If POPQstages II or higher are used, the prevalence may range from 47% to 63%. If the symptomatic P O P is defined as the leading edge of the prolapse presenting at the hymen or below, then 25.2% of postmenopausal women may have this condition (18). For racial differences, Caucasian women appear to have higher risk for P O P than African-American women (14,17,19), and the highest risk has been demonstrated in the Hispanic population (17). However, more investigations are needed before drawing any conclusion regarding the role of racial differences in POR Parity, obesity, and age are strongly associated with increased risk of P O P (13,17,18), although parity has been shown to be a considerably stronger risk factor than age alone (13).
V. PATHOPHYSIOLOGY A. Pelvic Support Defects The problems of pelvic support are often referred to as is more accurate than the term "pelvic relaxation" that has been used previously. Pelvic support defects can occur secondary to damage to the pelvic floor structures including muscles, endopelvic fascia, and their innervations. W h e n the pelvic floor structures are damaged or weakened, pelvic organs can descend and cause further attenuation in the endopelvic fascia. Eventually, these fasciae and their connections can break, allowing the pelvic organs to protrude through and contact the vaginal mucosa, thus causing prolapse in the apex, anterior, or posterior compartments. Most women with P O P have multiple defects; however, these defects can be described separately.
pelvic floor defects, which
1. APICAL DEFECTS
Defects in the apical compartment can produce uterine prolapse or vaginal vault prolapse. These defects usually occur at the insertions of cardinal-uterosacral ligament complexes,
748 pubocervical fascia, and rectovaginal fascia to the pericervical ring. Post-hysterectomy, the defects can also occur at the connection of the pubocervical and rectovaginal fasciae in the vaginal cuff. Although vaginal prolapse can occur without uterine prolapse, the uterus cannot descend without carrying some parts of the anterior or posterior vaginal walls with it. Due to theses associations, uterine prolapse is probably the most troubling type of POE These patients may have some degree of high cystocele, high rectocele, and enterocele together with the apical prolapse. In the patient whose uterus remains intact, apical enterocele usually occurs secondary to the detachment of rectovaginal fascia from the pericervical ring. The peritoneum is in contact with the vaginal mucosa, which can be stretched with time, forming an enterocele. The enterocele bulges out from the top of the vagina, when a woman is in the upright position, and the entrapped loops of small intestine can cause pain and pressure. These enterocele sacs are located posterior to the cervix and anterior to the rectum. For patients who have had a hysterectomy, an enterocele develops when the pubocervical and rectovaginal fasciae are separated, thus allowing the protrusion of the peritoneum and bowel. The enteroceles in these patients can occur anterior to, posterior to, or at the vaginal apex. 2. ANTERIOR VAGINALWALL DEFECTS
The intact pubocervical fascia and its attachments, which support the bladder and urethra, prevent anterior vaginal wall prolapse, such as cystocele or urethrocele, from occurring. Anterior compartment defects can be a result of overstretching and attenuation, which are often referred to as distention of the anterior vaginal wall. The underlying pubocervical fascia are usually very thin throughout or are lost in the midline, resulting in absence of rugal folds of the vaginal epithelium.
Ho AND BHATIA
The other type is the site-specific defects, which can be paravaginal, transverse, or midline (Fig. 52.6). Site-specific defects often coexist with some degree of distention in the anterior vaginal wall. The underlying breaks in the pubocervical fascia occur more commonly at peripheral attachments than in central locations. Paravaginal defects occur at the attachments of the pubocervical fascia to the arcus tendineus fascia pelvis and can be a partial disruption or a complete detachment from the fascial white lines. They can also occur unilaterally or bilaterally. Transverse defects occur when the pubocervical fascia is separated from its insertion into the pericervical ring and/or its connection to the cardinal-uterosacral ligament complexes. The midline (or central) defects represent anteroposterior break in the pubocervical fascia and usually occur around the bladder neck and proximal urethra. The vaginal epithelium is often smooth without rugae because there is no underlying pubocervical fascia. Defects in the anterior vaginal wall are often associated with stress urinary incontinence. An anterior enterocele due to the prolapse of the upper anterior wall can occur. This is usually seen in patients who have had a hysterectomy and retropubic colposuspension or sacrospinous fixation without attention to the reconstruction of the pericervical ring.
3. POSTERIOR VAGINALWALL DEFECTS
The rectovaginal fascia connects to the pericervical ring superiorly and extends downward from the cardinaluterosacral ligament complexes to the perineal body inferiorly and the fascial white line laterally. Rectocele is the bulging in the posterior vaginal wall and is caused by the breaks in the rectovaginal fascia and/or its connections,
FIGURE 52.6 Paravaginal,central, and transversedefects in the anteriorvaginalwall.A. Four areas of defects in pubocervicalfascia.B. Paravaginal repair. (Reproducedfrom ContemporaryOb/Gyn, 1990;35:100-109.)
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal Women with the anterior wall of the rectum in direct apposition to the vaginal epithelium. It is basically a defect of the rectovaginal fascia, not the rectum. Similar to the anterior compartment, the vaginal epithelium is smooth without rugae if there is no underlying rectovaginal fascia. The rectovaginal fascia is normally fused with the perineal body at the lower end, and the vast majority of rectoceles occur secondary to the tear along this attachment. In addition to the midline defects in the posterior vaginal wall, detachments of the rectovaginal fascia along the arcus tendineus fasciae rectovaginalis and breaks at the pericervical ring insertion are responsible for the lateral and transverse defects, respectively. All of these defects can form rectoceles. However, the detachment of the rectovaginal fascia from its posterior insertion into the pericervical ring also allows the peritoneum to herniate into the upper third of the vagina, resulting in the enterocele, which can mimic high rectocele and should be differentiated with a rectovaginal examination.
749 levator ani muscles, and perineal body. The pelvic fascia, ligaments, and muscles can become extensively attenuated from excessive stretching during pregnancy and labor. Labor and childbirth also cause injuries to the pudendal and perineal nerves, which lead to the atrophy or decreased tone of the pelvic floor muscles (24,25). When compared with nulliparous subjects, women with more than four vaginal deliveries have an l 1-fold increase in the risk of POP as shown by one study (12). Women who delivered larger babies also have an increased risk for POP (18,26). Although suggestions have been made to perform mediolateral episiotomies to decrease the risk, the data are conflicting and incomplete. The benefit of cesarean delivery is also unclear because several studies showed that pudendal nerve damage caused by vaginal birth can be avoided with cesarean section, whereas other investigations failed to demonstrate a reduction of POP incidence in women with this mode of delivery (27-30). 3. AGING
B. Etiology Although the cause of POP is not completely understood, it is probably multifactorial. In general, factors contributing to the development of POP can be referred to as congenital or acquired. The congenital factors include musculoskeletal, neurologic, and connective tissue defects. The acquired factors involve vaginal delivery, instrumental delivery, previous surgery, neurologic injury, estrogen deficiency, menopause, and aging. Other acquired factors such as obesity, smoking, lung disease, chronic constipation, chronic illness, diabetic neuropathy, occupational or recreational stressors, debilitation, and medications are also important. 1. CONGENITALDEFECTS
Both quantitative and qualitative defects in collagen may play a role in promoting POP (20,21). Collagen defects are suggested because patients with POP were observed to have greater degrees ofjoint hypermobility (22). Connective tissue disorders, such as Marfan's disease, have also been linked to urogenital prolapse. However, in women with Ehlers-Danlos syndrome, no relationship between joint mobility and POP was demonstrated (23). Other congenital defects that can affect the peMc muscles and nerves, such as muscular dystrophy, myelodysplasia, spina bifida, meningomyelocele, may result in POP, although they are not well documented. 2. CHILDBIRTH
Childbirth, especially with difficult vaginal delivery, forceps delivery, and vacuum extraction, can cause direct damage to the pelvic supports through tearing of the endopelvic fascia,
Atrophy of the supporting tissues with aging, especially after menopause, also plays an important role in the initiation or worsening of pelvic floor defects. Several studies demonstrated that the prevalence of POP in women increases with age and the incidence of POP roughly doubles with every decade of life (12,26,31). Swift and coworkers (26) also demonstrated that the incidence of severe POP increases 12% with each year of advancing age. 4. MENOPAUSE
The tissues of the pelvic floor and urogenital tract are known to be estrogen and androgen sensitive. Genital atrophy and hypoestrogenism after menopause are thought to play important roles in the pathogenesis of prolapse. However, the data are conflicting, as one investigation identified menopausal status as a risk factor for POP (31) but another study failed to demonstrate this risk (26). It is possible that the increase in POP prevalence is secondary to aging rather than menopausal status. The role of hormonal effects in POP is also unclear, and more studies are needed before we can draw any firm conclusion. 5. PREVIOUS SURGERIES
It has been shown that history of previous surgery for POP correction is the greatest risk for the subsequent development of severe prolapse (26). The recurrence rates of POP after surgical interventions range from 10% to 30% (32,33). This may reflect the inadequacies of current surgical treatments or preoperative evaluations of POP, or both. Some surgical procedures directed at correction of one compartment may predispose the other compartment to prolapse. For example, anterior repairs may compromise the posterior
750
Ho AND BHATIA
compartment and predispose to enterocele, whereas sacrospinous fixations may increase the incidence of cystocele (34). Other iatrogenic factors, including failure to correct all of the defects adequately, also contribute to the recurrence of the prolapse. Whether hysterectomy contributes to the subsequent development of POP is still unclear. Although one casecontrol study identified hysterectomy as a significant risk factor for developing of postoperative POP (26), other studies failed to demonstrate this risk (32,35,36). The disruption of the cardinal-uterosacral ligament complexes is believed to be responsible for post-hysterectomy vaginal vault prolapse, and reattaching these ligaments to the vaginal cuff as well as obliterating the cul-de-sac during the surgery have been shown to reduce the incidence of subsequent prolapse, particularly enterocele (35). Because supracervical hysterectomy preserves the cardinaluterosacral ligament complexes, the incidence of subsequent prolapse should be lower than total hysterectomy. However, the literature does not support this belief, and the roles of supracervical hysterectomy and uterosacralcardinal ligament preservation in preventing of postoperative POP are uncertain (37). Although the route of hysterectomy, whether abdominal or vaginal, appears to be unrelated to the incidence of subsequent POP, the initial indication for the hysterectomy correlates with the development of postoperative vault prolapse. The vaginal vault prolapse occurs more frequently in patients whose indication for hysterectomy is uterine prolapse (32). 6. OTHER ACQUIRED FACTORS
Although the incidence of urinary incontinence does not appear to increase in athletes (38), repeated lifting of heavy weights may contribute to the development of POE One study on nursing assistants who were exposed to heavy lifting at work demonstrated a 60% increase in the incidence of genital prolapse over the general population (39). Increased intra-abdominal pressure resulting from a chronic cough, chronic obstructive pulmonary disease, chronic constipation, smoking, or obesity can transmit to the pelvic organs and predispose to POE Pelvic tumors, sacral nerve disorders, and diabetic neuropathy may also cause POE
C. Regression of the Prolapse The natural history of POP after menopause was investigated in the participants of the W H I - H R T study, and the results showed that the prolapses are not always progressive in postmenopausal women but can be regressive as well (40). The progressions for grade 1 of cystocele, rectocele, and uterine prolapse occurred in 9.5%, 13.5%, and 1.9% of these women, respectively. The annual regression rates of 23.5% for
cystocele, 22% for rectocele, and 48% for uterine prolapse were also reported.
VI. EVALUATION AND DIAGNOSIS A. Clinical Presentation Mild degrees of POP are usually asymptomatic, although their signs are frequently observed (12). In general, patients with POP may complain of symptoms directly caused by vaginal protrusions or symptoms related to the prolapse. Direct symptoms include vaginal mass or bulge, pelvic pressure, low abdominal or back pain, and coital difficulty. Associated symptoms can be urinary frequency, urgency or incontinence, voiding difficulty, constipation, fecal incontinence, or inability to evacuate the rectum completely without straining or splinting. In many patients, the symptoms of POP are nonspecific. Often there is a feeling of heaviness or fullness in the pelvis or a feeling of something protruding from the vagina. Patients may describe something falling out or a bearing-down discomfort. The characteristic of nearly all symptoms is that they are less noticeable in the morning, worsen as the day progresses or after prolonged standing or walking, and relieved by lying down. Although symptoms associated with a specific type of POP are not well characterized, defects in the anterior vaginal wall often lead to bladder neck and urethral hypermobility, which may cause stress urinary incontinence. A large anterior vaginal prolapse can also compress the urethra, causing obstructional, incomplete, or intermittent bladder emptying. Some patients may require manual displacement of the prolapse in order to void. These patients may have recurrent urinary tract infections. Other patients may describe that their urinary incontinence has been resolved since the time their prolapse worsened. The resolution of urinary incontinence is secondary to the urethral kinking caused by advanced prolapse. Defects in the apical compartment of the vagina can also cause stress urinary incontinence because uterine prolapse and vaginal vault eversion often have a concomitant descent of the anterior vaginal wall. Obstructive voiding and urinary retention as well as masking of the urinary incontinence may occur in the uterine or vaginal vault prolapses. The protruding cervix or vagina can cause vaginal spotting or purulent discharge from their ulcerations. Neglected cases of procidentia may be complicated by bleeding and, rarely, carcinoma of the cervix. In the posterior vaginal wall defects, patients with a rectocele may have difficulty in emptying the rectum and complain of the need to splint the posterior vagina in order to evacuate the trapping stool.
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal Women
B. Examination 1. GENERALAND VAGINALEXAMINATIONS
The general physical examination should include height and weight for body mass index calculation because obesity is a promoting factor for POP. The examination should also include evaluations of the cardiac, pulmonary, abdominal, extremities, skin, and neurologic systems to detect medical conditions such as musculoskeletal disorders, connective tissue diseases, neurologic disorders, and pulmonary diseases that may affect the pelvic floor support. Vaginal examinations are conducted with the patient in the lithotomy position. External genitalia, estrogen status, and caliber of the introitus and vagina are assessed. Signs of hypoestrogenism, such as loss of rugae in the vaginal mucosa, decreased secretions, thin perineal skin, and easy perineal tearing should be inspected. Vaginal rugae are the irregular surfaces of the epithelium due to variations in its thickness caused by contraction and tension of the smooth muscles and elastin fibers within the collagen matrix of the underlying visceral fascia. In addition to the demonstration of a wellestrogenized epithelium, vaginal rugations also reflect the intact and healthy underlying endopelvic fascia. In postmenopausal women or patients with marked vaginal atrophy, topical estrogen can be used for 4 to 6 weeks to restore the prominent rugae before determining the status of underlying endopelvic fascia. In the bimanual examination, the uterus and adnexa are palpated for masses, mobility, and tenderness. 2. PELVIC ORGAN PROLAPSE EXAMINATIONS
a. Systematic Evaluation A systemic pelvic support examination should be part of the evaluation. The examination is usually carried out first in the dorsal lithotomy position. The prolapse is inspected, and if no bulging is apparent, the labia are spread to expose the vestibule and the hymen. After evaluation in the resting condition, the patient is asked to strain vigorously because observing maximal amount of prolapse is critical in the staging of POP and treatment planning. It is useful to ask the patient to confirm the maximal extent of her prolapse with a handheld mirror during the examination. The patient is then examined in the standing position because the degree of the prolapse is almost invariably worse when the patient is upright. Standing is also a position when symptoms of the prolapse occur in the patient's daily life. A standing pelvic examination can be conducted quickly and efficiently with one of the patient's feet elevated on a wellsupported footstool, and the examining gown is lifted slightly to expose the genital region. For the P O P Q staging, dorsal lithotomy can be used because studies have showed that the results obtained in this position are
751 highly correlated with the results obtained in the standing position (41). The traditional speculum examination of the vagina should be supplemented with a site-specific evaluation using a Sims retractor or the lower blade of a Graves speculum. Systematic examinations of the apical, anterior, and posterior compartments of the vagina as well as the perineal body and the perineum are important, particularly when the nature of a prolapse is unclear. Because multiple sites of defects can occur simultaneously, an examination of this kind will allow these defects to be identified before surgery, thus helping to facilitate concomitant repairs of all defects and, therefore, minimizing the chance of treatment failure. It is also important to palpate the thickness of the underlying endopelvic fascia. b. VaginalApex For the examination of apical defects, a bivalve speculum is used to visualize the cervix or vaginal apex, and with the patient straining maximally, the speculum is withdrawn slowly while observing for any descent of the cervix or vaginal cuff. The anterior and posterior fornices are inspected at rest and during Valsalva maneuvers. If the anterior and posterior fornices of the vagina bulge down significantly when the patient strains, the detachments of the pubocervical and rectovaginal fasciae as well as the cardinal-uterosacral ligament complexes from the pericervical ring are expected. Because of these detachments, the cervix demonstrates wide mobility. After inspections with the speculum, digital examination is performed. With gentle traction on the cervix along the longitudinal axis of the vaginal canal, the thick and firm uterosacral ligaments can normally be palpated as they insert onto the posterolateral aspect of the pericervical ring. These ligaments can also be appreciated during a gentle rectal examination, with cervical traction, as well as examination under anesthesia. When uterosacral ligaments become detached from the pericervical ring in patients with uterine or vaginal vault prolapse, a thin, empty line is felt in the examination. In the vaginal vault prolapse, the pubocervical and the rectovaginal fascia are detached from each other, and the underlying support feels very thin with no intervening visceral fascia. The overlying vaginal epithelium is smooth and without rugae. Vaginal vault prolapse usually involves apical enterocele. c. Anterior Vaginal Wall The anterior vaginal wall is examined with the patient in maximal straining, while the posterior wall is retracted, to allow optimal visualization of the bulging. Anterior compartment defects usually involve bladder descent with or without urethral protrusion. Scarring, tenderness, and rigidity of the urethra from previous vaginal surgeries or pelvic trauma should be noted. A bulging-down of vaginal apex and loss of anterior fornix may occur together with anterior compartment prolapse. The anterolateral sulci from the mid-vagina back toward the ischial spines should be
752 visualized, unless lateral defects are present. Palpation of the bulging tissues for their thickness is also important in the assessment of the prolapse. If the cystocele is observed, midline defects should be distinguished from paravaginal defects. One simple way to do this is to support the lateral walls with a ring forceps. With an opened ring forceps inserted over the speculum blade until the tips are against the bilateral ischial spines, the anterior vaginal wall is observed with maximal straining. If the cystocele resolves, the defect is paravaginal. If the cystocele persists, the defect is midline and exists either in isolation or with a paravaginal defect. The next maneuver is to relieve the lateral support and use a closed ring forceps to support the midline of the anterior wall. If the cystocele resolves while the patient is maximally strained, the defect is most likely midline. If the cystocele persists, a combination of central and paravaginal defects may be present. Usually, a loss of rugae of the anterior wall in a well-estrogenized vagina is suggestive of central defect, and collapsing side walls during bivalve speculum examination with maximal straining is suggestive of paravaginal defect. With maximal straining, a paravaginal defect can also be detected by observing the absence of one or both anterolateral sulci, whereas in the central defect, these sulci are usually well preserved. Because most of the central defects occur around the bladder neck and urethra, placing a small and well-lubricated dilator in the urethra can help to feel the thinness of the involved tissues. In the transverse defect caused by the detachment of pubocervical fascia from the anterior margin of the pericervical ring, a distinct outward bulging of the anterior vaginal fornix is observed. If the transverse defect is caused by the detachment of the uterosacral ligament from the perivercial ring, a significant cervical or vaginal vault descensus can be observed, with no uterosacral ligaments being palpable near the pericervical ring. The bulging in the transverse defect normally has poor rugations due to the loss of underlying pubocervical fascia. It should be noted that an anterior enterocele due to the prolapse of the upper anterior vaginal wall can mimic a cystocele on physical examination. After inspection, the anterior vaginal wall, bladder, and urethra should be palpated for tenderness or masses. d. Posterior Vaginal Wall The posterior vaginal wall is examined for defects while the anterior wall is retracted and the patient is in maximal straining. Sometimes the lateral walls need to be supported in order to provide a better view for these examinations. Rectocele is suspected by observation of posterior vaginal wall bulging and confirmed with an anterior displacement of the rectal wall on rectovaginal examination. On inspection of the lateral vaginal walls in a woman without pelvic prolapse, the anterolateral and posterolateral sulci travel together from the ischial spines to the mid-vagina, where they begin to separate. The anterolateral
Ho AND BHATIA
sulcus travels toward the pubic arch, while the posterolateral sulcus, which represents the line of attachment of the rectovaginal fascia to the arcus tendineus fasciae rectovaginalis, courses down toward the perineal body. In patients with posterior vaginal wall prolapse, the disappearance of these sulci suggests lateral defects and the loss of rugations indicates central defects. The upper part of the posterior vaginal wall should be checked with a rectovaginal examination for an enterocele, especially in patients who have undergone hysterectomy. Rectovaginal examination is useful to demonstrate a rectocele, by palpating the thickness and mobility of the rectovaginal fascia, and to distinguish it from an enterocele. Occult enterocele in the cul-de-sac can also be palpated between the thumb and forefinger during a rectovaginal examination while the patient is in maximal straining. The size and muscle tone of the pubococcygeus and puborectalis muscles are examined by placing one or two fingers within the lower third of the vaginal canal. The medial edge of the pubococcygeus muscle is usually felt at the level of the hymen ring. The patient is then asked to tighten the levator ani muscles around the examining fingers in order to grade the strength and quality of pelvic floor contractions. e. Anorectal Canal and Perineal Body Finally, the anorectal canal and external sphincter tone are assessed with rectal examinations. The strength and position of the levator plate should also be assessed. When contracted, the levator plate can be easily palpated with an impressive strength against the examining finger, and it should be almost horizontal in the standing position. However, in the dorsal lithotomy position, the levator plate presents in a more vertical fashion. If the levator plate is weak and inclined downward, the levator hiatus is widening, thus predisposing to POP (see Fig. 52.2). This also increases the shearing stress on the intact visceral pelvic support structures, especially the cardinal-ureterosacral ligament complexes. The perineal body can be palpated as the examining finger moves through the anal canal toward the rectum. The apex of the perineal body is usually at the level of the midvagina. An intact perineal body is attached to the rectovaginal fascia, which connects to the uterosacral ligament bilaterally, and makes the perineum concave in appearance. Any significant break in these connections will result in a descent and bulging of the perineal body that can be felt by rectal examinations. These examinations can also detect the perineal rectocele if the rectal muscularis is in direct contact with the perineal skin without any underlying fascia. Perineal rectocele occurs secondary to disruptions in the integrity of the perineal body and the involved muscles, such as the superficial transverse perinei, pubocavernosus, and levator ani. Significant bulging of the perineal rectocele can be demonstrated with a Valsalva maneuver, and the covering skin is usually smooth and stretching.
753
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal Women
f. PotentialIncontinence Potential incontinence (or masked incontinence) refers to the stress urinary incontinence that develops only after reduction of the prolapse. POP, particularly the prolapse of the apical and anterior vaginal wails, can kink the urethra and, therefore, mask the incontinence. In the preoperative evaluation of patients with POP for surgical treatment, the potential incontinence must always be ruled out. To test for potential incontinence, the patient should have a full bladder. If necessary, the bladder can be filled to maximum capacity (at least 300 mL) with sterile water or normal saline while reducing the prolapse with an appropriately fitted pessary in the vagina. The patient is then asked to cough or perform a Valsalva maneuver. If urinary leakage is observed, the potential incontinence is confirmed and urodynamic studies should be performed to determine whether the incontinence is secondary to bladder neck hypermobility or intrinsic sphincter deficiency. This determination is important because the intrinsic sphincter deficiency requires a different type of surgical treatment than the bladder neck hypermobility. Urodynamic studies also help to rule out the detrusor overactivity, which is often caused by bladder outlet obstruction secondary to the prolapse. Other ancillary tests, such as the cotton swab testing for the bladder neck mobility and the measurement of postvoid residual volumes, are also very useful (see Chapter 51). g. Other Investigations Other diagnostic tests may be needed in addition to a careful history and physical examination. In patients with complaints of urinary frequency, urgency, or incontinence, a urinalysis should be performed to rule out urinary tract infections. Cervical cultures are indicated in patients who present with ulcerations or purulent discharge. The Pap smear, endometrial biopsy, or biopsy of the ulceration may be necessary in rare cases of suspected carcinoma. Patients who presents with vaginal bleeding or spotting should also have an endometrial biopsy. Although rarely needed, cystourethroscopy can allow the examination inside the urethra, urethrovesical junction, bladder walls, and ureteral orifices to determine bladder stones, malignancies, diverticula, or suture from previous surgeries. Cystourethroscopy is usually reserved for patients with confounding factors such as microscopic or macroscopic hematuria, bladder pain on palpation, suburethral mass, or persistent urinary tract infections despite adequate therapy. Imaging studies including ultrasonography, computed tomography (CT) scan, and magnetic resonance imaging (MR1) are useful in the determination of POE The CT scan and the sector, real-time, or three-dimensional ultrasonographies have been employed to provide anatomic details of the pelvic floor support. MRI is an emerging technique for the study of pelvic floor dysfunctions and holds promise due to its excellent ability to differentiate soft tissues (42). However, its use is currently limited to research rather than
clinical practice due to the cost and the lack of standardized criteria for diagnosis of POP. It is also important to measure quality of life in women with pelvic floor disorders when planning the treatment and evaluating the efficacy of a particular therapy. Two instruments, the Pelvic Floor Distress Inventory (PFDI) and the Pelvic Floor Impact Q.uestionnaire (PFIQ), were developed and validated for these purposes (43,44). These questionnaires allow the inclusion of the patient's perception of POP into the overall assessment. They can be administered in clinical or office settings.
VII. TREATMENT Generally, patients with asymptomatic or mild POP do not require treatment. The treatments for symptomatic POP include conservative management and surgical interventions. The patient's baseline functions and medical conditions as well as the severity of the prolapse, rather than age alone, should be the guidelines in selecting the treatment.
A. Nonsurgical Treatment Although surgery has long been considered the mainstay of treatment for POP, nonsurgical interventions can alleviate or improve symptoms for a substantial number of patients. Unless extenuating circumstances exist, it is recommended that nonsurgical therapies should be attempted first, particularly with patients who are poor candidates for surgery. The cure rate of nonsurgical treatments is lower than surgical approaches; however, their costs, risks, and morbidities are low, whereas the patient satisfaction can be quite high. Nonsurgical interventions include conservative measures, pelvic muscle exercises, and the use of pessaries. 1. CONSERVATIVEMEASURES Prophylactic measures include diagnosis and treatment of chronic respiratory diseases, correction of constipation and other disorders that may cause chronic increase in intraabdominal pressures, and administration of topical estrogen to postmenopausal women. Significant medical problems such as obesity, asthma, and long-term steroid use may contribute to POP. These problems should be corrected before any surgical interventions because recurrences of the prolapse may be more likely if such conditions are not addressed. Topical estrogen is an important adjunct in the conservative management of patients with POP. Estrogen therapy may improve the quality of the pelvic tissues to better support the pelvic organs. In postmenopausal women or elderly patients with marked vaginal atrophy, topical estrogen can be used for 4 to 6 weeks to restore the prominent rugae
754 before examination to determine the underlying endopelvic fascia. The preoperative use of topical estrogen for 4 to 6 weeks can improve the quality of vaginal epithelium and other pelvic tissues at the time of surgery, thus contributing to the better outcomes. Although rare, ulcerations of the prolapsing tissues may be complicated by infections, and antibiotic therapy is indicated. 2. PELVIC MUSCLE EXERCISES
PeMc muscle exercises (Kegel exercises) may improve the strength and tone of the pelvic floor musculature. However, their effectiveness in the treatment of POP is unclear because there is no evidence to indicate that improvement of pelvic floor muscle strength can lead to regression of POE Most of the investigations examining the benefits of pelvic floor exercises have been focused on the treatment of urinary incontinence. Recently, a prospective, randomized and controlled study of 60 patients (30 with therapy and 30 controls) who underwent surgery for POP and urinary incontinence showed that preoperative and postoperative physical therapies significantly improved urinary incontinence, peMc floor muscle strength, and quality of life, although the postoperative followup was only 3 months (45). Nevertheless, Kegel exercises may strengthen levator ani muscles to provide better support for the pelvic organs and to enhance the patient's ability to retain the pessary in the vagina (46,47). It is believed that POP develops when the peMc floor muscles are weakened, thus compromising their ability to provide adequate support to the pelvic organs. The rationale behind Kegel exercises for POP is that they help to recover the peMc muscle functions, which may reverse the mild cases and/or prevent the progression of the prolapse. Because these exercises have practically no side effects and no contraindications, they can be incorporated into routine health maintenance for postmenopausal women to prevent the development of POR The Kegel exercises for POP are similar to the exercises for the management of urinary incontinence as described in Chapter 51. The verbal and written forms of instruction alone are not effective for these exercises in many women. It has been shown that up to 40% of women who received verbal instructions were unable to perform Kegel contractions sufficiently (48). Furthermore, up to 30% of women are unable to perform peMc muscle exercises correctly on their first attempt. Contractions of other muscle groups, such as the abdominal, gluteal, and hip abductor muscles, are the common errors. Identifying the correct muscles is a critical part, and this can be best achieved by using a feedback from a digital palpation. Perineometer provides accurate information on the strength of pelvic muscle contractions and visual feedback to the patients regarding their progress. In some patients, follow-up with repeated evaluations over several months may be necessary to ensure that
Ho AND BHATIA
the exercises are performed correctly. Several studies demonstrated that biofeedback techniques, such as perineometer or vaginal cones could enhance the successful rates of pelvic floor exercises (49,50). However, in a prospective, randomized study comparing supervised exercises with vaginal cone biofeedback and vaginal electrical stimulation, supervised exercises were found to have a highest successful rate (78% with supervised exercises, 30% with vaginal cone, and 13% with electrical stimulation) (51). 3. PESSARY
Pessaries are designed to restore and maintain the pelvic organs in their normal position. Although they are an effective alternative to surgery, physicians and patients are usually reluctant to use pessaries for the management of POP due to the lack of familiarity and the concerns of discomfort with the device.
a. Indications and Contraindications Pessaries can be used for relieving symptoms associated with the prolapse permanently as a final therapy or temporarily when undertaking activities or waiting for surgery. These devices can also be used when the patient is medically unfit for surgery, during pregnancy and the postpartum period, and to promote the healing of ulcered tissues before surgery. Pessaries may be employed preoperatively for several weeks in order for the patient to have a realistic expectation of whether surgical reduction of the prolapse will resolve her symptoms of pelvic pain, backache, urinary problems, and voiding dysfunctions. Contraindications to pessary use are few, including acute pelvic inflammation diseases and pain after insertion. Recurrent vaginitis is a relative contraindication and may require removal of the pessary. Prior hysterectomy and current sexual activity have also been considered as relative contraindications; however, the majority of clinicians do not found these conditions to be contraindicated for the use of pessaries (52). Wu and coworkers (53) reported that previous surgery, including hysterectomy, does not appear to affect the success rates of pessary fitting. b. Type of Pessary The selections of pessary for POP are suggested in Table 52.5. These selections are based on experiences rather than scientific data because formal evaluations of different types of pessaries for different kinds of prolapse have not been conducted. Choice of pessary depends on its efficacy for the management of POP and its acceptability by the patient. Although several types of pessaries are available, the less expensive and user-friendly device should be tried first. Ring, cube, and Gellhorn are the most common types of pessary for POP (Fig. 52.7). The ring pessary with or without a supportive membrane is the most versatile type because it can be folded in half for easy insertion, is less likely
CHAPTER 52 PeMc Organ Prolapse in Postmenopausal Women
755
TABLE 52.5 Suggested Pessary Selection for Pelvic Organ Prolapse
For anterior vaginal wall prolapse such as cystocele 9 Ring with/without support 9 Hodge 9 Gehrung 9 Donut
For apical prolapse such as vaginal vault prolapse or uterine prolapse A. First and second degrees 9 Ring with/without support 9 Shaatz B. Third degree and procidentia 9 Gellhorn 9 Donut 9 Cube 9 Inflatoball
For posterior vaginal wall prolapse such as rectocele 9 Donut 9 Ring 9 Risser
9 Gehrung 9 Gellhorn Note: There are generally two approaches to choosing a pessary for the management of symptomatic pelvic prolapse. Most commonly, selection of a pessary is based on the specific type of defects as described above. The second approach involves using the same pessary for all types of prolapse. Neither approach has been proven more effective than the other.
to sequester vaginal secretions, and works for most patients with different types and degrees of prolapse (52,53). W h e n the ring type failed, a second line of pessaries, such as Gellhorn, cube, and donut, can be used (53). Gellhorn pessary is the second-most common choice and has been successfully employed in grade 3 or 4 prolapse of the anterior vaginal wall, the vaginal cuff, or the uterus (52,54). For patients with concomitant prolapse and stress urinary incontinence, the pessary with ring support and knob can be a successful choice. For patients with significant uterine prolapse and a large introital diameter who have not obtained relief with other pessaries, the Gehrung type should be tried. For patients with a well-defined pubic notch and adequate vaginal width, the Smith-Hodge can be attempted. It should be noted that some patients may not tolerate pessary use, and the more severe forms of P O P are usually not relieved by the placement of these devices. c. Pessary Insertion Pessary fitting is far from an exact science, and a trial-and-error approach is the rule. The depth and width of the vagina determine the size of the first pessary and proper match, which may require several attempts, will avoid compression of the vaginal mucosa resulting in abrasion, bleeding, and erosion. The properly fitted pessary will
FIGURE 52.7 Different types of vaginal pessaries. A. Ring, B. Shaatz, C. Gellhorn, D. Gellhorn, E. Ring with support, F. Gellhorn, G. Risser, H. Smith, I. Tandem cube, J. Cube, K. Hodge with knob, L. Hodge, M. Gehrung, N. Incontinence dish with support, O. Donut, P. Incontinence ring, Q:. Incontinence dish, R. Hodge with support, and S. Inflatoball (latex). (Photo courtesy of Milex Products, Inc. and from Trowbridge ER, Fenner DE. Conservative management of pelvic organ prolapse. Clin Obstet Gyneco12005;48:668-681.)
also make the patient feels more comfortable. Ring pessary sizes 3, 4, or 5 are usually employed and can be successfully fitted in 70% of patients (53). Pessary insertion is a simple technique; however, it is usually overlooked. Women with an atrophic vagina or any sign of hypoestrogenism are pretreated locally with estrogen cream for 3 to 6 weeks. Lubricant is placed on the vaginal introitus, and the ring pessary, which has already been folded in half, is inserted with pressure maintained on the posterior vaginal wall while the labia minora are held separated. Once the pessary is inside the vagina, the index finger of the predominant hand is used to direct its leading edge into the posterior fornix and to ensure that the device lies behind the symphysis pubis with its diaphragm supporting the cervix. The pessary should not be too tight in the vagina, and it should be easy to slide up and down along the vaginal sidewall. However, a properly fitted and well-supported pessary should not be visible when the labia are separated and should not fall out or descend to the introitus with a Valsalva maneuver or cough. The patient should then be re-examined in the standing position to confirm the pessary support. If
756 the device falls out with a Valsalva maneuver or cough, a larger size should be tried. Pessary fitting should be carried out with a moderately full bladder to allow testing of its efficacy in improving voiding dysfunctions and to see whether reduction of the prolapse is accompanied by urinary incontinence. Also, the patient needs to void comfortably with the pessary in place before sending home. Patients are also instructed to perform normal activities such as walking, running, jumping, or straining in the clinic to ensure the device is best fitted and that the patient is comfortable with it. With all of these precautions, however, approximately 25% of patients will return for another fitting. For pessary removal, the index fingers of both hands are placed above and below the leading edge of the device to hook it, and tractions are applied along the vaginal axis to bring it down toward the introitus. In order to avoid discomfort, the pressure should be placed on the posterior introitus. Speculum examination is then performed to inspect the vaginal wall for any evidence of abrasions or erosion. The pessary is rinsed with Betadine, washed with water, and then dried before reinsertion. Patients should be taught to remove and insert the pessary themselves. Although a variety of recommendations are found in the literature, we usually schedule a follow-up visit at 2 weeks to examine the vagina for any inflammation, adjust the type and size of pessary if needed, and teach the patient again for insertion and removal of the device. In the first year, followup visits are scheduled every 3 months, and after the first year, every 6 months. This schedule is similar to other recommendations (53). At these visits, the pessary should be removed, inspected, and cleaned. Most of the silicone pessary can be used for up to 2 years, although it should be replaced when it becomes stiff or coated with thick secretion. The vagina is examined by speculum at every visit to ensure absence of abrasion or erosion. In some women, vaginal accommodation may occur, and increasing the pessary size at the time of follow-up visits may become necessary. For those patients who can remove and insert the pessary themselves, weekly overnight removal and cleansing with soap and warm water is recommended. For elderly patients who are unable to care for their own pessaries, visiting nurses can be valuable to visit the patient at home once a week to remove the pessary in the evening, clean it, and return in the morning to replace it.
d. Side Effects Modern pessaries are made from silicone, an inert and durable material, and produce less incidence of troublesome symptoms to patients. Although pessaries may cause vaginal irritation and ulceration, these adverse outcomes are rare if a patient is able to remove and reinsert the pessary on her own at least once a week. Any excessive foul-smelling discharge, abnormal vaginal bleeding, experience of pelvic pain, or leaking of urine indicates a
Ho AND BHATIA
need for medical follow-up. Failure to do so may result in serious consequences, including infections, impaction, and fistula formation by erosion into the bladder or rectum. These complications can also occur with a long-forgotten pessary. If patients complain of vaginal bleeding secondary to abrasion, the pessary should be removed and left out for several weeks to allow the vaginal epithelium to heal. Local estrogen can be given to atrophic epithelium. The cube pessaries may cause abrasions and erosion more often than the ring type, and they should be inspected more regularly. In advanced degrees of POP, urinary leakage may occur after placing the pessary due to unkinking of the urethra. In these cases, the ring pessary with support and knob can be used to manage both the prolapse and urinary incontinence. Cervical and vaginal malignancies have been reported in association with the use of pessaries, with the mean interval between insertion and cancer diagnosis being 18 years (range 1-41 years) (55). It is therefore important to follow patients with speculum examinations and maintain the regular schedule of Pap smears.
B. Surgical Treatment 1. SURGICAL APPROACH
For patients who have failed conservative management or who have severe POP, the treatment is often surgical intervention. The main objectives of surgery are to relieve symptoms; restore normal anatomic relationships; restore normal visceral functions; maximize satisfactory coital, bladder, and bowel functions; and attain a durable result. A careful preoperative evaluation should identify all concomitant defects that should be repaired simultaneously. For surgical decision-making, several important factors should be considered, including medical condition, age, severity of symptoms, degree and type of POP, patient's suitability for surgery, and the presence of other pelvic conditions requiring simultaneous treatment. The patient's choice of treatment and the desired postoperative caliber of the vagina and introitus should also be important parts of the decision making. With careful surgical planning and postoperative care, coupled with lifestyle changes (avoiding constipation and heavy rifting, performing pelvic floor exercises), the majority of patients can enjoy long-term results with satisfaction. Surgical repairs of POP can be carried out by abdominal, vaginal, or laparoscopic approaches. The selection of vaginal versus abdominal routes has been a subject of debate. Vaginal approach for the surgical treatment of POP allows shorter recovery time and is preferred when a concomitant vaginal procedure is planned for the correction of urinary incontinence. Paravaginal repair and sacral colpopexy can also be performed laparoscopically. The major advantages of laparoscopic approach are fast recovery, shorter hospitalization, less morbidity, better cosmetic results, and minimal invasiveness. Although results in short-term subjective re-
CHAPTER52 Pelvic Organ Prolapse in Postmenopausal Women ports are good, long-term and objective data are still lacking. Furthermore, laparoscopic surgery for pelvic reconstruction requires good operative skills and the learning curve is usually steep, with complications being more likely to occur in early experiences. 2. APICALPROLAPSE
The repair of a uterine or vaginal vault prolapse must address reconstruction of the pericervical ring and its attachments to the cardinal-uterosacral ligament complexes, pubocervical fascia, and rectovaginal fascia. For the repair of vaginal vault prolapse in posthysterectomy, it is also important to re-approximate the planes of pubocervical and rectovaginal fasciae together. A hysterectomy is not a mandatory part of the surgical treatment for uterine prolapse unless there is pathology involving the uterus. Performing a hysterectomy does not correct the apical defect, and there is no evidence to indicate that hysterectomy has any effect on the long-term success of apical suspension (56). However, in some cases, removal of a large and bulky uterus may be necessary for a better access to the apical reattachment sites. This decision may be made intraoperatively, and the patient should be appropriately counseled. In postmenopausal women with uterine prolapse, the rationale for concomitant hysterectomy may include the preventing of any uterine or cervical pathology, such as large fibroid, pelvic inflammatory diseases, endometriosis, endometrial hyperplasia, and carcinoma, that may develop. If the uterus is to be preserved, preoperative ultrasonographic study and endometrial biopsy are strongly recommended. a. McCall Culdoplasty The McCall culdoplasty can be used to correct apical defects by approaching the uterosacral ligaments together and reattaching the vaginal vault to these ligaments. This procedure may also be used as a prophylaxis against future development of enterocele after completion of the hysterectomy. The uretrosacral ligaments are plicated with two or three interrupted sutures that are placed in a progressively cephalad fashion (Fig. 52.8). The peritoneum between the uterosacral ligaments is also incorporated into these sutures to close the cul-de-sac. Additional sutures are then placed to suspend the vaginal cuff to both uterosacral ligaments. Because attaching the prolapsed vagina to a stretched uterosacral ligament is of little value, the segment nears the sacrum where the ligaments are still strong and undetached should be used. However, respecting and avoiding the neighboring ureter and other structures are critical, and intraoperative cystoscopy should be performed to be sure that the ureters have not been kinked or ligated. b. UterosacralSuspension This procedure involves anchoring of the vaginal cuff to the uterosacral ligaments without
757 plicating these ligaments in the midline (see Fig. 52.8). Studies have shown that the intermediate segment of the uterosacral ligament, which can be found at the level of the ischial spine, is the optimal location for suture placement because it balances between strength and safety (57). The ureter is in proximity to the anterior border of the ureterosacral ligament, being closest at the level of the cervix and farthest at the sacrum. The sutures, therefore, should be placed in the intermediate segment, at least 1 cm posterior to the anterior border, of the uterosacral ligament to avoid the potential of kinking or ligating the ureter. In this procedure, the first suture is placed through the uterosacral ligament and then incorporates the pubocervical fascia anteriorly and the rectovaginal fascia posteriorly before tying to anchor the vaginal cuff. In a similar fashion, a second suture is placed through the contralateral uterosacral ligament and then attached to the pubocervical and rectovaginal fasciae. On each side, an additional one or two sutures, through the ipsilateral uterosacral ligament in a progressively cephalad fashion and attaching to more medial aspects of the vaginal cuff, can be performed. Uterosacral suspension procedures can also be performed abdominally or laparoscopically. In these procedures, the uterosacral ligaments are not plicated in the midline because this may increase the risks of ureteral obstruction and may lead to narrowing of the upper vagina causing dyspareunia. However, if the uterosacral ligaments are highly attenuated, midline plication may be necessary to provide adequate suspension of the vaginal cuff. Success rates of 87% to 90% and the incidence rates of
FIGURE 52.8 McCall culdoplasty and uterosacral suspension. Relative position of two previously placed internal McCall sutures with external McCall suture is shown.The external suture is placed in a more cephalad position. (From Lee RA. Atlas of gynecologic surgery. Philadelphia: W.B. Saunders, 1992. Reprint from Lee RA. Vaginalhysterectomywith repair of enterocele, cystocele,and rectocele. Clin Obstet Gyneco11993;36:967-975.)
758
Ho
AND BHATIA
ureteral obstruction up to 11% with these procedures have been reported (58,59).
c. Iliococcygeus Suspension The iliococcygeus fascia suspension is useful when the uterosacral ligaments are not visualized or are extremely attenuated and weak. The posterior vaginal epithelium is first opened in the midline, and the rectovaginal spaces are dissected widely toward the ischial spines. The fascia overlying the iliococcygeus muscle lateral to the rectum and anterior to the ischial spine is identified, and a delayed absorbable suture is placed through this fascia to the pubocervical fascia anteriorly and the rectovaginal fascia posteriorly on the ipsilateral side. This suture can also be placed deeply into the iliococcygeus muscle (60). A similar suture is placed on the other side to suspend the vaginal cuff bilaterally. Alternatively, the attachments can be carried out by placing the sutures through the iliococcygeus and periosteum at the level of the ischial spine. Although the vaginal axis is not altered with this procedure, vaginal length may be reduced due to the fact that the ischial spines are considerably inferior to the normal position of the vaginal apex. Failure rates ranging from 4% to 19% have been reported for iliococcygeus suspension (60,61). d. Sacrospinous Ligament Suspension In the sacrospinous ligament suspension (sacrospinous fixation) procedure, the vaginal apex is attached to the sacrospinous ligament usually on the patient's right side, although bilateral fixation has also been described (62). This procedure is widely advocated for apical prolapse, and good results have been obtained. After entering the pararectal space through an incision in the posterior vaginal wall, the ischial spine is located, and the sacrospinous ligament, which runs from the ischial spine to the lower portion of the sacrum and coccyx, is identified. The Miya hook or the Capio device is used to place two nonabsorbable sutures through this ligament; the first is about 1 cm cephalad and about 2 cm medical to the ischial spine, and the second is about 1 cm medial to the first (Fig. 52.9). These sutures are then placed through the pubocervical fascia anteriorly and rectovaginal fascia posteriorly before tying, thus suspending the vaginal cuff to the sacrospinous ligament. The two most serious complications of the sacrospinous fixation are hemorrhage and nerve injury. The hemorrhage can result from injuries to the inferior gluteal vessels, hypogastric venous plexus, or internal pudendal vessels. If the suture is placed too close to the ischial spine, it may entrap the pudendal nerve. Because of its proximity to the sacrospinous ligament, injuries to the sciatic nerve can also occur, and reoperation may be necessary for removal of the sutures. Buttock pain is probably caused by injury of small nerves that run through the sacrospinous ligament. The fixation is not anatomic, and the posterior deflection of the
FIGURE52.9 Sacrospinousligamentsuspension(sacrospinousfixation). Vaginalwall is advancedtoward the sacrospinousligamentwhile tying sutures in order to bring the vaginalapexinto contactwith the ligamentand overlying muscle. (From MisheUDR, Jr, StencheverMA, DroegemueUer W, Herbst AL. Comprehensivegynecology,St. Louis:Mosby,1997:566.) vaginal axis may expose the anterior vaginal wall to increased forces during physical stress, thus predisposing this site to prolapse. It has been reported that the incidence of anterior vaginal wall prolapse increases after sacrospinous fixation (34); however, another retrospective study challenged these results (63).
e. Sacral Colpopexy In patients who have poor pelvic muscle strength, weak endopelvic fascia, attenuated ligaments, or a high level of occupational or recreational stress, sacral colpopexy is a better option for correction of apical prolapse. In this abdominal procedure, either synthetic mesh or biologic material is placed from the vaginal cuff (post-hysterectomy), or the amputated cervical stump (post-cervical-preserved hysterectomy), to the presacral fascia (Fig. 52.10). A variety of synthetic materials, such as Prolene, GoreTex, Mersilene, or Marlex, can be used as the graft. Autologous grafts such as harvested fascia lata from the thigh or rectus fascia in the abdomen, allografts such as cadaveric fascia lata or skin, and xenografts such as porcine small intestine submucosa can also be used. After dissecting the peritoneum and bladder off, the anterior and posterior straps at the lower end of the graft are attached to the vaginal apex. The upper end of the graft is then attached to the presacral ligament in the promontory with an undue tension. The peritoneum is closed over the graft to prevent any bowel entrapment. Sacral colpopexy provides sufficient vaginal length and high cure rates, which range from 78% to 100% (64). The most serious complication of this procedure is the bleeding from injuries to the presacral venous plexus or the
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal Women
759 vagina, which usually requires partial or complete removal of the material (65). Sacral colpopexy can also be performed laparoscopically in a similar manner as described for the abdominal procedure. Recently, one study demonstrated that laparoscopic approach is associated with shorter hospital stay, as well as clinical outcomes similar to the abdominal route (66). These results are usually from experienced laparoscopists.
3. ANTERIORCOMPARTMENTPROLAPSE
FIGURE 52.10 Sacralcolpopexy procedure. Grafts are attached to the posterior and anterior vaginalvault and then anchored to the sacral periosteum. In this figure the anterior and posterior grafts are anchored separately to avoid over-elevating the anterior wall, which may predispose to urinary incontinence. (From Van RooyenJB, Cundiff GW. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, eds. Ostergard's urogynecology and pelvic floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:421.)
middle sacral artery while operating in the presacral space. Although this intraoperative complication is uncommon, it can be life threatening because the bleeding from presacral vessels is difficult to control. Ureteral injury can also occur intraoperatively. The most common long-term complication is the erosion of synthetic mesh through the
a. Anterior Colporrhaphy Anterior colporrhaphy, which has been traditionally employed to correct cystocele and urethrocele, is a vaginal approach. The vaginal epithelium is first dissected off the underlying endopelvic fascia for the entire extent of the prolapse and to expose the area around the urethrovesical junction. The plicating sutures are then placed in the endopelvic fascia, lateral to the bladder neck, to augment the support of this structure in order to prevent stress urinary incontinence that may develop postoperatively. Finally, the vaginal muscularis including the pubocervical fascia are plicated in the midline without tension to close the defect and provide support to the bladder and urethra (Fig. 52.11). Modifications of anterior colporrhaphic techniques include plications ofparavaginal tissues to provide additional layers of support. These procedures provide better supports across the anterior vaginal wall; however, they may resttlt in narrowing the vagina, which must be considered when planning the surgery.
FIGURE52.11 Anteriorcolporrhaphy.A. Plication of bladder neck and repairof cystocele. B. Repairof vaginaovercystocele.(FromMisheUDR, Jr, StencheverMA, DroegemueUerW, Herbst AL. Comprehensivegynecology. St. Louis: Mosby,1997:586.)
760 As with other compartmental prolapses, the early failure of the anterior colporrhaphy is probably caused by inadequate repair of all of the defects, whereas late recurrences are associated with the subsequent weakening of the involved tissues secondary to aging, estrogen deficiency, and other factors. In addition to recurrence of the prolapse, sexual problems and dyspareunia are other concerns in anterior colporrhaphy, although dyspareunia is more common in posterior repair (67).
b. Site-Specific Repair Defects in the anterior compartment can occur laterally, medially, or transversely. In recent years, site-specific defect repairs have became more popular and may produce better long-term results (68). These procedures involve the precise identification of the defects preoperatively as well as intraoperatively. The repair of lateral (or paravaginal) defects is accomplished by reattaching the pubocervical fascia to the arcus tendineus fasciae pelvis using a vaginal or retropubic approach. If the white line is detached from the pelvic sidewall, is too attenuated, or is unrecognizable, the fascia overlying the obturator internus can be used for reattachment. Central (or midline) defects in the pubocervical fascia are easily approached from the vagina and can be repaired by approximating the edges of the defects together or by the traditional anterior colporrhaphy. There must be tissue-to-tissue approximation between the edges the defect. Transverse defects can be corrected by approximating the anterior pubocervical and posterior rectovaginal fasciae together. The long-term effectiveness of sitespecific repair remains unknown, and there is a need for randomized trials that compare outcomes of this approach to anterior colporrhaphy. c. Compensatory Repair with Graft Materials In some patients whose pubocervical fascia is too attenuated and weak for the anterior colporrhaphy or site-specific repairs to be successful, graft materials can be considered. These graft materials usually include biologic fascia or synthetic mesh. Biologic fascia can be autologous tissues (fascia lata or rectus fascia), aUografts (cadaveric fascia or dermal graft), or xenografts (porcine dermis or porcine intestine). Autologous grafts can eliminate the risks of tissue reactions and erosions; however, tissue harvesting increases morbidity, and the tissue size is often too small for anterior or posterior vaginal wall repairs. Furthermore, tissue failure can occur over time. AUografts and xenografts are associated with durability and rejection problems. Synthetic mesh (Prolene, Cortex, Mersilene) offer better uniformity, consistency, and durability than biomaterials, but it carries the risks of infection and erosion, which require subsequent removal. The selection of grafts remains controversial, and currently there is no ideal material (69).
Ho AND BHATIA The graft is placed over the anterior fascia and secured superiorly to the vaginal apex and laterally to the arcus tendineus fascia pelvis with interrupted sutures. A cure rate of 100% was reported with Marlex mesh, as compared with a 66% cure rate for traditional anterior colporrhaphy; however, there was a 25% incidence of mesh-related complications (70). In another study, a 75% success rate was obtained with synthetic mesh and a 57% cure rate with anterior repair alone (71). Recent studies indicate that whether graft use confers a significant advantage is unclear, and recommendations for using these materials in vaginal reconstructive surgery cannot be made due to a lack ofweU-designed prospective trials (72,73).
d. Concomitant Anti-incontinence Procedures It is common for cystocele and stress urinary incontinence to occur simultaneously, and anti-incontinence procedures are performed concurrently with anterior prolapse repairs. It should be noted that severe prolapse of the anterior or apex compartments can kink the urethra and mask the incontinence, and patients should be evaluated carefully before surgical treatment of POE Several anti-incontinence procedures, such as retropubic suspension, suburethral slings, and tension-free transvaginal tape (TVT) or transobturator tape (TOT), can be carried out concomitantly with POP repairs (see Chapter 51). 4. POSTERIOR COMPARTMENT PROLAPSE
a. Posterior Colporrhaphy In this procedure, the vaginal epithelium is first dissected off from the underlying rectovaginal fascia. The pararectal and rectovaginal fasciae are then identified and plicated in the midline, over the rectum. For a large rectocele, the puborectalis muscles can also be approximated to provide a stronger support; however, the plication of these muscles can create a transverse ridge in the posterior vaginal wall and decrease the caliber of the vagina, thus leading to dyspareunia. Perineoplasty is usually performed together with posterior colporrhaphy by approximating the perineal muscles with the deep sutures. The desired postoperative caliber of the vagina and introitus is an important factor in the repair of posterior vaginal wall prolapse using colporrhaphy. Very few studies have addressed the long-term effectiveness of posterior colporrhaphy. If the tissues overlying the rectum are still strong, posterior colporrhaphy can be used to close the defect with successful rates ranging from 76% to 96%. However, the plication of supporting tissues in this procedure often leads to dyspareunia, and narrowing of the vagina can occur with overzealous colporrhaphy or perineoplasty (74). Another study found that almost 38% of women had persistent dyspareunia 1 year or more after posterior repair and Butch procedure (75).
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal Women
b. Site-Specific Repair Although colporrhaphy has traditionally been employed for the correction of posterior vaginal wall prolapse, repairs of the site-specific defects may provide less discomfort for the patient. Although short-term results are similar to or better than posterior colporrhaphy (76), longer follow-up has shown that site-specific repair has higher rates of recurrent prolapse and similar rates of dyspareunia (77). In this approach, only the torn parts of the supporting tissue are repaired, resulting in much less scar tissue and less chance of painful intercourse. The most accurate way to identify the site-specific defects in the posterior vaginal wall is with the intraoperative examination. Midline defects are repaired by re-approximating the edges of the separated fascia with a series of interrupted sutures. Lateral defects should be repaired by reattaching the rectovaginal fascia to the fascial white lines of the pelvic sidewall. Transverse defects can occur at the connection of the rectovaginal fascia to the pericervical ring, at the attachment of the rectovaginal to the pubocervical fascia (in posthysterectomy patients), or at the midvagina. These defects can also be re-approximated with interrupted sutures. Because large numbers of the rectoceles are caused by the breaks between rectovaginal fascia and perineal body, these sites are identified and repaired individually in this approach. c. Compensatory Repair with Graft Materials In the correction of posterior compartment defects, the supporting tissues should be strong enough to make the colporrhaphy or site-specific repairs successful. If the native fascial tissues are extremely attenuated, graft materials can be considered. Similar to the repair of the anterior compartment, biologic or synthetic materials can be used. The graft is attached to the uterosacral ligaments and pubocervical fascia superiorly, to the arcus tendineus fascia rectovaginalis laterally, and to the perineal body inferiorly to provide a strong support for the posterior wall. A recent study reported that rectocele repair using porcine dermal graft was associated with unsatisfactory outcomes and high recurrence rates at a 3-year follow-up postoperatively (78). The use of permanent sutures in posterior repair has also been shown to increase the incidence of erosion and wound dehiscence (79). There is currently no evidence to recommend the routine use of graft materials for posterior repair. d. Sacral Colpoperineopexy If severe perineal descent and vaginal vault prolapse are associated with posterior compartment defects, abdominal sacral colpoperineopexy can be employed (80). This procedure is similar to the sacral colpopexy described earlier, except that the posterior strap of the graft extends down to the perineal body and attaches to it with interrupted sutures. Although the vaginal dissection is needed to expose the perineal body for attachment, the strap should be passed down from above because pushing it up from the vagina
761 is associated with high (up to 40%) rejection and erosion rates. In a series of 19 patients with severe posterior compartment prolapse, 63% reduced to stage 0, 21% to stage I, and the remaining 16% to stage II with sacral colpoperineopexy (80).
e. Enterocele Repair The surgical treatment of an enterocele follows the general principles of hernia repair, which includes identification of the enterocele sac, reduction of intraabdominal contents, and closure of the defect. The repair can be performed vaginally or abdominally. Enterocele rarely occurs in isolation but more often with concurrent apical prolapse, cystocele, and rectocele. The enterocele sac is first mobilized from the vaginal walls and rectum. Rectal examination is helpfial in distinguishing an enterocele from a rectocele. Sometimes the enterocele sac is diffictflt to identify in the presence of a large cystocele, and placement a transilluminating cystoscope may help. After the sac is entered and the contents are reduced, two or three circumferential purse-string sutures are used to close the enterocele sac. The defect is then supported by approximating the uterosacral ligaments. Enterocele repair can also be performed abdominally by obliteration of the cul-de-sac using the Moschcowitz or Halban techniques (81). Transverse plication of the uterosacral ligaments is another abdominal technique to obliterate the cul-de-sac (81). Laparoscopic approach for enterocele repair has been reported with a good success rate at 3 years after the surgery (82). 5. OBLITERATIVEPROCEDURE WITH COLPOCLEISIS
These procedures include partial or complete colpocleisis. In the presence of a uterus, partial colpocleisis provides lateral drainage channels for cervical secretions. LeFort's partial colpocleisis involves suturing the partially denuded anterior and posterior vaginal walls together in such a way that the uterus is supported above the partially occluded vagina (Fig. 52.12). The vaginal epithelium is first dissected off the underlying pubocervical fascia anteriorly and rectovaginal fascia posteriorly. After approximating the pubocervical and rectovaginal fasciae together on the lateral sides of the cervix to provide channels for drainage of cervical secretions, several layers of imbricating sutures that run from the pubocervical anteriorly to the rectovaginal fascia posteriorly are carried out until the prolapse is completely reduced. The vaginal edges are then closed with absorbable sutures. In the absence of the uterus, complete colpocleisis can be performed in a similar fashion as the partial procedure, but it involves total obliteration of the vagina without leaving lateral drainage channels. The success rates of colpocleisis in the treatment of prolapse range from 86% to 100% (83,84). Complications of this procedure are few, with postoperative urinary incontinence being the one of most concern. The advantages of this
762
Ho AND BHATIA
thesia. In patients with a uterus, vaginal hysterectomy should be considered before colpocleisis because this obliterative procedure precludes subsequent evaluation of the cervix and uterine cavity. If hysterectomy cannot be performed or is not desired, carefully preoperative evaluation including a Pap smear, pelvic ultrasonography, and endometrial biopsy should be carried out. 6. RECURRENCE OF PROLAPSE AFTER SURGERY
Postoperatively, patients should avoid any exercise or heavy lifting and refrain from intercourse for 6 weeks. After that, patients can progressively return to their usual daily activities but avoid causes of increased intra-abdominal pressure such as constipation, weight lifting, and cigarette smoking for at least another 3 months to facilitate adequate healing and prevent surgical failures. It has been shown that up to 30% of women who had surgical repair of POP may return for a subsequent procedure (32). The presence of neuromuscular damage of the pelvic diaphragm and attenuation of native tissues compromise the success of the repair. Other reasons for recurrence of prolapse after surgical treatment are congenital disorders in the supporting tissues of the pelvic floor and inadequate repairs of all defects. Although no data are available, ongoing occupational or recreational stresses or chronic illness may also compromise the repair.
FIGURE 52.12 Colpocleisis for obliterative repair. A. The cervix is totally inverted with the pfication of pubocervical fascia anteriorly and rectovaginal fascia posteriorly. B. Another layer of plication. C. The anterior and posterior vaginal mucosa is reapproximated. D. Completion of the procedure. (From Wheeless CR Jr. Atlas of pelvic surgery, ed. 3. Baltimore: Williams & Wilkins, 1997:81.)
procedure are the short operative time, which minimizes the blood loss, and the use of regional anesthesia or local anesthesia with sedation, which significantly decreases operative risks for patients with chronic illness. Colpocleisis can be considered for patients with comorbidity who cannot undergo long surgical procedures and who are not contemplating sexual activities. It is contraindicated for any women with a desire to maintain vaginal function, and the patient must be carefully counseled. This procedure may cause postoperative stress urinary incontinence because it flattens the posterior urethrovesical angle. Concurrent anti-incontinence procedures, such as TOT, TVT, or suburethral sling procedures (see Chapter 51) can be carried out at the time of colpocleisis with minimal additional morbidity. T O T is particularly useful in high-risk patients because it is less invasive, has fewer complications, requires less time, and can be performed with regional anes-
References 1. Hill K. The demography of the menopause. Matures 1996;23:113-127. 2. Rogers RM Jr. Anatomy of pelvic support. In: Bent AE, Ostergard DR, Cundiff GW, Swift SE, ed. Ostergard's urogynecology and pelvic floor dysfunction, ed. 5. Philadelphia: Lippincott, Williams & Wilkins, 2003:19-33. 3. DeLancey JOL, Delmas V. Gross anatomy and functional anatomy of the pelvic floor. In: Bourcier AP, McGuire EJ, Abrams P, eds. Pelvic floor disorders. Philadelphia: Elsevier Saunders, 2004:3-6. 4. DeLancey JO. Anatomic aspects of'vaginal eversion after hysterectomy. Am J Obstet Gyneco11992;166:1717-1724; discussion, 1724-1728. 5. Nichols DH, Milley PS, Randall CL. Significance of restoration of normal vaginal depth and axis. Obstet Gyneco11970;36:251-256. 6. Abrams P, Cardozo L, Fall M, et al. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn 2002;21:167-178. 7. Baden WF, Walker T. Fundamental, symptoms, and classification. In: Baden WF, Walker T, eds. Surgicalrepair of vaginal defects. Philadelphia: JP Lippincott, 1992. 8. Walters MD. Description and classification of lower urinary tract dysfunction and pelvic organ prolapse. In: Wakers MD, Karram MM, eds. Urogynecology and reconstructive pelvic surgery, ed. 3. Philadelphia: Mosby, 2007:55-61. 9. Bump RC, Mattiasson A, Bo K, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J
Obstet Gyneco11996;175:10-17.
CHAPTER 52 Pelvic Organ Prolapse in Postmenopausal W o m e n 10. Beecham CT. Classification of vaginal relaxation. Am J Obstet Gynecol 1980;136:957-958. 11. Hall AF, Theofrastous JP, Cundiff GW, et al. Interobserver and intraobserver reliability of the proposed International Continence Society, Society of Gynecologic Surgeons, and American Urogynecologic Society pelvic organ prolapse classification system. Am J Obstet Gynecol 1996;175:1467-1470; discussion, 1470-1471. 12. Samuelsson EC, Victor FT, Tibblin G, Svardsudd KE Signs of genital prolapse in a Swedish population of women 20 to 59 years of age and possible related factors. Am J Obstet Gyneco11999;180:299- 305. 13. Tegerstedt G, Maehle-Schmidt M, Nyren O, Hammarstrom M. Prevalence of symptomatic pelvic organ prolapse in a Swedish population. Int UrogynecolJPelvic Floor Dysfunct 2005;16:497-503. 14. Brown JS, Waetjen LE, Subak LL, et al. Pelvic organ prolapse surgery in the United States, 1997./IMJ Obstet Gyneco12002;186:712-716. 15. Bland DR, Earle BB, Vitolins MZ, Burke G. Use of the Pelvic Organ Prolapse staging system of the International Continence Society, American Urogynecologic Society, and Society of Gynecologic Surgeons in perimenopausal women. Am J Obstet Gynecol 1999;181:1324-1327; discussion, 1327-1328. 16. Swift SE. The distribution of pelvic organ support in a population of female subjects seen for routine gynecologic health care. Am J Obstet Gyneco12000;183:277-285. 17. Hendrix SL, Clark A, Nygaard I, et al. Pelvic organ prolapse in the Women's Health Initiative: gravity and gravidity. Am J Obstet Gynecol 2002;186:1160-1166. 18. Nygaard I, Bradley C, Brandt D. Women's Health Initiative. Pelvic organ prolapse in older women: prevalence and risk factors. Obstet Gyneco12004;104:489-497. 19. Kim S, Harvey MA, Johnston S. A review of the epidemiology and pathophysiology of pelvic floor dysfunction: do racial differences matter? J Obstet GynaecolCan 2005;27:251-259. 20. Moalli PA, Shand SH, Zyczynski HM, Gordy SC, Meyn LA. Remodeling of vaginal connective tissue in patients with prolapse. Obstet Gynecol 2005;106:953-963. 21. Jackson SR, Avery NC, Tarlton JF, et al. Changes in metabolism of collagen in genitourinary prolapse. Lancet 1996;347:1658-1661. 22. Norton PA, Baker JE, Sharp HC, Warenski JC. Genitourinary prolapse and joint hypermobility in women. Obstet Gynecol 1995;85: 225-228. 23. McIntosh LJ, Stanitski DF, Mallett VT, et al. Ehlers-Danlos syndrome: relationship between joint hypermobility, urinary incontinence, and peMc floor prolapse. Gynecol Obstet Invest 1996;41:135-139. 24. Tetzschner T, Sorensen M, Jonsson L, Lose G, Christiansen J. Delivery and pudendal nerve function. /tcta Obstet Gynecol &and 1997;76: 324-431. 25. Peschers UM, Schaer GN, DeLancey JO, Schuessler B. Levator ani function before and after childbirth. Br J Obstet Gynaecol 1997;104: 1004-1008. 26. Swift SE, Pound T, Dias JK. Case-control study of etiologic factors in the development of severe pelvic organ prolapse. Int UrogynecolJPelvic Floor Dysfunct 2001;12:187-192. 27. Smith AR, Hosker GL, Warrell DW. The role of partial denervation of the pelvic floor in the aetiology of genitourinary prolapse and stress incontinence of urine. A neurophysiological study. BrJ Obstet Gynaecol 1989;96:24-28. 28. Allen RE, Hosker GL, Smith AR, Warrell DW. Pelvic floor damage and childbirth: a neurophysiological study. BrJ Obstet Gynaeco11990;97: 770-779. 29. MacLennan AH, Taylor AW, Wilson DH, Wilson D. The prevalence of pelvic floor disorders and their relationship to gender, age, parity and mode of delivery. BJOG 2000;107:1460-1470. 30. Patel DA, Xu X, Thomason AD, et al. Childbirth and pelvic floor dysfunction: an epidemiologic approach to the assessment of prevention opportunities at delivery. ArnJ Obstet Gyneco12006;195:23-28.
763 31. Gurel H, Gurel SA. Pelvic relaxation and associated risk factors: the results of logistic regression analysis.Acta Obstet Gynecol&and 1999;78: 290-293. 32. Olsen AL, Smith VJ, Bergstrom JO, Colling JC, Clark AL. Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gyneco11997;89:501-506. 33. Benson JT, Lucente V, McClellan E. Vaginal versus abdominal reconstructive surgery for the treatment of pelvic support defects: a prospective randomized study with long-term outcome evaluation, zlmJ Obstet Gyneco11996;175:1418-1421; discussion, 1421-1422. 34. Shull BL, Capen CV, Riggs MW, Kuehl TJ. Preoperative and postoperative analysis of site-specific pelvic support defects in 81 women treated with sacrospinous ligament suspension and pelvic reconstruction. Am J Obstet Gyneco11992;166:1764-1768; discussion, 1768-1771. 35. Symmonds RE, Williams TJ, Lee RA, Webb MJ. Posthysterectomy enterocele and vaginal vault prolapse. Am J Obstet Gynecol 1981;140: 852-859. 36. Richter K. Massive eversion of the vagina: pathogenesis, diagnosis, and therapy of the "true" prolapse of the vaginal stump. Clin Obstet Gynecol 1982;25:897-912. 37. Thakar R, Sultan AH. Hysterectomy and peMc organ dysfunction. Best Pract Res Clin Obstet Gynaeco12005;19:403-418. 38. Nygaard IE. Does prolonged high-impact activity contribute to later urinary incontinence? A retrospective cohort study of female Olympians. Obstet Gyneco11997;90:718-722. 39. Jorgensen S, Hein HO, Gyntelberg E Heavy lifting at work and risk of genital prolapse and herniated lumbar disc in assistant nurses. Occup Med (Lond) 1994;44:47-49. 40. Handa VL, Garrett E, Hendrix S, Gold E, Robbins J. Progression and remission of pelvic organ prolapse: a longitudinal study of menopausal women. Am J Obstet Gyneco12004;190:27- 32. 41. Swift SE, Herring M. Comparison of pelvic organ prolapse in the dorsal lithotomy compared with the standing position. Obstet Gynecol 1998;91:961-964. 42. Hoyte L, Jakab M, Warfield SK, et al. Levator ani thickness variations in symptomatic and asymptomatic women using magnetic resonance-based 3-dimensional color mapping. Am J Obstet Gyneco12004;191:856- 861. 43. Barber MD, Waiters MD, Bump RC. Short forms of two conditionspecific quality-of-life questionnaires for women with peMc floor disorders (PFDI-20 and PFIQz7).AmJ Obstet Gyneco12005;193:103-113. 44. Barber MD, Waiters MD, Cundiff GW, PESSRI Trial Group. Responsiveness of the Pelvic Floor Distress Inventory (PFDI) and Pelvic Floor Impact Questionnaire (PFIQ) in women undergoing vaginal surgery and pessary treatment for pelvic organ prolapse. Am J Obstet Gyneco12006;194:1492-1498. 45. Jarvis SK, Hallam TK, Lujic S, Abbott JA, Vancaillie TG. Perioperative physiotherapy improves outcomes for women undergoing incontinence and or prolapse surgery: results of a randomised controlled trial. Aust N Z J Obstet Gynaeco12005;45:300-303. 46. Bo K. Can pelvic floor muscle training prevent and treat pelvic organ prolapse? Acta Obstet Gynecol &and 2006;85:263-268. 47. Boyington AR, Dougherty MC. Pelvic muscle exercise effect on pelvic muscle performance in women. Int UrogynecolJ Pelvic Floor Dysfunct 2000;11:212-218. 48. Bump RC, Hurt WG, Fantl JA, Wyman JE Assessment of Kegel peMc muscle exercise performance after brief verbal instruction. Am J Obstet Gyneco11991;165:322- 327; discussion, 327-329. 49. Burgio KL, Robinson JC, Engel BT. The role of biofeedback in Kegel exercise training for stress urinary incontinence. Am J Obstet Gynecol 1986;154:58-64. 50. Peattie AB, Plevnik S, Stanton SL. Vaginal cones: a conservative method of treating genuine stress incontinence. BrJ Obstet Gynaecol 1988;95:1049-1053.
764
51. Bo K, Talseth T, Holme I. Single blind, randomised controlled trial of pelvic floor exercises, electrical stimulation, vaginal cones, and no treatment in management of genuine stress incontinence in women. BMJ 1999;318:487-493. 52. Cundiff GW, Weidner AC, Visco AG, Bump RC, Addison WA. A survey of pessary use by members of the American Urogynecologic Society. Obstet Gyneco12000;95:931-935. 53. Wu V, Farrell SA, Baskett TF, Flowerdew G. A simplified protocol for pessary management. Obstet Gyneco11997;90:990-994. 54. Sulak PJ, Kuehl TJ, Shull BL. Vaginal pessaries and their use in pelvic relaxation. J ReprodMed 1993;38:919- 923. 55. Schraub S, Sun XS, Maingon P, et al. Cervical and vaginal cancer associated with pessary use. Cancer 1992;69:2505-2509. 56. Langer R, Ron-E1 R, Neuman M, et al. The value of simultaneous hysterectomy during Butch colposuspension for urinary stress incontinence. Obstet Gyneco11988;72:866-869. 57. Bullet JL, Thompson JR, Cundiff GW, et al. Uterosacral ligament: description of anatomic relationships to optimize surgical safety. Obstet Gyneco12001;97:873- 879. 58. Barber MD, Visco AG, Weidner AC, Amundsen CL, Bump RC. Bilateral uterosacral ligament vaginal vault suspension with site-specific endopelvic fascia defect repair for treatment of pelvic organ prolapse. Am J Obstet Gyneco12000;183:1402-1410; discussion, 1410-1411. 59. Shull BL, Bachofen C, Coates KW, Kuehl TJ. A transvaginal approach to repair of apical and other associated sites of pelvic organ prolapse with uterosacral ligaments. Am J Obstet Gyneco12000;183:1365-1373; discussion, 1373-1374. 60. Shull BL, Capen CV, Riggs MW, Kuehl TJ. Bilateral attachment of the vaginal cuff to iliococcygeus fascia: an effective method of cuff suspension. Am J Obstet Gynecol 1993;168:1669-1674; discussion, 1674-1677. 61. Meeks GR, Washburne JF, McGehee RP, Wiser WL. Repair of vaginal vault prolapse by suspension of the vagina to iliococcygeus (prespinous) fascia. Am J Obstet Gynecol 1994;171:1444-1452; discussion, 1452-1454. 62. Pohl JF, Frattarelli JL. Bilateral transvaginal sacrospinous colpopexy: preliminary experience. Am J Obstet Gyneco11997;177:1356-1361; discussion, 1361-1362. 63. Smilen SW, Saini J, Wallach SJ, Porges RE The risk of cystocele after sacrospinous ligament fixation. Am J Obstet Gynecol 1998;179: 1465-1471; discussion, 1471-1472. 64. Nygaard IE, McCreery R, Brubaker L, et al., Pelvic Floor Disorders Network. Abdominal sacrocolpopexy: a comprehensive review. Obstet Gyneco12004;104:805- 823. 65. Wu JM, Wells EC, Hundley AF, et al. Mesh erosion in abdominal sacral colpopexy with and without concomitant hysterectomy. Am J Obstet Gyneco12006;194:1418-1422. 66. Paraiso MF, Wakers MD, Rackley RR, Melek S, Hugney C. Laparoscopic and abdominal sacral colpopexies: a comparative cohort study. Am J Obstet Gyneco12005;192:1752-1758. 67. Haase P, Skibsted L. Influence of operations for stress incontinence and/or genital descensus on sexual fife. Acta Obstet Gynecol&and 1988; 67:659-661.
H o AND BHATIA
68. Cruikshank SH, Muniz M. Outcomes study: a comparison of cure rates in 695 patients undergoing sacrospinous ligament fixation alone and with other site-specific procedures - - a 16-year study. Am J Obstet Gyneco12003;188:1509-1512; discussion, 1512-1515. 69. Dwyer PL. Evolution of biological and synthetic grafts in reconstructive pelvic surgery. Int UrogynecolJ Pelvic Floor Dysfunct 2006;17: S10-S15. 70. Julian TM. The efficacy of Marlex mesh in the repair of severe, recurrent vaginal prolapse of the anterior midvaginal wall. Am J Obstet Gyneco11996;175:1472-1475. 71. Sand PK, Koduri S, Lobel RW, et al. Prospective randomized trial of polyglacfin 910 mesh to prevent recurrence of cystoceles and rectoceles. Am J Obstet Gyneco12001;184:1357-1362; discussion, 1362-1364. 72. Wheeler TL II, Richter HE, Duke AG, et al. Outcomes with porcine graft placement in the anterior vaginal compartment in patients who undergo high vaginal uterosacral suspension and cystocele repair. Am J Obstet Gyneco12006;194:1486-1491. 73. Huebner M, Hsu Y, Fenner DE. The use of graft materials in vaginal pelvic floor surgery. IntJ GynaecolObstet 2006;92:279-288. 74. Cuncliff GW, Fenner D. Evaluation and treatment of women with rectocele: focus on associated defecatory and sexual dysfunction. ObstetGynecol 2004;104:1403-1421. [Erratum in ObstetGyneco12005;105:222.] 75. Weber AM, Waiters MD, Piedmonte MR. Sexual function and vaginal anatomy in women before and after surgery for pelvic organ prolapse and urinary incontinence. Am J Obstet Gyneco12000;182:1610-1615. 76. Singh K, Cortes E, Reid WM. Evaluation of the fascial technique for surgical repair of isolated posterior vaginal wall prolapse. Obstet Gynecol 2003;101:320-324. 77. Abramov Y, Gandhi S, Goldberg RP, et al. Site-specific rectocele repair compared with standard posterior colporrhaphy. Obstet Gynecol 2005;105:314-318. 78. Altman D, Zetterstrom J, Mellgren A, et al. A three-year prospective assessment of rectocele repair using porcine xenograft. Obstet Gynecol 2006;107:59-65. 79. Luck AM, Galvin SL, Theofrastous JP. Suture erosion and wound dehiscence with permanent versus absorbable suture in reconstructive posterior vaginal surgery. Am J Obstet Gyneco12005;192:1626-1629. 80. Cundiff GW, Harris RL, Coates K, et al. Abdominal sacral colpoperineopexy: a new approach for correction of posterior compartment defects and perineal descent associated with vaginal vault prolapse. Am J Obstet Gyneco11997;177:1345-1353; discussion, 1353-1355. 81. Shull BL, Bachofen CG. Enterocele and rectocele. In: Waiters MD, Karram MM, eds. Urogynecologyand reconstructivepelvic surgery, ed. 2. St. Louis: Mosby, 1999:221-234. 82. Cook JR, Seman EI, O'Shea RT. Laparoscopic treatment of enterocele: a 3-year evaluation. Aust N Z J Obstet Gynaeco12004;44:107-110. 83. DeLancey JO, Morley GW. Total colpocleisis for vaginal eversion. Am J Obstet Gyneco11997;176:1228-1232; discussion, 1232-1235. 84. Fitzgerald MP, Richter HE, Siddique S, et al., for the Pelvic Floor Disorders Network. Colpocleisis: a review. Int UrogynecolJPelvic Floor Dysfunct 2006;17:261 - 271.
SECTION XII
Hormonal Therapy This section is not large in comparison to several other sections in this book. However, the relevance of hormonal therapy has been discussed in many sections of this text. Here the intent is to systematically discuss various hormonal options. At the same time, it has been assumed that the readers of this book will not require a "cookbook" approach for prescribing. In my view it is more important to discuss all the various options available while providing some basic information on the physiology and pharmacology of hormonal therapy. The first chapter of this Section is by Randall B. Barnes, who describes the pharmacology of estrogens. Following this, Frank Z. Stanczyk discusses the use of progestogens, the various classes of progestogens, and their effects in the body. Next, Susan R. Davis summarizes the data for the use of androgens in women. In the last chapter, I have asked G6ran Samsioe and Martina D6ren to think ahead and try to project what may lie ahead regarding various aspects of hormonal therapy.
This Page Intentionally Left Blank
; H A P T E R 5[
Pharmacology of Estrogens RANDALL B. BARNES
SETH G.
LEVRANT
Department of Obstetrics and Gynecology, Feinberg School of Medicine, Northwestern University, Chicago, IL 60633
TinleyPark, IL 60477
I. E S T R O G E N P R O D U C T I O N AND ACTION
estradiol has also been reported throughout the menstrual cycle (5,7). Luteal phase progesterone has been found to be both normal (6,7) and decreased (8), whereas urinary pregnanediol has been found to be decreased (5). The contradictory estrogen and progesterone findings may be due to small sample sizes together with greater cycle variability in perimenopausal women or to differences in sampling technique. Later in the perimenopause, estrogen levels rise and fall without evidence of ovulation, and cycles become irregular (5-7). FSH and luteinizing hormone (LH) levels increase, and eventually anovulation and amenorrhea occur. The ovary fails to respond to gonadotropin stimulation, and estrogen deficiency results. Postmenopausal women have serum estradiol levels below 15 pg/ml and mean estrone levels of about 30 pg/ml (9-11). In postmenopausal women the predominant estrogen is estrone, with a reversal in the estradiol to estrone ratio to less than 1. The primary source of estrone is from peripheral aromatization of androstenedione. Ninety-five percent of postmenopausal androstenedione production occurs in the adrenal gland, and 5% occurs in the ovaries. The conversion rate of androstenedione to estrone is 1% to 2% in normal premenopausal women (12-14). Increased conversion of androstenedione to estrone occurs with age (15) and with increasing weight (as high as 12% to 15% conversion rate in women who weigh 135 to 181 kg) (9,14,16). The conversion of androstenedione to estrone is catalyzed by the microsomal enzyme, aromatase cytochrome P450, which is a product of a single gene, CYP19. Tissue-specific
During the reproductive years, the main source of estrogen in women is the dominant follicle and the corpus luteum. The principal estrogen produced is estradiol (1). Normal ovulatory cycles result in serum estradiol levels ranging from 40 pg/mL in the early follicular phase to 250 pg/mL at midcycle and 100 pg/mL during the midluteal phase (2). Ovarian production of estradiol ranges from 60 to 600 rag/day during the menstrual cycle (3). The dominant follicle and corpus luteum account for more than 95% of the circulatory estradiol in the premenopausal women (1,4). Peripheral conversion of estrone to estradiol accounts for most of the remaining circulating estradiol (1). Estrone, the second major human estrogen, is derived principally from the metabolism of estradiol and from the aromatization of androstenedione in adipose tissue (1). A small quantity is secreted directly by the ovary and adrenal. Serum levels during the menstrual cycle vary from 40 to 170 pg/mL, paralleling estradiol levels. The estradiol to estrone ratio is greater than 1.0 in premenopausal women (1). In ovulatory, perimenopausal women the most consistent findings are a shortened follicular phase and elevated follicle-stimulating hormone (FSH) throughout the cycle compared with younger ovulatory women (1,5-7). Perimenopausal women have increased follicular and luteal phase urinary estrogen excretion (5) and increased follicular phase serum estradiol (8); however, decreased serum TREATMENT OF THE POSTMENOPAUSAL W O M A N
767
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
768
BARNES AND LEVRANT
promoters for CYP19 have been identified for the gonads, placenta, adipose tissue, bone, and brain, and each promoter has it characteristic regulatory factors (17). In adipose tissue, aromatase gene expression increases with age and is greater in gluteal than abdominal fat (18). The level of aromatase activity in stromal cells is greater than that in adipocytes (14). Aromatase activity in stromal cells can be increased up to 50-fold by glucocorticoid steroids (19). Weight loss may not decrease the aromatase activity in obese individuals, which suggests that the level of aromatase activity in stromal cells may be independent of the total mass of adipose tissue (14). Peripheral aromatization of testosterone to estradiol contributes minimally to estradiol production postmenopausally (20). The primary source of estradiol in the postmenopausal women is from the peripheral conversion of estrone. The conversion between estradiol and estrone is catalyzed by 1713-hydroxysteroid dehydrogenase (1713HSD) in the liver and other tissues (21). At least six human 17[3-HSD isoenzymes have been identified (22,23). The type I and 7 enzymes found in the ovary and placenta favor reduction and formation of estradiol, whereas the type 2 enzyme found in the liver and endometrium and the ubiquitous Type 4 enzyme favor oxidation and production of estrone. Thus, the peripheral conversion between estradiol and estrone favors the formation of estrone. Fifteen percent of estradiol is converted to estrone, whereas only 5% of estrone is converted to estradiol (Fig. 53.1). The amount of estrone derived from estradiol (approximately 3 pg/mL) is extremely low because of the low concentrations of estradiol in postmenopausal women. Another source of estrone is from the reversible metabolism of estrone-3-sulfate (estrone sulfate), which is the principle circulating estrogen in the body (24,25). Estrone sulfate is 90% bound to albumin and has a metabolic clearance rate (MCR) of 127 _+ 33 liters (L)/day/m 2 (range 102-180) (26). Levels of estrone sulfate follow the same general patterns as estrone and estradiol, ranging from 1000 pg/mL in the follicular phase to 1800 pg/mL in the
Estrone Sulfate~ ,,
,
C Estr~ ) --
/
0.15
• C Estr'di~
)
/
Estr~, epiestriols j~.Estrogen I conjugates Catechol estrogens/~ ] Excretion by kidney FIGURE 53.1 Estrogen metabolism and conversion ratios between estradiol, estrone, and estrone sulfate.
luteal phase (27-30). In postmenopausal women, estrone sulfate levels reach 320 pg/mL, which is many times higher than levels of serum estrone (30 pg/mL) and estradiol (13 pg/mL) (30). Sixty-five percent of estradiol and 54% of estrone are converted to estrone sulfate, whereas 21% of estrone sulfate is converted to estrone and only 1.5% is converted to estradiol (25). At the intracellular level, estradiol is the predominant estrogen, even in postmenopausal women in whom estrone is the predominant estrogen in the circulation (31). Differences in microvasculature permeability, plasma protein binding, and intracellular 17[3-HSD activity are major determinants of end-target estrogen activity (31,32). Hepatic extraction of all estrogens significantly exceeds that of the brain and uterus because the liver microcirculation does not have a basement membrane and its permeability barrier consists of only the hepatocyte plasma membrane (33). The extraction of estrogens by the liver, brain, endometrium, and myometrium exceed the free (dialyzable) fractions of the compounds, indicating that part of the protein-bound estrogens are available for tissue extraction (32,33). Influx of estrone sulfate and estrone into the endometrium exceeds that into the myometrium (32). Secretory endometrium, under the influence of progesterone, deactivates intracellular estradiol through type 2 17[3-HSD conversion to estrone (30). Estradiol and other unconjugated estrogens diffuse freely into cells, but estrogenic activity within a cell is dependent upon the presence of the estrogen receptor (ER). There are two ERs, alpha and beta (34,35), which are members of the nuclear receptor superfamily. The steroid hormone receptor members of the superfamily are hormone-activated regulators of transcription. When unbound, the ER is associated with a multiprotein complex that includes heat shock proteins. These proteins maintain the receptor in a conformation that allows it to easily bind to the estrogen ligand and may also prevent constitutive activity of the receptor (36). Binding of estrogen to its receptor results in receptor dimerization and binding to the palindromic ER response element of the promoter region of estrogen-responsive genes. The receptor then interacts with the transcription initiation complex to regulate transcription (36,37). The binding affinity of the physiologic estrogens for the ER parallel their in vivo activity with estradiol > estrone > estriol > catecholestrogens. The stilbene estrogens such as diethylstilbestrol have an affinity three to four times higher than that of estradiol, whereas the triphenylethylene estrogen agonist/antagonist vary, with 4-hydroxytamoxifen having almost twice the affinity of estradiol and clomiphene and tamoxifen having 25% or less affinity for the ER than does estradiol (38). The human ER-beta has 47% homology with ER-alpha (39,40), with 96% homology in the DNA binding domain and 58% in the ligand binding domain (41). In the presence of estradiol, both receptors bind to the same DNA estrogen
769
CHAPTER53 Pharmacology of Estrogens response element with similar affinity (42). The affinity of ER-beta for natural and synthetic ligands is similar to that of ER-alpha, except that the ER-beta has a higher affinity for phytoestrogens (38). Tamoxifen is both an agonist and antagonist for ER-alpha but only an antagonist for ER-beta (43). In the human ER-beta mRNA is expressed in classic estrogen-sensitive tissues such as the uterus, breast, and bone (41,44,45). It is also expressed in human granulosa cells and spermatids, tissues devoid of ER-alpha (41). In addition to their role as classic DNA binding nuclear receptors, both ER-alpha and ER-beta are present in mitochondria. The mitochondrial genome contains potentially estrogen-responsive sequences, and estrogen increases mitochondrial DNA-encoded gene transcript levels (43). Like peptide growth factors that act on membrane-bound receptors, estrogens activate various protein kinases and increase levels of second messengers such as cyclic adenosine monophosphate (AMP) within minutes. Membrane-bound forms of ER-alpha and ER-beta cause these nontranscriptional effects (43). Thus, estrogens acting via two receptors and through both genomic and nongenomic pathways are able to effect an incredibly complex nexus of cellular regulation.
II. E S T R O G E N METABOLISM Irreversible metabolism of estrogen proceeds primarily in the liver by oxidation of estrone via two different pathways (21). Hydroxylation at the 16 position on the D ring results in the formation of estriol, a biologically weak estrogen, and its isomers, epiestriols. Obesity, hypothyroidism, and cirrhosis favor the metabolism of estrone to estriol. Hydroxylation at the 2 or 4 positions on the A ring results in the formation of catechol estrogens (see Fig. 53.1). Catechol estrogens are also produced in the hypothalamus, where they may have important central nervous systems effects. Catechol estrogens competitively inhibit the enzymes tyrosine hydroxylase and catechol-o-methyl transferase. Catechol estrogens can modulate synthesis and degradation of catecholamines, dopamine, and norepinephrine, neurotransmitters important in the control ofgonadotropinreleasing hormone (GnRH) release (46). Catechol estrogen formation by A ring metabolism is favored in states of weight loss, such as anorexia nervosa and hyperthyroidism. Estrogen metabolism proceeds in the liver and kidney by conjugation of estriol, epiestriols, and catechol estrogens to glucuronides and sulfates, which are highly water soluble and are rapidly excreted by the kidney. The principle urinary estrogens are the conjugates of estriol and 2-hydroxyestrone (47). In addition to urinary excretion, there is a significant enterohepatic circulation of estrogen metabolites. After labeled estrone or estradiol is injected, approximately one-half is present in the urine excreted during the first 24 hours while the remainder is found in bile (48). The conjugated
biliary estrogens undergo hydrolysis in the gut, and approximately 80% are reabsorbed. They are returned to the liver, where they may escape reconjugation and enter the systemic circulation, or they may be reconjugated and excreted in the urine or bile. Only about 10% of an injected estrogen will ultimately be lost in the feces. The enterohepatic circulation of estrogen metabolites may be an important factor in the prolonged effect of orally administered estrogens. All estrogens, conjugated and unconjugated, circulate in the blood either protein bound or unbound (physiologically free). Estrogens are specifically bound with high affinity to sex hormone-binding globulin (SHBG) or loosely (nonspecifically) bound to serum albumin. Serum binding affects the availability of estrogen to diffuse across cell membranes and express biologic activity. Thirty-eight percent of estradiol is bound to SHBG, 60% is loosely bound to albumin, and approximately 2% to 3% is free to diffuse across cell membranes (49). Estrone, estriol, and estrone sulfate all bind poorly to SHBG but have a greater affinity for albumin than does estradiol. Estrone sulfate has the highest affinity for albumin, with more than 90% of circulating estrone sulfate bound to albumin (50). Alterations in SHBG levels change the concentration of unbound estradiol altering bioavailability. Estrogen therapy, pregnancy, and hyperthyroidism increase SHBG, and hypothyroidism, androgen excess, hyperinsulinemia, and obesity lower SHBG levels (14,51,52). Obese women who are ->50 lb (23 kg) above ideal body weight have serum SHBG binding capacity reduced by 20% to 30% compared with normal-weight postmenopausal women, which results in a twofold to threefold increase in the free and albumin-bound fractions of estradiol (13). In summary, the liver plays the central role in the metabolism and excretion of estrogens and is, just as importantly, influenced by estrogen status. The liver is affected by estrogens absorbed from the gastrointestinal tract (first-pass effect) and by reabsorption of estrogens secreted in the bile (enterohepatic circulation). The liver is a principal site of metabolic interconversion of estrogens as well as conjugation of estrogens in preparation for excretion into bile and by the kidney. Bioavailability of estrogens is affected by SHBG and albumin, products of the liver. Estrogens in turn influence liver carbohydrate metabolism, lipid metabolism, bile production, and protein production (binding proteins, clotting factors, and renin substrate).
III. PHARMACOLOGY OF E S T R O G E N PREPARATIONS Estrogens for hormone therapy may be separated by chemical composition (natural or synthetic) or by route of administration (oral or parenteral) (Tables 53.1 and 53.2). Natural estrogens include estradiol, estrone, estriol, and
770
BARNES AND LEVRANT
TABLE 53.1
Estrogen Replacement Therapies Available in the United States
Generic name Oral Conjugated equine estrogens Synthetic conjugated estrogens Estropipate (Piperazine estrone sulfate) Micronized estradiol Estradiol acetate Raloxifene (selective estrogen receptor modulator) Oral combined Conjugated equine estrogens (CEE) and medroxyprogesterone acetate (MPA) Esterified estrogens and methyltestosterone Estradiol and norethindrone acetate Estradiol and norgestimate
Brand name
Daily dose
Premarin Cenestin
0.3, 0.45, 0.625, 0.9, 1.25 mg 0.3, 0.45, 0.625, 0.9, 1.25 mg
Estrace Femtrace Evista
0.75, 1.5, 3 mg 0.5, 1, 2 mg 0.45, 0.9, 1.8 mg 60 mg
Prempro Estratest, Estratest HS Activella Prefest
0.3/1.5, 0.45/1.5, 0.625/2.5, 0.625/5 mg (CEE/MPA) 1.25/2.5, 0.625 mg/1.25 mg esterified estrogen/methyltestosterone 1/0.5 mg estradiol/norethindrone acetate 1 mg estradiol for 3 days alternating with 1 mg estradiol/0.09 mg norgestimate for 3 days
Transdermal
Estradiol patch
Estradiol gel
Alora (twice weekly) Climara (weekly) Menostar (weekly) Vivelle (twice weekly) Vivelle-Dot (twice weekly) Estrogel (daily)
0.014 (Menostar), 0.025, 0.0375, 0.05, 0.06, 0.075, 0.1 mg
Climara Pro (weekly) CombiPatch (twice weekly)
0.045/0.015 mg estradiol/levonorgestrel 0.05/0.14, 0.05/0.25 mg estradiol/norethindrone acetate
Premarin vaginal cream Estrace vaginal cream
0.625 mg/gram 1.0 mg/gram
Estring (3 months) Femring (3 months)
0.0075 mg 0.05, 0.10 mg
Vagifem
0.025 mg
Transdermal combined
Estradiol and levonorgestrel Estradiol and norethindrone acetate Vaginal cream CEEs Micronized estradiol Vaginal ring Estradiol Estradiol acetate Vaginal tablet Estradiol
their conjugates as well as conjugated equine estrogens (CEEs). Synthetic estrogens include ethinyl estradiol, mestranol, quinestrol, diethylstilbestrol, and raloxifene. Parenteral routes of administration include injection, transvaginal (creams, tablets, and silastic rings), transdermal patches, subcutaneous pellets, intranasal, and percutaneous gel administration. No matter the type of estrogen or the route of administration, large intraindividual and interindividual variations of serum concentrations occur (53,54).
A. Synthetic Estrogens Synthetic estrogens that have been used for menopause estrogen replacement include ethinylestradiol; its C-3 methylated derivative, mestranol; its cyclophenyl ether, quinestrol;
1.25-gram gel containing 0.75 mg estradiol
and the stilbene derivative diethylstilbestrol. Ethinyl estradiol, the estrogen used in most oral contraceptives, is rapidly absorbed in the gastrointestinal tract after oral administration. Comparing oral and intravenous doses of ethinyl estradiol, the bioavailability of oral ethinyl estradiol is 59.0 _+ 13% (55). An oral dose of 50 mg of ethinyl estradiol results in a blood level of 400 pg/mL. The ethinyl group at position 17e~ on ethinyl estradiol prevents oxidation by 17[3-HSD. With D-ring metabolism impeded, the principal inactivation pathway is ring A 2-hydroxylation to catecholestrogens. Liver cytochrome P-450 enzymes, the chief site of oxidative transformation of ethinyl estradiol, can become irreversibly inactivated by the intermediate compounds of ethinyl estradiol metabolism (56). The strong potency and pronounced effects on hepatic metabolism of ethinyl estradiol compared with natural estrogens are a consequence of the relatively slow
CHAPTER 53 Pharmacology of Estrogens
771
TABLE 53.2 Approximate Serum Estradiol and Estrone Levels after Administration of Various Estrogen Preparations Preparation Oral Conjugated equine estrogens Estropipate
Micronized estradiol Vaginal Conjugated equine estrogen Micronized estradiol Transdermal Estradiol patch Estradiol gel
Daily dose (mg)
Estradiol (pg/mL)
Estrone (pg/mL)
0.625 1.25 0.625 1.25 2.5 1 2
30-50 40-60 35 30-50 125 30-50 50-180
150 120-200 125 150-300 360 150-300 300-850
1.25 0.5
25 - 40 250
65 - 80 130
0.05 0.1 1.5 3.0
30-65 50-90 40-100 60-140
40-45 30-65 90 45-155
ethinyl estradiol that has an extremely long half-life. The pharmacokinetics of mestranol and quinestrol correspond to those of ethinyl estradiol. Raloxifene, a nonsteroidal benzothiophene derivative initially developed as a therapy for breast cancer, is approved in the United States for the prevention of osteoporosis in postmenopausal women. Raloxifene is a selective estrogen receptor modulator (SERM) that acts as an estrogen agonist in bone and the liver but as an estrogen antagonist in the uterus and breast (60,61). It binds to recombinant human ER with an affinity similar to estradiol. Raloxifene treatment maintained bone mineral content in postmenopausal women over a period of 2 years (62) and decreased total and low-density lipoprotein (LDL) cholesterol. Although raloxifene had no overall effect on high-density-lipoprotein (HDL) cholesterol, it increased HDL-2 cholesterol but not to the same extent as with CEEs (63). After oral administration, raloxifene undergoes extensive first-pass metabolism to its glucuronide conjugates, which are its primary circulating forms in plasma. It is excreted mainly in the feces, with less than 6% of the dose eliminated in urine as glucuronide conjugates (Evista Package Insert).
B. Conjugated Equine Estrogens transformation into inactive metabolites in the hepatocytes and resultant high local concentrations during the first liver passage. Ethinyl estradiol is approximately 75 to 1000 times more potent on a per weight basis than CEEs, piperazine estrone sulfate, or micronized estradiol in terms of increasing the production of hepatic proteins (SHBG, renin substrate, corticosteroid binding globulin, and thyroid binding globulin) (57). Potencies of estrogens including ethinyl estradiol vary according to the site of action; for example 10 mg is required for the normalization of the urinary calcium/ creatinine ratio, while 5 mg produces elevations in hepatic protein production (58). As with natural estrogens, the major circulatory form of ethinyl estradiol is its 3-sulfate (56). Ethinyl estradiol is excreted in the urine and feces as glucuronides and sulfates and undergoes enterohepatic circulation. Ethinyl estradiol is almost exclusively bound to albumin. Vaginal administration of 50 mg ethinyl estradiol results in circulating ethinyl estradiol levels equivalent to those achieved after oral ingestion of 10 mg ethinyl estradiol. Equivalent serum levels of ethinyl estradiol achieved after oral or vaginal administration result in similar effects on hepatic proteins (59). This suggests that the hepatic effects of ethinyl estradiol are more related to the chemical property of ethinyl estradiol than merely first liver pass after oral ingestion. Mestranol is a pro-drug that must be converted to ethinyl estradiol before becoming active. Q.uinestrol is an ester of
One of the most commonly prescribed estrogen preparations for postmenopausal hormone replacement are CEEs. This preparation is extracted from the urine of pregnant mares and contains classical estrogens (estrone, estradiol, and estrone sulfate) and ring B unsaturated estrogens that are not native to humans. CEE contains approximately 45% estrone sulfate, 25% equilin sulfate, 15% 17~t-dihydroequilenin sulfate, and lesser amounts of the sulfate esters of equilenin, 1713-dihydroequilin, 17~idihydroequilenin, 1713-dihydroequilenin, estradiol, 17ciestradiol, and delta-8-estrone (64,65). The pharmacokinetics of CEE is complex due to the various estrogenic components that constitute the preparation. The conversion of equilin and equilenin to 1713-dihydroequilin and 1713-dihydroequilenin by reduction of the 17-ketosteroid correspond to the conversion between estrone and estradiol. As with estrone, the major ring B unsaturated estrogens circulate in the blood as sulfate esters. Ingestion of 10 mg of CEE, containing 4.5 mg estrone sulfate and 2.5 mg equilin sulfate, results in peak levels of 560 pg/mL equilin and 1400 pg/mL estrone within 3 and 5 hours, respectively. After 24 hours, hormone levels decrease to 125 pg/mL equilin and 280 pg/mL estrone (66). Intravenous administration results in peak levels of equilin and estrone by 20 minutes because equilin sulfate and estrone sulfate are rapidly hydrolyzed to the unconjugated steroids (67). Oral ingestion of 1.25 mg of CEE results in peak levels of 120 to 180 pg/mL estrone and 40 pg/mL
772 estradiol within 6 to 10 hours. Plasma estrone and estradiol values return to baseline within 48 hours (68). A significant portion of equilin sulfate and estrone sulfate are hydrolyzed to the unconjugated steroids prior to absorption from the gut (67). After absorption the unconjugated estrogens undergo reconjugation in the liver to their sulfate, the main circulatory form of these hormones. The sulfated estrogens are also excreted in the bile and undergo enterohepatic recirculation (67,69). Equilin sulfate binds to albumin (74% of total) in a similar fashion to estrone sulfate (up to 88% of total) and does not bind to SHBG. Equilin sulfate is metabolized to its 1713metabolites. 1713-dihydroequilin, like 17[3-estradiol, is more potent than its parent compound equilin and has a binding affinity to SHBG similar to that of 1713-estradiol (67). Equilin sulfate has an M C R of 176 +_ 44 L/day/m 2 (range 93-342) and a half-life of 3.2 hours compared with an M C R of 94.1 +_ 22 L/day/m 2 (range 39-141) and a half-life of 5.3 to 9 hours for estrone sulfate (67). 1713-dihydroequilin and 1713-dihydroequilin sulfate have MCRs of 1252 + 103 L/day/m 2 and 376 + 93 L/day/m 2, respectively, and half-lives (slow component) of 45 _+ 9 minutes and 135 + 17 minutes, respectively (70). The binding affinities for human ER for various components of CEE follows the following order: 17[3-dihydroequilin > 1713-estradiol > 17~3-dihydroequilenin > estrone = equilin > 17ta-dihydroequilin > 17~-estradiol > 17pdihydroequilenin > equilenin (64,67). The pharmacokinetics of equilin sulfate is similar to that of estrone sulfate, with equilin and 17J3-dihydroequilin being analogous to estrone and 1713-estradiol. About 25% of equilin sulfate is converted to equilin which is similar to the estrone to estrone sulfate conversion rate of 15% to 20%. However, 15% of equilin sulfate is converted to 1713-dihydroequilin which is much higher than the 1% to 3% conversion rate of estrone to 1713-estradiol (64). Metabolites of equilin sulfate are excreted in the urine mainly conjugated to glucuronic acid (69). The biologic effects of CEE are therefore a combination of primarily estrone sulfate and equilin sulfate and their respective metabolites. CEE vaginal cream results in serum levels approximately one-half to one-fourth of those obtained with comparable oral dosages (Fig. 53.2) (71-73). Each gram of vaginal cream contains 0.625 mg of CEE. Intravaginal application of 1.25 mg CEE results in peak estradiol level of 33 + 6.6 pg/mL in 6 hours and peak estrone level of 73 + 9.2 pg/mL in 8 hours (73). Twenty-four hours after vaginal CEE, estradiol levels did not differ significantly from baseline and estrone levels were 50 + 8.7 pg/mL, twice the baseline level (73). Vaginal CEE (0.3 mg) is equivalent to the effect of 1.25 mg oral CEE on vaginal cytology. However, vaginal CEE (0.3 mg) results in minimal increases in estrone and estradiol levels that are lower than those found in the early follicular phase (72). Vaginal CEE (2.5 mg) is comparable to
BARNES AND LEVRANT
150 140
130
(lo)
120 110
(4)
A
E
v
_1
tu
100
i
p<0.03
90
ill .j zill
80
(14) oo
~u
8,
60
Ill z
, ~'~
ooO~~ ;~
50
I11
40
(s]~.."9"'~
O Z -
~oo~
p
a
30
20
p<0.00l
9
/~l)~'q' i~.....O Oral Estrogon 10 :~ T,'
0.300
! 9
9. . . . . 9Vaginal Estrogen All Values Means ~ SEM ( ) =N 0.625
1.250
ESTROGEN DOSE (mg/day) FmURE 53.2 Total plasma estrogen levels after 1 week of oral or vaginal conjugated equine estrogens. (From res 71, with permission.)
oral CEE (0.15-0.625 mg) in its action on hepatic markers (SHBG, thyroid-binding globulin (TBG), and renin substrate) and calcium to creatinine ratios (72). Comparable serum levels of estrone and estradiol but diminished hepatic effects have been reported for 2.5 mg vaginal CEE compared with 0.625 mg oral CEE (30). Vaginal absorption of estrogen is erratic, and in one report four of five women had higher estradiol and estrone levels after 1.25 mg of vaginal CEE than after 1.25 mg of oral CEE (68).
C. Estradiol and Estrone Oral estrone and estradiol preparations include estropipate (piperazine estrone sulfate), estradiol valerate and acetate, and micronized estradiol. All oral estrogens result in
773
CHAPTER 53 Pharmacology of Estrogens
higher serum estrone levels than estradiol levels because estradiol is oxidized by intestinal mucosa and liver 1713HSD during the first hepatic passage. Estradiol and estrone are better absorbed in their conjugated or micronized forms. Oral ingestion of 1.2 mg estropipate, 1 mg of micronized estradiol, and 0.625 mg CEE result in similar maximum serum levels of estradiol and estrone of 30 to 50 pg/mL and 150 to 300 pg/mL, respectively, after 4 to 6 hours (see Table 53.2) (74,75). 1.5 mg of oral estropipate and 2 mg estradiol valerate result in nearly superimposable levels of estradiol and estrone sulfate and slightly higher estrone values for estradiol valerate (Fig. 53.3) (76). Maximum levels of 300 pg/mL for estrone and 65 pg/mL for estradiol are achieved after 2 mg oral micronized estradiol (75,77). Pharmacokinetic studies of estropipate and estradiol valerate suggest similar half-life characteristics (78). Long-term use of 2 mg oral micronized estradiol results in estradiol levels of 114 _+ 65 pg/mL and 575 ___280 pg/mL for estrone (75,77). Estradiol and estrone levels after 4 weeks of 2.5 mg of estropipate daily were 126 pg/mL and 356 pg/mL, respectively (75,79). Serum levels achieved after chronic exposure to oral estrogens are slightly higher than those seen after acute dosing. This may be a result of increased levels of SHBG as well as sequestration of estrogens in adipose tissue. Most estrogens are readily absorbed across the vaginal epithelium. Vaginal administration of estrogen can be via a cream, saline suspension, tablet, or vaginal ring. Micronized estradiol suspended in saline results in a more rapid 11001000900A
"5 E
o..
800-
_~
700-
"~
600-
1~
5~-
=
400-
.o
g
m
o
ee%lel Iolll e~O~ O
e, 9
,,, I aJ=
o9
, , , l l ~ l l m l l , , t l l l l l eo
9 l
T
ell
300200-
Estradiol - 178
100-
I Pretreatment
I I 2 4 Hours after treatment
I 6
FIGURE 53.3 Mean serum concentrations (+ SE of mean)of estrone and estradiol before and at 2-hour intervals after oral piperazine estrone sulphate 1.5 mg (dotted line) and estradiol valerate 2 mg (solid line) in postmenopausal women. (From ref. 76, with permission.)
absorption than micronized estradiol in cream, with peak levels occurring at 2 and 4 hours, respectively (73,80). Vaginal administration of 2 mg micronized estradiol in cream and 0.5 mg micronized estradiol saline result in comparable estradiol and estrone serum levels. Vaginal administration of estradiol bypasses the intestinal-hepatic first-pass metabolism, resulting in higher serum estradiol values than estrone values. Vaginal micronized estradiol 0.5 mg in saline results in serum estradiol level of 860 pg/ mL for 3 hours, which then drops to 250 pg/mL after 6 hours. Increases in estrone are more modest: 210 pg/mL and 130 pg/mL after 3 and 6 hours, respectively (81). Vaginal administration of 0.5 mg estrone in saline results in a serum estrone level of 435 pg/mL after 6 hours and an estradiol level of 125 pg/mL (81). Micronized estradiol in tablet form is also well absorbed from the vagina. Vaginal rings consisting of crystalline estradiol mixed with a liquid polymer, polydimethylsiloxane, produce sustained and fairly constant levels of estradiol for 3 months after an initial burst effect (82,83). The 100 mg, 200 mg, and 400 mg estradiol vaginal rings produce serum levels of estradiol in the range of 40 to 50 pg/mL, 70 to 80 pg/mL, and 140 pg/mL, respectively (82). Serum estrone levels increase twofold to a peak value of 55 pg/mL with the 400-mg estradiol vaginal ring. A low-dose, estradiol-releasing intravaginal silastic ring (Estring), 7.5 ~g, is available for treatment of urogenital atrophy. After an initial burst of estradiol release, the ring releases about 8 tag of estradiol daily for 3 months. Plasma levels of estradiol and estrone increase by about one-third but remain within the normal postmenopausal range (84,85). The ring increases the vaginal maturation index and relieves urogenital atrophy but has no effect on FSH, SHBG, or endometrial thickness (85,86). A vaginal ring releasing 50 or 100 ~g of estradiol (Femring) is also available, which results in early follicular phase levels of estradiol (40-60 pg/mL). Intranasal administration of estrogen suspended in saline results in rapid dissipation and low estrogen levels (80). Estradiol suspended in dimethyl-b-cyclodextrin results in peak levels within 30 minutes with a half-life of 120 to 145 minutes (87). Nasal administration of 0.34 mg estradiol results in serum levels of approximately 408 pg/mL and 136 pg/mL after 2 and 5 hours, respectively. An open, non-randomized trial found decreased FSH and LH levels and symptomatic relief in 8 of 9 patients over 6 months (87). Estradiol may also be administered percutaneously in an alcohol-based gel. The gel is applied to the arms, shoulders, and abdomen; dries in 2 to 5 minutes; and leaves no sticky sensation or odor. The skin, particularly the stratum corneum, acts as a reservoir for continuous release of estradiol. Three milligrams of percutaneous estradiol result in an estradiol level of about 100 pg/mL within 4 to 6 hours and reaches a steady state after 3 to 5 days (54,88). Three milligrams and 1.5 mg of estradiol administered by percutaneous
774 gel daily for 14 days result in mean estradiol serum levels of 103 __+ 40 pg/mL and 68 __+27 pg/mL, respectively, compared with 114 _+ 65 pg/mL and 41 _ 14 pg/mL after 2 mg oral micronized estradiol and 50 mg/day transdermal estradiol (77). Mean serum estrone levels for percutaneous estradiol 1.5 mg and 3.0 mg are 90 pg/mL and 120 pg/mL, respectively, compared with 575 pg/mL and 43 pg/mL for 2 mg oral micronized estradiol and 50 mg/day transdermal estradiol (77). Estrogen levels may be extremely variable because absorption is proportional to the surface area of application, intensity of application in, and extent of removal by clothing (89). Another emulsion-based estradiol cream is also available (Estrasorb) which has been less extensively studied. Pellets containing crystalline estradiol, implanted underneath the skin in the subcutaneous tissue, deliver steady levels of estradiol for 6 months (90-92). Estradiol levels are highest 24 hours after insertion, reach a steady state after 1 week, and remain relatively constant for 24 weeks (92). Pellets containing 25 mg and 50 mg of estradiol result in serum estradiol levels of 50 to 70 pg/mL and 100 to 120 pg/mL, respectively (90,92). The estradiol/estrone ratio at 6 months is 1.45 and 1.59 after insertion of 25-mg and 50-mg estradiol pellets, respectively (91). Several transdermal delivery systems of estradiol are available in the United States (see Table 53.1). Estradiol is released from an alcohol gel reservoir or directly from an adhesive matrix. The patches are designed to release 0.014 to 0.10 mg of estradiol daily for 3.5 or 7 days. With transdermal delivery the estrogen reservoir is in the patch itself and not in the skin as it is with the percutaneous estradiol gel. The transdermal patch is applied to the abdomen or buttock once or twice a week. The half-life of estradiol delivered from a transdermal patch is 161 + 14 minutes in nonobese, fasting postmenopausal women (93). Maximal serum levels of estradiol are achieved within 2 to 8 hours (92,94,95). The 50-mg/day alcohol gel patch results in peak levels of 92 pg/mL of estradiol and 48 to 62 pg/mL of estrone within 2 to 8 hours after application (95). Long-term use results in mean serum estradiol and estrone levels of 33 to 62 pg/mL and 38 to 45 pg/mL, respectively (77,96,97). The 100-mg/day alcohol gel patch results in peak estradiol levels of 152 + 33 pg/mL and estradiol and estrone levels of 46 to 89 pg/mL and 32 to 64 pg/mL, respectively, with long-term use (92). The 7-day adhesive matrix patch produces similar mean estradiol levels as the 3.5-day patch at both the 0.05 and 0.10 mg doses, but with less variation in estradiol levels (94). The mean 7-day estradiol levels are approximately 50 pg/mL and 100 pg/mL for the 0.05 mg and 0.10 mg patch, respectively. The 0.014 mg E2 (ultra low dose) patch has been approved for use in older women for the prevention of osteoporosis. The increment in estradiol is only 5 pg/ml.
BARNES AND LEVRANT
IV. CLINICAL ASPECTS OF ESTROGEN THERAPY Estrogen replacement is beneficial for symptomatic relief of hot flushes and genitourinary symptoms and for prevention of osteoporosis. The lowest dosage that relieves symptoms and is effective in preventing osteoporosis should be used. The effective dosage for prevention of bone loss was thought to be 0.625 mg CEE or its equivalent (98). More recent studies have found lower doses of estrogen to be effective. In the Women's Health, Osteoporosis, Progestin, Estrogen (HOPE) trial, 0.3 mg of CEE together with 600 mg of calcium daily increased bone mineral density in the spine and hip after 24 months of therapy compared with baseline values and compared with placebo in younger, postmenopausal women. A dose response was observed in the spine but not in the hip, with a greater gain in bone mineral content in the 0.625-mg versus the 0.3-mg treatment group. Addition of medroxyprogesterone acetate (MPA) had little effect on bone mineral density (99). The HOPE study also found that 0.3 mg of CEE significantly reduced hot flushes, but to a lesser degree than 0.625 mg. However, 0.3 mg of CEE plus 1.25 mg MPA was as effective as 0.625 mg of CEE plus 2.5 mg MPA (100). In women over age 65, both CEE 0.3 mg plus MPA 2.5 mg and micronized estradiol 0.25 mg daily, together with calcium and vitamin D, significantly increased bone mineral density compared with calcium and vitamin D alone (101,102). Other data point to the prevention of bone loss in older women with the ultra low dose estradiol (0.014 mg) patch (102a). There is limited information on the interaction of diseases and drugs with estrogen in postmenopausal women receiving replacement therapy. Sex hormone binding capacity is increased in moderate smokers compared with nonsmokers (103). After ingestion of oral micronized estradiol, the change in serum estrone sulfate and estrone glucuronide is greater in postmenopausal smokers, suggesting a greater hepatic metabolism of estradiol than in nonsmokers (103). Thus, moderate smoking results in less biologically available estradiol in women taking estrogens. Postmenopausal smokers not taking estrogen replacement therapy have estrogen levels similar to nonsmokers (9,104). Acute alcohol ingestion resulted in a 20% increase in serum estradiol in women given estradiol transdermal patches (105), and the half-life of estradiol was increased by about 50% (106). In women taking oral estradiol, the serum estradiol level was tripled by acute alcohol ingestion (107), but there was no effect of alcohol on estradiol levels in women not taking estrogen. Interestingly, alcohol had no effect on estrone levels with transdermal or oral estradiol administration. In a case report, phenytoin was found to increase estrogen clearance and reduce estrogen levels (108). Postmenopausal women with end-stage renal disease had a higher baseline
CHAPTER 53 Pharmacology of Estrogens
s e r u m estradiol and a greater s e r u m estradiol response to oral estradiol a d m i n i s t r a t i o n t h a n did n o r m a l p o s t m e n o pausal controls (109).
References 1. Baird DT, Fraser IS. Blood production and ovarian secretion rates of estradiol-17 beta and estrone in women throughout the menstrual cycle.J Clin Endocrinol Metab 1974;38:1009-1017. 2. Kletzky OA, Nakamura RM, Thorneycroft IH, Mishell DR Jr. Log normal distribution of gonadotropins and ovarian steroid values in the normal menstrual cycle,d m J Obstet Gyneco11975;121:688-694. 3. Powers MS, Schenkel L, Darley PE, et al. Pharmacokinetics and pharmacodynamics of transdermal dosage forms of 17 beta-estradiol: comparison with conventional oral estrogens used for hormone replacement. Am J Obstet Gyneco11985 ;152:1099 -1106. 4. Lloyd CW, Lobotsky J, Weisz J, et al. Concentration of unconjugated estrogens, androgens and gestagens in ovarian and peripheral venous plasma of women: the normal menstrual cycle.J Clin EndocrinolMetab 1971;32:155-166. 5. Santoro N, Brown JR, Adel T, Skurnick JH. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab 1996;81:1495-1501. 6. Sherman BM, Korenman SG. Hormonal characteristics of the human menstrual cycle throughout reproductive fife. J Clin Invest 1975;55: 699-706. 7. Sherman BM, West JH, Korenman SG. The menopausal transition: analysis of LH, FSH, estradiol, and progesterone concentrations during menstrual cycles of older women. J Clin Endocrinol Metab 1976; 42:629-636. 8. ReyesH, Winter JS, Faiman C. Pituitary-ovarian relationships preceding the menopause. I. A cross-sectional study of serum follice-stimulating hormone, luteinizing hormone, prolactin, estradiol, and progesterone levels.Am J Obstet Gyneco11977;129:557-564. 9. Cauley JA, Gutai JP, Kuller LH, LeDonne D, Powell JG. The epidemiology of serum sex hormones in postmenopausal women. Am J Epidemio11989;129:1120-1131. 10. Judd HL. Hormonal dynamics associated with the menopause. Clin Obstet Gyneco11976;19:775- 788. 11. Laufer LR, DeFazio JL, Lu JK, et al. Estrogen replacement therapy by transdermal estradiol administration. Am J Obstet Gynecol 1983;146: 533-540. 12. Longcope C. Metabolic clearance and blood production rates of estrogens in postmenopausal women. Am J Obstet Gynecol 1971;111: 778-781. 13. Siiteri P, MacDonald P. Role of extraglandular estrogen in human endocrinology. Washington DC: American Physiology Society, 1973. 14. Siiteri PK. Adipose tissue as a source of hormones. Am J Clin Nutr 1987;45:277-282. 15. Hemsell DL, Grodin JM, Brenner PF, Siiteri PK, MacDonald PC. Plasma precursors of estrogen. II. Correlation of the extent of conversion of plasma androstenedione to estrone with age. J Clin Endocrinol Metab 1974;38:476-479. 16. Meldrum DR, Davidson BJ, Tataryn IV, Judd HL. Changes in circulating steroids with aging in postmenopausal women. Obstet Gynecol 1981;57:624-628. 17. Simpson ER, Misso M, Hewitt KN, et al. Estrogen m the good, the bad, and the unexpected. Endocr Rev 2005;26:322-330. 18. Misso ML, Jang C, Adams J, et al. Adipose aromatase gene expression is greater in older women and is unaffected by postmenopausal estrogen therapy. Menopause 2005;12:210-215.
775 19. Simpson ER, Ackerman GE, Smith ME, Mendelson CR. Estrogen formation in stromal cells of adipose tissue of women: induction by glucocorticosteroids. ProcNatl Acad Sci U S A 1981;78:5690-5694. 20. Judd HL, Shamonki IM, Frumar AM, Lagasse LD. Origin of serum estradiol in postmenopausal women. Obstet Gynecol 1982;59: 680-686. 21. Fishman J, Bradlow HL, Gallagher TE Oxidative metabolism ofestradiol.JBiol Chem 1960;235:3104-3107. 22. Labrie F, Luu-The V, Lin SX, Simard J, Labrie C. Role of 17 betahydroxysteroid dehydrogenases in sex steroid formation in peripheral intracrine tissues. Trends Endocrinol Metab 2000;11:421-427. 23. Penning TM. Molecular endocrinology of hydroxysteroid dehydrogenases. Endocr Rev 1997;18:281-305. 24. Pack BA, Tovar R, Booth E, Brooks SC. The cyclic relationship of estrogen sulfurylation to the nuclear receptor level in human endometrial curettings. J Clin Endocrinol Metab 1979;48:420-424. 25. Ruder HJ, Loriaux L, Lipsett MB. Estrone sulfate: production rate and metabolism in man.J Clin Invest 1972;51:1020-1033. 26. Jasonni M, Bulletti C, Franceschetti F, et al. Metabolic clearance rate of oestrone sulphate in post-menopausal women. Maturitas 1984;5: 251-257. 27. Hawkins RA, Oakey RE. Estimation of oestrone sulphate, oestradiol17beta and oestrone in peripheral plasma: concentrations during the menstrual cycle and in men. J Endocrino11974;60:3 - 17. 28. Loriaux DL, Ruder HJ, Lipsett MB. The measurement of estrone sulfate in plasma. Steroids 1971;18:463-472. 29. Wright K, Collins DC, Musey PI, Preedy JR. A specific radioimmunoassay for estrone sulfate in plasma and urine without hydrolysis. J Clin Endocrinol Metab 1978;47:1092-1098. 30. Lobo RA. Absorption and metabolic effects of different types of estrogens and progestogens. Obstet Gynecol Clin North Am 1987;14:143-167. 31. King RJ, Dyer G, Collins WP, Whitehead MI. Intracellular estradiol, estrone and estrogen receptor levels in endometria from postmenopausal women receiving estrogens and progestins. J Steroid Biochem 1980;13:377-382. 32. Bulletti C, Jasonni VM, Ciotti PM, et al. Extraction of estrogens by human perfused uterus. Effects of membrane permeability and binding by serum proteins on differential influx into endometrium and myometrium. Am J Obstet Gynecol 1988;159:509 - 515. 33. Steingold KA, Cefalu W, Pardridge W, Judd HL, Chaudhuri G. Enhanced hepatic extraction of estrogens used for replacement therapy. J Clin Endocrinol Metab 1986;62:761 - 766. 34. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA. Cloning of a novel receptor expressed in rat prostate and ovary. Proc NatlAcad Sci U S A 1996;93:5925-5930. 35. Mosselman S, Polman J, Dijkema R. ER beta: identification and characterization of a novel human estrogen receptor. FEBS Lett 1996; 392:49-53. 36. Beato M, Herrlich P, Schutz G. Steroid hormone receptors: many actors in search of a plot. Cell 1995;83:851-857. 37. Tsai MJ, O'Malley BW. Molecular mechanisms of action of steroid/ thyroid receptor superfamily members. Annu Rev Biochem 1994;63: 451-486. 38. Kuiper GG, Carlsson B, Grandien K, et al. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 1997;138:863- 870. 39. Green S, Walter P, Kumar V, et al. Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A. Nature 1986;320: 134-139. 40. Greene GL, Gilna P, Waterfield M, et al. Sequence and expression of human estrogen receptor complementary DNA. Science 1986;231: 1150-1154. 41. Enmark E, Pelto-Huikko M, Grandien K, et al. Human estrogen receptor beta-gene structure, chromosomal localization, and expression pattern. J Clin Endocrinol Metab 1997;82:4258-4265.
776 42. Pace P, Taylor J, Suntharalingam S, Coombes RC, Ali S. Human estrogen receptor beta binds DNA in a manner similar to and dimerizes with estrogen receptor alpha.J Bid Chem 1997;272:25832-25838. 43. Yager JD, Davidson NE. Estrogen carcinogenesis in breast cancer. N EnglJ Med 2006;354:270 - 282. 44. Arts J, Kuiper GG, Janssen JM, et al. Differential expression of estrogen receptors alpha and beta mRNA during differentiation of human osteoblast SV-HFO cells. Endocrinology 1997;138:5067- 5070. 45. Dotzlaw H, Leygue E, Watson PH, Murphy LC. Expression of estrogen receptor-beta in human breast tumors. J Clin Endocrinol Metab 1997;82:2371 - 2374. 46. Fishman J, Norton B. Brain catecholestrogens - - formation and possible function. ACdvBiosci 1975;15:123-131. 47. Fishman J. Role of 2-hydroxyestrone in estrogen metabolism. J Clin Endocrinol Metab 1963;23:207-210. 48. Sandberg AA, Slaunwhite WR Jr. Studies on phenolic steroids in human subjects. II. The metabolic fate and hepato-biliary-enteric circulation of C14-estrone and C14-estradiol in women. J Clin Invest 1957;36:1266-1278. 49. Wu CH, Motohashi T, Abdel-Rahman HA, Flickinger GL, Mikhail G. Free and protein-bound plasma estradiol-17 beta during the menstrual cycle.J Clin EndocrinolMetab 1976;43:436-445. 50. Rosenthal HE, Pietrzak E, Slaunwhite WR Jr, Sandberg AA. Binding of estrone sulfate in human plasma. J Clin Endocrinol Metab 1972;34: 805-813. 51. Anderson DC. Sex-hormone-binding globulin. Clin Endocrinol (Oxf) 1974;3:69-96. 52. Nestler JE. Sex hormone-binding globulin: a marker for hyperinsulinemia and/or insulin resistance? J Clin Endocrinol Metab 1993;76: 273-274. 53. Fotherby K. Intrasubject variability in the pharmacokinetics of ethynyloestradiol. J Steroid Biocbem Mol Bid 1991;38:733- 736. 54. Kuhl H. Pharmacokinetics of oestrogens and progestogens. Maturitas 1990;12:171-197. 55. Orme M, Back DJ, Ward S, Green S. The pharmacokinetics of ethynylestradiol in the presence and absence of gestodene and desogestrel. Contraception 1991;43:305-316. 56. Guengerich FP. Metabolism of 17 alpha-ethynylestradiol in humans. Life Sci 1990;47:1981-1988. 57. Mashchak CA, Lobo RA, Dozono-Takano R, et al. Comparison of pharmacodynamic properties of various estrogen formulations. Am J Obstet Gyneco11982;144:511- 518. 58. Mandel FP, Geola FL, Lu JK, et al. Biologic effects of various doses of ethinyl estradiol in postmenopausal women. Obstet Gynecol 1982;59: 673-679. 59. Goebelsmann U, Mashchak CA, Mishell DR Jr. Comparison of hepatic impact of oral and vaginal administration of ethinyl estradiol. Am J Obstet Gyneco11985;151:868-877. 60. Mitlak BH, Cohen FJ. In search of optimal long-term female hormone replacement: the potential of selective estrogen receptor modulators. Horm Res 1997;48:155-163. 61. SadovskyY, Adler S. Selective modulation of estrogen receptor action. J Clin Endocrinol Metab 1998;83:3-5. 62. Delmas PD, Bjarnason NH, Mitlak BH, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. NEnglJMed 1997;337:1641 - 1647. 63. Walsh BW, Kuller LH, Wild RA, et al. Effects of raloxifene on serum lipids and coagulation factors in healthy postmenopausal women. JAm MedAssoc 1998;279:1445-1451. 64. Bhavnani B R. Pharmacokinetics and pharmacodynamics of conjugated equine estrogens: chemistry and metabolism. Proc Soc Exp Biol Med 1998;217:6-16. 65. Bhavnani BR, Woolever CA, Benoit H, Wong T. Pharmacokinetics of equilin and equilin sulfate in normal postmenopausal women and men. J Clin Endocrinol Metab 1983;56:1048-1056.
BARNES AND LEVRANT 66. Bhavnani BR, Sarda IR, Woolever CA. Radioimmunoassay of plasma equilin and estrone in postmenopausal women after the administration of premarin. J Glin EndocrinolMetab 1981;52:741- 747. 67. Bhavnani BR. The saga of the ring B unsaturated equine estrogens. Endocr Rev 1988;9:396-416. 68. Englund DE, Johansson ED. Plasma levels of oestrone, oestradiol and gonadotrophins in postmenopausal women after oral and vaginal administration of conjugated equine oestrogens (Premarin). Br J Obstet Gynaeco11978;85:957-964. 69. Bhavnani BR, Woolever CA, Wallace D, Pan CC. Metabolism of [3H]equilin-[35S]sulfate and [3H]equilin sulfate after oral and intravenous administration in normal postmenopausal women and men. J Clin EndocrinolMetab 1989;68:757- 765. 70. Bhavnani BR, Cecutti A. Pharmacokinetics of 17 beta-dihydroequilin sulfate and 17 beta-dihydroequilin in normal postmenopausal women. J Clin EndocrinolMetab 1994;78:197-204. 71. Deutsch S, Ossowski R, Benjamin I. Comparison between degree of systemic absorption of vaginaUy and orally administered estrogens at different dose levels in postmenopausal women. Am J Obstet Gynecol 1981;139:967-968. 72. Mandel FP, Geola FL, Meldrum DR, et al. Biological effects of various doses of vaginally administered conjugated equine estrogens in postmenopausal women. J Clin EndocrinolMetab 1983;57:133-139. 73. Rigg LA, Hermann H, Yen SS. Absorption of estrogens from vaginal creams. NEnglJMed 1978;298:195-197. 74. Lobo RA, Brenner P, Mishell DR Jr. Metabolic parameters and steroid levels in postmenopausal women receiving lower doses of natural estrogen replacement. Obstet Gyneco11983;62:94-98. 75. Lobo RA, Cassidenti DL. Pharmacoldnetics of oral 17 beta-estradiol. J Reprod Med 1992;37:77-84. 76. Anderson AB, Sldovsky E, Sayers L, Steele PA, TurnbuU AC. Comparison of serum oestrogen concentrations in post-menopausal women taking oestrone sulphate and oestradiol. Br MedJ 1978;1:140-142. 77. Scott RT Jr, Ross B, Anderson C, Archer DE Pharmacokinetics of percutaneous estradiol: a crossover study using a gel and a transdermal system in comparison with oral micronized estradiol. Obstet Gynecol 1991;77:758-764. 78. Aedo AR, Landgren BM, Diczfalusy E. Pharmacokinetics and biotransformation of orally administered oestrone sulphate and oestradiol valerate in post-menopausal women. Maturitas 1990;12:333- 343. 79. Colvin PL Jr, Auerbach BJ, Koritnik DR, Hazzard WR, ApplebaumBowden D. Differential effects of oral estrone versus 17 beta-estradiol on lipoproteins in postmenopausal women. J Clin Endocrinol Metab 1990;70:1568-1573. 80. Rigg LA, Milanes B, Villanueva B, Yen SS. Efficacy of intravaginal and intranasal administration of micronized estradiol-17f3.J Clin Endocrinol Metab 1977;45:1261-1264. 81. Schiff I, Tulchinsky D, Ryan KJ. Vaginal absorption of estrone and 17[3-estradiol. Fertil Sterff 1977;28:1063 - 1066. 82. Stumpf PG. Selecting constant serum estradiol levels achieved by vaginal rings. Obstet Gyneco11986;67:91-94. 83. Stumpf PG, Maruca J, Santen RJ, Demers LM. Development of a vaginal ring for achieving physiologic levels of 17 beta-estradiol in hypoestrogenic women. J Clin Endocrinol Metab 1982;54:208-210. 84. Johnston A. Estrogens--pharmacokinetics and pharmacodynamics with special reference to vaginal administration and the new estradiol formulationmEstring. Acta Obstet Gynecol Scand Suppl 1996;163: 16-25. 85. Smith P, Heimer G, Lindskog M, Ulmsten U. Oestradiol-releasing vaginal ring for treatment of postmenopausal urogenital atrophy. Maturitas 1993;16:145-154. 86. Holmgren PA, Lindskog M, yon Schoultz B. Vaginal rings for continuous low-dose release of oestradiol in the treatment of urogenital atrophy. Maturitas 1989;11:55-63.
CHAPTER 53 Pharmacology of Estrogens 87. Hermens WA, Belder CW, Merkus JM, et al. Intranasal estradiol administration to oophorectomized women. Eur J Obstet GynecolReprod Bid 1991;40:35-41. 88. Whitehead M, Townsend P, Kitchin Y. Plasma steroid and protein hormone profiles in postmenopausal women following topical administration of oestradiol 1713. In: Mauvais-Jarvis P, ed. Percutaneous absorption of steroids. New York, Academic Press, 1980:231. 89. Jewelewicz R. New developments in topical estrogen therapy. Ferti! Steri11997;67:1-12. 90. Lobo RA, March CM, Goebelsmann U, Krauss RM, Mishell DR Jr. Subdermal estradiol pellets following hysterectomy and oophorectomy. Effect upon serum estrone, estradiol, luteinizing hormone, folliclestimulating hormone, corticosteroid binding globulin-binding capacity, testosterone-estradiol binding globulin-binding capacity, lipids, and hot flushes. Am J Obstet Gyneco11980;138:714- 719. 91. Notelovitz M, Johnston M, Smith S, Kitchens C. Metabolic and hormonal effects of 25-mg and 50-mg 17 beta-estradiol implants in surgically menopausal women. Obstet Gyneco11987;70:749- 754. 92. Stanczyk FZ, Shoupe D, Nunez V, et al. A randomized comparison of nonoral estradiol delivery in postmenopausal women. Am J Obstet Gyneco11988;159:1540-1546. 93. Ginsburg ES, Gao X, Shea BE Barbieri RL. Half-life of estradiol in postmenopausal women. Gynecol Obstet Invest 1998;45:45-48. 94. Gordon SE Clinical experience with a seven-day estradiol transdermal system for estrogen replacement therapy. Am J Obstet Gynecol 1995 ;173:998 - 1004. 95. Haas S, Walsh B, Evans S, et al. The effect of transdermal estradiol on hormone and metabolic dynamics over a six-week period. Obstet Gynecol 1988;71:671-676. 96. Ribot C, Tremollieres E Pouilles JM, Louvet JP, Peyron R. Preventive effects of transdermal administration of 17 beta-estradiol on postmenopausal bone loss: a 2-year prospective study. GynecolEndocrinol 1989;3:259-267. 97. Steingold KA, Laufer L, Chetkowski RJ, et al. Treatment of hot flashes with transdermal estradiol administration. J Clin Endocrinol Metab 1985;61:627-632. 98. Lindsay R, Hart DM, Clark DM. The minimum effective dose of estrogen for prevention of postmenopausal bone loss. Obstet Gynecol 1984;63:759-763. 99. Lindsay R, Gallagher JC, Kleerekoper M, Pickar JH. Effect of lower doses of conjugated equine estrogens with and without medroxyprogesterone acetate on bone in early postmenopausal women. JAm Med Assoc 2002;287:2668-2676.
777 100. Utian WH, Shoupe D, Bachmann G, Pinkerton JV, Pickar JH. Relief of vasomotor symptoms and vaginal atrophy with lower doses of conjugated equine estrogens and medroxyprogesterone acetate. Fertil Steri12001;75:1065-1079. 101. Prestwood KM, Kenny AM, Kleppinger A, Kulldorff M. Ultralowdose micronized 17[3-estradiol and bone density and bone metabolism in older women: a randomized controlled trial. JAm Med Assoc 2003;290:1042-1048. 102. Recker RR, Davies KM, Dowd RM, Heaney RP. The effect of lowdose continuous estrogen and progesterone therapy with calcium and vitamin D on bone in elderly women. A randomized, controlled trial. Ann Intern Med 1999;130:897-904. 102a. Ettinger B, Ensrud KE, Wallace R, et al. Effects of ultralow-dose transdermal estradiol on bone mineral density: a randomized clinical trial. Obstet Gyneco12004;104:443-451. 103. Cassidenti DL, Vijod AG, Vijod MA, Stanczyk FZ, Lobo RA. Short-term effects of smoking on the pharmacokinetic profiles of micronized estradiol in postmenopausal women. Am J Obstet Gynecol 1990;163:1953-1960. 104. Cassidenti DL, Pike MC, Vijod AG, Stanczyk FZ, Lobo RA. A reevaluation of estrogen status in postmenopausal women who smoke. A m J Obstet Gyneco11992;166:1444-1448. 105. Ginsburg ES, Walsh BW, Gao X, et al. The effect of acute ethanol ingestion on estrogen levels in postmenopausal women using transdermal estradiol. J Soc GynecolInvestig 1995;2:26-29. 106. Ginsburg ES, Walsh BW, Shea BF, et al. The effects of ethanol on the clearance of estradiol in postmenopausal women. Fertil Steri11995;63: 1227-1230. 107. Ginsburg ES, Mello NK, Mendelson JH, et al. Effects of alcohol ingestion on estrogens in postmenopausal women. J Am Med Assoc 1996;276:1747-1751. 108. Notelovitz M, Tjapkes J, Ware M. Interaction between estrogen and dilantin in a menopausal woman. N EnglJMed 1981;304:788-789. 109. Ginsburg ES, Owen WF Jr, Greenberg LM, et al. Estrogen absorption and metabolism in postmenopausal women with end-stage renal disease. J Clin Endocrinol Metab 1996;81:4414-4417.
This Page Intentionally Left Blank
~HAPTER 5 z
Structure- Functio n Relationships, Pharmacokinetics, and Potency of Orally and Parenterally Administered Progestogens FRANK Z . STANCZYK
Departmentsof Obstetrics & Gynecology, and Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
I. I N T R O D U C T I O N
progestin (1). In recent years, progestin has been often used when referring to synthetic progestogens such as norethindrone and levonorgestrel (1). This name excludes the "natural" progestogen, which is progesterone. The North American Menopause Society (NAMS) has been using this nomenclature (2). The use of different names to denote a compound with progestational activity can be confusing. It has been proposed that an international consensus be established to determine a single name for all progestational compounds. This would avoid ambiguity and would be in the best interest of communication and information retrieval. An obvious choice is the name progestogen, which would be consistent with the
A variety of progestogens are available for treatment of postmenopausal women, primarily for protecting the endometrium against hyperplasia during estrogen therapy. Progestogens are compounds that exhibit progestational activity. The most widely recognized progestational activity is transformation of proliferative to secretory endometrium in estrogenprimed uteri. Progestogens act primarily by activating the progesterone receptor. The term progestogen has been used interchangeably with other terms that include progestagen, gestogen, gestagen, and T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
779
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
780
FRANK Z. STANCZYK
name for a compound with estrogenic or androgenic activity (e.g., estrogen or androgen, respectively). In the present chapter, the NAMS nomenclature will be used when referring to progestational compounds. There is a concern that some progestogens cause adverse metabolic effects, including alterations in lipoprotein fractions and carbohydrate intolerance. The choice of progestogen, the dose, and the number of days of administration appear to be important. It is, therefore, helpful to understand the chemical structures of progestogens, what happens to them after they are administered, and their biologic activity. The objective of the present chapter is to discuss progestogens with respect to their classification, structure-function relationships, pharmacokinetics, and potency. Classification of progestogens separates these compounds on the basis of their chemical structures and enables us to have a clearer picture of which types are available for therapeutic use. Knowing the chemical structure of a progestogen allows us to understand better how modification of its structure affects its function. Knowledge about the pharmacokinetics of progestogens informs us about which of them are pro-drugs and provides us with information about two clinically relevant pharmacokinetics parameters, namely, their bioavailability and half-life. Finally, progestogen potency is poorly understood, and some of the misconceptions related to this topic will be addressed.
TABLE 54.1
Classification of Progestogens
Natural: progesterone Synthetic Structurally related to progesterone Pregnane derivatives Acetylated: medroxyprogesterone acetate, megestrol acetate, cyproterone acetate, chlormadinone acetate, medrogestone Nonacetylated: dydrogesterone, medrogestone 19-Norpregnane derivatives Acetylated: nomegestrol acetate, nesterone Nonacetylated: demegestone, promegestone, trimegestone Structurally related to testosterone Ethinylated Estranes: norethindrone, norethindrone acetate, ethynodiol diacetate, tibolone, norethynodrel, lynestrenol, 13-Ethylgonanes: levonorgestrel, desogestrel, norgestimate, gestodene Non-ethinylated: dienogest, drospirenone
II. CLASSIFICATION OF P R O G E S T O G E N S Table 54.1 shows a classification scheme for progestogens. Progestogens can be dMded into two types: natural and synthetic (3-6). There is only one natural progestogen, and that is progesterone; its chemical structure is shown in Fig. 54.1. The term natural, as used here, denotes that the compound is found in a living organism. It should be realized that progesterone is found in humans and certain animals, but not in plants. However, progesterone can be synthesized from plantderived sterols, such as diosgenin, by multiple chemical reactions (Fig. 54.2). A syntbetic progestogen (progestin) is defined
FIGURE 54.2
FIGURE54.1 The natural progestogen,progesterone. here as a progestational compound that is not found in living organisms. Diosgenin is also an important starting compound for the chemical synthesis of a variety of progestins. Because a substantial number of progestins are available for therapeutic use, it is sometimes convenient to
Conversionof diosgenin to progesterone.
CHAPTER 54 Structure-Function Relationships, Pharmacokinetics, and Potency of Progestogens classify them on the basis of their chemical structures into two groups: (a) those related to progesterone and (b) those related to testosterone (3-6). This classification scheme, however, does not necessarily imply the source of the steroid precursor used to synthesize the progestin. For example, the progestin levonorgestrel is synthesized chemically from estrone by multiple chemical reactions. On the other hand, norethindrone is synthesized from diosgenin. Progestins structurally related to progesterone can be subdivided into those with and without a methyl group at carbon 10 (i.e., pregnane and 19-norpregnane derivatives). These derivatives can be subdivided further into those that are acetylated and those that are not. In contrast to the progestins chemically related to progesterone, those structurally related to testosterone can be subdivided into those that contain an ethinyl group at carbon 17 versus those that do not (non-ethinylated). The ethinylated derivatives can be subdivided further into those that have an estrane structure and those with an 18-ethylgonane structure.
781
structure from MPA only in that it has a double bond between carbons 6 and 7. Chlormadinone acetate and cyproterone acetate also have a double bond between carbons 6 and 7, and in addition they both have a chloral group substituted for the methyl group at carbon 6. Cyproterone acetate differs from chlormadinone acetate in that it has a methylene group attached to carbons 1 and 2. A unique progestin that has a methyl group at carbon 10 but is not acetylated is dydrogesterone. This progestin belongs to a group of compounds called retroprogesterones. Their unique feature is that the methyl group at carbon 10 is present in the oL-orientation instead of the [3-orientation, as in progesterone and the pregnane subclass of progestins. Dydrogesterone also has a double bond between carbons 6 and 7. Compared with progesterone, this progestin apparently does not inhibit ovulation when given throughout the menstrual cycle and does not alter the basal body temperature. Thus, the change in spatial orientation of the methyl group at carbon 10 may result in altered peripheral and central effects. Another compound in the non-acetylated group of progestins that contain a methyl group at carbon 10 is medrogestone.
III. STRUCTURE-FUNCTION RELATIONSHIPS OF PROGESTOGENS It is often useful to consider how a change in the chemical structure of a progestogen relates to a change in its biologic activity. This is especially important when considering potency of progestogens, which will be discussed later in this chapter. Figs. 54.3 to 54.7 show the chemical structures of the progestins classified in Table 54.1.
A. Progestins Structurally Related to Progesterone l. PREGNANEDERIVATIVES
As stated earlier, the two reference compounds in the classification of synthetic progestins are progesterone and testosterone. If we begin with the progesterone molecule and add a hydroxyl group at carbon 17, the altered progesterone molecule loses its progestational activity. However, if we acetylate the hydroxyl group, the new molecule acquires some progestational activity and some oral activity. If we go one step further and add a methyl group at carbon 6, the altered molecule, medroxyprogesterone acetate (MPA), acquires relatively high progestational activity and oral activity. Chemical manipulation of the MPA molecule by addition of a double bond between carbons 6 and 7 and/or substitution of the methyl group at carbon 6 with a chloro substituent gives rise to three highly potent progestins that are more potent than MPA: megestrol acetate, chlormadinone acetate, and cyproterone acetate. Megestrol acetate differs in chemical
FIGURE 54.3 Progestinsstructurally related to progesterone: pregnane
derivatives-- acetylated and non-acetylated.
782
FRANK Z. STANCZYK
FIGURE 54.4 Progestins structurally related to progesterone: 19-norpregnane derivatives uacetylated and non-acetylated.
FIGURE 54.5
Progestins structurally related to testosterone: ethinylated derivatives - - estranes.
CHAPTER 54 Structure-Function Relationships, Pharmacoldnetics, and Potency of Progestogens
FIGURE 54.6
783
Progestins structurally related to testosterone: ethinylated derivatives m 13-ethylgonanes.
This progestin differs from progesterone in that it contains a methyl group at carbons 6 and 17 and a double bond between carbons 6 and 7.
B. Progestins Structurally Related to T e s t o s t e r o n e 1. ETHINYLATED DERIVATIVES
2. 19-NORPREGNANE DERIVATIVES
The norpregnane derivatives include nomogestrol acetate, demegestone, promegestone, trimegestone, and nesterone. All lack a methyl group at carbon 10. Except for the absence of this group, the chemical structure of nomogestrel acetate is identical to that of megestrol acetate. Nesterone differs from nomogestrol acetate in that it has a methylene group at carbon 16 and lacks the methyl group at carbon 6 and the double bond between carbons 6 and 7. In contrast, the structures of demegestone, promegestone, and trimegestone all have a double bond between carbons 9 and 10 and a methyl group substituted for the acetate group at carbon 17. Promegestone and trimegestone have an additional methyl group on the two-carbon side chain at carbon 17; the penultimate carbon is hydroxylated in the structure of trimegestone.
FIGURE 54.7 Progestins structurally related to testosterone: nonethinylated derivatives.
Using the testosterone molecule as the starting point, we can show how manipulation of its chemical structure alters its biologic activity dramatically. Addition of an ethinyl group to the molecule causes the steroid to lose androgenicity substantially and to acquire both progestational properties and oral activity. The new compound is 17~-ethinyltestosterone, which was given the common name etbisterone. Removal of the methyl group at carbon 10 of ethisterone further increases the progestogenic and oral activities of the molecule and virtually eliminates its androgenicity. The resulting product is norethindrone (the U.S. name), which is also called norethisterone (the name used in Europe and other countries). Substitution of an ethyl group for the methyl group at carbon 13 of norethindrone yields norgestrel, which consists of a racemic mixture of D-(-)-norgestrel (levonorgestrel) and L-(+)-norgestrel (dextronorgestrel) when norgestrel is synthesized (7). Levonorgestrel is the biologically active form of norgestrel and is one of the most potent orally active progestins. Progestins structurally related to norethindrone and levonorgestrel have been synthesized. The two groups are sometimes referred to as the norethindrone and levonorgestrelfamilies. The norethindrone family of progestins is often referred to as estranes because they all have the same 18-carbon steroid nucleus as the parent steroid, estrane. On the other hand, the levonorgestrel family of progestins is sometimes referred to as gonanes. The latter name, however, is not appropriate because all steroids by definition are gonanes since they contain the 4-ring carbon nucleus (gonane). A more appropriate name for these progestins is 13-ethylgonanes (8).
784 In addition to norethindrone, the norethindrone family of progestins also includes norethindrone acetate, ethynodiol diacetate, norethynodrel, and lynestrenol. Norethindrone acetate and ethynodiol diacetate differ from the parent compound by having an acetate group at carbon 3, and at carbons 3 and 17, respectively. Except for a shift in the double bond between carbons 4 and 5 of norethindrone to a double bond between carbons 5 and 10 in the norethynodrel molecule, both of these progestins have identical structures. Lynestrenol differs from norethindrone in structure by the absence of an oxygenated functional group at carbon 3. Another progestin that can be classified in the estrane group is tibolone. Tibolone differs structurally from norethindrone only in that it has a double bond between carbons 5 and 10, instead of 4 and 5, and a methyl group at carbon 7. Although tibolone is not classified in the norethindrone family of progestins, its chemical structure can be viewed as a derivative of norethindrone. In addition to levonorgestrel, other progestins in the levonorgestrel family (13-ethylgonanes) include desogestrel, norgestimate, and gestodene. These compounds are often referred to as the "new" progestins because they were marketed more recently compared with levonorgestrel, norethindrone, and progestins structurally related to norethindrone. In comparison to the chemical structure of levonorgestrel, desogestrel contains no oxygenated functional group at carbon 3 and has a methylene group at carbon 11, whereas norgestimate has an oxime group at carbon 3 and an acetate group at carbon 17. In contrast, gestodene differs from levonorgestrel only in that it has a double bond between carbons 15 and 16.
2. NoN-ETHINYLATEB DERIVATIVES
Two compounds can be included in the non-ethinylated subgroup of progestins; namely, dienogest and drospirenone. When compared with the chemical structure of norethindrone, dienogest contains a cyanomethyl group instead of an ethinyl group at carbon 17, and a double bond between carbons 10 and 11. Drospirenone has the basic chemical structure of the parent compound, androstane (19 carbons) and is an analog of spironolactone. It contains one methylene group attached to carbons 6 and 7 and another attached to carbons 15 and 16. In addition, a carbolactone group is present at carbon 17.
IV. PHARMACOKINETICS OF ORALLY ADMINISTERED PROGESToGENS Considering the wide use of progestogens for postmenopausal hormone therapy and steroidal contraception, surprisingly, there is generally little information about the pharmacokinetics of most orally administered progesto-
FRANKZ. STANCZYK
gens. A variety of pharmacokinetic parameters of a progestogen can be calculated from its serum or plasma levels obtained at frequent intervals after oral administration of the progestogen. The present discussion will focus on the pharmacokinetics of orally administered progestogens and on the following pharmacokinetic parameters: maximum concentration (Cmax), the time to reach Cmax (Tmax), bioavailability, and half-life. These parameters are determined by quantifying the progestogens in serum or plasma obtained at frequent intervals over a 24-hour period after dosing. Cmax and Tmax are obtained from the highest concentration of the drug. Bioavailability and half-life are two clinically relevant pharmacokinetic parameters. Bioavailability is defined as the extent to which an administered drug reaches the systemic circulation after undergoing hepatic first-pass metabolism. It is usually estimated by comparing the areas under the serum or plasma drug concentration-time curves (AUCs) obtained following oral and intravenous administration of a given dose of the drug. Half-life is the time required for a drug's blood level to fall to 50% of its maximal value. Table 54.2 contains a summary of the bioavailabilities and half-lives of different progestogens. There is large intrasubject variability and, to a lesser extent, intrasub]ect variability in circulating levels and pharmacokinetic parameters of all the progestogens that are administered orally. A fivefold difference between two women in serum levels of a progestogen, following oral administration of the same type and dose of progestogen, is not uncommon. It is also important to realize that the pharmacokinetics of progestogens in elderly women (>65 years) is altered compared with adult women and younger postmenopausal women. Elderly women have increased oral bioavailability of drugs due to decreased hepatic cytochrome P450 content; this results in decreased hepatic first-pass metabolism. They also have a more extensive volume of distribution of lipidsoluble drugs and a less extensive volume of distribution of water-soluble drugs. Perhaps the most important cause for altered pharmacokinetics of drugs in the elderly is the decline in renal clearance of a drug.
A. Progesterone Crystalline progesterone is poorly absorbed. However, when it is broken down to minute particles by the process of micronization, its absorption is improved substantially. The micronization process gives rise to a greater surface area of the compound, allowing it to be dissolved in the aqueous medium of the intestine. Increased surface area of a water-insoluble compound such as progesterone can also be achieved by dispersing the compound in watersoluble carriers such as sterols. For example, oral administration of a mixture of progesterone with cholesterol
CHAPTER
785
54 Structure-Function Relationships, Pharmacoldnetics, and Potency of Progestogens TABLE 54.2
Average Bioavailabilities and Half-Lives of Progestogens
Progestogen
Dose ( m g )
Progesterone Medroxyprogesterone acetate Megestrol acetate Cyproterone acetate Chlormadinone acetate Medrogestone Dydrogesterone Trimegestone Norethindrone Levonorgestrel Desogestrel Gestodene Dienogest Drospirenone Nomegestrol acetate
100, 200, 300 10 160 2 2 5 10 0.5 1 0.15-0.25 0.15 0.075 4 3 5
Bioavailability (%)
Half-Life (Hrs)
--* m
16.2-18.3 24 22.3 54.0-78.6 80.1 34.9 14-17 15 8 9.9, 13.2 11.9,23.8 12-14 10.8, 11.6 31.1-32.5 40
~ 100 m ~-100 64 89, 99 62, 76 87, 99 96.2 66 63
*Dashed line indicates that no data were available. pivolate on a lactose carrier increases plasma progesterone levels. Although the processes just described allow progesterone to be readily absorbed, its bioavailability is also limited by the extensive hepatic first-pass metabolism that it undergoes (9). This is due to the presence of a double bond and two ketone groups on the progesterone molecule, making it highly vulnerable to enzymatic reduction by reductases and hydroxysteroid dehydrogenases. These transformations occur in the small intestine by enzymes in bacterial flora and intestinal mucosa, and to a greater extent by hepatic enzymes. The bioavailability of orally administered micronized progesterone is generally considered to be <10%. The extensive intestinal and hepatic metabolism of progesterone requires administration of large doses of this compound compared with other progestins when used therapeutically. Dose proportionality of micronized progesterone has been evaluated in a three-way crossover design study in which doses of 100, 200, or 300 mg of micronized progesterone were ingested daily for 5 days by healthy, fasting, postmenopausal women (Fig. 54.8) (10). The Cmax and AUg0_24h values increased proportionately with increasing dose. Mean Cmaxvalues were 6.5, 13.8, and 32.3 ng/mL, and Tmax values ranged between 2.0 and 2.7 hours. These values meet or exceed the mean levels found in the midluteal phase of a spontaneous menstrual cycle, indicating that the doses administered in that study would be expected to achieve therapeutic circulating progesterone levels. However, after peaking, serum progesterone levels fell precipitously during the first 6 hours. After 6 hours post-treatment, mean progesterone levels were approximately 2, 4.5, and 8 ng/mL, and after 10 hours they were only 1.5, 2.5 and 3.5 ng/mL, for the 100, 200, and 300 mg doses, respectively. At 24 hours, the levels ranged from <1 to <2 ng/mL. The half-lives of elimination were estimated and shown to be 18.3, 16.8, and 16.2 hours, respectively. In the same study, it was shown that on the basis of AUG0_24h the absorption of
progesterone was enhanced twofold and Cmax approximately fivefold, with essentially no change in Tmax, when micronized progesterone was administered with food. Because oral micronized progesterone is metabolized so extensively, it was initially suggested that twice-daily doses were necessary to stabilize serum progesterone levels for endometrium protection (11). However, more recently a dose of 200 to 300 mg has been recommended when administered sequentially 10 to 14 days per month during estrogen therapy (12).
B. Medroxyprogesterone Acetate Considering that it has been used widely throughout the world by many women for many years, it is surprising that so little is known about the pharmacokinetics of MPA.
"~E 100 1 Ig
~
iJJ z 0a: LU
t 10
I-. UJ 0 :E
:::) i,tJ
K/)
o.1 0
4
o
12
16
20
24
TIME (hrs)
54.8 Serum progesterone concentrations (mean + standard error) prior to and followingoral administrationof micronizedprogesterone to postmenopausalwomen,usingthree different doses: 100 mg, 200 mg, and 300 mg. (From ref. 6, with permission.) FIGURE
786
FRANK Z. STANCZYK
Doses of 2.5 or 5 mg of MPA are used continuously combined with conjugated equine estrogens (CEEs), and 2.5, 5, and 10 mg doses are used sequentially with different estrogens. W h e n a single 10-mg tablet containing MPA was administered orally to each of three premenopausal women, mean peak levels ranged from 3 to 5 ng/mL and were attained between 1 and 4 hours (Fig. 54.9) (13). The levels then fell precipitously until 6 hours following dosing and then declined gradually until 24 hours. At this time the levels were in the range of 0.3 to 0.6 ng/mL. MPA levels were also measured in a triple-crossover design study in which four different drugs were administered orally to three
., :t "
C. Norethindrone
h
,i"
,,,
m
IItL1
qt
li' m
k,\
I-i t 1
o
to six postmenopausal women on three different occasions (14). Similar findings to those just described with the 10-mg dose of M P A were obtained after a tablet containing the same dose of M P A combined with 2 mg of estradiol valerate was administered. Cmax and 24-hour medroxyprogesterone values were a little lower when a tablet containing 5 mg of medroxyprogesterone combined with 2 mg of estradiol valerate was ingested. In the circulation, MPA is bound nonspecifically to albumin and undergoes extensive metabolism by A-ring reduction, hydroxylation (primarily at carbons 6 and 20), and conjugation (primarily glucuronidation) (15). The half-life of MPA following a 10-mg oral dose has been reported to be 24 hours (16).
X
",,i
o,
....... T
~4 ....
\
,Xi
!
.
\i
!
N
~, .............. r
#,
.......
Plasma norethindrone levels versus time profiles for doses of 1000, 500, and 300 pg of norethindrone administered orally to normal cycling women are shown in Fig. 54.10 (17,18). The highest dose of norethindrone was administered in combination with 120 pg of ethinylestradiol, whereas the other two doses were given without estrogen. A dose response was obtained with all three formulations. Mean peak norethindrone levels of approximately 16, 6, and 4 ng/mL were attained with the 1000-, 500-, and 300-pg doses of norethindrone, respectively, within I to 2 hours following the dosing. Thereafter, all three levels fell precipitously at first and then declined gradually until 24 hours. At that time, the mean levels were approximately 0.5 ng/mL or less. The AUCs calculated for the three different norethindrone doses were shown to be proportional to the administered doses.
m
FIO.I
14"
r
:]
|
4
?
,0 a
'
i;
rA
iX,
1"4
? .............
0
,
"
~,
24
".,j "
I
i:
'"
',
! " ,i,, !
--
411
I2-
, \ | \
Norelhindrone dose:
,
'X
!
10-
'!" ,
"f
.,!,
"'
9 1.0 mo (n-24) 0 05 rng (n,,5) " 0.3 mo (n-5)
,,
^,
8"
)
4,
72
.................
I
FIO.I
FIGURE 54.9 Serum concentrations of medroxyprogesterone acetate (MPA) prior to and following oral administration of a 10-mg dose of MPA to 3 premenopausalwomen. Frequentblood sampfingwas obtained on day 1 of treatment, but on subsequent days,blood sampleswere obtained at 2 and 24 hours after dosing. (From ref. 6.)
20
O" "O. " ~ ~ ' ~ " - ' - - - - ~ _ . _ . . ~
9 . 3' .
9~
9 9 r9 "--
12' '
" 1'5"
" 18"-" ' 21"
" 24
Hours after administration
FIGURE 54.10 Plasmanorethindrone level-time curves for three different doses of norethindrone administered orally to premenopausal women. (From ref. 6.)
787
CHAPTER 54 Structure-Function Relationships, Pharmacokinetics, and Potency of Progestogens Most of the information about the pharmacokinetics of norethindrone is based on a study in which I mg of norethindrone acetate in combination with 50 lag of ethinylestradiol was administered as a single dose orally and intravenously to a group of 6 women (19). The results show that the absolute bioavailability ranged from 47% to 73%, with a mean standard deviation of 64 +_ 10%. In the same study it was also shown that after the intravenous dose the values (mean ___ standard deviation) for the clearance, apparent volume of distribution, and half-life of elimination were, respectively, 355 _+ 68 mL/ hr/kg (range, 260 to 422 mL/hr/kg), 3.6 _+ 2.0 L/kg (range, 2.09 to 6.90 L/kg), and 8.0 _+ 3.3 hours (range, 5.2 to 12.8 hours). The values for these parameters were not significantly different from those obtained after the oral dosing. It has been shown that following the administration of norethindrone to 10 women, more than 95% of circulating norethindrone is bound to sex hormone-binding globulin (SHBG) and albumin (20). The following distribution of norethindrone in serum was found: SHBG-bound, 35.5 _+ 13.6%; albumin-bound, 60.8 ___ 12.9%; unbound, 3.7 ___ 0.9% (mean _+ standard deviation). Norethindrone undergoes extensive ring A reduction, forming dihydro- and tetrahydro-norethindrone metabolites, which undergo conjugation (7). It also appears to be aromatizable. Low serum ethinylestradiol levels have been measured in postmenopausal women, following oral administration of relatively large doses of norethindrone acetate or norethindrone (21). On the basis of AUCs determined for ethinylestradiol and norethindrone, it was shown that the mean conversion ratio of norethindrone to ethinylestradiol was 0.7% and 1.0% at doses of 5 and 10 rag, respectively. The authors calculated that this corresponds to an oral dose equivalent of about 6 txg of ethinylestradiol per milligram of norethindrone acetate. Similarly, the authors showed that a dose of 5 mg of orally administered norethindrone was equivalent to about 4 Ixg of ethinylestradiol per milligram of norethindrone. On the basis of these calculations, it was estimated that lower doses of norethindrone or its acetate (e.g., 0.5 to 1.0 mg) used in combination with ethinylestradiol would add between 2 and 10 Ixg of ethinylestradiol to the existing ethinylestradiol. It should be realized that the estimations for these lower doses were extrapolated from high doses of norethindrone and its acetate, which were not combined with ethinylestradiol. Nevertheless, the data show that significant amounts of ethinylestradiol were formed from norethindrone, and the amounts formed appear to be highly variable. Conversion of norethindrone acetate to ethinylestradiol was confirmed very recently in premenopausal women who were treated with 10, 20, or 40 mg of the progestin for 7 days in the early follicular phase of the menstrual cycle (22). On the basis of mean serum norethindrone levels of 5.5 ng/mL at 24 hours following oral administration of 10 mg of norethindrone acetate, the approximate conversion of this progestin to ethinylestradiol was calculated to be 0.2%.
D. P r o g e s t i n s S t r u c t u r a l l y R e l a t e d to N o r e t h i n d r o n e It is generally considered that the progestins structurally related to norethindrone are pro-drugs and that their progestational activity is due to norethindrone. After oral administration, norethindrone acetate and ethynodiol diacetare are rapidly converted to the parent compound by esterases during hepatic first-pass metabolism. Mthough less is known about the transformation of lynestrenol and norethynodrel (7), it appears that lynestrenol first undergoes hydroxylation at carbon 3 and then oxidation of the hydroxyl group, forming norethindrone. The possibility that some norethynodrel is metabolized by pathways not involving norethindrone as an intermediate has not been excluded, but there is no evidence in its favor. Thus, it generally appears that the pharmacokinetics of progestins structurally related to norethindrone is determined by the pharmacokinetics of norethindrone.
E. L e v o n o r g e s t r e l As depicted in Fig. 54.11, dose-response curves are obtained when doses of 250, 150, and 75 lag oflevonorgestrel are administered orally to normal cycling women (23-25). The 250-lag dose was administered in the form of the racemic mixture of norgestrel, that is, DL-(+)-norgestrel (500 lag). Also, all three levonorgestrel doses were given in combination with 30 to 50 lag of ethinyl-estradiol. Mean peak levels of approximately 6.0, 3.5, and 2.5 ng/mL were attained at 1 to 3 hours with the 250-, 150-, and 75-lag doses, respectively. At 24 hours, the mean levonorgestrel level was 1 to 2 ng/mL
Levonorgestreldose: 90.25 rng:(n,,~) 0 0.15 rng(n.24) 90 075 mo (n--,?4)
,,~ = 7 ~
i
-i %,--I._
o, \
O-o..~ "m-l----u_ 6
12
~q~-....,._ .
c:)
18 24 30 HoursafterMministration
-'m~~ll 38
42
48
FIGURE 54.11 Plasmalevonorgestrellevel-time curves for three differ-
ent doses of levonorgestrel administered orally to premenopausalwomen. (From ref. 6.)
788
FRANK Z. STANCZYK
with the highest dose and less than 0.5 ng/mL with the other two doses. Pharmacokinetic data have been obtained for the 150and 250-lag doses oflevonorgestrel (26,27) but are sparse for the 75-lag dose. The mean absolute bioavailability of levonorgestrel has been generally accepted to be virtually 100%. However, it should be realized that this conclusion is based on only two studies. In one of the studies (26), absolute bioavailabilities were determined for the 250- and 150-lag doses of levonorgestrel, each of which was administered in combination with ethinylestradiol to only five women. The absolute bioavailability for the 250-lag dose of levonorgestrel ranged from 72% to 125% (mean + standard deviation, 99 + 20%), and for the 150-lag dose the range was 65% to 108% (mean _+ standard deviation, 89 + 13%). Three of the five subjects who received the 250-lag dose of levonorgestrel had absolute bioavailabilities substantially greater than 100%. It is not clear why the AUCs in those subjects were greater by the oral route compared with the parenteral route, unless there were methodologic problems in the study. Nevertheless, it is incorrect to determine a mean absolute bioavailability using those values. In a second study (27), the absolute bioavailability of levonorgestrel was investigated using a 30-lag dose of the drug in three women; values of 80%, 83%, and 97% were obtained. On the basis of the very limited data on the bioavailability of levonorgestrel, it appears that, in general, this drug is not subject to an appreciable first-pass effect. However, a significant number of women receiving the 150- or 250-lag dose of levonorgestrel may undergo a substantial first-pass effect (absolute bioavailability <75%). Other pharmacokinetic parameters oflevonorgestrel have been calculated following either intravenous or oral dosing. In the same study in which the absolute bioavailabilities of levonorgestrel were determined for the 150- and 250-lag doses (26), the clearance, apparent volume of distribution, and half-life of elimination were found to be 105 + 36 and 113 + 31 mL/hr/kg, 1.9 _+ 0.7 and 1.6 + 0.7 L/kg, and 13.2 + 6.0 and 9.9 + 0.7 hours (mean + standard deviation), for the two doses, respectively, when administered intravenously. The half-life of elimination following oral dosing was similar to the values obtained after intravenous dosing. The following distribution of levonorgestrel in serum has been reported: SHBG-bound, 47.5 _+ 11.7%; albumin-bound, 50.0 ___ 11.0%; unbound, 2.5 +_ 0.7% (mean _+ standard deviation) (20).
F. N o r g e s t i m a t e Very little is known about the pharmacokinetics of orally administered norgestimate. However, it is known that norgestimate is a relatively complex pro-drug. After its oral
administration, the acetate group at carbon 17 is rapidly removed during hepatic first-pass metabolism. The product formed, levonorgestrel-3-oxime, has progestational activity. It has also been referred to as deacetylated norgestimate, and more recently it has been assigned the common name nordgestromin by the pharmaceutical company that markets this progestin. Rapid formation of norelgestromin from norgestimate was demonstrated in a study in which serum levels of norelgestromin were measured after single and multiple oral dose administration of 360 Ixg of norgestimate combined with 70 lxg ethinylestradiol to 10 women (28). Mean peak serum levels of norelgestromin were approximately 4 ng/mL and were attained after about 1 hour; the levels remained elevated as long as 36 hours after treatment. In contrast, peak levels of norgestimate were only about 0.1 ng/mL. It has also been shown that norgestimate is converted to levonorgestrel. In a randomized, comparative pharmacokinetic study, 12 women received single oral doses of 250 Ixg of norgestimate combined with 35 Ixg ethinylestradiol and 250 Ixg of levonorgestrel combined with 50 txg of ethinylestradiol (29). The levonorgestrel AUC ratios were determined after administration of both formulations and were used to calculate the bioavailability of norgestimate-derived levonorgestrel. The results showed that, on the average, about 22% of the dose of administered norgestimate became systemically available as levonorgestrel. More recently, it was demonstrated that substantial amounts (~ 2.5 ng/mL) of serum levonorgestrel levels are found in premenopausal women, following administration of 250 lag of norgestimate combined with ethinylestradiol (35 lag) orally or 150 tag/day of norelgestromin combined with ethinylestradiol (20 lag/day) transdermally (30). In addition to norelgestromin and levonorgestrel, a third progestationally active metabolite of orally administered norgestimate, namely levonorgestrel-17-acetate, is formed. However, it is barely detectable in serum (Wyeth-Ayerst Laboratories, personal communication).
G. D e s o g e s t r e l It is well established that desogestrel is a pro-drug and that its progestational action is mediated through one of its metabolites, namely 3-ketodesogestrel. This metabolite has recently been named etonogestrel by the pharmaceutical company that markets this progestin. Initial evidence for this finding came from a study in which a single 2.5-mg dose of desogestrel was administered to one woman and a peak circulating etonogestrel level of 12.7 ng/mL was attained within 1.5 hours after treatment (31). In contrast, the peak level of desogestrel was only about 0.7 ng/mL. This finding was supported by data showing that 3H-etonogestrel was a major metabolite resulting from incubation of human liver homogenates with 3H-desogestrel (32). Further evidence that
CHAPTER 54 Structure-Function Relationships, Pharmacokinetics, and Potency of Progestogens
desogestrel acts via etonogestrel came from a study in which 10 women received 150 lag of desogestrel in combination with 30 lag of ethinylestradiol and another 10 women received 150 lag of etonogestrel combined with 30 lag of ethinylestradiol (33,34). Each combination was ingested as a single dose, and serum samples were obtained at frequent intervals over a 24-hour period. The results show that desogestrel was undetectable following treatment. In contrast, etonogestrel levels rose and fell, and its mean levels were virtually superimposeable between the two groups of women, even though the intersubject variability in the levels was very high. The bioavailability of desogestrel was determined in a crossover study in which nine women received an oral dose of 150 lag of desogestrel in combination with 30 lag of ethinylestradiol and an intravenous dose of 150 lag etonogestrel combined with 30 lag of ethinylestradiol (35). The absolute bioavailability of etonogestrel was 76 __ 22% (mean _+ standard deviation). In a subsequent study (36), the absolute bioavailability of the same progestin was reported to be 62 _+ 7% (mean __ standard deviation). In the former study (35), the clearance of etonogestrel following its administration intravenously and desogestrel orally was 8.7 _+ 2.9 and 12.1 _+ 4.7 L/hr, respectively. Although the apparent volume of distribution of etonogestrel was not reported in that study, the data were utilized by others to calculate this parameter; a value of 3.0 __ 1.3 L/kg has been reported (37). In another study, the pharmacokinetics of 150 lag of desogestrel administered orally in combination with 30 lag of ethinylestradiol was investigated in 25 women (38). A mean Cm~x of 3.69 __ 0.97 ng/mL and Tm~x of 1.6 _+ 0.97 hours (mean _+ standard deviation) were reported for etonogestrel. In the same study, a reliable estimate of the elimination half-life of etonogestrel was obtained by following its serum levels for up to 72 hours; the mean elimination halflife was 23.8 _+ 5.3 hours. This value is approximately twice as high as that reported previously in a study in which blood samples were collected only up to 24 hours after oral or intravenous dosing (35). In most pharmacokinetic studies, blood sampling is carried out only for 24 hours, which does not provide an accurate half-life value. Serum levels of etonogestrel have also been quantified following multiple dosing with desogestrel (39). A dose of 150 lag of desogestrel in combination with 30 lag of ethinylestradiol was administered to 11 women during 12 continuous treatment cycles. Blood sampling was carried out at frequent intervals on days 1, 10, and 21 in cycles 1, 3, 6, and 12. This experimental design was part of a study (40) designed to compare circulating levels of ethinylestradiol following administration of ethinylestradiol with either desogestrel or gestodene. Comparison of the mean serum etonogestrel levels measured in the samples on days 1, 10, and 21 showed that the levels were relatively low on day 1 of treatment but rose progressively and were higher on day 21 of treatment in all study cycles, except cycle 12.
789
Increases in etonogestrel levels have been attributed to elevated serum levels of SHBG induced by the estrogenic component of the pill. In the study described in the previous paragraph (39), mean SHBG levels rose dramatically (131%) between days I and 10 of the first cycle but increased by only 10% between days 10 and 21 of the same cycle. On day 1 of subsequent cycles, mean SHBG levels were 84% to 115% higher than the SHBG level on day 1 of the first cycle. Also, in the subsequent cycles, the rise in mean SHBG levels between days 1 and 10 was considerably lower (25% to 54%) than the rise observed in the first cycle, and the rise in mean SHBG levels between days 10 and 21 did not exceed 20%. Mean SHBG levels on day 21 of subsequent cycles were 2.5- to 3.3-fold greater than the level on day 1 of the first cycle. The following distribution of etonogestrel in serum has been reported: SHBG-bound, 31.6 + 12.0%; albumin-bound, 65.9 + 11.9%, unbound, 2.5 _+ 0.2% (mean _+ standard deviation) (41).
H. Gestodene It has been shown that after oral administration of 14C-labeled gestodene to 3 women, the compound was converted to reduced and hydroxylated metabolites, but a substantial number of the metabolites were not identified (42). Gestodene was not excreted in urine to any significant extent in an unchanged form, and it was not converted to levonorgestrel. It is generally accepted that gestodene is not a pro-drug. Single-dose pharmacokinetics of gestodene after intravenous and oral administration was investigated in 6 women who received four different treatments: 75 lag of gestodene given intravenously and 50, 75, or 125 lag of gestodene administered orally, each in combination with 30 lag of ethinylestradiol (37). After oral administration of the 50-, 75-, and 125-lag doses, maximum plasma levels of 1.0, 3.6, and 7.0 ng/mL, respectively, were attained between 1.4 and 1.9 hours post-treatment. After the maximum levels were reached, the subsequent levels of gestodene showed two disposition phases with half-lives of approximately 1 hour and 12 to 14 hours for each of the three doses. The absolute bioavailability was 99 + 11% (mean ___ standard deviation) (range 86% to 111%) for the 75-lag oral dose of gestodene. A subsequent study (36) showed that the absolute bioavailability of gestodene was 87 + 19% (mean _+ standard deviation) (range 64% to 126%) when 75 lag of the progestin in combination with 30 lag of ethinylestradiol was administered orally and intravenously to a group of 10 women. The data from these studies show that gestodene is highly bioavailable. Both the clearance and volume of distribution of gestodene were calculated from data obtained in the latter two
790 studies (36,37). Values (mean _+ standard deviation) of 3.4 + 1.5 L/hr and 0.80 + 0.53 mL/min/kg were reported for the clearance and 47.3 _+ 24 L and 0.66 _+ 0.43 L/kg for the volume of distribution, respectively. The values for the clearance and volume of distribution obtained from the first study (36) can also be expressed on the basis of an estimated average weight of 60 kg per subject; they are 0.94 + 0.42 mL/min/kg and 0.79 ___0.40 L/kg, respectively. As mentioned earlier, a multiple-dosing study measured levels of gestodene in serum. The gestodene levels were quantified in samples obtained from 11 women at frequent intervals on days 1, 10, and 21 of several cycles during 12 continuous cycles of treatment with 75 lag of gestodene in combination with 30 lag of ethinylestradiol (43). The results showed a dramatic rise in the mean gestodene levels between day 1 and day 10 and a further rise between day 10 and day 21 in all study cycles. These findings are similar to those obtained when multiple dosing was performed with desogestrel. The multiple-dosing study with gestodene/ethinylestradiol (43) also showed that the increases in mean SHBG levels during the first cycle and in subsequent cycles of treatment were very similar to the increases observed during long-term treatment with desogestrel/ethinylestradiol. There was a 2.7- to 3.0-fold increase in mean SHBG levels on day 21 of each cycle relative to the mean SHBG level on day I of the first cycle. In a recent review, it was pointed out that after administration of 75 lag of gestodene in combination with 30 lag of ethinylestradiol to women, circulating levels of gestodene were high relative to levels of other progestins measured after treatment with combined oral contraceptives (44). These elevated levels occurred after both single and multiple doses of gestodene/ethinylestradiol. The finding is surprising because the 75-lag dose of gestodene is the lowest of any progestin in a combination pill. Two factors may contribute to high circulating levels of gestodene: elevated circulating SHBG levels and a high affinity of SHBG for gestodene. Elevated SHBG levels result from the estrogenic component of combination pills, which has been shown to increase SHBG as much as threefold from pretreatment levels (43). However, a similar increase in SHBG levels was found after treatment with desogestrel/ ethinylestradiol, although serum etonogestrel levels remain relatively low (39). It has been reported that gestodene is distributed in serum as follows: SHBG-bound, 75.3 ___9.1%; albumin-bound, 24.1 _+ 9.1%; unbound, 0.6 _+ 0.1% (mean ___standard deviation) (41). Thus, approximately 75% of total circulating gestodene is bound to SHBG, in contrast to 32% for etonogestrel, 35% for norethindrone, and 47% for levonorgestrel (37). As a consequence, gestodene has a lower metabolic clearance rate and a greater concentration in the circulation. It is probably the affinity of SHBG for gestodene rather than an increase in circulating SHBG levels that is responsible for elevating the serum levels of this progestin.
FRANK Z. STANCZYK
I. Dienogest Studies on the pharmacokinetics of dienogest have been carried out by Oettel and colleagues (45) following oral and intravenous administration of different doses of the drug. A dose response was observed in serum dienogest levels after oral administration of four single doses of dienogest (1, 2, 4, and 8 mg) in randomized order during 4 consecutive menstrual cycles in 12 women. Following administration of the 1 mg dose, the Cmax levels were 23.4 + 5.9 ng/mL and Tmax was 2.2 ___ 1.1 hours (mean _+ standard deviation); the half-life of elimination was 6.5 hours. The absolute bioavailability of dienogest was determined in 17 women who ingested a single dose of 2 tablets, each containing 2 mg of dienogest and 30 Ixg of ethinylestradiol; the average value reported for this parameter was 96.2%. In the same study the average terminal half-life was reported to be 10.8 and 11.6 hours after the oral and intravenous doses, respectively. In the studies by Oettel and colleagues (45), it became obvious that circulating levels of dienogest are relatively high compared with those found with similar doses of other progestins. In blood, dienogest is mostly weakly bound to albumin (45), yet its clearance appears to be lower than that of other progestins (4). No significant accumulation was observed in serum levels of dienogest during its daily intake (45).
J. Drospirenone Pharmacokinetic characteristics of drospirenone (3 mg) combined with ethinylestradiol (30 Ixg) were assessed in 13 women during 13 continuous cycles, each of which consisted of 21 continuous days of treatment, followed by a 7-day pill-free interval (46). Frequent blood sampling was carried out on day 21 of treatment cycles 1, 6, 9, and 13. After administration of the first tablet, the Cmaxof drospirenone was 36.9 _+ 4.8 ng/mL (mean _+ standard deviation). This value rose to 87.5 _+ 51.4 ng/mL on day 21 of the first cycle and ranged from 78.7 to 84.2 ng/mL on day 21 of the next three sampling cycles. The corresponding Tma~ values ranged from 1.6 to 1.8 hours, and the half-life of elimination values was 31.1 to 32.5 hours. Other pharmacokinetic characteristics of drospirenone, based on data obtained by the manufacturer of the oral contraceptive containing 3 mg of drospirenone combined with 30 Ixg of ethinylestradiol (Yasmin, Berlex Laboratories, Wayne, NJ), were provided in a review article (47). It was reported that a steady state in circulating drospirenone levels was achieved after 1 week of treatment. Also, a dose response in circulating drospirenone levels was obtained, following oral administration of doses ranging from 1 to 10 ng/mL. In addition, the absolute bioavailability was reported to be 66% on average.
CHAPTER 54 Structure-Function Relationships, Pharmacoldnetics, and Potency of Progestogens
V. PHARMACOKINETICS OF PROGESTOGENS ADMINISTERED PARENTERALLY Although the oral route of progestogen administration is most common, there is increasing interest in obtaining an effective parenteral route for progestogens, primarily to avoid the hepatic first-pass effect. Progestogens can be administered by a variety of parenteral routes, which include intramuscular, vaginal, percutaneous, intranasal, sublingual, and rectal. There is, however, a paucity of data on the pharmacokinetics of progestogens administered parenterally. Three of the more commonly used parenteral routes are discussed here.
A. Intramuscular Route Dose proportionality of progesterone was evaluated in a study of six premenopausal women who received intramuscular administration of 10, 25, 50, or 100 mg of progesterone in oil (48). The results showed that peak concentrations of plasma progesterone for the different doses increased proportionally with increasing dose and were attained within 8 hours of dosing; the peak levels were 7, 28, 50, and 68 ng/mL for each of the doses, respectively. Elevated progesterone levels persisted for 24 to 48 hours, which is consistent with a depot effect of progesterone after intramuscular administration. The data indicate that a single intramuscular injection of 25 mg of progesterone in oil can give rise to circulating progesterone levels comparable with those found during the luteal phase of a spontaneously occurring menstrual cycle. In another study, a comparison was made between intramuscular and oral administration of progesterone (49). Three premenopausal women received a single oral 200-mg dose of micronized progesterone during days 2 to 5 of the menstrual cycle. Two days later, the same women received an intramuscular injection of 25 mg of progesterone in sesame oil. Blood and urine were collected at baseline and at 1, 4, 8, 12, and 24 hours post-treatment. Progesterone was measured in both serum and urine, and pregnanediol glucuronide, the predominant metabolite of progesterone, was measured in urine. Following the oral dose, serum progesterone levels rose rapidly, and peak levels ranging from 8.6 to 11.7 ng/mL were attained at 3 to 4 hours after dosing. By 24 hours, the peak progesterone concentrations fell to approximately 10% of the peak values. In contrast, urinary progesterone and pregnanediol glucuronide levels peaked later (i.e., about 4 to 8 hours after dosing), and at 24 hours the levels of both compounds decreased to 5% to 8% of the peak values. In comparison to the oral dosing, the results of the intramuscular administration showed that serum pro-
791
gesterone levels rose rapidly and were elevated at a level of about 10 ng/mL after 12 hours in all three subjects. Urinary progesterone and pregnanediol glucuronide levels peaked later and were still elevated even 24 hours after dosing. Both the oral and intramuscular routes of administration showed that urinary pregnanediol glucuronide concentrations were much higher and much more variable than those of urinary progesterone. The clinical significance of this study is that luteal phase serum progesterone levels can be achieved by both routes of administration; however, the intramuscular route provides a prolonged effect on these levels.
B. Intravaginal Route Vaginal administration of micronized progesterone has been shown to be an effective, acceptable, and convenient alternative to intramuscular injections (50). In one study, 200 mg of micronized progesterone was administered intravaginally every 6 hours to one group of 15 women and 50 mg of progesterone in oil was injected intramuscularly twice into a second group of women during a 24-hour study period. The results show that after intramuscular administration a rapid rise in serum progesterone levels occurred, with a plateau in levels of about 16 ng/mL after 4 hours of treatment. In contrast, after vaginal administration of progesterone a slow rise in serum progesterone levels occurred, which reached a peak of about 7 ng/mL after 4 hours. In the same study, endometrial concentrations of progesterone were also measured after intravaginal and intramuscular progesterone treatments. The results showed that endometrial progesterone concentrations following intravaginal treatment were considerably higher than after intramuscular treatment, even though serum progesterone levels were higher after intramuscular injection. The high endometrial progesterone concentrations obtained by the intravaginal route indicate the potential importance of this route.
C. Percutaneous Route Progesterone can be administered percutaneously in the form of a topical cream or gel. A major concern about progesterone creams is that serum progesterone levels achieved with the creams are too low to have a secretory effect on the endometrium (51). However, antiproliferative effects on the endometrium have been demonstrated with progesterone creams when circulating levels of progesterone are low (<4 ng/mL). Thus, effects of topical progesterone creams on the endometrium should not be based on serum progesterone levels but rather on histologic examination of the endometrium. Despite the low serum progesterone levels achieved with the creams, salivary progesterone levels are very high, indicating that
792 progesterone levels in serum do not necessarily reflect those in tissues. The mechanism by which the serum progesterone levels remain low is not known. However, one explanation is that after absorption through the skin, the lipophilic ingredients of creams, including progesterone, may have a preference for saturating the fatty layer below the dermis. Because there appears to be rapid uptake and release of steroids by red blood cells passing through capillaries, these cells may play an important role in transporting progesterone to salivary glands and other tissues. In contrast to progesterone creams, progesterone gels are water soluble and appear to enter the microcirculation rapidly, thus giving rise to elevated serum progesterone levels with progesterone doses comparable to those used in creams. Preliminary data were obtained in one study in which one of two different doses (30 or 100 mg) of progesterone gel was applied on the inner portion of the arms of postmenopausal women daily for 1 month (52). Blood samples were obtained at frequent intervals on the first day of treatment. The results showed that following administration of the 30-mg dose of progesterone gel, serum progesterone levels increased by 50% to 100%, but remained in the follicular phase range (<0.5 ng/mL). With the 100-mg progesterone dose, peak progesterone levels of 5.8 to 8.0 ng/mL were formed at 2 to 3 hours after gel application. Similar levels of progesterone were found at 1, 2, and 4 weeks of treatment. Progesterone cream products are readily available over the counter. However, some of these products do not contain progesterone but contain wild yam extract instead. This extract contains diosgenin, which can be converted to progesterone by a series of chemical reactions (see Fig. 54.2) that do not occur in the body. It is important to inform patients using progesterone creams that creams containing wild yam extract without progesterone will not provide a progestational effect.
FRANK Z. STANCZYK
excretion. All these factors play an important role in determining the biologic response of a progestogen. The potency of a drug can be defined as an estimate of a specific biologic effect of the drug. From the technical point of view, a drug's potency is always related to that of a standard drug and is quantified by measuring the difference of the parallel dose-response curves produced by the reference drug and the test drug. A variety of qualitative and quantitative tests using either human or animal species to establish potencies of progestogens have been performed. The tests can be divided into the following three types: (1) in vitro receptor binding assays, (2) bioassays, and (3) clinical tests. Problems encountered in estimating potencies of progestogens using those tests are well recognized (53-56). A major source of the difficulties can be found in the variables associated with each test. In general, the variables include (1) type of animal species; (2) type of tissue or target organ; (3) the specific response of the tissue or target organ; (4) temporal considerations (e.g., time of test following dosing); (5) the route of, and vehicle for, drug administration; (6) the type of reference drug; and (7) assay conditions (e.g., incubation time and temperature). Interlaboratory assay variations preclude precise comparisons of relative potencies. Another important limitation of receptor binding tests or bioassays is that no estrogen is usually added to the progestogen being tested. Addition of estrogen to a progestogen may increase the potency of the progestogen substantially. Difficulties also arise from the fact that potency estimates from animal tests cannot be extrapolated to humans. Although these difficulties exist, there is a substantial amount of information pertaining to potencies of progestogens, and some generalizations can be made. An overview of some of the more relevant data pertaining to progestogen potency follows.
A. Progestational Activity VI. POTENCY OF P R O G E S T O G E N S The efficacy of a progestogen depends on a number of factors. Basic prerequisites for progestational activity of a steroid include adequate affinity for the progesterone receptor, induction of conformational change of the steroidreceptor complex, and duration of binding to DNA. The number of receptors occupied by the steroid is a function of its concentration in the target cell. Intracellular steroid concentration depends on how much steroid enters the cell and is metabolized and stored. The extent to which the steroid enters the cell is, in turn, dependent on its circulating level in a bioavailable (non-SHBG-bound) or free form. The blood level of the steroid depends on its pharmacokinetics; in other words, its absorption and metabolism during the hepatic first pass, rates of distribution and elimination, and
1. RECEPTOR BINDING TESTS
Uteri from various animal species, including humans, in various conditions of age and pretreatment have been used as a source of progesterone receptors for binding studies. In practice, the binding affinity of a test steroid for the progesterone receptor is determined by the concentration of the steroid that corresponds to 50% inhibition of the total binding (IC50) of a radiolabeled progestational marker (e.g., tritiated progesterone) to the receptor. Because progesterone is the natural progestational agent of all mammals, it is an obvious choice to be considered the prototype for comparison. Progesterone initially served as the reference steroid and was used in conjunction with 3Hprogesterone in competitive binding studies with progesterone receptors. However, because most progestins have
CHAPTER 54 Structure-Function Relationships, Pharmacokinetics, and Potency of Progestogens considerably greater progestational activity than progesterone, a highly potent synthetic progestin, referred to as R5020 (17,21-dimethyl- 19-nor-pregna-4,9-diene-3,20-dione), has replaced progesterone as the reference compound. Comparison of the binding affinity of a test steroid for a specific receptor relative to that of a steroid standard is expressed by the relative binding affinity (RBA). Thus, the RBA of a test steroid for uterine progesterone receptors can be calculated from the IC50 of a reference steroid divided by the corresponding IC50 of the test steroid. The ratio is multiplied by 100 and expressed as a percentage. Table 54.3 shows that when R5020 was employed as the reference steroid, etonogestrel, levonorgestrel, MPA, cyproterone acetate, and gestodene were all bound with high affinity to the human uterine progesterone receptors (57). The RBA values of these compounds ranged from 85% to 130% and were 2.1- to 3.2-fold higher than the RBA of progesterone (40%). No measurable binding affinity to the progesterone receptors was demonstrated for desogestrel or norgestimate. The lack of significant binding of norgestimate and desogestrel to the progesterone receptors supports the view that these compounds are pro-drugs and must be transformed to a biologically active form for progestogenic action. A different pattern in hierarchy of RBA values was obtained with the test substances when etonogestrel, gestodene, or levonorgestrel were used as reference steroids (57). On the basis of the data shown in Table 54.3, it appears that gestodene binds to the progesterone receptors with higher affinity than either etonogestrel or levonorgestrel and that etonogestrel has a slightly higher affinity than levonorgestrel. Profound differences in progestational activity are often observed between human and animal tissues used in progestogen potency tests. This is clearly evident in the results of two studies in which RBAs for uterine progesterone receptors were determined (58,59). In one of the studies (58), rabbit
TABLE 54.3
uterine tissue was used, and it was shown that norgestimate was bound to the progesterone receptor with high affinity (RBA, 124%). Also, norelgestromin and levonorgestrel had RBAs of 94% and 541%, respectively. In contrast, data from the other study (59), in which human uterine tissue was used, showed there was no significant binding of norgestimate to the progesterone receptor (RBA, 0.8%) and the binding of norelgestromin was low (RBA, 8%); the RBA of levonorgestrel was 250%. The discrepancy in the potency data of norgestimate and its principal active metabolites between human and rabbit tissues emphasizes the potential problems that may arise when extrapolating animal data to the human.
2. BIOASSAYS
The most common bioassays used to evaluate progestational potency determine the effect of the test compound on uterine glandular proliferation, pregnancy maintenance, delay of parturition, or inhibition of ovulation in rabbits or rats. The most widely used bioassay for progestational agents has been the Clauberg test. This test is based on initial observations made by Clauberg in the 1920s. In 1934 McPhail organized these observations into specific protocols (60). In general, the test procedure involves priming immature female rabbits with estrogen, followed by either oral or parenteral treatment with the test substance. Progestogens induce the development of complicated glandular structures in the estrogenic endometrium with simple glands. A standardized scale for grading glandular proliferation of the rabbit endometrium was provided by McPhail. This scale ranges from 0, which corresponds to no glandular development, to a value of +4, corresponding to maximal glandular development. Active progestational compounds are compared at a dose level that produces a +2 on the McPhail scale. Although the Clauberg test is the most widely used bioassay for progestational agents, it is subject to considerable
Relative Binding Affinities (%) of Different Progestogens for Human Uterine Progesterone Receptor
Compound
R5020
Progesterone Gestodene Levonorgestre1120 R5020 Org2058 Desogestrel Etonogestrel Norgestimate Cyproterone acetate Medroxyprogesterone acetate
40 85 90 100 350 1 130 < 0.1 90 115
ND, not determined. (From ref. 9, with permission.)
793
Etonogestrel
Gestodene
15 110 75 20 ND 0.3 100 < 0.1 ND ND
2 100 100 20 35 <0.1 55 < 0.1 ND ND
Levonorgestrel 18 150 22 ND 0.3 110 < 0.1 ND ND
794 divergence in estimates of potency (61). Interpretation of the test is complicated by the fact that the dose-response curves for commonly employed test substances are not parallel and the maximum response varies. The other commonly used bioassays also have limitations (61). Estrogens are ineffective in the pregnancy maintenance bioassays and will inhibit the active progestogens when administered at sufficient dose levels. The delay of parturition bioassay fails to distinguish among the various progestogens. Finally, the ovulation inhibition bioassay gives different progestogen potencies compared with those obtained in women. 3. CLINICALTESTS
A variety of clinical tests have been used to assess the relative potency of progestogens in women. These include tests based on induction of secretory changes in the endometrium, inhibition of ovulation, and changes in vaginal cytology and cervical mucus. Because endometrial effects of progestogens are relatively simple to assess clinically, the earliest efforts to compare the potency of progestogens relied on the delay-of-menses test, which was first described by Greenblatt et al. (62). The test is based on the fact that uterine bleeding is induced by withdrawal of hormonal support of the endometrium at the end of the menstrual cycle. In general, the assay protocol involves administration of the test substance beginning on the sixth or seventh day after ovulation and continuation of treatment for 3 weeks or more. An effective progestogen will delay menstrual bleeding until 2 to 3 days after treatment is discontinued. The delay-of-menses test was subsequently standardized by Swyer and used for comparative potency evaluations of progestogens; this test is sometimes referred to as the Swyer-Greenblatt test (63). In 1985 the potencies of progestogens were compared on the basis of available human data in the literature obtained from studies in which the effect of progestogens on the delay of menses, subnuclear vacuolization, and glycogen deposition, as well as on lipids and lipoproteins, was assessed (64). In the review of those data, the objective was to examine the scientific evidence, which generally supported the view that the potency of the progestogens used in oral contraceptive formulations marketed in the United States was similar. The conclusion of the review was that norethindrone, norethindrone acetate, and ethynodiol diacetate are approximately equivalent in potency, whereas norgestrel and its biologically active enantiomer, levonorgestrel, are about 5 to 10 and 10 to 20 times as potent as norethindrone, respectively. Another approach to determining progestogen potency from clinical data emerged from a series of studies by King and coworkers (65-71), in which progestogen effects were assessed by analyzing biochemical and morphologic features of endometria from estrogen-primed postmenopausal women. The postmenopausal women were treated daily
FRANKZ. STANCZYK
with either 0.625 or 1.25 mg of CEEs. Effects of at least three different doses of each of five orally administered progestogensmspecifically, norethindrone, levonorgestrel, MPA, dydrogesterone and progesteronemwere tested after 6 days of sequential progestogen treatment during the last 6 to 12 days of the month. The biochemical parameters analyzed included the nuclear estradiol receptor, DNA synthesis, and isocitric and estradiol dehydrogenases. Using data from these studies, King and Whitehead (72) recalculated the results to allow comparisons with corresponding premenopausal, secretory phase values. Relative to a value of 1 for norethindrone, it was shown that levonorgestrel had a potency that was eightfold greater, whereas the potencies ofMPA, dydrogesterone, and progesterone were 10, 50, and 500 times lower, respectively. The progestogen potencies just shown, based on studies by King and Whitehead (72), are consistent with oral progestogen doses that will provide endometrial protection (Table 54.4). In general, the doses are 1 mg, 0.15 mg, 2.5-10 mg, 20 mg, and 100-300 mg for norethindrone (or its acetate), levonorgestrel, MPA, dydrogesterone, and progesterone, respectively. Although these doses are commonly administered therapeutically, the specific dose used depends on whether the progestogen is given sequentially for 10 to 14 days per month or continuously combined, as well as on the type of estrogen administered. In addition to assessment of progestogen potency based on endometrial effects, ovulation inhibition has also been used to compare potency of progestogens. Progestogen doses required for ovulation inhibition have been determined for various combination oral contraceptives containing 30 to 35 p~g ethinylestradiol (73). The data show that high doses are required for norethindrone (400 pLg) and norgestimate (200 p.g) and relatively low doses for gestodene (30 ~g), levonorgestrel (60 ~zg), and desogestrel (60 ~g). On the basis of the clinical evaluations of progestogen potency that have been discussed here, certain generalizations can be made. Progestins structurally related to testosterone are considerably more potent than progesterone and TABLE 54.4
Comparison of Different Progestogen Potencies Determined Experimentally with Corresponding Therapeutic Oral Doses
Progestin Levonorgestrel Norethindrone Medroxyprogesterone acetate Dydrogesterone Progesterone
Experimental
Potency* Basedon dose
8 1
6.7 (0.15 mg) 1 (1 mg)
0.1 0.02 0.002
0.1-0.4 (2.5-10 mg) 0.05 (20 mg) 0.01-0.003 (100- 300 rag)
*Relativeto a valueof 1 for norethindrone.
CHAPTER 54 Structure-Function Relationships, Pharmacokinetics, and Potency of Progestogens
progestins structurally related to progesterone. Among the latter group of progestogens, progesterone is less potent than dydrogesterone, which in turn is less potent than MPA. In the 17ot-ethinylated 19-nortestosterone group, gestodene, levonorgestrel, and desogestrel are more potent than norgestimate, and norethindrone and its pro-drugs (e.g., norethindrone acetate). As pointed out earlier, desogestrel and norgestimate are pro-drugs and their progestational potency is exhibited through their active metabolites.
B. Other Biologic Activities of Progestogens In addition to the characteristic progestational activity of progestogens, they may also possess anti-estrogenic, androgenic, anti-androgenic, and/or anti-mineralocorticoid activities. The most controversial and confusing of these activities is the androgenicity of certain progestins. 1. ANDROGENICAND ANTI-ANDROGENIC ACTIVITY The two most notable androgenic progestins are levonorgestrel and norethindrone. Primary evidence for their androgenicity includes significant binding affinity for the rat prostatic androgen receptor, stimulation of ventral prostate growth in immature castrated rats, and suppression of serum SHBG and lipoprotein levels. However, there are deficiencies in studies of progestin androgenicity. First, data from studies using the rat ventral prostate may not be relevant to a woman's tissues. Also, the doses used in such studies are considerably higher than those required for ovulation inhibition in women, and the estrogenic component is usually not considered. It is well recognized that some progestins have a suppressive effect on hepatic proteins such as SHBG and high-density lipoprotein (HDL), and for this reason they are considered to be androgenic. However, proper interpretation of data from studies investigating androgenic effects of progestins using SHBG as the end point is often lacking. When two progestins are compared in such studies, the data may show that one has a lower percentage of free or bioavailable (non-SHBG-bound) testosterone than the other based on a higher SHBG level, suggesting that the former progestin is less androgenic. However, this suggestion is not valid unless the effect on the total testosterone concentration is taken into account. This has been shown very clearly in a study by Thorneycroft et al. (74)in which the effect of 100 Ixg oflevonorgestrel was compared with 1000 txg of norethindrone acetate, each combined with 20 Ixg of ethinylestradiol, on SHBG and bioavailable testosterone concentrations. The results showed that bioavailable testosterone was suppressed significantly and to the same extent from baseline with each formulation, but by different mechanisms. With the norethindrone acetate formulation, bioavailable testosterone concentrations were suppressed due to decreased SHBG levels. However, with the levonorgestrel formulation,
795
there was a lesser increase in SHBG levels, which was compensated for by a greater suppression of total testosterone levels compared with the norethindrone acetate formulation. It is important to realize that androgenic progestins such as levonorgestrel and norethindrone also have anti-androgenic effects. They suppress serum levels of androgens produced by the adrenal, ovarian, and peripheral compartments. In the study just described (74), the adrenal androgen marker, DHEAS1 and the markers of peripheral androgen production, dihydrotestosterone and 3ot-androstanediol glucuronide, were each similarly and significantly reduced from baseline in the two treatment groups. These data demonstrate that both levonorgestrel and norethindrone, in combination with ethinylestradiol, have profound anti-androgenic effects with respect to androgen production when administered orally. In addition, an important anti-androgenic effect of combined oral contraceptives, including those containing levonorgestrel and norethindrone, is observed clinically in women who show improvement in hirsutism and acne. 2. ESTROGENICACTIVITY
The binding affinity of a variety of progestogens for the uterine estrogen receptors has been investigated using incubation procedures similar to those described for the progesterone receptor and androgen receptor assays. Weak displacement oftritiated estradiol from rabbit uterine cytosol receptors has been demonstrated by norethindrone, norethynodrel, and ethynodiol diacetate but not by levonorgestrel, MPA, and megestrol acetate (56). In a similar study, relative binding affinities of a number of different progestogens were determined for binding to human estrogen receptors. The results showed no affinity of the estrogen receptor for gestodene, levonorgestrel, MPA, or cyproterone acetate (RBAs < 0.1%). The estrogenicity of norethindrone, norethynodrel, and ethynodiol diacetate has also been demonstrated in bioassays (56). Estrogenic effects of both norethynodrel and ethynodiol diacetate have been shown in short-term assays for vaginal cornification in rats (Allen-Doisy vaginal smear test). In contrast, norethindrone causes vaginal changes in the rat only when long-term protocols involving the oral route of administration are used (75). It is not effective by parenteral routes of administration. Clinical evaluation of estrogenicity, as measured by actions on the cervix and changes in vaginal cytology, shows that the estrogenic effects of norethynodrel and ethynodiol diacetate in humans parallel those in the rat (56). However, there is no clear-cut evidence that norethindrone exhibits an estrogenic effect clinically.
3. ANTI-MINERALOCORTICOIDACTIVITY
It is well recognized that progesterone has substantial anti-mineralocorticoid activity. It binds with high affinity to the mineralocorticoid receptor, acting as an antagonist, with
796
FRANK Z. STANCZYK
obvious significance for electrolyte homeostasis, and a variety of mineralocorticoid-related functions in the circulation and central nervous system. Drospirenone is the first progestin with anti-mineralocorticoid activity. This activity is well established at doses that exert progestational and antiovulatory effects in rats and women, respectively (76). T h e anti-aldosterone potency of drospirenone is in the range of 549% to 1095% of that of spironolactone (76).
VII. CONCLUSIONS In the pharmacokinetic studies discussed earlier, large intersubject variability was found in serum or plasma levels and pharmacokinetic parameters, including bioavailability and half-life, of all the progestogens. Higher bioavailability of a progestogen may result in a lower required dosage and less intersubject and intrasubject variability. A long serum half-life may be associated with a greater protective effect in the event of missed pills and may provide more consistent cycle control. Assessment of progestogen potency is problematic due to the large number of variables and assumptions associated with quantitative and qualitative tests of potency. Difficulties also arise when potency estimates from animal tests are extrapolated to humans. Nevertheless, certain clinical tests in which potency of orally administered progestogens is assessed in w o m e n can provide useful information from which optimal therapeutic progestogen doses can be determined. Biologic activities of progestogens, other than progestational effects, have been poorly studied. There is a misconception about progestin androgenicity. This is due primarily to extrapolation of data from rat studies to the human and misinterpretation of data that show effects of progestins on S H B G and free testosterone. A better understanding of the pharmacokinetics and potency of progestins will help clinicians to choose the optimal type and dose of progestogen for individualized treatment of patients.
References 1. Stanczyk FZ, Henzl MR. Use of the name "progestin". Contraception 2001;64:1-2. 2. Position Statement. Role of progestogen in hormone therapy for postmenopausal women: position statement of the North American Menopause Society.Menopause 2003;10:113-132. 3. Henzl MR. Contraceptive hormones and their clinical use. In: Yen SSC, Jaffe RE, eds. Reproductive endocrinology,physiology, pathophysiology, and clinical management. Philadelphia: W.B. Saunders, 1986: 643 -682. 4. StanczykFZ. Introduction: structure-function relationships,metabofism, pharmacokinetics and potency of progestins. Drugs Today 1996;32:1-14. 5. Stanczyk FZ. Pharmacokinetics and potency of progestins used for hormone replacement therapy and contraception. Rev Endocr Metabol Disord 2002;3:211-224.
6. Stanczyk FZ. All progestins are not created equal. Steroids 2003;68: 879-890. 7. Stanczyk FZ, Roy S. Metabolism of levonorgestrel, norethindrone, and structurally related contraceptive steroids. Contraception 1990;42:67-96. 8. Edgren RA, Stanczyk FZ. Nomenclature of the gonane progestins. Contraception 1999;60:313. 9. StanczykFZ. Pharmacokineticsof progesterone administered orally and parenterally. In: Sitruk-Ware R, Mishell DR Jr, eds. Progestins and antiprogestins in dinicalpractice. New York: Marcel-Dekker, 2000:393-400. 10. Simon JA, Robinson DE, Andrews MC, et al. The absorption of oral micronized progesterone: the effect of food, dose proportionality, and comparison with intramuscular progesterone. Fertil Steri11993;60: 26-33. 11. Padwick ML, Endacott J, Matson C, Whitehead MI. Absorption and metabolism of oral progesterone when administered twice daily. Fertil Steri11986;46:402-407. 12. Sitruk-Ware R. Progestins in hormonal replacement therapy and prevention of endometrial disease. In: Sitruk-Ware R, Mishell DR Jr, eds. Progestins and anti-progestins in clinical practice. New York: Marcel Dekker, 2000:269-287. 13. Hiroi M, Stanczyk FZ, Goebelsmann U, et al. Radioimmunoassay of serum medroxyprogesteroneacetate (Provera) in women following oral and intravaginal administration. Steroids 1975;26:373-386. 14. Svensson L-O, Johnson SH, Olsson S-E. Plasma concentrations of medroxyprogesterone acetate, estradiol and estrone following oral administration of Klimaxil,Trisequence/Provera and Divina. A randomized, single-blind, triple cross-over bioavailability study in menopausal women. Maturitas 1994;18:229-238. 15. Mathrubutham M, Fotherby K. Medroxyprogesterone acetate in human serum.J Steroid Biochem 1981;14:783-786. 16. VictorA, Johansson ED. Pharmacokineticobservations on medroxyprogesterone acetate administered orally and intravaginally. Contraception 1976;14:319-329. 17. Stanczyk FZ, Mroszczak EJ, Ling T, et al. Plasma levels and pharmacokinetics of norethindrone and ethinylestradiol administered in solution and as tablets to women. Contraception 1983;28:241-251. 18. Odlind V, Weiner E, Victor A, Johansson EDB. Plasma levels of norethindrone after single oral dose administration of norethindrone and lynestrenol. Clin Endocrino11979;10:29- 38. 19. Back DJ, Brackenridge AM, Crawford FE, et al. Kinetics of norethindrone in women. II. Single-dose kinetics. Clin Pharmacol Ther 1978;24: 448-453. 20. Hammond GL, Lahteenmaki PL, Luukkainen T, et al. Distribution and percentages of non-protein bound contraceptive steroids in human serum.J Steroid Biochem 1982;17:375-380. 21. Kuhnz W, Huener A, Humpel M, Seifert W, Michaelis K. In vivo conversion of norethisterone and norethisterone acetate to ethinyl estradiol in postmenopausal women. Contraception 1997;56:379- 385. 22. Chu MC, Nakhuda GS, Zhang X, Stanczyk FZ, Lobo RA. Formation of ethinyl estradiol in women during treatment with norethindrone acetate.J Clin Endocrinol Metab, in press. 23. Humpel M, Wendr H, Dogs G, et al. Intraindividual comparison of pharmacokinetic parameters of d-norgestrel, lynestrenol and cyproterone acetate in 6 women. Contraception 1977;16:199-215. 24. Goebelsmann U, Hoffman D, Chiang S. Woutersz T. The relative bioavailability of levonorgestrel and ethinyl estradiol administered as a low-dose combination oral contraceptive. Contraception 1986;34: 341-351. 25. Stanczyk FZ, Lobo RA, Chiang ST, Woutersz TB. Pharmacokinetic comparison of two triphasic oral contraceptive formulations containing levonorgestrel and ethinylestradiol. Contraception 1990;41:39-53. 26. Back DJ, Bates M, Breckenridge AM, Hall JM, Maclver M. Levonorgestrel and ethinylestradiol in women m studies with Ovran and Ovranette. Contraception 1981;23:229- 239.
CHAPTER 54 Structure-Function Relationships, Pharmacokinetics, and Potency of Progestogens 27. Humpel M, Wendt H, Pommerenke G, Weib CHR, Speck U. Investigations of pharmacokinetics of levonorgestrel to specific consideration of a possible first pass effect in women. Contraception 1978;17:207-220. 28. McGuire JL, Phillips A, Hahn DW, et al. Pharmacologic and pharmacokinetic characteristics of norgestimate. Am J Obstet Gynecol 1990; 163:2127-2131. 29. Kuhnz W, Blode H, Mahler M. Systemic availability of levonorgestrel after single oral administration of a norgestimate-containing oral contraceptive to 12 women. Contraception 1993b;47:283-294. 30. White T, Ozel B, Jain JK, Stanczyk FZ. Effects of transdermal and oral steroidal contraceptives on estrogen-sensitive hepatic proteins, levonorgestrel, serum-binding globulins, and C-reactive protein. Contraception 2006;74:293-296. 31. Viinikka L. Radioimmunoassay of a new progestogen, ORG 2969, and its metabolite.J Steroid Biochem 1978;10:353-357. 32. Viinikka L. Metabolism of a new synthetic progestogen, ORG 2969, by human liver in vitro. J Steroid Biochem 1979;10:353-357. 33. Hasenack HG, Bosch AMG, Kaar K. Serum levels of etonogestrel after oral administration of desogestrel and etonogestrel. Contraception 1986;33:591-596. 34. Hammerstein J. Prodrugs: advantage or disadvantage? Am J Obstet Gyneco11990;163:2198 - 2203. 35. Back DJ, Grimmer SFM, Shenoy N, Orme MLE. Plasma concentrations of etonogestrel after oral administration of desogestrel and intravenous administration of etonogestrel. Contraception 1987;35:619- 626. 36. Orme M, Back DJ, Ward S, Green S. The pharmacokinetics of ethinyl estradiol in the presence and absence of gestodene and desogestrel. Contraception 1991;43:305 - 316. 37. Tauber U, Tack JW, Matthes H. Single dose pharmacokinetics of gestodene in women after intravenous and oral administration. Contraception 1989;40:461-479. 38. Bergink W, Assendorp R, Koosterboer L, et al. Serum pharmacokinetics of orally administered desogestrel and binding of contraceptive progestogens to sex hormone-binding globulin. Am J Obstet Gynecol 1990;163:2132-2137. 39. Kuhl H, Jung-Hoffman C, Heidt E Serum levels of etonogestrel and SHBG during 12 cycles of treatment with 30 pg desogestrel. Contraception 1988a;38:381 - 390. 40. Jung-Hoffman C, Kuhl H. Interaction with the pharmacokinetics of ethinyl estradiol and progestogens contained in oral contraceptives. Contraception 1989;40:299- 312. 41. Kuhnz W, Pfeffer M, Al-Yacoub G. Protein binding of the contraceptive steroids gestodene, 3-keto-desogestrel and ethinylestradiol in human serum.J Steroid Biochem 1990;35:313-318. 42. Dusterberg B, Tack J-W, Krause W, Humpel M. Pharmacokinetics and biotransformation of gestodene in man. In: Elstein M, ed. Gestodene." development of a new gestodene-containing lozo-dose oral contraceptive. London: Parthenon, 1987:35-44. 43. Kuhl H, Jung-Hoffman C, Heidt E Alterations in the serum levels of gestodene and SHBG during 12 cycles of treatment with 30 mg ethinyl estradiol and 75 mg gestodene. Contraception 1988b;38:477-486. 44. Fotherby K. Potency and pharmacokinetics of gestagens. Contraception 1990;41:533-550. 45. Oettel M, Bervoas-Mart/n S, Elger W, et al. A 19-norprogestin without a 17a-ethinyl group. I: Dienogest from a pharmacokinetic point of view. DrugsToday 1995;31:499-516. 46. Blode H, Wuttke W, Loock W, R611G, Heithecker R. A 1-year pharmacokinetic investigation of a novel oral contraceptive containing drospirenone in healthy female volunteers. Eur J Contracept Reprod Health Care 2000;5:256-264. 47. Krattenmacher R. Drospirenone: pharmacology and pharmacokinetics of a unique progestogen. Contraception 2000;62:29-38. 48. Nillius SJ, Johansson EDB. Plasma levels of progesterone after vaginal, rectal, or intramuscular administration of progesterone. Am J Obstet Gyneco11971;110:470-477.
797
49. Stanczyk FZ, Gentzschein E, Ary BA, et al. Urinary progesterone and pregnanediol. Use for monitoring progesterone treatment. J Reprod Med 1997;42:216-222. 50. Miles RA, Paulson RJ, Lobo RA. Pharmacokinetics and endometrial tissue levels of progesterone after administration by intramuscular and vaginal routes: a comparative study. Fertil Steri11994;62:485-490. 51. Stanczyk FZ, Paulson R, Roy S. Percutaneous administration of progesterone: blood levels and endometrial protection. Menopause 2005; 12:232-237. 52. Bello SM, Merzow G, Shoupe D, Stanczyk FZ. Administration of progesterone by use of percutaneous gel in postmenopausal women. Proceedings of the Annual Meeting of the Pacific Coast Fertility Society, Indian Wells, CA, April 9-13, 1997. 53. Edgren RA, Sturtevant FM. Potencies of oral contraceptives. Am J Obstet Gyneco11976;125:1029-1038. 54. Edgren RA. Progestational potency of oral contraceptives: a polemic. IntJ Ferti11978;23:162-169. 55. Edgren RA. Relative potencies of oral contraceptives. In: Moghissi KS, ed. Controversiesto contraception. Baltimore: Williams &Wilkins, 1979: 1-19. 56. Edgren RA. Progestins. In: Givens JR, ed. Clinical use of sex steroids. Chicago: Year Book Medical, 1980:1-29. 57. Juchem M, Pollow K. Binding of oral contraceptive progestogens to serum proteins and cytoplasmic receptor. AmJ Obstet Gyneco11990;163: 2171-2183. 58. Phillips A, Demarest DW, Hahn F, Wong F, McGuire JL. Progestational and androgenic receptor binding affinities and in vivo activities of norgestimate and other progestins. Contraception 1990;41: 399-410. 59. Juchem M, PoUow E, Elger W, Hoffman G, Mobus V. Receptor binding of norgestimatema new orally active synthetic progestational compound. Contraception 1993;47:283-294. 60. McPhail MK. The assay of progestin. J Physio11934;83:145-156. 61. Edgren RA. Issues in animal pharmacology. In: Goldzieher JW, Fotherby K, eds. Pharmacology of the contraceptive steroids. New York: Raven, 1994;81-97. 62. Greenblatt RB, Jungck EC, Barfield WE. A new test of efficacy of progestational compounds. Ann NYAcad Sci 1958;71:717. 63. Swyer GIM, Little V. Clinical assessment of relative potency of progestogens. J Reprod Fertil Supp11968;5:63-68. 64. Dorflinger LJ. Relative potency of progestins used in oral contraceptives. Contraception 1985;31:557-570. 65. Whitehead MI, Townsend PT, Pryse-Davies J, Ryder TA, King RJB. Effects of estrogens and progestins on the biochemistry and morphology of the postmenopausal endometrium. N EnglJ ivied 1981; 305:1599. 66. Whitehead MI, Townsend PT, Pryse-Davies J, et al. Effects of various types and dosages of progestins on the postmenopausal endometrium. J Reprod Med 1982;27:539. 67. Lane G, Siddle NC, Ryder TA, et al. Dose dependent effects of oral progesterone on the oestrogenised postmenopausal endometrium. Br MedJ 1983;287:1241. 68. Whitehead M, Lane G, Siddle N, Townsend P, King R. Avoidance of endometrial hyperstimulation in estrogen-treated postmenopausal women. Semin Reprod Endocrino11983a;1:41. 69. Whitehead MI, Siddle NC, Townsend PT, Lane G, King RJB. The use of progestins and progesterone in the treatment of climacteric and postmenopausal symptoms. In: Bardin CW, Milgrom E, Mauvais Jarvis P, eds. Progesteroneandprogestins. New York: Raven, 1983b:277. 70. Lane G, Siddle NC, Ryder TA, et al. Effects of dydrogesterone on the oestrogenized postmenopausal endometrium. Br J Obstet Gynaecol 1986a;93:55.
798 71. Lane G, Siddle NC, Ryder TA, et al. Is Provera the ideal progestin for addition to postmenopausal estrogen therapy? Fertil Steril 1986b; 45:345. 72. King RJB, Whitehead MI. Assessment of the potency of orally administered progestins in women. Fertility and Sterility 1986;46: 1062-1066. 73. Teichman AT. Levonorgestrel. Stuart-George Thieme Verlag. 1990;43.
FRANK Z. STANCZYK 74. Thorneycroft IH, Stanczyk FZ, Bradshaw KD, et al. Effect of lowdose contraceptives on androgenic markers and acne. Contraception 1999;60:255-262. 75. Jones RC, Edgren RA. The effects of various steroids on the vaginal histology in the rat. Fertil Steri11973;24:284. 76. Elger W, Beier S, Pollow K, et al. Conception and pharmacodynamic profile of drospirenone. Steroids 2003;68:891-905.
_~HAPTER 5~
Intervention: Androgens SUSAN R. DAVIS
Women's Health Program, Monash University (CECS), Alfred Hospital, Prahran, Victoria 3181, Australia
I. I N T R O D U C T I O N
hormone therapy (HT) is being considered and again at subsequent consultations. The use of testosterone therapy in women is no longer restricted to those who have undergone either a spontaneous menopause in their middle years or who have had surgical menopause; it is increasingly being used in younger women who have experienced premature ovarian failure. A possible future indication includes women in their late reproductive years who experience loss of libido with no identifiable cause (Table 55.1).
Androgens have important physiologic effects in women. Not only are they the precursor hormones for estrogen biosynthesis in the ovaries and extragonadal tissues, but androgens also appear to directly act throughout the body via androgen receptors. Androgen levels decline with increasing age in women (1), and it is believed that many postmenopausal women experience a variety of physical symptoms secondary to androgen depletion as well as physiologic changes that affect their quality of life. Affected women complain of persistent fatigue, lack of well-being, and loss of libido, symptoms easily attributable to psychosocial and environmental factors. Although these symptoms are not clearly linked to low circulating total or free testosterone levels (2), testosterone therapy may result in significant improvement in symptomatology (3-6). In contrast to the acute fall in circulating estrogen at the time of menopause, the decline in circulating testosterone and the adrenal pre-androgens commences in the decade preceding the average age of natural menopause and most closely parallels increasing age (2). The symptoms of androgen insufficiency usually develop insidiously, and most women are not aware that such symptoms have a biologic basis. Therefore, it is important to directly question women about the presence of such symptoms when postmenopausal TREATMENT OF THE POSTMENOPAUSAL W O M A N
II. A N D R O G E N P H Y S I O L O G Y IN T H E R E P R O D U C T I V E YEARS Androstenedione (A) and dehydroepiandrosterone (DHEA) are produced by both the ovaries and the adrenals, with the adrenals also being the main site of production of DHEAS-sulfate (DHEA-S). Approximately half of the circulating testosterone is produced by peripheral conversion of these pre-androgens to testosterone, with A being the main precursor (7). The ovaries are a primary site of testosterone synthesis, and circulating DHEA-S is an important precursor for ovarian intrafollicular production of testosterone and dihydrotestosterone (DHT) (8). Whether there is direct secretion of testosterone by the adrenals is 799
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
SUSANR. DAVIS
800
TABLE 55.1
Clinical Indications for Testosterone Replacement in Women
9 Symptomatictestosterone deficiency following natural menopause 9 Symptomatic testosterone deficiency due to surgical menopause, chemotherapy, or irradiation 9 Premature ovarian failure 9 Premenopausal loss of libido with diminished serum free testosterone Potential indications for use in w o m e n :
9 Glucocorticosteroid-induced bone loss 9 Premenopausal bone loss 9 Premenopausal iatrogenic androgen deficiency states, including GnRH-analog treatment of endometriosis 9 Autoimmune diseases
controversial (9). Testosterone is further metabolized to the potent androgen D H T in various peripheral sites. There is significant cyclicity in plasma levels of A and testosterone in regularly ovulating women, with increases in the mean circulating levels of both of these hormones in the middle third of the menstrual cycle (10,11). This is followed by a second rise in A production by the corpus luteum during the late luteal phase. Ovarian androgens are produced by the thecal cells under the control of luteinizing hormone (LH). The blood level of D H E A in adulthood is higher than any other circulating steroid except cholesterol. D H E A secretion is acutely stimulated by adrenocorticotropic hormone (ACTH) (12,13); however, DHEA-S, which has a long plasma half-life, does not acutely increase following A C T H administration. Adrenal androgen and cortisol production are not always linked. Circulating adrenal androgen levels have been observed to be normal or suppressed in acute stress (14), severe systemic illness (15), anorexia nervosa (16), and Cushing's syndrome (17), which are otherwise characterized by elevated cortisol levels. Increased adrenal androgen production may also be seen in association with hyperprolactinemia (13), although the majority of patients with this disorder have normal androgen levels. D H E A and DHEA-S are converted peripherally into A and then into the potent androgens testosterone and D H T as well as to estrogens. Under normal physiologic conditions, only 1% to 2% of total circulating testosterone is free or biologically available. The rest is bound by sex hormone-binding globulin (SHBG) and albumin, with SHBG binding 66% of total circulating testosterone (18). The binding affinity for steroids bound by SHBG is D H T > testosterone > androstenediol > estradiol > estrone (18). SHBG also weakly binds DHEA, but not DHEA-S (18). Therefore, variations in the plasma levels of SHBG have a significant impact on the amount of free or bioavailable testosterone. Elevations in
estradiol and thyroxine increase SHBG, whereas increases in testosterone, glucocorticosteroids, growth hormone, and insulin suppress SHBG production. In women, androgens may act directly via the androgen receptor or indirectly after conversion to estrogen. Androgens are the precursor hormones for estrogen production, not only in the ovaries but also in extragonadal tissues, including bone, adipose, and brain (19). Therefore, maintenance of physiologic circulating androgen levels in women ensures an adequate supply of substrate for estrogen biosynthesis in extragonadal sites, such as bone, in which high tissue estrogen concentrations may be required physiologically (for example, maintenance of bone mineralization and prevention of bone loss). That the primary source of estrogen in extragonadal sites is that produced locally from androgens also explains the apparent threshold dose of estrogen replacement postmenopausally below which bone loss continues, and the observation that whereas standard estrogen replacement therapy has little effect on libido (20,21), most parameters of sexuality improve when extremely high doses of estrogen are administered (22). Furthermore, this hypothesis may also explain the gender discrepancy between the development of osteoporosis and dementia, which occur much later in life in men than in women because men maintain adequate circulating testosterone levels for the extragonadal production of estrogen in bone and brain well into their later years.
III. PHYSIOLOGIC AND NONPHYSIOLOGIC CAUSES OF T E S T O S T E R O N E DEFICIENCY IN W O M E N The mean circulating levels of total and free testosterone decline continuously with increasing age from the early reproductive years, such that the levels of women in their forties are approximately half of those of women in their twenties (1,23). Androstenedione and DHEA-S levels also fall linearly with age (1), and this may contribute to the decline in the level of their main metabolite, testosterone. In the late reproductive years there is failure of the midcycle rise in free testosterone that characterizes the menstrual cycle in young ovulating women (11). This occurs despite preservation of normal free testosterone levels at other phases of the cycle. The mean plasma concentrations of testosterone in women transiting the menopause are also significantly lower than younger ovulating women sampled in the early follicular phase. Acutely, across the perimenopausal period, neither A, DHT, nor the ratio of total testosterone to SHBG (the free androgen index, or FAI) appear to change (1,24). Following menopause, direct ovarian secretion appears to account for up to 50% of testosterone production, with the
CHAPTER 55 Intervention: Androgens
adrenal gland being a less important source (25). Ovarian stromal hypertrophy and hyperplasia may persist or develop after menopause, probably secondary to elevated LH levels and individual sensitivity, resulting in increased testosterone production (20). Alternatively, the ovaries may become fibrotic and a poor source of sex steroids, and in such women the adrenal gland becomes the main androgen source following menopause (20). A sudden decline in androgens is a feature of ovariectomy, with both testosterone and A decreasing acutely by about 50% (25). Other iatrogenic causes of testosterone deficiency include chemical ovariectomy; for example, the use of gonadotropin-releasing hormone (GnRH) agonists and antagonists for the treatment of fibroids or endometriosis, postchemotherapy or radiotherapy, and the administration of exogenous estrogens or glucocorticosteroids. In general, circulating free testosterone is suppressed in women using either the combined oral contraceptive pill or oral estrogen replacement therapy (26,27). This is a result of an increase in SHGB combined with suppression of LH production by the pituitary and, hence, decreased stimulus for the ovarian stromal production of testosterone. These effects may be amplified in older women whose overall androgen production is declining. Treatment with oral glucocorticosteroids results in ACTH suppression and, hence, reduced adrenal androgen production (28). In the premenopausal years, other pathophysiologic states such as hypothalamic amenorrhea and hyperprolactinemia are characterized by low circulating testosterone levels and bone loss. Similarly, women with premature ovarian failure experience significant loss of bone that not uncommonly persists despite adequate standard dose estrogen-progestin therapy (29), and it is likely that young women with either ongoing hypothalamic amenorrhea or premature ovarian failure require testosterone replacement to prevent progressive bone resorption.
IV. CLINICAL CONSEQUENCES OF A N D R O G E N INSUFFICIENCY A N D EVIDENCE FOR BENEFITS OF TESTOSTERONE REPLACEMENT IN W O M E N
A. Sexual Dysfunction Multiple interacting factors influence libido and the frequency and enjoyment of sexual activity in women. These include general health stares, the physical and social environment, education, past experiences, current expectations, and the cultural milieu. Prevailing myths are that sexuality in women declines with increasing age and that older women mostly continue to be sexually active in order to
801 please their partner. However, one study of women in nursing homes reported that 25% percent of women over the age of 70 masturbate (30). Most women who have a natural menopause do not report loss of sexual desire, erotic pleasure, or orgasm. However, there is an age-related reduction in sexual frequency among women and lessening of coital frequency associated with the menopausal transition independent of age (31). Androgens appear to be important in female sexuality, with reduced androgen levels in the late reproductive years and beyond contributing to the decline in sexual interest experienced by a proportion of women. Compared with premenopausal women, women in their postmenopausal years report fewer sexual thoughts or fantasies, have less vaginal lubrication during coitus, and are less satisfied with their partners as lovers (31). A low testosterone level is most closely correlated with reduced coital frequency (32). In the Melbourne Women's Midlife Health Study, 62% of women aged 45 to 55 years reported no change in sexuality in the preceding year, although 31% reported a decrease (33). In the latter group, the decline was significantly and adversely associated with menopause rather than with age. In a study of sexagenarian women, the only hormone positively correlated with sexual desire was circulating free testosterone (34). Estradiol and testosterone are both present in the human female brain. The highest concentrations of estradiol have been reported in the hypothalamus and the preoptic area, and the highest concentrations of testosterone have been reported in the substantia nigra, hypothalamus, and preoptic area (35). The concentration of testosterone is several times higher than estradiol in each of these regions, with the highest ratio of testosterone to estradiol demonstrated in the hypothalamus and preoptic area. This distribution corresponds with high aromatase activity found in these regions in animals (36). Thus, androgen effects within the central nervous system may be mediated directly via androgen receptors and as a consequence of local aromatization of androgen precursors to estrogen. Support for specific direct androgen actions within the brain comes from cross-sex H T studies in transsexuals (37). In one such study, the administration of androgens to female-to-male transsexuals led to increases in sexual motivation and arousability, whereas the combination of antiandrogens and high-dose estrogen given to male-to-female transsexuals had the opposite effect (37). Androgen treatment of female-to-male transsexuals also resulted in improved visuospatial ability, lessened verbal fluency, and increased anger readiness, while the reverse, including an increased tendency to indirect angry behavior, was seen in the male-to-female subjects. We have recently reported in a randomized controlled trial that aromatase inhibition does not interfere with the favorable effects of exogenous testosterone on libido and well-being in postmenopausal women (38).
802 Research addressing the effect of women anticipating an adverse influence of menopause on their sexuality has shown a weak correlation between anticipated changes in sexuality and what actually occurs. The greatest predictor of postmenopausal sexual satisfaction is the quality of the sexual aspects of a woman's life in her premenopausal years, although having a satisfying sexual relationship prior to menopause does not preclude a woman from experiencing androgen-responsive sexual dysfunction subsequently. Undoubtedly, the availability of a sexual partner and the effects of age on the health and sexual interest of that person contribute to the apparent decline in sexuality among some women with increasing age. Estrogen replacement improves vasomotor symptoms, vaginal dryness, and possibly general well-being, but it has little effect on libido (39,40). Oral estrogen replacement therapy improves sexual satisfaction in women with atrophic vaginitis as a cause of their dyspareunia, but women lacking this symptom benefit little or not at all (41). In some cases, the commencement of estrogen replacement may suppress free testosterone levels and cause relative testosterone deficiency in the postmenopausal years, as described earlier, and hence precipitate a decline in sexual desire. Several randomized controlled trials have demonstrated improvements in several aspects of sexuality in postmenopausal women treated with exogenous testosterone, over and above the effects achieved with estrogen alone (4-6,22). Sustained improvements in intensity of sexual drive, arousal, frequency of sexual fantasies, satisfaction, pleasure, and relevancy were observed in these studies of postmenopausal women. In contrast to the consistently observed increases in sexual motivation, improvements in coital frequency and orgasm vary considerably among studies. The failure for testosterone supplementation to improve coital frequency may be attributed to the participants in these studies being women with very long-term relationships, with established sexual patterns, such that change is less likely. Frequency of sexual activity as a measure of efficacy of androgen therapy in women can also be misleading, as often this is more a measure of a male partner's sexual appetite rather than the woman's. The only study in which the addition of testosterone replacement did not show any benefit over estrogen alone was in women being treated for generalized menopausal symptoms rather than for low libido (42). Testosterone therapy should also be considered as part of the management of young women with premature menopause, particularly Turner syndrome. Women who are sexually active when they develop premature ovarian failure are often very disturbed by their diminished libido. Alternatively, young women who have not become sexually active and have either primary or secondary premature ovarian failure should be fully informed about the option of androgen replacement or perhaps in some instances
SUSAN R. DAvis
offered low-dose androgen replacement as a component of their hormone replacement regimen. Whether premenopausal women who complain of loss of libido and have low bioavailable testosterone levels should be offered testosterone therapy is controversial. We have shown in a randomized, placebo-controlled crossover study that testosterone therapy improves libido, mood, and wellbeing in women age 35 to 45 years who present with low libido (43). Furthermore, such women appear to be at greater risk of premenopausal bone loss (44). At present this is a subgroup of women for whom clear recommendations cannot be made but for whom research is ongoing. Testosterone replacement is unlikely to benefit women in whom other factors play a dominant role in their sexual dysfunction. Clinically discerning this may be quite straightforward (e.g., postovariectomy) but may be extremely difficult in some women, and therefore an insightful psychosocial and sexual history is essential when evaluating the appropriateness of testosterone therapy in a woman.
B. Androgens and Bone Androgenic steroids have an important physiologic role in the development and maintenance of bone mineralization in women and men, although the mechanisms of androgen action on bone are still a matter of debate. The skeletal effects of androgens appear to be mediated in part via the estrogen receptor after local aromatization of androgens to estrogen, and mutations in either the estrogen receptor gene or the aromatase gene are associated with osteoporosis (45,46). Abundant aromatase activity has been reported in fetal osteoblasts and cell lines of osteoblastic origin (47). There is also evidence that androgens act directly on bone. Androgen receptors have been demonstrated in human osteoblast-like cell lines, and androgens have been shown to directly stimulate bone cell proliferation and differentiation (48,49). D H T has been shown to increase alkaline phosphatase activity, type 1 procollagen synthesis, and insulin-like growth factor II messenger RNA in the SAOS2 cell line (50). Androgen receptor mRNA is expressed in human osteoblasts, osteocytes, hypertrophic chondrocytes, marrow mononuclear cells, and vascular endothelial cells within bone, but not in osteoclasts (51). The pattern and number of cells expressing the androgen receptor are similar in female and male tissues (51). In addition, androgen receptors are upregulated in osteoblastic cells by both testosterone and D H T in vitro (52). Total and bioavailable testosterone and DHEA-S, not estradiol, are the greatest predictors of bone mineral density (BMD) and bone loss in premenopausal women (44,53). Women who experience bone loss confined to the hip prior to menopause, a not uncommon occurrence, have lower total and free testosterone concentrations, by
CHAPTER 55 Intervention: Androgens 14% and 22%, respectively, than those who do not significantly lose bone (44). Consistent with these findings, hyperandrogenic women have higher BMD, after correction for body mass index, than their normal female counterparts (53). In the premenopausal years, BMD is also strongly positively correlated with body weight (54). In obesity, SHBG is suppressed with a resultant increase in free testosterone (55), and this may partially explain the relationships between obesity, free testosterone, and increased BMD, with the greater endogenous levels of biologically active free testosterone in more corpulent women directly enhancing bone mass. Androgen insufficiency may also be a factor underlying bone loss in young women with premature ovarian failure. Despite adequate standard dose estrogen-progestin therapy, two-thirds of such women have significantly reduced BMD to levels associated with an increased risk of hip fracture. Of these, 47% have reductions in BMD within 18 months of their diagnosis (29). In postmenopausal women, low circulating free testosterone is predictive of subsequent height loss (a surrogate measure of vertebral compression fractures) and hip fracture (56,57). Circulating D H E A and DHEA-S are positively correlated with BMD in aging women (58-60), and the progressive decline in D H E A with increasing age is believed to contribute to senile osteoporosis. It is unlikely that these adrenal pre-androgens directly influence bone metabolism; rather, their effects are mediated indirectly following conversion to estradiol, A, or testosterone. Suppression of adrenal production of D H E A and DHEA-S with chronic glucocorticosteroid therapy may also contribute to the pathogenesis of osteoporosis and osteopenia, which are known complications of this therapy in women and men. D H E A or testosterone administration may be effective in preventing and/or treating this common and serious side effect of glucocorticosteroid therapy. Women treated with oral D H E A have restoration of circulating A, DHT, and testosterone to premenopausal levels, as well as increases in D H E A and DHEA-S, with no changes in circulating levels of estrone or estradiol from baseline (61). Similarly, the daily percutaneous administration of 10 mL of a 20% D H E A solution results in an increase in circulating total testosterone of approximately 50%, with no consistent effect on estradiol or estrone (62). Circulating DHEA-S, but not estradiol, in postmenopausal women is positively correlated with BMD, and daily application of a 10% D H E A cream has been reported to increase hip BMD in older women (62). Thus, D H E A therapy may prove in time to be an alternative to administration of testosterone replacement to androgen-deficient women. Studies of both oral and parenteral estrogen and estrogen + testosterone therapy in postmenopausal women have shown beneficial effects of testosterone replacement on BMD (22,63,64). Oral esterified estrogen-methyltestosterone
803 treatment has been shown to increase spinal BMD over a 2-year period, in contrast to estrogen-only therapy, which prevented bone loss (63). The estrogen-methyltestosterone combination not only suppresses biochemical markers of bone reabsorption (as seen with estrogen alone) but is also associated with increases in markers of bone formation (65). Treatment of postmenopausal women with nandrolone decanoate has been shown to increase vertebral BMD and has been used successfully for many years to treat osteoporosis (66). Combined estradiol and testosterone replacement with subcutaneous implant pellets increases bone mass in postmenopausal women (64,67), with the effects in the hip and spine being greater than with estradiol implants alone (22) (Figs. 55.1 and 55.2). Thus, it appears that estradiol alone has an antireabsorptive effect on bone in postmenopausal women, whereas the addition of testosterone, either orally or parentally, enhances bone formation. This idea is further supported by the previously mentioned studies of men, either with a mutation of the gene encoding the aromatase enzyme (45,46,68,69) or a mutation of the estrogen receptor (45). Increasing BMD is only clinically important if it is associated with enhanced mechanical strength and a reduced fracture rate. As yet, no studies have addressed the impact of androgens on fracture incidence. However, the effects of androgens on the mechanical properties of bone have been studied in female cynomolgus monkeys (70). In this primate model, increases in intrinsic bone strength and resistance to mechanical stress were associated with increased BMD following testosterone therapy. Treatment also resulted in increased bone torsional rigidity and bending stiffness. In summary, current data indicate that androgen therapy, in the form of testosterone, and possibly DHEA, is potentially an effective alternative for the prevention of bone loss and the treatment of osteopenia and osteoporosis, and warrants further investigation. As prospective data confirming a reduction in fracture rate with androgen therapy are lacking, specific guidelines cannot be given for this sole indication. The addition of androgen to standard hormone replacement regimens may be necessary to prevent bone loss in young hypogonadal women in whom androgen insufficiency is aggravated by exogenous estrogen. The use of testosterone to prevent bone loss in individuals on long-term glucocorticosteroids has not been studied but also warrants further research.
C. Testosterone and Autoimmune Disease Although immunogenetic factors are the major determinants of the development of immune-mediated diseases, gender and age also play a role. Androgens appear to suppress both cell-mediated and humeral immune responses, and it has been proposed that higher testosterone levels in men may be protective against autoimmune disease (71-75).
804
SUSANR. DAVIS DHEA, DHEA-S, and testosterone. Further evaluation of potential beneficial effects of testosterone administration in women with rheumatoid arthritis and systemic lupus erythematosus is needed.
D. Effects on Body Composition
FIGURE 55.1 and 55.2 The effects of hormonal implants on bone mineral density (BMD) (g/cm2): lumbar spine (L1-L4), femoral trochanter (Troc), estradiol (E), and estradiol plus testosterone (E+Te). Error bars represent standard error of the differences between means (SED). Inner error bars are used to compare means between times for the same treatment. The comparison between the treatment groups is made with the outer error bars. If error bars do not overlap (i.e., differ by more than 2 SEDs), the means are significantly different by a p value of at least 0.05.
The direct administration of testosterone replacement has resulted in symptomatic improvement in postmenopausal women with rheumatoid arthritis (74), and both premenopausal and postmenopausal women have been observed to have reductions in disease activity with D H E A therapy associated with increases in levels of circulating
In postmenopausal women, neither measured nor estimated free testosterone is associated with waist-to-hip ratio measurements, and there does not appear to be a direct relationship between androgens and visceral adiposity in this population (76). Contrary to popular belief, neither oral nor parenteral hormone therapy significantly increases body weight (77,78). There is also no evidence that the addition of testosterone replacement causes weight gain (4,5,22). With increasing age, women tend to lose muscle mass and replace it with increased fat mass, coupled with an overall tendency for weight gain in most Western societies. Increases in fat-free mass and a reduced-fat mass/fat-free mass ratio have been reported in postmenopausal women treated with concurrent estrogen-testosterone therapy (79,80). Gain in fat-free mass probably reflects increased muscle mass. As aging is associated with loss of muscle mass, this is a beneficial effect of testosterone therapy in the older woman and may contribute to preservation of muscle strength and skeletal stability. However, the ethics of treating an older woman who participates in competitive (Master's) sports with testosterone replacement for medical indications has become a significant issue that needs to be resolved. Testosterone levels are frequently lower in HIV-positive premenopausal women. The effects of physiologic testosterone replacement using a transdermal patch on fat-free mass, muscle strength, and weight maintenance have been studied in HIV-infected women with relative androgen deficiency and weight loss. Administration of the 300-~tg patch was found to be safe and effective in raising testosterone levels into the mid- to high-normal range, but it did not significantly increase fat-free mass, body weight, or muscle performance in HIV-infected women with low testosterone levels and mild weight loss (81).
V. POTENTIAL RISKS OF TESTOSTERONE REPLACEMENT IN W O M E N
A. Cardiovascular Disease Risk The differences in cardiovascular disease (CVD) morbidity and mortality between men and women and the increase in incidence in CVD in women in later life have been partially attributed to gender differences in absolute and
CHAPTER 55 Intervention: Androgens relative amounts of estrogens and androgens and the shift to a less favorable hormonal milieu in women following menopause. However, whether androgen levels in women contribute to the increased CVD risk in older women remains contentious (82-85). The particular concerns are the potential adverse metabolic effects of androgens with respect to lipid and lipoprotein metabolism and vascular function. Low SHBG levels have been reported as related to elevated CVD risk, as has a higher free androgen index (FAI: total testosterone/SHBG ratio) (84,85). However, the significance of the latter findings in terms of a direct adverse androgen effect is limited by the possibility that FAI is a proxy risk factor for CVD risk, where the real risk factor is SHBG (85a). Recent studies have demonstrated that inflammatory mechanisms underlie the process of arterial thrombus formation following atheromatous plaque rupture (86). Populationbased studies have shown that circulating inflammatory markers are strongly associated with cardiovascular events, even in individuals with normal lipoprotein profiles (87-89). Of the inflammatory markers, C-reactive protein (CRP) provides the strongest risk prediction for CVD in women (90,91). Elevated levels of CRP are also associated with obesity and weight gain (92) and predict the risk of developing type 2 diabetes mellitus (93). However, determination of CRP contributes more prognostic information to CVI) risk in women than serum lipids or features of the metabolic syndrome alone (89,94). 1. TESTOSTERONE TREATMENT AND CARDIOVASCULARRISK MARKERS
Menopause, both natural and surgically induced, is associated with the development of a more adverse lipoprotein profile that is unrelated to endogenous testosterone levels (95). Postmenopausal estrogen therapy lowers total and low-density lipoprotein (LDL) cholesterol (77,78), and these favorable effects are not diminished with either oral or parenteral testosterone therapy (5,22,38). Parenteral testosterone replacement does not affect high-density lipoprotein (HDL) cholesterol (22,38); however, H D L cholesterol and apolipoprotein A1 decrease significantly when oral methyl testosterone is administered with oral estrogen (96-98).Whereas oral estrogen replacement is associated with increased triglyceride levels, concomitant administration of oral methyltestosterone has resulted in a reduction in triglycerides (98). Measurement of circulating lipids is a clinical surrogate for lipoprotein-lipid metabolism. A more direct measurement of lipid metabolism is probably a better indicator of the effects of exogenous steroid therapy on cardiovascular risk. Combined oral esterified estrogen and methyltestosterone therapy reduces arterial LDL degradation and cholesterol ester content in cynomolgus monkeys, which does not
805 differ from the effects observed when estrogen is given alone (99). Combined estrogen and methyltestosterone therapy is also associated with reduced plasma concentrations of apolipoprotein B, reduced LDL particle size, and increased total body LDL catabolism (99). Small LDL particles are more susceptible to oxidation and are hence considered to be more atherogenic. However, because estrogens appear to increase oxidative modification of LDL in the arterial wall, the reduction in LDL particle size observed with both oral estrogen and combined therapy may not be deleterious, but rather may merely reflect selective removal of large LDL particles from the circulation. In a recent randomized controlled trial, testosterone gel therapy was not associated with any changes in highsensitivity CRP or lipoprotein (a) in postmenopausal women receiving concurrent estrogen therapy (38). 2. EFFECTS OF ANDROGENS ON THE MYOCARDIUM AND VASCULARFUNCTION
The incidence of coronary heart disease in postmenopausal women is not correlated with levels of circulating testosterone or DHEA-S (100,101). Intracoronary testosterone administration to anesthetized male and female dogs induces increases in coronary artery cross-sectional area peak flow velocity and calculated volumetric blood flow, which is blocked by pretreatment with an inhibitor of nitric oxide synthesis (102). The beneficial effects of estrogen therapy on coronary artery reactivity in cynomolgus monkeys is not lost with the addition of oral methyltestosterone (103). In postmenopausal women using long-term estrogen therapy, exogenous testosterone implants improve both endothelial-dependent (flow mediated) and endotheliumindependent (glyceryl trinitrate [GTN]-mediated) brachial artery vasodilation (104). Cardiac myocytes and fibroblasts contain functional alpha and beta estrogen receptors, and cardiac myocytes express cyp450 aromatase (105). Incubation of cardiac myocytes with A or testosterone results in transactivation of an estrogen receptor-specific reporter, with A resulting in a significantly higher induction than testosterone (105,106). Furthermore, both A and testosterone upregulate inducible nitric oxide synthase in cardiac myocytes (106). The capacity for cardiac myocytes to synthesize estrogen from androgenic precursors and activate downstream target genes is greater in cells from female animals compared with male animals (106). Thus, evaluating the data at hand, the administration of testosterone to women does not appear to affect the key events in coronary artery lipid metabolism or coronary artery function. Parenteral testosterone replacement does not negate the favorable effects exerted by exogenous estrogen on lipid and lipoprotein levels, whereas oral methyltestosterone use not only opposes the HDL-cholesterol elevating effects of oral estrogen but also reduces HDL-cholesterol
806
SUSAN R.
levels below baseline values. It is not known whether this effect of oral methyltestosterone in the setting of concomitant lowering of total cholesterol, LDL-cholesterol, and triglyceride levels is detrimental. Certainly a high cardiac risk profile, which includes a low HDL-cholesterol level, central obesity, and low SHBG, should be considered a relative contraindication to testosterone therapy.
DAVIS
TABLE 55.2 Contraindications to Testosterone Therapy Relative
Severe acne Moderate to severe hirsutism Androgenic alopecia Circumstances in which enhanced libido would be undesirable Very low sex hormone globulin level Absolute
B. Androgens and Breast Cancer Studies of possible relationships between endogenous androgen levels and breast cancer have been extensively reviewed (107). In premenopausal women, there are inconsistent results from two cross-sectional studies (108,109), and two prospective case-control studies have found no association between breast cancer risk and total testosterone levels (110,111). The risk of breast cancer is also not increased in women with polycystic ovarian syndrome (PCOS) who have chronic estrogen exposure and androgen excess (107). Results are inconsistent in both cross-sectional studies and prospective studies of postmenopausal women (107). Most studies are limited by measurement of only total (not free) testosterone, use of insensitive assays, and failure to consider the diurnal variation of testosterone or oophorectomy status. In experimental studies, testosterone action is antiproliferative and pro-apoptotic and is mediated via the androgen receptor (AR), despite the potential for testosterone to be aromatized to estrogen (107). Animal studies suggest that testosterone may serve as a natural, endogenous protector of the breast and limit mitogenic and cancer-promoting effects of estrogen on mammary epithelium. Androgen receptors are found in more than 50% of breast tumors (112) and are associated with longer survival in women with operable breast cancer and a favorable response to hormone treatment in advanced disease. There is also evidence that the mechanism by which high-dose medroxyprogesterone acetate (MPA) exerts a negative effect on breast cancer growth is mediated via the androgen receptor (113). No studies of testosterone therapy have evaluated breast cancer risk.
C. Testosterone Replacement and Clinical Side Effects The potential masculinizing effects of androgen therapy include development of acne, hirsutism, deepening of the voice, and excessive libido. These cosmetic side effects are uncommon if supraphysiologic hormone levels are avoided. Other adverse metabolic effects are also uncommon with judicious therapy. Fluid retention appears to be more idiosyncratic than dose related. Hirsutism, androgenic alopecia, and/or acne are relatively strong contradictions to androgen
Pregnancy or lactation Known or suspected androgen-dependent neoplasia
therapy. Enhancement of libido is currently the most common indication for testosterone therapy; however, circumstances in which this would be an undesirable effect are a relative contraindication to therapy. Absolute contraindications include pregnancy and lactation, as well as known or suspected androgen-dependent neoplasia (Table 55.2). Hepatocellular damage has been reported with high-dose oral 1713-alkylandrogens, but not in currently used doses or other oral formulations (114).
D. Summary of the Potential Risks
of Testosterone Therapy Syndromes of endogenous androgen excess are clearly associated with increased cardiovascular risk, perturbations in lipid and carbohydrate metabolism, a more android weight distribution, and virilization. In contrast, clinical data at hand do not indicate testosterone maintained close to, or within, the normal female reproductive range has any adverse metabolic consequences (Table 55.3). With respect to breast cancer risk, whether women who have endogenously high testosterone levels are at increased risk is controversial, and there is no evidence that there is an increased in risk with so-called "physiologic" exogenous androgen replacement therapy after menopause. Physicians should
TABLE 55.3 Potential Dose-Related Side Effects of Androgen Therapy in Women Masculinizafion: hirsutism, acne, temporal balding, voice deepening, clitoromegaly Fluid retention (more often idiosyncratic) Lipids: Oral testosterone may adversely affect serum levels of HDL-cholesterol and apolipoprotein A1. Drug interactions: C17 substituted derivatives of oral testosterone may decrease anticoagulant requirements. Androgens may elevate serum levels of oxyphenbutazone and may rarely affect insulin requirements in diabetic patients. Hepatocellular damage has been associated with high-dose 17otalkylandrogens given orally.
CHAPTER 55 Intervention:
Androgens
always be cognizant of the anxiety most women have about breast cancer when prescribing any hormone therapy.
VI. W H I C H W O M E N ARE M O S T LIKELYTO BENEFIT FROM TESTOSTERONE? A list of potential indications for testosterone therapy is provided. However, indications for androgen therapy for a woman is most often low libido and diminished well-being, and hence based on clinical assessment, with evaluation of the outcome of treatment based on the subjective selfassessment of response and reporting by the patient. Biochemical measurements are of limited value, as there is no cut-off level for total or free testosterone that distinguishes women with low libido from other women (2). The clinical settings in which the administration of testosterone is most likely to enhance a woman's health and well-being are listed in Table 55.1. Future indications may also include the prevention or treatment of bone loss, particularly iatrogenic bone loss from glucocorticosteroid therapy or premenopausal bone loss. Some women present seeking clinical assistance for low libido and/or inadequate restoration of well-being despite apparently sufficient estrogen/progestogen therapy. Others may volunteer such problems while attending their physician for another reason. A significant proportion of women will not raise the issue of diminished libido because they find the topic awkward. Sadly, many young women who suffer low/absent libido following treatment of a malignancy express difficulty discussing the issue with their physician as they feel their problem will be perceived as trivial relative to their disease recovery or remission. Following treatment with chemotherapy or radiotherapy, the symptoms of the iatrogenic menopause and androgen deficiency, including fatigue, loss of well-being, depression, and reduced libido, can be difficult to distinguish from the overall physical toll of cancer treatment, and frequently premature menopause and associated androgen insufficiency go undiagnosed and untreated. Therefore, all "at-risk" women should be directly questioned about the symptoms of androgen deficiency and made aware of the therapeutic possibilities available. Testosterone therapy is becoming a more common component of hormone therapy for the restoration of sexual function in women who have undergone a surgical menopause. Testosterone therapy in women who have undergone natural menopause and especially in women who have premature ovarian failure continues to be a controversial therapeutic issue. It is essential that physicians offering this therapy be sensitive to the enormous variations in women's knowledge, expectations, sexual practices, and needs and always tailor treatment to the needs of the individual.
807 The management of young women with premature menopause, particularly those with primary amenorrhea (e.g., Turner syndrome) who have never been sexually active, is more difficult. Testosterone replacement should be considered for such women experiencing persistent fatigue, lack of well-being, and low libido despite adequate estrogen and progestogen replacement. However, it is obviously difficult for a woman to identify herself as having inadequate libido when she has never experienced what may be "normal." Alternatively, young women who have not been sexually active should have the option of testosterone replacement explained to them. Whether young women with premature ovarian failure should be recommended to use testosterone replacement to prevent bone loss, specifically from the neck of femur, is yet to be established.
VII. ASSESSMENT OF POTENTIAL CANDIDATES FOR A N D R O G E N THERAPY All women should be carefully screened for depression and relationship issues as a cause of their sexual difficulties. Similarly, the presence of chronic illness does not exclude a hormonal cause. Indeed, a hormonal cause may be more likely in women with illness or therapy that causes adrenal suppression. A complete gynecologic history should be taken. History should also identify possible iron deficiency, thyroid disease, and galactorrhea. A general physical examination should include assessment of thyroid status and presence of anemia or galactorrhea. Gynecologic examination should include a pelvic examination with attention to signs of vaginal atrophy, size of introitus, presence of discharge or evidence of infection, vulvodynia, and deep tenderness. Women presenting with low libido and fatigue should have routinely have the following measured: iron stores, thyroid-stimulating hormone to exclude subclinical thyroid disease, and, on clinical suspicion, a screen for autoimmune disease causing chronic fatigue. A history of treatment for hirsutism, acne, and androgenic alopecia should be sought as well as these features on physical examination. Measurement of total or free testosterone or DHEA-S does not distinguish patients who will benefit from treatment but does enable exclusion of women with biochemical androgen excess. Free or bioavailable (non-SHBG-bound) testosterone measures are the most reliable indicators of tissue testosterone exposure. Ideally blood should be drawn between 8:00 and 10:00 AM due to the diurnal variation of testosterone, which results in higher levels at this time. In premenopausal women, testosterone is at its nadir during the early follicular phase, with small but less significant variation across the rest of the cycle. Thus, blood should be
808
SUSAN R. DAvis
drawn after day 8 of the cycle, and preferably before day 20. A serum sample is preferred to plasma.
VIII. H O W SHOULD TESTOSTERONE REPLACEMENT BE PRESCRIBED? Currently no form of testosterone therapy is approved by the U.S. Food and Drug Administration (FDA) for the treatment of loss of libido in women; however, oral methyltestosterone can be prescribed for menopausal symptoms that do not respond to estrogen therapy alone. Testosterone implants have been approved for therapy for women in the United Kingdom. The current international therapeutic options in terms of androgen therapy are listed in Table 55.4. To achieve a good therapeutic response in terms of enhanced libido with parenteral testosterone, levels usually need to be restored to within the normal physiologic range for young ovulating women. The oral estrogen/androgen preparation available in the United States is in two strengths, esterified estrogen 0.625 mg plus methyltestosterone 1.25 mg, or esterified estrogen 1.25 mg plus methyltestosterone 2.5 mg. Methyltestosterone is not available in most other countries because liver damage has been reported with long-term, high-dose therapy. More recent data do not support any short-term (12 months) detrimental effects of the available doses of methyltestosterone combined with estrogen on hepatic enzymes or blood pressure. It has been observed that women who use esterified estrogens combined with methyltestosterone report a lower instance of nausea compared with women receiving conjugated equine estrogen (CEE) alone (114). Testosterone undecanoate is an oral androgen prescribed for replacement therapy in hypogonadal men. Its clinical use in women has been little studied, although in some countries it is quite widely prescribed for women. It is believed to be absorbed via the lymphatics; hence, for optimal absorption it is recommended that testosterone undecanoate be ingested
TABLE 55.4
with fat, such as a glass of milk. However, supraphysiologic peaks of total testosterone have been reported with a dose as low as 20 mg (115). At present, testosterone undecanoate use cannot be recommended. There has been considerable clinical experience with the administration of testosterone implants in postmenopausal women, particularly in the Commonwealth countries, and these are approved for use in women in the United Kingdom. These implants are fused crystalline implants, 4 to 5 mm in diameter, containing testosterone BP (British Pharmacopoeia) as the active ingredient. A dose of 50 mg is extremely effective and does not cause virilizing side effects. This dose can be obtained by bisecting a 100-mg implant under sterile conditions. The implant is inserted subcutaneously (under local anaesthesia), usually into the lower anterior abdominal wall, using a trocar and cannula. This therapy provides a slow release of testosterone with an approximate duration of effect of 3 to 6 months for a 50-mg implant. There is marked individual variation in this period; therefore, testosterone levels must be carefully monitored, with a testosterone level measured prior to the administration of each subsequent implant. The author recommends that additional testosterone implants should not be inserted unless total testosterone is below the normal range for young women. Rarely is a 100-mg testosterone implant necessary to achieve an adequate therapeutic effect. Mean circulating testosterone levels approximately three times the upper limit of normal have been reported 4 weeks following the administration of 100-mg testosterone pellets. In contrast, 6 weeks after the insertion of a 50-mg testosterone implant, the mean circulating testosterone levels in postmenopausal women are just above the upper limit of normal for young ovulating women. Mixed testosterone esters, 50 to 100 mg, are occasionally administered 4 to 6 weekly as an intramuscular injection to women with androgen deficiency symptoms. Clinically this therapy results in a more rapid onset of effects, such that women report enhanced libido within 2 to 3 days of treatment. Some women report a heightened sense of agitation with this
Androgen Replacement Therapy Formulations Used for Women Dose range
Frequency
Route
Methyltestosterone* (in combination with esterified estrogen) Testosterone undecanoate** Nandrolone decanoate** Mixed testosterone esters** Testosterone implants
1.25-2.5 mg 40-80 mg 25-50 mg 50-100 mg 50 mg
Oral Oral Intramuscular Intramuscular Subcutaneous
Transdermal testosterone patcht
300 tag
Daily Daily or alternate days Every 6-12 weeks Every 4-6 weeks 3-6 Monthly Every 3.5 days
*Currently availablein the United States. **MeansNOT recommendedfor generaluse. *Undergoing clinical trial.
Topical
809
CHAPTER 55 Intervention: Androgens therapy. The pharmacokinetics of mixed testosterone esters administered intramuscularly to women have not been studied; however, many women report increased acne and occasional clitoromegaly with this therapy. The recent development of a transdermal testosterone matrix patch, specifically for use in women, may provide a new option for women requiring testosterone replacement. The patch, which has now been evaluated in phase III clinical trials, has been shown to improve the number of self-reported satisfactory sexual events experienced by surgically menopausal women (3,5). The patch delivers 300 Ixg of testosterone per day over a 3- to 4-day period (twiceweekly application). This therapy has now been approved for marketing in Europe. Nandrolone decanoate, a weakly aromatizable androgen, is approved in some countries for the treatment of postmenopausal osteoporosis. The dose, administered intramuscularly, should not exceed 50 mg, and the frequency of treatment is best titrated against the patient's gross build. It is prudent that treatment not be given more than every 6 weeks, and as these women are usually elderly and not on estrogen therapy, they should be very carefully monitored for masculinization. In most women, treatment with nandrolone decanoate results in cessation of bone loss over time and, in some women, an increase in BMD. Alternatives that are not currently or generally available include transdermal testosterone as a patch, spray, cream, gel, or parenteral testosterone via a vaginal ring. Although the transdermal preparations may be regionally available with a specific prescription from compounding pharmacists, there are no pharmacokinetic data available or published clinical experience for their preparations. The contraindications to and side effects of testosterone replacement are listed in Tables 55.2 and 55.3. Again, it is emphasized that with judicious dosing and careful patient monitoring, side effects of testosterone replacement are rare. All postmenopausal women treated with testosterone should be using concurrent estrogen therapy until such time as data are available for the safety of testosterone alone. The only clinical situation in which androgen therapy has been used in postmenopausal women without concurrent estrogen is the administration of the anabolic steroid, nandrolone decanoate. Women treated with this steroid must be very carefully monitored for virilization, which may occur when this compound is given unopposed by estrogen.
IX. CONCLUSIONS Testosterone levels in women fall substantially during the reproductive years, with little further change at the time of spontaneous menopause. A major fall in circulating testosterone occurs following bilateral oophorectomy. The biologic availability of the hormone is determined by the levels of
SHBG, with factors that raise SHBG levels leading to a fall in bioavailable testosterone. There is increasing interest in the contribution of androgen insufficiency to low libido and a variety of other symptoms. A substantial body of evidence indicates that testosterone is an important determinant of female sexuality and that androgen therapy enhances certain aspects of sexual function, particularly in women who have undergone oophorectomy. Many women are uneasy discussing their loss of sexual desire, particularly those who have undergone chemotherapy, and they often comment that they feel the issue will be viewed by their physician as trivial relative to their recovery or remission. Also, the symptoms of iatrogenic menopause can easily be attributed to side effects of chemotherapy or to other psychosocial factors in premenopausal women or in those who have undergone a premature menopause. Therefore, it is the treating physician's responsibility to facilitate discussion of sexuality in all "at-risk" women. Potential etiologic factors such as depression, poor relationship, and stress as well as other medical conditions should be addressed. The possibility of low androgen production as an underlying cause might then be considered. Physiologic-range androgen therapy may also have a role in the maintenance or restoration of bone mass, both following menopause and after pharmacologic glucocorticoid therapy. Further research into the biologic actions of androgens in bone and clinical studies of testosterone and fracture is needed to define the appropriate clinical application of androgens in the prevention and treatment of bone loss in women. Mthough more controversial, premenopausal women with either spontaneous or iatrogenic androgen deficiency also warrant consideration for androgen therapy, as do women experiencing glucocorticosteroid-induced bone loss and possibly premenopausal bone loss. At levels within or just above the physiologic range, testosterone in women does not appear to have major adverse effects on cardiovascular risk factors or on vascular function. Its actions, if any, in breast cancer risk remain uncertain. The inclusion of testosterone therapy in postmenopausal hormone therapy regimens is increasing but still limited by the lack of availability of preparations and formulations designed specifically for use in women. Advances, particularly involving transdermal technologies, are occurring in the effort to optimize the methods of androgen therapy in women, as well as in men.
References 1. Davison SL, Bell R, Donath S, Montalto J, Davis SR. Androgen levels in adult females: changes with age, menopause and oophorectomy. J Clin Endocrinol Metab 2005;90:3847. 2. Davis SR, Davison SL, Donath S, Bell R. Relationships between circulating androgen levels and self-reported sexual function in women. JAm MedAssoc 2005;294:91.
810 3. Buster JE, Kingsberg SA, Aguirre O, et al. Testosterone patch for low sexual desire in surgically menopausal women: a randomized trial. Obstet Gyneco12005;105:944. 4. Shifren JL, Braunstein G, Simon J, et al. Transdermal testosterone treatment in women with impaired sexual function after oophorectomy. N EngJ Med 2000;343:682. 5. Braunstein GD, Sundwall DA, Katz M, et al. Safety and efficacy of a testosterone patch for the treatment of hypoactive sexual desire disorder in surgically menopausal women: a randomized, placebo-controlled trial. Arch Intern Med 2005;165:1582-1589. 6. Simon J, Braunstein G, Nachtigall L, et al. Testosterone patch increases sexual activity and desire in surgically menopausal women with hypoactive sexual desire disorder.J Clin EndocrinolMetab 2005;90:5226-5233. 7. Labrie F, Belanger A, Luu-The V, et al. DHEA and the intracrine formation of androgens and estrogens in peripheral target tissues: its role during aging. Steroids 1998;63:322. 8. Haning JRV, Hackett R, Flood CA, et al. Plasma dehydroepiandrosterone sulfate serves as a prehormone for 48% of follicular fluid testosterone during treatment with menotropins. J Clin Endocrinol Metab 1993;76:1301. 9. Couzinet B, Meduri G, Lecce MG, et al. The post menopausal ovary is not a major androgen producing gland. J Clin Endocrinol Metab 2001;86:5060. 10. Judd HL, Yen SSC. Serum androstenedione and testosterone levels during the menstrual cycle.J Clin EndocrinolMetab 1973;36:475. 11. Mushayandebvu T, Castracane DV, Gimpel T, Adel T, Santoro N. Evidence for diminished midcycle ovarian androgen production in older reproductive aged women. Fertil Steri11996;65:721. 12. Vaitukaitis JL, Dale SL, Melby JC. Role of ACTH in the secretion of free DHA and its sulphate ester in man. J Clin Endo Metab 1969; 29:1443. 13. Vermeulen A, Ando S. Prolactin and adrenal androgen secretion. Clin Endocrinol (Oxf) 1978;8:295. 14. Parker LM, Eugene J, Farber D, Lifrale E, Lai M, Juler G. Dissociation of adrenal androgen and cortisol levels in acute stress. Horm Metab Research 1985;17:209. 15. Parker LN, Levin ER, Lifrat ET. Evidence for adrenocortical adaption to severe illness.J Clin Endo Metab 1985;60:947. 16. Zumoff B, Walsh B, Katz J. Subnormal plasma dehydroepiandrosterone to cortisol ratio in anorexia nervosa: a second hormone parameter of ontogenic regression. J Clin EndocrinolMetab 1983;56:668. 17. Cunningham SK, McKenna TJ. Dissociation of adrenal androgens and cortisol secretion in Cushing's syndrome. Clin Endo 1994;41:795. 18. Dunn JF, Nisula BC, Rodbard D. Transport of steroid hormones. Binding of 21 endogenous steroids to both testosterone-binding globulin and cortico-steroid-binding globulin in human plasma. J Clin Endocrinol Metab 1981;53:58. 19. Simpson ER, Davis SR. Minireview: aromatase and the regulation of estrogen biosynthesis--some new perspectives. Endocrinology 2001;142:4589. 20. Utian WH. The true clinical features of postmenopausal oophorectomy and their response to estrogen replacement therapy. S Afr MedJ 1972;46:732. 21. Campbell S, Whitehead M. Oestrogen therapy and the menopausal syndrome. Clin Obstet Gynaeco11977;4:31. 22. Davis SR, McCloud PI, Strauss BJG, Burger HG. Testosterone enhances estradiol's effects on postmenopausal bone density and sexuality. Maturitas 1995;21:227. 23. Zumoff B, Strain GW, Miller LK, Rosner W. Twenty-four hour mean plasma testosterone concentration declines with age in normal premenopausal women.J Clin Endocrinol Metab 1995;80:1429. 24. Burger HG, Dudley EC, Hopper DL. The endocrinology of the menopausal transition: a cross-sectional study of a population-based sample.J Clin Endo Metab 1995;80:3537.
SUSAN R. DAVIS 25. Judd HL, Lucas WE, Yen SSC. Effect of oopherectomy on circulating testosterone and androstenedione levels in patients with endometrial cancer. A n J Ob Gyneco11994;118:793. 26. Mathur RS, Landgreve SC, Moody LO, Semmens JP, Williamson HO. The effect of estrogen treatment on plasma concentrations of steroid hormones, gonadotropins, prolactin and sex hormone-binding globulin in post-menopausal women. Maturitas 1985;7:129. 27. Krug R, Psych D, Pietrowsky R, Fehm HL, Born J. Selective influence of menstrual cycle on perception of stimuli with reproductive significance. PsychosomMed 1994;56:410. 28. Abraham GE. Ovarian and adrenal contribution to peripheral androgens during the menstrual cycle.J Clin EndocrinolMetab 1974;39:340. 29. Anasti JN, Kalantaridou SN, Kimzey LM, Defensor RA, Nelson LM. Bone loss in young women with karyotypically normal spontaneous premature ovarian failure. Obset Gyneco11998;91:12. 30. Kaas MJ. Sexual expression of the elderly in nursing homes. Gerontologist 1978;18:372. 31. Frock J, Money J. Sexuality and the menopause. PsychotherPsychosom 1992;57:29. 32. McCoy NL, Davidson JM. A longitudinal study of the effects of menopause on sexuality. Maturitas 1985;7:203. 33. Dennerstein L, Smith AMA, Morse, Burger H. Sexuality and the menopause.J Psychsom Obstet Gyneco11994;15:56. 34. Bachmann GA, Leiblum SR. Sexuality in sexagenarian women. Maturitas 1991;13:45. 35. Bixo M, Backstrom T, Winblad B, Andersson A. Estradiol and testosterone in specific regions of the human female brain in different endocrine states. J Steroid Biochem Mol Bid 1995;55:297. 36. Roselli CE, Klosterman S, Resko JA. Anatomic relationships between aromatase and androgen receptor mRNA expression in the hypothalamus and amygdala of adult male cynomolgus monkeys. J ComplNeurol 2001;439:208. 37. Van Goozen SHM, Cohen-Kettensis PT, Gooren LIG, Frijda NH, Van de Poll NE. Gender differences in behaviour: activating effects of cross-sex hormones. Psychoneuroendocrinology 1995;20:343. 38. Davis SR, Goldstat R, Papalia MA, et al. Effects of aromatase inhibition on sexual function and wellbeing in postmenopausal women treated with testosterone: a randomized placebo controlled trial. Menopause 2006; 13:37-45. 39. Studd JWW, Chakravarti S, Cram D. The climacteric. Clin Obstet Gyneco11977;4:3. 40. Tazuke S, Khaw KT, Chir MBB, Barrett-Connor E. Exogenous estrogen and endogenous sex hormones. Medicine 1992;71:44. 41. Nathorst-Boos J, yon Schoultz H. Psychological reactions and sexual life after hysterectomy with and without oophorectomy. GynecolObstet Invest 1992;34:97. 42. Dow MGT, Hart DM, Forrest CA. Hormonal treatments of sexual unresponsiveness in postmenopausal women: a comparative study. BrJ Obstet Gynaeco11983;90:361. 43. Goldstat R, Briganti E, Tran J, Wolfe R, Davis SR. Transdermal testosterone improves mood, well being and sexual function in premenopausal women. Menopause 2003;10:390. 44. Slemenda C, Longcope C, Peacock M, Hui S, Johnston CC. Sex steroids, bone mass, and bone loss. A prospective study of pre-, peri- and postmenopausal women. J Clin Invest 1996;97:14. 45. Smith EP, Boyd J, Frank GR, et al. Estrogen resistance caused by a mutation in the oestrogen-receptor gene in a man. N Eng J Med 1994;331:1056. 46. Morishima A, Grumbach MM, Simpson ER. Aromatase deficiency in male and female siblings caused by a novel mutuation and the physiological role of estrogens. J Clin EndocrinolMetab 1995;80:3689. 47. Shozu M, Simpson ER. Aromatase expression of human osteoblast-like cells. Mol CellEndocrino11998;139:117.
CHAPTER 55 Intervention: Androgens 48. Colvard DS, Eriksen EF, Keeting PE. Identification of androgen receptors in normal human osteoblast-like cells. Proc NatlWcad Sci U S W 1989;86:854. 49. Kasperk CH, Wergedal JE, Farley JR, Llinkhart TA, Turner RT, Baylink DG. Androgens directly stimulate profiferation of bone cells in vitro. Endocrinology 1989;124:1576. 50. Kasperk CH, Faeling K, Borcsok I, Zieggler R. Effects of androgen on sub-populations of the human osteosarcoma cell-line SAOS2. Calcif Tissue Int 1996;58:376. 51. Abu EO, Homer V, Kusec V, Triffitt JT, Compston JE. The localization of androgen receptors in human bone. J Clin Endocrinol Metab 1997;82:3493. 52. Wiren KM, Keenan EJ, Orwoll ES, Zhang X, Chang C. Transcriptional U-regulation of the human androgen receptor by androgen in bone cells. Endocrinology 1997;138:2291. 53. Nilas L, Christiansen C. Bone mass and its relationship to age and the menopause. J Clin Endocrinol Metab 1987;65:697. 54. Simberg N, Titinen A, Silfvast A, Viinikka L, Ylikorkala O. High bone density in hyperandrogenic women: effect of gonadotropin-releasing hormone agonist alone or in conjunction with estrogen-progestin replacement. J Clin Endocrinol Metab 1995;81:646. 55. Heiss CJ, Sanborn CF, Nichols DL. Associations of body fat distribution, circulating sex hormones and bone density in postmenopausal women. J Clin Endocrinol Metab 1995;80:1591. 56. Jassal SK, Barrett-Connor E, Edelstein S. Low bioavailable testosterone levels predict future height loss in postmenopausal women. J Bone Miner Res 1995;10:650. 57. Davidson BJ, Ross RK, Paganini-Hill A, et al. Total and free estrogens and androgens in postmenopausal women with hip fractures. J Clin Endocrinol Metab 1982;54:115. 58. Nawata H, Tariaka S. Aromatase in bone cell: association with osteoporosis in post menopausal women. J Steroid Biochem ]Viol Biol 1995; 53:165. 59. Taelman P, Kayman JM, Janssens X, Vermeulen A. Persistence of increased bone resorption and possible role of dehydroepiandrosterone as a bone metabolism determinant in osteoporotic women in late menopause. Maturitas 1989;11:65. 60. Nordin B EC, Robertson A, Seamark RF, et al. The relation between calcium absorption serum DHEA and vertebral mineral density in postmenopausal women. J Clin Endo Metab 1985;60:651. 61. Morales AJ, Nolan JJ, Nelson JC, Yen SSC. Effects of replacement dose of dehydroepiandrosterone in men and women of advancing age. J Clin Endo Metab 1994;78:1360. 62. Labrie F, Diamond P, Cusan L, et al. Effect of 12-month dehydroepiandrosterone replacement therapy on bone, vagina and endometrium in postmenopausal women. J Clin Endocrinol Metab 1997;82:3498. 63. Watts NB, Notelovitz M, Timmons MC. Comparison of oral estrogens and estrogens plus androgen on bone mineral density, menopausal symptoms and lipid-lipoprotein profiles in surgical menopause. Obstet Gyneco11995;85:529. 64. Savvas M, Studd JWW, Fogelman I, et al. Skeletal effects of oral estrogen compared with subcutaneous oestrogen and testosterone in postmenopausal women. Br MedJ 1988;297:331. 65. Raisz LG, Wiita B, Artis A, et al. Comparison of the effects of estrogen alone and estrogen plus androgen on biochemical markers of bone formation and resorption in postmenopausal women. JCEM 1995; 81:37. 66. Need GA, Horowitz M, Bridges A, Morris H, Nordin C. Effects of nandrolone decanoate and antiresorptive therapy on vertebral density in osteoporotic women. Arch Int Med 1989;149:57. 67. Sawas M, Studd JWW, Norman S, Leather AT, Garnett TJ. Increase in bone mass after one year of percutaneous oestradiol and testosterone implants in post menopausal women who have previously received long-term oral oestrogens. BrJ Obstet Gynaeco11992;99:757.
811 68. Carani C, Qin K, Simoni M, et al. Effect of testosterone and estradiol in a man with aromatase deficiency. N E n g J M e d 1997;337:91. 69. Maffei L, Murata Y, Rochera V, et al. Dysmetabolic syndrome in a man with a novel mutation of the aromatase gene: effects of testosterone, alendronate, and estradiol. J Clin Endocrinol Metab 2004;89:61. 70. Kasra M, Grynpas MD. The effects of androgens on the mechanical properties of primate bone. Bone 1995;17:265. 71. Masi AT, Feigenbaum SL, Chatterton RT. Hormonal and pregnancy relationships to rheumatoid arthritis: convergent effects with immunological and microvascular systems. Semin Arthritis Rheum 1995;25:1. 72. Cutolo M, Seriolo B, Sulli A, Accardo S. Androgens in rheumatoid arthritis. In: Bijlsma JWJ, Linden S, van der Barnes CG, eds. Rheumatology highlights 1995. Rheumatol Eur. 1995;24:211. 73. Wilder RL. Adrenal and gonadal steroid hormone deficiency in the pathogenesis of rheumatoid arthritis. J Rheumatol Supp11996;44:10. 74. Booij A, Biewenga-Booij CM, Huber-Bruning O, et al. Androgens as adjuvant treatment in postmenopausal female patients with rheumatoid arthritis. Ann Rheum Dis 1996;55:811. 75. van Vollenhoven RE Dehydroepiandrosterone for the treatment of systemic lupus erythematosus. Expert Opin Pharmacother 2002;3:23. 76. Goodman-Gruen D, Barrett-Connor E. Total but not bioavailable testosterone is a predictor of central adiposity in postmenopausal women. IntJ Obes 1995;19:293. 77. Darling GM, Johns JA, McCloud PI, Davis SR. Estrogen and progestin compared with simvastatin for hypercholesterolemia postmenopausal women. N E n g J M e d 1997;337:595. 78. The Writing Group for the PT. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. JAm MedAssoc 1995;273:199. 79. Davis SR, Walker KZ, Strauss BJ. Effects ofestradiol with and without testosterone on body composition and relationships with lipids in postmenopausal women. Menopause 2000;7:395. 80. Dobs AS, Nguyen T, Pace C, Roberts CR Differential effects of oral estrogen versus estrogen-androgen replacement therapy on body composition in postmenopausal women. J Clin EndocrinolMetab 2002;87:1509. 81. Choi HH, Gray PB, Storer TW, et al. Effects of testosterone replacement in human immunodeficiency virus-infected women with weight loss. J Clin Endocrinol Metab 2005;90:1531-1541. 82. Tunstall-Pedoe H. Myth and paradox about coronary risk and the menopause. Lancet 1998;352:1425. 83. Colditz GA, Willett WC, Stampfer MJ, et al. Menopause and the risk of coronary heart disease in women. N E n g J M e d 1987;316:1105. 84. Golden SH, Ding J, Szklo M, et al. Glucose and insulin components of the metabolic syndrome are associated with hyperandrogenism in postmenopausal women: the atherosclerosis risk in communities study. Am J Epidemio12004;160:540. 85. Sutton-Tyrrell K, Wildman RP, Matthews KA, et al. Sex hormonebinding globulin and the free androgen index are related to cardiovascular risk factors in multiethnic premenopausal and perimenopausal women enrolled in the Study of Women Across the Nation (SWAN). Circulation 2005;111:1242. 85a. Bell RJ, Davison SL, Papalia MA, McKenzie DP, Davis SR. Endogenous androgen levels and cardiovascular risk profile in women across the adult life span. Menopause 2007. 86. Kaplan RC, Frishman WH. Systemic inflammation as a cardiovascular disease risk factor and as a potential target for drug therapy. Heart Dis 2001;3:326. 87. Ridker PM, Buring JE, Shih J, Matias M, Hennekens CH. Prospective study of C-reactive protein and the risk of future cardiovascular events among apparently healthy women. Circulation 1998;98:731. 88. Ridker P, Stampfer MJ, Rifai N. Novel risk factors for systemic atherosclerosis: a comparison of C-reactive protein, fibrinogen, homocysteine, lipoprotein(a), and standard cholesterol screening as predictors of peripheral artery disease. JAm MedAssoc 2001;285:2481.
812 89. Ridker PM, Buring JE, Cook NR, Rifai N. C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: an 8-year follow-up of 14719 initially healthy American women. Circulation 2003;107:391. 90. Albert MA. The role of C-reactive protein in cardiovascular risk. Curr Cardiol Rep 2000;2:274. 91. Ridker PM. High-sensitivity C-reactive protein: potential adjunct for global risk assessment in the primary prevention of cardiovascular disease. Circulation 2001;103:1813. 92. Barinas-Mitchell E, Cushman M, Meilahn EN, Tracy RP, Kuller LH. Serum levels of C-reactive protein are associated with obesity, weight gain, and hormone replacement therapy in healthy postmenopausal women. A m J Epidemio12001;15 3 :1094. 93. Pradham AD, Manson JE, Buring JE, Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. J A m MedAssoc 2001;286:327. 94. Ridker PM, Rifai N, Cook NR, Bradwin G, Buring JE. Non-HDL cholesterol, apolipoproteins A-I and B100, standard lipid measures, lipid ratios, and CRP as risk factors for cardiovascular disease in women. J A m MedAssoc 2005;294:326-333. 95. Wakatsuki A, Sagara Y. Lipoprotein metabolism in postmenopausal and oophorectomized women. Obstet Gyneco11995;85:523. 96. Lobo R, Rosen RC, Yang HM, Block B, Van der Hoop RG. Comparative effects of oral esterified estrogens with and without methyl testosterone on endocrine profiles and dimensions of sexual function in postmenopausal women with hypoactive sexual desire. Fertil Steril 2003;79:1341. 97. Barrett-Connor E, Young R, Notelovitz M, et al. A two-year, doubleblind comparison of estrogen-androgen and conjugated estrogens in surgically menopausal women. Effects on bone mineral density, symptoms and lipid profiles. J Reprod Med 1999;44:1012. 98. Chiuve SE, Martin LA, Campos H, Sacks FM. Effect of the combination of methyltestosterone and esterified estrogens compared with esterifled estrogens alone on apolipoprotein CIII and other apolipoproteins in very low densi~, low density, and high density lipoproteins in surgically postmenopausal women. J Clin EndocrinolMetab 2004;89:2207-2213. 99. Wagner JD, Zhang L, Williams JK, et al. Esterified estrogens with and without methyltesterone decrease arterial LDL metabolism in Cynomolgus monkeys, drteriosder Thromb VascBio11996;16:1473. 100. Barrett-Connor E, Goodman-Gruen D. Prospective study of endogenous sex hormones and fatal cardiovascular disease in postmenopausal women. BMJ 1995;311:1193. 101. Rexrode KM, Manson JE, Lee IM, et al. Sex hormone levels and risk of cardiovascular events in postmenopausal women. Circulation 2003; 108:1688-1693.
Susan R. DAVIS 102. Chou TM, Sudhir K, Hutchison SJ, et al. Testosterone induces dilation of canine coronary conductance and resistance arteries in vivo. Circulation 1996;94:2614. 103. Honore EK, Williams JK, Adams MR, Ackerman DM, Wagner JD. Methyltestosterone does not diminish the beneficial effects of estrogen replacement therapy on coronary artery reactivity in cynomolgus monkeys. MenopauseJ North Am Menopause Soc 1996;3:20. 104. Worboys S, Kotsopoulos D, Teede H, McGrath BP, Davis SR. Parental testosterone improves endothelium-dependent and independent vasodilation in postmenopausal women already receiving estrogen. J Clin Endocrinol Metab 2001;86:158. 105. Grohe C, Kahlert S, Lobbert K, et al. Cardiac myocytes and fibroblasts contain functional estrogen receptors. FEBS Lett 1997;416:107. 106. Grohe C, Kahlert S, Lobbert K, Vetter H. Expression of oestrogen receptor alpha and beta in rat heart: role of local oestrogen synthesis. J Endocrino11998;156:R1. 107. Somboonporn W, Davis SR. Testosterone and the breast: therapeutic implications for women. Endocrine Rev 2004;25:374. 108. Secreto G, Toniolo P, Pisani P, et al. Androgens and breast cancer in premenopausal women. CancerResearch 1989;49:471. 109. Yu H, Harris RE, Gao YT, et al. Comparative epidemiology of cancers in the colon, rectum, prostate and breast in Shanghai, China versus the United States. IntJEpidemio11991;20:76. 110. Thomas HV, Key TJ, Allen DS, et al. A prospective study of endogenous serum hormone concentrations and breast cancer risk in premenopausal women on the island of Guernsey. BrJ Cancer 1997;75:1075. 111. Wysowski DK, Freiman JR Tourtelot JB, et al. Fatal and nonfatal hepatotoxicity associated with flutamide. Ann Intern Med 1993;118:860. 112. Bryan RM, Mercer RJ, Rennie GC, Lie TH, Morgan FJ. Androgen receptors in breast cancer. Cancer 1984;54:2436. 113. Birrell SN, Roder DM, Horsfall DJ, Bentel JM, Tilley WD. Medroxyprogesterone acetate therapy in advanced breast cancer: the predictive value of androgen receptor expression.J Clin Onco11995;13:1572. 114. Barrett-Connor E, Timmons MC, Young R, Wiita B, Estratest Working Group. Interim safety analysis of a two-year study comparing oral estrogen-androgen and conjugated estrogens in surgically menopausal women. J I47omensHealth 1996;5:593. 115. Buckler HM, McElhone K, Durrington PN, et al. The effects of lowdose testosterone treatment on lipid metabolism, clotting factors and ultrasonographic ovarian morphology in women. Clin Endocrinol (Oxf) 1998;49:173.
; H A P T E R 5(
Future Strategies in
Climacteric
GORAN S A M
Medicine
SIOE
Department of Obstetrics & Gynecology, Lund University Hospital, 221 85 Lund, Sweden
MARTINA DOREN
Charit&Universit~itsmedizin Berlin, Campus Benjamin Franklin, Clinical Research Center of Women's Health, D-12200 Berlin, Germany
I. I N T R O D U C T I O N The various mechanisms of action of estrogen are still not fully understood, but major advances in our understanding of receptor-mediated effects have been achieved during the last decade. The discovery of a second estrogen receptor and an interaction between estrogen receptors and other steroid receptors, as well as the role of cytokines in modifying receptor responses, have been established. This research has led to an increased understanding as to why some substances could act as estrogen antagonists in some tissues and estrogen agonists in other tissues. With this knowledge it is possible to target the original molecular structure to create novel compounds with higher selectivity. These groups of substances, commonly referred to as selective steroid receptor modulators (SSRMs), have been developed to enhance desirable effects and avoid the drawbacks. For example, for corticosteroids, it would be ideal to have full corticosteroid effects without bone loss or the development of "moon face." Possible health hazards of long-term menopausal estrogen therapy, particularly of estrogen-progestogen therapy, have generated interest in developing alternatives to current regimens. The concept of pharmacotherapy, with the exception of the treatment of vasomotor and urogenital symptoms, has been questioned because of perceived health risks associated with short- and long-term use. Proposed alternatives include TREATMENT OF THE POSTMENOPAUSAL WOMAN
813
the use of naturally occurring compounds, many of which have estrogenic properties, often referred to as phytoestrogens. Several such agents have been introduced onto the market, but existing products are too "weak" to be a real alternative to menopausal hormone therapy for the treatment of hot flushes. Above all, there are virtually no long-term data available on efficacy and safety; this also applies for black cohosh, a substance which has been approved for the treatment of climacteric symptoms for many years in some countries. Many of the active substances in phytoestrogens have been classified based on their chemical structure, and "purified phytoestrogens" are under development. More than 100 plants, some of which are edible, which contain approximately 30 specific compounds, are known to exhibit estrogenic activity. Phytoestrogens consist of a number of classes: isoflavones, lignans, coumestans, and fungal estrogens. Various legumes contain the isoflavones daidzein, genistein, formononetin, biochanin A, and the coumestan coumestrol. It is possible to alter slightly the chemical structure of these compounds and thereby possibly change the pharmacokinetic and pharmacodynamic properties. It has become increasingly clear that the compulsory addition of progestogens to protect the endometrium from malignant transformations by estrogen stimulation is problematic. The addition of a progestogen increases the relative risk of breast cancer (1,2). Women may accept monthly Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
814 bleeds around the menopause and during the few years there after, but with advancing age, bleed-free regimens are clearly preferred in most cultures. Bleed-free regimens can be achieved in a great many women by using continuous combined regimens, but some women may bleed from an atrophic endometrium. Monitoring, surveillance, and treatment of bleeding from an atrophic endometrium remain a challenge because we do not fully understand the etiology behind these bleeding episodes, although much knowledge has been gained during the past few years. Whether endometrial ablation will alleviate or even abolish the problem of uterine bleeding remains to be proven in well-designed studies with appropriate long-term follow-up. The recent Women's Health Initiative (WHI) data augment an interest in finding a safe therapy and hence using estrogen alone avoiding progestogens. To induce endometrial atrophy and thereby a bleedfree regimen, the progestogen effect in combined therapy needs to be fairly high. This means that other progestogenic side effects are not uncommon in women using continuous combined regimens. Hence, we need to have more selective progestogens primarily designed as estrogen co-medications and also for the use in elderly postmenopausal woman. Unfortunately, it is still a fact that almost all data on pharmacokinetics and pharmacodynamics on estrogens, progestogens, and various estrogen/progestogen combinations, including the continuous combined regimens have been obtained in women around 50 years of age, and such data are probably not transferable to an older population. For these reasons, we have insufficient data to guide us in selecting the progestogen type, dose, or mode of administration for older postmenopausal women. Cost-benefit analyses seem to suggest that estrogens are cost-effective in women with symptoms and particularly in women without a uterus. That is to say that in the hysterectomized woman, estrogens can be given without a progestogen. This statement should not be used to encourage the medical community to perform hysterectomies around the time of the menopause. This type of surgery is performed too often worldwide for reasons which are not always convincing. However, the only proven positive effect of progestogen co-administration is its reduction of endometrial hyperplasia and the prevention of endometrial cancer.
II. DEVELOPMENT OF ESTROGEN COMPONENTS The predominant human natural estrogen during the fertile period is 17[3-estradiol. Hence, several hormone replacement therapies focus on the use of estradiol. Many of the same problems arise with natural estradiol as with natural progesterone. We need to know what the critical molecular structures are responsible for the positive effects on
SAMSIOE AND DOREN
symptoms and, thus, to avoid those molecular structures which may contribute to weight gain and influence breast tissues. Proliferation is thought to be involved in the development of breast cancer. Potentially, the estradiol molecule could be "tailored," in order to diminish or even abolish any interaction with breast tissue. Another feature to look at is the various components of conjugated equine estrogens (CEEs). Although the major components are estrone and estrone sulphate, CEE contains a variety of compounds that display different estrogenic activities. In particular, the so-called "alpha estrogens" contained in CEE are of interest as they possess qualifies different from the beta compounds. The further characterization of the various components of conjugated estrogens and their biologic effects should stimulate future research into estrogen pharmacokinefics and pharmacodynamics. It is also of pivotal importance to try to identify the type of molecular infrastructure that is responsible for the various cardiovascular effects, such as anfioxidative properties and vasodilatation, as well as the bone-sparing effects. New delivery systems have been introduced to improve the convenience of treatment and the potential for long-term use. A vaginal silicone ring that releases approximately 7.5 ~g estradiol per day constantly over a period of 3 months seems to be a valuable addition in the treatment of urogenital symptoms in estrogen-deprived women. Vaginal tablets containing 25 pg estradiol to be applied twice weekly are other alternatives to vaginal creams containing estriol or conjugated estrogens. Hormone-containing patches and gels are now established, but these need to be improved to avoid local skin irritation. Nasal spray continues to be an interesting line of development. Other forms of local treatment, although systemic absorption seems to be possible, include estradiol eye drops designed for the treatment of postmenopausal keratojunctivitis sicca. Long-acting delivery systems such as estradiol implants and intramuscular injections have been used for many years in various European countries. However, supraphysiologic blood levels are commonly encountered, sometimes associated with tachyphylaxis, and may therefore also carry health risks to a degree that is not well defined. Nonbiodegradable, silicone-based implants could improve this type of therapy by avoiding peak levels after insertion and maintaining lower, more constant hormonal levels that are in the range of the early to mid-follicular phase. An additional intriguing approach is the external regulation of body implants hormone-containing devices by a remote control. This can be achieved via the so-called "blue-tooth technique," which is under development in cooperation with the telecommunications industry. Many medical problems could benefit immensely from such techniques, including insulin administration in diabetics as well as sex steroids and anticoagulant therapy. The long-cycle therapy addresses a major problem, uterine bleeding, which is not accepted by most women. Although it is a very intriguing idea to space out the sequential administration of a progestogen to induce endometrial shedding, it
CHAPTER 56 Future Strategies in Climacteric Medicine
should be noted that the long-term endometrial safety of the few combinations tested remains to be demonstrated.
III. DEVELOPMENT OF P R O G E S T O G E N S AND PROGESTERONE An important task for future research is to try to delineate which structures in the progesterone molecule yield the different metabolic aspects. In other words, specific molecular structures influence the endometrium, and these molecular specifications give other potential negative effects such as premenstrual syndrome (PMS)-like effects and lowering of high-density lipoprotein (HDL) cholesterol? Thus far, progestogens have been developed mostly for use in contraception but also to treat various bleeding disorders in fertile life. More recently, progestogens to be used for postmenopausal women have become an important feature in the development of newer agents. The development of newer progestogens with fewer side effects and a more selective influence on the endometrium is an obvious goal in this respect. The patch administration of estrogens has been much in favor. However, there is less of an incentive to administer progestogens via the transdermal route because some of them have very little or no first-pass hepatic metabolism, as exemplified by levonorgestrel. Other progestogens including progesterone may show a first-pass metabolism. Natural progesterone displays a low oral bioavailability. Using newer microcrystallization techniques, it has become possible to achieve an acceptable bioavailability by the oral route. However, this needs to be substantially improved, as only a small fraction of orally administered progesterone is bioavailable. This means that progesterone has to be given in doses of approximately 200 mg per day to ensure endometrial protection. Work is underway to modify the molecule so that it may penetrate into the body, either via the skin or the oral route, and then rapidly degrade to natural progesterone. A vaginal progesterone gel, 45 mg, to be administered twice weekly for a period of I month, has been shown to induce a secretory endometrium in the presence of mean serum progesterone levels as low as some 2 mg/mL, which is suggestive of a so-called "first uterine pass" effect in women with premature ovarian failure and ovarian dysgenesis. The endometrial safety remains to be demonstrated in clinical trials of postmenopausal women subjected to treatment. Other developments to improve acceptance by women without compromising endometrial safety are to administer or use various progestogens in either implant form or as intrauterine devices (IUD). One approved IUD releases levonorgestrel 20 big daily for approximately 5 years. Two different IUDs with a release rate of 5 and 10 big/24 h were reported to induce acceptable bleeding patterns and showed no adverse effects on lipid and lipoprotein metabolism
815 when combined with 2 mg estradiol valerate orally administered (3). A disadvantage with this type of device is that the progestogen reservoir covered by the rate-limiting membrane could be too thick for insertion into the uterus of a postmenopausal woman. A smaller device with a thinner vertical arm is currently in clinical development for the use of postmenopausal women (4) (Table. 56.1). A release rate of 5 to 10 big/24 h is sufficient for endometrial protection. Another feature is the introduction of newer regimens in order to reduce the progestogen dosage. An example of this is the so-called "on-and-off therapy" in which estrogens are given continuously but progestogens are given 3 days on and 3 days off. In such clinical trials, bleeding discomforts were acceptable, but extensive clinical experience and controlled studies providing long-term efficacy and safety data are lacking. It is also disputable which type of progestogen should be preferred; in other words, one that has similar pharmacokinetics to the administered estrogens, such as medroxyprogesterone acetate (MPA), norethisterone, or dydrogesterone, or one that is longer acting, such as desogestrel, levonorgestrel, or norgestimate. Also under current investigation are combinations of estrogens and a variety of antihypertensive drugs to delineate perceived additive or even synergistic effects. This is particularly prudent because estrogen is about the only compound that has more effects (albeit small) on diastolic than systolic blood pressure. An angiotensin-converting enzyme (ACE)inhibiting effect and calcium channel blocking effects have been demonstrated for estrogens, and research that combines estrogens with different classes and compounds of antihypertensive drugs could improve the overall effect and possibly also decrease untoward effects by several of the compounds involved.
IV. FUTURE AVENUES FOR MIDLIFE WOMEN'S HEALTH For how long can women use hormones without adverse effects if vasomotor symptoms persist once therapy is discontinued and resumption of therapy is considered as an option? Whether the benefits of short-term use of hormones during TABLE 56.1
Intrauterine Menopausal Hormone Therapy
Levonorgestrel (LNG)-Intrauterine device 10 pg + 50 big E2 patch: Open, multicenter study of 294 postmenopausal women intended for 3 years First year results presented First insertion was 94% successful Bleeding and spotting mainly up to 2 months; decreased from 9 to 0 days at month 4 No adverse lipid effects (From ref. 4, with permission.)
816 the perimenopause and early menopausal period (5), the time with the highest likelihood of hot flushes, appear to outweigh the risks depends on available evidence, largely generated in trials including women well beyond 50 years old, and a personal benefit-risk assessment and perception. However, there are no such comprehensive controlled clinical trials results available to assess benefits and risks in women in the immediate perimenopausal period, during which vasomotor symptoms are most common. How should women be counseled if they seek medical advice once they experience health-related changes at midlife? Given that women (and men) face various physiologic changes as they age, evidence-based approaches to manage changes associated with midlife health are needed. At present, there is no clear, agreed-upon role for any natural or synthetic hormone with estrogenic properties in disease prevention, possibly except for osteoporosis in women at high risk, but this approach would be associated with accepting various health risks--and this does not seem to be prudent. Given the recent results of controlled studies with clinical end points as an outcome, the decreases in the risks of hip fracture and colon cancer do not appear to balance the increased risks of stroke, thromboembolic events, and breast cancer, not to mention the risks of cognitive impairment and urinary incontinence (6). Given the availability of other effective agents, the use of hormone therapy for the treatment or prevention of osteoporosis is not appropriate for most women. Until recently, it has been argued that many women, including (older) women who do not have vasomotor or urogenital symptoms, feel better when they take hormones, but the majority of placebocontrolled trials including W H I suggest that hormone therapy had no effect on measures of depression or cognition. However, there is no ultimate pharmacologic solution to meet the needs of symptom relief, particularly in women whose flushes may not disappear for many years. Nonhormonal medications for both vasomotor and urogenital symptoms are available, including a growing variety of soy products, black cohosh preparations, and vitamins, yet efficacy appears to be at least low (if existing) for many compounds, in particular for over-the-counter preparations (7,8); the exceptions (9) warrant further good-quality studies. Whether antidepressants such as serotonin and norepinephrine reuptake inhibitors provide relief from hot flushes (10,11) or are a "true" alternative to hormones remains to be established by carefully designed controlled trials that also provide data on longer-term safety. Also, there is no good (if any) evidence that nonpharmacologic approaches, including homeopathy or yoga, provide lasting benefits (7). In women with mild to moderate vasomotor symptoms, physical activity has repeatedly been shown to mitigate symptoms (12-14). Hence, we must increase our efforts to motivate and encourage physical activity in the community, including middle-aged women as well. To make people understand the importance of physical activity, it can be put on
SAMSIOE AND DOREN
prescription, such as in Sweden. The use of a drug prescription form to prescribe physical exercise augments the patient's comprehension and understanding that physical activity equates pharmacologic therapy. The annual increase in the risk of serious adverse events associated with postmenopausal hormone therapy is relatively small, but why should women be exposed at any risk? If the rates of diseases among 50-year-old women are estimated to be about half that reported for older women in the WHI, the net effect of hormone therapy in this age group will be about 1 serious adverse event per 1000 women treated per year. Hot flushes may be very disabling. Some women may choose to try other remedies or to live with their symptoms, whereas others will find the relief of symptoms afforded by hormone therapy worth the risk. Vasomotor symptoms resolve within several months in many women and within a few years in most women, so attempts should be made to taper the dose of hormones and to discontinue therapy. For women whose menopausal symptoms persist beyond the immediate postmenopausal period, estrogen (with or without a progestogen) may be an option, although it is associated with hazards given the inferiority of other currently available therapies. Women should be counseled that menopausal hormone therapy is only one option to improve certain aspects of midlife health. Evidence-based advice about smoking cessation, which may help with vasomotor symptoms (15), nutrition and weight control, physical exercise, management of stress, and possibilities to maintain cognitive abilities in old age in order to prevent diseases are the most essential cornerstones of preventive medicine. Regular (leisure time) physical activity can maintain cardiovascular health, diminishes the risk of osteoporosis, protects against obesity's many adverse health outcomes, and may protect against Alzheimer's disease (16). Moderate use of alcohol could be beneficial (17). However, data related to the consumption of alcohol and the onset of flashes are conflicting (18). Lung cancer in women, on the rise in many countries, may be targeted if women stop smoking and do not regard smoking as an expression of possessing equal rights of (smoking) men. Obesity appears to be one of the most challenging health problems for aging women and men. The prevalence of diabetes, which often accompanies obesity, appears to be associated with changing living conditions of women, who progressively often live alone in older age in Western societies (19). A balanced diet, not necessarily supplemented by calcium and vitamin D if daily nutrition provides amounts regarded as sufficient for bone protection, will help prevent osteoporosis and may reduce the risk of failing. Continued mental exercises appear to decrease the risk of Alzheimer's disease. Whether lowdose aspirin (50 mg/day) may have the potential to prevent strokes in women older than age 65 has been demonstrated in one large prospective trial (to date), yet this therapy also has side effects (20). However, all nonpharmaceufical approaches
CHAPTER 56 Future Strategies in Climacteric Medicine
are somewhat arduous and demand continuous, essentially lifelong, efforts, which are often not invested in maintenance of health by individuals and not often enough recommended and explained by doctors. The optimal prevention of disease remains a compelling, growing task given the demographic changes never before seen in history, with rapidly growing numbers of women and men at old and very old ages. Hormonal therapies to prevent symptoms and diseases with maximal benefits and minimal risks are one part of preventive health care yet to be optimized. In conclusion, there are several possibilities to address the age-dependent decrease of gonadal hormones in women, all of which need to offer more benefits than risks as a principle in order to be acceptable as a strategy to maintain or possibly improve (post)menopausal health. It is a challenge for today's clinical medicine to determine those women who have the most to gain from such therapy in order to make it cost-effective.
References 1. Chlebowski RT, Hendrix SL, Langer RD, et al. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women. JAm MedAssoc 2003;289:3243-3253. 2. Greiser C, Greiser EM, D6ren M. Menopausal hormone therapy and risk of breast cancer: a meta-analysis of epidemiological studies and randomized controlled trials. Hum Reprod Update 2005;11:561 - 573. 3. Wollter-Svensson LO Stadberg E, Andersson K, et al. Intrauterine administration of levonorgestrel 5 and 10 micrograms/24 hours in perimenopausal hormone replacement therapy. A randomized clinical study during one year. Acta Obstet GynecolScand 1997;76:449-454. 4. Sturdee DW, Rantala ML, Colau JC, Zahradnik H-P, Riphagen FE. The acceptability of a small intrauterine progestogens-releasing system for continuous combined hormone therapy in early postmenopausal women. Climacteric2004;7:404- 411. 5. MacLennan AH, Broadbent JL, Lester S, Moore V. Oral oestrogen and combined oestrogen/progestogen therapy versus placebo for hot flushes. CochraneDatabase System Rev 2004;4(CD002978). 6. Farquhar CM, Marjoribanks J, Lethaby A, et al., for the Cochrane HT Study Group. Long term hormone therapy for perimenopausal and postmenopausal women. CochraneDatabase System Rev 2005;3(CD004143). 7. Balk E, Chung M, Chew P, et al. Effects of soy on health outcomes. Evidence Report/Technology Assessment No. 126. AHRQ_Publication No. 05-E024-2. Rockville, MD: Agency for Healthcare Research and Quality, Aug. 2005.
817 8. Krebs EL, Ensrud KE, MacDonald R, Wilt TJ. Phytoestrogens for treatment of menopausal symptoms: a systematic review. Obstet Gyneco! 2004;104:824-836. 9. Wuttke W, Seidlova-Wuttke D, Gorkow C. The Cimicifuga preparation BNO 1055 vs. conjugated estrogens in a double-blind placebo-controlled study: effects on menopause symptoms and bone markers. Maturitas 2003;44:$67- $77. 10. Sterns V, Beebe KL, Iyengar M, Dube E. Paroxetine controlled release in the treatment of menopausal hot flashes: a randomised controlled trial. JAm MedAssoc 2003;289:2827- 2834. 11. Evans ML, Pritts E, Vittinghoff E, et al. Management of postmenopausal hot flashes with venlaflaxine hydrochloride: a randomized, controlled trial. Obstet Gyneco12005;105:161-166. 12. Aiello EY, Yasui Y, Tworoger SS, et al. Effect of a year-long, moderate-intensity exercise intervention on the occurrence and severity of menopause symptoms in postmenopausal women. Menopause 2004;11:382-388. 13. Lindh-Astrand L, Nedstrand E, Wyon V, Hammar M. Vasomotor symptoms and quality of life in previously postmenopausal women randomised to physical activity or estrogen therapy. Maturitas 2004;48: 97-105. 14. Gold EB, Sternfeld B, Kelsey JL, et al. Relation of demographic and lifestyle factors to symptoms in a multiracial/ethnic population of women 40-55 years of age. Am Epidemio12000;152:463-473. 15. Nelson HD, Haney E, Humphrey L, et al. Management of menopauserelated symptoms. Evidence Report/Technology Assessment No. 120. AHRQ_ Publication No. 05-E016-2. Rockville, MD: Agency for Healthcare Research and Q.uality, March 2005. 16. Rovio S, K&eholt I, Helkala E-L, et al. Leisure-time physical actMty at midlife and the risk of dementia and Alzheimer's disease. Lancet Neuro12005;4:705- 711. 17. Schilling C, Gallicchio L, Miller SR, et al. Current alcohol use is associated with a reduced risk of hot flashes in midlife women. Alcohol Alcoholism 2005;40:563- 568. 18. Freeman EW, Sammel MD, Grisso JA, et al. Hot flashes in the late reproductive years: risk factors for African American and Caucasian women. J WomensHealth Gend Based Med 2001;10:67- 76. 19. Lidfeldt J, Nerbrand C, Samsioe G, Agardh C-D. Women living alone have an increased risk to develop diabetes, which is explained mainly by lifestyle factors. Diabetes Care 2005;28:2531-2536. 20. Ridker PM, Cook NR, Lee IM, et al. A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women. N EnglJ Med 2005;352:1293-1304.
This Page Intentionally Left Blank
SECTION XIII
Some Alternative Medical Therapies This section covers a variety of other therapies that may be considered by postmenopausal women. Use of herbs and phytoestrogens were covered in an earlier section on Lifestyle (see Section X). Although not all potential therapies will be covered here, some of the major alternatives to traditional hormonal therapy will be discussed. Some of these therapies in reality are forms of hormonal therapy; however, they are not the traditional forms of estrogen and progestogens. It is clear that women require various options when considering treatment, and it is anticipated that these discussions will be useful in this regard. In the first chapter in this section, John E. Buster reviews the use of dehydroepiandrosterone (DHEA). Although D H E A is a hormone, it is available without a prescription and is not regulated by the Food and Drug Administration. It has been popularized as an anti-aging medicine that affects immune function and body composition and improves hormonal status. John will tell the reader whether this is true. Next, Ralf C. Zimmermann discusses the use of melatonin. This, too, is an interesting hormone available over the counter. Does it help with sleep and mood? The next chapter by Felicia Cosman and Robert Lindsay discusses the use of selective estrogen reuptake modulators (SERMs). It is appropriate that this chapter is authored by two "bone heads" in that the major indication for these agents is osteoporosis. Several products are available now, and more are in the pipeline. Finally, David Archer and Chris Gallagher discuss tibolone. This interesting progestogen has unique estrogenic, androgenic, and progestogenic properties and has been coined a "STEAR" (Selective Tissue Estrogenic Activity Regulator). It is available in many countries and is quite widely prescribed around the world for the indications of vasomotor symptom control and osteoporosis prevention. However, to date it is not available in the United States.
This Page Intentionally Left Blank
"m
t" J"
~ H A P T E R D,
Alternative Therapy: D e hydro ep i an dro s te ro n e for Menopausal Hormone Replacement PAUL S. DUDLEY Departmentof Obstetrics and Gynecology,Division of Reproductive Endocrinology and Infertility, Baylor College of Medicine, Houston, TX 77030 JOHN E. BUSTER Departmentof Obstetrics and Gynecology,Division of Reproductive Endocrinology and Infertility, Baylor College of Medicine, Houston, TX 77030
Dehydroepiandrosterone (DHEA) and its sulfoconjugate, dehydroepiandrosterone sulfate, DHEA(S), are androgen prohormones. Having no known receptors, these prohormones serve as precursors to downstream androgens such as testosterone (T) and dihydrotestosterone, both of which do have steroid receptors and are therefore biologically active. Because DHEA(S) production declines with advancing years, many aspects of aging have been attributed to waning DHEA(S) synthesis. Accordingly, DHEA is among the products touted by the supplement industry for its ability to defer many of ages' tyrannies. Statements trumpeting its potential are tempting: "Live better than ever. Maximize immunity. Stop stress in its tracks. Mend a broken heart." These assertions are not unfounded. As an androgen prohormone, DHEA(S) has the physiologic advantage of enhancing production of biologically active androgens within TREATMENT
OF T H E P O S T M E N O P A U S A L W O M A N
target tissues themselves. Because target tissues autoregulate their own intracellular production of bioactive androgens, replacement strategies involving DHEA are a tantalizingly natural method of restoring a youthful androgen milieu to postmenopausal women. DHEA as menopausal hormone replacement offers significant advantages over traditional estrogen and androgen therapy. A system employing such a strategy should approximate youthful D H E A production rates, allowing for a woman's natural physiology to produce downstream bioactive hormones at their site of action. Unfortunately, optimal development of a safe, effective DHEA delivery system remains elusive. This chapter reviews relevant literature on therapeutic DHEA and analyzes the rationale and controversies surrounding its clinical use as a menopausal hormone replacement. 821
Copyright 9 2007 by Elsevier, Inc. All rights of reproduction in any form reserved.
822
DUDLEY AND BUSTER
I. DHEA IN MENOPAUSE AND AGING DHEA and DHEA(S) are androgens, a class of C19 steroids produced by the adrenal glands and ovaries (1,2). Androgen syntheses proceed via two interrelated pathways invoMng five steroidogenic enzymes (Fig. 57.1). The preferred route entails cleavage of 17-OH pregnenolone to form DHEA. Within the highly specialized zona reticularis (ZR), or androgen cortex of the adrenal, DHEA is largely converted to and secreted as DHEA(S), which is secreted in considerable abundance. In young women, the production rate of DHEA(S) is 11 to 20 mg per day, an amount that considerably exceeds other steroids (2). Within the ovaries, DHEA is secreted, in far more modest amounts, as unconjugated DHEA. DHEA and DHEA(S) are extensively interconverted in peripheral tissues and secreted as various metabolites back into the circulation. Very high production rates of DHEA(S), combined with its slow clearance and long half-life, produce circulating concentrations of DHEA(S) that are 100 to 1000 times higher than any of the unconjugated steroids (3). DHEA(S) circulates in a pool with very slow turnover. This pool provides a large reservoir for peripheral conversion into biologically active androgens and estrogens in target tissue cells. Thus, steroids produced from DHEA(S) exert action inside the same cells in which their synthesis takes place ("intracrinology") (Fig. 57.2) (4). The quantitative importance of this mechanism is illustrated by estimates that approximately 50% of total androgens are derived from DHEA and DHEA(S) in adult men (5). In women, the intracrine contribution of DHEA and DHEA(S) to peripheral estrogen formation is even more substantial: approximately 75% before and close to 100% after menopause (Fig. 57.3) (6). CHOLESTEROL 20,22 hydroxylation 1
Side ChainCleavage (P450SCC) 17 Hydeoxylation PREONENOLONE- (P450 c17) ir 17-OHPREONENOLONE 17-20cleavagei (P450 c 17)
3/~-HSD
DHEA <
Steroid
Sulfatase . ,, ~" DHEAS
I
~
PARACRINE
AUTOCRINE
INTRACRINE
_
(oo, o() FIGURE 57.2 Schematic representation of endocrine, paracrine, autocrine, and intracrine secretion. Hormones may act on distant cells (endocrine activity), neighboring cells (paracrine activity), the same cell (autocrine activity), or the same cell without release into the periceUular compartment (intracrine activity). (From ref. 4, with permission.)
Thus, in women who have no testicular androgen production, the relentless decline in DHEA/DHEA(S) with age profoundly decreases androgen availability.
A. DHEA Deficiency and Aging The interest in DHEA as a menopausal therapy stems from its decline with age. In fetal life, the adrenal gland produces large amounts of DHEA(S) as a substrate for placental estrogen. After birth, synthesis rapidly dwindles. DHEA(S) remains low until adolescence, at which point a rapid increase in ZR mass and DHEA(S) production marks the onset of adrenarche. Serum DHEA(S) peaks during the twenties and thereafter declines 2% per year. Over time, an astounding 70% drop in DHEA(S) synthesis occurs between the ages of 20 and 60 years (Fig. 57.4) (5). Declining DHEA(S) is associated with a clinical sequence of reverse adrenarche (Table 57.1). Women experience reduced libido, diminution of pubic and axillary hair, deterioration of bone density and muscle mass, and immunosenescence. This postmenopausal senescence serves as further rationale for therapeutic DHEA replacement. In theory, artificial restoration of youthful DHEA(S) should avert or postpone reverse adrenarche.
...........
17 Hyckoxylatioa PROGESTERONE (P450c17) .. 17-OHPROGESTERONE
cleavage~
17-20 (P450 c17)
ANDROSTENEDIONE 17~HSD1 TESTOSTERONE 5evReductase~ DHT
FIGURE 57.1
END()CRINE
Overview of androgen synthesis.
B. Zona Reticularis Involution and Aging 3~HSD
The age-related decline in DHEA(S) production occurs in conjunction with selective involution of adrenal ZR architecture. As women age, the ZR undergoes programmed cell death (apoptosis). Fig. 57.5 illustrates the structural fragmentation that occurs with aging. A combination of apoptosis and cellular senescence with reduced replicative capacity are likely causes. Specifically, attenuated 17-20 bond cleavage by cytochrome p450 c17 is at least partially responsible for the DHEA(S) decline (7). As DHEA(S)
CHAPTER 57 Alternative Therapy: Dehydroepiandrosterone for Menopausal Hormone Replacement
823
FIGURE 57.3 Distributionof testosterone, dihydrotestosterone(DHT), and their metabolites (androsterone[ADT]-G, 3cx-diol-G,and 3~-diol-G)in the circulationand peripheral intracrine tissuesof women.The height of the bars is proportionalto the concentration of each steroid. (Fromref. 4, with permission.)
falls, so does peripheral intracellular production of bioactive androgens with its far-reaching physiologic consequences.
C. DHEA(S) as a Moderator and Biomarker of Aging DHEA(S) may moderate aging and its maladies when present in abundance. This logic follows first from the association of abundant DHEA(S) with events of youth, the decline of DHEA(S) with age, the association of this decline with events and diseases of aging, and the importance of DHEA(S) as an androgen prohormone. The logic evolves secondly from animal models, epidemiology, and human clinical trials in which endogenous or therapeutic DHEA(S) appears to moderate aging in the cardiovascular, immune, and central nervous systems, in bones, and in sexual functioning (8-10).
Elevated DHEA(S), as a circulating biomarker, may signal protection against heart diseases, diabetes, and cancer (11-13). Conversely, decreased DHEA(S) has been associated with insulin resistance and atherosclerosis (14,15). The concept finds support in both animal models and human epidemiology. In the rhesus monkey, caloric restriction is a robust and reproducible method to slow aging (16). In studies of calorically restricted monkeys, three "biomarkers of longevity" emerged (also common to long-lived humans): lower plasma insulin, reduced body temperature, and higher plasma DHEA(S) (17). The first two of these are in line with the thinking that caloric restriction prompts a fundamental metabolic shift from one with an emphasis on growth and reproduction to one of life maintenance. In contrast, the reduction in postmaturational decline in serum DHEA(S) appears to be a direct result of caloric restriction. In this way, the aging primate model fits reasonably well
TABLE 57.1
Postmenopausal Senescence as a Model of Reverse Adrenarche
Adrenarche
Menopausal Senescence
Increasing sex hair Increasing libido Increasing bone density Increasing stature Increasing muscle mass Immune maturation
Loss of sex hair Loss of libido Loss of bone density Loss of stature Loss of muscle mass Immunosenescence
FIGURE 57.4 Circulatingdehydroepiandrosteronesulfate,or DHEA(S),
in females from birth to death. (From ref. 3, with permission.)
From ref. 3, with permission.
824
DUDLEY AND BUSTER
sprays, implants, or patches hold the most promise for future studies. Unfortunately, efforts to develop a patch (19) have been problematic because an extremely large surface area is required to deliver the required doses of D H E A needed in clinical studies. In the case of DHEA, a gel, spray, or implant will ultimately provide the best delivery option.
A. Administration and Dosing
FIGURE 57.5 Fragmentation of adrenal zona reticularis with age. (From ref. 3, with permission.)
with human epidemiologic data to provide support for the notion that DHEA(S) is a biomarker of aging (18).
II. STRATEGIES FOR DHEA REPLACEMENT D H E A replacement should simulate the physiological D H E A and DHEA(S) secretion and metabolism of youth. Unfortunately, this goal is proving elusive because both oral and non-oral D H E A therapies reported to date fail to meet this standard. Most human trials have involved oral DHEA. Unfortunately, chronic, oral D H E A is innately suboptimal because orally ingested D H E A must undergo first-pass hepatic metabolism. Because of the first-pass effect, large doses of D H E A are needed to produce the desired physiologic effects. In addition, over time there is unpredictable acceleration of hepatic D H E A metabolism (necessitating increasing doses), liver function abnormalities, hyperlipidemia, and excess production of virilizing androgens disproportionate to the D H E A dose. In contrast, transdermal and transvaginal D H E A do not undergo first-pass metabolism. For this reason, transdermal strategies involving creams, gels,
Oral D H E A was first reported in men. Investigations began with very high (1600 mg/day) oral doses; in fact, the doses were more than 30 times those eventually chosen for women (20). These studies indicated that oral D H E A caused hyperlipidemia in men and could result in potentially virilizing androgen elevations if applied to women. Studies in women at these high doses resulted in similar findings. Subsequent studies in postmenopausal women investigated oral doses that were much smaller. Initially, doses of 300 mg and then 150 mg daily of D H E A were examined and then abandoned. Excessive T elevations and adverse lipid profiles were a major problem (21). Subsequently, lower doses of 50 mg/day were investigated. Unfortunately, even reductions to 50 mg/day continued to induce supraphysiologic levels of T (22,23). Further reduction to 25 mg/day uncovered other obstacles as well. In addition to continued adverse lipid effects, chronic administration at 25 mg/day resulted in accelerated liver metabolism over time (24). Consequently, unacceptably high doses were required to maintain therapeutic DHEA(S) levels. It is not likely that an acceptable oral strategy will emerge. The first study of non-oral administration compared oral micronized D H E A with vaginal micronized DHEA. The research was designed to ascertain whether physiologic circulating DHEA(S) could be obtained with minimal increases in T. This proved to be possible. The vaginal route was associated with substantially less bioconversion of D H E A to T and other virilizing androgens (25). A systemic examination of a transdermal system using a gel followed with similar results. In this study, 12 months of daily treatment with a 10% D H E A gel in a group of postmenopausal women had no significant effect on lipid profiles (26). Recent work confirms that percutaneous administration more closely mimics endogenous production. However, the amounts of gel required and the need for multiple sites to obtain therapeutic levels of DHEA(S) make this approach cumbersome for routine clinical use (27).
III. CLINICAL APPLICATIONS There is a large but disorganized body of evidence from animal models, human epidemiology, and limited clinical trials supporting the clinical application of D H E A as an agent to slow aging and its maladies.
825
CHAPTER57 Alternative Therapy: Dehydroepiandrosterone for Menopausal Hormone Replacement
A. DHEA and the Cardiovascular System In animal models, a link between D H E A and the cardiovascular system was first established using experimentally induced endothelial injury in a hypercholesterolemic rabbit. When compared with placebo, D H E A reduced atherosclerosis by 50% over 12 weeks (28). Reduction in plaque size was proportional to the serum DHEA(S) level. Subsequent work employing the same model confirmed that chronic D H E A administration significantly retards progression of atherosclerosis (29). In epidemiologic studies, the Rancho Bernardo study provided the sentinel data on the relationship between DHEA(S) and cardiovascular health. Three relevant analyses were published. In the initial reports, elevations in DHEA(S) were associated with reductions in cardiovascular mortality in older men (11). In a later report, however, the relationship did not extend to females (30). In a later report, after adjusting for lifestyle-related factors, the initial reports were not reconfirmed (31). No clinical trials examining D H E A and cardiovascular health have been reported. Disturbingly, oral D H E A (25 to 50 mg) did induce atherogenic serum lipid profiles. Both 3- and 6-month trials showed significant reductions in high-density lipoprotein (HDL) and apolipoprotein A1 (31,32). For this reason, oral D H E A is problematic. Transdermal D H E A has no adverse effect on the lipid profile (26). Thus, any future trials examining cardiovascular outcomes should use non-oral delivery.
B. DHEA and Glucose Metabolism In animal models, D H E A increases insulin sensitivity in genetically-diabetic mice (33). Similarly, D H E A has counteracted the age-related decline in glucose tolerance in older mice (34). In epidemiologic studies, there may be a positive effect of D H E A on glucose metabolism. An inverse correlation between DHEA(S) levels and the insulin response has been
observed in a small group of normal women subjected to the oral glucose tolerance test (OGTT) (35). In clinical trials, a 3-week treatment with oral D H E A (50 mg/day) in a randomized, placebo-controlled, doubleblind, crossover trial showed an improvement in Tlymphocyte insulin binding, a marker of insulin sensitivity (8). Indeed, in a subsequent study this benefit translated into a clinically evident improvement in insulin sensitivity (36). Similarly, transdermal use of 10% D H E A cream resulted in a statistically significant decline in fasting plasma glucose and insulin levels (26).
C. D H E A and Bone Metabolism There are neither animal models nor epidemiologic studies examining the relationship between D H E A and bone metabolism. Nonetheless, DHEA(S) may be linked to bone turnover, as its levels parallel declining bone mineral density (BMD) with advancing age. In clinical trials, oral D H E A for up to 12 months showed no effect on BMD or urinary collagen crosslink excretion (8,23,32,37). In contrast, a single trial of 10% D H E A cream administered daily for 12 months resulted in a small but significant increase in BMD (Table 57.2) (38).
D. DHEA and Memory, Mood, Cognition, and Libido In mouse models, D H E A functions as a neurosteroid via the N-methyl-D-aspartate (NMDA) receptor (39). Using this signaling pathway, D H E A and DHEA(S) stimulate axonal and dendritic growth from neocortical neurons (39). Additional work in mice suggests that decreased D H E A levels may contribute to the vulnerability of the aging brain to ischemic insult (40). In aging mice, improved memory, decreased anxiety, and improved maze performance have been reported with D H E A supplementation (41,42). Reports in epidemiologic studies are consistent with the in vitro findings. One was a 4-year study of more than
TABLE 57.2 Effects of Percutaneous Administrations of Dehydroepiandrosterone (DHEA) for 6 and 12 Months on Bone Mineral Density in Total Hip, Ward's Triangle, and Lumbar Spine (n = 14) Site
Pretreatment
6 months DHEA
12 months DHEA
Total hip Ward's triangle Lumbar spine
0.744 + 0.021 0.486 + 0.026 0.829 _+ 0.030
0.753 _+ 0.023* 0.500 ___0.026 0.835 _+ 0.032
0.758 _+ 0.025* 0.494 _+ 0.026 0.839 _+ 0.033
Values are expressedas grams per square centimeter. *p < 0.05, DHEA treatment versus pretreatmentvalue. From ref. 38, with permission.
826
DUDLEY AND BUSTER
600 individuals from an elderly, community-based population. It reported significantly lower values o f D H E A ( S ) in w o m e n with functional limitation, depressive symptomatology, and poor subjective perception of health and life satisfaction (43). In clinical trials, D H E A improves psychologic well-being and the sleep pattern (23,44). Uncontrolled studies also report sustained improvement in the Kupperman score, an index of vasomotor and psychologic symptoms (45-47). In a randomized, double-blind, placebo-controlled trial, D H E A 50 mg/day in 60 perimenopausal women did not result in any significant improvements in perimenopausal symptoms, mood, libido, cognition, memory, or well-being (31). In women with adrenal insufficiency, results were more impressive. Oral D H E A 50 mg given daily to 24 women in a randomized, double-blind trial showed a significant improvement in libido, as well as both physical and mental sexual satisfaction (Table 57.3) (48).
TABLE 57.3
Additional trials evaluating therapeutic D H E A have appeared recently in psychiatric journals. Unfortunately, none were carried out in postmenopausal women. Perhaps future studies in this unique population will find improvements in major depression, adaptation to stress, and decreased suicidality similar to those demonstrated in younger adults (49-51).
IV. SAFETY Animal studies point to side effects including stimulation of prostatic cell lines, hepatomegaly, and hepatic cancer (52-54). H u m a n trials of oral D H E A are consistently associated with androgenic side effects such as hirsutism, acne, and increased sebum production leading to significant dropout rates in more lengthy trials (55). T h e long-term effect of dyslipidemia and hepatic dysfunction will continue to be an
Scores on Visual-Analog Scales of Sexual Activity Before, During, and After Treatment with Dehydroepiandrosterone ( D H E A ) or Placebo*
Variable
Baseline
1 month
4 months
31 _+20
Frequency of sexual thoughts or fantasies
Placebo DHEA
40 _+ 21 29 -+ 14'
34 -+ 23 44 _+ 23*
27 + 9 42 _+ 23 w
33 _+ 24 31 _+ 15
35 _+ 19 48 -+ 234
27 _+ 20 45 _+ 2211
35 _+22
Degree of sexual interest
Placebo DHEA Level of mental satisfaction with sex Placebo DHEA Level of physical satisfaction with sex Placebo DHEA
Final examination
46 _+ 30 41 _+ 26 34 -+ 25
44 -+ 31 46 -+ 28
36 _+ 30 55 _+ 25** 42 _+ 27
44 _+ 26 32 _+ 20*
43 _+ 31 49 _+ 27
34 _ 29 51 _+ 26**
*Plus-minus values are means _+ SD. The total length of the scale was 100 mm; lower values indicate a lower frequency, degree of interest, or level of satisfaction. *P = 0.04 for the comparison with the placebo group at baseline. *P = 0.007 for the comparison with the absolute change from baseline in the placebo group. P = 0.07 for the comparison with the value at 1 month in the placebo group. w < 0.001 for the comparison with the absolute change from baseline in the placebo group. P = 0.006 for the comparison with the value at 4 months in the placebo group. ~P = 0.03 for the comparison with the absolute change from baseline in the placebo group. P = 0.06 for the comparison with the value at 1 month in the placebo group. lip = 0.01 for the comparison with the absolute change from baseline in the placebo group. P -- 0.002 for the comparison with the value at 4 months in the placebo group. **P = 0.002 for the comparison with the absolute change from baseline in the placebo group. P = 0.009 for the comparison with the value at 4 months in the placebo group. **P = 0.001 for the comparison with the absolute change from baseline in the placebo group. P = 0.02 for the comparison with the value at 4 months in the placebo group. From ref. 48, with permission.
CHAPTER 57 Alternative Therapy: Dehydroepiandrosterone for Menopausal H o r m o n e Replacement
issue with oral DHEA. Unfortunately, over-the-counter availability of D H E A is not evidence of safety. In studies to date involving transdermal administration, there are no reports of significant adverse effects.
V. C O N C L U S I O N S A cause-effect relationship between decreasing D H E A S and accelerated aging, although intriguing, remains unproven (43). However, some animal and many epidemiologic studies are promising. Together they drive a continued quest to develop a physiologic, non-oral delivery system. Because oral D H E A undergoes significant first-pass metabolism, high doses are required, placing individuals at risk for side effects. Continued investigation of transdermal D H E A is needed. Recently, D H E A cream has been shown to restore a hormonal milieu similar to that seen in the late follicular phase of a reproductive-age woman (27). Although effective, these creams require cumbersome (multiple sites of application) regimens to attain adequate serum levels. Likewise, a patch is not practical as the doses necessary would require an unacceptably large surface area. Consequently, a gel-based delivery system or a spray is the likely future solution. Oral D H E A cannot be recommended as an alternative treatment for menopausal women. Therapeutic non-oral D H E A , however, continues to hold substantial promise and is in great need of future investigation.
References 1. Abraham GE. Ovarian and adrenal contribution to peripheral androgens during the menstrual cycle.J Clin Endocrinol Metab 1974;39:340- 346. 2. Longcope C. Adrenal and gonadal androgen secretion in normal females. Clin EndocrinolMetab 1986;15:213-228. 3. Buster JE, Casson PR. Where androgens come from, what controls them, and whether to replace them. In: Lobo R, ed. Treatment of the postmenopausal woman: basic and clinical aspects. Philadelphia: Lippincott, Williams & Wilkins, 1999:141-156. 4. Labrie F, Belanger A, Luu-The V, et al. DHEA and the intracrine formation of androgens and estrogens in peripheral target tissues: its role during aging. Steroids 1998;63:322-328. 5. Labrie F, Belanger A, Cusan L, Gomez JL, Candas B. Marked decline in serum concentrations of adrenal C19 sex steroid precursors and conjugated androgen metabolites during aging.J Clin EndocrinolMetab 1997;82:2396- 2402. 6. Labrie E Intracrinology. Mol Cell Endocrino11991;78:Cl13-Cl18. 7. Liu CH, Laughlin GA, Fischer UG, Yen SS. Marked attenuation of ultradian and circadian rhythms of dehydroepiandrosterone in postmenopausal women: evidence for a reduced 17,20-desmolase enzymatic activity. J Clin Endocrinol Metab 1990;71:900-906. 8. Casson PR, Faquin LC, Stentz FB et al. Replacement of dehydroepiandrosterone enhances T-lymphocyte insulin binding in postmenopausal women. Fertil Steri11995;63:1027-1031. 9. Davis SR, Burger HG. Clinical review 82: androgens and the postmenopausal woman. J Clin Endocrinol Metab 1996;81:2759-2763.
827
10. Solerte SB, Fioravanti M, Vignati G, et al. Dehydroepiandrosterone sulfate enhances natural killer cell cytotoxicity in humans via locally generated immunoreactive insulin-like growth factor I.J Clin Endocrinol Metab 1999;84:3260- 3267. 11. Barrett-Connor E, Khaw KT, Yen SS. A prospective study of dehydroepiandrosterone sulfate, mortality, and cardiovascular disease. N E n g l J Med 1986;315:1519-1524. 12. Zumoff B, Levin J, Rosenfeld RS, et al. Abnormal 24-hr mean plasma concentrations of dehydroisoandrosterone and dehydroisoandrosterone sulfate in women with primary operable breast cancer. Cancer Res 1981 ;41:3360- 3363. 13. Small M, Gray CE, Beastall GH, MacCuish AC. Adrenal androgens in insulin-dependent diabetes mellitus. Diabetes Res 1989;11:93-95. 14. Barrett-Connor E, Friedlander NJ, Khaw KT. Dehydroepiandrosterone sulfate and breast cancer risk. Cancer Res 1990;50:6571-6574. 15. Suzuki M, Kanazawa A, Hasegawa M, Hattori Y, Harano Y. A close association between insulin resistance and dehydroepiandrosterone sulfate in subjects with essential hypertension. EndocrJ 1999;46:521-528. 16. Roth G S, Mattison JA, Ottinger MA, et al. Aging in rhesus monkeys: relevance to human health interventions. Science2004;305:1423-1426. 17. Roth GS, Lane MA, lngram DK, et al. Biomarkers of caloric restriction may predict longevity in humans. Science 2002;297:811. 18. Lane MA, Ingram DK, Ball SS, Roth GS. Dehydroepiandrosterone sulfate: a biomarker of primate aging slowed by calorie restriction. J Clin Endocrinol Metab 1997;82:2093 - 2096. 19. Minghetti P, Cilurzo F, Casiraghi A, Montanari L, Santoro A. Development of patches for the controlled release of dehydroepiandrosterone. Drug Dev Ind Pharm 2001;27:711-717. 20. Nestler JE, Barlascini CO, Clore JN, Blackard WG. Dehydroepiandrosterone reduces serum low density lipoprotein levels and body fat but does not alter insulin sensitivity in normal men. J Clin Endocrinol Metab 1988;66:57-61. 21. Buster JE, Casson PR, Straughn AB, et al. Postmenopausal steroid replacement with micronized dehydroepiandrosterone: preliminary oral bioavailability and dose proportionality studies. Am J Obstet Gynecol 1992;166:1163-1168. 22. Casson PR, Faquin LC, Stentz FB, et al. Replacement of dehydroepiandrosterone enhances T-lymphocyte insulin binding in postmenopausal women. Fertil Steri11995;63:1027-1031. 23. Morales AJ, Nolan JJ, Nelson JC, Yen SS. Effects of replacement dose of dehydroepiandrosterone in men and women of advancing age. J Clin Endocrinol Metab 1994;78:1360-1367. 24. Casson PR, Santoro N, Elkind-Hirsch K, et al. Postmenopausal dehydroepiandrosterone administration increases free insulin-like growth factor-I and decreases high-density lipoprotein: a six-month trial. Fertil Steri11998;70:107-110. 25. Casson PR, Straughn AB, Umstot ES, et al. Delivery of dehydroepiandrosterone to premenopausal women: effects of micronization and nonoral administration. Am J Obstet Gyneco11996;174:649-653. 26. Diamond P, Cusan L, Gomez JL, Belanger A, Labrie F. Metabolic effects of 12-month percutaneous dehydroepiandrosterone replacement therapy in postmenopausal women. J Endocrino11996;150:$43- $50. 27. Labrie F, Belanger A, Belanger P, et al. Metabolism of DHEA in postmenopausal women following percutaneous administration. J Steroid Biochem Mol Bio12007;103:178-188. 28. Gordon GB, Bush DE, Weisman HE Reduction of atherosclerosis by administration of dehydroepiandrosterone. A study in the hypercholesterolemic New Zealand white rabbit with aortic intimal injury. J Clin Invest 1988;82:712-720. 29. Eich DM, Nestler JE, Johnson DE, et al. Inhibition of accelerated coronary atherosclerosis with dehydroepiandrosterone in the heterotopic rabbit model of cardiac transplantation. Circulation 1993;87:261 - 269. 30. Barrett-Connor E, Khaw KT. Absence of an inverse relation of dehydroepiandrosterone sulfate with cardiovascular mortality in postmenopausal women. NEnglJMed 1987;317:711.
828 31. Barnhart KT, Freeman E, Grisso JA, et al. The effect of dehydroepiandrosterone supplementation to symptomatic perimenopausal women on serum endocrine profiles, lipid parameters, and health-related quality of life. J Clin Endocrino/Metab 1999;84:3896- 3902. 32. Casson PR, Santoro N, Elldnd-Hirsch K, et al. Postmenopausal dehydroepiandrosterone administration increases free insulin-like growth factor-I and decreases high-density lipoprotein: a six-month trial. Fertil Steri11998;70:107-110. 33. Coleman DL, Leiter EH, Schwizer RW. Therapeutic effects of dehydroepiandrosterone (DHEA) in diabetic mice. Diabetes 1982;31:830-833. 34. Gansler TS, Muller S, Cleary MR Chronic administration of dehydroepiandrosterone reduces pancreatic beta-cell hyperplasia and hyperinsulinemia in genetically obese Zucker rats. Proc Soc Ex]) Biol Med 1985;180:155-162. 35. Schriock ED, Buffington CK, Hubert GD, et al. Divergent correlations of circulating dehydroepiandrosterone sulfate and testosterone with insulin levels and insulin receptor binding. J Clin Endocrinol Metab 1988;66:1329-1331. 36. Bates G W Jr, Egerman RS, Umstot ES, Buster JE, Casson PR. Dehydroepiandrosterone attenuates study-induced declines in insulin sensitivity in postmenopausal women.Ann NYAcad Sci 1995;774:291-293. 37. Villareal DT, HolloszyJO, Kohrt WM. Effects of DHEA replacement on bone mineral density and body composition in elderly women and men. Clin Endocrinol (Oxf) 2000;53:561-568. 38. Labrie F, Diamond P, Cusan L, et al. Effect of 12-month dehydroepiandrosterone replacement therapy on bone, vagina, and endometrium in postmenopausal women. J Clin Endocrinol Metab 1997;82:3498-3505. 39. Compagnone NA, Mellon SH. Dehydroepiandrosterone: a potential signalling molecule for neocortical organization during development. Proc NatlAcad Sci U S A 1998;95:4678-4683. 40. Kimonides VG, Khatibi NH, Svendsen CN, Sofroniew MV, Herbert J. Dehydroepiandrosterone (DHEA) and DHEA-sulfate (DHEAS) protect hippocampal neurons against excitatory amino acid-induced neurotoxicity. Proc NatlAcad Sci U S A 1998;95:1852-1857. 41. Flood JF, Roberts E. Dehydroepiandrosterone sulfate improves memory in aging mice. Brain Res 1988;448:178-181. 42. Melchior CL, Ritzmann RE Dehydroepiandrosterone is an anxiolytic in mice on the plus maze. Pharmaco! Biochem Behav 1994;47:437-441. 43. Bert C, Lafont S, Debuire B, Dartigues JF, Baulieu EE. Relationships of dehydroepiandrosterone sulfate in the elderly with functional, psychological, and mental status, and short-term mortality: a French community-based study. Proc Nat! Acad Sci U S A 1996;93: 13410-13415.
DUDLEY AND BUSTER
44. Friess E, Trachsel L, Guldner J, et al. DHEA administration increases rapid eye movement sleep and EEG power in the sigma frequency range. Am J Physio11995;268:E107 - El13. 45. Stomati M, Monteleone P, Casarosa E, et al. Six-month oral dehydroepiandrosterone supplementation in early and late postmenopause. Gyneco/ Endocrino/ 2000;14:342- 363. 46. Stomati M, Rubino S, Spinetti A, et al. Endocrine, neuroendocrine and behavioral effects of oral dehydroepiandrosterone sulfate supplementation in postmenopausal women. Gyneco/Endocrino11999;13:15-25. 47. Genazzani AD, Stomati M, Bernardi F, et al. Long-term low-dose dehydroepiandrosterone oral supplementation in early and late postmenopausal women modulates endocrine parameters and synthesis of neuroactive steroids. Fertil Steri12003;80:1495-1501. 48. Arlt W, Callies F, van Vlijmen JC, et al. Dehydroepiandrosterone replacement in women with adrenal insufficiency. NEnglJMed 1999;341: 1013-1020. 49. Butterfield MI, Stechuchak KM, Connor KM, et al. Neuroactive steroids and suicidality in posttraumatic stress disorder. Am J Psychiatry 2005;162:380-382. 50. Charney DS. Psychobiological mechanisms of resilience and vulnerability: implications for successful adaptation to extreme stress. Am J Psychiatry 2004;161:195-216. 51. Wolkowitz OM, Reus VI, KeeNer A, et al. Double-blind treatment of major depression with dehydroepiandrosterone. Am J Psychiatry 1999; 156:646-649. 52. Schiller CD, Schneider MR, Hartmann H, et al. Growth-stimulating effect of adrenal androgens on the R3327 Dunning prostatic carcinoma. Urol Res 1991;19:7-13. 53. Rao MS, Subbarao V, Yeldandi AV, Reddy JK. Hepatocarcinogenicity of dehydroepiandrosterone in the rat. Cancer Res 1992;52:2977-2979. 54. Milewich L, Catalina F, Bennett M. Pleotropic effects of dietary D H E A . A n n N Y A c a d Sci 1995;774:149-170. 55. Labrie F, Diamond P, Cusan L, et al. Effect of 12-month dehydroepiandrosterone replacement therapy on bone, vagina, and endometrium in postmenopausal women. J Clin Endocrino/ Metab 1997;82:3498-3505.
2 H A P T E R .5{
Melatonin RALF C. ZIMMERMANN
Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology, Columbia University College of Physicians and Surgeons, New York, NY 10032
JAMESM. OLCESE Department of Biomedical Sciences, Florida State University College of Medicine, Tallahassee, FL 32306
I. CHEMISTRY, SYNTHESIS, AND METABOLISM
produces the hormone melatonin. Melatonin, which is lipophilic and hydrophilic, can leave and enter cells by simple diffusion. It is transported in the blood in an albuminbound form and cleared quickly (half-life less than 40 minutes). Melatonin (see Fig. 58.1) (see color insert) is metabolized in the liver to 6-hydroxymelatonin, which undergoes conjugation to either sulfate or glucuronide and is excreted mainly into urine (6). The enzyme arylalkylamine N-acetyltransferase controls the daily rhythm in pineal melatonin production and blood melatonin concentration. The activity of this enzyme increases substantially at night in the absence of light but is inhibited in the presence of light (6). Regulation of the expression of this gene occurs through [31-receptors, which are located in the membrane of pinealocytes. Binding of norepinephrine or a compound with ~-agonist activity will stimulate a Gstimulatory protein complex, which via adenylyl cyclase will produce cyclic adenosine monophosphate (cAMP), which will activate cAMP-dependent protein kinase to phosphorylate, a cAMP response element binding protein (CREB), to form phosphorylated CREB. It is likely that this process might be enhanced by norepinephrine's action on the ]31-adrenergic receptor located in the membrane of pinealocytes. The night-day ratios of pineal AANAT mRNA varies from more than 150:1 in the rat to 1.5:1 in sheep. Tissuespecific AANAT mRNA expression in the human is high in the pineal gland and retina, low in other areas of the brain,
The indole derivative N-acetyl-5-methoxytryptamine, which is better known as melatanin, was discovered by Lerner et al. (1) in 1958. This hormone is mainly synthesized by pinealocytes located in the pineal gland (2,3). The pineal gland is a small cone-shaped structure attached to the roof of the third ventricle between the superior colliculi, immediately adjacent to the habenular commissure (Fig. 58.1) (see color insert) (4). Melatonin is produced from the essential amino acid tryptophan, which is taken up by the gland passively from the bloodstream. Blood tryptophan shows a circadian rhythm (5), which very likely does not influence melatonin production as the concentration of the melatonin precursor 5-hydroxytryptamine in the pineal gland is high (6). Tryptophan is converted to 5-hydroxytryptophan in a reaction catalyzed by the enzyme tryptophan hydroxylase. The next step in melatonin production is the decarboxylation of 5-hydroxytryptophan by an L-amino-decarboxylase, which converts it to 5hydroxytryptamine, better known by the name serotonin. Serotonin is then converted to L-acetyl-5-hydroxytryptamine by an arylalkylamine N-acetyltransferase (AANAT), the rate-limiting step. This enzyme, which has been cloned, plays a key role in the regulation of melatonin production [7). The human AANAT gene is located on chromosome 17@5. O-methylation of N-acetyl-5-hydroxytryptamine TREATMENT OF THE POSTMENOPAUSAL WOMAN
829
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
830
ZIMMERMANN AND OLCESE
FIGURE 58.1
Regulationof melatonin secretion. (From res 97, with permission.)
and undetectable in peripheral tissues (7). It is of note that repeated stimulation of the pinealocyte with norepinephrine does not tonically elevate AANAT activity or AANAT mRNA. Therefore, stimulation of [31-receptors seems to induce a second mechanism, which counteracts AANAT activity. Inducible cAMP early response element (ICER) appears to function as such an inhibitory factor (8).
II. R E G U L A T I O N OF MAMMALIAN PINEAL MELATONIN RHYTHM AND
DETECTION
IN BLOOD
The rhythm of pineal melatonin secretion is regulated by a well-characterized neural circuit that drives the activity of pinealocytes to produce and secrete melatonin (9). Circadian stimulatory signals originate in the suprachiasmatic nucleus (SCN) (see Fig. 58.1) (see color insert), the principal circadian pacemaker in the mammalian brain; they are transmitted to the pineal gland through the hypothalamic paraventricular nucleus via the medial forebrain bundle to the intermediolateral cell column of the upper spinal cord, which in turn innervates the superior cervical ganglia (9). These ganglia innervate the pineal gland with norepinephrine-containing fibers. Release of this neurotransmitter increases AANAT mRNA, as described earlier (10). The SCN is the circadian pacemaker that is not dependent on the dark-light cycle, but rather functions autonomously. Input or entrainment path-
ways establish the precise period and phase of the pacemaker (i.e., it assures appropriate adjustment to the 24-hour lightdark cycle). Light, an entrainment signal (Zeitgeber), acts through the retino-hypothalamic pathway (see Fig. 58.1) (see color insert) to reset the clock (SCN), and via the SCN it makes subtle adjustments in the duration and intensity of stimulation of the pineal gland and its output signal melatonin. Blind individuals with abnormal retinal processing or a defective retino-hypothalamic tract can have free-running rhythms (11). The power of the input signal light in altering SCN activity and its impact on the pineal gland and melatonin secretion are exemplified by the following example: Light exposure at a time of high melatonin secretion, such as nighttime, blocks the SCN/pineal gland transmission, thereby terminating norepinephrine release into the pineal extracellular space, which promptly lowers melatonin production and release (12,13). Blood melatonin concentration is low during the daytime (20 to 25 pmoL/L), and increases at night with peak levels between 2 AM and 4 AM of 150 to 200 pmoL/L (1 pg/mL = 4.31 pmoL/L) (14). A similar pattern is seen when metabolites of melatonin are measured in urine; for example, 24-hour urine measurements of 6-hydroxymelatonin sulfate reflect 24-hour plasma melatonin secretion adequately (15). Melatonin production or secretion is not influenced by gender. Wide interindividual variation in melatonin secretion is seen, but the intraindividual secretion pattern is relatively stable (16,17). As mentioned, blind people whose retina is no longer competent to inform the SCN about light and darkness through the retinohypothalamic tract still secrete melatonin in a circadian
831
CHAPTER 58 Melatonin
fashion, but this secretion might no longer be entrained to light (i.e., the rhythm is free running). This is one of the reasons why these individuals might suffer from sleep-related problems (11).
III. MELATONIN RECEPTORS AND EFFECTS OF MELATONIN ON THE SUPRACHIASMATIC NUCLEUS Following the introduction of 2-(1251) iodomelatonin analogs, specific melatonin receptors were identified in the SCN but not in the pars mberalis of the pituitary gland in the human (18,19). Melatonin receptors, which have also been identified in the brain including the preoptic area, cerebral cortex, and thalamus of some mammals, could mediate the hypnotic effects of this hormone (20). Melatonin receptors in non-neuronal tissue are found in cerebral arteries, renal tubular cells, ovarian granulosa cells, uterine myometrium, Leydig cells of the testis, and prostate epithelial cells (6,21). The physiologic significance of these receptors in non-neuronal tissue is currently not clear. The dissociation constant (Kd) of this receptor is less than 100 pM (i.e., it is physiologically meaningful). The human melatonin receptors have been cloned by molecular biologic techniques (20,22). At least two receptor subtypes, MT1 and MT2, are coupled to Ginhibitory proteins. The chromosomal locations are 4q35.1 and 11q21-22, respectively (20). It has been shown recently that MT1 involves both inhibition of adenylyl cyclase and potentiation of phospholipase activation (23). Melatonin binding leads to a functional change in the SCN. SCN, cultured in vitro, continues 24-hour oscillations, with neuronal firing being maximal corresponding to the light phase and the firing being minimal corresponding to the dark phase of animals before sacrifice (24). The acute inhibitory activity of melatonin on neuronal firing seems to be mediated through activation of potassium channels activated by the f3y subunits or pertussis toxin-sensitive G proteins, which is consistent with melatonin signaling through Ginhibitoryproteins (24). This acute suppressive effect of melatonin on SCN multi-unit firing is abolished in MT1 receptor-deficient mice (24). In vitro application of physiologic melatonin concentrations to the SCN at specific time points can reset the neuronal circadian rhythms (25). Phase shifting activity by melatonin to reset the SCN is greatest prior to the onset of darkness, with robust advances of 2 to 4 hours occurring in subsequent melatonin rhythms (25). The phase shifting effect of melatonin, which is distinct from the acute inhibitory effect of melatonin on neuronal firing, might involve protein kinase C (26) and nitric oxide (27). This effect may be mediated by the MT2 receptor, as receptor knock-out mice (24) continue to show the phase shift phenomenon in response to melatonin despite the lack of MT1
receptors. Therefore, multiple signaling pathways might be used by the two melatonin receptors for different physiologic effects on the SCN. Detection ofmelatonin receptor-mediated alterations in gene expression, which might mediate these distinct effects ofmelatonin on the SCN, has recently begun to be analyzed (6). The expression of melatonin receptors is a dynamic process that might show a circadian variation in their binding capacity, as is exemplified in the SCN. Maximal binding of 125I-iodomelatonin to rat SCN melatonin receptors occurs late in the day and minimal binding in early morning around 4 AM (28). Gauer et al. (28) speculated that daily variations in plasma melatonin concentrations could be implicated in the regulation of the density of melatonin binding sites in the SCN and pars mberalis, possibly by a mechanism of desensitization of the melatonin binding sites by melatonin itself. If this is correct, administration of high doses of melatonin might not be innocuous as this could downregulate its own melatonin receptors. Also based on the observations discussed earlier (24,25), timing of exogenously administered melatonin might be crucial to cause a shift in circadian rhythms.
IV. PHYSIOLOGIC AND PATHOPHYSIOLOGIC ASPECTS OF MELATONIN SECRETION
A. Physiology In a natural setting, entrainment by melatonin may be most important during early development when retinamediated light information cannot be processed (29). Maternal melatonin secretion rhythms are maintained during pregnancy (30). During fetal life, at a time when the retina-SCN pathway has not yet formed, melatonin produced by the mother provides the developing SCN with entraining information. This fetal-maternal information keeps the fetal clock entrained and in rune with the outside world until the retinamediated entrainment becomes functional during postnatal life (29). Babies up to the age of 3 months do not have a welldeveloped circadian rhythm (31). Nighttime melatonin secretion peaks during early childhood and then drops sharply until early adulthood (32). It has been speculated that the sharp decline in melatonin secretion that occurs in late childhood might be involved in the initiation of puberty, but no strong data are available to support this assumption. Melatonin continues to decrease significantly with age; similarly, the time during which melatonin remains elevated at night also decreases with age (33). Studies in women during the perimenopause reveal that the decline in melatonin precedes follicle-stimulating hormone increase during menopause, with a decline of 41% in the age group 40 to 44 years and a
832
ZIMMERMANN AND
further decline of 35% between the age groups 50 to 54 years and 55 to 59 years (34). Whether this decline in melatonin secretion contributes to the development of menopause or to menopausal symptoms and signs is questionable. An association between the quality of sleep and the amount of melatonin secreted has been noted, especially in the elderly (35). Animal studies that included mice whose ovaries no longer functioned (similar to menopause) seemed to show that the aging process could be slowed down by the pineal grafting of young pineal glands from young animals onto aging animals. It was speculated that this inhibitory effect on aging might possibly be mediated by melatonin (36). Obviously, if these findings could be replicated, this would have important consequences in the management of menopause and aging (see later discussion). Body temperature fluctuates during the 24-hour time period, with temperature being higher during the daytime when compared with the nighttime. The drop in nighttime temperature can be reversed by exposure to bright light. Because this effect is mediated by melatonin, this hormone decreases core body temperature (37,38). The temperaturelowering effect of melatonin might of therapeutic interest because hot flushes are associated with an increase in temperature. No studies are available to answer this question. Melatonin secretion changes with alterations in the length of daytime encountered in the different seasons of the year. This has powerful effects on reproductive function in seasonal breeding animals, but if present, its effect on human reproduction seems to be minimal (39,40). Melatonin secretion does not change in different phases of the menstrual cycle or with the administration of oral contraceptives (14,17). Melatonin receptors have been identified in human granulosa cells (41), and this hormone stimulates progesterone production by human granulosa cells (42). Also, melatonin seems to concentrate in follicular fluid (43). The physiologic significance of these observations is not clear. A possible association of 24-hour melatonin secretion pattern and the initiation of the luteinizing hormone (LH) surge have been observed (43,44).
B. Pathophysiology Nocturnal melatonin secretion is amplified in women with functional hypothalamic amenorrhea (45,46). It is not clear whether increased melatonin concentration is a contributing factor in causing amenorrhea by influencing LHreleasing hormone secretion or is an epiphenomenon. Shortterm administration of supraphysiologic doses of melatonin (60 to 300 mg), which created supraphysiologic levels in the periovulatory, period did not prevent ovulation (44,47). Long-term use of high-dose melatonin (4 months, 300 mg/ day) seems to decrease the midcycle LH surge. It has been suggested that a combination of high-dose melatonin with a
OLCESE
progestin derivative (norethindrone) could be used as an estrogen-free oral contraceptive pill (47). Acute elevations of melatonin occur during fasting (48) and sustained exercise (49). The importance of these observations regarding possible impact on reproduction currently is not clear. Traveling across several time zones creates a situation in which the body's circadian rhythms, including melatonin secretion, are temporarily not properly synchronized to the prevailing light-dark cycle, which might cause jet lag (50). While readjusting their circadian rhythm to local time, individuals experience difficulty sleeping, among other signs and symptoms. Chronic circadian disturbance as encountered in shift work might cause many of the health and social problems reported by shift workers, including chronic sleep problems.
V. COMPOUNDS THAT ALTER MELATONIN SECRETION Because alterations in melatonin secretion might have an impact on the circadian rhythm and possibly sleep, it is important to be aware of substances that influence melatonin secretion. Nonsteroidal anti-inflammatory drugs (NSAIDs) (e.g., aspirin or ibuprofen) can lower melatonin secretion. These medications interfere with the production of prostaglandins, which can influence melatonin production in the pinalocyte (51). As would be expected from the regulation of melatonin secretion, [3-blockers (e.g., atenolol) abolish the nocturnal rise of melatonin secretion (15). Similarly, presynaptic blockage of norepinephrine production with cx-methyl-para-tyrosine attenuates the nocturnal melatonin peak (52). It is of note that subjects taking this compound complain of sleep disturbances for several days after stopping medication, which could be related to the disruption of the circadian rhythm of melatonin secretion. Some of the calcium channel blockers and the el2 receptor agonist clonidine decrease melatonin secretion (53,54). Decreased intracellular Ca 2+ levels seem to interfere with melatonin production, and presynaptic cx2 receptors decrease norepinephrine release. The sedative-hypnotics of the benzodiazepine type (e.g., diazepam and alprazolam), which act on the ~/-aminobutyric acid complex type A receptor complex, lower melatonin secretion (55,56). Antidepressants such as the norepinephrine reuptake inhibitor desipramine and monoamine oxidase (MAO) inhibitors seem to increase plasma melatonin secretion (57,58). Serotonin reuptake inhibitors can increase or decrease melatonin secretion (59,60). Therefore, different types of antidepressant medication might interfere with the circadian rhythm or magnitude of melatonin secretion. Caffeine and alcohol might decrease melatonin secretion, especially when consumed at nighttime (61,62). Dexamethasone decreases melatonin secretion (63). Alterations in plasma tryptophan concentration can increase or decrease melatonin secretion (64,65).
833
CHAPTER 58 Melatonin
VI. THERAPEUTIC INTERVENTIONS USING MELATONIN
A. Circadian Rhythms Jet lag is caused by traveling across several time zones within several hours. The internal clock (i.e., the SCN) is not synchronized with the daytime-nighttime rhythm at the place of arrival. Jet lag can cause insomnia, fatigue, irritability, and poor concentration. It is easier to adapt when flying west, when the day is stretched, than flying east when the day is compressed. A possible explanation is that the body clock operates on a 25-hour rhythm when no clues are given to the SCN (66). To alleviate jet lag symptoms, it is recommended to take low doses of melatonin I to 2 hours prior to going to bed in the place of destination for several days (67,68). In addition to other environmental clues, such as light, melatonin helps to reset the clock faster. Taking melatonin at a time corresponding to the beginning of the sleep period at the point of destination, starting a few days prior to departure, can inconvenience people significantly because of side effects such as drowsiness. Most of the data available seem to demonstrate that starting treatment after arrival seems to produce similar positive results when compared with starting to take melatonin several days prior to departure (69). Therefore, it is recommended to take a dose of 0.5 mg of melatonin I to 2 hours before bedtime after arrival at the destination (66,67,69). Exposure to sunlight after arrival also might help to reset the SCN (70). When traveling more than six time zones, avoiding bright light exposure at certain times might become important in facilitating readaptation, and a nomogram has been developed that helps to avoid exposure at these critical periods (71,72). A recent meta-analysis questioned the beneficial effect of treatment with melatonin for jet lag (73,74).
B. Shift Work Switching from working during the daytime to working at night is very taxing as the sleep-wake cycle is opposite to the light-dark cycle of the environment. A recent study on night workers (75) demonstrated that workers with an altered melatonin rhythm had a higher prevalence of sleep disorders, fatigue, and mental symptoms, as compared with workers with normal melatonin rhythms. To improve adjustment to these conditions, workers can expose themselves to artificial bright light at night, which resets the SCN and lowers melatonin levels (76-78). In addition, avoiding exposure to bright light during the daytime and taking melatonin (0.5 to 5 mg) or melatonin agonists such as agomelatine (79) prior to sleep seems to improve daytime sleep and alertness (66,77,80). It is not clear how beneficial an attempt to shift
the sleep-wake cycle is in the short term (--- 4 weeks). It is of note that air travelers do adjust their circadian rhythm faster than shift workers. Melatonin production that prevents the desired shift is suppressed by sunlight exposure, which is generally much brighter than the indoor light exposure for shift workers. Sunlight is also available to help shift the endogenous circadian pacemaker (70). A recent meta-analysis questioned the beneficial effect of treatment with melatonin for shift work (73,74).
C. Sleep Disturbances Sleep disturbances increase with age. Possible reasons for the high frequency of sleep complaints in elderly people are primary endogenous, age-related sleep disorders; increased likelihood of disease that interferes with sleep; and drug use, which can cause secondary sleep disorders (81). Therefore, it is important to review nonprescription and prescription drugs used should be reviewed before beginning treatment with melatonin in patients with insomnia due to physical disorders, mental health problems including depression and anxiety disorders, improper sleep environment, inadequate sleep habits, circadian cycle abnormalities, and substance abuse (82). It might also be prudent to collect nocturnal urine, preferably in 3-hour intervals, to detect a decrease in melatonin secretion or a shift in its peak secretion (81). To document improvement, a subjective sleep diary should be kept and if possible an actigraph should be used to estimate sleep variables (sleep quality, etc.) during melatonin use. Because of the rapid metabolism of orally administered melatonin, slow-release capsules as the one used by Garfinkel et al. (81) might be preferable (2 mg, Circadin, Neurim Pharmaceutical, Tel Aviv, Israel). Wurtman and Zhdanova (83) had similar positive results in sustaining sleep in elderly insomniacs using an oral dose of melatonin (0.3 mg). More recently the melatonin agonist ramelteon (Rozerem), a selective M T I / M T 2 receptor agonist, was approved by the Food and Drug Administration (FDA) for the treatment of insomnia (84). Therefore, in a carefully screened group of patients suffering from insomnia, a trial of melatonin might be indicated before switching to more powerful medications such as benzodiazepines, which have more side effects. Patients suffering from delayed sleep phase syndrome might possibly benefit from melatonin treatment (79,85). The effectiveness of melatonin in the treatment of secondary sleep disorders has been questioned in a recent meta-analysis (73,74).
D. Unproven Benefits from Melatonin Melatonin secretion can be decreased in patients with psychiatric disorders, especially depression (86). Some forms of depression seem to involve dysregulation in the
834 central nervous system of norepinephrine- and serotonincontaining neurons and their receptors (87,88). Therefore, at least from a theoretic standpoint, an attractive hypothesis is that alterations in melatonin secretion reflect disease states as both systems play an important role in the regulation of melatonin secretion (52,89). Unfortunately, no consistent changes in melatonin secretion have been found that reflect depression (32,89). Therefore, melatonin measurements cannot be used to diagnose depression, and treatment of depression with melatonin is not appropriate. Seasonal affective disorder, which is characterized by recurrent episodes of depression, hypersomnia, and augmented appetite in the autumn and winter, fulfills the criteria of a rhythm disorder (90). These patients show a change in core temperature rhythms, but melatonin rhythm and amplitude seem to be unchanged (90). The preferred therapy is timed exposure to bright light, and it is not clear whether melatonin administration might beneficial to this group of patients (90). Melatonin secretion does not seem to be altered in women with premenstrual syndrome (91). Patients suffering from anorexia nervosa might have elevated melatonin blood levels, but this has no therapeutic implications (92). Administration of 3 mg melatonin at bedtime to perimenopausal and postmenopausal women has been reported to improve thyroid function, causes a change of gonadotropins toward more juvenile levels, and might have some antidepressive activity (93,94). Use of melatonin alone for the relief of menopausal symptoms, including hot flushes, does not seem to be effective (95). In very high pharmacologic doses, melatonin works as an antioxidant, possibly through non-receptor-mediated mechanisms. It has been claimed that this property of melatonin might have a preventive effect on illnesses affected by free radicals (66). As stated by Reppert and Weaver (29), this antioxidant effect requires melatonin concentrations approximately 100 times greater than the physiologic melatonin secretion (1 nM). Therefore, an antioxidant effect of melatonin may have some therapeutic application, but definitely not to the extent claimed in self-help books (66). It has also been claimed that melatonin can reverse aging (96). Some researchers have studied strains of mice with a well-described genetic defect in pineal melatonin biosynthesis, which therefore could not make melatonin (29,36). In some of these melatonin-deficient mice strains, life span increased by 20%, but not in female C57BL/6 mice. The life span was actually shortened in the mouse strain C3H/He secondary to reproductive tract tumors (29,36). Thus, no evidence indicates that melatonin administered to melatoninproducing mice can increase longevity. The suggestion that melatonin may increase longevity in humans is based on pure speculation (29).
ZIMMERMANN AND OLCESE
VII. SUMMARY Molecular biology has been able to disperse some of the myths surrounding melatonin. Melatonin production sites can be clearly identified by detecting message and protein from AANAT in specific tissues. Specific melatonin receptors have been identified by detecting message and protein in specific tissues. Activation of specific signaling pathways used by different melatonin receptors can be detected. Also, knock-out models will help to further clarify the physiologic action of this hormone as well as its pathophysiologic states. This grounding in molecular biology will help to develop sound therapeutic applications for melatonin. It would be regrettable if melatonin did not find its proper place in the therapeutic armamentarium, given the well-defined positive actions of this fascinating hormone. References 1. Lerner BA, et al. Isolation of melatonin, pineal factor that lightens melanocytes.Jdm Chern Soc 1980;80:2587. 2. Cardinali DP. Melatonin. 1. A mammalian pineal hormone. Endocr Rev 1981;2:327-346. 3. Reiter RJ. The pineal and its hormones in the control of reproduction in mammals. Endocr Rev 1980;1:109-131. 4. Preslock JP. The pineal gland: basic implications and clinical correlations. Endocr Rev 1984;5:282-308. 5. Krahn LE, Lu PY, Klee G, et al. Examining serotonin function: a modified technique for rapid tryptophan depletion. NeuroDsychopharmacology 1996;15:325-328. 6. Olcese J. Cellular and molecular mechanisms mediating melatonin action. Aging Male 1998;1:113-129. 7. Klein DC, Roseboom PH, Coon SL. New light is shining on the melatonin rhythm enzyme. The first postcloning view. T E M 1996;7: 106-112. 8. Coon SL, Mazuruk K, Bernard M, et al. The human serotonin N-acetyltransferase (EC 2.3.1.87) gene (AANAT): structure, chromosomal localization, and tissue expression. Genomics 1996;34:76- 84. 9. Moore RY. Neural control of the pineal gland. Bebav Brain Res 1996; 73:125-130. 10. Roseboom PH, Coon SL, Baler R, et al. Melatonin synthesis: analysis of the more than 150-fold nocturnal increase in serotonin N-acetyltransferase messenger ribonucleic acid in the rat pineal gland. Endocrinology 1996;137:3033- 3045. 11. Lockley SW, et al. Relationship between melatonin rhythms and visual loss in the blind.d Clin EndocrinolMetab 1997;82:3763-3770. 12. Bispink G, et al. Influence of melatonin on the sleep-independent component of prolactin secretion, d Pineal Res 1990;8:97-106. 13. Lewy AJ, et al. Light suppresses melatonin secretion in humans. Science 1980;210:1267-1269. 14. Berga SL, Yen SS. Circadian pattern of plasma melatonin concentrations during four phases of the human menstrual cycle. Neuroendocrinology 1990;51:606-612. 15. Arendt J, et al. Immunoassay of 6-hydroxymelatonin sulfate in human plasma and urine: abolition of the urinary 24-hour rhythm with atenolol. J Clin Endocrinol Metab 1985;60:1166-1173. 16. Arendt J. Radioimmunoassayable melatonin: circulating patterns in man and sheep. Prog Brain Res 1979;52:249-258.
CHAPTER 58 Melatonin 17. Delfs TM, et al. Sex steroids do not alter melatonin secretion in the human. Hum Reprod 1994;9:49-54. 18. Reppert SM, et al. Putative melatonin receptors in a human biological clock. Science 1988;242:78- 81. 19. Weaver DR, et al. Melatonin receptors in human hypothalamus and pituitary: implications for circadian and reproductive responses to melatonin. J Clin Endocrinol Metab 1993;76:295- 301. 20. Reppert SM, Weaver DR, Godson C. Melatonin receptors step into the light: cloning and classification of subtypes. Trends Pharmacol Sci 1996;17:100-102. 21. Schlabritz-Loutsevitch N, et al. The human myometrium as a target for melatonin. J Clin Endocrinol Metab 2003;88:908-913. 22. Reppert SM, Weaver DR, Ebisawa T. Cloning and characterization of a mammalian melatonin receptor that mediates reproductive and circadian responses. Neuron 1994;13:1177-1185. 23. Godson C, Reppert SM. The Mella melatonin receptor is coupled to parallel signal transduction pathways. Endocrinology 1997;138:397-404. 24. Liu C, et al. Molecular dissection of two distinct actions of melatonin on the suprachiasmatic circadian clock. Neuron 1997;19:91-102. 25. Gillette MU, McArthur AJ. Circadian actions of melatonin at the suprachiasmatic nucleus. Behav Brain Res 1996;73:135-139. 26. McArthur AJ, Hunt AE, Gillette MU. Melatonin action and signal transduction in the rat suprachiasmatic circadian clock: activation of protein kinase C at dusk and dawn. Endocrinology 1997;138:627-634. 27. Starkey SJ. Melatonin and 5-hydroxytryptamine phase-advance the rat circadian clock by activation of nitric oxide synthesis. Neurosci Lett 1996;211:199-202. 28. Gauer F, et al. Daily rhythms of melatonin binding sites in the rat pars tuberalis and suprachiasmatic nuclei; evidence for a regulation of melatonin receptors by melatonin itself. Neuroendocrinology 1993;57:120-126. 29. Reppert SM, Weaver DR. Melatonin madness. Cell 1995;83: 1059-1062. 30. Kivela A. Serum melatonin during human pregnancy. Acta Endocrinol (Copenh) 1991;124:233-237. 31. Jaldo-Alba F, Munoz-Hoyos A, Molina-Carballo A, et al. Light deprivation increases plasma levels of melatonin during the first 72 h of life in human infants. Acta Endocrinol (Copenh) 1993;129:442-445. 32. Waldhauser F, Ehrhart B, Forster E. Clinical aspects of the melatonin action: impact of development, aging, and puberty, involvement of melatonin in psychiatric disease and importance of neuroimmunoendocrine interactions. Experientia 1993;49:671-681. 33. Nair NP, et al. Plasma melatonin--an index of brain aging in humans? Biol Psychiatry 1986;21:141-150. 34. Vakkuri O, et al. Decrease in melatonin precedes follicle-stimulating hormone increase during perimenopause. EurJEndocrinol 1996;135: 188-192. 35. Haimov I, et al. Sleep disorders and melatonin rhythms in elderly people. BMJ 1994;309:167. 36. Pierpaoli W, Regelson W. Pineal control of aging: effect of melatonin and pineal grafting on aging mice. Proc NadAcad Sci U S A 1994;91: 787-791. 37. Strassman RJ, et al. Elevated rectal temperature produced by all-night bright light is reversed by melatonin infusion in men. J Appl Physiol 1991;71:2178-2182. 38. Cagnacci A, Elliott JA, Yen SS. Melatonin: a major regulator of the circadian rhythm of core temperature in humans. J Clin Endocrinol Metab 1992;75:447-452. 39. Cagnacci A, Volpe A. Influence of melatonin and photoperiod on animal and human reproduction. J Endocrinol Invest 1996;19:382-411. 40. Kauppila A, et al. Inverse seasonal relationship between melatonin and ovarian activity in humans in a region with a strong seasonal contrast in luminosity. J Clin Endocrinol Metab 1987;65:823- 828. 41. Yie SM, Niles LP, Younglai EV. Melatonin receptors on human granulosa cell membranes. J Clin Endocrinol Metab 1995;80:1747-1749.
835
42. Webley GE, Luck MR. Melatonin directly stimulates the secretion of progesterone by human and bovine granulosa cells in vitro. J Reprod Ferti11986;78:711 - 717. 43. Brzezinsld A, et al. Melatonin in human preovulatory follicular fluid. J Clin EndocrinolMetab 1987;64:865-867. 44. Zimmermann RC, et al. Melatonin and the ovulatory luteinizing hormone surge. Fertil Steri11990;54:612-618. 45. Berga SL, Mortola JF, Yen SS. Amplification of nocturnal melatonin secretion in women with functional hypothalamic amenorrhea. J Clin Endocrinol Metab 1988;66:242-244. 46. Brzezinski A, et al. The circadian rhythm of plasma melatonin during the normal menstrual cycle and in amenorrheic women.J Clin Endocrinol Metab 1988;66:891 - 895. 47. Voordouw BC, et al. Melatonin and melatonin-progestin combinations alter pituitary-ovarian function in women and can inhibit ovulation. J Clin Endocrinol Metab 1992;74:108 - 117. 48. Beitins IZ, et al. Hormonal responses to short term fasting in postmenopausal women. J Clin Endocrinol Metab 1985;60:1120-1126. 49. Carr DB, et al. Plasma melatonin increases during exercise in women. J C/in Endocrinol Metab 1981;53:224-225. 50. Bellamy N. The jet lag phenomenon: etiology, pathogenesis, clinical features, and management. Modern Med Can 1986;41:717-732. 51. Cardinali DP, et al. Role of prostaglandins in rat pineal neuroeffector junction. Changes in melatonin and norepinephrine release in vitro. Endocrinology 1982;111:530-534. 52. Zimmermann RC, et al. Inhibition of presynaptic catecholamine synthesis with alpha-methyl-para-tyrosine attenuates nocturnal melatonin secretion in humans. J Clin Endocrinol Metab 1994;79:1110-1114. 53. Meyer AC, et al. Dihydropyridine calcium antagonists depress the amplitude of the plasma melatonin cycle in baboons. Life Sci 1986;39: 1563-1569. 54. Lewy AJ, et al. Clonidine reduces plasma melatonin levels. J Pharm Pharmaco11986;38:555-556. 55. Monteleone P, et al. Preliminary observations on the suppression of nocturnal plasma melatonin levels by short-term administration of diazepam in humans.J Pineal Res 1989;6:253-258. 56. McIntyre IM, Burrows GD, Norman TR. Suppression of plasma melatonin by a single dose of the benzodiazepine alprazolam in humans. Biol Psychiatry 1988;24:108-112. 57. Skene DJ, Bojkowski CJ, Arendt J. Comparison of the effects of acute fluvoxamine and desipramine administration on melatonin and cortisol production in humans. BrJ Clin Pharmaco11994;37:181-186. 58. Oxenkrug G, et al. Effect of selective monoamine oxidase inhibitors on rat pineal melatonin synthesis in vitro. J Pineal Res 1988;5:99-109. 59. Demisch K, et al. Melatonin and cortisol increase after fluvoxamine. BrJ Clin Pharmaco11986;22:620-622. 60. Childs PA, et al. Effect of fluoxetine on melatonin in patients with seasonal affective disorder and matched controls. Br J Psychiatry 1995;166:196-198. 61. Wright KP Jr, et al. Caffeine and light effects on nighttime melatonin and temperature levels in sleep-deprived humans. Brain Res 1997;747:78-84. 62. Ekman AC, et al. Ethanol inhibits melatonin secretion in healthy volunteers in a dose-dependent randomized double blind cross-over study. J Clin Endocrinol Metab 1993;77:780- 783. 63. Demisch L, Demisch K, Nickelsen T. Influence of dexamethasone on nocturnal melatonin production in healthy adult subjects. J Pineal Res 1988;5:317-322. 64. Hajak G, et al. The influence of intravenous L-tryptophan on plasma melatonin and sleep in men. Pharmacopsychiatry 1991;24:17-20. 65. Zimmermann RC, et al. Effects of acute tryptophan depletion on nocturnal melatonin secretion in humans. J Clin Endocrinol Metab 1993;76:1160-1164. 66. Reiter RJ, Robinson RP. Mdatonin. Breakthrough discoveries that can help you, vol 39. New York: Batam, 1995.
836 67. Petri KA, et al. A double-blind trial of melatonin as a treatment for jet lag in an international cabin crew. Biol Psychiatry 1993;33:526-530. 68. Claustrat B, et al. Melatonin and jet lag: confirmatory result using a simplified protocol. Biol Psychiatry 1992;32:705-711. 69. Lino A, et al. Melatonin and jet lag: treatment schedule. BiolPsychiatry 1993;34:587. 70. Lewy AJ, Ahmed S, Sack RL. Phase shifting the human circadian clock using melatonin. Behav Brain Res 1996;73:131-134. 71. Daan S, Lewy AJ. Scheduled exposure to daylight: a potential strategy to reduce "jet lag" following transmeridian flight. PsychopharmacolBu# 1984;20:566-568. 72. Arendt J, Skene DJ. Melatonin as a chronobiotic. Sleep Med Rev 2005;9:25-39. 73. Buscemi N, et al. Efficacy and safety of exogenous melatonin for secondary sleep disorders and sleep disorders accompanying sleep restriction: meta-analysis. BMJ 2006;332:385-393. 74. Arendt J. Does melatonin improve sleep? Efficacy of melatonin. BMJ 2006;332:550. 75. Butch JB, et al. Melatonin, sleep, and shift work adaptation. J Occup Environ Med 2005;47:893-901. 76. Bojkowsld CJ, et al. Suppression of nocturnal plasma melatonin and 6-sulphatoxymelatonin by bright and dim fight in man. Horm Metab Res 1987;19:437-440. 77. Eastman CI, et al. Dark goggles and bright fight improve circadian rhythm adaptation to night-shift work. Sleep 1994;17:535-543. 78. Dawson D, Encel N, Lushington K. Improving adaptation to simulated night shift: timed exposure to bright light versus daytime melatonin administration. Sleep 1995;18:11-21. 79. Leproult R, et al. Phase-shifts of 24-h rhythms of hormonal release and body temperature following early evening administration of the melatonin agonist agomelatine in healthy older men. Clin Endocrinol (Oxf) 2005;63:298- 304. 80. Folkard S, Arendt J, Clark M. Can melatonin improve shift workers' tolerance of the night shift? Some preliminary findings. ChronobiolInt 1993;10:315-320. 81. Garfinkel D, et al. Improvement of sleep quality in elderly people by controlled-release melatonin. Lancet 1995;346:541-544. 82. Sahelian R, Lovett PN, Bouvet JMC. Melatonin. Nature's sleepingpill. Garden City Park, NY: Avery, 1995:42-43.
ZIMMERMANN AND OLCESE
83. Wurtman RJ, Zhdanova I. Improvement of sleep quality by melatonin. Lancet 1995;346:1491. 84. Kato K, et al. Neurochemical properties oframelteon (TAK-375), a selective MT1/MT2 receptor agonist. Neuropharmacology2005;48:301-310. 85. Oldani A, et al. Melatonin and delayed sleep phase syndrome: ambulatory polygraphic evaluation. Neuroreport 1994;6:132-134. 86. Wetterberg L, et al. Circadian rhythms in melatonin and cortisol secretion in depression.Adv Biochem Psychopharmaco11984;39:197-205. 87. Schatzberg AF, Schildkraut JJ. Recent studies on norepinphrine systems in mood disorders. In: Bloom FE, Kupfer DJ, eds. Psychopharmacology. The fourth generation of progress. New York: Raven, 1995: 911-920. 88. Maes M, Meltzer HY. The serotonin hypothesis of major depression. In: Bloom FE, Kupfer DJ, eds. Psychopharmacology. Thefourth generation ofprogress. New York: Raven, 1995:933-944. 89. Rubin RT, et al. Neuroendocrine aspects of primary endogenous depression. XI. Serum melatonin measures in patients and matched control subjects. Arch Gen Psychiatry 1992;49:558 - 567. 90. Wirz-Justice A. Biological rhythms in mood disorders. In: Bloom FE, Kupfer DJ, eds. Psychopharmacology. The fourth generation of progress. New York: Raven, 1995:999-1018. 91. Mortola JE The premenstrual syndrome. In: Adashi EY, Rock JA, Rosenwaks Z, eds. Reproductive medicine. New York: Raven, 1996: 1635-1647. 92. Kennedy SH, et al. Pineal and adrenal function before and after refeeding in anorexia nervosa. Biol Psychiatry 1991;30:216-224. 93. Bellipanni G, et al. Effects of melatonin in perimenopausal and menopausal women: our personal experience. Ann N Y Acad Sci 2005; 1057:393-402. 94. Bellipanni G, et al. Effects of melatonin in perimenopausal and menopausal women: a randomized and placebo controlled study. Exp Gerontol 2001 ;36:297- 310. 95. Secreto G, et al. Soy isoflavones and melatonin for the relief of climacteric symptoms: a multicenter, double-blind, randomized study. Maturitas 2004;47:11-20. 96. Pierpaoli W, Regelson W. The melatonin miracle. New York: Simon and Schuster, 1995. 97. Brzezinski A. Melatonin in humans. NEnglJMed 1997;336:186-195.
--, ( _~HAPTER 5 .
Selective Estrogen Receptor Modulators FELICIA COSMAN
Helen Hayes Hospital, West Haverstraw, NY 10993; Columbia University College of Physicians and Surgeons, New York, NY 10032
ROBERT LINDSAY
Helen Hayes Hospital, West Haverstraw, NY 10993; Columbia University College of Physicians and Surgeons, New York, NY 10032
I. I N T R O D U C T I O N
increase in mammographic breast density. Tibolone also produces metabolites that have progestogenic and androgenic activities, including increasing libido. Tibolone is not a SERM and is not currently approved for use in the United States, so it will not be discussed further in this chapter. Aromatase inhibitors, which are currently approved for women with history of breast cancer and are in clinical trials for prevention of breast cancer, are a unique class and will not be considered in this chapter.
Selective estrogen receptor modulators (SERMS) are synthetic, nonsteroidal agents that bind to the estrogen receptor (ER), resulting in estrogen-agonist or estrogen-antagonist effects of varying magnitudes in different target tissues. Differences in target tissue effects are related in part to unique SERM-receptor conformations for each agent and subsequent binding of different adaptor proteins. These proteins can facilitate or inhibit transcription of different DNA sequences, producing unique physiologic effects. Only one SERM, raloxifene, is currently approved for prevention and treatment of osteoporosis. Tamoxifen, originally developed for treatment of breast cancer, is also approved for prevention of breast cancer in women at elevated risk of the disease. Several other SERMS (idoxifene, levormeloxifene, and droloxifene) failed during development due to an increased risk of pelvic organ prolapse and urinary incontinence. Three new SERMS, lasofoxifene (Pfizer), bazedoxifene (Wyeth), and arzoxifene (Lilly), are currently in Phase III development for osteoporosis, but published data are limited; they will be discussed briefly at the end of the chapter. Tibolone, a synthetic steroid, produces estrogen metabolites that exert estrogenic effects on hot flushes and night sweats, in addition to antiresorptive effects on bone, with modest increases in bone mass. In contrast, in the breast, tibolone's effects are distinct from those of estrogen, with no TREATMENT OF THE POSTMENOPAUSAL WOMAN
II. T A M O X I F E N A. Effects on the Skeleton and Bone Metabolism Observational studies in postmenopausal women with breast cancer indicate that tamoxifen produces modest benefits to bone mineral density (BMD) (1) and reduces indices of bone turnover by 20% to 55% both in postmenopausal women with and without prior breast cancer (1-7). One small bone biopsy study indicated that tamoxifen reduced activation frequency, the rate at which remodeling sites are initiated, while decreasing bone resorption cavity depth or the size of resorption pits (5,8). These effects are similar to those of estrogen, but of smaller magnitude. 837
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
838
COSMAN AND LINDSAY
Randomized clinical trials in postmenopausal women, with and without history of breast cancer, show that tamoxifen augments spine BMD more than 1% per year while slowing bone loss in the forearm (2-4,7,9,10). The effect of tamoxifen on hip BMD is either neutral or mildly positive (4,9). In contrast, in premenopausal women, tamoxifen is associated with greater losses of bone in both the spine and the forearm compared with placebo-treated women, suggesting that in women with high endogenous concentrations of estrogen, tamoxifen antagonizes estrogen's effects on bone. Only two randomized trials have reported the effect of tamoxifen on fracture risk (11,12). In the first, 1716 postmenopausal women with high-risk breast cancer were randomized (after mastectomy and radiotherapy) to tamoxifen 30 mg/day (or no additional treatment) for only I year (11). Hip fracture occurrence was not reduced over as long as 8 years of follow-up. The better fracture data come from the Breast Cancer Prevention Trial (BCPT), performed in 13,388 healthy premenopausal and postmenopausal women who were at increased risk of developing breast cancer (12). This study found a trend toward a reduction in the risk of selected fractures (hip, symptomatic spine, and Colles' fractures) in those women 50 years of age or older, but no reduction in fracture risk at other skeletal sites and no suggestion of reduction in fracture risk in premenopausal women. In fact, the overall incidence of all fractures was the same in tamoxifen-treated and placebo-treated women. Furthermore, spine radiographs were not done routinely to evaluate asymptomatic vertebral compression deformity. There were few bone density tests performed, such that the prevalence of osteoporosis in this group is unknown. Thus, there is still uncertainty about the effects of tamoxifen on fracture occurrence and how these effects might differ in a group of women with known prevalence of osteoporosis. B. Effects o n O t h e r O r g a n S y s t e m s 1. BREAST
Tamoxifen treatment of women with ER-positive breast cancer for 5 years reduces the risk of mortality, recurrent breast tumors, and contralateral new primary tumors, and it increases disease-free survival (13-16). In addition, tamoxifen has been shown to reduce risk of breast cancer in highrisk women without a prior history. In the BCPT trial, the largest of four randomized trials of tamoxifen efficacy against breast cancer in healthy women (12), tamoxifen administration over 4.5 years reduced the incidence of new breast cancers (invasive and noninvasive) by approximately 50% in both premenopausal and postmenopausal women who were at elevated risk for breast cancer (5-year risk of 1.66% or more, using the Gail Model). The incidence of ER-positive breast cancers was reduced by 69%, without any
effect on the risk of ER-negative cancer. The Italian Tamoxifen Prevention Study (17) showed no overall difference in the incidence of breast cancer between tamoxifen and placebo groups (18); however, when the group of women at higher risk for breast cancer were identified retrospectively (only 13% of total study population), tamoxifen did reduce incidence of breast cancer significantly in that group, but not in the lower-risk group (19). The reduced breast cancer risk was seen for ER-positive but not ER-negative tumors. In the Royal Marsden study (20), in which breast cancer risk was determined primarily by family history of breast cancer and hormone therapy was allowed during the trial, tamoxifen reduced the incidence of breast cancer versus placebo, but the difference was not statistically significant (21). The International Breast Cancer Study (IBIS) showed a 32% lower risk of breast cancer in the tamoxifen group compared with placebo groups (22), again with the effect limited to ER-positive breast tumors. Differences between the BCPT and IBIS trials might relate to differences in sample size, eligibility criteria for the trials, breast cancer risk, and use of hormone therapy. Overall, the results from all four tamoxifen breast cancer trials show a highly significant 38% lower risk of breast cancer incidence with tamoxifen (21). 2. EffECTS ON VASCULAR DISEASE AND INTERMEDIATES
Tamoxifen reduces serum total and low-density lipoprotein (LDL) cholesterol, lipoprotein(a), LDL oxidation, and fibrinogen, with no effect on high-density lipoprotein (1). Two European randomized trials of adjuvant tamoxifen treatment in breast cancer survivors (23-25) found that tamoxifen significantly reduced the incidence of cardiovascular events by approximately 30% to 60%. These results were similar to one U.S. trial in which a modest but insignificant reduction in cardiovascular mortality was seen (26). No reduction in the risk of stroke has been described (23-26). The larger BCPT trial (12), however, found no trend toward protection against cardiovascular disease but in contrast suggested that tamoxifen might increase the risk of cerebrovascular disease. Tamoxifen decreases serum levels of antithrombin III (1) and increases the risk of venous thromboembolic disease approximately threefold in breast cancer patients (24,27-31) and in premenopausal and postmenopausal healthy women (12). Approximately 1 in every 100 postmenopausal women who is taking tamoxifen for 5 years will have either deep vein thrombosis or pulmonary embolism.
3. GYNECOLOGIC EFFECTS AND MENOPAUSAL SYMPTOMS
Tamoxifen increases the incidence of uterine cancer as well as that of benign uterine disease, including fibroid tumors, adenomyosis, and hyperplasia (1,12,27,32-36). The
CHAPTER 59 Selective Estrogen Receptor Modulators relative risk of invasive uterine cancer in women older than 50 years was 4, but all cancers in the women carefully followed in the BCPT were localized stage 1 tumors. Although tamoxifen might increase occurrence of ovarian cysts, there are no data suggesting an increased risk of ovarian cancer (12,27,32). Increased vaginal discharge has been described (26). Tamoxifen increases hot flushes (26,37-38) but does not appear to affect mood, anxiety, or overall quality of life (12,38).
4. OTHER
There is an increase in cataract development with tamoxifen use and an increase in rate of cataract surgery (12).
III. RALOXIFENE A. Effects on the Skeleton and Bone Metabolism Raloxifene (200 mg) reduces bone turnover similarly to conjugated equine estrogen (CEE) (39), with estrogenic effects on bone remodeling (40) and calcium metabolism (41). Paired bone biopsy data in 65 normal postmenopausal women indicate that raloxifene doses of 60 to 120 mg/day reduce bone formation rate and remodeling activation frequency, consistent with an antiresorptive effect (42). Further analyses of these biopsies by quantitative microradiography revealed that the mean degree of mineralization of bone was increased compared with baseline and that the mineralization profile after raloxifene treatment of postmenopausal women was similar to that seen in premenopausal women (43). Biopsy data are also available from a small, randomized trial comparing raloxifene with hormone therapy and placebo. Remodeling frequency increased in cancellous and endocortical bone in the placebo group, whereas it decreased in the raloxifene and hormone therapy groups. These data confirmed that raloxifene appears to preserve bone mass by reducing bone turnover in postmenopausal women, similar to the mechanisms seen with hormone therapy (44). In osteoporosis prevention studies in healthy postmenopausal women with bone density above the osteoporosis range, raloxifene produces increments in bone mass at all skeletal sites, with increments of 1.4% to 2.8% versus placebo in the spine, hip, and total body. Increments in bone density are maintained in postmenopausal women over 5 years (45). Raloxifene reduces indices of bone formation by 7% to 21% and bone resorption by 12% to 26% versus placebo (40,45-49). In a trial of 619 healthy early postmenopausal women randomized to raloxifene, hormone therapy, or placebo, BMD effects in the spine and hip were
839 greatest with estrogen and stable with raloxifene, and losses were seen in the placebo group (50). The major evidence base confirming the antifracture efficacy of raloxifene derives from the Multiple Outcomes of Raloxifene Evaluation (MORE) study, originally designed as a 3-year trial in 7705 postmenopausal women (average age 66 years) with osteoporosis, defined as the presence of a radiographically confirmed vertebral fracture, or a BMD -< -2.5. Three years of treatment with raloxifene (60 and 120 mg/day) in M O R E reduced risk of vertebral fracture by approximately 40% (48), with similar relative risk reductions in those women with prevalent fractures as in those with osteoporosis by BMD criteria. As expected, women in the placebo group who had prevalent vertebral fractures at baseline had a higher incident fracture rate (21.2% cumulative incidence over 3 years) compared with women in the placebo group with osteoporosis who did not have prevalent fractures at entry (4.5% cumulative incidence over 3 years). The trial indicated that 46 women with osteoporosis, without previous vertebral fracture, would need to be treated for 3 years to prevent one vertebral fracture. The corresponding number for women with a previous vertebral fracture would be only 16 (because the baseline incidence is higher). In contrast to its effect on vertebral fracture, raloxifene did not reduce the risk of wrist, hip, or all nonvertebral fractures during the M O R E study. However, in a post hoc subgroup analysis of the small group of women who had severe vertebral fractures at baseline, raloxifene appeared to reduce the risk of common nonvertebral osteoporosisrelated fractures (hip, pelvis, lower leg, wrist, humerus, and clavicle) at 3 years (51). Furthermore, analysis of x-rays done for acute back symptoms in women enrolled in M O R E demonstrated that raloxifene reduced the risk of clinically apparent vertebral fractures by 68% at 1 year (52) and was effective at reducing new clinical vertebral fractures as early as 3 and 6 months after introducing the medication (53). The reductions in vertebral fracture seen with raloxifene use appear largely independent of magnitude of change in BMD at either the lumbar spine or femoral neck (54-55). In a subset of M O R E participants who had sampling for markers of bone turnover, raloxifene treatment resulted in significant reductions in bone markers of both formation and resorption (56). Patients with the greatest changes in bone formation markers at 6 and 12 months had the greatest reductions in vertebral fracture risk at 3 years (57), with P I N P being the most predictive marker (56). In contrast, there was no significant relationship between raloxifeneinduced change in urinary c-telepeptide and vertebral fracture risk (56-57). Furthermore, the effectiveness of raloxifene on vertebral fracture occurrence is largely independent of the presence of clinical risk factors (58). Women without prevalent vertebral fractures in M O R E (n = 3204) included those with osteopenia (enrolled before application of the
840 National Health and Nutrition Examination Surveys [NHANES] data to the normative range) as well as those with osteoporosis. There was no difference in relative risk reductions for vertebral fracture in women with osteopenia versus osteoporosis (59). The MORE study was extended for 1 additional year; however, during that fourth year, participants were allowed to take other bone-active medications, and the reported use of other bone-active agents was higher in the placebo group than in the raloxifene groups. Four-year data from this 1-year extension of the MORE study confirmed continued protection against vertebral fracture of similar magnitude to that seen after 3 years (60) (Fig. 59.1). Relative risk reduction in the fourth year alone was similar to that seen during the first 3 years. Also similar to the 3-year data, over 4 years, there was no evidence that raloxifene treatment lowered risk for nonvertebral fractures. The Continuing Outcomes Relevant to Evista (CORE) study (61) was an extension of the M O R E study, in which
FIGURE 59.1 The cumulative proportion of women with at least one incident vertebral fracture at 4 years. Women with or without prevalent vertebral fractures who had a baseline and any follow-up radiograph were treated with placebo or raloxifene 60 mg/day or 120 mg/day. A: Total study population (n = 6828). B: Women (n = 5077) who reported no use of other bone-active agents in year 4. The numbers above each bar indicate the number of women in each group who had a new vertebral fracture. The numbers of women in each group who reported use of other bone-active agents in year 4 who experienced a new vertebral fracture can be calculated by subtracting the numbers above the corresponding bars in panels A and B. (From ref. 60, with permission.)
COSMAN AND LINDSAY
women originally treated with placebo in M O R E continued placebo (n = 1286) and those in either of the original 60- or 120-mg raloxifene groups continued raloxifene at the standard treatment dose of 60 mg daily (n = 2725). The CORE study was a true randomized blinded clinical trial; however, as the main outcome of this study was the long-term effect of raloxifene on breast cancer, subjects were allowed to take other bone-active agents, as was true in the 1-year extension of MORE. Approximately 39% of the placebo group and 34% of the raloxifene groups used bisphosphonates during CORE. Furthermore, most participants experienced a gap of study medication (median gap 10.6 months) between the last dose of study medication in M O R E and the first dose of study medication in CORE. Moreover, of the 4011 participants enrolled in CORE, 20% did not take the study drug due to safety concerns that arose in M O R E or concomitant medication exclusions. In CORE (61), nonvertebral osteoporosis-related fractures were considered to be all fractures excluding those associated with motor vehicle accidents, beatings, trauma by moving objects, pathologic fractures, and fractures of the fingers, toes, and skull. The incidence of fractures over the full 8 years of M O R E and CORE was 22.9% in the placebo group and 22.8% in the raloxifene group. A subgroup of six common osteoporosis-related fractures was also evaluated (hip, pelvis, lower leg, wrist, humerus, and clavicle) and occurred in 17.5% of both the placebo and raloxifene groups over the 8-year M O R E / C O R E study. Nonvertebral fracture occurrence was also not different between placebo and raloxifene groups when analysis was restricted to those participants who did not take other osteoporosis medications. However, consistent with the post hoc exploratory analyses of the M O R E population, considering the entire 8-year M O R E / C O R E experience, those participants who had severe vertebral fractures at randomization had a significant reduction in risk of the six common nonvertebral fractures specified here. A subset of U.S. CORE participants also entered a bone density substudy in which BMD was assessed at the 7-year point from the beginning of MORE. Results indicated an increase of BMD in the spine of 1.7% versus placebo (not significant [NS]) and 2.4% in the femoral neck versus placebo (p < 0.05). When treatment with raloxifene is discontinued, even after 5 years of therapy, bone loss ensues (62). Significant bone loss was also seen during the off year at the end of MORE before the beginning of the CORE study (61). Raloxifene appears to decrease BMD loss in premenopausal women treated with gonadotropin-releasing hormone (GnRH) agonist administration for uterine leiomyomas (63) and decrease serum calcium levels in women with primary hyperparathyroidism (64). A recent meta-analysis of fracture data available for raloxifene indicates a consistent reduction in the occurrence of vertebral fracture in postmenopausal women treated with
CHAPTER 59 Selective Estrogen Receptor Modulators this agent (65). Additional information on the skeletal effect of raloxifene should come from the Evista Alendronate Comparison (EVA) trial. The EVA trial was designed to determine the effect of raloxifene compared with alendronate on the incidence of osteoporosis-related fracture in postmenopausal women with osteoporosis. The study was terminated early, and results should be available in late 2006 (66).
B. Effects on Other Organ Systems 1. BREAST The occurrence of invasive breast cancer in M O R E was 76% lower in women taking raloxifene for over 3 years, due to a reduction in ER-positive disease (67). A similar reduction was seen in women taking raloxifene for 4 years (68) (Fig. 59.2). Approximately 93 women would need to be treated with raloxifene for 4 years to prevent one case of invasive breast cancer. Women in the raloxifene trial were recruited because of osteoporosis, not elevated breast cancer risk. Nevertheless, the average age in M O R E was older than that in the tamoxifen BCPT, and therefore these women were probably at increased risk of breast cancer by age criteria alone. The potency of effects of raloxifene versus tamoxifen on breast cancer prevention, however, cannot be directly determined from the available data. A study comparing the two drugs regarding effects on breast cancer occurrence and other outcomes, the Study of Tamoxifen and Raloxifene (STAR), has recruited approximately 19,000 women 35 years of age and older who are at elevated risk for breast cancer and is in the latter years of the 5-year trial.
841 Results of the C O R E study were completely consistent with those of M O R E . In CORE, the risk of invasive breast cancer was reduced by 59%, and ER-positive invasive breast cancer was reduced by 66% in raloxifene versus placebo groups (69). There was no difference in the incidence of ER-negative breast cancer. Considering the 8-year experience of M O R E followed by CORE, the risks of invasive breast cancer and ER-positive invasive breast cancer were reduced by 66% and 76%, respectively (69).
2. EFFECTS ON VASCULAR DISEASE AND INTERMEDIATES
In some studies, raloxifene reduces serum total cholesterol, L D L cholesterol, lipoprotein a, and plasma fibrinogen while increasing the H D L subclass H D L - 2 (39,46-47,70-71), although total serum H D L is not increased (50). Raloxifene does not increase serum triglyceride levels to the same magnitude as does hormone therapy (50). Raloxifene also reduces homocysteine levels (72-73) and, in contrast to HRT, does not increase C-reactive protein levels (72). In the M O R E study, after 4 years, cardiovascular and cerebrovascular disease outcomes were decreased by 40% in patients with established disease or those at high risk (74). Furthermore, there was no early increase in risk as seen with hormone therapy (74-75). Venous thromboembolic disease was increased to a threefold degree, as was seen with tamoxifen and estrogen (48,76). Of approximately 155 women treated with raloxifene for 3 years, one case of venous thromboembolism would be diagnosed. An increased risk of venous thrombosis (about twofold) persisted during the C O R E study (69). More information on the vascular and breast outcomes of raloxifene are now available from the results of the Raloxifene Use for the Heart (RUTH) study (77). This trial enrolled 10,101 women at elevated risk for coronary heart disease and randomized them to receive raloxifene versus placebo. Invasive breast cancers were decreased, but coronary artery disease was not affected, and there was an increase in fatal stroke and thromboembolism (77a).
3. GYNECOLOGIC EFFECTS AND MENOPAUSAL SYMPTOMS
FIGURE 59.2 Breast cancer effects of raloxifene from the MORE trial (From ref. 68, with permission.)
In contrast to tamoxifen, raloxifene is not associated with an increased risk of uterine cancer or any benign disease. Serial transvaginal ultrasounds reveal no increase in endometrial proliferation or other benign uterine disease, including polyps (67,78). Long-term effects on the uterus after 5 years of raloxifene treatment indicate no increment in vaginal bleeding, mean endometrial thickness, endometrial hyperplasia, or carcinoma (45). Raloxifene does not increase the risk of ovarian cancer (79-80). Raloxifene does not increase the clinical development of pelvic organ prolapse and might actually reduce the need for
842
COSMAN AND LINDSAY
pelvic floor surgery (81); however, detailed comprehensive analyses of early quantitative effects on pelvic floor function have not been performed (82). Urinary incontinence was reported in more women receiving hormone therapy than raloxifene therapy in a comparative trial of 619 women (50,83). A prospective evaluation of a subgroup of participants from the M O R E study (n = 963 women) indicated that raloxifene had no effect on urinary incontinence during 3 years of therapy (84). In another subset of women from the M O R E trial, raloxifene did not affect any aspects of sexual function (85). Furthermore, raloxifene did not affect the improvement in vaginal atrophy symptoms associated with vaginal estrogen cream (86-87) or the vaginal estrogen ring (88). However, administration of raloxifene after discontinuation of transdermal hormone therapy is associated with worsening of vaginal atrophy (vaginal maturation value), and there are some increased reports of dyspareunia and urinary leaks (89). Raloxifene has been shown to increase hot flushes similarly to tamoxifen (67). However, not all studies indicate a significant impact of hot flushes and treatment satisfaction in women given raloxifene (90). Although hot flushes increase in severity and frequency upon discontinuation of systemic hormone therapy, raloxifene does not appear to worsen these symptoms compared with placebo (91). A systematic study of cognitive function in 143 women with osteoporosis revealed no differences in performance between raloxifene and placebo groups on tests of memory and mood (92). Cognitive function and risk of Alzheimer's disease were not significantly affected in postmenopausal women from the M O R E study who received the 60 mg/day dose (93-94). At the higher dose of raloxifene (120 mg/day in the M O R E study), a reduction in risk of cognitive impairment was suggested (94). Raloxifene does not produce any negative effects on mood and might actually reduce symptoms of depression and anxiety over 12 months of use (95).
4.
OTHER
In the M O R E study, raloxifene did not increase the risk for cataracts or gallbladder disease (78). Most (96-97), although not all (98), reports indicate neutral effects of raloxifene on insulin sensitivity. Minor side effects can include leg cramps, leg swelling, and an influenza-like syndrome. Some (99-100), but not all (101), studies indicate that raloxifene may exert a beneficial effect in premenopausal women with leiomyomas on uterine and leiomyoma size.
IV. SERMS IN DEVELOPMENT Arzoxifene, a SERM closely related chemically to raloxifene but with much higher affinity for the ER (102), has been shown to have strong breast anti-estrogen activity and
absence of uterine agonist activity (102-103). Pro-estrogenic activity is seen in bone and serum lipids (102-105). Arzoxifene is being studied as a breast cancer prevention agent as well as an agent for prevention and treatment of osteoporosis. Bazedoxifene has no uterine agonist activity but has a binding affinity to bone similar to that of raloxifene. As is true of all SERMS, it has estrogen antagonist activity at the breast (106). Changes in endometrial thickness with bazedoxifene were no different than those of placebo (107). It is being studied to determine whether increased uterine safety can be obtained in the face of beneficial effects on bone and breast. Bazedoxifene is also being studied in combination with estrogen to reduce menopausal symptoms without requiring the prescription of a progestin. Lasofoxifene is being developed for the prevention and treatment of osteoporosis as well as for the treatment of vaginal atrophy (108). However, in September 2005, the Food and Drug Administration issued a nonapprovable letter for the osteoporosis indication for unclear reasons. Further information will likely be available on lasofoxifene and bazedoxifene in the future. Little is known about the urogenital effects of these three SERMS currently in development (109).
V. SUMMARY Tamoxifen remains the gold standard for breast cancer prevention as of this writing; however, it is possible that raloxifene, which may be approved later in 2006, will be as effective and safer. The aromatase inhibitors may be even more effective as breast cancer prevention agents; however, their negative effects on bone status make them less desirable for most women except for those at particularly high risk. No SERMS have yet been developed that can treat menopausal symptoms such as hot flushes, although local estrogen application remains an effective treatment for vaginal atrophy, in women taking SERMs. Raloxifene is the only agent at present with consistent efficacy against vertebral fracture occurrence in patients with osteoporosis. All SERMS seem to have the potential risk of venous thrombosis, but effects on arterial disease need to be clarified. The new SERMs may have an even better benefit-to-risk ratio.
References 1. CosmanF, LindsayR. Selectiveestrogen receptor modulators:clinical spectrum. EndocrRev 1999;20:418-434. 2. KristensenB, Ejlertsen B, Dlagaard P, et al. Tamoxifen and bone metabolismin postmenopausallow-riskbreast cancerpatients.A randomized study.J Clin Onco11994;12:992-997. 3. LoveRR, Mazess RB, Barden HS, et al. Effects of tamoxifenon bone mineral densityin postmenopausalwomen with breast cancer. N Engl JMed 1992;326:852-856.
CHAPTER 59 Selective Estrogen Receptor Modulators 4. Grey AB, Stapleton JP, Evans MC, et al. The effect of the antiestrogen tamoxifen on bone mineral density in normal late postmenopausal women. A m J M e d 1995;99:636-641. 5. Wright CDP, Garrahan NJ, Stanton M, et al. Effect of long-term tamoxifen therapy on cancellous bone remodeling and structure in women with breast cancer. J Bone Miner Res 1994;9:153-159. 6. Kenny AM, Prestwood KM, Pilbeam CC, Raisz LG. The short term effects of tamoxifen on bone turnover in older women. J Clin Endocrinol Metab 1995;80:3287-3291. 7. Love RR, Barden HS, Mazess RB, Epsteine S, Chappell RJ. Effect of tamoxifen on lumbar spine bone mineral density in postmenopausal women after 5 years. Arch Intern Med 1994;154:2585-2588. 8. Wright CDP, Mansell RE, Gazet JC, Compston JE. Effect of long term tamoxifen treatment on bone turnover in women with breast cancer. BMJ 1993;306:429-430. 9. Powles TJ, Hickish T, Kanis JA, et al. Effect of tamoxifen on bone mineral density measured by dual energy x-ray absorptiometry in healthy premenopausal and postmenopausal women.J Clin Onco11996; 14:78-84. 10. Gotfredsen A, Christiansen C, PalshofT. The effect of tamoxifen on bone mineral content in premenopausal women with breast cancer. Cancer 1984;53:853-857. 11. Kristensen B, Ejlertsen B, Mouridsen HT, et al. Femoral fractures in postmenopausal breast cancer patients treated with adjuvant tamoxifen. Breast Cancer Res Treat 1996;39:321- 326. 12. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer. Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. JNatl Cancer Inst 1998;90:1371-1388. 13. Early Breast Cancer Trialists' Collaborative Group. Systemic treatment of early breast cancers by hormonal, cyctotoxic, or immune therapy. 133 randomised trials involving 31,000 recurrences and 24,000 deaths among 75,000 women. Lancet 1992;339:1-15, 71-85. 14. Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer. An overview of the randomised trials. Lancet 1998;351: 1451-1467. 15. Osborne CK. Drug therapy: tamoxifen in the treatment of breast cancer. N EnglJ Med 1998;339:1609-1618. 16. Hortobagyi GN. Treatment of breast cancer. N E n g l J M e d 1998;339: 974-984. 17. Veronesi U, Maisonneuve P, Costa A, et al. Prevention of breast cancer with tamoxifen: preliminary findings from the Italian randomised trial among hysterectomised women. Italian Tamoxifen Prevention Study. Lancet 1998;352:93-97. 18. Veronesi U, Maisonneuve P, Sacchini V, et al. Tamoxfien for breast cancer among hysterectomised women. Lancet 2002;359:1122-1124. 19. Veronesi U, Maisonneuve P, Rotmensz N, et al. Italian randomized trial among women with hysterectomy: tamoxifen and hormonedependent breast cancer in high-risk women.JNat! CancerInst 2003;95: 160-165. 20. Powles T, Eeles R, Ashley S, et al. Interim analysis of the incidence of breast cancer in the Royal Marsden Hospital tamoxifen randomised chemoprevention trial. Lancet 1998;352:98-101. 21. Cuzick J, Powles T, Veronesi U, et al. Overview of the main outcomes in breast-cancer prevention trials. Lancet 2003;361:296-300. 22. CuzickJ, Forbes J, Edwards R, et al. First results from the International Breast Cancer Intervention Study (IBIS-I): a randomized prevention trial. Lancet 2002;360:817- 824. 23. McDonald CC, Stewart HJ. Fatal myocardial infarction in the Scottish adjuvant tamoxifen trial. The Scottish Breast Cancer Committee. BMJ 1991;303:435-437. 24. McDonald CC, Alexander FE, Whyte BW, et al. Cardiac and vascular morbidity in women receiving adjuvant tamoxifen for breast cancer in a randomised trial. The Scottish Cancer Trials Breast Group. BMJ 1995;311:977-980.
843 25. Rutqvist LE, Mattsson A. Cardiac and thromboembolic morbidity among postmenopausal women with early stage breast cancer in randomized trial of adjuvant tamoxifen. The Stockholm Breast Cancer Study Group. J Natl Cancer Inst 1993;85:1398-1406. 26. Costantino JP, Kuller LH, Ives DG, et al. Coronary heart disease mortality and adjuvant tamoxifen therapy. JNatl CancerInst 1997;89:776-782. 27. Fisher B, Costantino JP, Redmond CK, et al. Endometrial cancer in tamoxifen-treated breast cancer patients. Findings from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14. J Natl Cancer Inst 1994;86:527-537. 28. Fisher B, Costantino JP, Redmond C, et al. A randomized clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have estrogen-receptor-positive tumors. N EnglJ Med 1989;320:479-484. 29. Fisher B, Redmond C. Systemic therapy in node-negative patients. Updated findings from NSABP clinical trials. National Surgical Adjuvant Breast and Bowel Project. J Natl Cancer Inst Monogr 1992;11: 105-116. 30. Saphner T, Tormey DC, Gray R. Venous and arterial thrombosis in patients who received adjuvant therapy for breast cancer. J Clin Oncol 1991;9:286-294. 31. "Nolvadex" Adjuvant Trial Organisation. Controlled trial of tamoxifen as a single adjuvant agent in the management of early breast cancer. Br J Cancer 1998;57:608-611. 32. Cook LS, Weiss NS, Schwartz SM, et al. Population-based study of tamoxifen therapy and subsequent ovarian, endometrial, and breast cancers. J Natl Cancer Inst 1995;87:1359-1364. 33. Rutqvist LE, Johansson H, Signomklao T, et al. Adjuvant tamoxifen therapy for early stage breast cancer and second primary malignancies. Stockholm Breast Cancer Study Group. J Natl Cancer Inst 1995;87: 645-651. 34. Neven P, DeMulder X, van Belle Y, et al. Tamoxifen and the uterus. BMJ 1994;309:1313-1314. 35. Jaiyesimi IA, Buzdar AU, Decker DA, Hortobagyi GN. Use of tamoxifen for breast cancer: twenty-eight years later. J Clin Oncol 1995;113:513-529. 36. Kedar RP, Bourne TH, Powles TJ, et al. Effects of tamoxifen on uterus and ovaries of postmenopausal women in a randomized breast cancer prevention trial. Lancet 1994;343:1318-1321. 37. Fisher B, Dignam J, Bryant J, et al. Five versus more than five years for tamoxifen therapy for breast cancer patients with negative lymph nodes and estrogen receptor-positive tumors. J Nat Cancer Inst 1996;88: 1529-1542. 38. Love RR, Cameron L, Connell B L, Leventhal H. Symptoms associated with tamoxifen treatment in postmenopausal women. Arch Intern Med 1991;151:1842-1947. 39. Draper MW, Flowers DE, Huster WJ, et al. A controlled trial of raloxifene (LY139481) HCI: impact on bone turnover and serum lipid profile in healthy postmenopausal women. J Bone Miner Res 1996;11: 835-842. 40. Product Circular 1997. Evista (Raloxifene Hydrochloride). Indianapolis: Eli Lilly and Company. 41. Heaney RP, Draper MW. Raloxifene and estrogen: comparative bone remodeling kinetics. J Clin Endocrinol Metab 1997;82:3425 - 3429. 42. Ott SM, Oleksik A, Lu Y, et al. Bone histomorphometric results of a 2-year randomized, placebo controlled trial of raloxifene in postmenopausal women. Bone 1998;23(S 1):TS295. 43. Boivin G, Lips P, Ott SM, et al. Contribution of raloxifene and calcium and vitamin D3 supplementation to the increase of the degree of mineralization of bone in postmenopausal women.J Clin EndocrinolMetab 2003 ;88:4199- 4205. 44. Weinstein RS, Parfitt AM, Marcus R, et al. Effects of raloxifene, hormone replacement therapy, and placebo on bone turnover in postmenopausal women. Osteoporosis Int 2003;14:814-822.
844 45. Jolly EE, Bjarnason NH, Neven P, et al. Prevention of osteoporosis and uterine effects in postmenopausal women taking raloxifene for 5 years. Menopause 2003;10:337- 344. 46. Delmas PD, Bjarnason NH, Mitlak BH, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. N Eng! J IVied 1997;337: 1641-1647. 47. Johnston CC, Bjarnason NH, Cohen FJ, et al. Long-term effects of raloxifene on bone mineral density, bone turnover, and serum lipid levels in early postmenopausal women: three-year data from 2 doubleblind, randomized, placebo-controlled trials. Arch Intern Med 2000; 160:3444-3450. 48. Ettinger B, Black DM, Mitlak BH, et al. Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene. Results from a 3-year randomized clinical trial. J A m Med Assoc 1999;282:637-645. 49. Lufldn EG, Whitaker MD, Nickelsen T, et al. Treatment of established postmenopausal osteoporosis with raloxifene. A randomized trial. JBone Miner Res 1998;13:11. 50. Reid IR, Eastrell R, Fogelman I, et al. A comparison of the effects of raloxifene and conjugated equine estrogen on bone and lipids in healthy postmenopausal women. Arch Intern Med 2004;164:871- 879. 51. Delmas PD, Genant HK, Grans GG, et al. Severity of prevalent vertebral fractures and the risk of subsequent vertebral and nonvertebral fractures: results from the MORE trial. Bone 2003;33:522-532. 52. Maricic M, Adachi JD, Sarkar S, et al. Early effects of raloxifene on clinical vertebral fractures at 12 months in postmenopausal women with osteoporosis. Arch Intern IVied2002;162:1140-1143. 53. Qu Y, Wong M, Thiebaud D, Stock JL. The effect of raloxifene therapy on the risk of new clinical vertebral fractures at three and six months: a secondary analysis of the MORE trial. Curr Med Res Opin 2005;21:1955-1959. 54. Sarkar S, Mitlak BH, Wong M, et al. Relationships between bone mineral density and vertebral fracture risk with raloxifene therapy. J Bone Min Res 2002;17:1-10. 55. Sarkar S, Reginster JY, Crans GG, et al. Relationship between changes in biochemical markers of bone turnover and BMD to predict vertebral fracture risk. J Bone Miner Res 2004;19:394-401. 56. Reginster JY, Sarkar S, Zegels B, et al. Reduction in PINP, a marker of bone metabolism, with raloxifene treatment and its relationship with vertebral fracture risk. Bone 2004;34:344- 351. 57. Bjarnason NH, Sarkar D, Duong T, et al. Six and twelve month changes in bone turnover are related to reduction in vertebral fracture risk during 3 years of raloxifene treatment in post-menopausal osteoporosis. Osteo Int 2001;12:922-930. 58. Johnell O, Kanis JA, Black DM, et al. Associations between baseline risk factors and vertebral fracture risk in the Multiple Outcomes of Raloxifene Evaluation (MORE) study. J Bone Miner Res 2004;19: 764-772. 59. Kanis JA, Johnell O, Black DM. Effect of raloxifene on the risk of new vertebral fractures in postmenopausal women with osteopenia and osteoporosis: a re-analysis of the multiple outcomes of raloxifene evaluation trial. Bone 2003;33:293-300. 60. Delmas PD, Ensrud KE, Adachi JD, et al. Efficacy of raloxifene on vertebral fracture risk reduction if postmenopausal women with osteoporosis: four-year results from a randomized clinical trial. J Clin Endocrinol Metabo12002;87:3609- 3617. 61. Siris ES, Harris ST, Eastrell R, et al. Skeletal effects of raloxifene after 8 years: results from the Continuing Outcomes Relevant to Evista (CORE) study.J Bone Miner Res 2005;20:1514-1524. 62. Neele SJM, Evertz R, DeValk-DeRoo G, Roos JC, Netelenbos JC. Effect of 1 year of discontinuation of Raloxifene or estrogen therapy on bone mineral density after 5 years of treatment in health postmenopausal women. Bone 2002;30:599-603.
COSMAN AND LINDSAY
63. Palomba S, Orio F Jr, Morelli M, et al. Raloxifene administration in women treated with GnRH agonist for uterine leiomyomas: effects on bone metabolism. J Clin Endocrinol Metab 2002;87:4476-4481. 64. Rubin MR, Lee KH, McMahon DJ, Silverberg SJ. Raloxifene lowers serum calcium and markers of bone turnover in postmenopausal women with primary hyperparathyroidism. J Clin Endocrinol Metab 2003;88:1174-1178. 65. Seeman E, Crans GG, Diez-Perez A, Pinette KV, Delmas PD. Antivertebral fracture efficacy of raloxifene: a meta-analysis. OsteoporosisInt 2006;17:313-316. 66. Lufldn EG, Sarkar S, Kulkarni PM, et al. Antiresorptive treatment of postmenopausal osteoporosis: review of randomized clinical studies and rationale for the Evista Alendronate Comparison (EVA) trial. Curr ivied Res Opin 2004;20:351 - 357. 67. Cummings SR, Eckert S, Krueger KA, et al. The effects of raloxifene on risk of breast cancer in postmenopausal women. Results from the MORE randomized trial.JAm MedAssoc 1999;281:2189-2197. 68. Cauley JA, Norton L, Lippman ME, et al. Continued breast cancer risk reduction in postmenopausal women treated with raloxifene: 4-year results from the MORE trial. Breast CancerRes Treat 2001;65:125-134. 69. Martino S, Cauley JA, Barret-Connor E, et al. Continuing outcomes relevant to Evista: breast cancer incidence in postmenopausal osteoporotic women in a randomized trial of raloxifene. J Natl Cancer Inst 2004;96:1751 - 1761. 70. Walsh BW, Kuller LH, Wild RA, et al. Effects of raloxifene on serum lipids and coagulation factors in healthy postmenopausal women. JAm MedAssoc 1998;279:1445-1455. 71. Cox DA, Sarkar S, Harper K, Barrett-Connor E. Effect of raloxifene on the incidence of elevated low density lipoprotein (LDL) and achievement of LDL target goals in postmenopausal women. Curr IVied Res Opin 2004;20:1049-1054. 72. Walsh BW, Paul S, Wild RA, et al. The effects of hormone replacement therapy and Raloxifene on C-reactive protein and homocysteine in healthy postmenopausal women: a randomized, controlled trial. J Clin Endocrinol Metab 2000;85:214-218. 73. DeLeo V, la Marca A, Morgante G, et al. Randomized control study of the effects of raloxifene on serum lipids and homocysteine in older women. AmJ Obstet Gyneco12001;184:350- 353. 74. Barrett-Connor E, Grady P, Sashegyi A, et al. Raloxifene and cardiovascular events in osteoporotic postmenopausal women. Four-year results from the MORE (Multiple Outcomes of Raloxifene Evaluation) randomized trial. JAm MedAssoc 2002;287:847- 857. 75. Keech CA, Sashegyi A, Barret-Connor E. Year-by-year analysis of cardiovascular events in the Multiple Outcomes of Raloxifene Evaluation (MORE) Trial. Curr Med Res Opin 2005;21:135-140. 76. Cosman F, Baz-Hecht M, Cushman M, et al. Short-term effects of estrogen, tamoxifen and raloxifene on hemostasis: a randomized-controlled study and review of the literature. Thromb Res 2005;116:1-13. 77. Wenger NK, Barret-Connor E, Collins P, et al. Baseline characteristics of participants in the RUTH Trial. Am J Cardio! 2002;90:1204-1210. 77a. Barrett Connor E, Mosca L, Collins R Effects of raloxifene on cardiovascular events and breast cancer in postmenopausal women. NEngl J Med 2006;355:125-137. 78. Grady D, Ettinger B, Moscarelli E, et al. Safety and adverse effects associated with raloxifene: Multiple Outcomes of Raloxifene Evaluation. Obstet Gyneco12004;104:837- 844. 79. Neven P, Goldstein SR, Ciaccia AV, et al. The effect of raloxifene on the incidence of ovarian cancer in postmenopausal women. Gynecik Onco12002;85:388- 390. 80. Martino S, Disch D, Dowsett SA, Keech CA, Mershon JL. Safety assessment of raloxifene over eight years in a clinical trial setting. Current Med Res Opin 2005;21:1441-1452. 81. Goldstein SR, Neven P, Zhou L, et al. Raloxifene effect on frequency of surgery for peMc floor relaxation. Obstet Gyneco12001:98:91-96.
CHAPTER 59 Selective Estrogen Receptor Modulators 82. Vardy MD, Lindsay R, Scotti RJ, et al. Short-term urogenital effects of raloxifene, tamoxifen, and estrogen. Am J Obstet Gynecol 2003;189: 81-88. 83. Goldstein SR, Johnson S, Watts NB, et al. Incidence of urinary incontinence in postmenopausal women treated with raloxifene or estrogen. Menopause 2005;12:160-164. 84. Waetjen LE, Brown JS, Modelska K, et al. Effect of raloxifene on urinary incontinence: a randomized controlled trial. Obstet Gynecol 2004;103:261 - 266. 85. Modugno F, Ness RB, Ewing S, Cauley JA. Effect of Raloxifene on sexual function in older postmenopausal women with osteoporosis. Obstet Gyneco12003;101:352- 361. 86. Kessel B, Nachtigall L, Plouffe L, et al. Effect of raloxifene on sexual function in postmenopausal women. Climacteric 2003;6:248-256. 87. Parsons A, Merritt D, Rosen A, et al. Effect of Raloxifene on the response to conjugated estrogen vaginal cream or nonhormonal moisturizers in postmenopausal vaginal atrophy. Obstet Gyneco12003;101:346-352. 88. Pinkerton JV, Shifren JL, La Valleur J, et al. Influence of Raloxifene on the efficacy of an estradiol-releasing ring for treating vaginal atrophy in PMW. Menopause 2003;10:45-52. 89. Checa MA, Garrido A, Prat M, et al. A comparison of raloxifene and calcium plus vitamin D on vaginal atrophy after discontinuation of longstanding postmenopausal hormone therapy in osteoporotic women. A randomized, masked-evaluator, one-year, prospective study. Maturitas 2005;52:70-77. 90. Palacios S, Farias ML, Luebbert H, et al. Raloxifene is not associated with biologically relevant changes in hot flushes in postmenopausal women for whom therapy is appropriate.dmJ Obstet Gyneco12004;191: 121-131. 91. Gordon S, Walsh BW, Ciaccia AV, et al. Transition from estrogenprogestin to raloxifene in postmenopausal women: effect on vasomotor symptoms. Obstet Gyneco12004;103:267-273. 92. Nickelsen T, Lifkin EG, Riggs BL, et al. Raloxifene hydrochloride, a selective estrogen receptor modulator: safety assessment of effects on cognitive function and mood in postmenopausal women. Psychoneuroendocrinology 1999;24:115-128. 93. Yaffe K, Krueger K, Sarkar S, et al. Cognitive function in postmenopausal women treated with roxifene. N Engl J Med 2001;344: 1207-1213. 94. Yaffe K, Krueger K, Cummings SR, et al. Effect of raloxifene on prevention of dementia and cognitive impairment in older women: the Multiple Outcomes of Raloxifene Evaluation (MORE) randomized trial. Am J Psychiatry 2005;162:683-690.
845 95. Jarkova NB, Martenyi F, Masanauskaite D, et al. Mood effect of raloxifene in postmenopausal women. Maturitas 2002;42:71- 75. 96. Cagnacci A, Arangino S, Renzi A, Zanni AL, Volpe A. Raloxifene does not modify insulin sensitivity and glucose metabolism in postmenopausal women. J Clin Endocrinol Metab 2002;87:4117- 4121. 97. Cucinelli F, Soranna L, Romualdi D, et al.The effect of raloxifene on glyco-insulinemic homeostasis in healthy postmenopausal women. J Clin Endocrinol Metab 2002;87:4186-4193. 98. Lee CC, Kasa-Vubu JZ, Supiano MA. Differential effects of raloxifene and estrogen on insulin sensitivity in postmenopausal women. JAm Geriatr Soc 2003;51:683-688. 99. Jirecek S, Lee A, Pavo I, et al. Raloxifene prevents the growth of uterine leiomyomas in premenopausal women. Fertil Steri12004;81:132-136. 100. Palomba S, Orio F Jr, Russo T, et al. Antiproliferative and proapoptotic effects of raloxifene on uterine leiomyomas in postmenopausal women. Fertil Steri12005;84:154-161. 101. Palomba S, Orio F Jr, MoreUi M, et al. Raloxifene administration in premenopausal women with uterine leiomyomas: a pilot study. J Clin Endocrinol Metab 2002;87:3603- 3608. 102. Sporn MB. The biology behind arzoxifene: a promising new selective estrogen receptor modulator for clinical chemoprevention of breast cancer. Clin Cancer Res 2004;10:5313-5315. 103. Fabian CJ, Kimler BF, Anderson J, et al. Breast cancer chemoprevention phase I evaluation ofbiomarker modulation by arzoxifene, a third generation selective estrogen receptor modulator. Clin Cancer Res 2004;10:5403 - 5417. 104. Munster PN. Arzoxifene: the development and clinical outcome of an ideal SERM. Expert Opin Investig Drugs 2006;15:317-326. 105. Ma YL, Bryant HU, Zeng Q~ et al. Long-term dosing of arzoxifene lowers cholesterol, reduces bone turnover, and preserves bone quality in ovariectomized rats. J Bone Miner Res 2002;17:2256-2264. 106. Komm BS, Kharode YP, Bodine PV, Harris HA. Bazedoxifene acetate: a selective estrogen receptor modulatory with improved selectivity. Endocrinology 2005;146:3999-4008. 107. Gruber C, Gruber D. Bazedoxifene. Curr ODinInvestig Drugs 2004;5: 1086-1093. 108. Gennari L. Lasofoxifene. Curr ODinInvestig Drugs 2005;6:1067-1078. 109. Albertazzi P, Sharma S. Urogenital effects of selective estrogen receptor modulators: a systematic review. Clirnacterk 2005;8:214-220.
This Page Intentionally Left Blank
; H A P T E R 6(
Tibolone: Selective Tissue Estrogen Activity Regulator Utilization in Postmenopausal Women DAVID F. A R C H E R
j. c.
Departmentof Obstetrics and Gynecology,CONRAD Clinical Research Center, Eastern Virginia Medical School, Norfolk, VA 23507
GALLAGHER Departmentof Medicine, Bone Metabolism Unit, Creighton UniversitySchool of Medicine, Omaha, NE 68178
hydroxytibolone, bind to the estrogen receptor (ER), whereas the A4 metabolite binds to the progesterone receptor (3). Tibolone itself binds with a low affinity to the androgen and progesterone receptors. These metabolites not only elicit an effect through receptor binding, but also modify specific tissue enzyme activity (9,10). The tissuespecific estrogen effect reflects the modulation of enzymatic activity in the tissues, specifically breast and uterus. This is shown schematically in Fig. 60.1, which indicates that tibolone and its metabolites in the breast actually alter the sulfokinase, sulfatase, and 17-hydroxysteroid dehydrogenase enzyme activities in breast tissue in vitro (11-13). The summation of these events is postulated to be a reduction in tissue estrogen concentrations. The mechanism of action of tibolone is described as a selective tissue estrogenic activity regulator (STEAR) (1,10). This is due to the effects of interacting with the receptors, secondarily altering tissue enzyme activity, and lastly the metabolites exerting a
Tibolone (Livial, NV Organon, The Oss, Netherlands) has been on the European market since 1990 as hormone therapy. It differs from standard estrogen plus progestin therapy, principally due to the fact that it is a single molecule that has differential effects on various organs in the body. Recent investigation, both basic and clinical, have demonstrated that tibolone is associated with clinical outcomes similar to the selective estrogen receptor modulators but differing in the mechanism of action and the scope of its therapeutic indications (1,2). Selective estrogen receptor modulators elicit conformational change in the estrogen receptor in the target tissues. Tibolone and its metabolites interact with estrogen, progestin, and androgen receptors (3). Tibolone, in contrast to SERMs, has a further modifying effect on tissue enzymes; tibolone is known to reduce menopausal symptoms and treat vaginal atrophy (4-7). Tibolone is metabolized rapidly and extensively into three metabolites (1,8). Two metabolites, the 3c~- and 3[3T R E A T M E N T OF T H E P O S T M E N O P A U S A L W O M A N
847
Copyright 9 2007 by Elsevier, Inc. All rights of reproduction in any form reserved.
848
ARCHER AND GALLAGHER
II. CLINICAL PROFILE OFTIBOLONE The tibolone metabolites have a variable effect on the target tissues depending on receptor content and metabolite concentration. The clinical profile appears similar but not identical to a SERM. Tibolone shows little effect on breast and endometrium while having a major effect on the central nervous system as measured by hot flashes (4,7,17-22). This is significantly different from SERM. Tibolone increases bone mineral density comparable to a SERM (23-28).
A. Menopausal Symptoms FIGURE 60.1 Effect of tibolone on enzymatic pathways in breast tissue in vitro. (Data are from refs. 11 and 12.)
differential effect in the respective target tissues. These attributes indicate that tibolone is not a true SERM.
I. PHARMACOKINETICS OFTIBOLONE Tibolone is a prodrug. It is metabolized in the body rapidly to two 3-hydroylated metabolites known as 3o~- and 3[3-hydroxytibolone (14). These two metabolites have estrogenic activity. Tibolone is metabolized in the endometrium to A4 tibolone, a compound that has progestational-like activity (3,15). The plasma concentration of tibolone and its metabolites are shown in Fig. 60.2 (14,16). The pharmacokinetic profile of tibolone after oral dosing is the 3cx-hydroxytibolone metabolite, which has the greatest concentration in blood, followed by the 3[3-hydroxytibolone metabolite. Tibolone and the A4 isomer are found in 10 to 100 times less concentration in the peripheral blood compared with the two hydroxylated metabolites.
The principal menopausal symptom is the hot flash, which occurs in perimenopausal and postmenopausal women. Approximately 85% of North American women experience a hot flash of varying degrees of intensity (29). The principal reason for using hormone therapy in postmenopausal women is relief of hot flashes (30). Tibolone has been shown to have a dose response effect, improving hot flashes with the highest dose of tibolone, 2.5 mg, having the greatest effect and the lowest dose, 0.625, having minimal to no effect compared with placebo (31). This action of tibolone could be due to its two estrogenic metabolites, 3o~- and 3[3-hydroxytibolone. These effects on hot flashes are comparable to those found with a standard estrogen plus progestin therapy (4,22,31,32). Vaginal and vulvar atrophy are associated with increasing time since menopause. The clinical symptoms are burning, itching, or discomfort in the vaginal area and discomfort or pain with intercourse. Vulvovaginal atrophy is a significant contributor to a reduction in coital frequency secondary to the discomfort associated with sexual intercourse. Tibolone has been shown to improve vulvovaginal symptoms, both objectively and subjectively, and reduce the dyspareunia, or pain with intercourse (20,33,34).
B. Effects on the Endometrium
FIGURE 60.2 Serum concentrations oftibolone and its metabolites after an oral dose of 2.5 mg tibolone. (Data from refs. 14 and 16.)
Approximately 70% of postmenopausal women have an intact uterus. Thirty percent have undergone a hysterectomy for a variety of medical reasons. Hormone therapy, estrogen plus progestin, is recommended in women who have an intact uterus to reduce the incidence of endometrial cancer. There is a high incidence of irregular, unpredictable endometrial bleeding associated with the use of estrogen plus progestin in postmenopausal women with an intact uterus (35-40). This has presented a problem for the clinician and the consumer. The physician is faced with a diagnostic problem regarding whether or not the endometrial bleeding is a reflection of the exogenous hormones or an indication of an underlying, more severe pathologic process. Postmenopausal
CHAPTER 60 Tibolone: Selective Tissue Estrogen Activity Regulator Utilization in Postmenopausal Women women do not wish to have any type of menstrual bleeding episodes, especially one that is irregular and unpredictable in nature (30,41-43). Tibolone has a very low incidence of endometrial bleeding in clinical trials (21,22,31,44-46). The incidence of bleeding or spotting with Tibolone is approximately 20% during the beginning of therapy, and the incidence of no bleeding increases with time. Clinical trials with tibolone have demonstrated an atrophic endometrium in the vast majority of cases (17,22). This effect is postulated to be due to the preferential formation of the A4 tibolone metabolite in the endometrium (15) (Fig. 60.3). A recent publication has associated an increased risk of endometrial cancer with tibolone use (47), But the current medical literature does not conclusively demonstrate this (48,49). A large multinational, multicenter clinical trial evaluated the effect of tibolone at two doses, 1.25 and 2.5 rag, versus conjugated equine estrogen and medroxyprogesterone acetate (CEE/MPA) at a dose of 0.625/5mg on endometrial safety. This trial, the Tibolone Histology of the Endometrium and Breast Endpoint Study (THEBES), has recently been completed without evidence of endometrial hyperplasia or endometrial cancer in either of the tibolone groups (50). Tibolone has been shown to have minimal to no effect on the endometrial thickness using ultrasonography. Clinical trials with ultrasound endometrial evaluation have not shown a significant change, or less than a 1 mm increase, in the endometrial thickness in women using tibolone (17,20,21,51). The combination of an atrophic endometrium by histologic evaluation with failure to find stimulation of the endometrial thickness or growth by ultrasound and the lack of endometrial bleeding in these women is a strong argument for its lack of effect on the endometrium. The low incidence of endometrial bleeding and spotting is thought to increase compliance and continuation with therapy in postmenopausal women who are experiencing menopausal symptoms.
849
C. Effects on t h e B r e a s t We do not have any adequate surrogates to indicate that a specific hormone therapy would be better or worse for women who are at risk for breast cancer. We do know that tamoxifen (a SERM) has been used for many years to reduce the recurrence of breast cancer. There has been recent publicity from the Study of Tamoxifen and Raloxifene (STAR) trial in a group of women at high risk for breast cancer. Raloxifene (a SERM) had a reduction in the incidence of breast cancer of approximately 50% in this high-risk group of women, comparable to tamoxifen (http://www.nsabp.pitt.edu/STAR/ Index.asp). Published data are not available at the time of this writing. A significant difference between hormone therapy and SERM has been the lack of an effect of the SERM on breast density as assessed by standard mammographic techniques (52). Tibolone has been shown to be similar to a SERM in this regard (15). Clinical studies show minimal to no effect on mammographic density in women using tibolone, whereas those women using a standard estrogen plus progestin therapy have an increase in breast density (33,53-56). Histologic studies of human breast tissue have demonstrated less mitotic activity with tibolone in contrast to estrogen and progestin (55,57,58). These findings of decreased mitotic activity in the human breast have been corroborated in a nonhuman primate model exposed to either tibolone or CEE/ MPA (59). After 2 years of treatment in the nonhuman primate with either low- or high-dose tibolone, minimal mitotic activity was found, in contrast to the CEE/MPA. These data have been used to suggest that tibolone, because of no increased mitotic activity, no impact on mammographic density, and evidence of inhibition of the enzymes in the breast leading to estradiol, could reduce the incidence of breast cancer in postmenopausal women. These clinical findings are similar to those reported with SERMs. The tibolone data are too limited to make any firm statement regarding this hypothesis. Tibolone had a comparable risk of breast cancer to estrogen alone in the Million Women study (60).
D. Effects on Sexual A c t i v i t y ( L i b i d o )
FIGURE 60.3 Conversionoftibolone to the A4 isomer,which takes place preferentially in the endometrium. (From ref. 14.)
There has been an increasing interest in the decline in libido or sexual activity in postmenopausal women (61-63). Postmenopausal women do suffer a reduction in their interest in and incidence of sexual activity (62). The evidence indirectly implicates decreased serum testosterone levels in the postmenopausal female as a possible cause of the reduction in sexual activity and decline in libido (64-66). Tibolone increases free testosterone in postmenopausal women. Tibolone does not increase sex hormone-binding globulin
850
ARCHER AND GALLAGHER
(SHBG) levels, whereas oral estrogens increase SHBG levels (34). Tibolone itself and its A4 isomer have slight binding activity to the androgen receptor, and this may also play a role in enhancing libido. Clinical trials, although they are small and limited, suggest that there is an increase in libido in the postmenopausal woman using tibolone (34).
E. Bone The pivotal study was performed in North America and randomized 767 women to four doses of tibolone and placebo (67). Spine density increased by 3% to 4.5% on the 2.5, 1.25, and 0.625 mg/day doses compared with placebo but not on the 0.3 mg/day dose. Total hip bone density increased 3% to 4% on the 2.5 and 1.25 mg doses compared with placebo but not on the 0.3 mg dose. In general the reduction in bone resorption markers data showed that better suppression of bone resorption (approximately 40%) was achieved with doses of 2.5 and 1.25 mg/day. For these reasons the recommended dose of tibolone for preventing osteoporosis is 1.25 mg/day. Long-Term Intervention on Fractures with Tibolone (LIFT), a treatment study, is just completing 3 years of examination of elderly subjects with osteoporosis. In the L I F T study, the Data and Safety Monitoring Committee recently reported an increased risk of stroke in this group of older women being treated in the osteoporosis study (68). This recent abstract reported a 50% reduction in vertebral fracture incidence on the 1.25 mg dose (68). Because the efficacy of tibolone has now been established in reducing fractures, the Safety and Monitoring Committee recommended that the L I F T trial be discontinued early because of the risk/benefit ratio; the full details of the trial have not yet been reported.
F. Cardiovascular Although tibolone decreases high-density lipoprotein (HDL) cholesterol, it also decreases lipoprotein(a) and triglycerides. In the T H E B E S trial, all treatment groups showed a decrease in low-density lipoprotein (LDL) cholesterol, but in contrast to Prempro, there was a decrease in serum triglycerides. A recent placebo-controlled study (Osteoporosis Prevention and Arterial Effects of Tibolone [OPAL]) showed that both tibolone and estrogen produced a faster increase in carotid intimal thickness over 3 years than the placebo group (69). Although there has been a reported increase in stroke in the elderly women in the L I F T study, there has not been an increased risk of heart attacks reported from the T H E B E S or L I F T trials. It may be that the studies were not large enough to detect an increase in heart attacks because on estrogen the event rate is low over 5 years; however, the same caution is needed in the
use of tibolone in elderly women as that recommended for estrogen.
Ill. C O N C L U S I O N S Tibolone is a safe and effective therapy in early menopausal women. Its indications for use are several. It is effective in treating vasomotor symptoms. It causes some endometrial bleeding in the first few months, but this decreases during the first year more than on estrogen. Despite the bleeding, endometrial biopsies show no evidence of endometrial stimulation and there has been no evidence of endometrial hyperplasia or cancer after 2 years. It also has the advantage over SERMs of increasing the vaginal maturation value. At the 1.25 mg dose, tibolone is highly effective in preventing bone loss in women in the early postmenopausal years, and early reports suggest that this dose can prevent fractures in older women. In general, tibolone is well tolerated, and there has been no increase in the risk of cardiovascular events or venous thrombosis in younger postmenopausal women. In women in their mid-60s, tibolone increases the risk of stroke in the 1.25 mg dose, and although it appears to reduce fractures, the risk/benefit ratio of the 1.25 mg dose has to be taken into consideration in older women.
References 1. Kloosterboer HJ. Tissue-selectivity:the mechanism of action of tibolone. Maturitas 2004;(48 suppl 1):$30-40. 2. Kloosterboer HJ. Tibolone: a steroid with a tissue-specific mode of action.J Steroid Biochem Mo! Bio12001;76:231-238. 3. Markiewicz L, Gurpide E. In vitro evaluation of estrogenic, estrogen antagonistic and progestagenic effects of a steroidal drug (Org OD-14) and its metabolites on human endometrium.J Steroid Biochem 1990; 35:535-541. 4. Landgren MB, Helmond FA, Engelen S. Tibolone relieves climacteric symptoms in highly symptomatic women with at least seven hot flushes and sweats per day.Maturitas 2005;50:222-230. 5. Di Lieto A, De FalcoM, Mansueto G, et al. Preoperativeadministration of GnRH-a plus tibolone to premenopausal women with uterine fibroids: evaluation of the clinical response, the immunohistochemical expression ofPDGF, bFGF and VEGF and the vascularpattern. Steroids 2005;70:95-102. 6. Hudita D, Posea C, Ceausu I, Rusu M. Efficacy and safety of oral tibolone 1.25 or 2.5 mg/day vs. placebo in postmenopausal women. Eur Rev Med Pharmacol Sci 2003;7:117-125. 7. Landgren MB, Bennink HJ, Helmond FA, Engelen S. Dose-response analysis of effects of tibolone on climacteric symptoms. BJOG 2002; 109:1109-1114. 8. Steckelbroeck S, Jin Y, Oyesanmi B, Kloosterboer HJ, Penning TM. Tibolone is metabolizedby the 3alpha/3beta-hydroxysteroiddehydrogenase activitiesof the four human isozymesof the aldo-keto reductase 1C subfamily:inversion of stereospecificitywith a delta5(10)-3-ketosteroid. Mol Pharmaco12004;66:1702-1711.
9. Kloosterboer HJ. Tissue-selective effects of tibolone on the breast. Maturitas 2004;49:$5- $15.
CHAPTER 60 Tibolone: Selective Tissue Estrogen Activity Regulator Utilization in Postmenopausal W o m e n 10. Reed MJ, Kloosterboer HJ. Tibolone: a selective tissue estrogenic activity regulator (STEAR). Maturitas 2004;48(suppl 1):$4-6. 11. Pasqualini JR. Role, control and expression of estrone sulfatase and 17 beta-hydroxysteroid dehydrogenase activities in human breast cancer. Zentralbl Gynako11997;119(suppl 2):48-53. 12. Pasqualini JR, Chetrite GS. Estrone sulfatase versus estrone sulfotransferase in human breast cancer: potential clinical applications. J Steroid Biochem Mol Bid 1999;69:287-292. 13. Pasqualini JR, Chetrite GS. Recent insight on the control of enzymes involved in estrogen formation and transformation in human breast cancer. J Steroid Biochem Mol Bid 2005;93:221-236. 14. Timmer CJ, Verheul HA, Doorstam DP. Pharmacokinetics of tibolone in early and late postmenopausal women. Br J Clin Pharmacol 2002; 54:101-106. 15. Tang B, Marldewicz L, Kloosterboer HJ, Gurpide E. Human endometrial 3 beta-hydroxysteroid dehydrogenase/isomerase can locally reduce intrinsic estrogenic/progestagenic activity ratios of a steroidal drug (Org OD 14).J Steroid Biochem Mol Bid 1993;45:345-351. 16. Timmer CJ, Huisman JA. Effect of a standardized meal on the bioavailability of a single oral dose of tibolone 2.5 mg in healthy postmenopausal women. Pharmacotherapy2002;22:310- 315. 17. Bruce D, Robinson J, Rymer J. Long-term effects of tibolone on the endometrium as assessed by bleeding episodes, transvaginal scan and endometrial biopsy. Climacteric 2004;7:261-266. 18. Wender MC, Edelweiss MI, Campos LS, de Castro JA, Spritzer PM. Endometrial assessment in women using tibolone or placebo: 1-year randomized trial and 2-year observational study. Menopause 2004;11: 423-429. 19. Volker W, Coelingh Bennink HJ, Helmond FA. Effects of tibolone on the endometrium. Climacteric 2001;4:203-208. 20. Morris EP, Wilson PO, Robinson J, Rymer JM. Long term effects of tibolone on the genital tract in postmenopausal women. Br J Obstet Gynaeco11999;106:954-959. 21. Egarter C, Huber J, Leikermoser R, et al. Tibolone versus conjugated estrogens and sequential progestogen in the treatment of climacteric complaints. Maturitas 1996;23:55-62. 22. Hammar M, Christau S, Nathorst-Boos J, Rud T, Garre K. A doubleblind, randomised trial comparing the effects of tibolone and continuous combined hormone replacement therapy in postmenopausal women with menopausal symptoms. Br J Obstet Gynaeco11998;105:904- 911. 23. Prelevic GM, Markou A, Arnold A, et al. The effect of tibolone on bone mineral density in postmenopausal women with osteopenia or osteoporosis--8 years follow-up. Maturitas 2004;47:229-234. 24. Rymer J, Robinson J, Fogelman I. Ten years of treatment with tibolone 2.5 mg daily: effects on bone loss in postmenopausal women. Climacteric 2002;5:390390- 390488. 25. Gallagher JC, Baylink DJ, Freeman R, McClung M. Prevention of bone loss with tibolone in postmenopausal women: results of two randomized, double-blind, placebo-controlled, dose-finding studies. J Clin Endocrinol Metab 2001;86:4717-4726. 26. Rymer J, Robinson J, Fogelman I. Effects of 8 years of treatment with tibolone 2.5 mg daily on postmenopausal bone loss. Osteoporos Int 2001;12:478-483. 27. Bjarnason NH, Bjarnason K, Haarbo J, Rosenquist C, Christiansen C. Tibolone: prevention of bone loss in late postmenopausal women. J C/in EndocrinolMetab 1996;81:2419-2422. 28. Gambacciani M, Ciaponi M, Cappagli B, et al. A longitudinal evaluation of the effect of two doses of tibolone on bone density and metabolism in early postmenopausal women. GynecolEndocrino12004;18:9-16. 29. Gold EB, Sternfeld B, Kelsey JL, et al. Relation of demographic and lifestyle factors to symptoms in a multi-racial/ethnic population of women 40-55 years of age. Am J Epidemio12000;152:463-473. 30. Ettinger B, Pressman A, Silver R Effect of age on reasons for initiation and discontinuation of hormone replacement therapy. Menopause 1999;6: 282-289.
851
31. Baracat EC, Barbosa IC, Giordano MG, et al. A randomized, openlabel study of conjugated equine estrogens plus medroxyprogesterone acetate versus tibolone: effects on symptom control, bleeding pattern, lipid profile and tolerability. Climacteric 2002;5:60-69. 32. Modelska K, Cummings S. Tibolone for postmenopausal women: systematic review of randomized trials. J Clin Endocrinol Metab 2002; 87:16-23. 33. Kenemans P, Speroff L. Tibolone: clinical recommendations and practical guidelines. A report of the International Tibolone Consensus Group. Maturitas 2005;51:21-28. 34. Davis SR. The effects of tibolone on mood and libido. Menopause 2002;9:162-170. 35. Archer DF, Dorin M, Lewis V, Schneider DL, Pickar JH. Effects of lower doses of conjugated equine estrogens and medroxyprogesterone acetate on endometrial bleeding. Fertil Steri12001;75:1080-1087. 36. Johnson JV, Davidson M, Archer D, Bachmann G. Postmenopausal uterine bleeding profiles with two forms of continuous combined hormone replacement therapy. Menopause 2002;9:16-22. 37. Simon JA, Liu JH, SperoffL, Shumel BS, Symons JP. Reduced vaginal bleeding in postmenopausal women who receive combined norethindrone acetate and low-dose ethinyl estradiol therapy versus combined conjugated equine estrogens and medroxyprogesterone acetate therapy. AmJ Obstet Gyneco12003;188:92-99. 38. Archer DF, Dorin MH, Heine W, Nanavati N, Arce JC. Uterine bleeding in postmenopausal women on continuous therapy with estradiol and norethindrone acetate. Endometrium Study Group. Obstet Gyneco11999;94:323- 329. 39. Archer DF, Furst K, Tipping D, Dain MP, Vandepol C. A randomized comparison of continuous combined transdermal delivery of estradiolnorethindrone acetate and estradiol alone for menopause. CombiPatch Study Group. Obstet Gyneco11999;94:498-503. 40. Archer DF, Thorneycroft IH, Foegh M, et al. Long-term safety of drospirenone-estradiol for hormone therapy: a randomized, doubleblind, multicenter trial. Menopause 2005;12:716-727. 41. Bjorn I, Backsrom T. Drug related negative side-effects is a common reason for poor compliance in hormone replacement therapy. Maturitas 1999;32:77-86. 42. Li C, Samsioe G, Lidfelt J, Nerbrand C, Agardh CD. Important factors for use of hormone replacement therapy: a population-based study of Swedish women. The Women's Health in Lund Area (WHILA) Study. Menopause 2000;7:273-281. 43. Manonai J, Theppisai U, Suchartwatnachai C, Jetsawangsri T, Chittacharoen A. Compliance with hormone replacement therapy in Thai women. Maturitas 2003;44:201-205. 44. Christodoulakos GE, Botsis DS, Lambrinoudaki IV, et al. A 5-year study on the effect of hormone therapy, tibolone and raloxifene on vaginal bleeding and endometrial thickness. Maturitas 2006;53:413-423. 45. Huber J, Palacios S, Berglund L, et al. Effects of tibolone and continuous combined hormone replacement therapy on bleeding rates, quality of life and tolerability in postmenopausal women. BJOG 2002;109:886-893. 46. Doren M, Rubig A, Coelingh Bennink HJ, Holzgreve W. Impact on uterine bleeding and endometrial thickness: tibolone compared with continuous combined estradiol and norethisterone acetate replacement therapy. Menopause 1999;6:299- 306. 47. Beral V, Bull D, Reeves G. Endometrial cancer and hormonereplacement therapy in the Million Women Study. Lancet 2005;365: 1543-1551. 48. de Vries CS, Bromley SE, Thomas H, Farmer RD. Tibolone and endometrial cancer: a cohort and nested case-control study in the UK. Drug Saf2005;28:241-249. 49. Lee KB, Lee JM, Lee JK, Cho CH. Endometrial cancer patients and tibolone: a matched case-control study. Maturitas 2006;55:264-269. 50. Ferenczy A, Archer D, Rymer J, et al. Tibolone and endometrial safety: results from THEBES, a randomized 2-year study comparing tibolone with CEE/MPA. Presented at EMAS, Istanbul, Turkey, June 3, 2006.
852 51. Kurtay G, Berker B, Demirel C. Transvaginal ultrasonographic assessment of the endometrium in asymptomatic, postmenopausal women using different HRT regimens containing tibolone or estrogen.JReprod Med 2004;49:893- 898. 52. Warren R. Hormones and mammographic breast density. Maturitas 2004;49:67-78. 53. Marchesoni D, Driul L, Ianni A, et al. Postmenopausal hormone therapy and mammographic breast density. Maturitas 2006;53:59-64. 54. Kutlu T, Ficicioglu C, Basaran T, Basaran E, Topaloglu T. Mammographic breast density changes after 1 year of tibolone use. Maturitas 2004;48:133-136. 55. Valdivia I, Campodonico I, Tapia A, et al. Effects of tibolone and continuous combined hormone therapy on mammographic breast density and breast histochemical markers in postmenopausal women. Fertil Steri12004;81:617-623. 56. Lundstrom E, Christow A, Kersemaekers W, et al. Effects of tibolone and continuous combined hormone replacement therapy on mammographic breast density. Am J Obstet Gyneco12002;186:717- 722. 57. Von Schoultz B. The effects of tibolone and oestrogen-based HT on breast cell proliferation and mammographic density. Maturitas 2004;49: $16-21. 58. Conner P, Christow A, Kersemaekers W, et al. A comparative study of breast cell proliferation during hormone replacement therapy: effects of tibolone and continuous combined estrogen-progestogen treatment. Climacteric 2004;7:50- 58. 59. Cline JM, Register TC, Clarkson TB. Effects of tibolone and hormone replacement therapy on the breast of cynomolgus monkeys. Menopause 2002;9: 422- 429. 60. Beral V. Breast cancer and hormone-replacement therapy in the Million Women Study. Lancet 2003;362:419-427.
ARCHER AND GALLAGHER
61. Leiblum SR, Koochaki PE, Rodenberg CA, Barton IP, Rosen RC. Hypoactive sexual desire disorder in postmenopausal women: US results from the Women's International Study of Health and Sexuality (WISHES). Menopause 2006;13:46- 56. 62. Dennerstein L, Alexander JL, Kotz K. The menopause and sexual functioning: a review of the population-based studies. Annu Rev Sex Res 2003;14:64- 82. 63. Kingsberg SA. Prevalence of hypoactive sexual desire disorder in postmenopausal women: results from the WISHES Trial. Menopause 2006;13:10-11. 64. Davis SR, Guay AT, Shifren JL, Mazer NA. Endocrine aspects of female sexual dysfunction. J Sex Med 2004;1:82- 86. 65. Nathorst-Boos J, Floter A, Jarkander-Rolff M, Carlstrom K, Schoultz B. Treatment with percutanous testosterone gel in postmenopausal women with decreased libido--effects on sexuality and psychological general well-being. Maturitas 2006;53:11 - 18. 66. Davis A, Gilbert K, Misiowiec P, Riegel B. Perceived effects of testosterone replacement therapy in perimenopausal and postmenopausal women: an internet pilot study. Health Care Women Int 2003;24: 831-848. 67. Gallagher JC, Baylink DJ, Freeman R, McClung M. Prevention of bone loss with tibolone in postmenopausal women: results of two randomized, double-blind, placebo-controlled, dose-finding studies.J Clin Endocrinol Metab 2001;86:4717-4726. 68. Cummings SR. LIFT study is discontinued. BMJ 2006;332:667. 69. Bots ML, Evans GW, Riley W, et al. The effect of tibolone and continuous combined conjugated equine oestrogens plus medroxyprogesterone acetate on progression of carotid intima-media thickness: the Osteoporosis Prevention and Arterial effects of tibolone (OPAL) study. Eur HeartJ 2006;27:746- 755.
SECTION X I V
Women's Centers, Information Needed for Clinical Trials, and the Future In the final three Chapters of the book, practical considerations are discussed. A focus on Menopausal Health ideally would require a specialized Center with an integrated multidisciplinary approach. Such a setting is described in detail by Morris Notelovitz who has had vast experience in the implementation of such Centers. This concept has led to many Menopause Centers being set up in the United States and around the world. Some have been successful but many have failed. In my view, unless various types of research activities are also carried out in conjunction with routine clinical care in the Center, the Center will fail economically. Serious consideration has to be given to "doing this right" before a Center is launched. The second Chapter is an update from Christian F. Holinka and James H. Pickar on FDA requirements for various types of clinical trials. Both authors have had a great deal of experience in planning and implementing clinical trials, and this information is important for those contemplating participating as investigators in clinical trials as well as to provide information to those providers whose patients are participating in clinical trials. Although this book has attempted to go beyond the issues of hormonal therapy for postmenopausal women, the last Chapter provides a practical view for the current status, and perhaps the future, of hormonal therapy. This field which has evolved over several decades has been sent on a tailspin in the last few years. My personal view is that things in the world of hormonal therapy have started to level off, and it is appropriate at this time to take a hard look at what we should consider doing now and in the future.
This Page Intentionally Left Blank
~HAPTER
6_
Integrated Adult Women's Medicine: A Model for Women's
Health Care Centers MORRIS NOTELOVITZ
Adult Women's Health & Medicine, Boca Raton, FL 33496
In welcoming remarks to participants at the 4th International Congress on the Menopause (October, 1984), it was noted that:
The model described in this article is derived from an appreciation of the various biologic events that have an impact on women during their premenopausal and postmenopausal life cycle and on the experience, research, and clinical observations from the Menopause Clinic Addington Hospital, Durban, RSA (1968), the Center for Climacteric Studies, University of Florida (1980), and 10 years of private practice experience at the Women's Medical & Diagnostic Center, Gainesville, Florida (1986-1996). This experience led to the concept of climacteric medicine being expanded to include the care of younger premenopausal women and conditions not associated with climacteric per se, hence, the revised descriptor, "Integrated Adult Women's Medicine," and more recently "adult women's health and medicine" (2a).
The needs of women vary during the climacteric. Latent changes in the early phase have the potential for significant morbidity in the later years; by early recognition and prevention, conditions such as osteoporosis and atherogenic cardiovascular disease may be ameliorated or even prevented. It is for this reason that the theme for the Congress--"Climacteric Medicine and Science: A Societal Need"nhas been chosen. By recognizing the needs of the total woman, the expertise of the primary care physician (gynecologist or family practitioner) is complemented by the skills and talents of nutritionists, exercise physiologists, psychologists and social counselors. Traditional health care specialists, in turn, need to work with the basic and social scientist. (1)
I. T H E P R I N C I P L E S
Climacteric medicine is still not recognized as a medical entity, but increasing awareness of the total health care needs of women has led to the opportunity to establish adult women's health centers. No agreed-on formula exists and, indeed, no meaningful literature relates to the subject (2). T R E A T M E N T OF THE POSTMENOPAUSAL W O M A N
The menopause is a natural life event that all women experience. W i t h advances in health care, the life expectancy of women in Western and Asian societies has
855
Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
856
MORRIS NOTELOVlTZ
increased to the point where most women can expect to live 30 years beyond their last menstrual period. The key to achieving longevity is to reach age 65 years. After this milestone women can anticipate living between 19.5 and 23.1 more years (3). With this increased longevity, however, conditions associated with aging have also increased, including cardiovascular disease (and related risk factors), osteoporosis, urinary incontinence, Alzheimer's disease, cancer (breast and colon), arthritis, and failure of the sensory organs of hearing and vision. It is the management of these and related conditions that forms the scientific basis for the concept of integrated adult women's medicine. In this context, three basic issues have an impact on the design of health care centers for women: 1. The health care needs of women involve organ systems remote from the pelvis~the domain of the traditional primary care physician for women (gynecologist). 2. The pathogenesis of diseases, such as cardiovascular disease (4) and osteoporosis (5) precede the menopause by at least three decades. Thus, for preventive measures to succeed, patient advice and care should commence in early adolescence (Fig. 61.1). 3. Alterable factors such as exercise, lifestyle, diet, and social habits can have an impact on the future health and well-being of women, as much as the physiologic hormone deficiency associated with the menopause. For example, in the Finnish Twin Study (6), level of physical activity and not familial factors influenced premature mortality. Compared with their sedentary cohorts, conditioned female twins reduced their risk of death by 76% (odds ratio 0.24). The concept of climacteric or adult women's medicine thus incorporates the following principles.
=
. . . . . . . . . . . . .
Climacteric
=
9 Postmenopausei Premenopause Perimenopause
. . . . . . . . . . . . . . . . . . . . .
P M DUB
S
~
Contraception
. . . . . . . . . . . . .
-/? - -
-
"-lt---
Vasomotor instability
__._.__.__a Pshycologic/psychosomatic m,
Genito-urinary problems . . . . . . . Cardiovascular disease - Osteoporosis Alzheimer's ........
. . . . . . . . . . . . . . . . . . .
~~r-
tt 16 years
35 years
General medical ~ffd~rifi~
45
50 Menopausal
,
[ ....
t 9. ,. . . . . . . . . . . .
Latent phase
: [ i .........[
FIGURE 61.1
Cancer Cervical Ovarian Breast Colon 55
medicine
1. Consumer education. Only an educated consumer can make informed decisions regarding her own health care needs. 2. Centralized technology. To make treatment decisions, certain diagnostic modalities should be available to individualize treatment regimens and monitor progress and treatment. This assists with treatment compliance (7,8). 3. Multidisciplinary staffing. This needs to include primary (family practitioner, nurse practitioner, physician assistant) and gynecologic care and medical subspecialists (radiologists, cardiologists, rheumatologists, and so forth); paramedical professionals (physical therapists; exercise physiologists; nutritionist; counselor); and appropriately trained nursing and clinical staff. 4. By housing all of these in one facility, the various professionals can function as a team supervising the health of one individual. Important cost savings are seen in terms of overhead expenses and, of course, convenience to the patient. Health care centers for women should avoid medicalization of the menopause. Nonmedical alternatives should be made available. However, the prevalence of various age and menopause-related diseases are a reality. The earlier they are detected and treated, the better. Only a comprehensive approach involving a multidisciplinary health care team and the selective use of appropriate diagnostic technology can optimize the health care of women.
disease
Hearing
II. THE PRACTICE /i 65
80+
A. PrimaryTarget Population
(45-55)
Ciii~acteric medicine
(35-65) (16-80+)
=
L
For this comprehensive and integrated approach to succeed, a center that provides for the patient's total health care has to be cost-effective and provide long-term benefits to both the individual and to society. The center also has to be economically viable. Experience at the Women's Medical & Diagnostic Center (WMDC) has found the following to be essential:
Vision .....
years
Adult women's medicine
Presenting symptom
Arthritis
1. Recognition of the woman as a total person. 2. Meeting her health care needs across the spectrum of her adult life cycle (age 16 onwards). 3. Inclusion of preventive as well as curative programs. Thus, services such as exercise prescription and nutritional advice should be available as an adjunct (or an alternative) to hormone therapy and other disease-specific medications. 4. Provision of diagnostic technologies for the early diagnosis of latent disease, for individualization of health care and monitoring of response to therapy.
]
Health care needs by life stage.
Climacteric medicine (as previously defined) involves the ambulatory care of women between 35 to 65 years of age and incorporates both preventive and curative health care. Adult women's medicine incorporates the management of
85 7
CHAPTER 61 Integrated Adult Women's Medicine: A Model for Women's Health Care Centers young adult women (i.e., the care of women from 16 to 35 years old) (Fig. 61.2). This provides the potential for optimal preventive medicine by emphasizing diet, exercise, and lifestyle modification from an early age. Many chronic conditions may be prevented by simple and inexpensive measures. For instance, where osteoporosis is concerned, it is now established that exercise and calcium plus vitamin D supplementation in teenagers leads to greater bone mineral accrual and the potential for an enhanced peak bone mass (9,10). The greater this peak bone mass, the lower is the risk for osteoporosis (5). In addition, osteoporosis risk factors, such as eating disorders and menstrual irregularities, can be readily detected and appropriately treated. Also, childhood obesity increases the risk of obesity in adulthood (11) and, therefore, the risk of diseases such as diabetes and coronary artery disease. A recent study confirmed that the children of parents with coronary artery disease were overweight in childhood (12). The implication for early lifestyle and exercise intervention is self-evident. Instruction in safe sexual habits and effective contraception is yet another benefit younger women derive, both for their future reproductive potential and for their later postmenopausal health. For example, long-term use of oral contraceptives significantly reduces the risk of ovarian and endometrial cancer and does not increase the risk of breast cancer (13). It has been the experience at the W M D C that obstetric care should not be included in an adult women's center. Geriatrics is a recognized speciality with specific patient and clinic requirements, and this practice was not incorporated into the W M D C model. However, many elderly women need to be evaluated (using dual-energy x-ray absorptiometry [DEXA] testing) and treated (with specific medications and physical therapy) for conditions such as osteoporosis. In addition, elderly women require annual physical examination with cancer surveillance (Pap smears and mammograms), screening for colon disease (using guaiac tests, sigmoidoscopies and selective colonoscopies), and other common health problems.
These include the evaluation and management of recurrent urinary tract infections and, particularly, urinary incontinence, which may require urodynamic testing. Urinary incontinence, which is probably the most common condition in older women, consistently and significantly reduces quality of life. A study confirmed that a simple questionnaire regarding voiding and leaking patterns serves as a valuable tool to detect this problem (14). Additional age-related problems that frequently occur are uterovaginal prolapse, arthritis, and visual and hearing impairment. Thus, in reality, senior citizen health care is an important extension of adult women's medicine. Therefore, in practice, mainly the frail elderly are referred for geriatric care.
B. Staffing 1. TYPE OF PRACTICE
In traditional women's curative medicine, patients are treated according to their presenting problem and the physician's area of interest. In contrast, climacteric and adult women's medicine targets the whole woman with a multidisciplinary, integrated approach, emphasizing (but not limited to) preventive care. Apart from routine gynecologic evaluations and management, women obviously need treatment for nonreproductive complaints such as influenza and bronchitis. Women are also subject to potentially life-threatening diseases such as cardiovascular disease, hypertension, osteoporosis, and metabolic and endocrine disorders, as well as common non-life-threatening conditions such as arthritis, urinary incontinence, and visual and hearing impairment. Screening for breast, colon, lung, and genital tract cancers forms an essential core of clinical services that should be offered. Essential, too, are the emotional, psychologic, and nonmedical needs of women, such as career guidance. 2. MEDICAL STAFF: ADULT WOMEN'S GENERALIST
FIGURE. 61.2 Defining the medical practice: integrated adult women's medicine. This model allows the potential for optimal preventive medicine by emphasizing diet and exercise and lifestyle modification from an early age.
A team approach is needed to optimize adult women's care. In practice, the medical services are best coordinated by a primary care adult women's generalist who functions as a gate keeper. As illustrated in Fig. 61.3, management of each case is determined by the adult women's generalist (AWG) after the screening and diagnostic procedure identifies either no problem or a primary (or latent) problem requiring medical attention. This process ensures cohesive integrated and cost-effective care. The AWG has to be well-rounded in primary care and internal medicine, and should be appropriately trained to perform gynecologic procedures such as colposcopy, endometrial biopsies, and pelvic ultrasound. The AWG could also be a board-certified gynecologist, provided the physician is experienced in general internal medicine.
858
MORRIS NOTELOVlTZ
1 NO PROBLEM [
,SCREENING
I
NO PROBLEM
I
I..,- l l
MANAGED BY AWG OR REFFEREDTO SPECIALIST .
FIGURE 61.3
I
W M D C . This approach was enthusiastically endorsed by patients, who appreciated receiving individualized and total health care at a single facility. A panel of additional consultants was formed that included specialists in rheumatology, endocrinology, neurology, psychiatry, general and cosmetic surgery, and physiatry. These services were provided in the consultants' private office but should be incorporated into a "definitive" adult women's health center.
.
.
.
.
.
.
A model for a team approach to adult women's medicine.
Physician assistants or nurse practitioners function very successfully and cost-effectively as AWGs. 3. SPECIALISTS AND CONSULTANTS
Summarized in Fig. 61.4 is the staffing structure of the W M D C from 1986 to 1996. The four ambulatory care providers were full-time employees. The consultants were contracted for their services, all of which were provided at
4. PARAMEDICAL STAFF
The paramedical personnel comprise an essential component of the adult women's health team. Professionals included psychologists (in general and sex counseling) who offered traditional therapies, including treatment with biofeedback for stress management and certain aspects of sexual dysfunction; a dietitian provided evidence-based advice regarding weight loss with (selectively) resting energy expenditure testing (using indirect calorimetry), DEXA body composition analysis and computer-assisted dietary analysis. The physical therapy department concentrated on the management of problems related to arthritis, osteoporosis, and urinary incontinence (caused by pelvic floor muscle dysfunction). Finally, individualized prescriptive exercise regimens were developed by an exercise physiologist (under medical supervision) for the treatment of special medical entities (e.g., hypertension, diabetes, osteoporosis). All patients had an on-site exercise stress test
Ambulatory
p~Jma~Care
1
I
Family Practitioner
I
.
.
.
.
.
.
I Medicine .... Climacteric & Urogynecology
I
Physician Assistant .
.
I
!
[Consu'tants
I
i
I
]
I I Paramedical [
I
Radiologist
Preventive Cardiologist Cardiologist
Ophthalmologist ENT Surgeon Gastroenterologist
I
"1 Gynecologist
J
,
i ......... Medical ]
I
] Assistants
Psychologist Sex Counselor
Dietician Exercise Physiologist Physical Therapist X-ray Technologist I,
Nursing X-ray Technologists Administrative
I
FIGURE 61.4 Implementing the practice: a model for integrated women's medicine at the Women's Medical & Diagnostic Center, Gainesville, Florida (1986-1996). ENT, ear, nose, and throat.
I I Metabolicl
[ cardi~
I
i
[B~
Hypertension HyperUpidemia Diabetes 1 Stroke Obesity
I
] DEXA- Markers - X-ray
I
I Lipid Profile
I
Osteopenia Osteoporosis Arthritis
-" C.H.D.
Anorexia
Hea,th
Glucose Tolerance I Exercise Stress Test I REE Flexibility I (Echo Stress Test) I DEXA Body Strength " Cardiac Ultrasound Composition Balance Carotid Doppler
I
I
'
Women'sl
Generalist
"
i
I
PRN Cardiologist
Endocrinologists
Physiatrist ]
F~
Exercise Physiologist Nutritionist Physical Therapist
[ Patient I
I
FIGURE 61.5 Model for a prescriptive exercise program. CHD, Coronary heart disease; PRN, as needed; REE, resting energy expenditure test.
859
CHAPTER 61 Integrated Adult Women's Medicine: A Model for Women's Health Care Centers
(under the supervision of a cardiologist) and additional selective tests according to their medical condition (e.g., DEXA testing if they had osteoporosis) (Fig. 61.5). Both aerobic and resistive exercises were prescribed. The exercise program was conducted off-site under the guidance of a trained physical therapist or exercise physiologist. The administrative staff, nurses, and x-ray technicians were critical to the success of the program at the W M D C . All were thoroughly schooled in both the philosophy and the practice of adult women's medicine. Their input was invaluable in reassuring patients that their care was in the hands of trained professionals. Much of the education and counseling of patients and their reassurance vis-d-vis treatment protocols (and the anticipated response or potential side effects) was undertaken by the nurses. X-ray technologists were trained to recognize abnormalities (e.g., in mammograms, x-ray films, or DEXA tests) and were encouraged to consult with the on-site physicians so that additional views or tests could be ordered prior to the patient leaving the center. Patients were often scheduled for physician consultations immediately after testing.
C. Patient Reporting and Compliance All patients received individualized written reports summarizing the outcome of their visit. Full reports were sent to referring physicians. Compliance was high, which was probably due to the patient's being appropriately educated to the point where she participated in her treatment decision. Ready telephonic access to trained and knowledgeable nurses was frequently favorably commented upon as well as the appointment reminders sent via mail and telephone.
III. CLINIC DESIGN Clinic design focused on respect for the individual and an understanding of her medical needs. The ambience and decor of a center sets the tone for the practice. Regional tastes vary. Thus, the colors, furnishings, and background, such as music, need to be tailored to the clientele. Research and clinical experience at the W M D C identified the type of customized patient examination furnishings and gowns, for example, that are universally acceptable and complementary to the overall relaxed nonmedical environment of the facility. Central to the design is a floor plan that is convenient for the patient and labor saving for the staff. To facilitate this goal, services relevant to a given condition should be clustered according to the diagnostic and therapeutic requirements of that condition. An example is illustrated in Fig. 61.6. Osteoporoticrelated fractures are associated with reduced bone mass, microarchitectural deterioration, and falls. Thus, the technology needed to evaluate bone health and strength (DEXA, bone
OSTEOPOROSlS
I
Bone M a s s
Fragility
I
I
Falls
I
DEXA - Markers - X-ray Ultrasound Balance 2 ~ Tests L - - Technicians J I Physical Therapy L ENT - Vision Adult Women's....] Generalist
I
I~
]
I at''nt I
Consult P.R.N. I Radiologist Endocrinologist Rheumatologist Orthopedist
FmURE 61.6
I
Physical Therapist Exercise Physiologist Nutritionist
Clinic design: cluster system.
ultrasound, and x-ray studies) is contiguous with the physical therapy department, where balance, posture, and muscle strength and flexibility are evaluated. The x-ray technologists are also trained in specimen collection techniques, for the evaluation of bone remodeling and other relevant biochemical tests, required either before or after the DEXA evaluation. Where relevant, impairment with hearing and inner ear dysfunction is evaluated by ear, nose, and throat (ENT) technicians; problems with visual acuity are evaluated by trained ophthalmic nurses. Both paramedicals are supervised by an ENT specialist and an ophthalmologist, who also consult with patients for hearing and visual problems unrelated to their risk for falling. Placement together of these seemingly unrelated diagnostic modalities also leads to patient-initiated crossreferral: a patient scheduled for an eye test realizes that she needs a bone density test and vice versa. Similar paradigms cluster technology and personnel for cardiovascular health evaluation, obesity management, cancer screening, urinary incontinence, and certain specialized team clinical services such as the premenstrual syndrome and eating disorder management.
IV. TECHNOLOGY: CLINICAL RESEARCH AND CONSUMER EDUCATION It is important to recognize that integrated adult women's medicine does not exist as an approved medical discipline. However, the development of various technologies allows
860 physicians to evaluate patients selectively and to acquire objective information, thus facilitating the practice of clinical evidence-based medicine. As such, specific and lower dose regimens can be prescribed for a given condition and the therapeutic response monitored with dosage adjustment made accordingly. Simply, prevention of disease for many conditionsmsuch as osteoporosismhas now become a reality. Hormone therapy (15) and some of the new antiresorptive drugs (16) have been shown to reduce vertebral fracture risk by 50% to 90%. Newer tissue-specific drugs are being developed and evaluated for total menopausal health care (17). The use of natural products, such as phytoestrogens, holds great promise. In certain societies (e.g., Japan), phytoestrogens have been linked with a much lower prevalence of cardiovascular disease and breast and endometrial cancer (18) when taken as part and parcel of their daily diet. The key is convenient and cost-effective technology for use in clinical medicine that identifies biologic markers that are closely linked to the pathogenesis of a given disease. Evaluation of evolving technology and participation in clinical drug trials are important activities that provide two advantages: early familiarity with cutting-edge technology and therapies, and research revenues that complement clinical income. Most women are unaware of the existence or the advantage of comprehensive health care. Even when available, services that are regarded as preventive may not be reimbursable. This was certainly the situation when the W M D C was first established. Yet, as the programs offered at the W M D C became known, women recognized the personal benefit they would derive, and they became paying patients. The message is clear: educated women can (and do) make informed decisions. The difficulty is how best to educate them about their health. Educational materials should be available and well-displayed in the office. The W M D C had a special lounge set aside for group discussions with a small library in the room. It is advisable to review the reading materials available on the subject of menopause and adult women's health. Much is written, but relatively few texts are authored by individuals qualified or experienced in climacteric care. At the W M D C additional educational services included a 1-800 call-in service with recorded messages on different aspects of the menopause, and a quarterly medical journal written for lay consumption (Fig. 61.7). Both of these services could now be made available on the Internet and customized to the practice. The importance of consumer education should not be underestimated. Approximately 95% of patients attending the W M D C were self-referrals~by friends who experienced the facility and through articles on the menopause in magazines that included information about the W M D C . In fact, women from 42 states sought consultation or ongoing care after learning about the W M D C through the national media.
MORRIS NOTELOVlTZ LOOKING FOR INFORMATION
' EDUCATIONAL SERVICES
I
]
EDUCATIONAL 9 PRODUCTS WOMEN TO WOMAN AMERICA (JOURNAL) WOMEN'S HEALTH DIGEST 1-800-MENOASK I ! EDUCATED CONSUMER
I
WHERE CAN SHE GET GOOD CARE? I
HOW CAN SHE HELP IMPROVE WOMEN'S HEALTHCARE I WOMEN'S MEDICAL & I DIAGNOSTIC CLINIC WOMEN'S RESEARCH 'CENTERS
FIGURE 61.7 WMDC Model for integrating consumer education with clinical services and research.
V. CONCLUSION There is a biologic limit to everyone's lifespan. Fries et al. (19) conceptualized the "rectangularization" of life, with the goal of maintaining physical and mental function as long as possible, until the inevitable and intended, acute point of decline and eventual death (Fig. 61.8). The practice of climacteric and adult women's medicine attempts to maximize
FIGURE 61.8 Compression of morbidity with aging: raising the functional health threshold. Reproduced from Notelovitz M. The Adult Woman's Health Plan. In Schneider HPG, ed. Menopause." the State of the Art-in Research and Management. London: Parthenon Publishing, 2003: 502-7. Adapted from Fries and Crapo, 1980.
861
CHAPTER 61 Integrated Adult Women's Medicine: A Model for Women's Health Care Centers
the active life expectancy of women and to allow them a full vigorous and independent life. A vital issue in this quest is preventive measures in young adulthood. By increasing the "baseline" health threshold, the level of wellness will be raised and extended into later life. As a result, the risk for at least two of the most prevalent chronic disabilities associated with the older womanmosteoporosis and cardiovascular diseasemcould be modulated, delayed, or even eliminated. Climacteric and adult women's medicine is a discipline waiting to be born. In short, it can be defined as preventive medicine for women (and men) in their middle years, and it has as its basic premise consideration of the whole individual, with the objective of achieving a healthy mind in a healthy body in a healthy environment. As such, it may be regarded as a national insurance policy. A healthy middle-age population will be a productive population; by preventing or ameliorating chronic illnesses much of the need for and cost of long-term geriatric care can be avoided (Fig. 61.9). We have noted the benefit of preventive medicine in obstetrics and dentistry. Why should the climacteric be any different? It is in this context that climacteric medicine has emerged as a science and has become a societal need (1). Preventive medicine, in fact, is practiced by a wide variety of health care professionals, both on an individual basis (e.g., doing Pap smears and ordering mammograms) and at a societal level (e.g., immunization). However, preventive medicine has not been the major thrust of health care in the United States and in most other countries. This stems in part from prioritization of the health care dollar and because of inadequate education of medical students and residents in training. With the current demand for cost-containment in health care and the obvious monetary savings associated with the prevention of disease and disability, preventive and primary care are now being recognized as essential components of a reformed health system (20). Women's health care centers provide the ideal environment for the practice of preventive medicine.
FmURE 61.9 Concept template for adult women's medicine: cost-benefit of health promotion and disease prevention. The greater the emphasis on primary prevention--nutrition, exercise and lifestyle--the lesser the need and cost of diagnostic testing (secondary intervention) and/or post-event therapy.
References 1. Notelovitz M. Climacteric medicine and science: a societal need. In: Notelovitz M, Van Keep P, eds. The climacteric in perspective. Boston, MTP Press, 1986:19- 21. 2. Notelovitz M. Is there a need for menopause clinics: In: Multidisciplinary Perspectives on Menopause. Ann N Y Acad Sci 1990;592: 239-241. 2a. Notelovitz M. Optimizing women's health: adult women's health & medicine. Climacteric 2005;8:205-209. 3. Hammond C. Menopause and hormone replacement therapy: an overview. Obstet Gyneco11996;87:2S-15S. 4. Van Horn L, Greenland E Prevention of coronary artery disease is a pediatric problem. JAMA 1997;278:1779-1780. 5. Notelovitz M. Post-menopausal osteoporosis. A practical approach to its prevention. Acta Obstet Gynecol Scand Supp11986;134:67-80. 6. Kujala UM, Kaprio J, Sarna S, Koskenvuo M. Relationship of leisuretime physical activity and mortality. The Finnish Twin Cohort. JAMA 1998;279:440-444. 7. Cook B, Notelovitz M, Rector C, Krischer J. An osteoporosis education and screening program: results and implications. Patient Education and Counseling 1991;17:135-145. 8. Silverman SS, Greenwald M, Klein RA, Drinkwater BL. Effect of bone density information on decisions about hormone replacement therapy: a randomized trial. Obstet Gyneco11997;89:321-325. 9. Lloyd T, Martel J, Rollings N, et al. The effect of calcium supplementation and Tanner stage on bone density, content and area in teenage women. OsteoporosisInt 1996;6:276-283. 10. Boot AM, de Ridar MAJ, Pols H, et al. Bone mineral density in children and adolescents: relation to puberty, calcium intake and physical activity. J Clin Endocrinol Metab 1997;82:57-62. 11. Whitaker RC, Wright JA, Pepe MS, et al. Predicting obesity in young adulthood from childhood and parental obesity. N EnglJ Med 1997; 337:869-873. 12. Rao W, Srinivasan SR, Valdez R, et al. Longitudinal changes in cardiovascular risk from childhood to young adulthood in offspring of parents with cardiovascular disease. The Bogalusa Heart Study. JAMA 1997;278:1749-1754. 13. Kaunitz AM, Benrubi GI. The good news about hormonal contraception and gynecological cancer. The Female Patient 1998;23:43-51. 14. Robinson D, Pearce KF, Preisser JS, et al. Relationship between patient reports of urinary incontinence symptoms and quality of life measures. Obstet Gyneco11998;91:224-228. 15. Notelovitz M. Estrogen therapy and osteoporosis: principles and practice. Am J Med Sci 1997;313:2-12. 16. Black DM, Cummings SR, KarpfDB, et al. Randomized trial of effect of alendronate on risk of fractures in women with existing vertebral fractures. Lancet 1996;348:1535-1541. 17. Delmas P, Bjarnason N, Mitlak B, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations and uterine endometrium in post-menopausal women. N Engl J Med 1997;337: 1641 - 1647. 18. Aldercreutz H, Mazur W. Phyto-oestrogens and western diseases../Inn Med 1997;29:95 - 120. 19. Fries HF, Green LW, Levine S. Health preservation and the compression of mortality. Lancet 1989;1:481. 20. Eggert RW, Parkinson MD. Preventive medicine and health system reform. Improving physician education, training and practice. JAM// 1994;272:688-693.
This Page Intentionally Left Blank
~ H A P T E R 6,
Clinical Trials in Postmenopausal Hormone Therapy CHRISTIAN
F. HOLINKA
JAMES
H. PICKAR
PharmConsult | New York, NY 10014 Wyeth Research, Collegeville, PA 19426
I. I N T R O D U C T I O N
regulations by the European Agency for the Evaluation of Medicinal Products (EMEA). In our age ofglobalization, the availability of worldwide standards for clinical development has become increasingly more relevant. This was formally recognized in 1990 when the International Conference on Harmonization (ICH), a group of regulatory and industry agencies, convened for the first time. The group has since met regularly with the aim to improve, through harmonization, the efficiency of developing and of registering new medicinal products in the United States, Europe, and Japan in order to make such products available to patients with minimal delay. The FDA has recognized the importance of working with international regulatory authorities in several ways, such as in its effort (Federal Register, October 1997) to develop guidelines for safety reporting that take into account ICH safety update report requirements. The first part of the chapter presents a brief overview of the historical development that has resulted in the stringent regulations of present-day drug development in the United States and elsewhere, followed by a description of different types of clinical studies and a discussion of regulatory requirements prior to start, during, and at the conclusion of clinical trials. The second part covers issues
As a result of extensive clinical trials, a variety of postmenopausal hormone therapy (HT) products are now available, thus offering women and their physicians the opportunity to individualize therapy. Currently available products contain different estrogens and progestins in different strengths and combinations, such as continuous combined or sequential regimens. Hormonal preparations are available in a variety of formulations, including tablets, patches, and creams. Despite this array of products tested for efficacy and safety in rigorous clinical trials, health care professionals and consumers are often unaware of the complexity of these trials required for marketing approval of drugs, including HT products. This chapter provides an introduction to the design and conduct of clinical trials of postmenopausal HT. The terms clinical trial, clinical study, and clinical investigation will be used interchangeably to designate any experiment in which a drug is administered or dispensed to one or more human subjects (1). The chapter will use the current U.S. regulatory requirements, as defined by the Food and Drug Administration (FDA), as a basis for discussion, but reference will be made to TREATMENT OF THE POSTMENOPAUSAL W O M A N
863
Copyright 9 2007 by Elsevier,Inc. All rights of reproductionin any form reserved.
864 specifically related to clinical studies in postmenopausal H T and includes a discussion of data from recent wellcontrolled clinical trials that have challenged some conclusions based on earlier results from observational studies.
II. HISTORICAL ASPECTS The present-day regulatory environment of preclinical and clinical drug development in the United States is the result of a gradual legislative evolution. Up to 1906 no comprehensive federal statutes for drug development and approval existed. As a result of increasing public concern regarding the lack of oversight and standards and the often haphazard handling of drug substances, the Food and Drugs Act (Wiley Act) was passed in 1906. The Act prohibited the interstate transportation of adulterated and misbranded drugs and food but otherwise provided no regulations with regard to drug safety and efficacy. The Act remained the principal legislative statute until 1937, when more than 100 people died after consuming a marketed product, Elixir of Sulfanilamide, which contained the antifreeze diethylene glycol as a solubilizer. This tragedy underlined the need for stricter legislation concerning drug safety and resulted in the passage of the Food, Drug and CosmeticsAct of 1938. The Act required premarketing approval of a drug based on its safety profile and further required that a drug be labeled with adequate directions for its safe use. The Act also regulated cosmetics and medical devices. However, it fell short in several other important aspects. For example, drugs marketed prior to 1938 were exempt from review; proof of the benefits claimed in a given drug's labeling was not required; and active approval of marketing applications by the FDA was not necessary. A drug was automatically approved unless the Agency objected to the new drug application within 60 days. Additionally, the Act provided enforceable food standards. The Durham-Humphrey Amendment of 1951 defined what constituted a prescription and an over-the-counter drug. It required yet another tragedy to advance the regulatory process of drug approval to contemporary standards. A sedative and antinausea drug, thalidomide, which was approved and sold in some European countries, but not approved in the United States, was prescribed to an estimated 20,000 people. Epidemiologic evidence soon emerged linking thalidomide use by pregnant women to birth defects, such as severely shortened limbs. In the wake of this catastrophe, an amendment to the Food, Drug and CosmeticsAct was passed in 1962, known as the Kefauver-Harris Amendment. The amendment represents the core regulatory basis of modern drug development. It requires that clinical trials demonstrate effectiveness as well as safety for all new drugs and that trials be well controlled. It provided for greater FDA control of drug trials, including the requirement for
HOLINKA AND PICKAR
informed consent for patients in those trials. Most importantly, the amendment provided for affirmative action by the FDA concerning the approval of a new drug for marketing instead of automatic approval within a certain period after submission of a new drug application. In addition to the developments described here, there exist a large number of other legislative statutes that contribute to the history of food and drug regulations in the United States.
III. CLINICAL RESEARCH A central goal of clinical research is to provide a solid basis for evidence-based medicine, which can be combined with a person's individual and medical needs for optimal therapy. The importance of different types of studies and their appropriate design to obtain the best evidence has been widely recognized (2,3). However, there is not total agreement regarding how broadly to define clinical research, and some may properly consider its definition to be narrower in scope than discussed here. The main types of clinical research are defined as follows. 1. Clinical studies, as contrasted with observational studies, generally randomize patients or subjects, which can help to minimize bias. Additionally, they also provide a stronger base for statistical analysis than do observational studies. Two types of clinical studies are regularly used in clinical research: 9 Double-blind, prospective, randomized controlled clinical trials provide the strongest evidence related to clinical safety and efficacy for preventive or therapeutic interventions. The different treatment groups are studied in parallel over an appropriate period. Subjects, investigators, and investigational personnel are unaware of the assigned treatment. 9 Double-blind, prospective, randomized crossover clinical trials are trials where patients or subjects in a given treatment group are studied for a time and then switched to the other treatment for the same period. The two phases are usually separated by a washout period. In this type of study, the individual serves as his or her own control, thus eliminating intersubject variability and potentiaUy reducing the sample size required to detect a significant effect. Period effect is an additional consideration with this design. Fig. 62.1 presents an example of a double-blind, placebo-controlled crossover study (4). Clinical studies are often considered limited because they assess efficacy and safety under highly controlled, thoroughly monitored conditions and thus may not fully reflect the effects of the treatment in the real world. Also, clinical studies are often conducted in a carefully selected population that may not be representative of
CHAPTER 62 Clinical Trials in Postmenopausal Hormone Therapy
FIGURE 62.1 A double-blind, prospective, randomized crossover study showing the effects of estrogen (solid line) and placebo (broken line) over a 3-month treatment period followed by another 3 months after doubleblind crossoverof treatments. (From ref. 4, with permission.)
subpopulations, and despite randomization the subject characteristics may not be equal in all treatment groups because some inequality of distribution may occur by chance alone. This becomes more important in studies of small sample size. Nevertheless, despite these limitations, clinical studies, when properly conducted, provide the strongest empirical evidence of the efficacy and safety of a clinical intervention. 2. Observational studies, unlike clinical trials, do not experimentally create the contrasts between study groups. They may provide a tool to study phenomena when it would be technically difficult or even unethical to assign treatment to a sufficient number of patients to resolve the question of interest. It therefore becomes necessary to observe the effects of an intervention, such as cigarette smoking, under conditions of "natural treatment," to assess how the exposure to risk factors influences the probability of disease. Observational studies provide weaker evidence regarding causality than do clinical studies, in part because of the potential for large confounding biases. Nevertheless, observational studies may provide valuable evidence for hypotheses that can, in some cases, then be tested in clinical studies. The following types of observational studies are often conducted. 9 Cohort studies are prospective investigations that follow a group or cohort of individuals who are initially free of the outcome of interest. The studies are
865 conducted for a sufficiently long period for some individuals of the cohort to develop the outcome of interest. These individuals are then compared with those who did not do so. Retrospective cohort studies are weaker than prospective studies because they do not allow the definition of the study population and the outcomes before the events occur. As in all retrospective studies, recall bias is a major concern. However, retrospective cohort studies may be appropriate if the investigators are blinded to study outcomes when formulating the hypotheses. 9 Case-control studies are retrospective studies that match individuals who have had an event with those who have not. To avoid selection bias, cases and controls are drawn from the same or similar populations matched to specific characteristics such as age, gender, or disease duration. 9 Nested case-control studies draw cases and controls from a cohort study. Nested case-control studies offer an advantage compared with case-control studies because the controls are selected from subjects who were at risk at the same period during which the cases arose in the cohort. 9 Cross-sectional (prevalence) studies are descriptive studies that assess the characteristics of a sample of subjects at a given time point. They lack information on the relationship between timing of exposure and outcome.
IV. GENERAL REGULATORY REQUIREMENTS Prior to the introduction of a new drug substance into humans, the U.S. Food and Drug Administration requires the submission of an investigational new drug application (IND) as an essential means to protect the participants in clinical studies. In addition, each study participant is required to provide informed consent. The I N D Rewrite of 1987 underlined the importance of the Investigator's Brochure as a significant part of the application. The F D A requires that the brochure provide the following information for clinical investigational personnel: 9 A brief description of the drug substance and formulation, including the structural formula, if known 9 A summary of the pharmacologic and toxicologic effects of the drug in animals and, to the extent known, in humans ~ A summary of the pharmacokinetics and biologic disposition of the drug in animals and, if known, in humans
866
9 A summary of information on safety and effectiveness in humans obtained from clinical studies, including foreign studies, with appended reprints of published reports of such studies 9 A description of possible risks and side effects to be anticipated on the basis of prior experience with the drug under investigation or with related drugs and of precautions and special monitoring to be done as part of the investigational use of the drug The IND Rewrite of 1987 also requires the following information for clinical trials: 9 A statement of the objectives and purpose of the study Q.ualifications, based on training and experience, of investigational personnel 9 Information on the facilities of study conduct 9 Criteria for the selection or exclusion of study participants 9 The estimated number of study participants 9 A description of the study design, including the kind of control group to be used, if any, and a description of the methods to be used to minimize bias on the part of study participants, investigators, and in data analysis 9 The method for determining the dose(s) to be administered, the planned maximal dosage, and the duration of drug exposure 9 A description of the observations and measurements to be made to fulfill the objectives of the study 9 A description of clinical procedures, laboratory tests, or other measures to be taken to monitor the effects of the drug in humans and to minimize the risk
9
Clinical protocols, informed consent forms, and some study-related materials must be submitted to and approved by an institutional review board (IRB), a committee formally charged to review, to approve the initiation of, and to conduct periodic review of biomedical research involving human subjects. The primary purpose of such reviews is to ensure the protection of the rights and welfare of study participants. Standardization of IRBs and informed consent requirements was established in 1991 for all federal agencies that regulate clinical research. Institutional review boards are responsible for the initial and continuing review and approval of all studies and for ensuring that the rights and welfare of the study participants are protected. Before entering a study, an individual must sign an informed consent form, which states that the person's participation in the trial is voluntary and that the person may refuse to participate, or withdraw from the trial at any time without penalty. The consent form must also explain the study in language the person can understand and must include a statement on reasonably expected benefits and reasonably foreseeable risks or inconveniences to the study participant. Each study site
HOLINKA AND PICKAR
must obtain approval from an IRB, which may be either local or, as in multicenter trials, central (responsible for multiple sites). Clinical studies of new substances are conducted only after an extensive preclinical program has been completed, including thorough testing of the drug in animals. Clinical studies are generally performed in three sequential phases, as follows, although there may be some overlap. Phase I studies introduce the drug for the first time under rigorously monitored conditions into humans. Phase I studies are intended to determine the side effects of the drug at increasing dose levels in humans, to evaluate its metabolism and pharmacologic action, and, if possible, to gain early evidence of effectiveness. Phase I studies are also expected to yield sufficient information about the drug's pharmacokinetics and its pharmacologic effects to allow for the design of phase II studies. Depending on the drug, phase I studies may require between 20 and 80 healthy volunteers or patients in the disease category for which the drug is intended. Phase II studies are early trials of the safety and effectiveness of the investigational drug at different dose levels. The studies are well-controlled, closely monitored, and conducted in a relatively small number of volunteers, usually involving several hundred patients. Phase II1 studies are performed after preliminary evidence of a drug's safety and effectiveness has been obtained in phase II. They are generally large and, based on clinical and statistical requirements, may include several thousand participants. These studies are conducted over an appropriate period (up to several years, depending on the drug under investigation). Phase 1II studies are scientifically rigorous and expected to provide sufficient information to evaluate the indication sought for the drug including its overall risk/ benefit relationship. They are further expected to provide an adequate basis for extrapolating the results to the general population. Phase IV studies are conducted after marketing approval of a product. They may be required by regulatory agencies or agreed to by a sponsor. Phase IV studies are intended to yield additional information about a drug's safety, efficacy, or optimal use. (Phase IIIB studies are similar to phase IV studies but are started prior to marketing approval.) The entire body of data from phase I to phase 111 studies, together with data from preclinical studies and other available information about the drug, such as that from scientific publications, is then submitted in the new drug application (NDA) for approval of the targeted indication. The clinical data on safety and efficacy, their analysis, and the discussion of their clinical relevance make up a major part of the NDA. Also included in the NDA is detailed information on the manufacturing of the drug, as well as the proposed labeling.
CHAPTER 62 Clinical Trials in Postmenopausal Hormone Therapy
V. SPECIFIC REQUIREMENTS FOR HORMONE THERAPY STUDIES
A. Vasomotor Symptoms/Vaginal Atrophy Studies The most recent FDA draft guidance for clinical studies of estrogen and estrogen/progestin drug products to treat vasomotor symptoms and vulvar and vaginal atrophy symptoms define postmenopausal status as at least 12 months of spontaneous amenorrhea or 6 months of spontaneous amenorrhea with serum follicle-stimulating hormone (FSH) levels of more than 40 mlU/mL, or 6 weeks postsurgical bilateral oophorectomy with or without hysterectomy (5). Apart from requirements to ensure postmenopausal status, the guidance specifies inclusion criteria: Trials in support of the indication for treatment of moderate to severe vasomotor symptoms must be conducted in highly symptomatic women who experience at least 7 to 8 moderate to severe hot flushes a day, or 50 to 60 flushes per week at baseline. Both moderate and severe flushes are defined as characterized by a sensation of heat and sweating but distinguished from each other in that moderate flushes do not interfere with a woman's activities, whereas severe flushes cause cessation of activity. (Mild flushes are defined as a sensation of heat not associated with sweating and without adverse effects on activity.) Alternatively, quantitative thermography, measuring objective endpoints as illustrated in Fig. 62.2, is acceptable to evaluate efficacy (6). As illustrated in the figure, quantifiable physiologic changes in skin conductance, skin temperature, and tympanic temperature occur at specific time intervals following the subjective experience of a hot flush. For the indication for treatment of moderate to severe symptoms of vulvar and vaginal atrophy, women may be
Onset Subjective Flash mho x t 0-"
5.51 0.5t 33.4
C Skin Temperature
29.4
H
27.4
G
_.K\ TympanicTemperature ..............
~C 36,3 C 35,9 I
0
4
!
8
I
I
12
t6
J
1
............. 1 ......
I
20 24 28 Time (min)
enrolled who have self-identified at least one moderate to severe symptom that is the most bothersome, have no more than 5% superficial cells on a vaginal smear, and have a vaginal pH greater than 5.0. The self-assessed symptoms include vaginal dryness, vaginal or vulvar irritation/itching, dysuria, vaginal pain associated with sexual activity, and vaginal bleeding associated with sexual activity. To ensure unbiased efficacy assessments, the guidance recommends that volunteers for a clinical study for these indications must not have taken estrogen or estrogen/progestin drug products for specific periods prior to baseline evaluation. The washout periods should be at least 1 week for vaginal hormonal products, at least 4 weeks for transdermal products, 8 weeks for oral products, 8 weeks for intrauterine progestin therapy, 3 months or longer for progestin implants and estrogen-alone injectable drug therapy, and 6 months or longer for estrogen pellet or progestin injectable therapy. To ensure safety, the guidance mandates that women older than 40 years have documentation of a negative mammogram, obtained at screening or within 3 months (for drug products containing estrogen plus progestin) or 9 months (for estrogen alone) of study enrollment, and have normal clinical breast examination results. Volunteers with a uterus must have a biopsy performed at screening. Findings of endometrial cancer or hyperplasia disqualify a volunteer from study participation. The study should be conducted over a period of 12 weeks, and primary efficacy analysis should show clinically and statistically significant reductions in both frequency and severity of hot flushes when compared with placebo within 4 weeks of therapy initiation. This effect should be maintained throughout the 12 weeks of therapy. The rapid decrease in vasomotor symptoms following treatment in response to each of three doses of conjugated estrogens/medroxyprogesterone acetate (CE/MPA) and the considerable placebo effect are illustrated in Fig. 62.3 (7). To document efficacy in the treatment of vulvar and vaginal atrophy, statistically significant improvement compared with placebo from baseline to week 12 of treatment should be shown in all three of the following parameters: the vaginal maturation index, that is, a decrease of immature (parabasal) vaginal cells and an increase in mature (superficial) vaginal cells; lowering of the vaginal pH; and the moderate to severe symptom identified by the women as most bothersome. Fig. 62.4 illustrates the decrease in parabasal cells and the increase in superficial cells following H T (8).
B. Endometrial Hyperplasia Studies
,
867
,!
1
I
32
36
40
FIGURE 62.2 Objective endpoints and their time course for the measurement of the hot flush. (From ref. 6, with permission.)
1. BACKGROUND
The association between endogenous estrogen and endometrial hyperplasia and the possible progression from hyperplasia to cancer were first recognized nearly 60 years
868
FIGURE 62.3 A double-blind, prospective, randomized parallel group study illustrating the therapeutic responses of hot flushes to several doses of estrogen/ progestin. Typically, the mean numbers of daily flushes declined rapidly during the first 4 weeks. Note the substantial response to the lowest hormonal dose and the marked placebo effect. (From ref. 7, with permission.)
ago (9). This association has since been well documented with regard to estrogen therapy. A meta-analysis found the risk of endometrial cancer to be proportional to duration of estrogen use. It remained elevated for 5 or more years after discontinuation of unopposed estrogen therapy. The risk of endometrial cancer death was also elevated (although not statistically significant) among unopposed estrogen users (10). Currently, endometrial hyperplasia is the only acceptable surrogate endpoint for endometrial cancer in clinical trials designed to evaluate the safety of l i T and to obtain marketing approval, as specified by regulatory authorities in the United States and in Europe. Other endometrial histology characteristics or data obtained by ultrasonography are not considered appropriate efficacy endpoints (5,11). However, the natural history of endometrial hyperplasia remains to be elucidated. The progression from hyperplasia to cancer was observed only in a relatively small percentage of women, and predominantly in women showing atypical hyperplasia (12). Thus, the presence ofhyperplasia cannot be considered a fully specific marker for adenocarcinoma. Moreover, the histologic diagnosis of hyperplasia is subject to a high degree of subjectivity (13). The extent of this variability was specifically addressed in a study designed to examine the reproducibility among five expert gynecologic pathologists to distinguish hyperplastic from nonhyperplastic endometrium (14). The readers were given 200 endometrial tissue slides
HOLINKA AND PICKAR
FIGURE 62.4 Data from a clinical study using vaginal cytology as an objective endpoint for estrogen action. Parabasal (immature) and superficial (mature) cells are expressed as percentage of the total vaginal epithelial cell population examined. Percentages during the early (EF) and late (LF) follicular phases of the regular menstrual cycle are presented for comparison. Under the action of estrogen there is a substantial shift from immature to mature cells both following hormone therapy (right columns) and during the menstrual cycle (left columns). No such shift is observed in the placebo group (middle column). (From ref. 8, with permission.)
selected from a large pool of archival slides from previous clinical HT studies. The slides represented approximately 50% hyperplastic and 50% nonhyperplastic tissues by previous diagnosis. The readers were blinded to this ratio. At individual reading, with each reader blinded to the diagnoses of the others, there was 55% concordance on the presence or absence of hyperplasia among all five readers. When a single divergent opinion was discarded, this percentage increased to 77%. The intraobserver concordance ranged from 47% to 84%. Among 15 cases where all five pathologists agreed on the presence of hyperplasia, there was agreement on the type of hyperplasia in only three cases. Moreover, the pathologists identified 16 of 200 samples in which no consensus could be achieved, even after simultaneous review at a multiheaded microscope and extensive discussion (14). Given the limitation in reproducibility of histologic diagnoses, further research to identify specific early biologic markers of endometrial cancer is needed. Historically, scientific knowledge in any area of biology, including reproductive science, has advanced in step with the development of new technologies (15). One promising approach applying new technologies to endometrial research involves clonal analysis
CHAPTER 62 Clinical Trials in Postmenopausal Hormone Therapy and computerized morphometry. Based on the observation that some endometrial hyperplasias, like cancers, are phenotypically monoclonal (16), a study was designed to combine clonal analysis with subjective histologic diagnoses and computerized morphometric analysis. Consistent with the data described earlier (14), the study found considerable variability in histologic diagnoses among the four pathologists, but a high degree of correlation was established between the diagnosis of atypical hyperplasia and monoclonal growth. In contrast, no such correlation was observed with nonatypical hyperplasias, which tended to be as likely to be monoclonal as polyclonal. Morphometric analysis of histologic sections precisely identified monoclonal precancers and thus may serve to classify accurately those lesions judged by pathologists as indeterminate--that is, as nonatypical hyperplasias. Although the authors note that the molecular assays for clonal analysis are impractical for clinical use, their approach encourages further research along similar lines to identify specific biologic markers for endometrial cancer to be used in clinical research and medical practice (17,18). 2. INCIDENCE OF HYPERPLASIA RELATED TO ESTROGEN THERAPY AND ITS PREVENTION BY ADJUNCTIVE PROGESTOGEN
The increased incidence of hyperplasia related to unopposed estrogen has been widely documented, and there is convincing evidence that adjunctive progestogen prevents estrogen-induced hyperplasia. Hyperplasia rates observed in large randomized clinical trials containing both unopposed estrogen and adjunctive progestogen treatment arms are summarized in Table 62.1 (19-23). As discussed earlier, endometrial hyperplasia is a relatively nonspecific surrogate marker for subsequent endometrial cancer. With regard to cancer itself as an endpoint, a large body of observational data exists describing the relationship between unopposed estrogen and increased endometrial cancer risk and the protection afforded by adjunctive progestogen (10,24-28). In terms of study design, this represents a relevant example of the value of observational studies in a situation where randomized clinical trials would be inappropriate. 3. HYPERPLASIA STUDY DESIGN
The draft FDA guidance recommends a 12-month, randomized, double-blind dose-ranging phase 11I clinical trial to document endometrial protection by the estrogen/progestin combination. The lowest effective dose of progestogen to support endometrial safety should be clearly identified by including an ineffective lower dose in the study. The lowest effective dose of estrogen for the indication should also be demonstrated. The revised guidance stipulates more stringent procedures for histologic evaluations of endometrium (5) compared
869 with those of the earlier guidance (29). For efficacy evaluation, three independent expert pathologists, blinded to treatment and to each other's readings, are required to supply the diagnosis, contrasting with earlier requirements of two pathologists plus a third pathologist only in cases of arbitration. The efficacy slides distributed to each of the pathologists (for the end-of-study review) should include a randomly selected set of 10% normal screening slides. For efficacy evaluation the concurrence of two of the three pathologists is accepted as the final diagnosis. If there is no agreement among the three pathologists, the most severe pathologic diagnosis will be the final diagnosis. It is now further recommended that the initial screening slides or any additional endometrial safety biopsy during the study, such as for uterine bleeding, be read by one of the three blinded pathologists. It is recommended that the study demonstrates a hyperplasia incidence of 1% or less with an upper bound of the one-sided 95% confidence interval that does not exceed 4%. The rates of atypical hyperplasia and cancer are considered important additional factors in determining approvability.
C. Osteoporosis Studies 1. BACKGROUND
Osteoporosis and its associated morbidity and mortality represent a major global health issue. The World Health Organization has estimated that worldwide 30% of women 50 years old or older suffer from osteoporosis, using the definition of bone mineral density (BMD) of greater than 2.5 standard deviations below the mean for young, healthy adult women (30). In the United States approximately 30 million women age 50 years or older have low bone mass, and of these, 8 million w e r e afflicted by osteoporosis in 2002, according to estimates by the National Osteoporosis Foundation. These numbers are expected to increase to 9 million osteoporotic women by 2010 and 10 million by 2020 (31,32). Apart from the physical and psychologic burden to the afflicted individual, the social and economic consequences of osteoporotic fractures are enormous, and related health care costs will steadily and markedly increase in the 21st century (33). Therefore, prevention and treatment of osteoporosis are of major importance to the actually or potentially afflicted individual and to public health in general. It has been known for several decades that H T prevents bone loss when started during the early postmenopausal period and that it is effective even after bone loss has occurred. This is illustrated in Fig. 62.5 by results from a classical study measuring metacarpal bone mineral content. Estrogen therapy entirely prevented bone loss when started around the period of menopause, reverted bone loss when started within 2 years thereafter, and prevented further erosion when started 5 years after the menopause (34). The conclusions of this
870
HOLINKA AND PICKAR TABLE 62.1.
Incidence of Endometrial Hyperplasia Following Therapy with Unopposed Estrogen, and Its Prevention by Adjunctive Progestogen a
Regimenb Estrogen/progestin (mg) CE/MPA
Placebo CE/MPA CE/P EJNGM
E2/NETA
Placebo CE/MPA
Follow-up (years) 0.625/0 0.625/2.5 cont 0.625/5.0 cont 0.625/5.0 cyclic 0.625/10.0 cyclic 0.625/0 0.625/2.5 cont 0.625/5.0 cont 0.625/5.0 cyclic 0.625/10.0 cyclic 3 0.625/0 0.625/2,5 cont 0.625/10.0 cyclic 0.625/200 1/0 1/0.03 cyclo 1/0.09 cyclo 1/0.18 cyclo 1/0 1/0.1 1/0.25 1/0.5
0.5
1
1
1
1 0.3/0 0.3/1.5 0.45/0 0.45/1.5 0.45/2.5 0.625/0 0.625/2.5
Hyperplasia/biopsies
Hyperplasia incidence(%)
21/298 1/295 0/291 1/293 0/292 57/283 2/279 0/271 3/277
7 <1 0 <1 0 20 <1 0 1
0/272
0
2/119 74/119 1/120 6/118 1/120 74/265 16/260 0/242 0/242 36/246 2/249 1/251 1/241 0/261 1/269 1/272 9/279 1/272 0/273 20/249 0/278
1.7
Ref. Woodruff, 1994 (19)
PEPI, 1996 (20)c
62.2
<1 5.1 5.0 28 6 0 0 14.6 <1 <1 <1 0 0.4 0.4 3.2 0.4 0 8 0
Corson, 1999 (21)
Kurman, 2000 (22)
Pickar, 2001 (23)
"Only studies including an unopposed estrogen arm are presented. bE2, estradiol; CE, conjugated estrogen; P, progesterone; MPA, medroxyprogesterone acetate; NETA, norethindrone acetate; NGM, norgestimate; cont, continuous combined estrogen/progestin; cyclic, adjunctive MPA for 14 days each 28-day cycle; cyclo; cyclophasic (continuous E2 plus phasic adjunctive NGM: 3 days on, 3 days off). cThe occurrence of hyperplasia was distributed evenly across the 3 yearsNthat is, 25, 29 and 20 cases in year 1, 2 and 3, respectively.
study have since been confirmed by a large body of clinical data. Meta-analyses of randomized trials have suggested that H T prevents vertebral (35) as well as nonvertebral fractures (36). The Women's Health Initiative study, a large randomized, placebo-controlled clinical trial, reported significant fracture reductions at multiple sites. Daily therapy with 0.625 mg CE plus 2.5 mg MPA (mean follow-up = 5.6 years) reduced the risk for hip fracture by 33% with a hazard ratio (HR) of 0.67 (95% nominal confidence interval [nCI], 0.47-0.96); the risk for vertebral fracture by 35%, HR 0.65 (95% nCI, 0.46-0.92); and the risk for total fracture by 24%, HR 0.76 (95% nCI, 0.69-0.83) (37). Daily therapy with 0.625 mg CE alone (mean follow-up = 6.8 years)
reduced the risk for hip fracture by 39%, HR 0.61 (95% nCI, 0.41-0.91); the risk for vertebral fracture by 38%, HR 0.62 (95% nCI, 0.42-0.93); and the risk for total fracture by 30%, H e 0.70 (95% nCI, 0.63-0.79) (38). More recently, two randomized, double-blind, placebocontrolled clinical trials found bone-sparing effects at doses markedly lower than those conventionally prescribed for women at risk of osteoporosis. Specifically, therapy with 0.3 mg and 0.45 mg CE alone or in combination with 1.5 mg MPA for 2 years resulted in significant increases from baseline in spine and hip bone mineral density (BMD) (39). The other trial evaluated effects on bone of ultra-low doses of transdermal 1713-estradiol, using patches delivering approximately 0.014 mg/day. The study found
871
CHAPTER 62 Clinical Trials in Postmenopausal Hormone Therapy
44
A
E
E ol
a2
E tli
a0
o ,,4D
tli
38
O III 36 O
34 0
2
a
6
8
t0
t2
~4
16
Years FIGURE 62.5 Data from a clinical study measuring metacarpal bone mineral as an objective endpoint of estrogen action. In this classical study, estrogen therapy prevented bone loss when started shortly after oophorectomy and had substantial bone-sparing effects when initiated 3 or 6 years after oophorectomy. (From ref. 34, with permission.)
significant differences from placebo in lumbar spine and total hip BMD 1 and 2 years after start of therapy. It also found significant increases from baseline in the lumbar spine BMD at 1 and 2 years, and in total hip BMD at year I but not year 2 (40). As apparent from these data, the bone-sparing effects of HT, including low-dose therapy, have been well documented at multiple skeletal sites using both BMD and fracture rates as endpoints. 2. OSTEOPOROSIS STUDY DESIGN
Prevention of postmenopausal osteoporosis is an indication for HT. However, when prescribing solely for the prevention of postmenopausal osteoporosis, therapy should only be considered for women at significant risk of osteoporosis and for whom nonestrogen medications are not considered to be appropriate. Use of bisphosphonates represents a major option for alternative therapy. But bisphosphonates may be associated with gastrointestinal disorders, such as dysphagia, esophagitis, and esophageal or gastric ulcers, and there is limited data regarding their long-term safety. Therefore, clinical trials evaluating bone-sparing effects of new H T compounds, such as selective estrogen receptor modulators (SERMs) and estrogen-SERM combinations, or of
approved compounds at new dosage levels, different combinations, or different modes of delivery remain well justified. The design of such studies is discussed in this section. The last FDA draft guidance related to agents used in the prevention or treatment ofpostmenopausal osteoporosis was issued in 1994. More recently the agency has indicated its intention to issue an updated draft guidance. Currently, the primary efficacy endpoints for phase III studies for the treatment of osteoporosis indication is the incidence of new fractures, whereas BMD is the primary endpoint for prevention studies. Studies for treatment of osteoporosis were to be initiated and interim data available prior to initiation of prevention studies for nonestrogen products. Because epidemiologic studies have demonstrated that estrogen therapy reduces the risk of vertebral and nonvertebral fractures, fracture evaluation was not required for H T products prior to studies for prevention (41). Well-controlled osteoporosis prevention studies are randomized, double-blind, placebo-controlled trials including multiple doses to enable the identification of the minimal effective dose. The minimal study duration is 2 years; the sample size must be sufficient to adequately demonstrate the safety and efficacy of the investigational drug. Studies for the treatment of osteoporosis in patients with a history of osteoporotic fractures or T-score osteoporosis should be double-blind, randomized, with either placebo or active drug as the control group. In studies using an active comparator, sample size calculations must be such that a sufficient number of subjects are enrolled to detect meaningful differences between the test drug and the active control--if such differences exist. The number of subjects required to test the efficacy of H T in the treatment of osteoporosis is likely to be greater than that required for prevention studies.
D. Safety Assessments Clinical trials in support of marketing approval for a new drug, or of approval for an additional indication for a marketed drug, must thoroughly document the safety of the drug in addition to its efficacy. Standard safety assessments in H T studies include physical and gynecologic examinations, mammography, Papanicolaou (Pap) smear, and laboratory tests (hematology, chemistry, and urinalysis). The FDA guidance specifically recommends measurement of antithrombin III, factor V Leiden, protein C, and protein S. Women with a uterus are closely monitored for endometrial hyperplasia. Prolonged uterine bleeding in women receiving unopposed estrogen may be an indication of endometrial hyperplasia (42). In cases of severe or prolonged bleeding, a biopsy may be indicated. Additionally, special metabolic parameters, such as lipid and lipoprotein profiles, coagulation factors, and carbohydrate
872 metabolism, are usually evaluated in H T studies because of favorable or potentially unfavorable hormone effects on these parameters. All treatment-emergent adverse events, whether categorized by the investigator as related to the investigational drug or not, are recorded throughout the study. An adverse event is considered serious if it is fatal or life-threatening, requires or prolongs inpatient hospitalization, or causes persistent or significant disability/incapacity or a congenital anomaly or birth defect (43).
VI. C H A N G I N G VIEWS: THE CASE OF C A R D I O P R O T E C T I O N A number of large randomized clinical trials (RCTs) conducted in the recent past in populations of women with average ages more than 10 years past the average age of menopause changed the landscape of HT, especially with regard to our understanding of the effects of estrogen on the cardiovascular system. These trials include the Heart and Estrogen/Progestin Replacement Study (HERS, HERS II) (44,45), the Estrogen Replacement and Atherosclerosis Study (ERAS) (46), and the Women's Health Initiative (WHI) study (47). Prior to these RCTs, the majority of observational studies had indicated substantial cardioprotective benefits of estrogen therapy (48). This conclusion was based on results from population-based or community-based case-control studies or prospective cohort studies. Although less information was available on the effects of adjunctive progestin, it was generally believed that the addition of progestin to estrogen did not attenuate cardioprotection (49). It is important to note that the majority of these observational studies, in contrast to the RCTs, were conducted in women who elected therapy when entering menopause~that is, in a relatively young age group of women. The primary outcome of the H T arm of W H I was coronary heart disease (nonfatal myocardial infarction and coronary heart disease [CHD] death), with invasive breast cancer, a secondary outcome, considered as the primary adverse outcome. The initial report on the estrogen plus progestin arm of W H I indicated an hazard ratio HR for coronary heart disease of 1.29 (95% nCI, 1.02-1.63) (47). Thereafter, the final report showed an HR of 1.24 (95% nCI, 1.00-1.54) (50). The magnitude of the HR decreased with decreasing years since menopause. For women whose menopause had begun less than 10 years previously, the HR was 0.89 (50). The HR for invasive breast cancer was 1.24 (95% nCI, 1.01 - 1.54) (51). In WHI, estrogen-only therapy during the mean of 6.8 years of follow-up had no adverse effects on C H D risk. The HR was 0.91 (95% nCI, 0.75-1.12), and there was a trend toward fewer cardiovascular events when therapy was initiated at an earlier age. Estrogen-only treatment did not
HOLINKA AND PICKAR
increase the risk of breast cancer. On the contrary, invasive breast cancer rates were lower in the CE group, an effect that barely missed statistical significance, with an HR of 0.77 (95% nCI, 0.59-1.01) (38). The W H I data, and the manner in which it was initially released, had an enormous impact on postmenopausal women, health care providers, and the public in general. The study was of unprecedented size and offered the advantages of a randomized controlled trial. Nevertheless, its relevance to younger postmenopausal women has been widely debated and often questioned. It was noted that women are most likely to elect H T as they enter menopause. In marked contrast, the W H I population was on average over a decade older, with two-thirds of the women being 60 years of age or older. By today's standards, the dose of l i T used in W H I was excessive considering the age of the patients. It has been emphasized that the advanced age of the women in W H I may be especially significant with regard to coronary heart disease inasmuch as cardiovascular lesions normally progress with age and thus H T may not be effective in that age group, and in fact may exert adverse effects when initiated in later years. It has also been argued that W H I was not truly a primary intervention study (52) because the study participants had greater risk factors for cardiovascular disease than did the women in most observational studies--based on their more advanced age, their high mean body mass index (BMI) of 28.5 kg/m 2 (34% had a BMI greater than 30), and their smoking habits (nearly 50% of the women were past or current smokers). It has been emphasized that H T should not be considered a cardiovascular drug but that the question still needs to be addressed whether HT, when used for the treatment of menopausal symptoms, provides any cardiovascular benefits in younger postmenopausal women (52).
VII. FUTURE DIRECTIONS Hormone therapy remains the principal option for the treatment of menopause-associated vasomotor symptoms and symptoms of vulvar and vaginal atrophy and offers a major option for the prevention ofpostmenopausal osteoporosis (53). With the efficacy of very low estrogen doses well established, further research efforts at lowering the doses may yield limited results. In contrast, further work on progestins to minimize overall exposure and to identify new molecular entities appears promising. Low-dose progestins that have so far been shown to be effective in continuous combination with estrogen may effectively confer endometrial protection and yield acceptable bleeding patterns when given intermittently at the same dose levels, thus reducing the overall progestin exposure. New selective estrogen receptor modulators (SERMs) have been successfully developed, and some have the potential to be combined with estrogen to provide endometrial
CHAPTER 62 Clinical Trials in Postmenopausal Hormone Therapy protection. Estetrol, a h o r m o n e that may be considered a natural S E R M , is currently under investigation (54,55). Estetrol is a unique h u m a n pregnancy h o r m o n e produced by the fetal liver early during pregnancy and found in the mother at exponentially growing concentrations during gestation, with fetal levels about 10 times higher than maternal levels. A forefront research question concerns the relationship between age at which H T is initiated in relation to its most optimal risk/benefit profile. Data from the W H I study indicate a trend toward fewer cardiovascular events w h e n H T is initiated in younger postmenopausal w o m e n (38). It has been suggested that age differences between the populations of observational studies and that of the W H I may explain some of the differences in cardiovascular results. Differences in risk/benefit profiles of different estrogens and progestogens, as well as different doses and routes of administration, also merit experimental attention (56). T h e need for research in different racial and ethnic groups has been emphasized by regulatory authorities in the U n i t e d States and abroad (57,58). Recent studies docum e n t e d that menopausal s y m p t o m s and lipid/lipoprotein profiles differ substantially a m o n g ethnic groups of Asian postmenopausal w o m e n ( 5 9 - 6 2 ) , as does responsiveness of lipids to H T (62) These findings and further clinical trials may allow individualization of H T in different ethnic groups.
References 1. U.S. Food and Drug Administration. Investigational new drug application." definitions and interpretations. 21CFR312.3(b). Washington, DC: U.S. Government Printing Office, April 1, 2004. 2. Holinka CE Design and conduct of clinical trials in hormone replacement therapy. Ann N YAcad Sci 2001;943:89-108. 3. Barton R. Which clinical studies provide the best evidence? BMJ 2000;321:255-256. 4. Coope J, Thomson JM, Poller L. Effects of'natural oestrogen' replacement therapy on menopausal symptoms and blood clotting. BMJ 1975;4:139-143. 5. U.S. Food and Drug Administration. Center for Drug Evaluation and Research (CDER). Estrogen and estrogen/progestin drugproducts to treat vasomotor symptoms and vulvar and vaginal atrophy symptoms-recommendations for clinical evaluation (draft guidance). Washington, DC: U.S. Government Printing Office, January 2003. 6. Meldrum DR. The pathophysiology of postmenopausal symptoms. Semin Reprod Endocrino11983;1:11-17. 7. Utian WH, Shoupe D, Bachmann G, Pinkerton JV, Pickar JH. Relief of vasomotor symptoms and vaginal atrophy with lower doses of conjugated estrogens and medroxyprogesteroneacetate. Fertil Steri12001;75: 1065-1079. 8. Laufer LR, DeFazio JL, Lu JKH, et al. Estrogen replacement therapy by transdermal estradiol administration. Am J Obstet Gynecol 1983; 146:533-540. 9. Gusberg SB. Precursors of corpus carcinoma: estrogens and adenomatous hyperplasia. Am J Obstet Gyneco11947;54:905- 927.
873 10. Grady D, Gebretsadik T, Kerlikowske K, Ernster V, Petitti D. Hormone replacement therapy and endometrial cancer risk: a meta-analysis. Obstet Gyneco11995;85:304-313. 11. European Medicines Agency. Committee for Medicinal Products for Human Use (CHMP). Guideline on clinical investigation of medicinal productsfor the treatment of hormone replacement therapy. (Draft). London: European Medicines Agency,January 2005. 12. Kurman RJ, Kaminski PF, Norris HJ. The behavior of endometrial hyperplasia: a long-term study of"untreated" hyperplasia in 170 patients. Cancer 1985;56:403-412. 13. Mutter GL, BaakJPA, Crum CP, et al. Endometrial precancer diagnosis by histopathology, clonal analysis, and computerized morphometry. J Patho12000;190:462-469. 14. Wright TC, Holinka CF, Ferenczy A, et al. Estradiol-induced hyperplasia in endometrial biopsies from women on hormone replacement therapy. Am J Surg Patho12002;26:1269-1275. 15. Holinka CE Female reproductive science in historical perspective.Ann NYAcad Sci 1997;828:1-16. 16. Jovanovic AS, Boynton KA, Mutter GL. Uteri of women with endometrial carcinoma contain a histopathological spectrum of monoclonal putative precancers, some with microsatellite instability. Cancer Res 1996;56:1917-1921. 17. Mutter G L. Histopathology of genetically defined endometrial precancers. IntJ Gynecol Patho12000;19:301 - 309. 18. Mutter GL. Diagnosis of premalignant endometrial disease. J Clin Patho12002;5 5:326- 331. 19. WoodruffJD, Pickar JH, for the Menopause Study Group. Incidence of endometrial hyperplasia in postmenopausal women taking conjugated estrogens (Premarin) with medroxyprogesterone acetate or conjugated estrogens alone. Am J Obstet Gyneco11994;170:1213-1223. 20. The Writing Group for the PEPI Trial. Effects of hormone replacement therapy on endometrial histology in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) trial. JAMA 1996;275:370-375. 21. Corson SL, Richart RM, Caubel P, Lim E Effect of a unique constantestrogen, pulsed-progestin hormone replacement therapy containing 1713-estradiol and norgestimate on endometrial histology. Int J Fertil 1999;44:279-285. 22. Kurman RJ, F61ix JC, Archer DF, et al. Norethindrone acetate and estradiol-induced endometrial hyperplasia. Obstet Gynecol 2000;96: 373-379. 23. Pickar JH, Yeh I-T, Wheeler JE, Cunnane MF, SperoffL. Endometrial effects of lower doses of conjugated equine estrogens and medroxyprogesterone actetate. Fertil Steri12001;76:25- 31. 24. Voigt LF, Weiss NS, Chu J, et al. Progestagen supplementation of exogenous oestrogens and risk of endometrial cancer. Lancet 1991;338: 274-247. 25. Persson I, Adami H-O, Bergkvist L, et al. Risk of endometrial cancer after treatment with oestrogens alone or in conjunction with progestogens: results of a prospective study. BMJ 1989;298:147-151. 26. Jick SS, Walker AM, Jick H. Estrogens, progesterone, and endometrial cancer. Epidemio11993;4:20-24. 27. Brinton LA, Hoover RN, and The Endometrial Cancer Collaborative Group. Estrogen replacement therapy and endometrial cancer risk: unresolved issues. Obstet Gyneco11993;81:265-271. 28. Archer DE The effect of the duration of progestin use on the occurrence of endometrial cancer in postmenopausal women. Menopause 2001;8:245-251. 29. U.S. Food and Drug Administration-- HRT Working Group. Guidance for clinical evaluation of combination estrogen/progestin-containing drug products usedfor hormone replacement therapy of postmenopausal women. Washington, DC: U.S. Government Printing Office, March 20, 1995. 30. WHO Study Group. Assessment of fracture risk and its application to screeningforpostmenopausal osteoporosis. W H O Technical Report Series 843. Geneva: WHO, 1994:5.
874 31. National Osteoporosis Foundation. America's bone health: the state of osteoporosis and low bone mass in our nation. Washington, DC: National Osteoporosis Foundation, 2002:1 - 55. 32. National Osteoporosis Foundation. America's bone health: the state of osteoporosis and low bone mass. http://www.nof.org/advocacy/ prevalence/index.htm (accessed August 22, 2005). 33. Johnell O. The socioeconomic burden of fractures: today and in the 21st century. Am J Med 1997; 103 (2A) :20S - 26S. 34. Lindsay R. Sex steroids in the pathogenesis and prevention of osteoporosis. In: Riggs BL, Melton LJ, eds. Osteoporosis. New York: Raven Press, 1988:333-358. 35. Torgerson DJ, Bell-Syer SEM. Hormone replacement therapy and prevention of vertebral fractures: a meta-analysis of randomised trials. BMC MusculoscelDisord 2001;2:7-10. 36. Torgerson DJ, BeU-Syer SEM. Hormone replacement therapy and prevention of nonvertebral fractures. A recta-analysis of randomized trials. JAMff 2001;285:2891-2897. 37. Cauley JA, Robbins J, Chen Z, et al. Effects of estrogen plus progestin on risk of fracture and bone mineral density. The Women's Health Initiative randomized trial. JAMA 2003;290:1729-1738. 38. The Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy. The Women's Health Initiative randomized controlled trial. JAMA 2004;291:1701 - 1712. 39. Lindsay R, Gallagher JC, Kleerekoper M, Pickar JH. Effect of lower doses of conjugated estrogens with and without medroxyprogesterone acetate on bone in early postmenopausal women. JAMA 2002;287: 2668-2676. 40. Ettinger B, Ensrud KE, Wallace R, et al. Effects of ultralow-dose transdermal estradiol on bone mineral density: a randomized clinical trial. Obstet Gyneco12004;104:443-451. 41. U.S. Food and Drug Administration. Division of Metabolic and Endocrine Drug Products. Guidelinesfor preclinical and clinical evaluation of agents used in the prevention or treatment of postmenopausal osteoporosis. U.S. Government Printing Office: April, 1994. 42. Pickar JH, Archer DF, for The Menopause Study Group. Is bleeding a predictor of endometrial hyperplasia in postmenopausal women receiving hormone replacement therapy? Am J Obstet Gynecol 1997; 177:1178-1183. 43. U.S. Food and Drug Administration. Center for Drug Evaluation and Research. Content and format of adverse reactions section of labeling for human prescription Drugs and biologics. Draft guidance. Washington, DC: U.S. Government Printing Office, May 2000. 44. HuUey S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMff 1998;280:605-613. 45. Grady D, Herrington D, Bittner V, et al. Cardiovascular disease outcomes during 6.8 years of hormone therapy. Heart and Estrogen/ Progestin Replacement Study follow-up (HERS II).JAMA 2002;288: 49-57. 46. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N EnglJ Med 2000;343:522 - 529. 47. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321- 333.
HOLINKA AND PICKAR
48. Grodstein F, Stampfer M. The epidemiology of coronary heart disease and estrogen replacement in postmenopausal women. Prog Cardiovasc Dis 1995;38:199-210. 49. Grodstein F, Stampfer MJ, Manson JE, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N E n g l J Med 1996;335:453-461. 50. Manson JE, Hsia J, Johnson KC, et al. Estrogen plus progestin and the risk of coronary heart disease. N E n g l J Med 2003;349:523-534. 51. Chlebowski, RT, Hendrix, SL, Langer, RD, et al. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women. The Women's Health Initiative randomized trial. JAMA 2003;289:3243- 3253. 52. Mikkola TS, Clarkson TB, Notelovitz M. Postmenopausal hormone therapy before and after the women's health initiative study: what consequences? Ann Med 2004;36:402-413. 53. U.S. Food and Drug Administration. Center for Drug Evaluation and Research. Labeling guidancefor noncontraceptive estrogen drug productsfor the treatment of vasomotor symptoms and vulvar and vaginal atrophy symptoms~prescribing informationfor health careproviders. Washington, DC: U.S. Government Printing Office, February 2004. 54. Coelingh Bennink HJT, Holinka CF, Bunschoten H. A new natural human estrogen/SERM. Climacteric 2002;5(suppl 1):46. 55. Coelingh Bennink HJT. Estetrol: a natural SERM. International Menopause Society. l lth World Congress on the Menopause. Buenos Aires, 2005. 56. Modena MG, Sismondi P, Mueck AO, Kutenn F, de Ligniere B, Verhaeghe J, Foidart M, Caufriez A, Genazzani AR. New evidence regarding hormone replacement therapy is urgently required. Transdermal postmenopausal hormone therapy differs from oral hormone therapy in risks and benefits. Maturitas, posted online June 15, 2005. 57. U.S. Food and Drug Administration. Center for Drug Evaluation and Research. Collection of race and ethnicity data in clinical trials. Draft guidance. Washington, DC: U.S. Government Printing Office, January 2003. 58. The European Agency for the Evaluation of Medicinal Products. Human Medicines Evaluation Unit. Ethnicfactors in the acceptability of foreign data. ICH Topic E 5. London: The European Agency for the Evaluation of Medicinal Products, September 1998. 59. Tan D, Haines CJ, Limpaphayom KK, Holinka CF, Ausmanas MK. Relief of vasomotor symptoms and vaginal atrophy with three doses of conjugated estrogens and medroxyprogesterone acetate in postmenopausal Asian women from 11 countries: The Pan-Asia menopause (PAM) study. Maturitas 2005;52:35- 51. 60. Haines CJ, Xing S-M, Park K-H, Holinka CF, Ausmanas MK. Prevalence of menopausal symptoms in different ethnic groups of Asian women and responsiveness to therapy with three doses of conjugated estrogens/medroxyprogesterone acetate: The Pan-Asia menopause (PAM) study. Maturitas 2005;52:264-276. 61. Limpaphayom KK, Darmasetiawan MS, Hussain RI, et al. Differential prevalence of quality-of-life categories (domains) in Asian women and changes after therapy with three doses of conjugated estrogens/ medroxyprogesterone acetate: The Pan-Asia Menopause (PAM) study. Climacteric 2006;9:204-214. 62. Taechakraichana N, Holinka CF, Haines CJ, et al. Distinct lipid/ lipoprotein profiles and hormonal responsiveness in nine ethnic groups of postmenopausal Asian women: The Pan-Asia Menopause (PAM) study. Climacteric 2007. in press.
2HAPTER
6;
The Future of Therapy and the Role of Hormone Therapy ROGERIO A.
LOBO
Departmentof Obstetrics & Gynecology, Columbia University College of Physicians and Surgeons, New York, NY 10032
In this final chapter of the third edition of Treatment of the Postmenopausa! Woman, the aim is to discuss where we are currently in the treatment of postmenopausal women and to look toward the future. Clearly this synthesis relies on all the previous chapters dealing with various therapies for specific conditions as well as the risks that postmenopausal women experience. As the age of menopause has remained fairly fixed and life expectancy has grown, women can expect to live 30 or more years after menopause. Ideally, these should be the best years, with an enhanced quality of life, free from diseases and with independence. Treatment per se may not be needed, particularly pharmacologic treatment. A greater understanding of emerging health risks after menopause and stressing lifestyle and wellness issues as well as survival from cancer can go a long way toward maintaining and enhancing health and quality of life. In this context, treatment implies a prescription for good nutrition, exercise, cessation of smoking, and so on. In other cases it may mean the addition of nutritional supplements such as calcium and vitamin D and possibly a trial of complementary therapies such as phytoestrogens and acupuncture. In other circumstances it may be necessary to intervene with antihypertensives, cholesterol-lowering medications, and hypoglycemics T R E A T M E N T OF T H E POSTMENOPAUSAL W O M A N
875
if dietary adjustments are not adequate. Nevertheless, notwithstanding the immense importance of these treatments noted, the rest of this chapter will focus on the issues surrounding hormonal types of treatment. The second edition of Treatment of the Postmenopausal Woman was published in 1999. Several randomized clinical trials (RCTs), but particularly the Women's Heath Initiative (WHI), turned the field on its head, and with the confusion and lack of clarity regarding these figures, it was decided to wait until now, until the "dust had settled" before embarking on this third edition. As the hormone trials of W H I were being implemented, I was asked by a reporter what I thought we would learn. At the time I, like others, was concerned about some aspects of the design and the immense costs involved, in that I felt we would not learn very much more than we had already known based on more than 20 years of data from observational trials. Although the RCT remains the gold standard in the clinical arena, as I have stated before in my preface to Section IX, Clinical Trial and Observational Data, a good study is a good study whether it is observational or an RCT, and the latter also has some limitations. After all, the observational data from several different cohorts were entirely consistent. At Copyright 9 2007 by Elsevier,Inc. All rights of reproduction in any form reserved.
876
ROGERIOA. LoBo
face value, however, I was wrong in that the results of W H I were opposite to conventional thinking (based on observational data) in a number of areas. But wait a minute: Except for two areas (at face value) there is remarkable consistency between the earlier observational data and the results of the RCTs (Table 63.1). Further, if one considers in more detail the characteristics of the patients involved in the observational studies compared with those in the trials, even these discrepancies can be resolved (Table 63.2). In several chapters in this volume, key differences between the observational data and the RCTs have been pointed out to explain the differences in the findings, namely in the areas of coronary heart disease (CHD) and cognition/dementia. Women in the RCTs were older, had more risk factors, were at least more than 10 years past the onset of menopause, and were asymptomatic. This population is clearly different from the typical younger symptomatic woman who needs hormone therapy (HT) for relief of menopausal symptoms. Abundant data are accumulating that treating younger women within a few years of the onset of menopause affords protective effects against C H D and dementia, unlike the findings in an older population. It appears that 10 or more years after menopause, in the absence of estrogen, progressive changes in the arterial system (accelerated atherosclerosis) and in the neuronal processes in the brain (loss of numbers and function) make these systems unresponsive to estrogen and potentially reacting in an adverse way when exposed acutely to standard doses of HT. These first few years after menopause, therefore, have been suggested to be a "window of opportunity" in which H T may be efficacious for symptoms and potentially beneficial for downstream risks of C H D and dementia/cognitive decline (although unproven). Recent data, including subgroup analyses from the existent RCTs, including W H I , would support these assertions, as will be reviewed later in this chapter.
TABLE 63.2
Baseline Characteristics: N H S vs. W i l l
Mean age or age range at enrollment (years) Smokers (past and current) BMI (mean) Aspirin users HT regimen
NHS .-c
WHI d
30-55
63
55%
49.9%
25.1 kg/m 2 43.9% Unopposed or sequential Predominant
Menopausal symptoms (flushing)
28.5 kg/m 2. 19.1% Continuous-combined Uncommon
*34.1% had BMI -> 30 kg/m2. "From res 13. bGrodstein F, et al. N E n g l J M e d 1996;335:453-461. cColditz GA, et al..dmJEpidemio11986;123:48-58. dWriting Group for the WHI Investigators.J.4MA 2002;288:321-333.
Recent analyses of the estrogen-only arm of W H I shows a significant protective effect in a global coronary score (1). A further analysis of the estrogen and progestogen arm of W H I shows a significant trend for reduced events in younger compared with older women; and in the Nurses' Health Study (NHS), the protective coronary effect was found to be confined only to those women who initiated treatment within 10 years of menopause (2). In a pooled analysis of 25 RCTs, women in the 50 to 59 years age group in these trials had a protective effect against coronary heart disease (CHD) (3) (Fig. 63.1). Reinforcing the notion of treating women during the "window of opportunity," a follow-up of women treated for 2 to 3 years in RCTs in Denmark shows that those women treated had significantly reduced cardiovascular (CV) mortality as well as all-cause mortality compared with women Study, Year
Odds ratio (fixed) 95% CI
(Reference)
TABLE 63.1 Consistency of the R C T with Observational Studies Agree Osteoporosis Breast cancer Colon cancer Uterine cancer VT and PE Stroke (dose) CHD* Alzheimer's*
X X X X X X
*Hypothesis: Age/menopause sensitive.
Disagree
Hall 1994 (24) Hall 1998 (21) Komulainen 1999 (8) Mosekilde 2000 (9) Nachtigalt 1979 (10) PEPi 1995 (26) Ravn t 999 (11 ) Samarus 1999 (t2) Utian 2001 (28)
t
9
i
l
.
.
.
.
.
.
="V:=:=:=:::=:i
.
3.03 [0.12, 75.28]
. . . . . 22,222 I2.Y...~ ................................... 0.1610.04,4.12] ......
777
7I7.Z.I
...........................
................ L S -
........2 . 2 2 Z 7 I
.................
......................................
2 2 2 7 2 L Z ] 2
....
................] 2 ] ] 2 . Z T T Z 7 7 7[ Z Z Z ] ...........................................................................
. . . . .. .
ToatW l HW t esetndorPWH(3)(2)CI 120022001)2000 (95%(13) ~
.
.
.
.
.
.
_
,
.
.
.
.
.
.
.
.
.
.
=
.
.
.
.
...................
.
*
3.03 [0.12, 75.06] 1.00 [0.06, 16.10] 0.33 [0.04, 3.19] 1.25 [0.06, 26.t01 0.33 [0.04, 8.12] 0.31 [0.01,8.291 0.05 [0.00, 1.16] 0.12 [0.00, 2.93] 0.87 [0.53, 1.41] I 0,56 [0.30, 1.03] 0.68 [0.48, 0.96]
0.001
0.01 0.1 Favors HT
1
10 100 Favors control
FIGURE 63.1 Pooled analysis from 23 randomized trials: Younger women (<10 years or <60 years). (From ref. 3, with permission.)
CHAPTER 63 The Future of Therapy and the Role of Hormone Therapy
assigned a placebo (4) (Fig. 63.2). Also, although the estrogen and progestogen trial in W H I and other RCTs, such as the Heart and Estrogen/Progestin Replacement Study (HERS) showed increased coronary events in the first year (early harm), this has not been shown in prospective trials in younger women (5). Although various meta-analyses of the observational trials showed a protective effect on Alzheimer risk with women receMng H T (6), results from W H I showed the opposite trend (7). In one of the more recent observational trials (8) it was reported that although there was an overall protective effect, if H T was initiated in older indMduals (just as in WHI), there was a trend for a worse outcome. An age effect was also demonstrated in the Multi-Institutional Research in Alzheimer Genetic Epidemiology (MIRAGE) trial (9); and in W H I , a recent analysis of prior users (at a younger age) of H T showed a reduction in dementia risk in this subgroup (10). Finally, in a similar analysis to that described for C H D from Denmark (4), women treated for 2 to 3 years at the onset of menopause had an improved "blessed score" (less cognitive decline) at follow-up more than 10 years later (11). Although all this discussion pertains to standard dose HT, there is clear evidence for reduced risks and similar effects with lower doses of l i T (12). Although stroke risk (ischemic) with standard dose estrogen therapy ( E T ) / H T is most likely relegated to older users, this risk does not appear to be increased with lower doses (0.3 versus 0.625 mg CEE) (13). This is possibly true for thrombosis risk as well, although the data are not definitive (12). More convincing are the findings of reduced thrombosis risk with the use of transdermal estrogen (13). Lower doses also have been shown to be sufficient to treat hot flashes and maintain bone mineral density (BMD) (12), as well as to provide the same coronary benefits as seen in the observational trials suggesting a protective effect (13). Thus, it is clear in my view that lower doses should be the standard approach in candidates for HT.
All-cause mortality n=174
i.................
[ .........
cv~ 1 n-39
aooe I
11
........i ! cancer
[ traumatic
l
n=5
FIGURE 63.2 Causes of death among the 174 cases included 61 cases (35.1%) of death due to cardiovasculardisease (CVD), 68 cases due to cancer (39.1%), and 45 cases of death (25.9%) due to other causes (including 5 cases [2.3%] due to trauma, and 13 cases [7.5%] of unknown reason). CHD, coronaryheart disease.
877
In the Western world, there has been a rapid increase in the prevalence of obesity and metabolic syndrome, leading to an increased incidence of diabetes and cardiovascular disease. The primary approach here should be early preventative medicine in terms of education and stressing healthy lifestyle, which includes changes in diet coupled with an increase in activity. In younger postmenopausal women, where many of these problems are quite prevalent, what is a symptomatic woman to do? There is a general perception that H T causes weight gain, and may increase insulin resistance and diabetes risk. Some of these issues have been tackled in Section V. A recent meta-analysis has provided more clarity around these issues for us (14). It is quite clear that H T does not cause weight gain, and in fact, studies on body composition have shown that abdominal fat mass decreases. Similarly as a rule insulin resistance may be reduced by as much a 12% and the risk of new onset diabetes may decrease by about 30% with H T (14). In diabetic women, there are also data to suggest improved glycemic control (14). While more data needs to be collected, there is evidence that in obese women with metabolic syndrome, it is preferable to use transdermal rather than oral therapy when H T is prescribed (15). Thus while it is not my perspective that we should be prescribing H T for the sole purpose of preventing metabolic syndrome, diabetes or cardiovascular disease, the point here is that there are data that if a woman chooses to use HT, these risks may be decreased, particularly if the appropriate therapy is chosen. The current evidence would favor lower doses of systemic estrogen, and minimizing progestogen exposure.
I. T H E DECISION TO TREAT OR N O T TREAT W I T H H T One can look at this issue physiologically, practically, or emotionally. From a physiologic standpoint, there are estrogen receptors in virtually every cell in the body. These receptors are substantially downregulated without their natural ligand, 17f3estradiol, after menopause. Is this natural? Women in prior centuries did not live much past the normal age of menopause. Because women now live for 30 or more years beyond menopause, do they need these estrogen receptors? The practical view regarding treatment is to treat it like any other drug and to prescribe it for a specific indication (e.g., hot flashes) for the shortest duration. This approach is in keeping with the current Food and Drug Administration (FDA) guidelines and does not encompass therapy for more nonspecific menopausal symptomatology or quality-of-life issues. It also falls to a second-line therapy for the prevention of osteoporosis, a position that is still widely debated. The emotional perspective that arose in the aftermath of W H I finds a very polarized world of prescribers. One camp would never ever consider E T / H T for any indication, with
878
ROGERIOA. LOBO
the other side totally trashing W H I as being fallacious (and a political plot, as some have told me) and discounting the findings. Several physician surveys have found that there is a great deal of confusion over the appropriateness of prescribing HT/ET. It was found in a survey of primary care physicians in Florida that most had misinterpreted many of the risks and benefits found in W H I (16) Many primary care physicians are concerned with prescribing H T even for symptoms (hot flashes) because of the perceived risks. Because of this, a recent primary care conference deliberated this issue and concluded that there should not be any concern prescribing to younger healthy women for symptomatic complaints at the onset of menopause (17). From the patient's perspective it is clear that most women receive their information from the media (18). Largely, at least in the past, the media have not accurately portrayed to the public the risks of HT, mixing up interpretations of relative risks (RR) and attributable risks for example. In various surveys of women in Europe and the U.S. regarding attitudes toward HT, women have said their primary reason for considering H T is for menopausal symptoms and that their greatest concern is that of breast cancer (19). However, in these surveys, the majority of women who were current users of H T were greatly satisfied and felt that, for them, the benefits outweighed the risks (Fig. 63.3). Earlier I wrote that I had told reporters, before the reports from W H I became public, that we would not learn much new information from WHI. Although much of this statement is still true, W H I and the secondary prevention trials for C H D did teach us some important lessons. Namely that older asymptomatic women should not be prescribed E T / H T and that there is no secondary coronary prevention
benefit to using ET/HT. Further, in older high-risk individuals there is the real concern of early harm when prescribing standard-dose therapy in women not receiving a statin. Therefore, although it is clearly appropriate to prescribe E T / H T for symptomatic relief in young healthy women, it should be initiated at lower doses. Also, it is not indicated in older women (more than 10 years from the onset of menopause) who do not have symptoms. In my view, it is also indicated in largely asymptomatic women, for the prevention of osteoporosis in younger postmenopausal women who are a high risk for osteoporotic fracture. In this situation, there are incomplete data for the appropriateness of bisphosphonates in younger women. Bisphosphonates are a good option in the older woman, and particularly in those women with the diagnosis of osteoporosis. Again, lower doses of estrogen should be used and in certain higher risk individuals (those who are obese, insulin resistant, hypertriglyceridemic, or hypertensive) transdermal therapy would be preferred. There are data suggesting that transdermal estrogen does not carry the thrombosis risk of oral estrogen, in standard doses. (20) In my view, progestogen exposure should be minimized. Although continuous combined regimens with lower doses of estrogen and progestogen may be a possible approach in individuals wishing not to bleed, my personal preference is to use continuous low-dose estrogen with intermittent progestogen. In this situation, depending on the degree of endometrial proliferation as witnessed by transvaginal ultrasound, progestogens could be used for 10 to 12 days monthly, bimonthly, or every 3 months. In this setting synthetic progestogens are also an option, although my preference is for micronized progesterone used orally or
Q: With which statement do you agree more?
FIGURE 63.3 U.S.:Agreement to statement about HT: Benefits vs. risks (distribution per user type).
CHAPTER 63 The Future of Therapy and the Role of Hormone Therapy vaginally. The currently available intrauterine progestogens are too high in dose for routine use in conjunction with low-dose estrogen. Although the length of therapy must be an individual decision, as long as a woman feels she is benefiting from therapy there need not be an arbitrary time line for stopping. Although substantially reduced or even eliminated by treating younger, healthy women with lower doses, the major risks of therapy (thrombosis, coronary events) occur early, in the first 1 to 2 years. Women on therapy beyond this point tend to continue to do well. Thus, the major concern of longer-term therapy, arbitrarily defined as more than 5 years, is the risk of breast cancer. The data reviewed here relate to standard-dose therapy (equivalent of 0.625 mg of conjugated equine estrogen [CEE]), and I would expect that lower doses would be safer (although this is unproven). In the estrogen-only arm of W H I , 0.625 CEE for over 6 years did not increase the risk of breast cancer and significantly decreased the total rate of breast cancer among adherent women, and specifically for cases of ductal carcinoma (21). In the NHS, estrogen alone at this dose did not increase the risk for up to 20 years of use (22). The addition of a progestogen, however, appears to increase this risk. The risk in W H I was of borderline significance after 5 years of CEE + medroxyprogesterone acetate (MPA) and when adjusted for other risks failed to reach statistical significance (23). It was clear in these latter analyses that CEE + MPA use in neverusers (i.e., women initiating therapy at menopause for the first time) was not significant for up to 5 years. Taken together these data suggest that standard-dose therapy of estrogen and progestogen is only a significant risk factor beyond 5 years of use. In that it is clear in my view that the risk of breast cancer is largely determined by the additional progestogen exposure, particularly for longer-term therapy, H T should be predominantly estrogen therapy, with progestogen added only as necessary to prevent uterine disease. This minimalist approach is particularly justified when using lower estrogen doses, which lead to less endometrial proliferation. Before leaving this topic, it is important to put this breast cancer risk into perspective as it is an ever-pervasive concern of women and has been sensationalized by the media. If we take the worst case scenario m that is, standard-dose therapy, with some prior use for 5 or more years m t h i s would be expected to result in a relative risk (RR) of breast cancer of 20% to 30% (RR: 1.2-1.3). This would be expected to increase the number of breast cancers in 100 women between the ages of 50 and 60 by less than 1% (an increase from the baseline risk of 2.8 to 3.4 per 100 women). This RR of breast cancer with HT, although we cannot trivialize any risk, is lower than that of some common exposures and occupationmfor example, it is less than being obese or being a flight attendant (Table 63.3).
879
TABLE 63.3
Putting Breast Cancer Risks into Perspective
CEE alone: 0.77 (0.59-1.01) CEE/MPA: 1.24 (1.01-1.54) W/H ratio > 0.8:3.3 (1.1-10.4) Fish intake: 1.14 (1.03-1.26) Antibiotic use: 2.07 (1.48-2.89) Flight attendant: 1.87 (1.15-2.23) Electric blanket use (African American): 4.9 (1.5-15.6)
II. LOOKING TO T H E FUTURE In that the most pertinent candidates for H T / E T (young healthy symptomatic women) have never been studied in RCTs, these data are sorely needed. However, another W H I in this population will never be done. Thus, we can look forward to results of some smaller trials. Early, Late Intervention Trial of Estrogen (ELITE) assesses CV risks in younger and in older women, and Knonos Early Estrogen Prevention Study (KEEPS) assesses the same risks as well as cognition in women initiating therapy within 3 years of the menopause. ELITE studies the use of oral micronized estradiol compared with placebo, with vaginal progestogen, and KEEPS compares oral CEE 0.45 mg versus transdermal E2, 0.05 mg versus placebo with sequential monthly oral micronized progesterone 200 mg for 12 days. We will hopefully have some of these results in 5 years. Further analyses of data from completed trials such as W H I will still be forthcoming, and although some further refinements in the subgroup analyses may be valuable, on balance it is not likely that there will be any major breakthroughs. One of the weaknesses of an RCT is that only one population can be studied and one specific therapy. Questions we need answered regarding different doses, routes of administration, and various progestogens cannot be extrapolated from these data. The long-term follow up of W H I subjects after they stopped the trials is something that would be of interest. More SERMs and SERM combinations may be forthcoming as well. The ideal SERM that would have benefits for hot flushes and brain activity has not been developed but is a potential for the future, particularly if this same SERM is beneficial for bone and inhibits breast and uterine proliferation. Newer SERMS such as lasofoxifene, which has purported beneficial vaginal effects, and bazedoxifene, which has an overall beneficial profile, may be marketed in the near future. The combination of CEE and bazedoxifene that avoids progestogen exposure and haves a beneficial effect on the breast and uterus (preventing withdrawal bleeding) is likely to be approved in the future as well. However, all selective estrogen receptor modulators (SERMS), like oral estrogen, still carry some risk of venous thrombosis.
880
ROGERIO A. LoBo
W h e t h e r there is a need for androgen therapy in w o m e n remains controversial. However, lower physiologic doses with transdermal therapy have been shown to be efficacious. Because of very theoretical unknown long-term sequelae (CV-D and breast cancer), this therapy has not been approved. Perhaps in the future selective androgen receptor modulators (SARMs) with the ability to target only the brain, and perhaps bone, may be of benefit. However, it needs to be appreciated that we still are unclear as to what actions androgens have on the brain in women. T h e future of various hormonal therapies appears bright. In my view the dust has just begun to settle p o s t - W H I , which had major consequences due to misinterpretation of the data and media attention with its exaggerated interpretations. W e are now just emerging again with a clearer understanding regarding who should be treated, for what, with what, and for how long. All this should be individualized and remain flexible; it should be well thought out, rational, evidence based, and with complete informed consent and understanding of the woman. It should not be an emotional issue.
References 1. Hsia J, Langer RD, Manson JE, et al. Conjugated equine estrogens and coronary heart disease: the Women's Health Initiative. Arch Intern Med 2006;166:357-365. 2. Grodstein F, Manson JE, Stampfer MJ. Hormone therapy and coronary heart disease: the role of time since menopause and age at hormone initiation.J WomensHealth (LarchmO 2006;15:35-44. 3. Salpeter SR, Walsh JM, Greyber E, Salpeter EE. Brief report: coronary heart disease events associated with hormone therapy in younger and older women. A meta-analysis.J Gen Intern Med 2006;21:363-366. 4. Alexandersen P, Tanko LB, Bagger YZ, Qin G, Christiansen C. The long-term impact of 2-3 years of hormone replacementtherapy on cardiovascular mortality and atherosclerosis in healthy women. Climacteric 2006;9:108-118. 5. Lobo RA. Evaluation of cardiovascular event rates with hormone therapy in healthy, early postmenopausal women: results from 2 large clinical trials. Arch Intern Med 2004;164:482-484. 6. LeBlanc ES, Janowsky J, Chan BK, Nelson HD. Hormone replacement therapy and cognition: systematic review and meta-analysis. JAMA 2001;285:1489-1499. 7. Shumaker SA, Legault C, Rapp SR, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651-2662.
8. Zandi PP, Carlson MC, Plassman BL, Welsh-Bohmer, et al. Hormone replacement therapy and incidence of Alzheimer disease in older women: the Cache County Study. JAMA. 2002; 288(17):2123-2129. 9. Henderson VW, Benke KS, Green RC, Cupples LA, Farter LA; MIRAGE Study Group. Postmenopausal hormone therapy and Alzheimer's disease risk: interaction with age.J Neurol Neurosurg Psychiatry 2005;76:103-105. 10. Espeland MA, Henderson VM. Preventing cognitive decline in usual aging. Arch Intern Med 2006;166:2433-2434. 11. BaggerYZ, Tanko LB, Alexandersen P, Qin G, Christiansen C. Early postmenopausal hormone replacement therapy may prevent cognitive impairment later in life. Menopause 2005;12:12-17. 12. Lobo RA. The rationale for low-dose hormonal therapy. Endocrine 2004;(3):217-221. 13. Grodstein F, Manson JE, Colditz GA, et al. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease.Ann Intern Med 2000;133:933-941. 14. Salpeter SR, Walsh JME, Ormiston TM, et al. Meta-analysis: effect of hormone-replacement therapy on components of the metabolic syndrome in postmenopausal women. Diabetes, Obesity and Metabolism 2006;8:538-554. 15. Chu MC, Cosper P, Nakhuda GS, Lobo RA.A comparison of oral and transdermal short-term estrogen therapy in postmenopausal women with metabolic syndrome. Fertil Steri12006;86:1669-1675. 16. Williams RS, Christie D, Sistrom C. Assessment of the understanding of the risks and benefits of hormone replacement therapy (HRT) in primary care physicians.Am J Obstet Gyneco12005. 193:551-558. 17. Lobo RA, Belisle S, Creesman WT. Should symptomatic menopausal women be offered hormonal therapy? MedGenMed 2006;8. 18. Hoffmann M, Hammar M, Kjellgren KI, Lindh-Astrand L, Brynhildsen J. Changes in women's attitudes towards and use of hormone therapy after HERS and WHI. Maturitas 2005;52:11-17. 19. Srothmann A, Schneider H, Schaefer M. Hormone therapy: the U.S. women'sperspective. Program and Abstracts of the 16th Annual Meeting of the North American Menopause Society. September 28-October 1, 2005,
San Diego, California. Abstract P-78. 20. Scarabin PY, Oger E, Plu-Bureau G; Estrogen and Thromboembolism Risk Study Group. Differential association of oral and transdermal oestrogen-replacement therapy with venous thromboembolism risk. Lancet 2003;362:428-432. 21. Stefanick ML, Anderson GL, Margolis KL, et al. Effects of conjugated equine estrogens on breast cancer and mammography screening in postmenopausal women with hysterectomy.JAMA 2006;295:164-257. 22. Chen WY, Manson JE, Hankinson SE, et al. Unopposed estrogen therapy and the risk of invasive breast cancer. Arch Intern Med 2006; 166:1027-1032. 23. Anderson GL, Chlebowski RT, Rossouw JE. Prior hormone therapy and breast cancer risk in the women's health initiative randomized trial of estrogen and progestin. Maturitas 2006;55:107-115.
Subject Index Note: Page numbers followed by an f refer to figures, those followed by t refer to tables.
A Acquired hypercoagulability, 494-495 ActMn hypothesis, 61-63 Acupuncture, 688 Acute coronary syndromes, 408,408f AD. See Alzheimer's disease
Age
at hormone replacement therapy, 633 at menopause, 78 breast cancer risk factors, 570 environmental influences, 83- 82 ethnic differences, 81 international differences, 81 menarche, 3 menopausal, 2 - 3 natural menopause, 79-80, 81 perimenopause, 79-80, 81 Aging bone loss, 331 DHEA in, 822-824, 822f, 823f, 823t versus menopause status, 272-273,273f, 274f menopause transition, changes associated with, 159 pelvic organ prolapse and, 749 of population, 6 - 7 and sexuality, 272 Alcohol, 90t CHD risk reduction, 438, 438f osteoporosis and, 328 Alendronate treatments for osteoporosis, 383- 384, 384f, 385t Alkaline phosphatase, 339 Alpha-adrenergics and stress urinary incontinence, 710 Alternative medicine. See Complementary and alternative medicine Alzheimer's disease (AD), 295-296, 302 dehydroepiandrosterone and, 301 dementia in women aged 65 and older, 288 -290 estrogen and, 287-288 brain function, 295-296, 296t cognition, 297- 298 future research, 291 hormone therapy in midlife, 297 in old age, 297-298,298t prevention of AD, observational studies, 298 prevention of AD, WHIMS, 298- 300, 299t treatment of AD, 300 menopause and, 287-288 phytoestrogens and, 302 progesterone and, 300-301 selective estrogen receptor modulators and, 301-302 testosterone and, 300-301 younger postmenopausal women, 290
Amenorrhea evaluations, 105 - 106 hypergonadotropic amenorrhea. See Premature ovarian failure menopause transition and, 158 AMH. See Antimfillerian hormone Amitriptyline for interstitial cystitis, 723 Anabolic steroid treatments for osteoporosis, 387, 387f Anal triangle, 741 Androgenic changes in menopause transition, 160-161 Androgens, 799, 800t, 809 brain, effects on, 211 breast cancer risks, 806 candidate assessments, 807-808 insufficiency, clinical consequences of, 801- 804 myocardium, effects on, 805- 806 osteoporosis treatments, 364- 365 physiology in reproductive years, 799-800 psychological functioning, neurobiologic effects of, 218-219 testosterone deficiency, causes of, 800-801 testosterone replacement benefits of, 801-804 most likely to benefit from, 807 prescribing, 808-809, 808t risks of, 804-807 vascular function, effects of, 805-806 Anorectal canal, support of and pelvic organ prolapse, 743 Anovulatory dysfunctional uterine bleeding, 155 Anterior vaginal wall prolapse, 744 Anticholinergic drugs and urge urinary incontinence, 715 - 716 Anticoagulation and hormone therapy, 483-484, 483f, 484f Antidepressants and hot flashes, 196 Antimtillerian hormone (AMH), 116, 130-131 Apical prolapse, 744 Aromatase deficiency and premature ovarian failure, 102 Aromatase mutations in females (46,XX) (CYP19), 41 Artificial urinary sphincter and stress urinary incontinence, 715 ART procedures, 111,112f Atarax for interstitial cystitis, 723 Atherogenic lipoproteins, 461-462, 462f Autonomic tone, regulation of and estrogen, 425 Autosomal chromosomal abnormalities, 37 Autosomal dominant POF, 43 Autosomal genes, 37, 38t aromatase mutations in females (46,XX) (CYP19), 41 blepharophimosis-ptosis-epicanthus (FOXL2), 40 carbohydrate-deficient glycoprotein, 42
881
cerebellar ataxia with XX gonadal dysgenesis, 39-40 fragile X syndrome (CGGn), 42 FSH-[3 mutations, 37- 38 galactosemia, 41- 42 GDF9 mutations, 39 genes postulated on basis of animal models, 42 germ cell failure in both sexes (nonsyndromic), 41 germ cell failure in both sexes (syndromic), 41 17(x-hydroxylase/17,20 Desmolase Deficiency (CYP17), 41 inhibin alpha (INHA) mutations, 38-39 LH receptor mutations, 38 malformation syndromes associated with XX gonadal dysgenesis, 40 Mendelian disorders characterized predominately by somatic features, 40, 40t myotonic dystrophy (CTGn), 41 Perrault syndrome, 39 undefined, 42 Autosomal recessive POF, 43 B
Baby boom, 6 Bacillus Calmette Gu4rin for interstitial cystitis, 725 Bacterial cystitis, 726 Barrier contraceptives, 177-180, 178t- 179t BASs. See Bile acid sequestrants Behavioral therapies, 688 hot flashes, 195 stress urinary incontinence, 708-709 urge urinary incontinence, 715 Berlin Study, 649-651,650f, 650t, 651f Bile acid sequestrants (BASs), 436-437, 437t Biochemical markers of bone turnover. See Bone turnover markers Biodentical hormones, 687 Biofeedback techniques for stress urinary incontinence, 709 Bisphosphonates treatments for osteoporosis, 358-359, 383-385,385t Black cohosh, 685-686 Bladder anatomy, 694, 695f filling and emptying assessment, 705 Bladder augmentation in interstitial cystitis, 725 Bladder neck hypermobility, 699 Bladder training for urge urinary incontinence, 715 Blepharophimosis-ptosis-epicanthus (FOXL2), 40 Blood flow and hot flashes, 187 melatonin, detection in, 830-831 viscosity, increases in and cardiovascular disease, 416 Body mass and composition age at natural menopause and onset of perimenopause, 82
882
SUBJECTINDEX
Body mass and composition (continued) breast cancer and body weight, effects of, 572-573,573t cardiovascular disease and hormone therapy, cardioprotective effect according to, 550-551
endometrial cancer and, 589 observational studies, 87t-88t presentation and symptoms of menopause, 83-86 testosterone replacement, benefits of, 804 Body temperature and hot flashes, 188-189, 189f Bone, 227-229, 228f, 229f changes, 321 density and strength. See Bone density and strength formation markers, 337-339, 338f and osteoporosis, 354-355 osteoporosis. See Osteoporosis pathogenesis of bone loss, 324-328, 325f, 326f, 327f remodeling. See Bone remodeling turnover. See Bone turnover turnover markers. See Bone turnover markers Bone density and strength, 331-332, 336 age-related bone loss, 331 androgens and, 802-803, 804f bone microarchitecture, 335- 336, 335f bone turnover, 324-328, 325f, 326f, 327f, 333-334, 334f, 334t, 353-355 changes in menopause transition, 163-164 DHEA and, 825, 825t DXA, 331-332, 333 environmental influences on bone mass, 353, 353t fractures, osteoporosis-related, 331 genetic influences on bone mass, 353, 353t low bone density and osteoporosis, 352 low bone mass, 332 measuring BMD, 332-333 and oral contraceptives, 173 osteopenia, 332 osteoporosis. See Osteoporosis peak bone mass and its determinants, 323-324 raloxifene, effects on, 839-841,840f tamoxifen, effects on, 837-838 tibolone, effects on, 850 trace elements in osteoporosis, 673-674 Bone remodeling, 324-328, 325f, 326f, 327f, 333-334, 334f, 334t markers. See Bone turnover markers and osteoporosis, 353- 355 estrogen-progestogen therapy, 379 estrogen treatments, 378-379, 379f Bone remodeling units, 324, 354 Bone resorption markers, 339-341,340f and osteoporosis, 354 Bone turnover, 324-328, 325f, 326f, 327f, 333-334, 334f, 334t markers. See Bone turnover markers and osteoporosis, 353- 355 estrogen-progestogen therapy, 379 estrogen treatments, 378-379, 379f Bone turnover markers, 337, 338t, 341,345-346 alkaline phosphatase, 339 bone formation markers, 337-339, 338f bone loss predictions, 344-345 bone resorption markers, 339-341,340f cathepsin K, 341 degradation products of helical region of type I collagen, 340
degradation products of telopeptide region of type I collagen, 339-340, 340f fractures, predictions of, 345 matrix metalloproteinases, 341 menopause, changes in markers at, 342-343 monitoring therapy, 345 osteocalcin, 339 osteoclastic enzymes, 341 osteoporosis, changes in markers at, 344 procollagen propeptides, 337- 338, 338f therapy responses changes in markers at, 344 predictions as to, 345 variability of, 341-342 Bony pelvis anatomy, 740, 740f Botanicals and vulvovaginal complaints, 268 Botulinum toxin, intravesical injection of interstitial cystitis, 725 urge urinary incontinence, 717 Brain, 277 aging of and reproductive senescence, 142-143 Alzheimer's disease. See Alzheimer's disease cognition. See Cognitive function estrogen and estrogen receptors, 202-204, 208-209, 281-283,295-296, 296t sex steroids, effect on, 199-200, 200t anatomic-functional correlations, 200- 202, 200f, 201f androgen, 211 and brain phenotype, 203-204 cognitive function and the climacteric, 207-210 estrogen and estrogen receptors, 202-204, 208-209, 281-283 mood and the climacteric, 205-207 progesterone, 210-211 progestins, 210- 211 sleep disorders and the climacteric, 204-205 vasomotor episodes and other brain dysfunction at start of climacteric, 204 Breast cancer, 569-570, 576-577 age at menarche and, 570 age at menopause and, 570 body weight, effects of, 572-573,573t exercise, effects of, 663 hormone therapy and, 573-576, 574f, 574t, 576t, 599-601,628t-629t, 630 mitotic activity of cell, 572, 572f oral contraceptives and, 174 parity, 570 phytoestrogens and, 684 raloxifene, effects on breast, 841,841f risk factors, 570 risk modeling, 570- 572, 571f tamoxifen, effects on breast, 838 testosterone replacement, risks of, 806 tibolone, effects on breast, 849 Women's Health Initiative dietary modification trial, 615 Breast surveillance, postmenopause, 579-583, 580f, 581f, 582f, 583f C Calcitonin treatments for osteoporosis, 363-364, 385-387, 386f Calcium osteoporosis and, 328, 359-360, 667-670, 672-673 Women's Health Initiative Calcium Vitamin D Trial, 615-616
CAM. See Complementary and alternative
medicine Cancer, 565-567, 566f breast cancer. See Breast cancer colorectal cancer hormone therapy and, 601-606, 602t, 604t-605t Women's Health Initiative Calcium Vitamin D Trial, 616 Women's Health Initiative dietary modification trial, 615 endometrial cancer. See Endometrial cancer lung cancer and hormone therapy, 606 oral contraceptives and breast cancer, 174 endometrial neoplasia, 172 - 173 ovarian neoplasia, 172-173 ovarian cancer. See Ovarian cancer Women's Health Initiative hormone therapy trial, 613-614 Capsaicin for interstitial cystitis, 724-725 Carbohydrate-deficient glycoprotein (CDG), 42 Carbohydrate metabolism and sex steroids, 502-504 Cardiac rhythm, regulation of and estrogen, 425 Cardiovascular disease (CV-D), 403-404, 405 -406 autonomic imbalance and, 416 - 417 blood viscosity, increases in, 416 cholesterol and, 656-659 compounded risk factors, 417 coronary artery atherosclerosis. See Coronary artery atherosclerosis coronary heart disease. See Coronary heart disease coronary syndromes, 408,408f depression and, 416 diabetes meUitus and, 413-414, 413f epidemiology of, 406-407, 406f, 407f exercise, role of, 659-663,661f, 662f gender and pathogenesis of cardiac ischemia, 409 gender-related differences in CHD, 417-418, 417f, 418f, 419f clinical presentation, 418- 419, 419f coronary findings, 420, 420t diagnostic invasive studies, use of, 419-420, 420f management of, clinical and surgical, 420-421 mechanisms of differences, 421- 423,422f outcome and prognosis, 421 genetics and, 415-416 hormone therapy and, 529-530, 530f, 559-560 arterial imaging trials, 537-541,538t cardioprotective effect according to body mass index, evidence for, 550- 551 cardioprotective effect according to duration of, evidence for, 542-545,543t, 544f, 544t cardioprotective effect according to timing of initiation of therapy, 545-547 cardioprotective effect requires healthy endothelium, evidence for, 547- 548, 547f, 548f CEE+MPA therapy, 548-550, 549f, 550f clinical perspective on, 555-559, 556t, 557t clinical trials and observational studies, reconciling of findings, 620-622, 621t, 622t Estrogen in the Prevention of Atherosclerosis Trial, 539-540, 539f, 540f
SUBJECTINDEX Estrogen in the Prevention of Reinfarction Trial, 532-533,533t Estrogen Replacement Atherosclerosis Trial, 538 Heart and Estrogen/Progestin Replacement Study, 530-532, 531f, 532f morbidity and mortality, 530-537, 531t, 554-555, 554f, 555f, 555t, 628t-629t, 631 Papworth Hormone Replacement Therapy Atherosclerosis Study, 532 Postmenopausal Hormone Replacement against Atherosclerosis Trial, 540-541 randomized controlled trials and observational studies, discordance between, 541-542, 541t stroke and, 551-552, 552t venous thromboembolism and, 552-554, 553t, 554t Women's Angiographic Vitamin and Estrogen Trial, 538- 539 Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial, 539 Women's Health Initiative, 533-537, 533f, 534f, 535f, 536t, 537f, 613 hyperlipidemia, 412, 412f hypertension and, 412-413 injury and inflammation, 407-408 life expectancy trends, 439-443,439f, 440f, 441f, 442f lipids and lipoproteins and hormone therapy, effects on, 466-467 menopause signal for the future, 9-10 nutrition, role of, 655-659, 656t, 657t, 658f obesity and, 409-411,410f, 411f, 432-434, 663-665 pathogenesis of cardiac ischemia, 407- 409 phytoestrogens and, 685 plasma lipid abnormalities, 411-412,411f, 412f risk factors, 409- 423 sedentary lifestyle and, 415 smoking and, 414-415 stress and, 416-417 testosterone replacement risks, 804-806 and use of contraceptives, 169, 170t Women's Health Initiative dietary modification trial, 615 Women's Health Initiative hormone therapy trial, 533-537, 533f, 534f, 535f, 536t, 537f, 613 Cardiovascular system changes in menopause transition, 163 DHEA and, 825 disease. See Cardiovascular disease estrogen, mechanism of action on, 453-454, 457-458,458f coagulation factors, effects on, 456 direct cardiovascular effects of, 454-456 estrogen-mediated signaling in, 455-456, 455f fibrinolytic factors, effects on, 456 inflammatory markers, effects on, 456-457 lipids, effects in, 456 metabolic state, effects on, 457 oxidative state, effects on, 457 progenitor cells, effects on, 457 systemic cardiovascular effects of, 456-457 vasoactive hormones, effects on, 457 tibolone, effects of, 850 Carotid arteries, 232-233,233f Carotid intima-media thickness, 432
883 Cataracts effects of raloxifene, 842 effects of tamoxifen, 839 Catecholamines and hot flashes, 191-192, 192f Cathepsin K, 341 CDG. See Carbohydrate-deficient glycoprotein Cell adhesion molecules, effects of hormone therapy on, 473-474, 473f, 474t Cellular basis of bone loss, 324-328,325f, 326f, 327f Central nervous system and estrogen, 310-311 Cerebellar ataxia with XX gonadal dysgenesis, 39 -40 Cerebrovascular disease and oral contraceptives, 173 Cervix, support of in pelvic organ prolapse, 743 Chaste-tree berry, 687 CHD. See Coronary heart disease Chemokines, effect of hormone therapy on, 473 -474 Chemotherapeutic agents and premature ovarian failure, 103 Chinese herbs, 687 Cholesterol and cardiovascular disease, 656-659, 661-662 Chronic diseases and osteoporosis, 328-329 Circadian rhythms and hot flashes, 194 and melatonin, 833 Climacteric definition of, 2 dementia and, 208 depression and, 206-207 Greene Climacteric Scale, 642-643 Clinical trials, 609 coronary heart disease and estrogen, 425-430 hormone therapy. See Hormone replacement therapy Leisure World Cohort Study, 627-628 Nurses' Health Study, 629-630 Women's Health Initiative. See Women's Health Initiative Coactivator protein, 24-25, 25f Coagulation hormone therapy and, 483-484, 483f, 484f, 486-487 hypercoagulability, VTE risks, 494-495 Cognitive function, 279,283 Alzheimer's disease and, 297-298 and the climacteric, 207-210 DHEA and, 825-826 estrogen and, 221 Alzheimer's disease and, 297-298 effects on brain, biologic plausibility for, 281-282 indirect effect on, 283 and objective measure of memory, 280-281 selective receptor modulators and, 283 hormone replacement therapy and, 632 memory and menopausal transition, 279-280 postmenopause, 287, 291 ,dementia in women aged 65 and older, 288 -290 estrogen and, 287-288 future research, 291 menopause and, 287-288 younger postmenopausal women, 290 Women's Health Initiative hormone therapy trial, 614
Coital incontinence, 698 Collagen degradation products of helical region of type I collagen, 340 degradation products of telopeptide region of type I collagen, 339-340, 340f Collagen markers, 229-230, 230f Colorectal cancer hormone therapy and, 601-606, 602t, 604t-605t, 628t-629t, 630 Women's Health Initiative Calcium Vitamin D Trial, 616 Women's Health Initiative dietary modification trial, 615 Complementary and alternative medicine (CAM), 683, 688, 819 acupuncture, 688 behavioral therapies, 688 biodentical hormones, 687 dehydroepiandrosterone. See Dehydroepiandrosterone exercise. See Exercise herbs. See Herbs melatonin. See Melatonin paced respiration and relaxation, 688 phytoestrogens. See Phytoestrogens progesterone creams, 687 selective estrogen receptor modulators. See Selective estrogen receptor modulators tibolone. See Tibolone vitamin E, 687-688 yam, 687 Compression of morbidity, 8 Computed tomography (CT), 432 Condoms, 179 Congenital hypercoagulability, 494-495 Connective tissue changes, 227, 233 bone, 227-228,228-229,228f, 229f carotid arteries, 232- 233,233f collagen markers, 229-230, 230f intervertebral discs, 233,234f skin, 228-229,228f, 229f, 230-232 Continuous urinary incontinence, 697 Contraceptives for older women, 169, 170t, 180 barrier contraceptives, 177-180, 178t- 179t condoms, 179 diaphragms, 178-179 etonorgestrel implants, 177 implantable contraceptives, 176-177 injectable contraception, 175-176 intrauterine devices, 175 levonorgestrel implants, 176-177 oral contraceptives. See Oral contraceptives spermicides, 180 steroidal vaginal contraceptives, 177 transdermal contraception, 177 Copper-containing IUDs, 175 Corepressor proteins, 25 Coronary artery atherosclerosis, 509- 510, 510f, 525 CHD compared, 511 estrogen and early menopause, effects in, 517-520, 517f, 518f endothelium, effects on, 519 ER expression or activity diminished, 523 late menopause, effects in, 521-524, 522f LDL oxidation prevention, 519 plaque instability, putative mechanisms for, 523-524, 523f, 524f vasodilation, cause of, 519, 523
884 Coronary artery atherosclerosis (continued) Estrogen in the Prevention of Atherosclerosis Trial, 539-540, 539f, 540f Estrogen Replacement Atherosclerosis Trial, 538 inflammation and inflammatory markers, 519-520, 520t, 521t ovarian dysfunction, role of, 512-514, 513f, 514f Papworth Hormone Replacement Therapy Atherosclerosis Study, 532 perimenopause transition and, 514-515, 515f animal model for studies of, need for, 516, 516f endothelial function and, 515-516, 516f plasma lipid and lipoprotein changes during, 515 plasma lipids atheroprotection mechanisms of estrogen, 518-519 role of, 512, 512f, 515 Postmenopausal Hormone Replacement against Atherosclerosis Trial, 540- 541 primary versus secondary prevention, 511 progression of in North American women, 510-511,511f premenopausal determinants, 511 - 514, 512f sex hormone receptor signaling, 519 timing hypothesis, 521,524-525 Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial, 539 Coronary heart disease (CHD) coronary artery atherosclerosis compared, 511 estrogen and, 430 autonomic tone, regulation of, 425 cardiac rhythm, regulation of, 425 cellular mechanisms, 424 clinical trials, randomized, 425-430 epidemiologic studies, 423-424, 423f estrogen actions on the cardiovascular system in females, 424-425 glucose, effects of, 424 molecular mechanisms, 424 myocardium protective effects, 425 observational studies, 423-424, 423f, 429-430 plasma lipids, effects of, 424 vasculo-protective effects, 425 risk assessment, 430-432, 431t risk reduction, 432, 439 active awareness, 438-439 alcohol and, 438, 438f early diagnosis and treatment of associated medical conditions, 434-435 goal of low risk, 438-439 lifestyle modifications, 432-434, 434f lipid lowering treatment, 435-438, 435t, 436t, 437t Corpus albicans of ovary, 52 Corpus luteum of ovary, 51 Cortex of ovary, 51, 52f C-reactive protein, effects of hormone therapy on, 472- 473, 474t CT. See Computed tomography CVD. See Cardiovascular disease Cystectomy and interstitial cystitis, 725 Cystistat for interstitial cystitis, 724 Cystitis, 726 Cystocele, 743 Cystourethrocele, 743 Cystourethroscopy, 707, 728
SUBJECT INDEX Cytogenetics of X chromosomal aberrations isochromosomes for Xq (i[Xq]), 34, 35t premature ovarian failure, 101-102 46,X, del(Xp) deletions, 33-34, 34f 46,X.del(Xq) deletions, 35-36 45,X/46,X,del(Xp) deletions, 33-34, 34f 45,X/46.X,del(Xq) deletions, 35-36 Cytokines, effects of hormone therapy on, 473 -474
D Deacetylated norgestimate, 788 Deep vein thrombophlebitis (DVT). See Venous thromboembolism Degradation products of helical region of type I collagen, 340 Degradation products of telopeptide region of type I collagen, 339-340, 340f Dehydroepiandrosterone (DHEA), 821, 827 administration of, 824 Alzheimer's disease and, 301 bone metabolism and, 825, 825t cardiovascular system and, 825 clinical applications, 824-826 cognition and, 825-826 dosing, 824 glucose metabolism and, 825 libido and, 825-826, 826t memory and, 825- 826 in menopause and aging, 822-824, 822f, 823f, 823t, 824f moderator and biomarker of aging, 823-824 mood and, 825-826 replacement strategies, 824 safety of, 826-827 zona reticularis involution and aging, 822- 823, 824f Dementia, 288 and the climacteric, 208 and estrogen, 208-209 hormone replacement therapy and dementia, 628t-629t, 632 HRT treatments for AD, 289 Women's Health Initiative Memory Study, 288 -289 Demographics, 79- 81 Depression, 307, 674-675 cardiovascular disease and, 416 and the climacteric, 206-207 course and treatment of, 309 defining, 309- 310 diagnosis of, 308 estrogen effects of on central nervous system, 310-311 efficacy of, 313-314 evaluation of perimenopause-related depression, 314-315 future research, 315 impact of, 307-308 management of perimenopause-related depression, 314- 315 menopausal transition association between, evidence of, 311- 312 risk factors, 312- 313 menopause, symptoms of, 5 methodological problems, 309- 310 pathophysiology of, 308 reproductive status, characterizing, 309 stress and, 313 Desogestrel, 788- 789 Detrusor overactivity (DO), 697
Detrusor overactivity incontinence, 697 DHEA. See Dehydroepiandrosterone Diabetes, 665-667, 666f cardiovascular disease and, 413-414, 413f hormone therapy and, 504-505, 505f Diaphragms, 178-179 Diary, voiding, 701 - 702, 719- 720 Dienogest, 790 Diet, 83, 86-90, 90t diabetes and, 667 osteoporosis and, 667-670, 669t Women's Health Initiative dietary modification trial, 615 Dimethyl sulfoxide for interstitial cystitis, 724 Distention of the anterior vaginal wall, 748 DO. See Detrusor overactivity Dong quai, 686 Drospirenone, 790 Duloxetine and stress urinary incontinence, 710 DVT. See Deep vein thrombophlebitis DXA, 331-332, 333
E Ectopic pregnancy, reduced incidence of and oral contraceptives, 172 Elavil for interstitial cystitis, 723 Electrical stimulation therapy stress urinary incontinence, 709 urge urinary incontinence, 716- 717 Electrophysiologic studies of urinary incontinence, 707 Elmiron for interstitial cystitis, 723, 724 Embryo donation in older women, 117-118, 118f Endocrinology of folficulogenesis, 52 hot flashes, 190-192 parameters and ovary reserve, 59-61, 59f, 60f, 61f perimenopause, changes during, 67, 75 clinical sequelae of, 73-74 dynamics of, 68-73, 70f, 71f, 72f, 73f hormone-induced changes in, 74 normal reproductive cycle, 67-68, 68f, 69f profile assessment, 74 Endometrial cancer, 585, 590 hormone therapy after treatment for, 589-590, 590t effects of, 587-589, 588t morbidity and mortality changes with, 628t-629t, 630 incidence, 585-586, 586f obesity and, 589 oral contraceptives and, 172-173 pathophysiology, 587 phytoestrogens and, 684 prognosis, 589 risk factors, 586-587 Endometrial hyperplasia study requirements, 867- 869, 870t Enterocele, 743 Environmental influences age at natural menopause and onset of perimenopause, 82- 83 alcohol use, 90t body mass and composition. See Body mass and composition physical activity. See Physical activity premature ovarian failure, 103 presentation and symptoms of menopause, 83-90 smoking. See Smoking
SUBJECT INDEX Enzymatic defects in premature ovarian failure, 102 EOC. See Epithelial ovarian cancer Epidemiology, 77, 90-91 age. See Age cardiovascular disease, 406-407, 406f, 407f coronary heart disease, 423-424, 423f demographics, 79- 81 environmental influences. See Environmental influences estrogen CHD and epidemiologic studies, 423-424, 423f osteoarthritis prevention and epidemiologic research, 397- 399 ethnic differences, 81 - 82 hot flashes, 187 international differences, 81 - 82 interstitial cystitis, 717- 718 methodology, 77- 78 osteoarthritis, 397- 399 pelvic organ prolapse, 747 physiological changes that accompany menopause, 78 -79 presentation and symptoms of menopause, 3-6, 78 demographic characteristics, 80- 81 environmental influences, 83-90 ethnic and international differences, 81- 82 psychological functioning, 217- 218 urinary incontinence, 698 urinary tract infections, 726 Epithelial ovarian cancer (EOC). See Ovarian cancer Estradiol preparations, 772-774, 773f Estradiol replacement for mood disorders, 313 -314 Estrogen Alzheimer's disease and the brain, 295-296, 296t and cognition, 297-298 breast cancer and hormone therapy, 573-576, 574f, 574t, 576t, 599-601 cardiovascular system, mechanism of action, 453-454, 457-458,458f coagulation factors, effects on, 456 direct cardiovascular effects of, 454-456 estrogen-mediated signaling in, 455-456, 455f fibrinolytic factors, effects on, 456 inflammatory markers, effects on, 456-457 lipids, effects in, 456 metabolic state, effects on, 457 oxidative state, effects on, 457 progenitor cells, effects on, 457 systemic cardiovascular effects of, 456-457 vasoactive hormones, effects on, 457 central nervous system, effects on, 310-311 CHD and, 430 autonomic tone, regulation of, 425 cardiac rhythm, regulation of, 425 cellular mechanisms, 424 clinical trials, randomized, 425-430 epidemiologic studies, 423-424, 423f estrogen actions on the cardiovascular system in females, 424-425 glucose, effects of, 424 molecular mechanisms, 424 myocardium protective effects, 425 observational studies, 423-424, 423f, 429-430
885 plasma lipids, effects of, 424 vasculo-protective effects, 425 clinical aspects of estrogen therapy, 774-775 cognitive functioning and, 221 Alzheimer's disease and, 295-298,296t biologic plausibility for effects of, 281-283 indirect effects on, 283 in postmenopause, 287- 288 selective receptor modulators, 283 and sex steroids, 202-203,202t, 203-204, 203f, 210 conjugated equine estrogen preparations, 771-772, 772f coronary artery atherosclerosis early menopause, effects in, 517-520, 517f, 518f endothelium, effects on, 519 ER expression or activity diminished, 523 late menopause, effects in, 521-524, 522f LDL oxidation prevention, 519 plaque instability, putative mechanisms for, 523-524, 523f, 524f vasodilation, cause of, 519, 523 dementia and, 208-209 depression effects of on central nervous system, 310-311 efficacy of, 313-314 development of components, 814- 815 estradiol preparations, 772-774, 773f estrone preparations, 772-774, 773f hot flashes and, 4, 190-191 lipids and lipoproteins, effects on, 424, 456, 463,464f memory, objective measure of, 280-281 metabolism of, 769 models of estrogen action, 18-19, 18f neurobiologic effects of and psychological functioning, 218-219 nonclassical models of estrogen action, 19-21, 20f osteoarthritis prevention, 394 cellular research, 394-395 endogenous estrogen and, 397 epidemiologic research, 397-399 estrogen receptor presence in chondrocytes, 394, 394f hormone therapy and, 397-399, 397t, 398t mediators of estrogen action on articular cartilage, 395,395t, 396f preclinical research, 395-396 osteoporosis, deficiency in, 328-329 osteoporosis treatments, 355-356, 378 adverse effects of, 381 androgens plus estrogen, 387-388 bone at different stages of menopause, effects on, 380-381,380f, 381f bone remodeling, effects on, 378-379, 379f combined hormone therapy, 379 definition of estrogen, 378 fractures, effects on, 381 low- and ultra-low-dose therapy, 381- 382 use of, 382 pharmacology of estrogen preparations, 769-774, 770t, 771t postmenopause and cognitive health, 287-288 production and action of, 767-769, 768f receptors. See Estrogen receptors stress urinary incontinence treatment, 709-710 synthetic estrogen preparations, 770-771 urinary incontinence treatments, 257-258
vulvovaginal complaints and estrogen loss, 264 Women's Angiographic Vitamin and Estrogen Trial, 538-539 Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial, 539 Women's Health Initiative Memory Study, 289 Estrogenic changes in menopause transition, 159 Estrogen in the Prevention of Atherosclerosis Trial, 539-540, 539f, 540f Estrogen in the Prevention of Reinfarction Trial, 532-533,533t Estrogen-progestogen therapy for osteoporosis, 378 adverse effects of, 381 androgens plus estrogen, 387-388 bone at different stages of menopause, effects on, 380-381,380f, 381f bone-remodeling, effect on, 379 combined hormone therapy, 379 definition of progestogen, 378 fractures, effects on, 381 low- and ultra-low-dose therapy, 381-382 use of, 382 Estrogen receptors, 17-18 coactivator protein association, 24-25, 25f corepressor protein association, 25 domain structures of, 18f ligand regulation of function, 22- 23, 23f models of estrogen action, 18-19, 18f nonclassical models of estrogen action, 19-21, 20f nongenomic responses, 19-20 osteoarthritis prevention, 394, 394f protein association, 24- 25 selective estrogen receptor modulators and Alzheimer's disease, 301-302 bone turnover marker changes, 344 osteoporosis treatments, 366- 367 subtypes, 21, 21f, 22 updated model of action, 25-26 Estrogen Replacement Atherosclerosis Trial, 538 Ethisterone, 783 Ethnic differences, 81- 82, 273 Etonogestrel, 788 Etonorgestrel implants, 177 Evening primrose, 686 Exercise, 655, 675-676, 688 cardiovascular disease CHD risk reduction, 432-434, 434f role in, 659-663,661f, 662f depression and, 674-675 diabetes and, 666 hot flashes, 195 osteoporosis and, 324, 365-366, 670-671,671f pelvic floor exercises and stress urinary incontinence, 708-709 pelvic organ prolapse treatment, 754 Exercise echocardiography, 432 F Factor V Leiden, 484 Fecundity, 111 Fibrates, 436 Fibrin formation and fibrinolysis and hormone therapy, 484-485,485f Fibrin monomer, 484 Fibrinogen degradation products, 485 Fibrinolysis markers, effects of hormone therapy on, 474-477, 475f, 476f, 477f
SUBj~CX INDEX
886 Finger temperature and hot flashes, 189-190 Fluoride treatments for osteoporosis, 362-363 Folliculogenesis, 52-54, 53f, 54f Food, Drug and Cosmetics Act of 1938, 864 Food and Drugs Act, 864 Fragile X syndrome (CGG"), 42 FSH chronobiology, 134 monotropic rise in FSH, 128, 128f ovarian anatomy and physiology, 54-56, 55f ovary reserve and endocrine parameters, 59-61, 59f, 60f, 61f perimenopause, changes during. See Endocrinology regulatory peptides, 129-131,130t, 131f FSH-[3 mutations, 37- 38 Functional incontinence, 697 G Gabapentin for interstitial cystitis, 723 Galactosemia, 41 - 42, 102 Gated myocardial perfusion single-photon emission computed tomography, 432 GDF9 mutations, 39 Gender cardiovascular disease and pathogenesis of cardiac ischemia, 409 CHD differences, 417-418, 417f, 418f, 419f clinical presentation, 418-419, 419f coronary findings, 420, 420t diagnostic invasive studies, use of, 419-420, 420f management of, clinical and surgical, 420-421 mechanisms of differences, 421 - 423,422f outcome and prognosis, 421 life expectancies by gender, 6 - 7 smoking differences, 8 Genetics autosomal chromosomal abnormalities, 37 autosomal dominant POF, 43 autosomal genes. See Autosomal genes autosomal recessive POF, 43 bone mass, influences on, 353 cardiovascular disease and, 415-416 cytogenetics of X chromosomal aberrations. See Cytogenetics of X chromosomal aberrations incidence of premature ovarian failure due to genetics, 42- 43 monosomy X, 30 gonads, 31, 31f secondary sexual development, 31- 32 osteoporosis, 323-324, 353t ovarian differentiation, 29-30 ovarian genes on the X, 36-37 ovarian maintenance genes, localizing to specific regions of the X, 32-33 polygenic control over oocyte number, 30 secondary sexual development, 31- 32 stochastic control over oocyte number, 30 X chromosomal mosaicism, 32, 32t, 33f Genital symptoms of incontinence, 258 Genuine stress incontinence (GSI), 696 Germ cell failure in both sexes (nonsyndromic), 41 Germ cell failure in both sexes (syndromic), 41 Germinal epithelium, 51 Gestagen. See Progestogens Gestodene, 789- 790 Gestogen. See Progestogens Giggle incontinence, 698 Ginseng, 686
Glucose and estrogen, 424 Glucose metabolism, 501,505- 506 DHEA and, 825 hormonal therapy and diabetes, 504-505, 505f metabolic syndrome, 410-411,410f, 501-504, 502f GnRH stimulation testing as probe of pituitary aging, 137-138, 138f Gonadotropin secretion or action chronobiology after menopause, 139-141,140f chronobiology of and sleep effects, 132-133 defects in premature ovarian failure, 103 hot flashes, 191 menopause transition changes in levels, 162 neuroreproductive axis in postmenopause, 138-139 synchronicity of other hormones with nighttime gonadotropin secretion, 133-134 in young women, 132-134 Gonads, 31, 31f cerebellar ataxia with XX gonadal dysgenesis, 39-40 malformation syndromes associated with XX gonadal dysgenesis, 40 Graafian follicles, 52-54, 53f Greene Climacteric Scale, 642-643, 644t, 645t GSI. See Genuine stress incontinence Gynecoid obesity, 409 H
Haversian systems, 354 HDL cholesterol and cardiovascular disease, 656-659, 661-662 Headaches and oral contraceptives, 174-175 Health care centers, 855, 860-861 adult women's generalist, 857-858, 858f clinic design, 859, 859f clinical research, 859- 860 consumer education, 859-860, 860f paramedical staff, 858-859, 858f patient reporting and compliance, 859 primary target population, 856-857, 857f principles of, 855-856 specialists and consultants, 858, 858f staffing, 857 Heart rate and hot flashes, 189 Height and age of menopause, 3 Helical peptide, 340 Hemostasis and hormone therapy, 474-477, 475f, 476f, 477f, 481,488 coagulation and anticoagulation, 483-484, 483f, 484f coagulation and clinical syndromes, 486-487 fibrin formation and fibrinolysis, 484-485,485f platelet factor, 482-483, 483f thrombosis clinical considerations, 487, 487f, 488f clinical markers of, 485-486 vessel wall and, 481-482, 482f Heparin for interstitial cystitis, 724 Herbs black cohosh, 685-686 chaste-tree berry, 687 Chinese herbs, 687 dong quai, 686 evening primrose, 686 ginseng, 686 kava, 686 licorice, 686- 687 red clover, 687
sage, 687 vitex, 687 High altitudes and age of menopause, 3 High-density lipoproteins, 462-463 Hilum of the ovary, 52 Historical view of menopause, 1 Hormone replacement therapy, 471-472, 477-478, 765 age at starting treatment, 633 Alzheimer's disease and midlife, 297 old age, 297-298, 298t prevention of AD, observational studies, 298 prevention of AD, WHIMS, 298-300, 299t treatment of AD, 300 benefits of, 634 breast cancer and, 573-576, 574f, 574t, 576t, 599-601,628t-629t, 630 cardiovascular disease and, 529 - 530, 530f, 559-560 arterial imaging trials, 537-541,538t cardioprotective effect according to body mass index, evidence for, 550-551 cardioprotective effect according to duration of, evidence for, 542-545, 543t, 544f, 544t cardioprotective effect according to timing of initiation of therapy, 545-547 cardioprotective effect requires healthy endothelium, evidence for, 547-548, 547f, 548f CEE+MPA therapy, 548-550, 549f, 550f clinical perspective on, 555-559, 556t, 557t clinical trials and observational studies, reconciling of findings, 620- 622, 621t, 622t morbidity and mortality, 530-537, 531t, 554-555,554f, 555f, 555t, 628t-629t, 631 randomized controlled trials and observational studies, discordance between, 541-542, 541t stroke and, 551-552, 552t venous thromboembolism and, 552-554, 553t, 554t Women's Health Initiative, 533-537, 533f, 534f, 535f, 536t, 537f, 613 cell adhesion molecules, effects on, 473-474, 473f, 474t chemokines, effect on, 473-474 clinical trial requirements, 863-864 cardioprotection and, 872 endometrial hyperplasia study requirements, 867- 869, 870t future directions, 872-873 history of, 864 osteoporosis studies, 869-871,871f regulatory requirements, 865- 866 research, 864-865, 865f safety assessments, 871-872 vasomotor symptoms/vaginal atrophy study requirements, 867, 867f, 868f clinical trials and observational studies, reconciling of findings, 619, 625,625t biologic explanations, 624-625,624f cardiovascular disease, effect of HT on, 620-622, 621t, 622t disease endpoints, 622 effects of liT, 619-620 explanations for discordant findings, 623-625 methodological explanations, 623-624, 623f
887
SUBJECT INDEX cognition and, 632 colorectal cancer and, 601-606, 602t, 604t-605t, 628t-629t, 630 connective tissue changes. See Connective tissue changes coronary artery atherosclerosis, effects of in early menopause, 517-518,517f, 518f C-reactive protein, effects on, 472-473,474t cytokines, effects on, 473-474 death and, 628t-629t, 632 decision to treat with, 877-879, 878f dementia and, 289, 628t-629t, 632 diabetes and, 504-505,505f dropout and drop-in rates, 634 endometrial cancer after treatment for, 589-590, 590t effects of, 587-589,588t morbidity and mortality changes with, 628t-629t, 630 endometrial hyperplasia study requirements, 867-869, 870t exercise, effects of, 662-663 fibrinolysis markers, effects on, 474-477, 475f, 476f, 477f future direction of therapy, 815-817, 876f, 876t, 879-880 future of, 875-877 healthy user effect, 632 hemostasis, effects on, 474-477, 475f, 476f, 477f, 481,488 coagulation and anticoagulation, 483-484, 483f, 484f coagulation and clinical syndromes, 486 -487 fibrin formation and fibrinolysis, 484-485, 485f platelet factor, 482-483,483f thrombosis, clinical considerations, 487, 487f, 488f thrombosis, clinical markers of, 485-486 vessel wall and, 481-482,482f hot flashes, 195 inflammation markers, effects on, 472-474, 472f Leisure World Cohort Study, 627-628 lipids and lipoproteins cardiovascular disease risks, effects on, 466-467 cholesterol levels, effects on, 464-466 long-term effects, 633 lung cancer and, 606 Menopause Rating Scale as outcome measure for, 651,651f in menopause transition, 164-165 morbidity and mortality, 627 differing results, 632-634 outcomes, 630-632 studies, 627-630 Nurses' Health Study, 629-630 osteoarthritis prevention, 397-399, 397t, 398t osteoporosis, 356-358, 357t, 378, 672-673 adverse effects of, 381 bone at different stages of menopause, effects on, 380-381,380f, 381f bone-remodeling, effect on, 379 combined hormone therapy, 379 definition of progestogen, 378 fractures, effects on, 381 low- and ultra-low-dose therapy, 381-382 morbidity and mortality changes with, 628t-629t, 630-631
use of, 382 Women's Health Initiative trial, 612-613, 612f, 613t Papworth Hormone Replacement Therapy Atherosclerosis Study, 532 Postmenopausal Hormone Replacement against Atherosclerosis Trial, 540-541 preexisting disease, 633 risks of, 634 role of in future, 875- 877, 876f, 876t, 879-880 short-term effects, 633 skin, effects on, 239t, 245-246 stroke and cardiovascular disease, 551-552, 552t morbidity and mortality changes, 628t-629t, 631 Women's Health Initiative hormone therapy trial, 613 tibolone and, 477 vasomotor symptoms/vaginal atrophy study requirements, 867, 867f, 868f venous thromboembolism morbidity and mortality changes, 628t-629t, 631 risks, 487, 488,496-498,496t, 497t, 552-554, 553t, 554t Women's Health Initiative trial, 613 vulvovaginal complaints, 267 - 268 Women's Health Initiative trial, 611-612 cancer, 613-614 cardiovascular disease, 613 cognitive decline, 614 menopausal symptoms, 614-615 morbidity and mortality changes, 627 osteoporosis, 612-613,612f, 613t stroke, 613 venous thrombosis, 613 Hot flashes, 3-4, 3t, 4f, 196 ambulatory monitoring, 190 antidepressants, 196 behavioral treatments, 195 blood flow, 187 catecholamines, 191-192, 192f circadian rhythms, 194 core body temperature, 188-189, 189f depression in the perimenopause, 313 descriptive physiology, 187 - 189 endocrinology, 190-192 epidemiology of, 187 estrogens, 190-191 exercise, 195 finger temperature, 189 - 190 gabapentin, 196 gonadotropins, 191 heart rate, 189 hormone therapy, 195 lifestyle modifications, 195 in menopause transition, 163 metabolic rate, 189 objective measurement, 189 - 190 opiates, 191 physiologic data, 187-189 phytoestrogens, 195,684 provocation techniques, 190 self-reported data, 187 skin conductance, 188, 188f, 189f, 190 skin temperature, 187 sleep, 194 - 195 sweating, 188, 188f, 189f thermoregulation, 192-194, 193f, 193t treatment of, 195-196
Hyaluronic acid for interstitial cystitis, 724 17o~-hydroxylase deficiency and premature ovarian failure, 102 17o~-hydroxylase/17,20 Desmolase Deficiency (CYP17), 41
Hydroxyzine for interstitial cystitis, 723 Hypercoagulability, VTE risks, 494-495 Hypergonadotropic amenorrhea. See Premature ovarian failure Hypergonadotropic hypogonadism. See Premature ovarian failure Hyperlipidemia and cardiovascular disease, 412, 412f Hypertension, 412-413,434-435
I Ibandronate treatments for osteoporosis, 384-385, 385t IC. See Interstitial cystitis Idiopathic detrusor overactivity (IDO), 697 Idiopathic premature ovarian failure, 105 IDO. See Idiopathic detrusor overactivity Immune disturbances and premature ovarian failure, 103-105, 104t Implantable contraceptives, 176 - 177 IMT, 432 Incontinence. See Urinary incontinence Infectious cystitis, 726 Inflammation markers coronary artery atherosclerosis and, 519 - 520, 520t, 521t hormone therapy on, effects of, 472-474, 472f Inhibin alpha (INHA) mutations, 38 - 39 Inhibin changes in menopause transition, 159-160 Injectable contraception, 175-176 Insulin resistance syndrome, 410-411,410f, 501-504, 502f Integrative medicine. See Complementary and alternative medicine International differences, 81 - 82 Interstitial cystitis (IC), 717 amitriptyline for, 723 Atarax for, 723 Bacillus Calmette Gudrin for, 725 bladder augmentation, 725 botulinum toxin injections, 725 Capsaicin for, 724-725 clinical presentation, 718 - 719 cystectomy, 725 Cystistat for, 724 cystoscopy, 720- 721 diagnosis, 719 - 722 dimethyl sulfoxide for, 724 Elavil for, 723 Elmiron for, 723, 724 epidemiology, 717- 718 gabapentin for, 723 heparin for, 724 hyaluronic acid for, 724 hydrodistention, 720- 721 hydroxyzine for, 723 intravesical instillation, 724-725 laboratory studies, 719 laser therapy, 725 local anesthetics for, 724 multiple agents, mixtures of, 725 Neurontin for, 723 NIDDK criteria, 719, 719t oral therapy, 723 pathophysiology, 718
888 Interstitial cystitis (continued) physical examination, 719 potassium sensitive test, 721-722 Resiniferatoxin for, 724-725 sacral neuromodulation, 725 sodium pentosan polysulfate therapy, 723, 724 steroids for, 724 supportive management, 723 surgical therapy for, 725 symptoms scale, 719-720, 720t, 721t, 722t transcutaneous electrical nerve stimulation, 725 treatment, 722- 725 urinary markers, 722 urodynamics, 722 voiding diary, 719- 720 Intervertebral discs, 233, 234f Intraovarian control, 57, 57f Intrauterine devices, 175 Intrinsic sphincter deficiency (ISD), 699-700 ISD. See Intrinsic sphincter deficiency Isochromosomes for Xq (i[Xq]), 34, 35t Isoflavone treatments for osteoporosis, 367-368 K
Kava, 686 Kefauver-Harris Amendment to Food, Drug and Cosmetics Act, 864 L Laser therapy for interstitial cystitis, 725 LDL cholesterol and cardiovascular disease, 656-659 Leiden mutation, 492 Leisure World Cohort Study, 627-628 Levator ani muscles, 740-741, 741f Levonorgestrel family, 783, 787-788, 787f Levonorgestrel implants, 176 - 177 Levonorgestrel-releasing intrauterine devices, 175 LH ovarian aging effects on, 131-132 role of, 56-57, 56f LH receptors induction, 55- 56 mutations, 38 Libido, 258-259 Licorice, 686-687 Life expectancies cardiovascular disease and trends in, 439-443, 439f, 440f, 441f, 442f historically, 6, 7-8, 50f Life span, 7-8 Lifestyle, 637 Ligands estrogen receptor function regulation, 22-23, 23f progesterone receptor function regulation, 22-23, 23f Lipids and lipoproteins, 461 abnormalities in cardiovascular disease, 411-412, 411f, 412f atherogenic lipoproteins, 461-462, 462f CHD risk reduction, 435-438, 435t, 436t, 437t coronary artery atherosclerosis atheroprotection mechanisms of estrogen, 518-519 role of, 512, 512f, 515 estrogen and, 424, 456, 463,464f coronary artery atherosclerosis, atheroprotection mechanisms of estrogen, 518-519
SUBJECT INDEX Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial, 539 high-density lipoproteins, 462-463 hormone therapy cardiovascular disease risks, effects on, 466 -467 cholesterol levels, effects on, 464-466 progestin, effect on, 464 Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial, 539 Lower urinary tract disorders. See Urinary tract disorders Lung cancer and hormone therapy, 606 M
Magnetic resonance imaging (MRI), 432, 708 Malformation syndromes associated with XX gonadal dysgenesis, 40 Masked incontinence, 698 Matrix metalloproteinases (MMP), 341 McCall culdoplasty, 757, 757f MCI. See Mild cognitive impairment Medroxyprogesterone acetate, 785-786, 786f Medulla of ovary, 51, 52 Melatonin, 834 chemistry of, 829-830, 830f circadian rhythms and, 833 compounds that alter secretion, 832 detection in blood, 830-831 metabolism of, 829-830, 830f pathophysiology of secretion, 832 physiology of secretion, 831- 832 receptors, 831 regulation of, 830-831 shift work and, 833 sleep disturbances and, 833 suprachiasmatic nucleus, effects on, 831 synthesis of, 829-830, 830f unproven benefits from, 833-834 Memory DHEA and, 825-826 effect of sex steroids on, 207 estrogen and objective measure of, 280-281 and menopause transition, 279- 280 Women's Health Initiative Memory Study, 288 -289 Menarche age of, 3 breast cancer risk factors, 570 Mendelian disorders characterized predominately by somatic features, 40, 40t Menopausal Quality of Life Scale (MQOL), 644t, 646t- 647t, 648- 649 Menopausal Symptom List (MSL), 644t, 646t, 647 Menopause, 10 age of, 2-3 bone turnover markers, changes in, 342-343 changes over time, 50f and cognition. See Cognitive function connective tissue changes. See Connective tissue changes defined, 2, 157 demographic characteristics, 80-81 depression and characterizing reproductive status, 309 DHEA in, 822-824, 822f, 823f, 823t environmental influences, 83-90 ethnic and international differences, 81-82
factor analysis, 642 glucose metabolism after. See Glucose metabolism Greene Climacteric Scale, 642-643, 644t, 645t hot flashes. See Hot flashes Menopausal Q.uality of Life Scale, 644t, 646t- 647t, 648- 649 Menopausal Symptom List, 644t, 646t, 647 Menopause Rating Scale, 644t, 646t, 647-648, 649t, 651,651f Menopause-Specific Q.uality of Life QFestionnaire, 644-647, 644t, 646t as opportunity, 9 origin of word, 2 Pan-Asia Menopause Study, 651-652, 652t pelvic organ prolapse and, 749 Q.ualifemme, 643-644, 644t, 645t signal for the future, 9-10 skin and. See Skin surgical menopause, 273- 274 symptom assessments, 640-652 symptoms of, 78 Utian Quality of Life Scale, 644t, 647t, 649 Women's Health Initiative hormone therapy trial and symptoms of, 614-615 Women's Health Questionnaire, 643, 644t, 645t Menopause Rating Scale (MRS), 644t, 646t, 649-648, 649t, 651,651f Menopause-Specific Q.uality of Life Q.uestionnaire (MENQ.OL), 644-647, 644t, 646t Menopause transition, 149-150, 150t, 155, 157-159, 165 aging, changes associated with, 159 amenorrhea and, 158 androgenic changes, 160-161 bone mineral density changes, 163-164 cardiovascular changes, 163 cycle length changes, 162 defined, 157-158 depression association between, evidence of, 311-312 risk factors, 312 - 313 estrogenic changes, 159 gonadotropin level changes, 162 hormone therapy, 164-165 hot flushes, 163 inhibin changes, 159-160 investigation and treatment, 154-155 and memory, 279-280 menstrual blood loss, 152-153,153f menstrual changes during, causes of, 153-154 menstrual patterns during, 150-152, 151f metabolic changes, 161-162 postmenopause bleeding episodes, 152 progestogenic changes, 159 smoking and, 158 somatotropic axis changes, 161 surgical management, 155 MENQOL. See Menopause-Specific Q.uality of Life Q.uestionnaire Menstrual cycle, 132, 133t changes during menopause transition, causes of, 153-154 length changes in menopause transition, 162 menopause transition, patterns during, 150-152, 15 If older cycling women, 134-136, 135f, 136f Menstrual function and osteoporosis, 324 Menstrual history, 80- 81
889
SUBJECTINDEX Mental health Alzheimer's disease. See Alzheimer's disease dementia. See Dementia depression. See Depression menopause, symptoms of, 5 psychological functioning. See Psychological functioning Metabolic changes in menopause transition, 161-162 Metabolic rate and hot flashes, 189 Metabolic syndrome, 410-411,410f, 501-504, 502f, 665-666, 666f Metropathia hemorrhagica, 151 Micturition, 253-254, 255f, 699 Mild cognitive impairment (MCI), 295 MIS. See Mtillerian inhibitory substance Mitosis and the ovary, 55 Mixed urinary incontinence, 697 pathophysiology, 700 treatment for, 717 MMP. See Matrix metalloproteinases Monosomy X, 30 gonads, 31, 31f secondary sexual development, 31- 32 Mood and the climacteric, 205-207 and sex steroids, 219-221,219f, 221f MQ_OL. See Menopausal Quality of Life Scale MRI. See Magnetic resonance imaging MRS. See Menopause Rating Scale MSL. See Menopausal Symptom List Miillerian inhibitory substance (MIS), 116, 130-131 Multichannel urodynamics, 705-707, 706f Myocardial infarction and oral contraceptives, 173 Myotonic dystrophy (CTG"), 41 N Natural medicine. See Complementary and alternative medicine Negative view of menopause, 1-2 Neoplasia. See Cancer Neurogenic detrusor overactivity, 697 Neurontin for interstitial cystitis, 723 Neuroreproductive axis in postmenopause, 138 gonadotropin chronobiology after menopause, 139-141,140f gonadotropin secretion, 138-139 ovarian steroid activity of the postmenopausal ovary, 141-142, 142f Neurosteroids, 300 NHS. See Nurses' Health Study Niacin, 436 NIDDK criteria in interstitial cystitis, 719, 719t Nocturnal enuresis, 697 Nongenomic responses, 19-20 19-Nonpregnane derivatives, 783 Non-traditional medicine. See Complementary and alternative medicine Norelgestromin, 788 Norethindrone family, 783, 786-787, 786f Norethisterone, 783 Norgestimate, 788 Nurses' Health Study (NHS), 629-630 Nutrition, 655, 675-676 cardiovascular disease, role in, 655-659, 656t, 657t, 658f depression and, 674 diabetes and, 667 osteoporosis and, 324, 328, 667-670
O OA. See Osteoarthritis OAB. See Overactive bladder Obesity cardiovascular disease and, 409-411,410f, 411f, 432- 434, 663- 665 diabetes and, 666 endometrial cancer and, 589 Obstetric management and delivery considerations in oocyte donation, 121f Occupational factors, 83 O'Leary-Sant Symptom Index, 720t Oligomenorrhea, 136-137, 137f Oocyte donation, 117-118 childbearing by perimenopausal and menopausal women, 118-119, 119f future applications, 121 obstetric management and delivery considerations in, 120-121, 121f recipient screening and preparation, 119-120, 119t, 120t Oocytes, 51 polygenic control over number, 30 stochastic control over number, 30 Opiates and hot flashes, 191 Oral contraceptives, 169-171,170t, 171f adverse events associated with, 173-175 benefits of in women over 35, 172-173 bone mineral density, 173 breast cancer, 174 cerebrovascular disease, 173 ectopic pregnancy, reduced incidence of, 172 efficacy, 171-172, 172f endometrial neoplasia, 172-173 headaches, 174-175 myocardial infarction, 173 ovarian neoplasia, 172-173 pelvic inflammatory disease, 172 uterine bleeding, 173 venous thromboembolism, 173 - 174 weight gain, 174 Osteoarthritis (OA), 399-400 defined, 393 estrogen and prevention of, 394 cellular research, 394- 395 endogenous estrogen and, 397 epidemiologic research, 397- 399 estrogen receptor presence in chondrocytes, 394, 394f hormone therapy and, 397-399, 397t, 398t mediators of estrogen action on articular cartilage, 395, 395t, 396f preclinical research, 395- 396 incidence of, 393- 394 prevalence of, 393-394 Osteocalcin, 339 Osteoclastic enzymes, 341 Osteons, 324, 326, 326f Osteopenia, 332 Osteoporosis, 323,377 alcohol and, 328 alendronate treatments, 383-384, 384f, 385t anabolic steroid treatments, 387, 387f androgen treatments, 364-365 bisphosphonate treatments, 358-359, 383-385,385t bone formation and, 354-355 bone remodeling, 324-328, 325f, 326f, 327f, 333-334, 334f, 334t, 353-355 bone resorption and, 354 calcitonin treatments, 363-364, 385-387, 386f
calcium and, 328, 359-360, 615,672-673 candidates for treatment, 389 cellular basis of bone loss, 324-328, 325f, 326f, 327f choice of treatment, 390 chronic diseases and, 328-329 classification of, 352t clinical studies, 869-871,871f defined, 332, 351-352 environmental influences on bone mass, 353,353t estrogen deficiency, 328-329 estrogen-progestogen therapy, 378 adverse effects of, 381 and androgens, 387-388 bone at different stages of menopause, effects on, 380-381,380f, 381f bone-remodeling, effect on, 379 combined hormone therapy, 379 definition of progestogen, 378 fractures, effects on, 381 low- and ultra-low-dose therapy, 381-382 use of, 382 estrogen treatments, 355-356, 378 adverse effects of, 381 androgen and estrogen, 387- 388 bone at different stages of menopause, effects on, 380-381,380f, 381f bone remodeling, effects on, 378-379, 379f combined hormone therapy, 379 definition of estrogen, 378 fractures, effects on, 381 low- and ultra-low-dose therapy, 381-382 use of, 382 exercise and, 324, 365-366, 670-671,671f fall-related fractures, 329, 331 fluoride treatments, 362-363 fractures calcitonin and, 386 estrogen and estrogen-progestogen replacement therapy, effects on, 381 fall-related fractures, 329, 331 fragility fractures and, 352 morbidity and mortality changes with, 628t-629t, 630- 631 fragility fractures and, 352 genetic factors, 323-324, 353t genetic influences on bone mass, 353 hormone replacement therapy, 356-358, 357t, 378, 672-673 adverse effects of, 381 bone at different stages of menopause, effects on, 380-381,380f, 381f bone-remodeling, effect on, 379 combined hormone therapy, 379 definition of progestogen, 378 fractures, effects on, 381 low- and ultra-low-dose therapy, 381-382 morbidity and mortality changes with, 628t-629t, 630- 631 use of, 382 Women's Health Initiative trial, 612-613, 612f, 613t ibandronate treatments, 384-385,385t isoftavone treatments, 367-368 length of treatment, 389- 390 low bone density and, 352 magnitude of diseases, 351 medications and, 329 menopause signal for the future, 9-10 menstrual function and, 324
890 Osteoporosis (continued) nutrition and, 324, 328, 667-670, 669t parathyroid hormone treatments, 361-362, 388, 388t pathogenesis of bone loss, 324-328, 325f, 326f, 327f peak bone mass and its determinants, 323-324 physical activity and, 328 phytoestrogen treatments, 367, 685 raloxifene treatments, 382-383 risedronate treatments, 384, 385t selective estrogen receptor modulator treatments, 366-367, 382-383 smoking and, 328 strontium ranelate treatments, 368-369, 388-389, 389t tibolone treatments, 364, 383 trace elements, role in, 673-674 vitamin D, 328, 360-361,615-616 Women's Health Initiative Calcium Vitamin D Trial, 615 Women's Health Initiative hormone therapy trial, 612-613, 612f, 613t Ovarian cancer, 593 oral contraceptives and, 172-173 risk assessment heritable factors, 594-595 modifiable risk factors, 593-594 screening, 595- 597 staging system for, 594t Ovarian failure autosomal chromosomal abnormalities, 37 autosomal dominant POF, 43 autosomal genes. See Autosomal genes autosomal recessive POF, 43 coronary artery atherosclerosis, role of, 512-514, 513f, 514f cytogenetics of X chromosomal aberrations. See Cytogenetics of X chromosomal aberrations incidence of premature ovarian failure due to genetics, 42- 43 monosomy X, 30 gonads, 31, 31f secondary sexual development, 31-32 ovarian differentiation, 29- 30 ovarian genes on the X, 36-37 ovarian maintenance genes, localizing to specific regions of the X, 32- 33 polygenic control over oocyte number, 30 secondary sexual development, 31-32 stochastic control over oocyte number, 30 X chromosomal mosaicism, 32, 32t, 33f Ovaries, 49-50, 63-64 anatomy, 51-52, 52f cancer. See Ovarian cancer corpus albicans, 52 corpus luteum, 51 cortex, 51, 52f determinants of reproductive aging, 128-132 failure. See Ovarian failure folliculogenesis, 52-54, 53f, 54f FSH, role of, 54-56, 55f hilum, 52 intraovarian control, 57, 57f LH, role of, 56-57, 56f LH receptor induction, 55-56 life expectancy and age of menopause, changes over time, 50f medulla, 51, 52 mitosis and, 55 oocytes, 51
SUBJECTINDEX ovarian steroid activity of the postmenopausal ovary, 141-142, 142f physiology, 52-54, 53f, 54f primordial follicle, 50, 5Of, 51f ramus ovaricus artery, 52 reserve. See Ovary reserve steroidogenic potential, 54- 55 stigma, 51 Ovary reserve, 57 accelerated loss, 61-63 activin hypothesis, 61-63 endocrine parameters, 59-61, 59f, 60f, 61f measurements, 115 regulation, 57-59, 58f, 58t Overactive bladder (OAB), 256, 697 Overflow urinary incontinence, 697 Ovulatory dysfunctional uterine bleeding, 155
P Paced respiration and relaxation, 688 Packets, 324, 326, 326f Pad tests, 704 PAM. See Pan-Asia Menopause Study Pan-Asia Menopause (PAM) Study, 651-652, 652t Papworth Hormone Replacement Therapy Atherosclerosis Study, 532 Parasympathetic pathways, 253 Parathyroid hormone treatments for osteoporosis, 361-362, 388, 388t Paraurethral collagen and stress urinary incontinence in women of reproductive age, 252 - 253 Paraurethral collagen and the menopause, 252 Pathogenesis of bone loss, 324-328, 325f, 326f, 327f Pelvic examinations and urinary incontinence, 702 Pelvic floor anatomy, 693-694, 694f Pelvic floor exercises and stress urinary incontinence, 708- 709 Pelvic inflammatory disease and oral contraceptives, 172 Pelvic organ prolapse (POP), 739 aging, 749 anorectal canal examinations, 752 support of, 743 anterior colporrhaphy, 759-760, 759f anterior compartment prolapse, 759-760 anterior vaginal wall defects, 748, 748f anterior vaginal wall examinations, 751- 752 apical defects, 747-748 apical prolapse, 757-759 bony pelvis anatomy, 740, 740f cervix, support of, 743 childbirth, 749 classification of, 743-745 clinical presentation, 750 compartmental classification of, 743-745,744t compensatory repair with graft materials, 760, 761 concomitant anti-incontinence procedures, 760 congenital defects, 749 conservative nonsurgical treatment, 753-754 enterocele, 743 enterocele repair, 761 epidemiology, 747 etiology, 749- 750 examinations, 751- 753 general examinations, 751 halfway system, 745, 745t
iliococcygeus suspension, 758 levator ani muscles, 740-741,741f McCall culdoplasty, 757, 757f menopause and, 749 modification system, 745, 745t nonsurgical treatment, 753- 756 obliteration procedure with colpocleisis, 761-762, 762f pathophysiology, 747- 750 pelvic muscle exercises as treatment, 754 pelvic support anatomy, 739-743 pelvic support defects, 747-749 pelvic support levels, 741-742 perineal body, 741 perineal body examinations, 752 perineum, 741,742f pessaries, 754-756, 755f, 755t posterior colporrhaphy, 760 posterior compartment prolapse, 760- 761 posterior vaginal wall defects, 748-749 posterior vaginal wall examinations, 752 potential incontinence, 753 quantification of, 745- 747 quantification system, 746, 746f, 747t recurrence of prolapse after surgery, 762 regression of prolapse, 750 sacral colpoperineopexy, 761 sacral colpopexy, 758-759, 759f sacrospinous ligament suspension, 758, 758f site-specific repair, 760, 761 surgeries, previous, 749- 750 surgical treatment, 756-762 systemic evaluations, 751 traditional classification of, 743,744f treatment, 753- 762 uterosacral suspension, 757-758, 758f vagina, support of, 743 vaginal apex examinations, 751 vaginal examinations, 751 Pelvic pain and urgency/frequency (PUF) Scale, 722t Pelvic support, 691 Pelvic support anatomy, 739-743 Perimenopause, 125 age at onset of, 79-80, 81 antimtillerian hormone as novel predictor of ovarian aging, 130-131 brain aging and reproductive senescence, 142-143 coronary artery atherosclerosis and, 514- 515, 515f animal model for studies of, need for, 516, 516f endothelial function and, 515 - 516, 516f plasma lipid and lipoprotein changes during, 515 defined, 127-128, 158 depression and. See Depression endocrine changes during, 67, 75 clinical sequelae of, 73-74 dynamics of, 68-73, 70f, 71f, 72f, 73f hormone-induced changes in, 74 normal reproductive cycle, 67-68, 68f, 69f profile assessment, 74 environmental influences, 83-82 ethnic differences, 81 FSH chronobiology, 134 FSH regulatory peptides, 129-131,130t, 131f future studies, 143-144, 144f GnRH stimulation testing as probe of pituitary aging, 137-138, 138f
891
SUBJECT INDEX gonadotropin changes during, 134-138 gonadotropin secretion in young women, 132-134 international differences, 81 LH, ovarian aging effects on, 131-132 monotropic rise in FSH, 128, 128f natural fertility in, 114-117, 115f, 116f, 117f neuroreproductive axis in postmenopause. See Neuroreproductive axis in postmenopause older cycling women, 134-136, 135f, 136f oligomenorrhea, age-related, 136-137, 137f ovarian determinants of reproductive aging, 128-132 synchronicity of other hormones with nighttime gonadotropin secretion, 133-134 transitional years, 2 Perineal body, 741 Perineum, 741,742f Periurethral bulking injections for stress urinary incontinence, 714-715, 714f Perrault syndrome, 39 Pessaries and pelvic organ prolapse, 754-756, 755f, 755t Pharmacologic treatment stress urinary incontinence, 255,709-710 urge urinary incontinence, 715- 716 Physical activity age at natural menopause and onset of perimenopause, 82- 83 exercise. See Exercise observational studies, 89t and osteoporosis, 328 presentation and symptoms of menopause, 86 Phytoestrogens, 88, 675, 683, 813 Alzheimer's disease and, 302 breast cancer risks, 684 cardiovascular disease risk factors, 685 endometrial cancer and, 684 hot flashes, 195, 684 osteoporosis treatments, 367, 685 risks, 685 vaginal epithelium and, 684 Pituitary aging, GnRH stimulation testing as probe of, 137-138, 138f Plasma lipids abnormalities in cardiovascular disease, 411-412, 411f, 412f coronary artery atherosclerosis atheroprotection mechanisms of estrogen, 518-519 role of, 512, 512f, 515 and estrogen, 424 Platelet factor and hormone therapy, 482-483, 483f POE See Premature ovarian failure Polygenic control over oocyte number, 30 POE See Pelvic organ prolapse Population rates, 6, 7t Posterior vaginal wall prolapse, 744 Postmenopausal Hormone Replacement against Atherosclerosis Trial, 540-541 Postmenopause, 185 bleeding episodes, 152 breast surveillance, 579-583,580f, 581f, 582f, 583f cardiovascular disease. See Cardiovascular disease cognitive health, 287, 291 dementia in women aged 65 and older, 288 -290 estrogen and, 287-288 future research, 291
menopause and, 287-288 younger postmenopausal women, 290 defined, 158 glucose metabolism. See Glucose metabolism gonadotropin chronobiology after menopause, 139-141,140f gonadotropin secretion, 138-139 hot flashes. See Hot flashes neuroreproductive axis in, 138-142 ovarian steroid activity of the postmenopausal ovary, 141-142, 142f Potassium sensitive test in interstitial cystitis, 721-722 Potential urinary incontinence, 698 Pregnane derivatives, 781-783 Premature menopause. See Premature ovarian failure Premature ovarian failure (POF), 97, 99 aromatase deficiency, 102 chemotherapeutic agents, 103 clinical features of, 99-101,100t cytogenetic abnormalities invoMng the X chromosome, 101 - 102 environmental insults, 103 enzymatic defects, 102 etiology of, 101-105,101t galactosemia, 102 genetic alterations causing, 102-103 gonadotropin secretion or action defects, 103 17c~-hydroxylase deficiency, 102 hypergonadotropic amenorrhea evaluations, 105-106 idiopathic premature ovarian failure, 105 immune disturbances, 103-105, 104t prevalence of, 101 radiation and, 103 resistant ovary syndrome, 105 smoking and, 103 treatment, 106 - 107 viral agents, 103 Presentation of menopause, 78 demographic characteristics, 80-81 environmental influences, 83-90 ethnic and international differences, 81- 82 Preventive health care, 8-9 Primary ovarian failure. See Premature ovarian failure Primary ovarian insufficiency. See Premature ovarian failure Primordial follicle, 50, 50f, 5 lf, 57 accelerated loss, 61- 63 activin hypothesis, 61-63 endocrine parameters, 59-61, 59f, 60f, 61f regulation, 57-59, 58f, 58t Procidentia, 743 Procollagen propeptides, 337-338, 338f Progestagen. See Progestogens Progesterone, 784-785, 785f Alzheimer's disease and, 300-301 development of, 815,815t effects of on brain, 210-211 progestins structurally related to, 781 - 783 Progesterone creams, 687 Progesterone receptors, 17-18 coactivator protein association, 24-25, 25f corepressor protein association, 25 domain structures of, 18f isoforms, 21 ligand regulation of function, 22-23, 23f models of progesterone action, 18-19, 18f nonclassical models of progesterone action, 19-21, 20f
protein association, 24-25 subtypes, 21, 21f updated model of action, 25-26 Progestins, 779 breast cancer and hormone therapy, 575-576 effects of on brain, 210-211 Heart and Estrogen/Progestin Replacement Study, 530-532, 531f, 532f lipids and lipoproteins, effects on, 464 neurobiologic effects of, 218- 219 structurally related to norethindrone, 787 structurally related to progesterone, 781-783 structurally related to testosterone, 783-784 Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial, 539 Women's Health Initiative Memory Study, 289 Progestogenic changes in menopause transition, 159 Progestogens, 779-780, 796 androgenic activity, 795 anti-androgenic activity, 795 anti-mineralocorticoid activity, 795- 796 bioassays of potency, 793- 794 classification of, 780-781,780f, 780t clinical tests of potency, 794-795, 794t defined, 378 desogestrel, 788- 789 development of, 815,815t dienogest, 790 drospirenone, 790 estrogenic activity, 795 estrogen-progestogen replacement therapy for osteoporosis. See Osteoporosis ethinylated derivatives, 783-784 gestodene, 789- 790 intramuscular administration, 791 intravaginal administration, 791 levonorgestrel, 787-788, 787f medroxyprogesterone acetate, 785-786, 786f non-ethinylated derivatives, 784 19-nonpregnane derivatives, 783 norethindrone, 786-787, 786f norgestimate, 788 percutaneous administration, 791-792 pharmacokinetics of orally administered progestogens, 784-790, 785t pharmacokinetics ofparenterally administered progestogens, 791- 792 potency of, 792-796 pregnane derivatives, 781 - 783 progestational activity, 792-795 progesterone. See Progesterone progestins structurally related to norethindrone, 787 progestins structurally related to progesterone, 781-783 progestins structurally related to testosterone, 783-784 receptor binding tests, 792-793, 793t structure-function relationships of, 781-784, 781f, 782f, 783f Proteins estrogen receptor association, 24- 25 progesterone receptor association, 24-25 Psychological functioning, 217, 223 - 224 androgen, neurobiologic effects of, 218-219 cognitive functioning. See Cognitive function dementia. See Dementia depression. See Depression epidemiology, 217- 218 estrogen
892 Psychological functioning (continued) and cognitive functioning, 221 neurobiologic effects of, 218-219 progestin, neurobiologic effects of, 218-219 self-image, 245 sex steroids and mood, 219-221,219f and sexuality, 221-223,222f skin, changes to, 245 PUF Scale. See Pelvic pain and urgency/frequency scale Pulmonary embolism. See Venous thromboembolism Pyelonephritis, 726
Q. QCT. See Quantitative computerized tomography Qztip tests and urinary incontinence, 703 Qpalifemme, 643-644, 644t, 645t Quality of life, 637, 639-640, 652-653 assessing, 640-652 Berlin Study, 649-651,650f, 650t, 651f factor analysis, 642 Greene Climacteric Scale, 642-643, 644t, 645t Menopausal Quality of Life Scale, 644t, 646t- 647t, 648- 649 Menopausal Symptom List, 644t, 646t, 647 Menopause Rating Scale, 644t, 646t, 647-648, 649t, 651,651f Menopause-Specific Quality of Life Qpestionnaire, 644-647, 644t, 646t Qualifemme, 643-644, 644t, 645t scales and subscales, construction of, 641-642 sexuality and, 652 urinary incontinence questionnaires, 702 Utian Quality of Life Scale, 644t, 647t, 649 Women's Health Q.uestionnaire, 643,644t, 645t Q.uantitative computerized tomography (Q.CT). 335 -336 R Race and age of menopause, 3 Radiation and premature ovarian failure, 103 Radiologic studies and urinary tract infections, 728 Raloxifene breast, effects on, 841,841f cataracts, effects of, 842 gallbladder disease, effects of, 842 gynecologic effects, 841- 842 insulin sensitivity, effects of, 842 leiomyomas, effects on, 842 menopausal symptoms and, 841-842 minor side effects, 842 osteoporosis treatments, 382- 383 skeleton and bone metabolism, effects on, 839-841,840f vascular disease and intermediates, effects of, 841 Ramus ovaricus artery, 52 Randomized clinical trials. See Clinical trials Rectangularization of life, 6, 7-8 Rectocele, 743 Recurrent infection, 726 Recurrent UTI, 726 Red clover, 687 Relapse of infection, 726 Relaxation therapy, 688 Reproductive history, 80-81 Reproductive options, 111-114,112f, 113f, 113t, 114f childbearing by perimenopausal and menopausal women, 118-119 embryo donation, 117-118, 118f
SUBJECT INDEX natural fertility in perimenopause, 114-117, l15f, ll6f, l17f oocyte donation. See Oocyte donation Reproductive senescence and brain aging, 142-143 Resiniferatoxin for interstitial cystitis, 724-725 Resistant ovary syndrome, 105 Risedronate treatments for osteoporosis, 384, 385t
S Sacral neuromodulation interstitial cystitis, 725 urge urinary incontinence, 716-717 Sage, 687 Secondary sexual development, 31-32 Secondary VTE, 495 Sedentary lifestyle and cardiovascular disease, 415 Selective estrogen receptor modulators (SERMs), 837, 842 Alzheimer's disease and, 301-302 bone turnover marker changes, 344 development of, 842 osteoporosis treatments, 366-367, 382-383 raloxifene. See Raloxifene tamoxifen. See Tamoxifen Selective steroid receptor modulators (SSRMs), 813 Self-image, 245 Sensory afferent innervation, 253,254f SERMs. See Selective estrogen receptor modulators Serotonin, 829 Serotonin-norepinephrine reuptake inhibitor (SNRI) preparations, 255-256 Serous epithelium, 51 Sex steroids, 274-275. See also specific hormones and the brain. See Brain carbohydrate metabolism and, 502-504 dehydroepiandrosterone. See Dehydroepiandrosterone estrogen. See Estrogen and mood, 205-207, 219-221,219f, 221f phytoestrogens. See Phytoestrogens progesterone. See Progesterone and sexuality, 221-223, 222f testosterone. See Testosterone Sexuality, 271-272 aging, effects of, 272 aging versus menopause status, 272-273,273f, 274f DHEA and, 825-826, 826t ethnicity and, 273 hormonal changes relative to other psychological factors, 273 libido, 258-259 menopause, symptoms of, 6 and quality of life, 652 secondary sexual development, 31- 32 and sex steroids, 221-223,222f sexual dysfunction and testosterone replacement, 801-802 surgical menopause, 273-274 tibolone, effects on, 849-850 vulvovaginal complaints, 265 Shift work and melatonin, 833 Single channel cystometry, 704f Skin, 237, 246 atrophy, 241-243,242f changes, 228-229, 228f, 229f, 230-232, 238t conductance and hot flashes, 188, 188f, 189f, 190 dry skin, 237-240
functional properties, 238t hormone replacement therapy, 239t, 245-246 self-image, 245 sweating, 241 temperature and hot flashes, 187 vitamin D, 244-245 wound healing, 243-245 Sleep disorders and the climacteric, 204-205 disturbances and melatonin, 833 gonadotropin chronobiology and, 132-133 hot flashes and, 194-195 menopause, symptoms of, 5 synchronicity of other hormones with nighttime gonadotropin secretion, 133-134 Smoking age at natural menopause and onset of perimenopause, 82 and cardiovascular disease, 414-415 CHD risk reduction, 432-434 menopause, age of, 3 menopause transition and, 158 observational studies, 84t-86t osteoporosis and, 328 premature ovarian failure and, 103 presentation and symptoms of menopause, 83 preventive health care, 8 SNRI. See Serotonin-norepinephrine reuptake inhibitor preparations Socioeconomic status, 80 Sodium pentosan polysulfate therapy for interstitial cystitis, 723, 724 Somatotropic axis changes in menopause transition, 161 Spermicides, 180 SSRMs. See Selective steroid receptor modulators Statin/niacin/fibrates, 436 Statin plus niacin, 436 Statins, 436 Steroidogenic potential and FSH, 5 4 - 5 5 Steroids interstitial cystitis treatments, 724 osteoporosis treatments, 387, 387f vaginal contraceptives, 177 Stigma of ovary, 51 Stochastic control over oocyte number, 30 STRAW, 2, 2f Stress cardiovascular disease and, 416 - 417 depression and, 313 Stress tests, 432, 703 Stress urinary incontinence (SUI), 696 abdominal approaches to surgery, 712-714 absorbent products and, 708 alpha-adrenergics, 710 anterior colporrhaphy with suburethral plication, 710, 711f artificial urinary sphincter, 715 behavioral intervention, 708- 709 biofeedback techniques, 709 Burch retropubic urethropexy, 712- 714 conservative management, 708 duloxetine, 710 electrical stimulation therapy, 709 estrogen treatment, 709-710 laparoscopic approaches to surgery, 714 Marshall-Marchetti-Krantz vesicourethral suspension, 712, 713f medical comorbidities, treatment of, 708 needle suspension procedures, 710-711 pathophysiology, 699- 700 pelvic floor exercises, 708-709
893
SUBJECTINDEX periurethral bulking injections, 714-715,714f pharmacologic treatment, 255, 709-710 prosthetic devices and, 708 retropubic suspension, 712, 713f suburethral sling procedures, 711,712f surgical treatment, 710-714, 711f, 712f, 713f tension-free-mid-urethral sling procedures, 711-712, 713f transient causes, treatment of, 708 treatment, 708- 715 vaginal approaches to surgery, 710-712 vaginal cones, 709 Stroke and hormone therapy cardiovascular disease and, 551-552, 552t morbidity and mortality changes, 628t-629t, 631 Women's Health Initiative hormone therapy trial, 613 Strontium ranelate treatments for osteoporosis, 368-369, 388-389, 389t Suburethral hammock support, 699 SUI. See Stress urinary incontinence Superinfection, 726 Surgery interstitial cystitis, 725 menopause transition, 155 mixed urinary incontinence, 717 pelvic organ prolapse and, 749-750, 756-762 stress urinary incontinence, 710-714, 711f, 712f, 713f urge urinary incontinence, 717 Surgical menopause, 273-274 SWAN, 2, 3, 311-312 Sweating, 188, 188f, 189f, 241 Swyer-Greenblatt test, 794 Sympathetic pathways, 253 Symptoms of menopause, 3-6, 78 demographic characteristics, 80- 81 environmental influences, 83-90 ethnic and international differences, 81-82 Syndrome X, 410-411,410f, 501-504, 502f T Tamoxifen breast, effects on, 838 cataracts, effects of, 839 gynecologic effects, 838-839 menopausal symptoms and, 838- 839 skeleton and bone metabolism, effects on, 837-838 vascular disease and intermediates, effects on, 838 Tartrate resistant acid phosphatase (TRAcP), 341 Teriparatide, 388 Testosterone Alzheimer's disease and, 300-301 autoimmune disease and, 803-804 deficiency in women, causes of, 800-801 progestins structurally related to, 783-784 replacement benefits of, 801-804 most likely to benefit, 807 prescribing, 808-809, 808t risks of, 804-807 sexual dysfunction, 801-802 Thermoregulation and hot flashes, 192-194, 193f, 193t Thrombosis and hormone therapy clinical considerations, 487, 487f, 488f clinical markers of, 485-486 Tibolone, 847-848, 848f, 850 bone, effects on, 850 breast, effects on, 849
cardiovascular effects of, 850 clinical profile of, 848-850 endometrium, effects on, 848-849, 849f hormone therapy and, 477 menopausal symptoms and, 848 osteoporosis treatments, 364, 383 pharmacokinetics of, 848, 848f sexual activity, effects on, 849-850 Timing hypothesis, 521,524-525 Trace elements in osteoporosis, 673-674 TRAcP. See Tartrate resistant acid phosphatase Traditional medicine. See Complementary and alternative medicine Traditional view of menopause, 1-2 Transcutaneous electrical nerve stimulation and interstitial cystitis, 725 Transdermal contraception, 177 Transient urinary incontinence, 697 Translocations, autosomal reciprocal, 37 Tricyclic antidepressants and urge urinary incontinence, 716 Trigonitis, 726 Trisomy, autosomal, 37 Tunica albuginea, 51 U Ultrasonography and urinary incontinence, 707-708 Unconventional medicine. See Complementary and alternative medicine University of Wisconsin Interstitial Cystitis Scale, 721t Unopposed estrogen therapy. See Estrogen UQOL. See Utian Quality of Life Scale Urethra anatomy, 694-695,695f Urethral pressure profile, 705- 706, 707f Urethral syndrome, 726 Urethritis, 726 Urethrocele, 743 Urge urinary incontinence (UUI), 256-257, 696-697 anticholinergic drugs, 715-716 behavioral therapy, 715 bladder training, 715 botulinum toxin, intravesical injection of, 717 electrical stimulation therapy, 716-717 fluid management, 715 pathophysiology, 700 pharmacologic treatment, 715-716 sacral neuromodulation, 716-717 surgical treatment, 717 treatment, 715 - 717 tricyclic antidepressants, 716 Urinary incontinence by age, 252f classification, 696-698 continuous urinary incontinence, 697 cystourethroscopy, 707 defined, 696-698 electrophysiologic studies, 707 epidemiology, 698 evaluation of, 700- 708 functional incontinence, 697 magnetic resonance imaging, 708 mechanism of micturition and continence, 699 medical history, 700-701 mixed urinary incontinence, 697 pathophysiology, 700 treatment for, 717 multichannel urodynamics, 705-707, 706f neurologic examinations, 702 nocturnal enuresis, 697
overflow urinary incontinence, 697 pad tests, 704 pathophysiology, 698-700 pelvic examinations, 702 pelvic organ prolapse and, 753 physical examinations, 702 postvoid residual volume, 703 potential urinary incontinence, 698 Qctip tests, 703 quality of life questionnaires, 702 single channel cystometry, 704-705, 704f stress tests, 703 stress urinary incontinence. See Stress urinary incontinence transient incontinence, 697 transient urinary incontinence, 697 treatment of, 708- 717 ultrasonography, 707- 708 urethral pressure profile, 705-706, 707f urge urinary incontinence. See Urge urinary incontinence urinalysis, 703 -704 urodynamics, 704-707 uroflowmetry, 707 videoradiographic urodynamics, 707 voiding cystourethrogram, 707 voiding diary, 701- 702 Urinary markers and interstitial cystitis, 722 Urinary tract disorders, 691,693 anatomy of lower urinary tract, 693-696 bladder anatomy, 694, 695f incontinence. See Urinary incontinence innervation of lower urinary tract, 695-696 interstitial cystitis. See Interstitial cystitis pelvic floor anatomy, 693-694, 694f sensory innervation, 696 somatic motor control, 695-696 sympathetic and parasympathetic innervation, 695, 696f urethra anatomy, 694-695,695f urinary tract infections. See Urinary tract infections Urinary tract infections (UTIs), 703-704, 725- 726 antibiotic treatment, 729, 730t, 731t asymptomatic bacteriuria, 729 catheter infection, 732-733 clinical presentation, 727 complicated infection, 732 cystourethroscopy, 728 diagnostic tests, 727-728 epidemiology, 726 general management, 729 host defense mechanism, 727 instrument-associated infection, 732-733 microscopic examination, 728 pathophysiology, 726-727 radiologic studies, 728 recurrent infection, 732 simple infection, 729-732 terminology, 726 urine analysis, 727-728 urine culture and sensitivity, 728 Urine storage and micturition, 253-254, 255f Urodynamics, 704- 707 interstitial cystitis, 722 stress incontinence, 696 Uroflowmetry, 707 Urogenital system, 251-252 aging symptoms, 252t continence, mechanism of, 253-254, 255f disorders. See Urinary tract disorders
894 Urogenital system (continued) estrogen treatments, 257-258 genital symptoms of incontinence, 258 innervation, 253,254f libido and, 258-259 parasympathetic pathways, 253 paraurethral collagen and stress urinary incontinence in women of reproductive age, 252-253 paraurethral collagen and the menopause, 252 pharmacologic treatment of stress incontinence, 255 sensory afferent innervation, 253, 254f serotonin-norepinephrine reuptake inhibitor preparations, 255-256 sympathetic pathways, 253 treatment modalities for female urinary incontinence, 254-259, 255t urge incontinence, 256-257 urine storage and micturition, 253-254, 255f Urogenital triangle, 741 Uterine bleeding and oral contraceptives, 173 Uterine prolapse, 743 UTIs. See Urinary tract infections Utian Quality of Life Scale (UQ.OL), 644t, 647t, 649 UUI. See Urge urinary incontinence V Vagina support of and pelvic organ prolapse, 743 vasomotor symptoms/vaginal atrophy study requirements, 867, 867f, 868f vulvovaginal complaints. See Vulvovaginal complaints Vaginal cones and stress urinary incontinence, 709 Vaginal vault prolapse, 743 Vasomotor episodes at start of climacteric, 204 Vasomotor symptoms/vaginal atrophy study requirements, 867, 867f, 868f VBB. See Virtual bone biopsy Venous thromboembolism (VTE) clinical entities, 491-492 clinical recommendations, 498 clinically identified risk factors, 495, 495t diagnosis, 492 hormone replacement therapy morbidity and mortality changes, 628t-629t, 631 risks, 487, 488, 496-498, 496t, 497t, 552-554, 553t, 554t
SuBJECXINDEX Women's Health Initiative trial, 613 hypercoagulability, risks of VTE with, 494-495 idiopathic VTE risk factors, 495-496 incidence, 493-498, 494t oral contraceptives and, 173-174 pathophysiology, 492-493 primary VTE risk factors, 495-496 prognosis, 493- 498 risk factors, 494-496 treatment, 492 Women's Health Initiative hormone therapy trial, 613 Videoradiographic urodynamics, 707 Viral agents and premature ovarian failure, 103 Virtual bone biopsy (VBB), 335 Vitamin D osteoporosis and, 328, 360-361,667-670, 669t and skin, 244-245 Women's Health Initiative Calcium Vitamin D Trial, 615-616 Vitamin E, 687-688 Vitex, 687 Voiding cystourethrogram, 707 Voiding diary, 701 - 702, 719- 720 VTE. See Venous thromboembolism Vulnerable Blood, 430, 431 Vulnerable Myocardium, 430, 432 Vulnerable Patient, 430 Vulnerable Women, 430 Vulvovaginal complaints, 263-264, 264t, 268 botanicals, use of, 268 diagnosis, 264, 264t estrogen loss, 264 evaluation of, 265-266, 266f hormonal therapy, 267- 268 nonhormonal therapy, 268 physical examination, 265,266f sexual complaints, 265 treatment, 267-268, 267f, 268f W Weight gain and oral contraceptives, 174 W H I . See Women's Health Initiative WHIMS. See Women's Health Initiative Memory
Study VVHQ~. See Women's Health Questionnaire
Wiley Act, 864 Women's Angiographic Vitamin and Estrogen Trial, 538-539 Women's Estrogen-Progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial, 539
Women's health care centers. See Health care centers Women's Health Initiative (WHI), 611,612f Calcium Vitamin D Trial, 615-616 dietary modification trial, 615 future direction of therapy, 815- 817 future reports, 617 hormone therapy trial, 611 - 612 cancer, 613-614 cardiovascular disease, 533-537, 533f, 534f, 535f, 536t, 537f, 613 cognitive decline, 614 menopausal symptoms, 614-615 morbidity and mortality changes, 627 osteoporosis, 612-613,612f, 613t stroke, 613 venous thrombosis, 613 Observational Trial, 616-617, 616t, 617t Women's Health Initiative Memory Study (WHIMS), 288-289, 298-300, 299t Women's Health Q.uestionnaire (WHO.), 643, 644t, 645t Wound healing, 243-245 X X chromosome abnormalities and incidence of premature ovarian failure due to genetics, 43 cytogenetics of X chromosomal aberrations. See Cytogenetics of X chromosomal aberrations mosaicism, 32, 32t, 33f ovarian genes on, 36-37 ovarian maintenance genes, localizing to specific regions of the X, 32- 33 premature ovarian failure and cytogenetic abnormalities involving, 101-102 Y Yam, 687 Z Zona reticularis involution and aging, 822-823, 824f
Spironolactone and Drospirenone
FIGURE 12.3
Drospirenone a progestational agent derived from spironolactone.
FIGURE 16.2 Bone density at L 2 - L 4 plotted against skin thickness in 1500 untreated postmenopausal women as control subjects (dots) and 250 untreated postmenopausal women who had suffered an osteoporotic bone fracture (squares). The accuracy of combining skin thickness measurements with bone density parameters in predicting bone fractures is shown.
FIGURE '17.1 (a) Skin tear due to fragility of forearm skin in an 87-year-old female without hormone replacement therapy. (b) Lesion taped closed to increase rate of healing by approximating the epidermal edges of the full-thickness wound. (c) Wound healed, with linear scar.
FIGURE 19.1 This photo demonstrates classic changes in an atrophic vulva, with narrowing of the introitus and inflamed mucosal surfaces. There is also loss of labial and vulvar fullness.
FIGURE 21.1 HT users show significant increases in hippocampal blood flow over a 2year interval. (Reprinted from Neurobiologyof 3ging, Volume 21, Maki PM, Resnick SM. Longitudinal effects of estrogen replacement therapy on PET cerebral blood flow and cognition, 373-383, 9 2000 with permission from Excerpta Medica, Inc.)
FIGURE 21.2 Estrogen modulates brain activation in frontal lobes in postmenopausal women. (From JAM_A,April 7, 1999, Page 1199, Copyright 9 1999 American Medical Association. All rights reserved.)
FIGURE 26.2 High-resolution pQCT images of the distal radius. A: Radiograph showing the site of imaging at the distal radius. The white line indicates the beginning of the joint space, and the red lines indicate the section of bone over which images are acquired. B: Representative crosssectional images from the stack of CT slices from proximal (top left) to distal (bottom right). C: Representative 3D image. (From ref. 26, with permission.)
FIGURE 32.1 Molecular mechanisms that mediate cardiovascular effects of estrogen and estrogen receptors (ERs). ERs regulate cardiovascular cell function by regulating gene expression. They do so by interacting with a variety of proteins important for these transcriptional regulatory pathways (point 1). The transcriptional activity of the receptor is also regulated by direct phosphorylation (point 3). ERs also regulate the expression of other transcription factors, including other sex steroid hormone receptors and non-sex hormone receptor transcription factors (points 4 and 6). Genetic polymorphisms in the receptors also contribute to interindividual responses to hormone treatment (point 5). Finally, in addition to these regulatory pathways that are all related to genomic effects, cell membrane-associated ERs can also regulate cellular functions by nongenomic pathways that involve the rapid activation of various ldnase cascades (point 2). CoA, coactivator; CoR, corepressor; GFR, growth factor receptor; NHR, non-SSHR nuclear receptors; SSHR, sex steroid hormone receptor; SSHRE, SSH response element; TFs, transcription factors. From ref. 11, with permission.
FIGURE 32.2 The importance of timing of initiation of hormone therapy on its cardiovascular effects. The "timing hypothesis" states that the effects of hormone therapy on cardiovascular disease differs between early and later stages of atherosclerosis. Atherosclerosis is characterized by the gradual loss of vascular protective mechanisms and the emergence of advanced, unstable lesions. Estrogen's effects on vascular cells differ depending on the state or progression of atherosclerosis in the exposed vessels. CAMs, cell adhesion molecules; Cox2, cyclooxygenase 2; LDL, low-density lipoprotein; MCP-1, monocyte chemoattractant protein 1; MMP, matrix metalloproteinase; TNF-c~, tumor necrosis factor-cx; VSMC, vascular smooth muscle cell. From ref. 11, with permission.
FIGURE 42.1 hyperplasia.
A typical photomicrograph of atypical endometrial
FIGURE 58.1
Regulation of melatonin secretion. (From res 97, with permission.)
