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