Diabetology of Pregnancy (Frontiers in Diabetes)

Diabetology of Pregnancy

Frontiers in Diabetes Vol. 17

Series Editors

M. Porta Turin F.M. Matschinsky Philadelphia, Pa.

Diabetology of Pregnancy

Volume Editors

J. Djelmiš Zagreb G. Desoye Graz M. Ivaniševic´ Zagreb

50 figures, 16 in color, and 22 tables, 2005

Basel · Freiburg · Paris · London · New York · Bangalore · Bangkok · Singapore · Tokyo · Sydney

Frontiers in Diabetes Founded 1981 by M. Belfiore, Catania

Josip Djelmiš, MD

Gernot Desoye, MD

Department of Obstetrics and Gynecology State Referral Centre for Diabetes in Pregnancy University of Zagreb, Zagreb, Croatia

Clinic of Obstetrics and Gynaecology Medical University of Graz, Graz, Austria

Marina Ivaniševic, ´ MD Department of Obstetrics and Gynecology Referral Centre for Diabetes in Pregnancy University of Zagreb, Zagreb, Croatia

Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents® and Index Medicus. Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher. © Copyright 2005 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel ISSN 0251–5342 ISBN 3–8055–7925–X

Contents

IX Preface 1 Epidemiology of Pregestational and Gestational Diabetes Pavlic-Renar, I.; Metelko, Z. (Zagreb) ´ ˆ

9 Fuel-Mediated ‘Functional Teratogenesis’ and Primary Prevention Plagemann, A. (Berlin) 18 Endocrine Changes in Diabetic Pregnancy Kautzky-Willer, A.; Bancher-Todesca, D. (Vienna) 34 Metabolic Changes in Diabetic Pregnancy Herrera, E. (Madrid) 46 Leptin in the Diabetic Pregnancy Hauguel-de Mouzon, S. (Cleveland, Ohio); Lepercq, J. (Paris); Catalano, P. (Cleveland, Ohio) 58 Maternal Diabetes and Embryonic Development De Hertogh, R. (Brussels) 72 Normal and Abnormal Fetal Growth Cetin, I.; Radaelli, T. (Milan) 83 Effect of Nutrition on Fetal Development: A View on the Pancreatic ␤-Cells Blondeau, B.; Breant, B. (Paris) 94 The Human Placenta in Diabetes Desoye, G. (Graz); Kaufmann, P. (Aachen)

V

110 Vasculogenesis and Angiogenesis in the Diabetic Placenta Leach, L.; Mayhew, T.M. (Nottingham) 127 Morphological Findings in Infants and Placentas of Diabetic Mothers Kos, M. (Zagreb); Vogel, M. (Berlin) 144 Principles of Evidence-Based Medicine. The Lesson for Diabetic Pregnancy Pieber, T.R. (Graz) 153 The Graz Approach to Diabetes in Pregnancy Weiss, P.A.M.; Desoye, G. (Graz) 161 Clinical Management of Pregnancies Complicated with Type 1/Type 2 Diabetes mellitus Djelmiš, J. (Zagreb) 174 Nutritional Management of Diabetes in Pregnancy Dornhorst, A.; Frost, G. (London) 195 Diabetes-Related Antibodies and Pregnancy Corcoy, R.; Mauricio, D.; de Leiva, A. (Barcelona) 206 Insulin and Oral Hypoglycemic Agents in Pregnancy Metelko, Z.; Pavlic´ -Renar, I. (Zagreb) ˆ

214 Continuous Glucose Monitoring for the Evaluation and Improved Treatment Adjustment in Gravid Women with Diabetic Pregnancy Hod, M.; Yogev, Y. (Petah Tiqva) 222 Fetal Hypoxia and Its Monitoring in Pregestational Diabetic Pregnancies Teramo, K.A.; Hiilesmaa, V.K. (Helsinki) 230 Ultrasonic Surveillance of the Diabetic Fetus Ivaniševic´ , M. (Zagreb) 254 Diabetic Pregnancy: Maternal Metabolic and Microvascular Complications in Type 1 Diabetes mellitus Pearson, D.W.M. (Aberdeen) 271 Pre-Eclampsia in Women with Type 1 Diabetes Mathiesen, E.R.; Damm, P. (Copenhagen) 278 Management of Delivery Miller, V.A.; Gillmer, M.D.G. (Oxford)

Contents

VI

288 Offspring of Diabetic Pregnancy Persson, B. (Stockholm); Eriksson, U.J.; Hanson, U. (Uppsala) 310 Long-Term Consequences of Gestational Diabetes mellitus Lauenborg, J.; Mathiesen, E.R.; Damm, P. (Copenhagen)

320 Author Index 321 Subject Index

Contents

VII

Preface

Diabetes mellitus has always been an active area of research. The complexity of the disease, its underlying pathogenetic mechanisms, progress in treatment strategies and recent insights into its potential and far-ranging complications have justified the establishment of diabetology as a medical and scientific specialty. In recent decades diabetes in pregnancy has evolved as a special subdiscipline. This book is published with the purpose of providing a comprehensive, but nevertheless concise, overview of the scientific and clinical characteristics/features of gestational diabetes as well as of type 1 and type 2 diabetes mellitus in pregnancy. It aims to address a wide range of specialists, health care professionals and academics, who are involved in tackling the medical and scientific problems of the disease and its influence on pregnancy and its outcome. The ramifications of diabetes in pregnancy cover a broad range of areas and thus call for a multidisciplinary approach. Throughout this book the authors have tried to review published literature in this field on aspects such as etiology, physiology, molecular biology and genetics, immunology, pathogenesis, diagnosis, treatment and management, pathology and evidence-based medicine. In addition all the authors were encouraged to discuss their own observations and opinions. Because of the population-dependent incidence and prevalence of diabetes in pregnancy the book begins with a chapter on its epidemiology. Two chapters deal with endocrine and metabolic changes in diabetic pregnancy, followed by a discussion on the role of leptin as the currently best-explored adipocyte hormone. Fuel-mediated functional teratogenesis and primary prevention introduce all IX

most recent aspects of the problem. Maternal influence on the embryonic development was the logical introduction to the chapter on normal and abnormal fetal growth, followed by discussions of nutritional effects on fetal development. A special characteristic of this book are three chapters on the placenta as a fetal organ at the interface between mother and fetus. Its development, vasculogenesis and angiogenesis and morphological characteristics in diabetes as well as its potential to provide a defense for the fetus are covered. Several illustrations will aid the readers in understanding the complex information on the sequences of developmental and pathologic alterations. The aim of several chapters in this book was to survey current concepts of the clinical management of the diabetic pregnancy including preconception diagnosis and care, and provide protocols for pregnancy surveillance. The recently introduced management of diabetic pregnancies with oral hypoglycemic agents is reviewed in a separate chapter. These articles were meant to form a basis for discussion among the medical professionals about which approach to choose. The editors hope the reader will gain some profound insight into the complexity of diabetes and pregnancy as well as potential pregnancy complications during diabetic pregnancy and delivery. In order to cover the rapid progress that has been made in recent years, especially in the basic sciences, this volume would require additional chapters so as to include results in the field of genetics and molecular biology as they relate to diabetes in pregnancy. However, space limitations and the attempt to find a balance between a comprehensive and a concise approach have resulted in the selection of the chapters included. It has been a great honor for us to collaborate with so many worldwide authorities in the field, several of whom had had the great opportunity to work and discuss with distinguished pioneers in the field such as Norbert Freinkel, Lars Molsted Pedersen, Lois Jovanovic and many others. Their particular expertise and their enthusiasm have helped to cover all the major topics of importance in diabetology of pregnancy. Without their dedicated contribution this project would not have been possible. Finally, the editors would like to thank the efficient and encouraging staff at Karger Publishers, who actively participated in our project. Josip Djelmiš, Gernot Desoye, Marina Ivaniševic´ Zagreb and Graz, April 2005

Preface

X

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 1–8

Epidemiology of Pregestational and Gestational Diabetes ˆ

Ivana Pavlic´-Renar, Z eljko Metelko Vuk Vrhovac University Clinic for Diabetes and Metabolic Diseases, Zagreb, Croatia

Introduction: Diabetes mellitus Incidence and Prevalence Trends

There are wide differences in diabetes prevalence in different ethnic groups giving regional estimates of prevalence varying from lowest in Africa (2.4%) to highest in Europe and North America (7.8 and 7.9% for the 20- to 79-year age group, respectively) [1]. One of the definitions states that there is an epidemic of a disease if the ratio (observed – expected)2/expected prevalence is less than 3.84. Global prevalence in 1995 was estimated to 135 million persons; the projection for 2025 revealed 300 million [2]. More recent estimates are in accordance with these projections [list in 1]. Based upon trends in the estimated global prevalence, the world is facing a pandemic of diabetes mellitus. Type 2 is the most frequent form of diabetes in the world (85% of the Caucasian population, at least 95% of other ethnic groups). Dramatic increases in the incidence and prevalence are observed in many newly industrialized and developing countries. The disease is often asymptomatic and hence undiagnosed. The incidence increases with age so any comparison needs to be ageadjusted. In most populations studied so far there is a tendency of decline of the ratio of impaired glucose tolerance:diabetes as the prevalence of diabetes increases. Areas with a ratio above 1 may be in the preepidemic phase of diabetes. In addition to a rise in prevalence, type 2 diabetes now presents in younger people including children: in parts of the USA 39% new diabetes cases in adolescents are type 2 [3], and in Japan newly diagnosed cases of type 2 diabetes outnumber type 1 in childhood and adolescence [4].

Prevalence and incidence rates of type 1 diabetes are known for only a small part of the global population [list in 1, 5–7]. The incidence varies in different populations from 1 or less to 50/100,000 persons/year. It has been increasing by about 3% per year in almost all countries studied. The worldwide incidence of type 1 diabetes is estimated to be slightly more than 218,000 persons/year, 40% of them children. There are only few population-based prevalence and mortality studies in type 1 diabetes. Substantial differences in life expectancy, demographic composition and incidence result in differences in prevalence between the world regions. It ranges from an estimated 1,265,000 persons with type 1 in Europe to slightly more than 100,000 in Africa. It is estimated that there were 5,350,000 persons with type 1 diabetes, 7.4% of them children, in mid-2000 [1].

Problems in Epidemiological Estimates of Gestational Diabetes

According to the classification of diabetes accepted by the World Health Organization (WHO) [8] which is a modified version of a proposal by the American Diabetes Association (ADA) [9], gestational diabetes mellitus (GDM) is glucose intolerance in pregnancy, with normal or impaired glucose tolerance after termination of pregnancy. This is somewhat different than the earlier definition which recognized GDM only if glucose tolerance was normal after termination of pregnancy. The physiological response to an increased insulin requirement during pregnancy (especially during the third trimester) is an up to 4 times increase of insulin secretion. With a potential beta cell failure already present different stages of glucose intolerance will become apparent [10, 11]. It has been shown that type 1 diabetes is diagnosed in pregnancy more often than expected [12]. The same is true for type 2. It has been suggested that ‘classical’ GDM (i.e. with normal glucose tolerance postpartum) is basically a form of type 2 diabetes [11]. Comparing epidemiological data on GDM is not an easy task since several diagnostic criteria are currently used. Moreover, over time these criteria have been changed [13]. Current ADA criteria [14] are based on two different protocols for oral glucose tolerance test (OGTT). A 3-hour test using a 100-gram glucose load is preferred. At least two blood glucose (BG) values equal to or greater than fasting 5.3 mmol/l, 1 h postload 10.0 mmol/l, 2 h postload 8.6 mmol/l, and 3 h postload 7.8 mmol/l (all for venous plasma) reveal GDM. For a 2-hour test using 75-gram glucose load thresholds in fasting, 1 and 2 h are the same, and again at least two measurements should be equal to or above the threshold. Criteria for GDM accepted by the WHO are the same as in nonpregnant individuals: with a 75-gram glucose load, BG equal to or greater than

Pavlic´-Renar/Metelko

2

7.8 mmol/l (8.9 mmol/l for capillary plasma) 120 min postload reveal GDM. No distinction between impaired glucose tolerance and diabetes is made in pregnancy [8, 9]. Comparison of ADA (using a 75-gram load) and WHO criteria in a large sample (n ⫽ 4,977) of the Brazilian population with a high perinatal mortality revealed that both criteria detect the risk of macrosomia, preeclampsia and perinatal death with a higher number of cases predicted by the WHO criteria, which thus might have a better potential for prevention [15]. In order to clarify unanswered questions on the association of maternal glycemia, which is less severe than overt diabetes mellitus, with risks of an adverse pregnancy outcome a 5-year investigator-initiated prospective observational study that will recruit approximately 25,000 pregnant women in 10 countries was designed: the Hyperglycemia Adverse Pregnancy Outcome Study (HAPOS) [16]. Glucose tolerance is assessed by a 75-gram 2-hour OGTT at 24–32 weeks’ gestation. Results are unblinded to the woman and her caregivers if fasting plasma glucose ⬎5.8 mmol/l, 2-hour plasma glucose ⬎11.1 mmol/l or any plasma glucose ⬍2.5 mmol/l. Random plasma glucose measurement is performed at 34–37 weeks or if symptoms suggest hyperglycemia; results are unblinded for values greater than or equal to 8.9 mmol/l. The results are expected.

GDM Incidence and Risk Factors

The GDM incidence is 2–14% [17–22]. It varies significantly by ethnicity [23]. Characteristically, its incidence reflects the incidence of type 2 diabetes in the background population. Besides being an additional indicator that GDM is mostly type 2 diabetes, this fact is a warning that a rapid increase in the incidence is expected concomitant with the already observed increase in the incidence of type 2, especially in developing countries [24]. There are already published data revealing an increase in the incidence of GDM [25]. The highest risk for GDM is maternal age. In a prospective study the relative risk for GDM rose by 4% for each year of age after 25 [26]. In women younger than 21 it is less than 1%, in those between 21 and 30 years of age less than 2%, whereas in those older than 30 it is 8–14%. The family history is a further significant risk factor. Unlike type 1 diabetes which is predominantly inherited from the father [27, 28], maternal inheritance is much stronger here [29]. This might be due to the importance of the intrauterine environment in early development [30]. The level of obesity is a further risk factor. Women with a BMI of 30 or more have GDM 3 times more frequently than women with a BMI of 20 or less [19]. A study which revealed the importance of BMI increase in early adulthood also revealed smoking as a significant risk factor for GDM with a relative risk of 1.43 [26].

Epidemiology of Diabetes mellitus

3

A poor obstetric history and a history of a macrosomic newborn are factors in a smaller percentage of women with GDM and are mostly explained by unrecognized GDM during a previous pregnancy [13].

Fetal Risks in Diabetic Pregnancy

Fetal malformation is related to metabolic disturbances in diabetes. In early pregnancy, during organogenesis, hyperglycemia seems to be the dominant risk factor for fetal malformation [31]. Thus it is primarily a problem in preexisting diabetes: preconceptional diabetes regulation is essential. An increased incidence of type 2 diabetes in younger age groups as well as increased maternal age in primiparous women make type 2 diabetes an emerging problem in this respect [32]. Outcomes of pregnancies complicated with unrecognized or poorly controlled type 2 diabetes are comparable to those with poorly controlled type 1 diabetes [33, 34]. The intrauterine environment seems to be a risk for insulin resistance in the child. Children of mothers with gestational and preexisting diabetes are less sensitive to insulin as adolescents than their age-matched controls [35, 36]. Studies in Pima Indians [37] and studies of discordant siblings who were born before and after the onset of maternal diabetes [38] distinguished between the effect of inheritance and of the intrauterine metabolic state. Animal studies also revealed a relatively diminished insulin response to glucose challenge in offsprings of females who were hyperglycemic during gestation [39, 40]. Children of mothers with GDM are more often obese in later life. It is difficult here also to distinguish inheritance from intrauterine effect. The aforementioned studies in Pima Indians [39] suggest an independent effect of inheritance and intrauterine exposure to hyperglycemia. A long-term follow-up revealed that children from pregnancies with GDM have worse general health at school age estimated by a need for hospitalization in comparison with their age-matched controls [41] and are more susceptible to inattention and fine and gross motor impairment [42].

Maternal Risks of GDM

Hyperglycemia in GDM is usually too mild and of too short duration to affect adversely women’s health, although there are reports of the occurrence of ketoacidosis and retinopathy [43, 44]. However, in the long term the risk of overt diabetes development is high in these women, ranging from 2.6 to 70% in studies following women from 6 weeks to 28 years postpartum. A systematic

Pavlic´-Renar/Metelko

4

review [45] with adjustment for the length of follow-up and cohort retention revealed an increase in the incidence in the first 5 years after index pregnancy and a plateau after 10 years. At least one third of women with GDM has a recurrence in a subsequent pregnancy [46]. These women are candidates for preventive actions, primarily for lifestyle modifications [47]. Around 5% women with GDM, characteristically with normal or low BMI, develop type 1 diabetes. They test positive for ICA and GAD [48, 49]. These pregnancies occur in the early prediabetic stage, so for some time after delivery, when the need for insulin is lower, normoglycemia may be maintained with no insulin [50]. Type 1 diabetes is suspected in lean women with early GDM [51].

Screening for GDM

Besides precise diagnostic criteria, another controversy in GDM is when and how to screen for it [52–54]. There are rational arguments both for universal screening and screening of risk groups only. By testing only women with a high risk, up to 50% GDM remains undiagnosed [52]. Caucasian women younger than 30 with a BMI less than 22 have a minimal risk of GDM (0.9%) and thus need not be tested [55]. However, with strict adherence to this strategy, 10% GDM would be undiagnosed, 16% of which would have a significantly high BG level, requiring insulin treatment [56]. Screening for GDM in the USA is performed using a 50-gram oral glucose load (glucose challenge test) with BG measurement 60 min later. A BG equal to or more than 7.8 mmol/l reveals a high risk for GDM and an OGTT is performed. With this two-step approach, 80% of cases are identified; if the threshold of 7.2 mmol/l in 60 min is employed the yield is 90% [14]. For twostep screening, an alternative to glucose challenge is a combination of a random BG equal to or greater than 8 mmol/l and a history of a large-for-gestation baby or previous GDM [57].

References 1 2 3 4

Global burden of diabetes; in Diabetes Atlas, ed 2. Bruxelles, International Diabetes Federation, 2003, pp 15–111; http://www.idf.org/e-atlas/ King H, Aubert RE, Herman WH: Global burden of diabetes 1995–2025. Diabetes Care 1998;21:1414–1431. Callahan ST, Mansfield JM: Type 2 diabetes mellitus in adolescents. Curr Opin Pediatr 2000;12: 310–315. Kitagawa T, Owada M, Urakami T, Yamauchi K: Increased incidence of non-insulin dependent diabetes mellitus among Japanese schoolchildren correlates with an increased intake of animal protein and fat. Clin Pediatr (Phila) 1998;37:111–115.

Epidemiology of Diabetes mellitus

5

5 6 7

8

9 10 11

12 13 14 15

16 17 18 19 20 21 22

23 24 25 26 27 28 29

Diabetes Epidemiology Research International Group: Geographic patterns of childhood insulindependent diabetes mellitus. Diabetes 1988;37:1113–1119. Green A, Gale EAM, Patterson CC: Incidence of childhood-onset insulin-dependent diabetes mellitus: The EURODIAB ACE study. Lancet 1992;339:905–909. Karvonen M, Tuomilehto J, Libman I, LaPorte R: A review of the recent epidemiological data on the worldwide incidence of type 1 (insulin-dependent) diabetes mellitus. World Health Organization DIAMOND Project Group. Diabetologia 1993;36:883–892. Definition, Diagnosis and Classification of Diabetes mellitus and Its Complications. Report of a WHO Consultation. Part 1. Diagnosis and Classification of Diabetes mellitus. WHO/NCD/NCS/99.2. Geneva, WHO, 1999. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997;20:1183. Kuhl C: Etiology and pathogenesis of gestational diabetes. Diabetes Care 1998;21:B19. Jovanovic L: The Norbert Freinkel Lecture: Glucose-mediated macrosomia: The over-fed fetus and the future. Program and abstracts of the 61st Scientific Session of the ADA, June 22–26, 2001, Philadelphia. Buschard K, Buch I, Molsted-Pedersen L, Hougaard P, Kuhl C: Increased incidence of true type I diabetes acquired during pregnancy. Br Med J 1987;294:275–279. American Diabetes Association: Position statement. Gestational diabetes mellitus. Diabetes Care 2004;27(suppl 1):S88. Ryan EA: What is gestational diabetes?; in Gernstein HC, Haynes RB (eds): Evidence-Based Diabetes Care. Hamilton, BC Decker, 2001, p 164. Schmidt MI, Duncan BB, Reichelt AJ, et al; Brazilian GDS Group: Gestational diabetes mellitus diagnosed with a 2 h 75-mg oral glucose tolerance test and adverse pregnancy outcome. Diabetes Care 2001;24:1151. HAPO Study Cooperative Research Group: The Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study. Int J Gynaecol Obstet 2002;78:69. Forsbach G, Contreras Soto JJ, Fong G, et al: Prevalence of gestational diabetes and macrosomic newborns in Mexican population. Diabetes Care 1988;11:235. Coustan DR, Nelson C, Carpenter MW, et al: Maternal age and screening for gestational diabetes: A population-based study. Obstet Gynecol 1995;18:1029. Engelgau MM, Herman WH, Smith PJ, et al: The epidemiology of diabetes and pregnancy in the US. Diabetes Care 1995;18:1029. Akhter J, Quershi R, Rahim F, et al: Diabetes in pregnancy in Pakistani women: Prevalence and complications in an indigenous South Asian community. Diabetes Med 1996;13:189. Erem C, Cihanyurdu N, Deger O, Karahan C, Can G, Teletar M: Screening for gestational diabetes mellitus in northeastern Turkey (Trabzon City). Eur J Epidemiol 2002;18:39. Jang HC, Yim CH, Han KO, Yoon HK, Han IK, Kim MY, Yang JH, Choi NH: Gestational diabetes mellitus in Korea: Prevalence and prediction of glucose intolerance early postpartum. Diabetes Res Clin Pract 2003;61:117. Tammaa A, Teich G, Scholl T, Salzer H: Ethnische Erbe als Risikofaktor für Diabetes gestationis in Mitteleuropa. Geburtshilfe Frauenheilkd 2001;61:142. WHO Prevalence Statistics: URL hppt://www.who.int/ncd/dia/databases.htm Ferrara A, Kahn HS, Quesenberry CP, Riley C, Hedderson MM: An increase in the incidence of gestational diabetes mellitus: Northern California, 1991–2000. Obstet Gynecol 2004;103:526. Solomon CG, Willett WC, Carey VJ, et al: A prospective study of pregravid determinants of gestational diabetes mellitus. JAMA 1997;278:1078. Vadheim CM, Rotter JI, MacLaren NL, et al: Preferential transmission of diabetic alleles within the HLA gene complex. N Engl J Med 1986;315:1314. Warram JH, Krolewski AS, Kahn CR: Determinants of IDDM and perinatal mortality in children of diabetic mothers. Diabetes 1988;37:1328. Harder T, Franke K, Kohlhoff R, Plagemann A: Maternal and paternal family history of diabetes in women with gestational diabetes or insulin-dependent diabetes mellitus type I. Gynecol Obstet Invest 2001;51:160.

Pavlic´-Renar/Metelko

6

30 31 32 33

34 35 36 37 38 39 40 41 42

43 44 45 46 47 48 49 50

51

52 53 54

Ryan EA, Enns L: Role of gestational hormones in the induction of insulin resistance. J Clin Endocrinol Metab 1988;48:1807. Starc¤evic´ V, Djelmiš J, Ivaniševic´ M, Mayer D: The effect of glycemia on early embryonic development in diabetic pregnancies. Prenat Neonatal Med 2001;6:208. Brydon P, Smith T, Proffitt M, Gee H, Holder R, Dunne F: Pregnancy outcome in women with type 2 diabetes mellitus needs to be addressed. Int J Clin Pract 2000;54:418. Towner D, Xiang A, Kjos SE, et al: Congenital malformations in pregnancies complicated by NIDDM. Increased risk from poor maternal metabolic control but not from exposure to sulphonylurea drugs. Diabetes Care 1995;18:1446. Cundy T, Gamble G, Townend K, Henley PG, MacPherson P, Roberts AB: Perinatal mortality in type 2 diabetes mellitus. Diabet Med 2000;17:33. Silverman BL, Metzger BE, Cho NH, et al: Impaired glucose tolerance in adolescent offsprings of diabetic mothers. Relationship to fetal insulinemia. Diabetes Care 1995;18:611. Plagemann A, Harder T, Kohlhoff R, et al: Glucose tolerance and insulin secretion in children of mothers with pregestational diabetes. Diabetologia 1997;40:1094. Pettit DJ, Aleck KA, Baird HR, et al: Congenital susceptibility to NIDDM. Role of intrauterine environment. Diabetes 1988;37:622. Dabelea D, Hanson RI, Linsday RS, et al: Intrauterine exposure to diabetes conveys risks for type 2 diabetes and obesity – a study of discordant sibships. Diabetes 2000;49:2208. Ryan EA, Liu D, Bell RC, et al: Long term consequences in offsprings of diabetes in pregnancy: Studies with synergic islet transplanted streptozotocin-diabetic rats. Endocrinology 1995;2:10. Linn T, Loewk E, Schneider K: Spontaneous glucose intolerance in the progeny of low dose streptozotocin-induced diabetic mice. Diabetologia 1993;36:1245. Aberg A, Westbom L: Association between maternal pre-existing or gestational diabetes and health problems in children. Acta Paediatr 2001;90:746. Ornoy A, Ratzon N, Greenbaum C, Wolf A, Dulitzky M: School-age children born to diabetic mothers and to mothers with gestational diabetes exhibit a high rate of inattention and fine and gross motor impairment. J Pediatr Endocrinol Metab 2001;14:681. Pitteleoud N, Binz K, Caufeld A, et al: Ketoacidosis during gestational diabetes. Diabetes Care 1998;21:1031. Hagay ZL, Schachter M, Pollack A: Development of proliferative retinopathy in a gestational diabetes patient following rapid metabolic control. Eur J Obstet Gynecol Reprod Biol 1994;57:211. Kim C, Newton KM, Knopp RH: Gestational diabetes and the incidence of type 2 diabetes – a systematic review. Diabetes Care 2002;25:1862. MacNeill S, Dodds L, Hamilton DC, Armson BA, VandenHof M: Rates and risk factors for recurrence of gestational diabetes. Diabetes Care 2001;24:659. Buchanan TA: Pancreatic B-cell defects in gestational diabetes: Implications for the pathogenesis and prevention of type 2 diabetes. J Clin Endocrinol Metab 2001;86:989. Dozio N, Beretta A, Belloni C, et al: Low prevalence of islet autoantibodies in patients with gestational diabetes mellitus. Diabetes Care 1997;20:81. Peterson JS, Dyrberg T, Damm P, et al: GAD65 autoantibodies in women with gestational or insulin dependent diabetes mellitus diagnosed during pregnancy. Diabetologia 1996;39:1329. Fuchstenbusch M, Ferber K, Standl E, et al: Prediction of type 1 diabetes postpartum in patients with gestational diabetes mellitus by combined islet cell autoantibody screening. Diabetes 1997;56:1459. Bartha JL, Martinez-del-Fresno P, Comino-Delgado R: Postpartum metabolism and autoantibody markers in women with gestational diabetes mellitus diagnosed in early pregnancy. Am J Obstet Gynecol 2001;184:965. Jovanovic L, Peterson CM: Screening for gestational diabetes: Optimum timing and criteria for retesting. Diabetes 1985;34:21. Khan NM, Hershey CO: Update on screening for type 2 diabetes – the why, who, how, and what of testing and diagnosing. Postgrad Med 2001;109:27. Homko CJ, Reece EA: To screen or not to screen for gestational diabetes mellitus. The clinical quagmire. Clin Perinatol 2001;28:407.

Epidemiology of Diabetes mellitus

7

55 56 57

Naylor CD, Sermer M, Chen E, et al: Selective screening for gestational diabetes mellitus. N Engl J Med 1997;337:1591. Moses RG, Moses J, Davis WS: Gestational diabetes: Do lean young Caucasian women need to be tested? Diabetes Care 1998;21:1803. Ostlund I, Hanson U: Repeated random blood glucose measurements as universal screening test for gestational diabetes mellitus. Acta Obstet Gynecol Scand 2004;83:46.

Ivana Pavlic´-Renar, MD, PhD Senior Assistant, Medical Faculty, University of Zagreb, Croatia Head Department of Diabetology, University Clinic Vuk Vrhovac Dugi dol 4a, HR–10000 Zagreb (Croatia) Tel. ⫹385 1 233 1408, Fax ⫹385 1 233 1515, E-Mail [email protected]

Pavlic´-Renar/Metelko

8

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 9–17

Fuel-Mediated ‘Functional Teratogenesis’ and Primary Prevention Andreas Plagemann Division of Experimental Obstetrics, Clinic of Obstetrics, Charité University of Medicine Berlin, Campus Virchow-Klinikum, Berlin, Germany

During recent years, the impact of the intrauterine and early postnatal environment on a lasting determination of fundamental processes of life has been more and more accepted. Especially, investigations and hypotheses by the groups of Barker [1] and Hales and Baker [2] led to the postulation of a so-called ‘small-baby syndrome’ which was explained by a ‘thrifty phenotype’, acquired by ‘poor fetal nutrition’. This concept has mainly contributed to attention being given worldwide to the phenomenon of early epigenetic conditioning, and terms like ‘nutritional programming’ or ‘imprinting’ were proposed to describe it. However, these concepts and observations are not so new. In particular, already in 1979 Freinkel and Metzger [3] have postulated the concept of ‘fuel-mediated teratogenesis’ of lasting deleterious consequences resulting from fetal exposure to a diabetic intrauterine environment [4]. In the same year, Aerts and Van Assche from Belgium provided fundamental experimental evidence for this assumption [5]. Some years before, it was Dörner [6] who was the first (1974) to postulate a general aetiological concept on ‘epigenetic’, perinatal ‘programming’ of the lifetime function of fundamental regulatory systems and, thereby, also of possible disorders and diseases throughout later life. Already in the early 1970s in a series of clinical as well as experimental studies he demonstrated that especially hormones are environment-dependent organizers of the neuroendocrine system, which finally regulates all fundamental processes of life. When present in nonphysiological concentrations, induced by alterations of the intrauterine and/or early postnatal environment, hormones can also act as ‘endogenous functional Article based upon the ‘Joseph Hoet Research Award’ Lecture delivered at the Diabetic Pregnancy Study Group Meeting in Luso, Coimpra, Portugal, September 2004.

teratogens’ by ‘malprogramming’ the ‘neuroendocrine-immune network’, leading to developmental disorders and diseases throughout life. This means, the classical science of ‘teratology’ as the discipline of exogenously induced macroscopic malformations should be supplemented by the science of ‘functional teratology’ as the discipline of perinatally acquired malfunctions [7]. Elevated insulin levels in fetal and perinatal life are pathognomic in children of mothers with diabetes during pregnancy (type 1 diabetes, type 2 diabetes, gestational diabetes), meanwhile affecting, e.g., about every 10th pregnant woman in Germany [8]. Epidemiological and clinical evidence has been accumulating from studies by a variety of authors, like Freinkel and Metzger [3], Dabelea et al. [9], Pettitt et al. [10], Silverman et al. [11] and Weiss et al. [12] as well as from our own group [13–18], showing that offspring exposed to maternal diabetes in utero are at increased risk of becoming obese and developing diabetes themselves. Most interestingly, in these studies it was shown that this acquired disposition may occur even irrespective of the genetic background but seems to depend, at least in part, on the fetal insulin levels and perinatal hyperinsulinism [11–13, 16]. Confirming these clinical observations on a critical role of perinatal insulin levels for a lasting ‘malprogramming’, even independent of birth weight [16], experimental evidence was accumulating showing that fetal and neonatal exposure to maternal diabetes may predispose to overweight and diabetes in later life. Interestingly, in rats [13, 15, 19–22] and even in rhesus monkeys [23] a lasting deleterious impact of fetal or neonatal insulin treatment could be demonstrated on the later risk of becoming overweight and developing diabetes and alterations typical for the metabolic syndrome X. With regard to hormones as dose-dependent self-organizers of their own neuroendocrine regulatory systems we, therefore, hypothesized that insulin itself, when occurring in elevated concentrations during critical perinatal periods of development, may contribute to a lasting ‘malprogramming’ of neuroendocrine systems regulating body weight and metabolism. In this respect note that insulin is an important modulator of central nervous development and growth. The ‘early experience’ of elevated insulin concentrations during ‘critical periods’ of neural coding might, therefore, lead to a ‘malprogramming’ of central nervous regulators of body weight and metabolism. To experimentally investigate this working hypothesis, we used and created different models of perinatal hyperinsulinism, namely offspring of diabetic mother rats, neonatally insulin-treated rats and neonatally overfed rats. In the first model (streptozotocin treatment at the day of conception in genetically indifferent mother rats), the typical perinatal hyperinsulinaemia occurred in the offspring of gestational diabetic mothers [13, 15, 24–26]. Crucial for our hypothesis was the observation that, indeed, hyperinsulinaemia

Plagemann

10

was accompanied by an elevation of insulin concentrations within the hypothalamus in perinatal life [24]. This is not self-evident, since the blood-brain barrier is characterized by a saturable transport system for insulin. However, during fetal and early postnatal life, these mechanisms are not yet mature, thereby obviously allowing increased insulin leakage from the circulation into the hypothalamus. During later life, lasting into old adult age, the offspring of gestational diabetic mother rats were characterized by hyperphagia, overweight, and impaired glucose tolerance [13, 15, 26]. All these findings were accompanied by persisting basal hyperinsulinaemia. This means the ‘experience’ of hyperinsulinism in early life obviously led to hyperinsulinism throughout life. In female F1 offspring, this perinatally acquired adipogenic and diabetogenic disposition resulted in spontaneous gestational diabetes during their own pregnancies, after mating with normal males. Consequently, the F2 offspring developed perinatal hyperinsulinism, accompanied again by basal hyperinsulinaemia and impaired glucose tolerance in later life. The same occurred in the maternalside F3 generation [13, 15]. These observations strongly argue for an ‘epigenetic’ non-hereditary mode of transmitting acquired ‘malprogramming’ maternofetally over successive generations mediated intergeneratively by the intrauterine environment provided by maternal diabetes to the next generation in each case (fig. 1). All these findings were also observed in rats treated neonatally with subcutaneous insulin [13, 15, 16]. Therefore, we wondered whether there really exists a long-term adverse effect explicitly caused by exposure of the hypothalamus to elevated insulin levels during critical developmental periods. To investigate this, insulin was applied only and directly into the mediobasal hypothalamus of newborn rats. By means of a stereotactic operation a long-acting insulin was applied, while in the control groups the same volume of an insulin-free, indifferent agar vehicle was given [21, 22]. In vitro studies revealed insulin release from the implants over a period of 4 days. Implants were topographically placed immediately neighbouring the ventromedial hypothalamic nucleus (VMN), which is well known to inhibit food intake as well as pancreatic insulin secretion, and the lateral hypothalamic area (LHA) which stimulates food intake and insulin release [27]. When those rats were followed up into adult age exciting data were obtained in animals intrahypothalamically treated with insulin on their 2nd or 8th day of life [21, 22]. Beginning at the latest at the age of 3 weeks, that means at the end of the critical hypothalamic differentiation period, rats treated neonatally with intrahypothalamic insulin became obese, which persisted throughout life. This overweight was accompanied by strongly impaired glucose tolerance, an increase in mean daily food intake, and lifelong persistent basal hyperinsulinaemia. These data clearly indicate that an only temporary, intrahypothalamic elevation of insulin levels during ‘critical windows’ of brain development may be a neuroendocrine teratogenic risk factor. This ‘epigenetic’ risk factor, as it

Fuel-Mediated ‘Functional Teratogenesis’ and Primary Prevention

11

Maternal diabetes mellitus during pregnancy

Primary prevention

Fetal and/or early postnatal overnutrition

Intrauterine growth retardation (‘low birth weight’)

Fetal and/or early postnatal hyperinsulinism (hyperleptinism, hypercortisolism)

Permanent malprogramming of the ‘neuro-endocrine-immune system’ (particularly of hypothalamic regulatory centres of food intake, body weight, and metabolism)

Perinatally acquired disposition to obesity, diabetes mellitus, and the Metabolic Syndrome X

Maternal phenotype of female offspring during their pregnancy: overweight and impaired glucose tolerance

Fig. 1. Concept on ‘functional teratogenesis’ and possible primary prevention of a nonhereditary, maternofetal transmission of an increased disposition to obesity, diabetes, and metabolic syndrome X, passed on epigenetically to succeeding generations of maternal descendance.

occurs due to perinatal hyperinsulinaemia in the offspring of gestational diabetic mothers, may cause a permanent, lifetime disposition to obesity and diabetes [13]. The question arises as to which mechanisms might underlie these phenomena of neuroendocrine ‘malprogramming’. As already mentioned, the VMN is well known to inhibit food intake and pancreatic insulin secretion. We, therefore, were interested in characterizing this important hypothalamic nucleus in our experimental rat models. Computer-aided morphometric analyses revealed decreased numbers and decreased sizes of neurons in the VMN of rats neonatally treated with intrahypothalamic insulin, while no morphometric alterations occurred in the LHA. Moreover, in rats neonatally treated with subcutaneous insulin and, most importantly, in the offspring of gestational diabetic mother

Plagemann

12

rats, exactly the same alterations were observed, i.e., hypotrophy and hypoplasia of the VMN, and these at weaning as well as at adult age [13, 15, 19, 25]. In conclusion, these observations in three different models of perinatal hyperinsulinism suggest that a perinatally acquired dysplasia of the VMN might contribute to the development of a disposition to hyperphagia, overweight and hyperinsulinaemia throughout life. Interestingly, similar hypothalamic alterations were also observed in early postnatally overfed rats. At this point we cannot but ask how all these findings might fit in with the ‘Barker hypothesis’ on the ‘small-baby syndrome’ and ‘thrifty phenotype’. It is noteworthy that there is no doubt about the crucial aetiogenetic role of overweight and obesity in the pathogenesis of the metabolic syndrome X. Interestingly enough, already in the mid 1970s, in their original study Ravelli et al. [28] showed that early fetal undernutrition was associated with becoming obese later in life, whereas late fetal and early postnatal caloric restriction led to decreased rates of obesity in young adults. These observations give rise to the suggestion that not (only) early fetal undernutrition but late fetal and early postnatal overfeeding may lead to a lasting disposition to obesity and consecutive metabolic and atherogenic risks. Rats reared in small litters have variously been proven to be an appropriate model to study consequences of early postnatal overfeeding. While rats reared in small litters rapidly develop obesity, rats undernourished due to rearing in large litters become underweight until weaning [13, 29, 30]. In a variety of studies, overweight in small litter rats was observed to persist from early postnatal through juvenile into adult age, although after weaning a standard diet was provided for all groups of rats. Interestingly, already 30 years ago Miller and Personage [31] have even demonstrated a strong inverse correlation between the body fat content in adult life and the neonatally adjusted litter size in early life. With regard to our main hypothesis, we first investigated whether both hyperinsulinaemia and elevated hypothalamic insulin levels also occur in this animal model of early neonatal overfeeding and, indeed, similar to the offspring of gestational diabetic mother rats, early postnatally overfed small litter rats displayed not only hyperinsulinaemia but also an increase of intrahypothalamic insulin concentrations during early postnatal life [29]. In accordance with other investigators, during later life, persisting into old adult age, in early postnatally overfed rats we observed hyperinsulinaemia, as well as hyperphagia, overweight, an impaired glucose tolerance, and an increase of systolic blood pressure [29]. Thus, a complex of symptoms characteristic of the metabolic syndrome X occurred in neonatally overfed rats. Interestingly, however, in rats underfed early postnatally due to nurturing in large litters neither overweight nor hyperinsulinaemia, impaired glucose tolerance or hypertension were observed in later life [13].

Fuel-Mediated ‘Functional Teratogenesis’ and Primary Prevention

13

In this context recent epidemiological data of Stettler et al. [32] are most noteworthy. In a series of impressive studies in different populations with thousands of participants they could clearly demonstrate that rapid weight gain in neonatal life is associated with an increased risk of overweight and obesity in later life, even independent of birth weight and weight at the age of 1 year. These data strongly confirm and expand earlier observations by Dörner et al. [33]. Rapid early weight gain, however, may mainly result from neonatal overfeeding. Looking for mechanisms possibly involved in neuroendocrine ‘malprogramming’ we were interested in further hypothalamic systems involved in the regulation of food intake and body weight control. Neuropeptide Y (NPY) plays a key role in these regulatory processes, especially by acting in the orexigenic arcuate-paraventricular axis. This hypothalamic system consists of NPYexpressing neurons in the hypothalamic arcuate nucleus (ARC), which project to the paraventricular nucleus (PVN). The expression and release of NPY within this axis is inhibited by circulating insulin and leptin, while fasting and a decrease of insulin and leptin lead to an activation of the NPY system, thereby stimulating food intake, in particular [34]. In our experiments [29, 30], as expected, hyperinsulinaemia and obesity in small litter rats were associated with a strong increase of leptin concentrations, while the hypoinsulinaemic, underweight rats reared in large litters displayed decreased leptin levels. As further expected, hypoleptinaemia, hypoinsulinaemia and underweight in early underfed rats were accompanied by a physiological stimulation of the hypothalamic NPY system indicated by an increased number of NPY neurons in the ARC and increased levels of NPY in the PVN. Most importantly, however, in the neonatally overnourished obese rats, characterized by elevated insulin and leptin levels, no decrease and suppression of NPY was observed, but even increases in both the number of NPY neurons in the ARC as well as in the NPY concentrations in the PVN occurred [30]. In our opinion, these findings strongly indicate a ‘malorganization’ and ‘malprogramming’ of the hypothalamic NPY system induced by overfeeding during the ‘critical period’ of early postnatal life. An acquired hypothalamic resistance to the circulating satiety hormones leptin and insulin could be suggested [29, 30]. If so, the most important question is whether this hypothalamic resistance to insulin and leptin may persist throughout later life. This would strongly indicate a neonatally acquired lasting ‘malprogramming’. Therefore, we investigated the electrophysiological responsiveness to leptin in arcuate neurons from hypothalamic brain slices of juvenile as well as adult rats reared in small litters compared to those reared in normal litters, and the following results were obtained [35]. Although all rats were fed a standard pellet diet after weaning on the 21st day of life, early postnatally overfed rats were hyperphagic and overweight throughout the study period. Interestingly, persisting hyperphagia and overweight in small

Plagemann

14

litter rats were accompanied by a nearly complete unresponsiveness of arcuate neurons to leptin. The number of neurons responding to leptin application with decreased firing rates was strongly reduced. Despite baseline activity similar to that in normal rats, arcuate neurons of small litter rats were not inhibited by leptin [35]. In our opinion, these data strongly indicate a neonatally acquired persisting hypothalamic leptin resistance in early overfed rats. We consequently wondered whether a ‘malorganization’ of the hypothalamic NPY system also occurs in the offspring of diabetic mothers. Indeed, from weaning until old adult age hyperphagia, overweight and persistent hyperinsulinaemia in the offspring of gestational diabetic mother rats were found to be associated with a persistently increased number of neurons expressing NPY in the ARC. Even positive correlations between the number of NPY neurons and the daily mean food intake as well as relative body weight were observed [26]. It is worth noting that similar findings were observed, e.g., with regard to hypothalamic neurons expressing galanin, a neuropeptide which especially stimulates the ingestion of fat [24, 26, 29]. Taken together, these exemplary observations in different models of fetal and neonatal hyperinsulinism and hyperleptinism suggest that a perinatally acquired ‘malorganization’ of orexigenic and/or anorexigenic neurons in the ARC might contribute to the occurrence of hyperphagia, overweight, and hyperinsulinaemia throughout later life. These persistent alterations seem to be a consequence, at least in part, of elevated insulin and leptin levels during ‘critical periods’ of early development. In our opinion, data strongly indicate a perinatally acquired, persisting hypothalamic resistance to insulin and leptin in perinatally hyperinsulinaemic and hyperleptinaemic rats. In summary, maternal diabetes during pregnancy and early postnatal overfeeding may lead to a complex of syndrome X-like alterations throughout life (fig. 1). Since mechanisms of early programming of obesity and diabetes are unclear, a complex ‘neuroendocrine malprogramming’ of the regulation of body weight and metabolism may provide a general aetiopathogenetic concept in this context. The association, e.g., between elevated insulin concentrations during early development and acquired alterations in hypothalamic regulatory areas might indicate processes of disturbed ‘hormone-dependent self-organization’ and ‘programming’ of these nuclei. Moreover, a modified neural coding and the resultant disposition to obesity and diabetes might be passed on to succeeding generations in a non-hereditary way, because of the resultant metabolic/’diabetic’ alterations creating the maternal intrauterine environment provided in each case by female offspring during their own gestation for the next generation (fig. 1). With regard to the widely reflected ‘small-baby syndrome’ and ‘thrifty phenotype hypothesis’ we additionally would like to propose that early postnatal overfeeding of underweight newborns may substantially contribute to their long-term risk,

Fuel-Mediated ‘Functional Teratogenesis’ and Primary Prevention

15

a potential mechanism which has rarely been considered so far in interpretations of the ‘Barker hypothesis’. However, from a clinical point of view all these observations point to the possibility of primary prevention of a lifelong increased disposition to obesity, diabetes, and consecutive risks by consequent screening for and treatment of maternal diabetes during pregnancy and by avoiding early postnatal overfeeding (fig. 1).

References 1 2 3

4 5 6 7 8

9

10 11 12 13

14

15

16

17

18

Barker DJP: In utero programming of chronic disease. Clin Sci 1998;95:115–128. Hales CN, Barker DJP: Type 2 (non-insulin-dependent) diabetes mellitus: The thrifty phenotype hypothesis. Diabetologia 1992;35:595–601. Freinkel N, Metzger BE: Pregnancy as a tissue culture experience: The critical implications of maternal metabolism for fetal development; in Pregnancy Metabolism, Diabetes, and the Fetus. Ciba Foundation Symposium 63. Amsterdam, Excerpta Medica, 1979, pp 3–23. Freinkel N: Of pregnancy and progeny. Banting lecture 1980. Diabetes 1980;29:1023–1035. Aerts L, Van Assche FA: Is gestational diabetes an acquired condition? J Dev Physiol 1979;1: 219–225. Dörner G: Hormones and Brain Differentiation. Amsterdam, Elsevier, 1976. Dörner G: Problems and terminology of functional teratology. Acta Biol Med Ger 1975;34: 1093–1095. Kleinwechter H: The government sponsored model project gestational diabetes (GDM) Schleswig-Holstein: Prevalence and foetal outcome in unselected women following the successful implementation of screening for GDM. Diabetologia 2000;43(suppl 1):A56. Dabelea D, Hanson RL, Lindsay RS, Pettitt DJ, Imperatore G, Gabir MM, Roumain J, Bennett PH, Knowler WC: Intrauterine exposure to diabetes conveys risks for type II diabetes and obesity: A study of discordant sibships. Diabetes 2000;49:2208–2211. Pettitt DJ, Baird HR, Aleck KA, Bennett PH, Knowler WC: Excessive obesity in offspring of Pima Indian women with diabetes during pregnancy. N Engl J Med 1983;308:242–245. Silverman BL, Metzger BE, Cho NH, Loeb CA: Impaired glucose tolerance in adolescent offspring of diabetic mothers. Diabetes Care 1995;18:611–617. Weiss PAM, Scholz HS, Haas J, Tamussino KF, Seissler J, Borkenstein MH: Long-term follow-up of infants of mothers with type I diabetes. Diabetes Care 2000;23:905–911. Dörner G, Plagemann A: Perinatal hyperinsulinism as possible predisposing factor for diabetes mellitus, obesity and enhanced cardiovascular risk in later life. Horm Metab Res 1994;26: 213–221. Dörner G, Plagemann A, Reinagel H: Familial diabetes aggregation in type I diabetics: Gestational diabetes an apparent risk factor for increased diabetes susceptibility in the offspring. Exp Clin Endocrinol 1987;89:84–90. Dörner G, Plagemann A, Rückert JC, Götz F, Rohde W, Stahl F, Kürschner U, Gottschalk J, Mohnike A, Steindel E: Teratogenetic maternofoetal transmission and prevention of diabetes susceptibility. Exp Clin Endocrinol 1988;91:247–258. Harder T, Kohlhoff R, Rohde W, Dörner G, Plagemann A: Perinatal ‘programming’ of insulin resistance in later life: Critical impact of neonatal insulin and low birth weight in a risk population. Diabet Med 2001;18:634–639. Plagemann A, Harder T, Kohlhoff R, Rohde W, Dörner G: Overweight and obesity in infants of mothers with long-term insulin-dependent diabetes or gestational diabetes. Int J Obes 1997;21: 451–456. Plagemann A, Harder T, Kohlhoff R, Rohde W, Dörner G: Glucose tolerance and insulin secretion in infants of mothers with pregestational insulin-dependent diabetes mellitus or gestational diabetes. Diabetologia 1997;40:1094–1100.

Plagemann

16

19 20 21

22

23

24

25

26 27 28 29

30

31 32 33

34 35

Harder T, Plagemann A, Rohde W, Dörner G: Syndrome X-like alterations in adult female rats due to neonatal insulin treatment. Metabolism 1998;47:855–862. Ogata ES, Collins JW, Finley S: Insulin injection in the fetal rat: Accelerated growth and altered fetal and neonatal glucose homeostasis. Metabolism 1988;37:649–655. Plagemann A, Heidrich I, Götz F, Rohde W, Dörner G: Lifelong enhanced diabetes susceptibility and obesity after temporary intrahypothalamic hyperinsulinism during brain organization. Exp Clin Endocrinol 1992;99:91–95. Plagemann A, Heidrich I, Rohde W, Götz F, Dörner G: Hyperinsulinism during differentiation of the hypothalamus is a diabetogenic and obesity risk factor in rats. Neuroendocrinol Lett 1992;14:373–378. Susa JB, Boylan JM, Sehgal P, Schwartz R: Impaired insulin secretion after intravenous glucose in neonatal rhesus monkeys that had been chronically hyperinsulinemic in utero. Proc Soc Exp Biol Med 1992;199:327–331. Plagemann A, Harder T, Rake A, Melchior K, Rittel F, Rohde W, Dörner G: Hypothalamic insulin and neuropeptide Y in the offspring of gestational diabetic mother rats. Neuroreport 1998;9: 4069–4073. Plagemann A, Harder T, Janert U, Rake A, Rittel F, Rohde W, Dörner G: Malformations of hypothalamic nuclei in hyperinsulinaemic offspring of gestational diabetic mother rats. Dev Neurosci 1999;21:58–67. Plagemann A, Harder T, Melchior K, Rake A, Rohde W, Dörner G: Elevation of hypothalamic neuropeptide Y-neurons in adult offspring of diabetic mother rats. Neuroreport 1999;10:3211–3216. Szabo AJ: CNS Regulation of Carbohydrate Metabolism. New York, Academic Press, 1983. Ravelli GP, Stein ZA, Susser MW: Obesity in young men after famine exposure in utero and early infancy. N Engl J Med 1976;295:349–353. Plagemann A, Harder T, Rake A, Voits M, Fink H, Rohde W, Dörner G: Perinatal elevation of hypothalamic insulin, acquired malformation of hypothalamic galaninergic neurons, and syndrome X-like alterations in adulthood of neonatally overfed rats. Brain Res 1999;836:146–155. Plagemann A, Harder T, Rake A, Waas T, Melchior K, Ziska T, Rohde W, Dörner G: Observations on the orexigenic hypothalamic neuropeptide Y-system in neonatally overfed weanling rats. J Neuroendocrinol 1999;11:541–546. Miller DS, Personage SR: The effect of litter size on subsequent energy utilization. Proc Nutr Soc 1971;31:30–31. Stettler NS, Zemel BS, Kumanyika S, Stallings VA: Infant weight gain in a multicenter, cohort study. Pediatrics 2002;109:194–199. Dörner G, Grychtolik H, Julitz M: Überernährung in den ersten drei Lebensmonaten als entscheidender Riskofaktor für die Entwicklung von Fettsucht und ihrer Folgeerkrankungen. Dtsch Gesundheitswes 1977;32:6–9. Kalra SP, Kalra PS: Nutritional infertility: The role of the interconnected hypothalamic neuropeptide Y-galanin-opioid network. Front Neuroendocrinol 1996;17:371–401. Davidowa H, Plagemann A: Decreased inhibition by leptin of hypothalamic arcuate neurons in neonatally overfed young rats. Neuroreport 2000;11:2795–2798.

Andreas Plagemann, MD Head of ‘Experimental Obstetrics’ Clinic of Obstetrics, Charité University of Medicine Berlin Campus Virchow-Klinikum, Augustenburger Platz 1 DE–13353 Berlin (Germany) Tel. ⫹49 30 450 52 40 41, Fax ⫹49 30 450 52 49 22, E-Mail [email protected]

Fuel-Mediated ‘Functional Teratogenesis’ and Primary Prevention

17

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 18–33

Endocrine Changes in Diabetic Pregnancy Alexandra Kautzky-Willer, Dagmar Bancher-Todesca Division of Endocrinology and Metabolism, Department of Internal Medicine III and Department of Obstetrics and Gynecology, University of Vienna, Vienna, Austria

The endocrinologic changes accompanying human pregnancy cause a physiological prodiabetogenic state. Pregnancy is a stress test for glucose homeostasis of women and precipitates glucose intolerance or diabetes in women with a preestablished latent metabolic syndrome or impaired glucose metabolism similar to a cortisol glucose tolerance test. Pregnancy is thus a physiological model of insulin resistance and ␤-cell stress and gives the opportunity to study physiological adaptive mechanisms and the defects underlying ␤-cell failure. Gestational diabetes mellitus (GDM) can serve as a model to assess early metabolic changes during the development of diabetes mellitus type 2 (DM2), which is an increasing general health problem in westernized populations.

Pregnancy-Associated Insulin Resistance

During pregnancy profound changes occur in the maternal metabolism in order to supply adequate nutrition to the fetus. In the first trimester a slight increase in insulin sensitivity has been found in normal and diabetic pregnancies in some studies although the characteristic feature of the pregnant state is insulin resistance, which emerges in the second trimester and is most prominent late in the third trimester. The development of insulin resistance leads to increased maternal and thus fetal concentrations of glucose, free fatty acids (FFAs) and amino acids. These changes are essential for the regulation of energy metabolism and fetal growth.

In normal pregnancy insulin sensitivity is decreased by 30–60% compared to healthy nonpregnant women as measured by intravenous and oral glucose tolerance tests (OGTT) or the clamp technique (table 1, fig. 1) [1–13]. Many studies compared insulin secretion and insulin action between women with GDM and healthy control women both during pregnancy and/or after delivery and found besides a persisting defect in first phase insulin secretion variable degrees in insulin resistance (table 1). In some reports insulin resistance abated postpartum while most studies show that insulin sensitivity increases in GDM but remains significantly lower compared to control women at the reinvestigation after delivery (table 1). Of note pregnancy often unmasks a preexisting (genetic) defect in insulin sensitivity in women with GDM. It was shown that GDM is characterized by an augmented glucose response to glucose administration and by a 50% reduction of insulin sensitivity compared to women with normal glucose tolerance (NT) [7]. Following delivery, the glucose response decreased markedly, being still higher than in control women, and insulin sensitivity improved but remained approximately 50% lower compared to weight-matched control women. These data are in line with data in obese DM2 patients who showed that approximately 60% of their insulin resistance was due to diabetes. Therefore, the persisting 50% decrease of insulin sensitivity in subjects with previous GDM might be attributed to the (pre-)diabetic state per se or to glucose toxicity. A few studies also investigated the potential role of hepatic insulin resistance as contributor to hyperglycemia in pregnancy. In human pregnancy both women with NT and those with GDM showed an increase in hepatic glucose production by 30% in late pregnancy but taking into account the marked increase in fasting plasma insulin in women with GDM compared to those with NT this observation seems to confirm the presence of hepatic insulin resistance in women with GDM [1]. As for DM2 the pathophysiological changes responsible for decreased insulin sensitivity in normal and diabetic pregnancy are not clarified but seem to be at a postreceptor level in the insulin signaling pathway. Most studies suggest that there is no significant insulin receptor defect in normal or diabetic pregnancy. In biopsies of the rectus abdominis muscle and of omental adipocytes derived during cesarean sections decreased GLUT-4 transporter concentrations were found in adipocytes but not in the muscles of women with GDM [14]. Given the fact that muscle is responsible for 80% of total insulindependent glucose uptake this defect does not seem to mainly contribute to insulin resistance. Insulin binding to skeletal muscle insulin receptors was also found to be diminished in pregnancy but no difference was found between pregnant women with normal or impaired glucose tolerance.

Endocrine Changes in Diabetic Pregnancy

19

Table 1. Insulin sensitivity and insulin secretion in women with GDM and healthy women with NT during pregnancy and in GDM postpartum (pGDM) compared to nonpregnant healthy controls Investigator

Method

Insulin sensitivity, % of controls

Insulin secretion (first phase insulin secretion), % of controls

GDM

NT

pGDM GDM

45

NM

Catalano et al. [1]

Euglycemic clamp

40

Ryan et al. [2]

Euglycemic clamp

27

Kirwan et al. [3]

Euglycemic clamp

35

66

NM

Homko et al. [4]

Hyperglycemic clamp

44

50

58

C-peptide kinetik study

NT

pGDM

⫹

⫹⫹

⌵⌴

67

74

170

200

85

Cousins [5]

FSIGT

31

43

65

Buchanan et al. [6]

FSIGT

36

30

NM

⫹(AIRg: 63) ⫹⫹(AIRg: 290) NM

Kautzky-Willer et al. [7]

FSIGT

17

36

51

⫹(66 of NT)

⫾(⫾)

⫺(62)

Bowes et al. [8]

FSIGT

47

77

75

⫹(⫾)

⫹(⫹)

⫾(⫾)

Ward et al. [9]

FSIGT

NM

NM 34

NM

NM

⫾(AIRg: 48)

Kousta et al. [10]

FSIGT

NM

NM 40

NM

NM

(AIRg: 95)

Kautzky-Willer et al. [11, 12]

OGTT

62

78

⫹(⫺)

⫹(⫹⫹)

⫾(⫺)

Bartha et al. [13] Insulin tolerance test

72 of NT

87 NM

Insulin sensitivity and insulin secretion of pregnant women with GDM or NT in the third trimester given as percent of the mean insulin sensitivity of nonpregnant healthy controls or women with NT (reference group of each study). ⫹ ⫽ Increased; ⫺ ⫽ decreased; ⫾ ⫽ unchanged compared to controls; NM ⫽ not measured; AIRg ⫽ acute insulin response to glucose.

Kautzky-Willer/Bancher-Todesca

20

p ⬍0.005

· m⫺2)

p⬍ 0.01

p⬍ 0.05

⫺1

10 8 6 4

OGTT: OGIS (ml· min

FSIGT: SI [104 · min⫺1 · (␮U/ml)⫺1]

12

p ⬍0.05 p ⬍0.05

2 0

p⬍ 0.005

600

p⬍ 0.01 p⬍ 0.05

500 400

p⬍ 0.01

p ⬍ 0.05

300 200 100 0

GDM

NT

GDMp

C

GDM

NT

GDMp

C

Fig. 1. Insulin sensitivity indices derived from FSIGT (SI) or OGTT (OGIS) in pregnant women with GDM or NT in comparison to nonpregnant control women (C) matched for age and BMI. Insulin sensitivity indices are also shown in the women with prior GDM at the postpartum investigation (GDMp).

In human skeletal muscle in late pregnancy normal levels of GLUT-4 transporters but impaired insulin receptor autophosphorylation was found. Obese women with GDM showed a decrease in tyrosine phosphorylation of the insulin receptor ␤-subunit associated with a further decrease in glucose transport activity. Furthermore, muscles of women with GDM had increased plasma cell membrane glycoprotein-1 content associated with reduced insulin receptor phosphorylation and insulin receptor tyrosine kinase activity [15]. In addition insulin’s ability to suppress FFA levels declined from early to late gestation in obese pregnant women with GDM and NT but was less in GDMs [16]. Abdominal subcutaneous adipose tissue biopsies revealed no change in insulin signaling protein expression in women with NT compared to the nonpregnant state but lower tissue IRS-1 protein levels and 2-fold higher p85␣ subunit of PI-3 kinase in adipocytes of GDM than in NT. Steady-state levels of PPAR␥ mRNA were decreased to a similar extent both in GDM and NT compared to nonpregnant control women. Thus, decreased IRS-1 but increased p85␣ protein levels in adipose tissue and their association with insulin signaling may contribute to impaired suppression of FFA by insulin in GDM. Reduced insulin action in late pregnancy could also be related to changes in fatty acid composition of muscles. Although no data on intramyocellular lipid content (IMCL) have been available up to now in pregnant women it has been shown by measurement with 1H nuclear magnetic resonance spectroscopy that increased IMCL in soleus muscles and in tibialis anterior muscles (IMCL-T) characterized insulin-resistant women with prior GDM 3 months after delivery. Furthermore, insulin-sensitive

Endocrine Changes in Diabetic Pregnancy

21

women with prior GDM requiring insulin therapy during pregnancy had higher IMCL than those on nutrition therapy only [17]. IMCL was inversely related to the gestational week of the diagnosis of GDM and related to body fat content and plasma leptin but only IMCL-T correlated with insulin sensitivity and glucose tolerance independent of adiposity. Women with prior GDM also had higher plasma FFA concentrations than women with NT during pregnancy at the postpartum investigation, which were closely related to their IMCL. At present, a cause-effect relationship between insulin resistance and IMCL and the mechanisms of potential interaction are not clear. Increased dietary fat supply and/or lipolysis in fat tissue may lead to increased plasma FFA, which could directly induce insulin resistance or could be channeled preferentially into triglycerides. Increased lipolysis of IMCL would increase cytosolic long-chain acyl-CoA which associate with insulin resistance and a decrease in insulin receptor substrate-1 phosphorylation. The exact mechanism underlying insulin resistance during pregnancy is still unclear but the mix of hormonal changes occurring during pregnancy seems to be responsible for changes in glucose and lipid homeostasis [18]. Human placental lactogen (hPL) is a protein hormone produced by the placenta with biological properties similar to growth hormone which might participate, directly or indirectly, in a number of important metabolic processes. Its putative insulin antagonistic actions include an increase of lipolysis with a rise of plasma concentrations of FFA, thereby providing a source of energy for maternal metabolism and fetal nutrition. Increased hPL concentrations also associate with maternal hyperinsulinemia, which favors protein synthesis and provides a mobilizable source of amino acids for transport to the fetus. hPL is supposed to be involved in the pathophysiology of disturbance in fetal metabolism in mothers with diabetes mellitus or GDM. hPL levels increase linearly in maternal plasma, peak in the third trimester of pregnancy and disappear immediately after the removal of the placenta. Its increase parallels the development of insulin resistance during pregnancy. Such a relationship is further supported by the observation that transgenic mice overexpressing the human placental growth hormone gene have markedly reduced insulin sensitivity. Another placental hormone which may affect glucose homeostasis in pregnancy is the steroid hormone progesterone. In fasting women with NT progesterone infusions increased insulin concentrations whereas plasma glucose remained unchanged indicating induction of insulin resistance. Furthermore, high progesterone levels have been shown to be associated with impaired glucose tolerance during pregnancy. This is in line with animal models where progesterone receptor knockout mice had increased insulin secretion and improved glucose tolerance. Thus, progesterone could modulate insulin sensitivity and the capacity of the ␤-cells to compensate for insulin resistance.

Kautzky-Willer/Bancher-Todesca

22

Estrogens have weak insulin antagonistic properties. This is based on their stimulating effect on the liver to increase production of cortisol-binding globulin, thereby enhancing maternal adrenal cortisol production to maintain free cortisol levels. Thus, free and bound cortisol both increase during human pregnancy with a peak in the third trimester. Glucocorticosteroids inhibit peripheral glucose uptake in muscles and increase liver gluconeogenesis, thus antagonizing insulin action and leading to insulin resistance. However, no direct correlation between free cortisol levels and insulin resistance were demonstrated. There does not seem to be an association between increasing prolactin levels and rising insulin resistance during pregnancy. Therefore, prolactin is not likely to play a role in the second half of pregnancy. It may, however, be involved in enhancing cell-to-cell communication among the ␤-cells of the pancreatic islets in early pregnancy. Whereas in the past such an increase of placental reproductive hormones was primarily linked to the development of insulin resistance in human pregnancy, recent studies challenge this historical view. Recently discovered hormones, adipocytokines, and cytokines are now suspected to play a primary role in pregnancy-driven changes in glucose metabolism. Thus, the cytokine tumor necrosis factor-␣ (TNF␣) was found to be a better predictor of insulin resistance in human pregnancy than placental reproductive hormones and cortisol [19]. TNF␣ increases in parallel to the reduction of insulin sensitivity and to the increase of maternal fat mass in pregnancy. TNF␣ and leptin are produced in the placenta and may be mediators of pregnancy-associated insulin resistance. TNF␣ can also induce leptin gene expression. A close association between longitudinal changes of TNF␣ and insulin sensitivity in women with NT and in those with GDM was described. TNF␣ was the most significant independent predictor of insulin resistance irrespective of body fat mass. A change in plasma leptin was the second best predictor but the inverse association between leptin and insulin sensitivity vanished after adjustment for adiposity. In that study cortisol further contributed to a variance in insulin sensitivity in lean and obese pregnant women. Leptin is an adipocyte-derived hormone that could be important in normal and in particular in diabetic pregnancies. It is supposed to regulate body weight through a negative feedback signal between adipose tissue and the hypothalamic center of satiety causing a decrease in food intake and an increase in energy expenditure and body temperature [20]. Furthermore, leptin is found to be higher in females than males and seems to be important for intact reproductive function. Short-term changes in glucose-insulin homeostasis like insulin hypersecretion following glucose ingestion (OGTT) do not affect leptin secretion in pregnant [12] women, whereas leptin is clearly associated with chronic hyperinsulinemia in the late second trimester of pregnancy. Thus, leptin seems to be

Endocrine Changes in Diabetic Pregnancy

23

primarily a long-term adiposity signal rather than a short-term feeding-related satiety signal. In nonpregnant females plasma leptin closely relates to the body mass index (BMI) and body fat mass potentially reflecting leptin resistance in obesity. Such a correlation is also observed during pregnancy, where plasma leptin is increased in all pregnant versus healthy nonpregnant women of comparable age and pregestational BMI [4]. Plasma leptin peaks at around 28 gestational weeks, reaching then a plateau, declines slightly before delivery and falls promptly postpartum. Such an elevation of plasma leptin in pregnancy goes along with both increased free plasma concentrations and alterations of leptinbinding proteins [21]. The cause and functional role of increased leptin release during pregnancy is unclear, although the placenta has been discussed as a major source of leptin synthesis and secretion both into the maternal and the fetal circulation. Recent studies provide evidence that in addition to estrogens insulin and glucocorticoids are the most important determinants for leptin synthesis and secretion. As an inverse relationship is found between plasma leptin concentrations and insulin sensitivity in many cross-sectional studies, leptin could modulate insulin sensitivity and promote fatty acid oxidation and/or decrease insulin secretion [22] also during pregnancy. Of note augmented placental leptin mRNA expression and increased maternal plasma leptin concentrations are in particular found in complicated pregnancies such as in preeclampsia or diabetic women. Higher plasma leptin concentrations are observed in markedly insulin-resistant women with GDM both during pregnancy and postpartum than in hyperglycemic diabetes mellitus type 1 (DM1) pregnant women and healthy females, despite their comparable pregestational weight (fig. 2). Furthermore, plasma leptin correlated with glycemic control, insulin resistance and with body weight and overall maternal weight gain during pregnancy. Although BMI before and during pregnancy was comparable between all groups and overall maternal weight gain even lowest in GDM, BMI postpartum remained significantly higher in GDM than in the women with NT and DM1. Thus, women with GDM were inable to normalize their body weight to the same extent as the other women postpartum when plasma leptin was again related to increased body weight in GDM, leading to the suggestion that plasma leptin at entry to prenatal care could predict postpartum weight retention. Adiponectin is an adipose tissue-specific plasma protein secreted by adipocytes, which decreases hepatic glucose production and increases insulin sensitivity by upregulation of fatty acid oxidation and suppression of TNF␣ secretion [23]. Hypoadiponectinemia was found in various insulin-resistant states and was a marker for subsequent development of DM2 in longitudinal studies. Thus, decreased adiponectin preceded a decrease in whole body insulin sensitivity and has been related to incident diabetes, whereas high plasma

Kautzky-Willer/Bancher-Todesca

24

30

p⬍ 0.0005 p⬍ 0.01

Fasting leptin (ng/ml)

p ⬍0.01

25 p ⬍0.05

20 15 10 5 0 GDM DM1 NT Pregnancy

GDMp DM1p NTp Postpartum

C

Fig. 2. Fasting plasma leptin concentrations in pregnant women with GDM or NT and in pregnant women with DM1 in comparison to nonpregnant control women (C) matched for age and BMI. Fasting plasma leptin is also shown at the postpartum investigation 3 months after delivery in women with prior GDM (GDMp), those with DM1 (DM1p) as well as in women with NT during and after pregnancy (NTp).

adiponectin concentrations were protective against DM2 [24]. In humans plasma adiponectin was found to be associated with skeletal muscle insulin receptor tyrosine phosphorylation. In pregnant women a decrease of adiponectin was shown with increasing insulin resistance in the third trimester and a further decrease in women with IGT or GDM compared to pregnant women with NT even after adjustment for varying degrees of adiposity [25]. In parallel decreased adiponectin mRNA levels were found in abdominal subcutaneous adipose tissue biopsies from women with GDM. Adiponectin was present in cord blood being slightly lower in the offspring of diabetic mothers than healthy mothers [26] but fetal adiponectin levels did not relate to fetal adiposity, insulin or leptin concentrations. Another study in healthy women, however, found higher adiponectin levels in umbilical than maternal serum and a positive correlation between umbilical adiponectin and birth weight [27]. Hypoadiponectinemia was also found in women with prior GDM 3 months after delivery independently of their body fat mass compared to women with NT during and after pregnancy [28]. Furthermore, among the women with a history of GDM the insulin-resistant women showed a significant decrease of plasma adiponectin concentrations 1 year after delivery, whereas the insulinsensitive women maintained their plasma concentrations. Thus, decreased

Endocrine Changes in Diabetic Pregnancy

25

plasma adiponectin concentrations could be involved in the pathophysiology of pregnancy-driven insulin resistance and in the pathogenesis of GDM and DM2. Resistin is a recently identified hormone emanating from fat tissue that could be another important link between insulin resistance, glucose intolerance and adiposity. In mice infusion of recombinant resistin deteriorated while infusion of antiresistin antibodies ameliorated insulin action and glucose metabolism, suggesting that resistin could modulate insulin sensitivity [29]. Resistin protein expression is higher in abdominal than in thigh fat in pregnant women. mRNA expression of resistin was also found in trophoblasts as well as an increase of resistin gene expression and secretion of resistin in the third trimester [30]. Both placenta-derived and fat-derived hormones seem to contribute to increased maternal plasma resistin concentrations in pregnancy. However, up to now the role of resistin during normal or diabetic pregnancy as well as in other insulin-resistant states has not been clear. ␤-Cell Activity in Pregnancy

Insulin Secretion Pregnant women have a 30–60% increase in insulin secretion to overcome pregnancy-associated insulin resistance. In GDM insulin resistance is inadequately compensated by insulin hypersecretion due to defective B cell function, which persists following delivery. Such a defect is seen both following oral glucose administration (OGTT) (fig. 3, 4) and following an intravenous glucose bolus (frequently sampled intravenous glucose tolerance test, FSIGT) (fig. 5). During the OGTT women with GDM show a delayed increase in plasma C-peptide and insulin concentrations despite marked hyperglycemia during the first hour. In parallel the early increment (␦0–30) in ␤-cell hormone concentrations is lower in GDM than in women with NT during pregnancy [12] despite greatly increased plasma glucose concentrations 30 min after 75 g glucose ingestion. The impaired early insulin response is even more evident when the data are analyzed by a mathematical model for reconstruction of the insulin secretion rate during the OGTT (fig. 4). GDM exhibits a significantly higher basal insulin secretion rate as a potential mechanism to compensate for insulin resistance, but the dynamic suprabasal insulin release was decreased in GDM, yielding reduced first and second phase sensitivity of the ␤-cell to glucose during FSIGT; this indicates inadequate B cell secretory capacity in GDM [7] (fig. 5a). This defect persisted after delivery (fig. 5b). Other studies investigating women with GDM before and after delivery or with a history of GDM also showed an inadequate initial insulinogenic response in GDM during FSIGT (first phase insulin secretion),

Kautzky-Willer/Bancher-Todesca

26

9

C-peptide (pmol/l)

Glucose (mmol/l)

12

6 3 0 0

60

120

6,000 4,000 2,000 0

180

0

Time (min)

120

180

Time (min) 25

800 600

IAPP (pmol/l)

Insulin (pmol/l)

60

400 200

20 15 10 5 0

0 0

60

120

180

0

60

120

180

Time (min)

Time (min)

Proinsulin (pmol/l)

100 75 50 25 0 0

60

120

180

Time (min)

Fig. 3. OGTT. Average time courses of glucose, insulin, C-peptide, proinsulin and IAPP plasma concentrations following an oral glucose load (OGTT). 䊊 ⫽ Women with GDM; 䊏 ⫽ pregnant women with NT; 䊉 ⫽ nonpregnant controls [from 6].

which also persisted postpartum despite a reversal of glucose disposition to the normal range. The finding of decreased insulin secretion for a given degree of insulin sensitivity in GDM compared with other insulin-resistant states like obesity or prediabetic conditions like IGT [7] supports the theory that the predominant defect in GDM is impaired insulin secretion, which leads to hyperglycemia when an exaggeration of insulin resistance is superimposed as during pregnancy. Thus, compared with obese subjects with IGT matched for the degree of insulin sensitivity, patients with GDM still showed a reduction in B cell sensitivity and total insulin secretion. Such a secretory defect is also seen postpartum

Endocrine Changes in Diabetic Pregnancy

27

pmol/L/min

300

300

200

200

100

100

0

0 0

a

60

120

180

0

b

Time (min)

60

120

180

Time (min)

pmol/L/min

300

200

100

0 0

c

60

120

180

Time (min)

Fig. 4. OGTT. Model-derived insulin secretion rate in pregnant women (between 24 and 28 gestational weeks) with GDM (a) or NT (b) and healthy nonpregnant women (c) [modified from 6].

when women with prior GDM but NT are compared with obese women with NT matched for insulin sensitivity. Insulin-deficient women with DM1 need an adequate increase in exogenous insulin during pregnancy to compensate for pregnancy-associated insulin resistance and to maintain normoglycemia. They have to be instructed in intensified insulin therapy and must be monitored carefully throughout pregnancy. The fact that insulin resistance returns to the pregestational state (or improves even further due to optimal metabolic control during pregnancy) immediately postpartum has to be considered. In the Diabetes in Early Pregnancy Study (DIEP) declining insulin requirements in the mid-first trimester of pregnancy were reported in women with DM1 [31]. The authors suggest as possible explanations besides overinsulinization in early pregnancy a transient decline in progesterone secretion during the late first-trimester luteoplacental shift in progesterone secretion. Of note such a drop

Kautzky-Willer/Bancher-Todesca

28

Glucose (mmol/l) C-Peptide (pmol/l)

20

20

15

15

10

10

5

5

0

0

2,500

2,500

2,000

2,000

1,500

1,500

1,000

1,000

500

500 0

a

10

20

30

60

120

180

Time (min)

240

0

b

10

20

30

60

120

Time (min)

Fig. 5. FSIGT. Average time course of glucose and C-peptide plasma concentrations following an intravenous glucose load (0.3 g/kg at zero time). a 䊉 ⫽ Women with GDM; 䊊 ⫽ pregnant women with NT (between 24 and 28 gestational weeks). b 䊏 ⫽ Women with GDM 3 months after delivery; 䊐 ⫽ nonpregnant control women. The time axis has been split to better show the patterns during the first half hour [from 5].

in insulin requirements was associated with de novo C-peptide release and thus endogenous insulin secretion in the first trimester in DM1 pregnant women in one study which could be ascribed to growth-promoting factors of pregnancy and potential rejuvenation of the pancreatic islets [32]. This observation together with a transient improvement of insulin sensitivity at this time could further explain the high risk for hypoglycemia in the first trimester in diabetic women. Women with pregestational DM2 are already markedly insulin resistant and have an insulin secretory defect in the nonpregnant state. If they are on nutrition therapy or oral antidiabetic drug therapy these women have to be instructed in insulin therapy before conception in order to achieve normoglycemia to avoid diabetic embriopathy and to be able to adapt insulin demand to the pregnancydriven exaggeration of their insulin resistance. Release of Proinsulin and Islet Amyloid Pancreatic Polypeptide In addition to the described changes in insulin secretion dynamics and insulin kinetics changes in insulin biosynthesis and changes in the release of other ␤-cell hormones or their ratio in comparison to insulin could affect the metabolic situation in normal and in particular in diabetic pregnancy. Thus,

Endocrine Changes in Diabetic Pregnancy

29

180

240

changes in the release of islet amyloid pancreatic polypeptide (IAPP) and/or proinsulin could play a role in the pathophysiology of pregnancy-driven insulin resistance and the interplay with insulin secretion. IAPP is costored and cosecreted with insulin in the ␤-cell. Plasma IAPP concentration is increased in obesity but relatively decreased in DM2 preceding impairment of insulin secretion [33]. In vitro and in vivo studies have shown that IAPP can decrease insulin secretion and in supraphysiological concentrations peripheral insulin sensitivity. Such insulin antagonist action of IAPP led to speculations about its possible role in the pathogenesis of glucose intolerance and DM2. Whereas most studies show that IAPP does not exert an effect on insulin resistance in man, IAPP could exert a physiological role in the regulation of insulin secretion in humans. The deposition of islet amyloid together with islet dysfunction induced by IAPP has also been hypothesized to contribute to a disproportionate proinsulin release and an increased proinsulin/insulin ratio as commonly found in prediabetic and diabetic subjects. In pregnant women markedly increased postprandial IAPP concentrations were found despite unchanged fasting levels (fig. 3) [11]. As IAPP secretion patterns were almost identical in pregnant women displaying either GDM or NT IAPP cannot be regarded a factor in the pathogenesis of GDM. Interestingly, however, in both pregnant groups IAPP secretion was 3–4 times greater than in nonpregnant healthy women following oral as well as intravenous glucose administration and, therefore, seems to be characteristic of pregnancy. Furthermore, the glucose-triggered IAPP and insulin release in pregnant women differed from that in obesity, which in the presence of a similar degree of insulin resistance had a 30% smaller total IAPP secretion than pregnant women despite an approximately 50% greater insulin secretion. This finding is reflected by an almost 5 times greater IAPP/insulin cosecretion factor in pregnant versus nonpregnant obese females, whose respective IAPP/insulin cosecretion factor was similar to that of the lean nonpregnant control women. From this it appears that increased IAPP secretion during pregnancy may serve as a mechanism to curb excessive hyperinsulinemia and protect from late maternal hypoglycemia. In women with GDM, however, a reduction in insulin secretion could further deteriorate the metabolic situation. A physiological effect of increased amylin concentrations in pregnancy after glucose loading both in normal and diabetic pregnancy might be the potential inhibiting effect of IAPP or amylin analogues on gastric emptying with consecutive delay and reduction in postprandial plasma glucose rise [34]. Fasting serum proinsulin concentrations as well as postprandial proinsulin were higher in GDM compared to women with NT or nonpregnant control women (fig. 3). Despite different clearance rates of the two peptides with a slower clearance of proinsulin, the description of the basal proinsulin/insulin

Kautzky-Willer/Bancher-Todesca

30

ratio is useful for characterizing the ␤-cell status. In GDM, in addition, the proinsulin/insulin ratio was increased compared to pregnant women with NT or nonpregnant control women while there was no difference in the proinsulin/ insulin ratio between pregnant and nonpregnant women with NT. Similarly, increased fasting serum proinsulin concentrations, molar proinsulin/insulin ratios and 32,33 split proinsulin levels as well as higher stimulated intact proinsulin levels after oral glucose loading have been reported in GDM. These findings fit the hypothesis that under conditions of increased insulin demand like in pregnancy-related insulin resistance and during hyperglycemia preexisting ␤-cell defects may be unmasked leading to a release of immature granules with an excessive proportion of insulin precursors. Three months after delivery postprandial IAPP levels fell to normal in women with prior GDM. The secretion and kinetic parameters for IAPP showed a clear tendency to normalization and were no more different from that of control women after delivery. In parallel, fasting proinsulin fell postpartum, whereas the fasting and postprandial molar proinsulin/insulin ratio did not change and remained elevated in prior GDM. Such an elevated fasting molar proinsulin/insulin ratio has also been reported in GDM patients despite normal insulin concentrations 3–4 years after pregnancy, but did not predict later impairment of glucose tolerance after 6–7 years [35]. Therefore, the increased proinsulin/insulin ratio could be a marker of B cell dysfunction only during events that lead to insulin resistance such as pregnancy. In summary, a mix of hormones and their interplay with cytokines and the insulin signaling pathway seem to be responsible for the profound metabolic changes occurring during pregnancy. Individual genetic background and environmental factors in particular obesity then determine the ability of a pregnant woman to balance these changes and to maintain a physiological milieu. References 1

2 3

4 5 6

Catalano PM, Tyzbir ED, Wolfe RR, Calles J, Roman NM, Amini SB, Sims EAH: Carbohydrate metabolism during pregnancy in control subjects and women with gestational diabetes. Am J Physiol 1993;264:E60–E67. Ryan EA, O’Sullivan MJ, Skyler JS: Insulin action during pregnancy. Studies with the euglycemic clamp technique. Diabetes 1985;34:380–389. Kirwan JP, Huston-Presley L, Kalhan SC, Catalano PM: Clinically useful estimates of insulin sensitivity during pregnancy: Validation studies in women with normal glucose tolerance and gestational diabetes. Diabetes Care 2001;24:1602–1607. Homko C, Sivan E, Chen X, Reece EA, Boden G: Insulin secretion during and after pregnancy in patients with gestational diabetes mellitus. J Clin Endocrinol Metab 2001;86:568–573. Cousins L: Insulin sensitivity in pregnancy. Diabetes 1991;40(suppl 2):39–43. Buchanon TA, Metzger BE, Freinkel N, Bergman R: Insulin sensitivity and B-cell responsiveness to glucose during late pregnancy in lean and moderately obese women with normal glucose tolerance or mild gestational diabetes. Am J Obstet Gynecol 1990;162:1008–1014.

Endocrine Changes in Diabetic Pregnancy

31

7

8

9

10

11 12 13

14

15

16

17

18 19

20 21

22

23 24

25 26

Kautzky-Willer A, Prager R, Thomaseth K, Pacini G, Streli C, Waldhäusl W, Wagner O, Ludvik B: Pronounced insulin resistance and inadequate B-cell insulin secretion characterize lean gestational diabetes mellitus. Diabetes Care 1997;20:1717–1723. Bowes SB, Hennessy TR, Umpleby AM, Benn JJ, Jackson NC, Boroujerdi MA, Sönksen PH, Lowy C: Measurement of glucose metabolism and insulin secretion during normal pregnancy and pregnancy complicated by gestational diabetes. Diabetologia 1996;39:976–983. Ward WK, Johnston CLW, Beard JC, Benedetti TJ, Halter JB, Porte D: Insulin resistance and impaired insulin secretion in subjects with histories of gestational diabetes mellitus. Diabetes 1985;34:861–869. Kousta E, Lawrence NJ, Godsland IF, Penny A, Anyaoku V, Millauer BA, Cela E, Johnston DG, Robinson S, McCarthy MI: Insulin resistance and beta-cell dysfunction in normoglycemic European women with a history of gestational diabetes. Clin Endocrinol 2003;59:289–297. Kautzky-Willer A, Thomaseth K, Ludvik B, Rabensteiner D, Waldhäusl W, Pacini G, Prager R: Elevated IAPP and proinsulin in lean gestational diabetes. Diabetes 1997;46:607–614. Kautzky-Willer A, Prager R, Ludvik B, Pacini G, Tura A, Bieglmeyer C, Schneider B, Waldhäusl W: Increased plasma leptin in gestational diabetes. Diabetologia 2001;44:164–172. Bartha JL, Comino-Delgado R, Martinez-Del-Fresno P, Fernandez-Barrios M, Bethencourt I, Moreno-Corral L: Insulin sensitivity index and carbohydrate and lipid metabolism in gestational diabetes. J Reprod Med 2000;45:185–189. Garvey WT, Maianu L, Hancock JA, Golichowski AM, Baron A: gene expression of GLUT4 in skeletal muscle from insulin-resistant patients with obesity, IGT, GDM, NIDDM. Diabetes 1992;41:465–475. Shao J, Catalano P, Yamashita H, Ruyter I, Smith S, Youngren J, Friedman JE: Decreased insulin receptor tyrosine kinase activity and plasma cell membrane glycoprotein-1 overexpression in skeletal muscle from obese women with gestational diabetes mellitus. Diabetes 2000;49:603–610. Catalano P, Nizielski SE, Shao J, Preston L, Qiao L, Friedman JE: Downregulated IRS-1 and PPARgamma in obese women with gestational diabetes: Relationship to FFA during pregnancy. Am J Physiol Endocrinol Metab 2002;282:E522–E533. Kautzky-Willer A, Krssak M, Winzer C, Pacini G, Tura A, Farhan S, Wagner O, Brabant G, Horn R, Stingl H, Schneider B, Waldhäusl W, Roden M: Increased intramyocellular lipid concentration identifies impaired glucose metabolism in women with previous gestational diabetes. Diabetes 2004;52:244–251. Chamberlain G, Pipkin FB: Clinical Physiology in Obstetrics, ed 3. Oxford, Blackwell Science, 1998, chap 7, pp 192–211. Kirwan JP, Hauguel-de Mouzon S, Lepercq J, Challier JC, Huston-Presley L, Friedman JE, Kalhan SC, Catalano PM: TNF␣ is a predictor of insulin resistance in human pregnancy. Diabetes 2002;51:2207–2213. Jeanrenaud FR, Jeanrenaud B: Obesity, leptin and the brain. N Engl J Med 1996;334:324–325. Lewandowski K, O’Callaghan CJ, Dunlop D, Medley GF, O’Hare P, Brabant G: Free leptin, bound leptin, and soluble leptin receptor in normal and diabetic pregnancies. J Clin Endocrinol Metab 1999;84:300–306. Emilsson V, Liu YL, Morton NM, Davenport M: Expression of the functional leptin receptor mRNA in pancreatic islets and direct inhibitory action of leptin on insulin secretion. Diabetes 1997;46:313–316. Chandran M, Phillips SA, Ciaraldi T, Henry RR: Adiponectin: More than just another fat cell hormone? Diabetes Care 2003;26:2442–2450. Tschritter O, Fritsche A, Thamer C, Haap M, Shirkavand F, Rahe S, Staiger H, Maerker E, Häring H, Stumvoll M: Plasma adiponectin concentrations predict insulin sensitivity of both glucose and lipid metabolism. Diabetes 2003;52:239–243. Retnakaran R, Hanley AJ, Raif N, Connelly PW, Sermer M, Zinman B: Reduced adiponectin concentration in women with gestational diabetes. Diabetes Care 2004;27:799–800. Lindsay RS, Walker JD, Havel PJ, Hamilton BA, Calder AA, Johnstone FD; Scottish Multicentre Study of Diabetes Pregnancy: Adiponectin is present in cord blood but is unrelated to birth weight. Diabetes Care 2003;26:2244–2249.

Kautzky-Willer/Bancher-Todesca

32

27

28

29 30 31

32

33 34

35

Chan TF, Yuan SS, Chen HS, Guu CF, Wu CL, Yeh YT, Chung YF, Jong SB, Su JH: Correlations between umbilical and maternal serum adiponectin levels and neonatal birthweights. Acta Obstet Gynecol Scand 2004;83:165–169. Winzer C, Wagner OF, Festa A, Schneider B, Roden M, Bancher-Todesca D, Pacini G, Funahashi T, Kautzky-Willer A: Plasma adiponectin, insulin sensitivity, and subclinical inflammation in women with prior gestational diabetes mellitus. Diabetes Care 2004;27:1721–1727. Steppan CM, Bailey SC, Bhat S, Brown EJ, Banerjee RR, Wright CM, Patel HR, Ahima RS, Lazar MA: The hormone resistin links obesity to diabetes. Nature 2001;409:307–312. Yura S, Sagawa N, Itoh H, Kakui K, Nuamah MA, Korita D, Takemura M, Fujii S: Resistin is expressed in the human placenta. J Clin Endocrinol Metab 2003;88:1394–1397. Jovanovic L, Knopp RH, Brown Z, Conley MR, Park E, Mills JL, Metzger BE, Aaron JH, Holmes LB, Simpson JL: Declining insulin requirement in the late first trimester of diabetic pregnancy. Diabetes Care 2001;24:1130–1136. Ilic S, Jovanovic L, Wollitzer AO: Is the paradoxical first trimester drop in insulin requirement due to an increase in C-peptide concentration in pregnant type 1 diabetic women. Diabetologia 2000;43:1329–1330. Ludvik B, Lell B, Hartter E, Schnack C, Prager R: Decrease of stimulated amylin secretion in type II diabetes. Diabetes 1991;40:1615–1619. Kolterman OG, Gottlieb A, Moyses C, Colburn W: Reduction of postprandial hyperglycemia in subjects with IDDM by intravenous infusion of AC137, a human amylin analogue. Diabetes Care 1995;18:1179–1182. Hanson U, Persson B, Hartling SG, Binder C: Increased molar proinsulin-to-insulin ratio in women with previous gestational diabetes does not predict later impairment of glucose tolerance. Diabetes Care 1996;19:17–20.

Alexandra Kautzky-Willer, MD Division of Endocrinology and Metabolism, Department of Internal Medicine III Department of Obstetrics and Gynecology University of Vienna, Währinger Gürtel 18–20, AT–1090 Vienna (Austria) Tel. ⫹43 1 40400 2126, Fax ⫹43 1 40400 4347 E-Mail [email protected]

Endocrine Changes in Diabetic Pregnancy

33

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 34–45

Metabolic Changes in Diabetic Pregnancy Emilio Herrera Facultad de Ciencias Experimentales y de la Salud, Universidad San Pablo-CEU, Madrid, Spain

During pregnancy, changes in carbohydrate, amino acid and lipid metabolism occur to ensure the continuous supply of nutrients to the fetus despite intermittent maternal food intake. While adaptations of carbohydrate and amino acid metabolism in pregnancy are reasonably well known, alterations in lipid physiology and their significance for fetal growth are still little understood. The purpose of this chapter is to review the overall metabolic changes that take place in pregnancy under normal and diabetic conditions, with special attention being given to those related to lipid metabolism. Changes Occurring in the Mother during Pregnancy Affecting Fetal Growth

The continuous supply of metabolites derived from the maternal circulation, across the placenta, sustain fetal development. Whereas the most abundant nutrient crossing the placenta is glucose, followed by amino acids, the transfer of lipid components is limited [1]. However, lipids play a major role in fetal development, as shown by changes in their availability, such as those produced by variations in dietary fat composition, which are known to have major implications on fetal and postnatal development [2]. In addition, deviations from normal maternal plasma lipid status, such as hypercholesterolemia, can trigger pathogenic events in the fetus and may influence atherosclerosis later in life [3]. During the first two thirds of pregnancy the mother develops hyperinsulinemia and normal or even enhanced insulin sensitivity [4, 5], which in combination with hyperphagia and limited fetal growth allows her to store a large proportion of the nutrients she eats causing an accumulation of fat stores. During the last

I.R. I.R.

Under fasti conditio ng ns ⫹ ⫹ Acyl-CoA Ketone TG bodies ⫹ ⫹ Glycerol-3-P Cholesterol Phospholipids Apo B100 I.R. ⫹ I.R. VLDL Liver ⫹ Glucose ⫹

FFA + Glycerol Adipose tissue I.R. ⫺ LPL

Acetyl-CoA

⫺

VLDL (TG)

⫺ I.R.

Glucose

HDL3 (EC)

⫺ HL

Proteins

TG EC

HDL2b (TG)

⫹

TG

CETP TG EC

EC

Aminoacids

LDL (EC,TG)

Muscle EFA, LCPUFA

⫹

⫹

⫹ ⫹

Fetus

Fig. 1. Schematic representation of major interactions of maternal metabolism during late pregnancy with special emphasis on lipid metabolism and with indication of their consequences for the availability of substrates to the fetus and the controlling role of insulin resistance (I.R.). ⫹ ⫽ Activated steps; ⫺ ⫽ inhibited steps; TG ⫽ triacylglycerols; EC ⫽ esterified cholesterol; CETP ⫽ cholesterol ester transfer protein; EFA ⫽ essential fatty acids; LCPUFA ⫽ long-chain polyunsaturated fatty acids.

third of gestation the mother switches from the previous anabolic condition to a catabolic one permitting an enhanced transfer of nutrients through the placenta to sustain the rapid fetal growth. This catabolic condition is enhanced under fasting conditions, and is specially noticed in terms of an enhanced breakdown of lipid stores by lipolysis in adipose tissue, and is facilitated by the development of an overt insulin-resistant condition [6] (fig. 1). Poorly controlled diabetes during the first 7 weeks of human pregnancy has been associated with a spectrum of developmental abnormalities including pre-implantatory embryo loss, increased resorption rate, induction of congenital anomalies, and embryonic developmental and growth delays. Although the mechanism of these effects in humans remains unknown, laboratory studies from diabetic animals have implicated maternal hyperglycemia as the principal teratogenic agent. Several closely interrelated pathways have been shown to be involved in the molecular mechanisms of hyperglycemia-induced tissue

Metabolism in Diabetic Pregnancy

35

damage: overproduction of reactive oxygen species, activation of protein kinase C isoforms, alteration in arachidonic acid metabolism leading to altered prostaglandin and nitric oxide production, increased hexosamine pathway flux, enhanced formation of advanced glycation end products and increased polyol formation. Distinct from the impaired development that results from poorly controlled diabetes during early pregnancy, events occurring as a result of poor control during the latter two thirds of gestation include accelerated fetal growth and a risk of large-for-gestational age infants, respiratory distress syndrome, neonatal hypoglycemia, neonatal hypocalcemia and neonatal hypomagnesemia. It was initially proposed that overgrowth of the fetus in maternal diabetes was the result of increased delivery of glucose to the fetus, which consequently develops premature maturation of pancreatic insulin secretion, and subsequent hyperinsulinemia which together with the excess availability of glucose results in overgrowth of the fetus [7]. Formerly, it was proposed that overgrowth of the fetus of the diabetic mother was the result of the integrated impact of multiple maternal nutrients on fetal development [8]. Reports relating birth weights in diabetic pregnancies to maternal amino acid levels and maternal triacylglycerol levels support the view that fetal growth is controlled by several metabolic factors, maternal glucose being one of them.

Carbohydrate Metabolism

Glucose is the primary energy source of fetoplacental tissues. Under normal conditions, during early pregnancy, basal glucose and insulin concentrations do not differ from nongravid values [9], and hepatic gluconeogenesis is unchanged [10]. However, during late pregnancy the mother tends to develop hypoglycemia, which is especially evident during fasting. Indirect studies in women [11] and direct experiments in rats [12] have shown that the rate of gluconeogenesis is enhanced during pregnancy under fasting conditions, the effect being especially manifest when glycerol is the studied substrate [13]. Thus, gestational hypoglycemia occurs despite the enhanced gluconeogenesis and decreased consumption of glucose by the insulin-resistant tissues (fig. 1), and is therefore the result of enhanced utilization of glucose as a consequence of the high rate of placental transfer of glucose [14]. The placental transfer of glucose is carried out by facilitated diffusion according to concentration-dependent kinetics, being therefore dependent on the positive maternal-fetal glucose gradient, which is maintained by the low concentration of glucose in fetal circulation. Carbohydrate metabolism has been studied in obese and nonobese women who were predisposed to and developed gestational diabetes mellitus (GDM) [15].

Herrera

36

In longitudinal studies of lean women with GDM a progressive decrease in first-phase insulin response in late gestation was found, whereas first-phase insulin response in obese women developing GDM did not change but secondphase insulin response to intravenous glucose challenge increased [16]. Basal glucose production increases similarly in patients with GDM and in control subjects throughout gestation, but in late pregnancy insulin suppression of hepatic glucose production is less in GDM patients than in controls. It was found that in women with insulin-treated GDM at 32–36 weeks of gestation the total energy expenditure, basal metabolic rate, whole-body net carbohydrate and exogenous (dietary) glucose oxidation did not differ from control subjects [17]. Basal glucose concentrations decrease with advancing gestation in women developing GDM, and although at late gestation they have increased fasting insulin levels and decreased hepatic insulin sensitivity, hepatic glucose production was found either increased or unchanged in women with GDM compared to control women. Endogenous hepatic glucose production was shown to remain sensitive to increased insulin concentration in normal pregnancy, but was less sensitive in GDM [10]. In overweight patients with GDM, similar rates of fasting glucose appearance are achieved, but with elevated insulin concentrations relative to pregnant control subjects [18]. Total energy expenditure, basal metabolic rate and whole-body glucose utilization did not differ between insulin-treated GDM patients and controls [17].

Amino Acid Metabolism

The accretion of protein is essential for fetal growth, and nitrogen retention and protein synthesis are increased in pregnancy in both maternal and fetal compartments [19]. Nitrogen balance is improved and dietary protein is used more efficiently in late pregnancy [20]. A decrease in most maternal amino acid concentrations occurs both during early pregnancy, before the accretion of maternal or fetal tissues, and during late pregnancy. Since insulin infusion in healthy adults decreases both plasma amino acid levels and protein breakdown, the decrease in plasma amino acid levels and the lower rate of appearance of leucine found during normal pregnancy [21] indicate that the pregnancyassociated resistance to insulin does not involve muscle protein breakdown. There are also studies in which a decreased insulin sensitivity manifested by a decreased suppression of leucine turnover during insulin infusion in late normal pregnant women, and an increase in basal leucine turnover in women with GDM were found [15]. Studies of protein metabolism in fasted pregnant diabetic subjects been have shown to have normal [18] or higher rates of protein breakdown and oxidation but protein synthesis rates similar to normal pregnant

Metabolism in Diabetic Pregnancy

37

subjects [22], whereas well-controlled type 1 diabetes was found to cause no abnormalities in protein breakdown, synthesis or oxidation [23]. Plasma levels of branched chain amino acids (leucine, isoleucine, and valine) were found to be higher in GDM women during late pregnancy, whereas no change was found in several other amino acids (aromatics, phenylalanine and tyrosine, and proline). Also, there were others like glycine and threonine that were found to be lower in GDM than in normal control women. Although quantification of leucine and phenylalanine kinetics using stable isotope-labeled tracers showed no difference between GDM and control subjects [18], there are reports showing a higher rate of leucine nitrogen turnover in GDM women compared to normal subjects [24]. All of this suggests a significant alteration in maternal protein and amino acid metabolism in GDM women. Distinct from glucose, the concentration of amino acids in fetal plasma is even higher than that found in the mother, because placental transfer of amino acids is carried out by an energy-dependent process, using selective transporters. This ensures the appropriate availability of these essential precursors to the fetus and may actively contribute to the tendency to maternal hypoaminoacidemia. Amino acids have a greater effect than glucose in stimulating fetal insulin secretion, and therefore changes in their delivery to the fetus may have profound consequences on fetal growth. The transport of neutral amino acids, which is mediated by the system A amino acid transporter, across the syncytiotrophoblast microvillous plasma membranes from placentas of women with diabetes has been shown to be either not affected [25], decreased [26] or even increased [27]. The uptake of leucine, but not of lysine or taurine was found increased in microvillous plasma membranes from placentas of GDM but not in those from type 1 diabetic women [27]. Most of these changes did not correlate with infant size, suggesting that they are not the primary cause for accelerated fetal growth in diabetic pregnancy.

Maternal Lipid Metabolism

Accumulation of lipids in maternal tissues as the result of major changes in adipose tissue metabolism and the development of maternal hyperlipidemia are the two most characteristic features of lipid metabolism during pregnancy. Adipose Tissue Metabolism Fat accumulation during pregnancy takes place during the first two thirds of gestation and occurs in both women [28] and experimental animals [29]. Body fat accumulation during early pregnancy appears to be the result of both hyperphagia and enhanced lipid (fatty acid and glyceride glycerol) synthesis in

Herrera

38

adipose tissue [14], and is driven by the enhanced adipose tissue insulin responsiveness found in this early stage of pregnancy [2]. The accumulation of maternal fat stops during the last third of gestation as a consequence of enhanced adipose tissue lipolytic activity. Increased lipolysis of adipose tissue fat stores occurs in both women and rats during the last third of gestation, the change being especially manifest under fasting conditions [30–32]. The products of adipose tissue lipolysis, free fatty acids (FFA) and glycerol, are released, in large part, into the circulation. Since the placental transfer of these products is quantitatively low [1], their main destination is the maternal liver where, after conversion into active forms, acyl-CoA and glycerol-3-phosphate, respectively, they are partially reesterified for the synthesis of triacylglycerols that are released into the circulation as part of very low density lipoproteins (VLDLs). In addition, glycerol may be used for glucose synthesis and FFA for ␤-oxidation to acetyl-CoA leading to energy production and ketone body synthesis; these pathways also increase markedly under fasting conditions in late pregnancy [12, 33]. Since insulin inhibits adipose tissue lipolytic activity, hepatic VLDL secretion, gluconeogenesis and ketogenesis, the insulin-resistant condition of late pregnancy contributes to the increased adipose tissue lipolysis and the increased hepatic VLDL production, gluconeogenesis and ketogenesis at late pregnancy under fasting conditions (fig. 1). Furthermore, since the underlying pathophysiology of GDM is a function of decreased maternal insulin sensitivity or increased insulin resistance, those pathways become further enhanced in this condition, explaining the increase in plasma FFA and ketone bodies consistently seen in diabetic pregnancy [34, 35]. Maternal Hyperlipidemia Enhanced maternal adipose tissue lipolytic activity during late gestation is associated with hyperlipidemia, mainly corresponding to increases in triacylglycerols, with smaller rises in phospholipids and cholesterol in the circulation. The greatest increase in plasma triacylglycerols corresponds to VLDL and results from enhanced production by the liver and decreased removal from the circulation as a consequence of reduced adipose tissue lipoprotein lipase (LPL) activity [36]. During normal pregnancy there is also an enrichment of triacylglycerols in other lipoprotein fractions that do not normally transport them, like low density lipoproteins (LDL) and high density lipoproteins (HDL) [36]. The abundance of VLDL triacylglycerols in the presence of an increase in cholesteryl ester transfer protein (CETP) activity which takes place at mid gestation contributes to this accumulation of triacylglycerols in LDL and HDL [37] (fig. 1). A further factor contributing to this effect is the decrease in hepatic lipase activity which

Metabolism in Diabetic Pregnancy

39

also occurs during late pregnancy [36], decreasing the conversion of buoyant HDL2b triglyceride-rich particles into small HDL3 triglyceride-poor particles, allowing a proportional accumulation of the former [36]. Both the insulin-resistant condition and the higher concentration of estrogen are thought to be responsible for the hypertriacylglycerolemia of pregnancy. As commented above, the insulin-resistant condition contributes both to the enhanced adipose tissue lipolytic activity which speeds up the transport of glycerol and FFA to the liver and their subsequent conversion into circulating VLDL triacylglycerols, and to the decreased LPL activity [38]. The increase in plasma estrogen concentrations during gestation also contributes to maternal hypertriacylglycerolemia since it enhances hepatic production of VLDL [39] and decreases the expression and activity of hepatic lipase [40]. Although exaggerated hypertriacylglycerolemia and lower LDL and HDL cholesterol have been found in diabetic pregnancy, there are reports which show no change in the maternal lipoprotein profile [34] or even decreased triacylglycerol levels [41]. Neither differences in the type of diabetes, degree of metabolic control or even the time of pregnancy studied explains this different response. As commented above, besides insulin resistance, hyperlipidemia occurring during gestation under normal conditions is driven by the increases in steroid hormones. Since plasma levels [34] as well as the level of sex hormone binding globulin [42] have been found decreased in diabetic women during early pregnancy or in those women in whom GDM subsequently developed, it is proposed that the degree of metabolic control and sex hormonal dysfunction may determine the development or lack of development of dyslipidemia in diabetic pregnant women, and this could explain the variety of reported findings. Placental Transfer of Lipid Metabolites Although triacylglycerols circulating in plasma lipoproteins do not directly cross the placental barrier [1], essential fatty acids from maternal diet, which are mainly transported as triacylglycerols in triacylglycerol-rich lipoproteins in maternal plasma [43], must be made available to the fetus. The presence of VLDL/apo E receptor, LDL receptor-related proteins and HDL receptors in placental trophoblast cells allow these lipoproteins to be taken up by the placenta. In addition, the trophoblasts also express different lipolytic activities including LPL, phospholipase A2 and an intracellular lipase. Thus, maternal triacylglycerols in plasma lipoproteins are either taken up intact through the placenta receptors or, after hydrolysis, their constituent fatty acids are taken up by the placenta, where the fatty acids are reesterified to synthesize glycerolipids to provide a reservoir of fatty acids. Subsequent intracellular hydrolysis of the glycerolipids releases fatty acids that diffuse to fetal plasma.

Herrera

40

Although smaller in proportion to lipoprotein triacylglycerols, maternal plasma FFA are also an important source of polyunsaturated fatty acids (PUFA) for the fetus. In human placenta there is a membrane fatty acid-binding protein (FABPpm) [44] which is responsible for the preferential uptake and transfer of certain PUFA: docosahexaenoic ⬎ ␣-linolenic ⬎ linoleic ⬎ oleic ⬎ arachidonic acid [45]. The selective uptake of certain fatty acids may also contribute to a degree of selective placental metabolism such as their conversion to prostaglandins and other eicosanoids, the incorporation of some fatty acids into membrane phospholipids, fatty acid oxidation and fatty acid synthesis. The combination of all these processes determines the actual rate of placental fatty acid transfer and its selectivity, resulting in the proportional enrichment of certain PUFA, such as arachidonic acid and docosahexaenoic acid, in the fetal compartment compared to the maternal compartment. Placental transfer of maternal cholesterol has been shown to be effective in different species, such as rat, guinea pig and rhesus monkey. Cholesterol synthesis in fetal tissues has also been shown to be highly active in some species. In humans, the comparison of maternal plasma concentrations of lipoprotein cholesterol and those in umbilical cord blood cholesterol gave positive correlations in some studies [46] but no correlation in others [47]. Gestational age seems to influence these comparisons, since in fetuses younger than 6 months, plasma cholesterol levels significantly correlate with the maternal ones [48], suggesting that, at these early stages of gestation, maternal cholesterol actively contributes to fetal cholesterol. At term, although there is delivery of cholesterol from placenta to the fetus, its contribution to the fetal plasma cholesterol pool is minor, and endogenous cholesterol synthesis appears to be the principal source of fetal cholesterol. Although intrinsic changes in placental capability to lipid transfer in diabetic women cannot be discarded, altered lipid profile on the maternal side affects the quantity and/or quality of lipids being transferred to the fetus. In fact, maternal dietary fatty acids influence fetal lipid metabolism and contribute to postnatal metabolic changes [2, 49]. GDM patients with macrosomic fetuses have been associated with high triglyacylglycerol, VLDL and HDL triacylglycerol levels [50]. Besides, macrosomic newborns of poorly controlled diabetic mothers have higher lipid and lipoprotein lipid levels than those found in controls [51], and levels of cholesterol, phospholipids and triacylglycerols in umbilical cord were found enhanced and correlated with FFA in fetuses of type 1 diabetes mellitus (DM) mothers [52]. The increased concentration of FFA in fetal blood of type 1 DM pregnancies is probably caused by increased delivery from maternal circulation, because an increased maternofetal gradient has been reported in diabetes [53]; this may drive the synthesis of cholesterol, triacylglycerols and phospholipids in the fetus of type 1 DM mothers. Moreover, the

Metabolism in Diabetic Pregnancy

41

placental transport of triacylglycerol fatty acids may be exaggerated in diabetes, where LPL declines in adipose tissue, contributing to maternal hypertriacylglycerolemia, but not in placenta [54]. During the perinatal period, maternal supplies of arachidonic acid and docosahexaenoic acid are the likely major sources of long-chain PUFAs to the fetus. Despite the same proportion of phospholipid arachidonic acid and docosahexaenoic acid in control and GDM women, fetal erythrocyte phospholipid arachidonic acid and docosahexaenoic acid are lower in women with GDM than in control subjects [55], suggesting an impairment in the maternal transfer of these fatty acids to the fetus.

Conclusion

The continuous supply of metabolites derived from maternal circulation, across the placenta, sustains fetal development. Under normal conditions, during the first two thirds of gestation, the mother develops hyperinsulinemia and normal or enhanced insulin sensitivity, contributing to an accumulation of fat stores. During the last third of gestation the mother switches to a catabolic condition, which is facilitated by her insulin-resistant condition and accelerates the availability of nutrients across the placenta. Poorly controlled diabetes during the first 7 weeks of human pregnancy has been associated with a spectrum of abnormalities in embryonic development, including embryo loss, induction of congenital anomalies and growth delay, maternal hyperglycemia being the principal teratogenic agent. However, poorly controlled diabetes during the latter two thirds of gestation accelerates fetal growth, and induces large-for-gestational age infants, respiratory distress syndrome and neonatal hypoglycemia. Studies on carbohydrate metabolism have shown that insulin suppression of hepatic glucose production is less decreased in GDM patients than in controls. An increase in basal leucine turnover and increased plasma branched amino acid levels were also found in GDM women at late pregnancy, suggesting a significant alteration in maternal protein and amino acid metabolism. Since enhanced adipose tissue lipolysis and liver production of VLDL triacylglycerol and decreased extrahepatic LPL activity during late pregnancy are caused by the insulin-resistant condition, a further decrease in insulin sensitivity in pregnant diabetic women would accelerate these changes and exaggerate the development of maternal hypertriacylglycerolemia. Since increased estrogen concentrations in late pregnant women actively contribute to the enhanced liver production of VLDL triacylglycerol, the decreased estrogen levels found in pregnant diabetic women may counteract such a change, avoiding the development of such exaggerated hypertriacylglycerolemia in certain diabetic subjects. These changes in maternal

Herrera

42

metabolism together with alterations in intrinsic placental function affect the quantity and quality of nutrients reaching the fetus, and consequently contribute to altered fetal growth.

Acknowledgements The author wish to thank Mr. Brian Crilly for his editorial help and Ms. Milagros Morante for her technical assistance.

References 1 2 3

4 5

6 7 8 9

10 11 12

13 14 15

16

Herrera E, Lasunción MA: Maternal-fetal transfer of lipid metabolites; in Polin RA, Fox WW, Abman SH (eds): Fetal and Neonatal Physiology, ed 3. Philadelphia, Saunders, 2004, pp 375–388. Herrera E: Implications of dietary fatty acids during pregnancy on placental, fetal and postnatal development – A review. Placenta 2002;23(suppl A):S9–S19. Napoli C, Witztum JL, Calara F, De Nigris F, Palinski W: Maternal hypercholesterolemia enhances atherogenesis in normocholesterolemic rabbits, which is inhibited by antioxidant or lipid-lowering intervention during pregnancy – An experimental model of atherogenic mechanisms in human fetuses. Circ Res 2000;87:946–952. Buch I, Hornnes PJ, Kuhl C: Glucose tolerance in early pregnancy. Acta Endocrinol (Copenh) 1986;112:263–266. Ramos P, Crespo-Solans MD, del Campo S, Herrera E: Fat accumulation in the rat during early pregnancy is modulated by enhanced insulin responsiveness. Am J Physiol Endocrinol Metab 2003;285:E318–E328. Cousins L: Insulin sensitivity in pregnancy. Diabetes 1991;40:39–43. Pedersen J: Weight and length at birth of infants of diabetic mothers. Acta Endocrinol 1954;16: 330–342. Freinkel N, Metzger BE: Pregnancy as a tissue culture experience: The critical implications of maternal metabolism for fetal development. Excerpta Med Int Congr Sec 1979;90:3–27. Catalano PM, Tyzbir ED, Wolfe RR, Roman NM, Amini SB, Sims EAH: Longitudinal changes in basal hepatic glucose production and suppression during insulin infusion in normal pregnant women. Am J Obstet Gynecol 1992;167:913–919. Butte NF: Carbohydrate and lipid metabolism in pregnancy: Normal compared with gestational diabetes mellitus. Am J Clin Nutr 2000;71(suppl):1256S. Kalhan S, Rossi K, Gruca L, Burkett E, O’Brien A: Glucose turnover and gluconeogenesis in human pregnancy. J Clin Invest 1997;100:1775–1781. Herrera E, Knopp RH, Freinkel N: Carbohydrate metabolism in pregnancy. VI. Plasma fuels, insulin, liver composition, gluconeogenesis and nitrogen metabolism during gestation in the fed and fasted rat. J Clin Invest 1969;48:2260–2272. Herrera E: Metabolic adaptations in pregnancy and their implications for the availability of substrates to the fetus. Eur J Clin Nutr 2000;54(suppl 1):S47–S51. Herrera E, Lasunción MA, Palacín M, Zorzano A, Bonet B: Intermediary metabolism in pregnancy. First theme of the Freinkel era. Diabetes 1991;40(suppl 2):83–88. Catalano PM, Kirwan JP, Haugel-de Mouzon S, King J: Gestational diabetes and insulin resistance: Role in short- and long-term implications for mother and fetus. J Nutr 2003;133:1674S– 1683S. Catalano PM, Huston L, Amini SB, Kalhan SC: Longitudinal changes in glucose metabolism during pregnancy in obese women with normal glucose tolerance and gestational diabetes. Am J Obstet Gynecol 1999;180:903–916.

Metabolism in Diabetic Pregnancy

43

17

18 19 20 21 22 23 24 25 26

27 28 29 30 31 32

33 34 35

36

37 38

39 40

Hsu HW, Butte NF, Wong WW, Moon JK, Ellis KJ, Klein PD, Moise KJ: Oxidative metabolism in insulin-treated gestational diabetes mellitus. Am J Physiol Endocrinol Metab 1997;272: E1099–E1107. Zimmer DM, Brechtel G, Golichowski AM, Baron AD, Karn CA, Denne SC: Glucose and amino acid turnover in untreated gestational diabetes. Diabetes Care 1996;19:591–596. Kalhan SC: Protein metabolism in pregnancy. Am J Clin Nutr 2000;71:1249S–1255S. Mojtahedi M, de Groot LCPGM, Boekholt HA, van Raaij JMA: Nitrogen balance of healthy Dutch women before and during pregnancy. Am J Clin Nutr 2002;75:1078–1083. Denne SC, Patel D, Kalhan SC: Leucine kinetics and fuel utilization during a brief fast in human pregnancy. Metabolism 1991;40:1249–1256. Kalhan SC, Denne SC, Patel DM, Nuamah IF, Savin SM: Leucine kinetics during a brief fast in diabetes in pregnancy. Metabolism 1994;43:378–384. Whittaker PG, Lee CH, Taylor R: Whole body protein kinetics in women: Effect of pregnancy and IDDM during anabolic stimulation. Am J Physiol 2000;279:E978–E988. Kalhan SC: Protein and nitrogen metabolism in gestational diabetes. Diabetes Care 1998;21(suppl 2): B75–B78. Dicke JM, Henderson GI: Placental amino acid uptake in normal and complicated pregnancies. Am J Med Sci 1988;295:223–227. Kuruvilla AG, D’Souza SW, Glazier JD, Mahendran D, Maresh MJ, Sibbley C: Altered activity of the system A amino acid transporter in microvillous membrane vesicles from placentas of macrosomic babies born to diabetic women. J Clin Invest 1994;94:689–695. Jansson T, Ekstrand Y, Björn C, Wennergren M, Powell TL: Alterations in the activity of placental amino acid transporters in pregnancies complicated by diabetes. Diabetes 2002;51:2214–2219. Villar J, Cogswell M, Kestler E, Castillo P, Menendez R, Repke JT: Effect of fat and fat-free mass deposition during pregnancy on birth weight. Am J Obstet Gynecol 1992;167:1344–1352. Lopez Luna P, Maier I, Herrera E: Carcass and tissue fat content in the pregnant rat. Biol Neonate 1991;60:29–38. Chaves JM, Herrera E: In vitro glycerol metabolism in adipose tissue from fasted pregnant rats. Biochem Biophys Res Commun 1978;85:1299–1306. Williams C, Coltart TM: Adipose tissue metabolism in pregnancy: The lipolytic effect of human placental lactogen. Br J Obstet Gynaecol 1978;85:43–46. Knopp RH, Herrera E, Freinkel N: Carbohydrate metabolism in pregnancy. VIII. Metabolism of adipose tissue isolated from fed and fasted pregnant rats during late gestation. J Clin Invest 1970;49:1438–1446. Zorzano A, Lasunción MA, Herrera E: Role of the availability of substrates on hepatic and renal gluconeogenesis in the fasted late pregnant rat. Metabolism 1986;35:297–303. Montelongo A, Lasunción MA, Pallardo LF, Herrera E: Longitudinal study of plasma lipoproteins and hormones during pregnancy in normal and diabetic women. Diabetes 1992;41:1651–1659. Jovanovic L, Metzger BE, Knopp RH, Conley MR, Park E, Lee YJ, Simpson JL, Holmes L, Aarons JH, Mills JL, et al: The diabetes in early pregnancy study – ␤-hydroxybutyrate levels in type 1 diabetic pregnancy compared with normal pregnancy. Diabetes Care 1998;21:1978–1984. Alvarez JJ, Montelongo A, Iglesias A, Lasunción MA, Herrera E: Longitudinal study on lipoprotein profile, high density lipoprotein subclass, and postheparin lipases during gestation in women. J Lipid Res 1996;37:299–308. Iglesias A, Montelongo A, Herrera E, Lasunción MA: Changes in cholesteryl ester transfer protein activity during normal gestation and postpartum. Clin Biochem 1994;27:63–68. Herrera E, Ramos P, Martín A: Control by insulin of adipose tissue lipoprotein lipase activity during late pregnancy in the rat; in Shafrir E (ed): Frontiers in Diabetes Research. Lessons from Animal Diabetes III. London, Smith-Gordon, 1990, pp 551–554. Knopp RH, Zhu X, Bonet B: Effects of estrogens on lipoprotein metabolism and cardiovascular disease in women. Atherosclerosis 1994;110(suppl):S83–S91. Brinton EA: Oral estrogen replacement therapy in postmenopausal women selectively raises levels and production rates of lipoprotein A-I and lowers hepatic lipase activity without lowering the fractional catabolic rate. Arterioscler Thromb Vasc Biol 1996;16:431–440.

Herrera

44

41

42

43 44

45 46

47 48

49

50

51

52

53 54

55

Peterson CM, Jovanovich-Peterson L, Mills JL, Conley MR, Knopp RH, Reed GF, Aarons JH, Holmes LB, Brown Z, Van Allen M, et al: The diabetes in early pregnancy study: Changes in cholesterol, triglycerides, body weight, and blood pressure. Am J Obstet Gynecol 1992;166:513–518. Thadhani R, Wolf M, Hsu-Blatman K, Sandler L, Nathan D, Ecker JL: First-trimester sex hormone binding globulin and subsequent gestational diabetes mellitus. Am J Obstet Gynecol 2003;189/1: 171–176. Herrera E: Lipid metabolism in pregnancy and its consequences in the fetus and newborn. Endocrine 2002;19:43–55. Campbell FM, Gordon MJ, Dutta-Roy AK: Plasma membrane fatty acid binding protein from human placenta: Identification and characterization. Biochem Biophys Res Commun 2000;209: 1011–1017. Haggarty P, Page K, Abramovich DR, Ashton J, Brown D: Long-chain polyunsaturated fatty acid transport across the perfused human placenta. Placenta 1997;18:635–642. Ortega RM, Gaspar MJ, Cantero M: Influence of maternal serum lipids and maternal diet during the third trimester of pregnancy on umbilical cord blood lipids in two populations of Spanish newborns. J Vitam Nutr Res 1996;66:250–257. Neary RH, Kilby MD, Kumpatula P, Game FL, Bhatnagar D, Durrington PN, O’Brien PMS: Fetal and maternal lipoprotein metabolism in human pregnancy. Clin Sci 1995;88:311–318. Napoli C, D’Armiento FP, Mancini FP, Postiglione A, Witztum JL, Palumbo G, Palinski W: Fatty streak formation occurs in human fetal aortas and is greatly enhanced by maternal hypercholesterolemia – intimal accumulation of low density lipoprotein and its oxidation precede monocyte recruitment into early atherosclerotic lesions. J Clin Invest 1997;100:2680–2690. Amusquivar E, Rupérez FJ, Barbas C, Herrera E: Low arachidonic acid rather than ␣-tocopherol is responsible for the delayed postnatal development in offspring of rats fed fish oil instead of olive oil during pregnancy and lactation. J Nutr 2000;130:2855–2865. Merzouk H, Bouchenak M, Loukidi B, Madani S, Prost J, Belleville J: Fetal macrosomia related to maternal poorly controlled type 1 diabetes strongly impairs serum lipoprotein concentrations and composition. J Clin Pathol 2000;53:917–923. Merzouk H, Madani S, Korso N, Bouchenak M, Prost J, Belleville J: Maternal and fetal serum lipid and lipoprotein concentrations and compositions in type 1 diabetic pregnancy: Relationship with maternal glycemic control. J Lab Clin Med 2000;136:441–448. Kilby MD, Neary RH, Mackness MI, Durrington PN: Fetal and maternal lipoprotein metabolism in human pregnancy complicated by type I diabetes mellitus. J Clin Endocrinol Metab 1998;83:1736–1741. Thomas CR: Placental transfer of non-esterified fatty acids in normal and diabetic pregnancy. Biol Neonate 1987;51:94–101. Knopp RH, Warth MR, Charles D, Childs M, Li JR, Mabuchi H, Van Allen MI: Lipoprotein metabolism in pregnancy, fat transport to the fetus and the effects of diabetes. Biol Neonate 1986;50:297–317. Wijendran V, Bendel R, Couch SC, Philipson EH, Cheruku S, Lammi-Keefe CJ: Maternal dietary intake, BMI and insulin resistance determine fetal DHA (22:6n3) status (abstract). FASEB J 1998;12:A970.

Emilio Herrera, MD Universidad San Pablo-CEU Ctra. Boadilla del Monte km 5.300, Boadilla del Monte ES–28668 Madrid (Spain) Tel. ⫹34 913724730, Fax ⫹34 913510496, E-Mail [email protected]

Metabolism in Diabetic Pregnancy

45

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 46–57

Leptin in the Diabetic Pregnancy Sylvie Hauguel-de Mouzona, Jacques Lepercqb, Patrick Catalanoa a

Department of Reproductive Biology, Obstetrics and Gynecology, MetroHealth Medical Center, Case Western Reserve University, Cleveland, Ohio, USA; bGynécologie-Obstétrique, Hôpital Cochin-Saint Vincent de Paul, Paris, France

Prior to its characterization as a protein secreted by the adipocyte [1], leptin was discovered as a circulating factor exhibiting satiety and antiobesity properties [2, 3]. Circulating leptin levels are proportional to adipose tissue mass and respond to changes in energy balance. Leptin signals elicited in fat are conveyed to the hypothalamus to maintain appropriate body energy stores by adjusting food intake and metabolism [4]. These specific interactions between adipose signals and brain targets gave rise to the dogma that the main function of leptin is to act as an adipostat. It is now recognized that expression of leptin is much more widespread than originally thought with direct evidence showing that the gastric fundus, skeletal muscle, bone, placenta and several fetal tissues synthesize leptin. Hence, the initial view of leptin-mediated actions at a central level has been expanded to a much more complex level which integrates a number of so-called peripheral effects. Leptin acts as a nutritional signal to the reproductive axis at the time of sexual maturation, stimulates angiogenesis and hematopoiesis, and regulates the extent of bone formation [5, 6]. Leptin expression and action are altered in a number of pathologies associated with metabolic disorders and insulin resistance such as obesity and diabetes [7]. A role for leptin in development and particularly in pregnancy has been proposed. This chapter addresses the known and unknowns of leptin in pregnancy and diabetes 10 years after the discovery of the hormone. Focus will be given to the functions and other potential roles of leptin within the fetoplacental unit.

Maternal plasma leptin (ng/ml)

80

60

40

20

0

Pregravid

1st trimester

3rd trimester

48 h postpartum

Fig. 1. Pregnancy-associated changes in maternal leptin concentrations. Leptin was measured in blood samples obtained from a maternal vein before pregnancy, at 12 weeks, at the time of delivery (38–41 weeks of gestation), then again 48 h postpartum [adapted from 8]. 6 subjects were analyzed longitudinally during pregnancy and 21 subjects postpartum.

The Origin of Circulating Leptin

Major modifications of circulating leptin levels occur during pregnancy [8, 9]. Leptin concentration increases significantly in early pregnancy, being 30% higher at 12 weeks compared to the pregravid state and resumes immediately after delivery (fig. 1). Differences in the gestation-induced leptin rise are noted among species. In the mouse, serum leptin increases 25-fold starting at mid gestation and the levels are sustained until term [10]. In the rat, there is a 2- to 3-fold rise at the end of the first week of gestation [11]. In addition to white adipose tissue, the main producer of leptin in nongravid individuals, the placenta becomes a primary source for leptin synthesis. The human placenta expresses high amounts of leptin mRNA and protein in early, mid and late gestation [9, 12]. Human placental leptin is identical to leptin of adipose origin on the basis of its size of 16,000 Da, charge and immunoreactivity [13]. Leptin is mainly localized within the syncytiotrophoblast cells which are in direct contact with maternal blood, and on fetal vascular endothelial cells, in contact with the fetal blood [14]. The origin of the pregnancy-induced rise in leptin is not known with certainty. There is strong evidence suggesting that the placenta rather than maternal adipose tissue makes a substantial contribution to the rise in maternal leptin concentrations. First, the leptin increase precedes the physiological increase in maternal BMI [8]. Second, circulating leptin rapidly declines after delivery in both mother and neonate (fig. 1). Cross-sectional comparisons of leptin mRNA concentration

Leptin in the Diabetic Pregnancy

47

in white adipose tissue from pregnant and nonpregnant women also suggest that adipose tissue leptin makes little contribution to the plasma rise because the mRNA content does not increase in early pregnancy [15]. Leptin synthesized within the placenta is released into both maternal and fetal circulation [16]. Ninety-five percent of the total leptin synthesized is released into the maternal side, and 5% into the fetal circulation. Although it may look modest compared to the proportion delivered to the mother, the rate of leptin delivery on the fetal side is much higher than the rate for other placentalderived hormones such as HPL and HCG [17]. Thus, it is conceivable that leptin from vascular endothelial cells is used as a complementing source of circulating leptin for the fetus.

The Biological Functions of Leptin

Function of Leptin in the Mother Ideally, the pregnant women should maintain a positive energy balance to sustain the energy requirements of fetal development and thus is not likely to reduce her food intake. From a pragmatic point of view, it has to be expected that the rise in maternal leptin is not accompanied by its classic central effect in regulating food intake. However, the increase above pregravid plasma values suggests that leptin may be needed for role(s) different from the hypothalamic regulation of appetite suppression. In contrast to weight loss achieved through starvation, weight loss with leptin administration is restricted to adipose tissue and does not involve lean mass [18]. In the second part of pregnancy, accretion of fat is the main caloric demand of the fetus and has to be supported by adequate nutrient supply. Therefore, one possible function of the increased maternal leptin may be to enhance the mobilization of maternal fat stores in order to increase availability and to support transplacental transfer of lipid substrates. It is anticipated that leptin activates lipid oxidation by inducing expression of enzymes of lipid oxidation as shown in nongravid humans and mice [19–21]. In addition, higher leptin levels reflect a state of leptin resistance analogous to obese individuals whose elevated leptin levels do not appear to successfully recruit their biological targets [22]. Another explanation for the higher leptin levels is the increase in the bound over free ratio in circulating leptin [23]. Leptin can bind to an extracellular (soluble) leptin receptor released in the circulation by placental membrane shedding [24]. The binding to this soluble isoform may delay clearance of leptin from the circulation, resulting in a peak in maternal plasma levels. Leptin binding to soluble receptors may also contribute to leptin resistance by slowing down intracellular leptin signal transduction through transmembrane leptin receptors either within the hypothalamus or peripheral tissues.

Hauguel-de Mouzon/Lepercq/Catalano

48

Function of Leptin in the Placenta The presence of leptin receptors within the placenta indicates that leptin can elicit a biological action locally either at the site or adjacent to the site of synthesis. Leptin receptors are primarily located on syncytiotrophoblast cells, which makes them accessible to maternal rather than to fetal circulating leptin [14]. Leptin of either placental or maternal origin can bind to placental leptin receptors, thus making it difficult to distinguish between a local paracrine action of placental-derived leptin versus an endocrine action of maternal leptin. Nevertheless, once bound to placental receptors, leptin triggers local or so-called peripheral effects as opposed to its central effects in the hypothalamus. The pleiotropic effects of leptin-modulating metabolic, immune, and angiogenic responses have been extensively characterized and some have been studied in the placenta. Leptin induces HCG production by trophoblast cells [25], stimulates amino acid uptake, synthesis of extracellular matrix proteins and metalloproteinases [26, 27] and enhances mitogenesis [28]. These effects are in keeping with a role of leptin to regulate placental growth, potentially leading to placental hypertrophy under conditions of leptin overproduction. Leptin also stimulates angiogenesis in primary cultures of human endothelial cells [29]. The human placenta is well equipped with the enzymatic machinery for transport, utilization and storage of lipids [30]. Leptin is able to modify lipid partitioning in muscle cells by favoring oxidation or to facilitate lipolysis and glycerol release as shown in adipocytes [31, 32]. These observations support the possibility that leptin provides a link between the immune and the endocrine system responses of the placental cell. Function of Leptin in the Fetus There is no correlation between maternal leptin levels and fetal weight or fat mass at the end of pregnancy, which is an indirect indication that maternal leptin does not cross the human placenta in significant amounts [33]. By contrast, umbilical leptin levels exhibit a very strong positive correlation with fat mass or ponderal index of the neonates and one that is not as good with birth weight [34]. The significant correlation between fetal circulating leptin and fetal fat mass cannot be taken as evidence that leptin regulates fetal growth. In contrast, it clearly indicates that fetal leptin levels are a good index of fetal fat mass. In other words leptin is a marker of adiposity in the fetus just like it is a marker of obesity after birth and throughout adulthood. This is further supported by the observation that fetal adipose tissue is able to synthesize leptin and that the leptin content of fetal adipose tissue is proportional to the neonatal ponderal index [35]. In contrast with postnatal life, in utero feeding requires a continuous nutrient supply which sometimes escapes regulation so that the fetus can get exposed to either too much or too little nutrients. Although some

Leptin in the Diabetic Pregnancy

49

Maternal plasma leptin (ng/ml)

80

Controls GDM

60 Type 1 40

20

0 15

25

35

45

50

Maternal BMI (kg/cm2)

Fig. 2. Relationship between plasma leptin and BMI in normal and diabetic pregnancy. 86 controls, 23 type 1 diabetics and 64 GDM subjects were included in the analysis. Leptin was measured in blood samples obtained from maternal vein at the time of delivery (38–41 weeks of gestation). r2 values were 0.43 for controls, 0.22 for GDM and 0.34 for type 1 pregnancy.

hypothalamic functions are mature in utero, there is no evidence that the hypothalamic targets of leptin receive metabolic signals to function prenatally in regulating food intake in the human fetus. To date, the only indication that leptin may have a biological function in the fetus is based on data obtained in mice and rats showing that leptin receptor mRNA is expressed in several fetal tissues including cartilage, bone, lung, kidney, testes [36] and in the hypothalamus [37]. This suggests that fetal leptin could serve endocrine or local functions. One example would be a role in the stimulation of fetal vasculogenesis, erythropoiesis or lymphopoiesis [38].

Maternal Leptin in Diabetes

Maternal leptin concentrations are essentially unchanged in diabetic pregnancies. When corrected for BMI, maternal plasma leptin concentrations are not different in type 1 and gestational diabetes (GDM) as compared with women with normal glucose tolerance. Interestingly, leptin levels of diabetic mothers are well correlated with their BMI, thus highlighting that circulating leptin levels are essentially reflecting fat mass as in nondiabetic pregnant women (fig. 2). By contrast, diabetic pregnancy is associated with dramatic changes in fetal and placental leptin levels. Placental leptin mRNA levels and protein content are higher in placenta from type 1 diabetic women as well as in placenta of

Hauguel-de Mouzon/Lepercq/Catalano

50

0.9

20

0.6

10

0.3

0 Maternal

Umbilical

Placental

0

Leptin (mg/␮g protein)

Plasma leptin (ng/ml)

30

Fig. 3. Diabetes-induced modifications of leptin concentrations in plasma and placenta. Plasma leptin was measured in maternal and umbilical cord at the time of delivery (left axis) and placental leptin content (right axis) assayed on whole placental homogenates was normalized for total protein content. ⵧ ⫽ 12 controls; 䊏, 䊏 ⫽ 7 GDM.

GDM women as compared to placenta of control pregnancies [39]. Umbilical leptin levels are also higher in pregnancy associated with diabetes in association with fetal macrosomia [40, 41]. Taken together, these data indicate that diabetes-induced changes in leptin are confined to the fetoplacental unit (fig. 3).

Placental Leptin in Diabetes

Regulation of Leptin Synthesis While there is literature available regarding the regulation of the leptin gene in adipose tissue, the data regarding regulation of leptin in placenta are meager by comparison. However, there is some evidence that regulation of placental leptin production occurs at the gene level. Placenta and white adipose tissue express the same gene except for the presence of a specific upstream enhancer (PLE) suggesting that leptin expression is regulated in a tissuespecific manner [42, 43]. Insulin is a major regulator of the adipose gene [44] but it is still debated whether the effects are transcriptional [45,46]. Increased leptin mRNA levels in diabetes and other situations associated with hyperinsulinemia and insulin resistance such as preeclampsia [39, 47] suggest that insulin acts as an activator of the placental gene. The adipose leptin gene is also under tight metabolic control. There is evidence for a nutrient-sensing pathway that involves a positive regulation by glucose and glucosamine and a negative regulation through FFA [48]. Leptin is also able to mediate autocrine regulation of its own expression in adipose tissue [32]. These are regulations that may occur in the placenta as a result of the ambient hyperglycemia, hyperlipidemia and hyperleptinemia associated with diabetic pregnancy [49]. Most of the regulations cited above involve transcriptional mechanisms and some of the transactivating factors have been described. Insulin effects involve

Leptin in the Diabetic Pregnancy

51

recruitment of C/EBPa through cAMP-dependent pathways [50]. Insulin and FFA recruit SREBP1c/ADD1 a key transcription factor linking changes in nutritional status and insulin levels to leptin expression [51]. Glucosamine increases the activity of the NF-␬B enhancer [52]. No data related to these transcriptional regulations are yet available for the placental leptin gene. Placental Leptin and Inflammation A comparison of the placental transcriptome in normal and diabetic pregnancy indicates that increased leptin synthesis is associated with a higher production of other cytokines, IL-6, IL-1, TNF-␣ in placentas of women with GDM [53]. The sum of these modifications creates a chronic inflammatory milieu that may further enhance leptin production as shown with the TNF-␣ system in adipocytes [54, 55]. In turn, the inflammatory responses participate to stimulate the expression of a subset of genes responsible for structural and vascular insults. These data provide additional support for convergence in the action of leptin and inflammatory cytokines [56–58]. Molecular Mechanisms for Placental Leptin Action Similarly to most cytokines, leptin transmits its biological action through binding to specific membrane receptors and activation of intracellular effectors [59, 60]. Four transmembrane and one soluble receptors lacking the region for attachment to the cell surface are generated by alternate splicing of a primary transcript [61]. All receptors contain a common extracellular domain analogous to that of class I cytokine receptors. They differ by the length and structure of their intracellular region responsible for distinct signaling capacities. None of these receptors have intrinsic kinase activity and they must activate janus kinase proteins (JAK) that initiate a phosphorylation cascade. Only the long receptor isoform bears a structural motif allowing JAK activation [62]. Both short and long leptin receptor isoforms are expressed in the placenta [14]. In placental BeWo cells, leptin does not stimulate the JAK-STAT pathways, which relay the classic activation of the long signaling isoform. In contrast, leptin stimulates MAPK phosphorylation and this is a requirement for stimulation of DNA synthesis through activation of the short isoform [28, 63]. Thus, only the short receptor isoform appears to be functional in placental cells. The MAPK phosphorylation cascade is also induced by insulin stimulation of placental cells, suggesting cross talks between the placental insulin and leptin receptors [64, 65]. In type 1 diabetes and GDM, the high placental leptin together with an hyperinsulinemic milieu is likely to enhance MAPK pathways. The convergence of mitotic activity and growth signals are potential mechanisms for diabetesinduced placental macrosomia. Besides its ability to activate the transmembrane receptors, placental leptin is able to bind the soluble receptor which does not

Hauguel-de Mouzon/Lepercq/Catalano

52

transduce signal. In contrast to the amount of transmembrane receptors which does not change, the concentration of soluble receptors is increased in type 1 diabetes and may participate to induce leptin resistance in the mother or to blunt leptin action in the placenta [14, 66].

Fetal Leptin in Diabetes

Origin of Fetal Hyperleptinemia It has been long recognized that fetuses grown in a diabetic environment are more likely to have an increased fat mass at birth [67–70]. Neonates of mothers with pre-GDM and GDM also have a higher cord leptin concentration than the control neonates [34, 39, 50, 70]. Interestingly, when cord leptin levels are adjusted for neonatal fat mass, the differences are no longer significant. This observation was interpreted to mean that the main reason why the infants of diabetic mothers have greater leptin levels is their increased adiposity. It corroborates the observation that fetal adipose tissue synthesizes leptin and that the leptin content is higher in adipose tissue of fetuses from diabetic mothers [16]. Role of Fetal Hyperleptinemia Based on the increased levels of leptin in neonates of diabetic women as well as on the significant positive correlations between leptin concentrations and fetal size, a recurrent assumption in the literature has been that ‘leptin stimulates fetal growth’. Factors controlling fetal fat accretion are not known but an increasing body of evidence has linked defective leptin action to the regulation of adiposity in the adult diabetic individual. Whether this could be extended to the situation in utero is a matter of speculation. A role for leptin in inhibiting insulin secretion is unlikely because the hyperleptinemic diabetic fetuses are also hyperinsulinemic [71]. However, fetal hyperleptinemia may promote increment of adipose tissue through enhancement of angiogenesis and the development of microvasculature in adipose tissue, a process which appears to precede the onset of adipogenesis in adults [56].

Consequences of High Leptin in the Fetoplacental Unit

The initial assumption that leptin acts as a growth factor for the fetus has been challenged by recent data. The fetoplacental unit of a diabetic pregnancy clearly develops in a high leptin milieu as a result of increased production by the placenta and fetal adipose tissue. Whether high leptin is beneficial or detrimental in utero remains to be determined. Leptin appears to be a central player

Leptin in the Diabetic Pregnancy

53

in the inflammatory response elicited by diabetes in the placenta. The placental inflammatory response is associated with major alterations of lipid storage and transport by the placenta that may provide a link between maternal diabetes and fetal obesity [53]. It is becoming evident that high fetal leptin results in the enlarged fat depot and that leptin is a marker of prenatal as well as postnatal obesity. Hyperleptinemia may be instrumental in inducing leptin resistance by chronic activation of leptin receptors in several fetal tissues. Hence, hyperleptinemia of diabetic fetuses may contribute to adverse in utero programming as it relates to the metabolic imbalance developing later in adulthood [72, 73].

Conclusion

The rise of leptin in maternal plasma which occurs during early pregnancy can be accounted for by the delivery of leptin synthesized within the placenta. Fetal adipose tissue also produces leptin and there is good evidence that leptin is a marker of prenatal as well as postnatal adiposity. However, there is no evidence that leptin directly regulates fetal growth. Diabetes does not modify the linear relationship between plasma leptin and maternal BMI. By contrast, diabetes exacerbates the production of leptin in the fetoplacental unit and generates an hyperleptinemic in utero environment. High placental leptin levels are part of the chronic inflammatory milieu associated with diabetes in pregnancy. References 1 2 3 4 5 6 7 8

9

Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman J: Positional cloning of the mouse ob gene and its human homologue. Nature 1994;372:425–432. Harvey G: The effects of lesions in the hypothalamus in the parabiotic rats. J Physiol 1958;145: 336–352. Coleman D: Effects of parabiosis of obese with diabetes and normal mice. Diabetologia 1973;9: 294–298. Friedman J, Halaas J: Leptin and the regulation of body weight in mammals. Nature 1998;395: 763–770. Ahima R, Flier J: Leptin. Annu Rev Physiol 2000;62:413–437. Harris RB: Leptin, much more than a satiety signal. Annu Rev Nutr 2000;20:45–75. Dagogo S: Regulation and possible significance of leptin in humans: Leptin in health and disease. Diabetes Rev 1999;7:23–38. Highman T, Friedman J, Huston L, Wong W, Catalano P: Longitudinal changes in maternal leptin serum concentrations, body composition and resting metabolic rate in pregnancy. Am J Obstet Gynecol 1998;178:1010–1015. Masuzaki H, Ogawa Y, Sagawa N, Hosoda K, Matsumoto T, Mise H, Nishimura H, Yoshimasa Y, Tanaka I, Mori T, Nakao K: Nonadipose tissue production of leptin: Leptin as a novel placentaderived hormone in humans. Nat Med 1997;3:1029–1113.

Hauguel-de Mouzon/Lepercq/Catalano

54

10

11

12 13 14 15

16

17 18

19

20 21

22 23 24

25

26

27

28 29 30 31

Tomimatsu T, Yamaguchi M, Murakami T, Ogura K, Sakata M, Mitsuda N, Kanzaki T, Kurachi H, Irahara M, Miyake A: Increase of mouse leptin production by adipose tissue after midpregnancy gestational profile of serum leptin concentration. Biochem Biophys Res Commun 1997;240:213–215. Chien EK, Hara M, Rouard M, Yano H, Phillippe M, Polonsky KS, Bell GI: Increase in serum leptin and uterine leptin receptor mRNA during pregnancy in rats. Biochem Biophys Res Commun 1997;237:476–480. Henson M, Swan K, O’Neil J: Expression of placental leptin and leptin receptor transcripts in early pregnancy and at term. Obstet Gynecol 1998;92:1020–1028. Senaris R, Garcia-Caballero T, Casabiell X, Gallego R, Castro R, Considine R, Dieguez C, Casanueva F: Synthesis of leptin in human placenta. Endocrinology 1997;138:4501–4504. Challier J, Galtier M, Bintein T, Cortez A, Lepercq J, Hauguel-de Mouzon S: Placental leptin receptor isoforms in normal and pathological pregnancies. Placenta 2003;24:92–99. Catalano P, Nizielski S, Shao J, Presley L, Qiao L, Friedman J: Down regulation of IRS-1 and PPARg in obese women with gestational diabetes: Relationship to free fatty acids during pregnancy. Am J Physiol 2002;282:E522–E533. Lepercq J, Challier J, Guerre-Millo M, Cauzac M, Vidal H, Hauguel-de Mouzon S: Pre-natal leptin production: Evidence that human fetal adipose tissue produces leptin. J Clin Endocrinol Metab 2001;86:2409–2413. Linnemann K, Malek A, Sager R, Blum W, Schneider H, Fusch C: Leptin production and release in the dually in vitro perfused human placenta. J Clin Endocrinol Metab 2000;85:4298–4301. Halaas J, Boozer C, Blair-West J, Fidahusein N, Denton D, Friedman J: Physiological response to long-term peripheral and central leptin infusion in lean and obese mice. Proc Natl Acad Sci USA 1997;94:8878–8883. Shimabukuro M, Koyama K, Chen G, Wang M, Trieu F, Lee Y, Newgard C, Unger R: Direct antidiabetic effect of leptin through triglyceride depletion of tissues. Proc Natl Acad Sci USA 1997;94:4637–4641. Soukas A, Cohen P, Socci N, Friedman J: Leptin-specific patterns of gene expression in white adipose tissue. Gene Dev 2000;14:963–980. Okereke N, Hutson-Presley L, Amini S, Kalhan S, Catalano P: Longitudinal changes in energy expenditure and body composition in obese women with normal and impaired glucose tolerance. Am J Physiol Endocrinol Metab 2004;287:E472–E479. Considine R, Sinha M, Heiman M: Serum immunoreactive leptin concentrations in normal-weight and obese humans. N Engl J Med 1996;334:292–295. Teppa R, Ness R, Crombleholme W, Roberts J: Free leptin is increased in normal pregnancy and further increased in preeclampsia. Metabolism 2000;49:1043–1048. Gavrilova O, Barr V, Marcus-Samuels B, Reitman M: Hyperleptinemia of pregnancy associated with the appearance of a circulating form of the leptin receptor. J Biol Chem 1997;272: 30546–30551. Chardonnens D, Cameo P, Aubert M, Pralong F, Islami D, Campana A, Gaillard R, Bischof P: Modulation of human cytotrophoblastic leptin secretion by interleukin-1alpha and 17beta-oestradiol and its effect on HCG secretion. Mol Hum Reprod 1999;5/11:1077–1082. Castellucci M, de Matteis R, Meisser A, Cancello R, Monsurro V, Islami D, Sarzani R, Marzioni D, Cinti S, Bishof P: Leptin modulates extracellular matrix molecules and metalloproteinases: Possible implications for trophoblast invasion. Mol Hum Reprod 2000;6:951–958. Jansson N, Greenwood S, Johansson B, Powell T, Jansson T: Leptin stimulates the activity of the system A amino acid transporter in human placental villous fragments. J Clin Endocrinol Metab 2003;88:1205–1211. Cauzac M, Czuba D, Girard J, Hauguel-de Mouzon S: Transduction of leptin growth signals in placental cells is independent of JAK-STAT activation. Placenta 2003;24:378–384. Bouloumie A, Drexler H, Lafontan M, Busse R: Leptin, the product of Ob gene, promotes angiogenesis. Circ Res 1998;83:1059–1066. Desoye G, Shafrir E: Placental metabolism and its regulation in health and diabetes. Mol Aspects Med 1994;15:505–682. Unger R: Minireview: Weapons of lean body mass destruction/the role of ectopic lipids in the metabolic syndrome. Endocrinology 2003;144:5159–5165.

Leptin in the Diabetic Pregnancy

55

32 33 34 35 36

37

38 39

40

41 42 43

44 45 46

47 48 49 50

51

52 53 54

Wang M-Y, Lee Y, Unger R: Novel form of lipolysis induced by leptin. J Biol Chem 1999;274: 17541–17544. Radaelli T, Uvena-Celebrezze J, Thomas A, Hauguel-de Mouzon S, Catalano P: Inflammatory cytokines are mediators of fetal growth and adiposity. Diabetes 2004;55(suppl 2):pA18. Lepercq J, Lahlou N, Timsit J, Girard J, Hauguel-de Mouzon S: Macrosomia revisited: Ponderal index and leptin delineate subtypes of fetal overgrowth. Am J Obstet Gynecol 1999;181:621–625. Okereke N, Uvena-Cellebrezze J, Hutson-Presley L, Amini S, Catalano P: The effect of gender and gestational diabetes mellitus on cord leptin concentration. Am J Obstet Gynecol 2002;187:798–803. Hoggard N, Hunter L, Duncan J, Williams L, Trayhurn P, Mercer J: Leptin and leptin receptor mRNA and protein expression in the murine fetus and placenta. Proc Natl Acad Sci USA 1997;94:11073–11078. Matsuda J, Yokota I, Tsuruo Y, Murakami T, Ishimura K, Shima K, Kuroda Y: Development changes in long-form leptin receptor expression and localization in rat brain. Endocrinology 1999;140:5233–5238. Bennett B, Solar P, Yuan J, Mathias J, Thomas G, Matthews W: A role for leptin and its cognate receptor in hematopoiesis. Curr Biol 1996;6:11701–11708. Lepercq J, Caüzac M, Lahlou J, Timsit J, Girard J, Auwerx J, Hauguel-de Mouzon S: Overexpression of placental leptin in diabetic pregnancies: A critical role for insulin. Diabetes 1998;47:847–850. Maffei M, Volpe L, Di Cianni G, Bertacca A, Ferdeghini M, Murru S, Teti G, Casadidio I, Cecchetti P, Navalesi R, Benzi L: Plasma leptin levels in newborns from normal and diabetic mothers. Horm Metab Res 1998;9:575–580. Persson B, Westgren M, Celsi G, Nord E, Ortqvist E: Leptin concentrations in cord blood in normal newborn infants and offspring of diabetic mothers. Horm Metab Res 1999;8:467–471. Bi C, Gavrilova O, Gong D-W, Mason M, Reitman M: Identification of a placental enhancer for the human leptin gene. J Biol Chem 1997;272:30583–30588. Grosfeld A, André J, Hauguel-de Mouzon S, Berra E, Pouyssegur J, Guerre-Millo M: Hypoxiainducible factor 1 (HIF-1) transactivates the human leptin gene promoter. J Biol Chem 2002;277: 42953–42957. Saladin R, De Vos P, Guerre-Millo M, leturque A, Girard J, Staels B, Auwerx J: Transient increase in obese gene expression after food intake or insulin administration. Nature 1995;377:527–529. Leroy P, Dessolin S, Villageois P, Moon B, Friedman J, Ailhaud G, Dani C: Expression of ob gene in adipose cells. Regulation by insulin. J Biol Chem 1996;271:2365–2368. Vidal H, Auboeuf D, De Vos P, Staels B, Riou JP, Auwerx J, Laville M: The expression of ob gene is not acutely regulated by insulin and fasting in human abdominal subcutaneous adipose tissue. J Clin Invest 1996;98:251–255. Lepercq J, Guerre-Millo M, Andre J, Cauzac M, Hauguel-de Mouzon S: Leptin: A potential marker of placental insufficiency. Obstet Gynecol 2003;55:151–155. Wang J, Liu R, Hawkins M, Barzilai N, Rossetti L: A nutrient-sensing pathway regulates leptin gene expression in muscle and fat. Nature 1998;393:684–688. Knopp RH, Montes A, Childs M, Li JR, Mabuchi H: Metabolic adjustments in normal and diabetic pregnancy. Clin Obstet Gynecol 1981;24:21–49. Miller SG, De Vos P, Guerre-Millo M, Wong K, Hermann T, Staels B, Briggs MR, Auwerx J: The adipocyte specific transcription factor C/EBPalpha modulates human ob gene expression. Proc Natl Acad Sci USA 1996;93:5507–5511. Kim J, Sarraf P, Wright M, Yao K, Mueller E, Solanes G, Lowell B, Spiegelman B: Nutritional and insulin regulation of fatty acid synthetase and leptin gene expression through ADD1/SREBP1. J Clin Invest 1998;101:1–9. James L, Tang D, Ingram A, Ly H, Thai K, Cai L, Scholey J: Flux through the glucosamine pathway is a determinant of nuclear factor ␬B-dependent promoter activation. Diabetes 2002;51:1146–1156. Radaelli T, Varastehpour A, Catalano P, Hauguel-de Mouzon S: Gestational diabetes induces placental genes for chronic stress and inflammatory pathways. Diabetes 2003;52:2951–2958. Mantzoros C, Moschos S, Avramopoulos I, Kaklamani V, Liolios A, Doulgerakis D, Griveas I, Katsilambros N, Flier J: Leptin concentrations in relation to body mass index and the tumor necrosis factor-alpha system in humans. J Clin Endocrinol Metab 1997;82:3408–3413.

Hauguel-de Mouzon/Lepercq/Catalano

56

55

56

57 58 59

60

61 62 63

64

65 66

67 68 69

70

71 72 73

Kirchgessner T, Uysal K, Wiesbrock S, Marino M, Hotamisligil G: Tumor necrosis factor-alpha contributes to obesity-related hyperleptinemia by regulating leptin release from adipocytes. J Clin Invest 1997;100:2777–2782. Sierra-Honigmann M, Nath A, Murakami C, Garcia-Cardena G, Papapetropoulos A, Sessa W, Madge L, Schechner J, Schwabb M, Polverini P, Flores-Riveros J: Biological action of leptin as an angiogenic factor. Science 1998;11:1683–1686. Guerre-Millo M: Adipose tissue and adipokines: For better or worse. Diabetes Metab 2004;30: 13–19. Bouloumie A, Marumo T, Lafontan M, Busse R: Leptin induces oxidative stress in human endothelial cells. FASEB J 1999;13:1231–1238. Tartaglia L, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards G, Campfield L, Clark F, Deeds J, et al: Identification and expression cloning of a leptin receptor, OB-R. Cell 1995;83: 1263–1271. Cioffi AJ, Shafer AW, Zupancic TJ, Smith-Gbur J, Mikhail A, Platika D, Snodgrass HR: Novel B219/OB receptor isoforms: Possible role of leptin in hematopoiesis and reproduction. Nat Med 1996;2:585–589. Barr V, lane K, Taylor S: Subcellular localization and internalization of the four human leptin receptor isoforms. J Biol Chem 2000;274:21416–21424. Bjorbaek C, Lavery H, Bates SH, Olson RK, Davis SM, Flier JS, Myers MG: SOCS3 mediates feedback inhibition of the leptin receptor via Tyr 985. J Biol Chem 2000;275:40649–40657. Bifulco G, Trencia A, Caruso M, Tommaselli G, Miele C, di Carlo C, Beguinot F, Nappi C: Leptin induces mitogenic effect on human choriocarcinoma cell line (JAr) via MAP kinase activation in a glucose-dependent fashion. Placenta 2003;4:385–391. Boileau P, Cauzac C, Pereira M, Girard J, Hauguel-de Mouzon S: Dissociation between insulinmediated signaling pathways and biological effects in placental cells: Role of PKB and MAPKinase phosphorylation. Endocrinology 2001;142:1–6. Szanto I, Kahn C: Selective interaction between leptin and insulin signaling pathways in a hepatic cell line. Proc Natl Acad Sci USA 2000;97:2355–2360. Lewandowski K, Horn R, O’Callaghan C, Dunlop D, Medley G, O’Hare Brabant G: Free leptin, bound leptin, and soluble leptin receptor in normal and diabetic pregnancies. J Clin Endocrinol Metab 1999;84:300–306. Freinkel N: The Banting lecture 1980. Of pregnancy and progeny. Diabetes 1980;29:1023–1035. Langer O, Mazze R: The relationship between large for gestational age infants and glycemic control in women with GDM. Am Obstet Gynecol 1988;159:1478–1483. Durnwald C, Huston-Presley L, Amini S, Catalano P: Evaluation of body composition of large-forgestational-age infants of women with gestational diabetes mellitus compared with women with normal glucose tolerance levels. Am J Obstet Gynecol 2004;191:804–808. Manderson J, Patterson C, Hadden D, Traub A, Leslie H, McCance D: Leptin concentrations in maternal serum and cord blood in diabetic and nondiabetic pregnancy. Am J Obstet Gynecol 2003;188:1326–1332. Seufert J, Kieffer T, Habener J: Leptin inhibits insulin gene transcription and reverses hyperinsulinemia in leptin-deficient ob/ob mice. Proc Natl Acad Sci USA 1999;96:674–679. Pettitt D, Aleck K, Baird H, Carraher M, Bennett P, Knowler W: Congenital susceptibility to NIDDM. Role of intrauterine environment. Diabetes 1988;37:622–628. Silverman BL, Metzger BE, Cho NH, Loeb CA: Impaired glucose tolerance in adolescent offspring of diabetic mothers. Relationship to fetal hyperinsulinism. Diabetes Care 1995;18:611–617.

Sylvie Hauguel-de Mouzon, PhD Department of Reproductive Biology, MetroHealth Medical Center 2550 MetroHealth Drive Cleveland, Ohio 44109 (USA) Tel. ⫹1 216 778 31 48, Fax ⫹1 216 778 15 74, E-Mail [email protected]

Leptin in the Diabetic Pregnancy

57

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 58–71

Maternal Diabetes and Embryonic Development R. De Hertogh † Physiology of Human Reproduction Research Unit and Endocrinology and Diabetology Department, University of Louvain-La Neuve, School of Medicine, Brussels, Belgium

Maternal diabetes constitutes an unfavorable environment for fetal development throughout pregnancy, resulting in multiple complications, from congenital malformations to neonatal morbidity and increased perinatal mortality. Congenital malformations obviously result from an impact of the maternal diabetic state on embryo development, occurring not later than the time at which such malformations may result from disturbed organogenesis, i.e. between 3 and 7 weeks of gestational age in humans [1]. The role of the disturbed maternal environment in the induction of diabetic congenital malformations was evidenced by the preventive effect of an early normalization of the diabetic state in the mother, even before the onset of pregnancy [2]. A later or incomplete normalization could explain the persistence of an increased occurrence of malformations in some studies [3]. Nevertheless, while the time of organogenesis can be described as the development step at which malformations occur, the time at which such diabetic malformations are induced during the embryonic development is still a matter of uncertainty. A large survey of congenital malformations has pointed out that diabetic malformations were characterized by a high level of one or more development field defects of blastogenic origin (70 vs. 17% in embryopathies from other origins) [4]. Development fields are ‘those parts of the embryo in which the processes of development of complex structure…are coordinated by a spatially ordered, temporally synchronized and epimorphically hierarchic manner’ [5]. Disturbances of field development may lead to abnormal differentiation and multiple congenital malformations [4, 6]. Hence, multiple development field defects of blastogenic origin imply that the teratogenic agent might be present from the beginning of pregnancy, or even the periconceptional period, i.e. before

organogenesis [4]. This observation would fit in with the results obtained in clinical practice with preconceptional versus postconceptional care [7]. The question regarding ‘when’ the triggering mechanism of congenital malformation occurs is linked to this second important query: how does it occur, i.e. what is the triggering mechanism? Here again, much data has been collected which concerns the organogenesis period. More recently, important data point to possible inducers acting during the preimplantation period [8]. We review those experimental data, and try to reconcile them in a unifying hypothesis, which possibly fits in with the continuous nature of the disturbed pathogenic maternal milieu.

Maternal Diabetes and Preimplantation Development

Embryonic Development before Implantation After fertilization of the oocyte, the ovum starts to divide to form a structure of equipotent cells called morula, floating free in the maternal genital tract. This morula soon enters into the first embryonic differentiation by compacting and thus forming outer and inner cells. The outer cells will constitute the outer layer of the embryo called trophectoderm (TE), while the inner cells will form the embryonic bud or inner cell mass (ICM), which is partially separated from the mural TE by a liquid-filled cavity called the blastocoele. The embryo is then called a blastocyst, which will expand and eventually implant into the adjacent decidualized uterine epithelium. The TE will give rise to most of the placental structures, while the ICM will generate all the embryonic tissues. This first, preimplantation, differentiation of the embryo is important, not only for the implantation phase, but also for postimplantation development, as key progenitor cells, present in the ICM, are necessary for further differentiation and growth [9]. Maternal Diabetes Concordant results were obtained in diabetic rats and mice showing the deleterious effect of maternal diabetes on the embryonic development during the preimplantation phase (table 1). In streptozotocin-induced diabetic rats, the implantation rate on day 6 of pregnancy was reduced by 25%. On day 5, the number of blastocysts floating in the uterine horns was reduced by 20%, while the number of morulae was slightly increased [10, 11]. The blastocysts also had a 15% cell deficit, mainly in the ICM (20%) and less in the TE (12%), exhibiting an abnormal distribution between these two cell lineages [12]. Similar observations were later made in spontaneously diabetic BB rats and in diabetic mice [13, 14]. In alloxan-diabetic mice also, fewer embryos were recovered on

Maternal Diabetes and Embryonic Development

59

Table 1. Effect of maternal diabetes or in vitro exposure to high glucose levels on rat and mouse blastocyst growth Species

Rat

Parameter

Morphological progression Average cell number (blastocysts) Implantation rate Chromatin degradation index Nuclear fragmentation index Clusterin mRNA expression Bcl-2 mRNA expression

Mouse

Morphological progression Average cell number (blastocysts) Implantation rate Chromatin degradation index Bax mRNA expression p53 expression FGF-4 protein expression ROS-induced fluorescence

Effects of Maternal diabetes

Glucose in vitro

20% decrease 20% decrease

13–25% decrease

25% decrease 6-fold increase 5-fold increase 1.7-fold increase

2- to 3-fold increase 3- to 4-fold increase 2.7-fold increase 8-fold increase

10–90% decrease 10% decrease

25–65% decrease

13% decrease 6.3-fold increase 2.5-fold increase ⫹

35% decrease 10-fold increase 2-fold increase ⫹ 32% decrease 2.8-fold increase

day 2 of pregnancy, and showed impaired development in vitro [15]. The abnormal morphology of the rat embryos recovered on day 5 from diabetic females was related to the severity of the diabetic state, and could be prevented by treating the mother with insulin from the time of fertilization, confirming the specificity of the diabetic environment in the observed development impairment [11, 16]. Interestingly, the blastocysts from diabetic rats contained more dead cells, suggesting the existence of an apoptotic process (see below) [12, 17]. Several parameters linked to the diabetic state have been explored to understand the underlying mechanism. Glucose In vitro, high levels of glucose impaired embryo development or induced degeneration of the embryos (table 1). Rat blastocysts exposed to 17 mM glucose showed a 16 and 33% deficit in ICM development after 24 and 48 h, respectively. TE development was less affected. The number of dead cells, with the morphological appearance of apoptotic cells, was increased in those blastocysts [18, 19].

De Hertogh

60

Tumor Necrosis Factor-a Tumor necrosis factor-␣ (TNF-␣) is present in the uterine wall at the time of implantation, and receptors to TNF-␣ are present on the blastocysts [20]. In vitro, TNF-␣ was shown to decrease ICM growth in mouse blastocysts [20]. Whether TNF-␣ may be involved in diabetic embryopathy is suggested by the increased synthesis of this factor in uterine explants from pregnant diabetic rats [21]. In vitro, high levels of glucose stimulate TNF-␣ synthesis by primary cultures of uterine cells [21]. A culture medium, conditioned by uterine explants from pregnant diabetic rats, decreases blastocyst ICM growth, unless TNF-␣ is neutralized in the medium by antibodies, or TNF-␣ receptors in the embryo by appropriate antisense oligonucleotide [22, 23]. Role of Apoptosis The role of apoptosis, suggested by an increased number dead cells with nuclear fragmentation in ICM from diabetic embryos or from embryos incubated in the presence of high levels of glucose, was further investigated (table 1). Chromatin degradation was evidenced by the TUNEL method in the ICM of blastocysts from diabetic rats and mice [17]. The incidence of malformations or embryo resorptions was linked to the presence of the proapoptotic effector Bax [24]. Also, transcripts for clusterin, a compound associated with apoptosis, were higher in blastocysts from diabetic rats [17]. In vitro, a high concentration of glucose increased the expression of Bax factor in mice blastocysts [24] and of clusterin in rat embryos [17]. Bcl2 mRNA, an antiapoptotic factor, was present in more cells of rat blastocysts incubated in the presence of 28 mM as compared to 6 mM glucose [25]. The inhibition of Bcl2 expression increased the incidence of chromatin degradation in those blastocysts [25]. Conversely, inhibition of caspase-3 or caspase-activated deoxyribonuclease, enzymes involved in DNA fragmentation, protected rat blastocysts from glucose-induced chromatin degradation [19]. These observations strongly suggest that apoptosis is involved in diabetes-induced embryo malformations or resorptions, possibly through the loss of key progenitor cells from the ICM. Role of Oxidative Stress Whether an oxidative stress may occur in preimplantation embryos has been suggested by studies performed by cyclic voltammetry on mouse blastocysts cultured in serum from diabetic pregnant women. The viability of those embryos was decreased as well as the antioxidative power of the surviving embryos [26]. It has also been suggested that hydrogen peroxide was a key mediator of naturally occurring apoptosis in mouse blastocysts [27]. Hence, disruption of this equilibrium by high glucose could involve excessive cell damage and impaired embryo development. A direct approach of the oxidative

Maternal Diabetes and Embryonic Development

61

stress has recently shown that a reactive oxygen species (ROS)-induced fluorescence was increased 3-fold in the ICM of mouse embryos exposed to 28 mM glucose compared to 6 mM, keeping the oxidative stress hypothesis as a possible player in the diabetes-induced embryopathy [28]. Glucose Metabolism Glucose is taken up by murine blastocysts through insulin-independent facilitation transport, by passive diffusion and possibly by an active transport system. Recently, an insulin-dependent transporter (Glut-8) was described [29]. Nevertheless, insulin does not increase glucose uptake by preimplantation rat blastocysts in vitro [30]. Although glucose transporters and insulin at low concentrations play a stimulating role in embryo growth, apparently independently of glucose uptake [18, 31–33], at high concentration, insulin may inhibit blastocyst development, possibly by downregulating IGF-1 receptors [18, 34]. Glucose uptake is decreased in mouse blastocysts incubated in a high glucose medium [28, 35], and Glut-1 transporter expression and function are also decreased [36]; similarly, Glut-1, 2 and 3 expressions are decreased in blastocysts from diabetic mice [24]. It has thus been suggested that glucose-induced lowering of glucose uptake would lead to intracellular glucose deficiency and energy shortage, responsible at last for apoptosis and genomic damage through the activation of pathways including p53, Bax and caspases [36, 37]. Previous works in rats and mice had, however, shown that the main glucose utilization pathway in the blastocyst, i.e. glycolysis, was saturated at glucose levels as low as 0.2 mM [30]. It was also shown that, despite a 4-fold reduction in glucose uptake, mice blastocysts incubated at high glucose levels took up glucose in amounts far beyond the saturation capacity of the glycolytic pathway [28]. Moreover, if glycolysis was indeed reduced by 25%, the pentose phosphate pathway was not affected [28]. Hence, it seems unlikely that the apparently specific slight reduction in the glycolytic pathway would be due to decreased glucose uptake, but possibly to the oxidative stress itself. Indeed, the glycolytic enzyme, glyceraldehyde-3-phosphate dehydrogenase, may be inhibited by ROS [38, 39]. From these observations it can be postulated that high glucose levels may induce oxidative stress in the ICM, contributing to lower energy production by decreased glycolysis. Lower glucose uptake, in some instances, could contribute to further lowering of energy production, and to facilitating the triggering of the apoptotic events. Fibroblast Growth Factor-4 Regulatory peptides from the FGF family are produced by the ICM and are involved in TE differentiation through a paracrine mechanism. Hence, fibroblast

De Hertogh

62

Table 2. Congenital malformation rates in transferred embryos from diabetic (NOD) or non-diabetic (ICR) mice [from 43] Embryo

Recipient mice

Total anomalies %

Diabetic NOD ICR ICR

ICR Diabetic NOD ICR

13.8 40 0

growth factor-4 (FGF-4) stimulates TE cell proliferation and represses their differentiation into giant trophoblastic cells [40]. ICM cells from mouse blastocysts, preincubated in high glucose concentrations, have a 30% decrease in their FGF-4 content. The trophoblast outgrowths of these embryos were increased by 18%. An addition of FGF-4 could prevent the abnormal trophoblast outgrowth [41]. This suggests that the abnormal ICM development induced by glucose might in turn, through decreased FGF-4 production and abnormal TE regulation, involve an abnormal TE development and differentiation, further impairing embryo implantation and growth

Maternal Diabetes and the Peri-Implantation Phase

Whether the diabetes-induced preimplantation defects could impair the postimplantation development is a matter of concern, which cannot be easily approached experimentally. The implantation phase might be impaired as suggested by the lower implantation rate in diabetic rats (table 1). Blastocysts from diabetic rats displayed growth impairment when cultured in vitro on fibronectin-coated dishes [42]. The ICM structure was frequently abnormal, and TE outgrowths contained more giant nuclei. In other studies, it was shown that the development of blastocysts, retrieved from diabetic mice and transferred to nondiabetic pseudopregnant recipient females, was impaired and produced a higher incidence of fetal malformations (table 2) [43]. In a similar model, TNF-␣-treated blastocysts from normal mice, transferred to recipient mothers together with untreated blastocysts, showed a higher rate of resorptions on day 15; surviving fetuses from TNF-␣-treated blastocysts had a 15% weight deficit compared to fetuses from untreated embryos (table 3) [44]. Respective placentas were not different. These indirect observations suggest that preimplantation insults could imply further postimplantation development anomalies.

Maternal Diabetes and Embryonic Development

63

Table 3. Development of mice embryos on day 15 after transfer of TNF-␣-treated blastocysts on day 3 [from 44] TNF-␣ ng/ml

Number of embryos transferred

Number of implantations

Number of resorptions

Number of surviving fetuses

Relative weight of surviving fetuses, %

0 50

190 191

117 (62) 98 (52*)

60 (51) 70 (74*)

57 (49) 28 (26*)

100 85*

Figures in parentheses represent percentage. *p ⬍ 0.05 vs. 0 ng/ml.

Maternal Diabetes and Postimplantation Phase

Organogenesis has been considered as the critical period when the diabetic teratogenic factors of maternal origin harm fetal development and produce congenital malformations. Hence, for several decades, early postimplantation embryos have been used in experimental studies in an attempt to disclose the underlying mechanisms, linking the disturbed maternal environment to the production of congenital defects. It soon appeared that multifactorial interplays could be responsible for this pathological linkage [45]. Not only high levels of glucose, but also ketone bodies, branched chain amino acids, somatomedin inhibitors and other diabetesrelated factors could impair in vitro fetal growth and organogenesis. The underlying mechanisms are still ill understood. Many metabolic pathways have revealed anomalies of potentially teratogenic consequences. Myoinositol Pathway High glucose levels induced myoinositol depletion in early somite embryos, resulting in a deficient phospholipid turnover and a reduction in PKC activity. Addition of myoinositol to the culture reduced the incidence of glucoseinduced malformations [46]. Arachidonic Acid and Prostaglandins Arachidonic acid and prostaglandin metabolism are disturbed in embryos from diabetic rats. Recently it was shown that the gene expression of cyclooxygenase-2 (COX-2), a key enzyme in the production of prostaglandin E2 (PGE2), was downregulated by high glucose. PGE2 concentration was decreased in day 10 embryos submitted to high levels of glucose or originating from diabetic

De Hertogh

64

animals [47]. The addition of arachidonic acid or PGE2 to the culture medium, or supplying diabetic females with these compounds, decreased the teratogenic effect of either glucose in vitro or maternal diabetes in vivo, respectively [48]. Contrariwise, inhibition of COX activity in vitro would increase the malformation rate, unless supplementation of arachidonic acid or PGE2 were provided [49]. These results strongly suggest that the prostaglandin pathway may be involved in the teratogenic mechanism induced by glucose or maternal diabetes. They also provide a possible approach to preventing diabetes-induced malformations. Oxidative Stress Increased ROS production and decreased ROS scavenging capacity have been evoked as teratogenic events secondary to maternal diabetes [50, 51]. By giving antioxidants to the diabetic mother (vitamins E or C, or others) it was possible to reduce the occurrence of congenital malformations in rodents [52, 53]. These results strongly suggest that free oxygen radicals may be involved in diabetic teratogenesis, and provide again a possible preventive approach. Moreover, it is suggested that ROS may contribute to the prostaglandin imbalance, resulting from the myoinositol and arachidonic pathways, linking together those three potentially teratogenic mechanisms (fig. 1) [54]. Other Factors Genetic factors have been involved in diabetic teratogenesis. Hence, Pax-3 gene expression, essential for neural tube development, was reduced by diabetes-induced oxidative stress [55]. Immunological factors have also been implicated, in view of the fact that immune-stimulated diabetic mice were less likely to give birth to malformed fetuses that nonstimulated animals [56]. More work is definitely needed to link the teratogenic metabolic factors and the impaired genetic expressions leading to malformations [57].

Diabetes-Induced Embryopathy: A Unifying Hypothesis

It is clear that no single process will explain the underlying mechanisms leading to diabetes-induced congenital malformations. The preimplantation events should not be overlooked, although their impact on later dysmorphogenesis has not yet been proven. Postimplantation disturbances of metabolic pathways, although partially convincing, are in many instances skewed by the

Maternal Diabetes and Embryonic Development

65

Glucose

Oxidative stress

Myoinositol

Arachidonic acid

Prostaglandins Genomic injury

Congenital malformations

Fig. 1. Interplay between the glucose-induced metabolic disturbances leading to congenital malformations.

unrealistic trigger used in the experimental approaches, i.e. glucose levels of 50 mM or more, rarely encountered even in severe diabetic states. Linking both sets of observations (pre- and postimplantation data) may open a new approach, which considers the teratogenic event as a multifactorial event, possibly operating at different development stages [58]. The multifactorial/threshold concept [59] may then possibly also reconcile experimental (pre- and postimplantation) and clinical observations. Following this concept, individuals undergo development stages according to a Gaussian distribution around a mean. For a certain development step to be passed smoothly, a certain threshold of development should be attained; if not, development anomaly will show up. Figure 2 shows the S-shaped distribution of the developmental stage (curve A). If threshold T1 is to be attained for further development (T1 is supposedly placed at mean ⫹3 SD), more than 99% of the individuals will reach the threshold and few malformations will occur. If the required threshold is increased by 1 SD (to T2), as a consequence of the presence of some intercurrent (teratogenic) event, about 2% of the individuals will not reach this new threshold and develop malformations. Alternatively, if for some reason (interfering with the earlier growth of the embryo), the development stage has been slowed down by 1 SD (curve B shifted to the right), the required threshold T1 will not be met by 2% of the individuals. If both events occur together: a shift to a lower development stage (curve B) and an increased

De Hertogh

66

Required threshold T2 T1

Attended requirement (%)

100

2%* 17%* 50

Curve A Curve B 0

⫺3

⫺2 ⫺1 M 1 2 Development stage (mean ⫾ SDs)

3

Fig. 2. The multifactorial/threshold concept of teratogenesis [after 59]. * ⫽ Percentages of required threshold not attended.

threshold (T2), the number of individuals unable to meet the required threshold (T2) will be 17%. Applying this hypothesis to diabetic pregnancy, one might consider that an early developmental disturbance (possibly during the preimplantation phase), that would alone produce only minor consequences, would however amplify considerably the teratogenic effect of a later (postimplantation) event increasing the required development threshold. This multiple-step (or multifactorial) event would also fit in with the development field defects of blastogenic origin observed preferentially in children from diabetic women [4]. Such a phenomenological building up of congenital anomalies may better explain the incidence of congenital malformations in relatively moderate diabetic states well below the metabolic disturbances produced in experimental diabetes or in in vitro studies.

Conclusions

The disruptive mechanism operating in diabetic pregnancy and leading to fetal malformations can best be described as multifactorial events, likely to influence embryo development throughout the periods of blastogenesis and organogenesis (table 4).

Maternal Diabetes and Embryonic Development

67

Table 4. A tentative unifying hypothesis for diabetes-induced teratogenesis Diabetic embryopathy Is not a casual event Is linked to development (development field defect concept) Involves early stages of blastogenesis (preimplantation, possibly periconceptional stages) and later stages (organogenesis) Fits in with the multifactorial/threshold concept of teratogenesis Is possibly linked to a disturbed maternofetal immunological interrelationship Involves genomic insult, leading to apoptosis and/or other genomic injuries, as a consequence of metabolic disturbances (prostaglandin imbalance, oxidative stress) Depends of a genetic background, influencing the sensitivity to the threshold

Experimental observations show that, already during the preimplantation period, embryos submitted to a diabetic environment undergo in their ICM a process of apoptosis, eventually leading to the loss of progenitor cells. Excess glucose decreases glucose uptake and glycolysis in blastocysts, thus decreasing energy production. TNF-␣ overproduction by the uterine wall, oxidative stress in the ICM, and decreased production of TE regulatory factors from ICM origin, like FGF-4, may contribute to the observed impairment of preimplantation embryo development and survival. After implantation of the surviving, although potentially abnormal embryos, the persistence of the diabetic environment would further impair organogenesis and produce malformations. Myoinositol deficiency, arachidonic acid and prostaglandins defects, and oxidative stress are factors observed contributing to the production of abnormal fetal differentiation. The multifactorial/threshold concept may reasonably link the pre- and postimplantation observations, and potentially explain their synergistically disruptive effects on embryo and fetal development in a diabetic environment, which are likely to occur clinically. Aside from further observations needed to acknowledge the above hypothesis, two practical aspects are to be put forward. First, it is clear that the development of the embryo should be protected from the beginning, which implies an early (pre- or even periconceptional) care of the pregnant woman. Having no clear indication as to the degree of normalization required to avoid embryopathy, the message should be to attend the most physiological metabolic situation as possible. Second, the observations made in the postimplantation phase show that correcting some of the anomalies, by supplying myoinositol, arachidonic acid, prostaglandins or antioxidants, might partially prevent the occurrence of congenital anomalies in diabetic animals. Although much remains to be done in terms of the necessary dosage and of pending toxicity of the compounds used,

De Hertogh

68

this line of research deserves to be further investigated as a possible complementary tool for preventing fetal malformations.

References 1 2 3

4 5 6 7 8 9 10

11

12 13 14 15

16 17

18

19

Mills JL, Baker I, Goldman AS: Malformations in infants of diabetic mothers occur before the seventh week. Implications for treatment. Diabetes 1979;28:292–293. Fuhrman K, Reiber H, Semmler K, Fisher F, Fisher M, Glöckner E: Prevention of congenital malformations in infants of insulin-dependent diabetic mothers. Diabetes Care 1983;6:219–223. Mills JL, Knopp RH, Simpson JL, Jovanovic-Peterson L, Metzger BE, Holmes LB, Aarons JS, Brown Z, Reed GF, Bieber FR, Van Allen M, Holzman I, Ober C, Peterson CM, Withiam MJ, Duckles A, Müller-Heubach E, Polk BF: Lack of relation of increased malformation rates in infants of diabetic mothers to glycemic control during organogenesis. N Engl J Med 1988;318: 671–676. Martinez-Frias ML: Development field defects and associations: Epidemiological evidence of their relationship. Am J Med Genet 1994;49:45–51. Opitz JM: The development field concept in clinical genetics. J Pediatr 1982;101:805–809. Martinez-Frias ML, Frias JL, Opitz JM: Errors of morphogenesis and development field theory. Am J Med Genet 1998;76:291–296. Kitzmiller JL, Gavin LA, Gin GD, Jovanovic-Peterson L, Main EK, Zigrang WD: Pre-conception care of diabetes: Glycemic control prevents congenital anomalies. JAMA 1991;265:731–736. Pampfer S, De Hertogh R: Pre-implantation embryopathy associated with maternal diabetes. Diabetes Rev 1996;10:90–113. Tam PL: Post-implantation development of mitomycin-C-treated mouse blastocysts. Teratology 1988;37:205–212. De Hertogh R, Vanderheyden I, Glorieux B, Ekka E: Oestrogen and progestogen receptors in endometrium and myometrium at the time of blastocyst implantation in diabetic pregnant rats. Diabetologia 1989;32:568–572. Vercheval M, De Hertogh R, Pampfer S, Vanderheyden I, Michiels B, De Bernardi P, De Meyer R: Experimental diabetes impairs rat embryo development during the pre-implantation period. Diabetologia 1990;33:187–191. Pampfer S, De Hertogh R, Vanderheyden I, Michiels B, Vercheval M: Decreased inner cell mass proportion in blastocysts from diabetic rats. Diabetes 1990;39:471–476. Lea RG, McCracken JE, McIntyre SS, Smiyh W, Baird JD: Disturbed development of the preimplantation embryo in the insulin-dependent diabetic BB/E rat. Diabetes 1996;45:1463–1470. Beebe LFS, Kaye PL: Maternal diabetes and retarded pre-implantation development of mice. Diabetes 1991;40:457–461. Diamond MP, Moley KH, Pellicer A, Vaughin WK, De Cherney AH: Effects of streptozotocin- and alloxan-induced diabetes mellitus on mouse follicular and early embryo development. J Reprod Fertil 1989;86:1–10. De Hertogh R, Vanderheyden I, Pampfer S, Robin D, Delcourt J: Maternal insulin treatment improves pre-implantation embryo development in diabetic rats. Diabetologia 1992;35:406–408. Pampfer S, Vanderheyden I, McCracken J, Vesela J, De Hertogh R: Increased cell death in rat blastocysts exposed to high glucose or tumor necrosis factor-␣ in vitro. Development 1997;124: 4827–4836. De Hertogh R, Vanderheyden S, Pampfer S, Robin D, Dufrasne E, Delcourt J: Stimulatory and inhibitory effects of glucose and insulin on rat blastocyst development in vitro. Diabetes 1991;40: 641–647. Hinck L, Van Der Smissen P, Heusterpreute M, Donnay I, De Hertogh R, Pampfer S: Identification of caspase-3 and caspase-activated deoxyribonuclease in rat blastocysts and their implication in the induction of chromatin degradation (but not nuclear fragmentation) by high glucose. Biol Reprod 2001;64:146–153.

Maternal Diabetes and Embryonic Development

69

20

21

22

23

24

25

26

27 28

29

30

31 32 33 34

35 36 37 38 39 40

Paampfer S, Wuu Y, Vanderheyden I, De Hertogh R: Expression of tumor necrosis factor-␣ receptors and selective effect of TNF-␣ on the inner cell mass of mouse blastocysts. Endocrinology 1994;134:206–212. Pampfer S, Vanderheyden I, De Hertogh R: Increased synthesis of tumor necrosis factor-␣ in uterine explants from pregnant diabetic rats and in primary cultures of uterine cells in high glucose. Diabetes 1997;46:1214–1224. Pampfer S, Vanderheyden I, Wuu Y, Baufays L, Maillet O, De Hertogh R: A possible role for tumor necrosis factor-␣ in the early embryopathy associated with maternal diabetes in the rat. Diabetes 1995;44:531–536. Pampfer S, Vanderheyden I, Vesela J, De Hertogh R: Neutralization of TNF-␣ action on cell proliferation in rat blastocysts by antisense oligonucleotides directed against TNF-␣ p-60 receptor. Biol Reprod 1995;52:1316–1326. Moley KH, Chi MM, Knudson CM, Korsmeyer SJ, Mueckler MM: Hyperglycemia induces apoptosis in pre-implantation embryos through cell death effector pathways. Nat Med 1998;4: 1421–1424. Pampfer S, Cordi S, Vanderheyden I, Van Der Smissen P, Courtoy PJ, Van Cauwenberghe A, Donnay I, De Hertogh R: Expression and role of Bcl-2 in rat blastocysts exposed to high D-glucose. Diabetes 2001;50:143–149. Ornoy A, Kimyagarov D, Yaffee P, Abir R, Raz I, Kohen R: Role of reactive species (ROS) in the diabetes-induced embryotoxicity: Studies on pre-implantation mouse embryos cultured in serum from diabetic pregnant women. Isr J Med Sci 1996;32:1066–1073. Parchment RE: The implication of a unifying theory of programmed cell death, polyamines, oxyradicals and histogenesis in the embryo. Int J Dev Biol 1993;37:75–83. Leunda-Casi A, Genicot G, Pampfer S, De Hertogh R: Increased cell death in mouse blastocysts exposed to high D-glucose in vitro: Implication of an oxidative stress and alterations in glucose metabolism. Diabetologia 2002;45:571–579. Carayannopoulos MO, Chi MMY, Cui Y, Pingsterhaus JM, McKnight RA, Mueckler M, Devaskar SU, Moley KH: Glut8 is a glucose transporter responsible for insulin-stimulated glucose uptake in the blastocyst. Proc Natl Acad Sci USA 2000;97:7313–7318. Dufrasnes E, Vanderheyden I, Robin D, Delcourt J, Pampfer S, De Hertogh R: Glucose and pyruvate metabolism in pre-implantation blastocysts from normal and diabetic rats. J Reprod Fertil 1993;98:169–177 Pinto AB, Carayannopoulos MO, Hoen A, Dowd D, Moley KH: Glucose transporter 8 expression and translocation are critical for murine blastocyst survival. Biol Reprod 2002;66:1729–1733. Pantaleon M, Harvey MB, Pascoe WS, James DE, Kaye PL: Glucose transporter GLUT3: Ontogeny, targeting and role in the mouse blastocyst. Proc Natl Acad Sci USA 1997;94:3795–3800. Augustin R, Pocar P, Wrenzycki C, Niemann H, Fisher B: Mitogenic and anti-apoptotic activity of insulin on bovine embryos produced in vitro. Reproduction 2003;126:91–99. Chi MM, Schlein AL, Moley KH: High insulin-like growth factor 1 (IGF-1) and insulin concentrations trigger apoptosis in the mouse blastocyst via down-regulation of the IGF-1 receptor. Endocrinology 2000;141:4784–4792. Moley KH, Chi MM, Mueckler MM: Maternal hyperglycemia alters glucose transport and utilization in mouse pre-implantation embryos. Am J Physiol 1998;275:E38–E47. Ogura K, Sakata M, Yamaguchi M, Kurachi H, Murata Y: High concentrations of glucose decreases transporter-1 expression in mouse placenta in vitro and in vivo. J Endocrinol 1999;160:443–452. Kleim AL, Chi MM, Moley KH: Hyperglycemia-induced apoptotic cell death in the mouse blastocyst is dependent on expression of p53. Mol Reprod Dev 2001;60:214–224. Knight RJ, Kofoed KF, Schelbert HR, Buxton DB: Inhibition of glyceraldehyde-3-phosphate dehydrogenase in post-ischemic myocardium. Cardiovasc Res 1996;32:1016–1023. Wentzel P, Ejdesjo A, Eriksson U: Maternal diabetes in vivo and high glucose in vitro diminish GAPDH activity in rat embryos. Diabetes 2003;52:1222–1228. Leunda-Casi A, De Hertogh R, Pampfer S: Control of trophectoderm differentiation by inner cell mass-derived fibroblast growth factor-4 in mouse blastocysts and corrective effects of FGF-4 on high glucose-induced trophoblast disruption. Mol Reprod Dev 2001;60:38–46.

De Hertogh

70

41

42 43

44 45 46

47

48 49 50 51

52 53 54

55

56

57

58

59

Leunda-Casi A, De Hertogh R, Pampfer S: Decreased expression of fibroblast growth factor-4 and associated dysregulation of trophoblast differentiation in mouse blastocysts exposed to high D-glucose in vitro. Diabetologia 2002;45:571–579. Pampfer S, Wuu Y, Vanderheyden I, De Hertogh R: In vitro study of the carry-over effect associated with early diabetic embryopathy in the rat. Diabetologia 1994;37:855–862. Otani H, Tanaka O, Tatewaki R, Naora H, Yoneyama T: Diabetic environment and genetic predisposition as causes of congenital malformations in NOD mouse embryos. Diabetes 1991;40: 1245–1250. Wuu Y, Pampfer S, Becquet P, Vanderheyden I, Lee KH, De Hertogh R: Tumor necrosis factor-␣ decreases the viability of mouse blastocysts in vitro and in vivo. Biol Reprod 1999;60:479–483. Reece EA, Homko CJ, Wu YK: Multifactorial basis of the syndrome of diabetic embryopathy. Teratology 1996;54:171–182. Wentzel P, Wentzel CR, Gareskog MB, Eriksson UJ: Induction of embryonic dysmorphogenesis by high glucose concentration disturbed inositol metabolism and inhibited protein kinase C activity. Teratology 2001;63:193–201. Wentzel P, Welsh N, Eriksson UJ: Developmental damage, increased lipid peroxidation, diminished cyclooxygenase-2 gene expression, and lowered PGE2 levels in rat embryos exposed to a diabetic environment. Diabetes 1999;48:813–820. Wiznitzer A, Furman B, Mazor M, Reece EA: The role of prostanoids in the development of diabetic embryopathy. Semin Reprod Endocrinol 1999;17:175–181. Wentzel P, Eriksson UJ: Anti-oxidants diminish developmental damage by high glucose and cyclooxygenase inhibitors in rat embryos in vitro. Diabetes 1998;47:677–684. Yang X, Borg LAH, Eriksson UJ: Altered metabolism and superoxide generation in neural tissue of rat embryos exposed to high glucose. Am J Physiol 1997;272:E173–E180. Trocino RA, Akasawa S, Ishibashi M, Matsumoto K, Matsuo H, Yamamoto H, Goto S, Urata Y, Kondo T, Nagataki S: Significance of glutathione depletion and oxidative stress in early embryogenesis in glucose-induced rat embryo culture. Diabetes 1995;44:992–998. Viana M, Castro M, Barbas C, Herrera E, Bonet B: Effect of different doses of vitamin E on the incidence of malformations in pregnant diabetic rats. Ann Nutr Metab 2003;47:6–10. Siman CM, Eriksson UJ: Vitamin C supplementation of the maternal diet reduces the rate of malformation in the offspring of diabetic rats. Diabetologia 1997;40:1416–1424. Eriksson UJ, Wentzel P, Hod M: Clinical and experimental advances in the understanding of diabetic embryopathy; in Hod M, Jovanovic L, Di Renzo JC, de Leiva A, Langer O (eds):Textbook of Diabetes and Pregnancy. London, Martin Dunitz, 2003, pp 262–275. Chang TI, Horal M, Jain SK, Wang F, Patel R, Loeken MR: Oxidant regulation of gene expression and neural tube development. Insight gained from diabetic pregnancy on molecular causes of neural tube defects. Diabetologia 2003;46:538–545. Punareewattana K, Sharova LV, Li W, Ward DL, Holladay SD: Reducing birth defects caused by maternal immune stimulation may involve increased expression of growth promoting genes and cytokine GM-CSF in the spleen of diabetic ICR mice. Int Immunopharmacol 2003;3:1639–1655. Chang TI, Loeken MR: Genotoxicity and diabetic embryopathy: Impaired expression of development control genes as a cause of defective morphogenesis. Semin Reprod Endocrinol 1999;17: 153–165. De Hertogh R, Leunda-Casi A, Hinck L: Pre-implantation embryopathy and maternal diabetes; in Hod M, Jovanovic L, Di Renzo JC, de Leiva A, Langer O (eds):Textbook of Diabetes and Pregnancy. London, Martin Dunitz, 2003, pp 240–252. Fraser FC: The multifactorial/threshold concept – Uses and misuses. Teratology 1977;14: 267–280.

R. De Hertogh, MD † 145 Rue du Monastère BE–1330 Rixensart (Belgium) Tel. ⫹32 2 6535904, Fax ⫹32 2 7645418, E-Mail [email protected]

Maternal Diabetes and Embryonic Development

71

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 72–82

Normal and Abnormal Fetal Growth Irene Cetin, Tatjana Radaelli Institute of Obstetrics and Gynecology ‘Luigi Mangiagalli’, University of Milano, Milan, Italy

The birth of a term, normal weight and healthy infant represents the main goal of pregnancy. This is achieved through a number of processes that regulate and determine intrauterine fetal growth. The increase in size over time of the fetus occurs simultaneously with maturation and developmental processes over the course of pregnancy. However, in case of abnormalities of fetal growth, such as intrauterine growth restriction or the excess of fetal growth associated with a diabetic pregnancy, the distinction between maturation and size becomes evident. Deviations from normal growth that occur during intrauterine life are not only associated with increased mortality and morbidity during the perinatal period, but carry pronounced effects on neonatal and adult health [1]. In particular, the offspring of diabetic pregnancies have an increased risk of obesity and type 2 diabetes [2, 3], conditions that might be associated with the increased fetal fat mass that has been reported in these pregnancies [4, 5].

Intrauterine Growth

Birth weight is the result of genetic potential, substrate availability and endocrine regulation that will determine the growth trajectory of the fetus in utero. Although the period of embryonic growth is characterized by very rapid cell proliferation, most of the fetal weight is gained during the second half of gestation. In this period, intrauterine growth of the fetus is accompanied by a large and exponential deposition of fat tissue [6] and at birth the human fetus is the mammal with the greatest amount of body fat [7]. Back in 1923, Moulton [8] showed that the variability in weight among mammalian species was explained by the amount of adipose tissue, whereas the

amount of lean body mass was relatively constant. These animal data were confirmed in humans by some chemical analysis studies [9]: a relatively comparable rate of accretion of lean mass was described in term fetuses with different patterns of growth, indicating fat accretion as the most important determinant of different weight in small-for-gestational age, appropriate-for-gestational age and large-for-gestational age (LGA) babies at birth. In addition, Catalano et al. [10] demonstrated that although constituting 12–14% of birth weight, fat mass accounts for 46% of the variance in neonatal weight. Maternal anthropometric variables are important factors related to fetal growth. Maternal pregravid weight and height, weight gain during gestation, parity and gestational age have been correlated with fetal weight [11]. Abrams and Laros [12] in particular reported a strong correlation between maternal weight gain and birth weight in women with a pregravid weight between underweight and moderately overweight, while in severely overweight women there was no correlation despite heavier fetuses. In contrast, paternal anthropometric factors have a limited effect on fetal birth weight compared to maternal ones. A recent study by Klebanoff et al. [13] reported that paternal birth weight, adult height and adult weight are able to account for only about 3% of the variance in neonatal birth weight.

Determinants of Fetal Nutrition

As noted by Sparks [9], genetic and anthropometric factors may influence birth weight and lean mass, whereas the in utero environment may correlate better with fetal fat mass. Principal determinants of fetal nutrition, i.e. of nutrient composition in the umbilical supply to the fetus, are summarized in figure 1. Fetal growth is regulated by the balance between the fetal nutrient demand and the maternal-placental supply that is strictly related to maternal nutrition and metabolism, uteroplacental blood flow, placental size and its transfer capabilities. There is considerable experimental evidence suggesting that in the second half of gestation fetal growth is controlled by both maternal and placental factors. It is difficult to estimate the relative influence of these two compartments in determining the rate of intrauterine fetal growth. Both maternal and placental factors are involved in alterations of fetal growth rate, by changing the amount of nutrients reaching the fetus. Maternal adaptation to pregnancy involves numerous changes that affect the metabolism as well as the function of the fetal placental unit through hemodilution, reduction of peripheral resistance to blood flow and reduction of blood pressure, most of all within the placental district, and hormonal changes. Throughout pregnancy, the mother adapts her metabolism in order to support

Fetal Growth

73

Maternal nutrition

Maternal metabolism

Placental substrate regulation

Placental transport and metabolism

Umbilical uptake

Fetal metabolism

Fig. 1. Principal determinants of fetal growth: the balance between the fetal nutrient demand and the maternal-placental supply is strictly related to maternal nutrition and metabolism, uteroplacental blood flow, placental size and its transfer capabilities.

the continuous draining of substrates by the fetoplacental unit. There is a marked increase in blood glycerol, free fatty acids and ketoacids induced by fasting of even moderate degrees and this phenomenon is known as ‘accelerated starvation of pregnancy’. Many studies have reported a good relationship between maternal metabolic parameters, i.e. 2-hour glucose concentration after OGTT [14], mean glucose concentration [15], postprandial glucose concentration [16], and neonatal birth weight. Furthermore, Schaefer-Graf et al. [17] showed that maternal BMI and a previous pregnancy with an LGA fetus appear to have the strongest influence on fetal growth in the late second trimester and early third trimester of gestation, while maternal glycemia predominates later in the third trimester. A strong correlation has also been demonstrated between the decreased maternal insulin sensitivity index in late pregnancy and neonatal birth weight and lean mass while insulin sensitivity before conception shows the best correlation with neonatal fat mass [18].

Cetin/Radaelli

74

The placenta has complex metabolic and endocrine activities and is essential for the growth and survival of the fetus in utero. There is a strong correlation between placental and fetal weight [19]. However, in the human placental growth follows an S-curve regression, while fetal growth follows an exponential pattern with maximum growth in the third trimester [20]. As a consequence, the fetal/placental weight ratio increases significantly during pregnancy [20]. Nutrients are transferred to the fetus by the placenta through a number of mechanisms involving transport systems present on the throphoblast microvillous and basal membranes and on the endothelial membranes of the fetal capillaries. The most important nutrients for the fetus are glucose, lactate, amino acids and fatty acids. Placental function varies though the pregnancy so that while at midgestation the placenta utilizes half of the oxygen and glucose uptake and transfers the rest to the fetal circulation, the proportion transferred to the fetus increases with progressing gestation.

Abnormal Fetal Growth

Changes that influence the supply of nutrients to the fetus might lead to alterations of the fetal growth trajectory. In the model of intrauterine growth restriction, maternal characteristics do not seem to be determinants of fetal growth rate, whereas it is generally accepted that the major constraining factor is the ability of the uteroplacental unit to supply oxygen and nutrients to the fetus [21]. Rather, in gestational diabetes, it has always been hypothesized that excess fetal growth is deriving from the increased availability of maternal nutrients to the placenta. The increased intrauterine growth and fetal fat mass deposition in these cases would then result from the combined effects of this excess of nutrients and the permissive environment of fetal hyperinsulinemia. Studies by Pettitt et al. [2] and by Silverman et al. [3] have shown that infants of women with gestational diabetes have an increased risk of adolescent obesity and glucose intolerance. Macrosomia is the classic feature of fetuses and infants born to diabetic mothers, but only few studies have extensively investigated the body composition of these infants at the time of delivery. Whitelaw [22] first and Enzi et al. [23] later reported that neonatal fat mass as estimated by skinfolds was significantly greater in infants from women with diabetes mellitus and gestational diabetes, respectively. More recently, Catalano et al. [5] studied neonatal body composition in 197 infants from diabetic mothers and 220 infants from a control group, using both skinfolds and TOBEC (totalbody electrical conductivity) (fig. 2). They confirmed that infants from diabetic mothers have an increased percent body fat compared to infants of women with normal glucose tolerance; this was independent of birth weight and could be

Fetal Growth

75

Fat-free mass (g) Fat mass (g) Body fat (%)

3,500 3,000 2,500 2,000 1,500 1,000 500 0 500 450 400 350 300 250 200 150 100 50 0 14 12 10 8 6 4 2 0

NS

GDM

NGT

*

GDM

NGT

**

GDM

NGT

Fig. 2. Fat-free mass, fat mass and percent body fat were evaluated in 197 infants from diabetic mothers and 220 infants from a control group using TOBEC. Neonates from gestational diabetes mellitus (GDM) mothers had a significantly increased fat mass and percent body fat compared to neonates from normal glucose-tolerant (NGT) mothers, despite a similar; this was independent of birth weight and fat-free mass [from 5]. *p ⫽ 0.0002; **p ⫽ 0.0001. NS ⫽ Nonsignificant.

seen as the early onset of adolescent obesity. A relationship has also been observed between fetal leptin levels and fat thickness measured by ultrasound in utero at the level of the abdominal circumference (fig. 3) [4]. According to the Pedersen’s hypothesis, fetal overgrowth or macrosomia are consequences of increased maternal glucose, which first stimulates fetal insulin production and probably other growth factors [24]. Although treatment of hyperglycemia can markedly decrease the risk of macrosomia, we are far from the definition of an optimal maternal treatment. First of all, we relate insulin resistance to glucose metabolism, but other nutrients are also involved in determining the peculiar in utero environment associated with gestational diabetes mellitus. Decreased maternal insulin sensitivity or increased insulin resistance in fact leads to decreased glucose uptake in skeletal muscle, white

Cetin/Radaelli

76

Umbilical venous leptin (ng/ml)

0.52cm

40 35 30 25 20 15 10 5 0

4.8 ng/ml 0

0.2

0.4

0.6

0.8

1

Abdominal fat thickness (cm)

Fig. 3. Leptin concentrations in fetuses from gestational diabetes mellitus pregnancies: relation to the in utero measurements of fetal fat abdominal thickness by ultrasound. The vertical and horizontal lines represent the mean abdominal fat thickness ⫹1 SD (0.52 cm) and the mean umbilical venous leptin concentration ⫹1 SD (4.8 ng/ml), respectively, measured in normal pregnancies (adequate-for-gestational age) [from 4].

adipose tissue and liver as well as decreased suppression of endogenous glucose production, lipolysis and amino acid turnover leading to an increased availability of nutrients for the fetus resulting in intrauterine overgrowth [25]. Normal maternal glucose levels have usually been considered as the main target of any protocol for the management of pregnancies complicated by gestational diabetes. Nevertheless, macrosomia complicates as many as 50% pregnancies with gestational diabetes despite an optimal glycemic control. Many factors in fact may affect fetal growth by influencing the exchange of nutrients and oxygen from maternal blood through the placenta into the fetal circulation. In diabetic mothers an increased availability of a number of nutrients has been reported, with increases in the plasma levels of glucose, free fatty acids, triglycerides and several amino acids. Changes in the placental nutrient transfer capacity also have to be considered. Increased placental weights and placental ratios (placental weight to birth weight ratio) [26] have been reported in pregnancies complicated by gestational diabetes even in the presence of optimal maternal glycemic control throughout the third trimester [27]. This was present independently of the type of treatment or the degree of glycemic control during treatment, probably because most of the placental growth occurs in the first half of gestation, well before the diagnosis of gestational diabetes and beginning of treatment. The increased placental mass could then augment placental nutrient exchange by increasing the surface for substrate transfer. Alterations in the activity and expression of placental nutrient transporters could also influence the relationship between maternal substrate levels and fetal growth in diabetic pregnancies. The protein expression of the primary placental

Fetal Growth

77

glucose transporter isoform GLUT1 as well as glucose transport activity has been shown to be increased in the basal membrane of type 1 diabetic pregnancies [28], but not in gestational diabetes [29]. In contrast, the transport of neutral amino acids across the syncytiotrophoblast membranes has been shown to be significantly increased in placentae obtained from both type 1 and gestational diabetic pregnancies [30]. Recent observations in vivo demonstrate indeed that the umbilical delivery of amino acids is significantly increased in gestational diabetic pregnancies [31], in contrast to what has been reported in pregnancies associated with intrauterine growth restriction [32].

Evaluation of Fetal Growth and Fetal Body Composition

Birth weight has been generally used as an indicator of intrauterine longitudinal growth. However, different patterns exist for fetal length and weight growth, probably due to differences in the growth curves of lean and fat mass. Lean mass has been associated with linear growth and peak velocity in the first stages of pregnancy, while a significant exponential increase in fetal body fat mass has been observed during the third trimester both at delivery [6] and in utero by longitudinal ultrasound evaluation of fetal fat and lean mass [33]. Infants from diabetic pregnancies undergo excess intrauterine growth, so their birth weight exceeds the normal ranges. Definitions of this condition at birth have been macrosomia, if birth weight exceeds 4,000 g or LGA, thus placing birth weight in relation to gestational age. However, as for the opposite condition of intrauterine growth restriction, the change in growth trajectory can be diagnosed in utero by ultrasound based on multiple records showing an increasing abdominal circumference, but the intrauterine pattern of growth has never been included in a more strict definition, although many authors now refer to it as ‘excess fetal growth’. Many studies have been published on the potential of ultrasound measurements to predict macrosomia. These studies have demonstrated that the sonographic measurement of the fetal abdominal circumference allows estimating the increase of fetal insulin-dependent tissue growth rate more than the biparietal diameter, a typical example of non-insulin-dependent tissue [34, 35]. The ultrasonographic determination of the abdominal circumference is able to predict 78% of macrosomic fetuses [36]. This measurement has been used to identify cases at risk for fetal macrosomia among gestational diabetic mothers [37]. Moreover, in the last years, nontraditional biometric parameters indicative of fetal fat and lean mass growth in utero have been identified [33]. In particular, fat thickness can be measured at the level of the abdominal circumference (fig. 4) and higher values for this parameter have been observed in those

Cetin/Radaelli

78

Fig. 4. A magnification of the anterior abdominal wall was obtained on a transversal section of the fetal trunk at the level of abdominal circumference. The subcutaneous fetal fat tissue can be evidenced as an external hyperechogenic surface.

gestational diabetic fetuses that had the highest leptin levels (fig. 3) [4]. Other measures of fetal fat can be obtained in utero by ultrasound at the level of the arm, thigh and subscapular. Bernstein et al. [33] proposed longitudinal normative values for both femoral fat and humeral fat in fetuses from nonobese normal glucose-tolerant patients; they suggested that, as a result of an accelerated rate of growth in late gestation, the measurement of fetal fat will provide a more sensitive and specific marker of abnormal fetal growth. However, although these measures can be useful in pathophysiological studies, they have not yet been standardized and at the moment they do not offer a valid alternative to the use of the abdominal circumference for the evaluation of the growth of fetal fat tissue in routine clinical practice.

Conclusions

Fetal Growth as a Criterion for Managing Diabetic Pregnancies Normalizing the rate of fetal growth is a primary goal in treating women with pregnancies complicated by gestational diabetes mellitus. With this in mind, it is difficult to understand why the evaluation of fetal growth has not been included in the criteria utilized for the clinical management of diabetes in pregnancy. Obviously, fetal growth is evaluated during gestation in developed countries, but only maternal glycemia is considered for decisions related to maternal metabolic treatment. While normalizing maternal glucose levels has

Fetal Growth

79

reduced neonatal morbidity in diabetic pregnancies, the macrosomia rate is still elevated compared with that in the normal obstetric population [38]. Management of gestational diabetes guided by fetal criteria can identify pregnancies at low risk for neonatal morbidity and limits intensive monitoring and/or therapy to those at high risk for excess fetal growth. Measurement of fetal abdominal circumference growth has demonstrated low rates of LGA newborns in infants whose abdominal circumference remains below the 70th to 75th percentile during pregnancy, even in those cases where glycemic criteria would have indicated insulin therapy [37]. In contrast, institution of insulin therapy in pregnancies with accelerated fetal growth without respect to maternal glycemia has been found to substantially decrease LGA rates [37]. Very recently, two prospective randomized studies have reported promising data proposing clinical management of diabetic patients based on maternal glycemic criteria combined with evaluation of fetal growth [39, 40]. Although outcomes were quite similar between the two randomized groups, Bonomo et al. [40] reported a reduction in the incidence of LGA in patients with a combined sonographic management. Moreover, the overall percent of appropriatefor-gestational age fetuses was significantly increased also due to the decrease of glucose testing and the use of insulin in low-risk patients, i.e. those with normally growing fetuses. Fetal growth should therefore be considered as one of the criteria for determining clinical management of diabetes in pregnancy, avoiding unnecessary intervention in low-risk pregnancies and intensifying therapy and controls in those showing alterations of fetal growth.

References 1 2 3

4

5 6 7 8 9

Barker DJ: Fetal and Infant Origins of Adult Disease. London, BMJ Press, 1992. Pettitt DJ, Baird HR, Aleck KA, Bennett PH, Knowler WC: Excessive obesity in offspring of Pima Indian women with diabetes during pregnancy. N Engl J Med 1983;5:446–451. Silverman BL, Rizzo TA, Cho NH, Metzger BE: Long-term effects of the intrauterine environment: The Northwestern University Diabetes in Pregnancy Center. Diabetes Care 1998;21(suppl): B142–B149. Cetin I, Morpurgo PS, Radaelli T, Taricco E, Cortellazzi D, Bellotti M, Pardi G: Fetal plasma leptin concentrations: Relationship with different intrauterine growth patterns from 19 weeks to term. Pediatr Res 2000;48:646–451. Catalano PM, Thomas A, Huston-Presley L, Amini SB: Increased fetal adiposity: A very sensitive marker of abnormal in utero development. Am J Obstet Gynecol 2003;189:1698–1704. Enzi G, Zanardo V, Caretta F, Inelmen EM, Rubaltelli F: Intrauterine growth and adipose tissue development. Am J Clin Nutr 1981;34:1785–1790. Widdowson EM, Dickerson JWT: Chemical composition of the body; in Comar CL, Bronner F (eds): Mineral Metabolism. New York, Academic Press, 1964, vol 2, pp 2–247. Moulton CR: Age and chemical development in mammalians. J Biol Chem 1923;57:79–97. Sparks JW: Human intrauterine growth and accretion. Semin Perinatol 1984;8:74–93.

Cetin/Radaelli

80

10 11 12 13 14 15

16

17

18 19

20 21 22 23 24 25

26 27 28 29 30 31

32

33

Catalano PM, Tyzbir ED, Allen SR, McBean JH, McAuliffe TL: Evaluation of fetal growth by estimation of neonatal body composition. Obstet Gynecol 1992;79:46–50. Love EJ, Kinch RAH: Factors influencing the birth weight in normal pregnancy. Am J Obstet Gynecol 1965;91:342–349. Abrams BF, Laros RK: Prepregnancy weight, weight gain and birth weight. Am J Obstet Gynecol 1986;154:503–509. Klebanoff MA, Mednick BR, Schulsinger C, Secher NJ, Shiono PH: Father’s effect on birth weight. Am J Obstet Gynecol 1998;178:1022–1026. Tallarigo L, Giampietro O, Penno G, Miccoli R, Gregori G, Navalesi R: Relation of glucose tolerance to complications of pregnancy in nondiabetic women. N Engl J Med 1986;315:989–992. Langer O, Levy J, Brustman L, Anyaegbunam A, Merkatz R, Divon M: Glycemic control in gestational diabetes – How tight is tight enough: Small for gestational age versus large for gestational age? Am J Obstet Gynecol 1989;61:646–653. Jovanovic-Peterson L, Peterson CM, Reed GF, Metzger BE, Mills JL, Knopp RH, Aarons JH: Maternal postprandial glucose levels and infant weight: The diabetes in early pregnancy study. Am J Obstet Gynecol 1991;164:103–111. Schaefer-Graf UM, Kios SL, Kilavuz O, Plageman A, Brauer M, Dudenhausen JW, Vetter K: Determinants of fetal growth at different periods of pregnancies complicated by gestational diabetes mellitus or impaired glucose tolerance. Diabetes Care 2003;26:193–198. Catalano PM, Drago NM, Amini SB: Maternal carbohydrate metabolism and its relationship to fetal growth and body composition. Am J Obstet Gynecol 1995;172:1464–1470. Molteni RA, Stys SJ, Battaglia FC: Relationship of fetal and placental weight in human beings: Fetal/placental weight ratio at various gestational ages and birth weight distribution. J Reprod Med 1978;21:327–334. Heinonen S, Taipale P, Saarikoski S: Weights of placentae from small-for gestational age infants revisited. Placenta 2001;22:399–404. Pardi G, Marconi AM, Cetin I: Placental-fetal interrelationship in IUGR fetuses – a review. Placenta 2002;23(suppl A):S136–S141. Whitelaw A: Subcutaneous fat in newborn infants of diabetic mothers: An indication of quality of diabetic control. Lancet 1977;i:15–18. Enzi G, Inelman EM, Caretta F, Villani F, Zanardo V, DeBiasi F: Development of adipose tissue in newborns of gestational diabetes and insulin diabetic mothers. Diabetes 1980;29:100–104. Pedersen J: The Pregnant Diabetic and Her Newborn: Problems and Management. Baltimore, Williams & Wilkins, 1967. Catalano PM, Kirwan JP, Haugel-de Mouzon S, King J: Gestational diabetes and insulin resistance: Role in short- and long-term implications for mother and fetus. J Nutr 2003;133/5(suppl 2): 1674S–1683S. Lao TT, Lee CP, Wong WM: Placental weight to birth weight ratio is increased in mild gestational glucose intolerance. Placenta 1997;18:227–230. Taricco E, Radaelli T, Nobile de Santis MS, Cetin I: Foetal and placental weights in relation to maternal characteristics in gestational diabetes. Placenta 2003;24:343–347. Gaither K, Quraishi AN, Illsley NP: Diabetes alters the expression and activity of the human placental GLUT1 glucose transporter. J Clin Endocrinol Metab 1999;84:695–701. Jansson T, Ekstrand Y, Wennergren M, Powell TL: Placental glucose transport in gestational diabetes. Am J Obstet Gynecol 2001;184:111–116. Jansson T, Ekstrand Y, Bjorn C, Wennergren M, Powell TL: Alterations in activity of placental amino acid transporters in pregnancies complicated by diabetes. Diabetes 2002;51:2214–2219. Cetin I, de Santis MS, Taricco E, Radaelli T, Teng C, Ronzoni S, Spada E, Milani S, Pardi G: Maternal and fetal amino acid concentrations in normal pregnancies and in pregnancies with gestational diabetes mellitus. Am J Obstet Gynecol 2005;192:610–617. Cetin I, Ronzoni S, Marconi AM, Perugino G, Corbetta C, Battaglia FC, Pardi G: Maternal concentrations and fetal-maternal concentration differences of plasma amino acids in normal and intrauterine growth retarded (IUGR) pregnancies. Am J Obstet Gynecol 1996;174:1575–1583. Bernstein IM, Goran MI, Amini SB, Catalano PM: Differential growth of fetal tissues during the second half of pregnancy. Am J Obstet Gynecol 1997;176:28–32.

Fetal Growth

81

34 35

36 37

38

39

40

Benson CM, Doubilet PM, Saltzman DM: Sonographic determination of fetal weights in diabetic pregnancies. Am J Obstet Gynecol 1987;156:441–444. Landon MB, Mintz MC, Gabbe SG: Sonographic evaluation of fetal abdominal growth: Predictor of the large-for-gestational-age infant in pregnancies complicated by diabetes mellitus. Am J Obstet Gynecol 1989;160/1:115–121. Shepard MJ, Richards VA, Berkowitz RL: An evaluation of two equations for predicting fetal weight by ultrasound. Am J Obstet Gynecol 1982;152:47–54. Buchanan TA, Kjos SL, Montoro MN, Wu PYK, Madrilejo NG, Gonzalez M, Nunez V, Pantoja PM, Xiang A: Use of fetal ultrasound to select metabolic therapy for pregnancies complicated by mild gestational diabetes. Diabetes Care 1994;17:275–283. Metzger BE, Coustan DR: Summary and recommendations of the Fourth International WorkshopConference on Gestational Diabetes Mellitus. The Organizing Committee. Diabetes Care 1998;21 (suppl 2):B161–B167. Schaefer-Graf UM, Kjos SL, Fauzan OH, Uhling KJ, Siebert G, Uhrer CB, Ladendorf B, Dudenhausen JW, Vetter K: A randomized trial evaluating a predominately fetal growth-based strategy to guide management of gestational diabetes in Caucasian women. Diabetes Care 2004;27:297–302. Bonomo M, Cetin I, Pisoni MP, Faden D, Mion E, Taricco E, Nobile de Santis M, Radaelli T, Motta G, Costa M, Solerte L, Morabito A: A flexible approach to the treatment of gestational diabetes modulated on ultrasound evaluation of intrauterine growth: A controlled randomized clinical trial. Diabetes Metab 2004;30:237–244.

Irene Cetin, MD Associate Professor Luigi Mangiagalli Institute of Obstetrics and Gynecology University of Milano, via della Commenda 12 IT–20122 Milano (Italy) Tel. ⫹39 02 503 20265, Fax ⫹39 02 503 20260, E-Mail [email protected]

Cetin/Radaelli

82

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 83–93

Effect of Nutrition on Fetal Development: A View on the Pancreatic ␤-Cells Bertrand Blondeau, Bernadette Breant INSERM U690, Hospital Robert Debré, Paris, France

Fetal growth is a complex, dynamic process taking place during a critical window of development which is dependent on a continuous supply of nutrients from the mother. Epidemiological studies have suggested that there is a strong relationship between fetal growth and the development of metabolic and cardiovascular disorders in adult life. Adults who were small at birth are at increased risk of ischemic heart disease, insulin resistance, glucose intolerance, hypertension and some types of hyperlipidemia, a combination called the metabolic syndrome. The prevalence of the metabolic syndrome has dramatically increased over the last 40 years and the World Health Organization predicts a further increase of more than 100% in the population of the whole world in the next 20 years. The epidemiological association between an adverse intrauterine environment and the later onset of adult diseases led to the new concept of the early-life programming whose mechanisms are still largely unknown. Various hypotheses have been proposed, such as the thrifty genotype, hypothesizing that genes that would favor survival during periods of famine would become detrimental when food supply is abundant [1]. Hales and Barker [2] proposed the thrifty phenotype hypothesis: a shortage of nutrients makes the fetus to adapt by diverting the nutrients to critical organs such as the brain at the expense of other organs such as the liver and pancreas. These changes may, however, prove detrimental if the organism is supplied with normal or supranormal nutrition later on. The association could also be the consequence of interactions between detrimental environmental factors and a genetic susceptibility [3], such as illustrated in studies performed on mono- and dizygotic twins [4]. To help dissecting the mechanisms underlying these human observations, various animal models of intrauterine growth retardation (IUGR) have been

created, mostly in rats and sheep. Here, we will concentrate on various models developed in the rat, a convenient small animal model. We first review the impact that these manipulations have on fetal and organ growth and in a second part we will focus on the influence of nutrition on fetal pancreas development in the same models and the late onset metabolic consequences at adult age.

Animal Models for Studying the Impact of Nutrition on Fetal Growth

The maternal capacity to supply nutrients and their placental transfer to the fetus are major determinants for optimal fetal growth. Glucose, the main substrate, cannot be synthesized by the fetus and can only be provided by the mother. The maternal metabolic condition is, therefore, critical to ensure proper fetal development. It is easily understandable that any perturbations, either higher glucose concentrations due to maternal diabetes, or hypoglycemia resulting from under- or malnutrition, would perturb fetal growth. Maternal Diabetes Glucose, while necessary for a proper development, can affect fetal growth when delivered in excess to the fetus. Models of maternal hyperglycemia have been generated using either streptozotocin (STZ), a diabetogenic drug that destroys the ␤-cells [5] or chronic glucose infusion [6], and the alterations of moderate or high maternal glucose concentrations have been studied. Mildly Diabetic Mothers STZ-induced diabetes induces dose-dependent consequences: fetuses from mildly diabetic mothers are hyperinsulinemic and macrosomic [5]. Interestingly, morphometric measurements showed a decreased size of neuronal nuclei and cytoplasm within the paraventricular and ventromedial hypothalamic nuclei in fetuses from mildly diabetic mothers [7], suggesting that possible dysfunctions of hypothalamic regulators of body weight and metabolism might contribute to the increased risk of developing diabetes and obesity later in life. Severely Diabetic Mothers High doses of STZ induce IUGR in the offspring [5]. At term, placentomegaly due to the maintenance of elevated protein synthetic rates is observed; growth of the liver and the skin appeared to be suppressed in proportion to fetal weight, whereas the lungs and brain were protected from growth retardation [8]. Despite normal size, lungs were immature [9], likely because the alveolar cells were less responsive to EGF [10]. Circulating insulin levels are very low and

Blondeau/Breant

84

insulin receptors are decreased in liver and lungs in fetuses from severely diabetic mothers [11]. Both after STZ-induced or glucose-infusion-induced diabetes, kidney development was impaired with reduced nephron numbers [12]. Additionally, recent morphometric data showed cardiac hypertrophy in the fetuses carried by diabetic mothers [13]. Both types of neonates (hypo- or hyperinsulinemic, IUGR or macrosomic, respectively) have an increased risk of becoming diabetic in adulthood. Pancreas development and the consequences for glucose homeostasis will be discussed later in this review. Fetal Malnutrition Fetal malnutrition in the rat can usually be of two different types affecting either maternal nutrition or the transfer of nutrients to the fetus. While in overall maternal undernutrition the total amount of food is reduced, malnutrition affects specifically one or more nutrients and vascular alterations such as uterine artery ligation reduce nutrient transfer. Maternal Malnutrition: The Low Protein Rat Model A maternal low protein (LP) diet given throughout pregnancy (isocaloric 8% protein instead of 20%) slightly decreased body weight in the offspring at birth [14] and profoundly affects the amino acid profile, especially the taurine levels, both in the maternal and fetal plasma [15]. Offspring of proteinmalnourished dams have permanent alterations in the key insulin-sensitive hepatic enzymes of glycolysis (decreased glucokinase activity) and gluconeogenesis (increased phosphoenolpyruvate carboxykinase activity) associated with a modified zonation of these enzymes in the liver [16]. Kidney weight was reduced in LP neonates and associated with reduced nephron numbers [17], an alteration which persists until adult age [18] and leads to hypertension, despite the introduction of a normal postnatal diet. The vascular structure of various organs is also altered in the offspring of LP-fed rat dams: pancreatic islets and cerebral cortex show a lower blood vessel density, a deficit which is irreversible at adult age in the brain, whether or not a normal diet was given after birth [14]. Interestingly, the use of a similar LP restriction diet in the mouse has recently been shown to increase life span when given postnatally during lactation but to reduce it when applied only during pregnancy [19]. During early adult life, the LP offspring have a better glucose tolerance compared to controls. However, they show a greater age-dependent loss of glucose tolerance than controls such that by 15 months of age they have an impaired glucose tolerance and by 17 months of age they have diabetes associated with insulin resistance [20].

Nutrition and Fetal Pancreas Development

85

Maternal Undernutrition: The Generally Food-Restricted Rat Maternal food restriction (50%) from day 14.5 of pregnancy resulted in IUGR in the offspring [21]. Neonates with moderate IUGR (birth weight between 2.5th and 10th percentile) normally nourished in the postnatal life showed normal body and organ weight and normal insulin contents in adulthood. Severely IUGR offspring (body weight below the 2.5th percentile) normally nourished in the postnatal life also rapidly recovered normal body and pancreatic weight, but liver and kidney weight were significantly reduced at adult age. Malnutrition until weaning in severely IUGR offspring induced deep growth retardation (50%) in body and organ weight at weaning, as well as alterations in hypothalamopituitary-adrenal (HPA) axis activity and delayed the onset of puberty [22]. Hence, maternal undernutrition profoundly activates the maternal HPA axis, leading to increased transplacental transfer of corticosterone which secondarily affects the fetal HPA axis [23]. Although pancreatic weight recovered at adult age, body, liver and kidney weight were irreversibly affected, despite several months of normal nutrition [21]. Furthermore, severe IUGR at birth resulted in decreased insulin content at adult age, irrespective of postnatal nutrition. Another model of severe maternal food restriction (⫺70%) during the entire pregnancy with normal nutrition from birth onwards, developed in the rat by Vickers et al. [24] leads to hyperphagia, hyperinsulinemia, hyperleptinemia and hypertension in the adult offspring, as a consequence of altered fetal development. Similarly to LP offspring, the consequences of general food restriction (GFR) for glucose metabolism are quite mild in young adults but they are clearly evidenced under conditions where there is an increased need for insulin such as ageing [25] and gestation [26] and the animals become intolerant to glucose. No signs of insulin resistance could be observed in this animal model, at least until 1 year of age.

Fetal Nutrition and Pancreas Development

Great attention has recently been given to the influence of fetal nutrition on pancreas development. This interest originates in studies that have suggested that diabetes in adults could originate from altered development of pancreatic ␤-cells in situations of disturbed fetal environment. In 1992, Hales and Barker [2] proposed in the thrifty phenotype hypothesis that poor development of the ␤-cells in utero, consecutive to under- or malnutrition, could be the primary event that could permanently program the onset of diabetes at adult age. This hypothesis was particularly attractive, given the extraordinary growth of the

Blondeau/Breant

86

10,000,000 1,000,000

␤-cell numbers

100,000 10,000 1,000 100 10 Birth 1 E14

E21

P7

P21

P90

P360

Fig. 1. ␤-cell numbers increase dramatically during late fetal life. They increase 10,000fold during the last third of fetal life in the rat but only 10-fold from birth till adult age.

␤-cells in fetal life (fig. 1). We measured ␤-cell growth in the rat from early embryonic stages to adulthood and showed that the most crucial period is the last week of fetal development when ␤-cells grow from 50 at embryonic day (E) 14 to 1,000,000 in the neonate, an expansion that represents a 20,000-fold increase within 1 week (fig. 1). The growth is much slower during postnatal life since it only increases 9 times between birth and 1 year of age. Such a formidable increase in ␤-cell numbers during fetal life together with the known effects of nutrients (glucose and amino acids) on islet growth in vitro strengthened the possibility that a perturbed nutritional environment in utero would impair fetal ␤-cell growth. To complement the studies done on humans, we and others have developed different animal models of altered fetal environment and studied the effects on pancreatic development. Diabetes during Pregnancy: The Consequences for the Fetal Pancreas Mild Hyperglycemia Fetuses of mildly hyperglycemic females have more ␤-cells and an increased insulin secretion in response to glucose, both in vivo and in vitro [27]. However, the offspring of mildly diabetic rats recover from the disturbed fetal environment since they exhibit a normal mass of endocrine tissue and a normal repartition of the islet cells at adult age [28]. Yet, their insulin secretion is reduced and they show an impaired glucose tolerance [27, 28]. Interestingly, the offspring of mildly diabetic rats show increased glucose levels during their

Nutrition and Fetal Pancreas Development

87

subsequent pregnancy, an abnormal milieu that leads to macrosomia, islet hypertrophy and hyperinsulinemia for their fetuses. When they reach adulthood, their offspring also show impaired glucose tolerance [28]. Severe Hyperglycemia Fetuses of severely diabetic mothers present fetal growth retardation as well as hyperglycemia and hypoinsulinemia [5]. Pancreatic endocrine mass is increased [5] but ␤-cells are overstimulated and show degranulation under electron microscopy [5, 27]. In the offspring of severely diabetic mothers, ␤-cells do not secrete insulin in response to glucose and other secretagogues [6, 27], except for arginine that could induce a sustained monophasic secretion [29]. Offspring of severely diabetic mothers at adult age present more ␤-cells because of increased numbers of small islets in the pancreas [30]. This suggested that neogenesis rather than ␤-cell replication was responsible for the observed increased mass. Insulin secretion was found to be increased both in vivo and in vitro [28]. Clamp studies showed insulin resistance in the offspring of severely diabetic mothers, both from the liver and skeletal muscle [31, 32]. During a subsequent pregnancy, the offspring of severely diabetic mothers show glucose intolerance. Effects of Fetal Malnutrition on the Endocrine Pancreas Bilateral Uterine Artery Ligation The rat model of uteroplacental insufficiency induced by bilateral uterine artery ligation leads to a limited supply of food, hormones and growth factors to the fetuses and results in IUGR. In the offspring, ␤-cell mass is normal during the first weeks of life but is then reduced by 7 weeks of age, the time at which body fat mass starts to increase, and from 4 months of age the animals are overtly diabetic and obese. When newborns were exposed to exendin-4, a GLP-1 analog that is known to stimulate ␤-cell neogenesis, ␤-cell mass reduction and diabetes were prevented [33]. Therefore, this model showed that the acquired defects can be prevented by stimulating ␤-cell growth postnatally. LP Diet Fetal protein deprivation is achieved by feeding pregnant female rats an isocaloric LP diet that contains 8% of protein instead of 20% in the control diet. LP fetuses present a reduced birth weight and altered structure and function of the endocrine pancreas characterized by a reduced mean islet size and islet cell proliferation, associated with a reduced density of the blood vessels and an increased apoptotic rate in the islets [14, 34]. The latter was associated with a lower expression of the antiapoptotic factors IGF-I and IGF-II in islets of LP fetuses [35]. Moreover, insulin secretion in response to secretagogues was

Blondeau/Breant

88

impaired in LP islets [34]. Thorough examination of the plasma amino acid profiles in LP mothers and fetuses showed that taurine was dramatically reduced. Supplementation of taurine in the maternal LP diet was sufficient to restore insulin secretion of cultured fetal islets but also islet cell proliferation, a normal apoptotic rate, and to normalize islet cell mass [36]. These crucial studies have thus demonstrated that the effect of an LP diet on islet function and structure was mainly mediated by taurine deficiency. GFR and Glucocorticoids To determine the effects of malnutrition of fetal pancreas development, we have developed a model of GFR based on a 50% food restriction during the last week of pregnancy in female rats. Offspring of underfed mothers have a decreased body weight at birth and an impaired ␤-cell mass due to fewer islets, indicating that differentiation from precursor to differentiated ␤-cells might be impaired [37]. When the restriction is maintained until 21 days of age, ␤-cell mass and islet number deficiency persist at weaning and in adult animals, in spite of their free access to food from weaning. ␤-cell proliferation was never affected at any stage in the GFR animals [38]. The offspring of underfed mothers then showed a lack of pancreatic endocrine mass adaptation during a subsequent pregnancy at the age of 8 months [26], a nonadaptation that was associated with an altered pancreatic development of the second generation fetuses [39]. Careful analysis of the ␤-cell mass during the period of malnutrition revealed that ␤-cell numbers at E17 were not different between GFR and control fetuses (fig. 2). However, from E19 till postnatal day 21, ␤-cell numbers were always lower in GFR animals compared to controls. We then sought to determine the mechanisms by which fetal malnutrition could regulate ␤-cell mass in utero. Since maternal undernutrition and fetal overexposure to glucocorticoids (GC) both give IUGR and lead to glucose intolerance in adulthood, we investigated whether the negative effects of these two abnormalities on fetal ␤-cell development were linked. We had observed that during the fetal period of malnutrition, the levels of both maternal and fetal GC were abnormally elevated. We next showed that the GC elevation was, for a great part, responsible for the decreased ␤-cell mass observed after malnutrition, since prevention of the maternal corticosterone elevation restored fetal ␤-cell mass and islet numbers to nearly normal levels in undernourished fetuses [40]. Moreover, fetal GC overexposure independently of malnutrition was also associated with decreased ␤-cell numbers (fig. 3, DEX and CARB) while a decreased GC fetal exposure led to increased ␤-cell numbers (fig. 3, ADX-MET). Interestingly, the variation in islet numbers followed those of ␤-cell numbers in the various groups (data not shown), indicating that GC could modulate pancreatic cell differentiation. All together, these findings allowed us to characterize GC hormones as inhibitory regulators of ␤-cell development and differentiation in the

Nutrition and Fetal Pancreas Development

89

␤-cell numbers · 106

5

**

4 3

2

1 0.5 0.1 0 E17 E21 E19 Birth

P4

P14 P7

P21

Age (days)

Fig. 2. Evolution of ␤-cell numbers during the undernutrition period in the GFR rat. Maternal undernutrition started at E14.5 and lasted until postnatal day 21 (P21). From E19 onwards, ␤-cell numbers are significantly reduced in GFR fetuses and neonates ( ) compared to controls (––––). **p ⬍ 0.01 (ANOVA) between GFR and controls.

1·106

** 0.8·106

0.6·106

0.4·106

**

**

*

0.2·106

0

Co

GFR

DEX

Maternal nutrition

N

50%

N

Maternal GC levels

N

CARB ADX-MET N

N

Fig. 3. Fetal ␤-cell numbers depend on GC levels. In all conditions of overexposure to GC such as maternal GFR, dexamethasone treatment (DEX) and inhibition of the 11␤-HSD2 with carbenoxolone (CARB), fetal ␤-cell numbers are significantly decreased; under conditions of underexposure to GC observed in fetuses of adrenalectomized and metyraponetreated mothers (ADX-MET), fetal ␤-cell numbers are dramatically increased. *p ⬍ 0.05; **p ⬍ 0.01 (ANOVA) vs. controls. N ⫽ Normal; Co ⫽ control.

Blondeau/Breant

90

Table 1. Postnatal consequences of IUGR in rats GFR

LP

Ut. Art. Lig.

30% UN

Birth Body weight, % of control

⫺15 to 20

⫺15 to 20

⫺15 to 20

⫺30

␤-cell mass, % of control

↓

↓

ND

ND

Adult (3–4 months) Body weight

↓

↓

↑, obesity

↑, obesity, hyperphagia

Fasting glycemia Fasting insulin

Normal ↓

Normal ↓

↑ ↑

Normal ↑

Glucose tolerance

Normal

Normal

IGT, diabetes IR

IR

␤-cell mass

↓

↓

↓

ND

Metabolic disorders

IGT, ↓insulin

IGT, IR, diabetes

Worsening

Obesity, hypertension, hypercaloric diet worsens

␤-cell mass

↓↓

↓

Nonadaptation to pregnancy

Yes

Yes, IR

ND

ND

Reference No.

25, 26, 37–40

14–16, 20, 34–36

33

24

Adult (12–15 months)

ND

The postnatal consequences for body weight, glucose homeostasis and ␤-cell mass are compared in the GFR (50% food intake during the last third of pregnancy and throughout lactation), LP (8% isocaloric instead of 20% protein), 30% undernourished (UN; 30% food intake throughout pregnancy) and uterine artery ligation from day 19 of pregnancy (Ut. Art. Lig.) rat models. ␤-cell mass decrease in adult offspring from uterine arteryligated rat dams is observed relative to body weight. ND ⫽ Not determined; IGT ⫽ impaired glucose tolerance; IR ⫽ insulin resistance.

rat. Relevant information concerning the different models and their characteristic effects on ␤-cell mass is summarized in table 1. Conclusion

In utero events have long-lasting consequences for the homeostasis of glucose regulation and insulin sensitivity in adult life. Despite some apparent discrepancies in the cellular mechanisms that take place in the pancreas in the various models, it seems that, in the rat, an early injury of pancreatic ␤-cell development is the primary event leading to impaired glucose tolerance or diabetes in adult

Nutrition and Fetal Pancreas Development

91

life: different events (gestational diabetes, LP diet or undernutrition) involving different molecules (maternal glucose elevation, specific amino acid deficiency, GC elevation) and different routes (differentiation, ␤-cell proliferation or apoptosis) lead to the same end point, which is fewer differentiated ␤-cells at birth. Whether one of these schemes also applies to human fetal growth restriction remains to be determined.

References 1 2 3 4 5 6 7 8 9 10 11

12 13 14 15

16 17 18 19 20

Neel JV: Diabetes mellitus: A ‘thrifty’ genotype rendered detrimental by ‘progress’? Am J Hum Genet 1962;14:353–362. Hales CN, Barker DJ: Type 2 (non-insulin-dependent) diabetes mellitus: The thrifty phenotype hypothesis. Diabetologia 1992;35:595–601. Jaquet D, Tregouet DA, Godefroy T, et al: Combined effects of genetic and environmental factors on insulin resistance associated with reduced fetal growth. Diabetes 2002;51:3473–3478. Poulsen P, Vaag A, Beck-Nielsen H: Does zygosity influence the metabolic profile of twins? A population based cross sectional study. BMJ 1999;319:151–154. Aerts L, van Assche FA: Rat foetal endocrine pancreas in experimental diabetes. J Endocrinol 1977;73:339–346. Bihoreau MT, Ktorza A, Kinebanyan MF, Picon L: Impaired glucose homeostasis in adult rats from hyperglycemic mothers. Diabetes 1986;35:979–984. Plagemann A, Harder T, Janert U, et al: Malformations of hypothalamic nuclei in hyperinsulinemic offspring of rats with gestational diabetes. Dev Neurosci 1999;21/1:58–67. Canavan JP, Goldspink DF: Maternal diabetes in rats. II. Effects on fetal growth and protein turnover. Diabetes 1988;37:1671–1677. Tyden O, Berne C, Eriksson U: Lung maturation in fetuses of diabetic rats. Pediatr Res 1980;14: 1192–1195. Thulesen J, Poulsen SS, Nexo E, Raaberg L: Epidermal growth factor and lung development in the offspring of the diabetic rat. Pediatr Pulmonol 2000;29:103–112. Mulay S, McNaughton L: Fetal lung development in streptozotocin-induced experimental diabetes: Cytidylyl transferase activity, disaturated phosphatidyl choline and glycogen levels. Life Sci 1983;33:637–644. Amri K, Freund N, Vilar J, Merlet-Benichou C, Lelievre-Pegorier M: Adverse effects of hyperglycemia on kidney development in rats: In vivo and in vitro studies. Diabetes 1999;48:2240–2245. Menezes HS, Barra M, Bello AR, Martins CB, Zielinsky P: Fetal myocardial hypertrophy in an experimental model of gestational diabetes. Cardiol Young 2001;11:609–613. Snoeck A, Remacle C, Reusens B, Hoet JJ: Effect of a low protein diet during pregnancy on the fetal rat endocrine pancreas. Biol Neonate 1990;57/2:107–118. Reusens B, Dahri S, Snoeck A, Bennis-Taleb N, Remacle C, Hoet J: Long-Term Consequences of Diabetes and Its Complication May Have a Fetal Origin: Experimental and Epidemiological Evidences. New York, Raven Press, 1995. Desai M, Byrne CD, Zhang J, Petry CJ, Lucas A, Hales CN: Programming of hepatic insulin-sensitive enzymes in offspring of rat dams fed a protein-restricted diet. Am J Physiol 1997;272:G1083–G1090. Merlet-Benichou C, Gilbert T, Muffat-Joly M, Lelievre-Pegorier M, Leroy B: Intrauterine growth retardation leads to a permanent nephron deficit in the rat. Pediatr Nephrol 1994;8/2:175–180. Langley-Evans SC, Welham SJ, Jackson AA: Fetal exposure to a maternal low protein diet impairs nephrogenesis and promotes hypertension in the rat. Life Sci 1999;64:965–974. Ozanne SE, Hales CN: Lifespan: Catch-up growth and obesity in male mice. Nature 2004;427: 411–412. Petry CJ, Dorling MW, Pawlak DB, Ozanne SE, Hales CN: Diabetes in old male offspring of rat dams fed a reduced protein diet. Int J Exp Diabetes Res 2001;2/2:139–143.

Blondeau/Breant

92

21 22

23

24

25 26

27 28 29 30 31 32

33 34 35

36

37 38 39

40

Garofano A, Czernichow P, Breant B: Postnatal somatic growth and insulin contents in moderate or severe intrauterine growth retardation in the rat. Biol Neonate 1998;73/2:89–98. Leonhardt M, Lesage J, Croix D, Dutriez-Casteloot I, Beauvillain JC, Dupouy JP: Effects of perinatal maternal food restriction on pituitary-gonadal axis and plasma leptin level in rat pup at birth and weaning and on timing of puberty. Biol Reprod 2003;68/2:390–400. Lesage J, Blondeau B, Grino M, Breant B, Dupouy JP: Maternal undernutrition during late gestation induces fetal overexposure to glucocorticoids and intrauterine growth retardation, and disturbs the hypothalamo-pituitary adrenal axis in the newborn rat. Endocrinology 2001;142:1692–1702. Vickers MH, Breier BH, Cutfield WS, Hofman PL, Gluckman PD: Fetal origins of hyperphagia, obesity, and hypertension and postnatal amplification by hypercaloric nutrition. Am J Physiol Endocrinol Metab 2000;279/1:E83–E87. Garofano A, Czernichow P, Breant B: Effect of ageing on beta-cell mass and function in rats malnourished during the perinatal period. Diabetologia 1999;42:711–718. Blondeau B, Garofano A, Czernichow P, Breant B: Age-dependent inability of the endocrine pancreas to adapt to pregnancy: A long-term consequence of perinatal malnutrition in the rat. Endocrinology 1999;140:4208–4213. Kervran A, Guillaume M, Jost A: The endocrine pancreas of the fetus from diabetic pregnant rat. Diabetologia 1978;15:387–393. Aerts L, Holemans K, Van Assche FA: Maternal diabetes during pregnancy: Consequences for the offspring. Diabetes Metab Rev 1990;6/3:147–167. Bihoreau MT, Ktorza A, Picon L: Gestational hyperglycaemia and insulin release by the fetal rat pancreas in vitro: Effect of amino acids and glyceraldehyde. Diabetologia 1986;29:434–439. Aerts L, Vercruysse L, Van Assche FA: The endocrine pancreas in virgin and pregnant offspring of diabetic pregnant rats. Diabetes Res Clin Pract 1997;38/1:9–19. Holemans K, Van Bree R, Verhaeghe J, Aerts L, Van Assche FA: In vivo glucose utilization by individual tissues in virgin and pregnant offspring of severely diabetic rats. Diabetes 1993;42:530–536. Ryan EA, Liu D, Bell RC, Finegood DT, Crawford J: Long-term consequences in offspring of diabetes in pregnancy: Studies with syngeneic islet-transplanted streptozotocin-diabetic rats. Endocrinology 1995;136:5587–5592. Stoffers DA, Desai BM, DeLeon DD, Simmons RA: Neonatal exendin-4 prevents the development of diabetes in the intrauterine growth retarded rat. Diabetes 2003;52:734–740. Dahri S, Snoeck A, Reusens-Billen B, Remacle C, Hoet JJ: Islet function in offspring of mothers on low-protein diet during gestation. Diabetes 1991;40(suppl 2):115–120. Petrik J, Reusens B, Arany E, et al: A low protein diet alters the balance of islet cell replication and apoptosis in the fetal and neonatal rat and is associated with a reduced pancreatic expression of insulin-like growth factor-II. Endocrinology 1999;140:4861–4873. Boujendar S, Reusens B, Merezak S, et al: Taurine supplementation to a low protein diet during foetal and early postnatal life restores a normal proliferation and apoptosis of rat pancreatic islets. Diabetologia 2002;45:856–866. Garofano A, Czernichow P, Breant B: In utero undernutrition impairs rat beta-cell development. Diabetologia 1997;40:1231–1234. Garofano A, Czernichow P, Breant B: Beta-cell mass and proliferation following late fetal and early postnatal malnutrition in the rat. Diabetologia 1998;41:1114–1120. Blondeau B, Avril I, Duchene B, Breant B: Endocrine pancreas development is altered in foetuses from rats previously showing intra-uterine growth retardation in response to malnutrition. Diabetologia 2002;45:394–401. Blondeau B, Lesage J, Czernichow P, Dupouy JP, Breant B: Glucocorticoids impair fetal beta-cell development in rats. Am J Physiol Endocrinol Metab 2001;281:E592–E599.

Bernadette Breant, MD INSERM U690, Hospital Robert Debré 48 Boulevard Sérurier, FR–75019 Paris (France) Tel. ⫹33 1 40 03 19 85, Fax ⫹33 1 40 40 91 95 E-Mail: [email protected]

Nutrition and Fetal Pancreas Development

93

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 94–109

The Human Placenta in Diabetes Gernot Desoyea, Peter Kaufmannb a

Clinic of Obstetrics and Gynaecology, Medical University of Graz, Graz, Austria; Department of Anatomy II, University of Technology Aachen, Aachen, Germany

b

The placenta is a complex and so far poorly understood organ, which plays the central metabolic role in pregnancy. In addition to synthesizing various hormones it regulates the transport of maternal fuels to the fetus and facilitates maternal metabolic adaptations to different stages of pregnancy [1]. The placenta of the diabetic woman has attracted much interest largely because it is thought that placental damage may be partially responsible for the unduly high incidence of fetal complications that occur in pregnancies complicated by diabetes mellitus [2]. Owing to its position between mother and fetus the placenta is exposed to metabolic endocrine derangements in both circulations. Intuitively, influences of the diabetic environment on the cells that are in contact with the maternal (syncytiotrophoblast) and fetal (endothelium) circulation may be expected, because of the presence of receptors, transporters, ion channels and other molecules on both placental surfaces. The central question has been whether the placenta adapts to the diabetic environment with the ultimate result of protecting the fetus from the adverse diabetic environment or whether the placenta contributes to the adverse fetal outcome associated with diabetic pregnancies despite improvement in the care of diabetic women. Placental development is a complex process of various coordinated differentiation steps that are mostly completed at the end of the second trimester. Thereafter placental growth is predominantly characterized by mass expansion. Thus development of the placenta precedes fetal development and growth, the latter being pronounced in the third trimester. Any insult of the maternal diabetic environment during the critical period of placental differentiation in the first and second trimester will likely result in placental changes that may have a profound effect on subsequent fetal growth. In contrast, any insult on the

placenta in the course of the third trimester will exert only gross effects on placental and fetal growth.

First Trimester Placenta

The effect of diabetes on the placenta at the beginning of pregnancy, i.e. in the first trimester, has not been studied yet. However, there are several lines of indirect evidence suggesting profound alterations of placental development already at this stage in pregnancy. The incidence of pregnancy complications that have been associated with inadequate placentation, in particular with trophoblast invasion into the maternal decidua, is increased in diabetic pregnancies. These include miscarriage, intrauterine growth restriction and pre-eclampsia. An influence of the quality of maternal metabolic control can be seen, but it is unclear whether hyperglycaemia itself is the causative factor, or whether insulin [3], leptin [4], isoprostanes [5] or others lead to this increased proportion of diabetes-associated invasion defects. Maternal levels of the placenta-derived placental lactogen are lower in the first trimester of a diabetic pregnancy, whereas those of an endometrium-derived hormone, PP14, are normal [6]. Since placental lactogen levels are determined by placental, i.e. trophoblast mass, an impaired trophoblast development can be inferred. This view is also supported by animal data and the clinical observation of growth retardation early in some diabetic pregnancies [7, 8].

Term Placenta

At term the placenta of a diabetic mother shows a number of variations as compared to placentas from normal non-diabetic mothers. These range from changes in morphology, blood flow, transport, metabolism, growth regulation to steroid synthesis; they include indications of pathology and oxidative stress. Because of space limitations this chapter will focus on morphology and development as well as on oxidative stress, since both may be interrelated to some extent [9]. The interested reader is referred to other reviews on the general subject [10–15]. Morphology and Development Despite improvement of the metabolic control of diabetic pregnancies over the past decades, a tendency towards heavier placentas in type 1 and gestational diabetes has been noticed, especially associated with heavier fetuses [16–18].

The Human Placenta in Diabetes

95

10 Amount in total placenta

Normal pregnancy GDM Type 1

*

8

* 6

Lipids

4

2

* *

**

**

0 DNA (g)

Glycogen (g) TG (nMol)

PL (g)

Cholesterol (g)

Fig. 1. Placental composition in normal, gestational diabetic (GDM) and pregestational type 1 diabetic pregnancies [data from 20]. TG ⫽ Triacylglycerols; PL ⫽ phospholipids. *p ⬍ 0.05 vs. normal pregnancy.

The occurrence of placentomegaly as a result of an increase in parenchymal tissue cellularity, reflected by higher DNA content [19], in cases of increased weight of the neonate confirms the close correlation of placental weight with that of the offspring. Such placentas also contain more triglycerides and phospholipids than normal placentas (fig. 1) [19, 20]. The morphological and histological placental changes were studied in gestational and type 1 diabetes with inconsistent and partly even controversial results. In view of the heterogeneity of ‘diabetes in pregnancy’ with regard to its nature and cause, modality of treatment and quality of glycaemic control, the relatively small sample sizes in the studies may have been the reason for the discrepancy. Other reasons may include the quality of the pathological assessment and differences in methodology, especially in morphometric studies. The literature has been reviewed comprehensively [21–24] and no further revelations have been made recently. If not otherwise stated, the structural features summarized below apply to placental villi in type 1 diabetes with poor metabolic control. In type 1 diabetes with improved to perfect metabolic control, despite the occasional appearance of placento- and fetomegaly the incidence of villi with typical ‘diabetic’ malformation is low and the majority of villi looks normal. The same is valid for gestational diabetes. Even in the severest cases of type 1 diabetes with poor metabolic control, only a certain percentage of villi – obviously mature intermediate villi as will be discussed later – reveal a characteristic ‘diabetic’ phenotype whereas stem villi and the majority of terminal villi are more or less inconspicuous.

Desoye/Kaufmann

96

100 ␮m

a

100 ␮m

Fig. 2. Serial paraffin sections of term placental villi in metabolically poorly controlled type 1 diabetes. Note the presence of a large paw-shaped villus, typical for diabetes mellitus. The stroma in both sections is composed of an unusually dense non-oriented population of large connective tissue cells (centre). a Staining with a cytokeratin-7 antibody preferably marks villous cytotrophoblast (red). Note that in this case nearly 100% of the villous surface is covered by cytotrophoblast (as compared to 14% in normal term villi). b Parallel section of that depicted in a, stained with the antibody Qbend10 which selectively stains endothelial cells. Note the presence of numerous fetal small-calibre capillaries (red) which are homogeneously distributed across the villous cross section (as opposed to a preferential superficial subtrophoblastic location in normal villi of comparable calibre). ⫻350 [courtesy of Dr. Mahmed Kadyrov and Dr. Berthold Huppertz, Aachen and Dr. Josip Djelmis, Dr. Marina Ivanisevic and Dr. Marina Kos, Zagreb].

The characteristic ‘diabetic phenotype’ of villi as mentioned above comprises the following features: (1) The villi are unusually large for non-stem villi in term placentas, exhibiting large diameters of several hundred micrometers. They are typically paw-shaped: the oval to roundish cross section on one side shows multiple surface indentations resulting in stubby finger-shaped side branches. The majority of the latter exhibit the same structural features as the parent villus, a few others may be normal terminal villi (fig. 2a, b). (2) The above-mentioned increase in parenchymal cellularity expressed by higher DNA content [19, 20] is histologically reflected by increased numerical density of nearly all cellular constituents of these villi (see fig. 2a, b and summarizing diagram in fig. 3). These include: villous stromal fibroblasts (fig. 2a, b, 3), villous macrophages [25; our own findings], endothelial cells [26; our own findings] (fig. 2b), villous cytotrophoblast [27–29] (see also fig. 2a), and even the number of syncytiotrophoblast nuclei seems to be increased, possibly as a consequence of increased syncytial fusion out of the rich stocks of villous cytotrophoblasts. Thickening of the syncytiotrophoblast and local accumulation of syncytial nuclei (syncytial knotting) [30] can be interpreted as the consequence of increased trophoblast fusion.

The Human Placenta in Diabetes

97

b

2nd trimester normal Syncytiotrophoblast Cytotrophablast Fetal capillary Fibroblast Basal lamina

Macrophage (Hofbauer cell) Fetal sinusoid 3rd trimester normal

3rd trimester diabetes mellitus

Fig. 3. Villous morphology in normal (left two villi) as compared to diabetic pregnancy (right villus). Placental villi from severe, poorly controlled type 1 diabetes mellitus differ from normal terminal villi by (1) larger size, (2) paw-shaped appearance, (3) thicker syncytiotrophoblast (grey), (4) increased numerical density of villous cytotrophoblast (light grey), (5) increased numerical density of stromal fibroblasts (green), (6) increased numerical density of macrophages (lilac), (7) increased numerical density of villous capillaries (blue), (8) reduced luminal width of villous capillaries, (9) homogeneous distribution rather than subtrophoblastic distribution of villous capillaries, and (10) thicker trophoblastic and endothelial basal laminas (green). In spite of large numbers of cytotrophoblasts and macrophages and large size, villi from placentas in diabetes mellitus cannot be considered simply immature; rather they differ from typical immature intermediate villi from the second trimester (upper left) by more numerous and more homogeneously distributed capillaries, by diffuse, highly cellular as compared to reticularly arranged connective tissue and by an increased size of fibroblastic nuclei.

(3) The most characteristic feature of ‘diabetic villi’ is the impressive degree of capillarization [25, 31–33] that may locally achieve angioma-like degrees (chorangiosis, chorangioma, chorangiomatosis) [34, 35]. This, however, does not necessarily result in an enlargement of the maternofetal exchange area and in improved maternofetal exchange conditions. The reasons are 3-fold: • The capillaries have mostly smaller diameters than normal (fig. 2b, 3) resulting in a reduced share in villous volume [33, 36]; the only contrasting publication [25] obviously does not reflect the usual picture.

Desoye/Kaufmann

98

•

Together with the number of capillary cross sections the villous diameter also increases [37–39], thus ‘diluting’ the numerical density of the capillaries. • Last but not least, different from similarly sized normal placental villi, the capillaries are not located in a peripheral position, just beneath the trophoblastic surface (fig. 3) but rather are homogeneously distributed across the villous cross section (fig. 2b, 3) resulting in rather long and inefficient maternofetal diffusion distances. Whether or not the total maternofetal exchange capacity is increased depends on whether or not the diabetes-typical mass increase of placental parenchyma can compensate for the loss of exchange capacity of the individual villus [40]. This may well be different for nutrients and respiratory gases. Transport of the former depends on transporter systems which are richly available in the large villous trophoblast surface. By contrast, diffusional transfer of respiratory gases depends on the availability of both, exchange surface and the increased mean maternofetal diffusion distance. (4) The structural features of villous capillarization in diabetes (fig. 2b) suggest the existence of richly branched capillary networks derived from branching angiogenesis [for a review, see 41]. Confocal laser microscopy data support this notion [42]. Branching angiogenesis very likely is driven by high levels of VEGF but only moderate levels of PlGF, a combination known to result from fetoplacental hypoxia [for a review, see 41]. The oxygen-dependent expression of angiogenic growth factor FGF-2 may further support this process [43]. It is a matter of discussion whether such hypoxic conditions are a consequence (1) of impaired maternofetal oxygen transfer, (2) of increased placental oxygen consumption due to an adverse combination of high relative cellularity with high placental mass, or (3) of excessive fetal oxygen demand in the wake of stimulated fetal aerobic metabolism, foremost of glucose. Data by Mayhew and Jairam [44] favour the latter assumption. It is further supported by higher erythropoietin levels in the fetal circulation [45, 46]. The elevated iron demand associated with the resulting stimulated erythropoiesis can be met by maternalfetal transfer of more iron, because placental transferrin expression is increased in diabetes [47]. Excessive branching angiogenesis with resulting structurally complex, but functionally less efficient capillary beds may be a typical feature of the placenta in type 1 diabetes and different from gestational diabetes, where no vascular or endothelial cellular remodelling was found [33, 48]. (5) The trophoblastic basement membrane is reportedly thicker in all types of diabetes [25, 27, 29, 30, 49], which is in agreement with our findings. Only Emmrich et al. [50] and Jirkovska [51] arrived at opposite conclusions which would be more in line with a shorter diffusion distance for oxygen [52–54]. The discrepancy can be resolved on the basis of where within the villus the

The Human Placenta in Diabetes

99

measurements have been made. In fibrotic and in large hypercellular villi (fig. 2a, b) the epithelial and endothelial basal laminae are thickest [50; our own findings], which may result from an increased synthesis or a decreased decomposition of the basal lamina constituents. The thinner basal laminae were observed particularly in small villi and in regions with vasculosyncytial membranes. Whether this is the result of a compensatory growth process with formation of ontogenetically younger villi providing better maternofetal exchange conditions remains to be established [50].Basement membrane thickening may be attributed to increased amounts of collagen, predominantly of collagen type IV. Some collagens, i.e. type IV, V and VI, contain a higher proportion of carbohydrates and it may be conceived that this is due to non-enzymatic glycation. (6) Term placentas in diabetes mellitus repeatedly have been described as showing ‘villous immaturity’ [25, 37–39]. We do not share this view. At first glance, the large ‘diabetic’ villi (fig. 3) show some phenotypic similarities with immature villi: the villous diameter is increased resembling those of immature intermediate villi (fig. 3), the number of macrophages is increased as compared to normal [25], villous cytotrophoblast can be found beneath 50–100% of the syncytiotrophoblast as compared to 14% in normal term villi (fig. 2a) [27–29], and the surface extent and number of vasculosyncytial membranes as preferred diffusional exchange areas are similarly low as in immature placentas [30]. On the other hand, there are important differences when comparing these villi with immature ones: typical ‘diabetic’ villi lack the characteristic reticular stroma of immature intermediate villi composed of stromal channels containing the macrophages, they do not show the typical peripheralization of villous capillaries, and the hypercellular connective tissue is composed of unusually voluminous, undifferentiated fibroblasts rather than of myofibroblasts as this is the case in immature intermediate villi. Moreover, topological analysis of the villous trees reveals that these pathognomonic large-calibre villi are located between largely normal stem villi and rarified but otherwise inconspicuous terminal villi in a position of mature intermediate villi. They clearly are pathologically altered, malformed mature intermediate villi, which due to abnormal local conditions (hypoxia?) have developed excessive branching angiogenesis rather than predominant nonbranching angiogenesis, the characteristic features of mature intermediate villi from 26 weeks postmenstruation onwards. These pathologically altered mature intermediate villi obviously depend in number and size on the type of diabetes and on the quality of metabolic control. Mayhew and Jairam [44] and Mayhew [33] found largely normal villous morphology in cases of good glycaemic control. In agreement with this notion, Bjork and Persson [55] described increased villous abnormalities with increased variation in maternal blood glucose levels up to week 32 of gestation,

Desoye/Kaufmann

100

suggesting a long-term priming effect of maternal hyperglycaemia on placental development [56].

Oxidative and Nitrative Stress of the Placenta in Diabetes Free radicals are constantly produced as reactive oxygen (reactive oxygen species) or nitrogen species. Once generated they are removed by antioxidant defence systems including the enzymes superoxide dismutases (SOD, converting superoxide anion into H2O2), glutathione peroxidase, catalase (converting H2O2 into water) as well as by antioxidant molecules like glutathione and vitamin E. Reactive oxygen species, including nitric oxide and related species, commonly exert a series of useful physiological effects. However, some conditions such as diabetes are associated with an imbalance between pro-oxidant and antioxidant defences in favour of pro-oxidants. The resultant oxidative stress is associated with the oxidative modification of biomolecules such as lipids, proteins, and nucleic acids and will impair their structure and function, thus imposing a threat to the biological systems. Oxidative stress during embryogenesis and in the placenta has been implicated in adverse pregnancy outcome such as birth defects, early pregnancy failure and pre-eclampsia [57–59]. In diabetes, oxygen-free radicals are thought to be produced as a result of prolonged periods of exposure to hyperglycaemia. This will (1) stimulate the mitochondrial electron transport chain to generate higher levels of superoxide [60] and (2) cause non-enzymatic glycation of plasma proteins [61] and increased formation of advanced glycation end products on the fetal aspect of the placenta [62]. These glycated products undergo further spontaneous reaction leading to the production and release of free radicals including superoxide. Superoxide in the absence of appropriate levels of scavengers may lead to an imbalance between pro-oxidants and antioxidants and produce a state of oxidative stress. With one exception [63] that may reflect different ‘severity’ of the diabetes, lower levels of all SOD isoforms and unchanged levels of glutathione peroxidase are reported in diabetes. No information on catalase activity or levels is available (table 1). In contrast to other embryonic tissues [64, 65] in which diabetesassociated generation of free oxygen radicals results in mitochondrial large amplitude swelling, the human trophoblast appears to be resistant to such changes regardless of the gestational age [66, 67], although trophoblast mitochondria can generate free oxygen radicals [67]. Despite evidence of oxidative stress in the placenta in diabetes, trophoblast mitochondria are morphologically normal [66].

The Human Placenta in Diabetes

101

Table 1. Alterations in antioxidant defence systems and marker of oxidative or nitrative stress in the trophoblast under hyperglycaemia and in the human term placenta in diabetes

Antioxidants Total SOD CuZn-SOD (SOD1) Mn-SOD (SOD2) GPx

Term trophoblast under hyperglycaemia

Gestational diabetes mellitus

Type 1 diabetes mellitus

↓1

⫽5 ↑2 ↓3

↓5 ↓3 ↓4

Markers of oxidative and nitrative stress Lipid peroxidation MDA, TBARSa ↑1 8-Isoprostanes Protein damage: protein carbonyls Nitrative stressb

⫽2

↑3 ↑2 ↑2 ↑5

↑3, 5

↑4, 5

1 ⫽ Ref. No. 99, 2 ⫽ Ref. No. 63, 3 ⫽ Ref. No. 100 and 4 ⫽ Ref. No. 77 predominantly in vascular endothelium, and 5 ⫽ Ref. No. 101. SOD ⫽ Superoxide-dismutase; GPx ⫽ glutathione peroxidase. a Measured as malone dialdehyde (MDA) or thiobarbituric acid-reactive (TBAR) substances. b Measured either as nitrotyrosine immunoreactivity [77] or as nitrate/nitrite ratio [101].

Antioxidant Capacity The placenta is equipped with antioxidants, such as SOD, catalase, glutathione peroxidase and glutathione S-transferase [68–74] as well as with ␣-tocopherol, glutathione [71] and ascorbic acid [75]. Activities of catalase and SOD increase throughout gestation whereas those of glutathione peroxidase and the ␣-tocopherol content remain virtually unchanged [68, 71, 76]. In response to increased pro-oxidant challenge, antioxidants may be increased as a compensatory response but then, subsequently, reduced in the chronic state as they are overwhelmed by oxidative agents. In the fetal placental vascular endothelium of diabetic pregnancies the manganese isoform of SOD was overexpressed [77]. Reductions in total radical trapping antioxidant capacities such as the scavenger activity of SOD and catalase, glutathione metabolism and/or vitamin E levels were seen in diabetic patients [78]. This was paralleled by an increase in lipid peroxide levels. Administration of antioxidants, vitamin C, and vitamin E reduced placental lipid peroxidation in the perfused human pre-eclamptic placenta [79].

Desoye/Kaufmann

102

Lipid Peroxidation Because lipids are a major component of living organisms and probably the first easy target of free radicals once they are produced, lipid peroxidation might play an important role in initiating and/or mediating some diabetes-associated changes in the placenta. These may put the fetus at risk, because both placental structure and function may be compromised so that it cannot adequately fulfil its pleiotropic roles related to fetal development. Cytotoxic intermediates of lipid peroxidation [80] may also be released into the umbilical circulation. The placenta itself is exposed to increasing concentrations of lipid peroxides in the maternal serum during the course of pregnancy [81], which at term are about 3-fold higher than those in the fetal circulation [71]. Lipid peroxidation can be induced by a variety of drugs and chemicals, either by an NADPH-dependent peroxidase or non-enzymatically by Fe2⫹ and ascorbic acid [82]. Placental lipids are susceptible to both mechanisms. Whereas NADPH-induced lipid peroxide formation is higher at term than in the first trimester [83], Fe2⫹-induced formation decreases from week 7 onward until term [71]. NADPH-dependent lipid peroxidation occurs in mitochondria and microsomes of human placental tissue [83–88]. In mitochondria, cytochrome P-450 may liberate superoxide ion, which then initiates NADPH-dependent lipid peroxidation [87]. Placental microsomes appear to have a low level of endogenous antioxidants rendering them particularly sensitive to lipid peroxidation [84]. When the small amounts of microsomal P450 [89] are destroyed by peroxidation, the propagation of lipid peroxidation is inhibited [86]. The cyclooxygenase component of prostaglandin H synthase was assumed to be involved in the formation of placental lipid peroxides, because this process can be blocked by aspirin [90]. It is unclear whether this may explain the imbalance of vasoconstricting placental thromboxane and vasodilating prostacyclin in favour of vasoconstriction in diabetes [91]. In normal pregnancies, the antioxidant systems efficiently protect the placenta and fetus from lipid peroxides. Overloading placental antioxidants with highly reactive oxygen intermediates and hydrogen peroxide or inefficient detoxification systems may result in overproduction of lipid peroxides. All available evidence demonstrates a higher activity of placental lipid peroxidation or activities of protecting enzymes in maternal diabetes (table 1). Surprisingly, no information on placental levels of the lipophilic antioxidant vitamin E (␣-, ␥-tocopherol) is available. In non-pregnant diabetics plasma levels of lipid peroxides are increased and the increment is higher in poorly controlled than in well-controlled patients. The elevated lipid peroxide levels may contribute to structural and functional disturbances of the placenta and fetus in this condition [92, 93].

The Human Placenta in Diabetes

103

Nitric Oxide and Superoxide in the Placenta Nitric oxide formation may be elevated by increased enzymatic activity of endothelial or inducible isoforms of nitric oxide synthase in diabetes [94] although this was not a consistent finding [95]. Superoxide can be increased, because of an overactive mitochondrial electron transport chain or generation from glycated proteins. At low concentrations nitric oxide can function as an antioxidant by inhibiting the mitochondrial electron transport chain [67], but nitric oxide is also inactivated by superoxide anion, which, therefore, limits the bioactivity of nitric oxide. This interaction, however, yields peroxynitrite anion (ONOO–) a powerful oxidant of a variety of biomolecules [96]. Peroxynitrite is known to cause lipid peroxidation, and to inhibit the mitochondrial electron transport system, nitrate tyrosine residues; moreover, it oxidizes sulfhydryl groups from proteins, hence altering their activity or disrupting signal transduction pathways [97]. An increased expression of nitrotyrosine residues, which are formed by the interaction of peroxynitrite with tyrosine moieties, was found in the fetal vasculature and villous stroma of placenta in type 1 diabetes [77, 98]. This location suggests local generation of peroxynitrite due to hyperglycaemiaassociated oxidative stress in stroma or diffusion of peroxynitrite produced in vascular endothelium at high concentrations into the underlying stroma. This was found despite good glycaemic control of the diabetic mothers and strongly suggests that optimal control in the fetal compartment was not achieved. The prevalent location of nitrotyrosine in the fetal aspect of the placenta with little formation in the trophoblast facing the maternal circulation can have at least two reasons: (1) the quality of glycaemic control of the mother is good enough not to allow for profound increases in superoxide generation or (2) the antioxidant defences are more efficient on the trophoblast than in the fetal aspects of the placenta. Overall, however, these findings suggested the involvement of peroxynitrite in the pathological processes of diabetic placental injury. It was suggested that lipid peroxides are a cause of diabetes-associated angiopathy [9].

Conclusion

The human placenta undergoes a number of structural changes which ultimately will facilitate the development of the fetus. Oxidative stress occurs in diabetes, particularly in type 1 diabetes. Some evidence suggests that the fetal side of the placenta, especially the vascular endothelium, is more susceptible. This strongly corroborates that the aim of an adequate treatment must be the normalization of the fetal metabolic status, which will not always be achieved by controlling the mother’ disease alone.

Desoye/Kaufmann

104

References 1 2 3

4

5 6

7

8 9 10 11 12

13 14 15 16 17 18 19 20 21 22 23

24 25

Munro HN, Pilistine HJ, Fant ME: The placenta in nutrition. Annu Rev Nutr 1983;3:97–124. Persson B, Geutz J, Lunell NO: Diabetes in pregnancy; in Scorpelli EM, Cosmi EV (eds): Reviews in Perinatal Medicine. New York, Raven Press, 1983, p 1. Shaw LM: Identification of insulin receptor substrate 1 (IRS-1) and IRS-2 as signaling intermediates in the alpha6beta4 integrin-dependent activation of phosphoinositide 3-OH kinase and promotion of invasion. Mol Cell Biol 2001;21:5082–5093. Castellucci M, De Matteis R, Meisser A, Cancello R, Monsurro V, Islami D, Sarzani R, Marzioni D, Cinti S, Bischof P: Leptin modulates extracellular matrix molecules and metalloproteinases: Possible implications for trophoblast invasion. Mol Hum Reprod 2000;6:951–958. Staff AC, Ranheim T, Henriksen T, Halvorsen B: 8-Isoprostaglandin f (2alpha) reduces trophoblast invasion and matrix metalloproteinase activity. Hypertension 2000;35:1307–1313. Pedersen JF, Sorensen S, Molsted-Pedersen L: Serum levels of human placental lactogen, pregnancy associated plasma protein A and endometrial secretory protein PP 14 in first trimester of diabetic pregnancy. Acta Obstet Gynecol Scand 1998;77:155–158. Brown ZA, Mills JL, Metzger BE, Knopp RH, Simpson JL, Jovanovic-Peterson L, Scheer K, Van Allen MI, Aarons JH, Reed GF: Early sonographic evaluation for fetal growth delay and congenital malformations in pregnancies and complicated by insulin-requiring diabetes. Diabetes Care 1992;15:613–619. Robinson J, Canavan JP, el Haj AJ, Goldspink DF: Maternal diabetes in rats. I. Effects on placental growth and protein turnover. Diabetes 1988;37:1665–1670. Skrha J: Pathogenesis of angiopathy in diabetes. Acta Diabetol 2003;40(suppl 2):S324–329. Desoye G, Shafrir E: Placental metabolism and its regulation in health and diabetes. Mol Aspects Med 1994;15:505–682. Desoye G, Shafrir E: The human placenta in diabetic pregnancy. Diabet Rev 1996;4:70–89. Desoye G, Hauguel-de Mouzon S, Shafrir E: The placenta in diabetic pregnancy; in Hod M, Jovanovic L, Di Renzo GC, De Leiva A, Langer O (eds): Diabetes and Pregnancy. London, Martin Dunitz, 2003, pp 126–147. Desoye G, Myatt L: The placenta; in Reece EA, Coustan DR, Gabbe SG (eds): Diabetes in Women, ed 3. Philadelphia, Lippincott, 2004, pp 147–157. Leach L, Mayhew TM: Vasculogenesis and angiogenesis in the diabetic placenta; in Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Basel, Karger, 2005, pp 110–126. Kos M, Vogel M: Morphological findings in infants and placentas of diabetic mothers; in Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Basel, Karger, 2005, pp 127–143. Clarson C, Tevaarwerk GJ, Harding PG, Chance GW, Haust MD: Placental weight in diabetic pregnancies. Placenta 1989;10:275–281. Makhseed M, Musini VM, Ahmed MA, Al-Harmi J: Placental pathology in relation to the White’s classification of diabetes mellitus. Arch Gynecol Obstet 2002;266:136–140. Taricco E, Radaelli T, Nobile de Santis MS, Cetin I: Fetal and placental weight in relation to maternal characteristics in gestational diabetes. Placenta 2003;24:343–347. Winick M, Noble A: Cellular growth in human placenta. II. Diabetes mellitus. J Pediatr 1967;71: 216–219. Diamant YZ, Metzger BE, Freinkel N, Shafrir E: Placental lipid and glycogen content in human and experimental diabetes mellitus. Am J Obstet Gynecol 1982;144:5–11. Fox H: Pathology of the Placenta. London, Saunders, 1978, p 223. Haust MD: Maternal diabetes mellitus – effects on the fetus and placenta. Monogr Pathol 1981;22:201–285. Whitsett JA, Brownscheidle CH: Aspects of placental structure and function in maternal diabetes; in Merkatz IR, Adams PAJ (eds): The Diabetic Pregnancy – A Perinatal Perspective. New York, Grune & Stratton, 1979, pp 124–143. Benirschke K, Kaufmann P: Pathology of the Human Placenta, ed 4. New York, Springer, 2000. Younes B, Baez-Giangreco A, al-Nuaim L, al-Hakeem A, Abu Talib Z: Basement membrane thickening in the placentae from diabetic women. Pathol Int 1996;46:100–104.

The Human Placenta in Diabetes

105

26

27 28 29 30

31 32 33

34 35

36 37 38 39 40 41 42

43

44

45

46 47

48

Pietryga M, Biczysko W, Wender-Ozegowska E, Brazert J, Bieganska E, Biczysko R: Ultrastructural examination of the placenta in pregnancy complicated by diabetes mellitus. Ginekol Pol 2004;75: 111–118. Jones CJ, Fox H: Placental changes in gestational diabetes. An ultrastructural study. Obstet Gynecol 1976;48:274–280. Kaufmann P, Gentzen DM, Davidoff M: The ultrastructure of Langhans cells in pathologic human placentas. Arch Gynäkol 1977;222:319–332. Szarvas Z, Treit P: Ultrastructural changes of the terminal placental villi in placental insufficiency. Zentralbl Gynäkol 1978;100:1508–1516. Honda M, Toyoda C, Nakabayashi M, Omori Y: Quantitative investigations of placental terminal villi in maternal diabetes mellitus by scanning and transmission electron microscopy. Tohoku J Exp Med 1992;167:247–257. Hoelzl VM, Luethje D, Dietrich H: Placental changes in diabetes mellitus with special reference to villi maturation disorders. Zentralbl Gynäkol 1975;97:846–858. Mayhew TM, Sorensen FB, Klebe JG, Jackson MR: Growth and maturation of villi in placentae from well-controlled diabetic women. Placenta 1994;15:57–65. Mayhew TM: Enhanced fetoplacental angiogenesis in pre-gestational diabetes mellitus: The extra growth is exclusively longitudinal and not accompanied by microvascular remodelling. Diabetologia 2002;45:1434–1439. Ogino S, Redline RW: Villous capillary lesions of the placenta: Distinctions between chorangioma, chorangiomatosis, and chorangiosis. Hum Pathol 2000;31:945–954. Guschmann M, Henrich W, Entezami M, Dudenhausen JW: Chorioangioma – new insights into a well-known problem. I. Results of a clinical and morphological study of 136 cases. J Perinat Med 2003;31:163–169. Geppert M, Peters FD, Geppert J: Histomorphometry of the vascularization of the placental villi in diabetic pregnant women. Geburtshilfe Frauenheilkd 1982;42:628–632. Stoz F, Schuhmann RA, Schmid A: Morphometric investigations of terminal villi of diabetic placentas in relation to the White classification of diabetes mellitus. J Perinat Med 1987;15:193–198. Stoz F, Schuhmann RA, Schultz R: Morphohistometric investigations of placentas of diabetic patients in correlation to the metabolic adjustment of the disease. J Perinat Med 1988;16:211–216. Arizawa M, Nakayama M, Kidoguchi K: Correlation of placental villous immaturity and dysmaturity with clinical control of maternal diabetes. Nippon Sanka Fujinka Gakkai Zasshi 1991;43:595–602. Teasdale F: Histomorphometry of the human placenta in class B diabetes mellitus. Placenta 1983;4:1–12. Charnock-Jones DS, Kaufmann P, Mayhew TM: Aspects of human fetoplacental vasculogenesis and angiogenesis. I. Molecular aspects. Placenta 2004;25:103–113. Jirkovska M, Kubinova L, Janacek J, Moravcova M, Krejci V, Karen P: Topological properties and spatial organization of villous capillaries in normal and diabetic placentas. Vasc Res 2002;39:268–278. Arany E, Hill DJ: Fibroblast growth factor-2 and fibroblast growth factor receptor-1 mRNA expression and peptide localization in placentae from normal and diabetic pregnancies. Placenta 1998;19:133–142. Mayhew TM, Jairam IC: Stereological comparison of 3D spatial relationships involving villi and intervillous pores in human placentas from control and diabetic pregnancies. J Anat 2000;197: 263–274. Widness JA, Teramo KA, Clemons GK, Voutilainen P, Stenman UH, McKinlay SM, Schwartz R: Direct relationship of antepartum glucose control and fetal erythropoietin in human type 1 (insulin-dependent) diabetic pregnancy. Diabetologia 1990;33:378–383. Salvesen DR, Brudenell JM, Snijders RJ, Ireland RM, Nicolaides KH: Fetal plasma erythropoietin in pregnancies complicated by maternal diabetes mellitus. Am J Obstet Gynecol 1993;168:88–94. Petry CD, Wobken JD, McKay H, Eaton MA, Seybold VS, Johnson DE, Georgieff MK: Placental transferrin receptor in diabetic pregnancies with increased fetal iron demand. Am J Physiol 1994;267:E507–E514. Mayhew TM, Charnock-Jones DS, Kaufmann P: Aspects of human fetoplacental vasculogenesis and angiogenesis. III. Changes in complicated pregnancies. Placenta 2004;25:127–139.

Desoye/Kaufmann

106

49 50 51 52

53 54

55 56 57 58 59 60

61 62

63 64 65 66

67

68 69 70 71 72 73

Liebhart M: The electron-microscopic pattern of placental villi in diabetes of the mother. Acta Med Pol 1971;12:133–137. Emmrich P, Fuchs U, Heinke P, Jutzi E, Godel E: The epithelial and capillary basal laminae of the placenta in maternal diabetes mellitus. Lab Invest 1976;35:87–92. Jirkovska M: Comparison of the thickness of the capillary basement membrane of the human placenta under normal conditions and in type 1 diabetes. Funct Dev Morphol 1991;1:9–16. Mayhew TM, Sorensen FB, Klebe JG, Jackson MR: The effects of mode of delivery and sex of newborn on placental morphology in control and diabetic pregnancies. J Anat 1993;183: 545–552. Mayhew TM, Sorensen FB, Klebe JG, Jackson MR: Oxygen diffusive conductance in placentae from control and diabetic women. Diabetologia 1993;36:955–960. Mayhew TM, Jackson MR, Boyd PA: Changes in oxygen diffusive conductances of human placentae during gestation (10–41 weeks) are commensurate with the gain in fetal weight. Placenta 1993;14:51–61. Bjork O, Persson B: Villous structure in different parts of the cotyledon in placentas of insulindependent diabetic women. A morphometric study. Acta Obstet Gynecol Scand 1984;63:37–43. Bjork O, Persson B: Placental changes in relation to the degree of metabolic control in diabetes mellitus. Placenta 1982;3:367–378. Loeken MR: Free radicals and birth defects. J Matern Fetal Neonatal Med 2004;15:6–14. Hempstock J, Jauniaux E, Grennwold N, Burton GJ: The contribution of placental oxidative stress to early pregnancy failure. Hum Pathol 2003;34:1265–1275. Poston L, Raijmakers MT: Trophoblast oxidative stress, antioxidants and pregnancy outcome – a review. Placenta 2000;25(suppl A):S72–78. Nishikawa T, Edelstein D, Du XL, Yamagishi S, Matsumura T, Kaneda Y, Yorek MA, Beebe D, Oates PJ, Hammes HP, Giardino I, Brownlee M: Normalizing mitochondrial superoxide production blocks three pathways of hyperglycaemic damage. Nature 2000;404:787–790. Tames F, Mackness M, Arrol S, Laing I, Durrington PN: Non-enzymatic glycation of apolipoprotein B in the sera of diabetic and non-diabetic subjects. Atherosclerosis 1992;93:237–244. Avila C, Tanji M, D’Agati V, Stern D, Schmidt A-M: Localization of advanced glycation endproducts (AGEs) and its receptor (RAGE) in placenta of gestational and pre-gestational diabetes. J Soc Gynecol Investig 1999;6(suppl):153A. Coughlan MT, Vervaart PP, Permezel M, Georgiou HM, Rice GE: Altered placental oxidative stress status in gestational diabetes mellitus. Placenta 2004;25:78–84. Yang X, Borg LAH, Eriksson UJ: Altered mitochondrial morphology of rat embryos in diabetic pregnancy. Anat Rec 1995;241:255–267. Yang X, Borg LAH, Siman CM, Eriksson UJ: Maternal antioxidant treatments prevent diabetesinduced alterations of mitochondrial morphology in rat embryos. Anat Rec 1998;251:303–315. Jones CJ, Weiss U, Siman CM, Desoye G: Mitochondria from human trophoblast and embryonic liver cells are resistant to hyperglycaemia-associated high-amplitude swelling. Diabetologia 2001;44:389–391. Goda N, Suematsu M, Mukai M, Kiyokawa K, Natori M, Nozawa S, Ishimura Y: Modulation of mitochondrion-mediated oxidative stress by nitric oxide in human placental trophoblastic cells. Am J Physiol 1996;217:H1893–H1899. Sekiba K, Yoshioka T: Changes of lipid peroxidation and superoxide dismutase activity in the human placenta. Am J Obstet Gynecol 1979;135:368–371. Awasthi YC, Dao DD: Glutathione-mediated detoxification mechanisms of human placenta. Placanta Suppl 1981;3:289–301. Misra S, Mukherjea M: A comparative study of human placental and fetal liver catalase during development. Eur J Obstet Gynecol 1985;19:81–88. Takehara Y, Yoshioka T, Sasaki J: Changes in the levels of lipoperoxide and antioxidant factors in human placenta during gestation. Acta Med Okayama 1990;44:103–111. Jendryczko A, Drozdz M, Wojcik A: Regional variations in the activities of antioxidant enzymes in the term human placenta. Zentralbl Gynäkol 1991;113:1059–1066. Church SL, Farmer DR, Nelson DM: Induction of manganese superoxide dismutase in cultured human trophoblast during in vitro differentiation. Dev Biol 1992;149:177–184.

The Human Placenta in Diabetes

107

74 75 76 77 78

79

80

81

82 83 84 85

86 87 88 89 90 91 92 93 94 95 96

97

Walsh SW, Wang Y: Secretion of lipid peroxides by the human placenta. Am J Obstet Gynecol 1993;169:1462–1466. Choi JL, Rose RC: Transport and metabolism of ascorbic acid in human placenta. Am J Physiol 1989;257:C110–C113. Van Hien P, Kovacs K, Matkovicx B: Properties of enzymes. I. Study of superoxide dismutase activity change in human placenta of different ages. Enzyme 1974;18:341–347. Lyall F, Gibson JL, Greer IA, Brockman DE, Eis AL, Myatt L: Increased nitrotyrosine in the diabetic placenta: Evidence for oxidative stress. Diabetes Care 1998;21:1753–1758. Maxwell SR, Thomason H, Sandler D, Leguen C, Baxter MA, Thorpe GH, Jones AF, Barnett AH: Antioxidant status in patients with uncomplicated insulin-dependent and non-insulin-dependent diabetes mellitus. Eur J Clin Invest 1997;27:484–490. Poranen AK, Ekblad U, Uotila P, Ahotupa M: The effect of vitamin C and E on placental lipid peroxidation and antioxidative enzymes in perfused placenta. Acta Obstet Gynecol Scand 1998;77: 372–376. Benedetti A, Comporti M, Fulceri R, Esterbauer H: Cytotoxic aldehydes originating from the lipid peroxidation of liver microsomal lipids. Identification of 4,5-dihydroxydecenal. Biochim Biophys Acta 1984;792:172–181. Yoshioka T, Kawada K, Shimada T, Mori M: Lipid peroxidation in maternal and cord blood and protective mechanism against activated-oxygen toxicity in the blood. Am J Obstet Gynecol 1979;135:372–376. Kappus H: A survey of chemicals including lipid peroxidation in biological systems. Chem Phys Lipids 1987;45:105–113. Diamant S, Kissilevitz R, Diamant YZ: Lipid peroxidation system in human placental tissue: General properties and the influence of gestational age. Biol Reprod 1980;23:776–781. Kulkarni AP, Kenel MF: Human placental lipid peroxidation. Some characteristics of the NADPHsupported microsomal reaction. Gen Pharmacol 1987;18:491–496. Kenel MF, Bestervelt LL, Kulkarni AP: Human placental lipid peroxidation. II. NADPH and iron dependent stimulation of microsomal lipid peroxidation by paraquat. Gen Pharmacol 1987;18: 373–378. Byczkowski JZ, Kulkarni AP: NADPH-dependent drug redox cycling and lipid peroxidation in microsomes from human term placenta. Int J Biochem 1989;21:183–190. Klimek J: Cytochrome: P-450 involvement in the NADPH-dependent lipid peroxidation in human placental mitochondria. Biochim Biophys Acta 1990;1044:158–164. Klimek J: The influence of NADPH-dependent lipid peroxidation on the progesterone biosynthesis in human placental mitochondria. J Steroid Biochem Mol Biol 1992;42:729–736. Chao ST, Juchau MR: Interactions of endogenous and exogenous estrogenic compounds with human placental microsomal cytochrome P-450 (P-450hpm). J Steroid Biochem 1980;13:127–133. Walsh SW, Wang Y, Jesse R: Peroxide induces vasoconstriction in the human placenta by stimulating thromboxane. Am J Obstet Gynecol 1993;169:1007–1012. Kuhn DC, Botti JJ, Cherouny PH, Demers LM: Eicosanoid production and transfer in the placenta of the diabetic pregnancy. Prostaglandins 1990;40:205–215. Eriksson UJ, Borg LA: Protection by free oxygen radical scavenging enzymes against glucoseinduced embryonic malformations in vitro. Diabetologia 1991;34:325–331. Eriksson UJ, Borg LA: Diabetes and embryonic malformations: Role of substrate-induced freeoxygen radical production for dysmorphogenesis in cultured rat embryos. Diabetes 1993;42:411–419. Schonfelder G, John M, Hopp H, Fuhr N, van Der Giet M, Paul M: Expression of inducible nitric oxide synthase in placenta of women with gestational diabetes. FASEB J 1996;10:777–784. Di Iulio JL, Gude NM, King RG, Li CG, Rand MJ, Brennecke SP: Human placental nitric oxide synthase activity is not altered in diabetes. Clin Sci (Lond) 1999;97:123–128. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA: Apparent hydroxyl radical production by peroxynitrite: Implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 1990;87:1620–1624. Ischiropoulos H, Zhu L, Chen J, van der Woerd M, Smith C, Chen J, Harrison J, Martin JC, Tsai M: Peroxynitrite-mediated tyrosine nitration catalyzed by superoxide dismutase. Arch Biochem Biophys 1992;298:431–437.

Desoye/Kaufmann

108

98 Myatt L, Rosenfield RB, Eis AL, Brockman DE, Greer I, Lyall F: Nitrotyrosine residues in placenta. Evidence of peroxynitrite formation and action. Hypertension 1996;28:488–493. 99 Li H, Gu Y, Zhang Y, Lucas MJ, Wang Y: High glucose levels down-regulate glucose transporter expression that correlates with increased oxidative stress in placental trophoblast cells in vitro. J Soc Gynecol Investig 2004;11:75–81. 100 Kinalski M, Telejko B, Kowalska I, Urban J, Kinalska I: The evaluation of lipid peroxidation products and antioxidative enzymes activity in cord blood and placental homogenates of pregnant diabetic women. Ginekol Pol 1999;70:57–61. 101 Pustovrh C, Jawerbaum A, Sinner D, Pesaresi M, Baier M, Micone P, Gimeno M, Gonzalez ET: Membrane-type matrix metalloproteinase-9 activity in placental tissue from patients with preexisting and gestational diabetes mellitus. Reprod Fertil Dev 2000;12:269–275.

Gernot Desoye, MD Clinic of Obstetrics and Gynaecology Auenbruggerplatz 14, AT–8036 Graz (Austria) Tel. ⫹43 316 385 4605, Fax ⫹43 316 385 2506 E-Mail [email protected]

The Human Placenta in Diabetes

109

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 110–126

Vasculogenesis and Angiogenesis in the Diabetic Placenta Lopa Leach, Terry M. Mayhew Centre for Integrated Systems Biology and Medicine, Institute of Clinical Research, Faculty of Medicine and Health Sciences, Queen’s Medical Centre, University of Nottingham, Nottingham, UK

As well as being supplied by the maternal circulation, the human placenta contains a vascular complex that is entirely fetal in origin and continuous with the vasculature of the developing fetus. Fetoplacental vessels are found in chorionic villi that are bathed in maternal blood and this close proximity permits efficient exchange of solutes and gases between the maternal and fetal circulations without intermingling of the two. The arrangement allows the development, growth and remodelling of fetoplacental vessels to be matched to fetal need but also renders them vulnerable to changes on both maternal and fetal sides of the placenta. Any pathological alterations in maternal haemodynamics, in maternal blood properties (such as hypoxia or hyperglycaemia) or in growth factors including vascular endothelial growth factor (VEGF), cytokines and inflammatory mediators may influence directly the growth, maintenance and functioning of fetoplacental vessels. Furthermore, changes in fetoplacental vessels may predict or reflect those in the vasculature of the developing fetus. Diabetes mellitus (DM) has been linked to accelerated microangiopathy (in diabetic retinopathy, nephropathy and neuropathy) and this, in turn, may be associated with capillary hypertension and changes in capillary permeability. The diabetic milieu includes hypoxia, hyperglycaemia, increased VEGF, free oxygen radicals, advanced glycation end products (AGEs), cytokines and inflammatory mediators. These factors are known to impair endothelial barrier function or to have angiogenic effects. In fact, increased angiogenesis is a common feature of diabetic vasculopathy, notably in the diabetic complication of proliferative retinopathy [1]. Maternal DM is linked to increases in morbidity and mortality in the embryonic, fetal and perinatal periods. Amongst the various abnormalities is a

range of cardiovascular problems which afflict the offspring of diabetic mothers. The vascular system is the first organ system to develop and its appearance is a stimulus to the development of other organs and, thereby, to organotypic differentiation of their own vascular beds. Just as organ differentiation during embryogenesis is promoted by fetal vascular morphogenesis, so placental growth and differentiation are promoted by establishing the fetoplacental vasculature. Indeed, throughout gestation, the increasing requirements of the growing fetal mass are associated with continuous adaptations in the structure and functional capacity of the placenta and its fetal vascular bed. The present chapter focuses on events which sculpt the fetoplacental vasculature and on perturbations which arise as a consequence of maternal DM.

Vascular Morphogenesis

This term embraces a number of processes which lead to the establishment of a mature vascular network. The processes include: vasculogenesis (creation of vessels from haemangioblastic precursor cells), angiogenesis (creation of vessels or vessel segments from existing vessels), network formation (formation of a primary vascular plexus), and network remodelling and vessel identity (formation of a mature network which contains arteries, capillaries and veins). In normal human pregnancy, placental vascular development occurs in phases and, as in other vasculatures, depends on the actions of angiogenic growth factors and their receptors produced by cells and extracellular matrix ingredients [2, 3]. Moreover cell-cell adhesion molecules, the cement that holds endothelial cells together, are directly or indirectly involved in molecular mechanisms driving vascular morphogenesis. Vasculogenesis Vasculogenesis of fetoplacental vessels occurs in the first month of gestation when stem cells in extra-embryonic mesoderm invade the developing chorionic villi. Ultrastructural studies on the macaque monkey and human placenta [4–6] have helped to identify the sequence of events. Mesenchymal stem cells aggregate into haemangioblastic cords within which the innermost cells are haematopoietic whilst outer cells (angioblasts) become endothelial cells. Dilation of paracellular clefts between pre-endothelial cells seems to be the main mechanism of lumen formation in these vessel precursors [3, 6]. Subsequently, the vascular plexus forms by tube coalescence, branching and anastomosis [7] but a fully functional circulation is not established until this is connected to the vasculature of the fetus.

Vasculogenesis and Angiogenesis in the Diabetic Placenta

111

Angioblasts and haematopoietic cells differentiate under the influence of vascular endothelial growth factor A (VEGF-A or VEGF) and fibroblast growth factors (FGFs). Both cell types express the receptor VEGFR-2 (KDR/flk-1) for which VEGF is the ligand. VEGFR-2 is the major mediator of the mitogenic, angiogenic and permeability enhancing effects of VEGF. The key role of this receptor is evidenced by a lack of vasculogenesis and failure to develop blood islands and organized blood vessels in Flk-1-null mice, resulting in death in utero between days 8.5 and 9.5 [8]. VEGFR-2-positive cells also give rise to the progenitors of perivascular or mural cells, which include pericytes and vascular smooth muscle cells. VEGFR-2 undergoes VEGF-dependent tyrosine phosphorylation and induces growth by activation of the Raf-Mek-Erk pathway. The PI-3 kinase-Akt pathway mediates the pro-survival activity. VEGFR-1 (flt-1), for which the ligands are VEGF and placenta growth factor (PlGF), is also expressed by angioblasts and implicated in their assembly into blood vessels. Its extracellular portion (the soluble VEGF antagonist, sVEGFR-1 or sflt-1) is sufficient for vascular development [9]. VEGFR-1 is expressed by villous endothelium, macrophages and trophoblast [7, 10–12] but the latter also expresses sVEGFR-1 [13, 14]. Pericyte induction within mesoderm requires transforming growth factor-␤ (TGF-␤) whilst recruitment to newly formed vessels requires platelet-derived growth factor (PDGF) and basic FGF (bFGF or FGF-2). Acidic FGF (aFGF or FGF-1) and FGF-2 are both expressed in human placental villi and thought to promote haemangioblast recruitment [2, 15, 16] but data on FGF receptors are sparse. In developing vessels, the distribution of PDGF and its receptor is consistent with the recruitment of perivascular cell congeners by endothelial cells [17]. VEGF is highly expressed in early pregnancy and associated with commitment, growth and aggregation of endothelial precursors. Whilst villous trophoblast and stromal macrophages (Hofbauer cells) are the main sources of VEGF [10, 18–20], a sequential expression suggests that cytotrophoblasts regulate early vasculogenic events whilst stromal macrophages help to control later angiogenic events [7]. Over time, this results in a spatial distribution in which, towards term, VEGF is expressed mainly, if not exclusively, in terminal villi (TV) [6]. Angiopoietins are required for angiogenesis, arteriogenesis and vessel stabilization. Angiopoietin-1 (Ang-1) and VEGF have complementary roles during vascular development. VEGF plays an early role in vessel formation whilst Ang-1 acts later during vessel remodelling, maturation and stabilization [21]. Ang-2 can act synergistically with VEGF in development of heart vessels but, in the absence of angiogenic signals, can lead to endothelial cell death and vessel regression [22]. In early placentas, weak immunoreactivity for Ang-1 and Ang-2 is observed because the nascent vessels are not yet stabilized or mature [6].

Leach/Mayhew

112

Vasculogenesis and angiogenesis are also regulated in part by the capacity of endothelial cells to adhere to each other and assemble into new vessels. Indeed, receptors for cell-matrix and cell-cell adhesion play pivotal roles. Vascular-endothelial cadherin (VE-cadherin) is the key transmembrane glycoprotein of adherens junctions (AJs). The calcium-dependent homotypic binding of its extracellular portion allows cell-cell adhesion and both the rigidity and clustering of extracellular domains influence adhesive strength. The cytoplasmic tail of VE-cadherin is an essential link to key signal transduction ligands such as those of the armadillo family, ␤-catenin, plakoglobin (␥-catenin) and p120. Binding to ␤-catenin allows the junctional complex to link to adjacent actin microfilaments [23, 24]. VEGF-induced phosphorylation of VE-cadherin (which contains tyrosine, serine and threonine residues) and ␤-catenin are associated in vitro with increased permeability and small-pore barrier dysfunction [25]. In mice, cytosolic truncation of the VE-cadherin gene (loss of the ␤-catenin adhesive domains) impairs VEGF-mediated endothelial survival and angiogenesis [26]. The survival function of VEGF appears to depend on formation of a complex containing VEGFR-2, ␤-catenin and VE-cadherin. Cells deficient in VE-cadherin cannot be rescued from apoptosis by VEGF [26]. Increasing evidence suggests that AJ molecules play a role in angiogenesis. Breakage of cell-cell adhesion, induction of proliferation and subsequent tubulogenesis all require efficient functioning of these molecules. Epithelial (and endothelial) studies link AJ molecules specifically with angiogenesis because unsequestering ␤-catenin from cell junctions results in active cytoplasmic pools of ␤-catenin. This free ␤-catenin can translocate to the nucleus and modulate cell transcription [27, 28]. An increase in VE-cadherin causes recruitment of cytoplasmic ␤-catenin by AJs and, thereby, inhibits signaling [24, 29]. Indeed, VE-cadherin expression and clustering at intercellular junctions (i.e. contact inhibition) block the proliferative response of endothelial cells to VEGF. Inhibition of VEGFR-2 tyrosine phosphorylation contributes to this effect [24]. Tight junctions (TJs) are dynamic, regulated structures forming zones of very close contact between adjacent cells. As in endothelium of other vessels, the TJs of placental endothelial cells are not fused but adjacent cells are separated by a gap of 4 nm [30, 31]. Occludin, claudins 1, 2 and 5/6 and junctional adhesion molecules are all transmembrane proteins localized at endothelial TJs [32, 33]. These molecules may also influence angiogenesis. In human placental microvascular endothelial cells, increases in cAMP-induced occludin correlate with inhibition of cell proliferation and increased transendothelial resistance [34]. SSeCKs, a protein kinase C substrate, regulates blood-brain barrier maturation (by altering TJ formation) and alters brain angiogenesis, mainly through decreased expression of VEGF and enhanced expression of Ang-1 [35]. Decreases in occludin content and increased retinal vascular permeability occur

Vasculogenesis and Angiogenesis in the Diabetic Placenta

113

in experimental diabetes in the rat [36]. VEGF can reduce the occludin content in retinal endothelial cells, whilst phosphorylation of ZO-1 has been implicated in decreased transcellular electrical resistance in brain endothelial cells [36, 37]. In vitro studies on embryonic stem cells [38, 39] and human umbilical vein cells (HUVECs) [40, 41], as well as studies of the murine vascular system [26], have clarified the importance of the adhesion molecules VE-cadherin and PECAM-1, and their sequential expression during cardiovascular development. Similar expression profiles are reported for the early placenta [6]. VE-cadherin is required for tube formation and is an early constitutive marker of cells committed to becoming endothelial (rather than haematopoietic) cells. Firsttrimester placental vessels contain VE-cadherin and ␤-catenin but lack plakoglobin, the component of fully differentiated AJs [6]. Furthermore, these vessels lack the TJ molecules, occludin, claudin 1 and claudin 2. This profile may reflect the phenotype of highly plastic, nascent or angiogenic vessels. Whether or not maternal DM affects placental vasculogenesis in the first trimester is not known. In streptozotocin-induced diabetic mice, and in murine conceptuses cultured under hyperglycaemic conditions, yolk sac vasculopathy may be associated with changes in expression and phosphorylation of platelet endothelial growth factor-1 and VEGF [42]. Obtaining experimental data on first-trimester vasculogenesis in maternal DM is an important avenue for future exploration. Angiogenesis and Vascular Remodelling Angiogenesis involves the creation of new vessels from existing vessels and, following vasculogenesis, leads to the appearance of primary, primitive, nascent or immature capillary networks which, gradually, are converted into mature networks. In the placenta, this may occur in three stages [3]. First (4–25 weeks of gestation), branching angiogenesis predominates within villi. Next (15–32 weeks), some peripheral capillary plexuses appear to regress leaving villi with central stem vessels (arteries and veins). Le Noble et al. [43] have shown that, during arteriovenous differentiation in the yolk sac, small-calibre vessels of arterial domains selectively disconnect from growing arterial trees and reconnect to the venous system. It remains to be seen whether this mechanism contributes to the apparent capillary regression in placental villi. Finally (25 weeks to term), terminal capillary loops are generated as part of a move towards greater non-branching angiogenesis. These events determine the vascularization indices of villi. For example, from 10 to 21 weeks (during the phase of branching angiogenesis), capillary:villus length ratios increase to a peak of about 4:1. They decline sharply during the capillary regression phase before gradually regaining peak values towards the end of the final phase of nonbranching angiogenesis [44].

Leach/Mayhew

114

Endothelial tubes formed by vasculogenesis transform into primitive capillary networks by branching angiogenesis. New branches and segments are created by two main mechanisms which may operate in isolation or in tandem [45, 46]. The first mechanism is sprouting angiogenesis and involves development of lateral sprouts from existing vessels. The second mechanism is intussusceptive angiogenesis and involves transcapillary endothelial pillars partitioning the lumen of a vessel into two or more lumina. Further work is required to establish the contributions made by these alternative types of branching angiogenesis. In the youngest (mesenchymal) villi, branching angiogenesis occurs less often than non-branching angiogenesis and capillary networks are poorly developed. However, as gestation proceeds, branching angiogenesis is stimulated and primitive capillary beds are transformed into a dense network below the trophoblast. Expressions of VEGF and VEGFR-2 are most intense early in gestation but decline sharply as pregnancy advances [7, 12, 47–51]. In contrast, there are higher expressions of PlGF and soluble VEGFR-1 towards term [2, 3, 13, 14, 51] as the incidence of non-branching angiogenesis rises. For both types of branching angiogenesis, it seems likely that regulation is influenced by VEGF and its receptors together with Ang-1 and Ang-2 and their receptor, Tie-2 (Tek). The role of PlGF in branching versus non-branching is uncertain. As part of vascular remodelling and stabilization, the coverage of vessels by perivascular cells also alters. At 8–12 weeks (early angiogenesis), less than 40% of fetoplacental vessels are associated with perivascular cells but the proportion rises to about 63% by term [52]. Whether this proportion is altered in the diabetic human placenta is not known. In diabetic mouse retina, the pericyte number declines by 50% [53] and may contribute to the pathogenesis of proliferative diabetic retinopathy. From about 25 weeks of gestation, there is a shift from producing the trunks and main branches of villous trees to generating finer branches at the periphery. In this context, vessel growth moves more towards non-branching angiogenesis and this accompanies the development of mature intermediate villi (MIV) at the periphery. MIV are long (⬎1,000 ␮m) and slender (80–120 ␮m) and harbour 1–2 poorly branched capillary loops [3]. They exhibit relatively low levels of trophoblast proliferation but greater levels of endothelial proliferation. Although suggestive of proliferative elongation [2], growth could involve also intercalative elongation where circulating endothelial progenitor cells are recruited into existing vascular endothelium. Differential growth of villous capillaries produces TV with coiled capillaries which obtrude into the overlying trophoblast. These are the principal sites for exchange of gases and nutrients between the mother and fetus by passive diffusion. Normally, the capillary loops of 5–10 TV are connected in series by the elongated capillaries of a single

Vasculogenesis and Angiogenesis in the Diabetic Placenta

115

MIV. With advancing gestation, terminal capillaries dilate focally into sinusoids (⬎40 ␮m diameter) which may compensate, to some extent, for the adverse effects on total vascular impedance of generating longer capillary loops [3, 50]. Molecular changes accompany these morphological events. The expressions of VEGF and VEGFR-2 are less intense than earlier in pregnancy [12, 47–51] whilst those of VEGFR-1 and PlGF are greater [11, 51]. PlGF is expressed in syncytiotrophoblast [19, 20] and the media of larger stem vessels [54, 55]. Shifting patterns of angiopoietin expression during gestation reflect the acquisition of larger immature villi containing differentiated vessels (arteries and veins) together with developing MIV and TV containing immature vessels. Ang-1 (which stabilizes vessels and constrains permeability) is found in most fetal vessels but expressed most intensely by large arteries. In contrast, Ang-2 (an Ang-1 antagonist) is present in TV and, to a lesser extent, in large vessels [6]. Ang-2 mRNA expression declines about 10-fold during gestation [50]. Associated with changes in vessel stability and villous maturity are increases in the percentage of vessels which have recruited perivascular cells [52]. There is differential expression of endothelial junctional molecules during gestation. Whilst vessels in the first-trimester placenta exhibit the dynamic junctional phenotype, those in term placenta exhibit a more stable phenotype with well-differentiated junctions. In the former, AJs contain VE-cadherin and ␤-catenin, whilst TJs appear to lack occludin and claudin 1 (both implicated in transendothelial resistance). At term, the more stable junctions display plakoglobin in AJs whilst TJs are rich in occludin and claudin 1 [41] and these changes may be associated with increased levels of Ang-1. However, even in term placenta, there are two distinct vascular populations. Vessels of stem and some intermediate villi contain the stable junctional phenotype whilst the capillaries of TV, the exchange vessels of the placenta, express the more labile phenotype (rich in ␤-catenin but lacking plakoglobin and occludin). The latter is reminiscent of the forming, or newly formed, vessels during the first trimester and this molecular phenotype may define labile junctions necessary for angiogenesis and paracellular solute transport, two key functions of TV capillaries in the last trimester. Certainly, in vitro and perfusion studies have shown that the molecular composition of endothelial junctions can dictate both paracellular permeability and angiogenesis [23, 25, 40, 56]. The more mature phenotype is linked to endothelial quiescence and restricted paracellular solute transport [23, 34, 41]. Perturbation of junctional molecules is a complication of DM in retinopathy [36]. The same appears to be true for fetoplacental vessels in pregestational DM (types 1 and 2) and in gestational diabetes (GDM). In type 1 DM [56], over 50% of placental microvessels show complete loss of VE-cadherin and ␤-catenin due to loss from designated microdomains (i.e. junctional regions)

Leach/Mayhew

116

a

b Fig. 1. a Confocal micrograph of a frozen section through a normal placental lobule perfused for 10 min with 76-kDa dextran conjugated with TRITC (1 mg/ml) [56]. No leakage of tracer from the fetal vascular compartment can be seen. b Confocal micrograph of a frozen section through a placental lobule similarly perfused with 76-kDa dextran-TRITC. The placenta (at term) was taken from a pregnancy complicated by type 1 diabetes [56]. Tracer can be seen trapped in the interstitial space outside fetal vessel suggesting increased leakiness of these vessels. Magnification ⫻1,000.

rather than loss of total protein. The depletion is associated with increased phosphorylation of these molecules as well as increases in the proportions of vessels showing junctional phosphotyrosine immunoreactivity. The immediate functional consequence of these perturbations is greater extravasation of tracers (76-kDa dextrans) which, in normal pregnancies, are retained in the vascular compartment of placental vessels (fig. 1). This suggests that, in type 1 DM, placental vessels, like those in the retina, are leaky. Preliminary findings in our laboratory reveal a similar downregulation of junctional expression in the type 2 diabetic placenta. VEGF levels are elevated in the type 1 diabetic placentas and this may be responsible in part for the phosphorylation and loss of adhesion molecules from AJs. In vitro, VEGF-induced tyrosine phosphorylation of VE-cadherin and ␤-catenin can lead to declustering of these molecules, junctional alterations and increased permeability to macromolecules [23, 25]. Perfusion of placental microvessels with VEGF reduces junctional immunoreactivity and extravasation of 76-kDa dextrans [57]. This observation strengthens the notion that increased VEGF levels contribute to the junctional perturbations observed in type 1 diabetic placentas. Although the signalling properties of VE-cadherin and ␤-catenin are largely VEGF-induced [24], the levels of other angiogenic growth factors, such as PlGF, FGF-2, Ang-1 and Ang-2, may also influence vascular development and maturity in the diabetic placenta and should be monitored.

Vasculogenesis and Angiogenesis in the Diabetic Placenta

117

Loss of junctional VE-cadherin and ␤-catenin are very early events in VEGF-induced angiogenesis in HUVECs [4]. VEGF elicits increased nuclear localization of ␤-catenin and increased cyclin D1 in HUVECs and this is correlated with increased proliferation [28]. Therefore, it is no surprise that increased endothelial proliferation is a feature of type 1 diabetic placenta [56]. Culturing human placental microvascular endothelial cells in the presence of elevated VEGF and FGF has resulted in increased proliferation and endothelial cells displaying poorly organized cell-cell junctions [34]. Indeed, this study also demonstrated that elevation of cAMP induces quiescence and formation of occludin-rich TJs. Stereological analyses have confirmed that the combined length of capillaries in intermediate and TV is significantly greater in diabetic placentas, clearly indicating that enhanced fetoplacental angiogenesis occurs in type 1 diabetic pregnancies [56]. In GDM (which affects the latter half of pregnancy, so sparing the development stages of vasculogenesis and early angiogenesis), the picture is less clear since there is junctional perturbation [58] but evidence for and against enhanced angiogenesis [58–60]. A study using three-dimensional visualization techniques has reported increased longitudinal growth of fetoplacental vessels and enhanced branching angiogenesis in GDM [59]. The glucose intolerance of GDM is usually mild, but nevertheless it means a higher incidence of complications during pregnancy and increased perinatal mortality and morbidity of the infants. Early studies of the placenta from GDM mothers revealed that there was significantly more surface area of exchange between mother and fetus, in terms of peripheral villous and capillary surface areas, and a greater intervillous space volume [40]. These structural changes were interpreted as successful adaptations to maternal metabolic disturbances. The permeability status of these vessels is not known. Functional Implications of Changes in Vascular Anatomy and Arrangement Doppler studies have shown that fetal vascular resistance decreases during normal pregnancy and may be compromised in complicated pregnancies [61]. Since there is no autonomic innervation, blood flow in the maturing vasculature must be regulated by local vasoactive effectors and by combining physiological adjustments (of perfusion pressure and vascular impedance) with anatomical changes (in vessel dimensions or spatial arrangements). Essentially, the latter involve generating either a few long vessel segments or many short segments. In a single uniform vessel, resistance to blood flow is directly proportional to its length and inversely proportional to the square of its cross-sectional area. Moreover, the total resistance (Rt) of a parallel arrangement of multiple vessel segments is less than the partial resistance of any individual segment

Leach/Mayhew

118

(1/Rt ⫽ 1/R1 ⫹ 1/R2 ⫹ 1/R3 ⫹ …1/Rn) but the total resistance of a serial arrangement is the sum of its partial resistances (Rt ⫽ R1 ⫹ R2 ⫹ R3 ⫹ …Rn). Other variables being constant, it follows that parallel arrangements of vessel segments are preferable to serial arrangements because they generate multiple segments of reduced mean length and, hence, of smaller overall impedance. The preferred option for minimizing segment lengths and resistances is branching angiogenesis. Whilst both sprouting and intussusceptive angiogenesis create parallel vascular arrangements in which there are numerous, but relatively short, vessel segments, intussusceptive angiogenesis [46] is the more economical (requiring less endothelial proliferation) and more versatile (contributing to several aspects of vascular morphogenesis). In contrast to branching, non-branching angiogenesis involves elongation of existing vessel segments and leads to increased resistance to flow. Elongation may be driven by proliferation of existing endothelial cells or intercalation of endothelial progenitor cells or a combination of proliferation and intercalation. From these considerations, it is reasonable to anticipate that the fetoplacental vasculature might grow principally by branching angiogenesis. In fact, it grows by a mixture of branching and non-branching angiogenesis and this pattern is maintained in the diabetic placenta. Enhanced fetoplacental angiogenesis in DM is not accompanied by changes in capillary calibre or cross-sectional shape, suggesting there is no vascular remodelling beyond that seen in nondiabetic controls. It is likely also that the mean surface area occupied by an endothelial cell is conserved in DM and, therefore, that increased lengths and volumes of capillaries are achieved by proliferation (possibly supplemented by intercalation) rather than remodelling of endothelial cells [62, 63]. Although not found consistently in all classes of DM, villous vascularization indices may be greater in diabetic pregnancies but this appears to be unrelated to the presence or otherwise of diabetic complications (retinopathy, nephropathy). Morphometric studies on subjects with type 1 pregestational DM have shown that increased fetoplacental angiogenesis can occur despite good control of glycaemia and glycated haemoglobin HbA1c levels [56, 62–66]. Why this is so warrants further scrutiny. DM is a syndrome and, consequently, factors beyond hyperglycaemia may influence the vascular changes seen, including insulin treatment. Chronic hypoxia may be partly responsible for the increased angiogenesis and predominance of leaky blood vessels induced by VEGF. Hypoxia is known to cause local elevation of VEGF and, in pregestational DM, there is evidence of fetal hypoxia. Adaptations in placental oxygen diffusive conductances affect principally the downstream tissues (villous stroma and fetal capillaries) of the intervascular barrier [67]. In addition, maternal haematocrit is normal but there are elevated fetal haematocrits, haemoglobin concentrations and levels of erythropoiesis [62, 67, 68]. Erythropoietin regulates

Vasculogenesis and Angiogenesis in the Diabetic Placenta

119

erythropoiesis during fetal life and reduced oxygen tension is the proximate stimulus for its production [69, 70]. Indeed, it has been suggested that erythropoietin itself has angiogenic properties [71]. Villous trophoblast secretes erythropoietin and its receptor is found on trophoblast and the endothelium of fetal vessels [72]. Moreover, the placenta may provide an important source of erythropoietin in cases of fetal hypoxia [73]. Amongst the consequences of hyperglycaemia are non-enzymic glycation of proteins and production of AGEs. Excessive glycation of extracellular matrix, produced by exposing umbilical vein endothelial cell cultures to high glucose concentrations, may explain impaired adhesion of pericytes [74]. AGEs increase VEGF expression in rat retina and act synergistically with hypoxia [75]. AGE receptors are present on various cells including endothelial cells, perivascular cells and macrophages. AGEs are toxic to pericytes and it has been shown that glycated albumin induces angiogenesis but the nascent vessels are deficient in pericytes [76]. This deficiency can stimulate endothelial proliferation [77] and lead to production of vessels which are more plastic (less mature) and more permeable. There is some uncertainty about the expression of angiogenic growth factors in the type 1 diabetic placenta. Janota et al. [78] reported no differences in mRNA expression levels of VEGF, Ang-1, Ang-2, FGF-2 or their receptors. However, at the protein level, differences in VEGF, VEGFR-1, FGF-2 and FGF-R2 have been detected [56, 79–81]. In maternal serum, levels of PlGF appear normal [80] but levels of FGF-2 are elevated [81]. Interestingly, increased PlGF expressions and local VEGF levels accompany retinal neovascularization in diabetic patients with proliferative retinopathy [82–84]. FGF-2 expression in diabetic pregnancies is greater in those complicated by retinopathy. Normally, FGF-2 is expressed in placental villi, appears in maternal serum and peaks in the second trimester [79, 81, 85]. Levels of FGF-2 in maternal and cord serum also correlate positively with fetal size and increase in pregestational DM [85]. These changes are accompanied by increased staining for FGF-2 and its high-affinity receptor in the villous syncytiotrophoblast [86]. FGF-2 expression is reported to increase in this compartment, and in cytotrophoblast and various stromal cells, in pregestational DM [87]. The morphological and molecular findings suggest that, in pregestational DM, increased vasculogenesis and angiogenesis could arise following hyperglycaemic episodes at early stages of pregnancy when oxygen tensions are low and VEGF expression is upregulated. This would be consistent with elevated levels of FGF-2 and VEGFR-1 which are implicated in vasculogenesis. Hyperglycaemia could also have direct toxic effects on pericytes and affect their adhesion to basal lamina and, hence, association with endothelial cells. These events would be expected to delay the stabilization of nascent capillaries.

Leach/Mayhew

120

After mid-gestation, the effects of high glucose might have less impact as the phase of elevated oxygen tensions and PlGF levels, together with reduced expression of VEGF, leads to non-branching angiogenesis. If fetoplacental angiogenesis is not elevated in GDM [58], an explanation might reside in the fact that this type of DM is usually diagnosed in the second half of pregnancy. Consequently, some of the metabolic confounders identified above might not appear until after the early period of branching angiogenesis has passed. To resolve the apparent discrepancies, further studies on control and diabetic placentas at different periods of gestation are warranted. The Wider Picture

The effects of DM impact on many organ systems and include vascular abnormalities in the retina and kidneys as well as the fetus and placenta. However, caution is advisable when extrapolating angiogenic events from one organ system to another because the responses to DM may vary between organs [88]. Enhanced angiogenesis occurs in diabetic retinopathy and nephropathy whereas reduced angiogenesis occurs in embryonic vasculopathy, compromised development of coronary artery collaterals and decreased wound healing capacity. Hypoxia upregulates VEGF and, indeed, levels of VEGF are elevated in patients with diabetic retinopathy and nephropathy [89]. Other factors which upregulate VEGF include hyperglycaemia, cytokines (e.g. TGF-␤) and growth factors (e.g. FGF-2, PDGF). In diabetic retinopathy, for example, there are elevated levels of FGF and neovascularization depends on detachment of pericytes from basal lamina and vascular endothelium [90, 91]. Later, endothelial cells secrete PDGF which stimulates recruitment of pericytes and activates TGF-␤ which stabilizes nascent vessels by inhibiting endothelial proliferation, promoting differentiation of perivascular cells and producing basal lamina and cell adhesion molecules [90, 92]. Ang-1 is also implicated in the process of vessel stabilization and leakage resistance but low oxygen tensions upregulate its antagonist, Ang-2 [91, 93]. The total interplay between these growth factors and the various growth, proliferation and survival signalling pathways will dictate the development of a mature vascular bed, capable of regulating solute transport that meets the specific requirement of the tissue it serves. Nowhere is it more important than at the maternofetal interface, i.e. the human placenta, which is developed solely for sustaining fetal growth. Acknowledgements LL gratefully acknowledges funding from The Wellcome Trust and Association of International Cancer Research for the research in her laboratory. TMM wishes to thank

Vasculogenesis and Angiogenesis in the Diabetic Placenta

121

The Medical Research Council (Development Grant Scheme, G9826907) and The Wellcome Trust for recent research funding.

References 1

2 3 4 5 6 7

8 9

10

11

12 13

14

15

16 17 18 19

Sander B, Larsen M, Engler C, Lund-Andersen H, Parving HH: Early changes in diabetic retinopathy: Capillary loss and blood-retina barrier permeability in relation to metabolic control. Acta Ophthalmol 1994;72:553–559. Charnock-Jones DS, Kaufmann P, Mayhew TM: Aspects of fetoplacental vasculogenesis and angiogenesis. I. Molecular regulation. Placenta 2004;25:103–113. Kaufmann P, Mayhew TM, Charnock-Jones DS: Aspects of fetoplacental vasculogenesis and angiogenesis. II. Changes during normal pregnancy. Placenta 2004;25:114–126. King BF: Ultrastructural differentiation of stromal and vascular components in early macaque placental villi. Am J Anat 1987;178:30–44. Demir R, Kaufmann P, Castellucci M, Erbengi T, Kotowski A: Fetal vasculogenesis and angiogenesis in human placental villi. Acta Anat 1989;136:190–203. Leach L, Babawale MO, Anderson M, Lammiman M: Vasculogenesis, angiogenesis and the molecular organisation of endothelial junctions in the early human placenta. J Vasc Res 2002;39:246–259. Demir R, Kayisli UA, Seval Y, Celik-Ozenci C, Korgun A, Demir-Weusten AY, Huppertz B: Sequential expression of VEGF and its receptors in human placental villi during very early pregnancy: Differences between placental vasculogenesis and angiogenesis. Placenta 2004;25:560–572. Shalaby F, Rossant J, Yamaguchi TP, Gertsenstein M, Wu X-F, Breitman M, Schuh AC: Failure of blood-island formation and vasculogenesis in Flk-1-deficient mice. Nature 1995;376:62–66. Hiratsuka S, Minowa O, Kuno J, Noda T, Shibuya M: Flt-1 lacking the tyrosine kinase domain is sufficient for normal development and angiogenesis in mice. Proc Natl Acad Sci USA 1998;95: 9349–9354. Ahmed A, Li XF, Dunk C, Whittle MJ, Rushton DI, Rollason T: Colocalisation of vascular endothelial growth factor and its Flt-1 receptor in human placenta. Growth Factors 1995;12: 235–243. Clark DE, Smith SK, Sharkey AM, Charnock-Jones DS: Localization of VEGF and expression of its receptors flt and KDR in human placenta throughout pregnancy. Hum Reprod 1996;11: 1090–1098. Vuckovic M, Ponting J, Terman BI, Niketic V, Seif MW, Kumar S: Expression of the vascular endothelial growth factor receptor, KDR, in human placenta. J Anat 1996;188:361–366. Clark DE, Smith SK, He Y, Day KA, Licence DR, Corps AC, Lammoglia R, Charnock-Jones DS: A vascular endothelial growth factor antagonist is produced by the human placenta and released into the maternal circulation. Biol Reprod 1998;59:1540–1548. He Y, Smith SK, Day KA, Clark DE, Licence DR, Charnock-Jones DS: Alternative splicing of vascular endothelial growth factor (VEGF)-R1 (FLT-1) pre-mRNA is important for the regulation of VEGF activity. Mol Endocrinol 1999;13:537–545. Ferriani RA, Ahmed A, Sharkey AM, Smith SK: Colocalization of acidic and basic fibroblast growth factor (FGF) in human placenta and the cellular effects of bFGF in trophoblast cell line JEG-3. Growth Factors 1994;10:259–268. Shams M, Ahmed A: Localization of mRNA for basic fibroblast growth factor in human placenta. Growth Factors 1994;11:105–111. Holmgren L: Potential dual roles of PDGF-B during human placental blood vessel formation. Endothelium 1993;1:167–171. Sharkey AM, Charnock-Jones DS, Boocock CA, Brown KD, Smith SK: Expression of mRNA for vascular endothelial growth factor in human placenta. J Reprod Fertil 1993;99:609–615. Shore VH, Wang T-H, Wang C-L, Torry RJ, Caudle MR, Torry DS: Vascular endothelial growth factor, placenta growth factor and their receptors in isolated human trophoblast. Placenta 1997;18:657–665.

Leach/Mayhew

122

20

21 22 23 24

25

26

27 28

29 30 31 32 33

34

35 36

37 38

39 40

Vuorela P, Hatva E, Lymboussaki A, Kaipainen A, Joukov V, Persico MG: Expression of vascular endothelial growth factor and placenta growth factor in human placenta. Biol Reprod 1997;56: 489–494. Thurston G: Complementary actions of VEGF and angiopoetin-1 on blood vessel growth and leakage. J Anat 2002;200:575–580. Ferrara N, Gerber H-P, LeCouter J: The biology of VEGF and its receptors. Nat Med 2003;9: 669–676. Dejana E: Endothelial adherens junctions: Implications in the control of vascular permeability and angiogenesis. J Clin Invest 1996;98:1949–1953. Lampugnani MG, Zanetti A, Corada M, Takahashi T, Balconi G, Breviario F, Orsenigo F, Cattelino A, Kemler R, Daniel TO, Dejana E: Contact inhibition of VEGF-induced proliferation requires vascular endothelial cadherin, ␤-catenin, and the phosphatase DEP-1/CD148. J Cell Biol 2003;161:793–804. Esser S, Lampugnani MG, Corada M, Dejana E, Risau W: Vascular endothelial growth factor induces VE-cadherin tyrosine phosphorylation in endothelial cells. J Cell Sci 1998;111: 1853–1865. Carmeliet P, Lampugnani MG, Moons L, Breviario F, Compernolle V, Bono F, Balconi G, Spagnuolo R, Oostuyse B, Dewerchin M, Zanetti A, Angellilo A, Mattot V, Nuyens D, Lutgens E, Clotman F, de Ruiter MC, Gittenberger-de Groot A, Poelmann R, Lupu F, Herbert JM, Collen D, Dejana E: Targeted deficiency or cytosolic truncation of the VE-cadherin gene in mice impairs VEGF-mediated endothelial survival and angiogenesis. Cell 1999;98:147–157. Gottardi CJ, Gumbiner BM: Adhesion signaling: How ␤-catenin interacts with its partners. Curr Biol 2001;11:R792–R794. Ilan N, Tucker A, Madri JA: Vascular endothelial growth factor expression, ␤-catenin tyrosine phosphorylation and endothelial proliferative behaviour: A pathway for transformation? Lab Invest 2003;83:1105–1115. Conacci-Sorrell M, Zhurinsky J, Ben-Ze’ev A: The cadherin-catenin adhesion system in signaling and cancer. J Clin Invest 2002;109:987–991. Firth JA, Bauman KF, Sibley CP: The intercellular junctions of guinea-pig placental capillaries: A possible structural basis for endothelial solute permeability. J Ultrastruct Res 1983;85:45–57. Leach L, Firth JA: Fine structure of the paracellular junctions of terminal villous capillaries in the perfused human placenta. Cell Tissue Res 1992;268:447–452. Furuse M, Hirase T, Itoh M, Nagafuchi A, Yonemura S, Tsukita S, Tsukita S: Occludin: A novel integral membrane protein localizing at tight junctions. J Cell Biol 1993;123:1777–1788. Furuse M, Fujita K, Hiiragi T, Fujimoto K, Tsukita S: Claudin-1 and -2: Novel integral membrane proteins localizing at tight junctions with no sequence similarity to occludin. J Cell Biol 1998;141:1539–1550. Dye JF, Leach L, Clark P, Firth JF: Cyclic AMP and acidic fibroblast growth factor have opposing effects on tight and adherens junctions in microvascular endothelial cells in vitro. Microvasc Res 2001;62:94–113. Lee SW, Kim WJ, Choi YK, Song HS, Son MJ, Gelman IH, Kim YJ, Kim KW: SSeCKS regulates angiogenesis and tight junction formation in blood-brain barrier. Nat Med 2003;9:900–906. Antonetti DA, Barber AJ, Khin SK, Lieth E, Tarbell JM, Gardner TW: Vascular permeability in experimental diabetes is associated with reduced endothelial occludin content. Diabetes 1998;47:1953–1959. Staddon JM, Herrenknecht K, Smales C, Rubin LL: Evidence that tyrosine phosphorylation may increase tight junction permeability. J Cell Sci 1995;108:609–619. Vittet D, Buchou T, Schweitzer A, Dejana E, Huber P: Targeted null-mutation in the vascular endothelial-cadherin gene impairs the organization of vascular-like structures in embryoid bodies. Proc Natl Acad Sci USA 1997;94:6273–6278. Gory-Faure S, Prandini MH, Pointu H, Roullot V, Pignot-Paintrand I, Vernet M, Huber P: Role of vascular endothelial cadherin in vascular morphogenesis. Development 1999;126:2093–2102. Wright TJ, Leach L, Shaw PE, Jones P: Dynamics of vascular endothelial-cadherin and ␤-catenin localisation by vascular endothelial growth factor-induced angiogenesis in human umbilical vein cells. Exp Cell Res 2002;280:159–168.

Vasculogenesis and Angiogenesis in the Diabetic Placenta

123

41

42 43

44 45 46 47 48 49 50

51

52

53 54

55 56

57 58

59

60 61 62

Leach L, Lammiman MJ, Babawale MO, Hobson SA, Bromolou B, Lovat S, Simmonds MJ: Molecular organization of tight and adherens junctions in the human placental vascular tree. Placenta 2000;21:547–557 Madri JA, Enciso J, Pinter E: Maternal diabetes: Effects on embryonic vascular development – A vascular endothelial growth factor-A-mediated process. Pediatr Dev Pathol 2003;6:334–341. Le Noble F, Moyon D, Pardanaud L, Yuan L, Djonov V, Matthijsen R, Breant C, Fleury V, Eichmann A: Flow regulates arterial-venous differentiation in the chick embryo yolk sac. Development 2004;131:361–375. Mayhew TM: Fetoplacental angiogenesis during gestation is biphasic, longitudinal and occurs by proliferation and remodelling of vascular endothelial cells. Placenta 2002;23:742–750. Augustin HG: Tubes, branches, and pillars. The many ways of forming a new vasculature. Circ Res 2001;89:645–647. Djonov V, Baum O, Burri PH: Vascular remodelling by intussusceptive angiogenesis. Cell Tissue Res 2003;314:107–117. Jackson MR, Carney EW, Lye SJ, Ritchie JW: Localization of two angiogenic growth factors (PDECGF and VEGF) in human placenta throughout gestation. Placenta 1994;15:341–353. Cooper JC, Sharkey AM, Charnock-Jones DS, Palmer CR, Smith SK: VEGF mRNA levels in placentae from pregnancies complicated by pre-eclampsia. Br J Obstet Gynaecol 1996;103:1191–1196. Shiraishi S, Nakagawa K, Kinukawa N, Nakano H, Sueishi K: Immunohistochemical localization of vascular endothelial growth factor in the human placenta. Placenta 1996;17:111–121. Geva E, Ginzinger DG, Zaloudek CJ, Moore DH, Byrne A, Jaffe RB: Human placental vascular development: Vasculogenic and angiogenic (branching and nonbranching) transformation is regulated by vascular endothelial growth factor-A, angiopoietin-1, and angiopoietin-2. J Clin Endocrinol Metab 2002;87:4213–4224. Kumazaki K, Nakayama M, Suehara N, Wada Y: Expression of vascular endothelial growth factor, placental growth factor, and their receptors Flt-1 and KDR in human placenta under pathologic conditions. Hum Pathol 2002;33:1069–1077. Zhang EG, Burton GJ, Smith SK, Charnock-Jones DS: Placental vessel adaptation during gestation and to high altitude: Changes in diameter and perivascular cell coverage. Placenta 2002;23: 751–762. Hammes HP, Lin J, Renner O, Shani M, Lundqvist A, Betsholtz C, Brownlee M, Deutsch U: Pericytes and the pathogenesis of diabetic retinopathy. Diabetes 2002;51:3107–3112. Khaliq A, Dunk C, Jiang J, Shams M, Li XF, Acevedo C, Weich H, Whittle M, Ahmed A: Hypoxia down-regulates placenta growth factor, whereas fetal growth restriction up-regulates placenta growth factor expression: Molecular evidence for ‘placental hyperoxia’ in intrauterine growth restriction. Lab Invest 1999;79:151–170. Khaliq A, Li XF, Shams M, Sisi P, Acevedo CA, Whittle MJ, Weich H, Ahmed A: Localisation of placenta growth factor (PlGF) in human term placenta. Growth Factors 1996;13:243–250. Leach L, Gray C, Staton S, Babawale MO, Gruchy A, Foster C, Mayhew TM, James DK: Vascular endothelial cadherin and beta-catenin in human fetoplacental vessels from pregnancies complicated by type 1 diabetes: Associations with angiogenesis and perturbed barrier function. Diabetologia 2004;47:695–709. Glover V, Leach L: Perfusion of human placental vessels with VEGF: Effect on leakage of 76 kDa dextran and junctional organisation (abstract). J Vasc Res 2003;40:306. Babawale MO, Lovat S, Mayhew TM, Lammiman MJ, James DK, Leach L: Effects of gestational diabetes on junctional adhesion molecules in human term placental vasculature. Diabetologia 2000;43:1185–1196. Jirkovská M, Kubínová L, Janácek J, Moravcová M, Krejcí V, Karen P: Topological properties and spatial organization of villous capillaries in normal and diabetic placentas. J Vasc Res 2002;39: 268–278. Teasdale F: Histomorphometry of the placenta of the diabetic woman: Class A diabetes mellitus. Placenta 1981;2:241–252. McParland P, Pearce JM: Doppler blood flow in pregnancy. Placenta 1988;9:427–450. Mayhew TM, Sørensen FB, Klebe JG, Jackson MR: Growth and maturation of villi in placentae from well-controlled diabetic women. Placenta 1994;15:57–65.

Leach/Mayhew

124

63

64 65 66 67 68

69 70 71 72 73 74

75

76 77

78

79

80

81 82

83

84

Mayhew TM: Enhanced fetoplacental angiogenesis in pre-gestational diabetes mellitus: The extra growth is exclusively longitudinal and not accompanied by microvascular remodelling. Diabetologia 2002;45:1434–1439. Teasdale F: Histomorphometry of the human placenta in class B diabetes mellitus. Placenta 1983;4:1–12. Teasdale F: Histomorphometry of the human placenta in class C diabetes mellitus. Placenta 1985;6:69–82. Mayhew TM, Kaufmann P, Charnock-Jones DS: Aspects of fetoplacental vasculogenesis and angiogenesis. III. Changes in complicated pregnancies. Placenta 2004;25:127–139. Mayhew TM, Sørensen FB, Klebe JG, Jackson MR: Oxygen diffusive conductances in placentae from control and diabetic women. Diabetologia 1993;36:955–960. Mamopoulos M, Bili H, Tsantali C, Assimakopoulos E, Mantalenakis S, Farmakides G: Erythropoietin umbilical serum levels during labor in women with preeclampsia, diabetes, and preterm labor. Am J Perinatol 1994;11:427–429. Caro J, Erslev AJ, Silver R, Miller O, Birgegard G: Erythropoietin production in response to anemia or hypoxia in the newborn rat. Blood 1982;60:984–988. Costa-Giomi P, Caro J, Weinmann R: Enhancement by hypoxia of human erythropoietin gene transcription in vitro. J Biol Chem 1990;265:10185–10188. Ribatti D, Vacca A, Roccaro AM, Crivellato E, Presta M: Erythropoietin as an angiogenic factor. Eur J Clin Invest 2003;33:891–896. Benyo DF, Conrad KP: Expression of the erythropoietin receptor by trophoblast cells in the human placenta. Biol Reprod 1999;60:861–870. Davis LE, Widness JA, Brace RA: Renal and placental secretion of erythropoietin during anemia or hypoxia in the ovine fetus. Am J Obstet Gynecol 2003;189:1764–1770. Beltramo E, Pomero F, Allione A, D’Alù F, Ponte E, Porta M: Pericyte adhesion is impaired on extracellular matrix produced by endothelial cells in high hexose concentrations. Diabetologia 2002;45:416–419. Lu M, Kuroki M, Amano S, Tolentino M, Keough K, Kim I, Bucala R, Adamis AP: Advanced glycation end products increase retinal vascular endothelial growth factor expression. J Clin Invest 1998;101:1219–1224. Okamoto T, Tanaka S, Stan AC, Koike T, Kase M, Makita Z, Sawa H, Nagashima K: Advanced glycation end products induce angiogenesis in vivo. Microvasc Res 2002;63:186–195. Hellström M, Gerhardt H, Kalén M, Li X, Eriksson U, Wolburg H, Betsholtz C: Lack of pericytes leads to endothelial hyperplasia and abnormal vascular morphogenesis. J Cell Biol 2001;153: 543–553. Janota J, Pomyje J, Toth D, Sosna O, Zivny J, Kuzel D, Stranak Z, Necas E, Zivny JH: Expression of angiopoietic factors in normal and type-I diabetes human placenta: A pilot study. Eur J Obstet Gynecol Reprod Biol 2003;111:153–156. Di Blasio AM, Carniti C, Vigano P, Florio P, Petraglia F, Vignali M: Basic fibroblast growth factor messenger ribonucleic acid levels in human placentas from normal and pathological pregnancies. Mol Hum Reprod 1997;3:1119–1123. Helske S, Vuorela P, Carpén O, Hornig C, Weich H, Halmesmäki E: Expression of vascular endothelial growth factor receptors 1, 2 and 3 in placentas from normal and complicated pregnancies. Mol Hum Reprod 2001;7:205–210. Hill DJ, Petrik J, Arany E: Growth factors and the regulation of fetal growth. Diabetes Care 1998;21(suppl 2):B60–B69. Aiello LP, Avery RL, Arrigg PG, Keyt BA, Jampel HD, Shah ST, Pasquale LR, Thieme H, Iwamoto MA, Park JE, et al: Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med 1994;331:1480–1487. Malecaze F, Clemens S, Simorre-Pinatel V, Mathis A, Chollet P, Favard C, Bayard F, Plouet J: Detection of vascular endothelial growth factor messenger RNA and vascular endothelial growth factor-like activity in proliferative diabetic retinopathy. Arch Ophthalmol 1994;112: 1476–1482. Khaliq A, Foreman D, Ahmed A, Weich H, Gregor Z, McLeod D, Boulton M: Increased expression of placenta growth factor in proliferative diabetic retinopathy. Lab Invest 1998;78:109–116.

Vasculogenesis and Angiogenesis in the Diabetic Placenta

125

85

86

87

88 89 90 91 92 93

Hill DJ, Tevaarwerk GJ, Caddell C, Arany E, Kilkenny D, Gregory M: Fibroblast growth factor 2 is elevated in term maternal and cord serum and amniotic fluid in pregnancies complicated by diabetes: Relationship to fetal and placental size. J Clin Endocrinol Metab 1995;80:2626–2632. Arany E, Hill DJ: Fibroblast growth factor-2 and fibroblast growth factor receptor-1 mRNA expression and peptide localization in placentae from normal and diabetic pregnancies. Placenta 1998;19:133–142. Burleigh DW, Stewart K, Grindle KM, Kay HH, Golos TG: Influence of maternal diabetes on placental fibroblast growth factor-2 expression, proliferation, and apoptosis. J Soc Gynecol Invest 2004;11:36–41. Martin A, Komada MR, Sane DC: Abnormal angiogenesis in diabetes mellitus. Med Res Rev 2003;23:117–145. Hovind P, Tarnow L, Oestergaard PB, Parving HH: Elevated vascular endothelial growth factor in type 1 diabetic patients with diabetic nephropathy. Kidney Int Suppl 2000;75:S56–S61. Speiser P, Gittelsohn AM, Patz A: Studies on diabetic retinopathy. III. Influence of diabetes on intramural pericytes. Arch Ophthalmol 1968;80:332–337. Darland DC, D’Amore PA: Blood vessel maturation: Vascular development comes of age. J Clin Invest 1999;103:157–158. Crocker DJ, Murad TM, Greer JC: Role of the pericyte in wound healing. An ultrastructural study. Exp Mol Pathol 1970;13:51–65. Beck L, D’Amore PA: Vascular development: Cellular and molecular regulation. FASEB J 1997;11:365–373.

Lopa Leach, MD School of Biomedical Sciences, E Floor, Queen’s Medical Centre University of Nottingham, Nottingham NG7 2UH (UK) Tel. ⫹44 115 9709430, Fax ⫹44 115 9709259 E-Mail [email protected] or [email protected]

Leach/Mayhew

126

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 127–143

Morphological Findings in Infants and Placentas of Diabetic Mothers Marina Kos, Martin Vogel Institute of Pathology, University of Zagreb Medical School, Zagreb, Croatia; Department of Pediatric Pathology and Placentology Charité, Humboldt University, Berlin, Germany

Infants of diabetic mothers have many problems, usually called diabetic embryopathy, i.e. fetopathy. Congenital malformations together with spontaneous abortions are usually summarized under the term diabetic embryopathy, while the term diabetic fetopathy encompasses increased incidence of perinatal death, increased somatic size (macrosomia) and hypertrophy of the islets of Langerhans with B cell hyperplasia and hyperinsulinaemia. Before the discovery of insulin and its widespread use in every day practice, many infants, but their mothers as well, have died during pregnancy complicated with diabetes. Close surveillance of diabetic pregnancy has resulted in a decrease of overall perinatal mortality to 2.8%, and even to less than 2% [1–3]. These figures are still 2- to 3-fold greater than in uncomplicated pregnancies. The risk of a lethal outcome for the infant depends upon the duration and intensity of the mother’s diabetes and also upon the degree of blood vessel damage it has caused. Increased perinatal mortality is the result of both increased intrauterine and neonatal mortality. Thirty years ago the main causes of neonatal death in infants from diabetic pregnancies were extreme prematurity, hyaline membrane disease, pulmonary changes unrelated to hyaline membranes, congenital malformations, infections (lung infections excluded) and changes due to asphyxia [3]. Despite many advances in the last 30 years, the rate of early and late fetal death and of neonatal mortality still remains higher than in uncomplicated pregnancies [2, 4–9]. The increased risk of fetal death begins during the third trimester, increasing gradually between 30 and 40 weeks’ gestation. The autopsy of stillborn infants often fails to reveal the cause of death [10]. Experiments have shown that hyperglycaemia in lambs causes hyperinsulinaemia, increased basal metabolism and hypoxaemia. We can assume that a similar reaction in human

fetuses leads to an increase in cardiac output and an abundant norepinephrine excretion together with a decrease in glucagon secretion. This pathophysiological mechanism may in humans also be the cause of unexplained intrauterine deaths in diabetic pregnancies [11]. However, strict control of maternal glycaemia has significantly reduced the stillbirth rate. In diabetic pregnancies there exists an increased risk of most congenital malformations, and the term ‘diabetic embryopathy and fetopathy’ has been long known in medicine [12]. In a large retrospective study carried out in Washington State, USA, the prevalence of congenital malformations was compared in infants of mothers with established pre-existing diabetes, gestational diabetes and the non-diabetic control group. In infants of mothers with pre-existing diabetes the prevalence of congenital malformations was 7.2%, in those of mothers with gestational diabetes 2.8%, while it was 2.1% in infants of healthy mothers [5]. These results differ from the results of studies in the United Kingdom and Poland: congenital malformations in infants whose mothers had pre-existing diabetes were found in 9.4 and 11.2%, respectively [7, 9]. On the other hand, a study from Germany showed congenital malformations in 2.1% of such infants [2]. A large prospective study done in 1990 has shown that infants of insulin-dependent mothers are 8 times more likely to have a major malformation [13]. A number of experts have tried to establish which phenotype is connected to diabetic embryopathy, but many of them came to the conclusion that this task is far from simple [14]. A more frequent appearance of cardiovascular malformations has been clearly proven (especially ‘double outlet right ventricle’ and common arterial trunk) [15], either isolated or in connection to vertebral defects and multisystem anomalies (VACTERL) [15–17]. ‘Double outlet right ventricle’ is a malformation where both great arteries (aorta and pulmonary artery) arise completely or mostly (⬎50%) from morphologically right ventricle. With rare exceptions, ventricular septal defects of different diameters and positions within the septum are also found; it is mostly perimembranous, followed in frequency by muscular and subarterial (doubly committed). In relation to blood vessels, these defects can be subaortic, subpulmonary, connected to both vessels or to none. There is always a muscular or fibrous connection between arterial and atrioventricular cusps; subvalvular, valvular or combined pulmonary stenosis is not rare as well as aortic arch defects. The common arterial trunk (truncus arteriosus communis) is a single arterial trunk that leaves the base of the heart giving rise to systemic and pulmonary arteries. The orifice of this vessel is usually closed by three semilunar valves, but their number can vary from two to six. There is always a ventricular septal defect, localized most frequently in such a position that the common arterial trunk overrides it. The coronary arteries are variable in respect to site and location of origin and connection. The general classification of the common arterial trunk

Kos/Vogel

128

is based on the origin of the pulmonary arteries. The VACTERL association consists of vertebral defects, anal atresia, cardiovascular anomalies, tracheooesophageal anomalies, and renal and limb defects [12]. These malformations are morphogenetically connected to ‘midline events’ that include fusion, segmentation, programmed cell death (apoptosis) with morphogenetic ‘necrosis’ and resorption, rotation and other morphogenetic events. Besides VACTERL and VATER association, caudal regression sequence (also known as ‘sirenomelia’) is also included in malformations that are caused by the disturbance of midline events. Caudal regression sequence usually consists of a fusion of lower limb buds during the 3rd gestational week, so that the body is reminiscent of a siren; there are always serious genitourinary malformations (hypoplasia or agenesis of internal and/or external genitalia, renal agenesis or multicystic renal dysplasia), imperforate anus, vertebral defects and also in 20–35% of the cases cardiovascular, respiratory and intestinal system malformations. For this reason, some consider VATER association to be just the lower stage of the caudal regression sequence [18]. Therefore, isolated congenital heart malformations, spina bifida, cleft lip, palate and facial clefts, isolated imperforate anus, holoprosencephaly, agenesis of corpus callosum and hypospadias are all included in the ‘midline defects’ [12]. Aforementioned malformations are considered ‘monotopic developmental field defects’, but ‘polytopic developmental field defects’ such as oculo-auriculovertebral anomalies that can be included into Goldenhar’s syndrome are also noticed in infants of diabetic mothers [19]. Goldenhar’s syndrome occurs sporadically and is caused by defects in the morphogenesis of the first and second branchial arch. It consists of hemifacial microsomia (caused by hypoplasia of the malar, maxillary or mandibular regions, especially the ramus and condyle of the mandible and temporomandibular joint), macrostomia and hypoplasia of the facial muscles. Microotia, accessory pre-auricular tags and pits, middle ear anomalies with various degrees of deafness, decreased secretion of saliva, anomalies of the tongue and soft palate and hemivertebrae also occur frequently [12]. The connection between major malformations and an increased level of HbA1c (that shows the glucose level, i.e. the quality of diabetes control) in maternal blood is proven [20]. Maternal hyperglycaemia during the early embryonal development represents a metabolically abnormal environment that can obviously influence processes during organogenesis. Experimental studies on mice have shown that hyperglycaemia during the pre-implantational phase acts on blastocysts by enhancing the expression of the gene that regulates apoptosis, i.e. programmed cell death. Morphological expression of this finding was a more pronounced fragmentation of DNA, while the apoptosis could be prevented by administering insulin to mice prior and immediately after conception [21]. Studies on animals have also shown the increase of enzymes that destroy free

Morphology of Infants and Placentas in Diabetes

129

oxygen radicals (responsible for cell damage) in the hyperglycaemic environment. It is thought that the oxidative stress (also held responsible for the occurrence of malformations) causes a greater need for these enzymes [22]. Subsequent administration of insulin could not stop the teratogenic influence of hyperglycaemic serum in rats [23]. Preconceptional diabetes in humans shows a strong connection to congenital cardiac malformations that arise early in pregnancy and also to cardiomyopathy (a fact that stresses the importance of adequate regulation of diabetes in the period before conception) [17]. Besides major malformations, minor malformations are also more frequent in infants of diabetic mothers. These malformations are, for example: pylonidal sinus, epicanthus, pre-auricular pendular fibromas, cleft tongue, accessory nipple, hypoplastic fingers or fingernails, and clinodactyly [24]. The prevalence of both major and minor malformations is connected to unsatisfactory control of maternal glycaemia during the first trimester of pregnancy [7, 20, 24]. A study showing almost the same prevalence of congenital malformations in infants of mothers with gestational diabetes as in those of healthy mothers has already been mentioned [5], but there exists quite a number of studies with similar results [7, 8, 25–27]. However, there are also studies whose results differ, showing that gestational diabetes can be a significant factor influencing the development of congenital malformations [28]. In gestational diabetes, as in pre-existing diabetes, it has to be stressed once again that the most important factor for the favourable outcome of pregnancy is the adequate control and surveillance of maternal glycaemia [29, 30]. Infants of mothers with gestational diabetes show some of the clinical features found in infants of mothers with pre-existing, insulin- dependent diabetes, such as increased perinatal mortality, macrosomia and hyperplasia of the islets of Langerhans. Macrosomia is a condition characterized by an increased quantity of fat tissue throughout the body and by enlarged viscera, particularly liver, heart and adrenals. The brain is often smaller than normal, while the thymus may show cortical involution (histologically ‘starry sky’ pattern because of the lysis of T cells that can be seen as increased rate of apoptosis in the cortical area while Hassal’s bodies become more apparent). Macrosomia does not appear before 25 weeks’ gestation, probably because of low fetal tissue sensitivity to insulin before this period. The causes of macrosomia and adiposity are the increased flow of glucose from the mother and the combination of fetal hyperglycaemia and fetal hyperinsulinaemia, leading to increased metabolic rate and hypoxaemia. Such a reaction by the fetus leads to the increased cardiac output, increased excretion of noradrenaline and obvious liberation of glucagon with conversion of glucose to fat and proteins. Fat cells store more and more fat, liver cells store more and more glycogen, asymmetric hypertrophy of the septum can develop in the heart, and the metabolic events in the lungs change,

Kos/Vogel

130

Fig. 1. Ischaemic necrosis in the myocardium. Myocardial fibres showing coagulative necrosis (arrows) are surrounded with preserved fibres. HE. ⫻400.

postponing the synthesis of mature surfactant. Because of all this, the aetiology of hypoglycaemia that develops in a macrosomic infant at the time of birth is complex. It involves hyperinsulinaemia, hypoglucagonaemia and, very probably, a reduced rate of gluconeogenesis and cortisol secretion. The increased rate of haemolysis can cause hyperbilirubinaemia, while the slower rate of parathyroid gland secretion increases the possibility of hypocalcaemia and hypomagnesaemia in the first 3 days of life. Abnormal levels of fibrinolytic activity inhibitors and PGE-like substance can cause thromboses, especially of the renal vein [31]. Because of the slower synthesis of pulmonary phospholipids that are important for surfactant activity respiratory distress syndrome can ensue [32]. Except for the lack of surfactant, pulmonary symptoms in infants of diabetic mother can be caused by pulmonary hypertension or heart failure [31]. Macrosomia can also complicate the birth process itself, causing dystocia or disproportion with asphyxia and birth traumas as consequences. Even in the absence of macrosomia, cardiomegaly is not rare in infants of diabetic mothers. Indeed, the transient hypertrophic cardiomegaly can be found in 30% of these neonates. There may be generalized ventricular hypertrophy with asymmetric hypertrophy of the septum which, if very severe, can lead to heart failure as the result of obstruction of outflow (hypertrophic obstructive cardiomyopathy). Histologically, the muscle fibres in the septum show disorganization and whorling as well as foci of necrosis (fig. 1). The precise reason for the development of this cardiomyopathy is not known, but it is thought that it might be caused by

Morphology of Infants and Placentas in Diabetes

131

a

b Fig. 2. Round, clear, enlarged and more numerous islets of Langerhans (arrows) within exocrine pancreatic parenchyma (a) and within periductal fibrous tissue (arrows) (b). HE. ⫻160.

the increased cardiac outflow in diabetes or by the direct action of insulin on fetal myocardium, which is very rich in insulin receptors [33]. The pancreas in infants of diabetic mothers shows hypertrophy of the islets of Langerhans with B cell hyperplasia. Maternal hyperglycaemia cannot be the only cause of this hypertrophy because it was found together with fetal hyperinsulinism as early as at 4 months’ gestation, while B cells become responsive to glucose levels after 24 weeks’ gestation [34]. However, ionic stimuli, leucine and arginine are known to stimulate fetal B cells from 14 weeks onwards, and maternal hyperaminoacidaemia has been reported even in the mildest forms of gestational diabetes [35, 36]. Interestingly, anencephalic infants of diabetic mothers fail to develop B cell hyperplasia, pointing to an ill-explained role of fetal hypothalamic-pituitary axis in B cell function [37]. Histological examination of the pancreas in infants of diabetic mothers shows an increased islet cell mass that is about 3 times normal (fig. 2a, b). This is caused by hypertrophy of the islets that are larger than normal, formation of new islets and hyperplasia of B cells (but also the increased number of non-B cells, but of a much lower degree). The B cells usually show an increase in the nuclear size and nuclear pleomorphism. Such findings can also be observed in cases of severe fetal erythroblastosis, but the ratio of B cells and other islet cells remains normal. Many necropsy studies have shown that even up to 65% of infants of diabetic mothers have cellular infiltration around, but not within some of the hypertrophied islets. The cells that infiltrate around the islets were lymphocytes, histiocytes (occasionally containing Charcot-Leyden crystals) and neutrophils, but eosinophils are thought to be specific for cellular infiltration around the pancreatic islets in these cases. In the aforementioned fetal erythroblastosis, haemopoetic cells can be found instead. Eosinophilic infiltrates are not found in

Kos/Vogel

132

other organs of infants of diabetic mothers, so this reaction seems to be organ specific, and it disappears a few days after birth. From about 11 days of age onwards, as the sequel of cellular infiltration, there may be fibrosis of the islets affecting up to 20% of the islet mass. Pancreatic changes are independent of the severity of maternal diabetes and may also be seen in infants whose mothers exhibit diabetes after many years [38]. One should always bear in mind that besides maternal diabetes and fetal erythroblastosis, there are some other conditions followed by endocrine pancreas hyperplasia. These conditions are: Beckwith-Wiedemann syndrome, sometimes Zellweger’s syndrome, neonatal hepatitis and, occasionally, intrauterine growth retardation. Diffuse and focal nesidioblastosis as well as pancreatic adenoma can also be associated with persistent hyperinsulinaemic hypoglycaemia of infancy [38]. Considering the fact that maternal diabetes is quite a frequent complication of pregnancy, placental alterations long ago caught the attention of morphologists. However, the results of a morphological examination of the placenta have to be viewed with caution, as substantial changes have occurred in diabetes control and the fetal outcome of diabetic pregnancies during the last five decades. We have to bear in mind that the morphological form of the placenta in diabetic mothers can vary considerably. This variability of the placenta has repeatedly been related to the stage of diabetes mellitus, the quality of the metabolic regime and, in particular, the glycaemic status during pregnancy [39–44]. An attempt has been made to find a clinical-morphological correlation not only for manifest diabetes mellitus, but also for clinically latent disturbances of the carbohydrate metabolism [45–47].

The Placenta in Overt Diabetes mellitus

The changes that can be determined in the placenta of diabetic mothers are not specific: they appear in other conditions as well. One especially frequent manifestation used to be the combination of an enlarged size of the placenta with an immaturity of the villi (fig. 3). This combination of findings was called ‘Placentopathia diabetica’ by Hormann [48]. Thirty years ago Vogel and Kloos [49] diagnosed this growth and maturation disturbance in a good half of all pregnancies of mothers with manifest diabetes. Today, however, these combined findings can only be seen rarely, a change that has probably resulted from the introduction of a strict metabolic regime for women with diabetes mellitus requiring insulin support. The placenta is considered to be of enlarged size when the weight and the basal plate are above the 90th percentile with respect to the week of pregnancy at birth [50–52]. Today, this enlarged size is rarely seen, and when it is diagnosed it is usually in diabetes stages White A, B and C

Morphology of Infants and Placentas in Diabetes

133

Fig. 3. A focus of immature placental villi that are not appropriate for gestational age (villous immaturity) (arrow). HE. ⫻400.

accompanied by an inadequate metabolic regime. In cases of vascularly complicated diabetes (stages D, R and H) the placenta can also be small and underweight (below the 10th percentile for the week of pregnancy at birth). The placental weight in cases of well-controlled maternal glycaemia does not differ from normal [53, 54]. The histological figure can differ from placenta to placenta, as well as within a single organ. Characteristic – but by no means pathognomonic – is the determination of an arrest in the process of villous maturation or a dissociated disturbance of villous maturation with the persistence of embryonal stromal structures (table 1). In the case of the arrest of villous maturation there is a focal disturbance of the development of the villi, with impaired branching, enlargement of the immature intermediate villi, frequent embryonic stroma and incomplete transformation of the stem villi. Here the grouped, enlarged villi are characteristic. They have a coarsely meshed ‘oedematous’ reticular or mesenchymal stroma: this is called ‘persisting embryonic stroma’. The vessels of the stem villi have only narrow media and the paravasal contractile sheath is very narrow and imperfectly developed [55, 56]. The number of myofibroblasts and smooth muscle cells is significantly reduced. The enlarged villi contain few capillaries, which only rarely have contact with the surface epithelium. The villous trophoblast, which is at many points only one layer thick, repeatedly contains nucleus-free epithelial plates and only few syncytiocapillary membranes. The term ‘villous oedema’, which is used in the literature for this disturbance of villi maturation is, in our estimation, inadequate. The changes described above cannot simply be explained as

Kos/Vogel

134

Table 1. Histological findings of the placenta in maternal diabetes mellitus (1992–2000) [unpubl. data from Vogel] Disorders of villous maturation, % Focal persistent embryonic villi Dissociated villous maturation, mostly immature Arrest of villous maturation, G2–G3 Retardation of villous maturation, G2–G3 Chorangiosis type I, G1–G3 Accelerated maturation

⬍20 about 10 ⬍5 ⬍5 ⬍3 about 10

Age-appropriate maturation, % Including dissociated villous maturation, mostly mature and chorangiosis type II

about 50

Additional findings Inter-, perivillous fibrin deposits, G2–G3 Syncytial knotting Villous fibrosis, G1–G3 Obliterative vasculopathy Perivillitis

about 20% about 25% about 155% ⬍5% ⬍5%

the result of a stromal oedema. We agree with Benirchke and Kaufmann [52] that this is a persisting villous immaturity, which frequently is present only in small foci. Further, the histological form changes from region to region in the majority of placentas. One can often find villi with a reduced vessel component, as well as villi with an excessive supply of vessels (chorangiosis) and still others with an age-appropriate maturity status, right next to each other [48, 50, 57, 58]. In a good 50% of the placentas no villous immaturity can be determined. The villi show normal development for the stage of pregnancy. In the placentas of premature children villi can even show accelerated maturation. The advanced maturation stage mentioned above, chorangiosis, as well as microfibrin deposits and fibrosis of the villous stroma, with and without endangiopathia obliterans, are encountered more frequently in the White stages D, R and H. On the surfaces of immature villi syncytial sprouts can occur; at a more advanced stage of maturation one finds regressively changed knots (apoptotic and syncytial knotting) [52]. The irregularly appearing cytotrophoblast cells in the villous trophoblasts are mainly of the intermediate type. However, mitoses cannot be regularly determined [58]. In recent years we have not observed an increased incidence of thrombosis in the fetal arteries (fig. 4). There was no significant incidence of acute funisitis in the placentas of diabetic mothers [59, 60]. We did, however, find in nearly 5% of the cases an obliterative vasculopathy in the chorionic plate and in vessels of the stem villi.

Morphology of Infants and Placentas in Diabetes

135

Fig. 4. Thrombosed blood vessel within the chorionic plate. The lumen is completely obliterated with fresh thrombus, consisting mainly of erythrocytes and fibrin (arrow). HE. ⫻400.

The results of histometric investigations give no uniform figure. Stolz et al. [61] found evidence of a delayed maturation of the villi, above all in the placentas of women with diabetes mellitus of the White stage D. Boyd et al. [62] and Teasdale [42] determined an increase in the villi surface. Teasdale also determined an increase in the villi vessel surface in the placentas of mothers with diabetes of the White stages B and C [41, 42]. Bjork and Persson [63] found an increase in the exchange surfaces. Jirkovska [64] reports a clear narrowing of the basal membrane of the villi capillaries in the placentas of women with type 1 diabetes as compared with that of healthy women. Our histometric findings showed an increase in the villi surface and the fetal capillaries only in placentas of term fetuses in women with type 1 diabetes mellitus; we obtained no such finding with premature births [65]. In addition, we found no significant correlation between the histometric placenta parameters and the health status of the children. Electron microscope investigation showed that the basal membrane of the fetal capillaries and the venules in terminal villi and intermediate villi of the peripheral type are more frequently irregularly widened. Furthermore, there was an increase in microvesicles along the basal membrane and regularly a significant increase in microfilaments in the cytoplasm of the endothelia. Jones and Fox [66] determined that there were necroses distributed throughout the syncytiotrophoblasts, as well as hyperplasia of the villous cytotrophoblasts; in addition they found a focal thickening of the basal membrane. In the endothelial cells and pericytes of the villar capillaries Jones and Desoye [67] discovered

Kos/Vogel

136

more glycogen in the placentas of women with diabetes than in the placentas of women with gestational diabetes or of women with normal metabolism. Laurini et al. [68] determined the persistence of ultrastructural changes in the diabetic placenta, with thickening of the basal membrane, proliferation of endothelial cells and an increased concentration in collagen fibres, despite strict monitoring and control of the condition. They drew the conclusion that the main change was a relative immaturity of the organ. The increase in cytotrophoblast thickness and syncytiotrophoblast knots as well as focal thickening of the basal membrane imply that the trophoblast is damaged in diabetic pregnancies, because the syncytiotrophoblast undergoes necrosis with subsequent proliferation of cytotrophoblasts [69]. The precise cause of this defect is, however, unknown. If uteroplacental blood flow is normal, it is probably related to a metabolically abnormal environment surrounding the villi (maternal blood in the intervillous space). It has been proved that the placental tissue produces free oxygen radicals that stimulate fibroblasts to an increased formation of intercellular matrix. This means that in these cases postplacental hypoxia is an important factor [70]. Further, in the cell membrane of the villous trophoblasts facing the intervillous space specific insulin-binding antibodies can be found. Normally, insulin does not pass the placental barrier. But, it has been shown that an increase in antibody-bound insulin is associated with fetal macrosomia [52]. Our experience shows that it is not possible to establish a significant statistical correlation between certain placental changes and clinical stages of diabetes mellitus according to White. Earlier studies showed that the severity of the villi immaturity increases over the White stages A–C and that it is also more frequent and strongly evident over these stages [41, 42, 54, 71]. But, this can no longer be found, or only minimally or irregularly, in placentas of women undergoing a strict metabolic regime with a glycolized haemoglobin value (HbA1) below 7% and normal glycaemia during pregnancy [65]. The change in therapy with a consistent monitoring of the metabolism and maintenance of a metabolic regime has led to improvement in the maturation of the placenta. Histologically, the placentas of women undergoing such therapy frequently do not differ from the fully mature placentas of non-diabetic mothers [44]. This means that normal maturation of the placenta is an important indication of an optimal metabolic regime with normoglycaemia, both preconceptually and during pregnancy. However, Laurini et al. [72] found no correlation between very strict treatment of diabetic mothers and changes in the placenta. They treated diabetic women requiring insulin of the White stages B, C and D with a continuous subcutaneous infusion of insulin and carried out very frequent metabolic tests. But they were unable to find any significant differences between the placentas of women whose blood sugar level was monitored very closely as compared with women who underwent only few such tests. It was also possible to determine pathological findings in the placentas of

Morphology of Infants and Placentas in Diabetes

137

Fig. 5. Chorangiosis, type II. Tertiary and terminal villi are mildly to moderately enlarged and contain too many (congested) capillaries (arrows), while formation of syncytiocapillary membranes is adequate. HE. ⫻160.

women with type 1 diabetes mellitus despite normoglycaemia of the mothers. One has to be careful, however, when comparing the different individual studies. The collectives investigated by different authors differ greatly with respect to stage of the diabetes, metabolism monitoring and stage of pregnancy. For statistical reasons as well, very heterogenous investigatory groups were formed. The minimal information required with regard to mothers consists of the stage of the disease and possible complications, the quality of the metabolic monitoring and the regime, the management and time of birth. Complicated pregnancy-induced hypertension and chronic hypertensive or diabetic angiopathy in the patient can alter the histological form. The form is determined by villi corresponding to the stage of pregnancy or prematurely matured. There are also cases of villi which are excessively vascularized but with adequate formation of syncytiocapillary membranes (type II chorangiosis) (fig. 5), while immature villi are lacking or can only be determined in residual amounts. An increase in syncytial knotting, intervillous and perivillous microfibrin deposits and micro-infarcts, as well as focal fibrosis of the villous stroma and endangiopathia obliterans, can also affect the form [51, 52, 56].

The Placenta in Gestational Diabetes

Despite the differences in the frequency of pathological placental changes in manifest diabetes mellitus, it is clear that disturbances in glucose metabolism

Kos/Vogel

138

Table 2. Percent of placentas with disorders of villous maturation in pregnancies with normal, borderline (IGT) and pathological glucose tolerance (GDM) (n ⫽ 325) [74] Glucose tolerance

Normal IGT GDM

n

95 76 154

Arrest of maturation (%)

9 24 32

Disorders of villous maturation (%) prevalence of immaturity

prevalence of maturity

14 13 20

77 63 48

in pregnancy result in an increase in disturbances of villous maturation of various types and forms [52, 54, 59, 65, 73, 74]. Both with manifest and latent metabolic disturbances, the incidence and forms of maturation defects depend on the blood sugar levels of the mother during pregnancy [44, 46]. This means that we should expect an increase in disturbances of villous maturation in cases of gestational diabetes as well [47, 60, 75, 76]. The most severe form of villous immaturity, an arrest of villous maturation, correlates with a high incidence of macrosomia (48%), intrauterine/subpartal hypoxia (38%), intrauterine death of the fetus (9%) and an increase in neonatal mortality (5%) [51]. Our work group determined that there was a close connection between disturbances of villous maturation and the result of the maternal oral glucose test (75 g OGTT). Table 2 shows that with normal OGTT we determined in 9% of the cases an arrest of villous maturation and in a further 14% a lesser degree of villous immaturity. In 77% of the cases the villi had matured corresponding to the stage of pregnancy. Even in cases with only a single pathological value in the OGTT (IGT ⫽ borderline maternal glucose tolerance), we found a 24% arrest in villi maturation, which was almost 3 times as high as in cases with a normal OGTT. The greatest proportion of moderately severe to severe cases of a disturbance of villous maturation was seen in gestational diabetes. Only 48% of these placentas were almost fully matured. There was a highly significant correlation between the severity of the mother’s glucose tolerance test, as diagnosed by means of OGTT, and the appearance of a disturbance in villi maturation in the placenta (p ⫽ 0.02) [74]. Further, the disturbance of maturation showed a correlation with the concentration of insulin in the amniotic fluid and in the blood of the umbilical cord, as well as with the fetal parameters of glucose metabolism [77, 78]. The fetal insulin level in the amniotic fluid had a significant correlation with the placenta morphology (p ⫽ 0.02), but the morphology did not correlate significantly with the level of insulin in the blood in the umbilical cord (p ⫽ 0.09). Since 50% of

Morphology of Infants and Placentas in Diabetes

139

the placentas of pregnant women with gestational diabetes demonstrated a disturbance in villous maturation and over one third of the placentas of mothers with borderline glucose tolerance also showed a disturbance of villous maturation, the possibility should be considered of monitoring pregnant women with IGT as closely as women with gestational diabetes, with respect to both the metabolism of the mother and the status of the fetus [74].

References 1 2

3 4 5

6 7

8 9 10 11 12

13 14

15

16 17 18

Somville T: Diabetes and pregnancy. Z Geburtshilfe Perinatol 1990;194:51–57. Von Kries R, Kimmerle R, Schmidt JE, Hachmeister A, Bohm O, Wolf HG: Pregnancy outcomes in mothers with pregestational diabetes: A population-based study in North Rhine (Germany) from 1988 to 1993. Eur J Pediatr 1997;156:963–967. Hubbell JP, Muirhead DM, Drorbaugh JE: The new-born infant of the diabetic mother. Med Clin North Am 1965;49:1035–1052. Damm P, Molsted-Pedersen L: Significant decrease in congenital malformations in newborn infants of an unselected population of diabetic women. Am J Obstet Gynecol 1989;161:1163–1167. Janssen PA, Rothman I, Schwartz SM: Congenital malformations in newborns of women with established and gestational diabetes in Washington State, 1984–91. Paediatr Perinat Epidemiol 1996;139:52–63. Botta RM: Congenital malformations in infants of 517 pregestational diabetic mothers. Ann Ist Super Sanita 1997;33:307–311. Mironiuk M, Kietlinska Z, Jezierska-Kasprzyk K, Piekosz-Orzechowska B: A class of diabetes in mother, glycemic control in early pregnancy and occurrence of congenital malformations in newborn infants. Clin Exp Obstet Gynecol 1997;24:193–197. Aberg A, Westbom L, Kallen B: Congenital malformations among infants whose mothers had gestational diabetes or preexisting diabetes. Early Hum Dev 2001;61:85–95. Casson IF, Clarke CA, Howard CV, et al: Outcomes of pregnancy in insulin dependent diabetic women; results of a five year population cohort study. BMJ 1997;315:275–278. Warren S, Le Compte PM, Legg MA: The Pathology of Diabetes mellitus, ed 4. Philadelphia, Lea & Febiger, 1966, pp 406–433. Milley JR, Rosenberg AA, Jones MD, Simmons MA: Circulating effects of hyperinsulinemia in the near term ovine fetus (abstract). Pediatr Res 1981;15:672. Gilbert-Barness E, Opitz JM: Congenital anomalies and malformation syndromes; in Dimmick JE, Kalousek D (eds): Developmental Pathology of the Embryo and Fetus. Philadelphia, Lippincott, 1992, pp 73–115. Becerra JE, Khoury MJ, Cordero JF, Erickson JD: Diabetes mellitus during pregnancy and the risks for specific birth defects: A population-based case-control study. Pediatrics 1990;85:1–9. Khoury MJ, Becerra JE, Cordero JF, Erickson JD: Clinical-epidemiologic assessment of pattern of birth defects associated with human teratogens: Application to diabetic embryopathy. Pediatrics 1989;84:658–665. Ferencz C, Rubin JD, McCarter RJ, Clark EB: Maternal diabetes and cardiovascular malformations: Predominance of double outlet right ventricle and truncus arteriosus. Teratology 1990;41: 319–326. Erickson JD: Risk factors for birth defects: Data from the Atlanta Birth Defects Case-Control Study. Teratology 1991;43:41–51. Loffredo CA, Wilson PD, Ferencz C: Maternal diabetes: An independent risk factor for major cardiovascular malformations with increased mortality of affected infants. Teratology 2001;64:98–106. Stocker JT, Heifetz SA: Sirenomelia: A morphological study of 33 cases and review of the literature. Perspect Pediatr Pathol 1987;10:7–50.

Kos/Vogel

140

19

20

21 22 23 24

25 26 27 28

29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Ewart-Toland A, Yankowitz J, Winder A, Imagire R, Cox VA, Aylsworth AS, Golabi M: Oculoauriculovertebral anomalies in children of diabetic mothers. Am J Med Genet 2000;90: 303–309. Miller E, Hare JW, Cloherty JP, Dunn PJ, Gleason RE, Soeldner JS, Kitzmiller JL: Elevated maternal hemoglobin A1c in early pregnancy and major congenital anomalies in infants of diabetic mothers. N Engl J Med 1981;304:1331–1334. Moley KH, Chi MM, Knudson CM, Korsmeyer SJ, Mueckler MM: Hyperglycemia induces apoptosis in pre-implantation embryos through cell death effector pathways. Nat Med 1998;4:126–127. Forsberg H, Borg LA, Cagliero E, Eriksson UJ: Altered levels of scavenging enzymes in embryos subjected to a diabetic environment. Free Radic Res 1996;24:451–459. Wentzel P, Eriksson UJ: Insulin treatment fails to abolish the teratogenic potential of serum from diabetic rats. Eur J Endocrinol 1996;134:459–466. Rosenn B, Miodovnik M, St John Dignan P, Siddiqi TA, Khoury J, Mimouni F: Minor congenital malformations in infants of insulin-dependent diabetic women: Association with poor glycemic control. Obstet Gynecol 1990;76:745–749. Kalter H: The non-teratogenicity of gestational diabetes. Paediatr Perinat Epidemiol 1998;12: 456–458. Bartha JL, Martinez-Del-Fresno P, Comino-Delgado R: Gestational diabetes mellitus diagnosed during early pregnancy. Am J Obstet Gynecol 2000;182:346–350. Kalter H: Gestational diabetes mellitus and congenital malformations (letter). Am J Obstet Gynecol 2001;184:246–247. Martinez-Frias ML, Bermejo E, Rodriguez-Pinilla E, Prieto L, Frias JL: Epidemiological analysis of outcomes of pregnancy in gestational diabetic mothers. Am J Med Genet 1998;78: 140–145. Thompson D, Densereau J, Creed M, Ridell L: Tight glucose control results in normal perinatal outcome in 150 patients with gestational diabetes. Obstet Gynecol 1994;83:362–366. Chia YT, Chua S, Thai AC, Kek LP, Ratnam SS: Gestational diabetes: Obstetric and neonatal outcome in 411 cases. Singapore Med J 1996;37:591–594. Morriss FH Jr: Infants of diabetic mothers. Fetal and neonatal pathophysiology. Perspect Pediatr Pathol 1984;8:223–234. Cunningham MD, Desai NS, Thompson SA, Greene JM: Amniotic fluid phosphatidyl glycerol in diabetic pregnancies. Am J Obstet Gynecol 1978;131:719–724. Gutgesell HP, Spear M, Rosenberg HS: Characterisation of the cardiomyopathy of infants of diabetic mothers. Circulation 1980;61:441–450. Bloodworth JBM: Endocrine Pathology. Baltimore, Williams & Wilkins, 1982, p 625. Milner RDG, Barson AJ, Ashworth MA: Human fetal pancreatic insulin secretion in response to ionic and other stimuli. J Endocrinol 1971;51:323–332. Milner RDG: Amino acids and beta cell growth; in Merkatz IR, Adam PAJ (eds): The Diabetic Pregnancy. New York, Grune & Stratton, 1979, pp 145–153. Van Assche FA, Gepts W, De Gasparo M: The endocrine pancreas in anencephalics: A histological and biochemical study. Biol Neonate 1969;14:374–388. Gray E: The endocrine system; in Keeling JW (ed): Fetal and Neonatal Pathology, ed 3. London, Springer, 2001. Bjork O, Persson B: Placental changes in relation to degree of metabolic control in diabetes mellitus. Placenta 1982;3:367–378. Emmrich P, Godel E: Morphologie der Plazenta bei mütterlichem Diabetes mellitus. Ergebnisse morphologischer Untersuchungen. Zentralbl Allg Pathol Pathol Anat 1972;116:56–63. Teasdale F: Histomorphometry of human placenta in class B diabetes mellitus. Placenta 1983;4:1–12. Teasdale F: Histomorphometry of the human placenta class C diabetes mellitus. Placenta 1985;6:69–82. Stoz F, Schuhmann RA, Schmid A: Morphometric investigations in terminal villi of placentas in diabetics in relation to White classification. J Perinat Med 1987;15:193–198. Semmler K, Emmrich P: Morphologie der Plazenta in Relation zur Glykämielage in der Schwangerschaft beim Diabetes mellitus. Z Geburtshilfe Perinatol 1989;193:124–128.

Morphology of Infants and Placentas in Diabetes

141

45

46

47 48 49

50 51 52 53 54 55 56 57 58 59

60 61 62 63 64 65

66 67 68 69

Schafer-Graf UM, Dupak J, Vogel M, Dudenhausen JW, et al: Hyperinsulinism, neonatal obesity and placental immaturity in infants born to women with one abnormal glucose tolerance test value. J Perinat Med 1998;26:27–36. Semmler K, Emmrich P, Fuhrmann K, Godel E: Reifungsstörungen der Plazenta in Relation zur Qualität der metabolischen Kontrolle während der Schwangerschaft beim insulinpflichtigen Gestationsdiabetes. Zentralbl Gynäkol 1982;104:1494–1502. Stoz F, Schuhmann RA, Schultz R: Morphohistometric investigations in placentas of gestational diabetes. J Perinat Med 1988;16:205–209. Hormann G: Zur Systematik einer Pathologie der menschlichen Plazenta. Arch Gynäkol 1958;191:297–344. Vogel M, Kloos K: Diabetes in der Schwangerschaft; neue morphologische Befunde. Plazenta und Fetus; in Dudenhausen JW, Saling E (eds): Perinatale Medizin. Stuttgart, Thieme, 1975, vol 6, pp 14–16. Vogel M: Pathologie der Schwangerschaft, der Plazenta und des Neugeborenen; in Remmele W (ed): Pathologie. Berlin, Springer, 1984, vol 3, p 510. Vogel M: Atlas der morphologischen Plazentadiagnostik, ed 2. Berlin, Springer, 1996, pp 82–93, 223–228. Benirchke K, Kaufmann P: Pathology of the Human Placenta, ed 4. New York, Springer, 2000, pp 437–460, 534–538. Clarson C, Tevaarwerk GJM, Harding PGR, Chance GW, Haust MD: Placental weight in diabetic pregnancies. Placenta 1989;10:275–281. Vogel M: Plakopathia diabetica. Entwicklungsstörungen der Plazenta bei Diabetes mellitus der Mutter. Virchows Arch (A) 1967;346:212–223. Graf R, Schonfelder G, Muhlberger M, Gutsmann M: The perivascular contractile sheath of human placental stem villi; its isolation and characterisation. Placenta 1995;16:57–66. Fox H: Pathology of the Placenta, ed 2. London, Saunders, 1997, pp 216–224. Altschuler G: Chorangiosis: An important placental sign of neonatal morbidity and mortality. Arch Pathol Lab Med 1984;108:71–74. Ogino S, Redline RW: Villous capillary lesions of the placenta: Distinction between chorangioma, chorangiomatosis and chorangiosis. Hum Pathol 2000;31:945–954. Salafia C: The fetal, placental and neonatal pathology associated with maternal diabetes mellitus; in Reece E, Coustan D (eds): Diabetes mellitus in Pregnancy. Principles and Practice. New York, Livingstone, 1988, pp 143–181. Salafia C, Weigl C, Silberman L: The prevalence and distribution of acute placental inflammation in uncomplicated term pregnancies. Obstet Gynecol 1989;73:383–389. Stoz F, Schuhmann RA, Schultz R: Morphohistometric investigations of placenta of diabetic patients in correlation to the adjustment of the disease. J Perinat Med 1988;16:211–216. Boyd P, Scott A, Keeling J: Quantitative structural studies on placentas from pregnancies complicated by diabetes mellitus. Br J Obstet Gynaecol 1986;93:31–35. Bjork O, Persson B: Villous structure in different parts of the cotyledon in placentas of insulindependent diabetic mothers. Acta Obstet Gynaecol Scand 1984;63:37–43. Jirkowska M: Comparison of the thickness of the capillary basement membrane of the human placenta under normal conditions and in type I diabetes. Funct Dev Morphol 1991;1:9–16. Steldinger R, Weber B, Jimenez E, Vogel M: Morphologische Untersuchungen von Plazenten bei Typ I Diabetes mellitus; in Dudenhausen JW, Saling E (eds): Perinatale Medizin. Stuttgart, Thieme, 1991, vol 13, pp 235–237. Jones CJP, Fox H: An ultrastructural and ultrahistochemical study of the placenta of the diabetic women. J Pathol 1976;119:91–99. Jones CJP, Desoye G: Gycogen distribution in the capillaries of the placental villus in normal, overt and gestational diabetic pregnancy. Placenta 1993;14:505–517. Laurini RN, Visser GHA, van Ballegoodie E: Morphological fetoplacental abnormalities despite well controlled diabetic pregnancy. Lancet 1984;i:800. Younes B, Baez-Giangreco A, al-Nuaim L, al-Hakeem A, Abu Talib Z: Basement membrane thickening in the placentae from diabetic women. Pathol Int 1996;46:100–104.

Kos/Vogel

142

70 71 72

73 74

75 76 77 78

Stanek J, Eis AL, Myatt L: Nitrotyrosine immunostaining correlates with increased extracellular matrix: Evidence of postplacental hypoxia. Placenta 2002;22(suppl A):56–62. Horky Z, Fischer U, Wappler E: Schwangerschaft bei Diabetes mellitus. Stuttgart, Fischer, 1969. Laurini RN, Visser GHA, van Ballegoodie E, Schoots CJF: Morphological findings in placentae of insulin-dependent diabetic patients treated with continuous insulin infusion (CSII). Placenta 1987;8:153–165. Diamant Y: The human placenta in diabetes mellitus. A review. Isr J Med Sci 1991;27:493–497. Schafer U, Vogel M, Unger M, Dupak J, Vetter K: Zottenreifungsstörungen der Plazenta beim Gestationsdiabetes in Abhängigkeit von mütterlichen und fetalen Parametern des Glukosestoffwechsels. Geburtsh Frauenheilk 1997;57:241–245. Desoye G: The Human Placenta in Gestational Diabetes. Springer, Vienna, 1988, pp 72–86. Figueroa R, Omar H, Tejani N, Wolin M: Gestational diabetes alters human placental vascular responses to changes in oxygen tension. Am J Obstet Gynecol 1993;168:1616–1622. Buchanan T, Kjos S, Montoro M, Wu P, et al: Use of fetal ultrasound to select metabolic therapy for pregnancies complicated by mild gestational diabetes. Diabetes Care 1994;17:275–283. Berkus MD, Langer O: Degree of glucose abnormality correlates with neonatal outcome. Obstet Gynecol 1993;81:344–348.

Marina Kos, MD, PhD Assistant Professor, Institute of Pathology University of Zagreb Medical School HR–10000 Zagreb (Croatia) Tel. ⫹385 1378 7437, Fax ⫹385 1378 7244, E-Mail [email protected]

Morphology of Infants and Placentas in Diabetes

143

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 144–152

Principles of Evidence-Based Medicine The Lesson for Diabetic Pregnancy

Thomas R. Pieber Department of Internal Medicine, Diabetes and Metabolism, Medical University Graz, Graz, Austria

Clinical evidence from modern research has become the major contributor to the clinical management of gestational diabetes. This chapter emphasizes the principles of evidence-based medicine, how evidence-based medicine is practiced and how it could be implemented for the diagnosis, prognosis, prevention, and treatment of gestational diabetes mellitus and its complications. Evidencebased medicine does not disparage the importance of basic research. Applied research clearly and necessarily builds on relevant studies of biology, physiological mechanisms, and pathophysiology [1, 2]. This basic research may provide some support for clinical decisions, but it is not designed to reveal the benefits and hazards of using one diagnostic test over another, estimate the incidence and risk of a consequence of disease in an individual from the clinic or the general population, or estimate the effect of an intervention on clinically important outcomes in typical individuals in the clinic or community. Such evidence can only be obtained from practical, clinical studies in patients, designed to answer these questions [3]. In the daily clinical work evidence from research does not make decisions, but clinicians and patients do. Clinical assessment and decision making should integrate the best external evidence, the medical experience of the clinician (also referred to as ‘internal evidence’) and the patient’s own preferences, expectations and values. In the clinical assessment of the patient, the clinician applies his or her expertise to collect and interpret evidence from the history, physical examination, diagnostic tests, social conditions and comorbidity, as well as other factors. It is one of the challenges for clinicians and health care providers in the future to practice medicine according to the principles of evidence-based medicine [4].

Where Does Evidence-Based Medicine Come from?

Evidence-based medicine is not new. In fact, it can be tracked down to the ancient times. The requirement that clinicians know what they are doing, more or less, is an ancient one and it has always been understood as a moral imperative. The Hippocratic Oath may be read as a request in this direction, for teachers and education, as well as a categorization of duties and virtues. It also includes the vow not to practice beyond one’s knowledge or capacity [5]. The English physician Thomas Beddoes (1760–1808) was not only recognized for his research in the medical use of gases, but also for his criticism of the medical practice in his time. In a letter to the authorities, Thomas Beddoes argued that 18th century medicine has become hidebound, stagnant, and secretive [6]. His proposed solutions (data sharing, collecting and archiving, analysis and reporting, publishing) are to a large extent identical with the goals and principles of evidence-based medicine. Beddoes was calling for such a system because he was convinced that the science of his day was often harming patients and could be better. He was suggesting a moral link between information management and medical practice, proposing outcome research and envisioning systematic reviews [6]. Some decades later, in 1834, Pierre Charles Alexandre Louis (1787–1872), published his ‘numerical method’, a keystone in the history of clinical evaluation. To get an accurate knowledge of any disease it is necessary to study a large series of cases, in all particular details. Louis performed the first chart reviews and thereby produced evidence to undermine beliefs about blood letting [7]. Louis’ method, which was simple and self-evident, still represents the base of modern outcome research. Despite the fundamental insights of Boddeos and Louis most publications for the next 100 years were related to case reports and observations. It was not before the middle of the 20th century that in medical science the gold standard of randomized controlled trials (RTC) was developed. But what failed to evolve was a system to make the information reliable and broadly available. ‘It is surely a great criticism of our profession that we have not organised a critical summary, by specialty or subspecialty, adapted periodically, of all the relevant randomized controlled trials’ Archie Cochrane (1909–1988) noted in 1979. Cochrane was not asking for more research or evidence, he was merely observing that the evidence we already have is removed or disconnected from the people who ought to be using it to take care of sick people. His core idea finally resulted in the Cochrane Collaboration, a worldwide network with the overall aim of exactly developing what Cochrane was asking for. Finally, the modern concept of evidence-based medicine evolved, based on the theoretical and practical work of David Sackett.

Principles of Evidence-Based Medicine

145

What Is Evidence-Based Medicine?

‘Evidence-based medicine is the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients’ [1]. The practice of evidence-based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research. Individual clinical expertise relates to the proficiency and judgement that individual clinicians acquire through clinical experience and clinical practice. Increased expertise and medical proficiency are reflected in different ways, but especially in more effective and efficient diagnoses, accurate treatment decisions and in the more thoughtful identification and compassionate use of individual patients’ preferences and wishes in making clinical decisions about their care. The best available external clinical evidence relates to clinically relevant research, often from the basic sciences of medicine, but especially from patientcentered clinical research. The best available external evidence relates to the accuracy and precision of diagnostic tests (including the clinical examination), the power of prognostic markers, and the efficacy and safety of therapeutic, rehabilitative, and preventive regimens. External clinical evidence both invalidates previously accepted diagnostic tests and treatments and replaces them with new ones that are more powerful, more accurate, more efficacious, and safer [4]. Good doctors use both individual clinical expertise and the best available external evidence, and neither on its own is enough. Without clinical expertise, practice risks becoming tyrannized by evidence, for even excellent external evidence may be inapplicable to or inappropriate for an individual patient. Without current best evidence, practice risks becoming rapidly out of date, to the detriment of patients.

What Is It Like to Practice Evidence-Based Medicine?

Evidence-based medicine is a process of lifelong, problem-based learning. The process involves: (1) converting the need for information into focused questions, (2) efficiently tracking down the best evidence with which to answer the question, (3) critically appraising the evidence for validity and clinical usefulness, (4) applying the results in clinical practice, and (5) evaluating the performance of the evidence in clinical application. Converting the Need for Information into Focused Questions The inability to ask a focused and precise clinical question can be a major impediment to evidence-based practice. Clinical questions should be

Pieber

146

clarified and best potential sources of evidence considered doing the following: (1) Ask: ‘Is this a question about foreground or background knowledge?’ Background knowledge questions are general questions about conditions, illnesses, syndromes and patterns of disease, and pathophysiology. They are usually composed of a question root (what, where, why, when, how) ⫹ a verb ⫹ a condition. For example, ‘What is the typical clinical presentation of preeclampsia?’ A novice more commonly asks this type of question in a particular knowledge area, in order to gain a general understanding of clinical issues. Best resources include evidence-based textbooks and reviews. Foreground questions are more often about issues of care. They query specialized and distinct knowledge needed for specific and relevant clinical decision making. Components of a well-built foreground question include ‘PICO’ (see below). Best resources may include an evidence-based abstraction service, guidelines, systematic reviews, or some evidence-linked textbooks, but may also include the primary literature. The need for skills in searching and critical appraisal is greatest when searching for evidence in the primary literature. (2) Pay attention to the question’s component parts, especially for foreground questions. A useful mnemonic is ‘PICO’: (i) P: patients or populations, (ii) I: interventions: (iii) C: comparison group(s) or ‘gold standard’, and (iv) O: outcome(s) of interest. (3) Classify the question into a domain: (i) therapy, (ii) diagnosis, (iii) prognosis, and (iv) harm or causality. (4) Ask: ‘How likely is it that there are high quality summaries or studies with valid and clinically important evidence specifically addressing this issue?’ Efficiently Tracking down the Best Evidence with Which to Answer the Question The most important structured review database is that produced by the Cochrane Collaboration (www.cochrane.org). Established in Britain in 1992 under the leadership of Dr. Iain Chalmers, it is presently composed of numerous centers. Its goals include the creation, maintenance, and dissemination of high quality systematic reviews of RCT. Reviews involve exhaustive searches for all RCT, both published and unpublished, on a particular topic. The studies are analyzed using standardized methodology and meta-analysis. The current database includes more than 2,100 completed reviews in internal medicine, and another 1,500 ‘under construction’. The reviews are mainly on therapies, although an increasing number of reviews on diagnostic topics are being developed. The second part of the Cochrane Library contains the Cochrane Controlled Trials Register (CCTR) with more than 420,000 RCT. The database also includes abstracts of non-Cochrane systematic reviews, a database on methodology for

Principles of Evidence-Based Medicine

147

conducting systematic reviews, and the Cochrane handbook, which contains information on how to form review groups, do systematic reviews and searches, and obtain information about existing groups. Bandolier is an independent journal about evidence-based healthcare, written by Oxford scientists (www.ebandolier.com). It appears monthly and is a source of evidence-based healthcare information. The impetus behind Bandolier was to find information about evidence of effectiveness (or lack of it), and put the results forward as simple bullet points of those things that worked and those that did not: a bandolier with bullets. Information comes from systematic reviews, meta-analyses, randomized trials, and from high quality observational studies. Another important source of secondary literature is ‘Best Evidence’, which contains the ACP Journal Club (a publication of the American College of Physicians-American Society of Internal Medicine), and the journal EvidenceBased Medicine (a joint publication of the ACP and the British Medical Journal Group, http://ebm.bmjjournals.com). Article selection criteria are based on evidence-based medicine principles and are explicitly stated in each issue. The articles are summarized in ‘value-added’ structured abstracts, and have appended commentary by content experts. With Best Evidence, the complete collection of structured abstracts and commentaries from 1991 onwards can be searched by keyword, topic, study type, year of publication or clinical activity. Several other initiatives provide relevant information for health care providers (and patients) at an accessible level. These sources should be comprehensive, should be frequently updated, have explicit links to evidence, and should be organized for easy and effective searching. A good example of an international source is Clinical Evidence, which is updated and extended monthly at www.clinicalevidence.com and has been translated into several languages. Of sources of primary literature, Medline, produced by the National Library of Medicine in Bethesda, Md., is the best-known bibliographic database of biomedical journal literature (including nursing and dentistry), which it indexes from 1966 on. Medline is searchable by controlled vocabulary (MeSH and subheadings), as well as author or journal. While the roughly 3,900 journals covered by Medline have been selected for their overall reliability and quality, the articles must be scrutinized carefully for their validity and quality as evidence. EMBASE is the Excerpta Medica database for biomedical and pharmaceutical journal articles. This database indexes 3,500 journals, a somewhat different spectrum compared with Medline, with more international (including Canadian) journals and fewer state, nursing, veterinary medicine, or dentistry journals than Medline. EMBASE has been available from 1988 on and is searchable by a controlled vocabulary, also with subheadings; the structure and organization of EMBASE are especially useful for searching for evidence

Pieber

148

regarding drugs and pharmaceuticals. As with Medline, the individual articles cited must be scrutinized for their validity and quality. Critically Appraising the Evidence for Validity and Clinical Usefulness Level of Evidence. For the clinician to summarize the strength of evidence surrounding particular recommendations, levels of evidence have been created as a tool (a useful system is presented in table 1). However, it is important to note that levels of evidence describe only the methodological quality of the study supporting the evidence; they do not describe treatment effect or relevance and have not themselves been validated. Odds Ratio. The odds ratio (OR) describes the odds of a patient in the treatment group experiencing an outcome relative to the odds of a patient in the control group experiencing that outcome. Numerically, odds are expressed by dividing the number of patients who experience the outcome by the number of patients in that treatment group who do not experience that outcome. The OR is a useful indicator of treatment effect size in examples such as case-control studies that assess harm associated with treatment, as these outcomes are rarer (e.g., risk of lactic acidosis in patients treated with metformin). Unfortunately, OR has limited usefulness clinically, as it does not convert to numbers needed to treat (NNT). Relative Risk. Relative risk reduction (RRR) expresses a proportional reduction in outcomes in the treatment group relative to the control group. The RRR has been described as an estimate of the amount of baseline risk that is removed owing to the effects of treatment. It is calculated by subtracting the risk of outcome in the treatment group from the risk of outcome in the control group and dividing the absolute difference by the risk of outcome in the control group. The RRR can also be calculated by subtracting the RR from 1. Absolute Risk Reduction and NNT. The absolute risk reduction (ARR) describes the absolute change in risk of an outcome between treatment and control groups. It is calculated by taking the absolute value of the difference in risk of the outcome in the treatment group as compared to the control group. The NNT describes the number of patients needed to treat to prevent one outcome. It is defined as the reciprocal of the ARR. Because NNT is quickly and easily generated in practice using the method described in this chapter, clinicians should simply be aware of the potential inaccuracy when calculating NNT for RCTs that use survival analysis. The NNT and the ARR will change if the baseline risk changes. Thus, if the risk for an event in a subgroup is higher, the ARR will be larger and the NNT will be less. This is not the case for the RRR, which does not generally change with the underlying risk of the population. NNT is arguably the most clinically relevant expression of outcome of trials. Where

Principles of Evidence-Based Medicine

149

Table 1. Classification of evidence for management (treatment or prevention): levels of evidence [adopted from 4] Level 1A

Systematic overviews or meta-analyses of multiple RCT Systematic overviews are a structured approach to synthesizing the results of previous research by starting with a clear question, comprehensively searching the literature, retrieving, critically appraising, and analyzing the evidence, and synthesizing the results. Meta-analysis is a statistical technique used to numerically combine the results of a number of trials to generate a more precise estimate of the treatment effect and to detect useful treatments that did not yield statistically significant results in previous underpowered trials of low sample size

Level 1A

Large RCT with adequate power to answer the question A randomized, placebo-controlled trial (RCT) with a statistically significant positive result for a clinical endpoint that answers a specific question would be assigned this level. The result should also be of sufficient magnitude to be clinically important. A negative RCT (one that does not support treating with a tested intervention) would also be assigned this level if it was large enough to exclude a clinically relevant effect of treatment

Level 1B

Nonrandomized clinical trial or cohort study with indisputable results This rare category would include studies in which all patients failed on the control therapy and some or all succeeded with the tested therapy (e.g., amputation for gangrene)

Level 2

RCT overviews that do not meet level 1 criteria An RCT that has insufficient power to exclude a clinically important result but that shows a trend toward a result would be assigned this level. Randomized trials with statistically significant results that are judged to be not clinically important or that are not reproducible may be included here as well

Level 3

Nonrandomized clinical trial or cohort study Studies in which the group receiving therapy was compared to a systematically selected contemporaneous group who did not receive therapy (but that was not excluded from therapy because of perceived unsuitability for the therapy) but were followed identically

Level 4

Other study designs and evidence, including consensus documents or expert opinions

harm is well studied, NNT can be weighed against calculated ‘number needed to treat to harm’ (NNH). For large complex trials with multiple outcomes, multiple NNTs and NNHs can be generated. Clinical experience dealing with NNT across a variety of interventions with patients is required to put these numbers into perspective, but we endorse their use in practice, in discussions with patients, and when considering the economics of various therapies.

Pieber

150

Applying the Results in Clinical Practice As clinicians look more often to systematic reviews or RCT to guide their clinical care, they must decide how to apply RCT results to individual patients in their practice setting. Applicability is closely related to concepts of generalizability and external validity, but is broader in its scope, including issues related to the overall impact of treatment in individual patients. In considering applicability, clinicians must decide whether the biology of the treatment effect will be similar in patients they are facing, their patients’ risk of the target event that treatment is designed to prevent and of the side effects that may accompany treatment, and their own ability to deliver the intervention in a safe and effective manner. Important pathophysiological differences can sometimes lead to diminished treatment responses due to a divergence in pathogenetic mechanisms or biologic differences in the causative agent. Hypertension in blacks, which has been observed to be relatively responsive to diuretics, and unresponsive to ␤-blockers, provides an example of the former. To the extent that groups of people exhibit different compliance with a treatment plan, clinicians may expect variations in treatment effectiveness. Variability in compliance between populations may stem from resource limitations in a particular setting, or less obvious attitudinal or behavioral idiosyncrasies. Both types of problems may, for example, affect the safety of outpatient administration of anticoagulants. Inadequate monitoring, whatever the reason, increases bleeding risk from overanticoagulation, shifting the balance between benefit and harm. Clinical practice guidelines (CPG) have been developed to reduce variations in physician practice, to incorporate related research outcomes, and to assist in the management and control of health care costs. It is believed that their incorporation into practice will help clinicians make better decisions that will ultimately improve the quality of health care for patients. When CPG are referred to as being evidence-based, the implication is that they are based on published research evidence that examined clinically important outcomes (where this evidence is available), and that the research itself is critically appraised according to established criteria. Typically, the body of scientific literature is gathered, critically appraised, and incorporated into a series of gradings and recommendations based on the strength of the published evidence. To develop evidence-based CPG from inception to dissemination requires significant resources. This includes the necessary expertise to gather and review the relevant literature, grade the literature according to the strength of the evidence, and develop recommendations. Dissemination and implementation strategies should be developed concurrently. Patient preferences, values, and attitudes are part of any decision-making process that involves therapy or harm. Quality of life, one of the surrogates for patient preferences, is used as an outcome in an increasing number of clinical

Principles of Evidence-Based Medicine

151

trials, although the effect of various treatments on quality of life is not yet fully understood. Evaluating Performance of the Evidence in Clinical Application The evaluation of performance in clinical application tries to assess the overall benefit for both, patients and health care providers. From audits carried out on clinical services it appears that evidence-based care is provided at least to some extent. Population-based ‘outcomes research’ has repeatedly documented that those patients who do receive evidence-based therapies have better outcomes than those who do not. For positive examples, myocardial infarction survivors prescribed aspirin or ␤-blockers have lower mortality rates than those who have not been prescribed these drugs, and where clinicians use more warfarin and stroke unit referrals, stroke mortality declines by ⬎20%. Conclusions

This chapter tries to emphasize on the principles of evidence-based medicine, how evidence-based medicine is practiced and how it could be implemented for the diagnosis, prognosis, prevention, and treatment of gestational diabetes and its complications. Clinicians always have sought to base their decisions on the best possible evidence. The practice of evidence-based medicine turns this process into lifelong, self-directed learning for the benefit of the patients. References 1 2 3 4 5 6 7

Sackett DL, Rosenberg WMC, Gray JA, et al: Evidence-based medicine: What it is and what it isn’t. BMJ 1996;312:71–72. Weiner JP, Parente ST, Garnick DW, et al: Variation in office-based quality. A claims-based profile of care provided to Medicare patients with diabetes. JAMA 1995;273:1503–1508. Haynes RB, Sackett DL, Gray JAM, et al: Transferring evidence from research into practice. 1. The role of clinical care research evidence in clinical decisions. ACP J Club 1996;125:A14–A16. Hertzel C, Gerstein R, Haynes B: Evidence Based Diabetes Care. Hamilton, Decker, 2001. Von Staden H: In a pure and holy way: Personal and professional conduct in the Hippocratic Oath. J History Med Allied Sci 1996;51:406–408. Porter R: The rise of medical journalism in Britain to 1800; in Bynum WF, Lock S, Porter R (eds): Medical Journals and Medical Knowledge: Historical Assays. London, Routledge, 1992. Osler W: The influence of Louis on American medicine; in McGovern JP, Roland CG (eds): The Collected Essays of Sir William Osler. Birmingham, Classics of Medicine Library, 1985.

Thomas R. Pieber, MD Associate Professor of Medicine Department of Internal Medicine, Diabetes and Metabolism Medical University Graz, Auenbruggerplatz 15, AT–8036 Graz (Austria) Tel. ⫹43 316 385 7703, Fax ⫹43 316 385 7704, E-Mail [email protected]

Pieber

152

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 153–160

The Graz Approach to Diabetes in Pregnancy Peter A.M. Weiss, Gernot Desoye Clinic of Obstetrics and Gynaecology, Medical University of Graz, Graz, Austria

In both pregestational and gestational diabetes the predominant focus of the obstetrician’s care is the growing fetus, which is at higher risk than the pregnant women herself [1, 2]. In gestational diabetes the treatment of the fetus, and here especially insulin treatment, is often instituted in the interest of the fetus, while the mother would require a less intensive care, if any at all, if the principles of internal medicine were applied. In a number of studies we could demonstrate on the basis of fetal insulin homeostasis that the quality of maternal metabolic control in gestational and pregestational diabetes is more important for the fetus than for the mother [1, 3]. The major risk for the fetus arises from fetal hyperinsulinism, which, depending on its duration and intensity, may result in diabetogenic fetopathy with well-known associated consequences for the fetus such as premature delivery, dysmaturation, acidosis, respiratory distress, alterations in electrolyte status or icterus [1]. The rate of caesarean sections is profoundly increased in hyperinsulinaemic fetuses because of the disproportionate fetal growth, fetal macrosomia and reduced tolerance towards the stress of vaginal delivery. A significant fetal hyperinsulinism at the time of functional programming at around week 28 of gestation may result in a permanent malfunctioning of the fetal pancreas in utero. When neonates of diabetic women were followed up to the age of 5–15 years we could demonstrate that those offspring with preceding fetal hyperinsulinism (amniotic fluid insulin levels ⱖ10 ␮U/ml) between weeks 27 and 34 of gestation had significantly higher serum levels of glucose and insulin in the basal state as well as after an oral glucose challenge. They also displayed reduced insulin sensitivity and a 13-fold higher incidence of obesity [4, 5]. These deviations closely correlated with the intrauterine diabetic milieu

as reflected by amniotic fluid insulin concentrations and are generally regarded as initial stages for the development of type 2 diabetes. In order to avoid the fetal, neonatal, adolescent and adult risks as defined above the primary goal for an adequate therapy must lie in an early diagnosis of fetal hyperinsulinism and its effective treatment to keep it as low as possible. In mother-fetus pairs with comparable maternal glucose levels there is a considerable idiosyncrasy of fetal glucose supply and thus insulin homeostasis, i.e. fetal glucose levels may differ greatly despite similar maternal blood glucose levels [6]. This is of particular relevance in diabetes. The quality of fetal metabolic control, of fetal glucose supply and, hence, of the fetal risk can, therefore, not be assessed reliably on the basis of maternal glycaemia only. The occurrence of fetopathy despite seemingly adequate glycaemic control of the mother is a well-known phenomenon to experienced obstetricians [7, 8]. Equally, women with apparently insufficient metabolic control can deliver metabolically normal fetuses. The fetal glucose gradient defined as the maternal-to-fetal glucose concentration ratio is determined by a number of variables [2]. It is the result of (1) maternal glucose levels, (2) uteroplacental and fetal-placental blood flow, (3) placental morphology, (4) placental glucose utilization, (5) fetal glucoseinsulin homeostasis and (6) backtransport of glucose from the fetus into the placenta [9]. Fetal glucose utilization is higher in situations of fetal stress and hyperinsulinism. In addition an imbalance between demand and supply will also affect the glucose gradient [10, 11]. However, three major factors are mainly responsible for the poor reflection of fetal glucose-insulin homeostasis by maternal blood glucose levels: • Individual differences in transplacental glucose transport and intraplacental glucose utilization • Different glucose sensitivity of individual fetuses even in twins [12] • The glucose steal phenomenon that results in a lowering of maternal glucose levels and smoothening of postprandial glucose peak levels particularly in the third trimester of gestation [4, 13, 14] Hence, especially in women with hints of diabetes in their case history and recorded only in late pregnancy, it can hardly be distinguished whether maternal euglycaemia is due to the absence of diabetes or to the fetal glucose steal phenomenon. On the other hand, in known diabetics, fetal siphoning can simulate euglycaemia and, thus, lead to undertreatment. The placenta undergoes a variety of structural and functional changes in diabetes [15–19]. The individual physiological differences in transplacental glucose transport are amplified in diabetes mellitus (fig. 1). Depending on its duration maternal hyperinsulinism (or hypoglycaemia) will stimulate GLUT1 mRNA expression and will increase transplacental

Weiss/Desoye

154

IDDM (N⫽119) GDM (N⫽119) Controls (N⫽ 140) Line of identity

Glucose in cord blood (mg/dl)

140

110

80

50

20 20

50

80

110

140

170

200

Maternal blood glucose (mg/dl)

Fig. 1. Paired glucose levels in maternal and cord blood in metabolically healthy controls, type 1 (IDDM) and gestational diabetic (GDM) women measured at delivery.

glucose transport, thus avoiding fetal hypoglycaemia [20]. On the other hand, persistent hyperglycaemia downregulates placental GLUT1 [20–22], thus resulting in reduced glucose transport to the fetus. The placenta has a variable demand for glucose to sustain its own activities [23]. This may reach levels of up to 50% of glucose taken up from the maternal circulation [24]. Maternal hyperglycaemia not only alters the kinetics of maternal-fetal glucose transport [24, 25] but also placental glucose metabolism including glycogen storage [26–28]. This will allow some storage of excess glucose in the placenta that may help in buffering excess glucose in the fetal circulation [24, 26, 27, 29]. Fetal glucose utilization amounts to 38–43 ␮mol/kg at a maternal glucose level of 100 mg/dl [24, 30]. This value will be higher in the presence of fetal hyperinsulinism. Because transplacental passage of glucose among other factors is directly proportional to the maternal-fetal glucose gradient [31], fetal hyperinsulinism will result in a lowering of fetal glucose levels with an ensuing increase in placental glucose transport in order to maintain fetal euglycaemia [32]. This notion is supported by the apparent independence of fetal glucose from fetal insulin levels [33, Weiss, unpubl. data]. Diabetes may lead to alterations in maternal-fetal and fetal-maternal glucose transfer that will make the assessment of the fetal glycaemic status only on the basis of maternal glucose levels even more problematic. Perfusion

The Graz Approach to Diabetes in Pregnancy

155

experiments on placental cotyledons with a fixed glucose concentration of 144 mg/dl (8 mmol/l) on the maternal side demonstrated that in gestational diabetic pregnancies that were treated with diet only placental glucose consumption (0.492 vs. 0.248 ␮mol/min/g), glucose utilization (0.255 vs. 0.129 ␮mol/min/g) and the glucose transfer into the fetal circulation (direct transfer 0.979 vs. 0.402 ␮mol/min/g; net transfer 0.269 vs. 0.118 ␮mol/min/g) were all reduced as compared to placentas from normal, non-diabetic pregnancies [34]. Hyperglycaemia reduced placental GLUT1 [22, 35]. This along with the fetal hyperinsulinism [36] may explain the lower fetal glucose levels after an overnight fast in diabetic women than in normal controls. In contrast, in placentas from women with pregestational diabetes the in vitro levels of GLUT1 and glucose uptake into trophoblast basal membrane vesicles are elevated by 40 and 59%, respectively [37]. Glucose transport across the human placenta in pregestational diabetes had not been measured yet, but an increased transfer was found in diabetic rats [25, 38]. In addition to the molecular changes an increase of total villus mass and surface area in diabetes as found by placental morphometry [15, 16, 39–43] may contribute to a potentially higher maternal-fetal glucose transfer, but relevant studies are pending. Collectively, the barrier function of the placenta for glucose appears to be different in gestational versus pregestational diabetes, because the glucose gradient is lower in gestational diabetes as compared to controls whereas in pregestational diabetes it is elevated (fig. 1). In parallel the cord blood HbF and C-peptide levels correlate in pregestational but not in gestational diabetes [44, 45]. Also various amino acids such as leucine, isoleucine and alanine have lower amniotic fluid levels in gestational but higher levels in pregestational diabetes [46]. The various placenta-related mechanisms can either synergize or antagonize each other, thus making the influence of barrier or transport function prevail [2]. Pregnancy can be regarded as a biological experiment in which the fetus serves as a sensor reflecting the balance of maternal homeostasis. In a range of studies about fetal insulin homeostasis we were able to demonstrate a strong correlation between insulin levels in the amniotic fluid [3, 47–50] or cord blood [51] and the intrauterine fetal development, neonatal outcome and development of the adolescent child [4]. Since the care for the growing fetus is the primary focus in the treatment of a diabetic pregnancy and because the effects of maternal glycaemia on the fetus cannot be predicted with certainty and may vary from case to case, it does not appear to us reliable enough to base maternal treatment exclusively on parameters reflecting maternal metabolic control and to ignore those of the fetus. The amniotic fluid insulin levels do in fact form a sensitive and representative fetal parameter that can easily be used for fetal surveillance and control whether the therapy instituted is adequate (fig. 2).

Weiss/Desoye

156

Number of cases 39

Amniotic fluid insulin (␮U/ml)

10

118

178

116

98

8 97% 6 90% 4 50% 2 10% 3%

0 20

24

28 32 36 Weeks of pregnancy

40

44

Fig. 2. Amniotic fluid insulin levels (3rd, 10th, 50th, 90th, and 97th centiles) in the course of pregnancy in 549 non-diabetic patients [taken from 51; with permission].

The measurement of amniotic fluid insulin levels from week 27 onward will allow the assessment of whether maternal glycaemic control was sufficient, too loose or too tight. This parameter has the advantage of being a more direct measure of fetal metabolic control and does not depend on placental function that may vary in each individual case. When amniotic fluid insulin levels exceed the 90th or fall below the 10th centile [Weiss et al., 1998] then maternal treatment must be changed. In gestational diabetic women that were only treated with diet alone an elevated amniotic fluid insulin level, i.e. levels ⬎6–8 ␮U/ml depending on the analytical method used, is a strong indication that insulin treatment must be instituted [3, 50]. Elevated amniotic fluid insulin levels in insulin-treated gestational diabetic women indicate that there is more insulin required for maintaining the fetal level of glycaemia within a normal or near normal range and, hence, the insulin dose should be raised. In general the maternal glucose levels will not change, because the higher insulin dose will result in reduced fetal insulin production and hence in a lower fetal glucose siphoning (glucose steal phenomenon) [3, 50]. In women with pregestational diabetes that depend on insulin therapy elevated amniotic fluid insulin levels indicate that mean maternal glucose levels should be lowered by at least 10 mg/dl [3]. In contrast, if amniotic fluid insulin levels are below the 10th centile then it is recommended to raise the maternal

The Graz Approach to Diabetes in Pregnancy

157

median glucose levels by an increment of at least 10 mg/dl, particularly in a situation where fetal growth is restricted. The incidence and prevalence of diabetes mellitus are dramatically increasing worldwide [52]. Non-genetic transmission of diabetes accounts for a significant proportion of this increase [4]. A wide range of studies have clearly shown that the non-genetic intrauterine transmission of diabetes is a result of fetal mal-programming [53–57] and may be avoided by normalization of the disturbed intrauterine glucose-insulin homeostasis of the fetus. Diagnosis and adequate treatment of fetal hyperinsulinism is, therefore, not only in the prime interest for the individual pregnancy but will affect future generations and will have a large socio-economic impact.

References 1 2 3

4 5

6 7 8

9 10 11 12 13 14

15 16

Weiss PAM, Walcher W, Scholz HS: Der vernachlässigte Gestationsdiabetes: Risiken und Folgen. Geburtshilfe Frauenheilkd 1999;59:535–544. Weiss PAM: Diabetes und Schwangerschaft. Vienna, Springer, 2002. Weiss PAM: Diabetes in pregnancy: Lessons from the fetus; in Dornhorst A, Hadden DR (eds): Diabetes and Pregnancy: An International Approach to Diagnosis and Management. Chichester, Wiley, 1996, pp 221–240. Weiss PAM, Haeusler M, Tamussino K, Haas J: Can glucose tolerance test predict fetal hyperinsulinism? Br J Obstet Gynaecol 2000;107:1480–1485. Weiss PAM, Scholz HS, Haas J, Tamussino K, Seissler J, Borkenstein M: Long-term follow-up of infants of mothers with type-1 diabetes: Evidence for hereditary and non-hereditary transmission of diabetes and precursors. Diabetes Care 2000;23:905–911. Fraser R: Diabetic control in pregnancy and intrauterine growth of the fetus. Br J Obstet Gynaecol 1995;102:275–277. Dandona P, Besterman HS, Freedman DB, Boag F, Taylor AM, Beckett AG: Macrosomia despite well-controlled diabetic pregnancy. Lancet 1984;i:737. Russell G, Farmer G, Lloyd DJ, Pearson DWM, Ross IS, Stowers JM, Sutherland HW, Visser GHA, van Ballegooie E, Sluiter WJ, Verhaaren HA, Craen M, De Gomme P, Rubens R, Parewijck W, Derom R: Macrosomy despite well-controlled diabetic pregnancy. Lancet 1984;i: 283–285. Schneider H, Reiber W, Sager R, Malek A: Asymmetrical transport of glucose across the in vitro perfused human placenta. Placenta 2003;1:27–33. Hay WW Jr: Fetal and neonatal glucose homeostasis and their relation to the small for gestational age infant. Semin Perinatol 1984;8:101–116. Lasuncion MA, Lorenzo J, Palacin M, Herrera E: Maternal factors modulating nutrient transfer to fetus. Biol Neonate 1987;51:86–93. Burke BJ, Sheriff RJ, Savage PE, Dixon HG: Diabetic twin pregnancy: An unequal result. Lancet 1979;i:1372–1373. Nolan CJ, Proietto J: The feto-placental glucose steal phenomenon is a major cause of maternal metabolic adaptation during late pregnancy. Diabetologia 1994;37:976–984. Weiss PAM, Scholz HS, Haas J, Tamussino KF: Effect of fetal hyperinsulinism on oral glucose tolerance test results in patients with gestational diabetes mellitus. Am J Obstet Gynecol 2001;184:470–475. Desoye G, Shafrir E: Placental metabolism and its regulation in health and diabetes. Mol Aspects Med 1994;15:505–682. Desoye G, Shafrir E: The human placenta in diabetic pregnancy. Diabetes Rev 1996;4:70–89.

Weiss/Desoye

158

17

18 19 20

21 22

23 24 25 26 27 28

29

30 31

32

33 34 35

36 37 38

39

Desoye G, Hauguel-de Mouzon S, Shafrir E: The placenta in diabetic pregnancy; in Hod M, Jovanovic L, Di Renzo GC, De Leiva A, Langer O (eds): Diabetes and Pregnancy. London, Dunitz, 2003, pp 126–147. Desoye G, Myatt L: The placenta; in Reece EA, Coustan DR, Gabbe SG (eds): Diabetes in Women, ed 3. Philadelphia, Lippincott, 2004, pp 147–157. Desoye G, Kaufmann P: The human placenta in diabetes; in Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Basel, Karger, 2005, pp 94–109. Gordon MC, Zimmerman PD, Landon MB, Gabbe SG, Kniss DA: Insulin and glucose modulate glucose transporter messenger ribonucleic acid expression and glucose uptake in trophoblast isolated from first-trimester chorionic villi. Am J Obstet Gynecol 1995;173:1089–1097. Illsley NP, Sellers MC, Wright RL: Glycaemic regulation of glucose transporter expression and activity in the human placenta. Placenta 1998;19:517–524. Hahn T, Barth S, Weiss U, Mosgoeller W, Desoye G: Sustained hyperglyceamia in vitro downregulates the GLUT1 glucose transport system of cultured human term placental trophoblast: A mechanism to protect fetal development? FASEB J 1998;12:1221–1231. Meschia G, Battaglia FC, Hay WW, Sparks JW: Utilization of substrates by the ovine placenta in vivo. Fed Proc 1980;39:245–249. Hauguel S, Desmaizieres V, Challier JC: Glucose uptake, utilization, and transfer by the human placenta as functions of maternal glucose concentration. Pediatr Res 1986;20:269–273. Thomas CR, Eriksson GL, Eriksson UJ: Effects of maternal diabetes on placental transfer of glucose in rats. Diabetes 1990;39:276–282. Barash V, Gurman A, Shafrir E: Mechanism of placental glycogen deposition in diabetes in the rat. Diabetologia 1983;24:63–68. Gewolb IH, Barrett C, Warshaw JB: Placental growth and glycogen metabolism in streptozotocin diabetic rats. Pediatr Res 1983;17:587–591. Desoye G, Hofmann HH, Weiss PA: Insulin binding to trophoblast plasma membranes and placental glycogen content in well-controlled gestational diabetic women treated with diet or insulin, in well-controlled overt diabetic patients and in healthy control subjects. Diabetologia 1992;35:45–55. Desoye G, Korgun ET, Ghaffari-Tabrizi N, Hahn T: Is fetal macrosomia in adequately controlled diabetic women the result of placental defect? – A hypothesis. J Matern Fetal Neonatal Med 2002;11:258–261. Kalhan SC, D’Angelo LJ, Savin SRM, Adam PAJ: Glucose production in pregnant women at term gestation. J Clin Invest 1979;63:388–394. Stembera ZK, Hodr J: I. The relationship between the blood levels of glucose, lactic acid and pyruvic acid in the mother and in both umbilical vessels of the healthy fetus. Biol Neonate 1966;10:227–238. Sutherland HW, Campbell-Brown BM, Fisher P, Treharne IAL: Heavy for date babies; in Sutherland HW, Stowers JM (eds): Carbohydrate Metabolism in Pregnancy and the Newborn. Berlin, Springer, 1978, pp 188–207. Paterson P, Page D, Taft P, Phillips L, Wood C: Study of fetal and maternal insulin levels during labor. J Obstet Gynaecol Br Commonw 1968;75:917–921. Osmond DTD, Nolan CJ, King RG, Brennecke SP, Gude NM: Effects of gestational diabetes on human placental glucose uptake, transfer, and utilisation. Diabetologia 2000;43:576–582. Hahn T, Hahn D, Blaschitz A, Korgun ET, Desoye G, Dohr G: Hyperglycaemia-induced subcellular redistribution of GLUT1 glucose transporters in cultured human term placental trophoblast cells. Diabetologia 2000;43:173–180. Oakley NW, Beard RW, Turner RC: Effect of sustained maternal hyperglycemia on the fetus in normal and diabetic pregnancies. Br Med J 1972;i:466–469. Jansson T, Wennergren M, Powell TL: Placental glucose transport and GLUT1 expression in insulin-dependent diabetes. Am J Obstet Gynecol 1999;180:163–168. Hadawi GL, Kihlström I, Eriksson UJ: Diabetes in pregnancy: Enhanced placental transport of glucose and neutral amino acids from manifest diabetic rats to their fetus (abstract). Diabetologia 1986;29:545A. Singer DB, Liu CT, Widness JA, Ellis RA: Placental morphometric studies in diabetic pregnancies. Placenta Suppl 1981;3:193–202.

The Graz Approach to Diabetes in Pregnancy

159

40 41 42 43 44 45 46

47 48

49 50 51 52 53

54

55

56

57

Teasdale F: Histomorphometry of the placenta of the diabetic woman: Class A diabetes mellitus. Placenta 1981;2:241–252. Teasdale F: Histomorphometry of the human placenta in class B diabetes mellitus. Placenta 1983;4:1–12. Teasdale F: Histomorphometry of the human placenta in class C diabetes mellitus. Placenta 1985;6:69–82. Boyd PA, Scott A, Keeling JW: Quantitative structural studies on placenta from pregnancies complicated by diabetes mellitus. Br J Obstet Gynaecol 1986;93:31–35. MacFarlane CM, Tsakalakos N, Taljaard JJF: Value of glycosylated haemoglobin determination in diabetic pregnancy. S Afr Med J 1985;68:310–312. MacFarlane CM, Tsakalakos N, Taljaard JJF: Acetylated fetal haemoglobin in neonates born to mothers with established and gestational diabetes. S Afr Med J 1985;68:571–574. Persson B, Pschera H, Lunell N-O, Barley J, Gumaa KA: Amino acid concentrations in maternal plasma and amniotic fluid in relation to fetal insulin secretion during the last trimester of pregnancy in gestational and type 1 diabetic women and women with small-for-gestational-age infants. Am J Perinatol 1986;3:98–103. Weiss PAM: Die Überwachung des Ungeborenen bei Diabetes mellitus an Hand von Fruchtwasserinsulinwerten. Wien Klin Wochenschr 1979;91:293–304. Weiss PAM, Hofmann HMH, Kainer F, Haas JG: Fetal outcome in gestational diabetes with elevated amniotic fluid insulin levels. Dietary versus insulin treatment. Diabetes Res Clin Pract 1988;5:1–7. Weiss PAM: Gestational diabetes: A survey and the Graz approach to diagnosis and therapy; in Weiss PAM, Coustan DR (eds): Gestational Diabetes. Vienna, Springer, 1988, pp 1–55. Weiss PAM, Pürstner P, Winter R, Lichtenegger W: Insulin levels in amniotic fluid of normal and abnormal pregnancies. Obstet Gynecol 1984;63:371–375. Weiss PAM, Kainer F, Haeusler M, Pürstner P, Haas J: Amniotic fluid insulin levels in nondiabetic pregnant women: An update. Arch Gynecol Obstet 1998;262:81–86. King H, Aubert RE, Herman WH: Global burden of diabetes, 1995–2025: Prevalence, numerical estimates, and projections. Diabetes Care 1998;21:1414–1431. Dörner G, Steindel E, Thoelke H, Sehliak V: Evidence for decreasing prevalence of diabetes mellitus in childhood apparently produced by prevention of hyperinsulinism in the foetus and newborn. Exp Clin Endocrinol 1984;84:134–142. Dörner G, Steindel E, Kohlhoff R, Reiher H, Anders B, Verlohren HJ, Hielscher K: Further evidence for a preventive therapy of insulin-dependent diabetes mellitus in the offspring avoiding maternal hyperglycaemia during pregnancy. Exp Clin Endocrinol 1985;86:129–140. Dörner G, Plagemann A, Reinagl H: Familial diabetes aggregation in type 2 diabetes: Gestational diabetes as apparent risk factor for increased diabetes susceptibility in the offspring. Exp Clin Endocrinol 1987;89:84–90. Dörner G, Plagemann A, Ruckert JC, Gotz F, Rohde W, Stahl F, Kurschner U, Gottschalk J, Mohnike A, Steindel E: Teratogenetic maternofoetal transmission and prevention of diabetes susceptibility. Exp Clin Endocrinol 1988;91:247–258. Plagemann A: Fuel mediated ‘functional teratogenesis’ and primary prevention; in Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Basel, Karger, 2005, pp 9–17.

P.A.M. Weiss, MD Elisabethstrasse 18 AT–8010 Graz (Austria) Tel. ⫹43 316 33 88 85

Weiss/Desoye

160

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 161–173

Clinical Management of Pregnancies Complicated with Type 1/Type 2 Diabetes mellitus Josip Djelmiš Department of Obstetrics and Gynecology, State Referral Centre for Diabetes in Pregnancy, Zagreb, Croatia

Diabetes in pregnancy causes many problems both for mother and child. Basic issues of a diabetic pregnancy are regulation of illness, congenital anomalies, spontaneous abortion and fetal growth. A significant question is the occurrence or worsening of diabetic nephropathy and pre-existent hypertension, as well as the problem of fetal and perinatal hypoxia, its diagnostics and treatment and pregnancy completion. There are also other problems of the newborn child: hypoglycaemia, hyperbilirubinaemia, hypocalcaemia and hypomagnesaemia, and respiratory distress syndrome.

Perinatal Mortality

Pregnancy in women with diabetes was made possible by the discovery and use of insulin, but the perinatal mortality during the 1950s and 1960s was still about 20% [1–5] and has significantly decreased (to less than 10%) since the 1970s, thanks to the simultaneous use of obstetric and monitoring methods in the neonatological period and the treatment of mother and child [3, 6]. Karlsson and Kjellmer [7] showed the dependence of perinatal death on the mother’s glycaemia. In 1965, Pedersen and Molsted-Pedersen [5, 8] put forward the hypothesis of excessive fetal growth due to the transplacental transport of glucose from the mother, which causes fetal hyperinsulinaemia. In recent years there has been a significant decrease of perinatal mortality, but in big centres it is still above 3% [9, 10].

Thanks to good metabolic control of pregnant diabetics, antenatal and neonatal care, monitoring the conditions of the fetus with cardiotocography and ultrasound, amniotic fluid analysis to prove fetal lung maturity and a more liberal attitude to the completion of the birth by Caesarean section in the last 4 years, the perinatal mortality at the Obstetrics and Gynaecology Clinic in Zagreb decreased to 1.8%. Pregnant women with type 1 diabetes have a high frequency of spontaneous abortions when glycosylated haemoglobin in the first trimester is above 12% or when the mean value of fasting blood glucose is above 6.7 mmol/l [11–13]. The frequency of spontaneous abortions correlates with the metabolic control.

Congenital Malformations

Congenital malformations most frequently occur in pregnant women who had hyperglycaemia during the first trimester of pregnancy [14, 15]. The congenital malformations reach a frequency of 8% and more if during the first 8 weeks of pregnancy there was hyperglycaemia of 3–4 times that in the general population [16, 17]. The most frequent are malformations of the heart and blood vessels, and central nervous system malformations, which are frequently lethal. Bigger congenital malformations are 2–4 times more frequent in pregnant women with type 1 diabetes than in non-diabetic pregnant women. The proportion of lethal malformations in perinatal death is 26% [18]. Risk factors for the occurrence of congenital malformations are no or bad preconception treatment and hyperglycaemia during the first 8 weeks of pregnancy [19–21]. Fetuses of diabetic pregnant women have a high risk of a neural tube defect. Less frequent complications are caudal regression syndrome (sacrum agenesis, phocomelia), more severe is agenesis of the lumbosacral spine and the most severe malformation is sirenomelia with the fusion of the lower extremities, so that the body looks like that of a siren and there are also genitourinary and gastrointestinal malformations [22]. The described malformation is pathognomonic for diabetes. Congenital malformations of the heart and big blood vessels are frequent: ventricular septum defect, transposition of big blood vessels and coarctation of the aorta. Among kidney anomalies there is kidney agenesis, and double ureter and gastrointestinal malformations are frequent. In the antenatal diagnosis of fetal malformations we determined the HbA1c level in the first trimester, ␣-fetoprotein serum between 15 and 18 weeks of pregnancy and did ultrasound examinations during the pregnancy.

Djelmiš

162

Influence of Hyperglycaemia of the Mother on the Growth of the Fetus

Characteristics of diabetic pregnancies are increased fetal growth and giving birth to a macrosomic child. The term macrosomia indicates increased body mass. Newborn macrosomia is arbitrarily defined as a body weight above 4,000 g at term and a large-for-gestational age baby above the 90th percentile for a certain gestational age. Macrosomia occurs because of fetal hyperinsulinaemia [23, 24]. Despite well-controlled glycaemia, the frequency of fetal macrosomia is still very high and amounts to between 13.2 and 37.5% [25]. At the Obstetrics and Gynaecology Clinic in Zagreb during the last 5 years, the frequency of giving birth to macrosomic children has decreased and amounts to 17%. The chronic hyperinsulinaemia in a fetus from a diabetic pregnancy results in increased total body mass with moderate increased trunk length and selective organomegaly, which is the result of hypertrophy of insulin-sensitive tissue. Usually, subcutaneous tissue, liver, lungs, spleen, adrenal gland, skeletal muscles, thymus and pancreas are hyperplastic. Due to organomegaly, a newborn baby is not only large, but it has a disproportionately big trunk in relation to the head measurements. The macrosomic newborn baby has wider shoulders and larger extremities diameter, a decreased ratio between the head circumference and shoulder width, higher values of skin folds and a higher proportion of fat in the total weight. Neonatal complications are more frequent in the group of disproportional/asymmetric/metabolic macrosomia in relation to proportional/ symmetric/constitutional macrosomia. A placenta with changes in its structure is significant for diabetes: hypo- and hyperramification of terminal villi, which can lead to the increased resistance in the fetoplacental circulation. Additionally, due to the mother’s vasculopathy, disturbances in the uteroplacental circulation are possible, although with the decreased perfusion this does not lead to a decrease of fetoplacental circulation. Good regulation of glycaemia ensures good fetoplacental circulation, while in conditions of hypoglycaemia and hyperglycaemia, the fetoplacental circulation is reduced. These changes do not occur with temporary changes of glycaemic status which do not cause a momentary vasoactive response of placenta, but they are the result of bad metabolic control over a long period of time and a resulting change in the placental structure. A frequent outcome of such pregnancies is a growth restriction of the fetus and in cases of expressed compromised fetoplacental transport, even intrauterine death. Insufficient nutrient transport, as in cases of too strict a control of glycaemia or diabetes complicated by nephropathy and/or hypertension, will cause hypoinsulinism in the fetus accompanied by growth restriction. After

Clinical Management of Diabetic Pregnancy

163

birth, these newborns have low glucose values and a low ratio between insulin and glucose. Placental factors, including its size and increased villous network, have been linked to increased fetal growth. The ratio between the birth weight and the placental weight together with macrosomia is significantly smaller than the ratio for newborn babies of normal or smaller growth for their age.

Complications of Diabetic Pregnant Women

Diabetic ketoacidosis, pre-eclampsia, pyelonephritis and bad antenatal care significantly increase the perinatal mortality. The perinatal outcome is also worsened by numerous other complications during pregnancy [26]. Infections are significantly more frequent in diabetic pregnant women than in healthy ones; there is a correlation between the frequency of infections and bad metabolic control. Pyelonephritis can be expected in 4% of diabetic pregnant women, and in only 1% of pregnant women in the non-diabetic population. In publications from the 1960s, a frequency from 5.7 to 39% [26] was mentioned. Cousins [27] mentioned a frequency of 4.3% for a total of 3,375 cases from 1965 to 1985. Significant bacteriuria is also frequent in diabetic pregnant women and amounts to about 40% [26–28]. In pregnancy, Pedersen and Molsted-Pedersen [8] found significant bacteriuria in 16% of 306 pregnant women. We [26] found significant bacteriuria in 39.2% of 508 pregnant women with type 1 diabetes at the first investigation. The most frequent cause was Escherichia coli (43.6%), followed by Enterobacter (27.1%). It is interesting – and also important – that in our sample of 508 patients we found significant bacteriuria more frequently in pregnant women with hypertension, i.e. there was significantly more frequently hypertension in pregnant women with significant bacteriuria than in those with sterile urine. Pre-eclampsia occurs in 20% of diabetic pregnant women, which is 3 times more frequently than in healthy pregnant women and significantly increases perinatal mortality and morbidity [27, 29]. The frequency of pre-eclampsia increases with the severity and duration of the illness according to White’s classification. It also occurs more frequently in diabetic pregnant women with worse metabolic control. According to the data of the Obstetrics and Gynaecology Clinic in Zagreb, the perinatal mortality of diabetic pregnant women with pre-eclampsia in a period of 40 years is high and amounts to 20.6% in relation to 3.5% of normotensive diabetic pregnant women. Chronic hypertension is a frequent complication of diabetes. In women with type 1 diabetes it occurs in about 15% in groups B to R according to White’s classification. During pregnancy there is frequently a worsening of

Djelmiš

164

hypertension and proteinuria, i.e. superimposed pre-eclampsia occurs, which increases perinatal mortality and morbidity. Diabetic nephropathy occurs in about 5% of diabetic pregnant women. If in the first trimester proteinuria above 300 mg/24 h is found, a bad perinatal outcome can be expected. Thirty percent of pregnant women with diabetic nephropathy will have hypertension in the first trimester, while 75% will develop hypertension before the birth. The frequency of premature deliveries is also high, due to the worsening of hypertension and proteinuria. In more than half of the diabetic pregnancies with pre-eclampsia, the delivery is completed before the 37th week of pregnancy. Other complications of the mother are the occurrence of serious oedema due to hypoalbuminaemia [28, 30]. The choice of antihypertensive drugs in diabetic pregnant women with nephropathy is not always an easy one. Angiotensin-converting enzyme inhibitor (ACE) and angiotensin receptor antagonists decrease the glomerular hyperfiltration, so they are a frequent therapy in diabetic persons. Teratogen effects which are expressed as fetal hypocalvaria and kidney defects are most frequently described. It is presumed that the cause of ACE inhibitor toxicity is the result of fetal hypotension and decreased kidney blood circulation, which leads to grave and sometimes even fatal anuria in fetuses and newborn babies. Apart from that, the use of this group of drugs is also associated with fetal growth restriction, premature birth and serious hypotension of the newborn. As a result of all this, ACE inhibitors in pregnancy are contraindicated. ␤-Adrenergic blockers can cause hypoglycaemia. Methyldopa and drugs which block calcium channels, like nifedipine, are the drugs of choice. Nevertheless, the successful outcome of pregnancy is above 90%. In preconception treatment it is important to notice and predict a bad perinatal outcome if proteinuria is above 3.0 g/24 h, and if serum creatinine is above 130 ␮mol/l. It must be borne in mind that even in a successful pregnancy maternal morbidity is very high. Proliferative retinopathy is characterized by neovascularization, i.e. capillary growth over the retina surface. Pregnancy deteriorated proliferative neuropathy. Macular oedema, which occurs due to changes in capillary permeability, also deteriorates in pregnancy, especially in pregnant women with hypertension [31]. Photocoagulation by laser is used before or during pregnancy not only as a treatment, but also in the prevention of proliferative retinopathy. It would be ideal to make the ophthalmological examination of pregnant women before the pregnancy, then in the first trimester of pregnancy and after that as needed. Peripheral and cranial neuropathy during pregnancy is rare. Autonomous neuropathy frequently causes problems in pregnancy due to the occurrence of orthostatic hypotension, a decreased response of catecholamines to hypoglycaemia

Clinical Management of Diabetic Pregnancy

165

and a decrease of symptoms connected with the stomach (gastropathy). Diabetic gastropathy in pregnancy occurs with symptoms of nausea, vomiting, and problems with food intake, which makes glycaemia control difficult. Treatment with H2 receptor antagonists is recommended. Diabetic ketoacidosis is rare; it occurs in less than 1% of cases, but it is a serious complication. Diabetic ketoacidosis can occur in pregnancy hyperemesis, diabetic gastropathy, in ␤-sympathomimetic therapy, during infection and taking of corticosteroids. Since insulin resistance occurs in pregnancy, lipolysis and ketogenesis increase, and so diabetic ketoacidosis can occur even with minimal hyperglycaemia. Diabetic acidosis is accompanied by a high fetal loss, which amounts to 20%. For fetal benefit, the mother’s acidosis must be quickly treated. The treatment is the same as in non-pregnant patients.

Care of Diabetic Pregnant Women

Preconception treatment is extremely important for diabetic women who plan a pregnancy. It should include advice on diet, adequate insulin therapy to achieve normoglycaemia, ophthalmological examination, and nephrological examination [32]. Before obtaining an HbA1c value ⬍7%, pregnancy is not recommended, since only in this way can the frequency of spontaneous abortions and congenital malformations be avoided. It is important to maintain normoglycaemia in pregnancy to decrease the perinatal morbidity. Ideal fasting blood glucose values in pregnancy are under 5.0 mmol/l and postprandial blood glucose less than 7.0 mmol/l. The frequency of determining the glucose concentration depends on the difficulty of glycaemia control. Four to eight glucose tests are necessary daily. Once a month the HbA1c value has to be determined. A woman with diabetes has to be informed in detail about the need to regulate diabetes before conception and before planning a pregnancy. Care of Diabetic Pregnant Women during the First Trimester Diabetic pregnant women should contact a doctor at the very beginning of pregnancy in order to achieve good metabolic control during the critical period of organogenesis and to predict diabetic complications like hypertension and nephropathy. Additional examinations are urine tests for proteins, determining of serum creatinine and glycosylated haemoglobin to estimate diabetic control and the risk of congenital malformations. It is necessary to determine the blood type, Rh factor, haemoglobin, antibodies to rubella, syphilis and hepatitis B. Examinations have to be made every 2 weeks, depending on diabetes control and presence of complications.

Djelmiš

166

Pregnant women with diabetes can only be conditionally healthy if they personally take an active role in regular self-control and care, which ensures a fetus of normal physical and mental health [33]. Every pregnant woman with diabetes needs instruction with regard to self-monitoring and self-care, should adhere to an appropriate diet and should exercise regularly. Pregnant women with diabetes should be taught about the significance of a diabetic diet; while monitoring the illness they should be permanently aware of changes of glucose in the blood to adjust the treatment adequately. Since the illness can only temporarily be controlled by a gynaecologist or a diabetologist, a diabetic pregnant woman must take over her treatment after adequate instructions and with the occasional help of a medical team (doctor, nurse, dietician). She must become familiar with the characteristics of diabetes, the dangers for the fetus, the possible complications of diabetes and the way of their prevention, i.e. treatment. Self-control means that a person with diabetes is in her daily life responsible for controlling her diet and body weight, taking care of body hygiene, performing everyday physical activities, monitoring the glucose level in blood and urine and according to the results changing the treatment [33, 34]. Self-monitoring of the blood glucose level is carried out by various test strips or electrodes. According to the results, the insulin dose can be changed; therefore, monitoring of the blood glucose is recommended frequently every day. Critical periods for testing blood glucose are: before breakfast, 2 h after breakfast, before lunch, 2 h after lunch, before dinner, 2 h after dinner, before bedtime and at about 3 a.m. Regulated glycaemia critically depends on the number of blood tests done during the day and on the success of changing the insulin dose according to the results self-monitoring. Today’s recommended diet for diabetics is a simple physiologically balanced diet which fulfils metabolic demands and whose composition (regarding the different sources of certain nutrients: carbohydrates, fat and proteins in food) can be changed and adapted to different local customs and habits [33]. The diet of a diabetic woman during pregnancy need not be changed. Taking folic acid 5 mg/day should start before pregnancy and should continue till the completion of 12 weeks of pregnancy in order to decrease the frequency of a neural tube defect. Necessary energy intake is about 30–35 kcal/kg of the ideal body weight, divided into six daily meals and it consists of 55% of carbohydrates, 20% of proteins and 25% of fat. Physical activity has an extremely positive effect on the exchange of substances in the organism of a diabetic pregnant woman, influencing metabolic processes in liver and muscles. Each diabetic pregnant woman should, depending on her weight and activities during the day, be prescribed (and carry out) a certain type of physical activity – as a kind of treatment.

Clinical Management of Diabetic Pregnancy

167

Today, human insulin which is produced by genetic engineering is most frequently used for treatment. Together with insulin treatment, glycaemia should be controlled throughout the day, especially at the so-called ‘critical points’. Based on the results of monitoring the level of glucose in the blood before using insulin, the dose of short-acting insulin can be changed daily. Intensive insulin treatment can be done in the following ways: • Intensive conventional insulin treatment: a diabetic pregnant woman takes long-acting insulin in the evening before going to bed or two injections of medium-acting insulin every 12 h to maintain the basal insulin value in the blood and also three injections of short-acting insulin immediately before the main meals by using a syringe in the form of a pencil. • Intensive insulin treatment using small devices for continuous insulin injection (so-called insulin pumps) is the second form of this treatment which is used more and more frequently since these devices have been perfected and are accessible to diabetic patients. Regulation of glycaemia should be evaluated every 1–2 weeks, according to the self-monitoring results of a diabetic pregnant woman obtained at home daily. In the first trimester of pregnancy, the need for insulin can be decreased, due to more frequent hypoglycaemia resulting from the proinsulin effect of a chorionic gonadotrophin. Later, the need for insulin can increase to a 3 times higher dose in relation to the dose before the pregnancy. Patients with type 2 diabetes [35] should initially be treated with a basic treatment principle and/or insulin before conception with satisfying regulation of glycaemia or immediately after determining pregnancy. Due to the presence of endogenous insulin, treatment has to start with one or two doses of insulin (combinations of medium-acting and short-acting insulin), and if necessary, treatment has to be continued with a more intensive use of insulin based on the results of the self-control of the blood glucose level done by the pregnant woman herself at home. The regulation of glycaemia should be assessed every 1–2 weeks and self-control should be done every day [36]. Care during the Second Trimester In the second trimester of pregnancy, there is an increased need for insulin. Till the end of the second trimester, the insulin dose is usually doubled. In the second trimester, vascular, kidney and retinal complications have to be followed and if they exist, they should be adequately treated. Hypertension should be treated with methyldopa and/or nifedipine. A detailed ultrasound examination should be performed between 18 and 20 weeks with an emphasis on structures and the risk of an abnormal development of spine, head, kidney and heart. Fetal ultrasound cardiography makes possible an early diagnosis of heart defects.

Djelmiš

168

Screening for Chromosome Anomalies. Pregnant women with type 1 diabetes do not have an increased frequency of giving birth to children with trisomy 21. If there is an indication for an early amniocentesis, it should be done since the risk of a spontaneous abortion is low and amounts to less than 1%. General Review of Obstetric Monitoring. Most of the obstetric problems are connected with diabetes such as pre-eclampsia, polyhydramnion, premature birth, macrosomia, intrauterine growth retardation and sudden intrauterine fetal death in the third trimester. The second trimester is accompanied by an increased frequency of so-called smaller complications in pregnancy due to diabetes including vaginal candidiasis, heartburn, urinary infection and carpal tunnel syndrome. Ultrasound Diagnostics. Ultrasound diagnostics answer three of the most important questions: estimating the gestational age, finding congenital anomalies and fetal growth. In diabetic pregnancies, due to more frequent late ovulations, the gestational age is shorter than the period of amenorrhoea; congenital anomalies are 3–4 times more frequent than in other populations. The duration of pregnancy as well as the expected time of birth should be determined as soon as possible even in women with irregular menstrual cycles. This is particularly important in diabetic pregnant women due to the risk of early births, since fetal macrosomia in the second trimester can lead to arriving at a wrong birth date. Care of Diabetic Pregnant Women in the Third Trimester The need for insulin increases till the 34th to 36th week of pregnancy when it usually slowly decreases. It is a physiological phenomenon, due to the increased glucose transfer to the fetus. More frequent ultrasound monitoring of fetal growth and measuring the quantity of amniotic fluid are necessary. The excessive fetal growth can be seen by an increase of abdomen circumference and biparietal diameter. Early recognition of macrosomic newborn babies is important for preventing birth complications such as shoulder dystocia, fractures, birth asphyxia and neurological damage. Clinical estimate is a method which has been, from a historical point of view, applied for the longest period of time. It is considered incorrect due to the various amniotic fluid volumes, adiposity of pregnant women or uterus anomalies [37]. The potential birth weight can be estimated by experience and practice, within ⫾500 g of the real birth weight. The ultrasound estimate is today the most frequently used method and is based on the biparietal diameter, head circumference, abdomen circumference and femur length. The body weight is approximated based on various mathematical formulae which take together the duration of the pregnancy [35, 36]. Around the birth date, the sensitivity of these methods is between 33 and 69%, specificity 77 and 98%, positive predictive value 40 and 83%, negative predictive value 72 and 90% and the success of the test 76 and 87% [38, 39]. Due to

Clinical Management of Diabetic Pregnancy

169

the frequent mistake of estimating the body weight as lower than it really is, especially in macrosomic fetuses, the head circumference, thoracic diameter, shoulder distance, cheek distance, cerebellum diameter, heart measurement, subcutaneous tissue thickness, independently or in ratios with the circumference of the abdomen or length of the femur are used. Generally, the ‘fatter’ the fetus is, the greater is the possibility of making a mistake with the body weight.

Methods of Fetus Monitoring

In Croatia, amnioscopy in monitoring pregnant women with diabetes was introduced already in 1965. Its success in diagnosing fetal suffering by proving meconium in the amniotic fluid is limited. It is known that a child can die in pregnancy without excreting meconium, probably due to a failure of a fetal regulation mechanism of a bloodstream centralization. Amniocentesis has been used since 1967 in diabetic pregnancies to monitor danger to the fetus and to evaluate fetal maturity. Regarding fetal maturity the lecithin-sphingomyelin (L/S) ratio did not significantly deviate from the L/S curve in healthy pregnant women and phosphate-diglycerol of the amniotic fluid is more reliable in estimating a neonatal respiratory distress. Since its introduction into the clinical practice, cardiotocography has been a standard method for evaluating danger to the fetus. We have had good experience with this method in diabetic pregnancies as well. Measurements of umbilical and fetal circulation in evaluating fetal hypoxia, i.e. ‘brain sparing effect’ – bloodstream centralization in diabetic pregnancies will have to show its value [40, 41].

Completion of Pregnancy

Pregnancy can be completed vaginally or by caesarean section. In the last few years, the number of caesarean sections has significantly increased. At the Obstetrics and Gynaecology Clinic, in the last 5 years, pregnancies of diabetics were in 60% of the cases completed by a Caesarean section. This increase in Caesarean sections in diabetic women is not only the result of the liberal attitude and the corresponding increase in the frequency of Caesarean sections in healthy pregnant women, but also due to frequent complications. Hypertension/ pre-eclampsia, fetal macrosomia, chronic fetal hypoxia, breech presentation, and immature cervix are frequent complications of diabetic pregnant women. For diabetic pregnant women with the previously mentioned complications and

Djelmiš

170

their newborn babies the caesarean section is undoubtedly a much gentler procedure than the vaginal birth. Puerperium

In women with diabetes mellitus, puerperium is an extremely sensitive period of time. The need for insulin is significantly decreased immediately after birth and the dose usually can be decreased to only 30% of the earlier dose during pregnancy. Pathology during puerperium includes pathological changes of the breasts, thrombo-embolic complications, bleeding, genitourinary infections and psychological disorders. Women with type 1 diabetes encounter numerous problems when starting lactation in comparison to normal pregnant women. However, despite this, today more and more women with type 1 diabetes decide to breastfeed. The most frequent problem that diabetic mothers have is an increased frequency of an early separation from their child after birth. This is, on one hand, the result of the frequent surgical completion of the birth in pregnant women with type 1 diabetes, and on the other hand, the condition of a newborn baby of a diabetic mother is usually such that intensive supervision lasting at least 6 h after birth is necessary. Namely, due to fetal hyperinsulinaemia, which occurs as a result of increased glucose values of the mother during pregnancy, such newborn babies are more inclined to hypoglycaemia in the first days after birth. All this makes early and frequent suckling difficult, which is necessary for optimum lactation. Apart from that, women with type 1 diabetes are more inclined to develop mastitis and breast abscess. Women with insulin-independent diabetes mellitus (type 2) very frequently tend to have increased weight [42]. Since breastfeeding is connected with an increased spending of energy and a loss of body weight, such women should be encouraged to breastfeed. Regarding the use of oral antidiabetics during breastfeeding, it has been shown that their excretion into the milk mostly depends on how strongly they are connected to plasma proteins [43]. It is a characteristic of drugs of the sulphonylurea derivative group that they are strongly linked to plasma protein, so their excretion in milk is minimal and, therefore, they are not contraindicated during breastfeeding. Conclusions

Today, regulation of the illness before conception, self-monitoring by the pregnant woman, a good cooperation of the pregnant woman, regional

Clinical Management of Diabetic Pregnancy

171

organization of care in big centres, early detection and treating of hypertension and bacteriuria, simultaneous monitoring of the fetus at the end of pregnancy, indications and safer Caesarean sections enable the prolongation of pregnancy almost till term, with maturing of the child in utero; this results in an elimination of congenital anomalies and very low perinatal mortality, almost close to that in the general perinatal population.

References

2 3 4 5 6 7 8 9 10

11

12

13

14 15 16 17

ˆ

´

Reece EA: The history of diabetes mellitus; in Reece EA, Coustan DR (eds): Diabetes mellitus in Pregnancy. New York, Churchill Livingstone, 1988, pp 3–16. Dra anc¤ic´ A: Diabetes mellitus and pregnancy. Gynaecol Perinatol 1994;3(suppl):15–29. Dra anc¤ic´ A, Djelmiš J, Metelko Z, Pavlic´-Renar I, Mayer D: Pregnancy care of diabetic women in Croatia. Book of Abstracts. 31st Annual Meeting of Diabetic Pregnancy Study Group, Brijuni, 1999. White PD: Pregnancy complicating diabetes. Am J Med 1949;7:609. Pedersen J, Molsted-Pedersen LM: Prognosis of the outcome of pregnancies in diabetics. A new classification. Acta Endocrinol 1965;50:70–78. Macfarlane A, Mugford M: Birth Counts, Statistics of Pregnancy and Childbirth. London, HMSO, 1984. Karlsson K, Kjellmer I: The outcome of diabetic pregnancies in relation to the mother’s blood sugar level. Am J Obstet Gyecol 1972;112:213. Pedersen J, Molsted-Pedersen L: Prognosis of the outcome of pregnancy in diabetes. Acta Endocrinol 1965;50:70. Gabbe SG, Graves CR: Management of diabetes mellitus complicating pregnancy. Obstet Gynecol 2003;102:857–868. Wylie BR, Kong J, Kozak SE, Marshall CJ, Tong SO, Thompson DM: Normal perinatal mortality in type 1 diabetes mellitus in a series of 300 consecutive pregnancy outcomes. Am J Perinatol 2002;19:169–176. Dicker D, Feldberg D, Samuel N, et al: Spontaneous abortion in patients with insulin-dependent diabetes mellitus: The effect of preconceptional diabetic control. Am J Obstet Gynecol 1988;158: 1161–1164. Miodovnik M, Skillman C, Holroyde JC, et al: Elevated maternal glycohemoglobin in early pregnancy and spontaneous abortions among insulin-dependent diabetic women. Am J Obstet Gynecol 1985;153:439–442. Drazanzic A, Blajic J, Djelmis J: Early spontaneous abortions and congenital malformations in relations to diabetes regulation in very early pregnancy (abstract). 24th Annual Meeting of Diabetic Pregnancy Study Group, Silkeberg, 1993. Fuhrman K, Reiher H, Semmler K, et al: Prevention of congenital malformations of infants of insulin-dependent diabetic mothers. Diabetes Care 1983;6:219–223. Key TC, Giuffrida R, Moore TR: Predictive value of early pregnancy glycohemoglobin in the insulin-treated diabetic patient. Am J Obstet Gynecol 1987;156:1096–1100. Baker L, Piddington R: Diabetic embryopathy: A selective review of recent trends. J Diabetes Complications 1993;7:204–212. Millner E, Hare JW, Cloherty JP, et al: Elevated maternal hemoglobin A1c in early pregnancy and major congenital anomalies in infants of diabetic mothers. N Engl J Med 1981;304:1331–1334. Djelmis J, Metelko Z, Pavlic´ -Renar I, Suresh Babu A: Pregestational diabetes mellitus and pregnancy. Diabetol Croat 1999;28:9–15. Freinkel N: Banting Lecture 1980: Of pregnancy and progeny. Diabetes 1980;29:1023–1035. Lucas JM, Leveno KJ, Williams ML, Raskin P, Whalley PJ: Early pregnancy glycosylated hemoglobin severity of diabetes, and fetal malformations. Am J Obstet Gynecol 1989;161: 426–431.

ˆ

1

18 19 20

Djelmiš

172

21

22 23 24

25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Hanson U, Persson B, Thunell S: Relationship between hemoglobin A1c in early type 1 (insulindependent) diabetic pregnancy and occurrence of spontaneous abortion and fetal malformation in Sweden. Diabetologia 1990;33:100–104. Stocker JT, Heifetz SA: Sirenomelia: A morphological study of 33 cases and review of the literature. Perspect Pediatr Pathol 1987;10:7–50. Jovanovic L: The Norbert Freinkel Lecture: Glucose-mediated macrosomia: The overfed fetus and the future (abstract). 61st Scientific Session of the ADA, Philadelphia, 2001. Delmis J, Drazancic A, Ivanisevic M, Suchanek E: Glucose, insulin, HGH and IGF-1 levels in maternal serum, amniotic fluid and umbilical venous serum: Comparison between late normal pregnancy and pregnancies complicated with diabetes and fetal growth retardation. J Perinat Med 1992;20:47–56. Small M, Cameron A, Lunan CB, MacCuish AC: Macrosomia in pregnancy complicated by insulin-dependent diabetes mellitus. Diabetes Care 1987;10:594–599. Drazancic A, Djelmis J, Blajic J, Kuvacic I, Latin V: Bacteriuria in diabetic pregnancies. Diabetol Croat 1997;26:175–181. Cousins L: Pregnancy complications among diabetic women. Obstet Gynecol Surv 1987;42: 140–149. Cousins L: Obstetric complications; in Reece EA, Coustan DR (eds): Diabetes mellitus in Pregnancy. New York, Churchill Livingstone, 1988, pp 455–468. Garner PR, D’Alton ME, Dudley DK, Huard P, Hardie M: Pre-eclampsia in diabetic pregnancies. Am J Obstet Gynecol 1990;163:505–508. Hare JW: Diabetic complications of diabetic pregnancies. Semin Perinatol 1994;18:451–458. Klein BEK, Moss SE, Klein R: Effect of pregnancy on progression of diabetic retinopathy. Diabetes Care 1990;13:34–40. Drazancic A: Prekoncepcijska regulacija bolesti trudnica s dijabetesom. Diabetol Croat 1983;12: 193–208. American Diabetes Association: Self-monitoring of blood glucose. Diabetes Care 1990;13 (suppl 1):41. Dunn SM, Turtle JR: Education: New definitions, new directions; Alberti KGMM, Kral LP (eds): Diabetes Annual 5. Amsterdam, Elsevier, 1990. European NIDDM Policy Group: A Desktop Guide for the Management of NonInsulin-Dependent Diabetes mellitus (NIDDM). Brussels, IDF, 1993. Powers AC: Diabetes mellitus; in Harrison’s Principles of Internal Medicine. New York, McGrawHill, 2001, pp 2109–2137. Chatfield J: ACOG issues guidelines on fetal macrosomia. American College of Obstetricians and Gynecologists. Am Fam Physician 2001;64/1:169–170. Chard T, Macintosh M, Yoong A: Customised antenatal growth charts. Lancet 1992;339:878. Kelly A, Kevany J, de Onis M, Shah PM: A WHO Collaborative Study of Maternal Anthropometry and Pregnancy Outcomes. Int J Gynaecol Obstet 1996;53:219–233. Bracero LA, Schulman H, Fleischer G, Farmakides G, Rochelson B: Umbilical artery velocimetry in diabetes and pregnancy. Obstet Gynecol 1986;68:654. Hütter W, Grab D, Ehman I, Stoz F, Wolf A: Die Wertigkeit dopplersonographischer Untersuchungen bei insulinpflichtigen Diabetikerinnen. Z Geburtshilfe Perinatol 1993;197:38–42. Dewey KG, Heining MJ, Nommsen LA: Maternal weight-loss patterns during prolonged lactation. Am J Clin Nutr 1993;58:162–166. Kacew S: Adverse effects of drugs and chemicals in breast milk of the nursing infant. J Clin Pharmacol 1993;33:213–221.

Josip Djelmiš, MD, PhD Department of Obstetrics and Gynecology Professor, State Referral Centre for Diabetes in Pregnancy, University of Zagreb Petrova 13, HR–10000 Zagreb (Croatia) Tel./Fax ⫹385 14604740, E-Mail [email protected]

Clinical Management of Diabetic Pregnancy

173

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 174–194

Nutritional Management of Diabetes in Pregnancy Anne Dornhorsta, Gary Frostb a Department of Metabolic Medicine and bDepartment of Nutrition and Dietetics, Hammersmith Hospital, London, UK

Maternal hyperglycaemia influences fetal growth and development. Fetal metabolism and metabolic programming are influenced by maternal hyperglycaemia resulting in changes that predispose to insulin resistance, obesity and diabetes in later life. Minimizing maternal hyperglycaemia by dietary management is an important part of the management of a diabetic pregnancy. However, glucose is not the only maternal nutrient that influences fetal growth. The optimal diet for a diabetic pregnant woman is not known. While some dietetic principles for pregnancy are scientifically founded, uncertainty and controversy surround many others (see table 1). The nutritional requirements and advice for women with pre-existing type 1 diabetes, type 2 diabetes and gestational diabetes mellitus (GDM) are different. As all dietary advice has to be individually tailored around a given patient to ensure it is socially and culturally acceptable no two diets are ever the same. This chapter on the nutritional management of diabetes in pregnancy covers the knowledge and doubts around which we give dietary advice today. This advice will undoubtedly need to be modified in the light of future research in this field. Reducing the Risk of Having a Diabetic Pregnancy

Obesity and type 2 diabetes are inextirpably linked [1]; reducing obesity in women of child bearing age will reduce the incidence of GDM and type 2 diabetes. Weight loss is recommended for all overweight [body mass index (BMI) 25.0–29.9 kg/m2] or obese (BMI 30.0 kg/m2) adults who have, or who are at risk for developing, type 2 diabetes [2]. This is especially important prior to any pregnancy given that both obesity [3] and diabetes are risk factors for

Table 1. Current areas of controversy Optimal weight gain Macronutrient composition Vitamin supplementation Fish oils Antioxidants Influence of maternal diet on fetal programming

poor pregnancy outcome. In addition there is now ample evidence that weight management coupled with increased physical activity can reduce the incidence of obesity and type 2 diabetes [4, 5]. Dietary modification between pregnancies can also reduce the risk of a GDM pregnancy recurring [6]. Reducing obesity in the population is a massive public health undertaking, but it will be the only way in which the incidence of maternal type 2 diabetes and GDM will fall and with it the lasting consequences of earlier onset diabetes in the subsequent generations [7].

General Nutritional Advice before Pregnancy

Ideally all known diabetic women contemplating pregnancy will have an opportunity to review their diet and receive general advice on healthy eating prior to becoming pregnant. Pre-pregnancy is a time to optimize body weight and adopt healthy eating patterns. Both under- and overnutrition affect fertility and therefore the likelihood of pregnancy. Adequate body fat stores are required for the onset of menarche and subsequent fertility [8], while excessive fat stores are associated with irregular menses and anovular cycles, especially if polycystic ovarian syndrome is also present [9]. It is important prior to pregnancy to identify and address any women with eating disorders. Both anorexia nervosa and bulimia are twice as frequent among women with type 1 diabetes than those without [10, 11]. Young type 1 diabetic women may also manipulate their insulin, sometimes omitting it altogether, to control their weight [12]. Dietetic assessment and advice for women with type 1 diabetes prior to pregnancy must address the importance of regular meals giving advice, when necessary, on how to avoid unnecessary weight gain while ensuing adequate insulin is given. Encouraging weight loss in obese type 2 diabetic women will not only reduce the risk during pregnancy, including the risk of congenital malformations

Nutritional Management of Diabetes in Pregnancy

175

Table 2. Main topics that need to be covered when giving preconception dietary advice Folic acid Alcohol Vitamin and antioxidant supplementation Fish oils Foods to avoid

[3, 13, 14] but will also help fertility in obese women with anovular cycles [15, 16]. Overweight women with a previous history of GDM also benefit from general dietetic advice as their likelihood of a recurrence in a subsequent pregnancy is reduced if they adhere to a low-fat diet and minimize their weight gain between pregnancies [6, 17]. Pregnancies among all obese and morbidly obese women, whether diabetic or not, are high-risk pregnancies, and all efforts should be made to reduce obesity prior to conception [3]. Minimizing obesity among women of childbearing age will have the long-term benefit of protecting against cardiovascular disease in later life [18].

Specific Preconception Dietary Advice for Women with Diabetes

The main dietary topics that need to be covered when giving preconception dietary advice are listed in table 2. Folic Acid Folate is a water-soluble B vitamin and enzymatic cofactor that is necessary for the synthesis of purine and thymidine nucteotides and for the synthesis of methionine from homocysteine. Impaired folate metabolism increases the risk of developmental anomalies including neural tube defects. Daily supplementation of 400 ␮g folic acid taken with a folate-rich diet before and during the first 12 weeks of gestation reduces the risk of neural tube defects [19]. As women with diabetes are at increased risk of giving birth to infants with neural tube defects, they like other high-risk groups are recommended to take 5 mg folate daily [19]. This should be done despite there being no direct evidence for abnormal folate metabolism in human diabetic pregnancies [20, 21]. The potential beneficial effect of folic acid supplementation is its antioxidant properties that can block the accumulation of the cellular oxidant homocysteine. Oxidative

Dornhorst/Frost

176

stress is increased in maternal diabetes and has been implicated in neural tube defects by inhibiting the expression of Pax-3, a gene critical for neural tube closure [22]. One caveat when prescribing the higher dose of folic acid to type 1 diabetic women is that these women are at increased risk of a low B12 level due to their increased risk of the autoimmune disease pernicious anaemia. It is therefore important to minimize the risk of undiagnosed vitamin B12 deficiency in these women [21]. Alcohol Diabetic women should be made aware of the dangers of excess alcohol in pregnancy [23]. Alcohol has known teratogenic effects that can partly be attributed to oxidative stress and generation of reactive oxygen species [24–26]. In a large Danish epidemiological study the rate of stillbirths was increased 6-fold for women consuming five or more alcoholic drinks a week compared to those consuming less than one drink a week (8.83 vs. 1.37 per 1,000 births) [27]. For women with type 1 diabetes alcohol has additional risk factors, the main being hypoglycaemia [28] and is therefore best avoided altogether in pregnancy when the risk of hypoglycaemia is already increased. For those with type 2 diabetes or GDM alcohol is an important, and unnecessary source, of calories and alcohol consumption should be restricted for this reason alone. General Vitamin Supplementation Oxidative stress is increasingly recognized as a unifying biochemical mechanism involved in fetotoxicity, including diabetic-related malformations [29]. Antioxidants given to experimental diabetic rodents reduce fetal malformations. Embryonic development is exquisitely sensitive to oxidants generated even during normal metabolism, perhaps because of the immaturity of the free radical scavenging pathways. The need for free radical scavenging is likely to be greater in embryos of diabetic mothers as increased fuel metabolism generates more oxidants [22]. Oxidative stress can lead to the peroxidation of lipids, nucleic acids, proteins, and carbohydrates that can cause chromosomal, enzymatic and cellular membrane dysfunction. A number of cellular enzymes and vitamins, including vitamin C, vitamin E, and ␤-carotene, possess antioxidant activity that helps to neutralize these reactive oxygen species. Supplementation of vitamin E and vitamin C to diabetic rats helps reduce fetal malformations as well as the levels of oxygen radicals in the fetal liver [30]. Recently dietary supplementation with high doses of vitamin C and E, both antioxidants, have been shown to reduce the incidence of pre-eclampsia in highrisk pregnancies [31]. Further studies are under way looking at high doses of vitamin C and E in diabetic pregnancies. As many type 1 diabetic women with proteinurea are at increased risk of pre-eclampsia, supplementation with high

Nutritional Management of Diabetes in Pregnancy

177

doses of vitamin C and E is likely to benefit them. It is hypothesized that these vitamins have their effect by reducing lipid peroxidation that is increased in preeclamptic placenta due in part to the decrease in activity of placental antioxidant enzymes [32]. During the first trimester the human placenta has limited antioxidant enzyme capacity of its own with very low levels of glutathione. Therefore, the placenta may be dependent on maternal vitamins C and E for their antioxidant capacity early in pregnancy when the fetus is most vulnerable to oxidative stress. It is currently unknown whether dietary supplementation of vitamins with antioxidant activities given in the periconceptional period can protect against birth defects in human diabetic pregnancies. To date the doses of vitamin C and vitamin E used in the pre-eclampsia trials have been 1,000 mg and 400 IU, respectively, both 10-fold higher than that found in over-the-counter multivitamin supplements sold for pregnancy. Calcium and vitamin D supplements both during pregnancy and lactation may be required for women from areas in the world, where due to cultural dressing or climate they have poor sunlight exposure or low calcium intakes [33, 34]. Fish Oils Long-chain n–3 fatty acids (long-chain polyunsaturated fatty acids, PUFA) are found in fish and fish oils as well as flaxseed oil, canola oil, soybean oil, and nuts. They comprise of the long-chain essential dietary PUFA linolenic acid (18:3), from which eicosapentaenoic acid 20:5 n–3 and docosahexaenoic acid 22:6 n–3 are derived. A number of clinical and epidemiological studies in pregnant and lactating mothers have reported benefits from a high n–3 fatty acid diet. A Danish prospective study of 8,729 pregnant women found that diets containing high levels of n–3 PUFA levels were protective against low birth weights and preterm deliveries [35]. In this study women who consumed fish or seafood once a week during the first trimester had a 3.6 times lower risk of giving birth to a low birth weight (⬍2,500 g) or premature infant (⬍259 days) than women who did not. Multicentre randomized clinical trials in high-risk pregnancies have also shown that fish oil supplementation reduces the risk of preterm delivery [36]. In other studies diets high in n–3 PUFA either during pregnancy, lactation or given as formula feeds have been reported to improve early retinal and infant intellectual development, and to lessen the risk of postpartum depression [37]. One of the potential beneficial effects of n–3 PUFAs in diabetic pregnancy is to lessen the long-term metabolic abnormalities associated with macrosomia [38]. General population studies suggest that foods containing n–3 PUFA are also cardioprotective [39, 40]. Caution is required, however, before universally advocating pregnant women to take fish oil supplements, as a high-fish oil diet given to pregnant and

Dornhorst/Frost

178

Table 3. Main aspects of dietetic management that need to be addressed Energy requirement Macronutrient composition of the diet Amount and type of dietary carbohydrate Amount and type of dietary fat Meal planning and avoidance of hypoglycaemia

lactating rats can inhibit fetal fatty acid metabolism resulting in low arachidonic acid levels [41]. Also in the follow-up studies of children born to mothers who were advised to follow a high-protein low-carbohydrate diet in pregnancy a high maternal fish consumption was associated with higher diastolic blood pressure in the offspring [42]. Finally the benefits of the n–3 fatty acids may be dependent on their ratio with the n–6 fatty acids, and without further research in this field the optimal maternal n–3 PUFA level in the diet remains uncertain. As the evidence for a precise ratio between saturated:poly:mono and fish oils is not known for pregnancy. A pragmatic view that reflects the recommendations for diabetes and coronary heart disease is to advise oily fish be included in the diet 3 times a week [43]. Foods to Avoid Vitamin A supplements and liver should be avoided in pregnancy due to the association of excess vitamin A levels with facial, ear, eye and palate deformities [21, 44]. Raw meat should be avoided because of the (small) risk of toxoplasmosis; soft and mould-ripened cheeses and pates are also best avoided as they may occasionally carry listeria.

Dietary Advice for Women with Diabetes during Pregnancy

Dietary advice is required throughout pregnancy for all women with diabetes. While certain general dietetic principles are applicable to all pregnant women with diabetes advice will vary depending on whether the mother has type 1, type 2 or GDM. The main aspects of dietetic management that need to be addressed are listed in table 3. It is essential that this advice is sound as there is increasing evidence that an unbalanced maternal diet during pregnancy may programme detrimental lifelong metabolic changes in the fetus [42, 45].

Nutritional Management of Diabetes in Pregnancy

179

Energy Requirements in Pregnancy Getting the dietary energy content right for pregnancy is important. Epidemiological evidence shows that both undernutrition and overnutrition of the mother have a detrimental effect on the pregnancy and the long-term health of the child. The concept of metabolic programming as the basis for the ‘fetal origins’ hypothesis as described by Barker [46] suggests that alterations in fetal nutrition and endocrine status cause developmental adaptations that predispose to cardiovascular, metabolic, and endocrine diseases in adulthood. It is postulated that inadequate or poor nutrition can change the fetal hypothalamiccortisol axis. These programmed effects on the hypothalamic-pituitary-adrenal axis may influence reactivity rather than resting secretion of cortisol and other stress hormones levels that may over the long term promote and propagate insulin resistance [47–49]. The nutrition of the fetus is dependent both on maternal nutrient fuels and uteroplacental blood flow. In type 1 diabetic mothers, with microvascular disease and nephropathy, placental blood supply can be severely limited and the fetus affectively starved of maternal nutrients due to poor placental blood flow rather than an inadequate maternal diet. The extra energy requirements for pregnancy are extremely modest, due to maternal metabolic adaptation. The UK dietary reference values for pregnancy are less than the American recommendations of an extra 300 kcal/day during the second and third trimesters [50, 51]. A number of studies have shown that well-nourished western woman need only to increase their food intake by 0.8 MJ/day (200 kcal/day) in the third trimester to meet the entire energy costs of pregnancy [52, 53]. Longitudinal studies have shown that a 20% increase in dietary input is all that is required for well-nourished pregnant women [54, 55]. Physical activity is reduced in most pregnancies and a 20% reduction by itself is sufficient to meet the entire energy costs for most pregnancies [52]. Under environmental pressure leading to inadequate energy intake maternal metabolic adaptation diverts nutrient fuels from the mother to the fetus. Under such circumstances there is a decrease in maternal fat stores and a fall in diet-induced thermogenesis and physical activity levels [53, 55–58]. Under extreme environmental pressure maternal basal metabolic rate falls in early pregnancy [54]. There are several examples in history of women continuing to give birth during famines. This is due to both maternal and fetal physiology being able to adapt to limited food supply through metabolic adaptation. Critical fetal metabolic pathways and endocrine changes occur in utero in response to limited nutrient supply that then persist into adult life. These changes undoubtedly favour short-term survival over long-term metabolic health ensuring essential fuels such as glucose are diverted away from adipose tissue and the pancreas to the developing brain. An example of intrauterine metabolic

Dornhorst/Frost

180

programming providing short-term survival advantages over long-term metabolic health is highlighted by the children born during the 5-month Dutch famine of 1944–1945. Children born to women who had access to only 800 kcal/day in late pregnancy were healthy and thin at birth remaining thinner during adolescence and early adulthood than their peers. However, by the time the women were 40 years old they had an increased incidence of glucose intolerance, diabetes and obesity [59–61]. Epidemiological studies across diverse populations have also shown that when infants fail to meet their growth potential they are at greater risk of developing the metabolic syndrome and all its associated conditions in later life. It is hypothesized that insulin resistance in muscles and fat is programmed in utero under conditions of limited maternal nutrients to allow glucose to be diverted to vital developing organs. Another potential adverse effect of energy deprivation is from fasting ketosis. Maternal ketonaemia, increased plasma ␤-hydroxybutyrate and free fatty acid levels have been linked with poor intellectual and neurophysiological development in early childhood [62, 63]. The degree of maternal ketosis required to pose a risk to the neurophysiological development of the child is unknown. It is, therefore, essential that all maternal diets provide sufficient calories to ensure fetal growth is not compromised and avoid unnecessary ketonaemia. While a lot of research has been done around the subject of fetal undernutrition it is now increasingly apparent that overnutrition is also detrimental to the pregnancy and the long-term health of the child. All pregnancies complicated with diabetes are at risk of delivering an excess of maternal fuel to the developing fetus resulting in a growth-promoted fetus. In addition to fetal macrosomia the lasting legacy of poor maternal diabetic control is one of childhood obesity, insulin resistance at puberty and early onset of type 2 diabetes [64, 65]. This is a consequence of maternal hyperglycaemia rather than a genetic susceptibility since the extent of the future metabolic derangement in the child appears proportional to the degree of maternal derangement in utero, and this is so for both children of type 1 and type 2 diabetic mothers. Preventing growth promotion and fetal overnutrition by not only targeting maternal hyperglycaemia but also by limiting excessive weight gain in obese and overweight pregnant women appears rational [66, 67]. The American guidelines currently endorse a modest 30–33% calorie restriction (to ⬎25 kcal/kg actual weight/ day) for obese women (BMI ⬎30 kg/m2) with GDM providing it is not associated with ketonuria [51, 68, 69]. As many women with type 2 diabetes and GDM are obese this advice seems sensible. However, without knowing the optimal weight gain for obese pregnant diabetic type 2 and GDM women caution is still required when advocating calorie restrictions, even to women with ample pre-pregnancy adipose stores to meet all the energy costs required.

Nutritional Management of Diabetes in Pregnancy

181

Macronutrients Composition

Protein The current American nutritional guidelines for diabetic women during pregnancy and lactation are for a daily protein intake of 0.75 g/kg plus an additional 10 g/day [70]. When advising on diet it is important to know that the advice given is beneficial and not harmful. The importance of getting the proportion of carbohydrate, fat and protein correct is underlined by the Motherwell experience. Between 1938 and 1977 all pregnant women in the relatively prosperous Motherwell and Wishaw district of Scotland were advised to consume a highprotein and low-carbohydrate diet with no specific instruction to reduce their energy intake [71, 72]. To achieve this they were advised to eat 1 lb (0.45 kg) of red meat per day. This advice was given in the belief that this diet would reduce the incidence of pre-eclampsia. As a consequence the babies born to primigravida mothers in Motherwell compared to those born to primigravida mothers from neighbouring Aberdeen were on average 400 g lighter (2,940 vs. 3,393 g). The typical Aberdeen primigravida diet at this time consisted of 54% carbohydrate, 34% fat and 12% protein, with a total daily calorie intake in the 2nd trimester of around 2,500 kcal as compared with the Motherwell mothers’ diet that was made up of 34% carbohydrate, 42% fat and 24% protein [71]. There is some evidence that the diet of the Motherwell women contained significantly fewer calories. Interestingly the children born to the mothers from Motherwell at this time had higher blood pressures as young adults [42] and raised cortisol levels in response to stress [45]. A high-protein low-carbohydrate diet is potentially metabolically unbalanced as it provides proportionally more essential amino acids without sufficient nutrients to utilize them. High-protein diets reduce the appetite, which may also explain the reduction in birth weight seen in a number of trials involving protein supplementation in pregnancy [42, 73]. Recently a high-protein diet containing processed meat has been linked with an increased risk of future type 2 diabetes, again suggesting that replacing carbohydrate for protein in our western diet may have long-term consequences for women with GDM [74]. The Motherwell experience of blanket dietary advice given to a generation of pregnant women shows that nutritional policies for pregnancy need to be based on evidence and not conjecture. The Amount and Type of Dietary Carbohydrate The current American dietary recommendations for a diabetic pregnancy are more restrictive in carbohydrates than the European guidelines, the latter advocating ⬎45% of the energy be in the form carbohydrates [75, 76] while the American recommendations suggest only 35–40% of calories be carbohydrate

Dornhorst/Frost

182

while increasing dietary fat to 40% [77]. Clearly if 35–40% of energy is carbohydrate 60–65% will need to be in the form of protein and fat. The potential problems of a low-carbohydrate high-protein diet are given above, while the problems of advocating a low-carbohydrate high-fat diet are that it is sending out the wrong dietary message for individuals already at greater risk of cardiovascular disease. What is becoming more apparent is that it is the type of dietary carbohydrate and fat, rather than the exact proportion that is probably most important. In a recent prospective study of non-diabetic pregnant women in the USA an increased fat intake in the second trimester was associated with glucose intolerance and GDM. In this study for each percentage increase in dietary fat there was an increased risk of IGT of 7% and GDM of 6% [78]. By contrast for each percentage increase in carbohydrates that substituted fat there was a 6% decrease in the risk of both IGT and GDM. There are two dietary ways to reduce postprandial glucose excursions and maternal hyperglycaemia through dietary manipulation. First by limiting dietary carbohydrate and second by replacing high for low-glycaemic-index carbohydrates, i.e. those that are slowly absorbed. On present day evidence it is the type not the quantity of carbohydrate that influences the degree of glycaemia [79–81]. Low-glycaemic-index carbohydrates, such as those found in pulse vegetables, whole fruits, oats and barley, have been shown to have metabolic benefits in pregnant and non-pregnant individuals including reducing postprandial glycaemia [81] and insulin resistance [79, 80], as well as reducing overall fat intake and weight gain [82]. Low-glycaemic-index diets have the added benefit for type 1 diabetic women that by ensuring a prolonged and attenuated postprandial glycaemic response they protect against preprandial hypoglycaemia and ketosis [83, 84]. When non-obese African women consume a traditional rural diet with ⬎60% of their energy as low glycaemic starches and soluble non-starch polysaccharides their glucose tolerance during pregnancy does not deteriorate [79]. By contrast western women consuming the high-glycaemic dietary carbohydrates typically found in highly processed foods and soft drinks do experience a degree of glucose intolerance during pregnancy [85]. Clinical studies have shown that by simply replacing high-glycaemic-index carbohydrate with low-glycaemic carbohydrates in the western diet of non-diabetic pregnant women glucose tolerance improves, and similar diets have also been reported to lower birth weights [81, 86]. On present-day evidence it appears reasonable to encourage obese women with GDM to avoid rapidly absorbed sugars in favour of the slowly absorbed starches and soluble non-starch polysaccharides. However, as low-glycaemic carbohydrates are associated with greater satiety, in practice it is difficult to achieve greater than 50% of the total energy in this form.

Nutritional Management of Diabetes in Pregnancy

183

The Amount and Type of Dietary Fat One important question to address is if dietary carbohydrate is to be reduced and dietary fat increased what type of fat should we be advocating: saturated fat (SF), PUFA, monounsaturated fatty acids (MUFA) or n–3 PUFA? Any advantage of a low-carbohydrate high-fat diet on glycaemic control needs to be considered against any secondary adverse effect an increase in fat may have on insulin resistance or hypertriglyceridaemia. The strong epidemiological evidence linking SF intake to cardiovascular disease has resulted in a general agreement that dietary SF and cholesterol should represent less than 10% of the total energy intake [87]. Whether this advice is appropriate for pregnancy is not known. However, animal studies suggest that in pregnancy high-SF diets promote cardiovascular disease in the progeny [88, 89]. When rats are fed a diet with 20% of the total energy as SF before, during and after pregnancy their female offspring have a more athrogenic lipoprotein profile as adults than offspring of control rats fed 4% fat during pregnancy and weaning. The arachidonic and docosahexaenoic acid in the fatty acid compositions of the aorta is also reduced in the offspring of rats fed high-SF diets in pregnancy [88], who also have evidence of vascular dysfunction [89]. Interestingly these finding are more apparent in rats born to diabetic rats fed a high-SF diet in pregnancy. Until we know how applicable this animal data is to human pregnancy, we need to be cautious promoting diets that replace carbohydrate with fat. Under test meal conditions a breakfast containing SF rather than an isocaloric meal containing monounsaturated fat gives lower 3-hour glycaemic and insulin responses [90]; however, encouraging SF over monounsaturated fat may also be unwise as epidemiological studies have linked severe hyperemesis gravidarum with high levels of dietary SF [91]. Another powerful argument for limiting dietary fat in women with a GDM is that recurrence of GDM in a subsequent pregnancy is increased in women on a high-fat diet [6]. As all diabetic women are at increased risk of future cardiovascular disease it seems prudent that pregnancy should be a time when all diabetic women are encouraged to adopt good dietary habits that will protect them in their future life; this should include advice that dietary SF and cholesterol should not exceed 10% of the total energy intake [87]. Recently epidemiological studies have shown that a high habitual intake of PUFA in the diet improves glucose tolerance and protects against developing GDM [92]. A study in Shanghai, China involving 8,002 consecutive pregnant women showed that abnormal glucose tolerance in pregnancy was inversely related to the dietary intake of PUFA [92]. The Chinese women in this study ate and prepared traditional meals in their homes using soybean and vegetable oils as the food sources of PUFA and unlike in the west the PUFA intake would not have correlated with SF intake. As high PUFA intakes in the west are usually

Dornhorst/Frost

184

accompanied by a parallel rise in SF intake any benefit from increasing dietary PUFA on glucose tolerance may be lost. American diets high in SF consumed in the second trimester are associated with an increased risk of glucose intolerance and the percentage of GDM [78]. The potential benefits of replacing dietary SF with MUFA in pregnancy have not been established. Improved insulin resistance and serum triglyceride and LDL levels have been reported in non-pregnant individuals when the MUFA in the diet is increased [93, 94]. There is, however, concern that when high-MUFA diets are eaten ad libitum overall energy intake may increase [95]. A small Danish study failed to show any improvement in insulin sensitivity when women with GDM eat high-MUFA diets in the third trimester compared with high-carbohydrate diets, although the high-MUFA diets did appear to have a favourable effect on blood pressure [96]. The optimal ratio between saturated:poly:mono and fish oils is not known for pregnancy. A pragmatic view would be to reflect the recommendations for diabetes and coronary heart disease aiming for a ratio of saturated:poly:mono of 1:1:1 with the specific advice to eat oily fish 3 times a week [43]. Meal Planning and Avoidance of Hypoglycaemia For women with pre-existing type 1 or type 2 diabetes nutritional advice needs to be given to help distribute the energy and carbohydrate intake throughout the day using a meal plan that accommodates the woman’s food and eating habits, her glycaemic control, and the stage in pregnancy. Regular meals and snacks are imperative to allow for the necessary insulin titration to be made to achieve euglycaemia without hypoglycaemia. Carbohydrate should be distributed throughout the day incorporated in both meals and snacks. Some form of low-glycaemic carbohydrate snack is usually required before bedtime to prevent overnight ketosis. Refined rapidly absorbed carbohydrate cereals at breakfast should be avoided as they cause high mid morning blood glucose. Specific food recommendations and meal schedules need to be assessed individually using information from the home glucose monitoring record, knowledge of the current insulin regimen and the women’s activity schedule. Structured education programmes such as the Düsseldorf Training for Diabetes endorsed by the German Diabetes Society and the UK equivalent [97] can help women to better understand the relationship between dietary carbohydrate and insulin requirement. Empowering an individual to adjust their own insulin dosage rather than the diet itself allows some freedom from a ridged traditional meal pattern and with the newer fast-acting insulin analogues this approach has become increasingly popular and easier. However, as so much in this field, direct evidence is lacking of whether this more relaxed approach to the diabetic diet management is appropriate for pregnancy.

Nutritional Management of Diabetes in Pregnancy

185

Table 4. Recommended total weight gain ranges for pregnant women by pregnancy BMI (kg/m2) [adapted from 99] Pre-pregnancy category

Total weight gain kg

Weight gain 2nd and 3rd trimester kg/4 weeks

Underweight (BMI ⬍19.8) Normal (BMI 26.1–26.0) Overweight (BMI 26.1–29.0) Obesity (BMI ⬎29.0) Twins

12.7–18.2 11.4–15.9 6.8–20.4 ⬎6.8 15.9–20.4

2.3 1.8 1.2 0.6 2.7

Recommended Maternal Weight Gain in Pregnancy Advising women on weight gain is particularly important in women with type 2 diabetes and GDM. Excessive weight gain is not only associated with problems in pregnancy but with glycaemic control and health risk postpartum. Advice regarding optimal weight gain is much less controversial among underweight and normal weight women than among obese women. The optimal weight gain for an obese and morbidly obese woman with type 2 diabetes or GDM has not been defined in randomized clinical studies. It is agreed, however, that weight gain for pregnancy should reflect the woman’s pre-pregnancy weight [98]. The guidelines on recommended maternal weight gains in non-diabetic women are based on large obstetric surveys in the United States [99] (see table 4). The maternal weight gain required to minimize the frequency of small-for-gestational age infants is higher for underweight (BMI ⬍19.8 kg/m2) than overweight or obese women (see table 4). When the pre-pregnancy BMI is ⬎35 kg/m2 little or no maternal weight gain does not increase the risk of a small-for-gestational age infant [100]. Overweight (BMI 26.1–29 kg/m2) and obese (BMI ⬎29 kg/m2) women are more susceptible to give birth to a large-for-gestational age infant and this risk increases as maternal weight gain increases. The US obstetric recommendation for a 7-kg minimum weight gain for all obese women [99, 101] may not be universally appropriate, especially for morbidly obese women (BMI is ⬎35 kg/m2) [100, 102]. As in well-nourished, inactive obese pregnant women minimal or no weight gain in pregnancy does not appear to jeopardize birth weight or other pregnancy outcomes [99, 100]. Among these obese women dietary intake and gestational weight gains correlate poorly with infant birth weight. As the majority of women with pre-existing type 2 and GDM are already obese it is important that the dietary advice given will not result in the postpartum weight exceeding the pre-pregnancy weight.

Dornhorst/Frost

186

Infant birth weights remain a crude reflection of infant health; this is especially true for infants of diabetic mothers. Large babies born to diabetic mothers can, and often are, malnourished in the sense that they are overnourished. A birth weight ⬎4,000 g or at or above the 95th percentile for the vast majority of babies born at term is completely normal, and will reflect that these babies have been at the 95th percentile for growth throughout the pregnancy. What distinguishes these healthy babies from inappropriately large babies of diabetic mothers is not the final absolute birth weight but the pattern of fetal growth in utero [67]. Infants of diabetic mothers frequently have evidence of gaining too much weight in utero. Serial ultrasounds show these infants can acquire excess visceral adipose stores resulting in accelerated fetal growth patterns that cross rather than travel along the normal percentile growth lines. Hence a fetus may start on the 40th percentile line at the start of the second trimester but have crossed the 90th percentile by the end of the third trimester. There is emerging evidence that a high birth weight due to accelerated fetal growth has a detrimental effect on the infant’s long-term health in contrast to a high birth weight in a normally developed infant, which in large epidemiological studies has been associated with a reduced longterm risk of diabetes [103, 104]. Low birth weight is less common in infants of mothers with GDM than in the background population or the population of women with type 1 diabetes; this probably reflects the degree of maternal obesity and that women with GDM have a lower risk of microvascular disease and poor placental function than type 1 diabetic women. The published literature to date would suggest that limiting maternal weight gain in the obese women with GDM and limiting maternal glycaemia can correct for accelerated fetal growth without increasing the risk of delivery of a small-for-gestational age infant [66, 67]. Low birth weight has been linked with adult diseases in later life including type 2 diabetes, the so-called ‘Barker hypothesis’ [105]. As with high birth weight a low term birth weight of less than the 10th percentile can be appropriate for an infant that has grown up this percentile line throughout pregnancy.

Weight Loss in the Obese Diabetic Women during Pregnancy

Weight management will become an increasing part of everyday clinical obstetrics as obesity becomes more prevalent among young women. Although there are genuine concerns around calorie restriction, as addressed above, obese glucose-intolerant women become less ketotic in response to a mild calorie restriction than glucose-tolerant women, due to their higher hepatic glucose output [106]. A modest degree of calorie constraint is probably safe in GDM [107, 108]. Ketosis is minimized when calorie-restricted diets are given as small frequent meals containing slowly absorbed carbohydrates. This is because the

Nutritional Management of Diabetes in Pregnancy

187

insulin response produced by this type of diet is attenuated and this helps delay lipolysis and ketogenesis between meals [109]. We have previously reported that modest calorie restriction in obese GDM women (20–25 kcal/kg/day) from the 24th gestational week results in fewer large-for-gestational age infants than when obese glucose-tolerant women do not diet. In this study the GDM women gained only half the weight of the controls from 28 weeks to term (1.7 ⫾ 1.6 vs. 4.1 ⫾ 3.1 kg) [110]. A similar degree of calorie restriction has been shown to improve glycaemic control [108]. On present evidence we recommend that weight gain in GDM pregnancies is limited to the bottom rather than the top end of the range of that recommended for average, overweight and obese women, with the proviso that when the BMI is ⬎34 kg/m2 no minimum weight gain should be set. This policy helps towards achieving no overall weight gain in the overweight women and provides the possibility of weight loss in the morbidly obese women.

Popular Weight Loss Diets in Pregnancy

The Atkins diet is currently very popular among young women trying to lose weight; this diet specifically advocates replacing dietary carbohydrate with fat and protein and as such it is potentially hazardous for diabetic and non-diabetic pregnant women. The Atkins diet is not dissimilar to the diet recommended for pregnant women in Motherwell in the mid 20th century [42], a diet now known to have resulted in reduced birth weights and adverse effects on blood pressure and the hypothalamic-cortisol axis in adult life (see above). The use of very-low-energy diets for the purpose of weight reduction has not been evaluated in human pregnancies; however, epidemiological evidence from the Dutch famine and animal experiments on extreme calorie restriction should guard us against their use [49]. Until proved otherwise any weight loss diet that produces a chronic rise in ketones should not be used in pregnancy.

Exercise

Regular aerobic exercise has been shown to lower fasting and postprandial glucose concentrations and may be used as an adjunct to nutritional therapy to improve maternal glycaemia [111]. For women with GDM exercise delays the need to initiate insulin and reduces the overall dose of insulin required [112]. There is insufficient evidence to recommend any specific type of exercise. Adjusting nutritional therapy around planned times of physical activity to avoid hypoglycaemia needs to be done on an individual basis, and should not be used as a barrier for not advising exercise. Encouraging women to continue an

Dornhorst/Frost

188

exercise programme postpartum is also important, especially for women with GDM as this will lessen the risk of developing GDM in subsequent pregnancies and type 2 diabetes later in life.

Lactation

Breastfeeding is recommended for women with pre-existing diabetes and GDM; however, successful lactation requires planning and coordination of care. Breastfeeding lowers blood glucose, and insulin-treated women may require a carbohydrate-containing snack either before or during breastfeeding. Energy requirements during the first 6 months of lactation are 200 calories in addition to the pregnancy meal plan. However, an energy intake of 1,800 kcal/day usually meets the nutritional requirements for lactation and this may allow for a gradual weight loss in overweight type 2 individuals or those who had GDM. Most mothers, however, increase their dietary intake over the period of lactation unintentionally.

Delaying Type 2 Diabetes in Women with Previous GDM

For women with GDM there is now compelling evidence that their risk of future diabetes can be reduced through aggressive lifestyle modification [4, 5]. The intensity of health care input and support provided in these intervention research studies is not achievable with current health resources in any country. However, this should not deter us from giving simple, unambiguous dietary and lifestyle advice. This advice should be to decrease high-fat diary products and animal fats while increasing fruits and vegetables and wholegrain cereals. A regular exercise regime of 30 min a day should be recommended to accompany the dietary advice for all women and a weight reduction target of 5–10% of body weight set for all women who are obese.

References 1 2

3

Mokdad A, Ford E, Bowman B, et al: Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA 2003;289:76–79. Klein S, Sheard N, Pi-Sunyer X, et al: Weight management through lifestyle modification for the prevention and management of type 2 diabetes: Rationale and strategies: A statement of the American Diabetes Association, the North American Association for the Study of Obesity, and the American Society for Clinical Nutrition. Diabetes Care 2004;27:2067–2073. Cedergren M: Maternal morbid obesity and the risk of adverse pregnancy outcome. Obstet Gynecol 2004;103:219–224.

Nutritional Management of Diabetes in Pregnancy

189

4 5 6 7 8 9 10 11 12

13 14

15 16 17 18 19 20 21

22

23 24 25 26 27 28 29

Tuomilehto J, Lindström J, Eriksson JG, et al: Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–1350. Diabetes Prevention Program Research Group: Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. Moses RG, Shand JL, Tapsell LC: The recurrence of gestational diabetes: Could dietary differences in fat intake be an explanation? Diabetes Care 1997;20:1647–1650. Dornhorst A: Maternal hyperglycaemia – Food for thought. Pract Diabetes Int 2003;20:283–289. Matkovic V, Ilich JZ, Skugur M, et al: Leptin is inversely related to age at menarche in human females. J Clin Endocrinol Metab 1997;82:1368–1372. Legro RS: Diabetes prevalence and risk factors in polycystic ovary syndrome. Curr Opin Endocrinol Diabetes 2002;9:451–458. Hoffman RP: Eating disorders in adolescents with type 1 diabetes. A closer look at a complicated condition. Postgrad Med 2001;109:67–69. Jones JM, Lawson ML, Daneman D, Olmsted MP, Rodin G: Eating disorders in adolescent females with and without type 1 diabetes: Cross-sectional study. BMJ 2000;320:1563–1566. Neumark-Sztainer D, Patterson J, Mellin A, et al: Weight control practices and disordered eating behaviors among adolescent females and males with type 1 diabetes: Associations with sociodemographics, weight concerns, familial factors, and metabolic outcomes. Diabetes Care 2002;25:1289–1296. Watkins ML, Rasmussen SA, Honein MA, Botto LD, Moore CA: Maternal obesity and risk for birth defects. Pediatrics 2003;111:1152–1158. Garci-Patterson A, Erdozain L, Ginovart G, et al: In human gestational diabetes mellitus congenital malformations are related to pre-pregnancy body mass index and to severity of diabetes. Diabetologia 2004;47:509–514. Poston WSC, Foreyt JP: Infertility and obesity: The effects of behavioral therapy, diet, and exercise. Infertility Reprod Med Clin North Am 2000;11:315–328. Fraser RB, Chan KL: Problems of obesity in obstetric care. Curr Obstet Gynaecol 2003;13: 239–243. Bo S, Menato G, Lezo A, et al: Dietary fat and gestational hyperglycaemia. Diabetologia 2001;44: 972–978. Krauss R, Winston M, Fletche R, Grundy S: Obesity: Impact of cardiovascular disease. Circulation 1998;98:1472–1476. The MRC Vitamin Research Study Group: Prevention of Neural Tube Defects: The Results of the Medical Research Council Vitamin Study. Lancet 1991;338:131–137. Kaplan JS, Iqbal S, England BG, Zawacki CM, Herman WH: Is pregnancy in diabetic women associated with folate deficiency? Diabetes Care 1999;22:1017–1021. Wilson R, Davies G, Desilets V, et al: Genetics Committee and Executive and Council of the Society of Obstetricians and Gynaecologists of Canada. The use of folic acid for the prevention of neural tube defects and other congenital anomalies. J Obstet Gynaecol Can 2003;25:959–973. Chang TI, Horal M, Jain SK, Wang F, Patel R, Loeken MR: Oxidant regulation of gene expression and neural tube development: Insights gained from diabetic pregnancy on molecular causes of neural tube defects. Diabetologia 2003;46:538–545. Janerich DT: Alcohol and pregnancy. An epidemiologic perspective. Ann Epidemiol 1990;1: 179–185. Kotch LE, Chen SE, Sulik KK: Ethanol-induced teratogenesis: Free radical damage as a possible mechanism. Teratology 1995;52:128–136. Amini SA, Dunstan RH, Dunkley PR, Murdoch RN: Oxidative stress and the fetotoxicity of alcohol consumption during pregnancy. Free Radic Biol Med 1996;21:357–365. Cohen-Kerem R, Koren G: Antioxidants and fetal protection against ethanol teratogenicity: Review of the experimental data and implications to humans. Neurotoxicol Teratol 2003;25:1–9. Kesmodel U, Wisborg K, Olsen SF, Henriksen TB, Secher NJ: Moderate alcohol intake during pregnancy and the risk of stillbirth and death in the first year of life. Am J Epidemiol 2002;155:305–312. Meeking DR, Cavan DA: Alcohol ingestion and glycaemic control in patients with insulindependent diabetes mellitus. Diabet Med 1997;14:279–283. Jauniaux E, Cindrova-Davies T, Johns J, et al: Distribution and transfer pathways of antioxidant molecules inside the first trimester human gestational sac. J Clin Endocrinol Metab 2004;89:1452–1458.

Dornhorst/Frost

190

30

31 32 33 34 35 36

37

38 39

40 41 42 43

44 45

46 47 48

49

50 51 52 53 54

Cederberg J, Siman C, Eriksson U: Combined treatment with vitamin E and vitamin C decreases oxidative stress and improves fetal outcome in experimental diabetic pregnancy. Pediatr Res 2001;49:755–762. Chappell LC, Seed PT, Briley AL, et al: Effect of antioxidants on the occurrence of pre-eclampsia in women at increased risk: A randomised trial. Lancet 1999;354:810–816. Poranen AK, Ekblad U, Uotila P, Ahotupa M: Lipid peroxidation and antioxidants in normal and pre-eclamptic pregnancies. Placenta 1996;17:401–405. Daaboul J, Sanderson S: Vitamin D deficiency in pregnant and breast-feeding women and their infants. J Perinatol 1997;17:10–14. Waiters B, Godel JC, Basu TK: Perinatal vitamin D and calcium status of northern Canadian mothers and their newborn infants. J Am Coll Nutr 1999;18:122–126. Olsen SF, Secher NJ: Low consumption of seafood in early pregnancy as a risk factor for preterm delivery: Prospective cohort study. BMJ 2002;324:447–450. Olsen SF, Secher NJ, Tabor A, Weber T, Walker JJ, Gluud C: Randomised clinical trials of fish oil supplementation in high risk pregnancies. Fish Oil Trials In Pregnancy (FOTIP) Team. BJOG 2000;107:382–395. De Vriese SR, Christophe AB, Maes M: Lowered serum n–3 polyunsaturated fatty acid (PUFA) levels predict the occurrence of postpartum depression: Further evidence that lowered n-PUFAs are related to major depression. Life Sci 2003;73:3181–3187. Merzouk H, Khan NA: Implication of lipids in macrosomia of diabetic pregnancy: Can n–3 polyunsaturated fatty acids exert beneficial effects? Clin Sci 2003;105:519–529. de Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J, Mamelle N: Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: Final of the Lyon Diet Heart Study. Circulation 1999;99:733–735. Daviglus ML, Stamler J, Orencia AJ, et al: Fish consumption and the 30-year risk of fatal myocardial infarction. N Engl J Med 1997;336:1046–1053. Herrera E: Implications of dietary fatty acids during pregnancy on placental, fetal and postnatal development. Placenta 2002;23(suppl 1):S9–S19. Shiell AW, Barker DJP: High-meat, low-carbohydrate diet in pregnancy relation to adult blood pressure in the offspring. Hypertension 2001;38:1282–1288. Lean MEJ, Brenchley S, Connor H, Elkeles RS, Govindji A, Hartland BV: Dietary recommendations for people with diabetes: An update for the 1990s. Nutrition Subcommittee of the British Diabetic Association’s Professional Advisory Committee. Diabet Med 1992;9:189–202. Chief Medical Officer: Vitamin A and Pregnancy. London, Department of Health, HMSO, 1993. Herrick K, Phillips DIW, Haselden S, Alistair W, Shiell M, Campbell-Brown, Godfrey KM: Maternal consumption of a high-meat, low-carbohydrate diet in late pregnancy: Relation to adult cortisol concentrations in the offspring. J Clin Endocrinol Metab 2003;88:3554–3560. Barker DJP: Fetal growth and adult disease. Br J Obstet Gynaecol 1992;99:275–276. Reynolds R, Walker B, Syddall H, et al: Altered control of cortisol secretion in adult men with low birth weight and cardiovascular risk factors. J Clin Endocrinol Metab 2001;86:245–250. Ward A, Syddall H, Wood P, Chrousos G, Phillips D: Fetal programming of the hypothalamicpituitary-adrenal (HPA) axis: Low birth weight and central HPA regulation. J Clin Endocrinol Metab 2004;89:1227–1233. Bloomfield F, Oliver M, Giannoulias C, Gluckman P, Harding J, Challis J: Brief undernutrition in late-gestation sheep programs the hypothalamic-pituitary-adrenal axis in adult offspring. Endocrinology 2003;144:2933–2940. Committee on Medical Aspects of Food Policy: Dietary Reference Values for Food Energy and Nutrients for the United Kingdom. London, HMSO, 1991. Food and Nutrition Board: Recommended Daily Allowances, ed 10. Washington, National Academy, 1989, pp 59–62. Hytten F, Leitch I: The Physiology of Human Pregnancy. Oxford, Blackwell Scientific, 1964. Butte NF, Hopkinson JM, Mehta N, Moon JK, Smith EO: Adjustments in energy expenditure and substrate utilisation during late pregnancy and lactation. Am J Clin Nutr 1999;69:299–307. Durnin J: Energy requirements of pregnancy: An integrated study in five countries. Background and methods. Lancet 1987;ii:895–897.

Nutritional Management of Diabetes in Pregnancy

191

55 56 57

58 59 60 61 62 63

64

65 66 67 68 69 70 71

72 73

74 75 76

77 78 79

Kopp-Hoolihan LE, van Loan MD, Wong WW, King JC: Longitudinal assessment of energy balance in well nourished, pregnant women. Am J Clin Nutr 1999;69:697–704. Illingsworth PJ, Jung RT, Howie PW: Reduction in post-prandial energy expenditure during pregnancy. Br Med J 1987;294:1573–1576. Poppitt SD, Prentice AM, Jequier E, Schutz Y, Whitehead RG: Evidence of energy sparing in Gambian women during pregnancy: A longitudinal study using whole-body calorimetry. Am J Clin Nutr 1993;57:353–364. Poppitt SD, Prentice AM, Goldberg GR, Whitehead RG: Energy-sparing strategies to protect human fetal growth. Am J Obstet Gynecol 1994;171:118–125. Ravelli GP, Stein ZA, Susser MW: Obesity in young men after famine exposure in utero and early infancy. N Engl J Med 1976;295:349–353. Ravelli ACJ, van der Meulen JHP, Michels RP, et al: Glucose tolerance in adults after prenatal exposure to famine. Lancet 1998;351:173–177. Ravelli AC, van der Meulen JH, Osmond C, Barker DJ, Bleker O: Obesity at the age of 50 y in men and women exposed to famine prenatally. Am J Clin Nutr 1999;70:811–816. Churchill JA, Berendez HW, Nemore J: Neuropsychological deficits in children of diabetic mothers. Am J Obstet Gynecol 1966;105:257–268. Rizzo T, Dooley SL, Metzger B, Cho N, Ogata E, Silverman B: Prenatal and perinatal influences on long-term psychomotor development in offspring of diabetic mothers. Am J Obstet Gynecol 1995;173:1753–1758. Weiss U, Cervar M, Puerstner P, et al: Hyperglycaemia in vitro alters the proliferation and mitochondrial activity of the choriocarcinoma cell lines BeWo, JAR and JEG-3 as models for human first-trimester trophoblast. Diabetologia 2001;44:209–219. Dabelea D, Hanson RL, Lindsay RS, et al: Intrauterine exposure to diabetes conveys risks for type 2 diabetes and obesity: A study of discordant sibships. Diabetes 2000;49:2208–2211. Buchanan TA, Kjos SL, Montoro MN, et al: Use of fetal ultrasound to select metabolic therapy for pregnancies complicated by mild gestational diabetes. Diabetes Care 1994;17:275–283. Buchanan TA, Kjos SL, Schafer U, et al: Utility of fetal measurements in the management of gestational diabetes mellitus. Diabetes Care 1998;21(suppl 2):B99–B106. American Diabetes Association: Nutritional management; in Medical Management of Pregnancy Complicated by Diabetes. Alexandria, American Diabetes Association, 1995, pp 47–56. American Diabetes Association: Gestational diabetes mellitus. Diabetes Care 2004;27:S88–S90. American Diabetes Association: Position statement: Nutrition principles and recommendations in diabetes. Diabetes Care 2004;27:S36. Kerr JF, Cambell-Brown BM, Johnstone FD: A study of the effect of a high protein low carbohydrate diet on birthweight on an obstetric population; in Sutherland HW, Stowers JM (eds): Carbohydrate Metabolism in Pregnancy and the Newborn 1978. Berlin, Springer, 1979, pp 518–534. Hill AJ, Blundell JE: Macronutrients and satiety: The effects of high-protein or high-carbohydrate meal on subjective motivation to eat and food preferences. Nutr Behav 1986;3:133–144. Rush D: Effects of changes in maternal energy and protein intake during pregnancy, with special reference to fetal growth; in Sharp F, Fraser R, Milner D (eds): Fetal Growth. London, Royal College of Obstetricians and Gynaecologists, 1989. Schulze M, Manson J, Willett W, Hu F: Processed meat intake and incidence of type 2 diabetes in younger and middle-aged women. Diabetologia 2003;46:1465–1473. Ha KK, Lean MEJ: Recommendations for the nutritional management of patients with diabetes mellitus. Eur J Clin Nutr 1998;52:467–481. Connor H, Annan F, Bunn E, et al: Nutrition Subcommittee of the Diabetes Care Advisory Committee of Diabetes UK. The implementation of nutritional advice for people with diabetes. Diabet Med 2003;20:786–807. Major C, Henry M, De Veciana M, Morgan M: The effects of carbohydrate restriction in patients with diet-controlled gestational diabetes. Obstet Gynecol 1998;91:600–604. Saldana T, Siega-Riz A, Adair L: Effect of macronutrient intake on the development of glucose intolerance during pregnancy. Am J Clin Nutr 2004;79:479–486. Fraser RB: The normal range of OGTT in the African female: Pregnant and non-pregnant. East Afr Med J 1981;58:90–94.

Dornhorst/Frost

192

80 Fraser R, Ford F, Lawrence G: Insulin sensitivity in third trimester pregnancy. A randomized study of dietary effects. Br J Obstet Gynaecol 1988;95:223–229. 81 Clapp J: Effect of dietary carbohydrate on the glucose and insulin response to mixed caloric intake and exercise in both nonpregnant and pregnant women. Diabetes Care 1998;21(suppl 2): B107–B112. 82 Frost G, Leeds A, Trew G, Margara R, Dornhorst A: Insulin sensitivity in women at risk of coronary heart disease and the effect of a low glycaemic diet. Metabolism 1998;47:1245–1251. 83 Gilbertson HR, Brand-Miller JC, Thorburn AW, Evans S, Chondros P, Werther GA: The effect of flexible low glycemic index dietary versus measured carbohydrate exchange diets on glycemic control in children with type 1 diabetes. Diabetes Care 2001;24:1137–1143. 84 Giacco R, Parillo M, Rivellese AA, Lasorella G, D’Episcopo L, Riccardi G: Long-term dietary treatment with increased amounts of fibre-rich low-glycemic index natural foods improves blood glucose control and reduces the number of hypoglycemic events in type 1 diabetic patients. Diabetes Care 2000;23:1461–1466. 85 Frost G, Dornhorst A: The relevance of the glycaemic index to our understanding of dietary carbohydrates. Diabet Med 2000;17:336–345. 86 Clapp J: Maternal carbohydrate intake and pregnancy outcome. Proc Nutr Soc 2002;61:45–50. 87 Franz MJ, Bantle JP, Beebe CA, et al: Evidence-based nutrition principles and recommendations for the treatment and prevention of diabetes and related complications. Diabetes Care 2002;25: 148–198. 88 Ghosh P, Bitsanis D, Ghebremeskel K, Crawford M, Poston L: Abnormal aortic fatty acid composition and small artery function in offspring of rats fed a high fat diet in pregnancy. J Physiol 2001;533:815–822. 89 Koukkou E, Ghosh P, Lowy C, Poston L: Normal and diabetic rats fed saturated fat in pregnancy demonstrate vascular dysfunction. Circulation 1998;98:2899–2904. 90 Ilic S, Jovanovic L, Pettitt D: Comparison of the effect of saturated and monounsaturated fat on postprandial plasma glucose and insulin concentration in women with gestational diabetes mellitus. Am J Perinatol 1999;16:489–495. 91 Signorello L, Harlow B, Wang S, Erick M: Saturated fat intake and the risk of severe hyperemesis gravidarum. Epidemiology 1998;9:636–640. 92 Wang Y, Storlien L, Jenkins A, et al: Dietary variables and glucose tolerance in pregnancy. Diabetes Care 2000;23: 460–464. 93 Georgopoulous A, Bantle JP, Noutsou M, Swaim WR, Parker SJ: Differences in the metabolism of postprandial lipoproteins after a high-monounsaturated-fat versus a high-carbohydrate diet in patients with type 1 diabetes mellitus. Arterioscler Thromb Vasc Biol 1998;18:773–782. 94 Garg A, Bantle JP, Henry RR, et al: Effects of varying carbohydrate content of diet in patients with non-insulin dependent diabetes mellitus. JAMA 1994;271:1421–1428. 95 Yu-Poth S, Zhao G, Etherton T, Naglak M, Jonnalagadda S, Kris-Etherton PM: Effect of National Cholesterol Education Program’s Step 1 and Step 11 dietary intervention programs of cardiovascular disease risk factors; a meta-analysis. Am J Clin Nutr 1999;69:632–646. 96 Lauszus F, Rasmussen O, Henriksen J, et al: Effect of a high monounsaturated fatty acid diet on blood pressure and glucose metabolism in women with gestational diabetes mellitus. Eur J Clin Nutr 2001;55:436–443. 97 DAFNE Study Group: Training in flexible, intensive insulin management to enable dietary freedom in people with type 1 diabetes: Dose adjustment for normal eating (DAFNE) randomized controlled trial. Diabet Med 2003;20(suppl 3):4–5. 98 Pitkin RM: Energy in pregnancy. Am J Clin Nutr 1999;69:583. 99 Food and Nutrition Board: Weight gain; in Nutrition during Pregnancy. Washington, National Academy of Sciences, 1990. 100 Bianco A, Smilen S, Davis Y, Lopez S, Lapinski R, Lockwood C: Pregnancy outcome and weight gain recommendations for the morbidly obese woman. Obstet Gynecol 1998;91:97–102. 101 American College of Obstetrician and Gynaecologist: Nutrition during pregnancy. ACOG Technical Bulletin, 1993, pp 1–7. 102 Feig D, Naylor CD: Eating for two: Are guidelines for weight gain during pregnancy too liberal? Lancet 1998;351:1054–1055.

Nutritional Management of Diabetes in Pregnancy

193

103 Pettitt D, Aleck K, Baird H, Carraher M, Bennett B, Knowler W: Congenital susceptibility to NIDDM: Role of intrauterine environment. Diabetes 1988;37:622–628. 104 Barker DJP, Hales CN, Fall CHD, Osmond C, Phipps K, Clark PMS: Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): Relation to reduced fetal growth. Diabetologia 1993;36: 62–67. 105 Barker DJP, Bull AR, Osmond C, Simmonds SJ: Fetal and placental size and risk of hypertension in adult life. Br Med J 1990;303:256–259. 106 Buchanan TA, Metzger BE, Freinkel N: Accelerated starvation in late pregnancy: A comparison between obese normal pregnant women and women with gestational diabetes. Am J Obstet Gynecol 1990;162:1015–1020. 107 Catalano PM, Huston L, Amini SB, Kalhan SC: Longitudinal changes in glucose metabolism during pregnancy in obese women with normal glucose tolerance and gestational diabetes mellitus. Am J Obstet Gynecol 1999;180:903–916. 108 Knopp RH, Magee MS, Raisys V, Benedetti T: Metabolic effects of hypocaloric diets in management of gestational diabetes. Diabetes 1991;40(suppl 2):165–171. 109 Wolever TM, Bentum-Williams A, Jenkins DJ: Physiological modulation of plasma free fatty acid concentrations by diet. Metabolic implications in nondiabetic subjects. Diabetes Care 1995;18:962–970. 110 Dornhorst A, Nicholls JSD, Probst F, et al: Calorie restriction for the treatment of gestational diabetes. Diabetes 1991;40(suppl 2):161–164. 111 Artal R: Exercise: The alternative therapeutic intervention for gestational diabetes. Clin Obstet Gynecol 2003;46:479–487. 112 Brankston G, Mitchell B, Ryan E, Okun N: Resistance exercise decreases the need for insulin in overweight women with gestational diabetes mellitus. Am J Obstet Gynecol 2004;190:188–193.

Anne Dornhorst, DM, FRCP, FRCPath Metabolic Medicine, Senior Lecturer Imperial College School of Medicine Hammersmith Hospital London W12 ONN (UK) Tel. ⫹44 208 383 3380, Fax ⫹44 208 383 3142, E-Mail [email protected]

Dornhorst/Frost

194

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 195–205

Diabetes-Related Antibodies and Pregnancy Rosa Corcoy, Dídac Mauricio, Alberto de Leiva Department of Endocrinology and Nutrition and Unit of Endocrinology and Nutrition, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain

The interest in diabetes-related antibodies (DRA) in pregnancy follows several lines. First, gestational diabetes mellitus (GDM) is the most frequent metabolic problem during pregnancy and type 1 diabetes mellitus (DM) is not uncommon in the fertile age range. Thus, several groups have studied the prevalence and prognostic importance of DRA to identify women with type 1 DM at a preclinical stage. Second, among the complex mechanisms inducing immune tolerance in pregnancy there is relative suppression of cell-mediated immunity with a relative decrease of Th1 in relation to Th2 cells [1], so that an influence of pregnancy on DRA can be contemplated. Third, several autoimmune diseases such as thyroid autoimmunity [2], systemic lupus erythematosus and antiphospholipid syndrome [3] have been associated with pregnancy loss so that some authors have analyzed the relationship of DRA with pregnancy outcome. Finally, DRA are transplacentally transferred potentially affecting the risk of type 1 DM in the offspring.

Prevalence of DRA in Pregnant Women

Gestational Diabetes mellitus (table 1) Autoantibodies against pancreatic ␤-cell antigens precede the clinical onset of type 1 DM, which is the end-point of a chronic immune-mediated destruction of insulin-producing ␤-cells [4]. As GDM is a risk factor of later DM [5] and autoantibodies against pancreatic ␤-cells are useful to characterize DM and predict type 1 DM in relatives of affected patients [6], there has been a long interest in the prevalence and implications of DRA in women with GDM.

Table 1. DRA in women with GDM Authors

Number of subjects

ICA prevalence %

Steel et al., 1980 [7] Ginsberg-Fellner et al., 1980 [8]a Fallucca et al., 1985 [9] Freinkel et al., 1985 [10]b Stowers et al., 1985 [11]c Catalano et al., 1990 [12]c Bell et al., 1990 [13] Stangenberg et al., 1990 [14] Mauricio et al., 1992 [15] Ziegler et al., 1993 [16] Damm et al., 1994 [17]d Tuomilehto et al., 1994 [31] Beischer et al., 1995 [32]c Mauricio et al., 1996 [29] Petersen et al., 1996 [33]d Lapolla et al., 1996 [18] Dozio et al., 1997 [19]d Fuchtenbusch et al., 1997 [20]b Fallucca et al., 1997 [34]d Whittingham et al., 1997 [21] Panczel et al., 1999 [22]c Kinalski et al., 1999 [23]b,c Mitchell et al., 2000 [35]b Bartha et al., 2001 [24] Kousta et al., 2001 [36]c Weng et al., 2002 [37]d,e Balaji et al., 2002 [38]b Lapolla et al., 2002 [25] Albareda et al., 2003 [26] Bo et al., 2003 [27]b

50 88 39 160 72 187 181 55 307 55 139 112 734 203 139 68 145 437 83 98 68 156 100 102 321 66 86 70 535 123

10 35 5 7.5 12.5 1.6 2.8 1.8 12.4 11 2.9

IAA prevalence %

GADA prevalence %

IA2A prevalence %

0 5.0 1.8 1 2.2

2.9 10 8.5

1.5 3.0

3 14.7 5.1 0.98

0 9.5 3.6 4

0 6.2 1

7.0 6 10.8 4.0 4.5

3.2

1.4 1.5 4.1

0 0.2

41 2.8 14 6.5

For groups with several papers on the subject, the first paper reporting the prevalence of each autoantibody is included. a The method was later shown to produce false-positive results. b p⬍0.05 vs. the control population. c Measurements were performed at different times after delivery. d Nonsignificant vs. the control population. e Women had both GDM and a positive family history for DM.

Corcoy/Mauricio/de Leiva

196

The prevalence of islet cell antibodies (ICA) has been reported to range between 1.0 and 35% [7–27], the highest figure being attributed to a method yielding false-positive results [28], so that the real prevalence would range between 1.0 and 14.7%. Only some authors have studied a control group, with ICA being reported to be higher in women with GDM [10, 20, 23, 27] except in studies with low statistical power [17, 19]. A characteristic of women with GDM and positivity for ICA is that titers are lower than in relatives of type 1 DM subjects or patients with new-onset DM [15, 17, 19, 21, 27]. The prevalence of insulin autoantibodies (IAA) in women with GDM has been addressed in a few papers and the reported prevalences range between 0 and 3% [17–19, 29] and, when positive, titers are also lower than in patients with type 1 DM or first-degree relatives [19, 29]. These low rates agree with the low prevalence of IAA in adults with type 1 DM. A different issue is the prevalence of insulin antibodies (IA) against exogenous insulin in insulin-treated mothers with GDM, which has been reported to be present in up to 44% of these women treated with insulin for a mean time of 9 weeks [30]. However, these antibodies could have important implications if they had a role in the pathogenesis of type 1 DM in the offspring. In the last 10 years, glutamic acid decarboxylase autoantibodies (GADA) are the DRA preferably measured in women with GDM [19–21, 23–27, 31–38]. The reason is the method, which is simpler than that for ICA and with a high rate of predicting the strong association of GADA with the development of type 1 DM. As in the case of ICA, their frequency has been reported to be higher than in the control population in some [20, 23, 27, 35] but not all papers [19, 33, 34, 37], probably due to the low statistical power of the latter. In most series measuring both ICA and GAD the prevalence of ICA is either higher or similar to that of GAD [20, 21, 23, 25–27] but there are exceptions where the rate for GADA [24] or GADA/IA2 [38] is strikingly higher than that for ICA. The GADA of patients with GDM recognize a reduced number of GAD epitopes compared to those of relatives of type 1 DM patients [39], a characteristic similar to that of patients with type 2 DM/LADA [40]. The prevalence of IA2 antibodies (IA2A) has been analyzed in five studies and ranges from 0 to 6.2% [19–21, 23, 25, 26]. This low prevalence is not surprising since these antibodies are not frequent in this age range [41] and are associated with rapid ␤-cell destruction [42], which is not usually the case in women with GDM. Titers are lower than those displayed by children after diagnosis [21]. General and High-Risk Populations Data on the prevalence of DRA in control pregnant women or in cord blood of their newborns come from either control groups of studies of diabetic

Diabetes-Related Antibodies and Pregnancy

197

pregnancies or studies in populations with a high incidence of type 1 DM, where there is specific interest to gain insight into the first stages of the disease. Ivarsson et al. [43] reported a prevalence of 0.8% for ICA and 0.8% for GADA in 123 pregnant control mothers. In cord blood, Tingle et al. [44] reported a prevalence of 0% ICA in 111 control subjects while a rate of 2.6% was reported in 2,249 newborns from Sardinia, a hot spot for type 1 DM [45]. In a control population for children developing type 1 DM Lindberg et al. [46] described the prevalence of all four DRA in cord blood as follows: 0.6% ICA, 2% GADA, 1% IA2A, and 0.3% IAA. Similar information is available from Finland, where 1,002 consecutive newborns were studied. Mothers with type 1 DM were specifically excluded, but information about type 1 DM among fathers was not available. ICA prevalence was 2.7%, GADA 0.6%, IA2A 0.3% and IAA 0.1% with an overall prevalence of ⱖ 1 antibody positivity of 2.9%, 0.5% rate of positivity for two antibodies and 0.1% for four antibodies [47]. In pregnant women with type 1 DM, Tingle et al. [44] reported a 28% rate of ICA positivity and Ziegler et al. [16] a 20% rate in the mothers and 21% in their offspring. Hämäläinen et al. [48] reported the prevalence of DRA to be 40% for ICA, 55% for GADA and 54% for IA2A in early pregnancy of women with type 1 DM, with the corresponding figures in cord blood being 33% for ICA, 50% for GADA and 51% for IA2A. In a recent paper, Koczwara et al. [49] reported a prevalence of 56% for GADA and 37% for IA2A in cord blood of 720 offspring of type 1 DM mothers.

Influence of Pregnancy on DRA

Maternal immune responses during pregnancy are biased towards antibody-mediated and away from cell-mediated. However, there is controversy in antibody-mediated diseases where autoantibody levels have been reported to decrease [50] or increase [51] depending on the autoimmune disorder. Longitudinal studies on DRA during pregnancy are scarce but agree on the fact that the influence of pregnancy is minor with a nonsignificant decrease in the third trimester for ICA, GADA and IA2 [48, 52]. However, seroconversion to negativity for GAD/IA2 has been described occasionally [53]. As to IA, which reflect autoimmunity in the insulin-naive patient but immunity against an exogenous protein in the insulin-treated patient, a decrease at delivery plus a rebound postpartum were reported in patients treated with insulin of animal origin [54, 55] but minor changes in a recent paper where diabetic pregnant women were probably treated with human insulin [52].

Corcoy/Mauricio/de Leiva

198

Table 2. Influence of DRA in pregnancy on morbidity Maternal Perinatal In women with GDM, higher rate of insulin treatment during pregnancy Increased risk of gestosis After delivery In women with GDM, increased risk of abnormal glucose tolerance/DM/type 1 DM at short and long term after delivery Offspring All DRA are transplacentally transferred and undergo postnatal elimination in most infants IA Conflicting results on birth weight Increased risk of hypoglycemia, hypocalcemia, respiratory distress and high cord hematocrit in some studies In a recent paper, maternally transmited GADA and IA2A but not IA have been associated at follow-up with a lower risk of islet cell autoimmunity and type 1 DM in infants of type 1 DM mothers

Influence of DRA on Mother and Fetus (table 2)

Transplacental Passage of DRA After the demonstration of insulin-binding activity in cord blood of infants of diabetic mothers [55], the description of ICA activity in the cord blood of an infant of a diabetic mother [56] and later reports including GADA and IA2A [43, 48], current information indicates that DRA in infants of diabetic mothers have been acquired transplacentally in most cases. First, maternal and fetal levels are highly correlated. This has been described for IA [53, 57–59], ICA [53], GADA [46, 53] and IA2A [53] although for some authors this was not the case for ICA [16, 46] and IA2 [46]. Second, DRA are present in the cord serum of infants of diabetic mothers but not of diabetic fathers [60]. Third, the mechanism of transfer is known since DRA are mainly IgGs [61] and these are actively transferred by the placenta [62]. Finally, antibodies detected in cord blood, usually disappear at follow-up with inverse seroconversion occurring occasionally [13, 47, 53]. Hämäläinen et al. [53] studied the postnatal elimination of transplacentally acquired DRA in 47 infants born to families with type 1 DM and positive for at least one antibody: 47% of these infants were positive for ICA, 68% for IA, 43% for GADA and 36% for IA2A. There was a progressive decline in positivity which at 12 months was 0% for ICA, IA and IA2A and 5.9% for GADA. The mean elimination time was 3.1 months for ICA and IA, 4.5 months for GADA and 4.3 months for IA2A, the pattern corresponding to that expected for IgGs [63].

Diabetes-Related Antibodies and Pregnancy

199

DRA and Perinatal Outcome Bauman and Yalow [64] described in 1981 the transplacental passage of insulin complexed to IA, demonstrating that an antigen otherwise restricted from transplacental passage can be transmitted to the fetus and in 1990 Menon et al. [57] published that the concentration of animal insulin in cord serum correlated with birth weight. However, such an association was found neither in a previous [58] nor in later papers [59, 65, 66] with some exceptions [23]. In women with GDM, the presence of DRA has been related to more frequent insulin therapy [20, 21, 27]. IA have been reported to be associated with toxemia [67] and DRA with HELLP syndrome [68]. Maternal IA have been associated with hypoglycemia [67, 69], hypocalcemia [67], respiratory distress [67] and high cord hematocrit levels [69] in the newborn. Other authors describe a nonsignificant association with morbidity [58] or the lack of differences in age at delivery, the rate of cesarean sections, birth weight distribution and Apgar scores according to DRA [27]. For morbidities related with fetal hyperinsulinism, IA would act facilitating the transfer of maternal insulin [57]. Other authors have suggested that IA could also buffer fetal insulin and thus stress the fetal pancreas to a greater degree than that attributable to hyperglycemia alone [70] but other authors disagree on the basis of a lack of a relationship between the maternal level of IA and C-peptide [58] or insulin and 32–33 split proinsulin [66] in cord blood. Other morbidities like toxemia could be related to damage induced by insulinIA complexes [71, 72]. DRA and Risk of Type 1 DM in the Fetus In the nonobese diabetic mouse, the transplacental passage of maternal autoantibodies increased the risk of autoimmune DM in the offspring [73] whereas the transmission of antibodies against autorreactive T cells is protective [74] and maternal immunity to insulin did not affect DM risk in the progeny [75]. Remarkably, in infants of diabetic mothers, the presence of GADA and IA2A but not of IA in cord blood samples was associated with a lower risk of developing multiple islet autoantibodies and type 1 DM [49]. This was most striking in offspring without the high-risk HLA DRB1*03/DRB1*04-DQB1*0302 genotype and was negligible in those with this genotype. Remarkably, DRA have been described in cord blood of children who developed type 1 DM before 15 years of age [46] and these DRA have been considered a potential risk factor for future DM [76]. This seems to imply that DRA in cord blood are only protective for later DM when maternally transferred. If this mechanism was operating in diabetic pregnancy, it could be partly responsible of the reduced risk of DM [77] and islet autoimmunity [78] in infants of type 1 diabetic mothers versus diabetic fathers

Corcoy/Mauricio/de Leiva

200

that has also been attributed to a higher transmission of risk genotypes when the affected progenitor is the father [79]. DRA and Risk of Maternal Abnormal Glucose Tolerance Most studies agree that DRA confer an increased risk of abnormal glucose tolerance/DM/type 1 DM to the mother. This has been reported for both ICA [7, 8, 15, 17], GADA [20, 33] and IA2A [20]. Positivity for multiple antibodies increases the risk for type 1 DM 2 years after delivery from 2.2% for 0 antibodies, 17% for one, 61% for two and 84% for three [20]. The association has also been reported by authors who have not measured DRA during pregnancy but at follow-up [22, 32]. However, some authors have not related the presence of autoantibodies to glucose tolerance at follow-up [25, 26]. Our group reported an increased risk of abnormal glucose tolerance at 1 year follow-up in women with GDM and ICA positiviy [15] as well as a decrease in the acute response of insulin to intravenous glucose tolerance test in ICA-positive GDM women with normal glucose tolerance at follow-up [80]. However, we did not confirm this finding at long-term follow-up [26]. Overall, GADA seems to be the best predictor of type 1 DM in women with GDM in terms of sensitivity, specificity and positive predictive value [17, 20, 33], which would be in agreement with GADA being the single best predictor for type 1 DM in the 20- to 40-year age range [31, 81].

References 1 2 3 4 5 6

7 8 9

10

Veenstra van Nieuwenhoven AL, Heineman MJ, Faas MM: The immunology of successful pregnancy. Hum Reprod Update 2003;9:347–357. Prummel MF, Wiersinga WM: Thyroid autoimmunity and miscarriage. Eur J Endocrinol 2004;150:751–755. Faussett MB, Branch DW: Autoimmunity and pregnancy loss. Semin Reprod Med 2000;18:379–392. Eisenbarth GS: Type I diabetes mellitus: A chronic autoimmune disease. N Engl J Med 1986;314: 1360–1368. Hagbard L, Svanborg A: Prognosis of diabetes mellitus with onset during pregnancy. A clinical study of seventy-one cases. Diabetes 1960;9:296–302. Bingley PJ, Christie MR, Bonifacio E, Bonfanti R, Shattock M, Fonte MT, et al: Combined analysis of autoantibodies improves prediction of IDDM in islet cell antibody-positive relatives. Diabetes 1994;43:1304–1310. Steel JM, Irvine WJ, Clarke BF: The significance of pancreatic islet cell antibody and abnormal glucose tolerance during pregnancy. J Clin Lab Immunol 1980;4/2:83–85. Ginsberg-Fellner F, Mark EM, Nechemias C, Hausknecht RU, Rubinstein P, Dobersen MJ, Notkins AL: Islet cell antibodies in gestational diabetics. Lancet 1980;ii:362–363. Fallucca F, Di Mario U, Gargiulo P, Iavicoli M, Galfo C, Contreas G, et al: Humoral immunity in diabetic pregnancy: Interrelationships with maternal/neonatal complications and maternal metabolic control. Diabete Metab 1985;11:387–395. Freinkel N, Metzger BE, Phelps RL, Dooley SL, Ogata ES, Radvany RM, Belton A: Gestational diabetes mellitus. Heterogeneity of maternal age, weight, insulin secretion, HLA antigens, and

Diabetes-Related Antibodies and Pregnancy

201

11 12 13 14

15

16

17

18 19 20

21

22 23

24

25

26 27

28 29

30

islet cell antibodies and the impact of maternal metabolism on pancreatic B-cell and somatic development in the offspring. Diabetes 1985;34(suppl 2):1–7. Stowers JM, Sutherland HW, Kerridge DF: Long-range implications for the mother. The Aberdeen experience. Diabetes 1985;34(suppl 2):106–110. Catalano PM, Tyzbir ED, Sims EA: Incidence and significance of islet cell antibodies in women with previous gestational diabetes. Diabetes Care 1990;13:478–482. Bell DS, Barger BO, Go RC, Goldenberg RL, Perkins LL, Vanichanan CJ, et al: Risk factors for gestational diabetes in black population. Diabetes Care 1990;13:1196–1201. Stangenberg M, Agarwal N, Rahman F, Sheth K, al Sedeiry S, De Vol E: Frequency of HLA genes and islet cell antibodies (ICA) and result of postpartum oral glucose tolerance tests (OGTT) in Saudi Arabian women with abnormal OGTT during pregnancy. Diabetes Res 1990;14/1:9–13. Mauricio D, Corcoy R, Codina M, Balsells M, Puig-Domingo M, Pou JM, de Leiva A: Islet cell antibodies identify a subset of gestational diabetic women with higher risk of developing diabetes mellitus shortly after pregnancy. Diabetes Nutr Metab 1992;5:1–5. Ziegler AG, Hillebrand B, Rabl W, Mayrhofer M, Hummel M, Mollenhauer U, et al: On the appearance of islet associated autoimmunity in offspring of diabetic mothers: A prospective study from birth. Diabetologia 1993;36:402–408. Damm P, Kuhl C, Buschard K, Jakobsen BK, Svejgaard A, Sodoyez-Goffaux F, et al: Prevalence and predictive value of islet cell antibodies and insulin autoantibodies in women with gestational diabetes. Diabet Med 1994;11:558–563. Lapolla A, Betterle C, Sanzari M, Zanchetta R, Pfeifer E, Businaro A, et al: An immunological and genetic study of patients with gestational diabetes mellitus. Acta Diabetol 1996;33/2:139–144. Dozio N, Beretta A, Belloni C, Castiglioni M, Rosa S, Bosi E, Bonifacio E: Low prevalence of islet autoantibodies in patients with gestational diabetes mellitus. Diabetes Care 1997;20:81–83. Fuchtenbusch M, Ferber K, Standl E, Ziegler AG: Prediction of type 1 diabetes postpartum in patients with gestational diabetes mellitus by combined islet cell autoantibody screening: A prospective multicenter study. Diabetes 1997;46:1459–1467. Whittingham S, Byron SL, Tuomilehto J, Zimmet PZ, Myers MA, Vidgren G, et al: Autoantibodies associated with presymptomatic insulin-dependent diabetes mellitus in women. Diabet Med 1997;14:678–685. Panczel P, Kulkey O, Luczay A, Bornemisza B, Illyes G, Halmos T, et al: Detection of antibodies against pancreatic islet cells in clinical practice. Orv Hetil 1999;140:2695–2701. Kinalski M, Kretowski A, Telejko B, Kowalska I, Bingley P, Kinalska I: Prevalence of ICA antibodies, anti-GAD and antylA-2 in women with gestational diabetes treated with diet. Przegl Lek 1999;56:342–346. Bartha JL, Martinez-del-Fresno P, Comino-Delgado R: Postpartum metabolism and autoantibody markers in women with gestational diabetes mellitus diagnosed in early pregnancy. Am J Obstet Gynecol 2001;184:965–970. Lapolla A, Fedele D, Pedini B, Dal Fra MG, Sanzari M, Masin M, et al: Low frequency of autoantibodies to islet cell, glutamic acid decarboxylase and second-islet antigen in patients with gestational diabetes mellitus: A follow-up study. Ann NY Acad Sci 2002;958:263–266. Albareda M, Caballero A, Badell G, Piquer S, Ortiz A, de Leiva A, Corcoy R: Diabetes and abnormal glucose tolerance in women with previous gestational diabetes. Diabetes Care 2003;26:1199–1205. Bo S, Menato G, Pinach S, Signorile A, Bardelli C, Lezo A, et al: Clinical characteristics and outcome of pregnancy in women with gestational hyperglycaemia with and without antibodies to beta-cell antigens. Diabet Med 2003;20/1:64–68. Yagihashi S, Suzuki H, Dobersen MJ, Onodera T, Notkins AL, Ginsberg-Fellner F: Autoantibodies to islet cells: Comparison of methods. Lancet 1982;ii:1218. Mauricio D, Balsells M, Morales J, Corcoy R, Puig-Domingo M, de Leiva A: Islet cell autoimmunity in women with gestational diabetes and risk of progression to insulin-dependent diabetes mellitus. Diabetes Metab Rev 1996;12/4:275–285. Balsells M, Corcoy R, Mauricio D, Morales J, Garcia-Patterson A, Carreras G, et al: Insulin antibody response to a short course of human insulin therapy in women with gestational diabetes. Diabetes Care 1997;20:1172–1175.

Corcoy/Mauricio/de Leiva

202

31

32

33

34 35 36

37

38

39

40

41

42

43

44 45

46

47

48

49 50

Tuomilehto J, Zimmet P, Mackay IR, Koskela P, Vidgren G, Toivanen L, et al: Antibodies to glutamic acid decarboxylase as predictors of insulin-dependent diabetes mellitus before clinical onset of disease. Lancet 1994;343:1383–1385. Beischer NA, Wein P, Sheedy MT, Mackay IR, Rowley MJ, Zimmet P: Prevalence of antibodies to glutamic acid decarboxylase in women who have had gestational diabetes. Am J Obstet Gynecol 1995;173:1563–1569. Petersen JS, Dyrberg T, Damm P, Kuhl C, Molsted-Pedersen L, Buschard K: GAD65 autoantibodies in women with gestational or insulin dependent diabetes mellitus diagnosed during pregnancy. Diabetologia 1996;39:1329–1333. Fallucca F, Tiberti C, Torresi P, Cardellini G, Sciullo E, D’Aliberti T, et al: Autoimmune markers of diabetes in diabetic pregnancy. Ann Ist Super Sanita 1997;33:425–428. Mitchell ML, Hermos RJ, Larson CA, Palomaki GE, Haddow JE: Prevalence of GAD autoantibodies in women with gestational diabetes mellitus. Diabetes Care 2000;23:1705–1706. Kousta E, Lawrence NJ, Anyaoku V, Johnston DG, McCarthy MI: Prevalence and features of pancreatic islet cell autoimmunity in women with gestational diabetes from different ethnic groups. BJOG 2001;108:716–720. Weng J, Ekelund M, Lehto M, Li H, Ekberg G, Frid A, et al: Screening for MODY mutations, GAD antibodies, and type 1 diabetes-associated HLA genotypes in women with gestational diabetes mellitus. Diabetes Care 2002;25/1:68–71. Balaji M, Shtauvere-Brameus A, Balaji V, Seshiah V, Sanjeevi CB: Women diagnosed with gestational diabetes mellitus do not carry antibodies against minor cell antigens. Ann NY Acad Sci 2002;958:281–284. Fuchtenbusch M, Bonifacio E, Lampasona V, Knopff A, Ziegler AG: Immune responses to glutamic acid decarboxylase and insulin in patients with gestational diabetes. Clin Exp Immunol 2004;135:318–321. Hawa MI, Fava D, Medici F, Deng YJ, Notkins AL, De Mattia G, Leslie RD: Antibodies to IA-2 and GAD65 in type 1 and type 2 diabetes: Isotype restriction and polyclonality. Diabetes Care 2000;23/2:228–233. Lohmann T, Sessler J, Verlohren HJ, Schroder S, Rotger J, Dahn K, et al: Distinct genetic and immunological features in patients with onset of IDDM before and after age 40. Diabetes Care 1997;20:524–529. Mayrhofer M, Rabin DU, Messenger L, Standl E, Ziegler AG: Value of ICA512 antibodies for prediction and diagnosis of type 1 diabetes. Exp Clin Endocrinol Diabetes 1996;104/3:228– 234. Ivarsson SA, Ackefors M, Carlsson A, Ekberg G, Falorni A, Kockum I, et al: Glutamate decarboxylase antibodies in non-diabetic pregnancy precedes insulin-dependent diabetes in the mother but not necessarily in the offspring. Autoimmunity 1997;26/4:261–269. Tingle AJ, Lim G, Wright VJ, Dimmick JE, Hunt JA: Transplacental passage of islet cell antibody in infants of diabetic mothers. Pediatr Res 1979;13:1323–1325. Olivieri A, Pinna G, Lai A, Velluzzi F, Pilo A, Atzeni F, et al: The Sardinian autoimmunity study. 4. Thyroid and islet cell autoantibodies in Sardinian pregnant women at delivery: A cross-sectional study. J Endocrinol Invest 2001;24:570–574. Lindberg B, Ivarsson SA, Landin-Olsson M, Sundkvist G, Svanberg L, Lernmark A: Islet autoantibodies in cord blood from children who developed type I (insulin-dependent) diabetes mellitus before 15 years of age. Diabetologia 1999;42/2:181–187. Hämäläinen AM, Ilonen J, Simell O, Savola K, Kulmala P, Kupila A, et al: Prevalence and fate of type 1 diabetes-associated autoantibodies in cord blood samples from newborn infants of nondiabetic mothers. Diabetes Metab Res Rev 2002;18/1:57–63. Hämäläinen AM, Savola K, Kulmala PK, Koskela P, Akerblom HK, Knip M; Finnish TRIGR Study Group: Disease-associated autoantibodies during pregnancy and at birth in families affected by type 1 diabetes. Clin Exp Immunol 2001;126/2:230–235. Koczwara K, Bonifacio E, Ziegler AG: Transmission of maternal islet antibodies and risk of autoimmune diabetes in offspring of mothers with type 1 diabetes. Diabetes 2004;53/1:1–4. D’Armiento M, Salabe H, Vetrano G, Scucchia M, Pachi A: Decrease of thyroid antibodies during pregnancy. J Endocrinol Invest 1980;3:437–438.

Diabetes-Related Antibodies and Pregnancy

203

51 52

53

54 55 56

57

58 59 60

61 62 63 64 65 66

67

68 69 70 71 72 73

74

Levy DL: Fetal-neonatal involvement in maternal autoimmune disease. Obstet Gynecol Surv 1982;37/2:122–127. Novak EJ, Ortqvist E, Nord E, Edwall L, Hampe CS, Bekris L, et al: Stability of disease-associated antibody titers in pregnant women with type 1 diabetes with or without residual beta-cell function. Diabetes Care 2000;23:1019–1021. Hämäläinen AM, Ronkainen MS, Akerblom HK, Knip M: Postnatal elimination of transplacentally acquired disease-associated antibodies in infants born to families with type 1 diabetes. The Finnish TRIGR Study Group. Trial to Reduce IDDM in the Genetically at Risk. J Clin Endocrinol Metab 2000;85:4249–4253. Exon PD, Dixon K, Malins JM: Insulin antibodies in diabetic pregnancy. Lancet 1974;ii:126–128. Spellacy WN, Goetz FC: Insulin antibodies in pregnancy. Lancet 1963;ii:222–224. Gamlen TR, Aynsley-Green A, Irvine WJ, McCallum CJ: Immunological studies in the neonate of a mother with Addison’s disease and diabetes mellitus. Clin Exp Immunol 1977;28/1: 192–195. Menon RK, Cohen RM, Sperling MA, Cutfield WS, Mimouni F, Khoury JC: Transplacental passage of insulin in pregnant women with insulin-dependent diabetes mellitus. Its role in fetal macrosomia. N Engl J Med 1990;323:309–315. Mylvaganam R, Stowers JM, Steel JM, Wallace J, MacHendry JC, Wright AD: Insulin immunogenicity in pregnancy: Maternal and fetal studies. Diabetologia 1983;24/1:19–25. Weiss PA, Kainer F, Purstner P, Zehetleitner G, Huttner U, Haas J: Anti-insulin antibodies and birth weight in pregnancies complicated by diabetes. Early Hum Dev 1998;53/2:145–154. Roll U, Christie MR, Fuchtenbusch M, Payton MA, Hawkes CJ, Ziegler AG: Perinatal autoimmunity in offspring of diabetic parents. The German Multicenter BABY-DIAB study: Detection of humoral immune responses to islet antigens in early childhood. Diabetes 1996;45:967–973. Omar MA, Srikanta S, Eisenbarth GS: Human islet cell antibodies: Immunoglobulin class and subclass distribution defined by monoclonal antibodies. Diabetes Res 1987;4/4:155–157. Pitcher-Wilmott RW, Hindocha P, Wood CB: The placental transfer of IgG subclasses in human pregnancy. Clin Exp Immunol 1980;41:303–308. Berg T: The immunoglobulin development during the first year of life. A longitudinal study. Acta Paediatr Scand 1969;58/3:229–236. Bauman WA, Yalow RS: Transplacental passage of insulin complexed to antibody. Proc Natl Acad Sci USA 1981;78:4588–4590. Schwartz R, Gruppuso PA, Petzold K, Brambilla D, Hiilesmaa V, Teramo KA: Hyperinsulinemia and macrosomia in the fetus of the diabetic mother. Diabetes Care 1994;17:640–648. Lindsay RS, Ziegler AG, Hamilton BA, Calder AA, Johnstone FD, Walker JD; Scottish Multicentre Study of Diabetes in Pregnancy: Type 1 diabetes-related antibodies in the fetal circulation: Prevalence and influence on cord insulin and birth weight in offspring of mothers with type 1 diabetes. J Clin Endocrinol Metab 2004;89:3436–3439. Fallucca F, Di Mario U, Gargiulo P, Iavicoli M, Galfo C, Contreas G, et al: Humoral immunity in diabetic pregnancy: Interrelationships with maternal/neonatal complications and maternal metabolic control. Diabete Metab 1985;11:387–395. Weitgasser R, Spitzer D, Kartnig I, Zajc M, Staudach A, Sandhofer F: Association of HELLP syndrome with autoimmune antibodies and glucose intolerance. Diabetes Care 2000;23:786–790. Murata K, Toyoda N, Sugiyama Y: The effects of insulin antibodies during diabetic pregnancy on newborn infants. Asia Oceania J Obstet Gynaecol 1990;16/2:115–122. Heding LG, Persson B, Stangenberg M: B-cell function in newborn infants of diabetic mothers. Diabetologia 1980;19:427–432. Tamas G Jr, Bekefi D, Gaal O: Insulin antibodies in diabetic pregnancy. Lancet 1975;i:521. Di Mario U, Fallucca F, Gargiulo P, Tiberti C, Scardellato A, Arduini P: Insulin-anti-insulin complexes in diabetic women and their neonates. Diabetologia 1984;27(suppl):83–86. Greeley SA, Katsumata M, Yu L, Eisenbarth GS, Moore DJ, Goodarzi H, et al: Elimination of maternally transmitted autoantibodies prevents diabetes in nonobese diabetic mice. Nat Med 2002;8:399–402. McKeever U, Khandekar S, Newcomb J, Naylor J, Gregory P, Brauer P, et al: Immunization with soluble BDC 2.5 T cell receptor-immunoglobulin chimeric protein:antibody specificity and

Corcoy/Mauricio/de Leiva

204

75 76 77 78 79 80

81

protection of nonobese diabetic mice against adoptive transfer of diabetes by maternal immunization. J Exp Med 1996;184:1755–1768. Koczwara K, Ziegler AG, Bonifacio E: Maternal immunity to insulin does not affect diabetes risk in progeny of nonobese diabetic mice. Clin Exp Immunol 2004;136/1:56–59. Lindberg B, Ivarsson SA, Lernmark A: Islet autoantibodies in cord blood could be a risk factor for future diabetes. Diabetologia 1999;42:1375. Warram JH, Krolewski AS, Gottlieb MS, Kahn CR: Differences in risk of insulin-dependent diabetes in offspring of diabetic mothers and diabetic fathers. N Engl J Med 1984;311/3:149–152 Yu L, Chase HP, Falorni A, Rewers M, Lernmark A, Eisenbarth GS: Sexual dimorphism in transmission of expression of islet autoantibodies to offspring. Diabetologia 1995;38:1353–1357. Vadheim CM, Rotter JI, Maclaren NK, Riley WJ, Anderson CE: Preferential transmission of diabetic alleles within the HLA gene complex. N Engl J Med 1986;315:1314–1318. Mauricio D, Corcoy R, Codina M, Morales J, Balsells M, de Leiva A: Islet cell antibodies and beta-cell function in gestational diabetic women: Comparison to first-degree relatives of type 1 (insulin-dependent) diabetic subjects. Diabet Med 1995;12:1009–1014. Zimmet PZ: The pathogenesis and prevention of diabetes in adults. Genes, autoimmunity, and demography. Diabetes Care 1995;18:1050–1064.

Rosa Corcoy, MD, PhD Department of Endocrinology and Nutrition Hospital de la Santa Creu i Sant Pau ES–08025 Barcelona (Spain) Tel. ⫹34 93 291 90 42, Fax ⫹34 93 291 92 70, E-Mail [email protected]

Diabetes-Related Antibodies and Pregnancy

205

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 206–213

Insulin and Oral Hypoglycemic Agents in Pregnancy Z¤ eljko Metelko, Ivana Pavlic´-Renar Vuk Vrhovac University Clinic for Diabetes and Metabolic Diseases, Zagreb, Croatia

Diabetes complicates 1–20% pregnancies [1]. Most frequently (more than 90%), it is gestational diabetes, the rest is preexisting type 1 and type 2 diabetes. Current classification proposed by the WHO [2] does not distinguish between these two entities. Pregnancy outcome, especially fetal outcome and the incidence of malformation, is related to maternal hyperglycemia [3]. Maintaining normoglycemia throughout pregnancy is thus of crucial importance, as well as planning pregnancy with ideal glucoregulation in women with preexisting diabetes [4]. Current guidelines recommend glycemic regulation in pregnancy only by basic treatment principles (education and self-monitoring, exercise, diet) and, if this is not sufficient, insulin [5, 6]. In nonpregnant diabetics several classes of oral hypoglycemics are used, as well as insulin analogues which have some pharmacodynamic advantages compared to conventional insulin. There is some evidence that some of these preparations might be safe for use in pregnant diabetics [7]. The aim of this chapter is an overview of current data on the use of insulin, insulin analogues and potential use of oral hypoglycemics in pregnancies complicated by diabetes mellitus. Insulin

The most physiological way of correcting hyperglycemia besides basic treatment is insulin supplementation. Due to their size, insulin molecules cannot cross the placenta, but a transplacental transfer of the insulin-insulin antibody complex is possible [8]. As antibody production is least with human insulin it is reasonable to

use it during pregnancy. However, although transplacental crossing of the insulin antibody complex has repeatedly been described, there is no evidence of any relationship between insulin antibodies in maternal serum and fetal outcome [9]. The aim of insulin replacement treatment is to mimic the physiological pattern of insulin secretion. This pattern is very complex, comprising a low secretion rate between meals and fasting and short high postprandial peaks. Hyperglycemia is the most potent stimulus for insulin secretion. Insulin replacement treatment might be considered the most complex replacement treatment in endocrinology since it requires full cooperation of the medical team and the patient who has to be educated and trained to adapt insulin doses to needs constantly monitoring blood glucose values. Best results can be obtained by the use of continuous subcutaneous insulin infusion using insulin pump devices (CSII) [10, 11]. The infusion rate is adapted to the needs: higher for meals and low for basal needs. Multiple insulin injection regimens of regular (short-acting) insulin for boluses and intermediate or long-acting insulin for basal when used properly can achieve results comparable to CSII in most patients. Conventional insulin preparations have in this respect some major pharmacodynamic pitfalls. The action of rapidly acting preparations is slower and longer in comparison with physiological secretion of insulin at mealtime: it takes around 30 min to begin, and lasts up to 6 h, with a maximum effect 2–3 h after subcutaneous application. Preparations for basal use have peaks of action: NPH (neutral protamine Hagedorn) or Lente 5–7 h after subcutaneous application and Ultratard or Ultralente type around 10 h later. There is a risk of these peaks leading to hypoglycemia, especially when they coincide with the action of rapidly acting insulin (fig. 1a, b)

Insulin Analogues

In order to obtain a more useful pharmacodynamic profile, insulin analogues have been developed [12]. Peakless insulins have been developed for basal needs, and others with rapid but short action for boluses. A combination of such profiles mimics the physiological secretion better than a combination of conventional insulins (fig. 1c). Alterations in the sequence of amino acids or the addition of fatty acid side chains change structural proprieties of the molecules and so the pharmacodynamics of such designed preparations. The affinity to insulin receptors and thus metabolic action should not be changed substantially. In addition, the affinity to other cross-reactive receptors (especially IGF) should remain comparable to insulin because of concerns of mitogenic potency. Commercially available short-acting insulin analogues as of today are lispro and aspart. Their pharmacodynamic profile is practically identical, mimicking insulin response to a meal. A third short-acting insulin (glulisin) is in phase III

Hypoglycemic Agents during Pregnancy

207

pl. insulin Basal: 1 NPH dose Boluses: 3 doses of regular insulin

b

Basal: 2 NPH doses Boluses: 3 doses of regular insulin

c

Basal: 2 NPH doses 6 12 Boluses: 3 doses of insulin analogue

pl. insulin

pl. insulin

a

18

24

Hours

Fig. 1. a–c Insulin replacement (approximation). pl. ⫽ Plasma.

clinical trials. Currently there is only one long-acting insulin analogue: glargin. Its action is without peaks with a duration of nearly 24 h, which makes possible the use of one injection per day for basal needs in most patients. Another longacting peakless insulin – detemir – is of shorter duration: around 12 h. It is in phase III clinical trials. In vitro studies showed a correlation of metabolic effects to insulin receptor affinity and mitogenic effects with IGF-1 receptor affinity [13]. Both shortacting insulin analogues have a comparable affinity to these receptors as the human insulin lispro has an insignificantly higher affinity to IGF-1. The longacting analogue – glargin – has a somewhat higher affinity to IGF-1; however, preclinical and clinical studies showed no relevance of that in a clinical setting. An anecdotal report of its use in pregnancy did not reveal any concerns [14]. Of all insulin analogues the short-acting insulin lispro has been in use longest and the number of studies of its use in pregnancy is highest. In comparison with regular human insulin, it provides the same [15–18] or better [19] glycemic control in terms of HbA1c with less hypoglycemia [15, 16, 19]. A multicentric randomized study of 33 pregnancies reported a higher number of ‘biochemical’ hypoglycemias with lispro, but a lower number of severe clinical hypoglycemias [18]. In all cited studies there were no differences in pregnancy termination, fetal or maternal complications or congenital abnormalities in lispro-treated pregnancies compared to those treated with regular human insulin. Some observations suggested an accelerated rate of retinopathy

Metelko/Pavlic´-Renar

208

progression [20]. However, two studies designed and conducted to investigate this issue did not confirm it [21, 22]. Similar results are obtained with the use of the insulin aspart [23]. The use of short-acting analogues in pregnancy is considered safe [6, 24].

Oral Hypoglycemics

Sulfonylureas are the oldest and most widely used oral hypoglycemic agents. Data on older preparations show moderate to significant transplacental crossing of these compounds and slow elimination from the newborn circulation. In most newborns clear signs of islet hyperplasia with variable grades of hypoglycemia were described [review in 7]. Data on transplacental crossing of newer sulfonylureas are somewhat controversial. Most data are on glibenclamide (glyburide). In animal models transplacental crossing is proved and fetal serum and tissue concentrations correlate with maternal serum concentrations [24]. However, in a model of an isolated human cotyledon, which is routinely used for transplacental drug crossing, only an insignificant transfer is revealed, even with high concentrations of drug in maternal blood [25]. This discrepancy might be explained by interspecies variations. By its molecular weight, solubility and dissociation constant, glibenclamide should cross the placenta. However, significant binding to plasma proteins (99.8%) and short half-elimination time (4–9 h) [26] as well as P glycoprotein effect [27] might explain its low crossing. P glycoprotein acts as a pump not allowing transplacental drug crossing. Glibenclamide is a substrate, but also an inhibitor of P glycoprotein, which is important since it might be responsible for drug interactions [27]. There are few data on the effect of new sulfonylurea preparations on the fetus [28]. These drugs are not recommended for use in pregnancy in Europe and North America, so from these countries there are only anecdotal reports. The first bigger prospective study, in which successive addition of metformin, glibenclamide and, only if glycemia was still unregulated, insulin in the treatment of gestational diabetes with poor control on the diet only, was performed in South Africa [29]. A total of 106 pregnancies was followed up: metformin was used in 102 and glibenclamide in 67. There was no more teratogenesis in pregnancies treated with oral hypoglycemic agents. Perinatal mortality was the same in all groups. Similar results in a smaller number of pregnancies and with glibenclamide only as the oral agent were observed in an Israeli study [30]. A more recent randomized trial comparing glibenclamide and insulin in a group of 404 women with gestational diabetes gave evidence for the safety of this approach [31] and provoked discussions on glibenclamide as an alternative in

Hypoglycemic Agents during Pregnancy

209

gestational diabetes in otherwise healthy women [32, 33]. It is important to notice that glibenclamide was used only after the 11th week of pregnancy, i.e. after organogenesis, and that the women included had no concomitant diseases or other complications during pregnancy. Metformin has recently been used as adjuvant treatment in polycystic ovary syndrome [34, 35]. Reports have appeared recently on the use of metformin during pregnancies suggesting a lower incidence of miscarriages, no teratogenesis and normal development in children up to 6 months [36, 37]. Together with earlier reports on the use of metformin in pregnancies [29] and animal experiments which revealed a higher incidence of malformations with fenformin but not metformin [38], this suggests a possible use of metformin in pregnancy. However, there are reports on adverse effects of metformin in pregnancy: a Danish cohort study [39] described a significantly higher incidence of preeclampsia in 50 women treated with metformin in pregnancy compared to 68 treated with sulfonylurea and 42 with insulin. In the same study, perinatal mortality was significantly higher in those treated with metformin in the last trimester. Regarding postnatal use, it seems to be safe in lactating mothers since it appears in milk in insignificant concentrations and does not appear in the circulation of breastfed infants [40]. There are virtually no data on the use of other classes of oral hypoglycemic agents in pregnancy: intestinal glycosidase inhibitors and paroxisome proliferatoractivated ␥-receptor (PPAR-␥) agonists (thiazolidinediones – ‘glitazones’). PPAR-␥ modulates differentiation depending on human trophoblast ligands: in trophoblast culture an enhanced biochemical and morphological differentiation is described with troglitazone [41]. Since there are recent reports on the use of drugs from this class in polycystic ovary syndrome [42] there is concern about the incidental use in early pregnancy although these drugs are contraindicated during pregnancy and lactation. Medline search from 1966 to February 2004 retrieved only one Mexican report on the use of acarbose in pregnancy – only 6 pregnancies are described, all ending in the normal birth of a healthy child [43]. Acarbose makes possible enhanced butyrate production in the intestine. Absorbed butyrate might induce prostaglandin E secretion; however, there are no clinical data of the significance of this effect [44]. Finally, it is worth noticing that a number of drugs taken frequently and continuously by diabetic patients are FDA X rated, i.e. contraindicated in pregnancy. These are hypolipemics from the ‘statin’ group (hydroxylmethylguanidine coenzyme A inhibitors), angiotensin-converting enzyme inhibitors and angiotensin receptor II inhibitors. It is important to stop these drugs during the preconception management of diabetic women, the same as aspirin which is also frequently used in small continuous doses in diabetics.

Metelko/Pavlic´-Renar

210

Conclusion

The target of glucose regulation in pregnancy is normoglycemia. In women with preexisting diabetes regulation of diabetes before conception is essential. Women with gestational diabetes should have immediate glucose regulation with the same target. In most of them it is achievable with basic treatment only. The safest treatment in women in whom basic principles do not achieve this target is insulin replacement: basal bolus with blood glucose monitoring. Shortacting insulin analogues seem to be safe and several ongoing prospective studies are designed to evaluate their effect in comparison with conventional human insulin. So far, there are no data on the use of long-acting insulin analogues in pregnancy. There is not enough evidence to recommend of glibenclamide or metformin in pregnancy. However, there does not seem to be a great risk in accidental use and more evidence is needed for its safe use on certain specific occasions.

References 1 2

3 4 5 6 7 8 9 10

11 12 13

Cheung NW, Byth K: Population health significance of gestational diabetes. Diabetes Care 2003;26:2005–2009. Definition, Diagnosis and Classification of Diabetes mellitus and Its Complications. Report of a WHO Consultation. Part 1. Diagnosis and Classification of Diabetes mellitus. WHO/NCD/NCS/ 99.2. Geneva, WHO, 1999. Starc¤evic´ V, Djelmiš J, Ivaniševic´ M, Mayer D: The effect of glycemia on early embryonic development in diabetic pregnancies. Prenat Neonat Med 2001;6:208. Ray JG, O’Brien TE, Chan WS: Preconception care and the risk of congenital anomalies in the offspring of women with diabetes mellitus: A meta-analysis. QJM 2001;94:435–444. American Diabetes Association: Gestational diabetes mellitus. Diabetes Care 2004;27(suppl 1): S88–S90. American Diabetes Association: Preconception care of women with diabetes. Diabetes Care 2004;27(suppl 1):S76–S78. Slocum JM, Soso MEB: Use of antidiabetes agents in pregnancy: Current practice and controversy. J Perinat Neonat Nurs 2002;16:40–53. Garcia-Bournissen F, Feig DS, Koren G: Maternal-fetal transport of hypoglycaemic drugs. Clin Pharmacokinet 2003;42:303–313. Weiss PA, Kainer F, Purstner P, Zehetleitner G, Huttner U, Haas J: Anti-insulin antibodies and birth weight in pregnancies complicated by diabetes. Early Hum Dev 1998;53:145–154. Simmons D, Thompson CF, Conroy C, Scott DJ: Use of insulin pumps in pregnancies complicated by type 2 diabetes and gestational diabetes in a multiethnic community. Diabetes Care 2001;24: 2078–2082. Pickup J, Keen H: Continuous subcutaneous insulin infusion at 25 years – Evidence base for the expanding use of insulin pump therapy in type 1 diabetes. Diabetes Care 2002;25:593–598. Heise T, Heinemann L: Rapid and long-acting analogues as an approach to improve insulin therapy: An evidence-based medicine assessment. Curr Pharm Des 2001;7:1303–1325. Kurtzhals P, Schaffer L, Sorensen A, Kristensen CM, Jonassen I, Schmid C, Trub T: Correlation of receptor binding and metabolic and mitogenic potencies of insulin analogues designed for clinical use. Diabetes 2000;49:999–1005.

Hypoglycemic Agents during Pregnancy

211

14 15 16 17

18

19 20 21 22

23 24 25

26 27

28 29 30 31 32 33 34 35 36 37 38

Devlin JT, Hothersalll L, Wilkis JL: Use of insulin glargine during pregnancy in a type 1 diabetic woman. Diabetes Care 2002;25:1095–1096. Jovanovic L, Ilic S, Pettitt DJ, Hugo K, Gutierrez M, Bowsher RR, Bastyr EJ 3rd: Metabolic and immunologic effects of insulin lispro in gestational diabetes. Diabetes Care 1999;22:1422–1427. Scherbaum WA, Lankisch MR, Pawlowski B, Somville T: Insulin lispro in pregnancy – retrospective analysis of 33 cases and matched controls. Exp Clin Endocrinol Diabetes 2002;110:6–9. Masson EA, Patmore JE, Brash PD, Baxter M, Caldwell G, Gallen IW, Price PA, Vice PA, Walker JD, Lindow SW: Pregnancy outcome in type 1 diabetes mellitus treated with insulin lispro (Humalog). Diabet Med 2003;20:46–50. Persson B, Swahn ML, Hjertberg R, Hanson U, Nord E, Nordlander E, Hansson LO: Insulin lispro therapy in pregnancies complicated by type 1 diabetes mellitus. Diabetes Res Clin Pract 2002;58: 115–121. Bhattacharyya A, Brown S, Hughes S, Vice PA: Insulin lispro and regular insulin in pregnancy. QJM 2001;94:255–260. Kitzmiller JL, Main E, Ward B, Theiss T, Peterson DL: Insulin lispro and the development of proliferative diabetic retinopathy during pregnancy. Diabetes Care 1999;22:874–875. Loukovaara S, Immonen I, Teramo KA, Kaaja R: Progression of retinopathy during pregnancy in type 1 diabetic women treated with insulin lispro. Diabetes Care 2003:26:1193–1198. Buchbinder A, Miodovnik M, McElvy S, Rosenn B, Kranias G, Khoury J, Siddiqi TA: Is insulin lispro associated with the development or progression of diabetic retinopathy during pregnancy? Am J Obstet Gynecol 2000;183:1162–1165. Buchbinder A, Miodovnik M, Khoura J, Sibai BM: Is the use of insulin lispro safe in pregnancy? J Matern Fetal Neonatal Med 2002;11:232–237. Sivan E, Feldman B, Dolitzki M, Nevo N, Dekel N, Karasik A: Glyburide crosses the placenta in vivo in pregnant rats. Diabetologia 1995;38:753–756. Elliot BD, Schenker S, Langer O, Johnson R, Prihoda T: Comparative placental transport of oral hypoglycemic agents in humans: A model of human placental drug transport. Am J Obstet Gynecol 1994;171:653–660. Koren G: Glyburide and fetal safety; transplancental pharmacokinetic considerations. Reprod Toxicol 2001;15:227–229. Smit JW, Huisman MT, van Tellingen O, Wiltshire HR, Schinkel AH: Absence or pharmacological blocking of placental P-glycoprotein profoundly increases fetal drug exposure. J Clin Invest 1999;104:1441–1447. Hod M, Shafrir E: Oral hypoglycemic agents as an alternative therapy for gestational diabetes. Isr J Med Sci 1995;31:640–643. Coetzee EJ, Jackson WPU: The management of non-insulin-dependent diabetes during pregnancy. Diabetes Res Clin Pract 1986;1:281–287. Ravina A: Non-insulin dependent diabetes of pregnancy treated with the combination of sulfonylurea and insulin. Isr J Med Sci 1995;31:623–625. Langer O, Conway DL, Berkus MD, Xenakis EMJ, Gonzales O: A comparison of glyburide and insulin in women with gestational diabetes mellitus. N Engl J Med 2000;343:1134–1138. Dornhorst A: A comparison of glyburide and insulin in women with gestational diabetes mellitus. Diabet Med 2001;18(suppl 3):12–14. Glueck CJ, Goldenberg N, Streicher P, Wang P: The contentious nature of gestational diabetes: Diet, insulin glyburide and metformin. Exp Opin Pharmacother 2002;3:1557–1568. Lord JM, Flight IHK, Norman RJ: Metformin in polycystic ovary syndrome: Systematic review and meta-analysis. Br Med J 2003:327:951–958. Harborne L, Fleming R, Lyall H, Norman J, Sattar N: Descriptive review of the evidence for the use of metformin in polycystic ovary syndrome. Lancet 2003;361:1894–1901. Jakubowicz DJ, Iurno MJ, Jakubowitz S, Roberts KA, Nester JE: Effects of metformin on early pregnancy loss in the polycystic ovary syndrome. J Clin Endocrinol Metab 2002;87:524–529. Glueck CJ, Wang P, Goldenberg N, Sieve-Smith L: Pregnancy outcomes among women with polycystic ovary syndrome treated with metformin. Hum Reprod 2002;17:2858–2864. Denno KM, Sadler TW: Effects of the biguanide class of oral hypoglycemic agents on mouse embryogenesis. Teratology 1994;49:260–266.

Metelko/Pavlic´-Renar

212

40 41

42

43 44

Hellmuth E, Damm P, Molsted-Pedersen L: Oral hypoglycaemic agents in 118 diabetic pregnancies. Diabet Med 2000;17:507–511. Hale TW, Kristensen JH, Hackett LP, Kohan R, Ilett KF: Transfer of metformin into human milk. Diabetologia 2002;45:1504–1514. Schaiff WT, Carlson MG, Smith SD, Levy R, Nelson DM, Sadovsky Y: Peroxime proliferatoractivated receptor-gamma modulates differentiation of human trophoblast in a ligand-specific manner. J Clin Endocrinol Metab 2000;85:3874–3881. Glueck CJ, Moreira A, Goldenberg N, Sieve L, Wang P: Pioglitazone and metformin in obese women with polycystic ovary syndrome not optimally responsive to metformin. Hum Reprod 2003;18:1618–1625. Zarate A, Ochoa R, Hernandez M, Basurto L: Effectiveness of acarbose in the control of glucose tolerance worsening in pregnancy (in Spanish). Ginecol Obstet Mex 2000;68:42–45. Kast RE: Acarbose related diarrhea: Increased butyrate upregulates prostaglandin E. Inflamm Res 2002;51:117–118.

ˆ

39

Zeljko Metelko, MD, PhD Professor, Medical Faculty, University of Zagreb Director, University Clinic Vuk Vrhovac, Dugi dol 4a HR–10000 Zagreb (Croatia) Tel. ⫹385 1 233 1408, Fax ⫹385 1 233 1515, E-Mail [email protected]

Hypoglycemic Agents during Pregnancy

213

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 214–221

Continuous Glucose Monitoring for the Evaluation and Improved Treatment Adjustment in Gravid Women with Diabetic Pregnancy Moshe Hod, Yariv Yogev Perinatal Division and WHO Collaborating Center, Department of Obstetrics and Gynecology, Rabin Medical Center, Petah Tiqva, Israel

The goal of management in diabetes is to maintain blood glucose and hemoglobin A1c (HbA1c) levels as near to normal as possible. This goal is difficult to achieve in all patients, and poses an even greater challenge in pregnant women. Gestational diabetes mellitus (GDM) defined as ‘carbohydrate intolerance of variable severity with onset or first recognition during pregnancy’ [1] occurs in nearly 4% of all pregnancies in the United States, but the actual prevalence may differ with ethnicity and maternal age [2]. Women with high blood glucose levels experience a greater risk of adverse maternal and fetal outcomes, including preeclampsia, cesarean delivery, macrosomia, congenital anomalies, and increased risk for future development of type 2 diabetes [3]. The most common and significant neonatal complication clearly associated with GDM is macrosomia, an oversized infant with a birth weight greater than the 90th percentile for gestational age and sex or a birth weight greater than two standard deviations (SD) above the mean of a normal population of neonates [4]. The greatest danger of macrosomia lies in its association with an increased risk of birth injuries and asphyxia. In untreated GDM the risk of macrosomia is as high as 40% of neonates [5]. Affected patients are prescribed a strict diet and exercise program alone or combined with insulin therapy if hyperglycemia persists. However, too-tight glycemic control may be accompanied by symptomatic or asymptomatic hypoglycemic events. According to current practice, morning fasting and postprandial glucose levels are measured by the self-blood glucose monitoring (SBGM) method [6, 7].

The optimal frequency of blood glucose measurements in women with diabetic pregnancy has not been established. This method has an important limitation, however, in that it provides only single values and not a longitudinal daily glycemic profile. As such, both hypoglycemic and hyperglycemic events may go undetected. Because intermittent blood glucose monitoring underestimates the number of hyperglycemic events, a more accurate determination of postprandial glucose levels is necessary to decrease the risk of macrosomia in GDM. Continuous glucose monitoring (CGM) may facilitate the detection of all postprandial peaks, including those due to unscheduled meals and may provide an opportunity for better intervention, by providing the complete glucose profile.

Continuous Glucose Monitoring

In order to improve glycemic monitoring, a new technology of CGM was recently developed [8]. The MiniMed CGM system (Sylmar, Calif., USA) is composed of a disposable subcutaneous glucose-sensing device and an electrode impregnated with glucose oxidase connected by a cable to a lightweight monitor, which is worn over clothing or a belt (fig. 1, 2). The system takes a glucose measurement every 10 s, based on the electrochemical detection of glucose by its reaction with glucose oxidase, and stores an average value every 5 min, for a total of 288 measurements per day. The system measures in subcutaneous tissue interstitial glucose levels within a range of 40–400 mg/dl. The data are stored in the monitor for later downloading and reviewing on a personal computer. The patients are unaware of the results of the sensor measurements during the monitoring period. Glucose values obtained with CGM have been shown to correlate with laboratory measurements of plasma glucose levels [8] and with home glucose meter values [9]. CGM may facilitate the detection of all postprandial peaks, including those due to unscheduled meals and hypoglycemic events (fig. 3, 4).

CGM for Clinical Use Not in Pregnancy

The reports using devices that measure interstitial fluid have shown that frequently monitored glucose levels can be used to adjust diabetes therapy and improve glucose control. Bode et al. [10] has reported in a pilot study that HbA1c levels improved using the feedback of the continuously monitored interstitial fluid glucose levels calibrated to plasma glucose levels. A total of 9 subjects with type 1 diabetes and HbA1c values greater than 8.5% completed the study. Subjects wore

CGM in Diabetic Gravid Women

215

Fig. 1. CGM mechanism.

Fig. 2. CGM device.

a continuous glucose monitor for two 1-week periods during the study. After each sensor use, changes to diet, insulin dosage and SMBG schedule were made. HbA1c decreased from 9.9% (SD ⫽ 1.1%) at baseline to 8.8% (SD ⫽ 1.0%) 5 weeks after baseline (p ⬍ 0.05), but daily insulin usage was unchanged over the same period of time (p ⫽ 0.4). The glucose sensors performed accurately, with a median correlation of 0.92 and a mean absolute difference of 19.1% (SD ⫽ 9.0%). The continuous glucose profiles allowed identification of glucose patterns and excursions that helped direct changes in therapy.

Hod/Yogev

216

Glucose concentration (mg/dl)

400 350 300 250 200 150 100 50 0 12:00 Midnight

4:00 a.m.

8:00 a.m.

12:00 Noon

4:00 p.m.

8:00 p.m.

12.00 Midnight

Glucose concentration (mg/dl)

Fig. 3. CGM demonstrating nocturnal hypoglycemia not diagnosed by SBGM.

400 350 300 250 200 150 100 50 0 12:00 Midnight

4:00 a.m.

8:00 a.m.

12:00 Noon

4:00 p.m.

8:00 p.m.

12.00 Midnight

Fig. 4. CGM demonstrating hyperglycemia not diagnosed by SBGM.

Recently, Kaufman et al. [11] showed that CGM could serve as a clinical tool for clinical decision making and glycemic control in children with type 1 diabetes. In another work, Hershkovitz et al. [12] demonstrated the clinical implications of CGM use to assess and manage asymptomatic hypoglycemic events in children with glycogen storage disease.

CGM in Pregnancy

Application of CGM in pregnancy may be the means whereby we finally answer the questions as to timing of the postprandial peak, the relationship between glucose and macrosomia, and achieving longitudinal data concerning the daily blood glucose profile.

CGM in Diabetic Gravid Women

217

The first pilot study utilized a CGM in 10 GDM women for a total of 30 days of continuous monitoring and observed that the total minutes/day of previously unknown hyperglycemia was a mean of 82.4 ⫾ 18 (SD) [13]. These events were discovered to be unscheduled meals not detected by conventional self-blood glucose protocols. Furthermore, these elevations of blood glucose often occurred shortly after patients took fasting and postprandial fingerstick glucose determinations that indicated that their glucose levels were in the target ranges. In another work by Yogev et al. [14] a comparison of the daily glycemic profile reflected by continuous and intermittent blood glucose monitoring in pregnant women with type 1 diabetes was evaluated. The study sample consisted of 34 gravid patients at gestational weeks 16–32 with type 1 diabetes, being treated by multiple insulin injections. Data derived by the CGM for 72 h were compared to fingerstick glucose measurements performed 6–8 times a day. An average of 780 ⫾ 54 glucose measurements was recorded for each patient with CGM. The mean total time of hyperglycemia (glucose level ⬎140 mg/dl) undetected by the fingerstick method was 192 ⫾ 28 min/day. Nocturnal hypoglycemic events (glucose level ⬍50 mg/dl) were recorded in 26 patients; in all cases, there was an interval of 1–4 h before clinical manifestations appeared or the event was revealed by random blood glucose examination. Based on the additional information obtained by continuous monitoring, the insulin therapeutic regimen was adjusted in 24 patients (70%). In order to examine the efficacy of a CGM system for treatment adjustment in patients with diabetic pregnancy, Yogev et al. [15] used data derived from the CGM system for 72 h for treatment adjustment. Two to 4 weeks later, the patients were reevaluated with CGM. In the first part of the study, the mean total time of hyperglycemia (glucose level ⬎140 mg/ml) undetected by the fingerstick method was 152 ⫾ 33 min/day. Nocturnal hypoglycemic events (glucose level ⬍50 mg/ml) were recorded in 7 patients. Based on the additional information obtained by continuous monitoring, the insulin regimen was changed in all patients. CGM reevaluation after treatment adjustment showed a reduction in undetected hyperglycemia to 89 ⫾ 17 min/day and in nocturnal hypoglycemic events, which were recorded in only 1 patient.

Detection of Hypoglycemia

Despite years of meticulous study, there is still a paucity of information regarding the optimal level of glycemia in diabetic pregnancy that clinicians should target to safely reduce maternal and perinatal morbidity. Strict metabolic control in this patient population has been associated with an increased risk of maternal hypoglycemia. In our study, CGM in women with diabetic pregnancies

Hod/Yogev

218

confirmed the high occurrence rate of nocturnal hypoglycemia suspected in earlier studies. Rosenn and Miodovnik [16] reported significant hypoglycemia, defined as hypoglycemia requiring assistance from another person, in 71% of gravid patients with type 1 diabetes with a peak incidence in the first trimester. The impact of maternal hypoglycemia on human fetal development and neonatal outcome has not been extensively studied. Although concern about the hazards of hypoglycemia is related primarily to the pregnant mother, the potential effects on the developing fetus need to be considered as well. These findings were not apparent with intermittent blood glucose levels measured by fingerstick capillary glucose concentrations. Data derived from CGM [14] led the evaluating physician to change the insulin regimen by decreasing the nighttime dose of intermediate-acting insulin. In over half of the patients, the hypoglycemic events were subclinical, diagnosed only by CGM, and in one fourth of the patients, more than one hypoglycemic events occurred during the night.

The Potential Use of CGM to Prevent Adverse Fetal Outcome

It seems clear that CGM is a superior tool over SBGM in detecting hypoglycemic events at nighttime, because most of them are asymptomatic. The actual clinical effect of these events is currently unknown and yet to be shown. On the other side of the glycemic control, the most common and significant perinatal complication clearly associated with diabetic pregnancy is macrosomia, which poses an increased risk of birth injuries and asphyxia. The risk of macrosomia rises as maternal glycemia increases. Intensified management of GDM reduces the rate of perinatal complications, normalizes birth weight and has a positive influence on the congenital malformation rate. Most likely, the high rate of macrosomia and perinatal morbidity persists despite intensified treatment protocols owing to the imperfect evaluation of the daily glucose profile, because intermittent blood glucose monitoring underestimates the hyperglycemic events. In addition, the most rigorous monitoring protocols only require postprandial glucose measurements 3 times a day. Many patients indulge in large betweenmeal snacks, and these may be the cause of the hidden hyperglycemia. The lack of a strong correlation between HbA1c and glucose levels by CGM [14] may indicate that plasma blood glucose levels vary significantly day by day, and although continuous monitoring is more informative than sporadic nonlongitudinal glucose monitoring, it cannot adequately describe daily glucose profiles over an 8- to 10-week period. It might be that HbA1c is a better predictor of preprandial than postprandial glucose levels, as more hours are spent in the interprandial and nocturnal periods than in the postprandial phases [17].

CGM in Diabetic Gravid Women

219

Conclusion

Many physicians have had the experience of managing women with GDM who appeared to be in good glycemia control based on their SBGM diaries and HbA1c, but, nonetheless, these women still delivered a macrocosmic infant. Our experience has shown that CGM in GDM women can reveal high postprandial blood glucose levels unrecognized by intermittent blood glucose determinations. CGM shows where and how hyperglycemia that might contribute to neonatal complications is occurring and provides a useful tool to help educate patients in behavior modifications that can improve compliance with the management regimen. CGM profiles allow the physician to identify glucose patterns and to better target insulin treatment. The treatment changes would not have been made on the basis of meter data alone. At this point, we do not recommend that CGM replace SBGM, but we suggest that intermittent application of CGM can be used to fine-tune glycemic control, assess patient compliance, if necessary, and prevent nocturnal hypoglycemic events. As in the first trimester of pregnancy the incidence of hypoglycemia in type 1 diabetes patients is higher and achieving strict glycemic control is more difficult, so that the frequency of CGM should be increased during this period. Later applications vary individually. Importantly, at this time point using CGM does not indicate a clinical difference in outcome from SBGM, and the clinical utility of CGM in improving perinatal outcome remains unknown. A large prospective study on maternal and neonatal outcome is needed to evaluate the clinical implications of this new monitoring technique.

References 1 2 3

4 5 6 7

Jovanovic L: Medical Management of Pregnancy Complicated by Diabetes. Alexandria, American Diabetes Association, 2000. Engelgau M, German R, Herman W, Aubert R, Smith J: The epidemiology of diabetes and pregnancy in the US, 1988. Diabetes Care 1995;18:1029–1033. McCance DR, Pettitt DJ, Hanson RL, Jacobsson LTH, Knowler WC, Bennett PH: Birth weight and non-insulin-dependent diabetes: Thrifty genotype, thrifty phenotype, or surviving small baby genotype? BMJ 1994;398:942–945. Langer O, Mazze R: The relationship between large-for-gestational-age infants and glycemic control in women with gestational diabetes. Am J Obstet Gynecol 1988;159:1478–1483. Persson B, Hanson U: Neonatal morbidities in gestational diabetes mellitus. Diabetes Care 1998;21(suppl 2):B79–B84. Homko CJ, Sivan E, Reece EA: Is self-monitoring of blood glucose necessary in the management of gestational diabetes mellitus? Diabetes Care 1998;S2:B118–B122. Langer O: Is normoglycemia the correct threshold to prevent complications in the pregnant diabetic patient? Diabetes Rev 1995;4:2–10.

Hod/Yogev

220

8 9

10

11

12 13 14 15 16 17

Mastortotoro J, Levy R, Georges LP, White N, Mestman J: Clinical results from a continuous glucose sensor multi-center study. Diabetes 1998;47:A61. Rebrin H, Steil GM, Van Antwep WP, Mastortotoro JJ: Subcutaneous glucose predicts plasma glucose independent of insulin: Implications for continuous monitoring. Am J Physiol 1997;277: E561–E571. Bode BW, Gross TM, Thornton KR, Mastrototaro JJ: Continuous glucose monitoring used to adjust diabetes therapy improves glycosylated hemoglobin: A pilot study. Diabetes Res Clin Pract 1999;46:183–190. Kaufman FR, Gibson LC, Halvorson M, Carpenter S, Fisher LK, Pitukcheewanont P: A pilot study of the continuous glucose monitoring system: Clinical decisions and glycemic control after its use in pediatric type 1 diabetic subjects. Diabetes Care 2001;24:2030–2034. Hershkovitz E, Rachmel A, Ben-Zaken H, Phillip M: Continuous glucose monitoring in children with glycogen storage disease type-1. J Inherit Metab Dis 2001;24:863–869. Jovanovic L: The role of continuous glucose monitoring in gestational diabetes mellitus. Diabetes Technol Ther 2000;2(suppl 1):S67–S71. Yogev Y, Chen R, Ben-Haroush A, Jovanovic L, Phillip M, Hod M: Continuous glucose monitoring for the evaluation of gravid women with type 1 diabetes mellitus. Obstet Gynecol 2003;101:633–638. Yogev Y, Ben-Haroush A, Chen R, Kaplan B, Phillip M, Hod M: Continuous glucose monitoring for treatment adjustment in diabetic pregnancies – a pilot study. Diabet Med 2003;20:558–662. Rosenn BM, Miodovnik M: Glycemic control in the diabetic pregnancy: Is tighter always better? J Matern Fetal Med 2000;9:29–34. Jovanovic-Peterson L, Peterson CM, Reed GF, Metzger BE, Mills JL, Knopp RH, Aarons JH: Maternal postprandial glucose levels and infant birth weight: The Diabetes in Early Pregnancy Study. The National Institute of Child Health and Human Development – Diabetes in Early Pregnancy Study. Am J Obstet Gynecol 1991;164:103–111.

M. Hod, MD Department of Obstetrics and Gynecology, Perinatal Division Rabin Medical Center, Beilinson Campus IL–49100 Petah Tiqva (Israel) Tel. ⫹972 3 9377400, Fax ⫹972 3 6024141, E-Mail [email protected]

CGM in Diabetic Gravid Women

221

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 222–229

Fetal Hypoxia and Its Monitoring in Pregestational Diabetic Pregnancies Kari A. Teramo, Vilho K. Hiilesmaa Department of Obstetrics and Gynecology, University Central Hospital, Helsinki, Finland

Perinatal Mortality

Perinatal mortality in type 1 diabetic pregnancies was 50% or more at the time of the discovery of insulin in 1921 and was still high five decades ago (table 1). Fetal death rate increased gradually from 30 weeks of pregnancy until term before modern electronic fetal monitoring [1]. After 36 weeks of pregnancy the stillbirth rate exceeded that of the neonatal death rate, which was the basis at that time for inducing labor 3–5 weeks before term in these pregnancies [2]. Over the last five decades perinatal mortality has decreased from 20–30% to 2–4% in pregnancies complicated by pregestational diabetes in centers specializing in the care of these pregnancies [3]. The main reason for this decrease was the understanding that poor glycemic control is associated with a high incidence of fetal and neonatal complications and that improvement of glycemic control throughout pregnancy decreases perinatal complications [2, 4]. Advances in electronic fetal monitoring and successful treatment of neonatal complications have also contributed to the improved results. However, unexplained fetal deaths still occur [5] and perinatal mortality is still 3–6 times higher than in the general population [6]. The incidence of type 1 diabetes is increasing worldwide [7], which is also seen in increasing numbers of children born to diabetic mothers. Since perinatal mortality in diabetic pregnancies has not changed during the last 25 years, the absolute number of perinatal deaths is actually increasing. At the Department of Obstetrics and Gynecology, Helsinki University Central Hospital, perinatal mortality was 2.13% between 1988 and 2002 among 1,032 infants of pregestational diabetic mothers (table 1). More than 95% of these mothers had type 1 diabetes.

Table 1. Perinatal mortality rates in IDDM pregnancies at the Department of Obstetrics and Gynecology, University Central Hospital Helsinki in 1951–2002 Time period

Infants

Stillbirths

Neonatal deaths

Perinatal mortality rates Finland1 %

IDDM

1951–1960 1959–1968 1970–1971 1975–1980 1988–1992 1993–1997 1998–2002

162 231 52 279 340 362 330

30 25 3 3 6 5 4

16 23 4 3 3 4 0

n

%

46 48 7 6 9 9 4

28.5 20.8 13.5 2.15 2.65 2.49 1.21

3.21 2.32 2.32 1.25 0.79 0.66 0.58

The data from 1951 to 1971 are from published reports, which explains why the years are not continuous. 1 Annual mean [8].

Of the 22 perinatal deaths 15 were stillbirths. During the last 5-year period (1998–2002) perinatal mortality had decreased to 1.21%. The main causes of perinatal deaths in diabetic pregnancies are congenital malformations, intrauterine asphyxia and prematurity. Roughly 1/3 of perinatal deaths results from fetal malformations. Of the 22 perinatal deaths during 1988–2002 (table 1), only 3 resulted from major malformations. However, there were 8 induced abortions because of fetal malformations during the same time period and at least 6 of the 8 fetuses would have resulted in a perinatal death. Thus, perinatal mortality figures are influenced by early detection of severe malformations and by the policy of interrupting such pregnancies. The definition of perinatal mortality varies between different published reports. Many centers use 28 weeks of gestation as the lower limit. In table 1 perinatal mortality was calculated from 22 completed weeks of gestation onwards, which is also the lower limit for the current definition of childbirth by WHO. Nine of the 22 perinatal deaths during the years 1988–2002 occurred before 28 weeks and 17 before 35 weeks of gestation. One of the reasons for prematurity is early delivery because of severe preeclampsia in pregnancies complicated by type 1 diabetes. Poor glycemic control is associated with an increased risk of preeclampsia in diabetic pregnancies [9, 10]. Improvement of glycemic control during the first half of pregnancy seems to decrease the risk of preeclampsia [10].

Fetal Hypoxia and Its Monitoring

223

Fetal Hypoxia

Clinical signs of fetal distress, abnormal fetal heart rate, low Apgar scores, and cord blood acidosis occur more often in diabetic than in uncomplicated pregnancies [11]. Elevated fetal plasma and amniotic fluid erythropoietin levels and an increased number of nucleated red cells are commonly observed in offspring of diabetic mothers suggesting that maternal diabetes can cause fetal chronic hypoxia [12, 13]. Depletion of iron stores in fetal liver, brain and heart in stillborn fetuses of diabetic mothers gives evidence that chronic intrauterine hypoxia precedes fetal death in diabetic pregnancies [14]. The exact pathogenetic mechanisms of fetal hypoxia in diabetic pregnancies are not fully understood. Placental ‘insufficiency’ has been suggested as a possible cause of fetal distress and hypoxia. This is probably true only in diabetic pregnancies complicated by preeclampsia and hypertension, in which the fetus becomes growth-restricted. Thickening of the basement membrane of chorionic villi has been observed in diabetic pregnancies and this could increase the diffusion distance of oxygen between the mother and the fetus. However, the placenta seems to compensate for these pathological changes by increasing the total area of the chorionic villi [15]. It is obvious that more than one factor can cause fetal hypoxia in the diabetic mother [16]. Maternal factors that can compromise fetal oxygen supply in diabetic pregnancies include hyperglycemia, increased levels of glycosylated hemoglobin and ketoacidosis. Since maternal hyperglycemia results in fetal hyperglycemia, the effects of maternal hyperglycemia on fetal oxygenation will be discussed under Fetal Factors below. Increased HbA1c levels can theoretically decrease the oxygen transport from the mother to the fetus, but does not seem to play an important role even in women with poor glycemic control [16]. Maternal ketoacidosis is a severe, life-threatening complication for both the mother and the fetus. The incidence of ketoacidosis during pregnancy is 1–3% [17]. Pregnancy predisposes to ketoacidosis, because it can occur at lower glucose levels during pregnancy than in nonpregnant diabetics. Vomiting, infection, insulin pump failure, noncompliance and use of sympathomimetic drugs or glucocorticoids can precipitate ketoacidosis in diabetic pregnancies. Fetal mortality is still 10% in maternal ketoacidosis [17]. The fetus is especially at risk during the recovery period from ketoacidosis because of reduced oxygen release from maternal red cells, which can markedly reduce placental oxygen transport to the fetus [16, 18]. Fetal Factors Experimental studies in fetal sheep have shown that both fetal hyperglycemia and fetal hyperinsulinemia can independently cause fetal hypoxemia

Teramo/Hiilesmaa

224

80 70

Number

60 50

1.8%

34.0%

40 30 20 10 0 ⫺5 ⫺4 ⫺3 ⫺2 ⫺1

0

1

2

3

4

5

6

7

8

Birth weight distribution of Finnish standard population (SD-units)

Fig. 1. Distribution of relative birth weights expressed as SD units of 599 consecutive newborn infants born to pregestational diabetic women at the Department of Obstetrics and Gynecology, Helsinki University Central Hospital in 1988–1997 compared with relative birth weights from a large Finnish standard population. 34% of the infants of the diabetic pregnancies were macrosomic (⬎2.0 SD) and 1.8% were growth-restricted (⬍⫺2.0 SD).

by increasing fetal oxygen consumption [19, 20]. Similar relationships have been reported in humans. Poor glycemic control in the third trimester increases the occurrence of abnormal fetal heart rate changes [21] and maternal HbA1c levels during the last month before delivery correlate with fetal plasma erythropoietin levels [22]. Elevated fetal erythropoietin levels, either in plasma or in amniotic fluid, are indicative of chronic fetal hypoxia [23]. It can be concluded that the worse the maternal glycemic control is, the more often the fetus suffers from chronic hypoxia. It has also been shown that amniotic fluid insulin levels correlate with fetal plasma erythropoietin levels independently of the degree of maternal glycemia in diabetic pregnancies [22]. Thus also in humans, both fetal hyperglycemia secondary to elevated maternal glucose levels and fetal hyperinsulinemia can cause fetal hypoxemia. Fetal macrosomia is one of the major complications in diabetic pregnancies. While many of the fetal and neonatal complications in diabetic pregnancies have decreased during the last three decades, fetal overgrowth is still extremely common (fig. 1). Fetal relative birth weight and macrosomia correlate with fetal insulin levels in diabetic pregnancies but not in nondiabetic pregnancies [24]. Fetal macrosomia correlates also with amniotic fluid erythropoietin levels, suggesting that the more macrosomic the fetus is, the greater the risk of chronic fetal hypoxia. This is also supported by observations from national birth registry

Fetal Hypoxia and Its Monitoring

225

data from the Nordic countries that fetal deaths increase with increasing birth weight above 4,000 g [25]. It has been suggested that maternal type 1 diabetes increases fetal growth in almost every case [26]. This can also be seen in the distribution of relative fetal birth weights from our institution (fig. 1): the distribution is normal Gaussian, but clearly shifted to the right towards large relative birth weights when compared with a normal standard population. This could explain why some of the stillbirths occur in diabetic mothers, whose fetus has a normal birth weight. Vaginal delivery of a macrosomic fetus can result in severe trauma both in the infant and the mother. Shoulder dystocia occurs in diabetic pregnancies roughly in 50% of infants with a birth weight over 4,500 g, of whom 1/4 get a brachial plexus injury [27]. In severe shoulder dystocia, in addition to brachial plexus injury, the newborn infant can suffer from hypoxic injury. A cesarean section should be considered when the estimated fetal weight is over 4,500 g.

Fetal Monitoring

Most of the ‘unexplained’ fetal deaths in diabetic pregnancies probably result from chronic fetal hypoxia. Strategies to prevent stillbirths in pregestational diabetic pregnancies include electronic fetal surveillance by cardiotocography and ultrasonic monitoring, and by elective delivery 2–3 weeks before term. Fetal lungs mature at a later gestational stage in diabetic pregnancies than in nondiabetic pregnancies. If delivery is planned before 39 weeks of gestation, an amniocentesis is recommended for an evaluation of fetal lung maturity [3]. At the Department of Obstetrics and Gynecology, Helsinki University Central Hospital, an amniotic fluid sample is usually obtained at 37 weeks of gestation for lecithin/sphingomyelin ratio and phosphatidylglycerol measurements [28]. Since 1996 erythropoietin has also been measured from the same amniotic fluid sample in order to detect possible chronic fetal hypoxia. Results of amniotic fluid erythropoietin measurements can now be obtained within a few hours, e.g. by an immunochemiluminescent method [13]. There is no general agreement on the best method to assess fetal wellbeing in diabetic pregnancies. A twice weekly nonstressed fetal heart rate testing with the backup of a fetal biophysical profile test or a uterine contraction stress test has been suggested to prevent stillbirths [29]. In the study by Kjos et al. [29], a cesarean section for fetal distress was performed in 15.5% (17/110) of mothers with pregestational diabetes and in 4.9% (68/1381) of mothers with gestational diabetes. Several studies have shown that Doppler measurements of umbilical artery flow are not a reliable method to detect

Teramo/Hiilesmaa

226

fetal distress in diabetic pregnancies except in cases with fetal growth restriction. At the Department of Obstetrics and Gynecology, Helsinki University Central Hospital, 198 (85%) out of 234 consecutive pregestational diabetic pregnancies had an amniocentesis for fetal lung maturity and erythropoietin assessment. Amniotic fluid erythropoietin levels obtained within 2 days before delivery were abnormal in 21 (14%) of the 156 pregnancies [13]. This gives an estimate of the frequency of fetal chronic hypoxia in pregestational diabetic pregnancies. The true number of cases with fetal hypoxia is actually higher, since in most of the cases without amniocentesis the indication to deliver by emergency cesarean section was fetal distress before the planned amniocentesis. When fetal lung maturity is observed, induction of labor or an elective cesarean section is performed. It is possible that pregnancy can safely be continued until term in diabetic mothers with normal amniotic fluid erythropoietin levels.

Prevention

There is evidence that admitting pregestational diabetics to a tertiary care center results in a better perinatal outcome than follow-up in small hospitals [30]. The risk of fetal hypoxia in diabetic pregnancies is highest in mothers with poor glycemic control, especially during the last trimester of pregnancy, in mothers with a macrosomic or a growth-restricted fetus, and in mothers who smoke. Even the occurrence of preeclampsia in diabetic pregnancies seems to depend on the degree of maternal glycemic control during the first half of pregnancy [10]. Therefore, the follow-up of these high-risk pregnancies should start earlier and be more frequent than the follow-up of low-risk diabetic pregnancies. Maintaining good glycemic control throughout pregnancy is extremely important in the prevention of fetal hypoxia in diabetic pregnancies. Twice weekly fetal monitoring with nonstressed fetal heart rate testing and ultrasound should probably begin at 32 weeks of gestation and even earlier in diabetics with preeclampsia or poor glycemic control [3]. When delivery is planned before 39 weeks of pregnancy, amniocentesis should be performed to assess fetal lung maturity. Measurement of amniotic fluid erythropoietin levels seems to help detect pregnancies with chronic fetal hypoxia.

References 1 2 3

Pedersen J: The Pregnant Diabetic and Her Newborn, ed 2. Copenhagen, Munksgaard, 1977. Hagbard L: Pregnancy and diabetes mellitus. Acta Obstet Gynecol Scand 1956;35(suppl 1):1–180. Gabbe SG, Graves CR: Management of diabetes mellitus complicating pregnancy. Obstet Gynecol 2003;102:857–868.

Fetal Hypoxia and Its Monitoring

227

4 5

6 7 8 9

10

11 12

13

14 15 16 17 18 19 20 21 22

23 24 25 26 27 28

Karlsson K, Kjellmer I: The outcome of diabetic pregnancies in relation to the mother’s blood sugar levels. Am J Obstet Gynecol 1972;112:213–220. Lauenborg J, Mathiesen E, Ovesen P, Westergaard JG, Ekbom P, Molsted-Pedersen L, Damm P: Audit on stillbirths in women with pregestational type 1 diabetes. Diabetes Care 2003;26: 1385–1389. Diabetes and Pregnancy Group, France: French multicentric survey of outcome of pregnancy in women with pregestational diabetes. Diabetes Care 2003;26:2990–2993. Onkamo P, Väänänen S, Karvonen M, Tuomilehto J: Worldwide increase in incidence of type I diabetes. Diabetologia 1999;42:1395–1403. Statistical Yearbook of Finland 2003. Palteisto K (ed), Jyväskylä 2003, pp 137 and 144. Hanson U, Persson B: Epidemiology of pregnancy-induced hypertension and preeclampsia in type 1 (insulin-dependent) diabetic pregnancies in Sweden. Acta Obstet Gynecol Scand 1998;77: 620–624. Hiilesmaa V, Suhonen L, Teramo K: Glycaemic control is associated with pre-eclampsia but not with pregnancy-induced hypertension in women with type I diabetes mellitus. Diabetologia 2000;43:1534–1537. Mimouni F, Miodovnik M, Siddiqi TA, Khoury J, Tsang RC: Perinatal asphyxia in infants of insulin-dependent diabetic mothers. J Pediatr 1988;113:345–353. Widness JA, Susa JB, Garcia JF, Singer DB, Sehgal P, Oh W, Schwartz R, Schwartz HC: Increased erythropoiesis and elevated erythropoietin in infants born to diabetic mothers and hyperinsulinemic rhesus fetuses. J Clin Invest 1981;67:637–642. Teramo K, Kari MA, Eronen M, Markkanen H, Hiilesmaa V: High amniotic fluid erythropoietin levels are associated with an increased frequency of fetal and neonatal morbidity in type 1 diabetic pregnancies. Diabetologia 2004;47:1695–1703. Petry CD, Eaton MA, Wobken JD, Mills MM, Johnson DE, Georgieff MK: Iron deficiency of liver, heart, and brain in newborn infants of diabetic mothers. J Pediatr 1992;121:109–114. Björk O, Persson B: Villous structure in different parts of the cotyledon in placentas of insulindependent diabetic women. Acta Obstet Gynecol Scand 1984;63:37–43. Madsen H: Fetal oxygenation in diabetic pregnancy. Dan Med Bull 1986;33:64–74. Kamalakannan D, Baskar V, Barton DM, Abdu TAM: Diabetic ketoacidosis in pregnancy. Postgrad Med J 2003;79:454–457. Ditzel J, Standl E: The oxygen transport system of red blood cells during diabetic ketoacidosis and recovery. Diabetologia 1975;11:255–260. Carson BS, Philips AF, Simmons MA, Battaglia FC, Meschia G: Effects of sustained insulin infusion upon glucose uptake and oxygenation of the ovine fetus. Pediatr Res 1980;14:147–152. Philips AF, Widness JA, Garcia JF, Raye JR, Schwartz R: Erythropoietin elevation in the chronically hyperglycemic fetal lamb. Proc Soc Exp Biol Med 1982;170:42–47. Teramo K, Ämmälä P, Ylinen K, Raivio K: Pathologic fetal heart rate associated with poor metabolic control in diabetic pregnancies. Obstet Gynecol 1983;61:559–565. Widness JA, Teramo KA, Clemons GK, Voutilainen P, Stenman UH, McKinlay SM, Schwartz R: Direct relationship of antepartum glucose control and fetal erythropoietin in human type 1 (insulin-dependent) diabetic pregnancy. Diabetologia 1990;33:378–383. Widness JA, Teramo K: Erythropoietin: Significance as an indicator of fetal pathology; in Kurjak A (ed): Textbook of Perinatology. London, Parthenon, 1998, pp 549–558. Schwartz R, Brambilla D, Gruppuso PA, Hiilesmaa V, Petzold K, Teramo K: Hyperinsulinemia and macrosomia in the fetus of the diabetic mother. Diabetes Care 1994;17:640–648. Births in the Nordic Countries. Registration and Outcome of Pregnancy 1979–1983. Reykjavik, NOMESKO, 1987. Bradley RJ, Nicolaides KH, Brudenell JM: Are all infants of diabetic mothers ‘macrosomic’? BMJ 1988;297:1583–1584. Rouse DJ, Owen J, Goldenberg RL, Cliver SP: The effectiveness and costs of elective cesarean delivery for fetal macrosomia diagnosed by ultrasound. JAMA 1996;276:1480–1486. Hallman M, Teramo K: Measurement of the lecithin/sphingomyelin ratio and phosphatidylglycerol in amniotic fluid: An accurate method for the assessment of fetal lung maturity. Br J Obstet Gynaecol 1981;88:806–813.

Teramo/Hiilesmaa

228

29

30

Kjos SL, Leung A, Henry OA, Victor MR, Paul RH, Medearis AL: Antepartum surveillance in diabetic pregnancies: Predictors of fetal distress in labor. Am J Obstet Gynecol 1995;173: 1532–1539. Hadden DR: How to improve prognosis in type 1 diabetic pregnancy. Diabetes Care 1999;22: B104–B108.

Kari A. Teramo, MD Department of Obstetrics and Gynecology University Central Hospital, Haartmanninkatu 2 FI–00290 Helsinki (Finland) Tel. ⫹358 9 4711, Fax ⫹358 9 471 74801, E-Mail [email protected]

Fetal Hypoxia and Its Monitoring

229

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 230–253

Ultrasonic Surveillance of the Diabetic Fetus Marina Ivaniševic´ Department of Obstetrics and Gynecology, State Referral Centre for Diabetes in Pregnancy, Medical School, University of Zagreb, Petrova, Zagreb, Croatia

Because of the high level of development in perinatal medicine today perinatal morbidity and mortality in pregnancies complicated by pregestational/gestational diabetes are comparable with those of healthy pregnant women. However, diabetic pregnancies are still burdened by abnormal fetal development and congenital malformations. The most frequent causes of perinatal mortality in diabetic women are congenital malformations, while the main cause of perinatal morbidity is abnormal fetal growth. With perinatal ultrasound diagnostics congenital malformations can be detected efficiently and early, with serial biometric measurements abnormal fetal growth can be diagnosed in time and biophysical tests and the Doppler method have made active monitoring of a fetal condition possible. Ultrasound examination of pregnant women with diabetes type 1 and 2 has to be available at a tertiary centre where other perinatal care can also be given. A centre for the care of diabetic pregnant women has to be technologically equipped for early discovery of malformations, while the ultrasound diagnosticians should be well trained and experienced.

Ultrasound in Estimating Gestational Age

In women with type 1 diabetes, pregnancy as well as gestational age of the fetus should be determined as soon as possible after the missed menstrual period in order to establish the gestational ultrasound age together with the menstrual age. It is well known that in diabetes disorders of menstruation are mostly due to an irregular cycle.

The determination of the exact gestational age in a diabetic pregnancy makes possible: (1) early detection of malformations due to a well-known entity of early embryonic growth restriction, (2) fetal growth acceleration or restriction can be noticed in time and (3) ending a diabetic pregnancy between the 38th and 40th week due to a certain hypoxia of the baby which appears after the 40th week. In early pregnancy, while determining the frequency of ultrasound examinations, the following should be taken into consideration: the quality of preconception diabetes regulation, the HbA1c level, diabetes complications and the previous reproductive history.

Ultrasound in Preimplantation Period and the First Trimester

Apart from measuring the thickness of the decidua, estimating the size and appearance of the luteal body cyst and Doppler measurement of the flow through uterine arteries, ultrasound in the preimplantation and implantation period does not play a more significant role today; in this period of a diabetic pregnancy there are, nevertheless, very important processes which will direct it to a regular or non-regular course [1]. Diabetic embryopathy is the main cause of neonatal death and/or malformations of children of diabetic mothers [2]. Although numerous efforts have been made, since the earliest stage of pregnancy when the gestational sac and the secondary yolk sac can be seen, to predict with high certainty the occurrence of miscarriage and/or future fetal anomalies in a diabetic pregnancy by ultrasound measurements and Doppler recordings, due to the specific quality of human placentation, it has still not been possible [3]. The yolk sac has been intensively researched up to now in disturbed diabetic pregnancies, since it is the target spot which is destroyed during periods of hyperglycaemia of the mother. Dysfunction or damaging of the cell membrane of the yolk sac leads to the inadequate differentiation of capillary network and blood cells in the visceral layer of the yolk sac wall and the result is the occurrence of embryopathy [4]. The gestational sac has to be examined by endovaginal ultrasound examination in early pregnancy (fig. 1a, b); then examination of the development, size and appearance of the secondary yolk sac (fig. 2) should follow as well as of the vitelline duct, the time of the appearance of embryonic echo (fig. 2), size of the embryo (till the 8th week a maximum longitudinal embryo line is measured and after the 8th week the crown-rump length, maximum longitudinal line and the maximum axial length), viability, including measuring the heart frequency, estimate of the primary quantity of the amniotic fluid and the longest diameter of the amniotic cavity.

Ultrasonography in Type 1 DM Pregnancy

231

a

b Fig. 1a. Gestational sac in early pregnancy (2-dimensional ultrasound). b Gestational sac in the same patient in 3-dimensional ultrasound reconstruction.

Fig. 2. 3-dimensional reconstruction of secondary yolk sac and embryonic echo at 9 weeks of gestation.

The first ultrasound examination in a diabetic pregnancy is carried out with an endovaginal ultrasonic probe between 5.5 and 9 weeks of gestation during which at least the following should be examined: (1) the yolk sac inside the gestational sac of a longest diameter of 8 mm and above, and (2) embryonic echo

Ivaniševic´

232

with a positive heart action within the gestational sac of a longest diameter of 16 mm and above. The yolk sac in non-regulated diabetic pregnancies can be hypertrophic in relation to the embryo or spontaneous regression can occur as well as the death of the embryo [5]. Between week 5.5 and 9, the measurement of the embryonic echo or later the crown-rump length gives one of the most precise estimates of the gestational age (⫾4 to 7 days). Then, the entity known as early embryonic growth restriction, which appears also in healthy pregnant women but is more frequent in diabetes type 1 up to 29%, can be discovered [6, 7]. Early embryonic growth restriction in type 1 diabetes is connected with the statistically significant occurrence of embryopathy [8, 9]. Between the 9th and 10th week the organogenesis stage is replaced by the fetal growth stage [10]; therefore, for diabetic pregnant women it is recommended to have another ultrasound examination during this period, as in less frequently checked diabetic controls, the first signs of fetal hypertrophy can be noticed. After the 9th week of gestation due to the beginning of fetal growth, the difference which occurred in those fetuses with an early embryonic growth restriction during organogenesis can be completely made up, if till then there was no miscarriage. The importance of such early ultrasound diagnostics is also obvious from the fact that only 84% of pregnancies where there was an early embryonic growth restriction have a chance of successful outcome, unlike healthy ones where the possibility of a successful pregnancy outcome is 94%. In pregnant women who had their first ultrasound examination after the 9th week of pregnancy, it is impossible, which was already limited, to diagnose diabetic embryopathy and the risk of fetal macrosomia. The adequate regulation of glycaemia after the 9th week of pregnancy except for a partial influence on fetal growth will not effect a reduction of diabetic embryopathy. From the 10th to 14th week on and later, nuchal fold thickness as an ultrasound marker for chromosomopathy can be measured (fig. 3), for which there are nomograms according to the crown-rump length for genetic fetal processing. Median and 95 centiles of nuchal fold thickness at a crown-rump length of 38 mm are 1.3 and 2.2 mm and these value increase linearly during the gestation to 1.9 and 2.8 mm at a crown-rump length of 84 mm [11]. From the 10th week on and later apart from the crown-rump length, biparietal diameter (BPD) must be measured. It is still recommended to do the examination with the endovaginal ultrasonic probe, so these measurements could be more exact, as well as to follow the development of the central nervous system, extremities (fig. 4), spine and internal organs. Today it is possible to do a 3-dimensional reconstruction of the whole fetal body by the end of the first trimester (fig. 5).

Ultrasonography in Type 1 DM Pregnancy

233

Fig. 3. Nuchal fold or nuchal translucency visualization.

Fig. 4. Femur measurement in the first trimester.

Ivaniševic´

234

Fig. 5. The same fetus in 3-dimensional ultrasound reconstruction in the first trimester of pregnancy.

Ultrasonic Biometry in the Second Trimester

BPD at the level of the thalamic nuclei is the base of routine ultrasound examination of the fetal head and measurements are within acceptable variations in the estimate of the gestational age till the 26th week of pregnancy (⫾7 to 10 days). At the same level it is possible to measure the frontooccipital diameter as well as the head circumference (HC). Transcerebellar Diameter Although a linear increase in the biometric parameters during gestation is confirmed both for BPD and HC, transversal cerebellar diameter and vermis cerebellum width are more stably connected to gestational age, so they are more reliable for the estimate of the gestational age in macrosomic children of diabetic mothers [12]. Abdominal Circumference The fetal liver is an organ which reacts most to the nutrient inflow; therefore, at the level of the liver and umbilical vein which is partly inside it, fetal abdominal circumference (AC) should be measured (fig. 6) and according to BPD and AC values, an estimate of fetal body weight can be obtained. In fetal biometry there are numerous other formulas with which the fetal body weight can be calculated, but there is no formula which is precise enough in estimating fetal weight if it is above 4,000 g [13].

Ultrasonography in Type 1 DM Pregnancy

235

Fig. 6. AC measurement.

Fetal Long Bone Measures Gestational age can be precisely calculated by one of the measures of any long bone: femur, humerus, radius and ulna, and these measures can be used in formulas for calculating the body weight of a fetus.

Disorders in Fetal Growth

Pregnancies complicated by type 1 diabetes have a risk of a significant deviation in fetal growth. Macrosomia affects one fourth of all newborns of diabetic mothers and it is connected to the mother’s hyperglycaemia, which induces fetal hyperglycaemia and hyperinsulinism. In diabetic pregnancies, due to metabolic disorders of the mother, there is an intrauterine retardation of fetal growth. The risk of such perinatal complications is higher if there are vascular complications of the basic disease in the mother, which can complicate pregnancy further, resulting in a failure of the mother’s renal function and hypertension.

Macrosomia

A fetus with an increased growth rate or macrosomia is defined by a birth weight of 4,000 g and above or a birth weight above the 90th centile fetal growth curve for the gestational age and sex. Macrosomia occurs in about 10% of all births. One of the following three causes is responsible for such a condition. The most common cause is the increased intrinsic growth potential, so 50–60% of macrosomia cases involve normal, big fetuses. The condition is characterized by an increased, but proportional, linear and circumferential growth. There is no

Ivaniševic´

236

risk of asphyxia in such fetuses, but only a risk of birth trauma. The abnormal homeostasis of glucose of the mother and the resulting variations of glucose values in fetuses are the cause of 35–40% of macrosomic fetuses. Although such fetuses can have a linear and circumferential growth rate which is a bit increased, they obviously have abnormal fat deposits, particularly in the buccal and paraspinal area [14]. Abnormal fat deposits and macrosomia are the precise indicators of neonatal consequences in diabetic pregnancies: the midarm fat mass and lean mass, midthigh fat mass and lean mass, the abdominal fat mass and the subscapular fat mass. All previously mentioned measurements can also be done by 3-dimensional ultrasound [15]. Reliable findings of macrosomia before the 36th gestational week would be of great importance to avoid a potential risk of birth trauma. Unfortunately, there are no reliable methods which would predict a fetus of a weight above 4,000 g, a weight level above which the birth trauma risk suddenly increases. Important and significant mistakes influence the fetal weight estimates within this range. Additional measurements of chest circumference or biacromial distance could serve as a predictor of shoulder dystocia. However, such a method, except for extreme cases, has a comparatively small significance in predicting dystocia. Wladimiroff et al. [16] found that the BPD/chest circumference ratio has a predictive value of increased fetal growth in fetuses big for their age in 47% cases. The same authors state that this predictive value is the same in diabetic and non-diabetic pregnant women. Cheek-to-cheek diameter by 3-dimensional ultrasound and cheek-to-cheek diameter and BPD ratio, in comparison with normograms, make possible ultrasound diagnostics of abnormal fetal growth. In case of metabolic macrosomia, the cheek-to-cheek diameter is higher due to fat tissue deposits and head growth is not significantly different from a fetus of the same gestational age. Therefore, cheek-to-cheek diameter and BPD ratio are higher than normal values. In cases of constitutional macrosomia, except for the increased cheek-to-cheek diameter for gestation, cheek-to-cheek diameter and BPD ratio will be close to the normal one, since both parameters are proportionally higher. Tamura et al. [17] examined the existence of correlations between BPD, HC and AC percentiles and the precise estimate of birth weight in the third trimester. They found that AC values within 90 percentiles precisely predicted macrosomia in 78% of born macrosomia children and AC above the 90th percentile predicted macrosomia in 74% of cases. Another two measures (BPD and HC) had a significantly lower predictive value. Numerous studies have analyzed the relation between the connection of the mother’s glycaemia, HbA1c, growth factor in fetal blood, insulin level in the blood of the mother, fetus and the amniotic fluid and correlated all these parameters with the increase in the accelerated fetal growth [18–20]. Correlations of biometric measurements with the occurrence of fetal hyperinsulinism were also

Ultrasonography in Type 1 DM Pregnancy

237

followed. In pregnant women with type 1 diabetes a study has been carried out in which ultrasound growth parameters were tested: BPD, AC and a femur length in relation to the insulin level in the amniotic fluid between 28 and 35 weeks of pregnancy. The AC as a biometric measure showed 80% sensitivity in discovering pathological insulin levels (higher than 20 ␮U/ml) in amniotic fluid of mothers suffering from type 1 diabetes [21]. It is just another confirmation that with AC, fetal macrosomia can be detected, but fetuses that have hyperinsulinism cannot be detected in a significant number of cases, unless it is extreme. The problem of detecting fetuses in diabetic pregnancies that, due to hyperinsulinism, have a risk of perinatal morbidity with a noninvasive method such as ultrasound has not been possible so far. Measurement of the increased fetal liver volume by 3-dimensional ultrasound may play a role in identifying fetal growth acceleration in diabetic pregnancies but a multicentric randomized study has to evaluate this hypothesis.

Intrauterine Growth Restriction

Diabetic pregnant women who have vascular complications, nephropathy and/or hypertension more frequently give birth to children of a birth weight under the 10th birth weight centiles. BPD and AC measured by ultrasound showed the highest sensitivity (73 and 97%) and specificity (79 and 60%) in a diagnosis of intrauterine fetal growth restriction. After intrauterine growth restriction has been diagnosed, it is important to establish whether it is of a symmetrical or asymmetrical type, which can be simply calculated from the head/body ratio, to exclude, if it had not already been done in early pregnancy, congenital anomalies and start with the intensive monitoring of fetal growth with the help of a series of biometrical measurements in weekly periods, by a biophysical profile according to Manning [22] and measuring blood flow through the uterine-placental-fetal compartment.

Monitoring Fetal Condition in Diabetic Pregnancy by Colour Doppler and Power Doppler

Several large studies carried out on pregnant diabetics regarding monitoring the placental function and blood flow through the umbilical artery correspond to findings in normal pregnancies, if diabetes was controlled well and there were no vascular complications [23, 24]. The resistance index in blood flow through the umbilical artery (fig. 7) shows a normal fall with the progress of pregnancy. Several authors described the flow through uterine arteries in

Ivaniševic´

238

Fig. 7. Umbilical artery blood flow evaluated by power Doppler.

diabetic pregnancy and resistance value indexes were 0.4 ⫾ 0.06 in the third trimester of pregnancy [25]. In some research, there was no connection between the glucose value and ratio systole/diastole in an umbilical artery during all three trimesters of pregnancy [26]. In an early study of Bracero et al. [27] there was a statistically significant connection between the systole and diastole ratio measured in an umbilical artery and glucose serum value. Although well-designed studies have not found statistically significant correlations between regulated glycaemia and chronic changes in the resistance of blood flow in placenta, acute or extreme changes in glucose concentrations of the mother’s blood still change the placental flow which can be registered by Doppler. Severe hyperglycaemia can cause occasional changes in the blood flow through the placenta, mediated by humoral processes, which can be detected only by simultaneous Doppler examination. Outside such a condition in well-controlled glycaemia, there are no significant increases in the resistance index in the blood flow through the umbilical artery [28]. A connection between the HbA1c and the blood flow value in the umbilical artery (2nd and 3rd trimester) has not been confirmed. A connection of changes in blood flow indexes in fetuses whose mothers had moderate hypoglycaemia [29] has also not been confirmed. Following the blood flow resistance index in the umbilical cord makes it possible to distinguish to a small extent the metabolic effect on the increase in growth rate. It is difficult to discover the pathological flow with this method, since it occurs most frequently in severe hyperglycaemic conditions and significantly more frequently the pathological

Ultrasonography in Type 1 DM Pregnancy

239

flow in the umbilical artery occurs due to the effect of vascular insufficiency and decreased fetal growth rate. Ultrasound Doppler monitoring is a necessary diagnostic examination in diabetic pregnancies with fetal growth restriction which occurs in about 30% of pregnant women with long-term diabetes and diabetic angiopathy. Vascular resistance can occur either due to physiological or structural changes in blood vessels. In a non-diabetic pregnancy, the increased resistance to blood flow through an umbilical artery is explained by a specific structural placental lesion: a decreased number of arterioles in tertiary villi [30] and generally, such a pathological flow is connected to the intrauterine fetal growth retardation. At the birth of a fetus with a such Doppler finding, low arterial pO2 is expected and high lactate values, so there is higher frequency of still births or neonatal deaths. The increased resistance to blood flow through uterine arteries reflects the abnormal invasion of trophoblast into spiral arteries, which increases the risk of pre-eclampsia and intrauterine fetal growth restriction in such pregnant women. In a diabetic pregnancy, placental anatomy and histology do not always correlate with the seriousness of the primary illness or the glucose regulation level in the mother’s blood. Blood vessels of the diabetic mother have been exposed to a disturbed glucose metabolism for years, but placental blood vessels need not show the mirror image of angiopathy [31]. The most frequent findings in placentas of diabetic pregnant women are villi immaturity which, together with the obliterating endarteritis, can be the cause of the abnormal distribution and decreased artery calibre of primary villi. In such cases, there is more frequent intrauterine fetal growth restriction and increased resistance to the blood flow can be measured through the umbilical artery. Many investigators found increased mean values of resistance index to the blood flow through the umbilical artery in diabetic pregnant women with hypertension and proteinuria [23, 26, 32] and pre-eclampsia [33]. There are little data on the values of flow through the uterine artery in pre-eclampsia and diabetes. Kofinas et al. [33] found a significant increase in the systole/diastole ratio in diabetic pregnant women with pre-eclampsia in comparison with healthy women. Salvesen et al. [34] found normal resistance indexes in the same blood vessels in diabetic women with serious nephropathy, but without pre-eclampsia. The authors concluded that pregnancy worsens nephropathy, proteinuria and hypertension, but since pre-eclampsia did not occur, there were no characteristic changes in the placental bed which occur in pre-eclampsia and due to that, blood flow values through uterine arteries and the umbilical artery remained normal. According to Sekizuka et al. [35], it is recommendable to practically classify flow values measured in the uterine artery and umbilical artery in verified intrauterine growth restriction (table 1), in order to spot an endangered fetus in time.

Ivaniševic´

240

Table 1. Scoring of uterine artery and umbilical artery flow in intrauterine growth restriction/pre-eclampsia [35] Blood flow characteristics through umbilical artery BFC 0: normal blood flow BFC 1: increased PI BFC 2: lack of flow at the end of diastole BFC 3: lack of flow through the whole diastole of reverse flow Uterine artery score UAS 0: normal blood flow through both uterine arteries UAS 1: one abnormal parameter in one of the uterine arteries (PI or diastolic notch) UAS 2: two abnormal parameters UAS 3: three abnormal parameters UAS 4: increased PI and diastolic notch in both uterine arteries BFC ⫽ Blood flow characteristic; PI ⫽ pulsatility index; UAS ⫽ uterine artery score.

Except for accepting certain guidelines on the risk assessment of the development of pre-eclampsia and/or intrauterine fetal growth restriction based on the estimate of placental vascular resistance, the application of power Doppler technologies available today should also be taken into consideration in the routine clinical work of tertiary perinatal care. The angiography by a power Doppler in examining the placenta and fetal organs helps in the evaluation of fetal tissue vascularization without the use of aggressive diagnostic methods. The information obtained in this way shows more the tissue vascularization than the existence of vascular resistance in that tissue [36]. The decrease of signal intensity in angiography by a power Doppler is a sign of chronic hypoxia. In the latest paper of Pretorius et al. [37] 3-dimensional ultrasound volume measurements give new information about the pathophysiology of the pregnancy. The development of methods for the estimate of the flow volume through big blood vessels will make possible a completely different attitude to monitoring all risk pregnancies including diabetic ones.

Congenital Malformations

Already in 1964 Molsted-Pedersen et al. [38] and a bit later (in 1971) Kucera [39] published that the incidence of congenital malformations of fetuses in pregnancies complicated by type 1 diabetes increased by 3–4 times [40]. In between 6 and 10% diabetic pregnancies, big congenital anomalies appear. In the Croatian population of diabetic pregnant women, the frequency of malformations has been

Ultrasonography in Type 1 DM Pregnancy

241

less than 3% in the last 5 years (author’s clinical data) due to the excellent preconception regulation of glycaemia and intensive perinatal care. Becerra et al. [41] based on an extensive study found an even stronger connection between type 1 diabetes and the absolute risk of developing fetal malformations related to the central nervous system, cardiovascular system and other bigger defects. An increased risk of structural anomalies has been found for skeletal malformations, anomalies of the face and urogenital and gastrointestinal tract [42]. Based on the increased risk of giving birth to a child with congenital anomalies, there is a high perinatal morbidity and mortality and for this reason, diabetic pregnant women have to be tested prenatally for the initial value of HbA1c and ␣-fetoprotein (AFP) in the serum; the heart of the fetus has to be examined by ultrasound, first at the level of 4-chamber view and later to make a detailed fetal echocardiogram regarding the existence of risk as a result of structural and functional anomalies of the fetal heart. The face and central nervous system should be examined by ultrasound, particularly paying attention to anencephaly, spina bifida and microcephaly; the gastrointestinal and genitourinary tract have to be examined in detail, an ultrasound antenatal diagnosis of caudal regression has to be made and the occurrence of polyhydramnios should be detected [43].

Central Nervous System

Anencephaly Anencephaly occurs in diabetic pregnancies at a rate of 0.57%, which is 3 times higher than in the general population and it is diagnosed by the 15th week of pregnancy. Due to the neural probe defect, badly developed cranial bones and the symmetrical absence of calvarium can be seen during the ultrasound examination. In 40–50% of cases, this malformation is accompanied by polyhydramnios. AFP values are increased in the serum of the mother. Termination of pregnancy is recommended since such a malformation is not viable. Spina bifida In diabetic pregnancies the incidence of a neural tube defect is impressively increased in relation to the general population (19.5 out of 1,000 in comparison with 1–2 out of 1,000) [44]. In a diabetic pregnancy, AFP screening in the mother’s serum with an ultrasound examination of the spine for the characteristic signs of spina bifida is obligatory. 3-dimensional ultrasound examination of the fetal skeleton is recommended (fig. 8). In the case of increased serum AFP values in the mother, but negative ultrasound findings, another step should be taken and AFP

Ivaniševic´

242

Fig. 8. Fetal skeleton at 12 weeks of gestation.

and n-acetylcholinesterase should be determined in amniotic fluid by amniocentesis. In spina bifida, parents have to decide about a possible termination of pregnancy and because of the potential trauma during a vaginal birth, pregnancy is usually completed by Caesarean section. Parents have to be informed that in a further pregnancy spina bifida occurs in 2% of cases and in a third pregnancy after two babies born with spina bifida, the risk of giving again birth to a child with the malformation is between 5.7 and 12% [45].

Face

The examination of the fetal face is an important part of the ultrasound examination in a high-risk pregnancy. Even under optimum conditions, it is difficult to show the complexity of form and roundness of the face with 2-dimensional ultrasound, so many sequential scans have to be taken in order to get a complete picture. A 3-dimensional ultrasound goes a step further in showing a fetal face, making it simpler, more visible and understandable, both for the experts and for the pregnant women. The ultrasonic examination of the head and face can be divided into a general view of forms and proportions and the specific examination of the face

Ultrasonography in Type 1 DM Pregnancy

243

Fig. 9. Fetal profile in 3-dimensional reconstruction (right lower part of the picture).

and intracranial build. The examination includes sequential images of sagittal, parasagittal, coronal and transversal planes. The sagittal image is the most useful for estimating the fetal profile including forehead, nose and lower jaw (fig. 9). Fetal face structures can be examined quite clearly and also the possible presence of genetic or structural anomalies is indicated. The fetal face is shown in a sagittal cross section (profile), serial coronary (frontal image) and serial transversal cross sections. Cleft faces from the cleft lip or palate to the big cleft of the face are best seen in frontal and transversal cross sections. The fetal lip and a hard palate can be shown in great details and enable a prenatal diagnosis of cleft lip or hard palate. The specificity and sensitivity of the ultrasound diagnostics of these conditions has not yet been completely determined. A 3-dimensional sonography can confirm the lip continuity at a higher percentage than the 2-dimensional ultrasound image in pregnancies till the 24th gestational week. In diabetic pregnancies, cleft lip/palate can occur, particularly in combination with anencephaly and spina bifida. Today, ultrasound, both 2-dimensional and especially 3-dimensional, easily detects this malformation. If there is a cushingoid appearance of a child, a cheek-to-cheek diameter can be easily measured for the estimate of fetal macrosomia.

Ivaniševic´

244

Heart Anomalies

Echocardiography of a fetus is necessary and part of the ultrasound examination after the 18th week of a diabetic pregnancy and heart anomalies can be detected even earlier by endovaginal ultrasound examination, somewhere around the 14th week [46]. Diabetic mothers have a 5 times higher risk of giving birth to children with structural heart anomalies or myocardial dysfunction, which represent 50% of all significant congenital malformations connected with diabetes and pregnancy [47]. The most frequent structural heart anomalies are defects of the ventricular septum and blood vessel anomalies. There are two abnormalities in the growth and development of the cardiovascular system and which of them will occur depends on whether the mother’s glycaemia was badly regulated. If hyperglycaemia occurs in the mother and fetus in the first trimester during the organogenesis (21–45 days after conception), big structural heart anomalies will occur. If the mother experiences hyperglycaemia during the third trimester, then there is a risk of a heart function disorder due to hypertrophy of the myocardium. Children can develop congestive heart disease which is manifested as respiratory distress syndrome, tachypnoea or tachycardia. In the study of Weber et al. [48] an enlarged heart was found as well as hypertrophy of the walls of both ventricles and septa regardless of good metabolic control. The disorder in pumping during the systole has to be checked by Doppler. Fetal hypertrophic cardiomyopathy in diabetic pregnant women is explained by hyperinsulinism, as a response to hyperglycaemia of the mother. Ultrasound examination techniques refer to the 4-chamber view during the fetal examination in the middle of the second trimester [49]. The authors of recommendations under the title of ‘routine antenatal diagnostic imaging with ultrasound study’ said that complex heart anomalies were detected in 43% of cases in fetuses whose 4 chambers were viewed in the examination of a randomized obstetric population [50]. In the diabetic population it is recommended to do a standard 4-chamber view together with additional cross sections in order to discover heart anomalies. Shields et al. [47] recommend that further fetal echocardiographic examination in diabetic pregnant women is carried out in all pregnant women whose HbA1c value was increased at first measurement. HbA1c is an indicator of the metabolic control of diabetes in the mother, but the correlation of HbA1c values and the frequency of congenital heart disease of the fetuses is not high [51]. The American literature recommends a strict preconception regulation of diabetes as the safest method to prevent heart disease [49], which is supported by our results, after more than 8 years of strict preconception control of diabetic patients and afterwards of their pregnancies. Heart anomalies can be functional, structural and mixed. The functional anomalies are shown as arrhythmia and they can be connected with fetal

Ultrasonography in Type 1 DM Pregnancy

245

congestive heart disease. B-mode dynamic ultrasound imaging is useful in observing arrhythmic characteristics, examining the harmony and disharmony in atrium and ventricle work, and in detecting signs of congenital heart disease: localized and general cardiomegaly, hepatomegaly, effusion and oedema. However, to find the etiological basis of heart dysrhythmia and choose an adequate therapy detailed M-mode echocardiographic and Doppler imaging is necessary. Structural heart diseases are examined in the longitudinal, transversal and coronal plane, paying special attention to certain chambers and valves. The right atrium is determined by the entrance of the upper and lower hollow veins. The intra-atrial septum can be seen well as well as the secondary intra-atrial orifice. The enlargement of the right atrium is connected with the tricuspid atresia, Ebstein’s anomaly, relative tricuspid insufficiency and initial congestive heart disease. Tricuspid and mitral valves, chorda tendineae and valve movements can be seen well. Endocardial cushion defects are characterized by inadequate development of atrioventricular valves and membranous ventricular septal defect. Due to the frequent occurrence of this heart anomaly in connection with trisomy 21 and trisomy 18, a fetal karyogram has to be done. With ultrasound examination two ventricles of the same size should be shown. The ventricle hypoplasia, mostly on the left side, is detected by the lack or very small size of the ventricle. In imaging the intraventricular septum, lower part (muscular part) defects of the septum can be noticed. The defects of the upper part of the septum (membranous part) are difficult to be viewed with B-mode, but they can be successfully viewed with Doppler technique. The lack of the intraventricular septum is typical for the ventricular syndrome. The origins of the aorta and pulmonary artery have to be shown. Riding aorta can also be shown. Dilatation of the aortic arch indicates the presence of riding aorta in Fallot tetralogy. The aortic arch can be seen well; the origins of blood vessels from the aortic arch also have to be shown. The diagnosis of coarctation of the aorta in utero is possible, but it has not been carried out till now. Although a pulmonary artery can be viewed well, it is practically impossible to see the ductus arteriosus. The aortic and pulmonary valves can also be viewed. The pulmonary valve stenosis and the pulmonary artery stenosis can be viewed in utero. The simplest and the most important imaging of fetal heart anatomy is the simultaneous view of 4 heart chambers. Such imaging includes both atria and both ventricles, intra-arterial and intraventricular septum and atrioventricular valves. The 4-chamber view is obtained by a cross section of the fetal abdomen at the level of the liver (the same level used for measuring the AC); the probe is turned to the fetal head. Such imaging is particularly useful for estimating the chamber size, examining the heart’s interior and for imaging the pericardial effusion. The 4-chamber view must be observed in relation to surrounding structures, stomach, spine, liver and hollow vein. In this

Ivaniševic´

246

way, the intrathoracic heart position is estimated. The heart top should be directed to the same side as the stomach. Disharmonization of directions can indicate the presence of very complex heart anomalies. The sagittal view of a fetal heart is obtained by placing the view plane between the left fetal hip and right shoulder. The rotation of the probe in a clockwise direction enables the view of a pulmonary artery and its way out from the right ventricle. If we continue the probe rotation, we will get a view of the ventricles and the intraventricular septum and frequently papillary muscles of the mitral valve can also be seen. It can be expected that a newborn child has an enlarged heart, hypertrophy of the walls of both ventricles and septum, regardless of good metabolic control. The sensitivity of this test is 33%, the specificity 100%, positive predictive value of the test is 100% and its negative predictive value is 97%. A 4-chamber heart view does not give false-positive abnormal findings. Most ‘wrong’ findings occur in septum defects and anomalies of blood vessel origins [43]. Fetal heart examination in M-mode will show the risk of mild heart hypertrophies and decrease of diastolic dimension of the left ventricle, even in well-regulated pregnancies from the 27th week on. The Fetal Heart Examination by Colour Doppler Additional confirmation of malformations discovered in the fetal heart is expected, since functional stenosis or dynamic obstruction can be discovered by Doppler echocardiography. Fetuses of poorly controlled diabetic mothers have a lower right atrioventricular ratio (early ventricular filling and active atrial filling ratio). This may be due to metabolic acidosis, non-hyperthrophic cardiac dysfunction or fetal polycythemia.

Anomalies of Gastrointestinal System

The most frequent anomalies of the gastrointestinal system in fetuses of diabetic pregnant women are duodenal atresia, colon atresia and anal atresia. The ultrasound finding of anechogenic dilated masses of double bubble images are characteristic for this malformation and can be rarely seen before the 24th week of pregnancy. This image is usually accompanied by polyhydramnios. Before the 22nd week of pregnancy, diagnosis is possible if the activity of disaccharidase, which is deficient in affected fetuses, is determined by amniocentesis from the amniotic fluid [52]. The diagnostic procedure in duodenal atresia is first to exclude aneuploidy by karyotyping. The prognosis of such anomalies depends on the period when it

Ultrasonography in Type 1 DM Pregnancy

247

was discovered, the length of a part of intestine which is not affected, the birth weight of a child and accompanying anomalies.

Anomalies of the Urogenital Tract

Anomalies of the genitourinary tract which have been more frequent in a diabetic pregnancy are a double-channel system and kidney agenesis. A doublechannel system is a malformation which can be connected by surgery. Kidneys can be viewed by ultrasound already at 12 weeks of pregnancy and at that time in more than 50% of fetuses a urinary bladder can be seen. The ultrasound can confirm kidney agenesis at about the 16th week of pregnancy when, apart from the absence of kidney and urinary bladder, the absence of the amniotic fluid will also be visible (oligohydramnios). Kidney agenesis is a lethal congenital anomaly and when a diagnosis is certain, abortion should be recommended.

Other Anomalies

Caudal Regression Syndrome The caudal regression syndrome occurs in 1 out of 200 to 1 out of 500 diabetic pregnancies as a result of the defect of mesoderm development in the 4th postconceptional week. A short spine and seriously malformed and shortened lower extremities can be seen on the ultrasound. The prognosis of such pregnancies depends on the period when the diagnosis was confirmed and if it was diagnosed later in pregnancy, the pregnancy proceeds while sensory functions are preserved. Polyhydramnios Polyhydramnios or the increased quantity of amniotic fluid can occur in diabetic pregnancies regardless of fetal anomalies. The mechanism of polyhydramnios in diabetes can be due to: (1) increased osmolality of the amniotic fluid due to high values glucose in the amniotic fluid, (2) polyuria of the fetus which is hyperglycaemic and (3) reduced function of fetal swallowing. Malformations of the central nervous system of the fetus occur most frequently together with polyhydramnios and as such have a share of 45% of all the malformations in diabetic pregnancy. Anencephaly occurs in 80% of the anomalies of the central nervous system, while gastrointestinal malformations have a share of 30% of all the malformations which accompany diabetes and pregnancy.

Ivaniševic´

248

It is simple to diagnose polyhydramnios by ultrasound, measuring one vertical pocket of the amniotic cavity filled with amniotic fluid of 8 cm or more or it is measured by a four-quadrant technique and the amniotic fluid index is obtained, which is pathological if the sum is above 24 cm [53].

Biophysical Profile according to Manning

Ultrasound monitoring of diabetic pregnancies should also include the estimate of the biophysical profile according to Manning [22]. Depending on the degree of fetal endangerment, the biophysical profile has to be determined starting from the 30th to 34th week to birth, once or twice a week with regular controls with the cardiotocograph.

Biochemical Screening Test in the First and the Second Trimester for Fetal Chromosomopathies and the Neural Tube Defect

Detection of fetal malformations is of great significance in diabetic pregnancy, particularly if the mother’s glycaemia is regulated suboptimally. The first biochemical marker which was used in the detection of malformations in a diabetic pregnancy was the mother’s serum AFP immediately at the beginning of the second trimester [54]. The frequency of the neural tube defect in fetuses of diabetic mothers is up to 10 times higher than in healthy pregnant women. It is necessary to determine the serum AFP in a mother with a detailed ultrasound examination till the end of the first or at the very beginning of the second trimester. The screening carried out by determining AFP, human chorionic gonadotropin (hCG) and non-conjugated oestriol (E3) in the mother’s serum in order to identify pregnant women with a risk of fetal chromosomopathy is today routinely carried out in the second trimester of pregnancy. Type 1 diabetes in pregnant women is the reason why this test has so far not been routinely used in our country in that group . The basic illness influences the significantly lower level of AFP and non-conjugated oestriol in the mother’s serum, so that a correction factor for AFP and E3 has to be included, in order to get an approximately precise risk calculation for the test that was carried out. The correction factor for hCG has still not been worked out, since hCG values in a diabetic pregnancy in the second trimester are higher than in healthy pregnant women. The triple test is most frequently carried out between the 15th and 18th week of pregnancy. The most frequently used correction factor for AFP is 0.77 and for non-conjugated oestriol 0.92 according to Wald and Kennard [55].

Ultrasonography in Type 1 DM Pregnancy

249

Table 2. MoM and mean ⫾ SD values of hCG, AFP and uE3 after adjustment for maternal weight in all type 1 diabetes mellitus pregnant patients and controls [Ivanisevic and Djelmis, unpubl. data] Serum parameters

AFP uE3 hCG

Controls

Type 1 DM

p values

MoM

mean ⫾ SD

MoM

mean ⫾ SD

0.97 0.97 0.92

1.01 ⫾ 0.63 1.00 ⫾ 0.29 1.07 ⫾ 0.63

0.87 0.91 0.83

0.90 ⫾ 0.29 0.92 ⫾ 0.31 0.86 ⫾ 0.30

0.09 0.18 0.24

MoM ⫽ Multiple of median; DM ⫽ diabetes mellitus.

In the Zagreb State Referral Centre for Diabetes in Pregnancy at the moment we are using a second trimester screening test in patients with type 1 diabetes mellitus who had optimal surveillance of blood glucose before conception and during pregnancy. Our results in such patients have proved that the general adjustment of AFP, uE3 and hCG except for weight is no longer necessary (table 2). Today, there is a need for earlier non-invasive tests; so now in the first trimester of pregnant women without diabetes pregnancy-associated plasma protein-A (PAPP-A) and free ␤-hCG are determined, in combination with the value of the nuchal fold [56]. The measurement of the placenta volume at the end of the first trimester by 3-dimensional ultrasound and placenta volume/crown-rump length should help in interpreting variables measured in a biochemical screening test and increase the sensitivity of the test itself [57].

References 1 2 3 4 5 6 7

Korgun ET, Demir R, Hammer A, et al: Glucose transporter expression in rat embryo and uterus during decidualization, implantation, and early postimplantation. Biol Reprod 2001;65:1364–1370. Dhanasekaran N, Wu YK, Reece EA: Signaling pathways and diabetic embryopathy. Semin Reprod Endocrinol 1999;17/2:167–174. Burton GJ, Hempstock J, Jauniaux E: Nutrition of the human fetus during first trimester – a review. Placenta 2001(suppl A):S70–77. Reece EA, Pinter E, Homko C, Wu YK, Naftolin F: The yolk sac theory: Closing the circle on why diabetes-associated malformations occur. J Soc Gynecol Invest 1994;1/1:3–13. Mara E, Foster GS: Spontaneous regression of a yolk sac associated with embryonic death. J Ultrasound Med 2000;19:655–656. Pedersen JF, Molsted-Pedersen L: Early growth retardation in diabetic pregnancy. Br Med J 1979;i:18. Reece EA: Maternal fuels, diabetic embryopathy: Pathomechanisms and prevention. Semin Reprod Endocrinol 1999;17:183–194.

Ivaniševic´

250

8 9 10

11

12 13 14 15

16 17 18

19

20 21 22 23 24

25

26 27 28 29

30

Starcevic V, Djelmis J, Ivanisevic M, Mayer D: The effect of glycemia on early embryonic development in diabetic pregnancies. Prenat Neonat Med 2001;6/4:208–213. Ivanisevic M, Bukovic D, Starcevic V, Djelmis J, Pfeifer D: Influence of hyperglycemia on early embryonal growth in IDDM pregnant women. Coll Antropol 1999;23/1:183–188. Deter RL, Buster JE, Casson PR, Carson SA: Individual growth patterns in the first trimester: Evidence for difference in embryonic and fetal growth rates. Ultrasound Obstet Gynecol 1999;13/2:90–98. Snijders RJM, Noble P, Sebire N, Souka A, Nicolaides KH: UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal translucency thickness at 10–14 weeks of gestation. Lancet 1998;352:343–346. Goldstein I, Tamir A, Reece EA: The fetal superior cerebellar vermian width in normal, growthrestricted and macrosomic fetuses. J Matern Fetal Med 2001;10:23–27. Reece EA, Goldstein I, Hobbins JC: Fundamentals of OB/GYN Ultrasound. Norwalk, Appleton & Lange, 1994. Djelmis J, Bukovic D, Pfeifer D, Ivanisevic M: Ponderal index and disproportionate fetal growth in IDDM pregnancies. Coll Antropol 1998;22:491–495. Greco P, Vimercati A, Hyett J, Rossi AC, Scioscia M, Giorgino F, Loverro G, Selveggi L: The ultrasound assessment of adipose tissue deposition in fetuses of ‘well controlled’ insulin-dependent diabetic pregnancies. Diabet Med 2003;20:858–862. Wladimiroff JW, Bloemsma CA, Wallenburg HCS: Ultrasonic diagnosis for the large-for-date infant. Obstet Gynecol 1978;52:285. Tamura RK, Sabbagha RE, Depp R: Diabetic macrosomia: Accuracy of third trimester ultrasound. Obstet Gynecol 1986;52:828. Delmis J, Drazancic A, Ivanisevic M, Suchanek E: Glucose, insulin, HGH and IGF-I levels in maternal serum, amniotic fluid and umbilical venous serum: A comparison between late normal pregnancy and pregnancies complicated with diabetes and fetal growth retardation. J Perinat Med 1992;20/1:47–56. Djelmis J, Blajic J, Bukovic D, Pfeifer D, Ivanisevic M, Kendic S, Votava Raic A: Glycosylated hemoglobin and fetal growth in normal, gestational and insulin dependent diabetes mellitus pregnancies. Coll Antropol 1997;21:621–629. Delmis J, Bukovic D, Ivanisevic M, Kendic S: Growth factors in macrosomic newborns of healthy mothers. Coll Antropol 1996;20/1:83–88. Kainer F, Weiss PA, Hutter U, Haas J: Ultrasound growth parameters in relation to levels of amniotic fluid insulin in women with diabetes type-I. Early Hum Dev 1997;49/2:113–121. Manning FA: The fetal biophysical profile score; in Aspects of Fetal Life. Norwalk, Appleton & Lange, 1994. Dicker D, Goldman JA, Yeshaya A, Peleg D: Umbilical artery velocimetry in insulin-dependent diabetes mellitus pregnancies. J Perinat Med 1990;18:391–395. Reece EA, Hagay Z, Assimakopoulos E, et al: Diabetes mellitus in pregnancy and the assessment of umbilical artery waveforms using pulsed Doppler ultrasonography. J Ultrasound Med 1994;13/2:73–80. Johnstone F, Steel JM: Use of Doppler ultrasound in the management of diabetic pregnancy; in Pearce JM (ed): Doppler Ultrasound in Perinatal Medicine. Oxford, Oxford University Press, 1992, pp 178–188. Landon MB, Gabbe SG, Bruner JP, Ludmir J: Doppler umbilical velocimetry in pregnancy complicated by insulin dependent diabetes mellitus. Obstet Gynecol 1989;73:961–965. Bracero LA, Jovanovic L, Rochelson B, Bauman W, Farmakides G: Significance of umbilical and uterine artery velocimetry in the well-controlled pregnant diabetic. J Reprod Med 1989;34:273–276. Ishimatsu J, Yoshiumura O, Manabe A: Umbilical artery blood flow velocity waveforms in pregnancy complicated by diabetes mellitus. Arch Gynecol Obstet 1991;248:123–127. Reece EA, Hagay Z, Roberts AB, et al: Fetal Doppler and behavioral responses during hypoglycemia induced with the insulin clamp technique in pregnant diabetic women. Am J Obstet Gynecol 1995;172:151–155. Giles WB, Trudinger BJ, Baird PJ: Fetal umbilical artery flow velocity waveforms and placental resistance: A pathological correlation. Br J Obstet Gynaecol 1985;92:31–38.

Ultrasonography in Type 1 DM Pregnancy

251

31

32

33 34 35

36

37

38 39 40 41 42 43

44

45 46 47 48 49

50

51 52 53

Fox H: The placenta in diabetes mellitus; in Sutherland HW, Stowers JM, Pearson DW (eds): Carbohydrate Metabolism in Pregnancy and the Newborn. London, Springer, 1989, pp 109–117. Zimmerman P, Kujansuu E, Tuimala R: Doppler velocimetry of the umbilical artery in pregnancies complicated by insulin-dependent diabetes mellitus. Eur J Obstet Gynecol Reprod Biol 1992;47:85–93. Kofinas A, Penry M, Swain M: Uteroplacental Doppler flow velocity wavefrom analysis correlates poorly with glycemic control in diabetic women. Am J Perinatol 1991;8:273–277. Salvesen DR, Higueras MT, Brudenell JM, Drury PL, Nicolaides KH: Doppler velocimetry and fetal heart rate studies in nephropathic diabetics. Am J Obstet Gynecol 1992;167:1297–1303. Sekizuka N, Hasegawa K, Takakuwa K, Tanaka K: Scoring of uterine artery flow velocity waveforms in the assessment of fetal growth restriction and/or pregnancy induced hypertension. J Matern Fetal Invest 1997;7:197. Gudmundsson S, Valenzin L, Pirhonen J, Olofsson PA, Dubiel M, Marsal K: Factors affecting color Doppler energy ultrasound recordings in an in-vitro model. Ultrasound Med Biol 1998;24:899. Pretorius DH, Nelson TR, Baergen RN, Pai E, Cantrell C: Imaging of placental vasculature using three-dimensional ultrasound and color power Doppler: A preliminary study. Ultrasound Obstet Gynecol 1998;12:45. Molsted-Pedersen LM, Tygstrup I, Pedersen J: Congenital malformations in newborn infants. Lancet 1964;i:1124–1126. Kucera J: Rate and type of congenital anomalies among offspring of diabetic women. J Reprod Med 1971;7:61–70. Chung CS, Myrianthopoulos NC: Factors affecting risks of congenital malformations. II. Effect of maternal diabetes on congenital malformations. Birth Defects 1975;11:23–38. Becerra JF, Khoury MJ, Cordero JF, Erickson JD: Diabetes mellitus during pregnancy and the risks for specific birth defects: A population-based controlled study. Pediatrics 1990;85:1–9. Reece EA, Hobbins JC: Diabetic embryopathy: Pathogenesis, prenatal diagnosis and prevention. Obstet Gynecol Surv 1986;41:325–335. Albert TJ, Landon MB, Wheller JJ, Samuels P, Cheng RF, Gabbe S: Prenatal detection of fetal anomalies in pregnancies complicated by insulin-dependent diabetes mellitus. Am J Obstet Gynecol 1996;174:1424–1428. Milunsky A, Alpert E, Kitzmiller JL: Prenatal diagnosis of neural probe defects. VIII. The importance of serum alpha fetoprotein screening in diabetic pregnant women. Am J Obstet Gynecol 1982;104:1030. Cowchock S, Ainbender E, Prescott G: The recurrence risk of neural probe defects in the United States: A collaborative study. Am J Med Genet 1980;5:309. Bronshtein M, Zimmer EZ, Gerlis LM, Lorber A, Drugan A: Early ultrasound diagnosis of fetal congenital heart defects in high-risk and low-risk pregnancies. Obstet Gynecol 1993;82:225–229. Shields LE, Gan EA, Murphy HF, Sahn DJ, Moore TR: The prognostic value of hemoglobin A sub 1c in predicting fetal heart disease in diabetic pregnancies. Obstet Gynecol 1993;81:954–957. Weber HS, Copel JA, Reece EA, Green J, Kleinman CS: Cardiac growth in fetuses of diabetic mothers with good metabolic control. J Pediatr 1991;118:103–107. Wigton TR, Sabbagha RE, Tamura RK, Cohen L, Minogue JP, Strasburger JF: Sonographic diagnosis of congenital heart disease: Comparison between the four-chamber view and multiple cardiac views. Obstet Gynecol 1993;82:219–224. Crane JP, LeFevre ML, Winborn RC, et al: A randomized trial of prenatal ultrasonographic screening: Impact on the detection, management, and outcome of anomalous fetuses. Am J Obstet Gynecol 1994;171:392–399. Gladman G, McCrindle BW, Boutin C, Smallhorn JF: Fetal echocardiographic screening of diabetic pregnancies for congenital heart disease. Am J Perinatol 1997;14:59–62. Morin PR, Potier M, Dallaire L: Prenatal detection of intestinal obstruction: Deficient fluid disaccharidases in affected fetuses. Clin Genet 1980;18:217. Carlson DE, Platt LD, Mederais AL: Quantifiable polyhydramnios: Diagnosis and management. Obstet Gynecol 1990;75:989.

Ivaniševic´

252

54

55 56

57

Milunsky A, Alpert E, Kitzmiller JL: Prenatal diagnosis of neural tube defects. VIII. The importance of serum alpha fetoprotein screening in diabetic pregnant women. Am J Obstet Gynecol 1982;104:1030. Wald NJ, Kennard A: Prenatal biochemical screening for Down’s syndrome and neural tube defects. Curr Opin Obstet Gynecol 1992;4:302–307. Spencer K, Souter V, Tul N, Snijders R, Nicolaides KH: A screening program for trisomy 21 at 10–14 weeks using fetal nuchal translucency, maternal serum free beta human chorionic gonadotropin and pregnancy associated plasma protein-A. Ultrasound Obstet Gynecol 1999;132:231–237. Metzenbauer M, Hafner E, Hoefinger D, et al: Three-dimensional ultrasound Measurement of the placental volume in early pregnancy: Method and correlation with placenta parameters. Placenta 2001;22:602–605.

Marina Ivaniševic´, MD, PhD Department of Obstetrics and Gynecology, State Referral Centre for Diabetes in Pregnancy Medical School, University of Zagreb Petrova 13, HR–10000 Zagreb (Croatia) Tel./Fax ⫹385 1 4604740, E-Mail [email protected]

Ultrasonography in Type 1 DM Pregnancy

253

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 254–270

Diabetic Pregnancy: Maternal Metabolic and Microvascular Complications in Type 1 Diabetes mellitus Donald W.M. Pearson Aberdeen Royal Infirmary, Foresterhill, Aberdeen, UK

With current management strategies a successful outcome of pregnancy can usually be anticipated in women with pre-existing diabetes; however, diabetic pregnancy is statistically a high-risk pregnancy with regard to fetal mortality and morbidity [1–5]. Meticulous blood glucose control before and during pregnancy is the cornerstone of management; in addition to metabolic supervision mothers require close obstetric and medical surveillance. Other chapters address obstetric surveillance. This chapter discusses the issues relating to the presence and progression of microvascular complications, such as retinopathy and nephropathy, and metabolic emergencies since severe hypoglycaemia and diabetic ketoacidosis (DKA) can develop quickly and have particular features in pregnancy. In recent years there has been a striking increase in type 1 diabetes mellitus (T1DM) in many European countries. T1DM is one of the most common chronic childhood illnesses affecting 18–20 children per 100,000 in the UK. The largest rate of increase is seen in children aged 0–4 years; girls diagnosed with T1DM in this age group may have been exposed to the effects of chronic hyperglycaemia for over 20 years when they plan or become pregnant. During pregnancy the average age of mothers with T1DM is just under 30 years of age and in Scotland women will have had T1DM on average for 12.9 years at the time of pregnancy [2]. Microvascular complications of diabetes mellitus are associated with duration of the condition and the presence and extent of microvascular complications will influence the progress of pregnancy.

Preconception Care: Glycaemic Control and General Health Issues

The importance of pregnancy planning should be discussed, in a sensitive manner, with all girls who have T1DM from menarche onwards. Adolescence can be a challenging time for everyone and the restrictions associated with diabetes can make life especially difficult. Support, education and encouragement for adolescents and young women with diabetes can help to achieve optimal glycaemic control. The importance of pregnancy planning should be regularly emphasized to women during the reproductive years and effective contraceptive advice should be part of a routine clinical care package. The apathetic teenager almost always becomes the interested and participative women when she is planning or already pregnant. The benefits of a planned pregnancy should be part of any diabetes education package for women. Infants whose mothers with diabetes received detailed multidisciplinary pre-pregnancy care show significantly fewer major congenital malformations with the rate approximating the non-diabetic population [1]. Attendance for structured pre-pregnancy care is also associated with a reduction in the rate of spontaneous abortion and in other complications of pregnancy. Infants of mothers attending pre-pregnancy care have fewer problems and are kept in special care for shorter periods than infants of non-attending mothers. The essential components of a pre-pregnancy care programme include: • Assessment of glycaemic control and methods to optimize control • Review of the medical, obstetric and gynaecological history • Screening for complications of diabetes Table 1 summarizes the components of pre-pregnancy care. Patients with type 2 diabetes should be considered for insulin treatment and should be assessed in a similar programme. The Diabetes Control and Complications Trial (DCCT) [6] proved the feasibility and importance of strict metabolic control in delaying and preventing microvascular complications. Microvascular changes can be reduced by 37% for each 1% reduction in HbA1c and by 13% for each 10 mm Hg reduction in systolic blood pressure. Thus girls with diabetes should aim for as good a control as possible from diagnosis, however motivation to achieve optimal control outwith pregnancy is not always present. The risk of hypoglycaemia is the major limiting factor which deters people with T1DM to strive for normal blood glucose levels. During pregnancy, good control is particularly important since it will reduce stillbirth rate, neonatal hypoglycaemia and respiratory distress syndrome. Women should strive to maintain blood glucose level as near the non-diabetic range as possible without excessive risk of hypoglycaemia. This

Metabolic and Microvascular Complications in Pregnancy with T1DM

255

Table 1. A pre-pregnancy care programme for T1DM Achieve optimal glycaemic control Aim for HbA1c level within the non-diabetic range (or as near to the non-diabetic range as possible while avoiding disabling hypoglycaemia) Review the home blood glucose monitoring technique and provide a blood glucose meter to test blood glucose 4–6 times daily Review the insulin regime: most women need intensive treatment with a multiple injection basal bolus regime or CSII Set blood glucose targets, e.g. fasting and premeal 4.0–5.5 mmol/l; 2 h postprandial ⬍7.0 mmol/l Discuss lifestyle issues which may affect glycaemic control, e.g. difficult work patterns Assess general health fitness for pregnancy and screen for factors which could disturb glycaemic control, e.g. urinary infection Medical history Review all medications and consider potential teratogens Arrange dietetic review and commence folic acid 5 mg daily (high dose recommended in view of the risk of neural tube defects) Ensure other medical conditions, e.g. thyroid disorders are optimally managed Screen for associated autoimmune disorders, e.g. coeliac disease Check rubella status Review menstrual and gynaecological factors which could impair fertility Reinforce antismoking advice Screen for evidence of microvascular complications and assess blood pressure Check retinal findings and treat proliferative retinopathy if identified Characterize urine protein excretion as normal, microalbuminuria or proteinuria Check baseline renal function Assess blood pressure at each visit Optimize blood pressure control

usually means targeting the blood glucose levels between 4 and 7 mmol/l. The diabetes team, but in particular the diabetes specialist nurses and specialist midwives, have an important role in educating women on the need for home blood glucose monitoring (usually 4–6 times a day) and introducing intensive insulin regimes if the women are not already on these programmes. Intensive insulin therapy attempts to mimic physiological insulin secretion [7]. Women inject a ‘bolus’ of short-acting insulin (or insulin analogue) before each meal and take a ‘basal’ injection of an intermediate or long-acting insulin (or long-acting insulin analogue) once a day to maintain the background level of insulin. For example, the basal insulin may be administered as an isophane preparation before bedtime to control the fasting glucose level. Some women need a second dose of isophane before breakfast to achieve good glycaemic control between meals during the daytime hours. The meal time ‘boluses’ of

Pearson

256

short-acting insulin are taken before the main meals, usually before breakfast, lunch and the evening meal. The dosages vary depending on the amount of carbohydrate in the meal and anticipated energy expenditure after the meal. Short-acting insulin analogues are increasingly used to achieve satisfactory postprandial glucose control. A basal bolus regime has the advantage of increased flexibility with regard to exercise, meal timing and meal size and it is most likely to work effectively in the appropriately educated and motivated patient. Some women cannot achieve acceptable glycaemic control on currently available insulin regimes. Continuous subcutaneous insulin infusion (CSII) can provide an alternative for this group. As with other regimes it is sensible to institute any changes a few months prior to conception. This ensures the woman is well educated about the regime and able to cope with varying situations. It is very important that the patient using CSII has a full education package and support to ensure the benefits from this type of regime are achieved. Midwifery and obstetric staff may need help and guidance for this type of approach during pregnancy and labour. In some European countries a significant percentage of women use this type of treatment during pregnancy but in many countries only a relatively small percentage of women use the technique. While an insulin regime is very important many factors will influence insulin action and effectiveness. Injections to the buttocks or thighs are more slowly absorbed than those in the abdomen. Injections too close to the skin surface or too deep into muscle will also affect absorption. Other factors such as skin temperature, exercise and massage can have an impact on absorption. Insulin suspensions need to be carefully mixed before injection and inspection of injection sites may reveal lipohypertrophic or lipoatrophic areas through which absorption will be variable. Thus it is prudent to consider and review all the practical aspects of insulin administration. Dietary advice is also essential before, during and after pregnancy. Such advice will encourage foodstuffs with a high level of complex carbohydrates, soluble fibre and vitamins. In general women are advised to reduce levels of saturated fat intake. Neural tube defects in high-risk pregnancies are associated with lower levels of folate. A large study of non-diabetic women has shown that the prescription of 4 mg of folate supplement pre- and periconceptionally provides protection against neural tube defects in women at high risk. In our practice all women are advised to take 5 mg of folic acid preconceptionally and continue it during pregnancy until around 12 weeks of gestation. In Scotland around 50% of women are documented as taking periconceptional folic acid although reports from Holland indicate a higher uptake of folate supplementation in the population prior to and during early pregnancy [1]. All people with diabetes should be strongly counselled against smoking since it is a significant reversible risk factor for cardiovascular disease which is the com-

Metabolic and Microvascular Complications in Pregnancy with T1DM

257

monest cause of death in people with diabetes. Diabetes and smoking are multiplicative risk factors for cardiovascular disease. Smoking has also been reported to increase the risk of development and progression of microvascular complications such as retinopathy. Smoking cessation is particularly important in pregnancy. Although smoking is not more common in women with diabetes than in other pregnant women, certain conditions such as hypertension and autoimmune diseases are increased in the diabetes population. Hypertension management is discussed in detail later in this chapter and also in the chapter by Mathiesen and Damm [this vol., pp. 271–277]. Thyroid dysfunction is the most common related autoimmune condition in women with diabetes. Some national studies have found a prevalence of thyroid peroxidase antibodies in 22% of patients with T1DM. Active hyperthyroidism and hypothyroidism should be controlled prior to conception. Approximately 1 in 10 patients with T1DM express transglutaminase IgA autoantibodies and more than half of these patients have coeliac disease on intestinal biopsy. Thus regular screening for thyroid and coeliac disease is recommended during childhood in people with T1DM. If pathology is identified, appropriate management should be initiated and the conditions stabilized prior to pregnancy.

Microvascular Complications of Diabetes

Outwith pregnancy regular review of individuals with diabetes should address both metabolic control and complication screening for the early detection of microvascular disease. Screening is important since effective interventions are available at an early stage or latent phase of the disease process. Outwith pregnancy complication screening is usually performed on an annual basis. Diabetic Retinopathy [8] Retinopathy is an ocular manifestation of a generalized endothelial and capillary disease. Capillary occlusion, a key factor in the development of retinopathy, happens through a variety of mechanisms involving coagulation factors, platelets, leucocytes and other components which adhere to damaged endothelium. The onset of T1DM precedes the development of its microvascular complications by several years. Prolonged exposure to high levels of blood glucose is recognized as the major factor contributing to the development of retinopathy but blood pressure, lipids and genetic factors also have a role in the pathogenesis of microvascular disease. Critical retinal ischaemia is preceded by other clinical features which are shown in table 2. Microaneurysms, which are recognized as dot haemorrhages, are the earliest clinical features of diabetic retinopathy. As the disease progresses

Pearson

258

Table 2. Diabetic retinopathy grading system Description

Outcome

R0

No diabetic retinopathy anywhere

Repeat examination in each trimester

R1 (mild)

BDR – mild At least one dot haemorrhage or microaneurysm with or without hard exudates

Repeat examination in each trimester

R2 (moderate)

BDR – moderate Four or more blot haemorrhages in one hemi-field only (Inferior and superior hemi-fields are delineated by a line passing through the centre of the fovea and optic disc)

Refer to ophthalmology

R3 (severe)

BDR – severe Any of the following features: Four or more blot haemorrhages in both inferior and superior hemi-fields Venous beading Intraretinal microvascular abnormalities

Refer to ophthalmology

R4 (proliferative)

Proliferative diabetic retinopathy Any of the following features: New vessels Vitreous haemorrhage

Refer to ophthalmology (urgent)

M1

Lesions within a radius of ⬎1 but ⱕ2 disc diameters of the centre of the fovea Any hard exudates

Refer to ophthalmology

M2

Lesions within a radius of ⱕ1 disc diameter of the centre of the fovea Any blot haemorrhages Any hard exudates

Refer to ophthalmology

Retinopathy

Maculopathy

BDR ⫽ Background diabetic retinopathy.

Metabolic and Microvascular Complications in Pregnancy with T1DM

259

other features are recognized. Cottonwool spots represent stasis in axoplasmic flow; hard exudates represent leakage of lipid and protein from small vessels and moderate background retinopathy is diagnosed by the finding of four or more blot haemorrhages in the hemi-field. Retinal neovascularization develops from postcapillary venules with surrounding areas of capillary closure. New vessels grow forward especially from the disk where there is no limiting membrane. Most guidelines recommend that all people with diabetes over 12 years of age should have retinal examination at diagnosis and thereafter annual screening. Visual acuity should be assessed as part of eye screening and a deterioration in visual acuity (by two lines or more on a Snellen’s chart) should merit referral to an ophthalmologist. In pregnancy retinal screening should be undertaken in each trimester. The methods for retinal screening include indirect ophthalmoscopy by a trained observer or a quality-controlled digital retinal screening programme. The findings of retinal screening should be described and graded with a plan of action (table 2). Tight blood pressure control has a dramatic effect on the progression of eye disease and the prevention of visual loss. Tight glycaemic control also has a beneficial effect on the progression of eye disease and the prevention of visual impairment. In some patients, however, if control is tightened up too quickly, this can lead to a progression of retinopathy [9]. In patients with no retinopathy or mild background changes, this usually manifests itself as the development of cottonwool spots. These will often disappear. In patients with severe background or worse retinopathy, rapid tightening of glycaemic control can lead to rapid progression to high-risk retinopathy. For this reason it is prudent to achieve optimal glycaemic control gradually prior to pregnancy in women with severe retinopathy and a high HbA1c. Contraception should be continued while control is gradually improved to the target HbA1c level. If screening reveals proliferative (new vessel) retinopathy, laser treatment is very effective and will prevent the majority of people from developing significant visual impairment. Laser treatment is given over several sessions. Thousands of high intensity laser burns are applied to the peripheral retina of the affected eye. Peripheral pan-retinal photocoagulation will lead to a regression of new vessels at the disc. New vessels elsewhere can be treated by focal laser therapy. Such treatment should be completed prior to pregnancy if possible. Retinopathy is common in pregnancy. In the Scottish National Audit [2] of those examined in the first trimester 21% had mild non-proliferative retinopathy, 3% preproliferative retinopathy and 6% proliferative disease. Such retinal disease can deteriorate during pregnancy. In another study 77.5% of women with baseline retinopathy showed progression during pregnancy with 22.5% requiring pan-retinal photocoagulation. Poor glycaemic control in the first trimester and pregnancy-induced chronic hypertension are independently associated with the progression of retinopathy. Fundal examination prior to conception and

Pearson

260

during each trimester is thus essential with more frequent assessment in those with poor glycaemic control or severe retinopathy. Early referral of pregnant women with moderate retinopathy or worse to an ophthalmologist is advised due to the potential for the rapid development of neovascularization. If macular oedema is present, particularly if aggressive, early delivery may be recommended. The pathophysiological mechanisms which contribute to the deterioration of retinopathy during pregnancy are poorly understood. The deterioration of retinopathy during pregnancy may be due to a combination of rapid onset of tight glycaemic control as well as the physiological effects of pregnancy and effective blood pressure control is important in modifying the effect of pregnancy on retinopathy. In the longer term, parous women with T1DM have significantly lower levels of all retinopathy compared with nulliparous women. The associated significant difference in HbA1c suggests that improved glycaemic control associated with pregnancy may be sustained over time with beneficial effects on long-term complications. Thus women should be reassured that tight glycaemic control during and immediately after pregnancy can effectively reduce the longterm risks of retinopathy in future. Renal Disease in Diabetes [10] Twenty percent of people with T1DM will develop diabetic nephropathy after 25 years of disease duration. Under the microscope diabetic glomerulopathy has a characteristic appearance and histological features include basement membrane thickening, glomerular hypertrophy, mesangial cell hypertrophy and expansion of the extracellular matrix. A deficiency of proteoglycans responsible for the polyionic nature of the basement membrane alters the function of the membrane which becomes leaky to proteins such as albumin. Albuminuria is thus the clinical hallmark of the development of diabetic nephropathy which is related to the duration of diabetes and glycaemic control. Other factors which contribute to the changes observed in diabetic glomerulopathy include hyperfiltration due to increased glomerular blood flow and non-enzymatic glycosylation of proteins. Hypertension exacerbates the changes of diabetic nephropathy; thus control of systemic blood pressure is a particularly important aspect of management in diabetes. In clinical practice established diabetic nephropathy is a severe long-term complication of diabetes; persistent proteinuria, hypertension and a progressive decline in renal function are the clinical characteristics of this condition. It is almost always associated with significant retinopathy. Persistent proteinuria outwith pregnancy confers an 80- to 100-fold increase in mortality due largely to cardiovascular disease. Unless intervention takes place someone with persistent proteinuria due to diabetic nephropathy will progress to end-stage renal failure

Metabolic and Microvascular Complications in Pregnancy with T1DM

261

Table 3. Routine urine testing for diabetic renal disease Spot urine sample (for screening) Albustix Albumin/creatinine ratio

Dipstick one plus or greater (⬎200 mg/l) indicates proteinuria The ratio of urinary albumin to creatinine concentration; this is a screening test for microalbuminuria Normal values: female ⬍3.5 mg/mmol, male ⬍2.5 mg/mmol

Timed overnight urine collections (for diagnosis) Normoalbuminuria Normal albumin excretion rate ⬍20 ␮g/min Microalbuminuria Albumin excretion rate 20–200 ␮g/min Proteinuria Albumin excretion rate ⬎200 ␮g/min Spot protein/creatinine ratio ⬎70 mg/mmol 24 h urinary protein ⬎500 mg

requiring renal replacement therapy with dialysis or transplantation. It is in the knowledge of these facts that many clinicians would not recommend pregnancy for women with established diabetic nephropathy; however, many women with diabetic nephropathy have had a successful pregnancy. These pregnancies must be recognized as high risk and require particularly careful monitoring. Sensitive methods for detecting protein in the urine have allowed recognition of earlier changes of diabetic nephropathy before the onset of persistent proteinuria. Microalbuminuria is an early indicator of diabetic renal disease [see chapter by Mathiesen and Damm, this vol., pp. 271–277] and precedes the onset of overt nephropathy (proteinuria) in both T1DM and T2DM. Microalbuminuria is also a significant independent risk factor for the development and progression of cardiovascular disease. Aggressive management of blood pressure can retard the progression of diabetic renal disease at all stages and thus control of blood pressure is important in patients with microalbuminuria [11]. People with diabetes over the age of 12 should have an annual screen for diabetic renal disease (table 3). A dipstick is performed for proteinuria (Albustix or equivalent) and a microalbumin estimation performed if this is negative. If the dipstick test is positive urinary infection and other reasons for a positive test should be considered. Persistent proteinuria, when confirmed on two consecutive occasions, should be quantified using a protein/creatinine ratio on a spot sample. The serum creatinine should be estimated to assess renal function. If the dipstick test is negative a screen for microalbumin by measurement of albumin/creatinine ratio should be performed on the first voided urine sample

Pearson

262

after sleep. Samples with evidence of infection should be rejected. A normal result in a female is an albumin/creatinine ratio ⬍3.5 ␮g/mmol creatinine. During pregnancy the measurement should be performed in each trimester. If the result is abnormal the test should be repeated in the first instance. If the repeat test is positive it is prudent to use a timed overnight collection to confirm the diagnosis (an abnormal result is an albumin excretion rate ⬎20 ␮g/min). Outwith pregnancy women with established diabetic renal disease need frequent and intensive monitoring. Aggressive management of blood pressure is essential to preserve renal function. Such patients are at higher risk of retinopathy, autonomic neuropathy and cardiovascular disease. For women not planning pregnancy angiotensin-converting enzyme (ACE) inhibitor therapy is indicated [11]. ACE inhibitors are indicated following the detection of microalbuminuria or proteinuria even if the blood pressure is normal. The dose should be increased gradually to the maximum tolerated dose. Those intolerant of ACE inhibitors should be given an angiotensin II antagonist (AIIA). Lowering blood pressure in microalbuminuric patients reduces albumin excretion and progression to nephropathy. All agents which lower blood pressure reduce albumin excretion, although ACE inhibitors and AIIA probably have a class-specific effect independent of their hypotensive effect and, outwith pregnancy, are the drugs of choice for initial therapy. In patients with microalbuminuria/proteinuria the target blood pressure is usually less than 125/75 mm Hg. For women planning pregnancy ACE inhibitors and AIIA are contraindicated and many centres would commence such women on methyldopa. ACE inhibitors and AIIA have an effect on the renin-angiotensin-aldosterone system of the fetus. Renal perfusion in the second trimester depends on an intact renin-angiotensinaldosterone system. Renal maldevelopment thus follows the use of these drugs in pregnancy, particularly if they are used in the second and third trimester. In pregnancy, diabetic nephropathy of any degree is less common than retinopathy. In the Scottish Audit 85% of women had normal urinary albumin secretion. Those with evidence of renal disease are an important subgroup since there is an association between pre-existing nephropathy (microalbuminuria or albuminuria) and a poorer pregnancy outcome [12, 13]. This is not due to any increase in congenital malformations. Proteinuria increases during pregnancy, returning to pre-pregnancy level within 3 months of delivery. In women with both insipient and overt nephropathy the incidence of worsening chronic hypertension or pregnancy-induced hypertension/pre-eclampsia is high (varying from 40 to 73% across series). This worsening nephropathy and superimposed pre-eclampsia are common causes of preterm delivery in women with diabetic nephropathy. The management of pregnant women with diabetic nephropathy should follow the general recommendations with the target blood pressure less than

Metabolic and Microvascular Complications in Pregnancy with T1DM

263

140/80 mm Hg. As mentioned above ACE inhibitor should be avoided as they may adversely affect the fetus. Appropriate antihypertensive agents which may be used during pregnancy include methyldopa and nifedipine. Very careful monitoring of blood pressure is essential to identify optimal timing of delivery. Preterm delivery and delivery by Caesarean section are more frequent in women with hypertension and diabetic nephropathy. Factors which lead to earlier delivery include superimposed pre-eclampsia leading to uncontrolled hypertension and progression of diabetic retinopathy despite treatment. In some series more than half of the women with nephropathy are delivered before 37 weeks’ gestation. In about two thirds of these cases uncontrolled hypertension precipitated the early delivery. There is controversy about the influence of pregnancy on the long-term progression of diabetic nephropathy. Some groups have described a mean decline in glomerular filtration rate around 1.8 ml per month during pregnancy and 1.4 ml per month postpartum. Any difference in the progression of diabetic nephropathy during and after pregnancy is usually explained by the increased severity of the hypertension during pregnancy especially in the third trimester. This can happen despite intensified antihypertensive therapy. Other groups have not found an adverse effect of pregnancy on long-term renal function. The infants of mothers with diabetic nephropathy are often small for gestational age. Although the short-term outcome for pregnancies in women with diabetic nephropathy has improved there is concern about the long-term maternal health. After delivery efforts should be made to continue to optimize blood glucose and blood pressure control. An ACE inhibitor with appropriate contraception can be reintroduced to try and achieve blood pressure levels as near the normal range as possible. A highly motivated mother during pregnancy should be encouraged to continue with optimal metabolic and blood pressure control to maintain her health so that she can enjoy the growth and development of her children. Neuropathy Peripheral neuropathy is common in T1DM but is rarely a clinical issue in pregnancy. Patients with long-standing diabetes may have autonomic neuropathy and, occasionally, this will produce significant clinical and management problems in pregnancy. Autonomic neuropathy is associated with postural hypotension, neuropathic oedema and bladder hypotonia but these are rare conditions during the reproductive years. Gastrointestinal problems associated with autonomic neuropathy, however, can pose particular problems. Gastroparesis may cause delay in gastric emptying and lead to variable food absorption and unpredictable swings in blood glucose. During pregnancy vomiting due to gastroparesis

Pearson

264

can be very challenging for the patient and her carers. Metoclopramide and domperidone may help by promoting gastric emptying but occasionally intravenous nutritional support is necessary to maintain glycaemic control.

Acute Metabolic Problems in Pregnancy

Hypoglycaemia [14] One of the major themes in this book is the importance of maintaining near normal blood glucose levels prior to and during pregnancy. Achieving this aim comes at a cost and the risk of hypoglycaemia is often the limiting factor in reaching optimal glucose control. Hypoglycaemia can be classified as mild, moderate or severe. Severe hypoglycaemia indicates the need for a third party to treat an episode and is common in pregnancy. Many series [15, 16] quote around 40% of pregnancies affected by severe hypoglycaemia which is most likely to occur early in pregnancy or after delivery. Hypoglycaemia refers to any episode of low blood glucose (usually ⬍3.5 mmol/l) with or without symptoms and mild hypoglycaemia is almost inevitable if a patient strives for normoglycaemia. Mild hypoglycaemia is usually recognized early and responds quickly to ingested carbohydrate; however, mild episodes may predispose to severe hypoglycaemia which is associated with a fear of developing hypoglycaemia. Outwith pregnancy many patients will avoid very strict blood glucose control in order to minimize their risk of experiencing hypoglycaemia. It is thus very important that women should be advised about how to minimize the risk of severe hypoglycaemia and recognize and treat mild hypoglycaemia if it develops. Hypoglycaemia is avoided in people who do not have diabetes by a series of counterregulatory responses [17]. In normal physiology a blood glucose level in the low/normal range leads to reduced insulin secretion and an increase in other hormones such as glucagon, adrenaline, cortisol and growth hormone. Glucagon is the most potent and important counterregulatory hormone while catecholamines provide a backup if glucagon is deficient (for example in T1DM). Cortisol and growth hormone are important mainly in prolonged hypoglycaemia. Hypoglycaemia also leads to a stimulation of the autonomic nervous system which comprises sympathetic and parasympathetic components. Activation of the autonomic nervous system will lead to haemodynamic changes including an increased heart rate, blood pressure, cardiac output and myocardial perfusion. Cerebral blood flow is increased but renal blood flow may be reduced. Many of the symptoms and signs of hypoglycaemia, e.g. sweating, tremor, and tachycardia, result from activation of the autonomic nervous system and help to warn the individual that the blood glucose level is low.

Metabolic and Microvascular Complications in Pregnancy with T1DM

265

In people with diabetes hypoglycaemia occurs when there is an imbalance between glucose delivery, disposal and utilization. Reduction of glucose delivery can be due to reduced appetite, missed meals or malabsorption. Increased disposal occurs if there is a relative or absolute insulin excess. Activity leads to increased muscle glucose utilization with an immediate and also a late effect on glucose levels. Additional factors which may contribute to the risk of hypoglycaemia include hyperemesis gravidarum, lipohypertrophy at injection sites (leading to erratic absorption of insulin), errors of insulin administration, alcohol excess (particularly if combined with reduced carbohydrate intake), gastroparesis in women with autonomic neuropathy and, very rarely, adrenal insufficiency. The most common reasons are reduced food intake or increased exercise. The symptoms of hypoglycaemia vary between patients and the same patient may experience different symptoms in different circumstances. All women should be warned that the symptoms of hypoglycaemia may change or indeed be lost during pregnancy. Symptoms are classified as autonomic, neuroglycopenic or non-specific. Autonomic symptoms are due to activation of the autonomic nervous system (sweating, tremor, anxiety, or palpitations). Neuroglycopenic symptoms are due to reduced glucose delivery to the brain (poor concentration or abnormal behaviour, dizziness, headache). Some patients recognize non-specific symptoms such as tingling around the lips as the first manifestation. In most patients, outwith pregnancy, the autonomic symptoms occur before the neuroglycopenic symptoms and thus provide a useful warning of hypoglycaemia and indicate the need for ingestion of refined carbohydrate. In some patients the order of onset of symptoms may be reversed. Neuroglycopenic symptoms such as confusion precede rather than follow autonomic symptoms such as tremor or sweating. This occurs especially in those with long duration of diabetes, very tight glycaemic control, recurrent episodes of severe hypoglycaemia or during pregnancy. It may result in hypoglycaemia unawareness when the patient is unable to recognize the onset of the hypoglycaemia. This can have serious consequences. In some people it may be possible to regain the normal symptoms of hypoglycaemia through meticulous avoidance of hypoglycaemia. All patients treated with insulin should be advised to carry carbohydrate on their person. As soon as symptoms are recognized or if a low blood glucose value is recorded (with or without symptoms), the patient should be advised to take one of the following sources of refined carbohydrate to correct hypoglycaemia: • Dextrose tablets or sugar lumps • Half of a small bottle (150 ml) of Lucozade or equivalent sugary drink, e.g. Cola (not diet or light) • Fruit juice – 1 glass • Chocolate, e.g. mini chocolate bar

Pearson

266

If symptoms are not improving after 10 min, the ingestion of refined carbohydrate should be repeated. If the patient is too drowsy to co-operate, a concentrated glucose solution such as ‘Hypostop’, honey or jam may be applied to the inside of the cheeks and massaged from the outside. It is advisable for patients to keep suitable remedies at home and to instruct family members and friends on their use. If the patient is unresponsive, 1 mg of glucagon should be given subcutaneously or intramuscularly (once only per episode). It is advisable to instruct partners, close relatives or friends of insulin treated people on the use of glucagon and to ensure a kit is kept at home. If professional help is available 25 g (50 ml 50%) of dextrose can be given intravenously. After initial treatment of severe hypoglycaemia it is essential for the patient to take long-acting carbohydrate such as biscuits and milk or a sandwich to prevent recurrence of hypoglycaemia. Nocturnal hypoglycaemia is a common occurrence in insulin-treated patients. The patients may or may not wake up during the night with symptoms, or sometimes wake up the following morning feeling ‘hung over’. Nocturnal hypoglycaemia may contribute to the development of hypoglycaemia unawareness. The risk can be minimized by ensuring a snack containing complex carbohydrate is taken at bedtime. Continuous glucose monitoring systems can be useful to identify nocturnal hypoglycaemia or the patient advised to check glucose at around 2–3 a.m. After an episode of severe hypoglycaemia patients can experience marked hyperglycaemia (rebound hyperglycaemia) which may be prolonged as a result of the release of conterregulatory hormones. Thus hyperglycaemia may be a signal of an earlier episode of hypoglycaemia. The risk of severe hypoglycaemia can be minimized by very regular self-monitoring of blood glucose, review of the insulin regimen and consideration of the quantity and timing of carbohydrate intake. Examination of injection sites may provide an explanation for variable control if hypertrophy is present. There are certain situations where severe hypoglycaemia must be avoided, e.g. car driving. Patients should be advised to check their blood glucose before and during long car journeys and should always carry refined carbohydrate in the car. If a patient has a hypo while driving she should stop the car, remove the keys from the ignition, leave the driver’s seat and take oral carbohydrate. Driving should not be resumed for at least 45 min. Women who have lost their warning symptoms of hypoglycaemia should be advised not to drive until the problem has resolved. After delivery insulin requirements fall to pre-pregnancy levels and insulin doses need to be adjusted to avoid hypos. Breast-feeding increases insulin sensitivity; insulin doses and carbohydrate intake should be modified to accommodate breast-feeding. In summary strict blood glucose control increases the risk of hypoglycaemia and warning signs are often lost in early pregnancy. All women and their partners should be educated about hypoglycaemia and its management.

Metabolic and Microvascular Complications in Pregnancy with T1DM

267

Ideally women should not sleep in the home alone at night because of the risk of nocturnal hypoglycaemia. Women who have lost awareness of hypoglycaemia during pregnancy should be advised not to drive until their warning symptoms of hypoglycaemia return. Diabetic Ketoacidosis [18, 19] Uncontrolled catabolism due to relative insulin deficiency leads to the development of ketoacidosis. Outwith pregnancy a modest elevation of insulin levels is sufficient to inhibit ketogenesis however the insulin resistance of pregnancy predisposes to the rapid development of ketosis. Additional factors which contribute to ketogenesis include an excess of counterregulatory hormone secretions associated with infection and fluid depletion. Administration of high-dose steroid for preterm labour can also precipitate DKA if the increased insulin resistance due to steroid is not identified and managed appropriately. In clinical practice most episodes of DKA are precipitated by intercurrent infection. The constellation of clinical and biochemical features associated with DKA can all be explained on the basis of insulin deficiency. Lack of insulin leads to increased hepatic glucose production and reduced peripheral uptake of glucose in muscle and other insulin-dependent tissues. The increased plasma glucose exceeds the renal threshold for glucose and an osmotic diuresis results in dehydration and loss of electrolytes. At the same time the lack of insulin fails to inhibit lipolysis. Raised free fatty acids in the circulation are transported to the liver where they are metabolized in the mitochondria to ketone bodies (e.g. acetone and ␤-hydroxybutyrate). Ketone bodies can be a normal product of metabolism especially in the fasting state, but excessive production of ketone bodies leads to a metabolic acidosis. Uncontrolled metabolic perturbations are exacerbated by reduced renal perfusion and vomiting which further accelerates loss of fluid and electrolytes. The rising hydrogen ion concentration (with a corresponding fall in pH) means that the pH-dependent enzyme systems do not function effectively. The clinical features of DKA are thus explained on the basis of these biochemical and physiological abnormalities. Dehydration of mucous membranes, dry skin and reduced skin turgor are due to loss of fluids. In association with this loss, a compensatory tachycardia develops and, if dehydration is severe, hypotension may occur. Excess ketones are excreted in the urine and also may be detectable in the breath. Respiratory compensation for the metabolic acidosis leads to hyperventilation and the characteristic ‘air hunger’ described by Kussmaul. Nausea and vomiting are due to the gastritis associated with ketonaemia and abdominal pain may cause diagnostic confusion. Coma can rarely develop if the condition is not recognized and managed effectively. The diagnosis is confirmed by demonstrating hyperglycaemia, ketonaemia and acidosis (bicarbonate level less than 17 mmol/l). Occasionally blood

Pearson

268

glucose levels may be only slightly elevated or very rarely within the normal range. The severity of DKA is defined by the extent of acidosis and dehydration rather than the absolute blood glucose level. DKA is a metabolic emergency which must be managed in a high dependency area by clinicians experienced in correcting the biochemical and physiological abnormalities. A local protocol should be carefully followed to ensure that regular clinical and biochemical monitoring dictate the appropriate fluid, insulin and electrolyte replacements to achieve a satisfactory outcome [19]. Metabolic acidosis can develop quickly in pregnancy but should be avoided by prompt recognition and action. The fetus can tolerate maternal hypoglycaemia but is very sensitive to maternal ketoacidosis which results in a high incidence of fetal loss at all gestations. It is usually avoidable. Pregnant women should have equipment for the measurement of urinary or blood ketones. Elevation of the blood glucose over 10 mmol/l together with the presence of non-fasting ketonuria is an indication for increased insulin dosage and urgent further clinical assessment. Women should be advised to contact their diabetes care team in such circumstances even if they feel well since a hospital admission for intravenous fluids and insulin may be necessary. The most common association with DKA during pregnancy is infection, especially urinary tract infection, but it can develop if the patient or her care team does not recognize the physiological increasing insulin requirements as pregnancy progresses. High-dose corticosteroid treatment given in anticipation of preterm labour is a possible precipitant of DKA since steroid produces significant insulin resistance. In-patient supervision of metabolism by an experienced team is essential since blood glucose has to be regulated and controlled. Insulin resistance will develop a few hours after the initial dose of steroids and remain for several days. It is particularly marked in the first 48 h. Many units use an intravenous insulin/dextrose infusion to control blood glucose in these circumstances. Conclusion

Clinicians caring for women with T1DM during pregnancy need to appreciate the metabolic and microvascular issues which influence the outcome for the mother and her baby. A meticulous approach to maternal metabolic control, careful screening for microvascular disease and management strategies to deal with any problems are rewarded by the delivery of a healthy child to grateful parents. Acknowledgement The author is very grateful to Miss Jillian Anderson for typing this manuscript.

Metabolic and Microvascular Complications in Pregnancy with T1DM

269

References 1 2 3

4

5

6

7 8 9 10

11

12 13

14 15

16

17 18 19

Evers IM, de Valk HW, Visser GHA: Risk of complications of pregnancy in women with type 1 diabetes: Nationwide prospective study in the Netherlands. BMJ 2004;328:915–918. Penney GC, Pearson D: A national audit to monitor and promote the uptake of clinical guidelines on the management of diabetes in pregnancy. Br J Clin Governance 2000;5/1:28–34. Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah POD, Platt MJ, Stanisstreet M, van Velszen D, Walkinshaw S: Outcomes of pregnancy in insulin dependent diabetic women: Results of a five year population cohort study. BMJ 1997;315:275–278. Hawthorne G, Robson S, Ryall EA, Sen D, Roberts SH, Ward Platt MP: Prospective population based survey of outcome of pregnancy in diabetic women: Results of the Northern Diabetic Pregnancy Audit, 1994. BMJ 1997;315:279–281. Penney GC, Mair G, Pearson DW; Scottish Diabetes in Pregnancy Group: Outcomes of pregnancies in women with type 1 diabetes in Scotland: A national population based survey. BJOG 2003;110:315–318. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med 1993;329:977–986. Nachum Z, Ben-Shlomo I, Weiner E, Shalev E: Twice daily versus four times daily insulin dose regimens for diabetes in pregnancy: Randomised controlled trial. BMJ 1999;319:1223–1227. Best RM, Chakravarthy U: Diabetic retinopathy in pregnancy. Br J Ophthalmol 1997;81:249–251. Chantelau E, Khoner EM: Why some cases of retinopathy worsen when diabetic control improves. BMJ 1997;315:1105–1106. Hovind P, Tarnow L, Rossing P, Jensen BR, Graae M, Torp I, Binder C, Parving HH: Predictors for the development of microalbuminuria and macroalbuminuria in patients with type 1 diabetes: Inception cohort study. BMJ 2004;328:1105–1108. Mathiesen ER, Hommel E, Hansen HP, Smidt UM, Parving HH: Randomised controlled trial of long term efficacy of captopril on preservation of kidney function in normotensive patients with insulin dependent diabetes and microalbuminuria. BMJ 1999;319:24–25. Ekbom P, Damm P, Feldt-Rasmussen B, Feldt-Rasmussen U, Molvig J, Mathiesen ER: Pregnancy outcome in type 1 diabetic women with microalbuminuria. Diabetes Care 2001;24:1739–1744. Dunne FP, Chowdhury TA, Hartland A, Smith T, Brydon PA, McConkey C, Nicholson HO: Pregnancy outcome in women with insulin-dependent diabetes mellitus complicated by nephropathy. QJM 1999;92:451–454. Amiel SA: Hypoglycaemia avoidance – technology and knowledge. Lancet 1999;352:502–503. Penney GC, Pearson DWM: The relationship between birth weight and maternal glycated haemoglobin (HbA1c) concentration in pregnancies complicated by type 1 diabetes. Diabet Med 2003;20:162–166. Evers IM, de Valk HW, Mol BWJ, ter Braak EWMT, Visser GHA: Macrosomia despite good glycaemic control in type 1 diabetic pregnancy; results of a nationwide study in The Netherlands. Diabetologia 2002;45:1484–1489. Rosenn BM, Miodovnik M, Khoury JC, Siddiqi TA: Counterregulatory hormonal responses to hypoglycaemia during pregnancy. Obstet Gynecol 1996;87:568–574. Marshall SM, Alberti KGMM: Management of hyperglycaemic emergencies. Proc R Coll Physicians Edinb 1995;25:105–117. Carr D: Devising a protocol for managing diabetic ketoacidosis in adults. Pract Diabetes Int 1995;12:164–168.

Donald W.M. Pearson, MD Consultant Physician/Diabetologist, Aberdeen Royal Infirmary Foresterhill, Aberdeen AB25 2ZN (UK) Tel. ⫹44 1224 55 22 58, Fax ⫹44 1224 55 11 86 E-Mail [email protected]

Pearson

270

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 271–277

Pre-Eclampsia in Women with Type 1 Diabetes Elisabeth R. Mathiesen, Peter Damm Endocrine and Obstetric Clinic, Rigshospitalet, University Hospital of Copenhagen, Copenhagen, Denmark

Women with type 1 diabetes attending a diabetes clinic for preconception care are heterogeneous in regard to age, duration of diabetes, presence of hypertension, microangiopathy and frank diabetic nephropathy. Pregnancy in women with type 1 diabetes is associated with increased prevalence of severe complications. Besides the risk of giving birth to a child with a major congenital malformation [1], development of pre-eclampsia, which occurs in around 20% of the women, is of major concern [2]. In general the risk of developing pre-eclampsia is lowest in patients with a short duration of diabetes and good metabolic control and highest in women with diabetic nephropathy [2]. Likewise the risk of preterm delivery is increased to 30–38% [3, 4] and perinatal mortality up to 5% [5–8]. Pre-eclampsia is characterized by the development of hypertension (⬎140/90 mm Hg) and proteinuria (0.3 g/24 h) later than 20 weeks of gestation. It is a pregnancy-specific syndrome and a serious complication associated with severe morbidity and increased mortality for the mother and child. Preterm delivery occurs in more than half of the women with diabetes and pre-eclampsia [6]. Respiratory distress syndrome is common in these infants leading to prolonged stays at the neonatal intensive care unit. Diabetes Complications and Pre-Eclampsia

The severity of diabetes, especially kidney involvement, has an impact on the increased risk of development of pre-eclampsia. Diabetic nephropathy is characterized by persistent proteinuria (urinary albumin excretion ⬎300 mg/24 h) elevated blood pressure, oedema and a decline in kidney function. In patients

with established diabetic nephropathy the risk of developing superimposed pre-eclampsia characterized by a further increase in blood pressure level and proteinuria is between 36 and 66% [2]. The onset of pre-eclampsia is often early and the course severe in these women and frequently associated with intrauterine growth restriction and preterm delivery [4]. Microalbuminuria, the forerunner of diabetic nephropathy, is an early manifestation of diabetic microvascular disease [9, 10]. It is associated with slightly elevated blood pressure within the normal range and subclinical oedema due to universal vascular leakage of albumin, and it predicts overt diabetic nephropathy with persistent proteinuria and hypertension [9, 11]. Recently our group has demonstrated that patients with microalbuminuria also have a high risk of developing pre-eclampsia [4]. The prevalence of microalbuminuria (30–300 mg/24 h) was 11% in an unselected group of 240 pregnant women with type 1 diabetes. The prevalence of pre-eclampsia in women with microalbuminuria was 42%. Presence of elevated urinary albumin excretion (⬎30 mg/24 h) before pregnancy was the best predictor of pre-eclampsia. In contrast, women with normal urinary albumin excretion before pregnancy had a prevalence of pre-eclampsia as low as 6%, comparable to the risk in the background population. The white class was also associated with an increased risk of pre-eclampsia [4], but was eliminated as an independent predictor in multivariate regression analysis including elevated urinary albumin excretion. If the level of urinary albumin excretion is not determined before pregnancy, the level in early pregnancy is useful [4]. More than half (64%) of the women who developed preeclampsia in our group were characterized by elevated urinary albumin excretion prior to pregnancy. Development of pre-eclampsia was the main cause of preterm delivery in women with elevated urinary albumin excretion (fig. 1). Hypertension and a high normal systolic blood pressure are associated with an increased risk of pre-eclampsia in women with type 1 diabetes [12, 13]. Twenty-four-hour day-time blood pressure measurement was a predictor of pre-eclampsia in nulliparous women [12], but whether 24-hour blood pressure measurement was superior to office blood pressure was not investigated. We measured 24-hour blood pressure in 74 pregnant women with type 1 diabetes. Microalbuminuria continued to be the best predictor of pre-eclampsia before pregnancy with no additive effect of 24-hour blood pressure measurement [13]. The method was resource intensive and rather unpleasant for the women.

Metabolic Control

A significant association between poor metabolic control in early pregnancy and pre-eclampsia is well described [14]. Recently Hiilesmaa et al. [15]

Mathiesen/Damm

272

100 Preterm delivery (%)

90 80 n⫽9

70 60

n ⫽6

50 40

n ⫽8

30 20 10 0 Normal UAE Microalbuminuria Nephropathy (n ⫽ 240) (n ⫽ 26) (n ⫽11)

Fig. 1. Prevalence of preterm delivery in relation to urinary albumin excretion (UAE) at baseline. ⵧ ⫽ Preterm delivery associated with pre-eclampsia; 䊏 ⫽ preterm delivery due to other causes [from 4].

50

40

%

30

20

10

0 Mean ⫹ 2 SD

Mean⫹ 2–6 SD

Mean⫹6–10 SD

Mean ⫹10 SD

Fig. 2. Prevalence of pre-eclampsia in relation to haemoglobin A1c levels early in pregnancy. Haemoglobin A1c mean ⫹ 2 SD equals 5.6 in this assay (n ⫽ 638) [adapted from 15].

investigated 683 pregnancies in women with type 1 diabetes from one centre in Scandinavia (fig. 2). Pre-eclampsia developed in 12.8% of the women and in 2.7% of the non-diabetic control women. Women with diabetic nephropathy were not included, and urinary albumin excretion was not measured; the impact of microalbuminuria could thus not be examined. Glycaemic control in early

Pre-Eclampsia in Women with Type 1 Diabetes

273

pregnancy, nulliparity, retinopathy, and duration of diabetes were independent predictors of pre-eclampsia. The adjusted odds ratio for pre-eclampsia was 1.6 for each 1% increment in the haemoglobin A1c value at 7 weeks of gestation. In addition, improvement in metabolic control during the first half of the pregnancy decreased the risk of pre-eclampsia. In a subgroup of 138 (20%) parous women without retinopathy or nephropathy and less than 15 years’ diabetes duration the prevalence of pre-eclampsia was 2.8%, comparable to the control women. Based on these results [4, 15] one may conclude that women with normal urinary albumin excretion and haemoglobin A1c levels close to normal (i.e. ⬍7.5%) have a low risk of developing pre-eclampsia comparable to the background population, while women with microalbuminuria or diabetic nephropathy in association with poor metabolic control prior to pregnancy (⬎8.5%) have a high risk of developing pre-eclampsia.

Pathogenesis

The pathogenesis of pre-eclampsia in women with type 1 diabetes has as in women without diabetes not been described well. Pre-eclampsia occurs only in the presence of a placenta and the placenta from non-diabetic women with pre-eclampsia is characterized by insufficient endovascular invasion and spiral artery remodelling associated with decreased perfusion. The levels of the placenta hormones activin A and inhibin A are elevated and the prevalence of growth-restricted infants is increased. However, in diabetic women developing pre-eclampsia our group found normal levels of activin A and inhibin A and a low frequency of intrauterine growth restriction [16]. Development of pre-eclampsia, however, requires more than a decreased placental perfusion. Other conditions such as intrauterine growth restriction and the fact that approximately a third of preterm births manifest abnormal modifications of the spiral arteries identical to that present in pre-eclampsia suggest that pre-eclampsia is a two-stage disorder [17]. Numerous maternal factors can predispose to the disorder; these may be genetic, behavioural, or environmental. The list of predisposing factors includes hypertension, diabetes, insulin resistance and increased blood homocysteine concentration. Interestingly these conditions are also risk factors for other endothelial diseases, particularly arteriosclerosis, and are associated with late diabetic complications. Pre-eclampsia, arteriosclerosis and diabetes also share dyslipidaemia. Increased triglycerides, decreased HDL and increased concentrations of small dense LDL are characteristic of these disorders [17]. Pre-eclampsia is characterized by hypertension, proteinuria and oedema. Diabetic patients with microalbuminuria or manifest diabetic nephropathy already have a slightly elevated blood pressure and oedema besides elevated

Mathiesen/Damm

274

urinary albumin excretion. Manifest endothelial dysfunction is described in women with type 1 diabetes especially when microalbuminuria is present [18]. Endothelial dysfunction in type 1 diabetes is related to albumin excretion rate and glycaemic control. It is, therefore, likely that the burden of pregnancy leads to pre-eclampsia in diabetic women with pre-existing dysfunctional endothelium. A current hypothesis explaining the endothelial alterations in arteriosclerosis involves oxidative stress as pathogenically important. Administration of antioxidants to women early in pregnancy decreased oxidative stress, endothelial activation, and the frequency of pre-eclampsia, which supports the potential role of oxidative stress in pre-eclampsia [19]. Pre-eclampsia is proposed to be a disorder secondary to impaired placental perfusion interacting with maternal constitutional factors and thereby resulting in oxidative stress, endothelial activation, and a multisystemic maternal disease [17]. The increased prevalence of pre-eclampsia in women with type 1 diabetes is confined to women with elevated urinary albumin excretion and/or poor metabolic control. We suggest that this increased prevalence of pre-eclampsia is mainly related to maternal constitutional factors by increased susceptibility to endothelial activation, while poor placentation is not a major pathogenetic factor.

Treatment Modalities

Strict metabolic control both preconceptionally and during pregnancy is imperative to reduce the risk of pre-eclampsia in diabetic women. Preconceptionally the aim should be blood glucose within the normal range and haemoglobin A1c values around 6.4%, which is the upper normal limit. It is important to pay attention to the natural decline in haemoglobin A1c during pregnancy and the aim for haemoglobin A1c values during pregnancy should therefore be reduced to 5.6%, which is the upper normal limit late in pregnancy [20]. However, the increased risk of experiencing serious hypoglycaemia in these women makes it difficult to obtain these goals, especially in women with a long duration of diabetes. Women with elevated urinary albumin excretion deserve special attention during pregnancy. Strict metabolic control and treatment with angiotensinconverting enzyme inhibitors 3–6 months prior to pregnancy seemed to reduce the incidence of pre-eclampsia and preterm delivery in a small series from Israel including 8 women with type 1 diabetes and diabetic nephropathy [21]. The ACE inhibitor treatment of the women was stopped at the time of the first missed menstrual bleeding. Theoretically, this treatment strategy could also be beneficial in women with microalbuminuria. During pregnancy antihypertensive treatment with e.g. methyldopa should be initiated when blood pressure

Pre-Eclampsia in Women with Type 1 Diabetes

275

exceeds 140/90 mm Hg in pregnant women with diabetes. A more aggressive treatment strategy might be beneficial in women with elevated urinary albumin excretion. We suggest that antihypertensive treatment with methyldopa is beneficial in pregnant diabetic women with elevated urinary albumin excretion, even if the blood pressure is within the normal range. A beneficial effect of low-dose aspirin on the development of pre-eclampsia has not been demonstrated in a controlled trial including 462 diabetic women [22] but may be useful in subgroups including women with diabetes [23]. Supplementation with antioxidants including folic acid during the whole pregnancy has also been described to reduce the risk of pre-eclampsia in high-risk women [19].

Conclusion

Pre-eclampsia occurs in approximately 20% of the pregnancies in women with type 1 diabetes and is associated with preterm delivery and increased perinatal morbidity. Pre-eclampsia develops mainly in women with elevated urinary albumin excretion and/or poor metabolic control. We suggest that this increased prevalence of pre-eclampsia is mainly related to maternal constitutional factors by increased susceptibility to endothelial activation, while poor placentation plays a minor role. In order to prevent pre-eclampsia a strict metabolic control is of uppermost importance and improved treatment strategies including aspirin, folic acid and antihypertensive drugs in normotensive women with elevated urinary albumin excretion rate must be considered.

References 1 2 3

4 5 6

Suhonen L, Hiilesmaa V, Teramo K: Glycaemic control during early pregnancy and fetal malformations in women with type 1 diabetes mellitus. Diabetologia 2000;43:79–82. Sibai BM: Risk factors, pregnancy complications, and prevention of hypertensive disorders in women with pregravid diabetes mellitus. J Matern Fetal Med 2000;9:62–65. Sibai BM, Caritis SN, Hauth JC, MacPhersin C, VanDorsten JP, Klebanoff M, Landon M, Paul RH, Meis PJ, Miodivnic M, Dombrowski MP, Thurnau GR, Moawad AH, Roberts J: Preterm delivery in women with pregestational diabetes mellitus or chronic hypertension relative to women with uncomplicated pregnancies. Am J Obstet Gynecol 2000;183:1520–1524. Ekbom P, Damm P, Feldt-Rasmussen B, Feldt-Rasmussen U, Molvig J, Mathiesen ER: Pregnancy outcome in type 1 diabetic women with microalbuminuria. Diabetes Care 2001;24:1739–1744. Cnattingius S, Berne C, Nordstrøm M-L: Pregnancy outcome and infant mortality in diabetic patients in Sweden. Diabet Med 1994;11:696–700. Sibai BM, Caritis SN, Hauth JC, MacPhersin C, VanDorsten JP, Klebanoff M, Landon M, Paul RH, Meis PJ, Miodivnic M, Dombrowski MP, Thurnau GR, Moawad AH, Roberts J: Risk of preeclampsia and adverse neonatal outcomes among women with pregestational diabetes mellitus. Am J Obstet Gynecol 2000;182:364–369.

Mathiesen/Damm

276

7

8

9 10

11 12

13

14 15

16

17 18

19 20 21

22

23

Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah PO, et al: Outcomes of pregnancy in insulin dependent diabetic women: Results of a five year population cohort study. BMJ 1997;315:275–278. Hawthorne G, Robson S, Ryall EA, Sen D, Roberts SH, Ward Platt MP: Prospective population based survey of outcome of pregnancy in diabetic women: Results of the Northern Diabetic Pregnancy Audit, 1994. BMJ 1997;315:279–281. Mathiesen ER: Prevention of diabetic nephropathy – The role of microalbuminuria and possibilities for intervention. Dan Med Bull 1993;40:273–285. Mogensen CE, Chachati A, Christensen CK, Deckert T, Hommel E, Kastrup J, Lefebvre P, Mathiesen ER, Feldt-Rasmussen B, Schmitz A, Viberti GC: Microalbuminuria, an early marker of renal involvement in diabetes. Uremia Invest 1986;9:85–95. Jensen T, Bjerre-Knudsen J, Feldt-Rasmussen B, Deckert T: Features of endothelial dysfunction in early diabetic nephropathy. Lancet 1989;i:461–463. Lauszus FF, Rasmussen OW, Lousen T, Klebe TM, Klebe JG: Ambulatory blood pressure as predictors of preeclampsia in diabetic pregnancies with respect to urinary albumin excretion rate and glycaemic regulation. Acta Obstet Gynecol Scand 2001;80:1096–1103. Ekbom P, Stage E, Norgaard K, Clausen P, Feldt-Rasmussen U, Feldt-Rasmussen B, MolstedPedersen L, Damm P, Mathiesen ER: Urinary albumin excretion and 24-hour blood pressure as predictors of pre-eclampsia in type 1 diabetes. Diabetologia 2000;43:927–931. Hanson U, Persson B: Epidemiology of pregnancy-induced hypertension and preeclampsia in type 1 (insulin-dependent) diabetic pregnancies in Sweden. Acta Obstet Gynecol Scand 1998;77:620–624. Hiilesmaa V, Suhonen L, Teramo K: Glycaemic control is associated with pre-eclampsia but not with pregnancy-induced hypertension in women with type 1 diabetes mellitus. Diabetologia 2000;43:1534–1539. Mathiesen ER, Ekbom P, Skakkebæk NE, Andersson A-M, Feldt-Rasmussen U, Damm P: Can activin A and inhibin A predict development of preeclampsia in women with type 1 diabetes. Diabetologia 2001;44(suppl A):247. Roberts JM, Cooper DW: Pathogenesis and genesis of preeclampsia. Lancet 2001;357:53–56. Dogra G, Rich I, Stanton K, Watts GF: Endothelium-dependent and independent vasodilatation studied at normoglycaemia in type 1 diabetes mellitus with and without microalbuminuria. Diabetologia 2001;44:593–601. Chappell LC, Seed PT, Briley A, et al: Effects of antioxidants on the occurrence of preeclampsia in women at increased risk: A randomized trial. Lancet 1999;354:810–816. Nielsen LR, Ekbom P, Damm P, Glumer C, Frandsen MM, Jensen DM, Mathiesen ER: HbA1c levels are significantly lower in early and late pregnancy. Diabetes Care 2004;27:1200–1201. Hod M, van Dijk DJ, Karp M, Weintraub N, Rabinerson D, Bar J, Peled Y, Erman A, Boner G, Ovadia J: Diabetic nephropathy and pregnancy: The effect of ACE inhibitors prior to pregnancy on fetomaternal outcome. Nephrol Dial Transplant 1995;10:2328–2333. Caritis S, Sibai BM, Hauth J, Lindheimer M, Klebanoff M, Thom E, et al: Low-dose aspirin to prevent preeclampsia in women at high risk. National Institute of Child Health and Human Development Network of Maternal-Fetal Medicine Units. N Engl J Med 1998;338:701–705. Coomarasamy A, Honest H, Papaioannou S, Gee H, Khan KS: Aspirin for prevention of preeclampsia in women with historical risk factors: A systematic review. Obstet Gynecol 2003;101:1319–1332.

Elisabeth R. Mathiesen, MD Endocrine Clinic, Rigshospitalet, University of Copenhagen Blejdamsvej 9, DK–2100 Copenhagen (Denmark) Tel. ⫹45 3545 8358, Fax ⫹45 3545 2240, E-Mail [email protected]

Pre-Eclampsia in Women with Type 1 Diabetes

277

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 278–287

Management of Delivery Veronica A. Miller, Michael D.G. Gillmer Women’s Centre, John Radcliffe Hospital, Oxford, UK

It has long been recognized that the infant of the diabetic mother is at increased risk during labour and delivery. Bennewitz [1] in 1824 described a case of obstructed labour and intrapartum death of a 12-lb fetus in a gestational diabetic. Even after insulin therapy had been available for more than 25 years, Peel and Oakley [2] in 1949 reported a perinatal mortality rate of 37% in diabetic women. In nearly 20% of these cases intrauterine death occurred in the last 4 weeks of pregnancy and before the onset of labour, while a further 10% were intrapartum stillbirths. Recent studies have shown that while improved diabetic control has greatly improved pregnancy outcome in diabetic women increased rates of late intrauterine death, perinatal morbidity and mortality are still evident compared to the results in non-diabetic women [3, 4]. The diabetic mother must therefore be considered a high-risk obstetric patient and delivered in a unit that can provide experienced obstetric care, continuous fetal monitoring, and neonatal intensive care. The challenge for the obstetrician caring for the diabetic woman is to ensure that her labour and delivery are conducted in a manner that minimizes intervention and ensures a safe delivery for both mother and child. There is, however, much debate as to when and how the diabetic patient should be delivered [5].

Premature Labour

Premature labour has recently been reported to occur in 16% of type 1 diabetic women [3]. Although the cause is uncertain, this complication is more common in women with poorly controlled diabetes and may be due to an increased incidence of polyhydramnios in association with fetal macrosomia.

Elective preterm delivery is more common because of an increased incidence of other complications of pregnancy, compared to the non-diabetic. These include fetal compromise and pre-eclampsia. Once delivered, the infant of the diabetic mother is at much higher risk of developing the respiratory distress syndrome than the infant of a non-diabetic mother. This risk is increased further by preterm delivery and Caesarean section [6]. The main benefit of tocolytic drugs when given to women in preterm labour is to reduce the numbers who deliver within 7 days of commencing the drug. This enables maternal administration of corticosteroids which have been shown to improve neonatal lung function for up to 7 days, when given before 34 weeks’ gestation. The traditional management of women presenting with premature labour has been to suppress uterine contractions using a tocolytic ␤-adrenergic and to give parenteral corticosteroids, such as betamethasone [7]. As ␤-adrenergic drugs stimulate hepatic glycogenolysis and corticosteroids cause severe insulin resistance, this combination causes major disruption of diabetic control [8]. The patient is therefore at risk of ketoacidosis, pulmonary oedema and hypokalaemia. Barnett et al. [9] reported that it was necessary to administer up to 30 units of insulin intravenously hourly to maintain normoglycaemia in patients treated with a combination of ␤-adrenergic agents and corticosteroids. They advised that this therapy should be avoided whenever possible even in non-insulin-dependent diabetic patients because of the risk of ketoacidosis. Others have, however, concluded that these agents can be safely used to treat premature labour in patients with insulin-dependent diabetes, provided they are administered in a strictly controlled clinical setting [6]. If a tocolytic agent is used ␤-adrenergic agents no longer seem to be the best choice, especially in diabetic women, as alternatives such as the calcium channel blocker nifedipine and the oxytocin antagonist atosiban appear to have comparable effectiveness and are associated with fewer maternal adverse effects [10]. Indirect comparison of these agents has, however, shown that nifedipine is more effective than atosiban. It is also much cheaper and has the advantage that it is administered orally, whereas atosiban requires an intravenous infusion [11]. Magnesium sulphate is popular for tocolysis in some parts of the world. A recent randomized trial in which magnesium sulphate was compared with other tocolytics and a placebo, however, found that the use of this drug was associated with worse perinatal outcome in a dose-response fashion [12]. As there is still no clear evidence that tocolytic drugs improve outcome following preterm labour it is reasonable not to use them.

Management of Delivery

279

Although a single course of antenatal corticosteroid, e.g. betamethasone 12 mg i.m. or i.v. administered twice at 12-hourly intervals, has been shown to improve newborn lung function after preterm birth and to reduce the risk of neonatal death in non-diabetic women, there is no current evidence of the effectiveness of this therapy in diabetic women. Repeat courses of prenatal corticosteroids should be avoided because of the growing evidence that this practice may adversely affect fetal growth and brain development and induce postnatal insulin resistance [13].

Prediction of Fetal Macrosomia

Fetal macrosomia is a frequent complication in pregnancies of women with diabetes. The exact incidence, however, depends on the definition used. National studies have reported a birth weight greater than the 90th centile affecting 20–45% of the infants of type 1 diabetic mothers. The incidence of complications during labour and delivery is, however, more closely related to actual weight and a birth weight of greater than 4,000 g has been reported in 20–25% of the infants of women with diabetes and a weight of greater than 4,500 g in 7–10% [14, 15]. The incidence of shoulder dystocia rises progressively with increasing birth weight and there is evidence that maternal diabetes is an independent risk factor for this complication. Langer et al. [16] reported a 21.8% incidence of shoulder dystocia in diabetic women delivering infants weighing more than 4,500 g compared to 7.5% in non-diabetics. The increased risk of shoulder dystocia associated with the fetus of the diabetic woman appears to be due to an increase in shoulder width and the shoulderhead ratio compared to the fetus of the non-diabetic [17]. Magnetic resonance imaging of fetal shoulder width has been shown to correlate with postnatal calliper measurements and birth weight [18, 19] but has not been evaluated for the management of diabetic pregnancy. Although prediction of macrosomia in diabetic pregnancy has been considered inaccurate Best and Pressman [20], using a gestation-adjusted prediction method, have recently reported comparable mean absolute percent errors in diabetic and non-diabetic pregnancies of 7.4 and 8.3%, respectively, corresponding to mean absolute errors in the predicted birth weight of 265 and 262 g in the two groups. Jazayeri et al. [21] reported that a fetal abdominal circumference of 35 cm or more identified 93% of infants weighing more than 4,000 g. Gilby et al. [22] also studied the fetal abdominal circumference measurement as a predictor of macrosomia in non-diabetic women. They demonstrated that if the abdominal

Miller/Gillmer

280

circumference was less than 35 cm, the risk of an infant birth weight of greater than 4,500 g was less than 1%. When the abdominal circumference exceeded 38 cm 78% of infants weighed more than 4,000 g and 37% weighed more than 4,500 g. Conway and Langer [23] performed a prospective study in which diabetic women with an ultrasonographic estimated fetal weight of equal to or greater than 4,250 g were delivered by elective caesarean section while those with an estimated fetal weight of less than 4,250 g underwent induction of labour. Ultrasonography correctly identified the presence of macrosomia in 87%. The rate of shoulder dystocia was 7.4% in macrosomic infants delivered vaginally. Although the caesarean section rate increased from 21.7 to 25.1%, they concluded that their protocol reduced the rate of shoulder dystocia without a meaningful increase in caesarean section rate. Cohen et al. [24] performed a retrospective analysis of the difference between the abdominal diameter and biparietal diameter, measured ultrasonographically within 2 weeks of delivery in diabetic women with an estimated fetal weight of 3,800–4,200 g. This revealed that no infant with an abdominalbiparietal diameter measurement of less than 2.6 cm suffered shoulder dystocia. In a later publication this group also demonstrated that the severity of shoulder dystocia correlates positively with the difference in the abdominal and biparietal diameters in the fetus of diabetic women [25].

Timing of Delivery

In the 1930s concern about late intrauterine death of the fetus in diabetic pregnancies led to a policy of early delivery, often well before 36 weeks’ gestation and frequently by elective caesarean section [26]. In the 1950s unexplained stillbirths remained relatively common near term and led to the practise of elective delivery of the diabetic patient usually between 36 and 38 weeks [27]. This was associated with an increase in the Caesarean section rate and an increased incidence of respiratory distress syndrome in the newborn. In the late 1960s and early 1970s amniocentesis to assess fetal pulmonary maturity by measuring the lecithin-sphingomyelin ratio and phosphatidylglycerol concentrations became routine in diabetic pregnancy [28]. As a result delivery was frequently delayed until 38 weeks with improved outcomes. A study by Robert et al. [29] published in 1976 indicated that the major causes of the respiratory distress syndrome in the offspring of diabetic women were delivery before 39 weeks and caesarean section. This led to the

Management of Delivery

281

recommendation that delivery should ideally occur after 39 weeks’ gestation and preferably without recourse to caesarean section [30]. In 1984 Murphy et al. [31] advocated delivery at 40 weeks or after. They were able to demonstrate an increase in mean gestational age from 37.4 to 39.4 weeks with this policy and described a reduced perinatal morbidity rate and an increase in vaginal delivery from 14.3 to 37.8%. There was, however, a significant increase in mean birth weight from 3,090 to 3,650 g and one avoidable stillbirth in the 44 women delivered after 40 weeks compared with none in the 35 women with earlier delivery. There is therefore little evidence to support the continuation of a type 1 diabetic pregnancy beyond 40 weeks and provided diabetic control has been good and fetal macrosomia avoided, induction of labour and delivery at 39 weeks appears to provide the best compromise of achieving a vaginal delivery and minimizing the risk of neonatal complications due to immaturity. While maternal diabetes alone is not an indication for caesarean section each case must be judged on its own merits. When the estimated fetal weight exceeds 4,250 g or there is ultrasound evidence of a significant difference in the ratio of the abdominal and head measurements delivery by elective caesarean section should be considered. Vaginal delivery may be attempted in diabetic women who have had a single caesarean section for a non-recurrent reason. The vaginal delivery rate in this group, however, appears to be much lower than that in non-diabetic women [32]. The decision to attempt vaginal birth after caesarean section must be taken by an experienced obstetrician and should only be considered in patients with good diabetic control and a normally grown fetus that can be managed in a unit with facilities for continuous fetal monitoring and emergency caesarean section.

Management of Diabetes during Labour

Several studies have shown that hypergylcaemia during labour predisposes to hypoglyceamia in the neonate [33, 34]. It is therefore important to maintain normoglycaemia during labour with blood glucose levels ideally between 4 and 6 mmol/l. A midwife experienced in the care of the diabetic patient should be in attendance. An intravenous insulin and dextrose regime should be commenced using an intravenous infusion of 10% dextrose administered at a rate of 10 g per hour. The insulin infusion should be given by means of a syringe pump using a 60-ml syringe containing 6 units of short-acting human insulin in normal saline. There is generally no need to add potassium to the infusion unless labour

Miller/Gillmer

282

Table 1. Insulin in labour Blood glucose mmol/l

Infusion rate ml/h

Insulin dose units/h

⬍4.0 4.0–6.0 ⬎6.0

5.0 10.0 20.0

0.5 1.0 2.0

is prolonged. The patient should remain fasted during labour and the insulin infusion should be commenced at a rate dependent on the initial capillary blood glucose concentration as shown in table 1. The dextrose infusion must be maintained at a rate of 10 g per hour throughout labour. The capillary blood glucose must thereafter be checked hourly taking the blood sample from the opposite arm to that into which the glucose and insulin are infused. If the blood glucose concentration is subsequently between 4 and 6 mmol/l, the infusion rate should not be changed. If the blood glucose concentration exceeds 6 mmol/l, then the infusion rate should be doubled. If the infusion rate is less than 4 mmol/l, the infusion rate should be halved and the blood glucose concentration checked after a further 30 min. The amount of insulin required during labour is often quite low. Jovanovic and Peterson [35] have reported that during the first stage of labour the insulin requirements may fall to zero but return to normal during the second stage. The insulin infusion rate must be halved immediately after delivery to prevent postnatal hypoglycaemia as there is a rapid increase in insulin sensitivity after separation of the placenta.

Induction of Labour

It is frequently necessary to induce labour in diabetic women [36]. Admission must be arranged the day before the planned induction for assessment of blood glucose control, blood pressure, and renal function. Fetal size and well-being should be assessed by means of an ultrasound examination including a biophysical profile. If the cervix is favourable for induction by artificial rupture of membranes, this should be performed early the following morning and the insulin and dextrose infusions commenced at this time. If the cervix is unfavourable for artificial rupture of the membranes, normal diet and subcutaneous insulin

Management of Delivery

283

injections should be maintained and a vaginal prostaglandin E2 regimen commenced to ‘ripen’ the cervix. The insulin infusion can be delayed until artificial rupture of the membranes is performed or labour becomes established. After this the fasting state must be maintained and the insulin and dextrose infusion regimen commenced. If an oxytocin infusion is required during or after labour this should be made up with normal saline.

Fetal Monitoring

It has long been recognized that there is an increased risk of fetal distress and intrapartum fetal death in diabetic pregnancy [27]. Studies by Salvesen et al. [37] have shown that fetal blood samples obtained by cordocentesis during pregnancy demonstrate a significant degree of acidaemia and lacticaemia in the absence of hypoxaemia. This finding may offer an explanation for the unexplained stillbirths observed in late diabetic pregnancy. Further studies by these workers and observations in pregnant sheep [38] suggest that the observed fetal acidaemia is most likely to be due to enhanced fetal glucose metabolism and lactate accumulation. Furthermore, animal studies suggest that while minor degrees of hyperglycaemia are associated with acidaemia in the absence of hypoxaemia, even mild fetal hypoxaemia together with mild hyperglycaemia may lead to severe acidosis and death [39]. It is therefore vital that not only should normoglycaemia be maintained throughout labour but the fetal heart rate and maternal uterine contractions should be monitored by continuous electronic cardiotocography. Measurement of the fetal blood pH by scalp blood sampling is indicated as in non-diabetic women. The contraction rate, dilation of the cervix and descent of the fetal head should be recorded on a partogram. Progress of labour can therefore be monitored and signs of obstructed labour noted and acted upon with early recourse to caesarean section if indicated.

Analgesia

Adequate pain relief during labour is vital in the diabetic patient [40]. Pain stimulates the release of catecholamines, which in turn causes hyperglycaemia. Good control of pain therefore improves control of blood glucose concentrations. The full range of obstetric analgesia should be made available to the diabetic patient. An epidural anaesthetic is especially useful.

Miller/Gillmer

284

Instrumental Delivery

Instrumental delivery is neither routinely indicated nor contra-indicated in the diabetic patient. However, because of the increased incidence of fetal macrosomia and risk of birth trauma due to cephalopelvic disproportion and shoulder dystocia, the patient must be assessed by an experienced obstetrician before commencing either a vacuum extraction or forceps procedure. A paediatrician should ideally be in attendance for instrumental deliveries in these women [41]. Diabetic women with mild proliferative retinopathy may be allowed to push during the second stage of labour, but if there is severe disease then an elective instrumental delivery may be preferable. This possibility should be discussed with the patient during her attendance at the antenatal clinic.

Caesarean Section

The caesarean section rate of type 1 diabetics remains high and in some units approaches 60% [3, 42]. The insulin and dextrose infusion regimen described for labour is also appropriate for elective and emergency caesarean section. The elective caesarean section should be preformed first on a morning list, as this is more likely to enable the patient to return to oral intake and her normal insulin regimen by the following morning. A regional anaesthetic such as a spinal or epidural should be used unless there is a contra-indication. This enables the patient to resume her normal diet and insulin regimen more quickly and will also reduce postoperative morbidity. A paediatrician should ideally be in attendance for both elective and emergency Caesarean sections performed on diabetic women [41]. Prophylactic antibiotics should be administered for both elective and emergency caesarean section [43]. Prophylaxis for deep vein thrombosis should also be given to obese women and following emergency caesarean section.

Postpartum

After delivery the insulin dose should be reduced to that used prior to the pregnancy as normal insulin sensitivity returns within 24 h of delivery. Breastfeeding should be encouraged but the insulin dose may need to be reduced by approximately 30% [30]. Appropriate contraceptive advice should be given before the patient is discharged from hospital.

Management of Delivery

285

References 1

2 3

4

5 6 7 8 9 10 11

12

13 14

15 16

17 18 19

20 21 22

Bennewitz HG: De diabete mellito, gravidatatis symptomate; MD thesis University of Berlin, 1824 (translated into English, deposited at the Wellcome Museum of the History of Medicine, London, 1987). Peel J, Oakley WG: Transactions of the 12th British Congress of Obstetrics and Gynaecology. Royal College Obstetrics and Gynaecology Press, London, 1949, p 161. Hawthorne G, Robson S, Ryall EA, Sen D, Roberts SH, Platt MPW: Prospective population based survey of outcome of pregnancy in diabetic women: Results of the Northern Diabetic Pregnancy Audit, 1994. BMJ 1997;315:279–281. Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah POD, Platt MJ, Stanisstreet M, van Velzen D, Walkinshaw S: Outcomes of pregnancy in insulin dependent diabetic women: Results of a five year population cohort study. BMJ 1997;315:275–278. Boulvain M, Stan C, Irion O: Elective delivery in diabetic pregnant women. Cochrane Database Syst Rev 2002;2:CD001997. Miodovnik M, Peros N, Holroyde JC, Siddiqi TA: Treatment of premature labour in insulin-dependent diabetic women. Obstet Gynecol 1985;65:621–627. Crowley P: Prophylactic corticosteroids for preterm birth (Cochrane Review). The Cochrane Library, Issue 1, 2004. Borberg C, Gillmer MD, Beard RW, Oakley NW: Metabolic effects of beta-sympathomimetic drugs and dexamethasone in normal and diabetic pregnancy. Br J Obstet Gynaecol 1978;85/3:184–189. Barnett AH, Stubbs SM, Mander AM: Management of premature labour in diabetic pregnancy. Diabetologia 1980;18:365–368. Tocolytic drugs for women in preterm labour. RCOG Clinical Guideline N0.1 (B), Oct 2002. Coomarasamy A, Knox EM, Gee H, Song F, Khan KS: Effectiveness of nifedipine versus atosiban for tocolysis in preterm labour: A meta-analysis with an indirect comparison of randomised clinical trials. BJOG 2003;110:1045–1049. Mittendorf R, Dambrosia J, Pryde PG, Lee KS, Gianopoulos JG, Besinger RE, Tomich PG: Association between the use of antenatal magnesium sulfate in preterm labor and adverse health outcomes in infants. Am J Obstet Gynecol 2002;186:1111–1118. Newnham JP, Moss TJ, Nitos I, Sloboda DM: Antenatal corticosteroids: The good, the bad and the unknown. Curr Opin Obstet Gynecol 2002;14:607–612. Hanson U, Persson B: Outcome of pregnancies complicated by type-1 insulin dependent diabetes in Sweden: Acute pregnancy complications, neonatal mortality and morbidity. Am J Perinatol 1993;4:330–333. Evers IM, De Valk HW, Visser GHA: Macrosomia despite good glycemic control in type-1 diabetic pregnancy; results of a nationwide prospective study. Diabetologia 2002;45:1484–1489. Langer O, Berkus MD, Huff RW, Samueloff A: Shoulder dystocia: Should the fetus weighing greater than or equal to 4000 grams be delivered by cesarean section? Am J Obstet Gynecol 1991;165:831–837. McFarland MB, Trylovich CG, Langer O: Anthropometric differences in macrosomic infants of diabetic and nondiabetic mothers. J Matern Fetal Med 1998;7/6:292–295. Kastler B, Gangi A, Mathelin C, Germain P, Arhan JM, Treisser A, Dietemann JL, Wackenheim A: Fetal shoulder measurements with MRI. J Comput Assist Tomogr 1993;17:777–780. Tukeva TA, Salmi H, Poutanen VP, Karjalainen PT, Hytinantti T, Paavonen J, Teramo KA, Aronen HJ: Fetal shoulder measurements by fast and ultrafast MRI techniques. J Magn Reson Imaging 2001;13:938–942. Best G, Pressman EK: Ultrasonographic prediction of birth weight in diabetic pregnancies. Obstet Gynecol 2002;99:740–744. Jazayeri A, Heffron JA, Phillips R, Spellacy WN: Macrosomia prediction using ultrasound fetal abdominal circumference of 35 centimeters or more. Obstet Gynecol 1999;93:523–526. Gilby JR, Williams MC, Spellacy WN: Fetal abdominal measurements of 35 and 38 cm as predictors of macrosomia. A risk factor for shoulder dystocia. J Reprod Med 2000;45:936–938.

Miller/Gillmer

286

23 24 25

26 27 28 29 30 31 32 33

34 35 36 37 38 39 40 41

42 43

Conway DL, Langer O: Elective delivery of infants with macrosomia in diabetic women: Reduced shoulder dystocia versus increased cesarean deliveries. Am J Obstet Gynecol 1998;178:922–925. Cohen B, Penning S, Major C, Ansley D, Porto M, Garite T: Sonographic prediction of shoulder dystocia in infants of diabetic mothers. Obstet Gynecol 1996;88:10–13. Cohen BF, Penning S, Ansley D, Porto M, Garite T: The incidence and severity of shoulder dystocia correlates with a sonographic measurement of asymmetry in patients with diabetes. Am J Perinatol 1999;16:197–201. Titus RS: Diabetes in pregnancy from the obstetric point of view. Am J Obstet Gynecol 1937;33: 386–392. Beard R, Lowy C: The British survey of diabetic pregnancies. Br J Obstet Gynaecol 1982;89: 783–785. Hare JW: Diabetes mellitus in pregnancy. Compr Ther 1977;3/11:23–28. Robert MF, Neff RK, Hubbell JP, Taeusch HW, Avery ME: Association between maternal diabetes and the respiratory-distress syndrome in the newborn. N Engl J Med 1976;294:357–360. Gillmer MDG, Bickerton NJ: Advances in the management of diabetes in pregnancy: Success through simplicity. Recent Adv Obstet Gynaecol 1994;18:51–78. Murphy J, Peters J, Morris P, Hayes TM, Pearson JF: Conservative management of pregnancy in diabetic women. Br Med J (Clin Res Ed) 1984;288:1203–1205. Blackwell SC, Hassan SS, Wolfe HM, Michaelson J, Berry SM, Sorokin Y: Vaginal birth after cesarean in the diabetic gravida. J Reprod Med 2000;45:987–990. Andersen O, Hestel J, Scholonker L, Kuhl C: Influence of the maternal plasma glucose concentration at delivery on the risk of hypoglycaemia in infants of insulin dependent diabetic mothers. Acta Paediatr Scand 1985;74:268–273. Taylor R, Lee C, Kyne-Grzebalski D, Marshall SM, Davison JM: Clinical outcome of pregnancy in women with type 1 diabetes (1). Obstet Gynecol 2002;99:537–541. Jovanovic L, Peterson CM: Insulin and glucose requirements during the first stage of labor in insulin-dependent diabetic women. Am J Med 1983;75:607–612. Penney GC, Mair G, Pearson DW: Outcomes of pregnancies in women with type 1 diabetes in Scotland: A national population based study. BJOG 2003;110:315–318. Salvesen DR, Brudenell JM, Snijders RJ, Ireland RM, Nicolaides KH: Fetal plasma erythropoietin in pregnancies complicated by maternal diabetes mellitus. Am J Obstet Gynecol 1993;16891/1:88–94. Robillard JE, Sessions C, Kennedy RL, Smith FG: Metabolic effect of constant hypertonic glucose infusion in well oxygenated fetuses. Am J Obstet Gynecol 1978;130:199–203. Shelley HJ, Bassett JM, Milner RDG: Control of carbohydrate metabolism in the fetus and newborn. Br Med Bull 1975;31:37–43. Rees GA, Hayes TM, Pearson JF: Diabetes, pregnancy and anaesthesia. Clin Obstet Gynecol 1982;9:311–332. Brown CJ, Dawson A, Dodds R, Gamsu H, Gillmer M, Hounsome B, Knopfler A, Ostler J, Peacock I, Rothman D, Steel J: Report of the pregnancy and neonatal care group. Diabet Med 1996;13:S43–S53. Jacob J, Pfenninger J: Cesarean deliveries: When is a pediatrican necessary? Obstet Gynecol 1997;89/2:217–220. Smaill F, Hofmeyr GJ: Antibiotic prophylaxis for cesarean section. The Cochrane Library, Issue 1, 2004.

Michael D.G. Gillmer, MD Women’s Centre, John Radcliffe Hospital Headley Way, Oxford, OX3 9DU (UK) Tel. ⫹44 1865 221624, Fax ⫹44 1865 221188, E-Mail [email protected]

Management of Delivery

287

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 288–309

Offspring of Diabetic Pregnancy Bengt Perssona, Ulf J. Erikssonb, Ulf Hansonc a

Department of Women and Child Health, Karolinska Institute, Stockholm, and Department of Medical Cell Biology, Uppsala University and cDepartment of Women and Child Health, Uppsala University Hospital, Uppsala, Sweden b

During recent years we have witnessed a marked improvement in the outcome of pregnancies complicated by type 1 diabetes [1–3]. The decline in perinatal mortality and morbidity rates can to a large extent be attributed to an improved quality of blood glucose control before and during pregnancy [4]. This successful development is dependent on a number of concurrent factors such as active participation of the pregnant mother and her family in the daily management, individualized educational and emotional support, availability of self-monitoring of blood glucose, continuous subcutaneous insulin infusion systems, insulin analogues and facilities for neonatal intensive care [5–8]. Despite the fact that major advances have been made in the clinical care of the pregnant mother with type 1 diabetes and her newborn we are still facing fetal and neonatal complications that may adversely influence the short and/or long-term prognosis for the offspring [9]. Thus the incidence of congenital malformations is 3- to 4-fold greater than in the offspring of nondiabetic mothers [10–13]. Fetal malformation has emerged as one of the leading causes of perinatal death [14]. The critical period of teratogenesis occurs during the first 8 weeks after the last menstrual period [15], i.e. during a period when the woman may not be aware that she has conceived. Preconceptional care is, therefore, an optimal method to improve pregnancy outcome in diabetic women [16–18]. The possibility to attain optimal glycemic control during the periconceptional period may be hampered by the fact that pregnancies are frequently unplanned [7, 19], and that many women have not received preconceptional advice. In addition, the exact cause of the increased rate of congenital malformations in diabetic pregnancy is still unclear [20]. To further elucidate the pathogenesis of malformation – a subject of intense research both experimental and clinical – is of great importance in order to provide effective preventive measures.

Another challenge is represented by the remarkable increase in the rate of fetal macrosomia seen in many centers [5, 21–24]. Insulin is an important modulator of fetal growth. The pathophysiology of fetal macrosomia in pregnancies complicated by diabetes is attributed to fetal hyperinsulinism. According to the Pedersen hypothesis [25] fetal hyperinsulinism is the consequence of chronic maternal hyperglycemia. Today most pregnant women with type 1 diabetes are subjected to a very rigid control of blood glucose. Therefore, the rising incidence of macrosomia is a puzzling and unexpected observation [26–28]. Why is this the case [29, 30], and is it possible to prevent it? The extent to which gestational diabetes mellitus (GDM) – defined as ‘carbohydrate intolerance of different severity with onset or first recognition during pregnancy’ – will adversely influence the perinatal outcome is still a matter of debate. GDM is characteristically associated with fetal macrosomia and sometimes difficult delivery though less frequently with severe neonatal morbidity. More serious neonatal complications including congenital malformations may occur in subgroups of GDM mothers who most likely entered pregnancy with undiagnosed type 2 diabetes [31]. There is general agreement that GDM represents a prediabetic condition. The subsequent development of type 2 diabetes in a great number of women with previous GDM adds to the worldwide epidemic of the disease [32]. Studies of populations with a very high incidence of type 2 diabetes have demonstrated that exposure of the fetus to GDM paves the way for early development of obesity, impaired glucose tolerance and type 2 diabetes in the genetically susceptible individuals [33, 34]. These findings lend support to the expression: ‘diabetes begets diabetes’. A recent estimate of the lifetime risk of developing diabetes for individuals born in the US in 2000 is more than 30% [35]. These alarming observations call for preventive efforts. This presentation focuses particularly on questions raised above but also on some controversies related to size at birth, neonatal hypoglycemia and longterm follow-up.

Embryopathy

Clinical Aspects Type 1 Diabetes. The first studies relating HbA1c to an increased rate of malformation demonstrated a marked increase when HbA1c exceeded a threshold value of 8–10 SD above the mean for control women [36–40]. These data supported the importance of a planned pregnancy but could be interpreted as suggesting that malformations could be avoided by keeping the HbA1c value below this threshold. However, more recent data have provided strong evidence

Offspring of Diabetic Pregnancy

289

that even a slightly raised level of HbA1c in early pregnancy is associated with an increased risk of malformations and that this risk increases with increasing HbA1c values [41]. This finding underlines the importance of preconceptional counseling to all women with type 1 diabetes. The incidence rate of congenital malformations is still increased compared to the background population [10–13, 31, 42]. The types of major malformation are with few exceptions – such as the caudal regression syndrome – mainly the same as in the background population. Ultrasonic examination performed at around 18 weeks of gestation can detect malformations with great accuracy – in particular those affecting the heart. Programs using ultrasonic examination and determination of ␣-fetoprotein as a marker of malformations in the central nervous system (CNS) have made it possible to detect antenatally more than 70% of major malformations in type 1 diabetic women [43]. Since the risk of malformation is also increased in women with HbA1c close to the normal range, all women with type 1 diabetes should be offered the available prenatal diagnostic techniques. Type 2 Diabetes. The rate of congenital malformations in type 2 diabetic pregnancy is comparable to that found in offspring of type 1 diabetic mothers [31, 44–46]. The use of oral hypoglycemic agents (OHA) during pregnancy is usually contraindicated despite limited information as to their possible teratogenic effects [47–49]. OHA can pass the placenta and enhance fetal insulin production with potential adverse consequences. Available data suggest that the increased malformation rate in type 2 diabetes is related to the degree of glycemic control and not to the use of OHA. Gestational Diabetes mellitus. Usually glycemic control is normal in early pregnancy and the rate of congenital malformation is not increased [50]. However, an elevated rate of malformation has been recorded in a subset of GDM patients, in whom type 2 diabetes was present but unrecognized before pregnancy [10]. Experimental Aspects: Genetic and Environmental Factors Studies of diabetic embryopathy have been performed in vivo and in vitro mainly using rodent models. Most of the developmental effects of diabetic pregnancy in human offspring have been mirrored in the different experimental models, such as perturbed maternal and embryonic metabolism, growth disturbances and malformations in several organ systems, e.g. skeletal and cardiac anomalies. Experimental studies have suggested that the major teratogen in diabetic pregnancy is hyperglycemia [51–54] (fig. 1), although other diabetes-related factors may also influence the fetal outcome, e.g. increased levels of ketone bodies [55–59], triglycerides [60, 61], and branched chain amino acids [61, 62]. Several teratological pathways in the embryonic tissues have emerged from the research

Persson/Eriksson/Hanson

290

a

b

c Fig. 1. Rat embryos cultured in low (a) and increased (b, c) glucose concentrations. They show no malformations (a), a minor malformation (open neural pore; b), and severe malformations (open neural tube and malrotation; c).

efforts, such as alterations in the metabolism of inositol [63–68], arachidonic acid/prostaglandins [63, 69, 70] and reactive oxygen species [62, 71–73]. The embryonic formation of sorbitol [64, 65, 67, 74, 75], glycated proteins [76–78], and the maternal and fetal genotypes [79–84] are also expected to influence the complex teratological events in diabetic pregnancy. The degree of severity of the diabetic state – e.g. estimated as glucose concentration in maternal blood [85] or glucose level in the culture medium of embryos – is also a major determinant of embryonic and fetal damage. The teratogenic period of a diabetic pregnancy is early during the first trimester [60, 86], up to gestational weeks 7–8 [15], and there is also a varied susceptibility to the teratogenicity of a diabetic pregnancy based on genetic predisposition [79–84]. Researchers are at present trying to identify genes involved in the susceptibility and/or resistance to diabetic embryopathy in different rodent models, where susceptible strains are compared to nonsusceptible strains in gene linkage analyses. One gene, which has been identified as involved in the pathogenesis of neural tube defects (NTDs) in offspring of diabetic mice, is Pax-3

Offspring of Diabetic Pregnancy

291

[87–90]. A diabetic state in the pregnant mouse causes reduced expression of Pax-3 [87] and NTD in the offspring [85]; the two latter events may be causally related since Pax-3 downregulation causes NTD in Pax-3-mutated Splotch mice [91, 92] and humans [93, 94]. Another gene, possibly involved in diabetic embryopathy, is vascular endothelial growth factor A (VEGF-A), since it has been shown that a decreased amount of this gene product in the yolk sac results from exposure to high ambient glucose concentration [95]. Furthermore, addition of exogenous rVEGF-A165 within a defined concentration range blunts the hyperglycemiainduced vasculopathy in the conceptus [96]. Based upon these findings it has been hypothesized that the cardiovascular malformations found in diabetic pregnancy may be, at least in part, due to alterations in the expression and phosphorylation state of adhesion molecules such as platelet endothelial growth factor-1 [97] and the expression of growth factors such as VEGF-A [98]. The experimental studies of etiological mechanism of diabetes-induced dysmorphogenesis also involve treatment attempts with the aim to prevent diabetic embryopathy and fetopathy. The first agent to be identified was arachidonic acid. The addition of arachidonic acid to the culture medium was shown to block the embryonic dysmorphogenesis elicited by high glucose concentration [99], a finding that has been repeated [100], and expanded [101] in subsequent studies. Intraperitoneal injections of arachidonic acid to pregnant diabetic rats diminished the rate of neural tube damage [99], as did enriching the diet of the pregnant diabetic rats with arachidonic acid [102, 103], thereby indicating a disturbance in the arachidonic acid cascade as a consequence of a diabetic environment [104]. The addition of PGE2 to the culture medium also blocks glucose-induced teratogenicity in vitro [63, 101], as well as maldevelopment of embryos cultured in diabetic serum [105]. Measurements of PGE2 have indicated that this prostaglandin is decreased in embryos of diabetic rodents during the period of neural tube closure [106, 107], in high glucosecultured embryos [107], as well as in the yolk sac of embryos of diabetic women [108]. The second anti-teratogenic agent identified was inositol. In embryos subjected to high glucose concentration in vitro, the inositol levels decrease due to impaired uptake [68], yielding an embryonic deficiency of inositol [65, 67, 109]. Supplementation of inositol to high glucose-cultured embryos [63, 64, 66], or dietary addition to pregnant diabetic rodents [103, 110–112] yields less embryonic maldevelopment. Adding the competitive inhibitor scyllo-inositol to the culture medium induces both inositol deficiency and embryonic dysmorphogenesis of a similar type as the damage caused by high glucose alone [109, 113]. Both the inositol deficiency and the embryo maldevelopment elicited by scyllo-inositol exposure can be diminished by the addition of inositol to the

Persson/Eriksson/Hanson

292

culture medium [109, 113]. These findings identify inositol deficiency as a likely component of diabetic teratogenesis [114]. The third set of anti-teratogenic agents were scavengers of radical oxygen species (ROS). The notion that diabetes is associated with oxidative stress has been suggested by several authors [115–117], and increased markers of oxidative stress have been reported in diabetic patients [118–124]. Increased lipid peroxidation and ROS generation were found in diabetic rats, measured as increased serum F2-isoprostane levels [125], and increased electron spin clearance rate [126]. Both these indicators of oxidative stress were normalized by vitamin E treatment of the diabetic rats [125, 126]. Cyclic voltammetric studies have indicated increased levels of lipid peroxidation in diabetic rats [127], and it has been demonstrated that mitochondria of vascular endothelial cells produce an excess amount of superoxide in response to hyperglycemia [128]. Diminishing this overproduction of ROS via inhibition of the electron transport, by uncoupling oxidative phosphorylation, or by addition of SOD blocked other markers of intracellular imbalance, such as activation of protein kinase C, formation of advanced glycation end products, sorbitol accumulation and NF-␬B activation. Furthermore, embryos subjected to a high glucose concentration show evidence of increased superoxide production, as measured in a Cartesian diver system [129]. By adding scavenging enzymes, e.g. SOD, catalase or glutathione peroxidase, to the culture medium it was first shown that antioxidative compounds protect rat embryos from dysmorphogenesis induced by high glucose concentration in vitro [71]. Teratogenic concentrations of ␤-hydroxybutyrate or the branched chain amino acid analogue ␣-ketoisocaproic acid can be blocked by addition of SOD to the culture medium [62], and addition of SOD or NAC diminishes the dysmorphogenesis caused by diabetic serum [130]. In a study of the early development of cranial neural crest cells, it was shown that high glucose inhibited, and NAC normalized, the migration and proliferation of these cells, and that control cells of nonneural crest origin were not affected by either treatment [131]. These findings have also been confirmed in vivo. Supplementation of antioxidants has been shown to be beneficial for the development of embryos in a diabetes-like environment in vivo and in vitro. Dietary addition of butylated hydroxytoluene [73], vitamin E [103, 132–138], vitamin C [139], combinations of vitamins E and C [80], glutathione ester [140], and lipoic acid [141] diminish perturbed embryonic development in vivo. Indeed, antioxidative treatment of hyperglycemic mice restores Pax-3 gene expression and diminishes the rate of NTD in the offspring [138]. The findings of diminished embryopathy by antioxidative treatment may also have therapeutic implications for human diabetic pregnancy, in particular since signs of enhanced oxidative stress have been reported in pregnant diabetic women [142]. It would be highly valuable to attempt to diminish

Offspring of Diabetic Pregnancy

293

developmental damage in diabetic pregnancy by clinical trials using antioxidative agents [143].

Diabetes Fetopathy

Perinatal Mortality Type 1 Diabetes. Studies from countries with excellent national figures for perinatal outcome have reported stillbirth rates that are 4–5 times higher than normal [11, 144–146]. Despite the lack of randomized studies overwhelming evidence suggests that improved blood glucose control is the main reason for the decrease in perinatal mortality rate in diabetic pregnancies. The exact reason behind intrauterine fetal death is not always clear. Strong evidence supports that hypoxia is an important pathogenic factor to explain the association between the increased rate of stillbirths and poor glycemic control. Animal experiments have shown that hyperglycemia induces hypoxemia in fetal sheep due to an increased fetal oxygen demand [147]. Chronic maternal hyperglycemia and accompanying chronic fetal hypoxia are associated with elevated erythropoietin levels in amniotic fluid and umbilical cord blood in infants of diabetic mothers [148]. It is important to note that concomitant hyperglycemia aggravates the fetal cerebral tissue damage due to hypoxia. Poor glucose control is also accompanied by increased interventricular septal thickness [149]. Cardiac hypertrophy could thus be an additional contributing factor to sudden intrauterine death. One important contributing factor to stillbirths is major malformation. The prevalence of malformations is still 2–3 times above normal [11]. Since fetal hypoxia may be an important factor routine supervision of fetal heart rate (CTG) is recommended. There is no evidence to support the routine use of Doppler measurements in pregnancies complicated by diabetes. The most important preventive measure is trying to obtain blood glucose values close to normal during the periconceptional period and throughout pregnancy. Type 2 Diabetes. Data regarding type 2 diabetes and pregnancy are scare. Available results suggest that perinatal outcome is close to that seen in type 1 diabetic pregnancies [46]. Gestational Diabetes mellitus. Some studies have reported increased perinatal morality rates. This can most likely be attributed to the inclusion of GDM patients with prior unrecognized type 2 diabetes. Newborn infants of diabetic mothers are traditionally classified as high-risk babies. This is understandable when considering the perinatal consequences of acute complications during pregnancy like preeclampsia or hypertension, severe diabetes angiopathy and above all variations in glycemic control during pregnancy. Perinatal outcome is modulated by the maternal metabolic

Persson/Eriksson/Hanson

294

control. Chronic fetal hyperglycemia is associated with fetal hypoxia and hyperinsulinism that significantly influence the fetal supply line of nutrients as well as oxygen. The fetal supply line may vary considerably. At one extreme, a markedly increased availability of nutrients to the fetus leads to accelerated fetal growth as may be seen in patients with a short duration of diabetes. At the other end of the clinical spectrum, a markedly reduced supply line may lead to fetal growth retardation, which is a frequent complication in mothers with severe diabetes angiopathy, in particular nephropathy. Size at Birth The interaction between genetic, placental, maternal and fetal growthpromoting factors is still not completely understood. Exaggerated secretion and function of insulin IGF-1 and IGF-2 and their carrier proteins lead to excessive transfer of nutrients from the mother and enhanced somatic growth. The classic diabetes fetopathy is characterized by macrosomia, an increased amount of adipose tissue and glycogen and a plethoric, cushingoid appearance. Macrosomia is often defined as a birth weight ⬎90th percentile for gestational age and sex. Others prefer a more stringent definition, i.e. a birth weight ⬎2 SD above normal. Strict glycemic control during pregnancy by intensive insulin therapy may normalize the fetal supply line, and the infant’s size and appearance at birth. This favorable outcome can to a large extent be attributed to a reduction of adipose tissue mass as indicated by significant relationships between the maternal level of glucose and the skinfold thickness and the average adipose tissue cell diameter. More recent studies have demonstrated weak but significant relationships between birth weight and either postprandial or fasting glucose values between gestational weeks 29–32 and 27 and 32, respectively [5, 150]. This period of gestation coincides with the time of rapid fetal growth as documented by serial ultrasound measurements. In our series of 113 consecutive pregnancies all mothers monitored their blood glucose values at least 6 times daily from around week 6 until delivery. Our analyses showed that both the maternal pre-pregnancy body weight and the fasting blood glucose levels between gestational weeks 27–29 were independently associated with the infant’s size at birth. These variables could, however, only explain around 12% of the variation in relative birth weight [5]. In recent years many centers have recorded a markedly increased rate of fetal macrosomia despite rigid control of maternal glucose during pregnancy [5, 150, 151]. The fact that around 30% of all infants are macrosomic is a puzzling observation. In Sweden the incidence of macrosomia increased from 20% (1983–1985) to 33.5% (1991–1996) and at the same time perinatal morality dropped from 3.1 to 2.4% [152, 153]. There is no reason to believe that this unexpected development could be attributed to a worsening of the quality of

Offspring of Diabetic Pregnancy

295

glycemic control in view of the parallel and continuous drop in perinatal mortality over the same time period. We hypothesized that fetal growth-restraining factors were more prevalent in the past: (1) It is well known that birth weight decreases progressively with the severity of maternal diabetes. Thus one factor that could retard fetal growth is the presence of diabetes microangiopathy, in particular overt and incipient nephropathy. These complications used to occur more frequently a few decades ago. In addition they are closely associated with acute pregnancy complications like hypertension and preeclampsia, which are also known to impair fetal growth. This view is supported by the significant decrease in the incidence of preeclampsia/hypertension from 20.6 to 13.8% in Sweden between the periods of 1983–1985 and 1991–1996 [152, 153]. (2) Another potential growth-restraining factor is poor glycemic control in early pregnancy [154]. Early intrauterine growth delay as assessed by ultrasound during the first trimester of pregnancy is associated with elevated HbA1c values, subsequent poor growth and lower birth weight than in offspring who were of normal size in early pregnancy. The incidence of early growth delay has decreased significantly with increasing awareness of the importance of good glycemic control during the periconceptional period. Excess levels of insulin and other growth factors like IGF-1 and IGF-2 accelerate fetal growth. Macrosomia seen in unusual conditions like Beckwith-Wiedemann and islet cell dysregulation syndromes is associated with hyperinsulinemia and neonatal hypoglycemia. Such cases as well as fetal overgrowth caused by experimentally induced hyperinsulinemia illustrate that excess insulin per se enhances the transfer of glucose and other nutrients from the mother to the fetus, although the mother is normoglycemic. It cannot be excluded that an exaggerated ␤-cell function may be established at an early stage of gestation due to transient elevation of substrates (glucose, amino acids) or other still undefined (genetic?) factors. It is of interest in this context that elevated amniotic fluid insulin levels have been recorded already at 14–20 weeks’ gestation in women who subsequently developed GDM [155]. Hyperinsulinemia may persist although near normoglycemia has been achieved during the remaining part of gestation explaining cases of unexpected macrosomia despite well-controlled diabetes. To this should be added that increased activity of the placental glucose transporter GLUT1 further enhances the glucose transfer to the fetus [156]. The typical diabetes fetopathy is characterized by selective organomegaly that in particular includes adipose tissue, heart and liver. The enlargement of these organs contributes to a disharmonious body composition that is not easily revealed by measurement of body weight and length. Information can be obtained by using noninvasive techniques such as echocardiography for the

Persson/Eriksson/Hanson

296

measurement of the thickness of the interventricular septum and skinfold thickness for assessment of the amount of adipose tissue. Previous studies have shown a prevalence of asymmetric septal hypertrophy of around 30% [157]. Recently, it was demonstrated that even appropriate-for-gestational age infants of gestational diabetic mothers had significantly greater amounts of body fat than controls of comparable size at birth [158]. Obviously birth weight/length alone is an incomplete index by which to evaluate the impact of the achieved level of metabolic control during pregnancy.

Birth Injury

Pregnancies complicated by type 1, type 2 and GDM are characterized by an increased rate of large-for-gestational age infants and as a consequence an increased rate of difficult and traumatic deliveries. Shoulder dystocia may result in brachial plexus injuries (like Erb’s palsy), clavicular and humeral fractures. High birth weight is a major predictive factor for Erb’s palsy. A contributing reason is the larger shoulder circumference frequently seen in infants of diabetic mothers, which represents a risk factor that is independent of the absolute birth weight. Much effort has been focused on identifying the macrosomic fetus in order to decide the appropriate mode of delivery. Proposed management programs are often based on register studies with exact information on the infant’s birth weight. Unfortunately ultrasound measurement of fetal weight is inexact. Hence most of these programs fail to accurately predict fetal size [159]. In addition it is important to recall that even if the rate of Erb’s palsy is elevated in birth weight classes of 4,500 g most affected infants have a birth weight below this limit. A recent Cochrane review concluded that there is little evidence to support either elective delivery or expectant management at term in pregnancies complicated by type 1 diabetes [160].

Neonatal Morbidity

Metabolic Abnormalities Most newborns of diabetic mothers (except those who are intrauterine growth retarded or those who have suffered severe intrauterine asphyxia) have considerable stores of glycogen in the liver or elsewhere as well as triglycerides in adipose tissue. Still plasma concentrations of glucose, free fatty acids (FFA) and ketone bodies are suppressed during the first hours after birth. These metabolic alterations are transient and are usually attributed to neonatal

Offspring of Diabetic Pregnancy

297

hyperinsulinism. At 60 min after birth we recorded the lowest mean plasma glucose value (1.2 vs. 2.8 mmol/l in controls); at 120 min values had increased (1.6 vs. 3.3 mmol/l) [161]. Profound hypoglycemia during the first hours after birth is almost invariably asymptomatic and is not accompanied by intellectual impairment later on in life [162]. Clinical signs and symptoms, which often accompany similarly low glucose values at a later age, may be associated with neuropsychological deficits. It is unlikely that this different response to hypoglycemia is due to a different energy demand of the CNS. Instead a more plausible explanation is an increased availability of alternate substrates like lactate [163], and possibly also local stores of glycogen in astrocytes in CNS in the immediate neonatal period. The physiological drop of blood glucose after birth is accompanied by a fall of insulin, a rise of glucagon and catecholamines concomitant with a marked increase of TSH. These hormonal responses favor hepatic glucose production, lipolysis and enhanced oxidation of FFA and ketogenesis. The FFA and ketone body levels remain lower than normal during the first couple of hours after birth. On the other hand, plasma glycerol concentrations increase significantly and are not different from controls. Lipolysis in offspring of diabetic mothers is not different from normal as also recently demonstrated using stable isotope techniques [164]. The pattern of response of high glycerol and low FFA concentrations seems difficult to reconcile with neonatal hyperinsulinism and the strong antilipolytic effect of insulin. In the newborn as opposed to the adult individual lipolysis is strongly activated by TSH [165]. We proposed that the low FFA and ketone body levels despite enhanced lipolysis could be explained by an increased rate of reesterification of liberated fatty acids with ␣-glycerol phosphate derived from glycogen stored within adipose tissue [161]. Neonatal Hypoglycemia Newborns of diabetic mothers are at increased risk of developing hypoglycemia and its occurrence is considered an important outcome variable. This topic has been surrounded by controversies related to methods used, the definition of hypoglycemia, monitoring and long-term consequences. With regard to methodology some issues should be recognized. Measurements of capillary and arterial samples are approximately 10% higher than of venous blood samples. Plasma values are approximately 15% higher than venous blood samples. The rate of glycolysis in red blood cells from newborns is significantly higher than in adults. Samples must be collected in tubes containing sodium fluoride (16 mg/ml) and kept in ice water until further processing in order to prevent falsely low glucose concentrations.

Persson/Eriksson/Hanson

298

Test strips and reflectance meters have caused much confusion. Results obtained using these techniques are unreliable and cannot be used for diagnostic purposes. Hypoglycemia is defined according to the following three factors: (1) Symptoms: One useful definition is Whipple’s triad: (i) low blood glucose, (ii) clinical symptoms and (iii) symptoms disappear following glucose administration. (2) Statistical data: Any value outside –2 or –3 SD of the normal mean for age. Recent data suggest that a blood glucose value less than 2.6 mmol/l is rarely recorded in term or preterm infants during the first 24–48 h after birth [166]. (3) Neurological outcome: Neurophysiological abnormalities – without clinical symptoms of hypoglycemia – have been recorded in a small group of infants when blood glucose fell below 2.6 mmol/l. A follow-up of large groups of preterm infants showed an association between blood glucose values below 2.6 mmol/l for a period of 3 days or more and poor neurological outcome, i.e. developmental delay, low scores for arithmetic and motor performance [167]. Prevention of Hypoglycemia Maternal blood glucose should be kept as close to normal as possible during pregnancy, and above all a glucose level around 4.5–5.5 mmol/l should be maintained during labor and delivery. The infant’s energy expenditure should be reduced by preventing unnecessary heat loss immediately after birth; the infant should be kept in the thermoneutral zone. Early feeding (enteral or intravenous) should be initiated, aiming at a full caloric intake at 3–4 days of age.

Long-Term Prognosis

Future Risk of Developing Diabetes and/or Obesity Offspring of parents with either type 1 or type 2 diabetes or GDM have an increased genetic susceptibility to developing impaired glucose tolerance and/or diabetes. The underlying genetic predisposition to diabetes and obesity of the offspring differs with the type of maternal diabetes. Hence the follow-up results will be considered separately. Type 1 Diabetes The incidence of type 1 diabetes is 1.3 and 6.1% in offspring of diabetic mothers and fathers, respectively [168]. The reason for this 5-fold difference in the incidence is unclear; it has been speculated that the diabetic environment protects the fetus from developing type 1 diabetes. Other results suggest that

Offspring of Diabetic Pregnancy

299

children born to mothers above 25 years of age have a significantly lower risk of developing type 1 diabetes by the age of 20 than those born to younger mothers [169]. There is no apparent explanation for this risk reduction that may have important implication for the planning of a pregnancy. Individuals at risk of developing type 1 diabetes can be identified by analysis of genetic (HLA types) and immunological (autoantibodies) markers. A number of studies have been performed to prevent type 1 diabetes in susceptible subjects. So far several large intervention studies (using e.g. parenteral or oral insulin, nicotinamide) conducted at the time of onset of diabetes or at the prediabetes stage have not been successful. Obesity. The incidence of obesity among children and adolescents increases worldwide as a consequence of an altered lifestyle with a low level of physical activity and/or too high an energy intake. Obesity in offspring of mothers with type 1 diabetes has been recorded by some authors [170, 171] but not all [172]. Available data suggest that heavy newborns tend to become big children during adolescence. It has been hypothesized that fetal exposure to maternal diabetes could lead to prenatal hypercellularity of adipose tissue and subsequent development. This view is, however, not supported by the observation that significantly oversized newborns of mothers with type 1 diabetes have normal values for body weight, total number of fat cells and total body fat at 18–26 years of age [173]. Neuropsychological Development. The high incidence of major cerebral handicaps that was seen in the past could be related to severe diabetic angiopathy in the mother, premature delivery and low birth weight. The outlook for the offspring has changed dramatically as a result of meticulous control of maternal blood glucose throughout pregnancy. Thus several authors have reported normal IQ scores determined between 3 and 12 years [162, 174–176]. Other tests and questionnaires have, however, disclosed minor problems like impaired fine and gross motor function, inattention and hyperactivity when the children reached early school age [176, 177]. Type 2 Diabetes Pima Indians who live in Arizona have an extremely high incidence of type 2 diabetes. Extensive studies of this ethnic group have demonstrated that offspring of mothers with type 2 diabetes – or GDM – run a considerable risk of developing type 2 diabetes themselves. Diabetes prevalence figures of around 20 and 50% at the age of 15–19 and 20–24 years, respectively, have been recorded [33]. Much lower diabetes prevalence among offspring of prediabetic mothers strongly suggests the impact of the diabetic intrauterine environment on the development of diabetes. This interpretation is further supported by the finding of a significantly lower risk of diabetes in siblings born before than

Persson/Eriksson/Hanson

300

after the mother developed diabetes [178]. Intrauterine exposure to diabetes thus carries a much increased risk of future diabetes in genetically susceptible offspring. ‘Diabetes begets diabetes’. The lifetime risk of developing diabetes (type 2, from birth to 80 years) in individuals born in the year 2000 in the United States has recently been estimated to 32.8% in males and 38.5% in females [35]. The estimated risk figures for Hispanics are even higher. It is of great importance to examine whether normalization of maternal blood glucose during pregnancy can reduce the excess risk of diabetes in the offspring. Obesity. The risk of later obesity in the offspring of mothers with GDM or type 2 diabetes is also much increased. The prevalence of obesity is around 70% in children aged 10–14 years born to Pima Indian women with type 2 diabetes as compared to around 30% in children born to prediabetic and nondiabetic mothers [33]. This increased risk of childhood obesity is – as is the case with diabetes – associated with fetal exposure to maternal diabetes. It should be emphasized that other studies have not confirmed an increased prevalence of obesity among offspring to GDM mothers treated with diet alone [179] or diet plus insulin [162]. This difference may be attributed to differences in the severity of maternal diabetes, the level of glycemic control during pregnancy or more likely genetic predisposition to obesity and diabetes.

References 1 2 3 4 5 6 7

8 9

10

Persson B, Hanson U: Insulin dependent diabetes in pregnancy: Impact of maternal blood glucose control on the offspring. J Paediatr Child Health 1993;29:20–23. Schwartz R, Teramo KA: Effects of diabetic pregnancy on the fetus and newborn. Semin Perinatol 2000;24:120–135. Gabbe SG, Graves CR: Management of diabetes mellitus complicating pregnancy. Obstet Gynecol 2003;102:857–868. Vaarasmaki M, Gissler M, Hartikainen AL: A uniform regimen enables decentralized care of diabetic pregnancies. Diabet Med 2001;18:871–876. Persson B, Hanson U: Fetal size at birth in relation to quality of blood glucose control in pregnancies complicated by pregestational diabetes mellitus. Br J Obstet Gynecol 1996;103:427–433. Hadden DR: How to improve prognosis in type 1 diabetic pregnancy. Old problems, new concepts. Diabetes Care 1999;22:B104–108. McElvy SS, Miodovnik M, Rosenn B, Khoury JC, Siddiqi T, Dignan PS, Tsang RC: A focused preconceptional and early pregnancy program in women with type 1 diabetes reduces perinatal mortality and malformation rates to general population levels. J Matern Fetal Med 2000;9: 14–20. Jovanovic L: Medical emergencies in the patient with diabetes during pregnancy. Endocrinol Metab Clin North Am 2000;29:771–787. Hanson U, Persson B, Thunell S: Relationship between haemoglobin A1c in early type 1 (insulindependent) diabetic pregnancy and the occurrence of spontaneous abortion and fetal malformation in Sweden. Diabetologia 1990;33:100–104. Aberg A, Westbom L, Kallen B: Congenital malformations among infants whose mothers had gestational diabetes or preexisting diabetes. Early Hum Dev 2001;61:85–95.

Offspring of Diabetic Pregnancy

301

11

12

13 14 15 16 17 18 19 20 21

22 23 24

25 26

27

28

29 30

31 32

33

Platt MJ, Stanisstreet M, Casson IF, Howard CV, Walkinshaw S, Pennycook S, McKendrick O: St Vincent’s Declaration 10 years on: Outcomes of diabetic pregnancies. Diabet Med 2002;19: 216–220. Queisser-Luft A, Stolz G, Wiesel A, Schlaefer K, Spranger J: Malformations in newborn: Results based on 30,940 infants and fetuses from the Mainz congenital birth defect monitoring system (1990–1998). Arch Gynecol Obstet 2002;266:163–167. Penney GC, Mair G, Pearson DW: Outcomes of pregnancies in women with type 1 diabetes in Scotland: A national population-based study. BJOG 2003;110:315–318. Lowy C: Maternal diabetes and its cell membrane abnormalities as they affect the foetus: Implications for the mother and birth defects. Nutr Health 2001;15:245–250. Mills JL, Baker L, Goldman AS: Malformations in infants of diabetic mothers occur before the seventh gestational week. Implications for treatment. Diabetes 1979;28:292–293. Dunne FP, Brydon P, Smith T, Essex M, Nicholson H, Dunn J: Pre-conception diabetes care in insulin-dependent diabetes mellitus. Q J Med 1999;92:175–176. Ray JG, O’Brien TE, Chan WS: Preconception care and the risk of congenital anomalies in the offspring of women with diabetes mellitus: A meta-analysis. Q J Med 2001;94:435–444. Korenbrot CC, Steinberg A, Bender C, Newberry S: Preconception care: A systematic review. Matern Child Health J 2002;6:75–88. Chia YT, Chua S, Thai AC, Kek LP, Ratnam SS: Congenital abnormalities and pregestational diabetes mellitus in pregnancy. Singapore Med J 1996;37:380–383. Eriksson UJ, Cederberg J, Wentzel P: Congenital malformations in offspring of diabetic mothers – animal and human studies. Rev Endocr Metab Disord 2003;4:79–93. Djelmis J, Blajic J, Bukovic D, Pfeifer D, Ivanisevic M, Kendic S, Votava-Raic A: Glycosylated hemoglobin and fetal growth in normal, gestational and insulin dependent diabetes mellitus pregnancies. Coll Antropol 1997;21:621–629. McAuliffe FM, Foley M, Firth R, Drury I, Stronge JM: Outcome of diabetic pregnancy with spontaneous labour after 38 weeks. Ir J Med Sci 1999;168:160–163. Lindsay RS, Hanson RL, Bennett PH, Knowler WC: Secular trends in birth weight, BMI, and diabetes in the offspring of diabetic mothers. Diabetes Care 2000;23:1249–1254. Jolly MC, Sebire NJ, Harris JP, Regan L, Robinson S: Risk factors for macrosomia and its clinical consequences: A study of 350,311 pregnancies. Eur J Obstet Gynecol Reprod Biol 2003;111: 9–14. Pedersen J: The Pregnant Diabetic and Her Newborn. Problems and Management. Copenhagen, Munksgaard, 1977. Lepercq J, Taupin P, Dubois-Laforgue D, Duranteau L, Lahlou N, Boitard C, Landais P, HauguelDe Mouzon S, Timsit J: Heterogeneity of fetal growth in type 1 diabetic pregnancy. Diabetes Metab 2001;27:339–344. Evers IM, de Valk HW, Mol BW, ter Braak EW, Visser GH: Macrosomia despite good glycaemic control in type I diabetic pregnancy; results of a nationwide study in The Netherlands. Diabetologia 2002;45:1484–1489. Greco P, Vimercati A, Hyett J, Rossi AC, Scioscia M, Giorgino F, Loverro G, Selvaggi L: The ultrasound assessment of adipose tissue deposition in fetuses of ‘well controlled’ insulin-dependent diabetic pregnancies. Diabet Med 2003;20:858–862. Lauszus FF, Klebe JG, Flyvbjerg A: Macrosomia associated with maternal serum insulin-like growth factor-I and -II in diabetic pregnancy. Obstet Gynecol 2001;97:734–741. Desoye G, Korgun ET, Ghaffari-Tabrizi N, Hahn T: Is fetal macrosomia in adequately controlled diabetic women the result of a placental defect? – a hypothesis. J Matern Fetal Neonatal Med 2002;11:258–261. Farrell T, Neale L, Cundy T: Congenital anomalies in the offspring of women with type 1, type 2 and gestational diabetes. Diabet Med 2002;19:322–326. Catalano PM, Kirwan JP, Haugel-de Mouzon S, King J: Gestational diabetes and insulin resistance: Role in short- and long-term implications for mother and fetus. J Nutr 2003;133:1674S– 1683S. Pettitt DJ, Nelson RG, Saad MF, Bennett PH, Knowler WC: Diabetes and obesity in the offspring of Pima Indian women with diabetes during pregnancy. Diabetes Care 1993;16:310–314.

Persson/Eriksson/Hanson

302

34 35 36 37

38 39

40

41 42

43

44 45 46 47 48 49 50

51 52 53 54

55

Pettitt DJ, Knowler WC: Long-term effects of the intrauterine environment, birth weight, and breast-feeding in Pima Indians. Diabetes Care 1998;21:B138–B141. Narayan KM, Boyle JP, Thompson TJ, Sorensen SW, Williamson DF: Lifetime risk for diabetes mellitus in the United States. JAMA 2003;290:1884–1890. Leslie RDG, Pyke DA, John PN, White JM: Hemoglobin A1 in diabetic pregnancy. Lancet 1978;ii:958–959. Miller E, Hare JW, Cloherty JP, Dunn PJ, Gleason RE, Soeldner JS, Kitzmiller JL: Elevated maternal hemoglobin A1c in early pregnancy and major congenital anomalies in infants of diabetic mothers. N Engl J Med 1981;304:1331–1334. Ylinen K, Raivio K, Teramo K: Haemoglobin AIc predicts the perinatal outcome in insulin-dependent diabetic pregnancies. Br J Obstet Gynaecol 1981;88:961–967. Greene MF, Hare JW, Cloherty JP, Benacerraf BR, Soeldner JS: First-trimester hemoglobin A1 and risk for major malformation and spontaneous abortion in diabetic pregnancy. Teratology 1989;39:225–231. Rosenn B, Miodovnik M, Combs CA, Khoury J, Siddiqi TA: Glycemic thresholds for spontaneous abortion and congenital malformations in insulin-dependent diabetes mellitus. Obstet Gynecol 1994;84:515–520. Suhonen L, Hiilesmaa V, Teramo K: Glycaemic control during early pregnancy and fetal malformations in women with type I diabetes mellitus. Diabetologia 2000;43:79–82. Boulot P, Chabbert-Buffet N, d’Ercole C, Floriot M, Fontaine P, Fournier A, Gillet JY, Gin H, Grandperret-Vauthier S, Geudj AM, Guionnet B, Hauguel-de-Mouzon S, Hieronimus S, Hoffet M, Jullien D, Lamotte MF, Lejeune V, Lepercq J, Lorenzi F, Mares P, Miton A, Penfornis A, Pfister B, Renard E, Rodier M, Roth P, Sery GA, Timsit J, Valat AS, Vambergue A, Verier-Mine O: French multicentric survey of outcome of pregnancy in women with pregestational diabetes. Diabetes Care 2003;26:2990–2993. Albert TJ, Landon MB, Wheller JJ, Samuels P, Cheng RF, Gabbe S: Prenatal detection of fetal anomalies in pregnancies complicated by insulin-dependent diabetes mellitus. Am J Obstet Gynecol 1996;174:1424–1428. Hellmuth E, Damm P, Molsted-Pedersen L: Congenital malformations in offspring of diabetic women treated with oral hypoglycaemic agents during embryogenesis. Diabet Med 1994;11:471–474. Towner D, Kjos S, Leung B, Montoro M, Xiang A, Mestman J, Buchanan T: Congenital malformations in pregnancies complicated by NIDDM. Diabetes Care 1995;18:1446–1451. Dunne F, Brydon P, Smith K, Gee H: Pregnancy in women with type 2 diabetes: 12 years outcome data 1990–2002. Diabet Med 2003;20:734–738. Piacquadio K, Hollingsworth DR, Murphy H: Effects of in-utero exposure to oral hypoglycaemic drugs. Lancet 1991;338:866–869. Denno K, Sadler T: Effects of the biguanide class of oral hypoglycemic agents on mouse embryogenesis. Teratology 1994;49:260–266. Hellmuth E, Damm P, Molsted-Pedersen L: Oral hypoglycaemic agents in 118 diabetic pregnancies. Diabet Med 2000;17:507–511. Garcia-Patterson A, Erdozain L, Ginovart G, Adelantado JM, Cubero JM, Gallo G, de Leiva A, Corcoy R: In human gestational diabetes mellitus congenital malformations are related to pre-pregnancy body mass index and to severity of diabetes. Diabetologia 2004;47:509–514. Cockroft DL, Coppola PT: Teratogenic effect of excess glucose on head-fold rat embryos in culture. Teratology 1977;16:141–146. Ellington SKL: In vivo and in vitro studies on the effects of maternal fasting during embryonic organogenesis in the rat. J Reprod Fertil 1980;60:383–388. Sadler TW: Effects of maternal diabetes on early embryogenesis. II. Hyperglycemia-induced exencephaly. Teratology 1980;21:349–356. Moley K, Chi M, Manchester J, McDougal D, Lowry O: Alterations of intraembryonic metabolites in preimplantation mouse embryos exposed to elevated concentrations of glucose: A metabolic explanation for the developmental retardation seen in preimplantation embryos from diabetic animals. Biol Reprod 1996;54:1209–1216. Forsberg H, Eriksson UJ, Melefors Ö, Welsh N: Beta-hydroxybutyrate increases reactive oxygen species in late but not in early postimplantation embryonic cells in vitro. Diabetes 1997;47:255–262.

Offspring of Diabetic Pregnancy

303

56 57 58

59 60 61 62 63

64

65

66

67 68 69 70

71 72

73 74 75 76 77

Horton WEJ, Sadler TW: Effects of maternal diabetes on early embryogenesis. Alterations in morphogenesis produced by the ketone body B-hydroxybutyrate. Diabetes 1983;32:610–616. Hunter ES, Sadler TW, Wynn RE: A potential mechanism of DL-beta-hydroxybutyrate-induced malformations in mouse embryos. Am J Physiol 1987;253:E72–E80. Ornoy A, Zaken V, Kohen R: Role of reactive oxygen species (ROS) in the diabetes-induced anomalies in rat embryos in vitro: Reduction in antioxidant enzymes and low-molecular-weight antioxidants (LMWA) may be the causative factor for increased anomalies. Teratology 1999;60:376–386. Sadler TW, Horton WEJ: Effects of maternal diabetes on early embryogenesis: The role of insulin and insulin therapy. Diabetes 1983;32:1070–1074. Eriksson RSM, Thunberg L, Eriksson UJ: Effects of interrupted insulin treatment on fetal outcome of pregnant diabetic rats. Diabetes 1989;38:764–772. Styrud J, Thunberg L, Nybacka O, Eriksson UJ: Correlations between maternal metabolism and deranged development in the offspring of normal and diabetic rats. Pediatr Res 1995;37:343–353. Eriksson UJ, Borg LAH: Diabetes and embryonic malformations. Role of substrate-induced freeoxygen radical production for dysmorphogenesis in cultured rat embryos. Diabetes 1993;42:411–419. Baker L, Piddington R, Goldman A, Egler J, Moehring J: Myo-inositol and prostaglandins reverse the glucose inhibition of neural tube fusion in cultured mouse embryos. Diabetologia 1990;33: 593–596. Hashimoto M, Akazawa S, Akazawa M, Akashi M, Yamamoto H, Maeda Y, Yamaguchi Y, Yamasaki H, Tahara D, Nakanishi T, Nagataki S: Effects of hyperglycaemia on sorbitol and myoinositol contents of cultured embryos: Treatment with aldose reductase inhibitor and myo-inositol supplementation. Diabetologia 1990;33:597–602. Hod M, Star S, Passonneau JV, Unterman TG, Freinkel N: Effect of hyperglycemia on sorbitol and myo-inositol content of cultured rat conceptus: Failure of aldose reductase inhibitors to modify myoinositol depletion and dysmorphogenesis. Biochem Biophys Res Commun 1986;140:974–980. Hod M, Star S, Passonneau J, Unterman TG, Freinkel N: Glucose-induced dysmorphogenesis in the cultured rat conceptus: Prevention by supplementation with myo-inositol. Isr J Med Sci 1990;26:541–544. Sussman I, Matschinsky FM: Diabetes affects sorbitol and myo-inositol levels of neuroectodermal tissue during embryogenesis in rat. Diabetes 1988;37:974–981. Weigensberg MJ, Garcia-Palmer F-J, Freinkel N: Uptake of myo-inositol by early-somite rat conceptus. Transport kinetics and effects of hyperglycemia. Diabetes 1990;39:575–582. Engström E, Haglund A, Eriksson UJ: Effects of maternal diabetes or in vitro hyperglycemia on uptake of palmitic and arachidonic acid by rat embryos. Pediatr Res 1991;30:150–153. Pinter E, Reece EA, Leranth CS, Sanyal MK, Hobbins JC, Mahoney MJ, Naftolin F: Yolk sac failure in embryopathy due to hyperglycemia: Ultrastructural analysis of yolk sac differentiation associated with embryopathy in rat conceptuses under hyperglycemic conditions. Teratology 1986;33:73–84. Eriksson UJ, Borg LAH: Protection by free oxygen radical scavenging enzymes against glucoseinduced embryonic malformations in vitro. Diabetologia 1991;34:325–331. Hagay ZJ, Weiss Y, Zusman I, Peled-Kamar M, Reece EA, Eriksson UJ, Groner Y: Prevention of diabetes-associated embryopathy by overexpression of the free radical scavenger copper zinc superoxide dismutase in transgenic mouse embryos. Am J Obstet Gynecol 1995;173: 1036–1041. Eriksson UJ, Simán CM: Pregnant diabetic rats fed the antioxidant butylated hydroxytoluene show decreased occurrence of malformations in the offspring. Diabetes 1996;45:1497–1502. Eriksson UJ, Naeser P, Brolin SE: Increased accumulation of sorbitol in offspring of manifest diabetic rats. Diabetes 1986;35:1356–1363. Eriksson UJ, Brolin SE, Naeser P: Influence of sorbitol accumulation on growth and development of embryos cultured in elevated levels of glucose and fructose. Diabetes Res 1989;11:27–32. Eriksson UJ, Wentzel P, Minhas HS, Thornalley PJ: Teratogenicity of 3-deoxyglucosone and diabetic embryopathy. Diabetes 1998;47:1960–1966. Thornalley PJ: Advanced glycation and the development of diabetic complications. Unifying the involvement of glucose, methylglyoxal and oxidative stress. Endocrinol Metab 1996;3: 149–166.

Persson/Eriksson/Hanson

304

78 79 80 81 82

83

84

85

86

87 88

89 90

91 92

93

94

95

96

97

98

Wolff SP, Jiang ZY, Hunt JV: Protein glycation and oxidative stress in diabetes mellitus and ageing. Free Radic Biol Med 1991;10:339–352. Cederberg J, Eriksson UJ: Decreased catalase activity in malformation-prone embryos of diabetic rats. Teratology 1997;56:350–357. Cederberg J, Galli J, Luthman H, Eriksson UJ: Increased mRNA levels of Mn-SOD and catalase in embryos of diabetic rats from a malformation-resistant strain. Diabetes 2000;49:101–107. Eriksson UJ: Importance of genetic predisposition and maternal environment for the occurrence of congenital malformations in offspring of diabetic rats. Teratology 1988;37:365–374. Otani H, Tanaka O, Tatewaki R, Naora H, Yoneyama T: Diabetic environment and genetic predisposition as causes of congenital malformations in NOD mouse embryos. Diabetes 1991;40: 1245–1250. Eriksson UJ, den Bieman M, Prins JB, van Zutphen LFM: Differences in Susceptibility for Diabetes-Induced Malformations in Separated Rat Colonies of Common Origin. 4th FELASA Symposium. Lyon, Fondation Marcel Mérieux, 1990, pp 53–57. Pani L, Horal M, Loeken MR: Rescue of neural tube defects in Pax-3-deficient embryos by p53 loss of function: Implications for Pax-3-dependent development and tumorigenesis. Genes Dev 2002;16:676–680. Fine EL, Horal M, Chang TI, Fortin G, Loeken MR: Evidence that elevated glucose causes altered gene expression, apoptosis, and neural tube defects in a mouse model of diabetic pregnancy. Diabetes 1999;48:2454–2462. Eriksson UJ, Bone AJ, Turnbull DM, Baird JD: Timed interruption of insulin therapy in diabetic BB/E rat pregnancy: Effect on maternal metabolism and fetal outcome. Acta Endocrinol (Copenh) 1989;120:800–810. Phelan SA, Ito M, Loeken MR: Neural tube defects in embryos of diabetic mice: Role of the Pax-3 gene and apoptosis. Diabetes 1997;46:1189–1197. Cai J, Phelan SA, Hill AL, Loeken MR: Identification of Dep-1, a new gene regulated by the transcription factor Pax-3, as a marker for altered embryonic gene expression during diabetic pregnancy. Diabetes 1998;47:1803–1805. Hill AL, Phelan SA, Loeken MR: Reduced expression of Pax-3 is associated with overexpression of cdc46 in the mouse embryo. Dev Genes Evol 1998;208:128–134. Phelan SA, Loeken MR: Identification of a new binding motif for the paired domain of Pax-3 and unusual characteristics of spacing of bipartite recognition elements on binding and transcription activation. J Biol Chem 1998;273:19153–19159. Wilson DB: Proliferation in the neural tube of the splotch (Sp) mutant mouse. J Comp Neurol 1974;154:249–255. Epstein DJ, Vekemans M, Gros P: Splotch (Sp2H), a mutation affecting development of the mouse neural tube, shows a deletion within the paired homeodomain of Pax-3. Cell 1991;67: 767–774. Tassabehji M, Read AP, Newton VE, Harris R, Balling R, Gruss P, Strachan T: Waardenburg’s syndrome patients have mutations in the human homologue of the Pax-3 paired box gene. Nature 1992;355:635–636. Hol F, Hamel B, Geurds M, Mullaart R, Barr F, Macina R, Mariman E: A frameshift mutation in the gene for PAX3 in a girl with spina bifida and mild signs of Waardenburg syndrome. J Med Genet 1995;32:52–56. Enciso JM, Gratzinger D, Camenisch TD, Canosa S, Pinter E, Madri JA: Elevated glucose inhibits VEGF-A-mediated endocardial cushion formation: Modulation by PECAM-1 and MMP-2. J Cell Biol 2003;160:605–615. Pinter E, Haigh J, Nagy A, Madri JA: Hyperglycemia-induced vasculopathy in the murine conceptus is mediated via reductions of VEGF-A expression and VEGF receptor activation. Am J Pathol 2001;158:1199–1206. Pinter E, Mahooti S, Wang Y, Imhof BA, Madri JA: Hyperglycemia-induced vasculopathy in the murine vitelline vasculature: Correlation with PECAM-1/CD31 tyrosine phosphorylation state. Am J Pathol 1999;154:1367–1379. Madri JA, Enciso J, Pinter E: Maternal diabetes: Effects on embryonic vascular development – a vascular endothelial growth factor-A-mediated process. Pediatr Dev Pathol 2003;6:334–341.

Offspring of Diabetic Pregnancy

305

99 Goldman AS, Baker L, Piddington R, Marx B, Herold R, Egler J: Hyperglycemia-induced teratogenesis is mediated by a functional deficiency of arachidonic acid. Proc Natl Acad Sci USA 1985;82:8227–8231. 100 Pinter E, Reece EA, Leranth CZ, Garcia-Segura M, Hobbins JC, Mahoney MJ: Arachidonic acid prevents hyperglycemia-associated yolk sac damage and embryopathy. Am J Obstet Gynecol 1986;155:691–702. 101 Wentzel P, Eriksson UJ: Antioxidants diminish developmental damage induced by high glucose and cyclooxygenase inhibitors in rat embryos in vitro. Diabetes 1998;47:677–684. 102 Reece EA, Wu YK, Wiznitzer A, Homko C, Yao J, Borenstein M, Sloskey G: Dietary polyunsaturated fatty acid prevents malformations in offspring of diabetic rats. Am J Obstet Gynecol 1996;175:818–823. 103 Reece AE, Wu YK: Prevention of diabetic embryopathy in offspring of diabetic rats with use of a cocktail of deficient substrates and an antioxidant. Am J Obstet Gynecol 1997;176:790–798. 104 Wiznitzer A, Furman B, Mazor M, Reece EA: The role of prostanoids in the development of diabetic embryopathy. Semin Reprod Endocrinol 1999;17:175–181. 105 Goto MP, Goldman AS, Uhing MR: PGE2 prevents anomalies induced by hyperglycemia or diabetic serum in mouse embryos. Diabetes 1992;41:1644–1650. 106 Piddington R, Joyce J, Dhanasekaran P, Baker L: Diabetes mellitus affects prostaglandin E2 levels in mouse embryos during neurulation. Diabetologia 1996;39:915–920. 107 Wentzel P, Welsh N, Eriksson UJ: Developmental damage, increased lipid peroxidation, diminished cyclooxygenase-2 gene expression, and lowered PGE2 levels in rat embryos exposed to a diabetic environment. Diabetes 1999;48:813–820. 108 Schoenfeld A, Erman A, Warchaizer S, Ovadia J, Bonner G, Hod M, Bonner G: Yolk sac concentration of prostaglandin E2 in diabetic pregnancy: Further clues to the etiology of diabetic embryopathy. Prostaglandins 1995;50:121–126. 109 Strieleman PJ, Connors MA, Metzger BE: Phosphoinositide metabolism in the developing conceptus. Effects of hyperglycemia and scyllo-inositol in rat embryo culture. Diabetes 1992;41:989–997. 110 Akashi M, Akazawa S, Akazawa M, Trocino R, Hashimoto M, Maeda Y, Yamamoto H, Kawasaki E, Takino H, Yokota A, Nagataki S: Effects of insulin and myo-inositol on embryo growth and development during early organogenesis in streptozocin-induced diabetic rats. Diabetes 1991;40: 1574–1579. 111 Reece EA, Khandelwal M, Wu YK, Borenstein M: Dietary intake of myo-inositol and neural tube defects in offspring of diabetic rats. Am J Obstet Gynecol 1997;176:536–539. 112 Khandelwal M, Reece EA, Wu YK, Borenstein M: Dietary myo-inositol therapy in hyperglycemia-induced embryopathy. Teratology 1998;57:79–84. 113 Strieleman PJ, Metzger BE: Glucose and scyllo-inositol impair phosphoinositide hydrolysis in the 10.5-day cultured rat conceptus: A role in dysmorphogenesis? Teratology 1993;48:267–278. 114 Baker L, Piddington R: Diabetic embryopathy: A selective review of recent trends. J Diabetes Complications 1993;7:204–212. 115 Oberley LW: Free radicals and diabetes. Free Radic Biol Med 1988;5:113–124. 116 Baynes JW: Role of oxidative stress in development of complications in diabetes. Diabetes 1991;40:405–412. 117 West IC: Radicals and oxidative stress in diabetes. Diabet Med 2000;17:171–180. 118 Guzel S, Seven A, Satman I, Burcak G: Comparison of oxidative stress indicators in plasma of recent-onset and long-term type 1 diabetic patients. J Toxicol Environ Health A 2000;59:7–14. 119 Hartnett ME, Stratton RD, Browne RW, Rosner BA, Lanham RJ, Armstrong D: Serum markers of oxidative stress and severity of diabetic retinopathy. Diabetes Care 2000;23:234–240. 120 Mezzetti A, Cipollone F, Cuccurullo F: Oxidative stress and cardiovascular complications in diabetes: Isoprostanes as new markers on an old paradigm. Cardiovasc Res 2000;47:475–488. 121 Ceriello A, Bortolotti N, Motz E, Lizzio S, Catone B, Assaloni R, Tonutti L, Taboga C: Red wine protects diabetic patients from meal-induced oxidative stress and thrombosis activation: A pleasant approach to the prevention of cardiovascular disease in diabetes. Eur J Clin Invest 2001;31: 322–328. 122 Manzella D, Barbieri M, Ragno E, Paolisso G: Chronic administration of pharmacologic doses of vitamin E improves the cardiac autonomic nervous system in patients with type 2 diabetes. Am J Clin Nutr 2001;73:1052–1057.

Persson/Eriksson/Hanson

306

123 Shin CS, Moon BS, Park KS, Kim SY, Park SJ, Chung MH, Lee HK: Serum 8-hydroxy-guanine levels are increased in diabetic patients. Diabetes Care 2001;24:733–737. 124 Ceriello A, Quagliaro L, Catone B, Pascon R, Piazzola M, Bais B, Marra G, Tonutti L, Taboga C, Motz E: Role of hyperglycemia in nitrotyrosine postprandial generation. Diabetes Care 2002;25: 1439–1443. 125 Palmer AM, Thomas CR, Gopaul N, Dhir S, Änggård EE, Poston L, Tribe RM: Dietary antioxidant supplementation reduces lipid peroxidation but impairs vascular function in small mesenteric arteries of the streptozotocin-diabetic rat. Diabetologia 1998;41:148–156. 126 Sano T, Umeda F, Hashimoto T, Nawata H, Utsumi H: Oxidative stress measurements by in vivo electron spin resonance spectroscopy in rats with streptozotocin-induced diabetes. Diabetologia 1998;41:1355–1360. 127 Elangovan V, Shohami E, Gati I, Kohen R: Increased hepatic lipid soluble antioxidant capacity as compared to other organs of streptozotocin-induced diabetic rats: A cyclic voltammetry study. Free Radic Res 2000;32:125–134. 128 Nishikawa T, Edelstein D, Du XL, Yamagishi S, Matsumura T, Kaneda Y, Yorek MA, Beebe D, Oates PJ, Hammes HP, Giardino I, Brownlee M: Normalizing mitochondrial superoxide production blocks three pathways of hyperglycaemic damage. Nature 2000;404:787–790. 129 Yang X, Borg LAH, Eriksson UJ: Altered metabolism and superoxide generation in neural tissue of rat embryos exposed to high glucose. Am J Physiol 1997;272:E173–E180. 130 Wentzel P, Thunberg L, Eriksson UJ: Teratogenic effect of diabetic serum is prevented by supplementation of superoxide dismutase and N-acetylcysteine in rat embryo culture. Diabetologia 1997;40:7–14. 131 Suzuki N, Svensson K, Eriksson UJ: High glucose concentration inhibits migration of rat cranial neural crest cells in vitro. Diabetologia 1996;39:401–411. 132 Simán CM, Eriksson UJ: Vitamin E decreases the occurrence of malformations in the offspring of diabetic rats. Diabetes 1997;46:1054–1061. 133 Sivan E, Reece EA, Wu YK, Homko CJ, Polansky M, Borenstein M: Dietary vitamin E prophylaxis and diabetic embryopathy: Morphologic and biochemical analysis. Am J Obstet Gynecol 1996;175:793–799. 134 Viana M, Herrera E, Bonet B: Teratogenic effects of diabetes mellitus in the rat. Prevention with vitamin E. Diabetologia 1996;39:1041–1046. 135 Yang X, Borg LAH, Simán CM, Eriksson UJ: Maternal antioxidant treatments prevent diabetes-induced alterations of mitochondrial morphology in rat embryos. Anat Rec 1998;251: 303–315. 136 Simán CM, Gittenberger-De Groot AC, Wisse B, Eriksson UJ: Malformations in offspring of diabetic rats: Morphometric analysis of neural crest-derived organs and effects of maternal vitamin E treatment. Teratology 2000;61:355–367. 137 Kinalski M, Sledziewski A, Telejko B, Zarzycki W, Kinalska II: Antioxidant therapy and streptozotocin-induced diabetes in pregnant rats. Acta Diabetol 1999;36:113–117. 138 Chang TI, Horal M, Jain SK, Wang F, Patel R, Loeken MR: Oxidant regulation of gene expression and neural tube development: Insights gained from diabetic pregnancy on molecular causes of neural tube defects. Diabetologia 2003;46:538–545. 139 Simán CM, Eriksson UJ: Vitamin C supplementation of the maternal diet reduces the rate of malformation in the offspring of diabetic rats. Diabetologia 1997;40:1416–1424. 140 Sakamaki H, Akazawa S, Ishibashi M, Izumino K, Takino H, Yamasaki H, Yamaguchi Y, Goto S, Urata Y, Kondo T, Nagataki S: Significance of glutathione-dependent antioxidant system in diabetesinduced embryonic malformations. Diabetes 1999;48:1138–1144. 141 Wiznitzer A, Ayalon N, Hershkovitz R, Khamaisi M, Reece EA, Trischler H, Bashan N: Lipoic acid prevention of neural tube defects in offspring of rats with streptozocin-induced diabetes. Am J Obstet Gynecol 1999;180:188–193. 142 Toescu V, Nuttall SL, Martin U, Nightingale P, Kendall MJ, Brydon P, Dunne F: Changes in plasma lipids and markers of oxidative stress in normal pregnancy and pregnancies complicated by diabetes. Clin Sci (Lond) 2004;106:93–98. 143 Persson B: Prevention of fetal malformation with antioxidants in diabetic pregnancy. Pediatr Res 2001;49:742–743.

Offspring of Diabetic Pregnancy

307

144 Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah PO, Platt MJ, Stanisstreet M, van Velszen D, Walkinshaw S: Outcomes of pregnancy in insulin dependent diabetic women: Results of a five year population cohort study. Br Med J 1997;315:275–278. 145 Hawthorne G, Robson S, Ryall EA, Sen D, Roberts SH, Ward Platt MP: Prospective population based survey of outcome of pregnancy in diabetic women: Results of the Northern Diabetic Pregnancy Audit, 1994. Br Med J 1997;315:279–281. 146 Lauenborg J, Mathiesen E, Ovesen P, Westergaard JG, Ekbom P, Molsted-Pedersen L, Damm P: Audit on stillbirths in women with pregestational type 1 diabetes. Diabetes Care 2003;26:1385–1389. 147 Philipps AF, Porte PJ, Stabinsky S, Rosenkrantz TS, Raye JR: Effects of chronic fetal hyperglycemia upon oxygen consumption in the ovine uterus and conceptus. J Clin Invest 1984;74: 279–286. 148 Widness JA, Teramo KA, Clemons GK, Voutilainen P, Stenman UH, McKinlay SM, Schwartz R: Direct relationship of antepartum glucose control and fetal erythropoietin in human type 1 (insulin-dependent) diabetic pregnancy. Diabetologia 1990;33:378–383. 149 Veille JC, Sivakoff M, Hanson R, Fanaroff AA: Interventricular septal thickness in fetuses of diabetic mothers. Obstet Gynecol 1992;79:51–54. 150 Combs CA, Gunderson E, Kitzmiller JL, Gavin LA, Main EK: Relationship of fetal macrosomia to maternal postprandial glucose control during pregnancy. Diabetes Care 1992;15:1251–1257. 151 Schwartz R, Gruppuso PA, Petzold K, Brambilla D, Hiilesmaa V, Teramo KA: Hyperinsulinemia and macrosomia in the fetus of the diabetic mother. Diabetes Care 1994;17:640–648. 152 Hanson U, Persson B: Outcome of pregnancies complicated by type 1 insulin-dependent diabetes in Sweden: Acute pregnancy complications, neonatal mortality and morbidity. Am J Perinatol 1993;10:330–333. 153 Hanson U: Major malformations, stillbirths and LGA-infants still a major problem in type-1-diabetic pregnancies (abstract). DPSG Meeting, Brioni, 1999. 154 Petersen MB, Pedersen SA, Greisen G, Pedersen JF, Molsted P L: Early growth delay in diabetic pregnancy: Relation to psychomotor development at age 4. Br Med J 1988;296:598–600. 155 Carpenter MW, Canick JA, Star J, Carr SR, Burke ME, Shahinian K: Fetal hyperinsulinism at 14–20 weeks and subsequent gestational diabetes. Obstet Gynecol 1996;87:89–93. 156 Jansson T, Wennergren M, Powell TL: Placental glucose transport and GLUT 1 expression in insulin-dependent diabetes. Am J Obstet Gynecol 1999;180:163–168. 157 Cooper MJ, Enderlein MA, Tarnoff H, Roge CL: Asymmetric septal hypertrophy in infants of diabetic mothers. Fetal echocardiography and the impact of maternal diabetic control. Am J Dis Child 1992;146:226–229. 158 Uvena-Celebrezze J, Fung C, Thomas AJ, Hoty A, Huston-Presley L, Amini SB, Catalano PM: Relationship of neonatal body composition to maternal glucose control in women with gestational diabetes mellitus. J Matern Fetal Neonatal Med 2002;12:396–401. 159 Combs CA, Jaekle RK, Rosenn B, Pope M, Miodovnik M, Siddiqi TA: Sonographic estimation of fetal weight based on a model of fetal volume. Obstet Gynecol 1993;82:365–370. 160 Boulvain M, Stan C, Irion O: Elective delivery in diabetic pregnant women. Cochrane Database Syst Rev 2001;97:CD001997. 161 Persson B, Gentz J, Kellum M: Metabolic observations in infants of strictly controlled diabetic mothers. Plasma levels of glucose, FFA, glycerol and B-beta-hydroxybutyrate during the first two hours after birth. Acta Paediatr Scand 1973;62:465–473. 162 Persson B, Gentz J: Follow-up of children of insulin-dependent and gestational diabetic mothers. Neuropsychological outcome. Acta Paediatr Scand 1984;73:349–358. 163 Vannucci RC, Vannucci SJ: Hypoglycemic brain injury. Semin Neonatol 2001;6:147–155. 164 Sunehag A, Ewald U, Larsson A, Gustafsson J: Attenuated hepatic glucose production but unimpaired lipolysis in newborn infants of mothers with diabetes. Pediatr Res 1997;42:492–497. 165 Marcus C, Ehren H, Bolme P, Arner P: Regulation of lipolysis during the neonatal period. Importance of thyrotropin. J Clin Invest 1988;82:1793–1797. 166 Aynsley-Green A, Hawdon JM: Hypoglycemia in the neonate: Current controversies. Acta Paediatr Jpn 1997;39(suppl 1):S12–S16. 167 Lucas A, Morley R, Cole TJ: Adverse neurodevelopmental outcome of moderate neonatal hypoglycaemia. BMJ 1988;297:1304–1308.

Persson/Eriksson/Hanson

308

168 Warram JH, Krolewski AS, Gottlieb MS, Kahn CR: Differences in risk of insulin-dependent diabetes in offspring of diabetic mothers and diabetic fathers. N Engl J Med 1984;311:149–152. 169 Warram JH, Martin BC, Krolewski AS: Risk of IDDM in children of diabetic mothers decreases with increasing maternal age at pregnancy. Diabetes 1991;40:1679–1684. 170 Weintrob N, Karp M, Hod M: Short- and long-range complications in offspring of diabetic mothers. J Diabetes Complications 1996;10:294–301. 171 Cho NH, Silverman BL, Rizzo TA, Metzger BE: Correlations between the intrauterine metabolic environment and blood pressure in adolescent offspring of diabetic mothers. J Pediatr 2000;136: 587–592. 172 Persson B, Gentz J, Moller E: Follow-up of children of insulin dependent (type I) and gestational diabetic mothers. Growth pattern, glucose tolerance, insulin response, and HLA types. Acta Paediatr Scand 1984;73:778–784. 173 Bjorntorp P, Enzi G, Karlsson K, Krotkiewski M, Sjostrom L, Smith U: The effect of maternal diabetes on adipose tissue cellularity in man and rat. Diabetologia 1974;10:205–209. 174 Sells CJ, Robinson NM, Brown Z, Knopp RH: Long-term developmental follow-up of infants of diabetic mothers. J Pediatr 1994;125:S9–S17. 175 Rizzo TA, Ogata ES, Dooley SL, Metzger BE, Cho NH: Perinatal complications and cognitive development in 2- to 5-year-old children of diabetic mothers. Am J Obstet Gynecol 1994;171: 706–713. 176 Ornoy A, Ratzon N, Greenbaum C, Peretz E, Soriano D, Dulitzky M: Neurobehaviour of school age children born to diabetic mothers. Arch Dis Child Fetal Neonatal Ed 1998;79: F94–F99. 177 Rizzo TA, Dooley SL, Metzger BE, Cho NH, Ogata ES, Silverman BL: Prenatal and perinatal influences on long-term psychomotor development in offspring of diabetic mothers. Am J Obstet Gynecol 1995;173:1753–1758. 178 Dabelea D, Hanson RL, Lindsay RS, Pettitt DJ, Imperatore G, Gabir MM, Roumain J, Bennett PH, Knowler WC: Intrauterine exposure to diabetes conveys risks for type 2 diabetes and obesity: A study of discordant sibships. Diabetes 2000;49:2208–2211. 179 Whitaker RC, Pepe MS, Seidel KD, Wright JA, Knopp RH: Gestational diabetes and the risk of offspring obesity. Pediatrics 1998;101:E9.

Bengt Persson, MD, PhD Logbacken 2 SE–13150 Salsjo-Duvnas (Sweden) Tel. ⫹46 8 7169590, Fax ⫹46 8 7169564, E-Mail [email protected]

Offspring of Diabetic Pregnancy

309

Djelmiš J, Desoye G, Ivaniševic´ M (eds): Diabetology of Pregnancy. Front Diabetes. Basel, Karger, 2005, vol 17, pp 310–319

Long-Term Consequences of Gestational Diabetes mellitus Jeannet Lauenborg, Elisabeth R. Mathiesen, Peter Damm Department of Obstetrics and Department of Endocrinology, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark

In the majority of women with gestational diabetes mellitus (GDM) glucose tolerance normalizes after pregnancy. However, it is well established that these women have a high risk of developing overt diabetes later in life. In the classical studies by O’Sullivan [1], diabetes was diagnosed in 36% of women 22–28 years after a pregnancy with GDM. This has later been confirmed in other populations [for a review, see 2]. Large variations in the risk of diabetes have been found among the different published studies. The trend has been that the reported risk for diabetes is higher in studies from the US compared to European studies. This has been ascribed to many factors, of which differences in ethnicity, degrees of obesity and diagnostic and screening criteria are the most important [1, 2]. Several predictive factors for later development of diabetes have been described, e.g. diagnosis of GDM early in pregnancy, high blood glucose levels at diagnosis, the need for insulin treatment during pregnancy, preterm delivery and an abnormal oral glucose tolerance test (OGTT) 2 months postpartum [3–8]. Overweight is a well-known risk factor for developing diabetes, also among women with previous GDM [9, 10]. This is notable since obesity is an increasing problem worldwide, also among younger Danish women [11]. Inspired by the present almost epidemic rise in obesity and overweight we will elaborate on the impact of overweight on the risk of diabetes in women with a history of GDM.

Glucose Tolerance after Pregnancy

Most long-term follow-up studies after GDM have been done in nonEuropean populations and do not differentiate between women treated with diet

or insulin during pregnancy. Since insulin-treated GDM in our population is a well-established and strong predictor for subsequent diabetes, this chapter mainly deals with the prognosis for women treated with diet only. We have previously published that 34% had abnormal glucose tolerance [3.7% type 1 diabetes, 13.7% type 2 diabetes, 17.0% impaired glucose tolerance (IGT)] 2–11 years after a pregnancy with diet-treated GDM compared to only 5% with IGT in a control group of women with a normal pregnancy [8]. We recently performed a study where we examined 481 women with previous diet-treated GDM: 151 diagnosed in 1978–1985, the old cohort, and 330 diagnosed in 1987–1996, the new cohort [12]. The incidence of abnormal glucose tolerance in the new cohort at the follow-up in 2000–2002 was compared with the results of the old cohort both at the first follow-up in 1988–1990 [8], where the duration of follow-up was comparable, and at the latest follow-up in 2000–2002. The latest follow-up took place at around 10 years (range 4–23 years) after the index pregnancy at age 43 years (range 25–61 years). The median body mass index (BMI) was 28 kg/m2 with more than 60% of the women being overweight (BMI ⱖ25 kg/m2) and more than half of the overweight being obese (BMI ⱖ30 kg/m2). The new cohort was older at index pregnancy, 33 versus 30 years, and had a markedly higher pre-pregnancy BMI, 26 versus 23 kg/m2. In the total cohort in 2000–2002, diabetes was present in 192 (40%), of whom 21 women had type 1 while 171 had type 2 diabetes. Half of the cases with diabetes were previously undiagnosed, which is in line with other studies [13]. Impaired fasting glucose (IFG) and/or IGT were diagnosed in 130 women (27%). Twelve of the 442 women (3%) examined 2 months postpartum had overt diabetes. All 12 had diabetes at follow-up. Figure 1 shows the incidence of diabetes and IFG/IGT in the new cohort in comparison with the old cohort at both follow-up examinations. It is noteworthy that the incidence of diabetes was more than 2-fold higher in the new cohort compared to the old cohort (41 vs.18%) at the first follow-up around 7 years after the index pregnancy. Furthermore, similar rates of diabetes were found in the two cohorts at the follow-up in 2000–2002 even though the old cohort was 10 years older and had been followed for more than 11 years longer. Among women from the old cohort, 15% had diabetes at the first follow-up and another 23% subsequently developed diabetes. Seventeen out of 27 (63%) with IFG and/or IGT at the first follow-up had developed diabetes at the latest follow-up.

The Impact of Overweight

Pre-pregnancy BMI increased during the years from a median of 23.6 kg/m2 around 1980 to 28.5 kg/m2 around 1995 (p ⬍ 0.0005). Accordingly

Diabetes following GDM

311

100 Normal GT IFG/IGT DM Type 2 Type 1

80

%

60 40 20 0 Old cohort Old cohort New cohort 1990, n ⫽241 2002, n ⫽ 151 2002, n ⫽ 330

Fig. 1. Glucose tolerance at follow-up for the old and new cohort. GT ⫽ Glucose tolerance; DM ⫽ diabetes diagnosed at the present follow-up.

100 80

Normal GT IFG/IGT Diabetes

%

60 40 20 0 Normalweight n⫽ 151

Overweight n⫽140

Obese n⫽187

Fig. 2. Glucose tolerance according to BMI at follow-up in 481 women with previous diet-treated GDM. Normal-weight BMI: ⬍25 kg/m2; overweight BMI: 25–30 kg/m2; obese BMI: ⬎30 kg/m2. GT ⫽ Glucose tolerance.

BMI was significantly higher in the new cohort compared to the old cohort around 7 years after pregnancy. Multiple logistic regression analysis identified pre-pregnancy overweight and obesity as independent predictors for the development of diabetes. The impact of weight on glucose tolerance at follow-up is shown in figure 2 and illustrates the severity of being obese as only 20% of the obese women had normal glucose tolerance. On the other hand, even among the normal-weight women the incidence of abnormal glucose tolerance was as high as 50%. Other studies have found similar results [1, 14]. Several follow-up studies in women with prior GDM have been published and some of these studies evaluating the risk of glucose intolerance (IFG, IGT or diabetes) after GDM are summarized in table 1. If available, data on weight, either in relation to pregnancy or at follow-up, are shown. The studies differ in

Lauenborg/Mathiesen/Damm

312

many features of their design. First of all the diagnostic tests applied in pregnancy and at follow-up varied between the studies and also within the studies with a long period of follow-up as in our latest study. This is almost inevitable when examining a population for more than two decades. Also the majority of studies did not differentiate between diet- and insulin-treated GDM in contrast to our study. Irrespective of the diagnostic test applied and length of follow-up, all studies found an increased risk of diabetes after GDM. Yet, only few studies had a follow-up length exceeding 10 years and the few long-term follow-up studies were carried out before the time when the worldwide increase in overweight took on epidemic proportions. One Australian study involving a large population (n ⫽ 881) published in 1991 equals our study regarding the length of follow-up and the test used at follow-up [15]. The authors found an incidence of 12% for diabetes and 16% for IGT and a mean BMI of 29.5 kg/m2 in the group with diabetes compared to 26 kg/m2 in glucose-tolerant subjects.

Predictors for Subsequent Diabetes

Our study in a predominantly white population found an incidence of diabetes that equals the risk in other populations with high incidences of type 2 diabetes in the background population. The long period of follow-up (up to 23 years) and the fact that part of the study population had been examined previously made it possible to detect a change in the incidence of diabetes. Thus, we found that the incidence of IFG/IGT and diabetes after GDM has more than doubled from our first follow-up in 1988–1990 (old cohort) to our second follow-up in 2000–2002 (new cohort). This difference remained after adjusting for the other predictive variables presented in table 2 and is referred to as the cohort effect. We believe the cohort effect reflects a change in lifestyle, e.g. less physical activity together with an excess caloric intake, which has been an increasing problem during the last decade. Though changes in lifestyle could explain the increasing BMI in the background population [16], BMI was an independent predictor in the present study together with cohort membership, which might be ascribed decreased physical activity. Table 2 presents independent predictors and their relevant odds ratios for the development of diabetes identified in our latest follow-up study [12] and by other studies. We found that the following variables were independently associated with diabetes in general and with type 2 diabetes in particular: new cohort membership, pre-pregnancy overweight (BMI ⱖ25 kg/m2), diagnosis of GDM before 24 weeks’ gestation, high fasting glucose at the diagnostic OGTT, and IGT 2 months postpartum. Type 2 diabetes was also predicted by the presence of a family history of

Diabetes following GDM

313

Table 1. Follow-up studies on the incidence of abnormal glucose tolerance after pregnancy in women with previous GDM Origin

GDM (control) n

Length of follow-up

Glucose tolerance (control)

Data on weight (control)

Boston, Mass., USA, 1979 [1, 10]

615 (328)

22–28 y

DM 36% (6%)b

Follow-up BW: ⬎120% of ideal BW: DM 55% (7%); ⬍120%: DM 31% (3%)

Melbourne, Australia, 1991 [15]

881

1–19 y

DM 12%; IGT 16%c

Pregnancy BMI: ⫹DM 29.5 kg/m2; ⫹IGT 27.5 kg/m2; NGT 26.0 kg/m2

Stockholm, Sweden, 1991 [23]

145 (41)

3–4 y

DM 3% (0%); IGT 22% (2%)c

Pre-pregnancy BW: ⫹DM/IGT 68 kg; NGT 60 kg (control no data)

Copenhagen, Denmarka, 1992 [8]

241 (57)

2–11 y

DM 17% (0%); IGT 17% (5%)c

Pre-pregnancy BMI: type 2 DM 25.7 kg/m2; type 1 DM 21.0 kg/m2; IGT 23.7 kg/m2; NGT 22.3 kg/m2 (21.5 kg/m2)

Providence, R.I., USA, 1993 [9]

350

⬍10 y

DM 7%; IGT 4%d

Pre-pregnancy BMI: ⫹DM 28.5 kg/m2; ⫺DM 25 kg/m2

Chicago, Ill., USA, 1993 [7]

274

⬍5 y

DM 41%; IGT 16%b

Follow-up BW: NGT 124%; IGT 119%; DM 135%

Los Angeles, Calif., USA, 1995 [3]

671

5y

Type 2 DM 47%d

No data

Madrid, Spain, 1999 [6]

788

3–6 m

DM 5%c

Pre-pregnancy BMI ⬎27 kg/m2: ⫹DM 39%; ⫺DM 23%

London, UK, 1999 [24]

192

⬍7 y

DM 25%; IGT 29%e

Follow-up BMI: ⫹IGT 28.5 kg/m2; ⫺IGT 26.3 kg/m2

Barcelona, Spain, 2000 [25]

120

⬍1 y

DM 2%; IGT 12%; IFG 3%c

Follow-up BMI: NGT 25.1 kg/m2; DM/IGT 28.5 kg/m2

Lund, Sweden, 2002 [5]

229 (60)

1y

DM 9% (0%); IGT 22% (2%)c

No data

Lauenborg/Mathiesen/Damm

314

Table 1 (continued) Origin

GDM (control) n

Length of follow-up

Glucose tolerance (control)

Data on weight (control)

Barcelona, Spain, 2003 [4]

696 (70)

0–14 y

DM 6% (0%); IGT 12%e

Pre-pregnancy BMI ⬎26.4 kg/m2: DM/IGT RR 3.0; follow-up BMI: 24.5 kg/m2 (24.8)

Copenhagen, Denmark, 2004 [12]

481

4–23 y

DM 40%; IGT/IFG 27%e

Pre-pregnancy BMI 25–30 kg/m2: DM OR 2.0; ⬎30 kg/ m2: DM OR 3.0

Studies presented are those applying an OGTT at follow-up (excluding postpartum follow-up) and with more than 100 subjects with prior GDM. The incidence of abnormal glucose tolerance is presented in the fourth column and data on weight in the fifth column. In studies including a control group the data are presented in parentheses. y ⫽ Years; m ⫽ months; NGT ⫽ normal glucose tolerance; DM ⫽ diabetes; BW ⫽ body weight; RR ⫽ relative risk; OR ⫽ odds ratio. a Only women with previous diet-treated GDM. b 100 g OGTT (National Diabetes Data Group). c 75 g OGTT (WHO 1985). d 75 g OGTT (National Diabetes Data Group). e 75 g OGTT (WHO 1999).

diabetes. Fasting glucose at the postpartum examination was not related to the diabetes status at follow-up. Ethnic origin other than Nordic Caucasian, maternal age and tobacco use did not predict diabetes. The only significant predictors for later type 1 diabetes were preterm delivery and IGT 2 months postpartum. Our study [12] is one of the few studies examining predictors for type 1 diabetes. One study by Albareda et al. [4] in women with prior GDM and type 1 diabetes at follow-up (n ⫽ 5) identified the ‘number of abnormal values or overt diabetes at diagnostic OGTT’ and ‘early diagnosis’ as independent predictors. Preterm delivery was a predictive factor in our study but not in the study by Albareda et al., probably due to a generally low proportion with type 1 diabetes as is often the case in populations other than Nordic Caucasians.

The Metabolic Syndrome after GDM

Impaired glucose regulation and overweight, frequent characteristics found in our previous diet-treated GDM population, are closely related to insulin

Diabetes following GDM

315

Table 2. Pregnancy-related independent predictors of subsequent diabetes identified in our latest study and in previously published studies Possible predictive factorsa For diabetes Pre-pregnancy BMI Early diagnosis of GDM High fasting BG at GDM diagnosis IGT postpartum Cohort membership For type 2 diabetes Pre-pregnancy BMI Family history of diabetes Early diagnosis of GDM High fasting BG at GDM diagnosis IGT postpartum For type 1 diabetes IGT postpartum Preterm delivery Not predictive Parity Maternal age at index pregnancy Ethnicity LGA infant

ORb

Confirming studies ref. no.

Non-confirming studies ref. no.

1.0/2.2/3.0c 3.6 2.3

4, 6, 7, 9 4 8

6, 8 6

4.4 3.1

6–8

1.0/2.6/4.2c 1.9 2.9 2.1

3, 4, 8d 3 3

3.5

3

2.8 3.2

8

10

5, 10

3, 6, 8, 10 3, 6, 8

3, 4, 6, 8, 10

2

8 4, 6, 8, 10

Early diagnosis of GDM ⫽ GDM diagnosis before 24 weeks of gestation; high fasting blood glucose (BG): blood glucose ⬎5.6 mmol/l; preterm delivery ⫽ delivery before 37 weeks of gestation; LGA ⫽ large for gestational age. a Factors examined by multiple logistic regression analyses in our latest follow-up [12]. b The results from the multiple logistic regression analyses presented as odds ratio (OR) for diabetes vs. not diabetes adjusted for other predictive factors with length of follow-up as a covariate. c Normal weight (BMI ⬍25 kg/m2)/overweight (BMI 25–30 kg/m2)/obesity (BMI ⬎30 kg/m2). d Excluding type 1 diabetic women.

resistance and may be accompanied by hypertension and dyslipidaemia. The presence of several of these pathophysiological features is called the metabolic syndrome X or the insulin resistance syndrome [17]. The prevalence of the metabolic syndrome in our study was almost 40% in the previous diet-treated

Lauenborg/Mathiesen/Damm

316

GDM group and this was 3 times higher than in the control group, even after adjusting for age and BMI [18]. Verma et al. [19] also found a 3-fold increased risk of the metabolic syndrome in a recent longitudinal study of prior GDM. In our study the prevalence of the metabolic syndrome was more than 2-fold increased among glucose-tolerant women with prior GDM compared to the control group, and for normal weight women with prior GDM the risk was 5fold increased.

Conclusion and Perspectives

Women with a history of GDM have a very high risk of developing diabetes later in life, especially type 2 diabetes. Furthermore, these women also show other characteristics of insulin resistance – central obesity, dyslipidaemia and hypertension – all associated with increased morbidity. The prevalence of abnormal glucose tolerance and the metabolic syndrome is much higher in women with prior GDM compared to the background population. From a population perspective it is very interesting that it has been estimated that up to one third of women with diabetes have previously had GDM [20]. The prevalence of diabetes and overweight is increasing worldwide and intervention strategies are needed to stop this increase. An obvious target for intervention are, therefore, women with prior GDM as they have already been identified at a young age as a high-risk group. Besides offering regular control of the glucose metabolism, examination of the lipid profile and for central obesity might be included in the years after pregnancy, to evaluate the overall risk profile in these women. It is well known that lifestyle intervention in subjects with IGT can reduce cardiovascular morbidity and the risk of progressing to overt diabetes [21]. The only intervention trial in women with previous GDM evaluated the effect of troglitazone in women with IGT postpartum and found a marked protection against development of diabetes [22]. Unfortunately troglitazone has later been withdrawn from the market due to side effects. Based on the available literature it is very likely but still needs to be confirmed that lifestyle or pharmacological intervention, also in women with a history of GDM, can reduce morbidity.

References 1

2

O’Sullivan JB: The Boston Gestational Diabetes Studies: Review and perspectives; in Sutherland HW, Stowers JM (eds): Carbohydrate Metabolism in Pregnancy and the Newborn IV. New York, Springer, 1989, pp 287–294. Kim C, Newton KM, Knopp RH: Gestational diabetes and the incidence of type 2 diabetes: A systematic review. Diabetes Care 2002;25:1862–1868.

Diabetes following GDM

317

3

4

5 6

7

8

9 10

11 12

13 14

15 16

17

18

19 20 21 22

23

Kjos SL, Peters RK, Xiang A, Henry OA, Montoro M, Buchanan TA: Predicting future diabetes in Latino women with gestational diabetes. Utility of early postpartum glucose tolerance testing. Diabetes 1995;44:586–591. Albareda M, Caballero A, Badell G, Piquer S, Ortiz A, de Leiva A, Corcoy R: Diabetes and abnormal glucose tolerance in women with previous gestational diabetes. Diabetes Care 2003;26: 1199–1205. Aberg AE, Jonsson EK, Eskilsson I, Landin-Olsson M, Frid AH: Predictive factors of developing diabetes mellitus in women with gestational diabetes. Acta Obstet Gynecol Scand 2002;81:11–16. Pallardo F, Herranz L, Garcia-Ingelmo T, Grande C, Martin-Vaquero P, Janez M, Gonzalez A: Early postpartum metabolic assessment in women with prior gestational diabetes. Diabetes Care 1999;22:1053–1058. Metzger BE, Cho NH, Roston SM, Radvany R: Prepregnancy weight and antepartum insulin secretion predict glucose tolerance five years after gestational diabetes mellitus. Diabetes Care 1993;16:1598–1605. Damm P, Kühl C, Bertelsen A, Mølsted-Pedersen L: Predictive factors for the development of diabetes in women with previous gestational diabetes mellitus. Am J Obstet Gynecol 1992;167: 607–616. Coustan DR, Carpenter MW, O’Sullivan PS, Carr SR: Gestational diabetes: Predictors of subsequent disordered glucose metabolism. Am J Obstet Gynecol 1993;168:1139–1144. O’Sullivan JB: Gestational diabetes: Factors influencing the rates of subsequent diabetes; in Sutherland HW, Stowers JM (eds): Carbohydrate Metabolism in Pregnancy and the Newborn. Berlin, Springer, 1979, pp 425–435. Heitmann BL: Ten-year trends in overweight and obesity among Danish men and women aged 30–60 years. Int J Obes Relat Metab Disord 2000;24:1347–1352. Lauenborg J, Hansen T, Jensen DM, Vestergaard H, Mølsted-Pedersen L, Hornnes P, Locht H, Pedersen O, Damm P: Increasing incidence of diabetes after gestational diabetes mellitus – a longterm follow-up in a Danish population. Diabetes Care 2004;27:1194–1199. Grant PT, Oats JN, Beischer NA: The long-term follow-up of women with gestational diabetes. Aust NZ J Obstet Gynaecol 1986;26:17–22. Kautzky-Willer A, Prager R, Waldhausl W, Pacini G, Thomaseth K, Wagner OF, Ulm M, Streli C, Ludvik B: Pronounced insulin resistance and inadequate beta-cell secretion characterize lean gestational diabetes during and after pregnancy. Diabetes Care 1997;20:1717–1723. Henry OA, Beischer NA: Long-term implications of gestational diabetes for the mother. Baillieres Clin Obstet Gynaecol 1991;5:461–483. Drivsholm T, Ibsen H, Schroll M, Davidsen M, Borch-Johnsen K: Increasing prevalence of diabetes mellitus and impaired glucose tolerance among 60-year-old Danes. Diabet Med 2001;18: 126–132. World Health Organization: Definition, Diagnosis and Classification of Diabetes mellitus and Its Complications. Report of a WHO Consultation. Part 1. Diagnosis and Classification of Diabetes mellitus. WHO/NCD/NCS/99.2. Geneva, World Health Organization, 1999. Lauenborg J, Mathiesen E, Hansen T, Glumer C, Jorgensen T, Borch-Johnsen K, Hornnes P, Pedersen O, Damm P: The prevalence of the metabolic syndrome in a Danish population of women with previous GDM is 3-fold higher than in the general population (abstract). Diabetologia 2004;47:A358. Verma A, Boney CM, Tucker R, Vohr BR: Insulin resistance syndrome in women with prior history of gestational diabetes mellitus. J Clin Endocrinol Metab 2002;87:3227–3235. Cheung NW, Byth K: Population health significance of gestational diabetes. Diabetes Care 2003;26:2005–2009. Qiao Q, Tuomilehto J, Borch-Johnsen K: Post-challenge hyperglycaemia is associated with premature death and macrovascular complications. Diabetologia 2003;46(suppl 1):M17–M21. Buchanan TA, Xiang AH, Peters RK, Kjos SL, Marroquin A, Goico J, Ochoa C, Tan S, Berkowitz K, Hodis HN, Azen SP: Preservation of pancreatic beta-cell function and prevention of type 2 diabetes by pharmacological treatment of insulin resistance in high-risk Hispanic women. Diabetes 2002;51:2796–2803. Persson B, Hanson U, Hartling SG, Binder C: Follow-up of women with previous GDM. Insulin, C-peptide, and proinsulin responses to oral glucose load. Diabetes 1991;40(suppl 2):136–141.

Lauenborg/Mathiesen/Damm

318

24

25

Kousta E, Lawrence NJ, Penny A, Millauer BA, Robinson S, Dornhorst A, de Swiet M, Steer PJ, Grenfell A, Mather HM, Johnston DG, McCarthy MI: Implications of new diagnostic criteria for abnormal glucose homeostasis in women with previous gestational diabetes. Diabetes Care 1999;22:933–937. Costa A, Carmona F, Martinez-Roman S, Quinto L, Levy I, Conget I: Post-partum reclassification of glucose tolerance in women previously diagnosed with gestational diabetes mellitus. Diabet Med 2000;17:595–598.

Jeannet Lauenborg, MD Snedronningvej 33, DK–2730 Herlev (Denmark) Tel. ⫹45 44943654, Fax ⫹45 35454471 E-Mail [email protected]

Diabetes following GDM

319

Author Index

Catalano, P. 46 Cetin, I. 72 Corcoy, R. 195

Gillmer, M.D.G. 278 Hanson, U. 288 Hauguel-de Mouzon, S. 46 Herrera, E. 34 Hiilesmaa, V. K. 222 Hod, M. 214 Ivaniševic´, M. IX, 230

Damm, P. 271, 310 De Hertogh, R. 58 de Leiva, A. 195 Desoye, G. IX, 94, 153 Djelmiš, J. IX, 161 Dornhorst, A. 174 Eriksson, U.J. 288 Frost, G. 174

Kaufmann, P. 94 Kautzky-Willer, A. 18 Kos, M. 127 Lauenborg, J. 310 Leach, L. 110 Lepercq, J. 46

Mayhew, T.M. 110 Metelko, Z. 1, 206 Miller, V.A. 278 ˆ

Bancher-Todesca, D. 18 Blondeau, B. 83 Breant, B. 83

Pavlic´-Renar, I. 1, 206 Pearson, D.W.M. 254 Persson, B. 288 Pieber, T.R. 144 Plagemann, A. 9 Radaelli, T. 72 Teramo, K.A. 222 Vogel, M. 127 Weiss, P.A.M. 153

Mathiesen, E.R. 271, 310 Mauricio, D. 195

Yogev, Y. 214

320

Subject Index

Abdominal circumference (AC) delivery risks 280, 281 ultrasound 235 Acarbose, safety in pregnancy 210 Adherens junctions, placental vasculogenesis 113, 116 Adiponectin, deficiency and gestational diabetes role 24–26 Advanced glycation end products (AGEs), fetoplacental effects 120 Alcohol, avoidance in pregnancy 177 Amino acid metabolism, gestational diabetes 37, 38, 156 Amniocentesis chromosome anomaly screening 169 fetal monitoring 170, 227, 281 Amniotic fluid, insulin levels 156–158 Anencephaly, fetal ultrasound 242 Angiopoietins, placental vasculogenesis 112, 116, 121 Angiotensin-converting enzyme (ACE) inhibitors, diabetic nephropathy management in pregnancy 165, 263, 264 Angiotensin II receptor blockers, diabetic nephropathy management 263, 264 Apoptosis, blastocysts 61 Aspart, insulin therapy in pregnancy 207–209 Autoantibodies, see Diabetes-related antibodies

Beta cell fetal growth 86, 87 glucocorticoids and development effects 89, 91 insulin secretion in pregnancy and gestational diabetes 26–29 islet amyloid pancreatic polypeptide secretion in pregnancy and gestational diabetes 29–31 maternal nutrition effects on fetal development bilateral uterine artery ligation 88 generally food-restricted rat model 89, 91 hyperplasia 132, 133 low-protein rat model 88, 89 mild hyperglycemia 87, 88 severe hyperglycemia 88 Biparietal diameter (BPD), ultrasound 233, 235 Body mass index (BMI), maternal weight gain recommendations 186 Brachial plexus injury, macrosomic infant delivery 226, 297 Carbohydrate metabolism gestational diabetes 36, 37 glucose metabolism in blastocysts 62 Caudal regression sequence fetal ultrasound 248 gestational diabetes association 129

321

Cesarean section diabetic mother approach 285 frequency in diabetic patients 170, 171 hyperinsulinemic infants 153 indications 226, 281 Crown-rump length, ultrasound 233 Delivery analgesia 284 cesarean section, see Cesarean section fetal monitoring 284 induction of labor 283, 284 instrumental deliveries 285 insulin therapy 171, 282, 283 macrosomia prediction and complications 280, 281 postpartum care 285 preterm labor incidence in diabetes 278, 279 tocolytic drugs 279, 280 timing 281, 282 Diabetes-related antibodies (DRA) morbidity in pregnancy fetal diabetes risks 200, 201 maternal abnormal glucose tolerance 201 perinatal outcome 199, 200 placental transfer 199 pregnancy effects on levels 198 prevalence in pregnancy control mothers 197, 198 gestational diabetes mellitus 195–197 type 1 diabetes 197, 198 Diabetes type 1 autoantibodies, see Diabetes-related antibodies complications in pregnancy, see specific complications congenital malformations 162, 223, 241, 242 embryopathy 289, 290 epidemiology 222, 223, 254 fetal monitoring 170 gestational diabetes mellitus and maternal risks 5 maternal care

Subject Index

completion of pregnancy 170, 171 diet, see Nutrition first trimester 166–168 preconception 166, 255–258 puerperium 171 second trimester 168, 169 third trimester 169, 170 maternal risks following gestational diabetes 310, 311 mortality in perinatal period 161, 162, 222, 223, 294 offspring risks later in life following diabetic pregnancy 299, 300 Diabetes type 2 complications in pregnancy, see specific complications embryopathy 290 fetal monitoring 170 global distribution 1 maternal care completion of pregnancy 170, 171 diet, see Nutrition first trimester 166–168 preconception 166 puerperium 171 second trimester 168, 169 third trimester 169, 170 maternal risks following gestational diabetes 310, 311, 317 mortality in perinatal period 294 offspring of diabetic pregnancies risk reduction in subsequent pregnancies 189 risks later in life 300, 301 trends 1, 2 Diabetic ketoacidosis (DKA) clinical features 268 diagnosis 268, 269 fetal mortality 224 frequency in pregnancy 166 management 269 pathophysiology 268 Diabetic nephropathy clinical features 261, 262 detection and screening 262, 263 fetal outcomes 264 frequency in pregnancy 165, 263

322

management in pregnancy 165, 263, 264 pregnancy effects on progression 264 Diabetic neuropathy clinical features 264, 265 management in pregnancy 165, 166 Diabetic retinopathy epidemiology in pregnancy 260, 261 grading 259 management in pregnancy 165, 260 pathophysiology 258 screening 260 Diet, see Nutrition Doppler ultrasound, see Ultrasound Duodenal atresia, fetal ultrasound 247, 248 Erythropoietin, fetal effects 119, 120 Estrogen, gestational diabetes role 23 Evidence-based medicine decision making 144 definition 146 origins 145 process application in clinical practice 151, 152 best evidence tracking for question answering 147–149 critical appraisal of evidence absolute risk reduction 149, 150 levels of evidence 149, 150 odds ratio 149 relative risk 149 need for information conversion to focused questions 146, 147 performance evaluation 152 Exercise diabetes prevention in later life after gestational diabetes 189 pregnant diabetics 167 weight management in pregnancy 188, 189 Facial defects, fetal ultrasound 243, 244 Fetal effects, gestational diabetes arachidonic acid and prostaglandin metabolism 64, 65 beta cells hyperplasia 132, 133 mild hyperglycemia effects 87, 88

Subject Index

severe hyperglycemia effects 88 diabetes risks later in life 289, 299 functional teratogenesis development of theory 9, 10 insulin-treated rat neonate studies 11, 12 neuropeptide Y system ‘malorganization’ 13, 14 overfeeding of neonatal rats 12, 13 primary prevention 15, 16 streptozotocin rat studies 10, 11 ventromedial nucleus 11, 12 general health in childhood 4 genetic and environmental factors 290–294 glucose gradient 154–156 hyperglycemia-induced tissue damage 35, 36 insulin resistance 4 macrosomia, see Fetal growth malformations 4, 58, 127–129, 162, 290 mortality 127, 161, 162 myocardial ischemic necrosis 131, 132 myo-inositol depletion 64 neonatal metabolic abnormalities 297–299 obesity 4, 10, 13, 153 peri-implantation phase 63 preimplantation development animal models 59, 60 apoptosis of blastocysts 61 blastocyst growth 60, 61 fibroblast growth factor-4 deficiency and inner cell mass development 62, 63 glucose metabolism 62 overview 59 oxidative stress role 61, 62 tumor necrosis factor alpha effects 61 teratogenic period in diabetic pregnancy 291 unifying hypothesis of diabetes-induced embryopathy 65–69 Fetal growth adiposity and birth weight 72, 73 animal models for nutrition impact studies

323

Fetal growth (continued) generally food-restricted rat model of maternal undernutrition 86, 89, 91 intrauterine growth retardation 83, 84 low-protein rat model of maternal malnutrition 85, 88, 89 streptozotocin diabetic rats 84, 85 beta cell growth 86, 87 birth weight distribution in diabetic pregnancy 225, 226, 295 recommendations 187 body composition evaluation 78, 79 critical period 288 evaluation 78, 79 fetal nutrition determinants 73–75 gestational diabetes effects fat mass 75, 76 monitoring of diabetic pregnancies 79, 80 nutrient availability 76–78 hyperinsulinemia and macrosomia 163, 164 macrosomia features 130, 131, 214 maternal anthropometric factors 73 maternal metabolic changes and fetal growth effects 34–36, 42, 43 restraining factors 296 ultrasound findings intrauterine growth restriction 238 macrosomia 236–238, 280, 281 Fetal hypoxia clinical features 224 fetal factors 224–226 maternal factors 224 monitoring 226, 227 mortality rates 222, 223 prevention 227 ultrasound evaluation 170 vascular endothelial growth factor induction 121 ␣-Fetoprotein (AFP), screening 249, 250 Fibroblast growth factor-2 (FGF-2), expression in pregnancy 120, 121 Fibroblast growth factor-4 (FGF-4), deficiency and inner cell mass development 62, 63 Fish oil, preconception nutrition 178, 179

Subject Index

Folic acid, preconception nutrition 176, 177, 257 Free faty acids (FFAs) insulin response in gestational diabetes 21, 22 placental transfer 40–42 Functional teratogenesis, see Fetal effects, gestational diabetes Gestational diabetes mellitus (GDM) definition 2, 214, 289 diet, see Nutrition epidemiology incidence 3, 214 problems in estimates 2, 3 risk factors 3, 4 fetal risks, see Fetal effects, gestational diabetes maternal risks 4, 5 mortality in perinatal period 127, 161, 162, 294, 295 screening 5 Glucocorticoids, beta cell development effects 89, 91 Glucose monitoring continuous glucose monitoring advantages in pregnancy monitoring 219, 220 adverse fetal outcome prevention 219 clinical use not in pregnancy 215–217 HbA1c comparison 219 hypoglycemia detection 218, 219 overview 215 pregnancy studies 217, 218 self-monitoring of blood glucose 167, 214, 215, 219 Glucose tolerance tests insulin resistance in pregnancy 19, 20 maternal diabetes risks following gestational diabetes 310–314 villous maturation correlation 139, 140 GLUT1, placental expression in gestational diabetes 77, 78, 155, 156 GLUT4 concentrations in gestational diabetes 19 pregnancy effects on muscle expression 21

324

Glyburide fetal effects 209, 210 placental transfer 209 Goldenhar’s syndrome, gestational diabetes association 129 HbA1c continuous glucose monitoring comparison 219 pregnant diabetic monitoring 166, 214 Heart, fetal ultrasound 245–247 Human chorionic gonadotropin (hCG), screening 249, 250 Human placental lactogen (hPL), gestational diabetes levels 22, 95 Hypertension, frequency in diabetic pregnancy 164 Hypoglycemia classification 265 clinical features 265, 266 continuous glucose monitoring 218, 219 definition 265 driving precautions 267, 268 management 266, 267 meal planning and avoidance 185 neonatal hypoglycemia 298, 299 nocturnal hypoglycemia 267, 268 Hypoxia, see Fetal hypoxia Infection, frequency in diabetic pregnancy 164 Inositol, embryopathy role 292, 293 Insulin amniotic fluid levels 156–158 placental transfer 206, 207 receptor autophosphorylation impairment in pregnancy 21 secretion in pregnancy and gestational diabetes 26–29 therapy for pregnant diabetics analogs 207–209 first trimester 167 goals 207 pharmacokinetics 207 preconception period in type 1 diabetes 256, 257 puerperium 171, 282, 283 second trimester 168

Subject Index

third trimester 169 Insulin receptor substrate-1 (IRS-1), concentrations in gestational diabetes 21 Insulin resistance fetal programming 153 pregnancy-associated 18–26 Intramyocellular lipid content (IMCL), gestational diabetes 21, 22 Islet amyloid pancreatic polypeptide (IAPP), secretion in pregnancy and gestational diabetes 29–31 JAK-STAT pathways, leptin receptor signaling 52 Kidneys, fetal ultrasound 248 Labor, see Delivery Lactation, nutrition requirements 189 Leptin adiposity signal in pregnancy 23, 24 fetal leptin consequences 53, 54 functions 49, 50, 53 origins 53 functional overview 46 gestational diabetes levels and role 24, 50, 51 maternal functions 48 placental leptin function 49 gene expression regulation 51, 52 inflammation correlation 52 receptor signaling 52, 53 synthesis 48 pregnancy changes 24, 47, 48 Lipid metabolism, gestational diabetes adipose tissue 38, 39 hyperlipidemia 39, 40 placental transfer of lipid metabolites 40–42 Lipid peroxidation, placenta in gestational diabetes 103 Lipoprotein lipase (LPL), gestational diabetes activity 39, 40, 42 Lispro, insulin therapy in pregnancy 207–209

325

Macrosomia, see Fetal growth Metabolic syndrome, maternal risks following gestational diabetes 315–317 Metformin, fetal effects 209, 210 Mitogen-activated protein kinase (MAPK), leptin receptor signaling 52, 53 Myocardial ischemic necrosis, gestational diabetes 131, 132 Myo-inositol, fetal depletion 64 Neuropeptide Y (NPY), hypothalamic system ‘malorganization’ in functional teratogenesis 13, 14 Neuropsychological development, diabetic pregnancy effects 300 Nitrative stress, placenta in gestational diabetes 101, 104 Nutrition diabetes prevention in later life after gestational diabetes 189 gestational diabetes risk reduction 174, 175 lactation 189 preconception dietary advice alcohol avoidance 177 fish oil 178, 179 folic acid 176, 177 foods to avoid 179 overview 175, 176 vitamin supplements 177, 178 pregnant diabetics carbohydrate 182, 183 energy requirements 180, 181 fat 184, 185 meal planning and hypoglycemia avoidance 185 overview 167 protein 182 weight management in pregnancy exercise 188, 189 extreme caloric restriction 188 maternal weight gain recommendations 186, 187 modest caloric restriction 187, 188 Obesity gestational diabetes mellitus risks 3

Subject Index

maternal diabetes risks following gestational diabetes 311–315 offspring risks after gestational diabetes 4, 10, 13, 153, 300, 301 weight management in pregnancy exercise 188, 189 extreme caloric restriction 188 maternal weight gain recommendations 186, 187 modest caloric restriction 187, 188 Oxidative stress fetal effects in gestational diabetes 65, 130, 293 placenta in gestational diabetes antioxidant capacity 102 lipid peroxidation 103 overview 101 superoxide 104 preimplantation development effects in gestational diabetes 61, 62 Pax-3, downregulation in neural tube defects 291, 292 Placenta amino acid transport 156 angiogenesis and vascular remodeling 114–118 diabetes mellitus effects chorangiosis type II 138 histological findings 133–138 history of studies 133 morphology 133, 134 staging 137, 138 ultrastructure studies 136, 137 villi changes 134–137 fetal vascular resistance 118, 119 gestational diabetes effects advanced glycation end products 120 fetoplacental angiogenesis 119, 120 first trimester 95 term placenta antioxidant capacity 102 lipid peroxidation 103 morphology and development 95–101 oxidative and nitrative stress 101–104 villi features 97–100, 138–140

326

glucose transport 77, 78, 154, 155, 156 leptin function 49 gene expression regulation 51, 52 inflammation correlation 52 receptor signaling 52, 53 synthesis 48 lipid metabolite transfer 40–42 nutrient transfer capacity in gestational diabetes 77 placenta growth factor expression in pregnancy 112, 115, 116 vasculogenesis 111–114 Polyhydramnios, fetal ultrasound 248, 249 Preeclampsia definition 271, 272 frequency in diabetic pregnancy 164, 276 hypertension and risks 272 management 275, 276 metabolic control in prevention 272–274 microalbuminuria and risks 272 pathogenesis 274, 275 Pregnancy-associated plasma protein-A (PAPP-A), screening 250 Preterm labor incidence in diabetes 278, 279 tocolytic drugs 279, 280 Progesterone, gestational diabetes role 22 Prolactin, gestational diabetes role 23 Prostaglandins embryopathy role 292 metabolism in gestational diabetes 64, 65 Resistin, gestational diabetes role 26 Shoulder dystocia, macrosomic infant delivery 226, 280, 297 Smoking, cessation 257, 258 Spina bifida, fetal ultrasound 242, 243 Sulfonylureas fetal effects 209, 210 placental transfer 209 Syndrome X, see Metabolic syndrome Thyroid disease, autoimmunity in type 1 diabetes 258 Tight junctions, placental vasculogenesis 113

Subject Index

Transcerebellar diameter, ultrasound 235 Tumor necrosis factor-alpha (TNF-␣) blastocyst effects 61 gestational diabetes role 23 Ultrasound biophysical profile 249 birth weight estimation 169, 170 congenital malformations anencephaly 242 cardiac anomalies 245–247 caudal regression syndrome 248 duodenal atresia 247, 248 facial defects 243, 244 incidence in diabetes 241, 242 kidneys 248 polyhydramnios 248, 249 spina bifida 242, 243 Doppler imaging of blood flow 238–241, 247 fetal hypoxia evaluation 170 first trimester 231–233 gestational age estimation 230, 231 intrauterine growth restriction findings 238 long bone measures 236 macrosomia findings 236–238, 280, 281 preimplantation period 231 second trimester 168, 169, 235, 236 VACTERL, gestational diabetes association 128, 129 Vascular endothelial cadherin diabetic perturbations 116, 118 placental vasculogenesis 113, 116 Vascular endothelial growth factor (VEGF) diabetic perturbations 117, 118, 120, 121 embryopathy role 292 gestational changes in expression 115, 116 hypoxia induction 121 placental vasculogenesis 112–114 receptors 112 Ventricular septal defect, gestational diabetes association 128 Villi, see Placenta Vitamins, supplements 177, 178

327