Special Section: Eatinn Disorders
Reproductive Functions in Eating Disorders
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
Donna E. Stewart
This article reviews current knowledge about the effects of anorexia nervosa, bulimia nervosa and partial syndromes on ovulation, menstruation, sexuality, fertility, pregnancy and fetal-infant health. Eating disorders may result in failure to ovulate, oligomenorrhea, amenorrhea, reduced sex drive, infertility, hyperemesis gravidarum, low maternal weight gain in pregnancy, small babies for gestational date, low birth weight infants, increased neonatal morbidity and problems in infant feeding. The available information suggests that clinicians should inquire about nutritional intake, a history of eating disorders and weight reducing behaviours as part of the routine assessment of patients with the disorders of reproductive function listed above. If an eating disorder is discovered before conception, the woman should be encouraged to delay pregnancy until the eating disorder is treated and effectively under control. If the woman is pregnant, early diagnosis and treatment are essential to reduce maternal and fetal complications. The infants of eating-disordered women should be carefully followed to ensure adequate nutritional intake. Problems in reproductive function related to eating disorders offer rich opportunities for multispecialty collaboration in primary and secondary prevention programmes directed toward both mother and infant. Key words: anorexia nervosa; bulimia nervosa; eating disorders; reproductive functions. (Annals of Medicine 24: 287-291,1992)
Introduction Despite the fact that eating disorders most commonly occur in young women, inadequate attention has been paid to their effects on reproductive functions. Although it has been known for over a century that anorexia nervosa can cause amenorrhea (I), only in the last decade has it become clear that approximately 50% of women with bulimia nervosa also have abnormal menstrual cycles (2, 3).This article will describe current knowledge about the effects of anorexia nervosa, bulimia nervosa and subclinical eating disorders (partial syndromes) on ovulation, menstruation,sexuality, fertility, pregnancy and fetal-infant health (summarized in Table 1).
Address and reprint requests: Donna E. Stewart, M.D., D. Psych., F.R.C.P.(C), Associate Professor of Psychiatry & Obstetrics and Gynaecology, University of Toronto, St. Michael's Hospital, 30 Bond Street, Toronto, Ontario, Canada M5B 1W8.
Ovulation and Menstruation There is a minimum or threshold weight for height, indicated by a fatness index, that is necessary for the onset and maintenance of ovulating cycles and menses (4). Young women who develop anorexia nervosa before puberty may suffer from primary amenorrhea. Women in whom menstruation has been established may develop irregular periods or cease menstruation if body fat falls below the minimum for this function (2). Amenorrhea is so frequent in anorexia nervosa that the absence of three consecutive menstrual cycles where otherwise expected to occur is a diagnostic criterion for this disorder (5). A number of studies have examined the effect of eating disorders on ovulation and menstruation. Starkey & Lee (4) looked at menstruation in 58 patients with a previous diagnosis of anorexia nervosa They found that most women became amenorrheic concurrent with the onset of the eating disorder. In the group who did not gain weight, none experienced menstrual return. Nillius reported that 34% of 287 women with amenorrhea seen in his practice had amenorrhea caused by self-induced weight loss, but no mention was made of the presence of
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
288
Stewart
clinical eating disorders (7).A Scandinavian study by Fries (8) found a high proportion of women with eating disorders among 30 women with secondary amenorrhea caused by self-induced weight loss. Copeland & Herzog (9) and Devlin et al. (10) described the endocrinological findings associated with menstrual cycle abnormalities in women with anorexia nervosa and bulimia nervosa. These women tend to have fewer luteinizing hormone (LH) secretory spikes and a trend toward lower mean 24-h LH levels than controls. Stimulation with gonadotropin releasing hormone produces elevated LH responses in the bulimic nervosa group and blunted LH responses in the anorectic group. Oestradiol levels were uniformly lower in women. with anorexia nervosa. Stimulation with oestradiol revealed diminished LH augmentative responses and a trend toward diminished follicle stimulating hormone (FSH) augmentative responses in both patients with anorexia nervosa and bulimia nervosa compared to those of controls. Pirke et al. (2, 3) showed that approximately 50% of normal weight women with bulimia nervosa have menstrual abnormalities with impaired follicular maturation caused by impaired gonadotropin secretion. They also reported that in normal young women with normal body weights who dieted for 6 weeks (800-1000 kc/day), a variety of changes in endocrinologicalfunction developed with disruption of menstrual cycles for 3-6 months after dieting in approximately 20% of women. McCluskey et al. (11) reported a higher prevalence of abnormal eating patterns in women with polycystic ovarian syndrome. However. replication is required (12,13). Eating disorders that fail to meet full diagnostic criteria for anorexia nervosa and bulimia nervosa may occur in at least 5% of the young female population (14, 15). These partial syndromes have been variably termed subclinical eating disorders or eating disorders not otherwise specified. Kreipe et al. (16) reported that women with eating disorders not otherwise specified also frequently have menstrual dysfunction. As this disorder occurs more often than criteria confirmed anorexia nervosa (1%) (17) or bulimia nervosa (1.7%) (18), the full extent of the contribution of eating disorders to the clinical symptom of disordered menstruation is still unknown but likely to be substantial in developed countries (Table 1).
Sexuality A number of clinicians have anecdotally reported a lack of sexual interest in female anorectics, describing them as ascetic, childlike and asexual (19,201. However, other papers have taken issue with this and claim there is a variety or little’ evidence for disturbed psychosexual function in eating-disordered.patients (20, 21). Part of this confusion may stem from a failure to distinguish between patients with anorexia nervosa and those with bulimia nervosa. Patients with bulimia nervosa are known to indulge in not only eating binges but also increased smoking, drinking and sexual behaviour. Men with anorexia nervosa are reported to show a loss of sexual interest and potency (22) associated with sex
Table 1. Effects of eatlng disorders (ED) on reproductlon.
Ovulation/mensfruation Amenorrhealoligomenorrhea in women with anorexia nervosa, bulimia nervosa. subclinical ED. Sexuality Women show disturbed sexual function. Men show loss of sexual interest and potency. Fertilify 17% of women in an infertility clinic had an ED. ED more common in women with disturbed menstruation. Pregnancy More hyperemesis gravidarum. Low maternal weight gain. Fetus/infant
Intrauterinegrowth retardation. Low birth weight. Congenital anomalies. Prematurity. Perinatal rnortality/low APGAR scores. Caloric restriction of babies.
hormone changes (23).More rigorous methodology is required to further understand any variation which may exist in the sexuality of eating-disorderedpatients.
Fertility It is estimated that 15-1 8% of all couples having regular unprotected intercourse in the childbearing years are unable to conceive within 1 year (24). Until recently it was assumed that among the eating disordered, only amenorrheic women with anorexia nervosa contributed to the overall rates of infertility. However, recent studies show that unexplained infertility in women may sometimes be due to an undisclosed eating disorder or severe weight control. The effect of eating disorders on male infertility is largely unknown, but sexual dysfunction and regression to prepubertal sex hormone levels have been reported (22,23). Bates et al. (25) found that 47 women with unexplained infertility or menstrual dysfunction had practiced weight control by caloric restriction in order to maintain ‘a fashionable body habitus’. When 36 of the women followed a dietary regimen designed to increase their weight to predicted ideal body weight, 73% conceived spontaneously, and 90% with secondary amenorrhea resumed menstruation. It is of interest that 97% of these infertile women and their husbands had been previously evaluated for infertility by a variety of diagnostic gynaecological studies without any cause being discovered. Nillius (7) found in a subset of his infertility patients with amenorrhea, that a number of costly and complicated gynaecological and endocrinological investigations and sometimes hormonal induction of ovulation occurred without their inadequate dietary intake ever coming to light. Allison et al. (26) employed the Eating Disorders Inventory as a screening instrument and found an increased prevalence of abnormal eating attitudes in a small sample of infertility clinic patients. Stewart et al.
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
Reproductive Functions in Eating Disorders (27) reported a 17% prevalence of eating disorders in 66 consecutive patients attending a reproductive biology unit for infertility. The patients were prospectively screened with a standardized eating attitudes test and women with positive screening tests were then interviewed to confirm or refute the diagnosis of an eating disorder: 7.6% of women met criteria for anorexia nervosa or bulimia nervosa, and 9.196 had an eating disorder not otherwise specified. Among infertile. women with amenorrhea or oligomenorrhea, 58% had eating disorders. None of these patients had previously disclosed their eating disorder to their infertility specialist. An Australian study by Abraham et al. (28) found 13 of 14 consecutive women who had ovulation induced met the criteria for an eating disorder at some time in the past, while five women currently fulfilled these criteria. They raised the question whether ovulation should be induced in women suffering or recovering from an eating disorder in view of their poorer prognosis in pregnancy. Several investigators have recommended that infertility specialists routinely ask questions about eating and dieting behaviour, a past history of eating disorder, exercising habits, and past and present body weight before proceeding with infertility investigations (27-29). Patients often fail to volunteer information about eating disorders and they sometimes appear to be of normal weight although they are below their personal set point (27). However, an awareness of the frequency of eating disorders in young women, followed by careful clinical inquiry about highest and lowest ever weights, dietary intake and dieting or other weight-losing behaviours should correctly identify most cases (Table 2).
289
Low Maternal Weight Gain. Low maternal weight before pregnancy and poor pregnancy weight gain correlate significantly with an increase in intrauterine growth retardation, low birth weight, congenital anomalies and perinatal mortality (30). Although much of this data was collected during famines or in concentration camps, eating disorders are also a cause of low pre-pregnancy weight and maternal failure to gain weight in pregnancy. Stewart et al. (31) described 15 women who had previously suffered from anorexia nervosa or bulimia nervosa and later conceived 23 pregnancies. Compared to women whose eating disorders were in remission, women who had an active eating disorder throughout pregnancy gained less weight and had more pregnancy complications. Lacey & Smith (32) examined the impact of pregnancy in a report of eating behaviour in 20 untreated bulimia nervosa patients with normal weight. Although they found the frequency of bulimic behaviour generally diminished as pregnancy advanced, symptoms tended to return in the puerperium, and in nearly half their sample eating patterns were more disturbed after delivery than before conception. It is clear that careful nutritional, weight and psychosocial histories should be obtained in women who fail to gain weight adequately in pregnancy and appropriate early referrals made when indicated (33). A variety of other complications have been described in eating-disordered pregnant women including exacerbation of the eating disorder, difficult labours, surgical deliveries, and hypertension (31, 32, 40).Larger samples are required to confirm these early findings.
Effects on the Fetus and Infant Pregnancy Hyperemesis Gravidarum. One study of hyperemesis gravidarum patients referred for psychiatric consultation found a number of these women had suffered from an eating disorder before conception (29). A comparison of these patients to hyperemesis gravidarum patients without a history of eating disorder found that those complicated by eating disorders responded less favourably to treatment and spent twice as many days in hospital during pregnancy. Fifty per cent of the eating-disorder patients with hyperemesis had ovulation induced in an infertility clinic. Although hyperemesis gravidarum may be caused by a variety of psychological, social, physical and physiological problems, an eating disorder may also play an important role in some women. The prevalence of eating disorders in women with hyperemesis gravidarum is still unknown. Table 2. Questions to identify an eating disorder (ED). 1. Current weight/height
2. LoweWhighestever body weight as adult. 3. Current nutritional intake. 4. Dieting behaviour or other weight-losingbehaviaurs. 5. Past history of ED. 6. Exercise habits. 7. Menstrual period regularity.
Several investigators (31, 32, 34-36) have described fetal problems associated with a maternal eating disorder. These difficulties may include intrauterine growth retardation, prematurity, low birth weight, low Apgar scores, increased risk of congenital anomalies and higher perinatal mortality. Van der Spuy has described the consequences to the infant of intrauterine growth retardation from any cause which may be considerable in causing antenatal or interpartum asphyxia, fetal death or an increase in fetal distress and damage (37). Once delivered, these small infants are at increased risk for a variety of problems including hypothermia, hypoglycemia and infection, and have an increased perinatal mortality rats (37). Stewart et al. (31) described smaller babies with lower 5-min Apgar scores in women who had an active eating disorder during their pregnancy compared to those whose eating disorders were in remission. Lacey & Smith (32) described a higher incidence of fetal abnormality including cleft lip and palate, multiple pregnancies, obstetrical complications including breech presentation and surgical intervention in 20 untreated normal weight bulimic women. Brinch et al. (35) followed 50 women with a past history of eating disorder and found double the rate of premature births in offspring and six times the expected rate of perinatal mortality. Treasure & Russell (36) described the outcome of seven pregnancies in women with anorexia nervosa who con-
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
290
Stewart
ceived despite low weight. The abdominal circumference of all seven babies was well below the third percentile at birth. These studies all suffer'from small sample sires and further work is required to replicate their findings in larger controlled studies. On a more positive note, Feingold et al. (38)report a single case of bulimia nervosa in which the diagnosis of eating disorder was recognized in early pregnancy but by implementing multidisciptinary management, including perinatology, nutrition and psychiatry, the woman was able to gain adequate weight in pregnancy and deliver a healthy baby. It will be necessary to determine if the results of such interventions can be generalized. Low birth weight infants who survive the early weeks of life may have long-term developmental consequences with continued delay in physical and neurological development and impaired intellectual ability, particularly when associated with low intrauterine growth before 26 weeks' gestation (37). No work has yet been reported on the long-term follow-up of low birth weight infants born to eating-disordered mothers. In view of the prevalence of eating disorders in young women, this work urgently needs to be undertaken. In addition to the above problems, some eating-disordered mothers worry that their infants will become obese, leading to early nutritional deprivation or caloric restriction. Lacey & Smith (32) found that 15% of bulimic mothers reported restricting calories in their child before I year of age. Stewart et al. (31) found a decreased rate of breastfeeding in women with an active eating disorder and more difficulties with postpartum adjustment. Brinch et al. (35) reported a 17% failure to thrive in the first year of life in the infants of eating-disordered mothers. Similarly, Treasure & Russell (36) reported that children of four patients attending their eating disorders clinic had been investigated for poor growth attributed to inadequaie feeding. They report that five of six mothers with anorexia nervosa had difficulty breast-feeding and introduced bottled feeds in the first few months of life. Stein & Fairburn (391,Fahy & Treasure (401, and Van Wezel-Meijler and Wit (41) also found poor nutrition in infants of women with eating disorders. Pediatricians have also reported failure to thrive in infants whose parents have been calorie-restricting their children because of fears that they might become overweight, although it is not clear that these parents actually suffered from an eating disorder (as opposed to overconcern about obesity) (42). Further investigations are required to establish the role of maternal eating disorders in nutritionally deprived children in developed countries.
Conclusions The information currently available suggests that a nutritional intake, and a history of eating disorders or weightreducing behaviours should be part of the routine assessment of patients with menorrhea, oligomenorrhea, infertility, prenatal patients with hyperemesis gravidarum, those who fail to gain weight adequately in pregnancy or who have small babies for their gestational
dates. Those women in whom eating disorders are discovered before conception should be counselled to delay pregnancy until the eating disorder is adequately treated and truly in remission. If the woman has already conceived, the earliest possible diagnosis of an eating disorder should be made to enable proper psychiatric treatment, dietary advice and weight monitoring to be implemented so as to reduce maternal and fetal complications. The infants of eating-disordered mothers should also be carefully observed for failure to thrive so that early corrective measures can be implemented. Similarly, the parents of infants who fail to thrive should be assessed for eating disorders,' overconcern about obesity or abnormal eating behaviours. Problems in reproductive function related to eating disorders offer rich opportunities for multispecialty collaboration in primary and secondary prevention programmes directed toward both mother and infant.
References Transactions of the Clinical Society of London 1674; 7: 22-8. 2. Plrke KY, Schwelger U, Lemmel W, Krieg JC, Berger, M. The influence of dieting on the menstrual cycle of healthy young women. d CIin Endocrinol Metab 1985; 60: 1174-9. 3. Plrke KM, Dogs M, Fichter MM, Tuschll RJ. Gonadotrophins, oestradiol, and progesterone during the menstrual cycle in bulimia nervosa. Clin Endocrinol Metab 1988; 60: 1174-9. 4. Ffisch RE, McArthur J. Menstrual cycles: fatness as a determinant of minimum weight for heighf necessary for their maintenance or onset Science 1974; 185: 949-52. 5. American Psychiatric Asaoclatlan. Diagnostic and Stafistical Manual of Mental Disorders, 3rd edn revised. Washington DC: American Psychiatric Association, 1967. 6. Starkey TA, Lee RA. Menstruation and fertility in anorexia nervosa. Am J Obsiet Gynecoll969;105: 374-9. 7. Nilliue SJ. Psycho-pathology of weight-related amenorrhea. In: Jacobs, US, ed. Advances in Gynaecolagical Endocrinology. London: Royal College of Obstetricians and Gynaecologists, 1978: 1 18-30. a. Frles H. Secondary amenorrhea, self-induced weight reduction and anorexia nervosa. Acta Psychiatr Scand 1974; 248 (SUPPI): 1-69. 9. Copeland PM, Herzog DB. Menstrual abnormalities. In: Hudson JI, Pipe, HG. eds. The Psychobiology of Bulimia. Washington DC: American Psychiatric Press, 1987:31 -54. 10. Oevlln MI), Walsh BT, K a k JL et al. Hypothalarnicpituitary-gonadal function in anorexia nervosa and bulimia. Psychiatry Res 1989; 28: 11-24. 11. McCluskey S, Evans C,Lacey HR, Pearce JB. Jambs H. Polycystic ovary syndrome and bulimia. Fertil and Steril 1991; 55: 287-91. 1 2 Chepdelaine FA. Bulimia and polycyslic ovarian syndrome? Fertil and Steril1991: 53: 585-613. Stewart DE. Eating disorders and inlertility. Am J ObStet Gyflecol 1991; 165: 1576-7. 14. Button EJ, Whltehouse A. Subclinica! anorexia nervosa. Psycho/ Med 1981; 11: 509-1 6. 15. King MB. Eating disorders in general practice. 8 r Med J 1986; 293;1412-14. 16. Krelpe RE, Strauss J, Hodgman CH, Ryan RM. Menstrual cycle abnormalities and subclinical eating disorders: a preliminary report. Psychosom Med 1989; 51: 1 -86. 17. Garflnkel PE, Garner 0. Anorexia Nervoss: A Multidimensional Perspective. New York, Bruner/Mazel, 1982: 307-26. 1. Gull W. Anorexia nervosa
Reproductive Functions in Eating Disorders 18. Ben-Tovim DI. DSM-Ill, draft DSM-Ill-R and the diagnosis and prevalence of bulimia in Australia. Am J Psychiatry
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
1988;145:1000-2. 19. Bruch H. Eating disorders. Basic Books, New York, 1973. 20. Abraham S,Beumont P. Varieties in psychosexual experience in patients with anorexia nervosa. Int J Eating Disorders 1982;1 : 10-1 9. 21. Scott D. The involvement of psychosexual factors in the causation of eating disorders: time for a reappraisal. lnt J Eating Disorders 1987;6:199-21 3. 22. Crisp A, Burns T. The clinical presentation of anorexia nervosa in the male. Int J Eating Disorders 1983;2:5-1 0. 23. Crisp A, Hsu L, Chen C, Wheeler M. Reproductive hormone profiles in male anorexia nervosa before, during and after restoration of body weight to normal. lnt J Eating Disorders 1982;1: 3-9. 24. Menning BE. The emotional needs of infertile couples. Fertil and Steril 1980;34:31 3-19. 25. Bates GW, Bates SR, Whitworth NS. Reproductive failure in women who practise weight control. Ferfil and Steril
1982;37:373-8. 26. Allison S,Kalucy R, Gilchrist P, Jones W. Weight preoccupation among infertile women. lnt J Eating Disorders
1988;7:743-8. 27. Stewart DE, Robinson GE, Goldbloom DS, Wrlght C. Infertility and eating disorders. Am J Obstet Gynecol 1990; 163:1 196-9. 28. Abraham S, Mira M, Llewellyn-Jones D. Should ovulation be induced in women recovering from an eating disorder or who are compulsive exercisers? Fertil and Steril 1990;52:
566-8. 29. Stewart DE, MacDonald OL. Hyperemesis gravidarum and eating disorders in pregnancy. In: Abraham S, LlewellynJones D, eds. Eating Disorders and Disordered Eating. Sydney: Ashwood House, 1987:52-5.
291
30. Abrams BF, Laros RK. Pregnancy weight, weight gain and birth weight. Am J Obstet Gynecol 1986;154:503-19. 31. Stewart DE, Raskin J, Garfinkel PE, MacDonald OL, Robinson GE. Anorexia nervosa, bulimia, and pregnancy. Am J Obstet Gynecoll987;157:1194-8. 32. Lacey JH, Smlth G. Bulimia nervosa: the impact of pregnancy on mother and baby. Br J Psychiatry 1987; 150:
777-81. 33. Stewart DE, Robinson GE. Pregnancy and eating disorders. In: Van Hall E, ed. The Free Woman. Carnforth, U.K.: Parthenon, 1989:81 2-1 7. 34. Strimling BS. Infant of a pregnancy complicated by anorexia nervosa. Amer J Dis Child 1984;138:68-9. 35. Brinch M, lsager T, Tolstrup K. Anorexia nervosa and motherhood: reproductional pattern and mothering behaviour of 50 women. Acta Psychiatr Scand 1988;77:
98-104. 36. Treasure JL, Russell GFM. Intrauterine growth and neonatal weight gain in babies of women with anorexia nervosa. Br Med J 1988;296:1038. 37. Van der Spuy 2. Nutrition and reproduction. Clin Obstet Gynecot 1985;12;579-604. 38. Feingold M, Kamlner Y, Lyonsk, Chaudhury AK, Costigan K, Cetrulo CL. Bulimia nervosa in pregnancy: a case report. Obstet and Gynecoll988; 77:1025-7. 39. Stein A, Fairburn C. Children of mothers with bulimia nervosa. Br Med J 1989;299:777-8. 40. Fahy T, Treasure J. Children of the mothers with bulimia nervosa. Br Med J 1989;299: 1031. 41. Van Wezel-Meijler G, Wit JM. The offspring of mothers with anorexia nervosa: a high-risk group for undernutrition and stunting? Eur J Pediatr 1989;149:130-5. 42. Pugliese MT, Weyman-Daum M, Moses M, Lifshitz F. Parental health beliefs as a cause of non-organic failure to thrive. Pediatrics 1987;80:175-82.
Reproductive Functions in Eating Disorders
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
Donna E. Stewart
This article reviews current knowledge about the effects of anorexia nervosa, bulimia nervosa and partial syndromes on ovulation, menstruation, sexuality, fertility, pregnancy and fetal-infant health. Eating disorders may result in failure to ovulate, oligomenorrhea, amenorrhea, reduced sex drive, infertility, hyperemesis gravidarum, low maternal weight gain in pregnancy, small babies for gestational date, low birth weight infants, increased neonatal morbidity and problems in infant feeding. The available information suggests that clinicians should inquire about nutritional intake, a history of eating disorders and weight reducing behaviours as part of the routine assessment of patients with the disorders of reproductive function listed above. If an eating disorder is discovered before conception, the woman should be encouraged to delay pregnancy until the eating disorder is treated and effectively under control. If the woman is pregnant, early diagnosis and treatment are essential to reduce maternal and fetal complications. The infants of eating-disordered women should be carefully followed to ensure adequate nutritional intake. Problems in reproductive function related to eating disorders offer rich opportunities for multispecialty collaboration in primary and secondary prevention programmes directed toward both mother and infant. Key words: anorexia nervosa; bulimia nervosa; eating disorders; reproductive functions. (Annals of Medicine 24: 287-291,1992)
Introduction Despite the fact that eating disorders most commonly occur in young women, inadequate attention has been paid to their effects on reproductive functions. Although it has been known for over a century that anorexia nervosa can cause amenorrhea (I), only in the last decade has it become clear that approximately 50% of women with bulimia nervosa also have abnormal menstrual cycles (2, 3).This article will describe current knowledge about the effects of anorexia nervosa, bulimia nervosa and subclinical eating disorders (partial syndromes) on ovulation, menstruation,sexuality, fertility, pregnancy and fetal-infant health (summarized in Table 1).
Address and reprint requests: Donna E. Stewart, M.D., D. Psych., F.R.C.P.(C), Associate Professor of Psychiatry & Obstetrics and Gynaecology, University of Toronto, St. Michael's Hospital, 30 Bond Street, Toronto, Ontario, Canada M5B 1W8.
Ovulation and Menstruation There is a minimum or threshold weight for height, indicated by a fatness index, that is necessary for the onset and maintenance of ovulating cycles and menses (4). Young women who develop anorexia nervosa before puberty may suffer from primary amenorrhea. Women in whom menstruation has been established may develop irregular periods or cease menstruation if body fat falls below the minimum for this function (2). Amenorrhea is so frequent in anorexia nervosa that the absence of three consecutive menstrual cycles where otherwise expected to occur is a diagnostic criterion for this disorder (5). A number of studies have examined the effect of eating disorders on ovulation and menstruation. Starkey & Lee (4) looked at menstruation in 58 patients with a previous diagnosis of anorexia nervosa They found that most women became amenorrheic concurrent with the onset of the eating disorder. In the group who did not gain weight, none experienced menstrual return. Nillius reported that 34% of 287 women with amenorrhea seen in his practice had amenorrhea caused by self-induced weight loss, but no mention was made of the presence of
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
288
Stewart
clinical eating disorders (7).A Scandinavian study by Fries (8) found a high proportion of women with eating disorders among 30 women with secondary amenorrhea caused by self-induced weight loss. Copeland & Herzog (9) and Devlin et al. (10) described the endocrinological findings associated with menstrual cycle abnormalities in women with anorexia nervosa and bulimia nervosa. These women tend to have fewer luteinizing hormone (LH) secretory spikes and a trend toward lower mean 24-h LH levels than controls. Stimulation with gonadotropin releasing hormone produces elevated LH responses in the bulimic nervosa group and blunted LH responses in the anorectic group. Oestradiol levels were uniformly lower in women. with anorexia nervosa. Stimulation with oestradiol revealed diminished LH augmentative responses and a trend toward diminished follicle stimulating hormone (FSH) augmentative responses in both patients with anorexia nervosa and bulimia nervosa compared to those of controls. Pirke et al. (2, 3) showed that approximately 50% of normal weight women with bulimia nervosa have menstrual abnormalities with impaired follicular maturation caused by impaired gonadotropin secretion. They also reported that in normal young women with normal body weights who dieted for 6 weeks (800-1000 kc/day), a variety of changes in endocrinologicalfunction developed with disruption of menstrual cycles for 3-6 months after dieting in approximately 20% of women. McCluskey et al. (11) reported a higher prevalence of abnormal eating patterns in women with polycystic ovarian syndrome. However. replication is required (12,13). Eating disorders that fail to meet full diagnostic criteria for anorexia nervosa and bulimia nervosa may occur in at least 5% of the young female population (14, 15). These partial syndromes have been variably termed subclinical eating disorders or eating disorders not otherwise specified. Kreipe et al. (16) reported that women with eating disorders not otherwise specified also frequently have menstrual dysfunction. As this disorder occurs more often than criteria confirmed anorexia nervosa (1%) (17) or bulimia nervosa (1.7%) (18), the full extent of the contribution of eating disorders to the clinical symptom of disordered menstruation is still unknown but likely to be substantial in developed countries (Table 1).
Sexuality A number of clinicians have anecdotally reported a lack of sexual interest in female anorectics, describing them as ascetic, childlike and asexual (19,201. However, other papers have taken issue with this and claim there is a variety or little’ evidence for disturbed psychosexual function in eating-disordered.patients (20, 21). Part of this confusion may stem from a failure to distinguish between patients with anorexia nervosa and those with bulimia nervosa. Patients with bulimia nervosa are known to indulge in not only eating binges but also increased smoking, drinking and sexual behaviour. Men with anorexia nervosa are reported to show a loss of sexual interest and potency (22) associated with sex
Table 1. Effects of eatlng disorders (ED) on reproductlon.
Ovulation/mensfruation Amenorrhealoligomenorrhea in women with anorexia nervosa, bulimia nervosa. subclinical ED. Sexuality Women show disturbed sexual function. Men show loss of sexual interest and potency. Fertilify 17% of women in an infertility clinic had an ED. ED more common in women with disturbed menstruation. Pregnancy More hyperemesis gravidarum. Low maternal weight gain. Fetus/infant
Intrauterinegrowth retardation. Low birth weight. Congenital anomalies. Prematurity. Perinatal rnortality/low APGAR scores. Caloric restriction of babies.
hormone changes (23).More rigorous methodology is required to further understand any variation which may exist in the sexuality of eating-disorderedpatients.
Fertility It is estimated that 15-1 8% of all couples having regular unprotected intercourse in the childbearing years are unable to conceive within 1 year (24). Until recently it was assumed that among the eating disordered, only amenorrheic women with anorexia nervosa contributed to the overall rates of infertility. However, recent studies show that unexplained infertility in women may sometimes be due to an undisclosed eating disorder or severe weight control. The effect of eating disorders on male infertility is largely unknown, but sexual dysfunction and regression to prepubertal sex hormone levels have been reported (22,23). Bates et al. (25) found that 47 women with unexplained infertility or menstrual dysfunction had practiced weight control by caloric restriction in order to maintain ‘a fashionable body habitus’. When 36 of the women followed a dietary regimen designed to increase their weight to predicted ideal body weight, 73% conceived spontaneously, and 90% with secondary amenorrhea resumed menstruation. It is of interest that 97% of these infertile women and their husbands had been previously evaluated for infertility by a variety of diagnostic gynaecological studies without any cause being discovered. Nillius (7) found in a subset of his infertility patients with amenorrhea, that a number of costly and complicated gynaecological and endocrinological investigations and sometimes hormonal induction of ovulation occurred without their inadequate dietary intake ever coming to light. Allison et al. (26) employed the Eating Disorders Inventory as a screening instrument and found an increased prevalence of abnormal eating attitudes in a small sample of infertility clinic patients. Stewart et al.
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
Reproductive Functions in Eating Disorders (27) reported a 17% prevalence of eating disorders in 66 consecutive patients attending a reproductive biology unit for infertility. The patients were prospectively screened with a standardized eating attitudes test and women with positive screening tests were then interviewed to confirm or refute the diagnosis of an eating disorder: 7.6% of women met criteria for anorexia nervosa or bulimia nervosa, and 9.196 had an eating disorder not otherwise specified. Among infertile. women with amenorrhea or oligomenorrhea, 58% had eating disorders. None of these patients had previously disclosed their eating disorder to their infertility specialist. An Australian study by Abraham et al. (28) found 13 of 14 consecutive women who had ovulation induced met the criteria for an eating disorder at some time in the past, while five women currently fulfilled these criteria. They raised the question whether ovulation should be induced in women suffering or recovering from an eating disorder in view of their poorer prognosis in pregnancy. Several investigators have recommended that infertility specialists routinely ask questions about eating and dieting behaviour, a past history of eating disorder, exercising habits, and past and present body weight before proceeding with infertility investigations (27-29). Patients often fail to volunteer information about eating disorders and they sometimes appear to be of normal weight although they are below their personal set point (27). However, an awareness of the frequency of eating disorders in young women, followed by careful clinical inquiry about highest and lowest ever weights, dietary intake and dieting or other weight-losing behaviours should correctly identify most cases (Table 2).
289
Low Maternal Weight Gain. Low maternal weight before pregnancy and poor pregnancy weight gain correlate significantly with an increase in intrauterine growth retardation, low birth weight, congenital anomalies and perinatal mortality (30). Although much of this data was collected during famines or in concentration camps, eating disorders are also a cause of low pre-pregnancy weight and maternal failure to gain weight in pregnancy. Stewart et al. (31) described 15 women who had previously suffered from anorexia nervosa or bulimia nervosa and later conceived 23 pregnancies. Compared to women whose eating disorders were in remission, women who had an active eating disorder throughout pregnancy gained less weight and had more pregnancy complications. Lacey & Smith (32) examined the impact of pregnancy in a report of eating behaviour in 20 untreated bulimia nervosa patients with normal weight. Although they found the frequency of bulimic behaviour generally diminished as pregnancy advanced, symptoms tended to return in the puerperium, and in nearly half their sample eating patterns were more disturbed after delivery than before conception. It is clear that careful nutritional, weight and psychosocial histories should be obtained in women who fail to gain weight adequately in pregnancy and appropriate early referrals made when indicated (33). A variety of other complications have been described in eating-disordered pregnant women including exacerbation of the eating disorder, difficult labours, surgical deliveries, and hypertension (31, 32, 40).Larger samples are required to confirm these early findings.
Effects on the Fetus and Infant Pregnancy Hyperemesis Gravidarum. One study of hyperemesis gravidarum patients referred for psychiatric consultation found a number of these women had suffered from an eating disorder before conception (29). A comparison of these patients to hyperemesis gravidarum patients without a history of eating disorder found that those complicated by eating disorders responded less favourably to treatment and spent twice as many days in hospital during pregnancy. Fifty per cent of the eating-disorder patients with hyperemesis had ovulation induced in an infertility clinic. Although hyperemesis gravidarum may be caused by a variety of psychological, social, physical and physiological problems, an eating disorder may also play an important role in some women. The prevalence of eating disorders in women with hyperemesis gravidarum is still unknown. Table 2. Questions to identify an eating disorder (ED). 1. Current weight/height
2. LoweWhighestever body weight as adult. 3. Current nutritional intake. 4. Dieting behaviour or other weight-losingbehaviaurs. 5. Past history of ED. 6. Exercise habits. 7. Menstrual period regularity.
Several investigators (31, 32, 34-36) have described fetal problems associated with a maternal eating disorder. These difficulties may include intrauterine growth retardation, prematurity, low birth weight, low Apgar scores, increased risk of congenital anomalies and higher perinatal mortality. Van der Spuy has described the consequences to the infant of intrauterine growth retardation from any cause which may be considerable in causing antenatal or interpartum asphyxia, fetal death or an increase in fetal distress and damage (37). Once delivered, these small infants are at increased risk for a variety of problems including hypothermia, hypoglycemia and infection, and have an increased perinatal mortality rats (37). Stewart et al. (31) described smaller babies with lower 5-min Apgar scores in women who had an active eating disorder during their pregnancy compared to those whose eating disorders were in remission. Lacey & Smith (32) described a higher incidence of fetal abnormality including cleft lip and palate, multiple pregnancies, obstetrical complications including breech presentation and surgical intervention in 20 untreated normal weight bulimic women. Brinch et al. (35) followed 50 women with a past history of eating disorder and found double the rate of premature births in offspring and six times the expected rate of perinatal mortality. Treasure & Russell (36) described the outcome of seven pregnancies in women with anorexia nervosa who con-
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
290
Stewart
ceived despite low weight. The abdominal circumference of all seven babies was well below the third percentile at birth. These studies all suffer'from small sample sires and further work is required to replicate their findings in larger controlled studies. On a more positive note, Feingold et al. (38)report a single case of bulimia nervosa in which the diagnosis of eating disorder was recognized in early pregnancy but by implementing multidisciptinary management, including perinatology, nutrition and psychiatry, the woman was able to gain adequate weight in pregnancy and deliver a healthy baby. It will be necessary to determine if the results of such interventions can be generalized. Low birth weight infants who survive the early weeks of life may have long-term developmental consequences with continued delay in physical and neurological development and impaired intellectual ability, particularly when associated with low intrauterine growth before 26 weeks' gestation (37). No work has yet been reported on the long-term follow-up of low birth weight infants born to eating-disordered mothers. In view of the prevalence of eating disorders in young women, this work urgently needs to be undertaken. In addition to the above problems, some eating-disordered mothers worry that their infants will become obese, leading to early nutritional deprivation or caloric restriction. Lacey & Smith (32) found that 15% of bulimic mothers reported restricting calories in their child before I year of age. Stewart et al. (31) found a decreased rate of breastfeeding in women with an active eating disorder and more difficulties with postpartum adjustment. Brinch et al. (35) reported a 17% failure to thrive in the first year of life in the infants of eating-disordered mothers. Similarly, Treasure & Russell (36) reported that children of four patients attending their eating disorders clinic had been investigated for poor growth attributed to inadequaie feeding. They report that five of six mothers with anorexia nervosa had difficulty breast-feeding and introduced bottled feeds in the first few months of life. Stein & Fairburn (391,Fahy & Treasure (401, and Van Wezel-Meijler and Wit (41) also found poor nutrition in infants of women with eating disorders. Pediatricians have also reported failure to thrive in infants whose parents have been calorie-restricting their children because of fears that they might become overweight, although it is not clear that these parents actually suffered from an eating disorder (as opposed to overconcern about obesity) (42). Further investigations are required to establish the role of maternal eating disorders in nutritionally deprived children in developed countries.
Conclusions The information currently available suggests that a nutritional intake, and a history of eating disorders or weightreducing behaviours should be part of the routine assessment of patients with menorrhea, oligomenorrhea, infertility, prenatal patients with hyperemesis gravidarum, those who fail to gain weight adequately in pregnancy or who have small babies for their gestational
dates. Those women in whom eating disorders are discovered before conception should be counselled to delay pregnancy until the eating disorder is adequately treated and truly in remission. If the woman has already conceived, the earliest possible diagnosis of an eating disorder should be made to enable proper psychiatric treatment, dietary advice and weight monitoring to be implemented so as to reduce maternal and fetal complications. The infants of eating-disordered mothers should also be carefully observed for failure to thrive so that early corrective measures can be implemented. Similarly, the parents of infants who fail to thrive should be assessed for eating disorders,' overconcern about obesity or abnormal eating behaviours. Problems in reproductive function related to eating disorders offer rich opportunities for multispecialty collaboration in primary and secondary prevention programmes directed toward both mother and infant.
References Transactions of the Clinical Society of London 1674; 7: 22-8. 2. Plrke KY, Schwelger U, Lemmel W, Krieg JC, Berger, M. The influence of dieting on the menstrual cycle of healthy young women. d CIin Endocrinol Metab 1985; 60: 1174-9. 3. Plrke KM, Dogs M, Fichter MM, Tuschll RJ. Gonadotrophins, oestradiol, and progesterone during the menstrual cycle in bulimia nervosa. Clin Endocrinol Metab 1988; 60: 1174-9. 4. Ffisch RE, McArthur J. Menstrual cycles: fatness as a determinant of minimum weight for heighf necessary for their maintenance or onset Science 1974; 185: 949-52. 5. American Psychiatric Asaoclatlan. Diagnostic and Stafistical Manual of Mental Disorders, 3rd edn revised. Washington DC: American Psychiatric Association, 1967. 6. Starkey TA, Lee RA. Menstruation and fertility in anorexia nervosa. Am J Obsiet Gynecoll969;105: 374-9. 7. Nilliue SJ. Psycho-pathology of weight-related amenorrhea. In: Jacobs, US, ed. Advances in Gynaecolagical Endocrinology. London: Royal College of Obstetricians and Gynaecologists, 1978: 1 18-30. a. Frles H. Secondary amenorrhea, self-induced weight reduction and anorexia nervosa. Acta Psychiatr Scand 1974; 248 (SUPPI): 1-69. 9. Copeland PM, Herzog DB. Menstrual abnormalities. In: Hudson JI, Pipe, HG. eds. The Psychobiology of Bulimia. Washington DC: American Psychiatric Press, 1987:31 -54. 10. Oevlln MI), Walsh BT, K a k JL et al. Hypothalarnicpituitary-gonadal function in anorexia nervosa and bulimia. Psychiatry Res 1989; 28: 11-24. 11. McCluskey S, Evans C,Lacey HR, Pearce JB. Jambs H. Polycystic ovary syndrome and bulimia. Fertil and Steril 1991; 55: 287-91. 1 2 Chepdelaine FA. Bulimia and polycyslic ovarian syndrome? Fertil and Steril1991: 53: 585-613. Stewart DE. Eating disorders and inlertility. Am J ObStet Gyflecol 1991; 165: 1576-7. 14. Button EJ, Whltehouse A. Subclinica! anorexia nervosa. Psycho/ Med 1981; 11: 509-1 6. 15. King MB. Eating disorders in general practice. 8 r Med J 1986; 293;1412-14. 16. Krelpe RE, Strauss J, Hodgman CH, Ryan RM. Menstrual cycle abnormalities and subclinical eating disorders: a preliminary report. Psychosom Med 1989; 51: 1 -86. 17. Garflnkel PE, Garner 0. Anorexia Nervoss: A Multidimensional Perspective. New York, Bruner/Mazel, 1982: 307-26. 1. Gull W. Anorexia nervosa
Reproductive Functions in Eating Disorders 18. Ben-Tovim DI. DSM-Ill, draft DSM-Ill-R and the diagnosis and prevalence of bulimia in Australia. Am J Psychiatry
Ann Med Downloaded from informahealthcare.com by Kungliga Tekniska Hogskolan on 08/25/15 For personal use only.
1988;145:1000-2. 19. Bruch H. Eating disorders. Basic Books, New York, 1973. 20. Abraham S,Beumont P. Varieties in psychosexual experience in patients with anorexia nervosa. Int J Eating Disorders 1982;1 : 10-1 9. 21. Scott D. The involvement of psychosexual factors in the causation of eating disorders: time for a reappraisal. lnt J Eating Disorders 1987;6:199-21 3. 22. Crisp A, Burns T. The clinical presentation of anorexia nervosa in the male. Int J Eating Disorders 1983;2:5-1 0. 23. Crisp A, Hsu L, Chen C, Wheeler M. Reproductive hormone profiles in male anorexia nervosa before, during and after restoration of body weight to normal. lnt J Eating Disorders 1982;1: 3-9. 24. Menning BE. The emotional needs of infertile couples. Fertil and Steril 1980;34:31 3-19. 25. Bates GW, Bates SR, Whitworth NS. Reproductive failure in women who practise weight control. Ferfil and Steril
1982;37:373-8. 26. Allison S,Kalucy R, Gilchrist P, Jones W. Weight preoccupation among infertile women. lnt J Eating Disorders
1988;7:743-8. 27. Stewart DE, Robinson GE, Goldbloom DS, Wrlght C. Infertility and eating disorders. Am J Obstet Gynecol 1990; 163:1 196-9. 28. Abraham S, Mira M, Llewellyn-Jones D. Should ovulation be induced in women recovering from an eating disorder or who are compulsive exercisers? Fertil and Steril 1990;52:
566-8. 29. Stewart DE, MacDonald OL. Hyperemesis gravidarum and eating disorders in pregnancy. In: Abraham S, LlewellynJones D, eds. Eating Disorders and Disordered Eating. Sydney: Ashwood House, 1987:52-5.
291
30. Abrams BF, Laros RK. Pregnancy weight, weight gain and birth weight. Am J Obstet Gynecol 1986;154:503-19. 31. Stewart DE, Raskin J, Garfinkel PE, MacDonald OL, Robinson GE. Anorexia nervosa, bulimia, and pregnancy. Am J Obstet Gynecoll987;157:1194-8. 32. Lacey JH, Smlth G. Bulimia nervosa: the impact of pregnancy on mother and baby. Br J Psychiatry 1987; 150:
777-81. 33. Stewart DE, Robinson GE. Pregnancy and eating disorders. In: Van Hall E, ed. The Free Woman. Carnforth, U.K.: Parthenon, 1989:81 2-1 7. 34. Strimling BS. Infant of a pregnancy complicated by anorexia nervosa. Amer J Dis Child 1984;138:68-9. 35. Brinch M, lsager T, Tolstrup K. Anorexia nervosa and motherhood: reproductional pattern and mothering behaviour of 50 women. Acta Psychiatr Scand 1988;77:
98-104. 36. Treasure JL, Russell GFM. Intrauterine growth and neonatal weight gain in babies of women with anorexia nervosa. Br Med J 1988;296:1038. 37. Van der Spuy 2. Nutrition and reproduction. Clin Obstet Gynecot 1985;12;579-604. 38. Feingold M, Kamlner Y, Lyonsk, Chaudhury AK, Costigan K, Cetrulo CL. Bulimia nervosa in pregnancy: a case report. Obstet and Gynecoll988; 77:1025-7. 39. Stein A, Fairburn C. Children of mothers with bulimia nervosa. Br Med J 1989;299:777-8. 40. Fahy T, Treasure J. Children of the mothers with bulimia nervosa. Br Med J 1989;299: 1031. 41. Van Wezel-Meijler G, Wit JM. The offspring of mothers with anorexia nervosa: a high-risk group for undernutrition and stunting? Eur J Pediatr 1989;149:130-5. 42. Pugliese MT, Weyman-Daum M, Moses M, Lifshitz F. Parental health beliefs as a cause of non-organic failure to thrive. Pediatrics 1987;80:175-82.
