Hormonal changes related to eating behavior in oligomenorrheic women

in European Journal of Endocrinology
Restricted access

Dumoulin SC, de Glisezinski I, Saint-Martin F, Jamrozik SI, Barbe P, Thouvenot J-P, Plantavid MM. Bennet AP, Louvet J-P Hormonal changes related to eating behavior in oligomenorrheic women. Eur J Endocrinol 1996;135:328–34. ISSN 0804–4643

The aim of this study was to determine those hormonal alterations in the gonadotropin-ovarian axis that are related to eating behavior in oligomenorrheic patients. We studied 74 oligomenorrheic women aged 26.2 ± 0.8 years, divided into group IA (N = 13) with eating disorders, group IB (N = 61) without eating disorders and 18 normally cycling controls aged 29.2 ± 1.6 years (group II). No subject had ovarian failure, pituitary disease, thyroid dysfunction or was taking any drug. Blood samples were taken on days 3–6 after the last menses. Luteinizing hormone (LH) was measured in two plasma pools, each made up of three samples taken at 30-min intervals, starting at 15.00 h (LH-1 5 h) and 09.00 h (LH-9 h), which allowed the mean LH (mLH) and variability in LH (V-LH: percentage increase from the lower to the higher of the two LH values) to be calculated. Follicle-stimulating hormone (FSH), sex steroids, and gonadotropin-releasing hormone-stimulated LH (sLH) and -FSH (sFSH) were also evaluated. Eating behavior was evaluated with the EAT questionnaire; the EAT 26 score, the dieting score (DS) and bulimia score (BS) were calculated. Dietary intake was evaluated in 35 group IB patients based on food diaries analyzed with the REGAL program, to evaluate daily calorie intake (Cal) and calories provided by carbohydrates (Carb), lipids (Lip) and proteins (Prot). Comparisons between groups were done by analysis of variance (followed by the Fisher PLSD test) and the KruskalWallis test. Groups IA. IB and II did not differ regarding age, body mass index, LH-9 h, LH-15 h, mLH, FSH, sLH, sFSH, estradiol or dehydroepiandrosterone sulfate; group IA had a higher V-LH than group II (p < 0.02) and a higher testosterone level than groups IB and II (p < 0.05). Positive correlations were found between V-LH and DS (p < 0.01) and BS (p < 0.05), and between testosterone and BS (p < 0.02) and DS (p < 0.05). The V-LH was negatively correlated with Cal and Carb, and testosterone was positively correlated with Cal and Lip. In patients referred for oligomenorrhea, it is concluded that testosterone levels and variability of LH levels are related to eating behavior.

Antoine P Bennet, Service d'Endocrinologie, Hôpital Purpan, Place du Dr Baylac, 31059 Toulouse-Cedex, France