Association of aromatase (CYP 19) gene variation with features of hyperandrogenism in two populations of young women.

BACKGROUND: Aromatase catalyses the conversion of androgens to estrogens and thus variation in the aromatase gene could contribute to female syndromes of androgen excess, such as precocious pubarche (PP) and polycystic ovarian syndrome (PCOS). METHODS: Two groups, one case-control containing girls from Barcelona, Spain with PP (n = 186) or healthy controls (n = 71), and the other a population study of young women from Oxford, UK, who volunteered for a study of normal women's health (n = 109), were genotyped at four aromatase gene haplotype-tag single nucleotide polymorphisms (SNP). Clinical features and hormone concentrations relevant to hyperandrogenism were compared across haplotypes or genotypes. RESULTS: Distributions of aromatase haplotypes (P < 0.0001) and aromatase SNP_50 genotype (P = 0.001) were significantly different between PP girls and Spanish controls. The AGGG haplotype was associated with an odds ratio (95% confidence interval) of 0.5 (0.3-0.9) (P = 0.005) for the presence of PP compared to GAGG. In 84 post-pubertal PP girls, aromatase haplotype was associated with functional ovarian hyperandrogenism (P < 0.05), independently of insulin sensitivity. In the Oxford population, SNP_50 was associated with variation in PCOS symptom score (P = 0.008) and circulating testosterone concentrations (P = 0.02). CONCLUSIONS: This study suggests that common variation at the aromatase gene (and not just rare loss-of-function mutations) is associated with androgen excess in girls and young women.

Polycystic ovaries and eating disorders: Are they related?

A cross-sectional observational study was used to investigate the reported association between polycystic ovarian syndrome and bulimia nervosa in a group of young, post-menarcheal women in the normal population. Volunteers aged 18-25 years were recruited from two universities and two general practice surgeries in Oxford. A total of 230 women completed an interviewer-based eating disorder examination, which was used to diagnose bulimia nervosa and its variants, and to assess eating behaviour. Transabdominal ultrasound was used to diagnose the presence or absence of polycystic ovaries. Symptoms of polycystic ovarian syndrome were assessed using menstrual history, anthropometric measurements, clinical observation of acne and hirsutism, and biochemical analysis of a fasting blood sample. A total of 30% of all participants described episodes of overeating and 4% had used extreme methods of weight control. Two women were diagnosed with bulimia nervosa, and five women with binge-eating disorder; however, these diagnoses were not associated with polycystic ovaries. Scores for dieting and overall eating disorder symptoms in the polycystic ovary groups were not significantly higher than those for women with normal ovaries, and therefore the suggestion that polycystic ovaries predispose towards the development of eating disorders is not supported by this study.

Polycystic ovaries and associated clinical and biochemical features in young women.

OBJECTIVE: To determine the prevalence of polycystic ovaries as identified by ultrasound in a group of young, postmenarcheal women in the normal population, and to investigate how polycystic ovaries are related to the spectrum of clinical and biochemical symptoms associated with the polycystic ovary syndrome (PCOS). DESIGN: Cross-sectional observational study. SUBJECTS AND METHODS: Volunteers were recruited from two universities and two general practice surgeries in Oxford. 230 women aged 18-25 years participated. Information collected and measurements performed included: a menstrual history, anthropometric measurements, clinical observation of acne and hirsutism, transabdominal pelvic ultrasound, and biochemical analysis of a fasting blood sample. MAIN OUTCOME MEASURES: Prevalence of polycystic ovaries and their association with symptoms of the polycystic ovary syndrome. RESULTS: Polycystic ovarian morphology was identified in 74 (33%, 95% CI = 27-39%) of the 224 women who attended for an ultrasound scan. In the non-users of hormonal contraception, irregular menstrual cycles were 20% more common in women with polycystic ovaries than in women with normal ovaries (P = 0.07). There were no significant differences in acne, hirsutism, body mass index or body fat percentage between women with polycystic and normal ovaries. Analysis of biochemical data showed that women with polycystic ovaries had higher total serum testosterone concentrations (P = 0.03). The prevalence of PCOS in this age group was as low as 8% or as high as 26% depending on which criteria were applied to define the syndrome. Sub-group analyses of women according to ovarian morphology and features of PCOS revealed greater mean BMI in women with PCOS, and also indicated lower fasting insulin concentrations and greater insulin sensitivity in polycystic ovary and PCOS groups when compared to women with normal ovaries. CONCLUSIONS: Polycystic ovaries are very common in this age group but are not necessarily associated with other symptomatology. The prevalence of polycystic ovary syndrome varies widely according to the definition applied. Sub-group analysis of women with polycystic ovaries according to the presence or absence of features of polycystic ovary syndrome does not reveal an increasing trend for progression of endocrine abnormalities usually associated with polycystic ovary syndrome.