Predictive markers for the FSH sensitivity of women with polycystic ovarian syndrome.

STUDY QUESTION Do parameters which are involved in pathogenesis of polycystic ovarian syndrome (PCOS) predict the dosage of recombinant FSH required to achieve monofollicular development for ovulation induction? SUMMARY ANSWER Anti-Mullerian hormone (AMH) appeared to be an independent predictor of the required dosage of FSH to achieve monofollicular development for ovulation induction in a study sample of clomiphene-resistant PCOS patients. WHAT IS KNOWN ALREADY AMH plays a key role in the pathogenesis of PCOS. This is the first study that has evaluated the association between AMH and the required FSH dosage to achieve the development of a large follicle of at least 18 mm, in the presence of additional predictors of ovarian responsiveness. In the few studies to date which have evaluated predictors of ovarian responsiveness in PCOS patients, fasting insulin has been shown to be a significant predictor. STUDY DESIGN, SIZE, DURATION A total of 48 infertile PCOS patients aged 18-43 years were enrolled in this prospective, observational study between 2009 and 2013. Study participants received between one and six cycles of ovarian stimulation with recombinant FSH using a step-up protocol. The mean total FSH dosage per cycle for reaching a monofollicular development for ovulation induction was evaluated to investigate its association with AMH, LH, FSH, LH/FSH-ratio, sex hormone-binding globulin (SHBG), androstendione, testosterone, free testosterone index, antral follicle count, ovarian volume, body mass index (BMI) and the age of patients. PARTICIPANTS/MATERIALS, SETTING, METHODS We used AMH-Gen-II ELISA (Beckman Coulter, Immunotech, Webster, TX, USA) for the assessment of AMH levels. Crude and multiple linear regression models were fitted to explore potential predictors of the required FSH dosage. MAIN RESULTS AND THE ROLE OF CHANCE An interquartile range (IQR) increase in AMH was associated with a 51.4% [95% confidence interval (CI): 24.7-79.0%; P = 0.0003] increase in the mean total FSH dosage per cycle (in IU) in a crude regression model, corresponding to a 7.2% increase in the mean total FSH dosage per cycle per ng/ml AMH. Adjustment for BMI augmented the effect of AMH, with a 58.3% (95% CI: 33.2-84.2%; P = 1.8 × 10(-5)) increase in FSH dosage per IQR AMH (corresponding to an 8.2% increase per ng/ml AMH) and a 46.2% (95% CI: 16.5-76.6%; P = 0.003) increase per IQR BMI (corresponding to a 3.7% increase per kg/m(2)). AMH was the only independent variable for which the effect on FSH dosage was statistically significant in the crude regression model as well as after adjustment for other promising predictors. The association of BMI with FSH dosage was statistically significant while adjusted for AMH, but not in the crude model. LIMITATIONS, REASONS FOR CAUTION The impact of metabolic parameters such as insulin resistance on the reported association between AMH and FSH dosage was not assessed. WIDER IMPLICATIONS OF THE FINDINGS Knowledge about the predictors of ovarian sensitivity to FSH can facilitate a physician's decision-making in providing the optimal infertility therapy for PCOS patients. STUDY FUNDING/COMPETING INTERESTS Funding was provided by the University Hospital of Essen.