Prognostic value of some ovarian reserve tests in poor responders.
BACKGROUND: The objective of this study was to determine the prognostic value of the basal estradiol (E2) and inhibin-B levels, the antral follicle count (AFC), and the clomiphene citrate challenge test (CCCT) of ovarian response in controlled ovarian hyperstimulation (COH), in an outcome with normal follicle-stimulating hormone (FSH) concentration in the early follicular phase of the menstrual cycle. METHODS: Fifty-two patients undergoing IVF treatment were included in the study. Blood samples were collected for assessment of basal E2, FSH, and inhibin-B levels. Transvaginal ultrasound of an unstimulated cycle was performed to determine the mean antral follicle count (AFC). Serum FSH concentration was measured again on day 10 for CCCT performance. RESULTS: The mean values of women's age, and basal and day 10 FSH levels were significantly higher in cancelled cycles than in the control group, whereas basal inhibin-B and AFC were significantly higher in the latter. The mean basal E2 concentration was similar in both groups. The results from the logistic regression analysis show that CCCT (cut-off point FSH > 12.5 mIU/ml; AUCROC = 0.90) was a better single predictor of poor ovarian response than AFC (AUCROC = 0.85) and inhibin-B (AUCROC = 0.79) with a correct prediction for CCCT (86.5%), antral follicle count (84.6%), and for inhibin-B (82.7%). CONCLUSIONS: In women with normal basal FSH level, the determination of E2 has no prognostic value for the outcome of poor responders. However, CCCT, AFC, and inhibin-B tests, when applied separately, produce good prognostic values. CCCT is the best single predictor of poor ovarian response, followed by antral follicle count and basal inhibin-B values. In spite of that, CCCT does not add significantly to the simpler AFC ultrasound test in the prediction of poor ovarian response.