[Serum anti-Müllerian hormone to predict ovarian response in assisted reproduction cycles]

PURPOSE: To compare serum anti-Mullerian hormone (AMH) levels on the seventh day of ovarian stimulation between normal and poor responders. METHODS: Nineteen women aged ≥35, presenting with regular menses, submitted to ovarian stimulation for assisted reproduction were included. Women with endometriosis, polycystic ovarian syndrome or those who were previously submitted to ovarian surgery were excluded. On the basal and seventh day of ovarian stimulation, a peripheral blood sample was drawn for the determination of AMH, FSH and estradiol levels. AMH levels were assessed by ELISA and FSH, and estradiol by immunochemiluminescence. At the end of the stimulation cycle patients were classified as normal responders (if four or more oocytes were obtained during oocyte retrieval) or poor responders (if less than four oocytes were obtained during oocyte retrieval or if the cycle was cancelled due to failure of ovulation induction) and comparatively analyzed by the t-test for hormonal levels, length of ovarian stimulation, number of follicles retrieved, and number of produced and transferred embryos. The association between AMH and these parameters was also analyzed by the Spearman correlation test. RESULTS: There was no significant difference between groups for basal or the seventh day as to AMH, FSH and estradiol levels. There was a significant correlation between seventh day AMH levels and the total amount of exogenous FSH used (p=0.02). CONCLUSIONS: AMH levels on the seventh day of the ovarian stimulation cycle do not seem to predict the pattern of ovarian response and their determination is not recommended for this purpose.