engleski

Should serum anti-Müllerian hormone (AMH) or antral follicle count (AFC) be used for prediction of oocyte yield in patients with AMH-AFC discordances?

STUDY QUESTION Wheather AMH or AFC should be used as a predictor of oocyte yield and in ovarian stimulation protocol planning in patients with discordances between measured AMH levels (AMHm) and AMH expected (AMHe) according to AFC (AMH-AFC discordances)? SUMMARY ANSWER It was demonstrated that AMH was superior to AFC as a predictor of oocyte yield, in patients with AMHm higher than AMHe. Conversely, in patients having AMHm lower than assessed by AFC, AFC should be used for IVF stimulation protocol planning. WHAT IS KNOWN ALREADY AFC represents a gonadotrophin responsive follicular cohort measuring 2-9 mm. AMH is predominantly secreted by preantral, ultrasonically invisible, and small antral follicles. AMH and AFC are both used for prediction of ovarian response. Although they are strongly related to each other, AMH-AFC discordances are encountered in certain number of patients. In such ambiguous circumstances, it is not clear to which ovarian response predictor the clinician should rely on when deciding on the IVF treatment strategy. STUDY DESIGN, SIZE, DURATION This is a retrospective study on medical records of 1088 patients aged 20-44 years who underwent the first IVF cycle with GnRH antagonist protocol between October 2010 and July 2012. AMHe was calculated as: Log(AMHe)=-0.2616+1.2089Log(AFC). AMH-AFC discordance was defined as AMHm more than residual standard deviation away from AMHe. PARTICIPANTS/MATERIALS, SETTING, METHODS Participants with AMH-AFC discordances (n=281) were devided in two groups: DG1 (n=143) and DG2 (n=138) comprising patients with AMHm>AMHe and AMHm<AMHe, respectively. Univariate and multivariate linear regression analysis were applied to asses the value of AMH and AFC for the prediction of oocyte yield in each group. MAIN RESULTS AND THE ROLE OF CHANCE AMH and AFC were found to be predictive for oocyte yield on univariate analysis in both study groups (R2= 0.198, P<0.001 and R2= 0.163, P<0.001 in DG1 for AMH and AFC, respectively ; R2= 0.300, P<0.001 and R2= 0.305, P<0.001 in DG2 for AMH and AFC, respectively). However, using multivariate regression analysis (MRA) with AMH and AFC as independent variables, significant predictive value was demonstrated only for AMH in DG1 (R2= 0.104, P<0.001) and for AFC in DG2 (R2= 0.314, P<0.001). The same was found even when MRA was applied after stratification by initial gonadotrophin dose and age. LIMITATIONS, REASONS FOR CAUTION The main limitations of this study are the retrospective design and relatively low number of patients in both study groups. The analysis included only IVF patients treated with the GnRH antagonist protocol and cannot be extrapolated to other ovarian stimulation protocols. WIDER IMPLICATIONS OF THE FINDINGS AFC and AMH are the most significant predictors of the number of oocytes retrieved. However, in patients in whom discordance between AMH results and AFC is observed, it is important to establish which test should be used for prediction of ovarian response and for the decision on starting dose of gonadotrophins. This study results, therefore, provide additional information which could improve optimization of ovarian stimulation strategy in IVF cycles. STUDY FUNDING/COMPETING INTEREST(S) No funding was used since all data were collected during the routine infertility work-up. The authors have no competing interest(s) to declare. TRIAL REGISTRATION NUMBER Not applicable. KEYWORDS AFC, AMH, GnRH antagonist, prediction of oocyte yield CATEGORY Clinical science TOPIC Reproductive endocrinology

AFC; AMH; GnRH antagonist; prediction of oocyte yield

nije evidentirano