Abstract
BACKGROUND: With the development of assisted reproduction technology (ART), many indicators have been proposed to evaluate ovarian response, and then predict pregnancy outcomes. In general, the predictive values remain limited. OBJECTIVE: To further explore the indicators to evaluate ovarian sensitivity to gonadotrophin (Gn) stimulation more accurately. METHODS: This retrospective cohort study included 330 women who underwent an entire ART cycle. We aimed to assess whether a new index, termed as average Gn dosage per follicle, could be used as a marker for ovarian response and pregnancy outcomes. It was calculated as the ratio of total Gn dose during ovarian stimulation and the number of pre-ovulatory follicles (PFC) on the trigger day. Patients were divided into three subgroups according to the average Gn dosage per follicle: below the 33rd percentile (Group A), between 33rd and 67th percentiles (Group B), and above the 67th percentile (Group C). Then stimulation data, laboratory and clinical outcomes were compared among the groups. RESULTS: The results showed patients in Group A had the best ovarian response, the number of retrieved oocytes was significantly higher than in Group B and C. A multivariate regression analysis showed that average Gn dosage per follicle was an independent predictor of cumulative live birth rates (CLBRs) [adjusted odds ratio (OR): 0.96, 95% confidence interval (CI): 0.95-0.98, P < 0.01]. CONCLUSIONS: The present study showed that average Gn dosage per follicle appears to be a highly reliable index of ovarian response to exogenous Gn and can be useful to estimate CLBR.