Prognostic value of a serum β-human chorionic gonadotropin cutoff, twelve days after fresh embryo transfer, on predicting live birth among Ugandan women

Abstract

Objective: This study aimed to evaluate the prognostic value of a serum β-hCG level cutoff, 12 days after embryo transfer, on predicting live births among Ugandan women. Methods: This is a retrospective cross-sectional study. Three hundred thirty-seven fresh IVF cycles with serum β-hCG ≥5 mIU/mL, at 12 days after embryo transfer, were eligible. We abstracted participant characteristics, IVF cycle characteristics, live birth, clinical pregnancy, and ongoing pregnancy data from each eligible cycle. We utilized the Youden Index metric and the maximize_boot_metric method to link serum β-hCG levels to outcome data and determine the optimal cutoff values. Results: The optimal serum β-hCG cutoff value for predicting live birth was 437.42 mIU/mL with a corresponding sensitivity and false positive rate of 72% and 31%, respectively. The cutoffs for clinical and ongoing pregnancy were 239.58 mIU/mL and 353.66 mIU/mL, respectively. These corresponded with a sensitivity of 83% and 77%, respectively, and a false positive rate of 27% and 33%, respectively. The serum β-hCG cutoff had poor discriminatory performance for predicting live births but moderate performance for predicting clinical and ongoing pregnancies. Conclusion: A single serum β-hCG 12 days after cleavage embryo transfer has poor discriminatory performance in predicting live birth, albeit performing modestly in predicting clinical pregnancy and ongoing pregnancy among Ugandan women.