Which factors are most predictive for live birth after in vitro fertilization and intracytoplasmic sperm injection (IVF/ICSI) treatments? Analysis of 100 prospectively recorded variables in 8,400 IVF/ICSI single-embryo transfers

Objective To construct a prediction model for live birth after in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI) treatment and single-embryo transfer (SET) after 2 days of embryo culture. Design Prospective observational cohort study. Setting University-affiliated private infertility center. Patient(s) SET in 8,451 IVF/ICSI treatments in 5,699 unselected consecutive couples during 1999–2014. Intervention(s) A total of 100 basal patient characteristics and treatment data were analyzed for associations with live birth after IVF/ICSI (adjusted for repeated treatments) and subsequently combined for prediction model construction. Main Outcome Measure(s) Live birth rate (LBR) and performance of live birth prediction model. Result(s) Embryo score, treatment history, ovarian sensitivity index (OSI; number of oocytes/total dose of FSH administered), female age, infertility cause, endometrial thickness, and female height were all independent predictors of live birth. A prediction model (training data set; n = 5,722) based on these variables showed moderate discrimination, but predicted LBR with high accuracy in subgroups of patients, with LBR estimates ranging from 40%. Outcomes were similar in an internal validation data set (n = 2,460). Conclusion(s) Based on 100 variables prospectively recorded during a 15-year period, a model for live birth prediction after strict SET was constructed and showed excellent calibration in internal validation. For the first time, female height qualified as a predictor of live birth after IVF/ICSI.