Abstract
OBJECTIVE: To compare the effect of P on live birth rate between hCG and GnRH agonist (GnRH-a) trigger cycles. DESIGN: Retrospective cohort study. SETTING: Large private assisted reproductive technology (ART) practice. PATIENT(S): A total of 3,326 fresh autologous ART cycles. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Live birth. RESULT(S): A total of 647 GnRH-a trigger cycles were compared with 2,679 hCG trigger cycles. Live birth was negatively associated with P in both the hCG trigger (odds ratio [OR] 0.62, 95% confidence interval [CI] 0.52-0.76) and the agonist trigger cohorts (OR 0.56, 95% CI 0.45-0.69). Interaction testing evaluating P and trigger medication was not significant, indicating that P had a similar negative effect on live birth rates in both cohorts. Progesterone ≥2 ng/mL occurred more commonly in GnRH-a trigger cycles compared with hCG trigger cycles (5.5% vs. 3.1%) and was negatively associated with live birth in both the hCG trigger (OR 0.28, 95% CI 0.11-0.73) and agonist trigger cohorts (OR 0.35, 95% CI 0.14-0.90). When P ≥2 ng/mL, the live birth rates were poor and similar in the hCG and GnRH-a cohorts (5.9% vs. 14.2%), indicating that P ≥2 ng/mL had a similar negative effect on live birth in both cohorts. CONCLUSION(S): Elevated serum P on the day of hCG was negatively associated with live birth rates in both hCG and GnRH-a trigger cycles.