OBJECTIVE In gonadotropin-releasing hormone analogue-pretreated in vitro fertilization-embryo transfer cycles, pregnancy rates are inversely related to serum progesterone levels on the day of administration of human chorionic gonadotropin. The relationship of the progesterone concentration on other days in the periovulatory period to pregnancy rates in such cycles is little studied. We therefore retrospectively analyzed the relationship between progesterone concentrations on the day after human chorionic gonadotropin and pregnancy in 114 cycles, 28 and 23 of which produced clinical and ongoing/delivered pregnancies, respectively. To assess the effect of the extent of follicular luteinization on success, we also studied the relationship between the progesterone concentration per oocyte retrieved and pregnancy for the day of and day after human chorionic gonadotropin. RESULTS Progesterone concentrations on the day after human chorionic gonadotropin were inversely associated with clinical pregnancy by multiple logistic regression analysis (P < 0.05). Progesterone/oocyte ratios were inversely associated with clinical pregnancy (P < 0.05) and ongoing/delivered pregnancy (P < 0.02) for both the day of and the day after human chorionic gonadotropin. CONCLUSIONS The study results extend the window of time during which elevated progesterone concentration is associated with poor outcome to at least 2 days. This finding is consistent with hypothetical mechanisms attributing the link between progesterone concentration and outcome to either endometrial or follicle/oocyte events. The association of lack of follicular luteinization (low progesterone per oocyte ratios) and favorable outcome suggests a predominant effect of progesterone on follicle/oocyte quality. Further studies are needed to clarify the mechanisms underlying the association between progesterone and in vitro fertilization-embryo transfer outcome.