The incidence of ectopic pregnancy in the United States is increasing annually and is approaching epidemic proportions. Serum progesterone measurements show promise for the early detection of abnormal gestational development. Unfortunately, examination of serum progesterone levels does not allow discrimination between an ectopic pregnancy and an intrauterine abortion. Additionally, discriminatory levels of progesterone fail to include all abnormal gestations in the abnormal group (false-negative result) and classify some normal pregnancies to the abnormal group (false-positive result). As such, no discriminatory level of progesterone has been identified that can diagnose abnormal gestations. Instead, the value of progesterone appears to be its ability to predict a gestational complication, in much the same fashion as human chorionic gonadotropin doubling times. The advantage of progesterone appears to be that a single test of serum progesterone during the first 8 weeks of pregnancy is as predictive as serial human chorionic gonadotropin measurements. Clearly, future investigations into early gestational complications will focus on testing procedures that can discriminate between an intrauterine first-trimester demise and an ectopic pregnancy. Medical treatment for ectopic pregnancy is dependent on early diagnosis and is not yet widespread. Prostaglandin F2 alpha and methotrexate continue to be effective agents and demonstrate approximately 95% efficacy. Reproductive performance after medical therapy is comparable to reproductive performance after surgical therapy, but the duration of time until cure is longer in patients treated medically than in those treated surgically. Nomograms for human chorionic gonadotropin surveillance after surgical treatment have now been described, but not such data exists for surveillance after medical therapy.(ABSTRACT TRUNCATED AT 250 WORDS)