Objective: To evaluate serial hormone concentrations in subjects treated with vaginally administered misoprostol for early pregnancy failure.
Design: As part of a randomized clinical trial, serum was collected on treatment days 1, 3, 8, and 15.
Setting: Multicenter clinical trial.
Patient(s): Women with a nonviable first-trimester pregnancy.
Intervention(s): Serum concentrations of human chorionic gonadotropin (hCG), progesterone, and sex hormone binding globulin (SHBG) were evaluated.
Main outcome measure(s): A logistic regression model was constructed to assess the associations of percent and complete expulsion of the gestational sac and/or successful management.
Result(s): The percent change from the day of treatment until the first follow-up visit was predictive for complete expulsion for progesterone (P) (P<.005) and hCG (P<.005), but not for SHBG. The actual value was not significantly associated with complete expulsion or successful management. A decrease (day 1-3) of 79% for both hCG and P was associated with a 90% probability of complete passage of the gestational sac. A 90% probability of successful management was noted if P decreased by 78% on day 3 or 59% on day 7, or hCG decreased by 74% on day 3 or 78% on day 7 compared with pretreatment values.
Conclusion(s): Percent change, but not absolute change, in serial hormone values are strongly associated with both the complete expulsion of the gestational sac with one dose of misoprostol and ultimate success.