Application of redefined human chorionic gonadotropin curves for the diagnosis of women at risk for ectopic pregnancy.

OBJECTIVE: To validate recently characterized curves of hCG rise and fall that are seen in viable and nonviable pregnancies in a population of women who have pregnancies of unknown location.
DESIGN: Historical cohort study.
SETTING: University hospital. PATIENT(S): One thousand two hundred forty-nine women with symptomatic early pregnancies. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Comparison of observed hCG values to predicted hCG values; time to diagnosis of ectopic pregnancy; accuracy of diagnosis. RESULT(S): Of the 1,249 patients included in this study, 196 had ectopic pregnancy (EP); 261, intrauterine pregnancy (IUP); and 792, spontaneous abortion (SAB). By determining the rate of change in hCG values from two consecutive visits and comparing them with a minimal rise of 35% in 2 days (the bound that is defined by the 99.9% confidence interval [CI] for the rise of hCG in an IUP) or a minimal fall of 21%-35% (the bound that is defined by the 90% CI for the fall of hCG in an SAB), we were able to make the diagnosis of EP an average of 2.5 days sooner than by standard clinical practice. Only 12% of patients had an EP go undiagnosed by using these rules, because the curve of rise or fall of their hCG mimicked that of a non-EP gestation. CONCLUSION(S): Recently redefined curves of rise and fall in hCG for IUP and SAB are valid for clinical use on the basis of our application to this large cohort of patients. Using them can shorten the time needed to make the diagnosis of EP. Use of a more conservative cutoff for minimal rise in hCG, one as slow as 35% over 2 days, to characterize a potentially viable gestation would minimize potential interruption of a desired pregnancy.