Serum human chorionic gonadotropin measurement in the diagnosis of ectopic pregnancy when transvaginal sonography is inconclusive.

OBJECTIVE: To assess the accuracy of initial and repeated serum hCG measurements in the diagnosis of ectopic pregnancy (EP) in patients in whom transvaginal sonography is inconclusive and to evaluate whether patient characteristics influence the accuracy of serum hCG measurements.
DESIGN: Prospective study.
SETTING: Two large teaching hospitals in Amsterdam, the Netherlands. PATIENT(S): Three hundred fifty-four consecutively seen pregnant patients with suspected EP and inconclusive transvaginal sonographic findings. INTERVENTION(S): Serum hCG measurements. MAIN OUTCOME MEASURE(S): The performance of repeated serum hCG measurements in the diagnosis of EP was evaluated through the analysis of receiver operating characteristic curves. RESULT(S): Initial serum hCG measurements were more diagnostic in conjunction with sonographic evidence of an ectopic mass or fluid in the pouch of Douglas than in the absence of sonographic abnormalities. On repeated measurement, the course of the serum hCG concentration provided more diagnostic information than did the absolute serum hCG concentration 2 and 4 days after the start of the diagnostic process. CONCLUSION(S): The interpretation of serum hCG measurements should depend on additional findings at transvaginal sonography. A cutoff level of 1,500 IU/L is recommended for patients with an ectopic mass or fluid in the pouch of Douglas; in patients without these findings, the cutoff level should be at least 2,000 IU/L. Four days after the start of the diagnostic process, any rise in the serum hCG concentration makes the diagnosis of EP very likely.