ABSTRACT: Women who present with first-trimester vaginal bleeding often undergo evaluation for ectopic pregnancy (EP). After a confirmatory pregnancy test, women who show no ultrasound evidence of an intrauterine pregnancy or an EP are said to have a pregnancy of unknown location (PUL). It is estimated that 7% to 20% of women who initially have a PUL will have a final diagnosis of EP. Practitioners evaluating a PUL often face a difficult management decision: premature intervention can result in termination of a viable pregnancy, but delayed intervention may lead to rupture of an untreated EP. In women with a PUL, 2 serial human chorionic gonadotrophin (hCG) measurements within the first 48 hours after initial presentation are commonly used to help predict the final diagnosis. Because initial outcome predictions based on 2 hCG values are often incorrect, many practitioners have patients with a PUL return 2 to 5 days after the second hCG measurement for an additional hCG evaluation. At present, there is no standard of care regarding time and frequency of hCG testing in women with a PUL.
The aim of this retrospective multicenter cohort study was to determine whether additional hCG values beyond the first 48 hours can improve diagnostic accuracy in women with a PUL. A total of 646 women with a PUL were recruited over 2 years at 3 academic centers in the United States; 146 of these women were ultimately diagnosed with EP. Participants presented to the emergency room with pain or bleeding in the first trimester, had a PUL on ultrasound, at least 2 hCG values, and a definitive final diagnosis after follow-up. Using standard clinical prediction rules, addition of a third hCG evaluation on day 4 after initial presentation significantly improved the accuracy of initial prediction from the first 2 hCG values by 9.3% (P = 0.015). Similarly, adding a third value on day 7 significantly improved the predicted diagnosis by 6.7% (P = 0.031). Assessing 4 hCG values (days 0, 2, 4, and 7) compared with 3 values (days 0, 2, and 4) produced a small insignificant improvement (1.3%) in the predicted diagnosis.
The data show that measurement of hCG values 48 hours (2 days) apart does not optimize the accuracy of diagnosis in women with a PUL. Adding a third hCG measurement on day 4 or 7 significantly improves the predicted diagnosis for 1 in 15 women. Although these data provide useful information for the prediction of outcomes for women with a symptomatic first-trimester PUL, the results may not be generalizable to all pregnant women.