Subclassification of indeterminate pelvic ultrasonography: prospective evaluation of the risk of ectopic pregnancy.

Abstract

We sought to determine the frequency of ectopic pregnancy among subclasses of indeterminate ultrasonographic examinations. A prospective observational study was performed from January 1, 1995, to August 31, 2000, on consecutive emergency department patients in the first trimester of pregnancy with a chief complaint of abdominal pain or vaginal bleeding and who had an indeterminate transvaginal ultrasonographic examination at the time of the ED visit. Patients were excluded if lost to follow-up. Ultrasonographic examinations were subclassified into 5 groups (ie, empty uterus, nonspecific fluid, echogenic material, abnormal sac, normal sac) on the basis of a previously published classification system. Patients were followed up until the diagnosis of ectopic pregnancy was either confirmed or excluded. The frequencies of ectopic pregnancy, along with 95% confidence intervals (CIs), were calculated for each of the subclasses. The relative risk of ectopic pregnancy was calculated when appropriate. Seven hundred eighty patients with indeterminate ultrasonographic examinations were identified. One hundred forty-five were lost to follow-up, and therefore, 635 were enrolled. The frequency of ectopic pregnancy for each subclass is as follows: empty uterus, 36 of 259 (13.9%; 95% CI 10.1% to 18.5%); nonspecific fluid, 6 of 127 (4.7%; 95% CI 1.9% to 9.6%); echogenic material, 4 of 93 (4.3%; 95% CI 1.4% to 10.5%); abnormal sac, 0 of 103 (0%; 95% CI 0.0% to 2.9%); and normal sac, 0 of 53 (0%; 95% CI 0.0% to 5.5%). The relative risk of ectopic pregnancy in patients with an empty uterus versus in those without an empty uterus was 5.2 (95% CI 2.6 to 10.2). In our sample, patients with an empty uterus at ultrasonography had the highest frequency of ectopic pregnancy, with a relative risk of ectopic pregnancy 5 times greater than that of the other 4 subclasses.