Utility of low-sensitivity and high-sensitivity urine pregnancy tests as indicators of gestational sac expulsion after medical abortion

Abstract

Introduction: We performed this trial to ascertain the accuracy of low-sensitivity (LS) and high-sensitivity (HS) urine pregnancy tests in determining gestational sac expulsion following treatment with mifepristone and misoprostol for medical abortion.Materials and Methods: This planned substudy was part of a multicenter medical abortion trial in women up to 63 days of gestation. One thousand eighty subjects received 200 mg of mifepristone po and were randomly assigned to take 800 ��g of misoprostol vaginally either 6���8 or 23���25 h later. Subjects returned 7 (��1) and 14 (��2) days after initiating treatment. A transvaginal ultrasound to determine the presence of a persistent gestational sac as well as HS and LS urine pregnancy tests were performed at each visit. We calculated the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of each urine pregnancy test.Results: The sensitivities, specificities, PPVs and NPVs with 95% CIs are as follows:First follow-up visit���LS test: sensitivity, 95.7 (87.3���100); specificity, 13.1 (10.8���15.5); PPV, 3.1 (1.8���4.3); NPV, 99.1 (97.2���100). HS test: sensitivity, 90.5 (77.9���100); specificity, 6.4 (4.7���8.1); PPV, 2.5 (1.4���3.6); NPV, 96.2 (91.1���100).Second follow-up visit���LS test: sensitivity, 78.9 (60.6���97.3); specificity, 38.5 (34.6���42.4); PPV, 4.0 (2���5.9); NPV, 98.3 (96.6���100). HS test: sensitivity, 83.3 (66.1���100); specificity, 33.7 (29.9���37.5); PPV, 3.7 (1.9���5.5); NPV, 98.5 (96.8���100).Discussion: The LS and HS urine pregnancy tests showed inconsistent sensitivities and specificities. The PPV was too low to accurately identify women who required additional treatment. The high NPV of both tests indicates the accuracy of a negative test result in detecting expulsion of the gestational sac after medical abortion. Introduction: We performed this trial to ascertain the accuracy of low-sensitivity (LS) and high-sensitivity (HS) urine pregnancy tests in determining gestational sac expulsion following treatment with mifepristone and misoprostol for medical abortion. Materials and Methods: This planned substudy was part of a multicenter medical abortion trial in women up to 63 days of gestation. One thousand eighty subjects received 200 mg of mifepristone po and were randomly assigned to take 800 ��g of misoprostol vaginally either 6���8 or 23���25 h later. Subjects returned 7 (��1) and 14 (��2) days after initiating treatment. A transvaginal ultrasound to determine the presence of a persistent gestational sac as well as HS and LS urine pregnancy tests were performed at each visit. We calculated the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of each urine pregnancy test. Results: The sensitivities, specificities, PPVs and NPVs with 95% CIs are as follows:First follow-up visit���LS test: sensitivity, 95.7 (87.3���100); specificity, 13.1 (10.8���15.5); PPV, 3.1 (1.8���4.3); NPV, 99.1 (97.2���100). HS test: sensitivity, 90.5 (77.9���100); specificity, 6.4 (4.7���8.1); PPV, 2.5 (1.4���3.6); NPV, 96.2 (91.1���100).Second follow-up visit���LS test: sensitivity, 78.9 (60.6���97.3); specificity, 38.5 (34.6���42.4); PPV, 4.0 (2���5.9); NPV, 98.3 (96.6���100). HS test: sensitivity, 83.3 (66.1���100); specificity, 33.7 (29.9���37.5); PPV, 3.7 (1.9���5.5); NPV, 98.5 (96.8���100). Discussion: The LS and HS urine pregnancy tests showed inconsistent sensitivities and specificities. The PPV was too low to accurately identify women who required additional treatment. The high NPV of both tests indicates the accuracy of a negative test result in detecting expulsion of the gestational sac after medical abortion.