Vaginal bleeding and early pregnancy outcome in an infertile population.

Abstract

Our purpose was to determine the prognostic value of vaginal bleeding in early pregnancy outcome as well as to analyze the role of cardiac activity in predicting pregnancy viability in the presence of vaginal bleeding. This was a cohort study of pregnancies obtained using either assisted reproductive technology (ART) or routine infertility treatment (RIT). Two hundred twenty-eight pregnant women were divided into two groups based on the presence or absence of vaginal bleeding. Successive measurements of beta-hCG levels were obtained every 2 days, starting on day 14 after ovulation or embryo transfer. All pregnancies underwent weekly transvaginal ultrasound (UTZ) examinations beginning on day 21. The occurrence of vaginal bleeding was monitored weekly. Seventy of the 228 patients (31%) had bleeding in early pregnancy, resulting in 31 (44%) pregnancy losses. Only 22 pregnancy losses (14%) were observed in 158 patients who did not have bleeding (P < 0.001). The abortion rate for the bleeding versus nonbleeding groups was 35 and 9%, respectively (P < 0.001). Vaginal bleeding was associated with a higher abortion rate in pregnancies following RIT than ART (51 vs 8%; P < 0.001). Fetal cardiac activity was noted by vaginal ultrasound in 189 patients. In this subpopulation, bleeding was also associated with a higher abortion rate than that in the nonbleeding group (17 vs 4%; P < 0.001). However this higher incidence was observed only in pregnancies following RIT, not ART (28 vs 5%; P < 0.001). Although bleeding significantly decreased the chance of a normal pregnancy outcome, more than 50% of the pregnancies did progress to term. The presence of cardiac activity in this population as a sign of fetal viability offered a better pregnancy prognosis. However, the predictive value of fetal cardiac activity was reduced in the presence of vaginal bleeding in an infertile population treated with RIT.