The incidence of spontaneous abortion once embryonic development has reached the stage of sonographically demonstrable cardiac activity has not been precisely determined in a large population. This retrospective study was designed to determine the risk of pregnancy loss after first trimester sonographic confirmation of fetal cardiac activity. A total of 840 patients were referred for ultrasound examination to the Perinatal Center, University of Cincinnati, between January 1, 1979 and December 31, 1985 who fulfilled the criteria of singleton pregnancy with sonographically visible fetal cardiac activity and crown-rump length consistent with gestational age less than or equal to 12 weeks. Patients were grouped into two categories based on the absence (control) or presence of vaginal bleeding (bleeding) prior to or at the time of the ultrasound examination. The abortion rates for the two groups were 5.2 and 16.4%, respectively (P less than 0.001). The relative risk of abortion in patients with vaginal bleeding was nearly four-fold greater than the control population. In the control group there was a significantly greater incidence of pregnancy loss in the greater than or equal to 34 years age group compared to less than 34 years age patients (4.4 versus 11.1%, P less than 0.05). Although there was no significant effect of race on abortion rates, low socioeconomic status almost doubled the relative risk of miscarriage. We conclude that the presence of visible fetal cardiac activity in the first trimester predicts a decreased risk for spontaneous abortion compared to generally reported rates. The significant risk factors for increased chance of abortion are vaginal bleeding, advanced maternal age, and low socioeconomic status.