The study was conducted to determine whether the administration of mifepristone followed by vaginal misoprostol can induce an abortion in early pregnancy when no gestational sac is present on sonogram. This report presents a prospective, pilot study of 30 healthy adult women, pregnant and seeking an abortion, and with no gestational sac on sonogram. All women had a baseline serum chorionic gonadotropin (hCG) level measured prior to using mifepristone 200 mg orally followed by misoprostol 800 mcg vaginally 48 h later, and then returned up to 4 days later for a repeat sonogram and serum hCG level. Women with initial hCG levels > 2000 IU/L were evaluated for ectopic pregnancy. At the first follow-up visit, if the hCG decreased by >50%, the women were followed with home pregnancy (25 IU/L) tests weekly until negative. If the levels did not decrease by 50%, a second dose of misoprostol was given. Surgical intervention was indicated for persistent hCG levels or excessive bleeding. Of the 30 women enrolled, the mean number of days of amenorrhea was 40 (SD 9) days. Two women had surgical intervention for continuing pregnancy, 2 had ectopic pregnancies, and 1 was lost to follow-up. Complete medical abortions occurred in 25/30 (88%) women, but when recalculated, in 25/27 (93%) women who completed the protocol and who did not have an ectopic pregnancy. There was 1 adverse event in a woman with an ongoing pregnancy who then received methotrexate. She was hospitalized a day later with a complicated pelvic infection and likely methotrexate-induced pneumonitis. Twenty-three women had a decrease in hCG at first follow-up visit of >50%. All 27 women who completed the protocol found the overall regimen acceptable. Mifepristone followed at 48 h by vaginal misoprostol were effective and acceptable in inducing an abortion in very early pregnancy. There may be a higher incidence of failure in very early pregnancies. Documentation of a complete abortion by hCG level is necessary to ensure the pregnancy is neither ongoing nor ectopic.