OBJECTIVE: To evaluate the effectiveness of prenatal prevention of Hb Bart's hydrops fetalis. STUDY DESIGN: The study was a prospective descriptive analysis of pregnant women attending an antenatal clinic between June 1990 and June 1998. The study consisted of two periods, the first half with no prenatal diagnosis (PND) (1990-1994) and the second half with PND. During the study period, all cases of lib Bart's hydrops fetalis were prospectively collected and postnatally confirmed. In the second period, prenatal strategy to control severe thalassemia was introduced. The strategy included (1) carrier identification by retrospective (history review for known risk) and prospective screening (simple erythrocyte osmotic fragility test) in women without known risks, (2) the couples at risk were offered genetic counseling and cordocentesis, (3) analysis of fetal blood for diagnosis, and (4) counseling for termination of pregnancy. RESULTS: During the first half of the study, the prevalence of Hb Bart's hydrops fetalis was 0.30% (89 in 29,399 deliveries). There were no fetuses with Hb Bart's hydrops fetalis among 16,360 screened pregnancies in the second half. However, of 6,856 pregnancies in the second half not screened due to a late first visit, 10 (0.15%) fetuses had Hb Bart's hydrops fetalis. Among the screened group, cordocentesis was performed in 361 pregnancies at risk, 170 and 191 from retrospective and prospective screening, respectively; and 75 (20.8%) were proven to have Hb Bart's disease, which was diagnosed before hydropic changes occurred. CONCLUSION: The strategy proved effective in preventing Hb Bart's hydrops fetalis, and extensive experience with it suggests that it be considered an effective way to control severe thalassemia.