The objective of this study was to investigate the clinical utility of umbilical artery Doppler velocimetry in the management of very premature growth-retarded fetuses. The study comprised 34 fetuses between 26 and 34 weeks' gestation, diagnosed ultrasonically as suffering from intrauterine growth retardation. Based on the umbilical artery Doppler findings, patients were divided into 3 groups: group 1 (10 fetuses) with normal Doppler systolic/diastolic (S/D) ratios (within 2 SD of the mean for gestational age); group 2 (9 fetuses) with significant abnormal umbilical artery S/D ratios (above 2 SD of the mean for gestational age), and group 3 (15 fetuses) with absent or reverse end-diastolic flow. Fetal outcome was assessed in terms of neonatal mortality and morbidity, i.e. low umbilical artery pH (pH < 7.20) and Apgar scores. In group 1, all fetuses had a stable, normal S/D ratio; in group 2, 2 fetuses (22%) showed deterioration in the Doppler findings and were transferred to group 3, while in group 3, 3 of 15 fetuses (20%) showed improvement in the absence of end-diastolic flow and were thus transferred to group 2. Emergency cesarean section due to antenatal deterioration of the biophysical profile occurred in only 1 patient (10%) in group 1, compared to 33 and 87% in groups 2 and 3, respectively (p < 0.05). The other 9 patients in group 1 had normal vaginal deliveries at 36-37 weeks' gestation, with no perinatal complications. The mean expectant interval from admission to delivery in group 3 was 8.2 days, while in groups 1 and 2 it was 23.8 and 22.2, respectively (p < 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)