THE ROLE OF DOPPLER VELOCIMETRY IN THE MANAGEMENT OF HIGH-RISK PREGNANCIES

Objective To determine whether knowledge of the result of Doppler velo cimetry of the umbilical artery is beneficial to the management of a h igh risk pregnancy. Design Randomised controlled trial. The trial was of the management type, designed to assess benefit accruing from addit ional information supplied by Doppler velocimetry. Setting Tygerberg H ospital, Cape Town, South Africa, The hospital serves a population fro m the lower socio-economic groups. Subjects Women with pregnancies 28 or more weeks gestation with hypertensive diseases and/or suspected sm all for gestational age fetuses were referred for Doppler velocimetry. From this population, three subsets were formed: 1. those with fetuse s with absent end-diastolic velocities (20 fetuses); 2. those with hyp ertension but with fetuses with end-diastolic velocities (89 fetuses); and 3. those with fetuses suspected of being small for gestational ag e but with end-diastolic velocities (104 fetuses). Interventions Doppl er velocimetry on all subjects. The study group consisted of 10 cases with absent end-diastolic velocities, 47 cases with hypertensive disea ses with end-diastolic velocities and 51 cases with suspected small fo r gestational age fetuses but with end-diastolic velocities in which t he result was revealed to the clinician. The control group consisted o f 10, 42 and 53 cases, respectively, in which the Doppler results were not revealed. All other routine investigations (sonar and antenatal f etal heart rate monitoring) were available to the clinicians. Standard management protocols were followed in all groups. Main outcome measur es Perinatal mortality and morbidity, antenatal hospitalisation, mater nal intervention, admission to the neonatal intensive care unit and ho spitalisation until discharge from the neonatal wards. Results In the study and control groups the gestational age at entry to the study, ma ternal age, parity and various complications were not significantly di fferent. In the subset with absent end-diastolic velocities, there was one neonatal death in the study group, but in the control group there were six deaths, five intrauterine and one perinatally related infant death (P = 0.029). Because of this significant finding, the study was stepped. There were no differences in outcome in the subset where the re was hypertensive disease with end-diastolic velocities between the study and control groups. In the subset in which small for gestational age fetuses were suspected, but in which end-diastolic velocities wer e present, the women in the study group had significantly fewer days i n hospital before delivery (P < 0.001) and tended to have fewer matern al interventions (study group = 27%, control group = 43%; P = 0.07; od ds ratio (OR) 0.49, 95% confidence limits (CL) 0.2 and 1.25) and caesa rean sections (study group = 13%, control group = 27%; P = 0.08; OR 0. 43, 95% CL 0.14 and 1.32). The infants of the study group in this subs et also spent significantly less time in the neonatal wards (P = 0.029 ).