Objective: Assess the prognostic value of ultrasound and Doppler parameters in severe preterm pre-eclampsia.Study design: Prospective cohort study in Cape Town, South Africa, involving 113 women with severe pre-eclampsia between 24 and 34 weeks of gestation and managed expectantly when suitable. Serial ultrasound and multivessel Doppler assessments were performed 2 to 3 times weekly. Observations included fetal weight estimation, growth pattern, amniotic fluid volume, appearance of the heart and bowel, pulsatility index of the uterine, umbilical and middle cerebral arteries and ductus venosus. Perinatal outcome (death or neurological compromise, n ¼ 12) and major morbidity (n ¼ 62) were related to findings at recruitment and on the last assessment preceding delivery.Results: In univariate analysis, fetal growth asymmetry and waveform analysis of all three fetal vessels significantly correlated with both end points. The initial fetal weight estimation with a cutoff weight of 1080 g had the highest sensitivity (83.3%) and negative predictive value (97.5%) for poor outcome (P<0.001) while the final ductus venosus pulsatility index had the highest specificity (92%) and positive predictive value (33%). The combination of a raised placentocerebral ratio and ductus venosus pulsatility index close to delivery had the highest (57.1%) positive predictive value. Logistic regression showed the best overall predictive model for poor outcome to be a combination of initial fetal weight estimation and final ductus venosus pulsatility index (overall accuracy 94.6%, RR 20.20 (7.36, 55.41)). The initial fetal weight estimation with a cutoff weight of 1283 g provided the best correct prediction of major morbidity (83.2% overall accuracy, sensitivity 79.0%, specificity 88.2%, positive and negative predictive values of 89.1 and 77.6% respectively). This was better than using the gestational age at recruitment or delivery, birth weight or any of the Doppler results. The prediction of morbidity was not improved by the addition of any other variable in logistic regression analysis.
Conclusion:For the short-term outcome measures assessed in this study, the estimated fetal weight at the time of diagnosis is the most important prognostic factor in severe pre-eclampsia with some additional value of ductus venosus assessment.