Ductus venosus blood flow velocity waveforms as a predictor for fetal outcome in isolated congenital heart disease

OBJECTIVE: To investigate whether pulsatility of ductus venosus (DV) flow velocity waveforms is of diagnostic value in predicting survival in fetuses with congenital heart disease (CHD). METHODS: In a cross-sectional study, Doppler investigation of DV and umbilical artery blood flow was performed in 58 fetuses with isolated structural CHD, without other sonographically detectable structural or chromosomal abnormalities or tachyarrhythmia. The pulsatility index for veins of DV (DV-PIV) waveforms was expressed as multiples of the 95th centile (Mo95th) of the reference ranges for gestational age. The PIV was related to intrauterine and neonatal mortality within the first 6 months of life. Terminations of pregnancies and neonates with additional abnormalities detected after birth were excluded from statistical analysis. For statistical analysis, the different types of heart defects were separated into atrial and/or ventricular (AV) septal defects, right or left ventricular in- and outflow tract abnormalities and others. RESULTS: After exclusion of 9 pregnancies (2 cases with failure of measurements of DV-PIV, 2 neonates with additional malformations, and 5 terminations of pregnancies), 49 cases were available for statistical analysis. The umbilical artery pulsatility index was within normal ranges in all but 1 case with AV canal and hydrops. In 7 pregnancies intrauterine fetal deaths occurred and 6 of them were hydropic. The median gestational age at birth for liveborn neonates was 39.0 weeks (range 27.8-41). There were 6 postnatal deaths, all but 1 within 28 days of delivery. The remaining fetuses survived for at least 6 months. The overall mortality rate was 27% (13/49). The DV-PIV was significantly higher in non-survivors than in survivors (median of Mo95th and interquartile ranges 1.48 (1.04-1.95) vs. 0.81 (0.70-1.15); p = 0.01). Analysis of subgroups showed significant differences for AV septal defects and abnormalities affecting predominantly the right ventricle (p = 0.046 and 0.043, respectively). Ten out of 13 non-survivors showed an abnormal DV-PIV (sensitivity 77%) as compared to 12 out of 36 survivors (specificity 67%). All hydropic fetuses (n = 6) showed an abnormal DV-PIV and ended in intrauterine deaths. CONCLUSIONS: Evaluation of the DV pulsatility is a useful additional variable for predicting the risk for mortality in fetuses with isolated structural CHD, in particular in fetuses with defects of the AV septum and with defects affecting predominantly right ventricular function. As there is no fetal hydrops without abnormal DV, this is another sign for the association of DV and cardiac failure.