Differential association of prenatal blood flow patterns with death and neurodevelopmental retardation in preterm infants with intrauterine growth restriction

Objectives:

To evaluate the association between fetal Doppler velocities with death and neurodevelopment in intrauterine growth restricted (IUGR) preterm infants (birth weight < 10^th percentile, umbilical artery (UA) pulsatility index (PI) > 95^th percentile).

Methods:

Retrospective single-center cohort study, including 176 live born IUGR infants 24 – 34 weeks of gestation. We assessed the association between UA (n = 156), middle cerebral artery (MCA, n = 146) and ductus venosus (DV, n = 106) blood flow measured within 48h before birth and death and neurodevelopmental retardation at one and two years' corrected age (Griffiths or Bayley scores one standard deviation below mean).

Results:

Twenty-one of 176 IUGR infants died (median [range] gestational age 25.6 [24.0 – 33.4] vs. 29.6 [24.6 – 34.0] weeks in survivors, p < 0.01). Death was associated with absent/reversed end-diastolic UA flow (11/16 nonsurvivors vs. 47/140 survivors, p = 0.0116) and absent/negative DV a wave (3/11 vs. 4/95, p = 0.02335), but these associations did not remain significant after controlling for gestational age. Neurodevelopmental retardation, observed in 38/101 infants examined, was not associated with gestational age but increased UA PI (2.21 [1.16 – 3.85] vs. 1.71 [0.82 – 5.73], p = 0.04) and decreased MCA PI (1.21 [0.9 – 1.82] vs. 1.39 [0.87 – 2.19], p = 0.01). After logistic regression analysis, the association was maintained for MCA PI (p = 0.019) but not for UA PI. Receiver operator characteristics curve analysis yielded an area of 0.685 (95% confidence interval 0.545 – 0.771) but no useful operational threshold.

Conclusions:

The associations between compromised UA and DV flow and extrauterine death are mediated by gestational age while an MCA brain sparing-like pattern is independently linked to neurodevelopmental retardation.