Very low-birth-weight infants with congenital cardiac lesions: is there merit in delaying intervention to permit growth and maturation?

BACKGROUND

Low birth weight and prematurity and are known risks for mortality in congenital heart lesions. It is not known whether risks of delayed intervention are offset by benefits of growth and maturation. We explored this question.

METHODS

All 1618 infants admitted to our institution within 30 days after birth for a congenital heart defect since 2000 were analyzed. Birth details and admission progress notes were detailed on all. For infants requiring cardiac interventions, clinical conference records and progress notes enabled their management to be classified as either USUAL (normal timing and mode of intervention) or DELAYED (intentional delay for growth/maturation). The survival implications of birth weight and prematurity were examined via parametric multiphase methodology with bootstrap resampling. Subsequently, the impact of DELAYED management was sought in propensity-adjusted and multivariable time-related models.

RESULTS

Low birth weight is a strong, robust and independent predictor of death within the first year of life (P < .0001; 99.6% bootstrap resamples). The relationship is nonlinear with an inflection point at approximately 2.0 kg, below which decrements in survival are increasingly pronounced. Prematurity is also associated with poor outcome but less reliably so (P < .0001; 53% resamples); its variance appears partially mitigated by colinearity with multiple factors including diagnosis and chromosomal aneuploidy. Of the 149 infants with birth weight less than 2.0 kg (highest risk and most likely to receive delayed care in this cohort), care was USUAL in 34 and DELAYED in 46. The remaining children received comfort care only (27), were not considered for intervention owing to severe noncardiac problems (12) or were routinely observed for nonurgent lesions (30). Survival between the children weighing less than 2.0 kg and receiving USUAL or DELAYED care was identical (78% ± 2% at 1 year; P = .88), even when adjusted via propensity score (P = 0.65) or multivariable analysis (P = 0.55). Major determinants of death in this very low-birth-weight population were antenatal diagnosis (P = .01), presence of congenital gastrointestinal defects (P = .07), or lesion type (all higher risk: anomalous pulmonary venous drainage, P = .03; pulmonary atresia and intact septum, P = .05; and truncus, P = .01).

CONCLUSIONS

For very low-birth-weight neonates (<2.0 kg) with congenital heart defects, imposed delays in intervention neither compromise nor improve survival. Other factors instead appear to account for survival differences, including lesion type, associated noncardiac congenital defects, and antenatal diagnosis.