100 Episodes of Support With Pulsatile Paracorporeal Ventricular Assist Device in 99 Neonates, Infants, and Children At a Single Institution.

Abstract

Unique challenges exist with providing ventricular assist device (VAD) support to neonates, infants, and children with congenital heart disease (CHD), functionally univentricular circulation, and those weighing <5 kg. We review our experience with all patients ever supported with Berlin Heart at University of Florida and assess risk factors for mortality and outcomes in high-risk subgroups. Overall outcomes of 99 patients who underwent 100 separate episodes of support with Berlin Heart were assessed; 47 patients weighed <5 kg. Patients with CHD (n =53) were compared with those with acquired heart disease (n = 45), and patients with functionally univentricular circulation (n = 40) were compared with those with biventricular circulation (n = 59). The primary outcome was mortality. Survival was modeled by the Kaplan-Meier method. Univariable Cox proportional hazard models identified prognostic factors for survival. Kaplan-Meier methods and log-rank tests were used to assess group differences in long-term survival. Risk factors for mortality were CHD (hazard ratio [HR], 2.25; 95% CI, 1.22-4.13; P = .009), liver dysfunction (HR, 3.46; 95% CI, 1.56-7.69; P = .002), and bleeding (HR, 3.91; 95% CI, 1.87-8.18; P = .0002). Biventricular heart (HR, 0.38; 95% CI, 0.18-0.8; P = .01) and biventricular assist device support (HR, 0.35; 95% CI, 0.16-0.76; P = .007) were protective. Overall survival at 1 and 5 years in all 99 patients was 74.5% (95% CI, 66.4%-83.7%) and 69.8% (95% CI, 61.1%-79.7%). Survival at 1 and 5 years in all 47 patients <5 kg was 63.4% (95% CI, 50.9%-78.9%) and 60.9% (95% CI, 48.3%-76.9%). Longitudinal survival is better in acquired heart disease vs CHD (log-rank P = .003) and biventricular vs univentricular patients (log-rank P = .026). Pulsatile VAD facilitates bridge to transplantation in neonates, infants, and children; however, survival is worse in smaller patients, those with CHD, and those with functionally univentricular circulation. These challenging subgroups represent important opportunities for improvement.