Objectives: Ventricular assist devices have gained popularity in the management of refractory heart failure in children listed for heart transplantation. Our primary aim was to compare the composite endpoint of all-cause pretransplant mortality and loss of transplant eligibility in children who were treated with a ventricular assist device versus a medically managed cohort.
Design: This was a retrospective cohort analysis.
Settings: Data were obtained from the Scientific Registry of Transplant Recipients.
Patients: The at-risk population (n = 1,380) was less than 18 years old, either on a ventricular assist device (605 cases) or an equivalent-severity, intensively medically treated group (referred to as MED, 775 cases).
Interventions: None.
Measurements and main results: The impact of ventricular assist devices was estimated via Cox proportional hazards regression (hazard ratio), dichotomizing 1-year outcomes to "poor" (22%: 193 deaths, 114 too sick) versus all others (940 successful transplants, 41 too healthy, 90 censored), while adjusting for conventional risk factors. Among children 0-12 months old, ventricular assist device was associated with a higher risk of poor outcomes (hazard ratio, 2.1; 95% CI, 1.5-3.0; p < 0.001). By contrast, ventricular assist device was associated with improved outcomes for ages 12-18 (hazard ratio, 0.3; 95% CI, 0.1-0.7; p = 0.003). For candidates 1-5 and 6-11 years old, there were no differences in outcomes between the ventricular assist device and MED groups (hazard ratio, 0.8 and 1.0, p = 0.43 and 0.9). The interaction between ventricular assist devices and age group was strongly significant (p < 0.001).
Conclusions: This is a comparative study of ventricular assist devices versus medical therapy in children. Age is a significant modulator of waitlist outcomes for children with end-stage heart failure supported by ventricular assist device, with the impact of ventricular assist devices being more beneficial in adolescents.