Third Annual Pediatric Interagency Registry for Mechanical Circulatory Support (Pedimacs) Report: Preimplant Characteristics and Outcomes

Abstract

Background: The influence of ventricular assist device (VAD) use in the care of children with end-stage heart failure is growing rapidly through increasing numbers, new devices, expanding indications, and improving outcomes. Pediatric Interagency Registry for Mechanical Circulatory Support (Pedimacs), a National Institutes of Health-sponsored United States database, provides a platform to understand this emerging population.

Methods: Between September 19, 2012, and December 31, 2017, 30 hospitals implanted 508 devices in 423 patients aged younger than 19 years. This past year was one of evolution for the database as its management was transitioned to The Society of Thoracic Surgery; therefore, this report does not include data from institutions not under contract by August 1, 2018.

Results: Of the 423 patients, the diagnosis was cardiomyopathy in 261 (62%), myocarditis in 48 (11%), other in 28 (7%), and congenital heart disease (CHD) in 86 (20%), with 52 of these patients having single-ventricle physiology. The two most common support strategies included left VAD in 342 (81%) and biventricular assist device in 64 (15%). Positive outcome (alive on device or bridge to transplantation/recovery) was 80% at 6 months (overall mortality of 20%). The patient cohort for implantable continuous-flow (IC) pumps (n = 197; age at implant, 13.4 ± 3.8 years; 19% Interagency Registry for Mechanically Assisted Circulatory Support [Intermacs] profile 1, 21% intubated at implant, and 12% with CHD) was significantly different from the paracorporeal continuous-flow (PC) pump cohort (n = 79; age, 3.9 ± 5.2 years; 49% Intermacs profile 1, 86% intubated at implant, and 38% with CHD) and the paracorporeal pulsatile (PP) pump cohort (n = 121; age, 3.3 ± 3.9 years; 41% Intermacs profile 1, 77% intubated at implant, and 21% with CHD). Consistent with their cohort composition, device type positive outcomes at 6 months were PC, 63%; PP, 77%; and IC, 92%. Parametric hazard modeling showed an early hazard for death was associated with Intermacs profile 1, biventricular assist device, percutaneous devices, PC devices, small-volume institutions, low age, and low weight, whereas a constant hazard was associated with intubation and liver dysfunction at time of implant.

Conclusions: IC VADs are the most common VAD type placed in children. The positive outcomes for the IC VADs exceed 90% at 6 months, which may represent our field's maturation in both patient selection and timing of implantation. The PP/PC devices are currently limited to supporting our most challenging patients, those weighing less than 20 kg and those with CHD. The introduction of new devices and our communities' commitment to shared learning and improvement will lead to more pediatric lives saved by VAD support and will also focus on improving the quality of life of children supported with VADs.