Impact of center volume on outcomes after ventricular assist device implantation in pediatric patients: An analysis of the STS-Pedimacs database

Abstract

To date, no pediatric studies have highlighted the impact of center's ventricular assist device (VAD) volumes on postimplant outcomes. Children (age <19) enrolled in Pedimacs undergoing initial left ventricular assist device implantation from 2012 to 2020 were included. Center volume was analyzed as a continuous and categorical variable. For categorical analysis, center volumes were divided as: low volume (1-15 implants), medium volume (15-30 implants), and high volume (>30 implants) during our study period. Patient characteristics and outcomes were compared by center's VAD volumes. Of 44 centers, 16 (36.4%) were low, 11 (25%) weremedium, and 17 (38.6%) werehigh-volume centers. Children at high-volume centers were least likely intubated, sedated, or paralyzed, and most likely ambulating preimplant (p<0.05 for all). Center's VAD volumes were not a significant risk factor for mortality postimplant when treated as a continuous or a categorical variable (p>0.05). Compared to low volume, children at high-volume centers had fewer early neurological events. Compared to medium volume, those at high-volume centers had fewer late bleeding events(p<0.05 for all). There were no significant differences in survival after an adverse event by hospital volumes (p>0.05). Althoughhospital volume does not affect post-VAD implant mortality, pediatric centers with higher VAD volumes have fewer patients intubated, sedated, paralyzed preimplant, and have lower adverse events. Failure to rescue was not significantly different between low, medium, and high-volume VAD centers.