Readmission to the Pediatric Cardiac Intensive Care Unit in Pediatric Patients with Durable Ventricular Assist Devices

Abstract

Purpose Pediatric patients implanted with durable ventricular assist devices (VADs) are managed in the pediatric intensive care unit (PICU) after implant but are eligible to be transferred to the ward once stable. Currently there is no literature regarding the impact of readmission to the PICU in this patient population. Therefore, the primary aim of this study was to characterize readmissions to the PICU, identifying possible risk factors for readmission and determining whether readmission is associated with increased mortality. Methods This was a retrospective study of all patients Results There were 44 patients who underwent durable VAD implantation during the study period. The median age of implant was 3.7 yrs (IQR 0.64-8.9), 57% were males and the most common etiology was cardiomyopathy (53%). The median time of VAD support was 110 days (IQR 42.3-212.3) with the median index ICU stay being 34 days (IQR 19.8-81). Thirty patients (68%) were discharged to the ward with 18 (60%) having at least 1 readmission with 46 total readmissions. The median time to first readmission was 18 days (IQR 14.8-110.3). The median number of readmissions per patient was 2 (IQR 1-3) with a readmission rate of 0.71 per 100 patient days of VAD support. The most common cause of readmission was pump thrombosis (34%) followed by neurologic dysfunction (23%). There were no statistically significant pre or post implant factors identified to be associated with readmission. Readmission was also not associated with mortality (p=0.6), with the majority of all deaths (7/8) occurring before ICU discharge. Pre-implant dialysis, ECMO, and bilirubin were all univariate factors associated with mortality (p Conclusion Readmissions to the PICU occurred in over 50% of patients on durable VAD support with first readmission occurring on average within the month post discharge. While, there were no clear factors identified associated with readmission, a sicker pre-implantation clinical status was associated with mortality.