LAUSR

Background: This study examines our institutional ventricular assist devices (VADs) experience over two decades to understand trends towards predictors of mortality. Methods: Retrospective study of patients aged 0–21 years supported with a VAD from January 1996 to May 2015. Patient data was examined pre and post‐VAD implant among survivors and non‐survivors. Results: Thirty‐six patients identified (8 supported by Thoratec® VAD and 28 supported by EXCOR Berlin Heart®). Patient's diagnosis included dilated cardiomyopathy (DCM) (n = 19,53%), congenital heart disease (CHD) (n = 12,33%), and other (n = 5,14%). Median age and body surface area (BSA) were 1.0 years[0–7 years] and 0.41[0.24–0.92], respectively. Survival to discharge was 75% with no deaths with DCM. The survival rate for patients with CHD was 42%. Univariate analysis showed diagnosis of CHD, smaller BSA and respiratory failure post‐implant (Intermacs criteria) as risk factors for mortality. Median duration of VAD support was lower in non‐survivors, 14 vs 63 days (p = 0.03). Renal function at time of transplant or death was normal/pRIFLE Risk category in 20(74%) of survivors and 2(22%) of non‐survivors (p = 0.06). Post‐implant, peak total bilirubin in the first week trended lower in survivors (p = 0.06). Conclusions: Persistent end‐organ impairment in the first 2 weeks after VAD placement could be a useful prognostic marker for survival to transplant. HighlightsEtiology of heart failure is a key determinant in VAD survival.Post‐VAD complications were similar between survivors and non‐survivors in this predominantly BiVAD subset.Persistent end‐organ impairment in the first 2 weeks after VAD placement could be a useful prognostic marker.Among the 13 patients in renal failure (pRIFLE), the longest need for RRT in a survivor was 22 days post‐VAD placement.