Preoperative prealbumin does not impact outcomes after left ventricular assist device implantation.

Abstract

This single-center, the retrospective study evaluates the impact of preoperative serum prealbumin levels on outcomes after left ventricular assist device (LVAD) implantation. Adults undergoing LVAD implantation, with a recorded preoperative prealbumin level, between 2004 to 2018 were included. Primary outcomes included rates of 1-year survival and secondary outcomes included rates of postimplant adverse events. Threshold regression and restricted cubic splines were utilized to identify a cut-point to dichotomize prealbumin level. Prealbumin was also evaluated as a continuous variable. Multivariable logistic regression was used for risk-adjustment. A total of 333 patients were included. Patients were dichotomized according to an optimal prealbumin threshold of 15 mg/dL: 47.4% (n = 158) had levels below and 52.6% (n = 175) had levels above this threshold, respectively. The rates of postimplant adverse events, including bleeding, infection, stroke, renal failure, and right heart failure, were similar between the groups (all P > .05). Furthermore, the rates of cardiac transplantation and device explantation were also similar (all P > .05). Unadjusted survival was comparable between the groups at 30-days, 90-days, and 1-year following LVAD implantation (all P > .05). In addition, lower prealbumin did not impact risk-adjusted 1-year mortality when modeled either as a categorical (OR, 1.08; 95% CI, 0.48-2.12; P = .82) or continuous variable (OR, 1.99; 95% CI, 0.73-2.34; P = .96). This study demonstrates that lower prealbumin levels were not predictive of increased post-LVAD morbidity or mortality. Although an established marker of nutritional and inflammatory status, the role of prealbumin in patient selection or prognostication appears limited in LVAD patients.