Aortic Insufficiency After Left Ventricular Assist Device Implantation: Predictors and Outcomes

Background. The aim of this study was to evaluate the incidence, risk factors, and clinical impact of progression of aortic insufficiency (AI) after left ventricular assist device (LVAD) implantation. Methods. Patients undergoing primary durable LVAD implantation between 2004 and 2018 were included. Significant AI was defined as more than mild AI. Clinical and echocardiographic data were collected. Patients were stratified by degree of pre-LVAD AI into 3 groups: no AI, group I; trace AI, group II; and mild AI, group III. Results. Three hundred sixteen eligible patients underwent LVAD implant, 229 patients (72.5%) in group I, 54 (17.1%) in group II, and 33 (10.4%) in group III. Median follow-up was 469 days. Forty-two patients (13.3%) progressed to significant AI during follow-up. Group III patients had the highest rate of developing significant AI (I, 10.0%; II, 13.0%; III, 36.4%; P = .002). Freedom from significant AI at 1 year after LVAD implant was 94.5% in group I, 86.1% in group II, and 62.4% in group III (P < .001). Predictors of developing significant AI included mild preoperative AI, ischemic cardiomyopathy, and shorter duration of LVAD support. Patients with significant AI after LVAD implantation had higher mortality compared with those without (59.5% vs 37.2%; P = .006). Conclusions. Although some patients supported with an LVAD develop significant AI, this risk is increased in those with mild AI preoperatively. This finding, in conjunction with the increased mortality risk once significant AI develops, suggests that closer follow-up and management of LVAD patients with mild preoperative AI may be prudent. (C) 2020 by The Society of Thoracic Surgeons