Aortic Valve Thickening as a Novel Risk Factor for Development of Aortic Incompetence after Left Ventricular Assist Device Implantation

Introduction The development of aortic incompetence (AI) after left ventricular assist device (LVAD) implantation occurs in 25-30% of all LVAD patients and contributes to increased morbidity. While the underlying pathophysiology is still unclear, previous pathology studies have shown that LVAD patients have accelerated aortic valve cusp remodeling when compared to other heart failure patients; this has been postulated to be the mechanism for AI in the LVAD population. A number of studies have attempted to identify populations that are at risk of developing AI, but none have assessed the effect of preoperative aortic valve sclerosis as a risk factor. We sought to assess the impact of preoperative aortic valve sclerosis on the development of AI in patients requiring LVAD support. Methods We retrospectively reviewed medical charts of patients implanted with Heartmate II (Abbott) and Heartware HVAD (Medtronic) LVADs at a large academic medical center between January 2011 and September 2018. We reviewed transthoracic echocardiography preoperatively and at 1, 3, 6, 9, and 12-months post LVAD implantation. Preoperative aortic valve thickening was predefined as mild or worse sclerosis on transthoracic echocardiogram. Patients with more than mild AI preoperatively or concomitant aortic valve replacement (AVR) were excluded from the analysis. Results A total of 660 echocardiography studies in 176 patients on LVAD support (mean age 55.1±14.0 years old, 84% male, 36.9% HVAD, 63.1% Heartmate2) were identified and analyzed. 25.5% were identified as bridge to transplant, 55.7% as destination therapy, 15.3% as bridge to decision and 3.4% as bridge to recovery. 21.6% were INTERMACS profile 1 and 71.0% were INTERMACS 2 and 3. 49.4% of the patients (n=87) had at least mild AI over the follow up period. The main risk factors identified for the development of AI post-operatively were aortic valve thickening on preoperative echocardiography (61.9% vs 42.4%, p = 0.014) and older age (58.9 ± 13.1 vs 51.4 ± 13.1, p = 0.003). On a multivariate analysis, patients with preoperative aortic valve thickening had a 2.1 times increased risk of developing mild or worse AI post LVAD implant (odds ratio of 2.1, p = 0.027). Conclusions Occurrence of aortic insufficiency is very frequent in patients on prolonged LVAD support. Our study is the first to show that preoperative aortic valve thickening is a significant risk factor for postoperative AI and should be included as a risk factor in decision making prior to LVAD implantation.