Percutaneous Transcatheter Interventions for Aortic Insufficiency in Continuous-Flow Left Ventricular Assist Device Patients: A Systematic Review and Meta-Analysis.

De novo progressive aortic insufficiency (AI) is a side effect frequently related to prolonged support with continuous-flow left ventricular assist devices (CF-LVAD). Its progression can result in recurrent clinical heart failure symptoms and significantly increased mortality. Recently, percutaneous intervention methods, such as transcatheter aortic valve replacement (TAVR) and percutaneous occluder devices, have emerged. However, given the very scarce global experience with these approaches, evidence in the literature is lacking. We sought to assess the outcomes of CF-LVAD patients who had undergone percutaneous intervention for AI. A systematic search of six databases from inception to April 2016 was performed by two independent reviewers. Eligible studies were those that included series or cases where patients had percutaneous transcatheter interventions for AI in CF-LVAD patients. Data were extracted and meta-analyzed from the identified studies. A total of 29 patients from 15 published studies and 3 unpublished records were included in the analysis. Mean patient age was 56.6 ± 13.7 years, and 72.4% were male. The etiology of heart failure resulting in LVAD placement was ischemic cardiomyopathy in 17.2%. The two intervention groups were TAVR (27.6%) and occluder devices (72.4%). A transfemoral approach (69%), apical approach (10%), brachial approach (7%), subclavian approach (3%), and mini-sternotomy (3%) were used. The preintervention AI grade was severe with a median grade of 4 (interquartile range, 4-4). Postoperatively, the AI grade improved significantly to a median grade of 0 (0-2). At long-term follow-up AI grade was still trivial with a median AI grade of 1 (0-1). Subgrouping the treatments into the occluder device and TAVR, it was found that both interventional techniques were similarly effective in reducing the AI grade from severe to trivial. In terms of complications, from the occluder group, two patients were complicated with device migration and another two with transient hemolysis. In the TAVR cohort, two patients experienced device migration and another had significant postimplant perivalvular leakage. Our results indicate that percutaneous interventions for AI in CF-LVAD patients with TAVR, and closure devices demonstrate similar efficacy in significantly reducing severe AI. Current results are encouraging, potentiating viability as a treatment option, particularly in nonsurgical candidates. Future research with larger patient cohorts and comparative controls is required to sufficiently evaluate the efficacy of this technique and promote its widespread acceptance as a mainstay treatment.