Evaluation of late aortic insufficiency with continuous flow left ventricular assist device†

The aim of this study was to evaluate late development of aortic insufficiency (AI) with continuous flow left ventricular assist device (CLVAD). Development of AI is an increasingly recognized important complication in CLVAD therapy, but there are still few reports about this topic.

We analysed data from 99 patients who underwent CLVAD implantation. De novo AI was defined as the development of mild or greater AI in patients with none or trace preoperative AI. Anatomic and functional correlates of de novo AI were investigated.

Among the 17 patients with preoperative mild AI, no improvements were observed in mitral regurgitation or LV end-systolic dimension. Of the remaining 82 patients, de novo AI was identified in 43 patients (52%), on the most recent follow-up echocardiography, and did not influence survival nor improvement of LV geometry. Rate of freedom from de novo AI at 1 year after CLVAD implantation was 35.9%. Development of significantly greater AI was observed in patients without valve opening (AI grade 1.3 ± 1.0 vs 0.7 ± 0.9; P = 0.005). By multivariate Cox hazard model, smaller body surface area (BSA) [hazard ratio: 0.83 [95% confidence interval (CI): 0.72-0.97], P = 0.018], larger aortic root diameter (AOD) [hazard ratio: 1.11 (95% CI: 1.02-1.22), P = 0.012] and higher pulmonary artery systolic pressure (PASP) [hazard ratio: 1.24 (95% CI: 1.10-1.41), P < 0.001] were identified as the independent preoperative risk factors for de novo AI. In a subset of patients with speed adjustments, increase of CLVAD speed worsened AI and led to insufficient LV unloading in patients with aortic dilatation (AOD ≥ 3.5 cm).

Any significant mortality difference related to preoperative or development of postimplant AI was not found. AI was associated with changes in LV size, and there appears to be an interaction between BSA, preoperative PASP, time since implant, aortic valve opening, aortic size and development of AI. Longitudinal clinical management in CLVAD patients, particularly in terms of CLVAD speed optimization, should include careful assessment.