Left ventricular recovery and outcomes in patients with a durable left ventricular assist device: the EUROMACS registry

Background and aims: Left ventricular assist device (LVAD) support has been shown to induce the reversal of neurohormonal changes and myocardial recovery. This study aimed to evaluate the prevalence, predictors, and clinical impact of left ventricular (LV) recovery after LVAD implantation using data from the European Registry for Patients with Mechanical Circulatory Support. Methods: Patients with at least one available left ventricular ejection fraction (LVEF) measurement at both baseline and post-LVAD implantation were enrolled. Left ventricular recovery was classified as 'no recovery' if LVEF increase was <5%, 'mild' if LVEF increased by 5%-15% with peak LVEF <50%, 'intermediate' if LVEF increased by >15% with peak LVEF <50%, and 'full' if LVEF increased by ≥5% with peak LVEF ≥50%. Results: A total of 1550 patients with a median age of 57 (48-64) years were included. LV recovery occurred mostly within the first year, with marginal probabilities of no, mild, intermediate, and full recovery at 4 years being 51% [95% confidence interval (CI) 48-54%), 41% (95% CI 38-43%), 8% (95% CI 7-9%), and 1% (95% CI 0-1%), respectively]. INTERMACS Classes 2, 3, and 4 [compared with Class 1; hazard ratio (HR) .72, .64, and .51, respectively] were associated with a lower hazard of attaining any LV recovery, whereas renin-angiotensin system inhibitor use was associated with a higher hazard of attaining any LV recovery (HR 1.17; 95% CI 1.00-1.37). Each 10% higher LVEF value during follow-up was associated with an estimated 14% lower instantaneous hazard of all-cause mortality (HR .86; 95% credible interval .75-.98). The cumulative incidence of the composite major adverse cardiovascular event endpoint, including cardiovascular death, late right heart failure, and arrhythmia beyond the first year, was lower across higher LV recovery categories defined according to the LVEF value at 1 year (Gray's test, P = .042). Conclusions: This multicentre European study shows that LV recovery mostly occurs within the first year after LVAD implantation and is associated with better clinical outcomes. Renin-angiotensin system inhibitor use was associated with both any and intermediate/full LV recovery, while combination neurohormonal blockade strategies yielded directionally favourable but imprecise estimates, warranting further study during LVAD support.