Long-term outcomes after valve-sparing root replacement in patients with reduced left ventricular ejection fraction

Valve-sparing aortic root replacement (VSRR) is an established operation for aortic root pathology with aortic insufficiency (AI), with limited long-term data available in patients with reduced left ventricular ejection fraction (LVEF).

Retrospective review of a database from 2 academic institutions identified 780 patients treated with VSRR from 2005 to 2020. In total, 745 patients had preoperative LVEF >40% (preserved-EF group), versus 35 patients with LVEF ≤40% (low-EF group). Conventional statistics were used for perioperative outcomes, Kaplan-Meier method for analysis of survival and reoperation, and Cox proportional hazard regression to examine independent predictors of late mortality and aortic root/valve reoperation. Propensity-score matching was also used to address selection bias.

Groups demonstrated similar preoperative characteristics, with the exception of greater frequency of moderate/severe AI (preserved EF 46.2% vs low EF 65.7%, P = .003) in the low-EF group. In-hospital mortality (preserved EF 1.5% vs low EF 2.9%, P = .426) was similar between groups. Overall survival (preserved EF 88.7% [87.7%-92.5%] vs low EF 82.5% [55.5%-95.6%], log-rank P = .332), and aortic valve/root reoperation (preserved EF 5.5% vs low EF 8.6%, P = .316) were similar at 10 years. Echocardiographic rates of late severe AI (preserved EF 2.2% vs low EF 3.5%, P = .705) were similar. Development of moderate or greater AI was independently associated with late reoperation (hazard ratio, 3.78 [2.22-7.97]; P < .001).

In well-selected patients, VSRR is an appropriate strategy for correcting significant aortic insufficiency in the setting of reduced LVEF and aortic root pathology. Overall survival, freedom from reoperation, and recurrent aortic insufficiency at 10 years were comparable with patients with preserved LVEF.