Valve-sparing root replacement in patients with bicuspid versus tricuspid aortic valves

Objectives: We sought to compare the outcomes of patients undergoing aortic valve-sparing root replacement with bicuspid versus tricuspid aortic valves. Methods: A total of 333 consecutive patients (bicuspid aortic valve, n = 45; tricuspid aortic valve, n = 288) underwent valve-sparing root replacement using the reimplantation technique from 1988 to 2012 at a single institution. The primary analysis was performed on a 1:3 bicuspid aortic valve: tricuspid aortic valve propensity-matched dataset to mitigate known differences between these 2 groups. In the matched, dataset, mean age (bicuspid aortic valve: 40 +/- 13 years; tricuspid aortic valve: 41 +/- 14) and rates of comorbidities were similar between groups. Patients with bicuspid aortic valves were less likely to have Marfan syndrome (bicuspid aortic valve: 9% vs tricuspid aortic valve: 53%, P<.001). Patients were followed prospectively with aortic root imaging for a median of 8.2 (5.3-12.2) years. Results: Primary cusp repair was required more often in patients with bicuspid aortic valves (bicuspid aortic valve: 79% vs tricuspid aortic valve: 45%, P<.001). A total of 3 operative deaths occurred (bicuspid aortic valve 0% vs tricuspid aortic valve 2%, P = .52). The probability of aortic insufficiency increased significantly over time in both groups (odds ratio, 1.106; 95% confidence interval, 1.033-1.185; P = .004), but there was no significant difference in this increase between the bicuspid aortic valve and tricuspid aortic valve groups (P = .08). Long-term freedom from mortality (P = .20), cumulative incidence of aortic valve reoperation (P = .42), and valve-related events (P = .69) were similar across groups. Conclusions: In well-selected patients with bicuspid aortic valves and favorable cusp morphology, valve-sparing root replacement offers excellent long-term clinical outcomes.