Long-term echocardiographic and clinical outcomes after aortic root replacement in patients with Marfan syndrome.

Abstract

We compared longitudinal echocardiographic data between patients with Marfan syndrome who underwent aortic valve-sparing (AVS) and valve-replacing (AVR) aortic root replacement procedures. Of 316 patients enrolled at 19 sites in 5 countries, 239 (75%) were treated with AVS and 77 (25%) with AVR. Transthoracic echocardiograms were collected at postoperative years 1, 3, 5, 7, 10, 12, 14, 16, 18, and 20. Mixed-effects models evaluated echocardiographic changes over time. Inverse-probability-weighted Cox regression models were used to compare the risk of persistent postoperative aortic regurgitation (AR) grade ≥2+ and AV reintervention between AVS and AVR. The echocardiographic studies showed higher AR grades in the AVS group during early-to-midterm follow-up (P<.001). The AVS group also had somewhat larger annular size (P<.001) and left ventricular end-systolic (P<.001) and end-diastolic (P=.003) diameters. AV mean gradient, AV systolic velocity, left ventricular ejection fraction, and sinus of Valsalva diameter did not have significantly different trajectories over time between AVS and AVR. Although AVS and AVR groups did not differ in overall survival (P=.2) or AV reintervention rates (P=.2), persistent AR grade ≥2+ was associated with AV reintervention (P=.01). Although patients with Marfan syndrome who underwent AVS had greater annular dilatation and higher AR grades than patients who underwent AVR, the likelihood of AV reintervention and death was similar between the groups.