Results
We identified 49 patients who underwent a valve-sparing root replacement during our study period. Twenty-four patients had a history of the Ross procedure, 17 patients had a history of connective tissue disease, and eight patients had a history of congenital heart defect. Group 1 was comprised of seven patients who underwent the aortic root remodeling technique, Group 2 was comprised of 32 patients who underwent the aortic root reimplantation technique, and Group 3 was comprised of ten patients who underwent an aortic root remodeling technique with a rigid aortic annuloplasty ring. All seven patients in Group 1 had undergone a prior Ross procedure. Group 2 and Group 3 were patients with a history of connective tissue disease, prior conotruncal heart surgery, congenital heart disease, or a prior Ross procedure. Mean follow-up was significantly different between groups (p < 0.001). Group 1 and Group 2 patients had 7- and 14-year follow-up respectively and Group 3 had 1-year follow-up. The traditional aortic root remodeling technique has been largely abandoned in our institution and replaced with the aortic root remodeling technique with rigid annuloplasty ring since 2018. Aortic root reimplantation has been largely replaced as well, but is still used for selected patients based on surgeon preference and patient anatomy. Patient demographics were similar among groups with a median age of 19 years (range 8-32) and 87% male sex (table 1). There were no significant differences between the cardiopulmonary bypass and aortic cross-clamp times between groups 1 and 2; however, cardiopulmonary bypass times and aortic cross-clamp times were significantly higher in group 3 (p < 0.001). Group 1 and 3 were more likely to have a history of the Ross procedure (p = 0.003), and group 2 was more likely to have a history of connective tissue disease (p = 0.21). Group 1 was more likely to require subsequent aortic valve replacement, with all Group 1 patients requiring aortic valve replacement over the study period (p < 0.001). Group 3 did not have any required valve reinterventions during the study period. Group 3 was also more likely to have undergone aortic valve leaflet plication as part of their procedure (p = 0.004). There were no significant differences between aortic annulus measurements preoperatively (p = 0.741), but there was a significant difference in postoperative aortic annulus measurement (p = 0.01) Preoperative AI grades were similar between groups and were 2, 1.7, and 2.4 respectively (p = 0.2). However, immediate postoperative AI grades were significantly different between groups: 2.2, 1.2, and 1.2 respectively (p = 0.3). The most recent AI grades in patients from group 2 and 3 who did not require subsequent aortic valve replacement were similar (p = 0.9). Survival was good in all groups with one early mortality due to hemorrhage and one late mortality from malignancy.
Thirteen patients required aortic valve replacement after their valve-sparing root replacement during the study period. According to procedure type, all seven patients (100%) in Group 1 required valve replacement, six patients (19%) in Group 2 required aortic valve replacement, and no patient required valve replacement in Group 3 (p < 0.001). Average time to aortic valve replacement after valve-sparing aortic root replacement between Group 1 and Group 2 was 4.7 and 3.4 years, respectively (p =0 .6). History of connective tissue disease, prior Ross procedure, concomitant procedures, and prior sternotomy were not associated with increased risk for reintervention. Patients who required subsequent valve reintervention had significantly longer follow-up (p < 0.001), larger postoperative aortic annulus measurements (p = 0.012), and longer cardiopulmonary bypass (p = 0.01) and aortic cross clamp times (p = 0.003). There was no significant difference in preoperative AI grades; however, patients requiring future aortic valve replacement had significantly higher immediate postoperative AI grades (p < 0.001). Six of seven patients with greater than mild aortic insufficiency on immediate postoperative echocardiogram required subsequent aortic valve replacement (p < 0.001). Postoperative mitral regurgitation was also higher in patients who eventually required aortic valve replacement (p = 0.05)