Background: We compared results of the Bentall procedure with valve-sparing aortic root replacement (VSRR) for aortic root aneurysm in Marfan syndrome.
Methods: Marfan syndrome patients who had the Bentall procedure or VSRR at our institution between April 1997 and September 2006 were identified. Follow-up information was obtained from hospital charts and contact with patients or their physicians. Kaplan-Meier survival and propensity score analyses were performed.
Results: One hundred forty Marfan syndrome patients had either the Bentall procedure (n = 56) or VSRR (n = 84; 40 remodeling and 44 reimplantation). Bentall patients were older than VSRR patients (38 versus 29 years; p = 0.0001) and had more aortic dissections (16% versus 1%; p = 0.0012); more urgent/emergent surgery (20% versus 2%; p = 0.0008); larger preoperative sinus diameter (5.7 versus 5.1 cm; p = 0.0004); and more preoperative 3+/4+ aortic insufficiency (59% versus 10%; p < 0.0001). There were no operative deaths. Postoperatively, 9% Bentall patients (5 of 56) and 1% of VSRR patients (1 of 84) suffered thromboembolic events (p = 0.03). Two percent (1 of 56) of Bentall patients required reoperation on the aortic root versus 6% of VSRR patients (5 of 84; p = 0.40). Eight-year freedom from aortic valve replacement was 90% for VSRR patients. Eight-year survival was 90% for Bentall and 100% for VSRR patients (p = 0.01). Propensity-adjusted regression showed that the Bentall procedure did not predict mortality (p = 1.00) and did not protect from reoperation (odds ratio = 0.28; 95% confidence interval: 0.01 to 4.33; p = 0.36).
Conclusions: The Bentall procedure and VSRR have similar operative results in Marfan syndrome. The procedures are distinguished by higher rates of thromboembolism among Bentall patients and higher rates of reoperation among VSRR patients. Lower late survival among Bentall patients probably reflects the preferential use of the Bentall procedure for higher risk patients.