Long-term evaluation of biological versus mechanical prosthesis use at reoperative aortic valve replacement

Objectives: Recent evidence indicated that the use of a bioprosthesis in young patients at first-time aortic valve replacement (AVR) is associated with an increased reoperation risk, but not with an increase in long-term mortality, when compared with the use of a mechanical valve. However, at reoperative AVR, follow-up data by prosthesis type have been lacking from the literature. Therefore, we examined long-term survival and valve-related complications according to the type of prosthesis used at reoperative AVR. Methods: We studied 437 patients who underwent reoperative AVR, at a mean age of 58.6 +/- 14.2 years, for failure of a previously implanted aortic valve prosthesis. Thirty-day mortality at reoperative AVR was 6% (n = 27). A bioprosthesis was used in 135 (31%) patients. Patients were subsequently followed up for a mean of 7.6 +/- 6.8 years after reoperative AVR. Results: The use of a bioprosthesis at reoperative AVR was not associated with impaired survival on adjusted analysis (hazard ratio [HR], 0.8 +/- 0.4; P = .6). Freedom from thromboembolism, and endocarditis were similar between valve types (both P > .05); however, late postoperative major hemorrhage occurred only in patients who received a mechanical prosthesis at reoperative AVR. Risk factors for third-time AVR included the use of a bioprosthesis (HR, 14.0) and younger age (HR, 1.05 per decreasing year) at reoperative AVR (both P < .001). Thirty-day mortality of third-time AVR was 4% (n = 1/27). Conclusions: At reoperative AVR, the use of a bioprosthesis is associated with equivalent long-term survival compared with a mechanical prosthesis. Patients who receive a bioprosthesis at reoperative AVR are less likely to experience major hemorrhage but more likely to require third-time AVR, albeit with an acceptable third-time perioperative mortality risk. Therefore, the patient's informed preferences regarding prosthesis choice should prevail, even in a reoperative context. (J Thorac Cardiovasc Surg 2012;144:146-51)