Outcomes of repeat mitral valve replacement in patients with prior mitral surgery: A benchmark for transcatheter approaches.

Objectives: With the emergence of transcatheter mitral valve-in-valve/ring replacement for deteriorated bioprostheses or failed repair, comparative clinical benchmarks for surgical repeat mitral valve replacement (re-MVR) are needed. We present in-hospital and survival outcomes of a 24-year experience with re-MVR. Methods: From January 1992 to June 2015, 520 adult patients underwent re-MVR; 273 had undergone prior mitral valve repair (pMVP) and 247 had undergone prior MVR (pMVR). A benchmark cohort of isolated re-MVR was defined based on potential eligibility for transcatheter mitral valve-in-valve/ring replacement, resulting in 73 pMVPs with previous annuloplasty rings and 74 pMVRs with previous bioprosthetic valves for comparison. Results: For the entire cohort, mean age was 64 +/- 12 years for pMVP patients and 63 +/- 15 years for pMVR patients (P = .281), which was similar for the benchmark cohort. Overall operative mortality was 14 out of 273 (5%) for pMVP versus 23 out of 247 (9%) for pMVR (P = .087). There were 3 operative deaths (4.1%) in both groups of the benchmark cohort (P = 1.0). For the benchmark cohort, median time to reoperation was 9.8 years for pMVP and 9.1 years for pMVR. Cox proportional hazard analysis showed that chronic kidney disease (hazard ratio [HR], 2.47; 95% CI, 1.77-3.44), endocarditis (HR, 1.49; 95% CI, 1.07-2.07), pMVR (HR, 1.45; 95% CI, 1.12-1.89), early reoperation <= 1 year (HR, 1.49; 95% CI, 1.02-2.17), and age (HR, 1.04/y; 95% CI, 1.03-1.05) were associated with decreased survival after re-MVR. Conclusions: A re-MVR is a high-risk operation, but in carefully selected patients such as our benchmark population, it can be performed with acceptable results. Patients undergoing pMVP also have better long-term survival compared with patients undergoing pMVR. These results will serve as a benchmark for transcatheter mitral valve-in-valve/ring replacement.