Long-Term Outcomes of Bioprosthetic Valves in the Mitral Position: A Pooled Meta-Analysis of Reconstructed Time-to-Event Individual Patient Data

Abstract

ObjectivesBioprosthetic MVR use is on the rise, but data regarding long term durability is lacking. We sought to perform a reconstructed individual patient data meta-analysis from published Kaplan Meier curves to ascertain survival, freedom from valve degeneration and reoperation in studies published since 2010. We explored the effects of age and valve type (bovine pericardial or porcine valve) on outcomes. MethodsWe searched MEDLINE, OVID, Embase and Cochrane CENTRAL for studies reporting at least 3 years of follow-up after bMVR and published since 2010. The ROBINS-I tool was used to assess methodological quality. Kaplan Meier curves were digitized to extract individual patient data and reconstructed estimates for overall survival, freedom from SVD and freedom from re-operation. Results20 studies (16465 patients) were included. 9 studies reported on porcine valves, 6 on bovine and 7 did not specify the valve type. Overall survival after bMVR at 15 years was 40% (CI 38%-42%), freedom from re-operation at 15 years was 79% (CI 76%-82%) and freedom from structural valve degeneration at 15 years was 64% (58%-70%). Freedom from SVD was improved in the 70+ age group (93% up to 25 years, HR 6.6(2.5-17) for 18-59 VS >70, p<0.0001). There was no difference in valve durability or survival between bovine pericardial or porcine valves. ConclusionIn this meta-analysis patients of receiving bioprosthetic mitral valve replacement using newer generation valves, the inverse relationship between age and SVD was re-iterated in the 70+ age group. Prosthesis type made no difference in outcomes.