Appraisal of Long-Term Outcomes of Tricuspid Valve Replacement in the Current Perspective
Résumé
Background We address the differential long-term results of tricuspid valve replacement and late valve-related events among bioprostheses versus mechanical prostheses. Methods In a single-institution investigation, we reviewed the patients’ prospectively collected data and performed a clinical follow-up. Both the Kaplan-Meier (actuarial) and the competing risks (actual) methodologies were used. Results Overall, 188 tricuspid valve replacements were performed during 1971 to 2012. In this complex population (reoperations, 48.1%; associated procedures in 71.3%), operative mortality was 27.6% (significantly declining in recent years). A bioprosthesis was used in 82.4%. Follow-up was 10.2 ± 9.1 years (as long as 37.3 years, or 1,270 patient-years; 91.2% complete). At 15 years, freedom from structural valve deterioration was 93.3% ± 6.4% (competing risks 94.7 ± 5.1%) in the mechanical group and 85.2% ± 5.4% (competing risks 92% ± 2.9%) in the bioprostheses group (p = 0.19). Freedom from any valve-related adverse events was lower among mechanical valves versus bioprostheses (although not statistically significant). Mechanical valves showed significantly lower freedom from thromboembolic events (actuarial 62.3% ± 14.3% versus 97.7% ± 1.6%; competing risks 74.1% ± 10% versus 98% ± 1.4%; p < 0.001) and earlier adverse events (4.9 ± 4.5 versus 11.1 ± 9.4 years) than bioprostheses. There were 11 reoperations for bioprosthetic structural valve deterioration (89.8% and 94.3% actuarial and actual freedom, respectively). Conclusions Bioprostheses for tricuspid valve replacement have a very good long-term durability. Mechanical valves display earlier and more severe morbidity at follow-up