Long-term outcomes of mechanical versus biological aortic valve prosthesis: Systematic review and meta-analysis.

Abstract

In recent years, the use of surgically implanted biological aortic valves has been favored over mechanical prosthesis in patients between 50 and 70 years of age. However, outcomes on long-term survival are contradictory. The objective of this study was to determine if patients with mechanical valves have worse long-term survival than patients with biological prostheses. We systematically searched published studies that: (1) were propensity score-matched or randomized controlled trials; (2) provided survival data with a minimum follow-up of 5 years; and (3) included patients older than 50 and younger than 70 years of age. Review articles, case reports, and editorials were excluded. We conducted a meta-analysis on the basis of 2 types of analysis. A reconstruction of the database of each study to simulate a patient-level meta-analysis was performed. Log rank test of Kaplan-Meier curves was recalculated. Hazard ratio (HR) was calculated using a univariate Cox regression. In addition, we calculated a pooled HR using the fixed-effect inverse variance method. Four propensity score-matched studies and 1 randomized controlled trial met the inclusion criteria. Data of 4686 patients were analyzed. Survival rates for mechanical versus biological valves at 10 and 15 years of follow-up were: 76.78% (95% confidence interval [CI], 74.72%-78.69%) versus 74.09% (95% CI, 71.96%-76.08%), and 61.58% (95% CI, 58.29%-64.69%) versus 58.04% (95% CI, 54.57%-61.35%). Log rank test was statistically significant (P = .012) and the pooled HR was 0.86 (95% CI, 0.76-0.97; P = .01). Compared with biological aortic valves, mechanical valves are associated with a long-term survival benefit for patients between 50 and 70 years.