The appropriate selection of prosthesis type—mechanical versus bioprosthetic—for mitral valve replacement has been largely based on long-term durability. It is well-known that younger patients are typically assigned mechanical prostheses to reduce the need for repeat surgeries, while older patients are often assigned bioprostheses to avoid prolonged anticoagulation therapy, at least with vitamin K antagonists. However, this approach has led to an increased use of mitral bioprostheses in younger patients, driven by the development of transcatheter “valve-in-valve” techniques, which may mitigate the risk of reintervention due to prosthetic degeneration.
Existing studies have compared mitral valve replacement outcomes between mechanical and bioprosthetic valves, often including patients undergoing concomitant cardiac procedures, mainly coronary artery bypass surgery and tricuspid and/or aortic valve replacement. To balance variations in morbidity associated with different age groups and prosthesis choices, studies like the present one, which focuses solely on isolated mitral valve replacement in propensity-matched populations, are essential for evaluating both early and long-term outcomes.
This study stratified patients into two age groups: under 65 and between 65 and 75 years. Initially, 1,536 patients who underwent isolated mitral valve replacement between 2000 and 2017 were included, of which 806 received mechanical prostheses and 730 bioprostheses. Propensity score matching was then performed based on 32 baseline variables, including demographics, sex, age at surgery, and comorbidities. For variables with missing data, such as ejection fraction or serum creatinine, multiple imputations assuming a multivariate normal distribution were used to estimate the missing values. The mean observational follow-up was 9.4 ± 5.8 years, during which postoperative complications, early and late morbidity, and both in-hospital and out-of-hospital mortality were evaluated. Additionally, logistic regression was employed to assess short-term outcomes, including in-hospital postoperative complications like stroke, gastrointestinal bleeding, and permanent pacemaker implantation. Cox proportional hazards model was used for long-term outcomes, which included 10-year mortality and the necessity for surgical or transcatheter reintervention. Inverse probability weighting was also applied to the results for comparative analysis.
The study successfully matched 226 patient pairs under 65 years and 171 pairs between 65 and 75 years, resulting in a total of 794 patients included in the final analysis. The findings from the propensity-matched cohorts indicated a higher stroke rate among patients with mechanical mitral valve replacement compared to those with bioprostheses, in both age groups. However, these differences were not statistically significant. Additionally, postoperative gastrointestinal bleeding rates were similar in both age groups. Regarding permanent pacemaker insertion, younger patients under 65 had a higher rate with mechanical valves, whereas among those over 65, the rate was higher with bioprostheses. Acute kidney injury requiring dialysis was significantly more common in patients under 65 with bioprosthetic valves (p = .011).
Long-term results revealed greater 10-year survival in patients under 65 with mechanical valves. Likewise, mechanical valves in younger patients were associated with a reduced reintervention rate compared to bioprostheses. However, this advantage did not extend to patients between 65 and 75 years, in whom reintervention rates were comparable between both valve types. Thus, the survival advantage of bioprostheses was more significant in patients aged 65 to 75.
COMMENTARY:
As supported by similar studies, the preference for mechanical mitral prostheses in patients under 65 stems from their durability and reduced need for reintervention. However, most of these studies include concomitant procedures that may affect overall outcomes. This study’s unique contribution lies in analyzing only isolated mitral valve replacements, thereby minimizing confounding factors and employing propensity matching across two age groups, reducing bias. However, the prosthesis selection based on biological age or expected survival may have influenced the better survival outcomes seen in younger patients with mechanical valves, which could reflect a cohort with a better preoperative condition not entirely controlled for by propensity analysis.
Although evidence seems robust, future studies should also consider long-term mortality and adverse events in patients with isolated mitral valve replacement with bioprostheses who cannot undergo prolonged anticoagulation, particularly among those under 65. Encouraging further research will help determine the individual impact of different prosthesis types and anticoagulation approaches associated with them, while considering that factors other than the prosthesis may dictate the need, type (vitamin K antagonists or direct oral anticoagulants), and intensity of anticoagulation therapy. It’s important to remember that the presence of an indication for oral anticoagulation due to valve disease should not dictate the choice of mechanical prosthesis (Class IIb recommendation in current clinical guidelines).
REFERENCE:
Rokui S, Gottschalk B, Peng D, Groenewoud R, Ye J. Long-term outcomes of isolated mechanical versus bioprosthetic mitral valve replacement in different age groups of propensity-matched patients. Eur J Cardiothorac Surg. 2024 Jul;66(1).
