To study the early and late results of mitral valve replacement (MVR) by Starr-Edwards caged-ball and bileaflet mechanical prosthesis.
We retrospectively analyzed 236 MVR performed in 236 patients: 127 by Starr-Edwards prosthesis (group 1) and 109 by bileaflet prosthesis (group 2).
During the early period (30 days), the mortality rate was higher in group 1 (6.3 % vs 1.8 % ; p = 0.0001), while hemorrhagic, thromboembolic and infectious complications were comparable in the two groups. In the late period (> 30 days) and with an average follow-up of 11.5 ± 5.7 years, mortality was higher in group 1 (9.4 % vs 4.6 % ; p < 0.0001). The same was true for thromboembolic complications (20.8 % vs 6.4 % ; p < 0.0001), hemorrhagic complications (13.4 % vs 7.3 % ; p = 0.02), infectious complications (3.1 % vs 0.9 % ; p = 0.02) and cardiac complications that were not due to the prosthesis (32.3 % vs 14.7 % ; p = 0.02). The hemodynamic profile of the bileaflet prostheses was better than that of the Starr-Edwards prostheses (average functional prosthetic surface area was 2.37 ± 0.44 cm2 and average pressure gradient was 5.6 ± 1.1 mmHg vs 2.04 ± 0.52 cm2 and 7.6 ± 4.9 mmHg).
Our work confirms the superiority of bileaflet mechanical prostheses, with rates of early and late mortality, thromboembolic and hemorrhagic complications lower than those of the Starr-Edwards prostheses in more than 11 years of follow-up. However, one should not forget that the prevention of infective endocarditis, good observance of oral anticoagulant treatment and early surgery before left ventricular dysfunction occurs remain the best guarantee a good result of the MVR.