ABSTRACT
BACKGROUND AND AIM OF THE STUDY: The present authors' experience with mitral valve reconstruction was reviewed to determine the influence of anatomical and functional abnormalities on mortality and reoperation. METHODS: Between 1991 and 2001, a total of 397 patients (mean age 59.0 +/- 13.4 years) underwent mitral valve reconstruction at the authors' institution. Patients were grouped according to the Carpentier functional classification as follows: type I, n = 88 (mean age 57.4 years; range: 24-81 years); type II, n = 169 (mean age 59.8 years; range: 16-81 years); type IIIa, n = 44 (mean age 52.8 years; range: 22-78 years); type IIIb, n = 85 (mean age 65.1 years; range: 33-82 years); and congenital (C), n = 11 (mean age 36.2 years; range: 18-65 years). RESULTS: The total follow up was 1,485.8 patient-years (pt-yr). Early mortality was: type I, 3.4% (n = 3); type II, 0; type IIIa; 2.3% (n = 1); type IIIb, 9.4% (n = 8); and C, 0 (p = 0.0028, IIIb > II and C). The late mortality was: type I, 2.1%/pt-yr (n = 6); type II, 1.1%/pt-yr (n = 7); type IIIa, 1.8%/pt-yr (n = 4); type IIIb, 4.3%/pt-yr (n = 12); and C, 1.7%/pt-yr (n = 1) (p = 0.0035 IIIb > II). The overall survival at 10 years was 79.6 +/- 4.2% (p <0.001, II > IIIb; p = 0.029, I > IIIb; p = 0.046, II > I). The reoperations were: type I, 1.4%/pt-yr (n = 4); type II, 1.9%/pt-yr (n = 12); type IIIa, 0.9%/pt-yr (n = 2); type IIIb, 0.4%/pt-yr (n = 1); and C, 0 (p = 0.0435 II > IIIb). Among the 19 reoperations there were 17 replacements, one re-repair, and one annuloplasty. The failures were predominantly type II, posterior (n = 5) and anterior (n = 5), or a combination (n = 2). The overall freedom from reoperation was 92.0 +/- 2.2% (p = NS between groups). CONCLUSION: Mitral valve reconstruction can be performed in all four functional categories with generally satisfactory results, except possibly for functional ischemic disease. The results of surgery for degenerative disease afford the opportunity for early surgical management, if proven techniques are adhered to in order to minimize the risk of reoperation. The repair of ischemic disease due primarily to ventricular dysfunction is in a state of evolution.