We compared clinical (30+/-24 months) and echocardiographic follow-up (22 +/- 20 months) data of 184 consecutive patients with myxomatous degenerative mitral regurgitation (Group A) and 85 consecutive patients with rheumatic mitral regurgitation (Group B) after repair. Selection criteria for rheumatic etiology was predominant mitral regurgitation with valve area >or= 2.0 cm(2) and with no significant calcification in valvular apparatus. Repair was successful in 93% of group A and in 92% of group B (p>0.05). There was no difference of operative mortality (1% vs 0%) and of the incidence of the second-pump valve replacement (4% vs 5%). The 4-yr survival, 4-yr event-free survival, and 4-yr mitral regurgitation-free survival rates in group A were 96 +/- 2%, 89 +/- 4%, and 76 +/- 5%, respectively, which were not different from those in group B (97 +/-2 %, 93 +/- 4%, and 68 +/-7 %, p >0.05). Independent determinants of development of at least moderate regurgitation in group A were no use of ring annuloplasty (hazards ratio 6.6, 95% CI 2.0 to 21.5) and new chordae formation (hazards ratio 3.5, 95% CI 1.4 to 8.7). In group B, no use of ring annuloplasty (hazards ratio 15.3, 95% CI 3.5 to 66.7) also was independent predictor. Valve repair is highly feasible in selected patients with rheumatic mitral regurgitation, and clinical course is not significantly different from that of patients with degenerative mitral regurgitation.