OBJECTIVE To evaluate the feasibility of mitral valve repair in patients with anterior leaflet (ALP) or bileaflet prolapse (BLP) and identify factors predisposing patients to replacement. To compare long-term survival of patients submitted to repair (Group Repair) against those submitted to replacement (Group Replacement), and investigate causes of early and late failures of repair. METHODS From January 1992 through December 2012, 768 patients with ALP or BLP were submitted to mitral valve surgery, of whom 501 had degenerative involvement [Myxomatous (Myx)-336 (67.1%) or fibroelastic deficiency (Fed)-165 (32.9%)] and constituted the study population. Isolated ALP was present in 274 patients (54.7%) and BLP in 227 (45.3%). Associated procedures were admitted. RESULTS Patients with Fed were significantly older (64.4 ± 12.1 vs 54.8 ± 15.5 years, P < 0.001), more symptomatic (63 vs 44.3%; P < 0.001) and with higher incidence of atrial fibrillation (43.6 vs 26.2%; P < 0.001). Repair was achieved in 94.8% of patients with an overall 30-day mortality rate of 1.2% (0.3% in the last decade) and no differences regarding aetiology. Age, moderate to severe left ventricular (LV) dysfunction, previous cardiac surgery, multiple segment prolapse, mitral calcification, leaflet retraction and the performing surgeon were independently associated with replacement. Group Repair patients had a greater adjusted 20-year survival by comparison with Group Replacement (43.4 ± 5.5 vs 13.6 ± 11.3%; P < 0.001) and similar to that of the age- and sex-adjusted general population (P = 0.10). Valve replacement, New York Heart Association (NYHA) class III-IV, pulmonary hypertension and LV dysfunction emerged as independent predictors of late mortality. Patients in NYHA class I-II experienced a higher repair rate (98.4%) and better survival than those in Class III-IV. Two repair patients were reoperated during the first year after surgery (early failure) and both were 'rerepaired'. Late failure was observed in 21 patients, mostly for progression of the disease. The 20-year rate of freedom from reoperation was 88 ± 2.7%, significantly worse in ALP patients (P = 0.040), and not different between Fed and Myx. CONCLUSIONS Patients with ALP or BLP can be submitted to surgery with low mortality and great probability of repair in expert hands. Patients should be operated on at an early phase (asymptomatic or mildly symptomatic), because there is a higher probability of repair and greater benefit on long-term survival.