Optimal timing of surgical intervention in infective endocarditis is important in reducing mortality. We prospectively studied 126 consecutive episodes of infective endocarditis treated in one institution over 5 years, with special emphasis on long-term results and on the effects on outcome of surgical interventions. Twenty-six patients (21 %) underwent acute surgery on median treatment day 14. Mortality during treatment was 8% for patients undergoing acute surgery vs. 11% for those not undergoing surgery, and the adjusted 5-year survival rate of acute surgically treated patients was 91%, compared with 69% for the medically treated patients. Using univariate analysis, excess mortality during 5 years follow-up was associated with new cardiac decompensation at entry p<0.01), age (p<0.01), no acute surgery p<0.05) and mitral valve involvement (p<0.05). Multivariate analysis showed new cardiac decompensation at entry to be an independent predictor of cardiac death at 5 years follow-up (relative risk 2.39; Cl 1.05–5.45), while no surgery during active disease implied a relative risk of 3.45, though not statistically significant. Patients undergoing surgery very early (≤10 days of treatment) did not have a poorer outcome. Acute valve replacement, as compared with medical therapy only, might be important to increase both short-term and long-term survival in infective endocarditis.