Since the introduction of effective antimicrobial therapy, the leading cause of death in patients with infective endocarditis is no longer sepsis but, rather, congestive heart failure. The mortality is higher in patients with severe heart failure due to infective endocarditis who are treated with medical therapy only than in those who additionally undergo cardiac valve replacement. The mortality is also higher in patients with severe heart failure due to aortic infective endocarditis (40 to 93%) than in those with heart failure due to mitral infective endocarditis (17 to 66%). In patients with and in those without infective endocarditis, surgical intervention can be carried out with comparable mortality not only for aortic valve replacement (9 vs 8.4%) but also overall for valve replacement (10 vs 12%). In patients with class IV heart failure, overall mortality of valve replacement was higher (17%) than in patients with class II (8%) or class III heart failure (7%) and, similarly, comparable with that of matched groups of patients without infective endocarditis. In patients with class IV disability, the mortality of valve replacement was higher in those with active infective endocarditis (19%) than in those with inactive infective endocarditis, possibly due to a higher incidence of sudden onset of severe aortic regurgitation and myocardial abscess. No patient with valve replacement for inactive infective endocarditis developed prosthetic valve endocarditis; a single case of prosthetic valve endocarditis occurred in a patient with active infective endocarditis. In general, early surgical intervention is preferable to procrastination in the management of patients with progressive or severe heart failure due to infective endocarditis. Although, in at least 70% of patients, blood cultures may be rendered sterile within one week of initiation of appropriate antimicrobial therapy, patients with infective endocarditis due to staphylococci, multiply-resistant gram-negative bacilli, fungi, Q-fever or those with myocardial abscess or multiple relapses may require surgical intervention. While the overall incidence of clinically apparent emboli has been reported to be as high as 30%, in a ten-year observation period at the Mayo Clinic, the rate was 5.6%. Patients with echocardiographic evidence of large or mobile vegetations and those with infective endocarditis cause by microorganisms associated with a high risk of embolization such as slow-growing fastidious gram-negative bacilli, fungi (especially Aspergillus) and nutritionally-variant viridans streptococci should be considered candidates for surgery irrespective of a history of emboli.(ABSTRACT TRUNCATED AT 400 WORDS)