Analysis of 109 cases of infective endocarditis in a tertiary care hospital

Early diagnosis and treatment can benefit the outcome of patients with infective endocarditis. The older diagnostic criteria (von Reyn criteria) relied upon tissue pathology and blood culture results, but the sensitivity was low. A newly proposed criteria, Duke criteria which apply echocardiographic findings as a major factor in diagnosis, have shown a better sensitivity in the diagnosis of infective endocarditis.

From 1984 to 1994, 120 episodes of endocarditis in 119 patients were reviewed retrospectively. One hundred and nine episodes fulfilling the Duke criteria were enrolled. The demographic data, antecedent disease, predisposing factors for infective endocarditis including systemic disease, prosthetic valve, intravenous drug abuse, dental or surgical manipulation, culture results, echocardiographic findings, complications and outcome were recorded and analyzed.

Sixty-six percent (72/109) was definite endocarditis, and 34% (37/109) was possible endocarditis, as categorized by Duke criteria. Eleven percent (12/109) of the patients were rejected by von Reyn criteria. The common possible predisposing factors were rheumatic heart disease, congenital heart disease, mitral valve prolapse, dental manipulation and intravenous drug abuse. Both patients with positive blood culture and negative blood culture results had one-fourth mortality rate. A notably higher mortality rate was found in patients with S. aureus endocarditis as compared with those of endocarditis caused by viridans streptococci(44% vs. 8%, p < 0.05). Patients with echocardiographic vegetations had a higher mortality rate (32% vs. 11%, p < 0.05) and more peripheral or organ embolic events (26% vs. 6%, p < 0.05) than those without vegetation.

The Duke criteria are more sensitive than the von Reyn criteria for diagnosis of infective endocarditis. S. aureus endocarditis carried a higher mortality rate than viridans streptococci endocarditis. The present study also indicated that patients with discernible valvular vegetation on echocardiogram had a high mortality rate and occurrence of peripheral or organ embolic events. However, there was no statistical significance in the development of congestive heart failure and CNS complications between the patients with and without vegetation.