Infective endocarditis in a tertiary-care hospital in southern Israel.

During the years 1980-1994, 84 patients were treated in our institution due to suspected infective endocarditis (IE). Seventy-one of these episodes occurring in 71 patients, classified definite or possible according to the Duke criteria, were retrospectively analyzed in this study. There were 52 cases of native valve endocarditis, 7 cases of early prosthetic valve endocarditis, and 12 cases of late prosthetic valve endocarditis. The incidence of IE did not change significantly during the study period. The overall mortality rate was 15%. Only one case of drug addiction appeared in our series despite its growing frequency in Israel. Rheumatic heart disease remained the main underlying cardiac condition and Streptococcus viridans remained the most common pathogen. Streptococcus bovis was found to be a significant pathogen causing IE in our patient population, while Staphylococcus aureus appeared to be less frequent. The Duke criteria significantly classified a greater proportion of cases as definitive, as opposed to the von Reyn criteria. Fewer cases were rejected by the Duke criteria, especially culture-negative cases, and those without histopathological confirmation. Application of the Duke criteria permits a more consistent approach to the diagnosis of IE, even in a non-drug-addict patient population.