Surgical Treatment of Active Infective Endocarditis: Determinants of Early Outcome

(Department of Cardiovascular Surgery, Hokkaido University School of Medicine, Sapporo, Japan)

Yasuhiro Kamikubo Toshifumi Murashita Hideyuki Kunishige
Norihiko Shiiya Keishu Yasuda
The purpose of this study was to review our experience in the treatment of active endocarditis and identify determinants of early outcome. Sixty-nine patients (mean age 47.3 years, range 5 months to 88 years) underwent surgery for active endocarditis. Native valve endocarditis was present in 59 (85.5%) and prosthetic valve endocarditis in 10 (14.9%). The aortic valve was infected in 26 (37.7%), the mitral valve in 24 (34.8%), both aortic and mitral valves in 13 (18.8%), and the tricuspid in 3 (4.3%). Paravalvular abscess was identified in 22 (31.9%). Streptococci (27.5%) and Staphylococci (23.3%) were the most common pathogens, but the pathogen was not identified in 36.2%. Hospital death occurred in 13 (18.8%), and causes of deaths included cardiac failure in 6 and sepsis in 5. There were 2 late deaths, and the causes of death were cerebral infarction and renal dysfunction. Univariate analysis indicated that older age (p0.02), New York Heart Association class III or IVip0.02), a preoperatively unidentified pathogenip0.02jand concomitant operation for abscess and fistula (p0.04) were significant risk factors in hospital mortality. Prosthetic valve infection was a relative risk factor in hospital mortality (p0.11). Multivariate analysis revealed that NYHA III -IV (p0.02, odds ratio18.1, 95% CI1.49 -220.1) and a preoperatively unidentified pathogen (p0.02, odds ratio7.45, 95% CI1.44 -38.5) were independent predictors of hospital mortality. To reduce hospital mortality in active endocarditis, early surgical intervention is recommended before the involvement of heart failure, particularly when the pathogen is not identified.
@Jpn. J. Cardiovasc. Surg. 33: 1 -5 (2004)