Active native valve endocarditis: determinants of operative death and late mortality.

BACKGROUND: In this report, we reviewed 247 patients who underwent operation by our team for active native valve endocarditis between January 1979 and December 1993.
METHODS: There were 201 male and 46 female patients (mean age, 45.4 +/- 6 years). The aortic valve was involved in 163 cases, the mitral valve in 36 cases, both mitral and aortic valves in 44 cases, and the tricuspid valve alone in 4 cases. The most common microorganisms were streptococci. Univariate Pearson (chi2 test) and multivariate (stepwise logistic regression [BMDPLR]) analyses were used to identify significant predictors of operative mortality, reoperation, and recurrent endocarditis. Cox proportional hazards regression model was used to study late survival.
RESULTS: Operative mortality was 7.6% (n = 19). Increased age, cardiogenic shock at the time of operation, insidious illness, and greater thoracic ratio (>0.5) were the predominant risk factors; the length of antibiotic therapy appeared to have no influence. Two hundred thirteen patients were followed up. Median follow-up time was 6 years (range, 2 to 19 years). Overall survival rate (operative mortality excluded) was 71.3% +/- 3.8% at 9 years. Increased age, preoperative neurologic complications, cardiogenic shock at the time of operation, shorter duration of the illness, insidious illness before the operation, and mitral valve endocarditis were the predominant risk factors for late mortality. The probability of freedom from reoperation (operative mortality included) was 73.3% +/- 4.2% at 8 years; risk factors were younger age and aortic valve endocarditis. The rate of prosthetic valve endocarditis was 7%. No significant risk factor was found.
CONCLUSIONS: Increased age, insidious illness, and hemodynamic failure are the main risk factors for operative mortality. Long-term survival is good except for patients with preoperative neurologic complications and mitral valve endocarditis.