Abstract

We carried out univariate and multivariate analysis of outcome among 122 patients with prosthetic valve endocarditis (PVE) admitted to our ICU between 1978 and 1992. The predominant pathogens were Staphylococcus aureus (33%), streptococci (20%), coagulase-negative staphylococci (12%), enterococci (10%), and Gram-negative bacilli (9%). At 4 months, overall survival was 66% (42 deaths). Staphylococcus aureus was the main predictor of death (75% vs 15% with other pathogens). In S aureus PVE, multivariate analysis identified the following predictors of death: prothrombin time < 30% (relative risk [RR]: 8.3), concomitant mediastinitis (RR: 4.9), heart failure (RR: 4.4), and septic shock (RR: 2.6). In PVE due to other pathogens, prothrombin time < 30% (RR: 32.26), renal failure (RR: 7.31), and heart failure (RR: 6.07) were associated with death. In S aureus PVE, survival was higher in patients who received medical-surgical therapy than in those who received medical therapy alone (9/20 [45%] vs 0/20) (p < 0.01). In PVE due to other pathogens, there was no difference in survival between patients who underwent prosthesis replacement (89%) and those who received only medical treatment (81%). Among the 65 patients who underwent heart surgery, the mortality rate and incidence of postoperative paravalvular leakage did not correlate with positive prosthesis cultures. We conclude that non-S aureus and uncomplicated PVE may be managed without valve replacement but that prompt surgical intervention should be required in all other situations.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call