Risk factors for mortality in patients with cardiac device-related infection.

BACKGROUND Because of the increased use of pacemakers and implantable cardioverter defibrillators, infection has become a complication with significant morbidity and mortality. Data on risk factors for mortality in patients with cardiac-device related infection are limited. We evaluated the prognostic significance of key clinical and echocardiographic variables in a large retrospective population of patients with cardiac-device related infection. METHODS AND RESULTS Two hundred ten patients with cardiac-device related infection were identified at the University of Michigan between 1995 and 2006. Data were abstracted on key clinical and echocardiographic variables, treatment strategy, and 6-month outcomes. We used multivariable Cox proportional hazards models to examine clinical and echocardiographic variables that were associated with 6-month mortality. Mean age for our study population was 63+/-17 years, and 72 (44%) were women. All-cause 6-month mortality was 18% (n=37). Independent variables associated with death were systemic embolization (hazard ratio 7.11; 95% CI 2.74 to 18.48), moderate or severe tricuspid regurgitation (hazard ratio 4.24; 95% CI 1.84 to 9.75), abnormal right ventricular function (hazard ratio 3.59; 95% CI 1.57 to 8.24), and abnormal renal function (hazard ratio 2.98; 95% CI 1.17 to 7.59). Size and mobility of cardiac device vegetations were not independently associated with mortality. CONCLUSIONS We identified several clinical and echocardiographic variables that identify patients with cardiac-device related infection who are at high-risk for mortality and may benefit from more aggressive evaluation.