ABSTRACT
BACKGROUND: A steady rise in the use of cardiovascular implantable electronic devices (CIEDs), particularly in the elderly, has led to an increase in device-related infections. Although often studied and reported as a single entity, these complications in fact comprise a heterogeneous group. Specific subgroups may be associated with distinct mortality risks. METHODS: Medical records of all patients who underwent device extraction for CIED-related infection at a single tertiary referral center between 1991 and 2007 were reviewed. Infections were divided into four subgroups: primary pocket site infection (PPSI), pocket site infection with bacteremia, primary/isolated bacteremia (PIB), and device-related infective endocarditis (DRIE). Clinical presentation, laboratory data, and mortality rates were obtained by chart review and by querying the Social Security Death Index. RESULTS: A total of 387 cases were analyzed. The overall in-hospital and 1-year all-cause mortality rates were 7.2% and 25.3%, respectively. Patients with PIB or DRIE had significantly higher mortality rates (hazard ratio [HR] 2.3; 95% confidence interval [CI] 1.2-4.6 and HR 2.5; 95% CI 1.6-4.1, respectively) when compared with patients in the PPSI group. Patients who did not receive a new device during the initial admission also had a higher 1-year mortality rate compared to those who did (HR 2.7; 95% CI 1.8-4.1). CONCLUSIONS: Our patients with CIED-related infections requiring extraction/hospitalization had a significant mortality risk. Presence of pocket site infection carried a more favorable prognosis, regardless of the presence of bacteremia. Early detection and prevention of CIED-related infections with PIB (i.e., no pocket site involvement), especially for high-risk populations, is needed.