Clinical Characteristics and Outcome of Infective Endocarditis Involving Implantable Cardiac Devices

Eugene Athan, Vivian Chu, Pierre Tattevin, Christine Selton-Suty, Philip Jones, Christoph Naber, Jose Miro, Salvador Ninot, Nuria Fernandez-Hidalgo, Emanuele Durante-Mangoni, Denis Spelman, Bruno Hoen, Tatjana Lejko-Zupanc, Enrico Cecchi, Frank Thuny, Margaret Hannan, Paul Pappas, Margaret Henry, Vance Fowler, Anna Lisa CrowleyAndrew Wang, David Gordon

    Research output: Contribution to journalArticlepeer-review

    192 Citations (Scopus)


    Context: Infection of implantable cardiac devices is an emerging disease with significant morbidity, mortality, and health care costs. Objectives: To describe the clinical characteristics and outcome of cardiac device infective endocarditis (CDIE) with attention to its health care association and to evaluate the association between device removal during index hospitalization and outcome. Design, Setting, and Patients: Prospective cohort study using data from the International Collaboration on Endocarditis-Prospective Cohort Study (ICE-PCS), conducted June 2000 through August 2006 in 61 centers in 28 countries. Patients were hospitalized adults with definite endocarditis as defined by modified Duke endocarditis criteria. Main Outcome Measures: In-hospital and 1-year mortality. Results: CDIE was diagnosed in 177 (6.4% [95% CI, 5.5%-7.4%]) of a total cohort of 2760 patients with definite infective endocarditis. The clinical profile of CDIE included advanced patient age (median, 71.2 years [interquartile range, 59.8-77.6]); causation by staphylococci (62 [35.0% {95% CI, 28.0%-42.5%}] Staphylococcus aureus and 56 [31.6% {95% CI, 24.9%-39.0%}] coagulase-negative staphylococci); and a high prevalence of health care-associated infection (81 [45.8% {95% CI, 38.3%- 53.4%}]). There was coexisting valve involvement in 66 (37.3% [95% CI, 30.2%- 44.9%]) patients, predominantly tricuspid valve infection (43/177 [24.3%]), with associated higher mortality. In-hospital and 1-year mortality rates were 14.7% (26/177 [95% CI, 9.8%-20.8%]) and 23.2% (41/177 [95% CI, 17.2%-30.1%]), respectively. Proportional hazards regression analysis showed a survival benefit at 1 year for device removal during the initial hospitalization (28/141 patients [19.9%] who underwent device removal during the index hospitalization had died at 1 year, vs 13/34 [38.2%] who did not undergo device removal; hazard ratio, 0.42 [95% CI, 0.22- 0.82]). Conclusions: Among patients with CDIE, the rate of concomitant valve infection is high, as is mortality, particularly if there is valve involvement. Early device removal is associated with improved survival at 1 year.

    Original languageEnglish
    Pages (from-to)1727-1735
    Number of pages9
    JournalJAMA-Journal of The American Medical Association
    Issue number16
    Publication statusPublished - 25 Apr 2012


    Dive into the research topics of 'Clinical Characteristics and Outcome of Infective Endocarditis Involving Implantable Cardiac Devices'. Together they form a unique fingerprint.

    Cite this