193. Infectious Complications Following Electrophysiology (EP) Procedures:  Defining Rates to Explore Automated Surveillance Possibilities
Session: Poster Abstract Session: Catheter-associated BSIs
Thursday, October 3, 2013
Room: The Moscone Center: Poster Hall C
Posters
  • IDSA poster 10-2013.pdf (311.4 kB)
  • Background:

    Infectious complication rates for implanted cardiac devices have varied widely, from <1% to greater than 5.6%.  The best method for infection surveillance following EP procedures is unclear. 

    Methods:

    All patients receiving pacemaker and implantable cardioverter-defibrillator procedures at Duke University Hospital from 1/1/2005-12/31/2011 were identified using ICD-9 procedure codes via automated queries.  We then identified infection-related complications among the cohort using ICD-9 diagnosis codes, including septicemia, bacteremia, infection of implantable cardiac device, lead infection, and endocarditis.  Microbiology data from blood, catheter/line and wound and tissue cultures for up to 365 days after the date of the procedure were analyzed.

    Results:

    6097 patients had 7137 procedures during the study timeframe.  61.6% of patients were male and 72.1% were Caucasian.  Using microbiology data, 188 procedures (2.6%) were defined as having potential infections within 365 days.  Alternatively, using diagnosis codes for surveillance, up to 1672 procedures (23%) had infection-related complications within 365 days.  Approximately 90% of positive microbiology data and diagnosis codes were identified within 30 days of the index procedure.  In combining the microbiologic and diagnosis code queries, a total of 1686 procedures (24%) were identified with a potential infectious complication (Figure). 174 procedures (2.4%) were identified with both microbiologic and diagnosis codes.

    Conclusion:

    Our data demonstrate that automated methods that use diagnosis codes, microbiologic culture results, or both would lead to variable measures of post-procedural complications. Microbiology data alone likely underestimate a ‘true' rate of infection, while diagnosis codes likely overestimate this rate.  Validation studies are needed to determine which automated method most closely aligns with clinically-diagnosed infection.

     

     

    Joel C. Boggan, MD, MPH1, Ann Marie Navar-Boggan, MD, PhD1, Lauren Knelson, MSPH2, Rebekah W. Moehring, MD, MPH2, Luke F. Chen, MBBS, MPH, CIC, FRACP3 and Deverick J. Anderson, MD, MPH3, (1)Department of Medicine, Duke University Medical Center, Durham, NC, (2)Division of Infectious Diseases, Duke University Medical Center, Durham, NC, (3)Duke Infection Control Outreach Network, Duke University Medical Center, Durham, NC

    Disclosures:

    J. C. Boggan, None

    A. M. Navar-Boggan, None

    L. Knelson, None

    R. W. Moehring, None

    L. F. Chen, None

    D. J. Anderson, None

    Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.