ABSTRACT
BACKGROUND: The isolation of a pathogen is vital in the diagnosis and treatment of a device infection. A swab culture, despite poor sensitivity, is the most common method used in specimen collection. OBJECTIVE: To determine the relative value of swab and tissue specimen cultures in patients with implantable cardiac pacemakers and defibrillators. DESIGN: Prospective patient cohort study. SETTING: A 1,000-bed tertiary referral center in Cleveland, Ohio. PATIENTS: Consecutive patients with implantable cardiac pacemaker or defibrillator presenting for lead extraction from October 1, 2000 to March 31, 2001. METHODS: Tissue and swab cultures were prospectively collected during pacemaker and implantable defibrillator surgeries that required lead extraction. Clinical manifestations, microbiology, and echocardiographic data were recorded in patients with and without a clinical diagnosis of device system infection. RESULTS: Seventy-one patients with implantable pacemaker (n = 49, 69%), implantable defibrillator (n = 18, 25%), or both devices (n = 4, 6%) requiring lead extraction had pocket swab and tissue cultures for analysis. Infection was evident clinically in 35 (49%) of the patients and absent in the remainder. The most common bacteria isolated were coagulase-negative Staphylococcus (37%) and Staphylococcus aureus (10%). Patients with clinical infection had positive cultures more frequently (P = 0.002) by pocket tissue culture (n = 24, 69%) than by swab culture (n = 11, 31%). However, patients without clinical infections had positive cultures at similar rates by pocket tissue culture (n = 10, 28%) and by swab culture (n = 8, 22%; P = 0.48). Patients without clinical infection were not treated with other than perioperative antibiotics, and did not develop clinical infections. CONCLUSION: Pocket tissue cultures are more effective than pocket swab cultures for the isolation and identification of the infectious pathogens in cardiac device infections. Positive cultures by pocket swab or tissue cultures in the absence of clinical signs and symptoms of infection does not imply infection or the need for specific therapy.
