Medicine Ward Patients in a Large Teaching Hospital: A High-Risk Population for Catheter-Associated Bloodstream Infections?

Abstract

BACKGROUND: Infection control surveillance for catheter-associated bloodstream infections (CA-BSIs) is often focused in intensive care units (ICUs), where national benchmark data is available. Limited data on CA-BSI rates are available for non-ICU areas. Increasing acuity among non-ICU patients with central venous catheters (CVCs) raised the need to evaluate these areas. OBJECTIVE: To determine the CA-BSI rate in selected non-ICU medical wards over a 13-month period. METHODS: An infection control specialist performed active surveillance for CA-BSIs on four medical wards (A, B, C, D) from April 2002 to April 2003. CA-BSI was defined using NNIS definitions. CVC days were obtained from an electronic report based on nurses' CVC charting. Type of CVC and CA-BSI microbiology was documented. RESULTS: The general medicine wards each average 5–10 admissions/day. Each ward has separate nursing staff. A medical director, nurse manager, and interns/residents are shared between wards A & B and between wards C & D. The mean monthly CVC days for the 13-month period was 141 (range 77-215) per ward. Device utilization ratios for the four wards were 0.21, 0.25, 0.19, 0.24, respectively. Forty-two episodes of CA-BSI were identified. The mean CA-BSI rate in Ward A was 5.3/1000 CVC days (n=9); Ward B: 8.0/1000 CVC days (n=16); Ward C: 4.3/1000 CVC days (n=7); and Ward D: 4.9/1000 CVC days (n=10). CA-BSI organisms were: polymicrobial, 12% (n=5); Enterococcus spp., 24% (n=10); Coagulase-negative staphylococci, 24% (n=10); Staphylococcus aureus, 7% (n=3); other gram-positive, 2% (n=1); Pseudomonas aeruginosa, 2% (n=1); Escherichia coli, 2% (n=1); Proteus mirabilis, 2% (n=1); other gram-negative, 10% (n=4); Candida albicans, 12% (n=5), and other Candida, 2% (n=1). Patients with CA-BSI had the following catheters: tunneled, 5% (n=2); non-tunneled, 83% (n=35); and implanted port, 7% (n=3). Two patients had more than one type of CVC. CONCLUSION: Non-ICU medical wards in our study had device utilization ratios that were considerably lower than medical ICUs, but CA-BSI rates were comparable to medical ICUs nationally. Studies of catheter utilization rates and knowledge of CVC care and insertion on medical wards should be performed. Interventions usually focused on ICU personnel may be indicated to prevent CA-BSI on medical wards.