Antibiotic-impregnated catheters associated with significant decrease in nosocomial and multidrug-resistant bacteremias in critically ill patients.

OBJECTIVE

To evaluate the impact of using central venous catheters (CVCs) impregnated with the combination of minocycline and rifampin on nosocomial bloodstream infections (BSIs), morbidity, and mortality in cancer patients in the ICU.

DESIGN

Prospective surveillance study consisting of the following two time periods: September 1997 through August 1998 (ie, fiscal year [FY] 1998); and from September 1998 through August 1999 (ie, FY 1999).

SETTING

ICUs of a tertiary care hospital in Houston, TX.

PATIENTS: Cancer patients in the medical ICU (MICU) and surgical ICU (SICU).

INTERVENTIONS: ICUs started using CVCs impregnated with the minocycline-rifampin combination at the beginning of FY 1999.

MEASUREMENTS AND MAIN

RESULTS

The rates of nosocomial BSIs and other patients' characteristics were compared for the two study periods to determine the impact of using the impregnated catheters in the ICU. Patients' characteristics, including antibiotic use, were comparable for the two study periods in both the MICU and the SICU. The rate of nosocomial BSIs in the MICU unit decreased from 8.3 to 3.5 per 1,000 patient-days (p < 0.01), and decreased in the SICU from 4.8 to 1.3 per 1,000 patient-days (p < 0.01) in FY 1999. Nosocomial vancomycin-resistant enterococcus (VRE) bacteremia also decreased significantly (p = 0.004). Length of stay in the MICU and SICU significantly decreased in FY 1999 (p < 0.01 and p = 0.03, respectively). The duration of hospitalization decreased for MICU and SICU patients (p = 0.06 and p < 0.01, respectively). The rate of catheter-related infections decreased from 3.1 to 0.7 per 1,000 patient-days in FY 1999 (p = 0.02). The decrease in infections resulted in net savings of at least $1,450,000 for FY 1999.

CONCLUSIONS

The use of antibiotic-impregnated CVCs in the MICU and SICU was associated with a significant decrease in nosocomial BSIs, including VRE bacteremia, catheter-related infections, and lengths of hospital and ICU stays.