HOUSTON—Intravascular catheters are a well-known source of nosocomial bacteremia in critically ill patients. In the late 1990s, catheters impregnated with minocycline and rifampin were introduced. Although their use has raised concerns about antibiotic resistance, a large study conducted by researchers at the University of Texas has demonstrated that these catheters significantly reduce the incidence of nosocomial bloodstream infections—including those caused by resistant organisms.[1]

TWO YEAR STUDY

The study was conducted between September 1997 and August 1999 in the surgical and medical ICUs of the M.D. Anderson Cancer Center in Houston. In the study’s first year, most of the central venous catheters used in the units were untreated. In the second year, most were impregnated with minocycline and rifampin. In both years, blood samples for culture were obtained from patients with unexplained fever, and positive findings were investigated for potential sources of infection.

In the first year, 653 patients were admitted to the medical ICU and 1,128 were admitted to the surgical ICU. In the second year, the numbers were 764 and 1,585, respectively. In the medical ICU, the use of antibiotic-impregnated catheters increased from 1% in the first year to 96% in the second year. In the surgical ICU, their use increased from 4% to 72%.

INFECTION RATES DECREASED

Between the first and second years of the study, the rate of nosocomial bloodstream infections in the medical ICU decreased from 8.3 to 3.5 per 1,000 patient-days, despite a longer duration of catheterization in the second year. In the surgical ICU, infection rates dropped from 4.8 to 1.3 per 1,000 patient-days. The decrease in the bacteremia rate was accompanied by significant decreases in the lengths of stay in both the medical and surgical ICUs. In addition, hospital length of stay was significantly decreased in the surgical ICU patients, and there was a definite trend toward decreased length of hospital stay among medical ICU patients.

When the medical and surgical patients were considered as one group, the rate of nosocomial bloodstream infections dropped from 6.3 to 2.2 per 1,000 patient-days between the first and second years, and the rate of confirmed catheter-related infections declined from 3.2 to 0.6 per 1,000 patient-days. In addition, the rate of local catheter site infections dropped from 1.8 to 1.2/1,000 patient-days, and the rate of vancomycin-resistant enterococcal infections decreased from more than 2 per 1,000 patient-days to near zero.

The Houston researchers noted that the reduction in nosocomial bloodstream infections could not be explained by any factor other than the introduction of antibiotic-impregnated catheters. The amount of antibiotics used in both ICUs did not differ between the first and second years, indicating that antibiotic administration did not contribute to the decreased rate of nosocomial bacteremia or vancomycin-resistant bacteremia.

One potential confounding factor was that the patients were moved to a new ICU in the second year. However, the decrease in the rates of nosocomial bacteremia and other types of infection had occurred before the move. According to the researchers, the decrease in nosocomial bloodstream infections resulted in a net savings of at least $1,450,000 in the study’s second year.

—Gale Jurasek

Reference
1. Hanna HA, Raad II, Hackett B, et al, and the M.D. Anderson Catheter Study Group. Antibiotic-impregnated catheters associated with significant decrease in nosocomial and multidrug-resistant bacteremias in critically ill patients. Chest. 2003; 124:1030-1038.

Return to table of contents