A randomized trial comparing povidone-iodine to a chlorhexidine gluconate-impregnated dressing for prevention of central venous catheter infections in neonates

Neonates who require a central venous catheter (CVC) for prolonged vascular access experience high rates of catheter-related bloodstream infection (CRBSI). Purpose. A multicenter randomized clinical trial was undertaken to ascertain the efficacy of a novel chlorhexidine-impregnated dressing (Biopatch Antimicrobial Dressing) on the CVC sites of neonates for the prevention of catheter tip colonization, CRBSI, and bloodstream infection (BSI) without a source. Setting. Six level III neonatal intensive care units. Patients Studied. Neonates admitted to study units who would require a CVC for at least 48 hours. Methods. Eligible infants were randomized before catheter placement to 1 of the 2 catheter site antisepsis regimens: 1) 10% povidone-iodine (PI) skin scrub, or 2) a 70% alcohol scrub followed by placement of a chlorhexidine-impregnated disk over the catheter insertion site. A transparent polyurethane dressing (Bioclusive Transparent Dressing) was used to cover the insertion site in both study groups. Primary study outcomes evaluated were catheter tip colonization, CRBSI, and BSI without an identified source. Results. Seven hundred five neonates were enrolled in the trial, 335 randomized to receive the chlorhexidine dressing and 370 to skin disinfection with PI (controls). Neonates randomized to the antimicrobial dressing group were less likely to have colonized CVC tips than control neonates (15.0% vs 24.0%, relative risk [RR]: 0.6 95% confidence interval [CI]: 0.5-0.9). Rates of CRBSI (3.8% vs 3.2%, RR: 1.2, CI: 0.5-2.7) and BSI without a source (15.2% vs 14.3%, RR: 1.1, CI: 0.8-1.5) did not differ between the 2 groups. Localized contact dermatitis from the antimicrobial dressing, requiring crossover into the PI treatment group, occurred in 15 (15.3%) of 98 exposed neonates weighing less than or equal to 1000 g. No neonates in the PI group developed contact dermatitis. Conclusion. The novel chlorhexidine- impregnated dressing, replaced weekly, was as effective as cutaneous disinfection with 10% PI and redressing the site every 3 to 7 days for preventing CRBSI and BSI without a source in critically ill neonates requiring prolonged central venous access. The risk of local contact dermatitis under the chlorhexidine dressing limits its use in low birth weight infants who require prolonged central access during the first 2 weeks of life.