Laurie Barclay, MD
April 6, 2004 Alcoholic povidone-iodine is better than aqueous in preventing central venous catheter colonization in the intensive care unit (ICU), according to the results of a randomized trial published in the March issue of Critical Care Medicine.
"Cutaneous microorganisms are predominant causes of central venous catheter (CVC)-related infection," write Jean-Jacques Parienti, MD, DTM&H, from the Côte de Nacre University Hospital in Caen, France, and members of the Groupe d'Etude Normand pour l'Antisepsie des Cathéters en Réanimation (NACRE Study Group). "Aqueous 2% chlorhexidine solution has significantly reduced nosocomial catheter-related infections compared with 10% aqueous povidone iodine (PVP-I), but aqueous PVP-I remains the most widely used agent for cleansing the catheter insertion site and for catheter-related care procedures."
From Jan. 1, 2001, to Jan. 1, 2002, consecutive patients requiring a central venous catheter were randomized to one of two similar 11-bed medical ICUs. The alcoholic PVP-I solution protocol was randomly assigned to one of two units when the study began, and it was switched from one unit to the other every three months. Depending on the unit and date of admission, catheters were inserted and cared for with 10% aqueous PVP-I solution or 5% PVP-I solution 70% ethanol-based combination.
Intent-to-treat analysis of 223 catheters revealed that the incidence of catheter colonization was lower with alcoholic PVP-I than with aqueous PVP-I (relative risk [RR], 0.38; 95% confidence interval [CI], 0.22 - 0.65; P < .001), and so was the incidence of catheter-related infection (RR, 0.34; 95% CI, 0.13 - 0.91; P < .04). Both protocols had similar incidence of catheter-related bacteremia.
After adjusting for other risk factors, time to central venous catheter colonization was longer with alcoholic PVP-I than with aqueous PVP-I solution (adjusted hazards ratio, 0.3; 95% CI, 0.2 - 0.6; P < .001). Analysis of 57 pairs of central venous catheters matched for age, duration, and site of insertion yielded similar results, demonstrating the superiority of alcoholic PVP-I solution in preventing central venous catheter colonization and infection.
Study limitations include lack of individual randomization; use of central venous catheter colonization rate rather than central venous catheterrelated bacteremia as a primary end point; failure to control for manipulations of the hubline catheters, which may have influenced the development of intraluminal colonization; greater severity of illness in the alcoholic PVP-I subjects of the subgroup pair-matched analysis; and the exclusion of 83 (27.1%) of 306 central venous catheters.
"The use of alcoholic PVP-I for skin disinfection reduced the incidence of catheter colonization and related infection compared with aqueous 10% PVP-I disinfection in an adult ICU," the authors write. "This study adds further evidence of the better antimicrobial potency of alcohol-based iodophor combinations, which should be preferred to water-based iodophors for intact and clean skin disinfection. Before short-term [central venous catheter] insertion and catheter-related care procedures, the use of alcoholic PVP-I rather than aqueous PVP-I can significantly reduce the incidence of catheter-tip colonization and nosocomial catheter-related infection in ICUs."
The French National Programme Hospitalier de Recherche Clinique supported this study.
