Current issues of ACP Journal Club are published in Annals of Internal Medicine


Etiology

Review: Transparent polyurethane-film catheter dressings for central or peripheral venous catheters compared with gauze dressings increases the risk for catheter-tip infections

ACP J Club. 1992 Sept-Oct;117:58. doi:10.7326/ACPJC-1992-117-2-058


Source Citation

Hoffmann KK, Weber DJ, Samsa GP, Rutala WA. Transparent polyurethane film as an intravenous catheter dressing. A meta-analysis of the infection rates. JAMA. 1992 Apr 15;267:2072-6.


Abstract

Objective

To compare the risks for infection associated with transparent polyurethane film and gauze dressings used on central and peripheral venous catheters in hospitalized patients.

Data sources

A MEDLINE search from 1966 to June 1991 was done using the terms occlusive or transparent dressings and wound infections. Bibliographies of relevant articles were scanned, and articles, abstracts, and letters were collected.

Study selection

All randomized controlled trials using hospitalized patients were included. Studies were excluded if they did not contain original data, outcome measures were not adequately described, subcutaneously tunneled catheters were used, or the control dressing was paper tape. 7 studies (5 articles, 2 abstracts) using central venous catheters and 7 (6 articles, 1 abstract) using peripheral venous catheters were included.

Data extraction

Risk ratios for infectious complications were calculated for each study, and their heterogeneity was assessed. Data were collected on catheter-tip infections (defined as > 15 colony-forming units using a semiquantitative culture method), bacteremia, and catheter-related sepsis for central venous catheters, and on phlebitis, infiltration, skin colonization, and catheter-tip infections for peripheral venous catheters.

Results

Topical antibiotic agents were used in 5 of the central venous catheter studies and in 2 studies of peripheral venous catheters. 1200 patients with central venous catheters were evaluated for catheter-tip infections and > 650 for bacteremia. The relative risk (RR) for catheter-tip infection comparing the transparent dressing with gauze was 1.78 (95% CI 1.38 to 2.30, P < 0.001); for bacteremia the RR was 1.63 (CI 0.76 to 3.47, P = 0.2); and for catheter sepsis the RR was 1.69 (CI 0.97 to 2.95, P = 0.06). More than 3000 patients with peripheral venous catheters were evaluated for all outcomes. The RR for catheter-tip infection for peripheral venous catheters was 1.53 (CI 1.18 to 1.99, P = 0.002). The transparent dressing group did not differ from the gauze group for phlebitis (RR 1.02, CI 0.86 to 1.20), infiltration (RR 1.12, CI 0.92 to 1.37), or skin colonization (RR 0.99, CI 0.90 to 1.09).

Conclusion

Transparent polyurethane-film catheter dressings for central or peripheral venous catheters compared with gauze dressings increases the risk for catheter-tip infections in hospitalized patients.

Source of funding: Not stated.

Address for article reprint: Ms. K.K. Hoffmann, Division of Infectious Diseases, CB#7030, Burnett-Womack 547, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7030, USA.


Commentary

Nosocomial blood infections are major causes of morbidity. Transparent venous catheter dressings have been implicated as a risk factor for these infections, but evidence from individual studies has been inconclusive. This excellent meta-analysis clarifies the conclusions we should draw from the evidence by compiling data from smaller studies of hospitalized patients in comparative trials of transparent and gauze venous catheter-site dressings. The increased relative risk that was found for all outcomes associated with central catheters and transparent dressings is important. Although statistical significance was only achieved for central and peripheral catheter-tip infections, the authors' suggestion that tip infections may predispose to more serious blood-stream infections is reasonable. The trend in the analysis toward increased relative risk for bacteremias and catheter sepsis with central catheters supports this suggestion.

One should not conclude, based on this information, that a simple switch to gauze dressing will result in a reduced rate of catheter-associated, nosocomial blood-stream infections. The results of the meta-analysis could be confounded by differences in the trials, such as application of topical antimicrobials, frequency of gauze dressing changes, type of transparent dressing, and perhaps, differences in venous catheters. Also the choice of catheter dressings will continue to involve considerations such as nursing time and product cost. Nevertheless, unless better studies are reported, this meta-analysis suggests that the use of gauze dressings will limit some infectious complications of venous catheter use compared with the use of transparent dressings. Infection control practitioners and other clinicians will find these data useful when establishing policies to control catheter-associated infections. With new, more permeable transparent dressings available, the results of this meta-analysis should be verified by a large trial, controlling for possible confounders.

P. J. Brennan, MD
University of PennsylvaniaPhiladelphia, Pennsylvania, USA

P. J. Brennan, MD
University of Pennsylvania
Philadelphia, Pennsylvania, USA