Abstract & Commentary: Landmark study bottom line: CR-BSIs are preventable

Landmark study bottom line: CR-BSIs are preventable

Model used adaptable to other patient safety issues

Synopsis: In a study involving 108 ICUs in Michigan, an evidence-based intervention led to a 66% reduction in catheter-related bloodstream infections that was sustained over an 18-month period.

Source: Pronovost P, et al. An Intervention to Decrease Catheter-Related Bloodstream Infections in the ICU. N Engl J Med 2006; 355:2,725-2,732.

Pronovost and colleagues conducted a multi-hospital cohort study aimed at reducing the rate of catheter-related bloodstream infections (CRBSIs) in ICUs. The intervention focused on increasing adherence to evidence-based CDC recommendations for catheter insertion. In summary, these are:

• hand hygiene;
• full-barrier precautions during insertion;
• skin cleansing with chlorhexidine;
• avoiding the femoral insertion site;
• removal of unnecessary catheters.

Each ICU designated a physician and nurse to serve as team leaders, who were given instruction in patient safety, the CDC guidelines, and data collection. In collaboration with the local infection control practitioners, the teams provided education to clinicians regarding the CDC guidelines. Incentives to compliance included providing a central-line insertion cart with all necessary supplies, discussion of catheter removal at daily rounds, a checklist to document compliance during catheter insertion, and regular feedback regarding the number and rate of CRBSIs in the unit. Unit staff were encouraged to stop providers from inserting catheters if insertion technique did not follow recommendations.

One hundred eight hospitals joined the study, and 103 participated. This represented 85% of all ICU beds in Michigan. The final analysis included 375,757 catheter days. During the baseline (pre-intervention) period the median rate of CRBSIs was 2.7/1000 patient days (range 0.6-4.8). Following intervention, the median rate was 0 (range 0-3.0); this was sustained for 18 months of follow-up. In fact, the rate of CRBSI fell progressively during follow-up. The incidence rate ratio was 0.62 during the first quarter after intervention, falling to 0.34 by the final quarter. To put it another way, at the end of the study, the rate of CRBSIs was 66% less than at baseline.

Commentary by Robert Muder, MD, hospital epidemiologist at the Pittsburgh VA Medical Center. Muder does research for Aventis and Pharmacia.

According to CDC estimates, approximately 50,000 CRBSIs occur in the United States annually. Attributable mortality is as high as 35%, and excess cost per episode is in greater than $40,000. The CDC has recently issued evidence based recommendations for catheter insertion. If these guidelines are indeed effective, and if they are widely implemented, the potential number of lives and dollars that would be saved is enormous.

I have little doubt that the infection control manuals of the vast majority of U.S. hospitals mandate adherence to the CDC guidelines for central catheter insertion. Whether providers actually do so is, of course, another matter entirely. There is ample evidence that purely educational endeavors don't change practice appreciably. Infection control practitioners have neither the time nor the authority to provide monitoring or enforcement. The study by Pronovost and colleagues demonstrates that a different approach can work admirably. Several key aspects of the strategy deserve emphasis. First, team leaders, consisting of a doctor and nurse, were recruited from ICUs in each hospital. These leaders would likely be highly visible on the units, and well known to the physicians and staff. The team leaders received training in the appropriate precautions, and were thus in a position to disseminate this knowledge to their colleagues. Second, compliance was made easier by providing all the necessary materials on a central-line insertion cart. Third, the unit staff received regular and timely feedback on infection rates from the team leaders. Finally, unit staff were given "permission" to prevent catheter insertion when appropriate precautions were not being taken.

The study has some shortcomings. First, it is a quasi-experimental, or "before and after" study, and thus there is no control group of ICUs. The authors note that finding a control group was difficult because all of the participating hospitals wanted to implement the intervention. However, there is no evidence that there was any other intervention or spontaneous trend that could account for the marked decrease in CRBSIs. The sheer number of hospitals and of catheter days, coupled with the fact that different hospitals started the intervention at different times, makes the results credible. One could have wished that the hospitals had reported rates of compliance with the CDC guidelines and correlated that with reduction in bacteremia rates. That is a rather minor flaw given the magnitude of the result.

I believe that two major points deserve emphasis. The first is that CRBSIs are clearly preventable. A major reduction in incidence can occur with a rather modest investment of time and money. There seems little reason all ICUs should not embark on a similar intervention. The second is that the intervention used in this study is a model that ought to be readily adaptable to a wide variety of patient safety issues. In many cases, we know what ought to be done; actually getting it done is has always been the problem.