Facebook
Twitter
LinkedIn
Email
 |
Checking for direct PDF access through Ovid | |
Excerpt
We read with particular attention the article by Raymond and colleagues (1) on the impact of antibiotic rotation in an intensive care unit (ICU) because this investigation adds an interesting contribution to support the potential role of this method, which is still the object of debate. Although the study was very well reported and gave promising results, in our opinion, it includes few, but remarkable, confounding factors that can make the conclusions controversial.
First, the authors identified, as catheter-related infection, the isolation of 15 or more colony-forming units from catheter tips using the semiquantitative roll plate technique in the setting of clinical infection. This definition does not meet the one provided by the Centers for Disease Control and Prevention (CDC) guidelines, which require the isolation of the same organism from a semiquantitative or quantitative culture of a catheter segment and from the blood in the presence of clinical symptoms of bloodstream infection and no other apparent source of infection (2). We think that as a consequence of an incorrect definition, the results reported in Table 10, showing a significant reduction of catheter infection/100 ICU admissions and of crude and attributable mortality between the no-rotation and rotation periods, appear less convincing. Furthermore, most authors refer to the CDC definitions, which makes these findings not comparable with the ones of other studies. These data are worthy of further explanations.
The second point we would like to discuss pertains to the decline in attributable mortality reported by the authors as being related to antibiotic rotation. Infectious mortality in the ICU was defined by Raymond and colleagues (1) as death during antibiotic treatment. We do not agree with this statement, because attributable mortality strictly defines the mortality directly associated with infections, apart from the mortality attributable to underlying conditions (i.e., co-morbidities, acute disturbances, and complications), which are very frequently present in critically ill patients. Nosocomial infections and mortality attributable to other causes often share common risk factors; thus, the assessment of mortality related to infections acquired in the ICU is neither simple nor straightforward. In fact the most reliable approach is to perform matched case-control studies (3). We, therefore, consider the definition adopted by the authors too simplistic, especially taking into account that a reduction in infectious mortality was the main result shown by this work.
During the period of study, alcohol handwashing was introduced. On the assumption that other studies did not demonstrate a significant increase of handwashing compliance after the implementation of this measure, the authors do not believe that this factor had a significant impact on the results. Nevertheless, they do not provide any information about handwashing compliance. We do not agree with this statement because authoritative studies demonstrated that the use of alcohol hand rubs significantly improved handwashing compliance and also contributed to reduced nosocomial infection rates (4). Furthermore, international expert consensus groups agree that hand hygiene is the single most important factor in preventing nosocomial infections (5). We can, therefore, reasonably suppose that introducing this new measure could have had a role in reducing infection rates and associated mortality.
