Good design vs. bad bugs

Graphic:  TM Osborn Associates

A new paper in Critical Care (full text here) attempts to evaluate whether improved ICU design has an impact on transmission of epidemiologically important pathogens, such as VRE and MRSA. The authors compare the proportion of patients acquiring these pathogens before and after an ICU moved from an open unit to one with all private rooms, and also compared this to another ICU that maintained an open floor plan during the study period. In the ICU that moved to all private rooms, the proportion of patients acquiring a targeted pathogen fell from 23% to 5%, whereas in the control ICU there was no change (20% in the corresponding pre-move period, and 18% in the post-move period). So the authors conclude that better ICU design leads to fewer infections.

The study has a number of problems--it's small and relatively short. Moreover, in the intervention ICU, prior to the move the patients were sicker, had received more antibiotics prior to ICU admission, and had a higher prevalence of multidrug-resistant organisms on admission. All of these factors could be responsible for the higher proportion of patients acquiring the bad bugs in the "old" ICU, making the "new" ICU look better than perhaps it was. Also, importantly, hand hygiene compliance was significantly higher in the intervention ICU than in the control ICU in the post-move period, though one could argue that improved hand hygiene was due to better design (e.g., better availability of sinks).

While I remain unconvinced regarding the conclusions of this particular study, it seems to me that for many reasons having ICUs with all private rooms is a good thing. Infection prevention, since time immemorial has been about separating clean from dirty and infected from uninfected. It just makes sense--kind of like not wearing the same dirty white coat to work every day.

Antibiotics: a market failure

Photo:  Summit Total Health

There's a very nice article in the October edition of the Atlantic on the problems of antibiotic development and antibiotic resistance viewed from an economics perspective. Full text of the article can be found here.

A four billion year head start

A recent article in Nature confirms that the sound you hear, late at night, when all else is quiet….is the sound of bacteria laughing…laughing at the idea that humans can win the antimicrobial resistance battle. An excerpt from the abstract describes their findings after isolating ancient bacterial DNA from Canadian permafrost:

“We report targeted metagenomic analyses of rigorously authenticated ancient DNA from 30,000-year-old Beringian permafrost sediments and the identification of a highly diverse collection of genes encoding resistance to β-lactam, tetracycline and glycopeptide antibiotics. Structure and function studies on the complete vancomycin resistance element VanA confirmed its similarity to modern variants. These results show conclusively that antibiotic resistance is a natural phenomenon that predates the modern selective pressure of clinical antibiotic use.”

Pulling out antibiotics that bacteria could defend against 30,000 years ago is kind of like….bringing a sword to gun battle.

Papers you should read, September 2011

We know. Postings have been light these past few weeks. Hey, we are getting ready for ICAAC and IDSA. So cut us some slack, already...(joking)

OK, in a matter of full disclosure, I have conflicts of interest with both of these papers, so following our standard practice of not critiquing papers where we have conflicts, I will leave the interpretation of each paper to you, our dear readers.

Click to enlarge: Frequency of non-prescription use of antimicrobials in the general population

Dan Morgan, who was our invited contributor on Veterans Day, has published a review of world-wide, non-prescription antimicrobial use in this months Lancet ID.  The review included 117 articles including 35 community surveys published between 1970 and 2009. I've pasted a summary figure above.

Co-blogger Dan has a paper coming out in ICHE (ahead of print) surveying IDSA's EIN Network of ID clinicians on their pre-op screening and decolonization practices for S. aureus. The survey included 488 respondents, who had knowledge of their hospital's pre-op screening. Overall, 60% screened for S. aureus with 47% screening for MRSA only and 13% for all S. aureus. Less than 20% screened extranasally. Screening method was led by PCR (36%) followed by standard culture (30%) and chromogenic agar (27%). Additionally, 52% decolonized with 31% decolonizing MRSA carriers, 8% decolonizing S. aureus carriers and 15% decolonized irrespective of colonization status. To find out more, you need to read the paper!

Clinical Microbiology Camp!

In February of this year, a large group of clinical microbiologists, employees of laboratory diagnostic companies, and an FDA representative met for a symposium in Houston (it was entitled, “Camp Clin Micro”!). The symposium consisted of a series of facilitated discussions of hot topics in diagnostic microbiology. I enjoyed the give-and-take during these sessions, and thought the interaction between hospital laboratory directors and industry scientists was especially valuable. The written summaries of this symposium are now available free as a supplement to the Journal of Clinical Microbiology, including a summary of the session on the role of clinical microbiology in infection prevention.

Hot topics in the news

Over the past week, there were 2 studies related to infection prevention that attracted a fair amount of attention by the mainstream media.

A paper in Psychological Science (full text here), looked at whether various messages on signs placed near sinks in hospitals could have an impact on handwashing compliance. Of note, the two authors of the study have no medical training--they're professors in business schools. They randomly assigned one of three signs to be posted near 66 soap dispensers, with the following messages:

• Hand hygiene prevents you from catching diseases
• Hand hygiene prevents patients from catching diseases
• Gel in, wash out (control message)

They hypothesized that healthcare workers are overconfident and overestimate their own invulnerability, thus a message that focused on protecting the patient rather than themselves would be more effective. Two experiments were performed: in the first, the effect of the messages was measured by the volume of soap used in the dispensers; in the second experiment, conducted 9 months later, covert observers who worked on the study units recorded hand hygiene compliance by direct observations. The results were a 33% increase in volume of product usage, and a 10% increase in hand hygiene compliance at sinks which had the sign that focused on protecting patients. Although these results were statistically significant, this study has many flaws. The methods section of the paper tells us little about the setting--the type of hospital, the size of the hospital, on how many patient care units the signs were posted, or the type of care units involved. The experiments were brief, each lasting only 2 weeks. We are not told how many healthcare workers were involved; this is particularly important since the number of hand hygiene observations was very small (a total of 322 opportunities before the intervention and 245 opportunities after the intervention). When the results were categorized by practitioner type (nurse, physicians, ancillary staff), the numbers of observations were  extremely small. All in all, it's hard to put much stock in this short, small study in a single hospital.

The other study in AJIC, cultured the clothing (white coats, nurse uniforms, and OR scrubs) of a convenience sample of 135 healthcare workers  in a hospital in Jerusalem. Cultures were set up to detect S. aureus, Enterobacteriaceae, Pseudomonas spp and Acinetobacter spp. Overall, 50% of the cultures were positive for one of the above pathogens. Acinetobacter was most common, found on 32% of white coats, 38% of nurses uniforms, and 43% of OR scrubs. So we have another study that documents what we already know: healthcare worker clothing is commonly contaminated. But the key question remains as to whether the organisms can be transferred to patients. More on that here.

Conflicts of Interest: Beware, you're human!

We've each posted frequently on conflicts of interest, financial and otherwise. Here, Dan Ariely talks briefly about scientific conflicts of interest - they aren't all financial - and what hope we might have to protect ourselves from other people's and our own conflicts. Dr. Ariely is the James B. Duke Professor of Psychology and Behavioral Economics at Duke University and a founding member of the Center for Advanced Hindsight.  I wonder if his Center is really good at determining if a CLABSI is secondary or not. Enjoy.