Stewardship, Stewardship, Stewardship

There has been a plethora of antimicrobial stewardship scholarship published these past few weeks. I'm currently on the inpatient medicine service and have even been harassed by the antimicrobial stewardship team (humor), so I only have a moment to briefly highlight three can't miss articles:

(1) Manisha Juthani-Mehta and co-authors just published an excellent JAMA Viewpoint discussing Antimicrobials at the End of Life.  It is open-access (free), so I hope you have a chance to read it thoroughly, but the main points include:

1. "Evidence-based and goal-directed counseling about infection management at the end of life must be a routine part of advance care planning and treatment discussions between clinicians and patients with advanced illness."
2. "Clinical algorithms aimed at improving antimicrobial stewardship from an infectious disease standpoint must also integrate treatment preferences when applied to patients near the end of life." 
3. "To the extent that inadequate outcome data hinder decision making, researchers should consider whether there is adequate clinical equipoise and need to justify a carefully designed randomized trial comparing symptom control and survival among patients with advanced illness who receive antimicrobials vs high-quality palliative care for suspected infections."

(2) Dan Livorsi and colleagues at the Sidney and Lois Eskenazi Hospital and the Richard Roudebush Veterans Affairs Medical Center in Indianapolis just published an important qualitative study in September's ICHE of factors that influence antibiotic prescribing among inpatient physicians (10 resident and 20 staff physicians). I'm happy to add that Dan Livorsi has just joined our group in Iowa City, where he is helping to jump-start our stewardship programs. Key findings of his study include:

1. "Antibiotic overuse is recognized but generally accepted; 
2. the potential adverse effects of antibiotics have a limited influence on physician decision making;
3. physicians-in-training are strongly influenced by the antibiotic prescribing behavior of their supervising staff physicians; and
4. other physicians’ prescribing decisions are sometimes questioned, but there is limited peer-to-peer feedback or critique."

(3) Nick Daneman and colleagues in Ontario examined antibiotic use and secondary harms in 607 nursing homes housing 110,656 residents in a recent JAMA Internal Medicine. Their findings are quite striking (if not surprising) in that antibiotic use varied from a low of 20.4 antibiotic days to a high of 192.9 antibiotic days per 1000 resident days. Antibiotic-related adverse events were higher in "high-use" nursing homes even among patients who did not receive antibiotics. An interesting finding (for someone in Iowa) was that rural facilities were overrepresented in the highest tertile of antibiotic use (see figure below), but after accounting for other nursing home– and patient-level characteristics, rurality was found to be protective against antibiotic-related harms." Would be interesting to figure out why rurality is associated with higher antibiotic use but fewer harms but my guess is that rural folks are just awesome. Of note, Lona Mody and Chris Crnich published an accompanying editorial that is worth reading.

Lovin' Contact Precautions (this time in nursing homes)

Contact precautions get very little love on our humble blog. So little in fact, that I've taken it upon myself to be the resident contact precautions fanboy. Just today on rounds, I was waxing sentimental about the poor yellow gowns that protect us from horrible pathogens and how we unceremoniously toss them into the trash after wearing them - we never even say goodbye...but I digress

There is a new study in the September ICHE by Mary-Claire Roghmann and colleagues from the University of Maryland and University of Michigan that sought to estimate the transmission of MRSA from nursing home residents to healthcare workers' gowns and gloves based on clinical activity and resident characteristics (i.e. skin integrity or stool incontinence).  They aimed to determine if there were certain situations where wearing gowns/gloves would be most protective of HCW contamination (and thus reduce MRSA transmission).  The logic - if gowns and gloves are contaminated, the underlying hands would be contaminated if gloves/gowns weren't worn and since no one has ever gotten hand hygiene compliance near 90-100% anywhere, including nursing homes, then gowns/gloves result in cleaner hands and less MRSA transmission. I know, much more complicated than a cluster-RCT, but important data...but I digress again...

RESULTS! Overall, they enrolled 401 nursing home residents from 13 facilities including 113 (28%) who were MRSA colonized. 62% were nasally colonized, 9% were colonized at the perianal skin and 28% were colonized at both sites. There were 954 HCW interactions (median 7 per patient) with MRSA+ patients with a median duration of 6 minutes. Overall, gowns were contaminated after 14% of the interactions and gloves were contaminated 24% of the time. Gown/Glove isolates were identical (Spa type) to patient isolates 89% of the time. Overall, the contamination rate ranged from zero to 24% for gowns and 8 to 37% for gloves based on activity. So as far as hand contamination goes, there were no safe interactions and thus, we wouldn't expect activity-based precautions to be effective. (See figure below)  Significant predictors of glove/gown contamination included dressing, transferring, patient hygiene, changing linens and changing diapers.

As far as patient characteristics, stool incontinence did not modify gown/glove contamination with MRSA, but skin breakdown was associated with higher contamination when healthcare workers transferred the patient, changed their diapers and helped dress the patient.

My interpretation of the study is that if we want to limit the substantial transmission of MRSA in nursing homes, we better up our game. And that game should probably include gloves and perhaps gowns for most of the analyzed patient care activities. Unless we can get hand hygiene compliance up to 100%, we better just learn to love the glove.

Screening, Decolonization and Environmental Decontamination for MRSA in Nursing Homes Doesn't Work

Just in the past couple of weeks, we've written about pneumonia prevention bundles, MDRO prevention bundles, and spread of S. aureus - all in nursing homes.  It's like no one cares about acute care facilities any more! (humor) There is now more great data for those charged with managing infection control in nursing homes.

Cristina Bellini and colleagues from Lausanne University Hospital in Switzerland just published the results of cluster-randomized trial of an MRSA prevention bundle in 104 nursing homes (53 intervention, 51 control) in the April ICHE. All residents in intervention and control nursing homes (NH), who gave consent, were screened for MRSA carriage at study entry and 12 months thereafter on a single day in each NH. Newly admitted or readmitted residents were screened when admitted to the NH. Screening included nasal, groin and ulcer swabs along with urine cultures if residents had an indwelling catheter. In the intervention NHs MRSA colonized residents underwent decolonization along with environmental decontamination.

The primary decolonization bundle included 5 days of nasal mupirocin, 5 days of CHG oral rinse twice per day, 5 days of CHG showers including CHG shampoo on day 1 and 5. Environmental disinfection included daily clothing changes for 5 days, new linens on day 1 and day 5, and daily bed/table/phone/remote/wheelchair/walker disinfection with 70% alcohol. A lot of steps.

Unfortunately, the MRSA decolonization and decontamination bundle was not successful. The baseline prevalence of MRSA was 8.9% in both groups. The rate declined in intervention units to 5.8% in the intervention unit and 6.6% on the control units after 12 months (p=0.66) Full stratified results are in Table 3 below, and as you can see, no matter how they analyzed the intervention, the MRSA bundle intervention did not reduce MRSA prevalence compared to controls. This was despite the fact that the participation rate was 87%.

A limitation of this study was that they only measured prevalence and not individual level acquisition of MRSA. It is possible that by measuring prevalence they missed detecting benefits of the intervention related to reduced patient-to-patient transmission of MRSA. In any case, as I said last week about the study in CID, congratulations to the authors and journal (this time ICHE) for publishing this important negative study.

MDRO Prevention Bundle in Nursing Homes: A Randomized Trial

In keeping with our long-term care theme (more on this later in the week), there is a new trial published in JAMA Internal Medicine by Lona Mody and colleagues at the University of Michigan. The team completed a cluster-randomized trial of a bundled intervention to prevent MDROs in nursing homes. Specifically, residents with indwelling urinary catheters, feeding tubes or both in 6 nursing homes were randomized to a targeted infection program (TIP) bundle that included (1) preemptive barrier precautions; (2) active surveillance for MDROs (baseline, day 15 and monthly at nares, oropharynx, feeding tube, supra-pubic catheter, groin, peri-rectal, and wound sites) and infections, with data feedback; and (3) NH staff education on key infection prevention practices, including minimum criteria for initiating antibiotics, and hand hygiene promotion. Inclusion required a signed inform consent. Six other nursing homes served as controls.

The outcomes assessed were quite broad. The primary outcome was "overall MDRO prevalence density rate, defined as each participant’s total number of MDRO-positive anatomic sites across all MDROs per visit averaged over the duration of his or her participation." This would result in residents "with persistent MDRO colonization (having) a higher prevalence than someone with intermittent or no colonization" and residents colonized at more sites (up to seven were tested) having a higher prevalence. Secondary outcomes included new MDRO acquisition and device-associated HAI both with 1000 device-day denominators.

For the primary outcome, 27% of swabs were positive in the intervention NH residents while 33% were positive in the control NH residents. The adjusted rate ratio was significant 0.77 (0.62-0.94). This outcome seems largely driven by lower MRSA colonization in residents with urinary catheters, feeding tubes, or both and lower ceftazidime-resistant GNR colonization in residents with urinary catheters in the intervention NHs. However, rates of VRE were higher in the feeding tube requiring residents but this increase was not significant. (see Table 3) Interestingly, new MRSA acquisition rates were lower in the intervention NH residents (see Table 4, below) and first new CAUTI rates were also lower in the intervention NH residents, HR 0.54 (0.30-0.97).

Overall, an important study and one that should be read closely. Clearly this was a very difficult RCT to undertake, especially with informed consent, and the research team should be congratulated. The primary outcome of MDRO prevalence density rate is an interesting choice and the authors make a compelling argument for why they chose it. However, it is unclear if the interventions in the TIP bundle are major components in the causal pathway for limiting MDRO colonization density or reducing CAUTI. However, the lower MRSA acquisition rate in the intervention NHs is an important outcome and does fit with how we expect barrier precautions to work. Minor quibbles aside, this RCT should be discussed widely and many components of it are worth testing in other settings in future trials. 

USA300 invades a nursing home

A study in this month's Emerging Infectious Diseases analyzes the molecular epidemiology of MRSA infections over a 10 year period (1997-2006) at a 1,000 bed long-term care facility in San Francisco. Of the nearly 1,300 patients with S. aureus clinical isolates, 58% were methicillin resistant, with an increase to 72% methicillin resistance in 2006. Most notably, in 2002 11% of MRSA isolates were the USA300 clone with a sharp rise to 64% in 2006. The majority of USA300 strains were associated with skin/soft tissue infections. Given the high rates of transfer of patients between LTCFs and acute care hospitals, the implications of this study are very important, and we must be concerned with how widespread this problem is beyond the LTCF studied.