February 2019: Antimicrobial Stewardship and Infection Control

Colonization with C. difficile on admission is a risk factor for C. difficile infection during the hospital stay.  The authors conducted C. difficile surveillance on all medical admissions to their hospital over a ten week period using rectal swabs and a PCR assay.  They compared the demographics and clinical characteristics of the carriers vs the non-carriers and stratified the patients based on whether they went on to develop CDI.

A total 3803 patients had a total 4601 hospitalizations; of these, 2368 patients had analyzable data, and 81 (3.4%) were C. difficile carriers. In multivariate analysis, carriers were older and more likely to have had a previous hospitalization than non-carriers; in addition, they were more likely to be prescribed antimicrobials during the current admission.  The incidence of CDI was much higher in the carriers vs non-carriers (76.7 vs 4.6 cases/10,0000 patient-days; RR 16.6, p = 0.002).

Of course, this doesn’t tell us whether decolonizing these folks would lower their risk of CDI (potentially this is just a marker of morbidity and susceptibility to CDI), but I think that’s biologically plausible and worth a clinical trial, assuming one hasn’t been done already. 30771530


Decolonizing inpatient MRSA carriers reduces risk of postdischarge MRSA infection. This was a multicenter RTC randomizing hospitalized adults being discharged to either hygiene education or decolonization with chlorhexidine mouthwash, chlorhexidine showers, and a 5-day course of nasal mupirocin twice monthly for six months.  The outcome of interest was whether patients developed an infection due to MRSA in the year following discharge.

A total 1063 participants were randomized to receive the education and 1058 to receive decolonization, and the rates of postdischarge MRSA infection were 9.2% and 6.3%, respectively.  MRSA infection led to a repeat hospitalization in 85% of cases.  In multivariate analysis, decolonization was associated with fewer MRSA infections versus education (HR 0.7; p=0.03), with a number needed to treat of 30 to prevent one MRSA infection.  The decolonization group also had overall lower hazard of infection (HR 0.82; 95% 0.7-0.99) and infection-related hospitalization (HR 0.76; 95% CI 0.62-0.93).  Decolonization was well-tolerated, with only 4.2% experiencing an adverse event, all of which were mild.

To conclude, a six-month MRSA decolonization regimen produced a 30% reduction in MRSA-related infections and a similar reduction in repeat hospitalizations, with NNTs in the 30s. Given the relative cost of another inpatient stay, this seems likely to be a cost-saving intervention. 30763195


Early ID follow-up reduces 30-day readmissions in patients receiving outpatient parenteral antibiotic therapy. This was a case-control study conducted at a single hospital over a two-year period.  Standard of care for OPAT at this center included weekly labs with at least a CBC, electrolytes, and BUN/Cr, and also including other labs – e.g. drug levels, CK, and liver enzymes – as indicated by the specific drug regimen. Patients, identified via the hospital’s OPAT database, were defined as cases if they were discharged from the hospital on OPAT had a 30-day readmission for any cause; controls were patients who received OPAT but did not have a readmission.  Matching cases and controls was based on age, sex, year of discharge, and discharge diagnosis.

A total 1102 patients participated in OPAT during the study period, of whom 201 had a 30-day readmission; forty-six percent of readmissions were related to OPAT, and forty percent occurred within a week of discharge.  In total the authors matched 194 cases to controls.  Risk factors for readmission included immunosuppression, use of quinolones, metronidazole, antifungals, or regimens with three or more antimicrobials, and not having a follow-up appointment in ID clinic within two weeks of discharge.  In multivariate analysis, ID follow-up within 2 weeks emerged as an important protective factor against readmission (aOR 0.33; 95% 0.19-0.59); interestingly, use of ceftriaxone was also associated with fewer readmissions (aOR 0.49), which might be due to those patients being treated for more straightforward infections.  Immunsuppression remained a risk factor for readmission (aOR 2.79).

These data are limited by their retrospective nature and the matched case:control design; that said, the sample size is large, the effect size is meaningful, and the association between early clinic followup and reduced readmission is plausible.  This is likely the highest quality data on the importance of timely ID clinic followup we’re going to get, so go take this paper to your boss and explain why it’s going to be good for your patients (and your hospital’s bottom line) to expand the ID clinic and make sure wait times are low.  30721936