By Richard R. Watkins, MD, MS, FACP
Division of Infectious Diseases, Akron General Medical Center, Akron, OH; Associate Professor of Internal Medicine, Northeast Ohio Medical University, Rootstown, OH
SYNOPSIS: A retrospective observational study found that among hospitalized patients, ward-level antibiotic prescribing was associated with a significantly increased risk for C. difficile infection beyond what would be expected with patient-level antibiotic use.
SOURCE: Brown K, et al. Hospital ward prescribing and the risks of Clostridium difficile infection. JAMA Intern Med 2015;175:626-633.
Clostridium difficile infection (CDI) is a prevalent and challenging disease in clinical practice. Although the incidence of community-acquired CDI is increasing, the majority of cases occur in health care facilities. The main risk factor for CDI is antibiotic exposure, with clindamycin, cephalosporins, and fluoroquinolones associated with greater risk compared with other antibiotic classes. Brown and colleagues aimed to determine the effect of ward antibiotic prescribing on ward CDI incidence and if there are factors that impact the risk of infection beyond the direct effects from the antibiotics.
The study was a retrospective cohort from a single institution that included patients aged 18 years and older during a four-year period. Individual risk factors for CDI were assessed, including age, sex, admission unit (surgical, medical, or oncology), number of previous admissions, use of a feeding tube, and inpatient medications including antibiotics, antacids, and chemotherapy. The authors also recorded characteristics of the ward population, including mean age, feeding tube use (tube in situ per 100 patient-days), medication usage (antibiotics, chemotherapy drugs, and antacids) in days of therapy per 100 patient-days, patient density, and hand hygiene adherence. Patients were excluded who received metronidazole, oral vancomycin, or fidaxomicin. Individual antibiotic risk was classified according to whether the patient received a high-risk antibiotic (cephalosporins, carbapenems, fluoroquinolones, or clindamycin), a medium-risk antibiotic (penicillins, sulfonamides, macrolides, or aminoglycosides), or a low-risk antibiotic (tetracyclines). Multivariate models were constructed to examine patient and ward factors that increased the risk for acquiring CDI.
A total of 255 patients developed CDI during the 46-month study period. The individual-level risk factors were found to be age, readmission, antibiotic usage, and use of a feeding tube. Admission to the oncology service led to an increased risk for CDI compared to medicine and surgery services. Receipt of a high-risk antibiotic was associated with the greatest relative risk for developing CDI (2.73, 95% confidence interval [CI], 2.05-3.63). At the ward level, each 10% increase in antibiotic use was associated with an increased incidence of CDI of 2.1 per 10,000 patient-days (P < .001). The other ward-level factors were not significantly associated with CDI incidence. Moreover, the multivariate model showed each 10% increase in antibiotic exposure was associated with a 1.34-fold increase in CDI infection risk (95% CI, 1.16-1.57).
The most interesting result of this study was that each 10% increase in antibiotic prescribing at the ward level was associated with a 34% increase in CDI incidence. This means that antibiotics increase the risk for acquiring CDI for every patient on a particular hospital ward, including those who never receive them. The authors argued that the high prevalence of antibiotic use led to an increase in the number of patients colonized with and shedding C. difficile, which subsequently increased environmental contamination. This hypothesis is supported by previous research that showed antibiotic exposure is the main risk factor for C. difficile, colonization. The finding that hand hygiene compliance, which ranged from 84.6% to 92.9%, was not associated with CDI incidence is surprising. Perhaps the hand hygiene rates were too close to produce meaningful difference or maybe a higher rate (e.g. 95%) would have been significant. Future studies are needed to clarify this issue.
The main clinical implication of the study by Brown and colleagues is that aggregate ward-level antibiotic prescribing should be monitored by infection control and antibiotic stewardship personnel. This strategy might be useful for institutions with CDI rates higher than national and regional averages. However, since the study was conducted at a single institution and might have been influenced by confounding factors, additional multi-center studies are warranted to confirm the findings and to justify the additional expenses that would occur in performing ward-level antibiotic monitoring.
Another limitation to the study is the way the antibiotic risk index was determined, which left out several agents that are frequently used on the wards, such as linezolid, daptomycin, and intravenous vancomycin. These would likely have been put in the medium-risk category. Brown and colleagues’ findings add further evidence of the benefits of antibiotic stewardship programs and support expanding their role in health care systems.
Finally, questions also remain about the role of asymptomatic C. difficile, colonization on admission, and how much of a risk it is for the development of CDI. At present, active surveillance testing is not recommended to control CDI. This is another topic that needs to be further elucidated through larger, multi-center studies.