Hospital patients who used the same beds as previous patients administered antibiotics had a 22% relative increase in risk for contracting Clostridium difficile infection (CDI), according to a new study.
Risk of contracting CDI for these "second-user" patients was still less than 1%, but Daniel E. Freedberg, MD, MS, of Columbia University Medical Center in New York City, and colleagues found a statistically significant increase in CDI rates relative to cases in which the prior bed occupant did not receive antibiotics.
"These data support the hypothesis that antibiotics given to one patient may alter the local microenvironment to influence a different patient's risk for CDI," Freedberg and colleagues wrote online in JAMA Internal Medicine. "Our results show that antibiotics can potentially cause harm to patients who do not themselves receive the antibiotics, and thus emphasize the value of antibiotic stewardship."
Among the 100,615 pairs of patients under study, 576 (0.57%) of the secondary patients developed CDI within 2 to 14 days after being assigned to the hospital bed (median 6.4 days). This relationship also showed statistical significance after adjusting for factors known to influence CDI risk, including receipt of antibiotics by the second patient (adjusted hazard ratio [aHR] 1.22; 95% CI, 1.02-1.45).
When first-round patients received antibiotics, new patients' cumulative risk of contracting CDI infection was 0.72% -- versus 0.43% when first-round patients did not receive antibiotics.
When the researchers excluded the 1,497 patient pairs featuring first-round patients who developed CDI, the relationship remained significant (aHR, 1.20; 95% CI, 1.01-1.43).
In addition, sensitivity analyses revealed a significant CDI-antibiotic connection even when the researchers excluded first-user patients who developed CDI themselves.
The researchers indicated that no factors besides prior-occupant antibiotic use were associated with an increased risk for CDI.
Freedberg's team monitored patients over 5 years across four hospitals in New York City, via a retrospective cohort study of adult patients. Excluded were patients who had recent CDI, who developed CDI within 48 hours of admission, and who had inadequate follow-up time -- or if the first-user patient was in the bed for less than 24 hours.
The researchers defined incident CDI as a positive result from a stool polymerase chain reaction test for the C. difficile toxin B gene, followed by treatment for CDI.
"These data imply that patient-to-patient transmission of C. difficile or other bacteria that mediate susceptibility to CDI takes place in the non-outbreak setting and in the face of a multifaceted effort seeking to prevent healthcare-associated CDI," the authors concluded.
CDI is the most common cause of diarrhea in the hospital and responsible for about 27,000 deaths annually in the U.S., according to a JAMA news release. Exposure is common in hospitals because spores can persist there for months.
Regarding consideration for future research, the authors wrote: "Antibiotics may affect the gastrointestinal microbiome more globally to decrease bacterial species that are protective against C. difficile or to increase bacterial species that facilitate C. difficile. Subsequent patient-to-patient transmission of these bacterial species may then drive risk for CDI in future patients. The specific mechanisms underlying the herd effects of antibiotics may be a fruitful area for future research."
Study limitations included the observational nature of the study, that it was confined to a single health care system, conducted in a non-outbreak setting, and had a small observed effect size.
Regarding the effect size, Freedberg and colleagues wrote: "Although this translates into a modest absolute risk associated with antibiotics in the prior bed occupant, it remains important because use of antibiotics in the hospital is so common."
They noted previous research indicating that antibiotic use is a known risk factor for CDI -- not only for the individual patient, but also "at the level of the hospital ward, the level of the institution, and the regional level."