Background: In randomized controlled trials and meta-analyses, as primary prevention, probiotics reduced C. difficile infection (CDI) among hospital patients. We performed a before-after evaluation of a Lactobacillus casei, L. acidophilus, and L. rhamnosus (BioK+) probiotic.
Methods: We administered probiotic to antibiotic recipients in a 694-bed teaching hospital. After we excluded pediatric, neonatal, and oncology units there were 575 eligible beds. We evaluated 12 month baseline (Oct. 1, 2012-Sep. 30, 2013) and intervention (Nov. 1, 2013-Oct. 31, 2014) periods; October 2013 was a 1-month run-in. Our protocol was to administer probiotic < 12 hours after initial antibiotic receipt, to continue daily for five days post final dose. We excluded patients on preoperative antibiotics or npo, or with pancreatitis, leukopenia, or post-transplant. Over several years pre-baseline, the hospital began PCR testing, antibiotic controls, enhanced cleaning, and hand hygiene campaigns; no interventions were added during the project. CDI was determined by a recommended infection surveillance definition. We compared baseline and intervention periods by segmented regression models.
Results: There were 283 CDI episodes among 360,016 patient days during the baseline and intervention periods, incidence = 7.9 per 10,000 patient-days. The incidence was similar during baseline and intervention periods (7.7 vs 8.0 per 10,000 patient-days; P=0.70). By segmented regression analysis, compared to the level baseline slope, there was a significant decrease in CDI during the intervention, P=0.05, Figure. The incidence rate ratio (IRR) in the final 6 months of intervention was 24% lower than baseline: IRR=0.76; 95% CI (0.54 to 1.1). Compared to the first 6 months of the intervention, the CDI rate in the final 6 months decreased 43%: IRR=0.57; 95% CI (0.40 to 0.80). Testing intensity was stable in baseline vs intervention periods, 19% vs 20% of stools tested were C. difficile PCR positive.
Conclusion: Probiotic administration targeted to antibiotic recipients was associated with a trend toward a reduced CDI incidence. The decline was evident 6 months post-implementation, possibly reflecting delayed impact due to combination of antibiotic and environmental risk factors.
J. Levato, None
R. Sweis, None
S. Veleeparambil, None
S. Johnson, BioK+: Scientific Advisor , Consulting fee
R. A. Weinstein, None
See more of: Poster Abstract Session