Epidemiology of Clostridium difficile infections in Japan

Background:

Population characteristics and antimicrobial prescribing practices suggest the hospitalized population in Japan should be at high risk for Clostridium difficile infection (CDI); however, few data exist on the epidemiology of CDI in Japan.

Methods: A prospective cohort study was performed on selected wards at 12 hospitals in Japan. Patients with clinically significant diarrhea were enrolled and stool tested for C. difficile at the hospital laboratories (toxin EIA, GDH, and/or culture). In addition, stool was shipped to a central lab and tested by nucleic acid amplification test (NAAT), stool culture, and toxigenic culture. CDI was defined as EIA+, NAAT+, or toxigenic culture+. CDI cases were stratified into hospital onset (HO), community-onset, healthcare facility associated (CO-HCFA), and community associated (CA). Data on CDI risk factors, management, and outcomes were collected and analyzed.

Results:

653 patients with diarrhea were enrolled, of whom 187 met the study definition of CDI (29%) (rate= 7.9 per 10,000 pt-days) and 171 were diagnosed clinically with CDI at the hospital. 83% of CDI cases were HO; 9% each were CO-HCFA or CA. 171 (26%) patients were given a clinical diagnosis of CDI at the hospital (rate=7.3 per 10,000 pt-days). 47 (27%) of clinical CDI diagnoses did not have a positive lab test at the study hospital; of these, 7 (15%) had CDI confirmed by the central lab. 650 patients had stool tested at the central lab: 127 (20%) were NAAT+, 209 (32%) were C. difficile culture+, and 157 (24%) were toxigenic culture+. 112 (60%) CDIs were concordant for study CDI and clinical CDI (Kappa=0.49). Admission from a long-term care facility, older age, low albumin, low creatinine, probiotic use in the previous 2 months, and longer time from hospitalization to diarrhea onset were risk factors for CDI (p<0.05 for all).

Conclusion: CDI is common among hospitalized patients with clinically significant diarrhea in Japan.