136 Community-onset Clostridium difficile Infection in a Veterans Affairs Medical Center

Friday, March 19, 2010
Grand Hall (Hyatt Regency Atlanta)
Jessica C. Becker, B.A. , Cleveland VA Medical Center, Cleveland, OH
Elizabeth C. Eckstein , Louis Stokes Cleveland VA Medical Center, Cleveland, OH
Dubert Guerrero, MD , University Hospitals of Cleveland/Case Med. Ctr., Cleveland, OH
Michelle Nerandzic, BS , Louis Stokes Cleveland VA Medical Center, Cleveland, OH
Curtis Donskey , Louis Stokes Cleveland VA Medical Center, Cleveland, OH

Background: In many cases of Clostridium difficile infection (CDI), the onset of symptoms occurs in the outpatient setting (i.e., community-onset CDI). Recent reports suggest that a significant proportion of patients with community-onset CDI have no recent inpatient healthcare exposures and no recent antibiotic exposure.

Objective: To determine the incidence of community-onset CDI in a Veterans Affairs Medical Center and to test the hypothesis that many of these cases have no recent inpatient healthcare or antibiotic exposure.

Methods: We performed a 2-year prospective study of all patients with CDI. Cases were classified as healthcare facility-onset, healthcare facility associated (HO-HCFA) (i.e., diagnosed more than 48 hours after admission) or community-onset (diagnosed in the community or within 48 hours after admission). Community-onset cases were further classified based on time from last discharge: within 4 weeks, community-onset, healthcare facility-associated (CO-HCFA); 4-12 weeks, indeterminate exposure; >12 weeks, community-associated. For a subset of community-associated cases, culture of stool samples for C. difficile and molecular typing was performed.

Results:   Of 355 CDI patients, 267 (75%) had HO-HCFA cases and 88 (25%) had community-onset cases. Of the community-onset CDI cases, 57 (65%) had CO-HCFA cases, 17 (19%) had indeterminate cases, and 14 (16%) had community-associated cases. Of 14 community-associated cases, 10 (71%) received antibiotics within 8 weeks before diagnosis and all except one received an antimicrobial with anti-anaerobic activity. Indications included dental (30%), skin and soft tissue infections (20%), urinary tract infections (20%), and upper respiratory infections (10%). The mean age of patients with community-associated CDI was 71 (range 51-89) and 8 (57%) had at least 2 chronic medical conditions. Half of the community-associated CDI cases required admission to the hospital and 4 (28%) were considered severe, with 1 (7%) death due to CDI and another death during treatment of CDI. Five stool samples from patients with community-associated CDI were analyzed and all had positive cultures for toxigenic C. difficile; only 1 (20%) was a PCR-ribotype 027 strain.

Conclusions: In a Veterans Affairs Medical Center, about a quarter of CDI cases had their onset in the community. However, only 16% of community-onset cases had no inpatient healthcare facility exposure in the prior 12 weeks and most of these patients had recent antibiotic exposure.