BACKGROUND/OBJECTIVES: During the first quarter of 2005, infection control (IC) personnel at Shands Hospital documented an increase in the occurrence of Clostridium difficile-associated disease (CDAD) in patients in the Bone Marrow Transplantation Unit (BMTU). Because of the medical complexity of these patients, medical personnel were unable to implicate specific CDAD risk factors. IC investigated to identify associated intrinsic risk factors and implement appropriate control measures.

METHODS: We conducted a retrospective case-control study. CDAD was confirmed by positive tests for toxins A and B. A case was defined as any BMTU patient who acquired CDAD during June 2004 through March 2005. Cases were ascertained through review of medical charts and microbiology records. Data were recorded in a standardized questionnaire and included demography, prior hospitalization, medical device usage, duration of hospitalization, service, exposure to medications (e.g., antimicrobials, laxatives, proton pump inhibitors), underlying disease and co-morbidities, type of cancer, gastrointestinal and invasive procedures, hyperalimentation, and outcome (death, survived, surgery). Observational studies were conducted to assess practices and procedures. Odds ratios (OR) and 95% confidence intervals (CI) were calculated. Fisher’s exact test was used, where appropriate.

RESULTS: CDAD incidence densities in the epidemic and pre-epidemic periods were 5.8 /1000 patient-days and 3.8/1000 patient-days, respectively. Nineteen patients met the case definition; 35 controls were randomly selected from patients who were admitted to the unit during the study period but were negative for CDAD. Cases comprised 17 adults and 2 children (five controls were children). Case characteristics: median age 51 (range 4-83) years; 13 (68%) male. Although cases were more likely to be >65 yrs of age, the difference was not statistically significant (OR: 3.1, CI: 0.4–30), Cases and controls were similar for race, sex, receipt of medical devices, exposure to antimicrobials, tube feedings, proton pump inhibitors or laxatives, chemotherapy, history of recurrent CDAD, and occurrence of diarrhea, vomiting, nausea, abdominal pain or distention, acute abdomen or rectal bleeding. No significant differences were noted for co-morbidities or the presence of underlying heart, lung, liver or renal diseases, or type of cancer. Cases were more likely to have undergone gastrointestinal surgery or procedures (p = 0.06).

CONCLUSIONS: Our data confirmed that there were no underlying intrinsic patient risk factors for acquiring CDAD in the BMTU. We hypothesize that the mode of transmission was most likely environmental or via healthcare workers hands. Epidemiologic studies like this may assist infection prevention and control programs by identifying areas for targeting interventions and control measures.

L. K. Archibald, Regeneration Technologies Inc. Employee.

Regenerations Technologies, Inc.