Analyses of fluoroquinolones and Clostridium difficile-associated diarrhoea in tuberculosis patients

SETTING: Systematic studies of fluoroquinolones (FQs) and Clostridium difficile-associated diarrhoea (CDAD) are scarce among tuberculosis (TB) patients, in whom fluoroquinolones (FQs) are increasingly used. OBJECTIVE: To evaluate the relationship between FQs and CDAD among TB patients. DESIGN: Retrospective cohort and nested case-control analyses were conducted among 3319 hospital patients on anti-tuberculosis treatment from 1999 to 2005. Each case of CDAD was matched by three sex- and age-matched controls randomly selected from the rest of the cohort. Not every case was confirmed by C. difficile cytotoxins. RESULTS: Among 38 cases studied, the incidence of CDAD, which was 28.2 (95%CI 20.3-38.3) per 100 000 patient-days overall, increased from 12.9 (95%CI 5.8-25.3) for patients aged <60 years to 26.6 (95%CI 15.5-42.8) for those aged between 60 and 79 years, and 66.9 (95%CI 39.8-106.1) for those aged >79 years. Univariate analysis showed a significant association between CDAD and age, FQs, non-FQ antibiotics, serum albumin level, duration of hospital stay and nasogastric feeding. Only duration of hospital stay and nasogastric feeding remained significant on multivariable analysis. CONCLUSION: The risk of CDAD due to FQs among TB patients is probably modest after controlling for sex, age, non-FQ antibiotics, serum albumin level, duration of hospital stay and nasogastric feeding.