Introduction Several studies have highlighted the risks in patients with a flare of colitis and concomitant Clostridium difficile infection: including increased mortality and increased rates of colectomy in patients with ulcerative colitis (UC). Infection control policies, including antibiotic prescribing, have had an impact on rates of Clostridium difficile associated diarrhoea (CDAD) in the general in-patient population but rates of CDAD in inflammatory bowel disease (IBD) have been reported to be as high as 20%. This study was conducted to investigate the current rates of CDAD in IBD in our Trust.

Methods For January 2007–2009, case notes for patients with a joint diagnosis of CDAD and IBD (UC and all Crohn's disease (CD)), and hospital audit database for patients with IBD alone, were examined for length of stay, outcomes and medication. Patients were identified by the in-hospital reporting system (Infection Control), as well as discharge diagnoses (Coding). CDAD was diagnosed by positive A/B toxin in standard stool sample testing.

Results A total of 217 patients with IBD (86 UC) were identified for analysis. 20 patients (12 UC) had CDAD on a pre-existing diagnosis of IBD. This was associated with a dramatically increased average length of stay (38 vs 5.6 days, p<0.001), but not with rates of colectomy (1/12 with UC plus CDAD vs 21/74 with UC alone, p=0.28). Patients with CDAD were older (55.5±20.4 vs 26.8±16.8 years, p<0.05), and 12/20 had received antibiotics for either IBD-related or other pathology, but only 6/20 were on concomitant PPI therapy. Mortality in the CDAD plus UC group was high (2/12), but the deaths occurred in one patient with rectal cancer and another with concomitant chest sepsis. All patients with CDAD received initial therapy with oral or IV metronidazole (in line with Trust practice), and changed to oral vancomycin if diarrhoea had not settled after 10 days (after discussion with the Microbiologist) or if clinical deterioration before this. The use of vancomycin was associated with a longer mean in-patient stay (49.6 vs 28.3 days, p<0.001).

Conclusion Our rates for CDAD compare favourably with other case series, but do indicate an increased length of in-patient stay for such patients. Neither increased mortality directly attributable to CDAD, nor increased colectomy rates were detected and may reflect early diagnosis and treatment of CDAD in this “at-risk” patient group. Older patients and those receiving antibiotic therapy for other diagnoses remain the highest risk groups identified.