P529 Longer duration of vancomycin prevents recurrence of C. difficile in patients with inflammatory bowel disease

Abstract

Background: Compared with the general population, IBD patients acquire CDIs at higher rates, have greater mortality, and experience longer hospitalisations. The current treatment recommendation for severe CDI is a 14‐day course of oral vancomycin, but the supporting studies have specifically excluded IBD patients. IBD patients are 33% more likely to have a recurrence of CDI than the general population, suggesting that different protocols are necessary. We assessed the association between length of oral vancomycin therapy and rates of CDI recurrence and reinfection in IBD patients. Methods: A single‐centre retrospective review was performed on patients with IBD (CD or UC) and an index positive C. difficile toxin assay by PCR between 2010 and 2016. Patients treated with oral metronidazole or combined with vancomycin were excluded. Data gathered included patient demographics, surgical history, medications, incident CDI therapy length, and CDI assay results. Distinction between CDI and IBD relapse was defined as resolution of symptoms without change in IBD therapy. Long duration (LD) and short duration (SD) therapies were a priori defined as oral vancomycin treatment of 21‐42 days and 10‐14 days, respectively. Primary outcomes were recurrence or reinfection of CDI, defined as a positive C. difficile result within 8 weeks of the end of antibiotic therapy; and a positive C. difficile result after 8 weeks of the end of antibiotic therapy, respectively. The Fisher's exact test was used to test for significance and multivariate logistic regression models were constructed to control for other covariables. Results: We identified 135 IBD patients (58 UC, 77 CD; median age 29.3, IQR 21.4‐50.3) with index CDI, and median follow‐up 685 days (IQR 321‐1238). SD vancomycin had a 11.69% incidence of CDI recurrence, compared with 1.72% in the LD vancomycin group (OR = 7.544, p = 0.043). CDI reinfection rates and time to reinfection were not significantly different (SD 17% vs. LD 19%, p = NS). Time to recurrence or reinfection: SD median 126 days, IQR 36.3‐253.8 vs. LD median 274 days, IQR 151‐460 p = NS). Multivariate logistic regression model controlling for patient characteristics (sex, age) and clinical history (UC vs. CD, concomitant drug use at incident CDI, surgical history of partial colectomy) found that patients treated with LD vancomycin had lower odds for recurrence than those in the SD group (OR = 0.0604, p = 0.0299). Conclusions: In IBD patients with CDI, oral vancomycin for 21‐42 days is associated with significantly less CDI recurrence and numerically longer time to reinfection than shorter duration therapy. These results will guide clinical decisions and the development of a prospective trial.