Case-control study of factors that trigger inflammatory bowel disease flares.

Abstract

To explore the association between inflammatory bowel diseases (IBD) flares and potential triggers. Patients evaluated for an acute flare of IBD by a gastroenterologist at the Dallas VA Medical Center were invited to participate, as were a control group of patients with IBD in remission. Patients were systematically queried about nonsteroidal anti-inflammatory drug use, antibiotic use, stressful life events, cigarette smoking, medication adherence, infections, and travel in the preceding 3 mo. Disease activity scores were calculated for each patient at the time of enrollment and each patient's chart was reviewed. Multivariate regression analysis was performed. A total of 134 patients with IBD (63 with Crohn's disease, 70 with ulcerative colitis, and 1 with indeterminate colitis) were enrolled; 66 patients had flares of their IBD and 68 were controls with IBD in remission (for Crohn's patients, average Crohn's disease activity index was 350 for flares vs 69 in the controls; for UC patients, Mayo score was 7.6 for flares vs 1 for controls in those with full Mayo available and 5.4p for flares vs 0.1p for controls in those with partial Mayo score). Only medication non-adherence was significantly more frequent in the flare group than in the control group (48.5% vs 29.4%, P = 0.03) and remained significant on multivariate analysis (OR = 2.86, 95%CI: 1.33-6.18). On multivariate regression analysis, immunomodulator use was found to be associated with significantly lower rates of flare (OR = 0.40, 95%CI: 0.19-0.86). In a study of potential triggers for IBD flares, medication non-adherence was significantly associated with flares. These findings are incentive to improve medication adherence.