ASSOCIATION BETWEEN ANTIBIOTIC USE AND INFLAMMATORY BOWEL DISEASE OUTCOMES

Abstract

Abstract OBJECTIVE Benefit of antibiotic use in Inflammatory Bowel Disease (IBD) is unclear and has been a subject of research. This study investigated the association between antibiotic use and treatment outcomes among IBD patients using real-world data. METHODS Data from Study of a Prospective Adult Research Cohort with IBD (SPARC IBD) of Crohn’s & Colitis Foundation’s IBD Plexus program were used. Patients diagnosed with Crohn’s disease (CD) or ulcerative colitis (UC) since 2000 were included. The use of antibiotics (ciprofloxacin, metronidazole, rifaximin and vancomycin) for ≥7 days after enrollment was identified from electronic health records and case report forms. Disease activity scores were assessed at the time of enrollment (baseline) and during follow-up using 6-point Mayo score (6-pt MS) and short Clinical Disease Activity Index (sCDAI) for UC and CD, respectively. Patients’ demographic and clinical characteristics were also collected. Remission was defined as 6-pt MS of 0-1 for UC and sCDAI<150 for CD. The association between antibiotic use and outcome during follow-up was assessed in a subset of patients without remission at baseline using a multivariable Cox proportional hazards model. The model included the following covariates for adjustment: age, gender, race, family history of IBD, IBD surgery, disease activity at enrollment and other IBD drug use during follow-up. RESULTS A total of 2,959 IBD patients (UC=1,060; CD=1,899) were included. Mean age at diagnosis was 30±14 years (UC=32±15; CD=28±14). Forty-four percent of the patients were male (UC=49%; CD=41%) and 79% were White (UC=82%; CD=78%). Antibiotic use after enrollment was observed in 29% of the patients (UC=26%; CD=30%). About 60% of UC or CD patients were in remission at baseline. The antibiotic use group had significantly higher baseline disease activity (UC 6-pt MS: 1.7±1.8; CD sCDAI: 167±96) than patients without antibiotic use (UC 6-pt MS: 1.3±1.7; CD sCDAI: 137±89). In the subset of patients with mild, moderate or severe baseline IBD (6-pt MS>1 for UC; sCDAI≥150 for CD), the baseline disease activity scores were largely comparable between patients who used antibiotic and who did not (UC 6-pt MS: 3.6±1.3 vs 3.3±1.3, p-value=0.063; CD sCDAI: 248±83 vs 238±83, p-value=0.169), while patients who used antibiotics were less likely to experience remission during follow-up compared to those who did not use antibiotics (UC adjusted remission hazard ratio [aRHR] 0.39, 95% confidence interval [CI] 0.26-0.60; CD aRHR 0.60, 95% CI 0.46-0.79). CONCLUSION In this study, antibiotic use in patients with IBD was negatively associated with remission during follow-up. The effect of antibiotics on gut microbiome and optimal use of antibiotics in IBD patients need to be further explored.