DOP090 Risk factors for bacteraemia in hospitalised patients with inflammatory bowel disease

Abstract

Patients with inflammatory bowel diseases (IBD) may be at increased risk for invasive bacterial infections due to altered bowel integrity, surgical complications and medications’ side effects. We aimed to identify risk factors that were associated with bacteraemia in hospitalised patients with IBD. We evaluated hospitalisation data of IBD patients, over 16 years old in both medical and surgical departments at a large tertiary hospital over an 8 years duration. Patients over 80 years of age and those with significant baseline comorbidities (Charlson Comorbidity Index > than 1) were excluded. Of the remaining, IBD patients with at least one positive blood culture were identified and compared with those with negative cultures. Mortality data were obtained from the National Israeli Population Registry. Logistic regression was used to evaluate associations between bacteraemia and medications used before admission. For each medication class, we calculated the odds ratios (OR) using uni-variable models and the adjusted odds ratio (aOR) in multi-variable analysis, adjusted for age, gender and comorbidity differences (using Charlson Index and Norton Scale). A total of 5467 hospitalised IBD patients' records were evaluated, of which, 129 (2%) patients developed bacteraemia (80 Crohn's disease, 37 ulcerative colitis and 12 IBD unclassified). Patients with bacteraemia had significantly longer hospitalisations (mean Length of stay (LOS) 23.4 ± 34 vs. 6.4 ± 16 days; p < 0.001), were older (mean age 45.5 ± 17 vs. 40.2 ± 15 years old, p < 0.0001). The most common pathogen was Escherichia coli (27 of 129 patients with a 15% 30-days mortality) followed by coagulase-negative Staphylococci and Staphylococcus Aureus. Use of opioids (aOR 5.2, 95% CI 2.7–9.2 p < 0.0001), anti-propulsive (aOR 6.0; 95% CI 2.4–13, p < 0.0001), anti-cholinergics (aOR=5.0, CI 1.3–13.3, p = 0.003), vitamin K antagonists (aOR 3.5, 95% CI 1.5–7, p = 0.001) and parenteral iron (OR 5.8, 95% CI 1.7–15, p = 0.001) were associated individually with significant increased odds for bacteraemia. The use of anti-TNF, purine-analogues, steroids and aminosalicylates were not associated with bacteraemia. The relative risk of bacteraemia was greatest in IBD patients older than 65 years of age (RR 2.22, 95% CI 1.4–3.4, p = 0.0002, relative to those 65 years or younger). Based on a large-scale retrospective registry, bacteraemia in hospitalised patients with IBD is relatively rare. Anti-inflammatory medications used to treat IBD are not associated with increased risk of bacteraemia, whereas opioids, antipropulsives, anticholinergics, vitamin K antagonists and parenteral iron are associated with increased risk for bacteraemia.