Early Colonoscopy for Diverticular Bleeding Does Not Reduce Risk of Postdischarge Recurrent Bleeding: A Propensity Score Matching Analysis

BACKGROUND & AIMS: Colonoscopy within 24 hours (early colonoscopy) is recommended for patients with colonic diverticular bleeding, but it is unclear if this strategy improves postdischarge outcomes. We aimed to determine whether early colonoscopy is associated with decreased risk of rebleeding and hospital re-admission within 30 days. METHODS: We performed a retrospective cohort study using Marketscan (Truven Health Analytics, Inc, Ann Arbor, MI), a nationwide insurance claims database. From January 2004 through September 2015, patients with a primary diagnosis of diverticular bleeding who underwent inpatient colonoscopy were included. We used propensity score matching to account for differences between recipients of early vs delayed colonoscopy. Multivariable logistic regression was performed to determine the association between early colonoscopy and rebleeding or hospital re-admission within 30 days of discharge. RESULTS: In total, 20,010 patients underwent colonoscopy for diverticular bleeding; 11,690 underwent early colonoscopy. After propensity matching, 8320 pairs of patients were analyzed. In the matched analysis, higher proportions of patients who received early colonoscopy underwent additional colonoscopies (73%), compared with patients who did not receive early colonoscopy (4%) (P < .0001), but lower proportions received endoscopic interventions (3% vs 8%; P < .0001). On multivariable analysis, early colonoscopy (odds ratio [OR], 1.34; 95% CI, 1.08-1.66; P = .007), transfusion requirement (OR, 2.31; 95% CI, 1.88-2.83; P < .0001), and baseline chronic kidney disease (OR, 2.13; 95% CI, 1.49-3.04; P < .0001) were associated with increased risk of rebleeding within 30 days. Early colonoscopy (OR, 1.18; 95% CI, 1.02-1.36; P = .03), endoscopic intervention (OR, 1.37; 95% CI, 1.03-1.81; P = .03), transfusion requirement (OR, 2.17; 95% CI, 1.88-2.51; P < .0001), coronary artery disease (OR, 1.27; 95% CI, 1.06-1.51; P = .009), and chronic kidney disease (OR, 1.98; 95% CI, 1.54-2.54; P < .0001) were associated with increased re-admission to the hospital within 30 days. CONCLUSIONS: In a propensity-matched analysis, we associated early colonoscopy with increased risk of rebleeding events and hospital re-admissions. However, these observations might be due to confounding factors.