S654 Early Colonoscopy Is Associated With Lower Mortality in Stable Diverticular Bleeding: A National Database Study

Abstract

Introduction: Major GI societies recommend early colonoscopy (within 24 to 48 hours) for acute, high risk, lower GI bleeds, of which diverticular bleeds comprise about 20 to 40 percent. There is paucity of evidence regarding the timing of colonoscopy for stable presumed diverticular bleeds. Majority of these patients are managed conservatively with outpatient follow up colonoscopy at a later date. Methods: This was a retrospective case-control study. We queried a large national database (TriNetX, LLC.) which aggregates data from 59 healthcare organizations across the United States comprising more than 70 million patients using ICD-10 and CPT codes. All patients were adults ≥ 18 years with known diverticulosis experiencing their first ever lower GI bleed and deemed to be clinically sTable: Cases had a colonoscopy within 7 days of the lower GI bleed, whereas controls had a colonoscopy after 7 days following the lower GI bleed. Patients requiring ICU admission were excluded from both groups. Additionally, patients with upper GI bleeds or presence of any condition that would predispose to a GI bleed (esophageal varices, peptic ulcers, angiodysplasias, GI malignancies, platelet disorders, complement factor deficiencies, etc) were excluded. We also excluded all patients with acute diverticulitis. The cohorts were propensity matched for demographics and comorbidities as shown in Table. Cases and controls were compared for outcomes of interest and odds ratio (OR) and 95% confidence interval (CI) were calculated. Results: There were 13,890 cases of stable diverticular bleed who had a colonoscopy within 7 days and 103,718 controls with a stable diverticular bleed who had a colonoscopy after 7 days. After propensity matching, there were 12,502 patients in each group (Table). Patients in the late colonoscopy group ( > 7 days) had a significantly higher risk of 30-day mortality (HR = 1.80, 95% CI = 1.33, 2.41). On the other hand, the controls had a lower risk of 30-day re-admissions (OR = 0.88, 95% CI = 0.82, 0.95). There were no significant differences between cases and controls with respect to need for blood transfusion (OR = 1.0, 95% CI = 0.72, 1.40). Conclusion: Early colonoscopy within 7 days seems to confer a survival benefit in patients with stable diverticular bleeds when compared to performing colonoscopy after 7 days. Large prospective randomized controlled trials would be necessary to confirm these findings. Table 1. - Baseline characteristics of patients with stable diverticular bleed who underwent colonoscopy Unmatched cohorts Propensity matched cohorts Colonoscopy < 7 days (N = 13,890) Colonoscopy > 7 days (N = 103,718) Colonoscopy < 7 days (N = 12,502) Colonoscopy > 7 days (N = 12,502) Age at index 63.4 +/- 15.2 years 64.2 +/- 15.9 years 63.4 +/- 15.2 years 63.4 +/- 15.3 years White race 8,752 (70%) 67,415 (68.7%) 8,752 (70%) 8,889 (71.1%) Male 5,717 (45.7%) 43,326 (44.1%) 5,717 (45.7%) 5,648 (45.2%) Nicotine dependence 1,536 (12.3%) 11,435 (11.7%) 1,536 (12.3%) 1,414 (11.3%) Alcohol abuse 496 (4%) 3,196 (3.3%) 496 (4%) 391 (3.1%) Hypertension 6,547 (52.4%) 50,258 (51.2%) 6,547 (52.4%) 6,611 (52.9%) Chronic kidney disease 1,351 (10.8%) 11,523 (11.7%) 1,351 (10.8%) 1,272 (10.2%) Type 2 diabetes 2,579 (20.6%) 20,162 (20.5%) 2,579 (20.6%) 2,532 (20.5%) BMI ≥ 30 4,424 (35.4%) 29,734 (30.3% 4,424 (35.4%) 4,090 (32.7%) Atherosclerosis 633 (5.1%) 5,812 (5.9%) 633 (5.1%) 544 (4.4%) NSAID use 3,921 (31.4%) 29,896 (30.5%) 3,921 (31.4%) 4,017 (32.1%) Anticoagulant use 2,810 (22.5%) 21,327 (21.7%) 2,810 (22.5%) 2,792 (22.3%) Anti-platelet agent use 3,532 (28.3%) 24,942 (25.4%) 3,532 (28.3%) 3,459 (27.7%)