Urgent colonoscopy for evaluation and management of acute lower gastrointestinal hemorrhage: a randomized controlled trial.

We hypothesized that early intervention in patients with lower gastrointestinal bleeding (LGIB) would improve outcomes and therefore conducted a prospective randomized study comparing urgent colonoscopy to standard care. Consecutive patients presenting with LGIB without upper or anorectal bleeding sources were randomized to urgent purge preparation followed immediately by colonoscopy or a standard care algorithm based on angiographic intervention and expectant colonoscopy. A total of 50 patients were randomized to each group. A definite source of bleeding was found more often in urgent colonoscopy patients (diverticula, 13; angioectasia, 4; colitis, 4) than in the standard care group (diverticula, 8; colitis, 3) (the odds ratio for the difference among the groups was 2.6; 95% CI 1.1-6.2). In the urgent colonoscopy group, 17 patients received endoscopic therapy; in the standard care group, 10 patients had angiographic hemostasis. There was no difference in outcomes among the two groups-including: mortality 2%versus 4%, hospital stay 5.8 versus 6.6 days, ICU stay 1.8 versus 2.4 days, transfusion requirements 4.2 versus 5 units, early rebleeding 22%versus 30%, surgery 14%versus 12%, or late rebleeding 16%versus 14% (mean follow-up of 62 and 58 months). Although urgent colonoscopy identified a definite source of LGIB more often than a standard care algorithm based on angiography and expectant colonoscopy, the approaches are not significantly different with regard to important outcomes. Thus, decisions concerning care for patients with acute LGIB should be based on individual experience and local expertise.