Randomized trial of normal saline solution injection versus bipolar electrocoagulation for treatment of patients with high-risk bleeding ulcers: Is local tamponade enough?

Abstract

Background: Endoscopic injection of normal saline solution is reportedly an effective treatment for bleeding ulcers. If it is as effective as standard therapy, low cost, wide availability, and lack of injury would make saline solution injection an attractive option. Methods: Patients with clinical evidence of major bleeding from an ulcer with active bleeding or a nonbleeding visible vessel were randomly assigned to injection with normal saline solution (1-2 mL boluses; mean volume 30 mL) or bipolar electrocoagulation (20 W, 10-sec applications; mean time 100 sec). Patients, those caring for patients, and those collecting data were blinded to therapy. Results: Further bleeding occurred in 14 (29%) of 48 patients in the saline solution group versus 6 (12%) of 52 patients in the bipolar group (95% CI [2%, 33%]; p = 0.04). Significantly more units of blood were transfused in the saline solution group (median 2 units vs. 0 units; p = 0.01). Hospital days (median 4 vs. 3) and mortality (6% vs. 2%) were not significantly different in the 2 groups. Independent risk factors for further bleeding were saline solution injection (p = 0.02), units transfused before therapy (p = 0.02), and ulcer size (p = 0.03). Conclusion: Local tamponade with saline solution injection is less effective than bipolar electrocoagulation for the treatment of bleeding ulcers. (Gastrointest Endosc 2002;55:6-10.)