Endoscopic thermocoagulation hemostasis for acute non-varicose upper gastrointestinal hemorrhage: a randomized controlled study.

Abstract

This study evaluated the application of the bipolar electrocoagulation catheter via a peripheral-to-central (3 + 1) technique, relative to directly pressing only at the center of the bleeding site (direct-stroke), to effect endoscopic hemostasis of acute non-varicose gastrointestinal bleeding (ANVUGIB). Patients (n = 148) with endoscopically diagnosed ANVUGIB were randomly apportioned to receive treatment by 3 + 1 (n = 78) or direct-stroke (n = 70) application of the bipolar electrocoagulation catheter. The 3 + 1 strategy required pressing at3 narrowly restricted sites equidistant peripheral and center to the site of bleeding. The rates of initial success, hemostasis time, and number of compressions were compared according to intention-to-treat (ITT) or per protocol (PP). The ITT (PP) rate of initial hemostatic success in patients receiving the 3 + 1 catheter was 91.02% (95.9%); and for the direct-stroke group was 71.42% (76.9%). For Forrest IIa lesions specifically, these rates were respectively 91.70% (97.1%) and 63.9% (67.6%). The ITT (PP) hemostasis times of the 3 + 1 and direct-stroke groups were 10.96 ± 3.28 (10.65 ± 2.90) and 14.27 ± 6.58 (14.12 ± 6.67) min; and the number of compressions numbered 5.73 ± 1.98 (5.42 ± 1.46) and 6.47 ± 2.82 (6.16 ± 2.47). During thermocoagulation treatment of ANVUGIB via bipolar electrocoagulation catheter, the 3 + 1 strategy showed a significantly higher rate of successful initial hemostasis relative to the direct-stroke technique, and shorter hemostasis time, with no increase in total procedural steps.