Endoscopic management and long-term follow-up of Dieulafoy's lesions in the upper GI tract

Background: The aim of this study was to retrospectively evaluate the short-and long-term effectiveness of different methods of endoscopic treatment for bleeding Dieulafoy's lesions. Methods: Patients were allocated into 2 groups according to the hemostatic method applied: (1) injection group (epinephrine and/or ethanolamine oleate), and (2) thermal coagulation group (heat probe), either alone or combined with epinephrine injection. The combination of epinephrine and ethanolamine oleate was used in 5 patients, epinephrine alone in 3, ethanolamine oleate alone in one, heat probe and epinephrine in 8, and heat probe alone in 1 patient. Results: Dieulafoy's lesions were found in 18 (1%) of 1750 patients with acute nonvariceal upper GI bleeding. Comorbid conditions were present in 5 (28%) patients. Initial hemostasis was achieved endoscopically in 13 patients (72%) and permanent hemostasis in 17 patients (94%). Bleeding recurred in 5 patients (2 with concomitant disease) in the injection group; 3 were successfully retreated by heat probe coagulation and epinephrine injection, 1 with hemoclip application and 1 by surgery. There was no recurrent bleeding in thermal treatment group. Thermal treatment was significantly superior to injection (p = 0.0029). Conclusions: Endoscopic thermal coagulation with or without epinephrine injection should be the initial treatment of choice for Dieulafoy's lesions. Mortality is lowest in patients with no significant comorbidity and an unremarkable medical history. (Gastrointest Endosc 2002;55:527-31.)