A prospective, randomized trial of large- versus small- , volume endoscopic injection of epinephrine for peptic ulcer bleeding Hwaideng Lin, MD, FACG,Yu-Hsi Hsieh, MD, Guan-Ying Tseng, MD, Chin-Lin Perng, MD, Full-Young Chang, MD, Shou-Dong Lee, MD, FACG Taiwan, Republic of China Background: Endoscopic injection of epinephrine in the treatment of bleeding peptic ulcer is con- sidered highly effective, safe, inexpensive, and easy to use. However, bieeding recurs in 6% to 36% of patients.The aim of this study was to determine the optimal dose of epinephrine for endoscopic injection in the treatment of patients with bleeding peptic ulcer. Methods: One hundred fifty-six patients with active bleeding or nonbleeding visible vessels were randomized to receive small- (5-10 mL) or large-volume (13-20 mL) Injections of a 1 :10,000 solu- tlon of epinephrine. Results: The mean volume of epinephrine injected was 16.5 mL (95% CI [15.7, 17.3 mL]) in the large-v6lume group and 8.0 mL (95% CI [7.5,8.4 mL]) in the small-volume group. Initial hemosta- sis was achieved in all patients studied. The number of episodes of recurrent bleeding was small- er in the large-volume group (12/78,15.4%) compared with the small-volume group (24/78,30.8%, p = 0.037).The volume of blood transfused after entry into the study, duration of hospital stay, num- bers of patients requiring urgent surgery, and mortality rates were not statistically different between the 2 groups. Conclusions: Injection of a large volume (>I3 mL) of epinephrine can reduce the rate of recurrent . - bleeding In with high-risk peptic ulcer and is superior to injection of lesser volumes of epinephrine when used to achieve sustained hemostasis. (Gastrointest Endosc 2002;55:615-9.) Although peptic ulcer bleeding ceases sponta- Injection therapy is also effective, safe, relatively neously in 70% to 80% of patients, it can be cata- inexpensive, and easy to use.2 Impressive rates of strophic in the remainder.1 The mortality rate asso- hemostasis have been obtained by using solutions of ciated with such bleeding has been reduced to 2% to various agents such as epinephrine, absolute alco- 6% through the endoscopic use of various methods of hol, hypertonic saline solution, 50% glucose/water, therapy.2 Endoscopic hemostatic techniques are thrombin, and fibrin sealants.3-13Of these, epineph- thus a major advance in the treatment of these rine is the most commonly used agent for endoscopic patients. By reducing both morbidity and mortality, injection. There is a wide range in the volume (2-25 endoscopic therapy has become the treatment of mL) of epinephrine injected in various studies to first choice for ulcer hemorrhage.2 Of the various achieve hemostasis.3-12 The reported rates of initial possible techniques, the efficacy and safety of heat hemostasis obtained with epinephrine injection probe coagulation and multipolar probe electrocoag- range from 80% to 100%.3-12 However, bleeding ulation are well established.1 Hemostatic rates of recurs in 6% to 36% of patients.3-12 The reason for 80% to 95% can be achieved with these modalities. this wide variation in rates of recurrent bleeding is However, bleeding recurs in 10%to 30% of patients.1 unknown. Additionally, the optimal volume of epi- Received May 14, 2001. For revision July 12, 2001. Accepted nephrine needed to obtain hemostasis has not been September 21,2001. established. Current affiliations: Division of Gastroenterology,Department of The mechanisms that underlie hemostasis in Medicine, VGH-TAIPEI, and the School of Medicine, National response to endoscopic injection of epinephrine are Yang-Ming University, Taipei, Buddhish Tzu Chi Dalin General Hospital, Chia-Yi, and Ton-Yen Geneml Hospital, Hsin-Chu, vasoconstriction, vessel compression, and latel let lbiwaa ROC. aggregation.14~15Among these, the immediate Presented at Digestive Diseases Week, May 20-23, 2001, in mechanical compression of the bleeding vessel is Atlanta, Georgia (Gastrointest Endosc 2001;53:AB206). thought to be the most important for the initial con- Reprint requests: Hwai-Jeng Lin, MD, Division of trol if bleeding.16,17 Therefore, initial endoscopic Gastroenterology Department of Medicine, VGH-TAIPEI, Shih- Pai Rd, Sec 2, lbipei, Taiwan, 11217, R.O.C. injection of a larger volume of epinephrine is theo- Copyright 0 2002 by the American Society for Gastrointestinal retically better than injection of a smaller volume in Endoscopy 0016-5107/2002/$35.00 + 0 37111123271 patients at high risk for ulcer hemorrhage. The aim doi:lO.1067/mge.2002.123271 of this study was to compare the rates of recurrent VOLUME 55, NO. 6,2002 GASTROINTESTINfi ENDOSCOPY 615