Hu ML, Wu KL, Chiu KW, Chiu YC, Chou YP, Tai WC, Hu TH, Chiou SS, Chuah SK. Predictors of rebleeding after initial hemostasis with epinephrine injection in high-risk ulcers. World J Gastroenterol 2010; 16(43): 5490-5495
Corresponding Author of This Article
Dr. Seng-Kee Chuah, Division of Hepatogastroenterology, Department of Internal Medicine, Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of Medicine, 123, Ta-Pei Road, Niao-Song Hsiang 833, Kaohsiung, Taiwan. firstname.lastname@example.org
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Ming-Luen Hu, Keng-Liang Wu, King-Wah Chiu, Yi-Chun Chiu, Yeh-Pin Chou, Wei-Chen Tai, Tsung-Hui Hu, Shue-Shian Chiou, Seng-Kee Chuah, Division of Hepatogastroenterology, Department of Internal Medicine, Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of Medicine, Niao-Song Hsiang 833, Kaohsiung, Taiwan
Author contributions: Chuah SK designed the research; Hu ML performed the research and wrote the paper; Tai WC, Chiu YC, Chou YP, Hu TH and Chiou SS contributed equally to this work; Wu KL and Chiu KW and Chuah SK performed critical revision.
Supported by No Financial Interests or Grants support that might have an impact on the views expressed in this study
Correspondence to: Dr. Seng-Kee Chuah, Division of Hepatogastroenterology, Department of Internal Medicine, Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of Medicine, 123, Ta-Pei Road, Niao-Song Hsiang 833, Kaohsiung, Taiwan. email@example.com
Telephone: +886-7-7317123 Fax: +886-7-7322402
Received: July 31, 2010 Revised: August 25, 2010 Accepted: September 1, 2010 Published online: November 21, 2010
AIM: To identify the predictors of rebleeding after initial hemostasis with epinephrine injection (EI) in patients with high-risk ulcers.
METHODS: Recent studies have revealed that endoscopic thermocoagulation, or clips alone or combined with EI are superior to EI alone to arrest ulcer bleeding. However, the reality is that EI monotherapy is still common in clinical practice. From October 2006 to April 2008, high-risk ulcer patients in whom hemorrhage was stopped after EI monotherapy were studied using clinical, laboratory and endoscopic variables. The patients were divided into 2 groups: sustained hemostasis and rebleeding.
RESULTS: A total of 175 patients (144, sustained hemostasis; 31, rebleeding) were enrolled. Univariate analysis revealed that older age (≥ 60 years), advanced American Society of Anesthesiology (ASA) status (category III, IV and V), shock, severe anemia (hemoglobin < 80 g/L), EI dose ≥ 12 mL and severe bleeding signs (SBS) including hematemesis or hematochezia were the factors which predicted rebleeding. However, only older age, severe anemia, high EI dose and SBS were independent predictors. Among 31 rebleeding patients, 10 (32.2%) underwent surgical hemostasis, 15 (48.4%) suffered from delayed hemostasis causing major complications and 13 (41.9%) died of these complications.
CONCLUSION: Endoscopic EI monotherapy in patients with high-risk ulcers should be avoided. Initial hemostasis with thermocoagulation, clips or additional hemostasis after EI is mandatory for such patients to ensure better hemostatic status and to prevent subsequent rebleeding, surgery, morbidity and mortality.
Citation: Hu ML, Wu KL, Chiu KW, Chiu YC, Chou YP, Tai WC, Hu TH, Chiou SS, Chuah SK. Predictors of rebleeding after initial hemostasis with epinephrine injection in high-risk ulcers. World J Gastroenterol 2010; 16(43): 5490-5495
Bleeding peptic ulcer is a common and life-threatening medical emergency. With regard to endoscopic hemostasis, epinephrine injection (EI) monotherapy is a common and effective endoscopic method of hemostasis because of its low risk, low cost and high accessibility[2-4]. Although EI monotherapy has good efficacy in the hemostasis of bleeding peptic ulcers, bleeding recurs in about 10%-30% of the population[5-8]. Undoubtedly, recurrent bleeding remains the most important adverse independent prognostic factor[9,10]. Recent studies have revealed that thermocoagulation, sclerosant injection or clips alone, or combined with EI are superior to EI alone for preventing rebleeding, surgery and mortality. EI alone is not recommended[11-15]. However, the reality is that EI monotherapy to arrest ulcer bleeding is still commonly practiced, in part, because it is a relatively simple and effective hemostatic method and because some patients are intolerant of time-consuming EI combination therapy[16-18]. As sustained hemostasis is the goal of endoscopic therapy, we tried to identify factors which predicted recurrent bleeding after achieving initial hemostasis with EI monotherapy. Therefore, an additional hemostatic method to EI, or initial thermocoagulation, sclerosant injection or clips is warranted in such high-risk patients to ensure better hemostatic efficacy.
MATERIALS AND METHODS
Based on previous validations, those ulcers with spurting bleeding (Ia), oozing bleeding (Ib), non-bleeding visible vessel (NBVV, IIa) and adherent clots (IIb) were high-risk ulcers according to the Forrest classification and should be considered for endoscopic therapy[4,19]. From October 2006 to April 2008 in the Medical Center, Chang Gung Memorial Hospital, Kaohsiung, Taiwan, we enrolled patients with high-risk ulcers (I, IIa and IIb) who achieved initial hemostasis with EI monotherapy (epinephrine solution, 1:10 000) into this study. Patients who failed to achieve hemostasis during EI monotherapy or received endoscopic combination therapy were excluded. Moreover, patients with malignant ulcers were also excluded. Pharmacotherapy with proton pump inhibitors (PPIs) in these patients was also recorded.
Clinical, laboratory and endoscopic assessments
We analyzed the possible factors which predicted recurrent bleeding in patients with successful initial hemostasis. The clinical, laboratory and endoscopic variables investigated included age, sex, Forrest class, ulcer size, ulcer location, hemoglobin level, blood transfusion, difficulty in the injection approach, injection dose, patient status including outpatient (in the emergency room due to ulcer bleeding) and inpatient (development of ulcer bleeding during admission) and bleeding presentation including severe bleeding signs (SBS; hematemesis or hematochezia) or mild bleeding signs (MBS; coffee ground vomitus or melena). Shock status was defined as either a systolic blood pressure of less than 90 mmHg, or less than 100 mmHg plus a pulse rate of more than 100 beats per minute. The clinical risk status of the patients was assessed by means of the American Society of Anesthesiology (ASA) classification. That is, ASA I = healthy patient, ASA II = patient with mild systemic disease without functional limitation, ASA III = severe systemic disease with definite functional limitation, ASA IV = severe systemic disease that is a constant threat to life, and ASA V = moribund patient not expected to survive for more than 24 h with or without surgery. The hemoglobin level was recorded before blood transfusion and endoscopic therapy. The total amount of packed red blood cells (pRBC) transfused was recorded until endoscopic hemostasis was achieved. Use or non-use of nonsteroidal anti-inflammatory drugs (NSAID) and aspirin was also recorded. The accessibility of injection therapy was categorized into an easy or difficult approach depending on the location of the ulcers. Peptic ulcers located in the lesser curvature and posterior wall of the gastric body and posterior wall of the duodenum were regarded as those that required a difficult approach for injection therapy. Peptic ulcers located in areas other than the above-mentioned were regarded as easy approach. Rebleeding was defined as the recurrence of bleeding within 2 wk of initial hemostasis.
Univariate analysis for possible factors predicting recurrent bleeding was performed using the Pearson χ2 test and Fisher’s exact test for categorical variables and the Student’s t test for continuous variables. Variables with a P < 0.05 were considered statistically significant. Variables with a P < 0.4 in the univariate analysis were used for multiple logistic regression analysis with backward stepwise correction and were considered independent predictors of recurrent bleeding with a P < 0.05.
From October 2006 to April 2008, a total of 662 sessions of EI-based procedures to treat upper gastrointestinal bleeding were recorded in our computerized medical record system. After excluding patients who underwent endoscopic combination therapy, endoscopic hemostasis failure, and those who had malignant ulcer bleeding or Mallory-Weiss tear bleeding, a total of 175 patients with high-risk ulcers (I, IIa and IIb) who achieved initial hemostasis were enrolled. All these patients received intravenous PPI therapy during the acute bleeding period followed by oral PPIs to maintain hemostasis. Our records indicated that 144 patients (82.3%) achieved sustained hemostasis and 31 patients (17.7%) suffered from recurrent bleeding. Univariate analysis revealed that older age (age ≥ 60 years), advanced ASA status (category III, IV and V), shock, severe anemia (hemoglobin < 80 g/L), injection dose ≥ 12 mL and SBS were risk factors of recurrent bleeding (Table 1). However, backward stepwise correction (Table 2) revealed that only older age ≥ 60 years, odds ratio (OR) 5.11, 95% confidence interval (CI): 1.34, 19.48, hemoglobin < 80 g/L (OR 13.44, 95% CI: 4.29, 42.13), injection dose ≥ 12 mL (OR 5.72, 95% CI: 1.69, 19.38) and SBS (OR 5.46, 95% CI: 1.89, 15.79) were independent predictors. All 31 rebleeding patients received repeated endoscopic therapies, and only 14 of them achieved permanent hemostasis. The rest suffered from recurrent bleeding and 10 underwent surgery. In summary, among the 31 patients who re-bleeded after initial endoscopic hemostasis with EI alone, 15 (18/31, 48.4%) encountered delayed hemostasis causing major complications such as sepsis, hypovolemic shock, and renal and respiratory failure, and 13 (13/31, 41.9%) died of these complications (Table 3). The overall clinical course of the 175 patients is listed in Figure 1.
Table 1 Predictors of recurrent bleeding after epinephrine injection therapy in the univariate analysis.
1Some patients suffered 2 or more major complications related to recurrent bleeding.
Due to medical progress in the management of ulcer hemorrhage, pharmacotherapy with PPIs and endoscopic hemostasis are standard treatments. PPIs are the first choice of pharmacotherapy to control ulcer bleeding at present due to their strong inhibition of acid secretion and promotion of platelet aggregation. PPIs have also been demonstrated to reduce rebleeding, the need for surgery and repeated endoscopic therapy. With regard to endoscopic therapy, this treatment also reduces the occurrence of rebleeding, the need for surgery and the morbidity and mortality of patients. When bleeding recurs, however, repeated endoscopic therapy may either prevent patients from undergoing surgery or delay surgical hemostasis[23,24]. Therefore, the most important goal of endoscopic therapy is to initially achieve permanent hemostasis. Recently, numerous meta-analyses have indicated that adding a second procedure, such as a second injectate (alcohol, thrombin, sclerosant or fibrin glue), thermocoagulation or clips to EI significantly reduced rebleeding, surgery and mortality compared with EI alone in high-risk ulcer patients[11-15]. EI alone is not recommended in the management of bleeding ulcers; however, the reality is that endoscopic hemostasis with EI alone is still commonly practiced[16-18]. There are several reasons for this: first, injection monotherapy is a simple and effective hemostatic method (82.3% in the current study). Second, EI combination therapy is a relatively time-consuming procedure, so patients may not tolerate or complete the process, particularly those in a hemodynamically unstable status. Third, thermotherapy such as argon plasma coagulation (APC) or mechanical hemostasis with clips may not be available in every endoscopic unit, particularly in local hospitals. Therefore, to identify predictors of recurrent bleeding in high-risk ulcer patients after EI alone may justify the benefit of endoscopic combination therapy or initial replacement by thermocoagulation or clips for a better treatment outcome. Those patients with such risk factors treated with EI only should be closely monitored or referred to other hospitals with a well-equipped endoscopic unit. From the high-risk ulcer patients treated by PPIs pharmacotherapy and EI alone in our study, we found older age (≥ 60 years), advanced ASA categories (III, IV and V), shock, severe anemia (hemoglobin < 80 g/L), injection dose ≥ 12 mL and SBS were the factors which predicted rebleeding. Older age, severe anemia, high injection dose and SBS were independent predictors in multivariate analysis.
Compared to ulcers with NBVV or black spots, active ulcer bleeding tends to rebleed if left untreated. After successful endoscopic hemostasis and administration of PPIs, there were no statistical differences observed in this study. Severe anemia and SBS were relevant to the amount of blood loss and the severity of ongoing bleeding, and were risk factors for recurrent bleeding. The amount of blood transfused before endoscopic therapy may not correlate well with the severity of blood loss. In our hospital, a 24-h emergency endoscopy service is provided and patients often undergo early therapeutic endoscopy within 24 h of either visiting the emergency room or the occurrence of bleeding after admission. Thus, the amount transfused prior to early endoscopy cannot be considered as a risk factor of rebleeding. As for the use of large and small volumes of EI in bleeding ulcers, there are two prospective studies which refer to the hemostatic effectiveness of injection volume. In these studies, the injection of a larger volume might reduce the rate of recurrent bleeding compared to a smaller volume in a prospective design[25,26]. On the other hand, the need for a higher injection dose to arrest bleeding might imply a difficult hemostasis and a higher risk of recurrent bleeding from a retrospective viewpoint. In our study, an injection dose ≥ 12 mL was an independent risk factor of recurrent bleeding.
NSAID and aspirin also cause peptic ulcer bleeding. Prior use of NSAID/aspirin was reported to increase the risk of rebleeding in bleeding ulcer patients[27,28]. If the use of NSAID is discontinued, PPI therapy is very effective in treating NSAID-related ulcers and preventing further bleeding.
It is well known that Helicobacter pylori (H. pylori) infection can cause peptic ulcers. The detection of H. pylori was not performed in our study as these patients were in an acute bleeding phase. The rapid urease test during the acute bleeding phase is unreliable for the detection of H. pylori infection. Furthermore, Schilling et al also revealed that H. pylori infection does not affect the early rebleeding rate in patients with peptic ulcer bleeding after successful endoscopic hemostasis, however, patients should be tested and treated for H. pylori infection once their condition has stabilized to prevent recurrent ulcers.
EI monotherapy is commonly practiced in endoscopic hemostasis due to its good efficacy, and because it is a simple and time-saving technique[2-4]; however, it should be replaced by combination therapy, clips, sclerosant injection or thermocoagulation based on more recent evidence. Sclerosant injection, clips or thermocoagulation alone or in combination with EI are more effective methods than EI alone. Although we achieved sustained hemostasis in 82.3% of our patients with bleeding ulcers treated by endoscopic EI monotherapy, 31 patients suffered from recurrent bleeding. The outcome of the patients with rebleeding was discouraging as 15 patients (48.4%) suffered from major complications and 13 patients (41.9%) died of complications. Among the 31 rebleeding patients, 10 patients underwent surgical hemostasis but only 4 patients survived. Although these rebleeding patients often had underlying major diseases, to achieve sustained hemostasis using sclerosant, thermocoagulation, clips or EI combination as soon as possible would allow these patients to have a better prognosis. Therefore, we suggest that all patients with high-risk ulcers should undergo sclerosant injection, thermocoagulation or clips in index endoscopy because they are more effective and time-saving. EI followed by thermocoagulation, clips or another injectate should also be considered, however, these are more time-consuming and some patients cannot tolerate or complete the procedures. Unless sclerosant, APC or clips are unavailable, EI alone is not recommended, particularly in older patients (> 60 years), SBS, severe anemia, and those in need of a higher EI dose to arrest bleeding during endoscopic therapy. Additional endoscopic therapies to EI are mandatory or these patients should be referred to a hospital with a well-equipped endoscopic unit for close monitoring.
In conclusion, endoscopic EI monotherapy in patients with high-risk ulcers should be avoided. Initial hemostasis with thermocoagulation, clips or additional hemostasis after EI is mandatory, particularly in older patients, SBS, severe anemia and where a higher injection dose in need. Then, it is possible to ensure that such patients have a better hemostatic status and will avoid subsequent rebleeding, surgery, morbidity and mortality.
Recent studies have revealed that thermocoagulation, sclerosant injection or clips alone, or combined with epinephrine injection (EI) are superior to EI alone in preventing rebleeding, surgery and mortality. EI alone is not recommended. However, the reality is that EI monotherapy to arrest ulcer bleeding is still common practice, in part, because it is a relatively simple and effective hemostatic method and because some patients are intolerant of time-consuming EI combination therapy.
Sustained hemostasis is the goal of endoscopic therapy. This study identified the factors which predicted rebleeding if patients with high-risk ulcers were treated using EI alone to achieve initial hemostatic status.
Innovations and breakthroughs
Endoscopic EI monotherapy in patients with high-risk ulcers should be avoided. Initial hemostasis with thermocoagulation, clips or additional hemostasis after EI is mandatory, particularly in older patients, SBS, severe anemia, and where a higher injection dose is needed to prevent subsequent rebleeding, surgery, morbidity and mortality.
This article further emphasized the need to follow the existing guidelines for the treatment of high-risk ulcers to possibly avoid a poor outcome when EI is performed alone.
High-risk ulcers refer to ulcers with active bleeding, or with non-bleeding visible vessels or adherent clots, and endoscopic therapy is necessary in these ulcers to prevent further bleeding.
The clinical study focused on identification of predictors of rebleeding after initial epinephrine hemostasis in the upper gastrointestinal tract. The statistical approach revealed age, severe bleeding signs, hemoglobin, and epinephrine injection dose more than 12 mL as predictors.
Peer reviewer: Nikolaus Gassler, Professor, Institute of Pathology, University Hospital RWTH Aachen, Pauwelsstrasse 30, 52074 Aachen, Germany
S- Editor Sun H L- Editor Webster JR E- Editor Ma WH
Church NI, Palmer KR. Ulcers and nonvariceal bleeding.Endoscopy. 2003;35:22-26.
Sugawa C, Joseph AL. Endoscopic interventional management of bleeding duodenal and gastric ulcers.Surg Clin North Am. 1992;72:317-334.
Villanueva C, Balanzó J, Espinós JC, Fábrega E, Sáinz S, González D, Vilardell F. Endoscopic injection therapy of bleeding ulcer: a prospective and randomized comparison of adrenaline alone or with polidocanol.J Clin Gastroenterol. 1993;17:195-200.
Chung SC, Leung JW, Steele RJ, Crofts TJ, Li AK. Endoscopic injection of adrenaline for actively bleeding ulcers: a randomised trial.Br Med J (Clin Res Ed). 1988;296:1631-1633.
Oxner RB, Simmonds NJ, Gertner DJ, Nightingale JM, Burnham WR. Controlled trial of endoscopic injection treatment for bleeding from peptic ulcers with visible vessels.Lancet. 1992;339:966-968.
Balanzó J, Sainz S, Such J, Espinós JC, Guarner C, Cussó X, Monés J, Vilardell F. Endoscopic hemostasis by local injection of epinephrine and polidocanol in bleeding ulcer. A prospective randomized trial.Endoscopy. 1988;20:289-291.
Cook DJ, Guyatt GH, Salena BJ, Laine LA. Endoscopic therapy for acute nonvariceal upper gastrointestinal hemorrhage: a meta-analysis.Gastroenterology. 1992;102:139-148.
Rockall TA, Logan RF, Devlin HB, Northfield TC. Risk assessment after acute upper gastrointestinal haemorrhage.Gut. 1996;38:316-321.
Wong SK, Yu LM, Lau JY, Lam YH, Chan AC, Ng EK, Sung JJ, Chung SC. Prediction of therapeutic failure after adrenaline injection plus heater probe treatment in patients with bleeding peptic ulcer.Gut. 2002;50:322-325.
Park CH, Joo YE, Kim HS, Choi SK, Rew JS, Kim SJ. A prospective, randomized trial comparing mechanical methods of hemostasis plus epinephrine injection to epinephrine injection alone for bleeding peptic ulcer.Gastrointest Endosc. 2004;60:173-179.
Chung SS, Lau JY, Sung JJ, Chan AC, Lai CW, Ng EK, Chan FK, Yung MY, Li AK. Randomised comparison between adrenaline injection alone and adrenaline injection plus heat probe treatment for actively bleeding ulcers.BMJ. 1997;314:1307-1311.
Calvet X, Vergara M, Brullet E, Gisbert JP, Campo R. Addition of a second endoscopic treatment following epinephrine injection improves outcome in high-risk bleeding ulcers.Gastroenterology. 2004;126:441-450.
Barkun AN, Martel M, Toubouti Y, Rahme E, Bardou M. Endoscopic hemostasis in peptic ulcer bleeding for patients with high-risk lesions: a series of meta-analyses.Gastrointest Endosc. 2009;69:786-799.
Marmo R, Rotondano G, Piscopo R, Bianco MA, D'Angella R, Cipolletta L. Dual therapy versus monotherapy in the endoscopic treatment of high-risk bleeding ulcers: a meta-analysis of controlled trials.Am J Gastroenterol. 2007;102:279-289; quiz 469.
Andriulli A, Loperfido S, Focareta R, Leo P, Fornari F, Garripoli A, Tonti P, Peyre S, Spadaccini A, Marmo R. High- versus low-dose proton pump inhibitors after endoscopic hemostasis in patients with peptic ulcer bleeding: a multicentre, randomized study.Am J Gastroenterol. 2008;103:3011-3018.
Adamsen S, Bendix J, Kallehave F, Moesgaard F, Nilsson T, Wille-Jørgensen P. Clinical practice and evidence in endoscopic treatment of bleeding peptic gastroduodenal ulcer.Scand J Gastroenterol. 2007;42:318-323.
Tseng GY, Fang CT, Lin HJ, Yang HB, Tseng GC, Wang PC, Liao PC, Cheng YT, Huang CH. Efficacy of an intravenous proton pump inhibitor after endoscopic therapy with epinephrine injection for peptic ulcer bleeding in patients with uraemia: a case-control study.Aliment Pharmacol Ther. 2009;30:406-413.
Bleau BL, Gostout CJ, Sherman KE, Shaw MJ, Harford WV, Keate RF, Bracy WP, Fleischer DE. Recurrent bleeding from peptic ulcer associated with adherent clot: a randomized study comparing endoscopic treatment with medical therapy.Gastrointest Endosc. 2002;56:1-6.
Owens WD, Felts JA, Spitznagel EL Jr. ASA physical status classifications: a study of consistency of ratings.Anesthesiology. 1978;49:239-243.
Patchett SE, Enright H, Afdhal N, O'Connell W, O'Donoghue DP. Clot lysis by gastric juice: an in vitro study.Gut. 1989;30:1704-1707.
Leontiadis GI, Sharma VK, Howden CW. Proton pump inhibitor therapy for peptic ulcer bleeding: Cochrane collaboration meta-analysis of randomized controlled trials.Mayo Clin Proc. 2007;82:286-296.
Swain P. What should be done when initial endoscopic therapy for bleeding peptic ulcer fails?Endoscopy. 1995;27:321-328.
Lin HJ, Hsieh YH, Tseng GY, Perng CL, Chang FY, Lee SD. A prospective, randomized trial of large- versus small-volume endoscopic injection of epinephrine for peptic ulcer bleeding.Gastrointest Endosc. 2002;55:615-619.
Park CH, Lee SJ, Park JH, Park JH, Lee WS, Joo YE, Kim HS, Choi SK, Rew JS, Kim SJ. Optimal injection volume of epinephrine for endoscopic prevention of recurrent peptic ulcer bleeding.Gastrointest Endosc. 2004;60:875-880.
Godil A, DeGuzman L, Schilling RC 3rd, Khan SA, Chen YK. Recent nonsteroidal anti-inflammatory drug use increases the risk of early recurrence of bleeding in patients presenting with bleeding ulcer.Gastrointest Endosc. 2000;51:146-151.
Vreeburg EM, de Bruijne HW, Snel P, Bartelsman JW, Rauws EA, Tytgat GN. Previous use of non-steroidal anti-inflammatory drugs and anticoagulants: the influence on clinical outcome of bleeding gastroduodenal ulcers.Eur J Gastroenterol Hepatol. 1997;9:41-44.
Archimandritis A, Tzivras M, Sougioultzis S, Papaparaskevas I, Apostolopoulos P, Avlami A, Davaris PS. Rapid urease test is less sensitive than histology in diagnosing Helicobacter pylori infection in patients with non-variceal upper gastrointestinal bleeding.J Gastroenterol Hepatol. 2000;15:369-373.
Schilling D, Demel A, Nüsse T, Weidmann E, Riemann JF. Helicobacter pylori infection does not affect the early rebleeding rate in patients with peptic ulcer bleeding after successful endoscopic hemostasis: a prospective single-center trial.Endoscopy. 2003;35:393-396.