Saudi Journal of Gastroenterology

: 2002  |  Volume : 8  |  Issue : 1  |  Page : 17--21

Predictive factors for failure of endoscopic management therapy in peptic ulcer bleeding

Radhakrishnan Siva, Ghazi Al Zubaidi, Al Kashoob Masoud, Mohan Nihar 
 Department of Medicine, Royal Hospital, Muscat, Oman

Correspondence Address:
Radhakrishnan Siva
Senior Consultant Gastroentrologist, Royal Hospital, P. O. Box 1331, CPO Seeb-111, Muscat


Background: After endoscopic therapy for peptic ulcer bleeding, rebleeding occurs in up to 20% of patients. Objective: The aim of this retrospective analysis was to identify the factors responsible for failure to achieve hemostasis or rebleeding after endoscopic therapy. Methods: Seventy six patients who bled from peptic ulcers and received endoscopic therapy were identified in a retrospective analysis of six years, from 1993 to 1998, in a tertiary care hospital in Muscat, Oman. All patients were endoscoped within 24 hours and received endoscopic treatment, either injection of 1/10,000 adrenaline alone or both injection and thermocoagulation. We looked at the following factors, which could have influenced the outcome of endoscopic therapy. The clinical and endoscopic parameters used to assess the outcome of endoscopic therapy were: age, sex, blood pressure and hemoglobin on admission, number of units of blood transfused , use of NSAID, comorbid conditions, ulcer site, ulcer size. Forrest grade, injection alone and injection plus thermocoagulation. Results: Endoscopic therapy failed in 16 patients (21%). Twelve patients received a second endoscopic treatment , but 13 patients eventually required surgery. Six patients died as a result of bleeding (mortality 8%). Among the parameters, hemoglobin on admission, more than six units of blood transfusion, shock, co-morbid diseases, ulcer site and size and Forrest grade la predicted the possibility of failure of endoscopic therapy. Conclusion: In patients presenting with peptic ulcer bleeding, hemoglobin less than 10 grams, more than six units of blood transfusion, shock on admission, co-morbid disease, posterior wall duodenal ulcer, large ulcer size of > I cm size and Forrest la predicted failure of endoscopic therapy.

How to cite this article:
Siva R, Al Zubaidi G, Masoud A, Nihar M. Predictive factors for failure of endoscopic management therapy in peptic ulcer bleeding.Saudi J Gastroenterol 2002;8:17-21

How to cite this URL:
Siva R, Al Zubaidi G, Masoud A, Nihar M. Predictive factors for failure of endoscopic management therapy in peptic ulcer bleeding. Saudi J Gastroenterol [serial online] 2002 [cited 2020 Jun 6 ];8:17-21
Available from:

Full Text


An NIH consensus conference on therapeutic endosocpy and bleeding ulcers in 1990 concluded that endoscopic hemostasis therapy should be used in patients at high risk of recurrent bleeding and death [1] . The outcome of bleeding from peptic ulcer is partly dependent on the endoscopic stigmata of bleeding as described by Forrest [Table 1]. It is recommended that, patients with Forrest la, Ib, Ila and lib should receive endoscopic therapy [2] . Two meta-analysis studies reviewing the data from 25-30 prospective endoscopic trials, demonstrated a reduction in mortality in 30%-45%, and a 60%-70% decrease in the rate of recurrent bleeding and need for emergency surgery [3],[4] . Adrenaline injection (1:10,000 dilution) and/or coaptive thermal coagulation using heater probe has been established as the standard treatment for actively bleeding ulcers [5] . Nevertheless, rebleeding occurs in 10%­ 30% of patients after initial hemostasis [6],[7],[8] . A number of other studies have reported factors predicting failure of endoscopic therapy in bleeding peptic ulcer [9],[10],[11],[12] . Despite this, the conclusion from various reports remains uncertain. Almost all the reports are from the West and from the Far East. This retrospective analysis was carried out to identify the predictive factors for failure of endoscopic therapy in bleeding peptic ulcers in a teaching hospital in the Middle East

 Patients and Methods

Patients who presented to the Royal hospital, Muscat with active upper GI bleeding and received endoscopic therapy were reviewed retrospectively. Ninety-five patients were identified. Nineteen were excluded since they had bleeding from sources other than peptic ulcer (maliganancy-5; Mallory- Weiss tear-6; vascular ectasia-8). Seventy-six patients who had bleeding from peptic ulcer received endoscopic treatment at the initial endoscopy session.

All patients were admitted to a high dependency unit and endoscoped within 24 hours. Patients admitted with fresh hemostosis and those who were haemodynamically unstable were endoscoped as soon as they had been resuscitated. Informed consent was obtained from all patients before the procedure. Endosocpy was carried out using Olympus GIF Q200; GIF Q30 and GIF IT endoscopes. Endoscopic treatment consisted of injection of adrenaline (1:10,000 dilution) in all patients using a 21-gauge, six mm long needle, submucosally in I to 2 ml aliquots around the bleeding point until the bleeding was controlled. Mean amount of injection was 8 ml (6-18 ml). Twenty patients received coaptive coagulation using heater probe (Olympus CD- 20Z - 2.8 mm) at 30 Joules of energy, in addition to adrenaline injection as described previously [13] . Four endoscopists were involved and since this is a retrospective analysis, the decision to use additional heater probe was taken by the individual endoscopist. However, all six patients with Forrest la lesion (spurting hemorrhage) received combined therapy. All patients received intravenous H2 receptor blockers for the following 48 hours. Patients were monitored for persistent bleeding or rebleeding. Recurrence of bleeding was considered if the patients had vomited fresh blood or fresh blood reappeared in the naso-gastric tube. Other signs of recurrence of bleeding were, development of hypotension (systolic pressure 90 mm of Hg or less or pulse rate of 120/ minute or more), passage of fresh blood per rectum or reappearance of melena and requirement of more than four units of blood in the 48-hour period following endoscopic treatment. Most patients, who rebled during the observation period were repeated endoscopic therapy. The mean observation period was 72 hours (48 hours to 7 days). Referral to surgery was made for patients failed the second session of endoscopic therapy.


From January 1993- December 1998, ninety-five patients with bleeding from the upper GI tract who had received endoscopic therapy were identified. Nineteen patients who had non-ulcer bleeding were excluded. Seventy six patients were identified to have bleeding from peptic ulcers and had Forrest la, lb, Ila or IIb stigmata of bleeding. The Forrest classification is shown in [Table 1]. The demographic data of these patients are as in [Table 2]. After the first session of endoscopic therapy, bleeding was controlled in 60 patients. One patient had perforation following combined therapy with adrenaline injection and coaptive coagulation with heater probe. A 2.8 mm Olympus heater probe with 30 joule energy was applied as in all other patients.

Of the 16 patients who had failure of endoscopic therapy, four patients had surgery because of persistent bleeding. Second endoscopic treatment was done in twelve patients. Bleeding was controlled in three (25%). Nine patients in this group also had surgery. Nine patients, among the 76 patients bleeding from peptic ulcer, died. Three of these deaths were unrelated to the bleeding. One patient died following rebleeding, ten days after second session of endoscopic treatment, but his condition was considered too grave for surgery.

Thus, total thirteen patients had surgery for uncontrolled bleeding or rebleeding following endoscopic therapy. Among the patients who had surgery, five died (30 day operative mortality of 38%). The cause of death in these patients were, uncontrolled bleeding in one , septicemia and DIC in two and underlying coronary vascular disease in the other two patients.

Based on the clinical and the endoscopic parameters which could have influenced the outcome of endoscopic therapy [Table 3],[Table 4], it was found that posterior wall duodenal ulcer , ulcer size more than one cm in size, Forrest grade la bleeding , admission hemoglobin less than 10 gms /dl, requirement of more than six units of blood transfusion, shock on admission and co-morbid conditions predicted failure of endoscopic treatment.


Majority of the peptic ulcer bleeding ceases spontaneously in 70%-80% of patients, however, further bleeding in the remaining may be catastrophic with a mortality rate of about l0% [14],[15],[16] . Endoscopic treatment is warranted in patients with active bleeding and visible vessel or fresh clot on the ulcer base [17] . Although, high initial hemostatic rate can be achieved with endoscopic injection of adrenaline alone, the rebleeding rate is reported to be 15%-36% [5],[7] . Chung et al, showed that addition of heater probe decreases the rebleeding in patients with spurting hemorrhage [5] . Our numbers were small, and such a benefit for combined treatment could not be demonstrated in this review.

Endoscopic treatment, in general fails to control bleeding in 15-20% patients, and these patients have high mortality . A number of studies addressing the risk factors for failure of endoscopic treatment failed to reach a consensus as to which were important, but ulcer location on the posterior wall of the duodenal bulb, age over sixty five and presence of other co­morbid conditions have frequently cited as being significant [9],[10],[11],[12] . As in other published series, endoscopic hemostasis failed in 16 of the 76 patients in this series. We found posterior wall duodenal ulcer, ulcer size more than one cm, Forrest grade I a bleeding, admission hemoglobin less than 10 gms/dl, requirement of more than six units of blood transfusion, shock and co-morbid conditions predicted failure of endoscopic treatment. Repeat endoscopic treatment has succeeded to control the bleeding and has reduced the need for surgery in about 50% of patients [6],[18] . In this series as well, the initial endoscopic treatment failed in 16 patients. Twelve of these patients had a second endoscopic treatment and the bleeding was controlled in three, thus eliminating the need for emergency surgery . This once again confirms that a second endoscopic treatment can be effective if the first session fails.


Peptic ulcer bleeding has high mortality and consumes substantial resources. A knowledge of risk factors for poor prognosis should lead to active resuscitation, prompt endoscopic therapy and intensive care for this high risk patients. Endoscopic treatment is effective in controlling the bleeding in the majority and these measures should improve the treatment outcomes.


1NIH consensus conference. Therapeutic endosocpy and bleeding ulcers. Gastrointest Endoscopy 1990;36:S 1-65.
2Lame L. Peterson W. Bleeding peptic ulcer. N Engl. J Med 1994:331:721-7.
3Sacks HS, Charmers TC, Blum AL. Endoscopic hemostasis: an effective therapy for bleeding peptic ulcers. JAMA 1990:264:494-9.
4Cook DJ, Guyatt GH, Salene BJ, Lame LA. Endoscopic therapy for acute non-variceal upper GI hemorrhage: a metanalysis . Gastroenterology 1992;102:139-48.
5Chung SCS, Lau JYW, Sung JJY. Randomized comparison between adrenaline injection alone and adrenaline injection and heater probe treatment for actively bleeding ulcers. BMJ 1997:314:1307-11,
6James YW, Lau MB, Joseph JY, Sung MD, Chung SCS. Endoscopic retreatment compared with surgery in patients with recurrent bleeding after initial endoscopic control of bleeding ulcers. N Engl. J med. 1999;340:751-6.
7Villanueva C. Balanzo J. Torras X. Value of second-look endoscopy after injection therapy for bleeding peptic ulcer: a prospective and randomized trial. Gastrointest Endosc 1994:40: 34-9.
8Llach J. Bordas JM, Salmeron JM. A prospective randomized trial of heater probe thermocoagulation versus injection therapy in peptic ulcer hemorrhage. Gastrointest Endosc 1996:43:117-20.
9Brullet E. Calver X. Campo R. Factors predicting failure endoscopic injection therapy in bleeding duodenal ulcer. Gastrointest Endosc 1996;43: 111-6.
10Saeed ZA, Ramirez FC, Graham DY. Prospective validation of Baylor bleeding score for predicting the likelihood of rebleeding after endoscopic hemostasis of peptic ulcers. Gastrointest Endosc 1995; 41:561-5.
11Choudhuri CP. Rajagopal C, Elton RA. Palmer KR. Failures of endoscopic therapy for bleeding peptic ulcer : an analysis of risk factors. Am J gastroenterol 1994:89: 1968-72.
12Lin HJ, Tseng GY. Lo WC, Lee FY, Perng CL. Predictive factors for rebleeding in patients with peptic ulcer bleeding after multipolar electrocoagulation: a retrospective analysis. J gastroenterol 1998;26: 113-16.
13Leung JW, Chung SCS. Practical management of nonvariceal upper gastrointestinal bleeding. In: Tytgat GNJ, Classen M. Practice of therapeutic endosocpy. Longman group, 1994:1-16.
14Rockall TA. Logan RFA. Devlin HB. Northfield TC Steering Committee and Members of national Audit of Acute Upper GI Haemorrhage. Risk assessment after acute upper GI haemorrhage Gut 1996; 38:316-21
15Gilbert DA. Silversterine FE, Tedesco FJ Buenger NK, Persing J. The national ASGE survey on upper gastrointestinal bleeding. Gastrointest Endosc 1981; 94:94­-102.
16Rockall TA, Logan RF, Devlin HB, Northfield TC. Incidence of and mortality from upper gastrointestinal haemorrhage in the United Kingdom. BMJ 1995: 311 : 222-­6.
17Garripoli A, Ferrari A, De Angelis A. Sambataro A Emergency endoscopy in upper gastrointestinal bleeding at a second level unit: clinical results and quality assurance. three year experience. Giorn Ital. End Dig 1996;19: 261-85.
18Saeed ZA, Cole RA, Ramirez FC, Schneider FE, Graham DY. Endoscopic retreatment after successful initial hemostasis prevents ulcer rebleeding: a prospective randomised trial. Endosocpy 1996;28:288-94.