Saudi Journal of Gastroenterology
Home About us Instructions Submission Subscribe Advertise Contact Login    Print this page  Email this page Small font sizeDefault font sizeIncrease font size 
Users Online: 1084 

ARTICLES Table of Contents   
Year : 1995  |  Volume : 1  |  Issue : 1  |  Page : 31-36
Endoscopic management of serious non-variceal upper GI bleeding with local injection therapy

Division of Gastroenterology, Department of Medicine, Dr. Erfan & Bagedo General Hospital, P.O. Box 6519, Jeddah, Saudi Arabia

Click here for correspondence address and email


Massive non-variceal upper GI bleeding is a very common and serious problem. especially if bleeding comes from a vessel. In these circumstances, unless therapeutic endoscopy or emergency surgery is per­formed, the bleeding may lead to a very high mortality. Endoscopic injection therapy can be performed by injecting diluted epinephrine [1:10,000] or normal saline; alone or with a sclerosant. Adding a scleros­ant is preferred if the bleeding is massive or has a spurt pattern. This technique proved to be highly effec­tive and safe and almost comparable to other types of therapeutic endoscopies like bipolar electro­coagulator (BICAP) heat probe or laser photocoagulation therapy. The clinical studies performed in this field, the technique of injection and the types and amount of the substance to be injected will be discussed in details.

How to cite this article:
Subei lM. Endoscopic management of serious non-variceal upper GI bleeding with local injection therapy. Saudi J Gastroenterol 1995;1:31-6

How to cite this URL:
Subei lM. Endoscopic management of serious non-variceal upper GI bleeding with local injection therapy. Saudi J Gastroenterol [serial online] 1995 [cited 2022 Jun 30];1:31-6. Available from:

   Natural History of Non-Variceal Upper Gastroin­testinal Bleeding Top

Non-variceal upper gastrointestinal bleeding is a common and serious problem with a high mor­bidity and an overall mortality of 10%; a figure that has not changed during the last four dec­ades [1] and until recently about 20% of the patients needed emergency surgery [2] which car­ried a much higher mortality than the elective one. [3] Schiller et al [3] showed that the mortality rate of emergency surgery reached 19.2% com­pared to 7.8% mortality rate in the elective surgery. On the other hand, it has been found that around 80% of the patients with non-variceal upper GI bleeding stops spontaneously by the time endoscopy is performed or even by the arri­val to the hospital. However, 20% of the patients continue to bleed or rebleed in spite of intensive medical therapy. Several risk factors have been identified as predictors of poor outcome of the non-variceal upper GI bleeding in which cases emergency surgery may have to be performed [Table - 1]. The first group of factors are related to the severity of hemorrhage: i.e. massive bleeding, continuous bleeding in spite of intensive medical therapy and rebleeding. [1],[2],[3],[4],[5],[6],[7],[8],[9],[10],[11],[12] Gilbert, et al [1] found in their analysis of the reports of the par­ticipating ASGE members on 2,427 upper GI bleeders that massive and active bleeding at the time of endoscopy carried a mortality rate of 2-3 times the mortality of the whole group of patients. Fleischer [7] found the mortality rate of severely and continuously bleeding patients to be 33% as compared to the overall mortality rate of 6.9%. Schiller, et al. [3] found the mortality rate of the rebleeders to be 10% as compared to 5% mortal­ity rate of the self-limited single episode bleeders. The other group of factors that predict a poor out­come of the non-variceal upper GI bleeding are the presence of other comorbid problems e.g. age ≥ 60 and the presence of other systemic illnesses such as heart failure, shock or being on mechani­cal ventilator. [2]

Three endoscopic stigmata have been found to predict a poor outcome [7],[8],[9],[10],[11],[12] and has been called stigmata of hemorrhage [SH] which are: 1) actively bleeding lesion at the time of endoscopy 2) the presence of a fresh blood clot on the lesion 3) the presence of a visible vessel [VV].

All investigators agreed on the fact that actively bleeding lesions at the time of endoscopy, espe­cially those with spurting bleeding carry a very high morbidity and mortality rate and unless therapeutic measures are performed, the need for emergency surgery is inevitable. [5] On the other hand, the role of the non-bleeding visible vessel or the fresh clot as predictors of high incidence of rebleeding has been subjected to controversies. [7],[8],[9],[10],[11],[12],[13] In the study of Griffith, et al., [9] 100% of the 28 patients with active bleeding from a visible vessel needed surgical intervention because of uncon­trolled bleeding or rebleeding, while bleeding stopped and did not recur in patients with bleed­ing peptic ulcer that had no visible vessel. Wara, [10] however questioned the role of the visi­ble vessel as a sole factor in predicting rebleeding; where in the visible vessel that was bleeding or covered with a fresh clot at the time of endoscopy was found to predict a high rebleeding rate. These controversies, however, could be attributed to two major reasons. The first one is that the defini­tion of a visible vessel v.s. a sentinel clot was unclear and many authors used these two descrip­tions interchangeably while they are indeed two different entities. [11],[12] Freeman, et al. [12] made it clear that a visible vessel is a pale or colorless pro­tuberance in the base of the ulcer while the sen­tinel clot is a colored one i.e. either red or black. In that study, it was found that the real visible ves­sel was accompanied with a 71% incidence of re­bleeding while the red sentinel clot had a 38% re­bleeding rate and the black one had 0% rebleed­ing rate. Furthermore, the sentinel clot disap­peared within 1-3 days of the first endoscopy while on medical treatment, wherein the real visible ves­sel stayed apparent until rebleeding occurred. The second reason that contributed to the controver­sies regarding the role of the visible vessel was found to be related to the location of the lesion that had the visible vessel i.e. the presence of the visible vessel in the gastric or duodenal ulcers pre­dicted a higher incidence of rebleeding compared to the visible vessel seen in pre-pyloric ulcer. [10] Nevertheless, all the authors agreed on the fact that when none of the stigmata of hemorrhage is present, the bleeding is usually self-limited and rarely rebleeds once it stops. [8],[9],[10],[11],[12] The size of the bleeding ulcer is also found to be a significant fac­tor that is, ulcers larger than 2, cm. carry a much higher risk than the smaller ulcers. [14] In this dilemma it is important also to notice that the experience of the endoscopist plays a very impor­tant role in interpreting the endoscopic findings, especially, if he/she has not had a chance to learn the endoscopic characteristics of these stigmata of hemorrhage. [13]

Furthermore, it seems that the combination of endoscopic stigmata of hemorrhage and the clini­cal features of major hemorrhage provides a more reliable prediction of a poor outcome than having either of the two groups of factor alone, [10],[11] espe­cially if this occurred in the elderlies or in those who have serious co-morbid illnesses. [1],[21],[3],[4]

   Role of Endoscopy in Non-Variceal Upper GI Bleeding Top

Endoscopy has revolutionized the diagnostic evaluation of patients with various digestive dis­eases including an accurate diagnosis of various causes of upper GI bleeding [4] however, Graham [5] and peterson, et al. [6] found that diag­nostic endoscopy by itself did not improve the out­come of these patients. Nevertheless, the observa­tion of these two groups can easily be criticized now due to the fact that at the time of this observa­tion, the management of upper GI bleeding was limited to pharmacotherapy which later on proved to be ineffective [1],[2],[3],[4] or to surgical intervention which has a very high morbidity and mortality. [2],[3],[4]

In the last decade, the scope proved to be a suc­cessful tool in performing various therapeutic interventions and can achieve hemostasis in patients with non-variceal upper GI bleeding who did not respond to medical management. Therapeutic endoscopy resulted in a drastic decrease in the number of emergency surgery per­formed and a decrease in the mortality, as well. [2],[4],[14],[15],[16],[17],[18],[19],[20],[21],[22],[23],[24],[25],[26],[27],[28],[29],[30] The principle of these endoscopic therapeutic modalities is to obliterate the vessel that is responsible for this massive bleeding. By using the thermal modalities, heat energy can be applied locally to the bleeding vessel which causes coagulation of that vessel leading to obliteration of its lumen. [14],[15],[16],[17],[18] This heat energy can be deliv­ered in a form of laser photocoagulation energy using, the neodymium yag laser monopolar or bipolar electrocoagulators or heat probe. All these three modalities proved to be highly and almost equally effective in inducing permanent hemostasis in more than 90% of the patients, which also resulted in a significant reduction in the amount of blood transfused, the hospital stay and the need for emergency surgery, all of which indi­rectly decreased the mortality rate in patients with massive non-variceal upper GI bleeding. [14],[15],[16],[17],[18]

   Endoscopic Injection Sclerotherapy in Serious Non-Variceal Upper Gastrointestinal Bleeding Top

It was noticed that endoscopic injection sclerotherapy in esophageal variceal bleeding was very effective not only in controlling the bleeding but if properly performed, the varices could be obliterated completely and replaced by a fibrous tissue; this observation tempted the clinicians to try it in non-variceal bleeding since the mid 70's [17] and it has become popular in the late 80's.

Various substances have been used to achieve hemostasis in significant non- variceal upper GI bleeding: the most commonly used one is diluted epinephrine [1:10,000], normal saline alone, [22],[29],[30],[31],[32] or normal saline followed by injection of sclerosing substances like hypertonic saline, [20],[21] dextrose 50%, [27] polidocanol, [23] sodium tet­radecyl sulphate [25],[30] or absolute dehydrated alcohol. [24],[27] Even distilled water was locally injected with almost similar rate of efficacy. [29]

Furthermore, endoscopic injection sclerotherapy was used together with other hemostatic modalities like heat probe or BICAP. Initial hemostasis was achieved with injection sclerotherapy then the bleeding vessel was sub­jected to the thermal coagulator. [28]

   Mechanisms of Action of Endoscopic Injection Therapy Top

The main effect of the endoscopic injection therapy is achieved by creating a significant tissue edema around the bleeding vessel constricting it via a tamponade effect. [25],[29],[32] This mechanism was more substantiated by the study of Lai et al [29] who found that local injection of distilled water was as effective as injection of epinephrine. Epinep­hrine also causes vasoconstriction, increases platelets aggregation and causes low degree of local tissue damage leading to sclerosis and fibrosis and of the vessel, [32] while the main mechanism of action of absolute alcohol and other sclerosants is directed towards creating tissue fibrosis and sclerosis similar to their effect on varices. [25]

   Clinical Studies on the Efficacy and Safety of Endos­copic Injection Therapy Top

Initial non-randomized clinical studies proved that endoscopic injection therapy was highly effec­tive in achieving permanent hemostasis in about 90% of patients who had actively bleeding non-vari­ceal upper GI lesions. [20],[21],[22],[23],[24],[25]

One of the pioneer randomized controlled study was performed by Chang et al. [26] who proved that patients who received endoscopic injection therapy aimed at stopping an actively bleeding peptic ulcers did significantly better than the control group received conservative medical treatment. Injection therapy resulted in a higher rate of hemostasis, a less amount of blood transfusion needed and a far lower rate of emergency surgery compared to con­servative medical management (i.e 15% v.s. 41%, respectively). Lin et al. [25] randomized 250 patients with actively bleeding vessels at the base of peptic ulcer into five groups each consisted of 50 patients; Four groups received endoscopic injection therapy with four different substances i.e. normal saline, hypertonic saline, dextrose 50% or absolute alcohol, and the fifth group was a control one treated only medically. The outcome of the four groups that had injection therapy was almost similar to each other and permanent hemostasis was achieved in about 90% but all had far better out­come when compared to the control group who had a rebleeding rate of almost 85%.

When compared to other therapeutic endoscopic modalities, endoscopic injection therapy is proved to have the same effectiveness in achieving long term hemostasis. [14] In all the studies discussed above endoscopic injection therapy proved to be a very safe technique since no significant local or sys­temic complications has been reported [19],[20],[21],[22],[23],[24],[25],[26],[27],[28],[29],[30],[31] Chung, et al [32] measured the serum level of epinephrine in 18 actively bleeding ulcer patients subjected to endos­copic injection therapy of epinephrine solution 1:10,000 dilution. Epinephrine serum level increased immediately up to three folds of the baseline level but returned to normal within 20 minutes of the injection. Nevertheless, no car­diovascular complication could be detected in any of these patients. They postulated that Xylocaine which had been sprayed locally in the pharynx could have prevented the occurrence of any arrhythmias in these patients.

   Methods and Materials used in Endoscopic Injec­tion Therapy Top

One of the advantages of endoscopic injection therapy is that it is readily available at the time of performing diagnostic endoscopy. [25] Once the etiology and the location of the bleeding is deter­mined, diluted epinephrine 1:10,000 is injected 5­10 mm away from the bleeding vessel, usually each site is injected with 0.5 to 1 ml of the epinep­hrine solution. Even pinpointing the exact bleed­ing point is not initially needed, since upon inject­ing epinephrine hemostasis will occur which ena­bles the endoscopist to clear the field and locate exactly the bleeding point. [25]

In general, the size and the location of the lesion [S], as well as the severity of the hemorrhage all dictate the total amount of the solution used and the number of sites to be injected. [25],[27] In most of the instances, 3-10 ml are used per lesion per ses­sion. Once hemostasis is achieved, few seconds post injection the vessel can then be directly injected with a sclerosant, e.g. alcohol, polidocanol or STD 0.2-0.5 ml or coagulated with thermal modalities. [25],[26],[27],[28]

Most of the endoscopists use 23 or 25 gauge reg­ular sclerotherapy injectors for this purpose, i.e. the same injector catheter used for sclerotherapy of the varices. However, a new type of double lumen injector catheter has been used [33] which allows simultaneous washing of the site of bleed­ing with water while injecting the sclerosant. Another advantage of this catheter is that it can penetrate the hard base of the ulcer or the hard tumor tissue much easier than the regular one since it is much stiffer. [33] Villanueva, et al. [31] prospectively studied the value of performing 24 hours of hemostasis and then to perform another session of injection therapy if a visible vessel is seen. They found no significant difference of the outcome between those who were randomized to be re-endoscoped and those who were not. So, it was concluded that routine second look endos­copy is not recommended in most of the patients with non- variceal upper GI hemorrhage and the other sessions of endoscopy should be performed only if the bleeding recurs. Most of the authors agreed that if the patient rebleeds after two ses­sions of injection therapy, further sessions are unlikely to help and the patient should be referred for surgical intervention. [25]

   Types of Bleeding Lesions that can be Injected Top

Most the studies focussed on peptic ulcers that have a bleeding vessel since they represent the majority of the non-variceal upper GI bleeding lesions. [20],[21],[23],[25],[27],[28],[29],[30],[31],[32] It is also recommended to perform injection therapy whenever a visible ves­sel is seen even if it is not bleeding at the time of endoscopy. [12],[22],[24],[27],[31] . If bleeding is seen oozing from the edge of the ulcer without a visible vessel, it is almost agreed on that there is no need to per­form any sort of therapeutic endoscopy including injection therapy since the bleeding is usually mild and self-limited. [1],[2],[3],[4], [ 7],[8],[9],[10],[11],[12],[13],[14],[15],[16],[17],[18],[19],[0],[21],[22],[23],[24],[25],[26], [27,[28],[29],[30],[31] Other studies have also found that this technique is very effective in vari­ous other causes of massive non variceal upper GI bleeding beside peptic ulcer i.e. Mallorey-Weiss lesion, erosive gastritis and bleeding tumors. [19],[21],[22],[25]

Finally, few questions still needed to be agreed upon among endoscopists in relation to endos­copic injection therapy i.e.: a) which is the best substance?; b) what is the optimal amount of that substance needed to be injected to achieve the highest rate of hemostasis with the least complica­tions?; c) whether to combine two or more sub­stances; one to achieve hemostasis and the other to sclerose the bleeding vessel. [34]

   Conclusion Top

Endoscopic injection therapy seems to be very effective and safe method in the management of massive non-variceal upper gastrointestinal hemorrhage. The clinical studies have proved that its efficacy and safety features are almost similar to other forms of therapeutic endoscopy, how­ever, the advantages of injection therapy are: a. it is costless, readily available at the time of diagnos­tic endoscopy and easy to perform. b. at the begin­ning of the procedure, there is no need to locate exactly the bleeding point since the injection can be started as near to the bleeding lesion as possi­ble, then once transient hemostasis is achieved and the field becomes clear, further injections will be directed to the bleeding point.

   References Top

1.Gilbert DA, Silverstein FE. Tedesco FJ, et al: The National ASGE Survey on upper Gastrointestinal Bleeding. Gastrointest Endosc 1981; 27:94-103.  Back to cited text no. 1    
2.Johnston J. Natural History of Upper Gastrointestinal Bleeding and Determination of Outcome. In: Fleischer D, Jansen D, Bright-Asare P (eds). Therapeutic Laser Endoscopy in Gastrointestinal Disease, Boston. Mar­tinus Nijhoff publishers 1983: 3:29-37.  Back to cited text no. 2    
3.Schiller KFR, Truelove SC, Williams OG. Haematemesis and Melena, with Special Reference to Factors influencing the Outcome. Br Med J 4 1970; 1:7­-14.  Back to cited text no. 3    
4.Fleischer D. Endoscopic Therapy of Upper Gastrointes­tinal Bleeding in Humans. Gastroenterology 1986; 90:217-34.  Back to cited text no. 4  [PUBMED]  
5.Graham DY. Limited Value of Early Endoscopy in the Management of Acute Upper Gastrointestinal Bleeding: prospective Controlled Trial. Am J Surg 1980; 140:284­-90.  Back to cited text no. 5  [PUBMED]  
6.Peterson WL. Routine Early Endoscopy in Upper Gas­trointestinal-Tract Bleeding. A Randomized Controlled Trial. N Engl J Med 1981; 304:925-9.  Back to cited text no. 6    
7.Fleischer D. Etiology and prevalence of Severe persis­tent Upper Gastrointestinal Bleeding. Gastroenterology 1983; 84:538-43.  Back to cited text no. 7  [PUBMED]  
8.Foster DN, Miloszewski KJA, Losowsky MS. Stigmata of Recent Hemorrhage in Diagnosis and Prognosis of Upper Gastrointestinal Bleeding. Br Med J 1978; 1:1173-7.  Back to cited text no. 8    
9.Griffith WJ, Newmann DA, Welsh JD. The Visible Ves­sel As an Indicator of Uncontrolled or Recurrent Gas­trointestinal Hemorrhage. N Engl J Med 1979; 300:1411­-3.  Back to cited text no. 9    
10.Wara P. Endoscopic Predication of Major Rebleeding - A Prospective Study of Stigmata of Hemorrhage in Bleeding Ulcer. Gastroenterology 1985; 88:1209-14.  Back to cited text no. 10  [PUBMED]  
11.Johnston J. The Sentinel Clot/Visible Vessel Revisited. Gastrointest Endosc 1986; 32:238.  Back to cited text no. 11    
12.Freeman ML, Cass OW, Peirre CJ. Oustad GR. The Non-bleeding visible vessel v.s. The sentinel clot: Natu­ral history and risk of rebleeding. Gastrointest Endosc 1993; 39:359-66.  Back to cited text no. 12    
13.Laine L. Freeman M, Cohen H. Lack of Uniformity in Evaluation of Endoscopic Prognostic Features of Bleed­ing Ulcer; Gastrointest Endosc 1994; 40:411-7.  Back to cited text no. 13    
14.Waring JP. Sarrowski RA. A Randomized comparison of multipolar electrocoagulator and injection sclerosis for the treatment of bleeding peptic ulcer. Gastrointest Endosc 1991; 37:295-8.  Back to cited text no. 14    
15.Rutgeerts P, Vantrappen G, Von Hootegam PH. Neodymium-YAG Laser Photocoagulation Versus Mul­tipolar Electrocoagulation for the treatment of severely bleeding ulcers: A randomized comparison. Gastrointest Endosc 1987; 33:199- 202.  Back to cited text no. 15    
16.Mathewson K, Swain CP, Bland M et al. Randomized Comparison of ND-YAG Laser, Heater Probe and No Endoscopic Therapy for Bleeding Peptic Ulcer. Gas­troenterology 1987; 98:1239-44.  Back to cited text no. 16    
17.Laine L. Multipolar Electrocoagulation in the treatment of active upper Gastrointestinal Tract Hemorrhage. N Engl J Med 1987; 316:1613-7.  Back to cited text no. 17  [PUBMED]  
18.Hui WM, M.M.T.; LOK, A.S.F. et al. - A randomized comparative study of laser photocoagulation, heat probe and bipolar electrocoagulation in the treatment of actively bleeding ulcers. Gastrointest Endosc 1991; 37:299-304.  Back to cited text no. 18    
19.Schuman BM. Endoscopic injection therapy for non-variceal upper gastrointestinal hemorrhage - Is it too good to be true? Gastrointest Endosc 1987; 33:121-2.  Back to cited text no. 19  [PUBMED]  
20.Hirao M, Kobayashi T, Masuda K. Endoscopic local injection of hypertonic saline- epinephrine solution to arrest hemorrhage from the upper gastrointestinal tract. Gastrointest Endosc 1985; 31:313-7.  Back to cited text no. 20    
21.Chen PC, Wu CS, C.S, Liaw YF. Hemostatic effect of endoscopic local injection with hypertonic saline epinep­hrine solution and pure ethanol for digestive tract bleed­ing. Gastrointest Endosc 1986; 32:319-23.  Back to cited text no. 21    
22.Leung JWC, Chung SCS. Endoscopic injection of Adrenal in bleeding peptic ulcers. Gastrointestinal Endoscopy 1987;33:73-5.  Back to cited text no. 22    
23.Soehendra N, Grimm H, Stenzel M. Injection of Non­Variceal Bleeding lesions of the Upper Gastrointestinal Tract. Endoscopy 1985; 17:129-32.  Back to cited text no. 23  [PUBMED]  
24.Sugawa C, Fujita Y, Ikeda W. Endoscopic Hemostasis of bleeding of the Upper Gastrointestinal Tract by local injection of Ninety-Eight percent Dehydrated Ethanol Surgery. Gyn and Obstet 1986; 162:159-63.  Back to cited text no. 24    
25.Subei IM, Merdad A. Endoscopic injection therapy for non-variceal upper gastrointestinal bleeding: Effective and Safe. Ann Saudi Med 1993; 13:255-8.  Back to cited text no. 25  [PUBMED]  
26.Chung SCS, Leung JWC, Steel RTC. Epinephrine injec­tion for actively bleeding ulcers: A Randomized Control­led Study. Gastrointest Endosc 1987; 33:146(A).  Back to cited text no. 26    
27.Lin HJ, Perug CL, Lee FY, et al. Endoscopic injection for the arrest of peptic ulcer hemorrhage: Final results of a prospective randomized comparative trial. Gastroin­test Endosc 1993; 39:15-9.  Back to cited text no. 27    
28.Laine L. Injection therapy for bleeding ulcers: Which sol­ution is best. Gastroenterology 1992; 103:1358-9.  Back to cited text no. 28  [PUBMED]  
29.Lai KH, Perug SN, Guo WS. Endoscopic injection for the treatment of bleeding ulcers: local tamponade or drug effect? Endoscopy 1994; 26:339-41.  Back to cited text no. 29    
30.Chung SES, Leung JWC, Leong HT, et al. Adding sclerosant to endoscopic Epinephrine injections in actively bleeding ulcers. A randomized trial. Gastroin­test Endosc 1993; 39:611-5.  Back to cited text no. 30    
31.Villaneous C, Balanzo J, Torras X, et al. The value of second look after injection therapy for bleeding peptic ulcer. Gastrointest Endosc 1994; 40:34-9.  Back to cited text no. 31    
32.Sung JY, Chung SCS, Low JM, et al. Systemic absorp­tion of Epinephrine after endoscopic submucosal injec­tion of patients with bleeding peptic ulcers. Gastrointest Endosc 1993; 39:20-22.  Back to cited text no. 32    
33.Laine L. A new double lumen, direct irrigation injection therapy catheter. Gastrointest Endoscopy 1992; 38:594.  Back to cited text no. 33    
34.Lawson JM. Epinephrine injection of peptic bleeding; is it sclerosant needed for that "special sauce?" Am J Gas­troenterol 1994; 89:948-9.  Back to cited text no. 34    

Correspondence Address:
lyad M Subei
Division of Gastroenterology, Department of Medicine, Dr. Erfan & Bagedo General Hospital, P.O. Box 6519, Jeddah
Saudi Arabia
Login to access the Email id

Source of Support: None, Conflict of Interest: None

PMID: 19864865

Rights and PermissionsRights and Permissions


  [Table - 1]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
    Email Alert *
    Add to My List *
* Registration required (free)  

    Natural History ...
    Role of Endoscop...
    Endoscopic Injec...
    Mechanisms of Ac...
    Clinical Studies...
    Methods and Mate...
    Types of Bleedin...
    Article Tables

 Article Access Statistics
    PDF Downloaded1    
    Comments [Add]    

Recommend this journal