The programs listed are provided for reference purposes only. They were current when produced, but are no longer maintained and may now be outdated. Persons with disabilities having difficulty accessing information on this page may contact us for assistance. Please select the ODP's home page to access current information.
Endoscopy in Upper GI Bleeding

National Institutes of Health
Consensus Development Conference Statement
August 20-22, 1980

Conference artwork, figure of s stomach with an endoscope inserted.

This statement is more than five years old and is provided solely for historical purposes. Due to the cumulative nature of medical research, new knowledge has inevitably accumulated in this subject area in the time since the statement was initially prepared. Thus some of the material is likely to be out of date, and at worst simply wrong. For reliable, current information on this and other health topics, we recommend consulting the National Institutes of Health's MedlinePlus

This statement was originally published as: Endoscopy in Upper GI Bleeding. NIH Consens Statement 1980 Aug 20-22 3(5):1-5.

For making bibliographic reference to the statement in the electronic form displayed here, it is recommended that the following format be used: Endoscopy in Upper GI Bleeding. NIH Consens Statement Online 1980 Aug 20-22 3(5):1-5.


A National Institutes of Health
Consensus Development Conference, held at NIH August 20-22, 1980, addressed the issues that have arisen over the use of endoscopy in the management of upper gastrointestinal (GI) bleeding.

At NIH, consensus development conferences bring together biomedical research scientists, practicing physicians, consumers, and others as needed to provide scientific evaluation of a technology and reach general agreement on the safety and effectiveness of that medical technology. The technology may be a drug, device, or medical or surgical procedure.

Consensus Development Panel and members of the audience considered evidence presented to answer the following questions:

  • What are the benefits of endoscopy in upper GI bleeding?
  • What is the place of other diagnostic approaches to upper GI bleeding?
  • What are the considerations in the decision as to whether to perform endoscopy?
  • What are the risks of endoscopy in the bleeding patient?

The members of the Panel included individuals with many of the disciplinary skills involved in the evaluation and treatment of bleeding in the upper gastrointestinal tract. After hearing formal presentations and discussion, the Panel met to assess the issues based on the evidence heard. The results of the Panel's deliberations are presented in the summary which follows.  


Upper GI bleeding is a common and serious medical problem. It has been estimated that in the United States each year approximately 115 patients per 100,000 population are hospitalized with episodes of upper GI bleeding. These events lead to approximately 250,000 new hospital admissions per year. The mortality from such bleeding, approximately 10%, has not decreased substantially over the past years in spite of improvement in diagnostic accuracy and innovations in therapy.

Upper GI bleeding presents complex problems in management for a number of reasons: (1) it may occur as an unexpected emergency; hence, diagnostic and therapeutic facilities must be proximally located and available on short notice; (2) clinical signs are imperfect guides to the magnitude of hemorrhage, since they neither signal whether bleeding has stopped, nor indicate what level of cardiovascular reserve remains to compensate for any additional blood loss; (3) patients may require supportive therapy to maintain functional blood volume and to ensure adequate perfusion of vital organs; (4) supportive therapy and diagnostic evaluation must proceed simultaneously. (In some patients, therapy for hemorrhage can be definitive, e.g., resection of a bleeding gastric ulcer with antrectomy.)

The first step in the management of upper GI bleeding is evaluation of cardiovascular status, i.e., pulse, blood pressure, orthostatic changes, and other physical signs of hypovolemia (low blood volume). If evidence of hypovolemia is found, volume replacement must be prompt and adequate. By history and physical examination, clues that will lead to likely diagnostic candidates should be sought, i.e., physical evidence of liver disease, history of ingestion of aspirin or other drugs, ingestion of anticoagulants, history of dyspepsia, or previous bleeding episodes.

On the basis of this initial evaluation, an estimate as to the severity of the bleeding episode should be developed as well as a plan for the diagnostic procedures that are most likely to provide useful information at minimal risk and discomfort to the patient. One of these procedures, upper GI endoscopy, was the subject of this consensus conference.

Upper GI endoscopy was chosen for consideration because it is a rapidly evolving technology. Its use has increased rapidly in recent years. Questions have been expressed as to whether this technique is being used either too frequently or not enough. In addition, it should be asked whether use of the procedure incurs an expense in excess of that associated with more traditional management, and if so, whether this cost is warranted by improvement in coping with the problems in upper GI bleeding.

It is generally agreed that in the presence of acute upper GI bleeding expertly performed endoscopy appears, in most instances, to offer the best available method for identifying the site of the bleeding lesion. The technology may not be applicable when bleeding is so massive that immediate surgery appears to provide the only hope of saving the patient's life. This consideration should not preclude use of endoscopy in the operating room to aid the surgeon in locating the lesion. In addition, if bleeding is so rapid as to reduce greatly the chance of visualizing the esophagus and esophago-gastric junction, the technology may not be applicable and the risk of aspiration in these circumstances is significant. Therefore, endoscopy should be delayed until a more appropriate time.

As in all clinical situations, the primary physician caring for the individual patient with upper GI bleeding has an important role in assessing that patient's needs and candidacy for endoscopy. The decision as to what diagnostic or therapeutic steps to take will, of course, be directed mainly by the patient's immediate clinical status, but may often be modified by the immediate availability of various consultants who are expert in certain diagnostic techniques.

The diagnostic accuracy of endoscopy in gastrointestinal bleeding is dependent upon the skill and experience of the endoscopist and the adequacy of facilities, equipment, and supporting personnel. Individuals with minimal training should be discouraged from performing endoscopy, as lack of experience may lead to diagnostic error, complications, and inappropriate decisions regarding patient care. The best interest of the public, in terms of the quality of care related to endoscopy and its attendant cost, is assured by the application of well-defined professional standards for endoscopy.

Although performing upper GI panendoscopy is difficult in the bleeding patient, the well-trained endoscopist can, in the majority of circumstances, locate the bleeding lesion, and in those cases in which there is more than one potentially bleeding lesion present, identify the one responsible for most of the bleeding.

Before choosing any diagnostic or therapeutic course, the physician must weigh the risks and costs of each procedure against its possible benefits.  

What Are the Benefits of Endoscopy in Upper GI Bleeding?

Precise determination of location and type of lesion may be of singular value to the surgeon operating upon the patient with upper GI hemorrhage. Since the nature and timing of the surgical procedure is dictated to a major extent by the specific lesion being treated, preoperative knowledge of the exact lesion enhances placement of the appropriate incision, increases the rapidity and smoothness with which the procedure can be accomplished, and averts the need for unnecessary intraoperative maneuvers which might increase morbidity and mortality.

One of the major benefits of precise endoscopic diagnosis is avoidance of an inappropriate operation on the patient who is bleeding from esophageal varices, or perhaps on patients with diffuse erosive gastritis.

Although it seems likely that such improvement in precision of diagnosis will reduce mortality and morbidity resulting from surgery, this has not yet been clearly demonstrated. Carefully controlled studies in this area are urgently needed.

Although the advent of fiberoptic endoscopy has improved the precision with which an accurate diagnosis of the source of upper GI hemorrhage can be made, there is, as yet, no clear evidence that mortality is significantly altered by such information. It is possible that this situation results from our inability to identify specific subsets of patients with upper GI hemorrhage who would benefit from early and specific diagnosis. In addition, definitive and effective therapy for identifiable lesions is not yet fully developed.

Despite a lack of demonstrable improvement in mortality, other somewhat less obvious benefits may derive from specific diagnosis of the bleeding lesion. The physician can avoid the use of potentially harmful techniques, may discontinue ineffective therapies, or may identify a lesion whose course may be affected by therapy. Counseling of the patient or the family of the patient may be more rational, and a better estimate of long-term prognosis may be possible.

Therapeutic endoscopic methods for control of GI bleeding need to be investigated using large-scale clinical trials. There are several possible methods for coagulating bleeding lesions through endoscopy; these have now reached a suitable state of development, so that the method most likely to be readily and safely adoptable for use by the average endoscopist should now be chosen for randomized comparison with conventional therapy. Uncontrolled use of these modalities should be discouraged until the results of such studies are known.  

What Is the Place of Other Diagnostic Approaches to Upper GI Bleeding?

Contrast Radiography

The single contrast upper GI series has been found in comparison studies to be about one half as accurate as endoscopy in identifying upper GI lesions responsible for bleeding. There is evidence that expertly performed double contrast techniques approach the accuracy of endoscopy. Drawbacks of barium studies include obscuring the field for endoscopy, or angiography; complicating surgery; creating difficulty in obtaining an adequate examination in an uncooperative patient; causing inaccuracy of diagnosis in the presence of retained blood clots or dilution of the barium by large amounts of blood or other material in the stomach; and requiring repeated radiation exposure resulting from the numerous studies frequently used.


Angiography is generally not as accurate as endoscopy, and rarely identifies bleeding unless blood loss is 0.5 cc/min or greater. The clearest indication for angiography is a situation in which intra- arterial therapy is likely to be effective. Drug infusion or embolization may be effective for occlusion of varices and stanching bleeding from other lesions.

Radionuclide Scanning

Radionuclide scanning with technetium sulfurcolloid or tagged red blood cells is a promising non invasive screening technique for the localization of GI bleeding; this requires further evaluation. So far, the use of this method has not been as well demonstrated in patients with upper gastrointestinal bleeding as it has been in subjects with lower tract bleeding.  

What Are The Considerations in the Decision Whether to Perform Endoscopy?

It is important for the physician to assess all relevant clinical factors, including the severity of the hemorrhage, to determine benefits to be expected from having a precise diagnosis for the origin of the bleeding. The utility of this diagnostic information should be weighed against the potential risks of performing endoscopy. While endoscopy is usually the best procedure for identifying the bleeding lesion, it need not automatically be performed in all instances of upper GI bleeding, nor is it required following each episode of repeated bleeding in an individual patient. Differences in the expertise of the locally available specialists in endoscopy and radiography should also be considered in choosing the diagnostic approach. An uncooperative patient is a poor candidate not only for endoscopy but also for most other diagnostic procedures.

There are clinical settings in which bleeding is minor and apparently self-limited, and in which careful prolonged observation is the best course to follow.

Continuous or recurrent bleeding during hospitalization in the absence of a clear-cut diagnosis is usually a relatively strong indication for endoscopy if it has not already been done. Such rebleeding in a patient previously endoscoped but without firm diagnosis should be cause to consider repeating endoscopy.

Endoscopy may be indicated when there is doubt as to the preferred medical or surgical therapy and when improved diagnostic information would be expected to clarify this issue. For example, in patients known to have cirrhosis of the liver with esophageal varices, bleeding episodes are often related to factors other than the varices, such as peptic ulcers or erosive gastritis. It is important to ascertain the exact cause of bleeding, since the therapeutic approach to bleeding varices is completely different from that used to treat other causes of bleeding.

It is very important to identify bleeding from esophageal varices related to extra-hepatic portal hypertension, since the beneficial effects of decompressive surgery are so much more definitive and long-lasting than they are in hepatic cirrhosis. Unexplained upper GI bleeding in the presence of splenomegaly or other physical findings suggesting portal hypertension would prompt a search for varices.  

What Are The Risks of Endoscopy in the Bleeding Patient?

Although complications are generally rare, they are more common in patients with actual or potential heart, lung, renal, and liver diseases and in patients on immunosuppressive regimens. Patients with any of those conditions must be carefully evaluated beforehand and observed during the endoscopic procedure for hypoxemia, arrhythmia, hypotension, or other untoward events. Resuscitation facilities must be in place where the procedure is performed.

Complications from endoscopy in gastrointestinal hemorrhage when done in an emergency are increased over those to be expected when the procedure is done electively. Aspiration is more common and more serious when endoscopy is performed in the presence of active upper GI bleeding. The most common complications of endoscopy, regardless of the indication for its use, are cardiopulmonary and result from the procedure itself or from premedication. Thrombo-phlebitis may occur following intravenous medication. Perforations, although rare, occur particularly in the proximal esophagus in the area of the cricopharyngeus. Perforations also occur through Zenker's diverticula and through areas of tumor during or following endoscopic manipulation. Cervical spondylosis sometimes makes endoscopy difficult. Other complications include aggravated bleeding, particularly if a biopsy is performed, and sepsis.

The transmission of infection by endoscopy is extremely rare and is avoidable by appropriate mechanical cleansing and disinfection.


Endoscopy is an excellent tool for the differential diagnosis of upper GI bleeding. The lack of demonstrated effect on overall mortality, however, strongly suggests that the major current task is to use the diagnostic information gleaned by endoscopy as a stimulus for further vigorous investigations of different and newer therapies for the different lesions causing upper gastrointestinal bleeding.

There is need for a standardized coding system for diagnosis and services rendered so that adequate data for planning resource commitment to the problem of gastrointestinal hemorrhage can be made on a rational basis. Physicians should specify upper GI bleeding as the reason for hospitalization, if that be the case, in addition to any other indicated diagnoses.

Financial Implications

In spite of the increased diagnostic accuracy of endoscopy, its high cost is an impediment to abandoning contrast radiography for the conventional investigation of acute upper GI hemorrhage. This is especially true since the increased diagnostic information obtained from endoscopy has not resulted in definitively improving mortality. Charges for GI endoscopy vary considerably across the country. The Panel urges those doing endoscopy to closely examine their costs and fees to help cost containment for this procedure.  

Consensus Development Panel

Sherman M. Mellinkoff, M.D. (Chairman)
University of California at Los Angeles School of Medicine
Los Angeles, California
William R. Best, M.D.
University of Illinois Hospital
Chicago, Illinois
Wilma Diner, M.D.
University of Arkansas for Medical Sciences
Little Rock, Arkansas
Faith Fitzgerald, M.D.
University of Michigan
Ann Arbor, Michigan
James A. Fry, M.D.
Western Clinic
Tacoma, Washington
Thomas R. Hendrix, M.D.
The Johns Hopkins University School of Medicine
Baltimore, Maryland
Benjamin M. Kagan, M.D.
Cedars Sinai Medical Center/UCLA
Los Angeles, California
Louis Lazar, M.D.
Millard Fillmore Hospital
Buffalo, New York
Paul Sherlock, M.D.
Cornell University Medical College and Memorial
Sloan Kettering Cancer Center
New York, New York
William Silen, M.D.
Harvard Medical School and Beth Israel Hospital
Boston, Massachusetts


Abass Alavi, M.D.
"Utility of Radionuclide Scanning"
Chief, Division of Nuclear Medicine
Associate Professor of Radiology
Hospital of the University of Pennsylvania
Philadelphia, Pennsylvania
Marvin E. Ament, M.D.
"Risk of Upper Endoscopy for Bleeding in Children"
Professor of Pediatrics
Chief, Division of Pediatric Gastroenterology
Department of Pediatrics
University of California School of Medicine
Los Angeles, California
Stuart Danovitch, M.D.
"Indications and Contraindications to Endoscopy"
Professor of Medicine
The George Washington University
Washington, D.C.
Gregory L. Eastwood, M.D.
"Does the Patient with Upper Gastrointestinal Bleeding Benefit form Endoscopy: Reflections and Discussion of Recent Literature"
Director, Division of Gastroenterology
University of Massachusetts Medical School
Worcester, Massachusetts
Harvey V. Fineberg, M.D., Ph.D.
"A Model for Cost-Effectiveness Analysis of Endoscopy in Upper Gastrointestinal Bleeding"
Harvard School of Public Health
Center for the Analysis of Health Practices
Boston, Massachusetts
David A. Gilbert, M.D.
"ASGE Prospective Study: Risks of Endoscopy in Upper Gastrointestinal Bleeding"
Assistant Professor of Medicine
Department of Medicine
University of Washington
Seattle, Washington
Peter B. Gregory, M.D.
"Clinical Assessment"
Associate Professor of Clinical Medicine
Department of Medicine
Division of Gastroenterology
Stanford School of Medicine
Stanford, California
Joyce Gryboski, M.D.
"Benefits of Endoscopy in Children with Upper Gastrointestinal Bleeding"
Professor of Pediatrics
Department of Pediatrics
Yale University School of Medicine
New Haven, Connecticut
Hans Herlinger, M.D., F.R.C.R.
"Utility of Contrast Radiology"
Associate Professor of Radiology
Hospital of the University of Pennsylvania
Philadelphia, Pennsylvania
Paul H. Jordan, Jr., M.D.
"Surgical Perspectives on Diagnosis in Upper Gastrointestinal Bleeding"
Professor of Surgery
Baylor College of Medicine
Texas Medical Center
Houston, Texas
Ronald M. Katon, M.D.
"Risks of Endoscopy Including Medications; Patient Monitoring: Review of the Literature"
Associate Professor of Medicine
Director of Gastroenterology
University of Oregon Health Sciences Center
Portland, Oregon
Fredrick S. Keller, M.D.
"Utility of Angiography in the Diagnosis and Treatment of Upper Gastrointestinal Bleeding"
Assistant Professor of Cardiovascular Radiology
University of Oregon Health Sciences Center
Portland, Oregon
Raymond S. Koff, M.D.
"Benefits of Endoscopy in Upper GI Bleeding in Gastrointestinal Patients with Liver Disease"
Professor of Medicine
Chief, Hepatology Section
Boston University School of Medicine
Chief, Hepatology Section
Veterans Administration Medical Center
Boston, Massachusetts
Philip Kramer, M.D.
"What are the Benefits of Endoscopy in Upper Gastrointestinal Bleeding"
Professor of Medicine
Boston University Medical Center
Boston, Massachusetts
Jeffrey L. Lichtenstein, M.D.
"Accuracy of Endoscopy in Upper GI Bleeding"
Director of Medical Education/Medical Staff
Meriden-Wallingford Hospital
Meriden, Connecticut
Albert I. Mendeloff, M.D.
"Nature and Magnitude of the Problem of Upper Gastrointestinal Bleeding"
Professor of Medicine
Johns Hopkins University School of Medicine
Sinai Hospital of Baltimore
Baltimore, Maryland
John F. Morrissey, M.D.
"Diagnostic Endoscopy in Patients with Upper Gastrointestinal Bleeding"
Professor and Vice Chairman
Department of Medicine
University of Wisconsin Medical School
Clinical Science Center
Madison, Wisconsin
Bergein F. Overholt, M.D.
"The Costs of Endoscopy and Alternatives: The Current State of Knowledge"
Gastrointestinal Associates, P.C.
Knoxville, Tennessee
John P. Papp, M.D.
"Electrocoagulation in Upper Gastrointestinal Bleeding"
Associate Clinical Professor of Medicine
Michigan State University
School of Human Medicine and Blodgett Memorial
Medical Center
Grand Rapids, Michigan
Walter L. Peterson, M.D.
"Experiences in Conducting a Controlled Trial of Diagnostic Endoscopy in Upper Gastrointestinal Bleeding"
Assistant Professor of Internal Medicine
University of Texas Southwestern Medical School
Assistant Chief of Gastroenterology
Dallas Veterans Administration Medical Center
Dallas, Texas
Theodore R. Schrock, M.D.
"Does Endoscopy Affect the Surgical Approach to the Patient with Upper Gastrointestinal Bleeding"
Associate Professor of Surgery
University of California Medical Center
San Francisco, California
Fred E. Silverstein, M.D.
"Experimental Work and Clinical Data on Photocoagulation and Electrocoagulation"
Associate Professor of Medicine
University of Washington
Seattle, Washington
Stephen E. Silvis, M.D.
"What are the Risks of Endoscopy in the Bleeding Patient"
Chief, Special Diagnostic and Treatment Unit
Veterans Administration Medical Center
Minneapolis, Minnesota
Howard M. Spiro, M.D.
"Other Diagnostic Approaches to Upper Gastrointestinal Bleeding"
Professor of Medicine
Yale University
New Haven, Connecticut
Francis J. Tedesco, M.D.
"Experiences in the Design and Implementation of a Therapeutic Trial of Endoscopy in Upper Gastrointestinal Bleeding"
Department of Medicine
Medical College of Georgia
Augusta, Georgia
Jack A. Vennes, M.D.
"The Risk of Transmitting Infection and Methods of Instrument Cleansing"
Professor of Medicine
University of Minnesota
Staff Physician
Veterans Administration Medical Center
Minneapolis, Minnesota

Conference Sponsors

National Institute of Arthritis, Metabolism, and Digestive Diseases
Harold P. Roth, M.D.

National Center for Health Care Technology

Office of Medical Applications of Research

Privacy Notice | Disclaimer | Accessibility | Contact NIH
Department of Health and Human Services HomepageNational Institutes of Health Home PageFirstGov Home Page - The U.S. Government's Official Web Portal  National Guideline ClearingHouse - a public resource for evidence-based clinical practice guidelines