Version July 2025
INTRODUCTION —
Patients with acute upper gastrointestinal (GI) bleeding commonly present with hematemesis (vomiting of blood or coffee-ground-like material) and/or melena (black, tarry stools), though patients with large-volume upper GI bleeding may also present with hematochezia (red or maroon blood with stool). The initial evaluation of patients with acute upper GI bleeding involves an assessment of hemodynamic stability and resuscitation if necessary. Diagnostic studies (usually endoscopy) follow, with the goals of diagnosis, and when possible, treatment of the specific disorder.
The diagnostic and initial therapeutic approach to patients with acute upper GI bleeding will be reviewed here. While there is variability among guidelines, this approach is generally consistent with a multidisciplinary international consensus statement updated in 2019, a 2012 guideline issued by the American Society for Gastrointestinal Endoscopy, a 2021 guideline issued by the American College of Gastroenterology, a 2015 guideline issued by the European Society of Gastrointestinal Endoscopy, and a 2021 update issued by the European Society of Gastrointestinal Endoscopy [1-5]. The causes of upper GI bleeding, the endoscopic management of acute upper GI bleeding, and the management of active variceal hemorrhage are discussed separately. (See "Causes of upper gastrointestinal bleeding in adults" and "Overview of the treatment of bleeding peptic ulcers" and "Overview of the management of patients with variceal bleeding" and "Methods to achieve hemostasis in patients with acute variceal hemorrhage".)
A table outlining the major causes, clinical features, and emergency management of acute severe upper gastrointestinal bleeding in adults is provided (table 1).
INITIAL EVALUATION —
The initial evaluation of a patient with acute upper gastrointestinal bleeding includes a history, physical examination, and laboratory tests. The goal of the evaluation is to assess the severity of the bleeding, identify potential sources of the bleeding, and determine if there are conditions present that may affect subsequent management. The information gathered as part of the initial evaluation is used to guide decisions regarding triage, resuscitation, empiric medical therapy, and diagnostic testing.
Factors that are predictive of bleeding coming from an upper GI source identified in a meta-analysis included a patient-reported history of melena (likelihood ratio [LR] 5.1-5.9), melenic stool on examination (LR 25), blood or coffee grounds detected during nasogastric lavage (LR 9.6), and a ratio of blood urea nitrogen to serum creatinine greater than 30 (LR 7.5) [6]. On the other hand, the presence of blood clots in the stool made an upper GI source less likely (LR 0.05). Factors associated with severe bleeding included red blood detected during nasogastric lavage (LR 3.1), tachycardia (LR 4.9), or a hemoglobin level of less than 8 g/dL (LR 4.5-6.2).
Bleeding manifestations — Hematemesis (either red blood or coffee-ground emesis) suggests bleeding proximal to the ligament of Treitz. Traditionally, it has been taught that frankly bloody emesis suggests moderate to severe bleeding that may be ongoing, whereas coffee-ground emesis suggests more limited bleeding. However, at least one study suggests that these findings may not accurately predict the severity of bleeding [7].
The majority of melena (black, tarry stool) originates proximal to the ligament of Treitz (90 percent), though it may also originate from the oropharynx or nasopharynx, small bowel, or colon [8]. Melena may be seen with variable degrees of blood loss, being seen with as little as 50 mL of blood [9]. It takes at least 14 hours for bleeding to manifest as melena, and melena may persist for up to five days after bleeding has stopped [10,11].
Hematochezia (red or maroon blood in the stool) is usually due to lower GI bleeding. However, it can occur with massive upper GI bleeding [12], which is typically associated with evidence of hemodynamic instability. (See 'Physical examination' below.)
Past medical history — Patients should be asked about prior episodes of upper GI bleeding, since up to 60 percent of patients with a history of an upper GI bleed are bleeding from the same lesion [13]. In addition, the patient's past medical history should be reviewed to identify important comorbid conditions that may lead to upper GI bleeding or may influence the patient's subsequent management.
Potential bleeding sources suggested by a patient's past medical history include:
●Peptic ulcer disease
•History of Helicobacter pylori (H. pylori) infection
•Nonsteroidal anti-inflammatory drug (NSAID) use
•Antithrombotic use
•Smoking
●Varices or portal hypertensive gastropathy
•History of liver disease
•Excess alcohol use
●Gastrointestinal angiodysplasia
•Kidney disease
•Aortic stenosis
•Hereditary hemorrhagic telangiectasia
●Aorto-enteric fistula
•Abdominal aortic aneurysm
•Aortic graft
●Malignancy
•History of smoking
•Excess alcohol use
•H. pylori infection
●Marginal ulcer (ulcer at an anastomotic site)
•Gastroenteric anastomosis
Comorbid illnesses may influence patient management in the setting of an acute upper GI bleed. Comorbid illnesses may:
●Make patients more susceptible to adverse effects of anemia (eg, coronary artery disease, pulmonary disease). Such patients may need to be maintained at higher hemoglobin levels than patients without these disorders. (See 'Blood product transfusions' below.)
●Predispose patients to volume overload in the setting of vigorous fluid resuscitation or blood transfusions (eg, kidney disease, heart failure). Such patients may need more invasive monitoring during resuscitation. (See 'General support' below.)
●Result in bleeding that is more difficult to control (eg, coagulopathies, thrombocytopenia, significant hepatic dysfunction). Such patients may need additional hemostatic therapies. (See 'Blood product transfusions' below.)
●Predispose to aspiration of GI contents into the lungs (eg, dementia, hepatic encephalopathy). Endotracheal intubation should be considered in such patients. (See 'General support' below.)
Medication history — A thorough medication history should be obtained, with particular attention paid to drugs that:
●Predispose to peptic ulcer formation, such as aspirin and other NSAIDs, including COX-2 selective NSAIDs (see "NSAIDs (including aspirin): Pathogenesis and risk factors for gastroduodenal toxicity").
●Are associated with pill esophagitis (see "Pill esophagitis").
●Increase the risk of bleeding, such as anticoagulants (including warfarin and the direct oral anticoagulants) and antiplatelet agents (eg, P2Y12 inhibitors and aspirin).
●Have been associated with GI bleeding, including selective serotonin reuptake inhibitors (SSRI), calcium channel blockers, and aldosterone antagonists.
●May alter the clinical presentation, such as bismuth, charcoal, licorice, and iron, which can turn the stool black.
Symptom assessment — Patients should be asked about symptoms as part of the assessment of the severity of the bleed and as a part of the evaluation for potential bleeding sources. Symptoms that suggest the bleeding is severe include orthostatic lightheadedness, confusion, angina, severe palpitations, and cold/clammy extremities.
Specific causes of upper GI bleeding may be suggested by the patient's symptoms [8]:
●Peptic ulcer – Upper abdominal pain
●Esophageal ulcer – Odynophagia, gastroesophageal reflux, dysphagia
●Variceal hemorrhage or portal hypertensive gastropathy: jaundice, abdominal distention (ascites)
●Mallory-Weiss tear – Emesis, retching, or coughing prior to hematemesis
●Malignancy – Dysphagia, early satiety, unintentional weight loss, cachexia
Physical examination — The physical examination is a key component of the assessment of hemodynamic stability. Signs of hypovolemia include [8]:
●Mild to moderate hypovolemia (less than 15 percent of blood volume lost) – Resting tachycardia.
●Blood volume loss of at least 15 percent – Orthostatic hypotension (a decrease in the systolic blood pressure of more than 20 mmHg and/or an increase in heart rate of 20 beats per minute when moving from recumbency to standing).
●Blood volume loss of at least 40 percent – Supine hypotension.
Examination of the stool color may provide a clue to the location of the bleeding, but it is not a reliable indicator. In a series of 80 patients with severe hematochezia (red or maroon blood in the stool), 74 percent had a colonic lesion, 11 percent had an upper GI lesion, 9 percent had a presumed small bowel source, and no site was identified in 6 percent [12].
The presence of abdominal pain, especially if severe and associated with rebound tenderness or involuntary guarding, raises concern for perforation. If any signs of an acute abdomen are present, further evaluation to exclude a perforation is required prior to endoscopy. (See "Overview of gastrointestinal tract perforation", section on 'Diagnosis'.)
Finally, as with the past medical history, the physical examination should include a search for evidence of significant comorbid illnesses. (See 'Past medical history' above.)
Laboratory data — Laboratory tests that should be obtained in patients with acute upper gastrointestinal bleeding include a complete blood count, serum chemistries, liver tests, and coagulation studies. In addition, serial electrocardiograms and cardiac enzymes may be indicated in patients who are at risk for a myocardial infarction, such as older adults, patients with a history of coronary artery disease, or patients with symptoms such as chest pain or dyspnea. (See "Diagnosis of acute myocardial infarction".)
The initial hemoglobin level in patients with acute upper GI bleeding may be at the patient's baseline because the patient is losing whole blood. With time, the hemoglobin level will decline as the blood is diluted by the influx of extravascular fluid into the vascular space and by fluid administered during resuscitation. The hemoglobin level should initially be monitored every two to twelve hours. The frequency of the assessment will depend on factors such as whether there is hemodynamic instability (more frequent assessment).
Acute bleeding does not alter the mean corpuscular volume (MCV). If the MCV is low, it suggests possible iron deficiency, which could be caused by chronic bleeding. Anemia or other abnormalities on the CBC that persist after recovery from the acute bleeding event should be evaluated. (See "Diagnostic approach to anemia in adults".)
Because blood is absorbed as it passes through the small bowel and patients may have decreased kidney perfusion, patients with acute upper GI bleeding typically have an elevated blood urea nitrogen (BUN)-to-creatinine or urea-to-creatinine ratio. Values >30:1 or >100:1, respectively, suggest upper GI bleeding as the cause [6,14-16]. The higher the ratio, the more likely the bleeding is from an upper GI source [14].
Nasogastric lavage — The routine use of nasogastric tube (NGT) placement in patients with suspected acute upper GI bleeding is not recommended, as studies have failed to demonstrate a benefit with regard to clinical outcomes [5,17,18]. We instead suggest that erythromycin be used before endoscopy to help clear the stomach of particulate matter, fresh blood, or clots. (See 'Prokinetics' below.)
A retrospective study looked at whether there were clinical benefits from NGT lavage in 632 patients admitted with gastrointestinal bleeding [19]. Patients who underwent NGT lavage were matched with patients with similar characteristics who did not undergo NGT lavage. NGT lavage was associated with a shorter time to endoscopy. However, there were no differences between those who underwent NGT lavage and those who did not with regard to mortality, length of hospital stay, surgery, or transfusion requirement. Similarly, in a randomized trial with 280 patients with upper GI bleeding, there were no differences in rebleeding rates or mortality between patients who underwent NGT lavage and those who did not [20].
GENERAL MANAGEMENT
Hemodynamically unstable patients — While the principles behind the management of all patients with upper gastrointestinal bleeding are similar, there are some special considerations when it comes to patients presenting with hemodynamic instability (shock, orthostatic hypotension, tachycardia) (table 1).
Intravenous access — For patients who are hemodynamically unstable, two 16 gauge intravenous catheters and/or a large-bore, single-lumen central cordis should be placed.
Fluid resuscitation — Fluid resuscitation should begin immediately and should not be delayed pending transfer of the patient to an intensive care unit. The approach to fluid resuscitation in patients who are hemodynamically unstable is discussed in detail elsewhere. (See "Evaluation of and initial approach to the adult patient with undifferentiated hypotension and shock", section on 'Hemodynamic support'.)
Transfusion — For patients with active/brisk bleeding and hypovolemia, transfusion should be guided by hemodynamic parameters (eg, pulse and blood pressure), the pace of the bleeding, estimated blood loss, and the ability to stop the bleeding, rather than by serial hemoglobin measurements. If the initial hemoglobin level is low transfusions should be initiated [21,22]. The specific threshold for transfusion will vary based on factors such as whether the patient has coronary artery disease (algorithm 1). In an acutely hemorrhaging patient, however, transfusion support should not be delayed while awaiting laboratory test results. (See "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Acute bleeding' and 'Blood product transfusions' below.)
Medications and endoscopy — The approach to medications (eg, proton pump inhibitors) and endoscopy are similar for patients with hemodynamic instability compared with patients who are hemodynamically stable. It is particularly important to ensure that these patients are adequately resuscitated prior to undergoing upper endoscopy. (See 'Medications' below and 'Upper endoscopy' below.)
Triage — All patients with hemodynamic instability or active bleeding (manifested by ongoing hematemesis, bright red blood per nasogastric tube, or hematochezia) should be admitted to an intensive care unit for resuscitation and close observation with automated blood pressure monitoring, electrocardiographic monitoring, and pulse oximetry.
A table outlining the emergency management of acute severe upper gastrointestinal bleeding is provided (table 1).
Other patients can be admitted to a regular medical ward, though we suggest that all admitted patients, with the exception of low-risk patients, receive electrocardiographic monitoring. Outpatient management may be appropriate for some low-risk patients. Determining the appropriate site of care for a patient can be facilitated using risk stratification scores, such as the Glasgow-Blatchford score. Use of these scores is recommended in the International Consensus Group, United States, and European guidelines [1]. (See 'Risk stratification' below.)
General support — Patients should receive supplemental oxygen by nasal cannula and should receive nothing per mouth. Two large caliber (18 gauge or larger) peripheral intravenous catheters or a central venous line should be inserted. For patients who are hemodynamically unstable, two 16 gauge intravenous catheters and/or a large-bore, single-lumen central cordis should be placed.
We reserve endotracheal intubation for patients deemed high-risk for aspiration, including those with massive upper gastrointestinal (GI) bleeding or altered mental status. Elective endotracheal intubation in patients with ongoing hematemesis or altered respiratory or mental status may facilitate endoscopy and decrease the risk of aspiration. However, among patients who are critically ill, elective endotracheal intubation has been associated with worse outcomes.
A case control study with 200 patients with upper GI bleeding who were critically ill found that patients who had elective endotracheal intubation were more likely than patients who were not intubated to have adverse cardiopulmonary outcomes based on a composite outcome that included pneumonia, pulmonary edema, acute respiratory distress syndrome, and cardiac arrest [23]. Of note, the presence of respiratory distress prior to intubation was not reported and the mean Glasgow Coma Scale score was 14.7 (+/- 0.95), indicating that altered mental status was absent in the majority of patients. Patients who were electively intubated were more likely to suffer cardiopulmonary complications compared with patients who were not intubated (20.0 versus 6.0 percent). In particular, patients who were intubated were more likely to be diagnosed with pneumonia within 48 hours (14.0 versus 2.0 percent).
Fluid resuscitation — Adequate resuscitation and hemodynamic stabilization is essential prior to endoscopy to minimize treatment-associated complications [24]. Patients with active bleeding should receive intravenous fluids (eg, 500 mL of normal saline or lactated Ringer's solution over 30 minutes) while being typed and cross-matched for blood transfusion. The rate of fluid resuscitation will in part depend on whether the patient is hemodynamically unstable. Patients at risk of fluid overload may require intensive monitoring.
If the blood pressure fails to respond to initial resuscitation efforts, the rate of fluid administration should be increased. In some patients, temporary support with vasopressor drugs may be required. (See "Evaluation of and initial approach to the adult patient with undifferentiated hypotension and shock", section on 'Hemodynamic support' and "Treatment of severe hypovolemia or hypovolemic shock in adults", section on 'Initial rate of fluid repletion'.)
Blood product transfusions
Anemia
General approach — The decision to initiate blood transfusion must be individualized (algorithm 1). For most hemodynamically stable patients, our approach is to initiate blood transfusion if the hemoglobin is <7 g/dL (<70-g/L) and to maintain the hemoglobin at a level ≥7 g/dL (70 g/L), rather than at a higher level [1,25-28]. Higher levels may be appropriate for patients at increased risk of adverse events in the setting of significant anemia, such as those with coronary artery disease or in those with evidence of ongoing active bleeding. (See "Overview of the acute management of non-ST-elevation acute coronary syndromes", section on 'Anemia'.)
Using a restrictive transfusion strategy for most hemodynamically stable patients was supported by a meta-analysis of five randomized trials with a total of 1965 patients with acute upper gastrointestinal bleeding [21]. Patients assigned to a restrictive transfusion strategy had a lower all-cause mortality than those assigned to a liberal transfusion strategy (absolute risk reduction [ARR] 2.2 percent, relative risk [RR] 0.65, 95% CI 0.44-0.97) and rebleeding (ARR 4.4 percent, RR 0.58, 95% CI 0.40-0.84). While two of the studies used a cutoff of 7 g/dL (70 g/L) for a restrictive transfusion strategy and three used a cutoff of 8 g/dL (80 g/L), all of the studies favored a restrictive transfusion strategy. There were no differences between patients with cirrhosis and those with non-variceal bleeding. On subgroup analysis, there were non-statistically significant trends toward a higher risk of mortality with a restrictive transfusion strategy among patients with ischemic heart disease (RR 4.4, 95% CI 0.27-22) and a lower risk of mortality among patients without ischemic heart disease (RR 0.58, 95% CI 0.86-1.3). Rebleeding risk was similar between those with and without ischemic heart disease (RR 0.50 and 0.69, respectively).
The meta-analysis did not detect a difference between a restrictive and a liberal transfusion strategy in the risk of myocardial infarction (RR 0.79, 95% CI 0.33-1.89), stroke (RR 0.49, 95% CI 0.12-2.01), or acute kidney injury (RR 0.77, 0.56-1.05), but not all of the included trials reported these outcomes.
Active bleeding and hypovolemia — For patients with active/brisk bleeding and hypovolemia, decisions about transfusion are guided by hemodynamic parameters (eg, pulse and blood pressure), the pace of the bleeding, estimated blood loss, and the ability to stop the bleeding, rather than by serial hemoglobin measurements. Patients who require massive transfusion (defined by institutional protocols, often >3 units of RBCs in an hour or 10 units of RBCs in 24 hours) may also need replacement of coagulation factors and/or platelets. (See "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Acute bleeding' and "Massive blood transfusion", section on 'Approach to volume and blood replacement'.)
Thrombocytopenia — Patients with critical or life-threatening bleeding and a low platelet count (eg, <50,000/microL) should be transfused with platelets. Limited data suggest that proceeding with upper endoscopy in patients with thrombocytopenia is generally safe [29], though whether there is a lower limit below which endoscopy should be delayed is unclear [30]. Our approach is to perform an upper endoscopy if the platelet count is >20,000/microL, though if the patient is suspected to have active bleeding, we attempt to raise the platelet count to >50,000/microL prior to endoscopy.
In the past, platelet transfusions were considered in non-thrombocytopenic or mildly thrombocytopenic patients with life-threatening bleeding who had been taking antiplatelet agents such as aspirin or clopidogrel [31]. However, high-quality evidence regarding the benefit of platelet transfusion is lacking, and some evidence suggests that platelet transfusion may be deleterious [32]. Because these cases can be complex, an individualized approach based on the complete clinical picture is required. Potential adverse effects of platelet transfusion are discussed separately. (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'Complications'.)
Managing anticoagulants, antiplatelet agents, and coagulopathies
Anticoagulants and antiplatelet agents — The approach to management of anticoagulants and antiplatelet agents depends on the medications being used and their indications, how severe the bleeding is, and how quickly reversal of anticoagulation is needed. The medications and products that may be used to reverse anticoagulation are discussed in the tables and separate topic reviews:
●Warfarin (table 2) (see "Management of warfarin-associated bleeding or supratherapeutic INR", section on 'Treatment of bleeding')
●Direct oral anticoagulants (DOACs) (table 3) (see "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Major bleeding')
●Heparins (see "Heparin and LMW heparin: Dosing and adverse effects", section on 'Reversal')
For most patients, endoscopy should not be delayed because of anticoagulant or antiplatelet agent use [1]. Provided the patient is hemodynamically stable, urgent endoscopy can usually proceed simultaneously with management of antithrombotic medications. However, for patients undergoing upper endoscopy, we wait until the INR is <2.5 to perform the endoscopy, if possible [33]. This approach is based on data that suggest endoscopy is safe and endoscopic therapy effective in patients who are mildly to moderately anticoagulated [34].
If a patient is taking antiplatelet monotherapy such as aspirin for secondary cardiovascular prevention (prior acute coronary syndrome, coronary artery stent), the agent may be continued [35]. If the patient is taking dual antiplatelet therapy, then one of the antiplatelet agents is generally discontinued [5]. Meta-analyses of randomized trials suggest that patients may continue just monotherapy with a P2Y12 inhibitor if acute coronary syndrome or stent placement was >1 to 3 months earlier [36]. In patients taking low-dose aspirin for primary cardiovascular prophylaxis aspirin should be permanently discontinued. In complicated cases, consultation with the provider who prescribed the antiplatelet medication should be considered. (See "Management of anticoagulants in patients undergoing endoscopic procedures", section on 'Urgent procedures' and "Management of antiplatelet agents in patients undergoing endoscopic procedures" and "Gastrointestinal endoscopy in patients with disorders of hemostasis".)
When possible, anticoagulants should be held in patients with acute upper GI bleeding. In patients with severe, ongoing bleeding who are taking an anticoagulant, administration of a reversal agent or intravenous prothrombin complex concentrate may be indicated. However, the thrombotic risk of reversing anticoagulation should be weighed against the risk of continued bleeding without reversal, and thus the decision to discontinue medications or administer reversal agents needs to be individualized.
When to resume these medications once hemostasis has been achieved will depend on the patient's risks for thrombosis and recurrent bleeding. (See "Management of anticoagulants in patients undergoing endoscopic procedures", section on 'Resuming anticoagulants after hemostasis' and "Overview of the treatment of bleeding peptic ulcers", section on 'Risk factors for persistent or recurrent bleeding'.)
Coagulopathies related to cirrhosis — The management of coagulopathies in patients with cirrhosis is particularly complicated. In patients with cirrhosis, the INR only reflects changes in procoagulant factors, but both procoagulant and anticoagulant factors are reduced, and thus INR is not a reliable indicator of bleeding risk or coagulation status. These issues are discussed separately. (See "Hemostatic abnormalities in patients with liver disease", section on 'Bleeding'.)
Other bleeding disorders — For individuals with other bleeding disorders, consultation with hematology (the patient's primary hematologist if possible) is prudent. They can advise regarding specific products, dosing, and monitoring based on the individual's specific disorder and medical history.
Dilutional coagulopathy — Patients who require massive transfusion (defined by institutional protocols, often >3 units RBCs in an hour or 10 units RBCs in 24 hours) may also need replacement of coagulation factors and/or platelets. (See "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Acute bleeding' and "Massive blood transfusion", section on 'Approach to volume and blood replacement'.)
Medications
Acid suppression — Patients admitted to the hospital with acute upper GI bleeding are typically treated with a proton pump inhibitor (PPI). Guidelines vary with regard to recommending for or against use of pre-endoscopic PPIs [1-5]. The optimal approach to PPI administration prior to endoscopy is unclear. Options include giving an IV PPI every 12 hours or starting a continuous infusion.
Our approach is to give a high-dose bolus (eg, esomeprazole 80 mg) to patients with signs of active bleeding (eg, hematemesis, hemodynamic instability). If endoscopy has not been performed after 12 hours, we give a second dose of an IV PPI 12 hours after the first dose was given (eg, esomeprazole 40 mg). Subsequent dosing will depend on the endoscopic findings. Oral formulations (eg, esomeprazole 40 mg orally twice daily) are a reasonable alternative if IV formulations are not available. Pantoprazole and esomeprazole are the only intravenous formulations available in the United States, and intravenous lansoprazole has been removed from the world market.
The role of acid suppression, including administration options and studies of its efficacy, is discussed elsewhere.(See "Overview of the treatment of bleeding peptic ulcers", section on 'Acid suppression'.)
Prokinetics — Both erythromycin and metoclopramide have been studied in patients with acute upper GI bleeding [37-46]. The goal of using a prokinetic agent is to improve gastric visualization at the time of endoscopy by clearing the stomach of blood, clots, and food residue. We suggest that erythromycin be used before endoscopy. A reasonable dose is 250 mg intravenously over 10 to 30 minutes. Endoscopy is performed 20 to 90 minutes following completion of the erythromycin infusion. Patients receiving erythromycin need to be monitored for QTc prolongation. In addition, drug-drug interactions should be evaluated before giving erythromycin because it is a cytochrome P450 3A inhibitor (table 4).
Erythromycin promotes gastric emptying based upon its ability to be an agonist of motilin receptors. Using erythromycin to improve gastric visualization has been studied in several randomized controlled trials and meta-analyses [37-45]. The randomized trials were included in a 2016 meta-analysis that examined the role of pre-endoscopic erythromycin [44]. The meta-analysis included eight randomized trials with 598 patients with upper gastrointestinal bleeding and compared patients who received erythromycin with those who did not. Patients who received erythromycin were more likely to have adequate gastric visualization (77 versus 51 percent, odds ratio [OR] 4.14; 95% CI 2.01-8.53), were less likely to require second-look endoscopy (15 versus 26 percent, OR 0.51; 95% CI 0.34-0.77), and had shorter hospital stays (mean difference -1.75 days, 95% CI -2.43 to -1.06). There were no differences in units of blood transfused, endoscopy duration, or need for emergent surgery between those who received erythromycin and those who did not. A second meta-analysis also found better visualization with the use of erythromycin [45]. In two trials with 195 patients, patients who received erythromycin scored higher than patients who received placebo on a 16-point ordinal scale, with higher scores indicating better visualization (mean difference 3.63 points, 95% CI 2.20- 5.05).
Some trials have compared pre-endoscopy erythromycin with nasogastric lavage. In one trial, 253 patients were assigned to receive erythromycin alone, nasogastric lavage alone, or nasogastric lavage plus erythromycin. It found that the quality of visualization did not differ significantly among the three groups [41]. In addition, there were no differences among the groups with regard to procedure duration, rebleeding rates, need for second endoscopy, number of transfused units of blood, and mortality. A meta-analysis also failed to show a significant difference between erythromycin and nasogastric lavage [45]. (See 'Nasogastric lavage' above.)
Metoclopramide has also been studied as an approach to improving gastric visualization. In a randomized trial with 62 patients with acute upper GI bleeding, intravenous metoclopramide (10 mg administered over five minutes) given 30 to 120 minutes prior to endoscopy was not different from placebo with regard to the primary endpoint of adequate visualization (77 versus 62 percent, OR 2.16 [95% CI 0.71-6.58]), p=0.16) or secondary endpoints of second-look endoscopy (10 versus 23 percent, OR 0.36 [95% CI 0.08-1.57], p=0.16) and length of stay (5.4 vs. 6.7 days; p=0.46) [46]. Whether these modest differences in favor of metoclopramide would be maintained and achieve significance in a larger trial is unknown.
Vasoactive medications — Somatostatin, its analog octreotide, and terlipressin are used in the treatment of variceal bleeding (See "Methods to achieve hemostasis in patients with acute variceal hemorrhage", section on 'Somatostatin and its analogs'.)
Octreotide is not recommended for routine use in patients with acute nonvariceal upper GI bleeding, but it has been used as adjunctive therapy on occasion. Its role is generally limited to settings in which endoscopy is unavailable or as a means to help stabilize patients before definitive therapy can be performed. (See "Overview of the treatment of bleeding peptic ulcers", section on 'Somatostatin and octreotide'.)
Antibiotics for patients with cirrhosis — Bacterial infections are present in up to 20 percent of patients with cirrhosis who are hospitalized with gastrointestinal bleeding; up to an additional 50 percent develop an infection while hospitalized. Such patients have increased mortality. Patients with cirrhosis who present with acute upper GI bleeding (from varices or other causes) should be given prophylactic antibiotics, preferably before endoscopy (although effectiveness has also been demonstrated when given after endoscopy). (See "Overview of the management of patients with variceal bleeding".)
Ineffective treatments — Tranexamic acid is an antifibrinolytic agent that has been studied in patients with upper GI bleeding and does not appear to be beneficial [47,48].
Consultations
●Gastroenterology – Gastroenterological consultation should be obtained in all patients with suspected clinically significant acute upper GI bleeding.
●Transfusion medicine (if available) – The transfusion medicine service or clinical pathology/blood bank physician should be alerted if the patient may require a massive transfusion protocol so they can ensure that adequate blood products are available.
●Hematology – Hematology consultation can help if emergency reversal of anticoagulation is indicated and/or if abnormal clotting tests are unexplained.
●Surgery/interventional radiology – The decision to obtain surgical and interventional radiology consultations prior to endoscopy should be based upon the likelihood of persistent or recurrent bleeding, or risks/complications stemming from endoscopic therapy (perforation, precipitation of massive bleeding).
●Clinician who prescribed an anticoagulant or antiplatelet agent (if relevant) – If an anticoagulant or antiplatelet agent needs to be discontinued and/or reversed due to serious or life-threatening bleeding, the clinician who prescribed the medication should be contacted to discuss plans for resuming it.
As a general rule, we obtain surgical and interventional radiology consultation if endoscopic therapy is unlikely to be successful, if the patient is deemed to be at high risk for rebleeding or complications associated with endoscopy, or if there is concern that the patient may have an aorto-enteric fistula. In addition, a surgeon and an interventional radiologist should be promptly notified of all patients with severe acute upper GI bleeding, such as those presenting with hemodynamic instability that fails to respond to resuscitation [49].
DIAGNOSTIC STUDIES —
An algorithm providing an overview of the diagnostic approach to patients with suspected upper gastrointestinal bleeding is provided (algorithm 2).
Upper endoscopy — Upper endoscopy is the diagnostic modality of choice for acute upper GI bleeding (algorithm 2) [50,51]. Endoscopy has a high sensitivity and specificity for locating and identifying bleeding lesions in the upper GI tract. In addition, once a bleeding lesion has been identified, therapeutic endoscopy can achieve acute hemostasis and prevent recurrent bleeding in most patients. Early endoscopy (within 24 hours) is recommended for most patients with acute upper GI bleeding. For patients with suspected variceal bleeding, United States and international guidance suggest endoscopy within 12 hours, although available evidence generally supports a 24-hour threshold in this group as well [52-55]. (See 'Early endoscopy' below and "Methods to achieve hemostasis in patients with acute variceal hemorrhage", section on 'Initial management' and "Overview of the treatment of bleeding peptic ulcers", section on 'Endoscopic therapy' and "Methods to achieve hemostasis in patients with acute variceal hemorrhage", section on 'Management of esophageal varices'.)
Endoscopic findings in patients with bleeding peptic ulcers are described using the modified Forrest classification [56]. Findings include spurting hemorrhage (class Ia) (picture 1), oozing hemorrhage (class Ib), a nonbleeding visible vessel (class IIa) (picture 2), an adherent clot (class IIb) (picture 3), a flat pigmented spot (class IIc) (picture 4), and a clean ulcer base (class III) (picture 4). The endoscopic appearance helps determine which lesions require endoscopic therapy. (See "Overview of the treatment of bleeding peptic ulcers", section on 'Endoscopic therapy'.)
It may be helpful to give a prokinetic agent such as erythromycin or to irrigate the stomach prior to endoscopy to help remove residual blood and other gastric contents. However, despite prokinetic administration or irrigation, the stomach can be obscured with blood, potentially making it difficult to establish a clear diagnosis and/or perform therapeutic maneuvers. In patients in whom blood obscures the source of bleeding, a second endoscopy may be required to establish a diagnosis and to potentially apply therapy, but routine second-look endoscopy is not recommended. (See 'Nasogastric lavage' above and "Overview of the treatment of bleeding peptic ulcers", section on 'Second-look endoscopy'.)
Early endoscopy — Our approach is to perform upper endoscopy within 24 hours for most patients with upper GI bleeding, but only after adequate resuscitation has been provided. For patients with suspected variceal bleeding, we perform endoscopy within 12 to 24 hours of presentation.
Some retrospective studies have suggested that endoscopy within 24 hours may reduce length of stay, need for surgery, and possibly even mortality [3,57,58]. Randomized trials have shown that among hemodynamically stable patients without severe comorbidities endoscopy within two to six hours after initial evaluation identifies low-risk endoscopic findings which may allow early discharge in ≥40 percent of cases, thereby reducing costs [59,60].
While some retrospective studies have suggested that very early endoscopy (within 6 to 12 hours) may be associated with poor outcomes in higher-risk patients [61-63], possibly due to inadequate resuscitation or attention to other comorbidities, a randomized trial with 516 patients with upper GI bleeding who were at increased risk for death or further bleeding (Glasgow-Blatchford score ≥12) did not find a difference in outcomes between those who underwent "urgent" endoscopy (within six hours of gastroenterology consultation) and "early" endoscopy (between 6 and 24 hours after gastroenterology consultation), though there was a trend toward worse outcomes in the urgent endoscopy group [64]. Outcomes examined in the study included 30-day mortality (8.9 versus 6.6 percent with urgent and early endoscopy, respectively; hazard ratio 1.35; 95% CI 0.72 to 2.54) and further bleeding within 30 days (10.9 versus 7.8 percent; hazard ratio 1.45; 95% CI 0.83 to 2.58). Of note, because there was a lag between presentation and gastroenterology consultation, the patients in the urgent endoscopy group underwent endoscopy a mean of 10 hours after presentation and those in the early endoscopy group underwent endoscopy a mean of 25 hours after presentation (with 55 percent undergoing endoscopy >24 hours after presentation). In addition, patients with hemodynamic instability who could not be stabilized were excluded from the study, so the results may not apply to this group of patients.
Risks of endoscopy — Risks of upper endoscopy include pulmonary aspiration, adverse reactions to medications used to achieve sedation, GI perforation, and increasing bleeding while attempting therapeutic intervention.
While patients need to be hemodynamically stable prior to undergoing endoscopy, data suggest that patients do not need to have a normal hematocrit in order to safely undergo endoscopy [65]. In addition, endoscopy appears to be safe in patients who are mildly to moderately anticoagulated [34]. In a retrospective study of 920 patients with upper GI bleeding undergoing upper endoscopy, patients with low hematocrits (<30 percent) were similar to those with high hematocrits (>30 percent) with regard to cardiovascular complications and mortality [65]. In another retrospective study with 233 patients with upper GI bleeding who received endoscopic therapy, an elevated INR was not associated with an increased risk of rebleeding, transfusion requirement, surgery, length of stay, or mortality [34]. However, 95 percent of the patients had an INR between 1.3 and 2.7, so the results of the study may only apply to patients who are mildly to moderately anticoagulated.
The risks versus benefits of upper endoscopy should be considered in high-risk patients, such as those who have had a recent myocardial infarction. In one study, for example, 200 patients who underwent endoscopy within 30 days after myocardial infarction (MI) were compared with 200 controls matched for age, sex, and endoscopic indication [66]. Complications (including fatal ventricular tachycardia, near respiratory arrest, and mild hypotension) occurred more often in patients who had a recent MI (8 versus 2 percent) and in patients who were very ill (Apache II score >16 or hypotension prior to endoscopy; 21 versus 2 percent). However, such patients are at increased risk for complications even without endoscopy and may be particularly vulnerable to complications from continued bleeding without endoscopy. (See "Predictive scoring systems in the intensive care unit".)
Other diagnostic tests — Other diagnostic tests for acute upper GI bleeding include CT angiography (CTA using an overt GI bleeding protocol) and angiography, which can detect active bleeding [67,68], deep small bowel enteroscopy, and rarely, intraoperative enteroscopy (algorithm 2). Upper GI barium studies are contraindicated in the setting of acute upper GI bleeding because they will interfere with subsequent endoscopy, angiography, or surgery [50]. There is also interest in using wireless capsule endoscopy for patients who have presented to the emergency department with suspected upper GI bleeding. An esophageal capsule (which has a recording time of 20 minutes) can be given in the emergency department and reviewed immediately for evidence of bleeding. Confirming the presence of blood in the stomach or duodenum may aid with patient triage and identify patients more likely to benefit from early endoscopy [69-73]. Small bowel capsule endoscopy has also been employed to help localize bleeding in patients with acute gastrointestinal bleeding without hematemesis. (See "Angiographic control of nonvariceal gastrointestinal bleeding in adults" and "Evaluation of suspected small bowel bleeding (formerly obscure gastrointestinal bleeding)" and "Wireless video capsule endoscopy", section on 'Esophageal and gastric capsule endoscopy' and "Wireless video capsule endoscopy", section on 'Indications' and "Wireless video capsule endoscopy", section on 'Acute gastrointestinal bleeding'.)
A colonoscopy is generally required for patients with hematochezia and a negative upper endoscopy unless an alternative source for the bleeding has been identified. In addition, patients with melena and a negative upper endoscopy frequently undergo colonoscopy to rule out a colonic source for the bleeding, as right-sided lesions may present with melena. In a study that included 1743 colonoscopies performed for the evaluation of melena following a non-diagnostic upper endoscopy, a suspected bleeding source was identified in 5 percent of patients, a rate that was higher than that seen in 194,979 average-risk screening controls (1 percent) [74]. Despite the relatively low yield in patients with melena, we routinely perform a colonoscopy in patients with melena and a negative upper endoscopy, as well as in patients with hematochezia. (See "Approach to acute lower gastrointestinal bleeding in adults", section on 'Colonoscopy'.)
RISK STRATIFICATION —
Endoscopic, clinical, and laboratory features may be useful for risk stratification of patients who present with acute upper GI bleeding (table 5 and picture 1 and picture 2 and picture 3 and picture 4) [75-84], and the use of risk stratification tools is recommended by International, United States, and European guidelines, although these are primarily used to identify very low-risk patients at presentation and are less reliable for predicting subsequent risk to guide management [1]. Factors associated with rebleeding identified in a meta-analysis included [85]:
●Hemodynamic instability (systolic blood pressure less than 100 mmHg, heart rate greater than 100 beats per minute)
●Hemoglobin less than 10 g/L
●Active bleeding at the time of endoscopy
●Large ulcer size (greater than 1 to 3 cm in various studies)
●Ulcer location (posterior duodenal bulb or high lesser gastric curvature)
An increase in the blood urea nitrogen (BUN) level at 24 hours compared with baseline may be another predictor of poor outcomes. A study of 357 patients with acute nonvariceal upper GI bleeding found that an increase in the BUN at 24 hours was a predictor of a composite outcome that included rebleeding and mortality [86]. The authors speculate that the association of poor outcomes with an increase in BUN may be the result of inadequate resuscitation.
Several investigators have developed decision rules and predictive models that permit identification of patients who are at low risk for recurrent or life-threatening hemorrhage [87]. Such patients may be suitable for early hospital discharge or even outpatient care. The effectiveness of such rules has been evaluated in a variety of clinical settings, with most studies suggesting that patients deemed to be low-risk can safely be discharged early or treated as outpatients [59,60,75-81,87-93]. In addition, this approach is associated with reduced resource utilization compared with universal hospitalization of patients with acute upper GI bleeding.
Risk scores — Two commonly cited scoring systems are the Rockall score and the Blatchford score. The International Consensus Group suggests using a Glasgow Blatchford score (GBS) of ≤1 to identify patients who are very low risk for rebleeding or mortality and who can be considered for outpatient management [1].
●The full Rockall score is calculated after endoscopy and is based upon age, the presence of shock, comorbidity, diagnosis, and endoscopic stigmata of recent hemorrhage (calculator 1) [75]. In one validation study, only 32 of 744 patients (4 percent) who scored 2 or less (out of a maximum of 11) rebled and only one died.
On the other hand, in a later study of 247 patients who underwent endoscopic therapy for bleeding peptic ulcers, the model performed poorly when predicting recurrent bleeding, underscoring the need for validation of this model [94].
●The GBS, unlike the Rockall score, does not take endoscopic data into account and thus can be calculated when the patient first presents (calculator 2) [80]. The score is based upon the blood urea nitrogen, hemoglobin, systolic blood pressure, pulse, and the presence of melena, syncope, hepatic disease, and/or cardiac failure. The score ranges from zero to 23 and the risk of requiring endoscopic intervention increases with increasing score. One meta-analysis found that a Blatchford score of zero was associated with a low likelihood of the need for urgent endoscopic intervention (likelihood ratio 0.02, 95% confidence interval [CI] 0-0.05) [6]. A second study with 3012 patients found that a score ≤1 could be used to identify a low-risk cohort [95].
A simpler version of the score, known as the modified GBS, is calculated using only the blood urea nitrogen, hemoglobin, systolic blood pressure, and pulse. The score ranges from 0 to 16. A prospective study of the modified score found that it performed as well as the full GBS and that it outperformed the Rockall score with regard to predicting the need for clinical intervention, rebleeding, and mortality [96].
AIMS65 is another scoring system that uses data available prior to endoscopy (serum albumin, INR, presence of altered mental status, systolic blood pressure, and age), but it is less sensitive than the Blatchford and preendoscopic Rockall scores for identifying low-risk patients [95,97]. A newer score, the Age, Blood tests and Comorbidities (ABC) score, was developed to predict mortality in patients with upper GI bleeding and lower GI bleeding [98]. Initial data from the validation cohort, which included 4019 patients with upper GI bleeding and 2336 patients with lower GI bleeding suggests good performance for the score (AUROC 0.81 to 0.84).
Implementation — The data presented above suggest that risk stratification is feasible and permits identification of patients who can be managed safely without hospitalization. However, for these systems to be successful, the risk stratification system must be tied directly to decisions regarding patient discharge. None of the published risk scores has yet been adopted widely.
As a general rule, we discharge patients following endoscopy if they have a likely bleeding source identified on upper endoscopy that is not associated with a high risk of rebleeding provided they:
●Have no comorbidities
●Have stable vital signs
●Have a normal hemoglobin level
High-risk bleeding sources include variceal bleeding, active bleeding, bleeding from a Dieulafoy's lesion, or an ulcer bleeding with high-risk stigmata (table 5).
Prior to endoscopy, we will discharge patients with a GBS of 0-1 provided they have no significant comorbidities, have stable vital signs, have a normal hemoglobin level, do not live far from medical care, and have a mechanism for prompt outpatient gastroenterology evaluation/endoscopy (ie, within three days). We discharge the patient on a proton pump inhibitor (eg, omeprazole or esomeprazole 20 mg once daily, pantoprazole 40 mg once daily).
Ultimately, the decision to discharge the patient also depends on individual-patient factors, such as reliability for follow-up and confidence of the endoscopists in the diagnosis.
If patients do not meet the above criteria, we admit them to a monitored setting or intensive care unit (depending upon the severity of bleeding, comorbidities, and stability of vital signs). Most patients who have received endoscopic treatment for high-risk stigmata should be hospitalized for 72 hours to monitor for rebleeding, since most rebleeding occurs during this time [99].
TREATMENT —
The treatment of patients with upper GI bleeding is discussed separately:
●(See "Overview of the treatment of bleeding peptic ulcers".)
●(See "Methods to achieve hemostasis in patients with acute variceal hemorrhage".)
●(See "Angiodysplasia of the gastrointestinal tract", section on 'Treatment'.)
●(See "Portal hypertensive gastropathy", section on 'Management'.)
●(See "Causes of upper gastrointestinal bleeding in adults", section on 'Vascular lesions'.)
●(See "Causes of upper gastrointestinal bleeding in adults", section on 'Aortoenteric fistulas'.)
●(See "Causes of upper gastrointestinal bleeding in adults", section on 'Upper gastrointestinal tumors'.)
●(See "Argon plasma coagulation in the gastrointestinal tract".)
●(See "Angiographic control of nonvariceal gastrointestinal bleeding in adults".)
SOCIETY GUIDELINE LINKS —
Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Gastrointestinal bleeding in adults".)
INFORMATION FOR PATIENTS —
UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Upper endoscopy (The Basics)" and "Patient education: GI bleed (The Basics)")
●Beyond the Basics topics (see "Patient education: Upper endoscopy (Beyond the Basics)" and "Patient education: Peptic ulcer disease (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Rapid overview – A table outlining the emergency management of acute severe upper gastrointestinal bleeding is provided (table 1). (See 'Introduction' above.)
●History and examination – The history and physical examination can help identify potential sources of upper GI bleeding, assess the severity of the bleeding episode, and identify comorbid conditions that may influence the patient's subsequent management. (See 'Initial evaluation' above.)
The presence of abdominal pain, especially if severe and associated with rebound tenderness or involuntary guarding, raises concern for perforation. If any signs of an acute abdomen are present, further evaluation to exclude a perforation is required prior to endoscopy. (See 'Physical examination' above.)
●Laboratory testing – Laboratory tests that should be obtained in patients with acute upper GI bleeding include a complete blood count, serum chemistries, liver tests, and coagulation studies. In addition, we suggest ruling out a myocardial infarction in older adult patients and those with known cardiovascular disease who have severe bleeding, especially if there has been hemodynamic instability. (See 'Laboratory data' above.)
●Risk stratification and triage – We suggest incorporation of a validated risk score for upper GI bleeding into routine clinical practice to facilitate optimal triage decisions. Patients with a Glasgow Blatchford Score (GBS) of 0-1 may be considered for out-patient management. Others should be admitted to a monitored bed or intensive care unit depending upon the severity of bleeding. (See 'Risk scores' above and 'Triage' above.)
●Pre-endoscopic management – Prior to endoscopy, patients should receive general supportive measures, including:
•Provision of supplemental oxygen by nasal cannula
•Nothing per mouth
•Two large caliber (18 gauge or larger) peripheral catheters or a central venous line
Nasogastric lavage is NOT a routine part of the management of acute upper GI bleeding. (See 'Nasogastric lavage' above.)
●Pharmacologic therapy – Medications that should be started prior to endoscopy include a proton pump inhibitor, erythromycin, antibiotics (for patients with cirrhosis), and somatostatin or one of its analogs (for patients with suspected variceal bleeding).
•We suggest that patients admitted to the hospital with acute upper GI bleeding receive an IV proton pump inhibitor (PPI) (Grade 2B). The optimal approach to PPI administration prior to endoscopy is unclear. Our approach is to give a high-dose bolus (eg, esomeprazole 80 mg) to patients with signs of active bleeding (eg, hematemesis, hemodynamic instability). If endoscopy is performed beyond 12 hours, a second dose of an IV PPI should be given (eg, esomeprazole 40 mg). For patients who may have stopped bleeding (eg, patients who are hemodynamically stable with melena), we give an IV PPI every 12 hours (eg, esomeprazole 40 mg). Subsequent dosing will then depend on the endoscopic findings. (See 'Acid suppression' above and "Overview of the treatment of bleeding peptic ulcers", section on 'Acid suppression'.)
We suggest that erythromycin be given prior to endoscopy to help improve visualization (Grade 2C). A reasonable dose is 250 mg intravenously over 20 to 30 minutes. Endoscopy is performed 20 to 90 minutes following completion of the erythromycin infusion. Patients receiving erythromycin need to be monitored for QTc prolongation. In addition, drug-drug interactions should be evaluated before giving erythromycin because it is a cytochrome P450 3A inhibitor (table 4). (See 'Prokinetics' above.)
•Patients with cirrhosis who present with acute upper GI bleeding (from varices or other causes) should be given prophylactic antibiotics. (See "Overview of the management of patients with variceal bleeding" and "Overview of the management of patients with variceal bleeding", section on 'Antibiotic prophylaxis'.)
•A vasoactive medication (somatostatin, its analog octreotide, or terlipressin) should be started if variceal bleeding is suspected. (See "Methods to achieve hemostasis in patients with acute variceal hemorrhage" and "Methods to achieve hemostasis in patients with acute variceal hemorrhage", section on 'Pharmacologic therapy'.)
●Blood transfusion – Transfusion is guided by the hemoglobin level and the presence of comorbid conditions:
•The decision to initiate blood transfusion must be individualized (algorithm 1). Our approach is to initiate blood transfusion if the hemoglobin is <7 g/dL (70 g/L) for most patients. Higher transfusion thresholds may be indicated for patients at increased risk of adverse events in the setting of significant anemia or those with coronary artery disease. (See "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Overview of our approach'.)
•For patients with active/brisk bleeding and hypovolemia, decisions about transfusion are guided by hemodynamic parameters (eg, pulse and blood pressure), the pace of the bleeding, estimated blood loss, and the ability to stop the bleeding, rather than by serial hemoglobin measurements. (See "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Acute bleeding' and "Massive blood transfusion", section on 'Approach to volume and blood replacement'.)
●Anticoagulants, antiplatelet agents, and coagulopathies – The approach to management of anticoagulants and antiplatelet agents depends on the specific medications being used, the reason they are being used, and how quickly reversal of anticoagulation is needed. Management of coagulopathy depends on the underlying etiology. Hematology input may be advisable. (See 'Managing anticoagulants, antiplatelet agents, and coagulopathies' above and 'Consultations' above.)
●Endoscopy – We recommend upper endoscopy within 24 hours for the evaluation and management of patients admitted with acute upper GI bleeding. For patients with suspected variceal bleeding, we perform endoscopy within 12 to 24 hours of presentation.
Additional diagnostic tests may be required for some patients. An algorithm providing an overview of the diagnostic approach to patients with suspected upper gastrointestinal bleeding is provided (algorithm 2).
●Treatment of the underlying lesion – The approaches to achieve hemostasis for variceal bleeding and bleeding peptic ulcers are discussed separately.
ACKNOWLEDGMENT —
We are saddened by the death of Mark Feldman, MD, who passed away in March 2024. UpToDate gratefully acknowledges Dr. Feldman's role as Section Editor on this topic and his dedicated and longstanding involvement with the UpToDate program.
