Hematemesis is upper gastrointestinal (GI) bleeding, typically described as a hemorrhage proximal to the ligament of Treitz. Symptoms consist of coffee ground emesis, vomiting frank blood, melena, and hematochezia. Symptoms of anemia may also be present, including chest pain, weakness, fatigue, syncope, or dyspnea.
II. Diagnostic Approach
A. What is the differential diagnosis for this problem?
Ulcerative or erosive
Peptic Ulcer Disease (55%)
Drugs: Acetylsalicylic acid (ASA), nonsteroidal anti-inflammatory drugs (NSAIDS)
Infectious: Helicobacter pylori, CMV, HSV
Infectious: Candida albicans, HSV, CMV
Pill induced: Tetracycline, Alendronate, Quinidine, Potassium Chloride, ASA, NSAIDS
Esophageal Varices (14%)
Portal hypertensive gastropathy
Vascular defects (7%)
Radiation induced telangiectasia
Mallory-Weiss tear (5%)
Post endoscopy: polypectomy
Polyp: hyperplastic, adenomatous, hamartomatous
1. Historical information important in the diagnosis of this problem.
Basic history including complete HPI elements and timing of hematemesis are essential to determining the cause. Although attention should be given to the consistency and color of stool, hematochezia and melena can be seen in upper GI bleeding. Patients with hematochezia and upper GI bleeding typically have massive blood loss and have significant hemodynamic instability.
Other important elements to be sure to cover include:
Aspirin or other NSAID use
History of alcohol use
History of liver disease, cirrhosis or variceal bleeding
History of GERD or ulcer disease
Recent weight loss
Dysphagia or odynophagia
2. Physical Examination maneuvers that are likely to be useful in diagnosing the cause of this problem.
Basic physical exam should be performed with special emphasis on:
Vital Signs: Pay close attention to blood pressure and pulse. Obtain orthostatic vitals for hemodynamic stability.
Mouth and Nose: Make sure bleeding is not coming from teeth or epistaxis.
Abdomen: Look for focal tenderness, AAA, signs of cirrhosis (caput medusa, ascites).
Rectal: Guaiac stools.
3. Laboratory, radiographic and other tests that are likely to be useful in diagnosing the cause of this problem.
Complete Blood Count (CBC): Looking for recent past haemoglobin and hematocrits (H/H) to compare to current value. If patient has acute blood loss as in variceal bleeding, there may not be a significant drop in H/H. Also, look at indices for long-term blood loss and iron deficiency in cases such as gastritis or esophagitis. Platelet count can be quite low in patients with cirrhosis.
Chemistry: May show elevated BUN/Creatinine ratio.
PT/PTT: To evaluate possible coagulopathy, especially in patients with cirrhosis or on anticoagulants.
Type and crossmatch: To ensure availability of blood products.
Gastroccult: Sensitive, but not specific.
Upper endoscopy: Diagnostic test of choice for acute hematemesis due to the high sensitivity in ability to locate and treat specific locations of active bleeding lesions. Therapeutic endoscopy can be performed with esophageal variceal banding, clips, sclerosing agents, epinephrine injection, and thermocoagulation. Biopsies can be taken to confirm diagnoses such as benign tumor, malignancy, H. pylori, or other infectious causes.
D. Over-utilized or “wasted” diagnostic tests associated with the evaluation of this problem.
Abdominal X-ray may be a waste of time in obtaining a diagnosis of hematemesis, it often does not give specific information unless you are worried about perforation.
Barium X-rays are contraindicated in acute hematemesis and can interfere with endoscopy, angiography and surgery.
III. Management while the Diagnostic Process is Proceeding
A. Management of Hematemesis.
Patient with shock, hypotension, or active bleeding should be closely monitored in ICU setting with continuous monitoring of vital signs and clinical status. Adequate IV access should be obtained, preferably with 2 large bore IV’s. IV fluids should be started to maintain adequate perfusion while awaiting blood products.
Packed red blood cells should be transfused to maintain hemoglobin at 7 gm/dL in high-risk patients or patients that have lost 30 percent of circulating volume. In patients with recent ischemic heart disease, a transfusion goal of 8-9 gm/dl should be targeted. Also, be aware of the possible need for platelets with patients having a platelet count less than 50 or plasma needs in patients with elevated INR over 1.5 and continued bleeding.
Pressor support can be used as necessary to maintain adequate perfusion if fluids and blood products do not bring pressure up to an acceptable level.
Gastroenterology should be involved soon to be aware of need for emergent endoscopy. Surgery should also be called in cases of severe acute upper GI bleeding where endoscopic intervention is unlikely to succeed.
Naso-gastric (NG) lavage
Naso-gastric (NG) lavage: May be used to remove blood clots to facilitate endoscopy and confirm upper GI source of bleeding, but there is a risk of aspiration that must be weighed against the clinical utility of performing this procedure. NG lavage that reveals bright red blood or coffee ground material confirms upper GI as source. It can also be useful to predict the severity of the lesion. NG lavage may be negative if bleeding has stopped or if the bleeding is occurring after the pylorus and the pylorus closes. If the NG lavage reveals no blood and shows bilious fluid, it is highly suggestive that there is no active upper GI bleeding.
Octreotide should be started and maintained for 5 days in patients with variceal GI bleed.
Proton pump inhibitors (PPI) should be given intravenously in patients with known major peptic ulcer bleeding, including active bleeding or non-bleeding visible vessel after endoscopic hemostasis to lower the risk of rebleeding. Continuation of PPI can be decided once cause of bleeding is determined.
Antibiotics should be initiated in patients with chronic liver disease who present with acute upper GI bleeding.
Patients found to have peptic ulcer disease should be tested and treated for H. pylori if positive. Treatment should consist of one-week eradication therapy and three weeks’ ulcer healing therapy. Maintenance antisecretory therapy is not needed in non-NSAID ulcers.
Early endoscopic intervention
Endoscopic therapy should be initiated within 24 hours and only be delivered to actively bleeding lesions, non-bleeding visible vessels and to ulcers with an adherent blood clot. Combination therapy with epinephrine and thermal or mechanical treatment is recommended over single therapy. Esophageal varices should be treated with band ligation.
If the initial endoscopic evaluation is not optimal, repeat endoscopy should be performed within 24 hours. Routine second-look endoscopies are not recommended.
B. Common Pitfalls and Side-Effects of Managing Hematemesis
Aspirin and NSAIDS should be discontinued in patients with peptic ulcer bleeding. Aspirin and NSAIDS should only be prescribed if there is a clear indication and should be administered with a PPI. ASA for cardiovascular protection should be restarted when the risk of cardiovascular complication outweighs risk of bleeding.
Oral anticoagulants and steroids should be used with caution in patients taking NSAIDS or aspirin.
Erythromycin can be given immediately prior to endoscopy to promote gastric emptying and help with endoscopic visualization, but should not be routinely used.
IV pantoprazole should be given as an 80mg bolus followed by 8mh/hour infusion. May be transitioned to po if no re-bleeding occurs in 24 hours, and may be discontinued if alternate cause is identified.
Octreotide should be infused as a 50mcg bolus followed by 50mcg/hour infusion for 3-5 days. May be discontinued if alternate cause is identified.
Remember to give antibiotics prior to endoscopy to patients with advanced liver disease to prevent increased mortality from infectious complications.
Risk scores should be utilized for patient stratification during triage to determine the need for admission, the level of care, and the predicted mortality risk. Three scores are commonly used: Blatchford, Rockall, and AIMS65.
Use the following formal risk assessment scores for all patients with acute upper gastrointestinal bleeding:
Blatchford score at first assessment, and
Full Rockall score after endoscopy
Consider early discharge for patients with a pre-endoscopy Blatchford score of 0.
(Online calculators are available.)
The AIMS65 is a quick means of determining at bedside the factors associated with increased inpatient mortality:
– Albumin less than 3.0 g/dL (30 g/L)
– INR greater than 1.5
– Altered Mental status (Glasgow coma score less than 14, disorientation, lethargy, stupor, or coma)
– Systolic blood pressure of 90 mmHg or less
– Age older than 65 years
The mortality rate increases significantly as the number of risk factors present increase. Zero risk factors: 0.3%; 1 risk factor: 1%; 2 risk factors: 3%; 3 risk factors: 9%; 4 risk factors: 15%; and five risk factors: 25%.
Barkun, AN, Bardou, M, Kuipers, EJ. “International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding”. Ann Intern Med. vol. 152. 2010. pp. 101
Gralnek, IM, Dumonceau, JM, Kuipers, EJ. “Diagnosis and management of nonvariceal upper gastrointestinal hemorrhage: European Society of Gastrointestinal Endoscopy (ESGE) Guideline”. Endoscopy. vol. 47. 2015. pp. a1
Pallin, DJ, Saltzman, JR. “Is nasogastric tube lavage in patients with acute upper GI bleeding indicated or antiquated”. Gastrointest Endosc. vol. 74. 2011. pp. 981
Villanueva, C, Colomo, A, Bosch, A. “Transfusion strategies for acute upper gastrointestinal bleeding”. N Engl J Med. vol. 368. 2013 Jan. pp. 11-21.
Altraif, I, Handoo, FA, Aljumah, A. “Effect of erythromycin before endoscopy in patients presenting with variceal bleeding: a prospective, randomized, double-blind, placebo-controlled trial”. Gastrointest Endosc. vol. 73. 2011. pp. 245
Saltzman, JR, Tabak, YP, Hyett, BH. “A simple risk score accurately predicts in-hospital mortality, length of stay, and cost in acute upper GI bleeding”. Gastrointest Endosc. vol. 74. 2011. pp. 1215
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- I. Problem/Condition.
- II. Diagnostic Approach
- A. What is the differential diagnosis for this problem?
- 1. Historical information important in the diagnosis of this problem.
- 2. Physical Examination maneuvers that are likely to be useful in diagnosing the cause of this problem.
- 3. Laboratory, radiographic and other tests that are likely to be useful in diagnosing the cause of this problem.
- D. Over-utilized or “wasted” diagnostic tests associated with the evaluation of this problem.
- III. Management while the Diagnostic Process is Proceeding
- A. Management of Hematemesis.
- B. Common Pitfalls and Side-Effects of Managing Hematemesis