by Dan O’Brien
Introduction
Despite advances in diagnosis prevention and treatment, nonvariceal upper gastrointestinal bleeding is still a serious problem in clinical practice. The incidence ranges from 48 to 160 cases per 100,000 population per year. Upper GI bleeding causes mortality ranges from 5% to 14%.
Case Presentation
A 68-year-old female presents to the emergency room at midnight, with a chief complaint of vomiting “coffee grounds” earlier that evening. She has a history of congestive heart failure, hypertension, and a mild stroke. Her medications include lisinopril 20 mg, Lasix 20 mg, aspirin 325 mg, and clopidogrel 75 mg daily. Recently, she has taken ibuprofen several times daily for arthritic pains. Family history is significant for peptic ulcer in her mother and a brother. Pertinent Exam: blood pressure is 98/65 mmHg, heart rate 110 bpm and regular, respiratory rate 14, non-labored, and temperature 37 ºC. She appears pale, has a poor capillary refill and is mildly confused but oriented to person, place and time. Heart and lung sounds are normal; her abdomen is soft, non-tender, without organomegaly, and without bruits. She has trace pedal and pretibial edema. Her neurological exam is grossly normal. While being examined, she asks for a bedpan and vomits a cup full of bright red blood. Her blood pressure systolic is now 85 mm Hg.
She has hemodynamically significant upper GI bleeding. Her hypotension and tachycardia indicate loss of more than 20% total blood volume. The most likely working diagnosis is active upper GI bleeding likely from peptic ulceration secondary to nonsteroidal anti-inflammatory drugs (NSAIDs) with likely Helicobacter infection.
Bedside testing reveals hemoglobin of 6 g/dL. Transfusion with packed red blood cells is begun with a goal of hemoglobin of 8.0-10.0 g/dL. Additional labs demonstrate a normal albumin and prothrombin time (PT). Her creatinine is 2.0 mg/dL. Old records document baseline hemoglobin of 11 g/dL and creatine of 1.0 mg/dL.
Initial resuscitation is successful. Upper endoscopy reveals a bleeding duodenal ulcer that is successfully stopped with hemoclips. The patients H. pylori stool antigen is positive. Cardiology and Neurology agree to stop aspirin and clopidogrel. Oral iron was started, and the patient was discharged. With the avoidance of NSAIDs and with H. pylori eradication the risk of rebleed is less than 5%.
Critical Bedside Actions and General Approach
“She has hemodynamically significant upper GI bleeding.”
Indicators of Major Blood Loss
| Acidosis Anticoagulation Antiplatelet medications Azotemia (BUN > 40 mg/dL Chest pain or dyspnea Continued bleeding or re-bleeding End stage renal disease Hematochezia from upper GI source Hemoglobin <8 g/dL Liver cirrhosis, coagulopathy Orthostatic Resting tachycardia (>100 bpm) Syncope (systolic < 90 mmHg) Transfusion > 1 unit/8hrs or 6 units total |
Initiate resuscitation: insert two large bore intravenous catheters infuse lactated ringers, type and crossmatch, obtain complete blood count with platelets, PT, and INR, as well as routine blood chemistries to assess for renal and hepatic function. Start intravenous octreotide, a somatostatin analog, at 50 mcg/hour. Proton pump inhibitor by continuous drip, as a pH of 7 or greater is needed for platelet function and clot adherence.
Intravenous erythromycin 250 mg if given within 30 minutes of planned endoscopy can improve visualization. If there is a history of liver cirrhosis also give one gram IV Ceftriaxone or 400 mg norfloxacin orally twice daily. Antibiotics are of benefit in cirrhosis by decreasing infectious sequelae as well as the incidence of encephalopathy. The reduction of bacterial products in the portal circulation results in less vasodilation, which lowers the rebleeding risk. After initial stabilization efforts, consult an endoscopist.
“The most likely working diagnosis is active upper GI bleeding likely from peptic ulceration secondary to nonsteroidal anti-inflammatory drugs (NSAIDs) with likely Helicobacter infection.”
Differential Diagnosis
While it is important to know what is bleeding to determine prognosis and guide management, it is most crucial to think of anatomy and pathophysiology: larger vessels bleed faster and more often require urgent intervention. The internal diameter and pressure in vessels above the ligament of Treitz are greater than vessels associated with lower GI bleeding.
Comparison of Upper and Lower GI Bleeding
| Upper GI Bleeding | Lower GI Bleeding |
|---|---|
| 35% present with shock | 19% present with shock |
| 65% require transfusion | 36% require transfusion |
| 30% require intervention to stop | >90% stop spontaneously |
Presentations of Upper and Lower GI Bleeding
| Upper GI Bleeding | Lower GI Bleeding |
|---|---|
| Abdominal pain may or may not be present in peptic ulcer | Large volume hematochezia or maroon stool with orthostasis indicates bleeding from right sided diverticulae or ateriovenous malformations |
| Chest pain with esophageal ulcer | Small volume hematochezia without orthostasis indicates hemorrhoidal bleeding (usually painless) if painful with dyschezia indicates anal fissures or proctitis |
| Sudden fullness with nausea due to blood in GI tract | Bloody loose stools with low abdominal pain present in infectious colitis, inflammatory bowel disease or ischemic colitis |
| Hematemesis or coffee ground emesis followed by melena | |
| Hematochezia in 10% of rarpidly bleeding upper lesions | |
| Coughing followed by hematemesis in Mallory Weiss tear | |
| Valsalva may prompt bleeding from esophageal or gastric varices |
Upper GI bleeding represent 65 to 80% of all GI bleeding, and includes
- esophageal or gastric varices,
- duodenal or gastric ulcer,
- erosive gastritis,
- erosive or ulcerative esophagitis,
- Mallory Weiss tears,
- gastrointestinal cancers,
Rarer causes are
- hemobilia,
- splenic artery pseudoaneurysms,
- Dieulafoy lesions,
- gastrin-secreting tumors (Zollinger-Ellison syndrome),
- arteriovenous fistulae,
- penetrating foreign bodies,
- gastric antral vascular ectasias (GAVE or “watermelon stomach”) or arteriovenous
- malformations above the ligament of Treitz.
Worldwide, upper GI bleeding from peptic ulcer is most prevalent, although persons with portal hypertension may represent the majority who present with massive upper GI bleeding. With advanced age and atherosclerotic disease, more patients are using anticoagulants or antiplatelet medications that impair clot formation, and augment bleeding. Gastrointestinal bleeding from non-steroidal induced peptic ulcers is on the rise, with up to 1 in 2 adults taking these medicines. Although with industrialization and improved hygiene the prevalence of Helicobacter pylori has declined, the infection and associated conditions are still major causes of upper GI bleeding in many parts of the world including the Middle East, Asia, and South America.
Lower GI bleeding is most often caused by right-sided diverticula, arteriovenous malformations, colonic adenocarcinoma, ischemic colitis, inflammatory bowel disease, infectious colitis, or anorectal lesions including hemorrhoids, fissures, and proctitis.
History and Physical Examination Hints
Elderly patients and those with valvular heart disease or renal failure have an increased risk for arteriovenous malformations anywhere in the GI tract. Chronic NSAID use causes gastric mucosal erosions in at least one-third of daily users or significant ulceration in 2%. Alcohol consumption, chronic viral hepatitis, non-alcoholic hepatitis (NASH) can result in cirrhosis with portal hypertension. A family or prior history of peptic ulcer suggests Helicobacter pylori infection. H. pylori is a spiral-shaped flagellated bacterium that lives in the human stomach and interrupts the protective mucous bicarbonate layer, thus exposing the epithelium to hydrochloric acid leading to chronic inflammation. Most persons with duodenal ulcer report sharp epigastric pain worsened by eating while less than 50% of patients with gastric ulcer report abdominal pain that improves post meals, as the acid is then buffered by the food and duodenal bicarbonate secretion. The color of vomitus or stool is also predictive of severity: hematemesis suggests ongoing bleeding, whereas “coffee grounds” indicate partially digested hematin or “old blood.” Abrupt symptom onset associated with hypotension suggests acute bleeding whereas a history of weeks of intermittent dark melanic stools suggests chronic blood loss. Presenting vitals signs are most predictive of the magnitude of blood loss; other important physical clues for portal hypertension include abdominal ascites, enlarged liver or splenomegaly. Evidence of hyperestrogenemia in males with cirrhosis includes gynecomastia, testicular atrophy, and spider telangiectasias on the chest or upper body. Palmer erythema and bounding pulses in the fingers from peripheral vasodilation are indicators of advanced cirrhosis. The bedside physical exam is unreliable in females as palmer erythema and telangiectasias are normal findings. Ascites determination is difficult in every patient unless it is massive. If available, a bedside ultrasound may confirm suspected ascites, coarse echotexture of the liver, or show reduced or reversed (hepatopedal) flow in the hepatic veins in advanced cirrhosis.
Emergency Diagnostic Tests and Interpretation
When abdominal pain is present, a plain film with upright chest x-ray may reveal significant atherosclerotic disease, ingested foreign bodies, or subdiaphragmatic free air from a perforated ulcer.
“Bedside testing reveals hemoglobin of 6 g/dL. Transfusion with packed red blood cells is begun with a goal of hemoglobin of 8.0-10.0 g/dL. Additional labs demonstrate normal albumin and PT. Her creatinine is 2.0 mg/dL. Old records document baseline hemoglobin of 11 g/dL and creatine of 1.0 mg/dL.”
Patients with low albumin, prolonged PT or INR, or platelet count less than 150K, consider underlying cirrhosis and avoid transfusion above hemoglobin of 7.5 to 8.0, to avoid increasing portal pressure and increasing the risk of rebleeding. A platelet transfusion would be indicated for a critically low platelet count (<50K) from consumptive coagulopathy or splenic sequestration. For significant coagulopathy, (INR >1.8) transfusing fresh frozen plasma (10-15 cc/kg) corrects factor seven deficiency to allow coagulation to occur. Cryoprecipitate or other specific factors may be indicated for patients with known factor deficiencies. Red blood cell indices are useful as a low mean corpuscular volume (MCV) suggests iron deficiency from chronic GI blood loss whereas in acute bleeding MCV is often increased due to the release of reticulocytes from the bone marrow.
Emergency Treatment Options
“Initial resuscitation is successful.”
Emergency Management of GI Bleeding
| Treatment Options | Comments |
|---|---|
| Resuscitation with packed red blood cells | Goal: Hgb 7.5 to 8.0 in those with portal HTN Hgb 9.5 to 10 in the elderly, MI, CHF, stroke |
| IV octreotide | 50mcg/hour for active bleeding anywhere in the GI tract (both upper and lower bleeding) |
| IV continuous PPI for all upper GI bleeders | give IV erythromycin 250mg 30 minutes prior to EGD to improve visualization |
| Antibiotics | ceftriaxone 1 gram IV or norfloxacin 400mg IV twice daily for all GI bleeders with cirrhosis |
| Timing of endoscopy | It depends on magnitude of bleeding; goal is to stop bleeding as soon as possible to prevent end organ ischemic damage. Involve consultants early! |
| Other | Less than 5% cases require angiography, less than 1% surgery |
It is not necessary to insert a nasogastric tube (NG) as the information obtained rarely changes management and may compound problems by causing pain, gagging, and epistaxis. A negative aspirate does not exclude active bleeding, and a positive aspirate does not affect the timing of endoscopy or additional interventions. Gastric lavage is no longer considered useful. Consider endotracheal intubation to decrease aspiration risk before elective endoscopy for any patient with upper GI bleeding who is unconscious, in significant respiratory distress, or with recurrent witnessed hematemesis.
Urgent Endoscopy: Call as soon as possible for endoscopy in patients with hemodynamically significant bleeding. Endoscopy is portable and can be safely performed in the emergency department providing immediate information regarding diagnosis, treatment, prognosis, and disposition.
Large studies have shown that endoscopy can safely be performed in patients with bleeding leading to acute myocardial ischemia or infarction, with improved outcomes for interventions that stop further bleeding. The magnitude of bleeding, signs of continued bleeding, or suspicion of varices predicate the timing of endoscopy in upper GI bleeding. Outcomes for endoscopic intervention have shown reduced rebleeding and transfusion requirements, with improved morbidity. With excellent clinical care and combined with endoscopic therapies, mortality from non-variceal GI bleeding is 10%, and mortality from variceal hemorrhage is 25-30%. Colonoscopy for diagnosis of lower GI bleeding cause is most often performed electively after resuscitation; usually within 24 to 48 hours. Visualization of the lower GI tract requires cleansing with large volume balanced electrolyte solutions taken orally or by NG for quicker delivery. Colonoscopy is used less often as an interventional therapeutic technique to stop bleeding from hemorrhoids, fissures, arteriovenous malformations or diverticuli.
Interventional Radiographic Techniques are required in less than 5% of all cases of non-variceal GI bleeding. Angiography can arrest bleeding from arteriovenous malformations of the upper or lower GI tract, as well as selective embolization of arterioles from bleeding tumors or ulcers. Early use of transjugular intrahepatic portosystemic shunts (TIPS) to definitively reduce portal pressure is beneficial after initial endotherapy, and as first-line therapy for select patients with high risk of variceal bleeding.
Emergency Surgery for GI bleeding is required in less than 1% of all cases including surgery for a peptic ulcer, and total or subtotal colectomy for shock associated with bleeding diverticuli. Emergency shunt surgery for liver cirrhosis is almost never performed, as the mortality is unacceptably high compared to endoscopic and angiographic techniques.
Pregnant Patients and Other Considerations
In pregnant patients with GI bleeding, monitor for fetal distress, and consult Obstetrics. Avoid erythromycin in the third trimester otherwise treat the same as any other adult with GI bleeding. Emergency upper endoscopy is safe in all trimesters. Lower endoscopy may be difficult depending on the size of the fetus/uterus but is not contraindicated. The endoscopist will use safe sedation medications for pregnancy. For GI bleeding in patients with acute MI, significant heart disease, stroke or significant neurovascular disease consult cardiology or neurology for help with management. Often they will agree to urgent endoscopy to clarify and treat bleeding lesions without interruption or reversal of anticoagulation.
Disposition Decisions
There are several valuable bleeding scoring systems help to guide disposition.
Glasgow-Blatchford Risk Score
| Category | Score |
|---|---|
| BUN in mg/dL | |
| 18.2 to 22.4 | 2 |
| 22.5 to 28 | 3 |
| 28.1 to 70 | 4 |
| 70.1 or greater | 6 |
| Hemoglobin, men g/dL | |
| 12 to 13 | 1 |
| 10 to 11.9 | 3 |
| 9.9 or less | 6 |
| Hemoglobin, women g/dL | |
| 10 to 12 | 1 |
| 9.9 or less | 6 |
| Systolic Blood Pressure, mmHg | |
| 100-109 | 1 |
| 90-99 | 2 |
| <90 | 3 |
| Heartrate >100 peats per minute | 1 |
| Melena | 1 |
| Syncope | 2 |
| Hepatic Diseases | 2 |
| Heart failure | 2 |
AIM65 Bleeding Score
| Risk Factor | Score |
|---|---|
| Albumin <3.0 | 1 |
| INR > 1.5 | 1 |
| Altered mental status | 1 |
| SPB < 90mm Hg | 1 |
| Age > 65 | 1 |
| MAXIMUM SCORE | 5 |
| Point | Mortality % |
| 0 | 0% |
| 1 | 0.9% |
| 2 | 7.4% |
| 3 | 42% |
| 4 | 75% |
| 5 | 100% |
Admission Criteria
Patients with GI bleeding presenting in shock, requiring transfusion or with bleeding scores (AIM65>2 or Glasgow-Blatchford (GB) >10) have significant predictable morbidity and mortality requiring ICU admission and treatment. Patients with AIM65 of 1 or less or GB score 2 or less have predictably mild GI bleeding (melena without hematemesis and who are hemodynamically stable) and may be cautiously admitted to a medical floor.
Discharge Criteria
Patients with AIM or GB score of zero may be discharged home without outpatient gastroenterology evaluation within two weeks. Prescribe twice daily PPI, avoidance of NSAIDS and alcohol for those with upper GI symptoms. Instruct them to return immediately for syncope or signs of bleeding. Those with history and findings consistent with minor lower GI bleeding and stable hemoglobin should also be referred for outpatient consultation within 2 weeks.
“Upper endoscopy reveals a bleeding duodenal ulcer that is successfully stopped with hemoclips. The patients H. pylori stool antigen is positive. Cardiology and Neurology agree to stop aspirin and clopidogrel. Oral iron was started, and the patient was discharged. With the avoidance of NSAIDs and with H. pylori eradication the risk of re-bleed is less than 5%.
References and Further Reading
- Sostres C, Lanas A. Epidemiology and demographics of upper gastrointestinal bleeding: prevalence, incidence and mortality. Gastrointest Endosc Clin N Am. 2011 Oct;21(4):567-8
- Rotonado G. Epidemiology and diagnosis of acute nonvariceal upper gastrointestinal bleeding. Gastroenterology Clinics of North America, 2014-12-01, Volume 43, Issue 4, Pages 643-663
- Gotzsche P, Hrobjartsson A, Somatostatin analogues for acute bleeding oesophageal varices. Cochrane Database Syst Rev 2005;(1):CD000193. http//dx.doi.org/10.1002/12651858.CD000193.pub2.
- Barkun A, Bardou M, Kuipers E, et al. International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2010;152:101-113
- Lau J, Sung J, Lee K, et al. Effect of intravenous omeperazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers. N Engl J Med 2000;343:310-316.
- Barkun A, Bardou M, Gralnek I, Sung J. Prokinetics in acute upper GI bleeding: a meta-analysis. Gastrointest Endosc 2010;72:1138-1145.
- Fernandez J, Ruiz del Arbol L, Gomez C, et al. Norfloxacin vs. Ceftriaxone in the prophylaxis of infections in patients with advanced cirrhosis and hemorrhage. Gastroenterology 2006;131:1049-56.
- Longstreth G. Epidemiology of hospitalization for acute upper gastrointestinal hemorrhage: a population based study. Am J Gastroenterol 1995;90:206-210.
- van Leerdam M. Epidemiology of acute upper gastrointestinal bleeding. Best Pract Res Clin gastroenterol 2008;22:209-224.
- Ahsberg K, Hoglund P, Kim W, von Holstein C. Impact of aspirin, NSAIDS, warfarin, corticosteroids and SSRIs on the site and outcome of non-variceal upper and lower gastrointestinal bleeding. Scand J Gastroenterol 2010;45:1404-1415.
- Lanas A, Perez-Aisa M, Feu F, et al. A nationwide study of mortality associated with hospital admission due to severe gastrointestinal events and those associated with nonsteroidal anti-inflammatory drug use. Am J Gastroenterol 2006;100:1685-1693.
- Perez-Perez G, Rothenbachert D, Brenner H. Epidemiology of Helicobacter pylori infection. Helicobacter 2004;9 (Suppl.1)1-6
- Kok T, van der Jagt E, Haagsma E, et al. The value of Doppler ultrasound in cirrhosis and portal hypertension. Scand J Gastroenterol 1999;230:82-88.
- Jairath V, Hearnshaw S, Brunskill S, et al. Red cell transfusion for the management of upper gastrointestinal haemorrhage. Cochrane Database Syst Rev. 2010;CD006613.
- Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute gastrointestinal bleeding. N Engl J Med 2013;368:11-21.
- Rudolph S, Landsverk B, Freeman M. Endotracheal intubation for airway protection during endoscopy for severe GI hemorrhage. Gastrointest Endosc 2001;57:58-61
- Yachimski P, Hur C. Upper endoscopy in patients with acute myocardial infarction and upper gastrointestinal bleeding: Results of a decision analysis. Dig Dis Sci 2009;54:701-711
- Holster I, Kuipers E. Update on the endoscopic management of peptic ulcer bleeding. Curr Gastroenterol Rep 2011;13:525-531
- Laine L, McQuaid K. Endoscopic therapy for bleeding ulcers:an evidence-based approach based on meta-analysis of randomized controlled trials. Clin Gastroenterol Hepatol. 2009;7:33-47
- Crooks C, Card T, West J. Reductions in 28-day mortality following hospital admission for upper gastrointestinal hemorrhage. Gastroenterol 2011;141:62-70
- Cardenas, A. Management of Acute Variceal Bleeding: Emphasis on endoscopic therapy. Clin Liver Dis 2010;14:251-262
- Jensen D, Machicado G. Colonoscopy for diagnosis and treatment of severe lower gastrointestinal bleeding: routine outcomes and cost analysis. Gastrointest Endosc Clin N Am. 1997;7:477-498
- Walker T, Salazar G, Waltman A. Angiographic evaluation and management of acute gastrointestinal hemorrhage World J Gastroenterol 2012;18(11):1191-1201
- Garcia-Pagan J, Caca K, Bureau C, et al. Early use of TIPS in patients with cirrhosis and variceal bleeding. N Engl J Med 2010;362:2370-2379
- Garcia-Pagan J, Di Pascoli M, Caca K, et al. Use of early TIPS for high risk variceal bleeding: results of a post-RCT surveillance study. J Hepatol 2013;58:45-50
- Jalan R, John T, Redhead D, et al. A comparative study of emergency transjugular intrahepatic portosystemic stent shunt and esophageal transaction in the management of uncontrolled variceal hemorrhage. Am J Gastroenterol 1995;90:1932-1937
- Savas N. Gastrointestinal endoscopy in pregnancy. World J. Gastroenterol 2014;20(41):15241-15252
- Hyett B, Abougergi M, Carpenter J, et al. The AIM 65 score compared with the Glasgow-Blatchford score in predicting outcomes in upper GI bleeding. Gastrointest Endosc 2013;77:551-557
- Stanley A. Update on risk scoring systems for patients with upper gastrointestinal haemorrhage. World J Gastroenterol 2012;18(22):2739-2744
- Chan F, Chung S, Suen B, et al. Preventing recurrent upper gastrointestinal bleeding in patients with Helicobacter pylori infection who are taking low-dose aspirin or naproxen. N Engl J Med 2001;344:967-973
- Lai K, Hui W, Wong W, et al. Treatment of Helicobacter pylori in patients with duodenal ulcer hemorrhage–a long-term randomized, controlled study. Am J Gastroenterol 2000;356:1318-1321