Bleeding from the gastrointestinal (GI) tract may present in five ways. Hematemesis is vomitus of red blood or “coffee-grounds” material. Melena is black, tarry, foul-smelling stool. Hematochezia is the passage of bright red or maroon blood from the rectum. Occult GI bleeding (GIB) may be identified in the absence of overt bleeding by a fecal occult blood test or the presence of iron deficiency. Finally, patients may present only with symptoms of blood loss or anemia such as lightheadedness, syncope, angina, or dyspnea.
Sources of Gastrointestinal Bleeding
Upper Gastrointestinal Sources of Bleeding
(Table 41-1) The annual incidence of hospital admissions for upper GIB (UGIB) in the United States and Europe is ∼0.1%, with a mortality rate of ∼5–10%. Patients rarely die from exsanguination; rather, they die due to decompensation from other underlying illnesses. The mortality rate for patients <60 years in the absence of major concurrent illness is <1%. Independent predictors of rebleeding and death in patients hospitalized with UGIB include increasing age, comorbidities, and hemodynamic compromise (tachycardia or hypotension).
Table 41-1 Sources of Bleeding in Patients Hospitalized for Upper GI Bleeding |Favorite Table|Download (.pdf)
Table 41-1 Sources of Bleeding in Patients Hospitalized for Upper GI Bleeding
|Sources of Bleeding||Proportion of Patients, %|
|No source identified||5–14|
Peptic ulcers are the most common cause of UGIB, accounting for up to ∼50% of cases; an increasing proportion is due to nonsteroidal anti-inflammatory drugs (NSAIDs), with the prevalence of Helicobacter pylori decreasing. Mallory-Weiss tears account for ∼5–10% of cases. The proportion of patients bleeding from varices varies widely from ∼5 to 40%, depending on the population. Hemorrhagic or erosive gastropathy (e.g., due to NSAIDs or alcohol) and erosive esophagitis often cause mild UGIB, but major bleeding is rare.
In addition to clinical features, characteristics of an ulcer at endoscopy provide important prognostic information. One-third of patients with active bleeding or a nonbleeding visible vessel have further bleeding that requires urgent surgery if they are treated conservatively. These patients clearly benefit from endoscopic therapy with bipolar electrocoagulation; heater probe; injection therapy (e.g., absolute alcohol, 1:10,000 epinephrine); and/or clips with reductions in bleeding, hospital stay, mortality rate, and costs. In contrast, patients with clean-based ulcers have rates of recurrent bleeding approaching zero. If there is no other reason for hospitalization, such patients may be discharged on the first hospital day, following stabilization. Patients without clean-based ulcers should usually remain in the hospital for three days because most episodes of recurrent bleeding occur within three days.
Randomized controlled trials document that a high-dose, constant-infusion IV proton pump inhibitor (PPI) (e.g., omeprazole 80-mg bolus and 8-mg/h infusion), designed to sustain intragastric pH > 6 and enhance clot stability, decreases further bleeding and mortality in patients with high-risk ulcers (active bleeding, nonbleeding visible vessel, adherent clot) when given after endoscopic therapy. Institution of PPI therapy at presentation in all patients with UGIB decreases high-risk ulcer characteristics (e.g., active bleeding) but does not significantly improve outcomes such as further bleeding, transfusions, or mortality as compared to initiating therapy only when high-risk ulcers are identified at the time of endoscopy.
Approximately one-third of patients with bleeding ulcers will rebleed within the next 1–2 years if no preventive strategies are employed. Prevention of recurrent bleeding focuses on the three main factors in ulcer pathogenesis, H. pylori, NSAIDs, and acid. Eradication of H. pylori in patients with bleeding ulcers decreases rates of rebleeding to <5%. If a bleeding ulcer develops in a patient taking NSAIDs, the NSAIDs should be discontinued, if possible. If NSAIDs must be continued or reinstituted, a cyclooxygenase 2 (COX-2) selective inhibitor (coxib) plus a PPI should be used. PPI co-therapy alone or a coxib alone is associated with an annual rebleeding rate of ∼10% in patients with a recent bleeding ulcer, while combination of a coxib and PPI provides a further significant decrease in recurrent ulcer bleeding. Patients with cardiovascular disease who develop bleeding ulcers while taking low-dose aspirin should restart aspirin as soon as possible after their bleeding episode (e.g., ≤ 7 days). A randomized trial showed that failure to restart aspirin was associated with a nonsignificant difference in rebleeding (5% vs. 10% at 30 days), but a significant increase in mortality at 30 days (9% vs. 1%) and 8 weeks (13% vs. 1%) as compared to immediate reinstitution of aspirin. Patients with bleeding ulcers unrelated to H. pylori or NSAIDs should remain on full-dose antisecretory therapy indefinitely. Peptic ulcers are discussed in Chap. 293.
The classic history is vomiting, retching, or coughing preceding hematemesis, especially in an alcoholic patient. Bleeding from these tears, which are usually on the gastric side of the gastroesophageal junction, stops spontaneously in 80–90% of patients and recurs in only 0–7%. Endoscopic therapy is indicated for actively bleeding Mallory-Weiss tears. Angiographic therapy with embolization and operative therapy with oversewing of the tear are rarely required. Mallory-Weiss tears are discussed in Chap. 292.
Patients with variceal hemorrhage have poorer outcomes than patients with other sources of UGIB. Endoscopic therapy for acute bleeding and repeated sessions of endoscopic therapy to eradicate esophageal varices significantly reduce rebleeding and mortality. Ligation is the endoscopic therapy of choice for esophageal varices because it has less rebleeding, a lower mortality rate, fewer local complications, and it requires fewer treatment sessions to achieve variceal eradication than sclerotherapy.
Octreotide (50-μg bolus and 50-μg/h IV infusion for 2–5 days) further helps in the control of acute bleeding when used in combination with endoscopic therapy. Other vasoactive agents such as somatostatin and terlipressin, available outside the United States, are also effective. Antibiotic therapy (e.g., ceftriaxone) is also recommended for patients with cirrhosis presenting with UGIB because antibiotics decrease bacterial infections and mortality in this population. Over the long term, treatment with nonselective beta blockers decreases recurrent bleeding from esophageal varices. Chronic therapy with beta blockers plus endoscopic ligation is recommended for prevention of recurrent esophageal variceal bleeding.
In patients who have persistent or recurrent bleeding despite endoscopic and medical therapy, more invasive therapy with transjugular intrahepatic portosystemic shunt (TIPS) is recommended. Older studies indicate that most patients with TIPS developed shunt stenosis within 1–2 years and required reintervention to maintain shunt patency. The use of coated stents appears to decrease shunt dysfunction by ∼50% in the first 2 years. A randomized comparison of TIPS (with uncoated stents) and distal splenorenal shunt in Child-Pugh class A or B cirrhotic patients with refractory variceal bleeding revealed no significant difference in rebleeding, encephalopathy, or survival, but had a much higher rate of reintervention with TIPS (82% vs. 11%). Therefore, decompressive surgery may be an option in patients with milder, well-compensated cirrhosis.
Portal hypertension is also responsible for bleeding from gastric varices, varices in the small and large intestine, and portal hypertensive gastropathy and enterocolopathy.
Hemorrhagic and Erosive Gastropathy (“Gastritis”)
Hemorrhagic and erosive gastropathy, often labeled gastritis, refers to endoscopically visualized subepithelial hemorrhages and erosions. These are mucosal lesions and, thus, do not cause major bleeding. They develop in various clinical settings, the most important of which are NSAID use, alcohol intake, and stress. Half of patients who chronically ingest NSAIDs have erosions (15–30% have ulcers), while up to 20% of actively drinking alcoholic patients with symptoms of UGIB have evidence of subepithelial hemorrhages or erosions.
Stress-related gastric mucosal injury occurs only in extremely sick patients: those who have experienced serious trauma, major surgery, burns covering more than one-third of the body surface area, major intracranial disease, or severe medical illness (i.e., ventilator dependence, coagulopathy). Significant bleeding probably does not develop unless ulceration occurs. The mortality rate in these patients is quite high because of their serious underlying illnesses.
The incidence of bleeding from stress-related gastric mucosal injury or ulceration has decreased dramatically in recent years, most likely due to better care of critically ill patients. Pharmacologic prophylaxis for bleeding may be considered in the high-risk patients mentioned above. Multiple trials document the efficacy of intravenous H2-receptor antagonist therapy, which is more effective than sucralfate but not superior to a PPI immediate-release suspension given via nasogastric tube. Prophylactic therapy decreases bleeding but does not lower the mortality rate.
Other less frequent causes of UGIB include erosive duodenitis, neoplasms, aortoenteric fistulas, vascular lesions [including hereditary hemorrhagic telangiectasias (Osler-Weber-Rendu) and gastric antral vascular ectasia (“watermelon stomach”)], Dieulafoy's lesion (in which an aberrant vessel in the mucosa bleeds from a pinpoint mucosal defect), prolapse gastropathy (prolapse of proximal stomach into esophagus with retching, especially in alcoholics), and hemobilia or hemosuccus pancreaticus (bleeding from the bile duct or pancreatic duct).
Small-Intestinal Sources of Bleeding
Small-intestinal sources of bleeding (bleeding from sites beyond the reach of the standard upper endoscope) are difficult to diagnose and are responsible for the majority of cases of obscure GIB. Fortunately, small-intestinal bleeding is uncommon. The most common causes in adults are vascular ectasias, tumors (e.g., adenocarcinoma, leiomyoma, lymphoma, benign polyps, carcinoid, metastases, and lipoma), and NSAID-induced erosions and ulcers. Other less common causes in adults include Crohn's disease, infection, ischemia, vasculitis, small-bowel varices, diverticula, Meckel's diverticulum, duplication cysts, and intussusception.
Meckel's diverticulum is the most common cause of significant lower GIB (LGIB) in children, decreasing in frequency as a cause of bleeding with age. In adults <40–50 years, small-bowel tumors often account for obscure GIB; in patients >50–60 years, vascular ectasias and NSAID-induced lesions are more commonly responsible.
Vascular ectasias should be treated with endoscopic therapy if possible. Surgical therapy can be used for vascular ectasias isolated to a segment of the small intestine when endoscopic therapy is unsuccessful. Although estrogen/progesterone compounds have been used for vascular ectasias, a double-blind trial found no benefit in prevention of recurrent bleeding. Isolated lesions, such as tumors, diverticula, or duplications, are generally treated with surgical resection.
Colonic Sources of Bleeding
The incidence of hospitalizations for LGIB is ≥20% that for UGIB. Hemorrhoids are probably the most common cause of LGIB; anal fissures also cause minor bleeding and pain. If these local anal processes, which rarely require hospitalization, are excluded, the most common causes of LGIB in adults are diverticula, vascular ectasias (especially in the proximal colon of patients >70 years), neoplasms (primarily adenocarcinoma), and colitis—most commonly infectious or idiopathic inflammatory bowel disease, but occasionally ischemic or radiation-induced. Uncommon causes include post-polypectomy bleeding, solitary rectal ulcer syndrome, NSAID-induced ulcers or colitis, trauma, varices (most commonly rectal), lymphoid nodular hyperplasia, vasculitis, and aortocolic fistulas. In children and adolescents, the most common colonic causes of significant GIB are inflammatory bowel disease and juvenile polyps.
Diverticular bleeding is abrupt in onset, usually painless, sometimes massive, and often from the right colon; minor and occult bleeding is not characteristic. Clinical reports suggest that bleeding colonic diverticula stop bleeding spontaneously in ∼80% of patients and rebleed in about 20–25% of patients. Intraarterial vasopressin or embolization by superselective technique should stop bleeding in a majority of patients. If bleeding persists or recurs, segmental surgical resection is indicated.
Bleeding from right colonic vascular ectasias in the elderly may be overt or occult; it tends to be chronic and only occasionally is hemodynamically significant. Endoscopic hemostatic therapy may be useful in the treatment of vascular ectasias, as well as discrete bleeding ulcers and post-polypectomy bleeding, while endoscopic polypectomy, if possible, is used for bleeding colonic polyps. Surgical therapy is generally required for major, persistent, or recurrent bleeding from the wide variety of colonic sources of GIB that cannot be treated medically, angiographically, or endoscopically.
Approach to the Patient: Gastrointestinal Bleeding
Measurement of the heart rate and blood pressure is the best way to initially assess a patient with GIB. Clinically significant bleeding leads to postural changes in heart rate or blood pressure, tachycardia, and, finally, recumbent hypotension. In contrast, the hemoglobin does not fall immediately with acute GIB, due to proportionate reductions in plasma and red cell volumes (i.e., “people bleed whole blood”). Thus, hemoglobin may be normal or only minimally decreased at the initial presentation of a severe bleeding episode. As extravascular fluid enters the vascular space to restore volume, the hemoglobin falls, but this process may take up to 72 h. Patients with slow, chronic GIB may have very low hemoglobin values despite normal blood pressure and heart rate. With the development of iron-deficiency anemia, the mean corpuscular volume will be low and red blood cell distribution width will increase.
Differentiation of Upper from Lower GIB
Hematemesis indicates an upper GI source of bleeding (above the ligament of Treitz). Melena indicates that blood has been present in the GI tract for at least 14 h (and as long as 3–5 days). The more proximal the bleeding site, the more likely melena will occur. Hematochezia usually represents a lower GI source of bleeding, although an upper GI lesion may bleed so briskly that blood does not remain in the bowel long enough for melena to develop. When hematochezia is the presenting symptom of UGIB, it is associated with hemodynamic instability and dropping hemoglobin. Bleeding lesions of the small bowel may present as melena or hematochezia. Other clues to UGIB include hyperactive bowel sounds and an elevated blood urea nitrogen level (due to volume depletion and blood proteins absorbed in the small intestine).
A nonbloody nasogastric aspirate may be seen in up to 18% of patients with UGIB—usually from a duodenal source. Even a bile-stained appearance does not exclude a bleeding postpyloric lesion because reports of bile in the aspirate are incorrect in ∼50% of cases. Testing of aspirates that are not grossly bloody for occult blood is not useful.
Diagnostic Evaluation of the Patient with GIB
(Fig. 41-1) History and physical examination are not usually diagnostic of the source of GIB. Upper endoscopy is the test of choice in patients with UGIB and should be performed urgently in patients who present with hemodynamic instability (hypotension, tachycardia, or postural changes in heart rate or blood pressure). Early endoscopy is also beneficial in cases of milder bleeding for management decisions. Patients with major bleeding and high-risk endoscopic findings (e.g., varices, ulcers with active bleeding or a visible vessel) benefit from endoscopic hemostatic therapy, while patients with low-risk lesions (e.g., clean-based ulcers, nonbleeding Mallory-Weiss tears, erosive or hemorrhagic gastropathy) who have stable vital signs and hemoglobin, and no other medical problems, can be discharged home.
Suggested algorithm for patients with acute upper gastrointestinal bleeding. Recommendations on level of care and time of discharge assume patient is stabilized without further bleeding or other concomitant medical problems. ICU, intensive care unit; PPI, proton pump inhibitor.
(Fig. 41-2) Patients with hematochezia and hemodynamic instability should have upper endoscopy to rule out an upper GI source before evaluation of the lower GI tract. Patients with presumed LGIB may undergo early sigmoidoscopy for the detection of obvious, low-lying lesions. However, the procedure is difficult with brisk bleeding, and it is usually not possible to identify the area of bleeding. Sigmoidoscopy is useful primarily in patients <40 years with minor bleeding.
Suggested algorithm for patients with acute lower gastrointestinal bleeding.*Some suggest colonoscopy for any degree of rectal bleeding in patients <40 years as well. †If massive bleeding does not allow time for colonic lavage, proceed to angiography.
Colonoscopy after an oral lavage solution is the procedure of choice in patients admitted with LGIB unless bleeding is too massive or unless sigmoidoscopy has disclosed an obvious actively bleeding lesion. 99mTc-labeled red cell scan allows repeated imaging for up to 24 h and may identify the general location of bleeding. However, radionuclide scans should be interpreted with caution because results, especially from later images, are highly variable. In active LGIB, angiography can detect the site of bleeding (extravasation of contrast into the gut) and permits treatment with embolization or intraarterial infusion of vasopressin. Even after bleeding has stopped, angiography may identify lesions with abnormal vasculature, such as vascular ectasias or tumors.
Obscure GIB is defined as persistent or recurrent bleeding for which no source has been identified by routine endoscopic and contrast x-ray studies; it may be overt (melena, hematochezia) or occult (iron-deficiency anemia). Current guidelines suggest angiography as the initial test for massive obscure bleeding, and video capsule endoscopy, which allows examination of the entire small intestine, for all others. Push enteroscopy, with a specially designed enteroscope or a pediatric colonoscope to inspect the entire duodenum and part of the jejunum, also may be considered as an initial evaluation. A systematic review of 14 trials comparing push enteroscopy to capsule revealed “clinically significant findings” in 26% and 56% of patients, respectively. However, in contrast to enteroscopy, lack of control of the capsule prevents its manipulation and full visualization of the intestine; in addition, tissue cannot be sampled and therapy cannot be applied.
If capsule endoscopy is positive, management (e.g., enteroscopy, laparoscopy) is dictated by the finding. If capsule is negative, current recommendations suggest patients may be either observed, or if their clinical course mandates (e.g., recurrent bleeding, need for transfusions or hospitalization), undergo further testing. Newer endoscopic techniques (e.g., double-balloon, single-balloon, and spiral enteroscopy) allow the endoscopist to examine, obtain specimens from, and provide therapy to much or all of the small intestine. Newer imaging techniques (CT and MR enterography) are now frequently being used in place of older specialized small-bowel radiographic exams (e.g., enteroclysis). Other tests include 99mTc-labeled red blood cell scintigraphy; angiography, which may be useful even if bleeding has subsided because it may disclose vascular anomalies or tumor vessels; and 99mTc-pertechnetate scintigraphy for diagnosis of Meckel's diverticulum (especially in young patients). When all tests are unrevealing, intraoperative endoscopy is indicated in patients with severe recurrent or persistent bleeding requiring repeated transfusions.
Positive Fecal Occult Blood Test
Fecal occult blood testing is recommended only for colorectal cancer screening and may be used in average-risk adults (beginning at age 50) and in adults with a first-degree relative with colorectal neoplasm at ≥60 years or two second-degree relatives with colorectal cancer (beginning at age 40). A positive test necessitates colonoscopy. If evaluation of the colon is negative, further workup is not recommended unless iron-deficiency anemia or GI symptoms are present.