Lower Gastrointestinal Bleeding in Adults
Lower Gastrointestinal Bleeding in Adults
Authors: John Sarko, MD, Clinical Attending Physician, Department of Emergency Medicine, Maricopa Medical Center, Phoenix, AZ; and Leigh Barrow, DO, Resident Physician, Emergency Medicine Residency, Maricopa Medical Center, Phoenix, AZ.
Peer Reviewer: Philip L. Henneman, MD, Professor of Emergency Medicine, Tufts–Baystate Medical Center, Boston, MA.
Many odors are noxious, but few are as repellent as the foul smell of massive hematochezia. Everybody in the emergency department knows something is wrong. Once you get past the smell, you realize you often have a very ill patient with a complex medical history and underlying comorbidities. Disposition decision is often easy ("ADMIT"), but to whom and where?
I had the pleasure of discussing the proposal of this article with the author before and during its creation. I challenged him to search the literature and find those nuggets of information from which the practicing clinician can obtain true guidance. I think he has done well and trust you will find this article useful.
J. Stephan Stapczynski, MD, Editor
Definition of the Problem and Epidemiology
Lower gastrointestinal bleeding (LGIB) is a complaint that all emergency physicians will face during their careers. Lower gastrointestinal bleeding is defined as bleeding below the Ligament of Treitz, and the spectrum ranges from minor hemorrhoidal bleeding to life-threatening hemorrhagic shock. Massive LGIB has been defined in various ways; most authors define it as an episode requiring a transfusion of 2 or more units of blood, a drop in hematocrit of 20% or more, or some combination of factors.1,2
Epidemiological data from a large health maintenance organization suggest that LGIB severe enough to warrant hospitalization occurs with an annual incidence of 20.5/100,000 people, with an average length of stay of 5.6 days.3 It appears to be more frequent in men (24.2/100,000 men versus 17.2/100,000 women). It is also predominately a disease of the elderly, with an average age of 63-66 years.4,5 In the 20- to 29-year-old age group, only 1/100,000 was hospitalized for LGIB, but that incidence increased to 205/100,000 in patients 80 years of age and older. In a Canadian study of LGIB, mean length of stay was 7.5 ± 12 days at an average cost of care of $4382 ± 7187.6 Older age was predictive of both higher costs and longer lengths of stay.
Mortality varies, depending on the population studied. Longstreth found a 2.4% mortality rate in patients who were admitted because of LGIB, but it increased to 23% in patients who developed an LGIB while hospitalized for other reasons.3 Other studies reported in-hospital mortality rates of 3.6%-18%7-12 for all patients hospitalized because of an LGIB. Patients with massive bleeding have death rates between 2.8-20.5%.4,5 Similar to those with upper gastrointestinal bleeding (UGIB), mortality for LGIB appears to be related to the patient's comorbidities, exacerbated by the hemorrhage.3,7,9,12
The majority of bleeding stops spontaneously, but rebleeding can be a problem for 6-20% of patients.4,10,12,13
Etiology
While many studies have evaluated the location and cause of LGIB, there are limitations from the published data for definite conclusions. First, all studies involved hospitalized patients; therefore, the frequency of minor causes, such as bleeding from internal hemorrhoids, is underestimated. Second, some studies limited the evaluation to patients with massive hemorrhage or bleeding requiring intensive care unit (ICU) treatment. Third, in some studies, the diagnosis was not certain, but the abnormality identified was presumed to be the cause.
Given these limitations, diverticular bleeding from the colon accounts for the greatest number of published cases, between 26-56% of identified causes.3-7,10,11,13,14 (See Table 1.) The right colon was involved most often, followed by the descending colon.4 Vascular ectasias and cancer are the two next most common causes, accounting for 2.7-30% and 7-26% of cases, respectively. Hemorrhoids are a common cause of LGIB in all ages, but the incidence increases with age.3-5,7,10,14 Lower gastrointestinal bleeding can also arise in the small bowel, accounting for 4.8-29% of cases of LGIB.4,10,15
 |
Risk Factors. In contrast to analyses of UGIB, risk factors for developing an LGIB are less clearly defined. Older age, however, is a risk factor for LGIB; the epidemiological study by Longstreth3 found a 200-fold rise as age increased from the third through the ninth decades of life, and as noted, most other studies found an average age in the seventh decade.
Nonsteroidal anti-inflammatory drugs (NSAIDs) may be a risk factor. A case control study of patients with LGIB found that almost twice as many patients had used NSAIDs in the week before the bleed compared to hospitalized controls (60% vs. 34%). Second, after adjusting for age, race, and gender, the odds ratios (OR) for NSAID use was 2.6 (95% confidence interval 1.7-3.9).16 Other case control studies have found similar results.13 A systematic review of this topic found that NSAID use is associated with many lower gastrointestinal (GI) lesions, and most studies, though not all, did show an association with bleeding.17 This risk may be reduced some by the use of selective cyclooxygenase-2 inhibitors.18,19
Alcohol, a factor in UGIB, may also be a risk factor for LGIB. In the American College of Gastroenterology Bleeding Registry13 the use of alcohol was associated with LGIB (OR of 2.1 (95% confidence interval 1.5-2.9)), and the use of both alcohol and NSAIDs increased the adjusted (OR of 4.5 (95% confidence interval 2.7-7.3)).
Research efforts have focused on which patients are at increased risk for adverse outcomes, such as rebleeding, severe bleeding, or death. Rebleeding, which is bleeding that occurs after the initial episode has ceased, has been defined in several ways: the passage of red stool after the patient has been stabilized, a drop of hematocrit of greater than 10% after stabilization, or a need for transfusions after stabilization. Defined these various ways, rebleeding occurs in 6-20% of patients.
Some studies tried to identify risk factors present on initial assessment that predict severe bleeding. A hematocrit of less than 35%, systolic blood pressure of under 100 mm Hg, heart rate greater than 100 per minute 1 hour after initial medical evaluation, and gross blood on rectal examination were associated with severe bleeding.2 In this study, though, the definition of severe bleeding included the abnormal vital signs identified as predictors.
Similar findings were noted in derivation and validation studies by Strate et al.1,20 Heart rate over 100/minute, systolic blood pressure less than 115 mm Hg, a nontender abdomen, syncope, bleeding in the first 4 hours of evaluation, aspirin use, and more than 2 comorbidities were all independent risk factors. Gross blood on rectal examination was a significant predictor in the univariate analysis, but not in the multivariate analysis. Dividing patients into low (no risk factors), moderate (1-3 risk factors), and high risk (> 3 risk factors) groups predicted increased risk of rebleeding.
Mortality of patients with GI bleeding (either upper or lower) was studied in 103 patients who were divided into high and low risk groups. The low risk group, identified as patients with mean arterial pressure of over 60 mm Hg and systolic blood pressure of over 100 mm Hg before ICU admission; no need for the use of vasopressors; no evidence for other organ system dysfunction that would require the use of an ICU, eg., acute myocardial infarction; and no active GI hemorrhage at the time of emergency department (ED) presentation (defined as the absence of hematemesis, red blood in the nasogastric lavage, and no non-hemorrhoidal bright red blood per rectum), had a 0% mortality rate. In contrast, the high-risk group (any patient not meeting all of these criteria) had a 21.3% mortality rate.21
Pathophysiology
Diverticular Disease. Diverticulosis occurs in approximately 30-50% of adults in industrialized nations and is the most common cause of LGIB, with rare occurrences of bleeding in diverticulitis.4 A diverticulum is a small protrusion of the colonic mucosa, usually where a blood vessel perforates the muscular layer of the colonic wall. Diverticular disease is most common in the left colon, but bleeding is more common from diverticula located in the right colon.4 Diverticular hemorrhage is characterized by either painless bleeding or mild discomfort due to colonic spasm from the irritation of intraluminal bleeding. The bleeding is self-limited in 70-80% of cases but has a rebleeding rate of approximately 25%.22 Patients prone to diverticulosis include those with chronic nonsteroidal anti-inflammatory drug and aspirin use, constipation, diets lacking in fiber, and advanced age.
Angiodysplasia. Angiodysplasia is a term that describes tortuous and dilated submucosal vessels thought to form because of chronic episodic partial obstruction of the submucosal veins from colonic muscle contraction.22 They are found most commonly in the cecum or ascending colon. The term angiodysplasia includes arteriovenous malformations (AVMs), angiomas, and vascular ectasia. Bleeding from angiodysplastic lesions is normally self-limited, painless, and episodic. As opposed to the arterial bleeding seen with diverticular disease, angiodysplastic bleeding is venous in origin and, therefore, usually involves a smaller volume of blood loss. There is, however, an approximately 80% chance of rebleeding in untreated angiodysplasia. Although initial reports suggested a possible reduction in rebleeding with estrogen and progesterone therapy, randomized trials suggest that there is no benefit.23 Long-term use of octreotide may prevent rebleeding from gastrointestinal angiodysplasia.24
Neoplasm. Neoplasm is most common in patients over 50 years old with recurrent low grade or occult LGIB. The bleeding, a result of overlying ulceration or erosion of the neoplasm itself, is normally self-limited. Patients often present with iron deficiency anemia, fatigue, a change in the caliber or frequency of stools, and weight loss. Physical examination findings might include a palpable tumor mass, a Sister Mary Joseph nodule of the umbilicus (a metastatic umbilical mass due to an intra-abdominal neoplasm), or left supraclavicular lymphadenopathy. Patients with a history of inflammatory bowel disease are at greater risk for developing colonic neoplasms. A thorough family history may identify familial cancer syndromes. Rectal carcinomas can be palpated on digital rectal examination approximately 40% of the time.25
Colitis. In contrast to the causes of LGIB mentioned above, colitis typically presents as painful bleeding, often associated with dehydration and fever. Colonoscopy reveals bowel mucosal edema, erythema, friability, and ulceration. Colitis can be categorized as infectious, ischemic, or inflammatory in nature. Recent travel suggests infectious colitis and is most commonly associated with Shigella, Salmonella, Campylobacter, invasive Escherichia coli 0157, and Clostridium difficile. Recent antibiotic usage would suggest antibiotic-associated colitis (Clostridium difficile). Inflammatory bowel disease refers to both ulcerative colitis and Crohn's disease. These diseases both present with similar diffuse lower abdominal pain and hematochezia, and are difficult to distinguish from each other. It is not mandatory to pinpoint a definitive diagnosis in the ED. The transmural inflammation and deep ulcers characteristic of Crohn's disease are thought to be the reasons massive hemorrhage is more common in Crohn's disease than ulcerative colitis.
Ischemic Colitis. Ischemic colitis is most common in elderly patients with a history of cardiac dysrhythmia, heart failure, sepsis, or hypovolemia who sustain an episode of hypotension.26 These patients often present with significant and continuous left-sided abdominal pain with sparing of the rectum, followed by an episode of mild to moderate bloody diarrhea associated with leukocytosis. Severe bleeding should prompt consideration of another etiology. Ischemic colitis should also be considered in all patients with atherosclerosis or vasculitis, patients with polycythemia vera (due to increased blood viscosity), and marathon runners.
One study reported that as many as 16% of runners had at least one bout of bloody diarrhea after a hard run or race.27 The "watershed areas" of the colon, such as the rectosigmoid junction or splenic flexure, are the most commonly affected due to reduced collateral circulation. Computerized tomography scan findings often include areas of low attenuation consistent with edema, wall thickening with heterogeneous enhancement, loss of colonic haustra, enhancement of the mucosa, and various degrees of pericolic streakiness. Large mesenteric vessels are not involved and, therefore, angiography is of little use in making this diagnosis. Treatment includes correcting the underlying cause of the ischemia. Although recovery is the expected result, approximately 15% of patients will develop an acute abdomen due to bowel necrosis. Antibiotics should be considered in patients with ischemic colitis and severe or recurrent bleeding.26
Anorectal Bleeding. Anorectal bleeding is commonly referred to as "outlet-type bleeding."28 The most common presentation is intermittent, minor bleeding associated with bowel movements. Hemorrhoids are the most common cause of painless anorectal bleeding (see below). Less commonly anorectal bleeding occurs from anal fissures, Dieulafoy's lesions (shallow ulcers around thick-walled arterial vessels), and rectal ulcers. Anal pain associated with bleeding is often due to anal fissures or thrombosed external hemorrhoids but can also be present with infectious causes, such as herpes, rectal trauma (including trans-rectal prostate biopsy), anal intercourse, and rectal carcinoma. Rectal ulcers often present with pain while straining during defecation, bleeding, the passage of mucus, and a feeling of incomplete evacuation. Anoscopy is often diagnostic and is recommended in all patients with anorectal bleeding.
Hemorrhoids. It has been estimated that 10.4 million Americans are afflicted by hemorrhoidal symptoms annually (4.4% prevalence).29 They are more common in pregnancy, the post-partum period, patients with constipation, and Caucasians. Hemorrhoids are most common in the 45-65 year age group, and are the most common cause of rectal bleeding in adult patients younger than 30 years old. Hemorrhoids are cushions of connective tissue in direct contact with arteriovenous communications of the superior rectal arteries and the superior, inferior, and middle rectal veins. They are classified as external (below the dentate line) or internal (above it). External hemorrhoids are usually painless until they become thrombosed, at which time they cause a painful palpable perianal lump. Internal hemorrhoids are often asymptomatic but can cause pruritus and bright red blood coating stool at the end of defecation. Blood mixed in with stool suggests an alternative diagnosis such as diverticulosis.30 Internal hemorrhoids cause the greatest degree of hematochezia due to their blood supply from the superior and middle hemorrhoidal arteries. Asking the patient to bear down during the anal examination can induce prolapse of a hemorrhoid or bleeding of a surface lesion.
Radiation Telangiectasia or Proctitis. Patients who have undergone radiation therapy of pelvic and abdominal cancers are prone to mild diarrhea, intermittent rectal bleeding, the passage of mucus, and rectal urgency following radiation. This can occur anywhere from 6 weeks following radiation up to 30 years after exposure. Cancer recurrence should be considered if a patient presents with LGIB years after radiation therapy. Treatment or radiation proctitis includes sucralfate enemas, steroids, iron supplementation for chronic blood loss, topical formalin, hyperbaric oxygen, and antioxidant therapy with either vitamin E or C. Formalin produces chemical cauterization to stop active bleeding.
Polypectomy and Post-biopsy Bleeding. Bleeding occurs in 4.8% of removed or biopsied polyp cases.31 Usually self-limited, acute arterial bleeding can occur due to inadequate coagulation of the stalk of the polyp or disruption of an underlying artery. Bleeding as late as 1 week following polypectomy can be attributed to sloughing of the coagulated eschar, and is usually self-limited.
Varices. Portal venous hypertension can cause anorectal varices to bleed, requiring resuscitative measures and blood transfusions. Patients with portal venous hypertension may have colonic varices, which are located proximal to the internal hemorrhoids, between the superior hemorrhoidal veins and the middle and inferior hemorrhoidal veins, several centimeters above the dentate line. Colonic varices are very rare but should be considered in patients with LGIB and signs of portal venous hypertension (spider angiomas, palmar erythema, splenomegaly, or ascites). Treatment is similar to that of esophageal varices, including sclerotherapy or portosystemic shunts.
Small Bowel Hemorrhage. Small bowel hemorrhage is an uncommon cause of apparent LGIB. Small bowel hemorrhage is most often due to angiodysplasia (70-80% of cases).22 Other causes include neoplasia, jejunoileal diverticula, Crohn's disease, Meckel's diverticulum, Dieulafoy's lesion (a gastrointestinal submucosal artery that has ruptured into the lumen and caused massive hemorrhage), and aortoenteric fistulas. The small bowel is technically difficult to evaluate for a source of bleeding because of its length and inaccessibility. Small bowel sources are investigated only after the colon and upper GI tracts have been thoroughly evaluated.32
Other Rare Causes of Lower GI Bleed. Aortoduodenal fistulas occur in patients with abdominal aortic aneurysms, most often after surgical repair. Bleeding typically starts with a small, herald bleed, followed by a massive hemorrhage. The classic triad of symptoms (abdominal pain, GI bleeding, and a pulsating abdominal mass) is found in only a minority of patients. Lower GI bleeding also occurs in patients with HIV, most often due to CMV colitis, lymphoma, and idiopathic colitis.33
Clinical Features
History. Patients with LGIB may be hemodynamically unstable, so it is often necessary to obtain a history and physical while simultaneously beginning resuscitation. Ask the patient about the duration, frequency, and color of the contents passed through the rectum. Hematochezia and maroon stool are most commonly associated with colonic bleeding, whereas melena is more often a result of an upper GI source (although small bowel or right sided colon bleeding can produce melena if the colon has delayed motility). Black stool is strongly associated with an upper gastrointestinal source (OR 16.6, 95% CI 7.7-35.734; likelihood ratio for presence of black stool = 5.1). Ingestion of charcoal, licorice, bismuth, and iron preparations can also cause black stool that can be confused with melena. Hemorrhoidal bleeding is typically bright red and may be described as dripping, squirting into, or filling the toilet bowl.
Also ask about prior abdominal surgery or radiation to the abdomen or pelvis. The presence of chest pain, dyspnea, light headedness or syncope, and current or recent medication use (especially NSAIDs, anticoagulants, beta-blockers), and comorbid diseases are also important to determine.
Physical Examination. Perform a directed physical examination starting with assessing the stability of the patient and severity of the bleed. Vital signs, preferably with orthostatics, are important to obtain. Note that patients taking beta-blocking drugs may not mount a tachycardic response to hypovolemia. Assess mental status to help determine the adequacy of perfusion. Observe skin color to assess for the presence of severe anemia or advanced liver disease. Cardiac and pulmonary examinations should be carried out. A murmur or arrhythmia noted on cardiac examination raises suspicion for ischemic colitis. An abdominal examination should focus on the presence of tenderness (suggests an inflammatory process or ischemic colitis), scars suggestive of prior abdominal vascular surgery, and masses. Perform an anorectal examination including a digital rectal examination and note stool color and consistency. Test the stool for blood with a guaiac card if gross blood is not present. Ask the patient to bear down because this can cause hemorrhoids to prolapse. Anoscopy will identify the presence of hemorrhoids and fissures and is usually well tolerated.
Nasogastric Aspiration. It is important to localize the bleeding site to the lower GI system. Blood in the stool (either red or maroon) may also be the presenting sign of an upper GI source. Melena typically indicates a UGIB, although bleeding in the ascending colon, coupled with a slow transit time, can also produce it. Traditionally, red or maroon blood from the rectum due to an upper GI source has been believed to be indicative of severe, brisk bleeding. However, few studies have been done to support or refute this. One such study examined patients known to have a UGIB.36 Fourteen percent of 727 patients with a known upper GI source had hematochezia or maroon stool on presentation. Nine of these patients (8.3%) had neither hematemesis nor a positive nasogastric (NG) aspirate for blood. Peptic ulcer disease was the most common reason for the UGIB. Compared to patients with melena and with brown stool, no differences were noted in admission vital signs, percent of patients in shock, hemoglobin concentration, or mortality. However, patients with hematochezia had a greater transfusion requirement and a higher rate of rebleeding than those with melena. Thus, this study suggests that patients with hematochezia and a known upper GI source are not necessarily sicker or more unstable on admission, but have a greater rate of complications during hospitalization.
Unfortunately, this study does not help the emergency physician decide if a patient with blood in the stool has an upper GI source. Only two studies have attempted to answer this question.34,37 Byers and colleagues studied 114 adults with bright red rectal bleeding or maroon stools without obvious bleeding hemorrhoids or anal fissures who had an NG aspirate tested and inspected for blood.37 Upper GI bleeding was identified in 11 patients (9.6%). These patients had lower initial hemoglobin levels and were more likely to have had a history of UGIB, but importantly, did not have a different blood pressure or heart rate. Symptoms of dyspepsia or epigastric pain also did not differ between the groups.37
The study by Witting, et al., attempted to identify predictors of an upper GI source in ED patients without hematemesis.34 Patients with bloody or Hemoccult® positive stools were included. Of 325 patients in the study, 128 had an upper GI source confirmed. The results are summarized in Table 2. The presence of black stool, age under 50 years, and a BUN/Cr ratio of greater than 29 were the strongest predictors of a UGIB. A history of a LGIB argued strongly against an upper GI source. As expected, the presence of a positive NG aspirate was the best predictor. Surprisingly, the presence of epigastric pain, hemodynamic instability, and the consistency of the stool (clots and tarry) were not independent predictors. It appears that an important percentage of patients with rectal blood have an upper GI source, and that they do not appear significantly different on presentation from those with a lower GI etiology. These two studies support the routine insertion of an NG tube to assess for a UGIB in patients presenting with hematochezia.
 |
Definitive Diagnosis
The emergency physician's primary goal is preserving or restoring vital organ perfusion, using intravenous fluids and blood products as necessary. Once an upper GI source has been ruled out with as much certainty as possible, and anorectal disease is considered to be unlikely, the precise localization of the source is usually left to the consultant, most often a gastroenterologist, surgeon, and/or radiologist. The emergency physician, though, can be instrumental in coordinating the patient's care and arranging for the proper test. The modalities available include colonoscopy, radionuclide imaging, and angiography. Newer possibilities include the use of helical CT scanning and wireless capsule endoscopy.
Colonoscopy. Colonoscopy is considered by many to be the initial diagnostic test.11,38-40 It has a success rate in identifying the cause of bleeding of 69-97%11,38,41 and is safe when performed in bleeding patients once they have been stabilized.38 It has traditionally been performed after a delay to allow for the proper preparation of the colon for endoscopy. However, data exist to show that urgently performed colonoscopy (within about 24 hours) leads to a definitive diagnosis more often than a delayed colonoscopy42-44 and to a shorter hospital stay.44-46 Preparation of the colon is with a polyethylene glycol-based purgative (e.g., GoLYTELY) given orally unless the patient is hemodynamically unstable or unable to take oral fluids. It can be given via an NG tube for those who are unable to take oral fluids. Four to six liters are given over 3-4 hours (approximately 250 cc every 15 minutes) orally or via NG tube to completely clear the colon. Colon preparation can be initiated in the ED, especially if there is a delay in getting the patient to the inpatient bed.
Radionuclide Scans. Radionuclide scanning is available at most hospitals. The most common type of scintigraphy uses technetium labeled red blood cells (TRBC scan). These tagged red blood cells have a long half life and can be detected on imaging 12-24 hours after administration. If the bleeding is very slow or intermittent, the tagged red blood cells will sometimes accumulate to allow a determination of the site of bleeding on a delayed image 24 hours after injection. The TRBC scan is believed to be able to detect hemorrhage rates as low as 0.1 cc/minute. Sensitivity and specificity of the test varies in the literature, from a sensitivity of 23-97%, a specificity of 76-95%, and overall accuracy from 41-94%.22,47
Even when positive, a site contradictory to that identified by the TRBC scan will be found in 10% of patients.47 A negative scan is not reassuring, as 27% of patients can rebleed after a negative scan.48 The yield may be increased in patients requiring larger transfusions47 and those who are hemodynamically unstable,49 and the false positive rate may be reduced by the use of subtraction scintigraphy.50 Because of this most authors use the results of a radionuclide scan to help make decisions about further diagnostic testing (e.g., angiography) rather than to help guide treatment.22
Conventional Arteriography. Conventional arteriography is also available at most centers. Advantages of arteriography are that it can localize the bleeding and allow treatment at the same time. It can also be done when colonoscopy is unsuccessful because of too much bleeding. Active bleeding at a rate of at least 0.5 cc/minute is required to be visible on arteriography. Complications include contrast induced renal failure, dislodgment of thrombus, hematoma at the puncture site, and contrast allergy. Arteriography does not appear to be as accurate as colonoscopy in localizing the bleeding sitereviews suggest a rate of 40-86%.22,38,43,51 Success rates are improved in patients with the more severe bleeds, as indicated by hemodynamic instability, a larger drop in hemoglobin, and a larger requirement for transfused blood.52
Multidetector CT Scans and Wireless Capsule Endoscopy. Future modalities for the assessment of LGIB may include multidetector CT scanning (see Figures 1 and 2) and wireless capsule endoscopy. The use of these modalities in LGIB has been evaluated in uncontrolled studies only, and are currently too new to be recommended for routine use by emergency physicians.
 |
 |
Emergency Department Resuscitation
Assessment for circulatory shock, using vital signs and evidence of poor organ perfusion, is the initial step in resuscitating patients with LGIB. Monitor the patient's cardiac function continuously, and administer supplemental oxygen at no less than 5 liters per minute. Patients who are hypotensive or tachycardic should have at least 2 large-bore peripheral IVs (at least 18 gauge) placed. Blood should be obtained for blood type and cross match, complete blood cell count (CBC), prothrombin time (PT), electrolytes and renal function (e.g., basic metabolic panel, BMP). Obtain an electrocardiogram for all patients with an abnormal cardiac examination, age greater than 50 years, risk factors for coronary artery disease, chest pain or palpitations, or severe anemia. Once the IVs are in place, unstable patients should receive 1-2 liters of crystalloid solution wide-open pending the arrival of the crossmatched blood. Reassess the patient, and if the patient remains or becomes unstable at this point, begin transfusing packed red cells. If crossmatched blood is unavailable and the patient is in shock, consider the use of universal donor (O-negative) red blood cells. Also correct any coagulopathy or thrombocytopenia with fresh frozen plasma and platelets, respectively. Patients on warfarin may need fresh frozen plasma and vitamin K to reverse their anticoagulation.
If the patient is in shock, place a urinary catheter for close monitoring of the urinary output. Keep the patient "nil per os" (NPO). A nasogastric (NG) tube should be placed to determine if the source of bleeding is the upper GI tract. The return of fluid believed to be bile (yellow or green color) makes an upper GI source less likely. Once the gastric sample is obtained, the NG tube can be removed. False negative gastric lavages may be seen, however, if no bile is noted, because a closed pylorus may prevent aspiration of duodenal blood and/or bile, or if bleeding has ceased. If a strong suspicion for an upper GI source still exists (based upon history), upper GI endoscopy should be performed. Obtain consultation early (see Consultation, below). Admission to an ICU is appropriate in all unstable patients.
Stable Patients. If the patient appears stable, perform a thorough history and physical examination first, along with anoscopy. Consider a nasogastric lavage if it cannot be determined if the bleeding is upper or lower. Obtain blood for a CBC, BMP, and PT/INR, and type and screen if the patient shows signs or symptoms of anemia, or has a source other than minor ano-rectal bleeding. Stable patients with minor anorectal bleeding can be discharged (see Clinical Decision Making, below).
Definitive Treatment
Definitive treatment of identified lesions will be undertaken by the consulting physician. Colonoscopy allows for cautery of bleeding lesions and biopsy of suspicious abnormalities. Lower gastrointestinal bleeding, like UGIB, usually stops spontaneously (approximately 80%11) and colonoscopy allows for visualization of these sites. Treatment through the colonoscope may reduce rebleeding, need for surgery, and length of stay,53 although other studies suggest no differences in the need for surgery or death.42,44
Superselective transcatheter embolization is a technique gaining support in the treatment of actively bleeding lower GI lesions. A 2.5-3 French catheter is placed into a small artery, such as the vasa recta or marginal artery that has been identified as the bleeding vessel, and hemorrhage is arrested with the use of embolic substances such as metal coils, polyvinyl alcohol, or gel foam.54 Ideally, blood flow to the bleeding site is reduced enough to allow thrombosis to occur, but not reduced so much that ischemic bowel develops.
Studies have found this procedure to be generally safe and effective. Technical success, the immediate cessation of bleeding, occurs in 85-100% of cases in which active bleeding is seen.55-59 Rebleeding occurs, though, in 7-26% of cases.55-59 Intestinal ischemia is reported in 4.5-25% of patients, with about one-quarter requiring surgery.55-59 Of course, a bleeding site must be identified in order for an embolization procedure to be performed.
Surgery is generally reserved for cases of continued or recurrent bleeding that is not amenable to treatment with colonoscopy or angiography. It may also be an initial therapeutic option when patients are experiencing a severe, exsanguinating hemorrhage.38 Ultimately, surgery is needed in 10-25% of patients.22 Prior studies suggest that surgery is more likely to be necessary if more than 4 units of blood are transfused in a 24-hour period.60 It is important to attempt to identify a site of bleeding before operation. If one is found, a subtotal colectomy can be preformed, with a low incidence of rebleeding and death.60 On the other hand, if no site is located, a total colectomy may be required, with a 27% mortality rate.61
Disposition
Clinical Decision-Making. Several algorithms have been proposed to assist with decision making in patients with LGIB (see Figures 3A and 3B).32,40,62-64 For the emergency physician, the initial decisions to be made are to determine if the patient is unstable or having a severe bleed. When this is so, resuscitation is the first priority. Two large-bore intravenous lines should be placed and crystalloid infusions started. A nasogastric tube should be passed, and if the return is positive for blood, the patient should be admitted and a gastroenterologist should be consulted for a UGIB. If the patient is determined to have a lower GI source and can be stabilized, consultation with a gastroenterologist is warranted (at some hospitals surgeons may also be consulted). If the patient is persistently unstable even after 2 liters of crystalloid fluid and cannot be stabilized with fluids and blood, a general surgeon should be promptly consulted. Management of the patient after this point occurs in conjunction with the specialist.
 |
Figure 3b (PDF).
If the patient has intermittent bleeding, or only has a small amount of blood, is hemodynamically stable, and has a stable hematocrit, a mild LGIB is the probable source, and anoscopy can be performed.38 If the patient is under 50 years of age and minor hemorrhoidal or anal fissure bleeding is identified, the patient can be discharged with stool softeners, dietary fiber, Sitz baths, and instructed to follow up with a primary care physician. Although a patient over 50 years of age may still be bleeding from hemorrhoids, the chances of a more serious etiology are increased, so these patients should be referred for outpatient colonoscopy.64
Consultation. The admitting physician, if not a gastroenterologist, should consult a gastroenterologist when a patient is hospitalized with an LGIB. Outpatients are referred to either a primary care physician or a gastroenterologist, depending on institutional practices. Consultation with a surgeon is recommended whenever more than 2 units of packed red blood cells are given in the ED, or there is clinical evidence of an acute surgical abdomen with fever, abdominal distension, guarding, and/or rebound.65 If a patient with a history of an abdominal aortic aneurysm, repaired or not, presents with a GI bleed, consultation with a vascular surgeon is prudent because of the possibility of an aortoenteric fistula.
Disposition. It is recommended to admit most patients with anything besides minor anorectal LGIB to the hospital. The specific unit the patient is admitted to depends upon institutional factors, such as the level of nursing care needed, hospital policies on colonoscopy and endoscopy, and available ancillary support, in addition to patient factors. Few evidence-based criteria exist to assist in this decision-making, and recommendations are based upon opinions, experience, and knowledge of the risk factors for an adverse outcome or severe bleed discussed earlier.1,2,4,10,12,13,20,21,66,67
One study published in 2000 considered a "legitimate" ICU admission to be a patient with a GI bleed (upper or lower) who developed a recurrent GI bleed, need for emergency surgery to control the bleeding, or death.67 Of the 411 patients, 81 had a poor outcome. The presence of hepatic cirrhosis (OR 2.27), active bleeding (OR 3.11), end organ dysfunction (angina, electrocardiographic evidence of infarction or ischemia, or syncope; OR 3.10), and Apache II scores derived from ED values (OR 1.08) were predictive of these outcomes. In this study, however, 37% of patients were admitted to an ICU setting, and the possible effect of this intensive treatment was not accounted for in the analysis.
Patients without any of the risk factors listed for severe bleeding, rebleeding, or death can usually be admitted to an unmonitored floor bed. It is not usually necessary to admit those with anemia by itself, if it is believed to be chronic, there is no other need for transfusion, there is no evidence for hypovolemia, and the vital signs are normal.66 Conversely, admit those with an LGIB and a fever or leukocytosis, unless there is definite cause to believe the patient has an infectious diarrhea. Admit to the ICU those who remain hypotensive despite ED resuscitation, are moderately or severely tachycardic, require more than 4 units of transfused blood, have unstable comorbid diseases, or are at high risk for rebleeding. Most physicians also admit a patient with ongoing bleeding to the ICU. Patients who fall in between these criteria can be admitted to either an ICU or step-down/intermediate care bed, depending on hospital and individual patient factors.
When a patient with minor bleeding is being discharged home, it is important to be sure the patient understands the signs and symptoms of severe bleeding and has access to transportation to return to the ED if necessary. The side effects of any medications prescribed should be explained, and the patient should be instructed to arrange for follow up in 1-2 days. Unless absolutely necessary, NSAIDs or aspirin should not be used.
Summary
Lower GI bleeding is a disease all emergency physicians will treat during their careers. Resuscitation is the key to a successful outcome, and this is where emergency physicians are the most skilled, and the most positioned to make a difference. An assessment of the severity of the bleeding and the risk for rebleeding should be made, and the source of the bleeding, upper GI or lower GI, should be determined. Definitive treatment is a multidisciplinary task, with gastroenterologists, surgeons, and interventional radiologists all playing vital roles.
References
1. Strate LL, Saltzman JR, Ookubo R, et al. Validation of a clinical prediction rule for severe acute lower intestinal bleeding. Am J Gastroenterol 2005;100:1821-1827.
2. Velayos FS, Williamson A, Sousa KH, et al. Early predictors of severe lower gastrointestinal bleeding and adverse outcomes: A prospective study. Clin Gastroenterol Hepatol 2004;2:485-490.
3. Longstreth GF. Epidemiology and outcome of patients hospitalized with acute lower gastrointestinal hemorrhage: A population-based study. Am J Gastroenterol 1997;92:419-424.
4. Leitman IM, Paull DE, Shires GT, 3rd. Evaluation and management of massive lower gastrointestinal hemorrhage. Ann Surg 1989;209:175-180.
5. Farrands PA, Taylor I. Management of acute lower gastrointestinal haemorrhage in a surgical unit over a 4-year period. J R Soc Med 1987;80:79-82.
6. Comay D, Marshall JK. Resource utilization for acute lower gastrointestinal hemorrhage: The Ontario GI bleed study. Can J Gastroenterol 2002;16:677-682.
7. Akhtar AJ. Lower gastrointestinal hemorrhage in African-American and Hispanic elderly patients. Ethn Dis 2002;12:379-382.
8. Wilcox CM, Clark WS. Association of nonsteroidal antiinflammatory drugs with outcome in upper and lower gastrointestinal bleeding. Dig Dis Sci 1997;42:985-989.
9. Rios A, Montoya MJ, Rodriguez JM, et al. Severe acute lower gastrointestinal bleeding: Risk factors for morbidity and mortality. Langenbecks Arch Surg 2007;392:165-171.
10. Wilcox CM, Clark WS. Causes and outcome of upper and lower gastrointestinal bleeding: The Grady Hospital experience. South Med J 1999;92:44-50.
11. Kok KY, Kum CK, Goh PM. Colonoscopic evaluation of severe hematochezia in an Oriental population. Endoscopy 1998;30:675-680.
12. Anthony T, Penta P, Todd RD, et al. Rebleeding and survival after acute lower gastrointestinal bleeding. Am J Surg 2004;188:485-490.
13. Peura DA, Lanza FL, Gostout CJ, et al. The American College of Gastroenterology Bleeding Registry: Preliminary findings. Am J Gastroenterol 1997;92:924-928.
14. Akhtar AJ. Lower gastrointestinal bleeding in elderly patients. J Am Med Dir Assoc 2003;4:320-322.
15. Rios A, Montoya MJ, Rodriguez JM, et al. Acute lower gastrointestinal hemorrhage originating in the small intestine. Rev Esp Enferm Dig 2006;98:196-203.
16. Wilcox CM, Alexander LN, Cotsonis GA, et al. Nonsteroidal antiinflammatory drugs are associated with both upper and lower gastrointestinal bleeding. Dig Dis Sci 1997;42:990-997.
17. Laine L, Smith R, Min K, et al. Systematic review: The lower gastrointestinal adverse effects of non-steroidal anti-inflammatory drugs. Aliment Pharmacol Ther 2006;24:751-767.
18. Laine L, Connors LG, Reicin A, et al. Serious lower gastrointestinal clinical events with nonselective NSAID or coxib use. Gastroenterology 2003;124:288-292.
19. Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: The CLASS Study: A randomized controlled trial. JAMA 2000;284:1247-1255.
20. Strate LL, Orav EJ, Syngal S. Early predictors of severity in acute lower intestinal tract bleeding. Arch Intern Med 2003;163:838-843.
21. Kollef MH, Canfield DA, Zuckerman GR. Triage considerations for patients with acute gastrointestinal hemorrhage admitted to a medical intensive care unit. Crit Care Med 1995;23:1048-1054.
22. Hoedema RE, Luchtefeld MA. The management of lower gastrointestinal hemorrhage. Dis Colon Rectum 2005;48:2010-2024.
23. Junquera F, Feu F, Papo M, et al. A multicenter, randomized, clinical trial of hormonal therapy in the prevention of rebleeding from gastrointestinal angiodysplasia.[see comment]. Gastroenterology 2001;121:1073-1079.
24. Junquera F, Saperas E, Videla S, et al. Long-term efficacy of octreotide in the prevention of recurrent bleeding from gastrointestinal angiodysplasia. Am J Gastroenterol 2007;102:254-260.
25. Steer ML, Silen W. Diagnostic procedures in gastrointestinal hemorrhage. N Engl J Med 1983;309:646-650.
26. Green BT, Tendler DA. Ischemic colitis: A clinical review. South Med J 2005;98:217-222.
27. Sullivan SN, Wong C. Runners' diarrhea. Different patterns and associated factors. J Clin Gastroenterol 1992;14:101-104.
28. Marderstein EL, Church JM. Classic "outlet" rectal bleeding does not require full colonoscopy to exclude significant pathology. Dis Colon Rectum 2008;51:202-206.
29. Johanson JF, Sonnenberg A. The prevalence of hemorrhoids and chronic constipation. An epidemiologic study. Gastroenterology 1990;98:380-386.
30. Madoff RD, Fleshman JW, Clinical Practice Committee AGA. American Gastroenterological Association technical review on the diagnosis and treatment of hemorrhoids. Gastroenterology 2004;126:1463-1473.
31. Levin TR, Zhao W, Conell C, et al. Complications of colonoscopy in an integrated health care delivery system. Ann Intern Med 2006;145:880-886.
32. Vernava AM, 3rd, Moore BA, Longo WE, et al. Lower gastrointestinal bleeding. Dis Colon Rectum 1997;40:846-858.
33. Bini EJ, Weinshel EH, Falkenstein DB. Risk factors for recurrent bleeding and mortality in human immunodeficiency virus infected patients with acute lower GI hemorrhage. Gastrointest Endosc 1999;49:748-753.
34. Witting MD, Magder L, Heins AE, et al. ED predictors of upper gastrointestinal tract bleeding in patients without hematemesis. Am J Emerg Med 2006;24:280-285.
35. Zuckerman GR, Trellis DR, Sherman TM, et al. An objective measure of stool color for differentiating upper from lower gastrointestinal bleeding. Dig Dis Sci 1995;40:1614-1621.
36. Wilcox CM, Alexander LN, Cotsonis G. A prospective characterization of upper gastrointestinal hemorrhage presenting with hematochezia. Am J Gastroenterol 1997;92:231-235.
37. Byers SE, Chudnofsky CR, Sorondo B, et al. Incidence of occult upper gastrointestinal bleeding in patients presenting to the ED with hematochezia. Am J Emerg Med 2007;25:340-344.
38. Zuccaro G, Jr. Management of the adult patient with acute lower gastrointestinal bleeding. American College of Gastroenterology. Practice Parameters Committee. Am J Gastroenterol 1998;93:1202-1208.
39. Savides TJ, Jensen DM. Acute Lower Gastrointestinal Bleeding. In: Friedman SL, Grendell J, McQuaid K, eds. Current Diagnosis and Treatment in Gastroenterology. 2nd ed. New York:McGraw-Hill; 2003.
40. Eisen GM, Dominitz JA, Faigel DO, et al. An annotated algorithmic approach to acute lower gastrointestinal bleeding. Gastrointest Endosc 2001;53:859-863.
41. Chaudhry V, Hyser MJ, Gracias VH, et al. Colonoscopy: The initial test for acute lower gastrointestinal bleeding. Am Surg 1998;64:723-728.
42. Green BT, Rockey DC, Portwood G, et al. Urgent colonoscopy for evaluation and management of acute lower gastrointestinal hemorrhage: A randomized controlled trial. Am J Gastroenterol 2005;100:2395-2402.
43. Richter JM, Christensen MR, Kaplan LM, et al. Effectiveness of current technology in the diagnosis and management of lower gastrointestinal hemorrhage. Gastrointest Endosc 1995;41:93-98.
44. Strate LL, Syngal S. Predictors of utilization of early colonoscopy vs. radiography for severe lower intestinal bleeding. Gastrointest Endosc 2005;61:46-52.
45. Strate LL, Syngal S. Timing of colonoscopy: Impact on length of hospital stay in patients with acute lower intestinal bleeding. Am J Gastroenterol 2003;98:317-322.
46. Schmulewitz N, Fisher DA, Rockey DC. Early colonoscopy for acute lower GI bleeding predicts shorter hospital stay: A retrospective study of experience in a single center. Gastrointest Endosc 2003;58:841-846.
47. Olds GD, Cooper GS, Chak A, et al. The yield of bleeding scans in acute lower gastrointestinal hemorrhage. J Clin Gastroenterol 2005;39:273-277.
48. Hammond KL, Beck DE, Hicks TC, et al. Implications of negative technetium 99m-labeled red blood cell scintigraphy in patients presenting with lower gastrointestinal bleeding. Am J Surg 2007;193:404-407.
49. Feingold DL, Caliendo FJ, Chinn BT, et al. Does hemodynamic instability predict positive technetium-labeled red blood cell scintigraphy in patients with acute lower gastrointestinal bleeding? A review of 50 patients. Dis Colon Rectum 2005;48:1001-1004.
50. Currie GM, Towers PA, Wheat JM. Improved detection and localization of lower gastrointestinal hemorrhage using subtraction scintigraphy: Clinical evaluation. J Nucl Med Technol 2007;35:105-111.
51. Edelman DA, Sugawa C. Lower gastrointestinal bleeding: A review. Surg Endosc 2007;21:514-520.
52. Abbas SM, Bissett IP, Holden A, et al. Clinical variables associated with positive angiographic localization of lower gastrointestinal bleeding. ANZ J Surg 2005;75:953-957.
53. Jensen DM, Machicado GA, Jutabha R, et al. Urgent colonoscopy for the diagnosis and treatment of severe diverticular hemorrhage. N Engl J Med 2000;342:78-82.
54. Funaki B. Superselective embolization of lower gastrointestinal hemorrhage: A new paradigm. Abdom Imaging 2004;29:434-438.
55. Sheth R, Someshwar V, Warawdekar G. Treatment of acute lower gastrointestinal hemorrhage by superselective transcatheter embolization. Indian J Gastroenterol 2006;25:290-294.
56. Kuo WT, Lee DE, Saad WE, et al. Superselective microcoil embolization for the treatment of lower gastrointestinal hemorrhage. J Vasc Interv Radiol 2003;14:1503-1509.
57. d'Othee BJ, Surapaneni P, Rabkin D, et al. Microcoil embolization for acute lower gastrointestinal bleeding. Cardiovasc Intervent Radiol 2006;29:49-58.
58. Rafique MZ, Ul Haq T, Ud Din GN, et al. Transcatheter embolization of acute non-variceal gastrointestinal hemorrhage. J Coll Physicians Surg Pak 2005;15:81-84.
59. DeBarros J, Rosas L, Cohen J, et al. The changing paradigm for the treatment of colonic hemorrhage: Superselective angiographic embolization. Dis Colon Rectum 2002;45:802-808.
60. McGuire HH Jr. Bleeding colonic diverticula. A reappraisal of natural history and management. Ann Surg 1994;220:653-656.
61. Bender JS, Wiencek RG, Bouwman DL. Morbidity and mortality following total abdominal colectomy for massive lower gastrointestinal bleeding. Am Surg 1991;57:536-540.
62. Green BT, Rockey DC. Lower gastrointestinal bleedingmanagement. Gastroenterol Clin North Am 2005;34:665-678.
63. Rockey DC. Lower gastrointestinal bleeding. Gastroenterology 2006;130:165-171.
64. Savides TJ, Jensen DM. Acute Lower Gastrointestinal Bleeding. In: Friedman SL, McQuaid K, and Grendell J, eds. Current Diagnosis & Treatment in Gastroenterology. 2nd ed. New York:McGraw-Hill; 2003.
65. Henneman PL. Gastrointestinal Bleeding. In: Marx JA, Hockberger RS, Walls RM, eds. Rosen's Emergency Medicine. 6th ed. Philadelphia:Mosby; 2006:220-227.
66. Terdiman JP, Lindenauer PK. Acute Gastrointestinal Bleeding. In: Wachter RM, Goldman L, Hollander H, eds. Hospital Medicine. 2nd ed. Philadelphia:Lippincott, Williams, and Wilkins; 2005:767-779.
67. Afessa B. Triage of patients with acute gastrointestinal bleeding for intensive care unit admission based on risk factors for poor outcome. J Clin Gastroenterol 2000;30:281-285.
Many odors are noxious, but few are as repellent as the foul smell of massive hematochezia. Everybody in the emergency department knows something is wrong. Once you get past the smell, you realize you often have a very ill patient with a complex medical history and underlying comorbidities. Disposition decision is often easy ("ADMIT"), but to whom and where?Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.