Section 5 Gastrointestinal Tract and Abdomen
6 Lower Gastrointestinal BleedingMichael J. Rosen, MD
Cleveland Clinic Foundation
Cleveland Clinic Health Sciences Center of Ohio State University
Director, Section of Endoscopic Surgery, Department of General Surgery
Cleveland Clinic Foundation
Approach to Lower GI Bleeding
Lower gastrointestinal bleeding is defined as abnormal hemorrhage into the lumen of the bowel from a source distal to the ligament of Treitz. In the majority of cases, lower GI bleeding derives from the colon; however, the small bowel is identified as the source of bleeding in as many as one third of cases,
1,2 and the upper GI tract is identified as the source in as many as 11% of patients presenting with bright-red blood per rectum.3Lower GI bleeding is more common in men than in women. The incidence rises steeply with advancing age, exhibiting a greater than 200-fold increase from the third decade of life to the ninth. This increase is largely attributable to the various colonic disorders commonly associated with aging (e.g., diverticulosis and angiodysplasia).4–6 The exact incidence of lower GI bleeding is not known, because there is no standardized technique for localizing it. Several investigators, however, estimate the incidence to be in the range of 20 to 27 cases per 100,000 adults.4,7 A 1997 survey of GI bleeding from the American College of Gastroenterology found that lower GI hemorrhage accounted for 24% of all GI bleeding events.8 Another study published the same year found that 0.7% of 17,941 discharges from a Veterans Affairs hospital were for patients who had had lower GI bleeding.9
The basic components of management are (1) initial hemodynamic stabilization, (2) localization of the bleeding site, and (3) site-specific therapeutic intervention. There are many conditions that can cause lower GI hemorrhage [see Discussion, Etiology of Lower GI Bleeding, below]; accordingly, successful localization depends on timely and appropriate use of a variety of diagnostic tests. Despite the abundance of diagnostic modalities available, attempts to localize the source of the hemorrhage fail in as many as 8% to 12% of patients.10,11 Once the bleeding site is localized, the appropriate therapeutic intervention must be carried out as expeditiously as possible.
Lower GI bleeding can be acute and life-threatening, chronic, or even occult. In what follows, we focus on severe, life-threatening hematochezia, reviewing the wide array of possible causes of lower GI bleeding and outlining the diagnostic and therapeutic modalities available for treating this difficult clinical problem.
Initial Evaluation and ResuscitationInitial evaluation of a patient with lower GI bleeding should include a focused history and physical examination, to be carried out simultaneously with resuscitation. Of particular importance in taking the history is to ascertain the nature and duration of the bleeding, including stool color and frequency. The patient should also be asked about any associated symptoms of potential significance (e.g., abdominal pain, changes in bowel habits, fever, urgency, tenesmus, or weight loss), as well as about relevant past medical events (e.g., previous GI bleeding episodes, injuries, surgical procedures, peptic ulcer disease, inflammatory bowel disease [IBD], and abdominal or pelvic irradiation). Any complicating comorbid conditions (e.g., heart or liver disease and clotting disorders) should be investigated. A comprehensive review of medications—in particular, nonsteroidal anti-inflammatory drugs (NSAIDs) and anticoagulants—is mandatory.12
The physical examination should include determination of postural vital signs so that intravascular volume status can be accurately estimated. A drop in the orthostatic blood pressure greater than 10 mm Hg or an increase in the pulse rate greater than 10 beats/min indicates that more than 800 ml of blood (> 15% of the total circulating blood volume) has been lost. Marked tachycardia and tachypnea in association with hypotension and depressed mental status indicates that more than 1,500 ml of blood (> 30% of the total circulating blood volume) has been lost. A complete abdominal examination, including digital rectal examination and anoscopy, should be performed.
Laboratory evaluation should include a complete blood count, measurement of serum electrolyte concentrations, a coagulation profile (prothrombin time and partial thromboplastin time) [see 1:4 Bleeding and Transfusion], and typing and crossmatching.
A nasogastric tube should be placed for gastric lavage. If lavage yields positive results (i.e., the aspirate contains gross blood or so-called coffee grounds), esophagogastroduodenos copy (EGD) is indicated [see 5:18 Gastrointestinal Endoscopy]. An aspirate that contains copious amounts of bile is strongly suggestive of a lower GI source of bleeding, and the workup proceeds accordingly [see Investigative Studies, below]. The choice is less clear-cut with a clear aspirate. In the absence of bile, such an aspirate cannot rule out a duodenal source for the bleeding. Accordingly, there is some degree of latitude for clinical judgment: depending on the overall clinical picture, the surgeon may choose either to perform EGD to rule out a duodenal bleeding source or to proceed with colonoscopy on the assumption that the source of the bleeding is in the lower GI tract.
Resuscitative efforts should begin immediately, with the aim of maintaining the patient in a euvolemic state. Two large-bore peripheral intravenous catheters should be inserted and isotonic I.V. fluid administered. A Foley catheter should be placed to facilitate monitoring of intravascular volume status. Whether and in what form to administer blood products is determined on an individual basis, with appropriate weight given to the presence or absence of comorbid conditions, the rate of blood loss, and the degree of hemodynamic stability. Severe hemodynamic instability may necessitate monitoring in the intensive care unit.
Investigative StudiesA number of diagnostic techniques are available for determining the source of lower GI hemorrhage, the most useful of which are colonoscopy [
see 5:18 Gastrointestinal Endoscopy], radionuclide scanning, computed tomography, and angiography (in the form of selective mesenteric arteriography). The goal of these tests is to locate the site of bleeding accurately so that definitive therapy can be properly directed. Which diagnostic test is chosen for a specific patient depends on several factors, including the hemodynamic stability of the patient, the bleeding rate, the comorbid conditions present, and the local expertise available at the physician's hospital.Colonoscopy
Several large series that evaluated the diagnostic utility of colonoscopy in patients with lower GI bleeding found this modality to be moderately to highly accurate, with overall diagnostic yields ranging from 53% to 97% [
see Table 1].3,13–17 Those studies that reported morbidity found colonoscopy to be safe as well, with an average complication rate of 0.5%. Colonoscopy has both a higher diagnostic yield and a lower complication rate than arteriography in this setting and thus would appear to be a more attractive initial test in most circumstances.3,18 An argument has been made—one with which we agree—that colonoscopy should be considered the procedure of choice for structural evaluation of lower GI bleeding and that arteriography should be reserved for patients with massive, ongoing bleeding in whom endoscopy is not feasible or colonoscopy fails to reveal the source of the hemorrhage.12The merits of colonic purging have been extensively debated in the literature.3,11,14 Although no firm conclusion has been reached, we feel that adequate colonic purging can improve both the diagnostic yield and the safety of colonoscopy. Given the absence of any definitive data suggesting that colonic purging either reactivates or increases bleeding,12 it is our practice to administer an oral purge after the patient has been adequately resuscitated.
If the entire colon has been adequately visualized and no source for the bleeding has been identified, the ileum should be intubated; fresh blood in this region suggests a possible small bowel source. If no active bleeding is observed in the ileum, upper GI endoscopy should be performed to rule out an upper GI bleeding site.
When colonoscopy and routine upper GI endoscopy fail to locate a bleeding source, push enteroscopy may be helpful. This procedure can be carried out in several ways. It can be performed purely endoscopically with a pediatric colonoscope. This approach generally requires a high level of skill on the part of the endoscopist, in that the lack of retroperitoneal attachments of the small intestine makes endoscopic navigation extremely challenging. In most cases, only the proximal 150 cm of the small intestine can be evaluated in this way. Alternatively, push enteroscopy can be performed in the operating room at the time of exploratory laparotomy. The surgeon can manually 'milk' the small bowel over the scope to evaluate its distal portion. In addition, an enterotomy can be made, and the scope can be passed in both a retrograde and an antegrade fashion so that the entire small intestine can be evaluated. Depending on the indication and on the technique employed, the diagnostic yield from push enteroscopy has ranged from 13% to 78%.19 Typically, yields are highest (40% to 60%) in patients with significant GI hemorrhage.
Radiolabeled Red Blood Cell Scanning
 |
| Figure 1. Collection of tracer at hepatic flexure |
Radionuclide scanning is highly sensitive for lower GI hemorrhage: it is capable of detecting bleeding at rates as slow as 0.1 to 0.4 ml/min.20 Two imaging tracers, both labeled with technetium-99m (99mTc), are currently available for radionuclide scanning in this setting: 99mTc-labeled sulfur colloid (99mTc-SC) and 99mTc-labeled red blood cells (RBCs). 99mTc-SC requires no preparation time and can be injected immediately into the patient; however, its rapid absorption into the liver and the spleen can often hinder accurate localization of overlying bleeding sites.9 At our institution, we prefer to use 99mTc-labeled RBCs. This agent requires some preparation time, but it has a much longer half-life than 99mTc-SC does, it is not taken up by the liver and spleen, and it can be detected on images as long as 24 to 48 hours after injection [see Figure 1].21,22
One study directly compared these two techniques and found 99mTc-labeled RBC scanning to have an accuracy of 93%, compared with an accuracy of only 12% for 99mTc-SC scanning.23 The high sensitivity of 99mTc-labeled RBC scanning—80% to 98%—is well attested, but there is considerable disagreement in the literature with regard to its specificity in identifying the anatomic site of bleeding.24–27 For example, on one hand, a 1996 study found radiolabeled RBC scanning to be 97% accurate for localizing bleeding in 37 patients undergoing surgical resection27; on the other hand, a 1990 study reported a 42% rate of incorrect resection when surgical therapy was based solely on this modality.26 In 2005, one group retrospectively reviewed 127 bleeding scans in an effort to identify factors that might predict a positive scan.28 The investigators found that tagged RBC scans were 48% accurate in localizing bleeding sites later confirmed by endoscopy, surgery, or pathologic evaluation. Multivariate analysis demonstrated that both the number of units of blood transfused in the 24 hours preceding the scan and the lowest recorded hematocrit differed significantly between patients with positive scans and those with negative scans. However, the clinical significance of a positive scan was unclear in this study, in that the rate of endoscopy was not significantly different between patients who had positive scans and those who did not.
To date, no prospective, randomized trials have compared radionuclide scanning with colonoscopy as the initial diagnostic procedure for patients with lower GI hemorrhage. In our view, however, given that radionuclide scanning (unlike colonoscopy and angiography) has no therapeutic intervention capabilities, its best use is in patients with non—life-threatening lower GI bleeding as a prelude and a guide to mesenteric angiography after active hemorrhage has been confirmed.
Computed Tomography
With the ongoing improvements in high-speed abdominal CT scanning, there has been growing interest in the evaluation of GI bleeding with CT.
29 Helical CT scanners can provide direct or indirect evidence of the source of GI bleeding. Typical findings that can facilitate localization of bleeding sites include spontaneous hyperdensity of the peribowel fat, contrast enhancement of the bowel wall, vascular extravasation of the contrast medium, thickening of the bowel wall, polyps, tumors, and vascular dilatation.CT evaluation of GI bleeding has several noteworthy advantages: the scanners typically are readily available, mobilization of special teams or units is not required, the scans can be completed rapidly in the emergency department, and bowel preparation is unnecessary. In one experimental study, CT scanners were able to detect arterial bleeding at rates as low as 0.07 ml/min, which suggests that CT scanning is more sensitive than angiography for this purpose.30 In addition, CT scans are noninvasive and carry little morbidity. Unfortunately, like radionuclide scanning, CT has no therapeutic capability.
A 2003 study of 19 patients with GI hemorrhage compared triphasic helical CT evaluation with colonoscopy and surgery for localization of bleeding sites.30 In this series, five patients had small bowel bleeding sites, and 14 had colonic sites. Helical CT scanning correctly identified four of the five small bowel lesions and 11 of the 14 colonic lesions. These findings, though preliminary, suggest that CT is a potentially valuable evaluation method in certain cases of GI bleeding. Perhaps CT scanning can eventually replace radionuclide scanning, which is often inaccurate. One potential drawback to the use of CT in this setting is the excessive dye load if angiography is employed as well.
Angiography
Selective Mesenteric Arteriography
 |
| Figure 2. Extravasation of contrast |
 |
| Figure 3. Injection of methlyene blue dye |
Selective mesenteric arteriography is somewhat less sensitive than radionuclide scanning for lower GI hemorrhage: bleeding must be occurring at a rate of at least 1.0 to 1.5 ml/min to be detect able with this test.31 The procedure involves percutaneous placement of a transfemoral arterial catheter for evaluation of the superior mesenteric, inferior mesenteric, and celiac arteries. A positive test result is defined as extravasation of contrast into the lumen of the bowel [see Figure 2]. Once the bleeding vessel has been localized angiographically, the area must be marked so that it can be successfully identified intraoperatively; this is commonly accomplished by infusing methylene blue into the bleeding artery [see Figure 3].32,33
In several large series [see Table 2], the overall diagnostic yield of arteriography ranged from 27% to 67%.27,34–38 The complication rate for arteriography performed for lower GI bleeding ranges from 2% to 4%.2,38 Reported complications include contrast allergy, renal failure, bleeding from arterial puncture, and embolism from a dislodged thrombus.12
Unlike radionuclide scanning, arteriography provides several therapeutic options, including vasopressin infusion and embolization of bleeding vessels. Nonetheless, given that arteriography has a lower diagnostic yield and a higher complication rate than colonoscopy does, it is reasonable to attempt colonoscopy first in patients with lower GI hemorrhage and to reserve angiography for patients in whom the volume of bleeding is such that colonoscopy would be neither safe nor accurate.
Provocative Angiography for Continued Obscure Bleeding
In a minority of patients, obscure bleeding persists despite negative findings from endoscopy, mesenteric arteriography, and radio labeled RBC scanning. This obscure bleeding presents a considerable diagnostic challenge, which some investigators have proposed addressing by means of so-called provocative angiography.39,40 Provocative angiography involves the use of short-acting anticoagulant agents (unfractionated heparin, vasodilators, thrombolytics, or combinations thereof) in association with angiography. Once the bleeding point has been localized, methylene blue is injected and the patient is immediately brought to the OR for surgical treatment. To date, unfortunately, little has been published on this technique, but it does appear to be a promising approach to this difficult problem.
ManagementAlthough, in the majority of cases, lower GI bleeding stops spontaneously, in a significant number of cases, hemorrhage continues and necessitates therapeutic intervention. Treatment options include endoscopic therapy, angiographic therapy, and surgical resection.
Endoscopic Therapy
When colonoscopy identifies a bleeding source, endoscopic treatment may be an option [
see 5:18 Gastrointestinal Endoscopy]. Endoscopic modalities used to treat lower GI bleeding include use of thermal contact probes,41,42 laser photocoagulation,43 electrocauterization,44 injection of vasoconstrictors, application of metallic clips,45 and injection sclerotherapy.46 The choice of a specific modality often depends on the nature of the offending lesion and on the expertise and resources available locally. A 1995 survey of members of the American College of Gastroenterology found that endoscopic therapy was used in 27% of patients presenting with lower GI bleeding.8Diverticular hemorrhage can be difficult to treat endoscopically because of the high bleeding rate and the location of the bleeding point within the diverticulum. In 2000, one group of investigators reported their experience with endoscopic therapy for severe hematochezia and diverticulosis in a prospective series of 121 patients.47 In this series, none of the patients treated endoscopically with epinephrine injections, bipolar coagulation, or both required surgery and none experienced recurrent bleeding episodes. A 2001 study from another group, however, reported high rates of recurrent bleeding episodes in both the early and the late posttreatment periods.48 In the absence of prospective, randomized trials, it is difficult to draw definitive conclusions about the utility of endoscopic therapy in treating diverticular hemorrhage.
Angiodysplasias resulting in GI hemorrhage typically are amen able to endoscopic treatment. That these lesions are frequently found in the right colon makes perforation a concern; this complication is reported in approximately 2% of patients.49 Good success rates have been reported with both injection and thermal methods.50 In one series, endoscopic fulguration was successful in 87% of patients, and no rebleeding episodes occurred over a 1- to 7-year follow-up period.50 Bleeding from multiple telangiectatic lesions in the distal colon resulting from radiation injury can be treated with thermal contact probes, lasers, or noncontact devices such as the argon plasma coagulator.51
Postpolypectomy hemorrhage can often be successfully treated by endoscopic means. Methods used include simple resnaring of the stalk while pressure is maintained52; electrocauterization, with or without epinephrine injection; endoscopic band ligation; and placement of metallic clips. For patients whose bleeding is attributable to benign anorectal causes, endoscopic therapy may include epinephrine injection, sclerosant injection, or band ligation of internal hemorrhoids.53
Angiographic Therapy
Diagnostic use of angiography in patients with lower GI bleeding can often be followed by angiographic therapy. The two main angiographic treatment options are intra-arterial injection of vasopressin and transcatheter embolization.
Vasopressin acts to control bleeding by causing arteriolar vasoconstriction and bowel wall contraction.
9 Once the bleeding site has been localized angiographically, the catheter is positioned in the main trunk of the vessel. Infusion of vasopressin is initiated at a rate of 0.2 U/min and can be increased to a rate of 0.4 U/min. Within 20 to 30 minutes, another angiogram is performed to determine whether the bleeding has ceased. If the bleeding is under control, the catheter is left in place and vasopressin is continuously infused for 6 to 12 hours. If the bleeding continues to be controlled, infusion is continued for an additional 6 to 12 hours at 50% of the previous rate. Finally, vasopressin infusion is replaced by continuous saline infusion, and if bleeding does not recur, the catheter is removed.54,55The vasoconstrictive action of vasopressin can have deleterious systemic side effects, including myocardial ischemia, peripheral ischemia, hypertension, dysrhythmias, mesenteric thrombosis, intestinal infarction, and death.9,36 Occasionally, simultaneous I.V. administration of nitroglycerin is necessary to counteract these systemic effects. The reported success rate of vasopressin in controlling lower GI bleeding ranges from 60% to 100%, and the incidence of major complications ranges from 10% to 20%.56–58 Rebleeding rates as high as 50% have been reported.57,58
An alternative for patients with coronary vascular disease, severe peripheral vascular disease, or other comorbidities that prevent safe administration of vasopressin is transcatheter embolization. In this technique, a catheter is superselectively placed into the identified bleeding vessel and an embolizing agent (e.g., a gelatin sponge, a microcoil, polyvinyl alcohol particles, or a balloon) is injected. Several small series found this technique to be 90% to 100% successful at stopping bleeding.59–63 Equally impressive was the finding that the rebleeding rates in these series were 0%. The complication rates of this procedure are generally reasonable as well; however, intestinal infarction has been reported.36,64
The use of small microcatheters and the ability to superselectively embolize individual vessels have reduced the potential for ischemic perforation. It is possible that as more experience is gained with these techniques, superselective embolization may replace catheter-directed vasoconstrictive therapy, thus obviating the potential deleterious systemic effects of vasopressin administration. Some researchers have suggested that with the exception of cases of diffuse bleeding lesions or cases whose demands exceed the technical limitations of superselective catheterization, embolization therapy should be the first choice for angiographic treatment of lower GI bleeding.65,66
Surgical Therapy
Although there are no absolute criteria for surgical treatment of lower GI bleeding, there are several factors—including hemodynamic status, associated comorbidities, transfusion requirements, and persistent bleeding—that are instrumental in making an appropriate and timely decision whether to operate. In general, patients who require more than 4 units of blood in a 24-hour period to remain hemodynamically stable, whose bleeding has not stopped after 72 hours, or who experience rebleeding within 1 week after an initial episode should undergo surgery.9
If the patient's hemodynamic status permits, surgical treatment should be undertaken after accurate localization of the bleeding site. When possible, directed segmental resection is the procedure of choice: it is associated with rebleeding rates ranging from 0% to 14% and mortality rates ranging from 0% to 13%.10,36,67 Blind segmental colectomy should never be performed: it is associated with rebleeding rates as high as 75% and mortality rates as high as 50%.68 If hemodynamic compromise and ongoing hemorrhage make it necessary to perform surgical exploration before the bleeding site can be localized, every effort should be made to identify the source of bleeding intraoperatively before embarking on resection. Intraoperative options for bleeding-site localization include colonoscopy (to allow for this option, patients should always be placed in the lithotomy position), EGD, and transoral passage of a pediatric colonoscope for enteroscopy with simultaneous intraperitoneal assistance for small bowel manipulation.9 If the bleeding site still cannot be accurately localized, subtotal colectomy is the procedure of choice. This procedure is associated with mortality rates ranging from 5% to 33%,69,70 which underscores the importance of accurate preoperative localization of bleeding before surgical intervention.
DiscussionEtiology of Lower GI Bleeding
As noted, lower GI bleeding has a wide array of possible causes [see Table 3].9,71 Of these, diverticular disease is the most common, accounting for 30% to 40% of all cases.72 Arteriovenous malformations (AVMs), though extensively described in the literature, are considerably less common causes, accounting for 1% to 4% of cases.73,74 Other significant causative conditions are IBD, benign and malignant neoplasms, ischemia, infectious colitis, anorectal disease, coagulopathy, use of NSAIDs, radiation proctitis, AIDS, and small bowel disorders.
Diverticular Disease
 |
| Figure 4. Bleeding diverticulum |
The reported prevalence of colonic diverticulosis in Western societies is 37% to 45%.75 The vast majority of colonic diverticula are actually false diverticula (pseudodiverticula) that contain only serosa and mucosa [see 5:12 Diverticulitis]. They occur at weak points in the colonic wall where the vasa recta penetrate the muscularis to supply the mucosa9; as the diverticulum expands, these vessels are displaced. A 1976 anatomic study of colonic specimens from patients with diverticular bleeding used angiography to demonstrate that in all cases, the vasa recta overlying the diverticulum ruptured into the lumen of the diverticulum, not into the peritoneum [see Figure 4].76
It has been estimated that approximately 17% of patients with colonic diverticulosis experience bleeding, which may range from minor to severe and life-threatening.77 As many as 80% to 85% of diverticular hemorrhages stop spontaneously.78 In one series, surgery was unlikely to be necessary if fewer than 4 units of packed RBCs were transfused in a 24-hour period, whereas 60% of patients receiving more than 4 units of packed RBCs in a 24-hour period required surgical intervention.5 The risk of a second bleeding episode is approximately 25%.3 Semielective surgical therapy is usually offered after a second diverticular bleeding episode because once a second such episode has occurred, the risk that a third will follow exceeds 50%.79 In a series of 83 conservatively managed cases of diverticular disease, the predicted yearly recurrence rates were 9% at 1 year, 10% at 2 years, 19% at 3 years, and 25% at 4 years.4
Colitis
The broad term colitis includes IBD, infectious colitis, radiation colitis, and idiopathic ulcers. IBD, in turn, includes Crohn disease [
see 5:11 Crohn Disease]. Patients with IBD usually present with bloody diarrhea that is not life-threatening; however, 6% to 10% of patients with ulcerative colitis have lower GI bleeding severe enough to necessitate emergency surgical resection,80,81 and 0.6% to 1.3% of patients with Crohn disease have acute life-threatening lower GI bleeding.80,82 In one review, 50% of patients with intestinal hemorrhage from IBD experienced spontaneous cessation of bleeding.80 Approximately 35% of patients whose bleeding stops without intervention will have another bleeding episode. Because of this high recurrence rate, semielective surgery is recommended after the first episode of severe GI bleeding secondary to IBD.Colitis caused by various infectious agents (e.g., Salmonella typhi,83,84Escherichia coli O157:H7,85Clostridium difficile,86 and cytomegalovirus) can result in severe lower GI bleeding, but this is a relatively rare occurrence.
Increasing use of radiation therapy to treat pelvic malignancies has led to a corresponding increase in the incidence of chronic radiation proctitis.87 Radiation therapy damages bowel mucosa, resulting in the formation of vascular telangiectases that are prone to bleeding.88 From 1% to 5% of cases of acute lower GI bleeding from radiation-induced proctocolitis are severe enough to necessitate hospitalization.4,14 In a survey of patients with prostate cancer who underwent pelvic irradiation, 5% of the patients reported hematochezia daily.89 Initial therapy for clinically significant hematochezia related to radiation proctitis should include some form of endoscopic treatment (e.g., argon-beam coagulation). Surgery should be reserved for unstoppable hemorrhage or other major complications, such as fistulas and strictures.87
Neoplasia
Significant GI bleeding from colorectal neoplasia [
see 5:15 Adenocarcinoma of the Colon and Rectum] accounts for 7% to 33% of cases of severe lower GI hemorrhage.3,11,14,36,90 Such bleeding is believed to result from erosions on the luminal surface.91 One report identified ulcerated cancers as the cause in 21% of cases of hematochezia.14 Adenomatous polyps are implicated in 5% to 11% of cases of acute lower GI bleeding.7,8,14,92,93 Lower GI hemorrhage, either immediate or delayed, is the most common reported complication after endoscopic polypectomy, occurring in 0.2% to 6% of cases.3,4,94,95 Immediate postpolypectomy bleeding is believed to result from incomplete coagulation of the stalk before transection.52 Delayed bleeding has been reported as long as 15 days after polypectomy and is thought to be secondary to sloughing of the coagulum; it is less common than immediate bleeding, occurring in only 0.3% of cases.14,52Coagulopathy
Lower GI bleeding can be a presenting symptom both for patients with iatrogenic coagulopathy from heparin or warfarin therapy and for patients with a hematologic coagulopathy from thrombocytopenia [
see 1:4 Bleeding and Transfusion]. It is unclear, however, whether severe coagulopathy leads to spontaneous hemorrhage or whether it predisposes to bleeding from an existing lesion.96,97 In an early series of leukemic patients with thrombocytopenia and severe GI hemorrhage, 50% of bleeding patients had platelet counts lower than 20,000/mm3 without any identifiable mucosal lesions; furthermore, when the platelet count rose above 20,000/mm3, the incidence of bleeding decreased to 0.8%.96 The investigator concluded that severe thrombocytopenia led to spontaneous GI hemorrhage. Other investigators subsequently challenged this conclusion, arguing that spontaneous bleeding from coagulopathy is in fact rare.98 In one report, the distribution of pathologic lesions in patients with GI bleeding who were taking heparin or warfarin was essentially equivalent to that in the general population.98 Regardless of what the precise relation between coagulopathy and GI hemorrhage may be, a thorough investigation for an anatomic lesion is imperative in the workup of patients with lower GI bleeding even in the face of coagulopathy or thrombocytopenia.Benign Anorectal Disease
Hemorrhoids, ulcer/fissure disease, and fistula in ano [
see 5:17 Benign Rectal, Anal, and Perineal Problems ] must not be overlooked as causes of GI hemorrhage: in one review comprising almost 18,000 cases of lower GI bleeding, 11% were attributable to ano rectal pathology. It is crucial to remember that identification of a benign anorectal lesion does not eliminate the possibility of a more proximal cause of hemorrhage. In general, patients with hemorrhoids identified on physical examination should still undergo thorough endoscopic evaluation of the colon to rule out other pathologic conditions.Portal hypertension [see 5:10 Portal Hypertension], congestive heart failure, and splenic vein thrombosis can cause colonic or anorectal varices, which can result in massive lower GI hemorrhage.99 The reported incidence of anorectal varices in patients with portal hypertension ranges from 78% to 89%.100,101 If local measures fail to control hemorrhage, some form of portosystemic shunting is indicated.
Colonic Arteriovenous Malformations
The term arteriovenous malformation includes vascular ectasias, angiomas, and angiodysplasias. AVMs are ectatic blood vessels seen in the mucosa and submucosa of the GI tract. They are degenerative lesions of the GI tract, occurring more frequently with advancing age.
9 In autopsy series, the reported incidence of colonic AVMs is 1% to 2%.102 In patients older than 50 years, the incidence of colonic AVMs is estimated to range from 2% to 30%.103–106 In healthy asymptomatic adults, the prevalence is estimated to be approximately 0.8%.107Colonic AVMs are believed to derive from chronic colonic wall muscle contraction, which leads to chronic partial obstruction of the submucosal veins, causing the vessels to become dilated and tortuous. This process eventually renders the precapillary sphincters incompetent, resulting in direct arterial-venous communication.108,109 Colonic AVMs are most commonly found in the cecum.10 They have been associated with several systemic diseases, including atherosclerotic cardiovascular disease, aortic stenosis, chronic renal disease, collagen vascular disease, von Willebrand disease, chronic obstructive pulmonary disease, and cirrhosis of the liver; to date, however, no definite causal relationship to any of these conditions has been established.6,21,44,110
 |
| Figure 5. Arteriovenous malformation |
The diagnosis of a colonic AVM is made at the time of angiography or colonoscopy. During angiography, visualization of ectatic, slow-emptying veins, vascular tufts, or early-filling veins establishes the diagnosis.111 During endoscopy, angiodysplasias appear as red, flat lesions about 2 to 10 mm in diameter, sometimes accompanied by a feeding vessel [see Figure 5].6,41,44,72
Typically, the bleeding caused by colonic AVMs is chronic, slow, and intermittent.9 Although these lesions can cause severe lower GI hemorrhage, they are a relatively uncommon cause: in most large series, they account for only about 2% of cases of acute bleeding.74,104 The bleeding stops spontaneously in 85% to 90% of cases,10 but it recurs in 25% to 85%.112 Accordingly, definitive surgical or colonoscopic treatment should be rendered once the lesion has been identified.
Colonic Ischemia
Acute lower GI bleeding can also be a presenting symptom of colonic ischemia. In several large series, colonic ischemia accounted for 3% to 9% of cases of acute lower GI hemorrhage.
4,7,8,14,92 Other vascular diseases reported as potential causes are poly arteritis nodosa, Wegener granulomatosis, and rheumatoid vasculitis.113,114 The resultant vasculitis can cause ulceration, necrosis, and ultimately hemorrhage.115Small Intestinal Sources
 |
| Figure 6a. Intraoperative specimen of small-bowel tumor (a) |
 |
| Figure 6b. Intraoperative specimen of small-bowel tumor (b) |
Small intestinal sources account for 0.7% to 9% of cases of acute lower GI bleeding.3,4,116–118 About 70% to 80% of cases of small bowel hemorrhage are attributable to AVMs; other, less common causes are jejunoileal diverticula, Meckel's diverticulum,119 neoplasia, regional enteritis, and aortoenteric fistulas [see Figures 6a and 6b].90,120,121 Accurate localization of a bleeding site in the small intestine can be highly challenging: the length and the free intraperitoneal position of the small bowel make endoscopic examination difficult, and the nature of the overlying loops makes angiographic localization imprecise. For these reasons, the small intestine is usually left for last in the attempt to localize the source of lower GI bleeding and is examined only after sources in the colon, the upper GI tract, and the anorectum have been ruled out.9
AIDS
The etiology of lower GI bleeding in patients with AIDS differs from that in the general population.
91 In AIDS patients, lower GI bleeding is caused predominantly by conditions related to the underlying HIV infection. Cytomegalovirus colitis is the most common cause of such bleeding in this population, occurring in 39% of cases.122 AIDS patients with hemorrhoids or anal fissures often experience significant bleeding as a result of HIV-induced thrombocytopenia.122 A 1998 study reported that in 23% of AIDS patients hospitalized for lower GI bleeding, benign anorectal disease was the cause.123 Other significant causes of lower GI hemorrhage in this population are colonic histoplasmosis, Kaposi sarcoma of the colon, and bacterial colitis.123,124NSAID Use
The association between NSAID use and upper GI hemorrhage is well known.
125 Current data suggest that NSAIDs have a toxic effect on colonic mucosa as well.126 An epidemiologic study estimated the incidence of NSAID-associated large bowel bleeding to be 7/100,000.127 A retrospective review found that patients who had experienced lower GI bleeding were twice as likely to have taken NSAIDs as those who had not.128 NSAIDs have also been linked to diverticular hemorrhage: in one study, 92% of patients with diverticular bleeding were taking NSAIDs.107 The exact mechanism of NSAID-induced colonic injury is unknown; nevertheless, heightened clinical awareness of this potential cause of lower GI bleeding is warranted.91References1. Briley CA Jr, Jackson DC, Johnsrude IS, et al: Acute gastrointestinal hemorrhage of small-bowel origin. Radiology 136:317, 1980 [PMID 6967614]
2. Koval G, Benner KG, Rosch J, et al: Aggressive an giographic diagnosis in acute lower gastrointestinal hemorrhage. Dig Dis Sci 32:248, 1987 [PMID 3493124]
3. Jensen DM, Machicado GA: Diagnosis and treatment of severe hematochezia: the role of urgent colonoscopy after purge. Gastroenterology 95:1569, 1988 [PMID 3263294]
4. Longstreth GF: Epidemiology and outcome of patients hospitalized with acute lower gastrointestinal hemorrhage: a population-based study. Am J Gas troenterol 92:419, 1997
5. McGuire HH Jr: Bleeding colonic diverticula: a reappraisal of natural history and management. Ann Surg 220:653, 1994 [PMID 7979613]
6. Foutch PG: Angiodysplasia of the gastrointestinal tract. Am J Gastroenterol 88:807, 1993 [PMID 8389094]
7. Bramley PN, Masson JW, McKnight G, et al: The role of an open-access bleeding unit in the management of colonic haemorrhage: a 2-year prospective study. Scand J Gastroenterol 31:764, 1996 [PMID 8858744]
8. Peura DA, Lanza FL, Gostout CJ, et al: The American College of Gastroenterology. Bleeding Registry: preliminary findings. Am J Gastroenterol 92:924, 1997 [PMID 9177503]
9. Vernava AM 3rd, Moore BA, Longo WE, et al: Lower gastrointestinal bleeding. Dis Colon Rectum 40:846, 1997 [PMID 9221865]
10. Boley SJ, DiBiase A, Brandt LJ, et al: Lower intestinal bleeding in the elderly. Am J Surg 137:57, 1979 [PMID 310250]
11. Caos A, Benner KG, Manier J, et al: Colonoscopy after Golytely preparation in acute rectal bleeding. J Clin Gastroenterol 8:46, 1986 [PMID 3486210]
12. Zuccaro G Jr: Management of the adult patient with acute lower gastrointestinal bleeding. American College of Gastroenterology Practice Parameters Committee. Am J Gastroenterol 93:1202, 1998 [PMID 9707037]
13. Richter JM, Christensen MR, Kaplan LM: Effectiveness of current technology in the diagnosis and management of lower gastrointestinal hemorrhage. Gastrointest Endosc 41:93, 1995 [PMID 7721024]
14. Rossini FP, Ferrari A, Spandre M, et al: Emergency colonoscopy. World J Surg 13:190, 1989 [PMID 2658365]
15. Goenka MK, Kochhar R, Mehta SK: Spectrum of lower gastrointestinal hemorrhage: an endoscopic study of 166 patients. Indian J Gastroenterol 12:129, 1993 [PMID 8270291]
16. Ohyama T, Sakurai Y, Ito M, et al: Analysis of urgent colonoscopy for lower gastrointestinal tract bleeding. Digestion 61:189, 2000 [PMID 10773724]
17. Chaudhry V, Hyser MJ, Gracias VH, et al: Colo n oscopy: the initial test for acute lower gastrointestinal bleeding. Am Surg 64:723, 1998 [PMID 9697900]
18. Cohn SM, Moller BA, Zieg PM, et al: Angiography for preoperative evaluation in patients with lower gastrointestinal bleeding: are the benefits worth the risks? Arch Surg 133:50, 1998 [PMID 9438759]
19. Lin S, Branch MS, Shetzline M, et al: The importance of indication in the diagnostic value of push enter oscopy. Endoscopy 35:315, 2003 [PMID 12664388]
20. Alavi A, Dann RW, Baum S, et al: Scintigraphic detection of acute gastrointestinal bleeding. Radiology 124:753, 1977 [PMID 302010]
21. Gupta N, Longo WE, Vernava AM 3rd: Angio dysplasia of the lower gastrointestinal tract: an entity readily diagnosed by colonoscopy and primarily managed nonoperatively. Dis Colon Rectum 38:979, 1995 [PMID 7656748]
22. McKusick KA, Froelich J, Callahan RJ, et al: 99mTc red blood cells for detection of gastrointestinal bleeding: experience with 80 patients. AJR Am J Roentgenol 137:1113, 1981 [PMID 6976077]
23. Bunker SR, Lull RJ, Hattner RS, et al: The ideal radiotracer in gastrointestinal bleeding detection. AJR Am J Roentgenol 138:982, 1982 [PMID 6282098]
24. Kester RR, Welch JP, Sziklas JP: The 99mTc-labeled RBC scan: a diagnostic method for lower gastrointestinal bleeding. Dis Colon Rectum 27:47, 1984 [PMID 6317306]
25. Suzman MS, Talmor M, Jennis R, et al: Accurate localization and surgical management of active low er gastrointestinal hemorrhage with technetium-labeled erythrocyte scintigraphy. Ann Surg 224:29, 1996 [PMID 8678614]
26. Hunter JM, Pezim ME: Limited value of technetium 99m-labeled red cell scintigraphy in localization of lower gastrointestinal bleeding. Am J Surg 159:504, 1990 [PMID 2334015]
27. Ng DA, Opelka FG, Beck DE, et al: Predictive value of technetium Tc 99m-labeled red blood cell scintigraphy for positive angiogram in massive lower gastrointestinal hemorrhage. Dis Colon Rectum 40:471, 1997 [PMID 9106699]
28. Olds GD, Cooper GS, Chak A, et al: The yield of bleeding scans in acute lower gastrointestinal hemorrhage. J Clin Gastroenterol 39:273, 2005 [PMID 15758618]
29. Yamaguchi T, Yoshikawa K: Enhanced CT for initial localization of active lower gastrointestinal bleeding. Abdom Imaging 28:634, 2003 [PMID 14628865]
30. Ernst O, Bulois P, Saint-Drenant S, et al: Helical CT in acute lower gastrointestinal bleeding. Eur Radiol 13:114, 2003 [PMID 12541118]
31. Baum S, Athanasoulis CA, Waltman AC: Angio-graphic diagnosis and control of large-bowel bleeding. Dis Colon Rectum 17:447, 1974 [PMID 4546724]
32. Athanasoulis CA, Moncure AC, Greenfield AJ, et al: Intraoperative localization of small bowel bleeding sites with combined use of angiographic methods and methylene blue injection. Surgery 87:77, 1980 [PMID 6965332]
33. Schrodt JF, Bradford WR: Presurgical angiographic localization of small bowel bleeding site with methylene blue injection. J Ky Med Assoc 94:192, 1996 [PMID 8935401]
34. Pennoyer WP, Vignati PV, Cohen JL: Management of angiogram positive lower gastrointestinal hemorrhage: long term follow-up of non-operative treatments. Int J Colorectal Dis 11:279, 1996 [PMID 9007623]
35. Rantis PC Jr, Harford FJ, Wagner RH, et al: Technetium-labelled red blood cell scintigraphy: is it useful in acute lower gastrointestinal bleeding? Int J Colorectal Dis 10:210, 1995 [PMID 8568406]
36. Leitman IM, Paull DE, Shires GT 3rd: Evaluation and management of massive lower gastrointestinal hemorrhage. Ann Surg 209:175, 1989 [PMID 2783842]
37. Casarella WJ, Galloway SJ, Taxin RN, et al: 'Lower' gastrointestinal tract hemorrhage: new concepts based on arteriography. Am J Roentgenol Radium Ther Nucl Med 121:357, 1974
38. Colacchio TA, Forde KA, Patsos TJ, et al: Impact of modern diagnostic methods on the management of active rectal bleeding: ten year experience. Am J Surg 143:607, 1982 [PMID 6979266]
39. Bloomfeld RS, Smith TP, Schneider AM, et al: Provocative angiography in patients with gastrointestinal hemorrhage of obscure origin. Am J Gastroenterol 95:2807, 2000 [PMID 11051352]
40. Shetzline MA, Suhocki P, Dash R, et al: Provocative angiography in obscure gastrointestinal bleeding. South Med J 93:1205, 2000 [PMID 11142458]
41. Krevsky B: Detection and treatment of angiodysplasia. Gastrointest Endosc Clin N Am 7:509, 1997 [PMID 9177150]
42. Foutch PG: Colonic angiodysplasia. Gastro enterologist 5:148, 1997
43. Rutgeerts P, Van Gompel F, Geboes K, et al: Long term results of treatment of vascular malformations of the gastrointestinal tract by neodymium Yag laser photocoagulation. Gut 26:586, 1985 [PMID 3874122]
44. Rogers BH: Endoscopic diagnosis and therapy of mucosal vascular abnormalities of the gastrointestinal tract occurring in elderly patients and associated with cardiac, vascular, and pulmonary disease. Gastrointest Endosc 26:134, 1980 [PMID 6969679]
45. Binmoeller KF, Thonke F, Soehendra N: Endoscopic hemoclip treatment for gastrointestinal bleeding. Endoscopy 25:167, 1993 [PMID 8491134]
46. Jaspersen D, Korner T, Schorr W, et al: Diagnosis and treatment control of bleeding intestinal angio dysplasias with an endoscopic Doppler device. Bildge bung 62:14, 1995
47. Jensen DM, Machicado GA, Jutabha R, et al: Urgent colonoscopy for the diagnosis and treatment of severe diverticular hemorrhage. N Engl J Med 342:78, 2000 [PMID 10631275]
48. Bloomfeld RS, Rockey DC, Shetzline MA: Endoscopic therapy of acute diverticular hemorrhage. Am J Gastroenterol 96:2367, 2001 [PMID 11513176]
49. Naveau S, Aubert A, Poynard T, et al: Long-term results of treatment of vascular malformations of the gastrointestinal tract by neodymium YAG laser photocoagulation. Dig Dis Sci 35:821, 1990 [PMID 2364836]
50. Santos JC Jr, Aprilli F, Guimaraes AS, et al: Angio dysplasia of the colon: endoscopic diagnosis and treatment. Br J Surg 75:256, 1998
51. Eisen GM, Dominitz JA, Faigel DO, et al: An annotated algorithmic approach to upper gastrointestinal bleeding. Gastrointest Endosc 53:853, 2001 [PMID 11375617]
52. Habr-Gama A, Waye JD: Complications and hazards of gastrointestinal endoscopy. World J Surg 13:193, 1989 [PMID 2658366]
53. Trowers EA, Ganga U, Rizk R, et al: Endoscopic hemorrhoidal ligation: preliminary clinical experience. Gastrointest Endosc 48:49, 1998 [PMID 9684664]
54. Athanasoulis CA, Baum S, Rosch J, et al: Mesenteric arterial infusions of vasopressin for hemorrhage from colonic diverticulosis. Am J Surg 129:212, 1975 [PMID 1078946]
55. Rahn NH 3rd, Tishler JM, Han SY, et al: Diagnostic and interventional angiography in acute gastrointestinal hemorrhage. Radiology 143:361, 1982 [PMID 6978500]
56. Levinson SL, Powell DW, Callahan WT, et al: A current approach to rectal bleeding. J Clin Gastro enterol 3:9, 1981
57. Clark RA, Colley DP, Eggers FM, et al: Acute arterial gastrointestinal hemorrhage: efficacy of trans catheter control. AJR Am J Roentgenol 136:1185, 1981 [PMID 6786033]
58. Browder W, Cerise EJ, Litwin MS: Impact of emergency angiography in massive lower gastrointestinal bleeding. Ann Surg 204:530, 1986 [PMID 3094466]
59. Matolo NM, Link DP: Selective embolization for control of gastrointestinal hemorrhage. Am J Surg 138:840, 1979 [PMID 315725]
60. Encarnacion CE, Kadir S, Beam SA, et al: Gastrointestinal bleeding: treatment with gastrointestinal arterial embolization. Radiology 183:505, 1992 [PMID 1561358]
61. Bookstein JJ, Chlosta EM, Foley D, et al: Trans catheter hemostasis of gastrointestinal bleeding using modified autogenous clot. Radiology 113:277, 1974 [PMID 4547597]
62. Peck DJ, McLoughlin RF, Hughson MN, et al: Percutaneous embolotherapy of lower gastrointestinal hemorrhage. J Vasc Interv Radiol 9:747, 1998 [PMID 9756061]
63. Gady JS, Reynolds H, Blum A: Selective arterial embolization for control of lower gastrointestinal bleeding: recommendations for a clinical management pathway. Curr Surg 60:344, 2003 [PMID 14972273]
64. Gomes AS, Lois JF, McCoy RD: Angiographic treatment of gastrointestinal hemorrhage: comparison of vasopressin infusion and embolization. AJR Am J Roentgenol 146:1031, 1986 [PMID 3485897]
65. Funaki B: Microcatheter embolization of lower gastrointestinal hemorrhage: an old idea whose time has come. Cardiovasc Intervent Radiol 27:591, 2004 [PMID 15578134]
66. Darcy M: Treatment of lower gastrointestinal bleeding: vasopressin infusion versus embolization. J Vasc Interv Radiol 14:535, 2003 [PMID 12761306]
67. Wright HK, Pelliccia O, Higgins EF Jr, et al: Controlled, semielective, segmental resection for massive colonic hemorrhage. Am J Surg 139:535, 1980 [PMID 6966142]
68. Eaton AC: Emergency surgery for acute colonic haemorrhage—a retrospective study. Br J Surg 68:109, 1981
69. McGuire HH Jr, Haynes BW Jr: Massive hemorrhage for diverticulosis of the colon: guidelines for therapy based on bleeding patterns observed in fifty cases. Ann Surg 75:847, 1972
70. Setya V, Singer JA, Minken SL: Subtotal colectomy as a last resort for unrelenting, unlocalized, lower gastrointestinal hemorrhage: experience with 12 cases. Am Surg 58:295, 1992 [PMID 1622010]
71. Jensen DM, Machicado GA: Colonoscopy for diagnosis and treatment of severe lower gastrointestinal bleeding: routine outcomes and cost analysis. Gastrointest Endosc Clin N Am 7:477, 1997 [PMID 9177148]
72. Foutch PG, Rex DX, Lieberman DA: Prevalence and natural history of colonic angiodysplasia among healthy asymptomatic people. Am J Gastroenterol 90:564, 1995 [PMID 7717311]
73. Heer M, Ammann R, Buhler H: Clinical significance of colonic angiodysplasias. Schweiz Med Wochenschr 114:1416, 1984 [PMID 6333723]
74. Sebastian JJ, Lucia F, Botella MT, et al: Diffuse gastrointestinal angiodysplasia associated with cryptogenic hepatic cirrhosis and coagulopathy simulating von Willebrand disease. Rev Esp Enferm Dig 88:631, 1996 [PMID 8962779]
75. Hughes LE: Postmortem survey of diverticular disease of the colon: I. Diverticulosis and diverticulitis. Gut 10:336, 1969 [PMID 5771665]
76. Meyers MA, Alonso DR, Gray GF, et al: Pathogenesis of bleeding colonic diverticulosis. Gas troenterology 71:577, 1976
77. Rushford AJ: The significance of bleeding as a symptom in diverticulitis. Proc R Soc Med 49:577, 1956 [PMID 13359418]
78. Bokhari M, Vernava AM, Ure T, et al: Diver ticular hemorrhage in the elderly—is it well tolerated? Dis Colon Rectum 39:191, 1996
79. Luk GD, Bynum TE, Hendrix TR: Gastric aspiration in localization of gastrointestinal hemorrhage. JAMA 241:576, 1979 [PMID 310892]
80. Robert JR, Sachar DB, Greenstein AJ: Severe gas trointestinal hemorrhage in Crohn's disease. Ann Surg 213:207, 1991 [PMID 1998401]
81. Binder SC, Miller HH, Deterling RA Jr: Emergency and urgent operations for ulcerative colitis: the procedure of choice. Arch Surg 110:284, 1975 [PMID 1078773]
82. Cirocco WC, Reilly JC, Rusin LC: Life-threatening hemorrhage and exsanguination from Crohn's disease: report of four cases. Dis Colon Rectum 38:85, 1995 [PMID 7813353]
83. Reyes E, Hernandez J, Gonzalez A: Typhoid co-litis with massive lower gastrointestinal bleeding: an unexpected behavior of Salmonella typhi. Dis Colon Rectum 29:511, 1986 [PMID 3488182]
84. Maguire TM, Wensel RH, Malcolm N: Massive gastrointestinal hemorrhage cecal ulcers and Salmonella colitis. J Clin Gastroenterol 7:249, 1985 [PMID 3894495]
85. Cohen MB, Giannella RA: Hemorrhagic colitis associated with Escherichia coli O157:H7. Adv Intern Med 37:173, 1992 [PMID 1557995]
86. Gould PC, Khawaja FI, Rosenthal WS: Anti bio tic-associated hemorrhagic colitis. Am J Gastro enterol 77:491, 1982
87. Tagkalidis PP, Tjandra JJ: Chronic radiation proctitis. ANZ J Surg 71:230, 2001 [PMID 11355732]
88. den Hartog Jager FC, van Haastert M, Batterman JJ, et al: The endoscopic spectrum of late radiation damage of the rectosigmoid colon. Endoscopy 17:214, 1985 [PMID 4065051]
89. Crook J, Esche B, Futter N: Effect of pelvic radiotherapy for prostate cancer on bowel, bladder, and sexual function: the patient's perspective. Urology 47:387, 1996 [PMID 8633407]
90. Ellis DJ, Reinus JF: Lower intestinal hemorrhage. Crit Care Clin 11:369, 1995 [PMID 7788537]
91. Zuckerman GR, Prakash C: Acute lower intestinal bleeding. Part II: etiology, therapy, and outcomes. Gastrointest Endosc 49:228, 1999 [PMID 9925703]
92. Wagner HE, Stain SC, Gilg M, et al: Systematic assessment of massive bleeding of the lower part of the gastrointestinal tract. Surg Gynecol Obstet 175:445, 1992 [PMID 1440174]
93. Makela JT, Kiviniemi H, Laitinen S, et al: Diagnosis and treatment of acute lower gastrointestinal bleeding. Scand J Gastroenterol 28:1062, 1993 [PMID 8303208]
94. Geenen JE, Schmitt MG Jr, Wu WC, et al: Major complications of coloscopy: bleeding and perforation. Am J Dig Dis 20:231, 1975 [PMID 804812]
95. Macrae FA, Tan KG, Williams CB: Towards safer colonoscopy: a report on the complications of 5000 diagnostic or therapeutic colonoscopies. Gut 24:376, 1983 [PMID 6601604]
96. Gaydos LA, Freireich EJ, Mantel N: The quantitative relation between platelet count and hemorrhage in patients with acute leukemia. N Engl J Med 266:905, 1962 [PMID 13897369]
97. Wilkinson JF, Nour-Eldin F, Israels MC, et al: Haemophilia syndromes: a survey of 267 pa tients. Lancet 2:947, 1961 [PMID 14007059]
98. Mittal R, Spero JA, Lewis JH, et al: Patterns of gastrointestinal hemorrhage in hemophilia. Gastroenterology 88:515, 1985 [PMID 3917418]
99. Cappell MS, Price JB: Characterization of the syndrome of small and large intestinal variceal bleeding. Dig Dis Sci 32:422, 1987 [PMID 3549204]
100. Chawla Y, Dilawari JB: Anorectal varices—their frequency in cirrhotic and non-cirrhotic portal hypertension. Gut 32:309, 1991
101. Goenka MK, Kochhar R, Nagi B, et al: Rec to sigmoid varices and other mucosal changes in patients with portal hypertension. Am J Gastro enterol 86:1185, 1991
102. Baer JW, Ryan S: Analysis of cecal vasculature in the search for vascular malformations. AJR Am J Roentgenol 126:394, 1976 [PMID 175709]
103. Danesh BJ, Spiliadis C, Williams CB, et al: Angiodysplasia—an uncommon cause of colonic bleeding: colonoscopic evaluation of 1,050 pa tients with rectal bleeding and anaemia. Int J Colorectal Dis 2:218, 1987
104. Heer M, Sulser H, Hany A: Angiodysplasia of the colon: an expression of occlusive vascular disease. Hepatogastroenterology 34:127, 1987 [PMID 2440788]
105. Richter JM, Hedberg SE, Athanasoulis CA, et al: Angiodysplasia: clinical presentation and colonoscopic diagnosis. Dig Dis Sci 29:481, 1984 [PMID 6609803]
106. Zuckerman G, Benitez J: A prospective study of bidirectional endoscopy (colonoscopy and upper endoscopy) in the evaluation of patients with occult gastrointestinal bleeding. Am J Gas tro enterol 87:62, 1992
107. Foutch PG: Diverticular bleeding: are non steroidal anti-inflammatory drugs risk factors for hemorrhage and can colonoscopy predict outcome for patients? Am J Gastroenterol 90:1779, 1995 [PMID 7572894]
108. Boley SJ, Sammartano R, Adams A, et al: On the nature and etiology of vascular ectasias of the colon: degenerative lesions of aging. Gastro enterology 72:650, 1977
109. Mitsudo SM, Boley SJ, Brandt LJ, et al: Vascular ectasias of the right colon in the elderly: a distinct pathologic entity. Hum Pathol 10:585, 1979 [PMID 527961]
110. Imperiale TF, Ransohoff DF: Aortic stenosis, idiopathic gastrointestinal bleeding, and angio dysplasia: is there an association? A methodologic critique of the literature. Gastroenterology 95:1670, 1988 [PMID 3053316]
111. Boley SJ, Sprayregen S, Sammartano RJ, et al: The pathophysiologic basis for the angiographic signs of vascular ectasias of the colon. Radiology 125:615, 1977 [PMID 928682]
112. Helmrich GA, Stallworth JR, Brown JJ: Angio dysplasia: characterization, diagnosis, and advances in treatment. South Med J 83:1450, 1990 [PMID 2251535]
113. Burt RW, Berenson MM, Samuelson CO, et al: Rheumatoid vasculitis of the colon presenting as pancolitis. Dig Dis Sci 28:183, 1983 [PMID 6825538]
114. Moses FM: Gastrointestinal bleeding and the athlete. Am J Gastroenterol 88:1157, 1993 [PMID 8338081]
115. Sokol RJ, Farrell MK, McAdams AJ: An unusual presentation of Wegener's granulomatosis mimicking inflammatory bowel disease. Gastro enterology 87:426, 1984
116. Klinvimol T, Ho YH, Parry BR, et al: Small bowel causes of per rectum haemorrhage. Ann Acad Med Singapore 23:866, 1994 [PMID 7741501]
117. Gilmore PR: Angiodysplasia of the upper gastrointestinal tract. J Clin Gastroenterol 10:386, 1988 [PMID 3047221]
118. Netterville RE, Hardy JD, Martin RS Jr: Small bowel hemorrhage. Ann Surg 167:949, 1968 [PMID 5301245]
119. Lu CL, Chen CY, Chiu ST, et al: Adult intussuscepted Meckel's diverticulum presenting mainly lower gastrointestinal bleeding. J Gastroenterol Hepatol 16:478, 2001 [PMID 11354290]
120. Longo WE, Vernava AM 3rd: Clinical implications of jejunoileal diverticular disease. Dis Colon Rectum 35:381, 1992 [PMID 1582362]
121. Buchman TG, Bulkley GB: Current management of patients with lower gastrointestinal bleeding. Surg Clin North Am 67:651, 1987 [PMID 3495893]
122. Chalasani N, Wilcox CM: Gastrointestinal hemorrhage in patients with AIDS. AIDS Patient Care STDS 13:343, 1999 [PMID 10842854]
123. Chalasani N, Wilcox CM: Etiology and outcome of lower gastrointestinal bleeding in patients with AIDS. Am J Gastroenterol 93:175, 1998 [PMID 9468236]
124. Becherer PR, Sokol-Anderson M, Joist JH, et al: Gastrointestinal histoplasmosis presenting ashematochezia in human immunodeficiency virus—infected hemophilic patients. Am J Hematol 47:229, 1994
125. Allison MC, Howatson AG, Torrance CJ, et al: Gastrointestinal damage associated with the use of nonsteroidal antiinflammatory drugs. N Engl J Med 327:749, 1992 [PMID 1501650]
126. Davies NM: Toxicity of nonsteroidal anti-inflammatory drugs in the large intestine. Dis Colon Rectum 38:1311, 1995 [PMID 7497845]
127. Langman MJ, Morgan L, Worrall A: Use of anti-inflammatory drugs by patients admitted with small or large bowel perforations and haemorrhage. Br Med J (Clin Res Ed) 290:347, 1985 [PMID 3917814]
128. Holt S, Rigoglioso V, Sidhu M, et al: Non steroidal antiinflammatory drugs and lower gastrointestinal bleeding. Dig Dis Sci 38:1619, 1993 [PMID 8359073]
Section 5 Gastrointestinal Tract and Abdomen
6 Lower Gastrointestinal BleedingMichael J. Rosen, MD
Approach to lower gastrointestinal bleeding, initial evaluation and resuscitation, investigative studies, and management are described, and the etiology of lower GI bleeding is discussed.
Role of Radionuclide Scanning for Lower GI HemorrhageRadionuclide scanning is highly sensitive for lower GI hemorrhage: it is capable of detecting bleeding at rates as slow as 0.1 to 0.4 ml/min. Two imaging tracers, both labeled with technetium-99m (99mTc), are currently available for radionuclide scanning in this setting: 99mTc-labeled sulfur colloid (99mTc-SC) and 99mTc-labeled red blood cells (RBCs). 99mTc-SC requires no preparation time and can be injected immediately into the patient; however, its rapid absorption into the liver and the spleen can often hinder accurate localization of overlying bleeding sites. At our institution, we prefer to use 99mTc-labeled RBCs. This agent requires some preparation time, but it has a much longer half-life than 99mTc-SC does, it is not taken up by the liver and spleen, and it can be detected on images as long as 24 to 48 hours after injection.
The high sensitivity of 99mTc-labeled RBC scanning—80% to 98%—is well attested, but there is considerable disagreement in the literature with regard to its specificity in identifying the anatomic site of bleeding. In 2005, one group retrospectively reviewed 127 bleeding scans in an effort to identify factors that might predict a positive scan.1 The investigators found that tagged RBC scans were 48% accurate in localizing bleeding sites later confirmed by endoscopy, surgery, or pathologic evaluation. Multivariate analysis demonstrated that both the number of units of blood transfused in the 24 hours preceding the scan and the lowest recorded hematocrit differed significantly between patients with positive scans and those with negative scans. However, the clinical significance of a positive scan was unclear in this study, in that the rate of endoscopy was not significantly different between patients who had positive scans and those who did not.
To date, no prospective, randomized trials have compared radionuclide scanning with colonoscopy as the initial diagnostic procedure for patients with lower GI hemorrhage. In our view, however, given that radionuclide scanning (unlike colonoscopy and angiography) has no therapeutic intervention capabilities, its best use is in patients with non-life-threatening lower GI bleeding as a prelude and a guide to mesenteric angiography after active hemorrhage has been confirmed.
1. Olds GD, Cooper GS, Chak A, et al: The yield of bleeding scans in acute lower gastrointestinal hemorrhage. J Clin Gastroenterol 39:273, 2005 [PMID 15758618]
Diagnostic use of angiography in patients with lower GI bleeding can often be followed by angiographic therapy. The two main angiographic treatment options are intra-arterial injection of vasopressin and transcatheter embolization.
Vasopressin acts to control bleeding by causing arteriolar vasoconstriction and bowel wall contraction. Once the bleeding site has been localized angiographically, the catheter is positioned in the main trunk of the vessel. Infusion of vasopressin is initiated at a rate of 0.2 U/min and can be increased to a rate of 0.4 U/min. Within 20 to 30 minutes, another angiogram is performed to determine whether the bleeding has ceased. If the bleeding is under control, the catheter is left in place and vasopressin is continuously infused for 6 to 12 hours. If the bleeding continues to be controlled, infusion is continued for an additional 6 to 12 hours at 50% of the previous rate. Finally, vasopressin infusion is replaced by continuous saline infusion, and if bleeding does not recur, the catheter is removed.
The vasoconstrictive action of vasopressin can have deleterious systemic side effects, including myocardial ischemia, peripheral ischemia, hypertension, dysrhythmias, mesenteric thrombosis, intestinal infarction, and death. An alternative for patients with coronary vascular disease, severe peripheral vascular disease, or other comorbidities that prevent safe administration of vasopressin is transcatheter embolization. In this technique, a catheter is superselectively placed into the identified bleeding vessel and an embolizing agent (e.g., a gelatin sponge, a microcoil, polyvinyl alcohol particles, or a balloon) is injected. Several small series found this technique to be 90% to 100% successful at stopping bleeding.1 Equally impressive was the finding that the rebleeding rates in these series were 0%. The complication rates of this procedure are generally reasonable as well; however, intestinal infarction has been reported.
The use of small microcatheters and the ability to superselectively embolize individual vessels have reduced the potential for ischemic perforation. It is possible that as more experience is gained with these techniques, superselective embolization may replace catheter-directed vasoconstrictive therapy, thus obviating the potential deleterious systemic effects of vasopressin administration. Some have suggested that with the exception of cases of diffuse bleeding lesions or cases whose demands exceed the technical limitations of superselective catheterization, embolization therapy should be the first choice for angiographic treatment of lower GI bleeding.2,3
1. Gady JS, Reynolds H, Blum A: Selective arterial embolization for control of lower gastrointestinal bleeding: recommendations for a clinical management pathway. Curr Surg 60:344, 2003 [PMID 14972273]
2. Funaki B: Microcatheter embolization of lower gastrointestinal hemorrhage: an old idea whose time has come. Cardiovasc Intervent Radiol 27:591, 2004 [PMID 15578134]
3. Darcy M: Treatment of lower gastrointestinal bleeding: vasopressin infusion versus embolization. J Vasc Interv Radiol 14:535, 2003 [PMID 12761306]
In a minority of patients, obscure bleeding persists despite negative findings from endoscopy, mesenteric arteriography, and radiolabeled RBC scanning. This obscure bleeding presents a considerable diagnostic challenge, which some investigators have proposed addressing by means of so-called provocative angiography. Provocative angiography involves the use of short-acting anticoagulant agents (unfractionated heparin, vasodilators, thrombolytics, or combinations thereof) in association with angiography. Once the bleeding point has been localized, methylene blue is injected and the patient is immediately brought to the OR for surgical treatment. To date, unfortunately, little has been published on this technique, but it does appear to be a promising approach to this difficult problem.
Lower Gastrointestinal Bleeding
Question 1 A 47-year-old man presents with complaints of nonspecific abdominal pain, which he has been experiencing for 2 days. On examination, the patient is found to be hypotensive and anemic. He denies having experienced any weight loss, and he reports that there have been no changes in his stool.
For this patient, which of the following statements is true?
Please choose the single most appropriate answer to the question - Given the abundance of diagnostic modalities, the source of the hemorrhage is identified in almost 100% of such patients
Sorry, this is incorrect. Try again!
- A drop in the orthostatic blood pressure greater than 10 mm Hg indicates that more than 800 ml of blood has been lost
This is correct.
Objective: To understand the diagnosis of lower gastrointestinal bleedingLower GI bleeding is more common in men than in women. The incidence rises steeply with advancing age, exhibiting a greater than 200-fold increase from the third decade of life to the ninth. Despite the abundance of diagnostic modalities available, attempts to localize the source of the hemorrhage fail in as many as 8% to 12% of patients. A drop in the orthostatic blood pressure greater than 10 mm Hg or an increase in the pulse rate greater than 10 beats/min indicates that more than 800 ml of blood (> 15% of the total circulating blood volume) has been lost. An aspirate that contains copious amounts of bile is strongly suggestive of a lower GI source of bleeding, and the workup should proceed accordingly. The choice is less clear-cut with a clear aspirate. In the absence of bile, such an aspirate cannot rule out a duodenal source for the bleeding. - A clear aspirate on gastric lavage almost always signifies a lower GI source of bleeding
- None of the above
Question 2 A 56-year-old woman presents with symptoms of abdominal pain, weight loss, and rectal bleeding. She is anemic and hypotensive, but she is stable.
For this patient, which of the following should be done first if the gastric lavage yields copious amounts of bile? Please choose the single most appropriate answer to the question - Colonoscopy
Objective: To understand the investigative studies for patients with lower GI bleedingAn aspirate that contains copious amounts of bile should suggest a lower GI source of bleeding. A number of diagnostic techniques are available for determining the source of lower GI hemorrhage. Colonoscopy has both a higher diagnostic yield and a lower complication rate than arteriography in this setting and thus would appear to be a more attractive initial test in most circumstances. When colonoscopy and routine upper GI endoscopy fail to locate a bleeding source, push enteroscopy may be helpful. - Esophagogastroduodenoscopy
- Arteriography
- Emergency laparotomy
Question 3 A 66-year-old man is suspected of having lower GI bleeding. Colonoscopy does not reveal the source of the bleeding.
For this patient, which of the following statements regarding radionuclide scanning is true? Please choose the single most appropriate answer to the question - Technetium-99m (99mTc)-labeled red blood cells (RBCs) can detect bleeding at rates as slow as 0.1 to 0.4 ml/min
- 99mTc-labeled sulfur colloid (99mTc-SC) requires no preparation time
- Radionuclide scanning is best used in patients with non-life-threatening lower GI bleeding
- All of the above
Objective: To understand the role of radionuclide scanning in the diagnosis of lower GI bleedingRadionuclide scanning is highly sensitive for lower GI hemorrhage: it is capable of detecting bleeding at rates as slow as 0.1 to 0.4 ml/min. Two imaging tracers, both labeled with 99mTc, are currently available for radionuclide scanning in this setting: 99mTc-SC and 99mTc-labeled RBCs. 99mTc-SC requires no preparation time and can be injected immediately into the patient. There is considerable disagreement in the literature with regard to its specificity in identifying the anatomic site of bleeding. Given that radionuclide scanning (unlike colonoscopy and angiography) has no therapeutic intervention capabilities, its best use is in patients with non-life-threatening lower GI bleeding as a prelude and a guide to mesenteric angiography after active hemorrhage has been confirmed.
Question 4 A 39-year-old man presents with lower GI bleeding. He has no abdominal discomfort and has experienced no loss of weight.
For this patient, which of the following statements regarding the etiology of lower GI bleeding is true? Please choose the single most appropriate answer to the question - Arteriovenous malformations (AVMs) are the most common cause of lower GI bleeding
- Lower GI bleeding from diverticulosis often requires surgery
- 6% to 10% of patients with ulcerative colitis have lower GI bleeding severe enough to necessitate emergency surgical resection
Objective: To understand the etiology of lower GI bleedingLower GI bleeding has a wide array of possible causes. Of these, diverticular disease is the most common, accounting for 30% to 40% of all cases. AVMs are considerably less common, accounting for 1% to 4% of cases. As many as 80% to 85% of diverticular hemorrhages stop spontaneously. Semielective surgical therapy is usually offered after a second diverticular bleeding episode. The term colitis includes inflammatory bowel disease (IBD), infectious colitis, radiation colitis, and idiopathic ulcers. IBD, in turn, includes Crohn disease and ulcerative colitis. Patients with IBD usually present with bloody diarrhea that is not life-threatening. However, in 6% to 10% of patients with ulcerative colitis, lower GI bleeding is severe enough to necessitate emergency surgical resection; 0.6% to 1.3% of patients with Crohn disease have acute life-threatening lower GI bleeding. - None of the above
Không có nhận xét nào:
Đăng nhận xét