Lower Gastrointestinal Bleeding
Anthony MacLean, MD, FRCSC, FACS
Clinical Associate Professor
Department of Surgery
University of Calgary
Lower gastrointestinal bleeding (LGIB) is defined as bleeding distal to the ligament of Treitz. It can range in severity from trivial to massive, but when significant represents a serious and potentially life threatening situation. LGIB accounts for approximately 20% of all major GI bleeds. The incidence of LGIB requiring hospital admission is approximately 21 cases per 100,000 adults. It is predominantly a disease of the elderly, with a mean age at presentation of 63 to 77 years. Although 80-90% of cases will stop bleeding spontaneously, as many as 25% will re-bleed either during or after their hospital admission. While most patients have a self-limited illness, the reported mortality ranges from 2-4%.
Bleeding from colonic diverticula, vascular ectasias, and colonic ischemia are the most common cause of significant acute LGIB in patients over the age of 65, while in younger patients, infectious or inflammatory conditions are most common.
Diverticulosis: Prevalence of diverticulosis increases with age, and in Western society, approximately two thirds of the population will have diverticulosis by age 85. Twenty percent of patients with diverticulosis will present with bleeding during their lifetime, and 5% will have severe hemorrhage. While the majority of diverticula are located in the sigmoid and descending colon, diverticular bleeding is distributed fairly equally between the right and left sides of the colon. The pathogenesis is felt to be injury to the submucosal branch of the vasa recta. Observation alone is generally recommended following the first episode of diverticular hemorrhage. However, following a second episode, the risk of subsequent episodes appears to approximate 50%, and thus elective resection has been recommended. Patients should be counselled to avoid non steroidal anti-inflammatory drugs, and these have been shown to increase the risk of diverticular bleeding three-fold.
Vascular ectasias: Also known as angiodysplasia, vascular ectasia can occur anywhere in the GI tract, but are most common in the cecum and ascending colon. Lesions can be multiple, and are found incidentally at colonoscopy in approximately 2% of non-bleeding patients over the age of 65. The typical endoscopic appearance is that of a flat, red lesion usually less than 1cm in diameter, with ectatic blood vessels radiating from a central feeding vessel. The pathophysiology is unclear, but is felt to be due to intermittent obstruction of the submucosal veins because of the colonic wall tension, which is highest in the cecum. A high incidence of associated cardiac disease, particularly aortic stenosis, has been observed in patients with vascular ectasias. Only about 15% of patients with vascular ectasia will develop gastrointestinal hemorrhage.
Ischemic colitis: Ischemic colitis results from a sudden and often temporary reduction in mesenteric blood flow, typically caused by hypoperfusion, vasospasm, or occlusion. The usual areas affected are the “watershed” areas of the colon: the splenic flexure and the rectosigmoid junction. A precipitating event or vascular lesion often cannot be identified. Clinically, patients present with abdominal pain, usually accompanied within 24 hours by bloody diarrhea. Patients tend to be elderly, often with significant atherosclerosis or cardiac disease.
Other causes of bleeding are less common, and include inflammatory bowel disease, neoplasms, anorectal disease, infectious colitis or enteritis, radiation colitis/proctitis, trauma, and hematologic disorders. Most of these present with more insidious bleeding, and allow time for thorough investigation in an outpatient setting.
Initial Evaluation and Resuscitation
Because many patients presenting with LGIB have hemodynamic changes, appropriate resuscitation must accompany the initial evaluation. The basics of cardiopulmonary resuscitation must be followed, including placement of large bore intravenous catheters and resuscitation with crystalloid solution. Standard bloodwork should be performed including cross match, and efforts should be made to ensure that patients remain warm. Monitoring strategies will depend upon the degree of the LGIB and the patients’ comorbidities.
Early on a nasogastric tube should be placed. As many as 10% of patients initially felt to have a lower GI bleed will turn out to have an upper GI source for their bleeding. If clear bile is not returned on the nasogastric aspirate, and upper GI endoscopy should be included in the diagnostic evaluation. It should be noted that while a bile aspirate makes an upper source less likely, it does not completely rule it out; thus upper GI endoscopy should always be considered.
A complete history and physical examination is important, and should typically include anoscopy, proctoscopy and/or sigmoidoscopy to rule out an anorectal source for the bleeding.
The three primary traditional diagnostic tools for the evaluation of LGIB are: radionuclide scanning, mesenteric angiography, and colonoscopy. In cases where the bleeding source cannot be identified with these modalities, capsule endoscopy, small bowel enteroscopy, and helical CT scans may have a role in establishing the etiology of the bleeding.
Technetium-labelled red blood scanning
Technetium-labelled red blood scanning has supplanted sulphur colloid scanning as the radionuclide technique of choice. The technique is able to detect bleeding rates as low as 0.1mL per minute, but has a false localization rate of about 25% in reported series. Performing dynamic scintigraphy and basing results on early blushing may give a more accurate localization of the site of bleeding. Its use has been suggested as a screening test to establish bleeding prior to going on to angiography. However, in practice this is difficult to endorse, because the delay in obtaining the scan may allow the window of opportunity for a positive angiogram to pass.
Selective mesenteric angiography
Selective mesenteric angiography has become widely used in the evaluation of LGIB because it has the benefit of being both diagnostic and potentially therapeutic. Bleeding can be detected at rates as low as 0.5-1.0mL per minute. While extravasation is obviously unequivocal evidence of a bleeding source, angiography can also demonstrate other lesions such as tumour blush or angiodysplasias. When bleeding is identified, transcatheter therapy can be instituted to stop the bleeding, be means of either intra-arterial vasopressin or transcatheter arterial embolization (TAE). Since vasopressin has a re-bleeding rate approaching 50%, most centers have adopted TAE, particularly superselective TAE using microcoils or other agents. With available superselective techniques, the rate of bowel ischemia following embolization is low.
Colonoscopy is an extremely valuable tool in the evaluation and management of LGIB, and like angiography, has the advantage of being potentially both diagnostic and therapeutic. Options include immediate colonoscopy with only a rectal washout vs. delayed urgent colonoscopy with a bowel prep. The colonoscopy with only a rectal washout has the advantage of allowing an immediate examination. In patients with a left-sided bleed, the unprepped right colon can generally be visualized, and if no blood is seen on the right, this is a strong indicator that the left side of the bowel was the source of bleeding. However, immediate unprepped colonoscopy has the disadvantage of residual blood and stool, making visualization of mucosal abnormalities such as vascular ectasias difficult. When a bleeding site is identified, it can often be treated. Diverticular bleeding is typically treated with coagulation, hemoclip application, or injection therapy, typically epinephrine. Vascular ectasia is usually treated with coagulation. It should be remembered that even patients who have had their bleeding controlled angiographically should have a colonoscopy, as 5-30% will have neoplastic lesions found.
Newer modalities including CT angiography and MRI angiography have been studied more recently. They are promising, seem to provide good sensitivity and specificity, and may add additional anatomic information, but are diagnostic and not therapeutic. They may have a role as screening, to be followed by mesenteric angiography. Lastly, provocative angiography is occasionally used in a small subset of patients with obscure bleeding. This involves the use of vasodilators, anticoagulants, and/or thrombolytics. It has a sensitivity range of 29-40%, and typically positive tests go on to have embolization or surgery.
The majority of patients with LGIB will stop spontaneously and never require surgery. However, approximately 10-25% of patients will require operative intervention. The indications for surgery include continued or recurrent hemorrhage despite nonoperative attempts, ongoing hemodynamic instability, or transfusion requirement >6 units. It should be remembered that when patients receive 10 or more units of blood, their morbidity and mortality increase significantly. While it has been shown that patients who require 4 or more units of blood in the first 24 hours have a 50% chance of requiring surgery, there are no absolute predictors of who will require an operation, and all patients should be thoroughly evaluated and treated.
Every effort should be made to localize the bleeding source pre-operatively. If patient stability allows it, as many diagnostic tests as necessary should be performed to identify the source of bleeding. Emergency colectomies for non-localized bleeding have a mortality rate of 10-30%, and if a segmental colectomy is preformed in this setting, the risk of re-bleeding is approximately 35-75%% or more, with a mortality rate of 20-50%.
When the bleeding site has been localized, segmental resections are the treatment of choice, with rates of re-bleeding ranging from 0-15% and mortality less than 10% in reported series. The decision of whether to perform an anastomosis depends on patient stability and their comorbidity. In cases where the bleeding has not been localized, a thorough exploration will sometimes identify the source of bleeding. On-table colonoscopy and enteroscopy when possible will occasionally identify the source. To facilitate this, patients should be placed in the lithotomy position. If localization techniques are not successful, and the bleeding appears to be originating from the colon, the procedure of choice is a total abdominal colectomy, which has a re-bleeding rate of less than 1%.
Post-polypectomy hemorrhage occurs in 0.3% to 6.1% of polypectomies, which can either present immediately, or can be delayed as long as a month. These cases are treated with standard endoscopic techniques, including injection therapy, electrocoagulation, band ligation, EndoLoop application, or endoscopic clipping. If these methods fail, angiographic embolization and/or surgery may be required. When surgery is required, if the histology of the polyp is not available, and if the condition of the patient permits it, consideration should be given to performing an oncologic resection of the involved colonic segment. If the polyp is known to be benign, suture of the base of the polyp may suffice.
Small bowel sources account for 3-5% of all cases of LGIB. Diagnosis and treatment of these lesions is often difficult because of the long length and relative inaccessibility of the small bowel. The most common cause of small bowel hemorrhage is angiodysplasia (70-80%), followed by small bowel diverticula, Meckel’s diverticula, neoplasia, Crohn’s disease, and aorto-enteric fistulas. Wireless capsule endoscopy and double balloon enteroscopy are relatively new techniques that has been more successful than other modalities in identifying small bowel sources of hemorrhage.
Rockey D. Lower Gastrointestinal Bleeding. Gastro. 2006;130:165-171
Vernava AM, Moore BA, Longo WE, and Johnson FE. Lower Gastrointestinal Bleeding. Dis Colon Rectum. 1997; 40: 846-858
Faigel DO, Dominitz JA, Eisen GM, et al. Complications of colonoscopy. Gastro Endosc. 2003; 57:441-445
Hoedema RE, Luchtefeld MA. The management of lower gastrointestinal hemorrhage. Dis
Jensen DM, Machicado GA, Jutabha R, et al. Urgent colonoscopy for the diagnosis and treatment of severe diverticular hemorrhage. NEJM. 2000;342:78-82
Davila RE, Rajan E, Adler DG, et al. ASGE Guideline: the role of endoscopy in the patient with lower-GI bleeding. Gastrointest Endosc 2005; 62: 656-660
McGuire HH and Haynes BW. Massive hemorrhage from diverticulosis of the colon: Guidelines for therapy based on bleeding patterns observed in fifty cases. Ann Surg. 1972; 175(6): 847-853
Sabharwal R, Vladica P, Chou R, and Law WP. Helical Ct in the diagnosis of acute gastrointestinal hemorrhage. Eur J Rad 2006;58:273-278
Yoon W, Jeong YY, Shin SS, et al. Acute massive gastrointestinal bleeding: Detection and localization with arterial phase multi-detector row helical CT. Radiology 2006;239:160-167
Chan FP and Chhor CM. Active Lower gastrointestinal hemorrhage diagnosed by magnetic resonance angiography: case report. Abdominal imaging. 2003; 28(5): 637-639
Bloomfeld RS, Smith TP, Schneider AM and Rockney DC. Provocative angiography in patients with gastrointestinal hemorrhage of obscure origin Am J Gastroenterol 2000;95:2807-2812
Kuo WT, Lee DE, Saad WE, et al. Superselective microcoil embolization for the treatment of lower gastrointestinal hemorrhage. J Vasc Interv Radiol 2003;14:1503-9
Koh DC, Luchtefeld MA, Kim DG, et.al. Efficacy of transarterial embolization as definitive treatment in lower gastrointestinal bleeding. Colorectal Disease. 2009; 11: 53-59
Bender JS, Wiencek RG, Bouwman DL. Morbidity and mortality following total abdominal colectomy for massive lower gastrointestinal bleeding. Am Surg 1991; 57:536-541
Milewski PJ and Schofield PF. Massive colonic hemorrhage - the case for right hemicolectomy. Ann R Coll Surg Engl. 1989; 71(4): 253-259