Diagnosis and Treatment of Inflammatory Bowel Disease–Associated Bleeding

List of Abbreviations


Argon plasma coagulation


Balloon-assisted enteroscopy


Crohn’s disease


Confidence interval




Colorectal cancer




Inflammatory bowel disease


Ileal pouch-anal anastomosis


Nonsteroidal antiinflammatory drugs


Odds ratio


Primary sclerosing cholangitis


Tumor necrosis factor


Ulcerative colitis


Video capsule endoscopy


The author is supported by the Ed and Joey Story Endowed Chair.


While bloody bowel movement is one of the most common presentations of active ulcerative colitis (UC), bleeding as a predominant clinical symptom in Crohn’s disease (CD) is not common. The majority of patients with UC or CD, however, would present with some degree of bleeding, resulting from lesions from active mucosal inflammation, ulcer, or polyps. Therefore, stool tests for occult bleeding, which was designed for screening for colorectal cancer is not recommended for the assessment of gastrointestinal (GI) bleeding or disease activity in inflammatory bowel disease (IBD). In contrast, the commonly used noninvasive stool tests for the assessment disease activity in IBD are calprotectin and lactoferrin.

The pattern of GI bleeding in UC or CD ranges from hidden blood in the stool to massive to life-threatening hemorrhage. The reported definitions of GI bleeding as a presenting symptom in IBD, particularly in CD, varied in the literature, and there has been on published consensus. The drop in hemoglobin value and/or the number of units of blood transfusion, during a set of time periods, have used to define the GI bleeding. Based on the published data in the literature, this author would suggest that the definition of GI bleeding in UC or CD is a clinical symptom of bloody stools, a drop of hemoglobin of more than 2 g (for UC within 7 days; for CD within 24 h), and/or requirement of blood transfusion. The patients may or may not be hemodynamically stable. The degree of GI bleeding in IBD can be classified into mild, moderate, and severe, depending on the requirement of blood transfusion and hemodynamic status ( Table 26.1 ).

Table 26.1

Classification of Bleeding From IBD

Category Subcategory/Definition Example
Location Esophagus and stomach
Small bowel
Colon and rectum
Anal area
Etiology Inflammation Ulcerative colitis (UC)
Crohn’s disease (CD)
Ulcer Disease associated UC or CD
Anastomosis related
Infectious Cytomegalovirus
Polyps Inflammatory/pseudopolyps
Adenomatous or dysplastic polyps
Malignancy Cancer
Anal area disease Hemorrhoids
Anal fissure
Concurrent systemic disease Primary sclerosing cholangitis Variceal bleeding
Bleeding disorders Amyloidosis, idiopathic thrombocytopenic purpura
Transplanted gut
Iatrogenic Diagnostic endoscopy Mucosal biopsy
Therapeutic endoscopy Balloon dilation, polypectomy, endoscopic stricturotomy
Severity Mild Nontransfusion required
Moderate Transfusion-required
Severe Transfusion-required hemodynamically unstable
Speed Acute


One of the earliest studies in the disease course in UC reported a frequency of bleeding of 17%. In contrast, it is estimated that the frequency of severe GI bleeding as a main presenting symptoms in CD ranged from 0.6% to 4%. In addition, various definitions of GI bleeding have been used. An early study of 177 patients with acute lower GI bleeding from Mayo Clinic between 1989 and 1996 showed that 31 (18%) had IBD, with three had UC and 21 had CD, representing 0.1% of all admissions for UC and 1.2% for CD. A retrospective case-controlled study from Korea reported a cumulative probability of acute GI bleeding was 1.7%, 3.6%, 6.5%, and 10.3% after 1, 5, 10, and 20 years of CD diagnosis respectively. In a historical cohort study of 1374 CD patients, 73 patients (5%) were identified with main presenting symptom of acute lower GI bleeding A multivariable study of the 73 CD patients presenting with acute lower GI bleeding versus 146 matched CD controls without bleeding showed the left colon involvement (odds ratio [OR] 2.26, 95% confidence interval [CI] 1.04–4.91), azathioprine use ≥1 year (OR 0.44, 95% CI 0.20–0.99), and previous hemorrhage history (OR = 13.04, 95% CI 5.66–30.04) were risk or protective factors for bleeding.

The reported frequency may not truly reflect the prevalence or incidence of GI bleeding in UC or CD, as the studies were retrospective in nature and suffered from the tertiary care–based patient population, selection, and referral bias. The mortality from bleeding appears to be low.

Etiology and Risk Factors

Obviously rectal bleeding, along with urgency and diarrhea, are the presentation of active disease flare-up in UC. In addition, rectal bleeding can be a sign of relapse of CD. In a study of 34 CD patients from Belgium, presented with acute lower GI bleeding as defined by the need of transfusion of at least 2 units of red blood cells within 24 h showed that the bleeding occurred during a flare-up of the disease in 35% of the cases. The results in active disease flare-up may only explain a minority of patients with the bleeding symptoms. The etiopathogenesis of bleeding in the majority of patients with CD is not entirely clear. In most cases, the bleeding is thought to result from ulcer lesions eroding blood vessels ( Fig. 26.1 ).

Figure 26.1

Large ulcer of the terminal ileum in Crohn’s disease. (A) Discrete large ulcer with a visible vessel while the patient on biological therapy; (B and C) deployment of endoclips to the visible vessel; (D) subsequent injection of 50% glucose in the area.

It is estimated that 75% of the patients with CD and 25% of the patients with UC would need at least one surgery during their lifetime. Ulcer and stricture at the surgical anastomosis site are common. Anastomotic or suture line ulcers are one of the common sources of bleeding in both CD ( Figs. 26.2 and 26.3 ) and UC (with restorative proctocolectomy and ileal pouch-anal anastomosis [IPAA]; Fig. 26.4 ), in short and long terms. In addition to bowel resection and anastomosis, surgical stricturoplasty has been routinely used to treat small bowel CD. In rare occasion, major GI bleeding can occur at the stricturoplasty site.

Figure 26.2

Bleeding from ileocolonic anastomosis from a diagnostic colonoscopy in a Crohn’s disease patient. (A) The normal anastomosis ( green arrow ); (B) bleeding started by passage of a pediatric colonoscopy; (C) bleeding for another 48 h. Ulcer and a visible vessel seen on the anastomosis; (D) endoclips were used to treat.

Figure 26.3

Ileocolonic anastomotic ulcer bleeding in Crohn’s disease. (A) Anastomotic ulcer with bleeding; (B) application of argon plasma coagulation.

Figure 26.4

Bleeding from mid-ileal pouch staple line causing severe anemia in two patients. (A and B) Small inflammatory polyp along the staple line causing intermittent bleeding, controlled by endoclip; (C and D) staple line bleeding controlled by endoclips.

Emphasis has been placed on the avoidance of the use of nonsteroidal antiinflammatory drugs, as this category of medicines associated with disease flare-up and bleeding in patients with IBD ( Fig. 26.5 ). Selective cyclooxygenase 2 inhibitors, such as rofecoxib, have been reported with major GI bleeding complication in CD.

Figure 26.5

Nonsteroidal antiinflammatory drugs colopathy in a patient with ileal Crohn’s disease. (A and B) Mild ileal Crohn’s disease with stricture and evidence of use of mesalamines; (C) NSAID-induced colopathy with bleeding; (D) otherwise normal colon.

Patients with IBD, especially those with UC, are prone to the development of inflammatory polyps, which can cause bleeding. It is believed that the presence of inflammatory pseudopolyps is associated with an increased risk for development of colitis-associated neoplasia. However, routine surveillance biopsy of inflammatory pseudopolyps is controversial, mainly due to the risk of procedure-associated bleeding.

Patients with UC and Crohn’s colitis can have concurrent chronic liver diseases with portal hypertension, low platelet counts, and deficiency in coagulating factors. Primary sclerosing cholangitis (PSC) is a classic example ( Fig. 26.6 ). CD or UC can be associated with bleeding disorders such as Henoch-Schönlein purpura (immunoglobulin A [IgA] vasculitis) and other vasculopathy ( Fig. 26.7 ). It is speculated that both HSS and CD are associated an IgA-mediated mechanism, to various degrees. On the other hand, HSS can be an adverse effect after antitumor necrosis factor therapy for IBD.

Dec 30, 2019 | Posted by in GASTROENTEROLOGY | Comments Off on Diagnosis and Treatment of Inflammatory Bowel Disease–Associated Bleeding

Full access? Get Clinical Tree

Get Clinical Tree app for offline access