It is common for patients to present with occult or obscure gastrointestinal bleeding, but meaningful diagnostic criteria have not been delineated for either type of bleeding. Occult gastrointestinal bleeding refers to the initial presentation of a patient with a positive fecal occult blood test (FOBT) and/or iron deficiency anemia (IDA), without visible fecal blood. Obscure bleeding is defined as bleeding of unknown origin that persists or recurs after negative initial or primary endoscopic evaluation (colonoscopy or upper endoscopy). The Clinical Practice and Practice Economics Committee of the American Gastroenterological Association (AGA) reviewed the literature and developed recommendations for the evaluation and treatment of occult and obscure gastrointestinal bleeding. This discussion is limited to their recommendations for the evaluation of bleeding.

The bedside examination may be helpful in providing clues to the cause of bleeding. A history can reveal ingestion of medications known to cause bleeding (e.g., aspirin, nonsteroidal anti-inflammatory drugs, alendronate, potassium chloride, anticoagulants). A family history might suggest a hereditary vascular problem. Other rare causes of bleeding may be detected on physical examination, including Plummer-Vinson syndrome, acquired immunodeficiency syndrome (AIDS), neurofibromatosis and other diseases with typical cutaneous manifestations. Symptoms specific to the upper or lower intestinal tract may direct the initial endoscopic procedure, but data do not support limiting the evaluation to the symptomatic region.

Colonoscopy and upper endoscopy remain the major investigative methods in the evaluation of occult bleeding, although some lesions found on colonoscopy are a chance finding, because not all polyps or cancers have a propensity to bleed. The contribution of non-bleeding upper gastrointestinal lesions to occult bleeding cannot be determined with certainty. In a review of the causes of occult bleeding, IDA and FOBT are more commonly linked to a bleeding source in the upper gastrointestinal tract (29 to 56 percent) than in the lower gastrointestinal tract (20 to 30 percent). In addition, the likelihood of finding a bleeding source was higher in patients with IDA (61 to 71 percent) than in those with positive FOBT (48 to 53 percent). For a list of causes of occult bleeding associated with FOBT and IDA, see Table 1 on page 3398.

Double-contrast barium studies appear to be less accurate in identifying bleeding sources than endoscopy, but they may be a reasonable alternative when considering factors such as patient preferences, risks of conscious sedation, comorbid disease and local expertise. Small bowel studies should be reserved for use in patients who have persistent IDA or positive FOBT results, which, by definition, are categorized as obscure bleeding.

The overall incidence and location of lesions responsible for obscure bleeding have not been well studied. Obscure bleeding requires evaluation of the small bowel and may require repeat upper and lower bowel endoscopy. For a list of possible causes of obscure bleeding, see Table 2. A number of methods can be used to view the small bowel. Biopsy can be useful in detecting celiac sprue as a cause of IDA. Other methods include enteroscopy, in an attempt to view the entire small bowel; radiographs using small bowel follow through after ingestion of a barium suspension; or enteroclysis, in which contrast material is instilled through a small tube placed in the proximal intestine. Radioisotope bleeding scans may be helpful in identifying the site of bleeding if the volume is greater than 0.1 to 0.4 mL per minute. However, positive findings in this type of testing must be verified with an alternative test because of a relatively high number of false-positive results. Angiography may be useful in patients with active bleeding greater than 0.5 mL per minute and can identify highly vascular nonbleeding lesions such as angiodysplasia and neoplasms.

The AGA Committee concludes that no single technique has emerged as the most efficient way to evaluate gastrointestinal bleeding. Most patients will benefit from careful evaluation that includes as much visualization of the small bowl as possible. For some patients, this may mean multiple procedures. Additional studies are needed to clearly identify the most efficient approach to evaluation and definitive diagnosis of gastrointestinal bleeding.