Obscure gastrointestinal bleeding: How to find, and treat, the source

It accounts for 5% of all cases of gastrointestinal hemorrhage. It requires an average of 7.3 diagnostic tests, five hospitalizations and 46 units of blood per patient to evaluate.

It accounts for 5% of all cases of gastrointestinal hemorrhage. It requires an average of 7.3 diagnostic tests, five hospitalizations and 46 units of blood per patient to evaluate. And after all that, a doctor has only a 50-50 chance of figuring out what's causing it.

Obscure gastrointestinal bleeding, which involves bleeding of unknown origin that persists or recurs after an initial negative endoscopic exam, has long been a head-scratcher for physicians. After all, it's difficult to treat bleeding when you're not sure where it's coming from.


“It's a big problem,” said Jonathan A. Leighton, MD, associate professor at the Mayo Clinic in Scottsdale, Ariz., during a talk at the American College of Gastroenterology's annual meeting in Philadelphia in October. “It takes a median period of two years to evaluate these patients, and even then, you wind up with a definitive diagnosis only about 50% of the time.”

But new technology has emerged in the last decade that is making evaluation and treatment a little easier and, in some cases, less burdensome to patients. The key to success is knowing which technique to use in a given situation.

First steps

When evaluating a patient with obscure GI bleeding, physicians start with the patient's age, Dr. Leighton said. For patients under age 40, the cause is most likely to be tumors, followed by Crohn's disease, Meckel's diverticulum, polyposis syndromes and angiodysplasia. For patients older than 40 years, the most common causes are angiodysplasia, gastric antral vascular ectasia, tumors, drug-induced small-bowel injury and Dieulafoy lesion.

“In addition to age, we have to take into account the amount of bleeding, the necessity of blood transfusions, concomitant diseases and associated symptoms like weight loss. We also have to rule out other causes of anemia, particularly in patients who present with iron deficiency anemia,” he said.

In addition to these factors, a second endoscopy should always be considered before proceeding with a small-bowel evaluation, Dr. Leighton said. That's because studies have shown that 35% to 75% of lesions are missed in the first endoscopy.

“Multiple studies have shown the second-look endoscopy will pick up lesions of the stomach and colon, including lesions in the upper tract like peptic ulcer disease and in the colon like angiodysplasia and neoplasia,” Dr. Leighton said.

Moving on

If the second-look endoscopy proves unhelpful in revealing the source of bleeding, one should move on to small-bowel evaluation, where the main choices are between capsule endoscopy (CE), double-balloon enteroscopy (DBE) and CT enterography/enteroclysis (CTE), Dr. Leighton said.

The likelihood of discovering lesions through CE or DBE is about the same, according to a meta-analysis published last April in Gastrointestinal Endoscopy. Physicians should thus decide which test to use based on individual clinical presentation, Dr. Leighton advised.

Patients who present with intermittent overt or occult bleeding—which are the majority of those with obscure bleeding—should have CE as their first-line test, he said.

“The benefit of capsule endoscopy is that it's painless and non-invasive, it potentially visualizes the entire small bowel, and it has an excellent diagnostic yield,” Dr. Leighton said.

Fairly stable patients with acute, ongoing overt bleeding that physicians are reasonably certain is coming from the small bowel should probably have DBE as the first approach, he advised.

“If you give some of these patients capsule endoscopy, you will see blood in the small bowel and you can't do anything about it. With the double balloon, you can treat it,” Dr. Leighton said.

For patients with obstructive symptoms, CT enterography is a reasonable way to start out, he said, because the patient will need to go directly to surgery if he or she has mass lesions. Drawbacks to CT enterography include inadequate bowel distention, the absolute need for IV contrast, the fact that it doesn't directly evaluate mucosa or offer therapeutic capability, and the potential for patients to experience nausea or diarrhea from the barium solution, he said.

Treating the problem

Once the source of bleeding is identified, the physician must then determine how to treat it. Research suggests angiodysplasia will be a likely cause, Dr. Leighton said. Since there isn't much evidence yet about sclerotherapy, the main treatment options for angiodysplasia are thermal contact coagulation and argon plasma coagulation (APC).

There are no randomized, controlled trials comparing the two treatments. But a controlled comparison of APC and bipolar electrocoagulation, published in the 1999 Gastrointestinal Endoscopy, found APC less convenient and slower to use for bleeding arteriovenous malformation (AVM) patients, with APC therapy failing in 14%. Yet a 2006 study in the American Journal of Gastroenterology reported that 100 AVM and gastric antral vascular ectasia patients treated with APC saw improved median hemoglobin and no complications. As well, only 23% required blood transfusions.

“My preference is to use APC, but for those more comfortable with the heat probe or Gold probe, either one works well,” Dr. Leighton said. “One thing to remember is that perforation is a risk, so you want to avoid over-sedation, and decompress the bowel lumen to give some protection. Also avoid firm pressure with the probe tip and overtreatment.”

For removing small-bowel polyps, the snare technique is most effective, Dr. Leighton said. Because the risk of perforation is similar to that of the right colon, however, doctors may want to consider a surgical approach if the polyps are very large, he added.

In general, surgery for obscure GI bleeding is indicated for patients who continue to bleed and have had negative endoscopic and radiologic workups, and who are clearly operative candidates. Operative candidates include those who need blood transfusions despite optimal iron replacement, patients with associated symptoms like abdominal pain and patients with a history of tumors that could metastasize to the small bowel. Rebleeding can occur in up to 50% of surgical patients.

Medical therapy

In a good chunk of patients, a physician simply can't find the source of the bleeding. For them, the best management is medical therapy, Dr. Leighton said.

“I think the data is out there that endoscopic therapy is by far the best, but there are situations where you might want to use medical therapy, and that can include supportive care and blood transfusions,” Dr. Leighton said. “Obviously, you want to get patients off anticoagulants, antiplatelets and NSAIDs if you can.”

Starting with iron supplementation isn't a bad idea, he added.

“I see a lot of patients with obscure GI bleeding referred to me who are not on iron. Iron may be the only therapy necessary for some, though the source of the blood loss should still be identified,” Dr. Leighton said. “We don't need to be overly aggressive with every patient we see, particularly those with a negative capsule endoscopy.”

Another medical option is hormonal therapy, for which studies have shown mixed results. The main concern is that as many as 57% of those who take hormones have adverse side effects, such as gynecomastia, breast tenderness, vaginal bleeding, fluid retention and heart failure.

“In women, we recommend a mammography and a gynecological exam before starting hormonal therapy,” Dr. Leighton said.

For patients in whom hormonal therapy is contraindicated, octreotide acetate may work to reduce splanchnic blood flow and inhibit angiogenesis. There is a bit of evidence for thalidomide as well, he said.