**1. Introduction**

260 Gastrointestinal Endoscopy

Veldhuyzen van Zanten SJ, Bartelsman JF, Schipper ME, Tytgat GN. (1986) Recurrent

Yanar, H. (2007) Dolay K, Ertekin C, Taviloglu K, Ozcinar B, Guloglu R, Barbaros U. An

cases. *Gut* 27: 213-222

Hepatogastroenterology. 54: 1013-7

massive hematemesis from Dieulafoy vascular malformations- a review of 101

infrequent cause of upper gastrointestinal tract bleeding: "Dieulafoy's lesion".

Obscure gastrointestinal bleeding (OGIB) is defined as occult or overt bleeding of unknown origin that persists or recurs despite negative primary radiological and endoscopic studies. It can be classified into two different clinical forms: obscure-overt OGIB, defined as visible passage of blood (ie, melena or hematochezia) and obscure-occult OGIB, manifested by irondeficiency anemia or positive fecal occult blood test without other evidence of bleeding.1

Since the source of bleeding is not readily identifiable by upper GI endoscopy and colonoscopy, OGIB is therefore, by definition, recurrent. Approximately, 5% of GI bleeding occurs between the ligament of Treitz and the ileocecal valve. Angiodysplasias of the small bowel account for 30% to 40% of OGIB and are the most common source of bleeding in patients over 60 years.1,2 They can be found as a primary disease or a gastrointestinal manifestation of systemic diseases such as hereditary haemorrhagic telangiectasia (HHT), von Willebrand (vW) disease, cardiac valvular disease, radiation enteritis, end-stage renal disease, portal hypertension, connective tissue diseases or vasculitis. Other causes include non-steroidal anti-inflammatory drugs enteropathy, inflammatory bowel disease, small bowel tumors (ie, leiomyomas, carcinoid, lymphomas, adenocarcinomas), Meckel´s diverticulum or Dieulafoy´s lesion.

Over the last decade, the diagnostic yield and therapeutic capabilities of small bowel endoscopy have dramatically changed with the development of video capsule endoscopy and deep enteroscopy systems (single balloon, double balloon or spiral). Nonetheless, the diagnostic yield is 75% at best combining both techniques, so a quarter of patients lack a diagnosis of the source of bleeding despite exhaustive evaluation and may be at high risk of rebleeding.1 Additionally, a variable percentage of patients with a diagnosis may not respond to endoscopic therapy or may not be tributary to aggressive endoscopic or surgical management due to severe comorbidities or diffuse distribution of lesions throughout the GI tract. In this particular subset of patients, medical therapy is commonly required to stop, or at least, ameliorate bleeding, which usually leads to high transfusional requirements, exacerbations of medical conditions and subsequent hospital admissions. Indications for medical therapy in OGIB, as approved in the latest American Gastroenterology Association technical review, are listed in Table 1.2

Pharmacological Therapy for Recurrent Obscure Gastrointestinal Bleeding 263

**Depending on the type of bleeding** 

**Depending on the source of bleeding** 

Portal hypertension: beta-blockers

**2. Hormonal therapy** 

**3. Somatostatin analogues** 

Life-threatening: vasoactive medication, rFVIIa

Angiodysplasias: long-acting somatostatin analogues, antiangiogenic drugs (thalidomide, lenalidomide)

Hereditary hemorrhagic telangiectasia: antifibrinolytics, tamoxifen, antiangiogenic drugs (thalidomide, bevacizumab)

Table 3. Suggested therapeutic algoritham for pharmacological therapy in OGIB

displace hormonal therapy as first-line therapeutic option for OGIB.

von Willebrand disease: desmopressin, vW factor, antifibrinolytics

Estrogen-progesterone combination was proposed for OGIB because of preliminary reports of improvement of epistaxis in patients with HHT during pregnancy and further relapse in the puerperium.3 Its effect, which is not immediate, seems to be estrogen dose-dependent and acts by enhancing microvascular circulation, coagulation, and vascular endothelial integrity. The most common combination schedule has been ethynil estradiol 0.01-0.05 mg and noresthisterone 1-3 mg.4 This therapy should be used over six-month periods with pauses to reduce the incidence of adverse effects, mostly due to the estrogen component (vascular thrombosis, gynecomastia and loss of libido in men, breast tenderness and vaginal bleeding in women). However, the two largest placebo-controlled studies addressing the impact of hormonal therapy on GI bleeding from angiodysplasias failed to demonstrate any significant benefit. In the first study, patients with out-of-reach bleeding small-bowel angiodysplasias were treated using high-dose estrogens, estrogen-progesterone or placebo, but no statistical improvement of transfusion requirements was observed amongst the groups.5 Additionally, in the second study, the authors failed to identify any significant effect of hormonal therapy compared to placebo in 72 non-cirrhotic patients bleeding from documented angiodysplasia6. This latter study, however, has setbacks such as the use of low doses of ethynil estradiol and the exclusion of patients with vascular ectasia associated to cirrhosis and HHT. Overall, the effectiveness of hormonal therapy remains unclear and both negative controlled trial results and serious and frequent side effects strongly limit its use in OGIB. Recent reports on the effectiveness of other agents with an improved safety profile

In 1993, octreotide was first reported for the treatment of bleeding small bowel angiodysplasias, in a small series of three patients successfully treated for 10 to 40 months.7 The rationale for the use of somatostatin analogues is based on its effects on splanchnic circulation, as they induce a marked reduction of portal and mesenteric blood flow mediated through inhibition of vasodilator peptides. Additionally, experimental studies have shown that octreotide has antiangiogenic effects, by downregulation of vascular


Table 1. Indications for pharmacological therapy in OGIB

The aim of this chapter is to give an overview of current scientific evidence supporting the use of pharmacological therapy in these, often difficult to treat, OGIB patients. The standard diagnostic and therapeutic approach involving endoscopic and radiological techniques, the management of concomitant antiplatelet and anticoagulant drugs and the supportive care of anemia in OGIB patients, are described elsewhere. Up to date, data regarding pharmacological agents for OGIB are scarce and exclusively based on case reports and small uncontrolled studies. The available evidence suggests a potential role for pharmacological therapy as an adjunctive measure in patients with either multiple comorbidities or in whom lesions are inaccessible or refractory to endoscopic therapy. However, their clinical utility remains to be proven in randomized controlled trials. Similarly, the appropriate dose and schedule required for long-term therapy are also unknown. The different pharmacological agents used for OGIB reported in the literature are listed in Table 2.


Table 2. Pharmacological agents used in OGIB

**Depending on the type of bleeding**  Life-threatening: vasoactive medication, rFVIIa **Depending on the source of bleeding**  Angiodysplasias: long-acting somatostatin analogues, antiangiogenic drugs (thalidomide, lenalidomide) Hereditary hemorrhagic telangiectasia: antifibrinolytics, tamoxifen, antiangiogenic drugs (thalidomide, bevacizumab) Portal hypertension: beta-blockers von Willebrand disease: desmopressin, vW factor, antifibrinolytics

Table 3. Suggested therapeutic algoritham for pharmacological therapy in OGIB
