Abstract
Bleeding from the upper gastrointestinal (GI) tract remains common, with a reported annual incidence of up to 172 per 100000[1], which has if anything increased from earlier series. Case fatality was recently reported as 14%[2], which has probably not changed over several decades. These figures may reflect a rising proportion of elderly patients and increasing non-steroidal anti-inflammatory use, but occur despite apparently better treatments and understanding of the underlying pathophysiology of peptic ulcer disease. Of patients in whom a diagnosis is confirmed, more than 90% suffer from Peptic Ulcers, oesophageal or gastric malignancy, varices, Mallory-Weiss syndrome, erosive disease and oesophagitis[1,2]. This report will focus on the some of the less common aetiologies of upper GI bleeding which are sometimes difficult to identify at endoscopy. Angiodysplasia Gastrointestinal angiodysplasia are the most common cause of obscure chronic blood loss from the digestive tract with small bowel angiodysplasia accounting for up to 40% of obscure GI bleeding[3]. The pathophysiology is unknown, but has been suggested to result from low grade venous obstruction of submucosal veins as they cross muscle layers[4]. It is said to be more prevalent in chronic renal failure patients[5]and in patients with aortic stenosis, although, recent reports have failed to confirm this link[6]. Isolated gastric angiodysplasia commonly occurs on the greater curve of the body (Figure (Figure1),1), whereas small bowel angiodysplasia are often multiple and widespread but may cluster in the proximal jejunum. Figure 1 Cutaneous angiodysplasia in Osler-Weber-Rendu Syndrome. Osler-Weber-Rendu Syndrome is an autosomal dominant condition characterized by angiodysplastic lesions involving the skin, mucosal membranes and organs other than the GI tract (Figure (Figure2).2). Patients present as children or adults with epistaxis and up to 40% develop chronic iron deficiency secondary to GI bleeding usually after the age of 50. Their endoscopic appearance is indistinguishable from other angiodysplastic lesions, but they tend to be more widespread. Figure 2 Large gastric angiodyslasia. Acquired angiodysplastic like lesions such as those due to radiation damage can also lead to upper GI bleeding. Various thermal coagulation devices, including heater probes, bipolar probes, the Nd:YAG laser and the argon plasma coagulator appear to be successful in treating these lesions. Coagulation should begin at the central feeding arteriole and work peripherally. Bleeding is common during treatment and usually self limiting. The depth of injury should be minimized, especially in the small bowel and right colon, in order to avoid either frank perforation, or the post-coagulation syndrome, in which patients develop rebound tenderness without detectable intraperitoneal gas. Laser treatment can cause deep injury relatively easily and must be used carefully. Our primary treatment modality is the bipolar probe because it causes more superficial injury than other thermal methods. Complication rates are low for gastric lesions and although no data exists, are also thought to be low in the small bowel. Colonic complications after laser therapy have been reported in up to 10% cases and include partially treated lesions and perforation. Treatment is not required for incidental lesions. Treatment of isolated gastric lesions will often terminate bleeding, whereas treatment of small bowel lesions can more often only hope to reduce transfusion requirements since many lesions are not reached and treated and new lesions will develop with time. The frequency of further endoscopic treatment sessions depends on clinical assessment, rate of recurrence of anaemia and transfusion requirements. Some patients will maintain a stable haemoglobin on iron therapy alone. A placebo-controlled trial mainly in patients with Osler Weber Rendu disease demonstrated decreased transfusion requirements in patients taking 0.05 mg ethinyl oestradiol and 1.0 mg of norethindrone daily[7].
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have