The Forrest classification SB525334 is often used to distinguish endoscopic appearances of bleeding ulcers (Ia spurting active bleeding; Ib oozing active bleeding; IIa visible vessel; IIb adherent clot; IIc flat pigmented spot; III ulcer with a clean base) [116]. In PUB, patients with active bleeding ulcers or a non-bleeding visible vessel in an ulcer bed are at highest risk of re-bleeding and therefore need prompt endoscopic hemostatic therapy. Patients with low-risk stigmata (clean-based ulcer or a pigmented spot

in ulcer bed) do not require endoscopic therapy [81]. Two small randomised trials, and a meta-analysis suggested that a clot should be removed in search of an artery and, when it is selleckchem present, endoscopic treatment should be given, although the management of peptic Thiazovivin ulcers with overlying adherent

clots that are resistant to removal by irrigation is still controversial [98, 117–119]. Endoscopic treatment can be divided into injection (including epinephrine, sclerosants and even normal saline solution), thermal (including monopolar or bipolar cautery and argon plasma coagulation) and mechanical methods (including hemoclips). Often, the choice of which endoscopic therapy employ is based on local preference and expertise. Injection of diluted epinephrine alone is now judged to be inadequate [94]. Cushions of fluid injected into the submucosa compress the artery to stop or slow down bleeding and allow a clear view of the artery. A second modality should be added to induce thrombosis of the artery. Calvet et al. pooled

the results of 16 randomised controlled trials that compared injection of diluted adrenaline alone with injection followed by a second modality, and showed that combination treatment led to substantial reductions in rate of recurrent bleeding (risk reduction from 18,4% to 10,6%), surgery (from 11,3% to 7,6%) and mortality (from 5,1% to 2,6%) [120]. The investigators also compared studies 6-phosphogluconolactonase with or without second look endoscopies after initial endoscopic treatment. Rebleeding was higher in the group given adrenaline injection alone than in the combination treatment group (15,7% vs. 11,4%). Two other meta-analyses that summarised studies of monotherapies versus dual therapies also concluded that a second modality should be added to injection treatment [108, 121]. The observation suggested that if combination treatment had been instituted at index endoscopy, a second look endoscopy would have been unnecessary, so routine second look endoscopy after initial endoscopic haemostasis is not recommended [122].