The aim of the study is to compare the outcomes of trans-catheter arterial embolization (TAE) and surgery as salvage therapy of peptic ulcer bleeding after failed endoscopic therapy.

The aim of the study is to examine the hypothesis that trans-catheter arterial embolization (TAE) is safer than and probably as effective as surgery in the control of bleeding from ulcers after failed endoscopic therapy. Patients with major arterial bleeding that cannot be stopped by endoscopic therapy will be randomly assigned to receive immediate TAE or surgery. Primary outcome will be death within 30 days of randomization. Secondary outcomes include recurrent bleeding after assigned treatment, need for additional intervention either in the form of interventional radiology or surgery, and post procedural morbidities.

Inclusion Criteria:

• Patients with bleeding peptic ulcers documented during endoscopic therapy AND anyone of the following:
• Forrest I bleeding that fails to be controlled by therapy during first endoscopy
• Check endoscopy after clinical re-bleeding of Forrest I bleeding
• Forrest IIa or an initial IIb but a sizable artery unveiled upon clot elevation in a high risk ulcer defined by:
• posterior bulbar duodenal ulcer > 2cm, or
• an angular notch / lesser curve gastric ulcer > 5cm that fail an attempt at endoscopic therapy together with evidence of a significant bleed (hypotension with SBP < 90mmHg, fresh hematemesis or hematochezia).

Exclusion Criteria:

• Refusal to participate in trial
• No consent
• Age < 18
• Pregnancy
• Moribund patients
• Patients with terminal malignancy