The presence of blood clot in stomach limited quality of endoscopic view ,which affect successful rate of hemostatic endoscopy in patient with acute upper gastrointestinal bleeding. The study is aimed to evaluate the efficacy of metoclopramide, as pro-kinetic agent ,for gastric visualization in the patient with acute UGIB; double-blind randomized controlled trial and two centers study. The patient were randomly assigned to receive either metoclopramide (10mg) intravenously or placebo before endoscopy 30-120 min. The primary endpoint was endoscopic yield, assessed by objective gastric visualized scoring systems. Secondary end points include duration of endoscope, technical success rate, the need for second-look EGD, units of blood transfusion, length of hospital stay and 30-day rebleeding rate.

• Double-blind, Double centers, RCT

• All endoscopists at two participating sites attend the pre-study meeting for standardization of protocol and scoring system.

• Eligible patients were randomly assigned to either metoclopramide or placebo group in a 1:1 conceal allocation according to a computer-generated randomization list with block randomization in size of four.

• Assigned treatment was kept in opaque sealed envelopes.

• Before EGD, the patients had adequate resuscitation to maintain stable hemodynamics(SBP ≥ 90 mmHg and/or HR < 100 bpm) and blood transfusion to reach Hb> 7g/dL and correct coagulopathy.

• EGD is performed in left lateral position, under local lidocaine anesthesia alone or combined with IV anesthetics

• The standardized set of endoscopic landmarks, including the esophagus, gastro-esophageal junction, fundus, gastric body, incisura, antrum, duodenal bulb, and second part of the duodenum were examined.

• To assess endoscopic gastric visualization, we applied simple validate objective scoring system and estimated coverage of blood clot on mucosa on reference endoscopies view in 4 locations, including fundus, corpus, antrum and duodenal bulb.

• Each location is scored between 0 and 2; Score 0 (worst vision) < 25% of the surface was visible; Score 1 25-75% visible surface; Score 2 (best vision), > 75% visible surface ( total score range 0-8)

• 'Adequate visualization' defined as total score six or higher (out of 8).

• All photos of endoscopic landmark, including the reference endoscopic views of four locations (fundus, corpus antrum and duodenal bulb), were taken and internally validated by another endoscopist who was blinded to the randomization allocation.

• The duration of the endoscopy was recorded from beginning to end of the procedure in minute.

• The 72-hour recurrent GI bleeding was documented and re-EGD was performed to confirm rebleeding if the following conditions were met

  : I) hematemesis or bloody NG > 6 hours after endoscopy; II) melena after normalization of stool color; III) hematochezia after normalization of stool color or melena; IV) development of tachycardia (HR ≥ 110 bpm) or hypotension(SBP ≤ 90 mmHg) after ≥ 1 hour of vital sign stability without other cause; V) hemoglobin drop of ≥ 2 g/dL after two consecutive stable hemoglobin values ( < 0.5 g/dL decrease) ≥ 3 hours apart; VI) tachycardia or hypotension that does not resolve within 8 hours after index endoscopy despite appropriate resuscitation (in the absence of an alternative explanation), associated with persistent melena or hematochezia or; VII) persistently dropping hemoglobin of > 3 g/dL in 24 hours associated with persistent melena or hematochezia.

• 30-day rebleeding is accessed by direct phone call to patient.

Inclusion Criteria:

1. Age >= 18 years
2. Active upper GI bleeding ( defined as fresh or bright red hematemesis within 24hr. or presented of blood via NG aspiration )
3. Underwent upper GI endoscopy within 12hr
4. Informed consent obtained

Exclusion Criteria:

1. Known allergy of metoclopramide
2. History of gastric or duodenal surgery
3. Known case esophageal, gastric or duodenal cancer
4. Diagnosed with advanced HIV infection (defined as CD4 cell count <200 cells/mm3 or WHO clinical stage 3 or 4)
5. Pregnancy or lactating
6. NG lavage was done with solution > 50 ml.

All participating endoscopists and patients were blinded to the randomization allocation.