Early endoscopy is an integral part of the management plan for patients presenting with clinical signs of severe or ongoing UGIB. An accurate endoscopic diagnosis and successful endoscopic hemostasis is highly dependent on adequate visualization of the entire gastric mucosa. Metoclopramide has previously been investigated as a prokinetic agent to aid gastric emptying prior to endoscopy, but its widespread adoption is limited by a lack of high quality clinical evidence as well as concerns regarding side effects. Erythromycin is currently the only prokinetic agent recommended by the American and the European guidelines for use in selected patients in order to reduce the need for second endoscopy. Its clinical application, however, is limited by risk of arrhythmia, significant drug interactions, and frequent drug shortages. Azithromycin is structurally related to erythromycin, but is devoid of most adverse side effects associated with erythromycin use. Early evidence suggests that azithromycin may be an effective alternative to erythromycin in the treatment of gastroparesis. The current study, an interventional, randomized, triple-blinded, placebo-controlled clinical trial, is primarily aimed to evaluate the effectiveness of azithromycin as a prokinetic agent in the management of UGIB. It is also aimed to further evaluate the role of metoclopramide as a prokinetic agent in this setting. Outcome measures to be collected in this study include the need for secondary endoscopy, overall mortality, transfusion requirement, length of stay, requirement for surgery, and incidence of adverse side effects. Results from this study would help identify a safe, effective, and readily available prokinetic agent to be used prior to endoscopy.

Inclusion Criteria:

• 1. Adult patients ≥ 18 years of age at the time of presentation;
• 2. Admitted to Rhode Island Hospital (RIH) emergency room or inpatient services;
• 3. Presented with hematemesis, coffee ground emesis, or melena;
• 4. Upper endoscopy is planned within 24 hours of presentation or onset of bleeding.

Exclusion Criteria:

• 1. Known anaphylactic allergic reaction to erythromycin, azithromycin, and/or metoclopramide;
• 2. Concurrent use of certain medications associated with tardive dyskinesia (TD):
• a. Fluphenazine,
• b. Haloperidol,
• c. Loxapine,
• d. Paliperidone,
• e. Perphenazine,
• f. Pimozide,
• g. Risperidone,
• h. Thiothixene,
• i. Trifluoperazine;
• 3. Concurrent use of certain medications associated with torsade de pointes:
• a. Amiodarone,
• b. Chlorpromazine,
• c. Disopyramide,
• d. Dofetilide,
• e. Methadone,
• f. Procainamide,
• g. Quinidine,
• h. Sotalol;
• 4. Known history of TD, ventricular arrhythmias , or long QT syndrome;
• 5. Already received erythromycin and/or azithromycin within the past 10 days, or metoclopramide within the past 4 days for other indications;
• 6. Recipient of hematopoietic stem cell transplant;
• 7. History of Neisseria gonorrhoeae infection;
• 8. Pregnancy;
• 9. Prior gastrectomy.

Initial management of UGIB would be performed by the primary team without interference from the investigator. If a gastroenterology consult were placed by the primary team, the endoscopist on-call would evaluate the patient per routine care. If an early endoscopy is deemed appropriate and necessary based on the patient's clinical status, the participant is registered to a central electronic database, which would assign participants to the intervention arm or the placebo arm using a permuted block randomization process. The appropriate medication or placebo is prepared and distributed by the central inpatient pharmacy. The investigator, the endoscopist, the study participant, and the primary team are all blinded to the assignment.