Background: The diagnostic and therapeutic value of EGD in patients with upper GI bleeding is often limited by the presence of residual blood or clots. When infused before EGD, erythromycin, a potent gastrokinetic drug, might improve the quality of EGD in patients with upper GI bleeding. The aim of this study was to assess the effect of erythromycin on the quality of the EGD in patients with acute upper GI bleeding. Methods: Forty-one patients admitted to the intensive care unit because of acute upper GI bleeding were included in a randomized, endoscopist-blinded, controlled trial, comparing erythromycin (3 mg/kg intravenously over 30 minutes) to no treatment (control patients). EGD was performed 30 to 90 minutes after the end of the infusion. The primary study outcome was quality of EGD examination, as evaluated on a scale of 0 to 3, and the need to perform a second diagnostic EGD within the next 8 days. Secondary outcomes were efficiency of the endoscopic hemostatic therapy and intensity of esophagogastroduodenal contractions, as evaluated on a scale of 0-3. Results: The characteristics of the patients and lesions found by EGD were similar in both groups. Quality of the EGD examination was significantly better in the erythromycin group (n = 19) compared with the control group (n = 22) (2.5 [0.8] vs. 1.5 [1.3]; p = 0.02). Three patients receiving erythromycin required a second EGD compared with 10 control patients (p = 0.089). Erythromycin did not interfere with the performance of EGD or the efficiency of hemostatic procedures. The intensity of gastroduodenal contractions was similar in both groups. No adverse event was observed. Conclusions: In patients with acute upper GI bleeding, infusion of erythromycin before endoscopy significantly improved the quality of EGD and tended to reduce the need for second-look endoscopy. (Gastrointest Endosc 2002;56:174-9.)