Purpose: Elective esophageal variceal ligation (EVL) is a commonly performed procedure done to decrease the risk of variceal hemorrhage. Side effects of elective EVL include hemorrhage, chest pain, dysphagia and odynophagia. Because gastric acid may exacerbate the residual ulcers seen after EVL and delay healing, proton pump inhibition may decrease the side effects associated with EVL. The aim of this study was to assess the efficacy of pantoprazole, a proton pump inhibitor, as an adjunct to elective EVL. Methods: We performed a double-blinded randomized placebo controlled trial of pantoprazole in the setting of elective EVL. Subjects in the pantoprazole arm received pantoprazole 40 mg intravenously immediately after elective EVL and oral pantoprazole 40 mg for nine days following. Subjects in the control arm received identical-appearing intravenous and oral placebo medications. Subjects underwent repeat upper endoscopy 10–14 days after initial banding. Primary outcome variables included the size of the banding ulcers, number of ulcers present after controlling for the number of bands initially placed, and the subjects' reports of dysphagia, chest pain, and heartburn symptoms. Results: Forty-four subjects were randomized, and forty-two successfully completed the protocol. Four subjects, all in the placebo group, had adverse outcomes, including 3 who suffered variceal bleeding (2 after the initial banding session and 1 within 30 days following the second banding session), and one with sepsis. At 10 days post banding, the mean number of ulcers was similar in the pantoprazole and placebo groups. However, the ulcers in the pantoprazole group were on average half as large as those in the placebo group (3.7 mm vs 8.2 mm, p <0.01). Chest pain, dysphagia and heartburn scores were not different between the two groups. Conclusions: Subjects receiving pantoprazole therapy after elective EVL had significantly smaller post-banding ulcers on follow-up endoscopy than those receiving placebo. However, total ulcer number, as well as patient symptoms, were not different between the groups.