High-Dose Proton Pump Inhibitors Are Superior to Standard-Dose Proton Pump Inhibitors in High-Risk Patients With Bleeding Ulcers and High-Risk Stigmata After Endoscopic Hemostasis

INTRODUCTION: To define the best cutoff of the Glasgow-Blatchford score (GBS) for identifying high- and low-risk rebleeding patients with bleeding ulcers and high-risk stigmata after endoscopic hemostasis and compare the efficacy of high-dose and standard-dose intravenous proton pump inhibitors (HD-IVPs and SD-IVPs, respectively) in this patient population. METHODS: We retrospectively reviewed the data of 346 patients with bleeding ulcers and high-risk stigmata who underwent endoscopic hemostasis between March 2014 and September 2018 in our center and were divided into an HD-IVP group and an SD-IVP group. Propensity score–matching analysis was performed to control for selection bias and other potential confounders. Recurrent bleeding rates were calculated according to the GBS. RESULTS: Overall, 346 patients meeting the inclusion criteria were enrolled, with 89 patients in the SD-IVP group and 89 patients in the HD-IVP group after matching with all baseline characteristics balanced (P > 0.05). GBS = 8 was the best cutoff for identifying high-risk rebleeding patients (GBS ≥ 8) with a significant difference (P = 0.015) in recurrence rate between the SD-IVP (17/61, 27.9%) and HD-IVP (7/65, 10.8%) groups and low-risk rebleeding patients (GBS < 8) with no difference (P = 1) in recurrence rate between the SD-IVP (2/28, 7.1%) and HD-IVP (2/24, 8.3%) groups. DISCUSSION: The best cutoff for identifying high-risk and low-risk rebleeding patients with bleeding ulcers and high-risk stigmata after endoscopic hemostasis was GBS = 8. Although HD-IVP is more effective than SD-IVP in high-risk patients, they are equally effective in low-risk patients.