Answers are grounded in this paper and cite their sources.
Retrospective cohort study reveals increased PPI deprescription rates in UGIB patients, indicating adherence to guidelines.
Objective Proton pump inhibitors (PPIs) are widely prescribed but inappropriate indications and concerns over long-term side effects have led to recommendations to deprescribe PPIs in certain patients. We previously found a 4-fold increase in PPI deprescription in patients with esophageal strictures. This study aims to assess the PPI deprescription rate in patients with upper gastrointestinal bleeding (UGIB). Methods All patients from 2 gastroenterology practices who received endoscopic treatment for UGIB during the years of 2015-2022 were identified using physician billing codes. We defined PPI deprescription as either a 50% dose reduction, frequency reduction, or complete medication discontinuation at the time of endoscopic intervention compared to the established PPI therapy from the 3 months prior. We compared the rate of PPI deprescription between 2 time periods 2015-2018 (group 1) and 2019-2022 (group 2). Results Three hundred one UGIB managed with endoscopy were analyzed. Patients in group 2 had a significantly higher rate of PPI deprescription than group 1 (15% vs 4%; P < .002). Patients with peptic ulcer disease (PUD) had a significantly higher PPI deprescription during the second time period (16% vs 0%; P = .028). Among patients with repeat UGIB, 10% had their PPI deprescribed. Conclusions Proton pump inhibitor deprescription in patients with UGIB treated with endoscopic intervention was more common in the second time period. This corresponds to when PPI deprescription guidelines were distributed. Physicians should ensure the appropriate application of PPI deprescription guidelines and continuation of PPI therapy for patients with strong indications.
No takes yet. Share an insight, caveat, or question.
Kecskemeti et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: