Non-Variceal Upper GI Bleeding in Established Users of Proton-Pump Inhibitors – Higher Incidence but Better Outcomes.

Objectives: Non-variceal upper gastrointestinal bleeding (NVUGIB) still occurs despite taking proton-pump inhibitors (PPIs). We tested the hypothesis that PPI efficacy fades with time. Methods: This is a prospective and controlled observational comparison of cases of NVUGIB arising from the adult PPI-users’ and non-users’ subpopulations of a well-defined community. Patients presenting with NVUGIB were grouped into PPI-users (index group) and non-users (control group). Results: The incidence of NVUGIB in our overall adult population was 89.7 in 2019 and 115.1 per 100,000 in 2022 (P=0.01). Over the two years, the incidence of NVUGIB in the non-PPI subpopulation was 59.3 vs 364.4 in the PPI subpopulation (P<0.0001). In the index group (n=213), 21 (9.9%) patients developed duodenal ulcers vs 44 (20.2%) in the control group (n=211; P=0.002); 162 (76.1%) were admitted vs 188 (89.1%), P<0.001; 54 (25.5%) required transfusion vs 74 (35.2%), P=0.034; and there was a 30-day mortality of 8 (3.8%) vs 18 (8.5%), P=0.044, respectively in controls. Of 15 cases in the index group with gastric ulcer, 9 (60%) developed in the fourth quartile of duration of PPI use (76-180 months). The duration of PPI use grouped by quartiles was a significant predictor of a finding of gastric ulcer (odds ratio 2.39, 95% CI 1.29-4.46, P=0.006). Conclusions: Apart from gastric ulceration, there was no consistent evidence of a decline in PPI efficacy with time. The need for and long-term use of PPI may identify those at a higher risk of NVUGIB. Registration: International Standard Randomised Clinical Trial number ISRCTN14750381.