Adverse effects of proton pump inhibitors
Menée à partir de données 1994-2000 de registres nationaux dans cinq pays nordiques (Danemark, Finlande, Islande, Norvège et Suède) portant sur 172 297 témoins et 17 232 patients atteints d'un adénocarcinome gastrique (non cardia), cette étude analyse l'association entre une utilisation prolongée d'inhibiteurs de la pompe à protons (au moins 12 mois) et le risque de développer la maladie
Proton pump inhibitors are among the most used drugs worldwide.1 Given how common proton pump inhibitors are, understanding potential harms is essential. Even though rare adverse effects associated with long term proton pump inhibitor use have received considerable attention, uncertainty around the causality of many of these effects remains.23 This is not a trivial matter. Although overuse of proton pump inhibitors, particularly long term use without an indication, is a recognised problem, low quality evidence that unjustly suggests that a useful medicine is harmful could discourage its rational use when indicated and appropriate.
One frequently cited concern is an increased risk of gastric cancer associated with proton pump inhibitor use. However, existing studies have had major methodological limitations.4 To clarify this uncertainty, in a linked paper (doi:10.1136/bmj-2025-086384), Duru and colleagues did a multinational, population based, case-control study evaluating the association between more than one year of proton pump inhibitor use and risk of gastric adenocarcinoma.5 The authors overcame key limitations from previous research, most notably reverse causation. Proton pump inhibitors may be used because of symptoms caused by an as-yet undiagnosed cancer. Failure to consider this will lead to a spurious association between proton pump inhibitor use and cancer. To minimise this bias, Duru and colleagues excluded exposure to proton pump inhibitors in the 12 months before the index date. They found no association between proton pump inhibitor use and risk of gastric cancer. Through a series of illustrative sensitivity analyses, the authors intentionally reintroduced sources of bias seen in other studies. In doing so, they were able to recreate the previously reported, but spurious, findings of an increased risk of gastric adenocarcinoma.
BMJ , éditorial en libre accès, 2026