An observational study found no clear reduction in overall gastrointestinal bleeding among selected patients taking dual antiplatelet therapy.

A study reporting no benefit from proton pump inhibitors might appear to offer a straightforward prescribing message. This one requires a closer look. Patients with previous peptic ulcers or gastrointestinal bleeding—the very people whose protection often concerns clinicians most—were excluded.

That boundary is essential when interpreting new research on PPI use after ischaemic stroke. The findings raise questions about routine prescribing in a selected population, but cannot establish that gastroprotection is ineffective in patients at higher bleeding risk.

Xinyu Zhou and colleagues reviewed 3,122 patients with acute ischaemic stroke attributed to large artery atherosclerosis at the Affiliated Hospital of Qingdao University, China. All received aspirin with either clopidogrel or ticagrelor. The study was published online in the Journal of Stroke and Cerebrovascular Diseases on October 2, 2026.

Of these patients, 1,404 received a PPI alongside their antiplatelet medicines and 1,718 did not. Gastrointestinal bleeding occurred in 29 PPI users and 23 non-users, corresponding to rates of 2.1% and 1.3%.

After adjustment for measured patient characteristics, PPI use was not associated with a statistically significant reduction in bleeding. The adjusted odds ratio was 1.51, with a 95% confidence interval of 0.87–2.62. A propensity-matched analysis involving 1,222 pairs produced a similarly inconclusive result.

These estimates do not prove equivalence, benefit or harm. Treatment was selected in clinical practice rather than randomly assigned, leaving the possibility that doctors prescribed PPIs to patients with concerns not fully captured in the records. Statistical matching can balance recorded characteristics; it cannot remove every prescribing difference.

There is another important distinction. The investigators assessed gastrointestinal bleeding overall without separating upper from lower tract events. Consequently, the study cannot determine whether PPIs affected bleeding from acid-related upper gastrointestinal lesions differently from bleeding elsewhere.

The outcome required documented bleeding manifestations or faecal occult blood positivity accompanied by a haemoglobin decline of at least 2 g/dL. This was therefore broader than an endpoint restricted to endoscopically confirmed ulcer bleeding.

An unexpected subgroup finding deserves particular restraint. Among patients with posterior circulation infarction who had not undergone endovascular treatment, PPI use was associated with more bleeding. However, event numbers were small, and the matched estimate had a very wide confidence interval. This is a hypothesis for further investigation, not evidence that PPIs cause bleeding in this subgroup.

Other gaps limit bedside interpretation. Specific PPI agents, doses and treatment durations were incompletely available. Medication adherence could not be quantified, and potentially relevant information, including H. pylori status and stroke severity, was not fully accounted for.

The exclusions also narrow applicability beyond previous ulcers and bleeding. Patients with severe hepatic or renal dysfunction and those using chronic NSAIDs or glucocorticoids were excluded. The findings should not be extended automatically to these groups.

For practising doctors, the useful action is to revisit the indication written beside the PPI prescription. Establish the patient’s gastrointestinal history, concurrent medicines and reason for acid suppression. A negative observational finding in a selected cohort should not override an established individual indication.

Equally, a PPI entry should not create false reassurance when bleeding develops. Investigate the source and review antithrombotic management with the treating team. The clinical decision remains patient-specific: prescribe gastroprotection for a documented reason, and reassess that reason as treatment changes.

Tags:    

Disclaimer: This website is primarily for healthcare professionals. The content here does not replace medical advice and should not be used as medical, diagnostic, endorsement, treatment, or prescription advice. Medical science evolves rapidly, and we strive to keep our information current. If you find any discrepancies, please contact us at corrections@medicaldialogues.in. Read our Correction Policy here. Nothing here should be used as a substitute for medical advice, diagnosis, or treatment. We do not endorse any healthcare advice that contradicts a physician's guidance. Use of this site is subject to our Terms of Use, Privacy Policy, and Advertisement Policy. For more details, read our Full Disclaimer here.

NOTE: Join us in combating medical misinformation. If you encounter a questionable health, medical, or medical education claim, email us at factcheck@medicaldialogues.in for evaluation.

Our comments section is governed by our Comments Policy . By posting comments at Medical Dialogues you automatically agree with our Comments Policy , Terms And Conditions and Privacy Policy .