Meta‐Analysis: Prevalence of Non‐
                    <scp>
                      <i>Helicobacter pylori</i>
                    </scp>
                    , Non‐
                    <scp>NSAID</scp>
                    Peptic Ulcer Disease

doi:10.1111/apt.70906

DOI: 10.1111/apt.70906 ISSN: 0269-2813

Meta‐Analysis: Prevalence of Non‐ Helicobacter pylori , Non‐ NSAID Peptic Ulcer Disease

Jun Watanabe, Hiromi Sekiguchi, Takeshi Kanno, Keigo Misawa, Satoshi Sato, Yuhong Yuan, Paul Moayyedi, Tomonori Yano, Atsushi Masamune, Tomoari Kamada

Show PDF Cite

ABSTRACT

Background

Non‐ Helicobacter pylori ( H. pylori ), non‐nonsteroidal anti‐inflammatory drug (NSAID) peptic ulcer disease (PUD) is increasingly recognized as a distinct subgroup, but its proportion across clinical presentations remains unclear.

Aim

We aimed to estimate this proportion.

Methods

We searched MEDLINE, Embase and CENTRAL from their inception to March 2026 for observational studies. The primary outcome was the proportion of non‐ H. pylori , non‐NSAID PUD among adults with overall, bleeding and perforated PUD. Random‐effects meta‐analysis of logit‐transformed proportions was performed using restricted maximum likelihood. Meta‐regression assessed associations with study year, country‐level H. pylori prevalence and region. Protocol registered in PROSPERO (CRD420251174202).

Results

Overall, 92 studies (60,816 patients) from 32 countries were included. The pooled proportions of non‐ H. pylori , non‐NSAID PUD cases were 13.0% (95% confidence interval [CI], 10.4–16.2) in overall PUD, 11.3% (95% CI, 8.8–14.4) in bleeding PUD, and 22.0% (95% CI, 13.1%–34.7%) in perforated PUD. In univariable meta‐regression, a more recent study year was associated with a higher proportion of bleeding PUD ( β = 0.076; 95% CI, 0.031–0.120). This association remained statistically significant after adjustment for country‐level H. pylori prevalence ( β = 0.070; 95% CI, 0.026–0.113). Lower H. pylori prevalence was also associated with a higher proportion after further adjustment for region ( β = −0.025; 95% CI, −0.046 to −0.005).

Conclusion

Non‐ H. pylori , non‐NSAID PUD represents a non‐negligible but heterogeneous component of PUD. In bleeding PUD, exploratory ecological analyses suggest a temporal increase, potentially reflecting evolving etiologic patterns beyond H. pylori infection and NSAID exposure.

Outline