Associations of Helicobacter pylori infection across the cardiometabolic and steatotic liver disease spectrum: an umbrella review with targeted reanalysis of selected source reviews
Background Whether H. pylori infection is associated with cardiometabolic and steatotic liver diseases remains uncertain. Methods We searched five information sources from inception to July 12, 2026, for systematic reviews with meta-analyses; no language restriction was applied during searching, but non-English reports were excluded at screening. For five ratio-measure outcomes, we reanalyzed study-level data from one selected source review per outcome with extractable data (searches ending 2016–2023), using restricted maximum-likelihood random-effects models with Hartung–Knapp adjustment, prediction intervals, bias diagnostics, Ioannidis credibility criteria (classes I, convincing, to IV, weak), and an exploratory GRADE-based certainty assessment. Sensitivity analyses addressed the model, unit of analysis, adjustment, and source-review choice. Results ROBIS rated 30 of 37 reviews at high risk of bias, and all had critically low confidence on AMSTAR 2, a tool designed for intervention reviews. H. pylori infection was associated with prevalent nonalcoholic fatty liver disease [NAFLD; odds ratio (OR): 1.28; 95% confidence interval (CI): 1.14–1.44], prevalent metabolic dysfunction-associated steatotic liver disease (MASLD; OR: 1.12; 95% CI: 1.02–1.22), diabetes of any type (OR: 1.27; 95% CI: 1.09–1.47; type 2 diabetes, 1.43), metabolic syndrome (source-defined ratio, 1.34; 95% CI: 1.16–1.54), and hypertension (OR: 1.35; 95% CI: 1.04–1.76). Prediction intervals included 1.0 for four outcomes, so effects in new settings could be null. Most estimates were crude or of uncertain adjustment status, and adjustment attenuated the NAFLD association (OR: 1.21). No outcome met class I. NAFLD met class II criteria under the conventional DerSimonian–Laird model but class III under the primary model, under which no outcome exceeded class III; estimates and classes also changed with newer source reviews. The liver syntheses shared most primary studies (corrected covered area, 65.6%), and the published insulin-resistance estimate could not be reproduced. Certainty was very low throughout, and eradication evidence was inconsistent. Conclusions Associations were modest, heterogeneous, and based largely on overlapping, inconsistently adjusted cross-sectional evidence; NAFLD's higher credibility class does not indicate a stronger or causal association. In our interpretation, current evidence does not support testing for or eradicating H. pylori solely for a metabolic indication. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420261435859 , PROSPERO, identifier CRD420261435859.
Authors
- Yang Won Min (ORCID: https://orcid.org/0000-0001-7471-1305)
- Hyosoon Yoo
- Tae Jun Kim
- Hyuk Lee
Institutions
- Samsung Medical Center (KR)
- Hallym University Dongtan Sacred Heart Hospital (KR)
- Sungkyunkwan University (KR)
Publication Details
- Journal
- Frontiers in Medicine
- Published
- 2026-11-20
- DOI
- https://doi.org/10.3389/fmed.2026.1964694
- Primary Topic
- Helicobacter pylori-related gastroenterology studies
- Type
- article
- Field-Weighted Citation Impact
- 0.00