Erschienen in:
24.03.2023 | Original Contribution
Association between salt intake and gastric atrophy by Helicobacter pylori infection: first results from the Epidemiological Investigation of Gastric Malignancy (ENIGMA)
verfasst von:
Viktoria Knaze, Heinz Freisling, Paz Cook, Katy Heise, Johanna Acevedo, Marcos Cikutovic, Karl-Heinz Wagner, Rodrig Marculescu, Catterina Ferreccio, Rolando Herrero, Jin Young Park
Erschienen in:
European Journal of Nutrition
|
Ausgabe 5/2023
Einloggen, um Zugang zu erhalten
Abstract
Purpose
Gastric atrophy (GA), usually linked to chronic infection with Helicobacter pylori (H. pylori), may over time evolve into gastric malignancy. Besides H. pylori, high salt intake may play a role in GA development. This study evaluates cross sectionally the association between salt intake and GA in Chilean adults.
Methods
Population-based samples were recruited from two sites, Antofagasta and Valdivia, partaking in the Epidemiological Investigation of Gastric Malignancies. At recruitment, participants answered questionnaires and provided biospecimens. Salt intake (g/day) was estimated from casual spot urine samples using the Tanaka equation. GA was determined by serum pepsinogen levels. Only participants ≥ 40 to 70 years of age were considered in this analysis, n = 565. For the association between salt intake (as sex-specific quartiles) and GA, odds ratios (ORs) and the corresponding 95% confidence intervals (CI) were estimated through multivariable logistic regression.
Results
In women, the multivariable-adjusted OR for GA comparing quartile 4 of the estimated salt intake (12.8 g/day) to quartile 1 (6.6 g/day) was 1.18 (95% CI 0.52–2.68, P-trend = 0.87). The corresponding OR in men was 0.49 (95% CI 0.19–1.27, P-trend = 0.17) with salt intakes of 12.8 g/day and 7.1 g/day for quartiles 4 and 1, respectively.
Conclusion
There was little evidence for an association between salt intake estimated from spot urine and GA risk in our cross-sectional analysis of middle aged and older adults in Chile. Reverse causation bias cannot be ruled out and the sample size was limited to provide more precise estimates.