Ting Zhang

Ph.D.

National Cancer Institute

Postdoctoral Fellow

PLCO (Learn more about this study)

PLCO-1673

Sep 26, 2024

Dietary sodium consumption and oral microbiome

Background: High sodium intake is known to raise blood pressure and has been associated with risks of cardiovascular disease and gastric cancer. The underlying mechanisms are not clear. Sodium has known antibacterial properties through osmosis, which may affect the microbiome thus affecting host metabolome and biological prosesses. High sodium intake has been shown to affect the gut microbiome in mice. Population-based studies have suggested that habitual excessive sodium intake was associated with microbial between-person diversity (beta diversity), pathogenic Staphylococcus and Moraxellaceae, and host metabolites. A controlled feeding study of sodium intake showed changed plasma levels of multiple metabolites related to the gut microbiome. Although sodium has been widely used for mouthwash, the effect of dietary sodium intake on the oral microbiome is poorly understood, which may provide insights into mechanisms of sodium-associated conditions. Building upon the previous work, we propose to evaluate the associations of dietary sodium consumption and sodium-rich food consumption with the oral microbiome in 3 cohorts, including the NIH-AARP, the Agricultural Health Study (AHS), and PLCO.
Study populations: 2402 (1347 cases, 1055 non-cases)), 1790 (1100 cases, 690 non-cases), and 2341 (1272 cases, 1069 non-cases) participants from case-cohort studies of any cancer (lung, colorectal, pancreas, head-neck, stomach, esophagus, hepatobiliary, small intestine) within the NIH-AARP, AHS, and PLCO, respectively.
Exposures: Dietary intakes of sodium, salt and foods known to be high in sodium, assessed by using FFQ at baseline.
Outcomes: Oral microbiome in oral wash samples using 16S rRNA gene sequencing.
Statistical analysis:
We will assess differences in baseline characteristics across different categories of sodium intake (e.g., ≤ and >2300 mg/d) using Wilcoxon’s Signed Rank or Kruskal-Wallis tests for continuous variables and chi-square or Fisher’s exact tests for categorical variables.
All statistical analyses will be adjusted for complex sampling design to generate unbiased estimates and appropriate variances by weighted regression using R survey package or SAS SUDAAN package. Within each study, we will use linear regression to examine the association between sodium intake and each continuous oral microbiome diversity metrics (the four alpha diversity metrics, and the top five PCoA vectors). For genus presence/absence, we will use logistic regression models, while associations with genus relative abundance will be evaluated using a pseudo-Poisson regression model to account for the extra variability. In the latter, genus-level sequence counts will serve as the outcome, with total sequence counts included as an offset. The regression models will be adjusted for age at blood collection (continuous and quartiles), sex (male or female), education, body mass index (kg/m2), smoking status, self-reported hypertension, alcohol intake, and case-control status. Study-specific estimates will be meta analyzed using fixed effects models. Heterogeneity by study will be assessed using the Cochrane Q statistic. To account for multiple testing, we will use a false discovery rate of <0.05 as the threshold of statistical significance.