SRF301 - Noise Exposure and Glucose Metabolism: Evidence from Epidemiological and Mendelian Randomization Studies

This study explored how occupational, recreational, and firearm use influence glucose metabolism. Plasma fasting glucose levels (PFG) and glucose clearance rates were assessed in a sample of 2035 and 1435 individuals, respectively. Linear regression analysis, controlling for demographic factors, indicated that noisy occupations were linked to higher PFG levels. Mendelian Randomization (MR)analysis hinted at a modest causal connection between noise and PFG levels. Participants with firearm use and noisy jobs displayed faster glucose clearance, but MR analysis did not support a causal link between genetically predicted noise exposure and glucose clearance. Mechanistic research is needed for deeper insights.

Summary:

Rationale: Understanding the potential influence of noise exposure on glucose metabolism is a vital research endeavor, given the widespread prevalence of noise in our modern society and its established links to various adverse health effects. While prior studies produced mixed findings, recent research hints at a link between noise and elevated plasma fasting glucose levels. Moreover, mechanistic studies in model organisms suggest noise's impact on metabolic health markers, warranting further investigation in a diverse, representative population. Traditional epidemiological studies often face limitations like confounding and bias. To overcome these challenges and provide stronger evidence, this study utilizes MR. By using genetic variants associated with noise exposure as instrumental variables, MR minimizes confounding and reverse causation, enabling a more reliable assessment of the causal relationship between noise and glucose metabolism. This approach aligns with randomized control trial principles, enhancing our understanding of how noise affects metabolic health. 

Design: The analysis involved four distinct cohorts: 1) all participants (N=2035), 2) participants with normal PFG levels (N=879), 3) all participants who underwent OGTT (N=1435), and 4) participants with normal PFG levels who underwent OGTT (N=668). Three noise exposure variables were extracted: firearm use, noisy job, and recreational noise. We used two glucose variables PFG and OGTT. A new variable called Oral Glucose Tolerance Percentage (OGTP) was introduced which captures the percentage change in blood glucose levels before and after glucose consumption. We also considered race, sex and age, physical activity, and smoking history in our analysis. In the Two-sample one-direction MR analysis, publicly available GWAS summary statistics were used for noisy workplace exposure and glucose metabolism outcomes. SNPs with p-values < 10-5 were included, and multiple MR methods were employed to assess causality while accounting for potential heterogeneity. 

Results:  Noisy job participants exhibited higher PFG levels, and MR analysis confirmed a modest causal connection between noisy jobs and elevated PFG levels, consistent with NHANES data. Firearm use, and noisy jobs (With normal PFG) were linked to faster glucose clearance, but no causal connection was found via MR. Noisy job, firearm use, and recreational noise exposure were associated with higher physical activities and a higher likelihood of reporting smoking. The MR analysis revealed complementary findings. The results revealed significant main effects of sex (male >female), ethnicity (non-Europeans > Europeans), and age (older > younger) on PFG levels, consistent with the literature. The results revealed significant main effects of age (older < younger) on OGTT. 

Conclusions: The study found a link between occupational noise exposure and slightly elevated PFG levels, supported by MR analysis indicating a modest causal connection. Firearm use and noisy jobs in participants with normal PFG were linked to faster glucose clearance, but MR analysis didn't confirm a causal link. The study relied on self-reported noise exposure, suggesting potential accuracy improvement through objective measures like noise dosimetry. The p-value threshold was set at 10-5 due to limited genetic associations with noise exposure, suggesting the need for future replication with more robust data. Mechanistic research is needed to understand the underlying molecular pathways driving these findings.