L-shaped association of triglyceride glucose index and sensorineural hearing loss: results from a cross-sectional study and Mendelian randomization analysis.

BACKGROUND

The association between the sensorineural hearing loss (SNHL) and triglyceride-glucose (TyG) index remains inadequately understood. This investigation seeks to elucidate the connection between the TyG index and SNHL.

METHODS

In this cross-sectional study, we utilized datasets sourced from the National Health and Nutrition Examination Survey (NHANES). A comprehensive analysis was conducted on 1,851 participants aged 20 to 69, utilizing complete audiometry data from the NHANES database spanning from 2007 to 2018. All enrolled participants had accessible hearing data, and the average thresholds were measured and calculated as both low-frequency pure-tone average and high-frequency pure-tone average. Sensorineural hearing loss (SNHL) was defined as an average pure tone of 20 dB or higher in at least one better ear. Our analysis involved the application of multivariate linear regression models to examine the linear relationship between the TyG index and SNHL. To delineate any non-linear associations, we utilized fitted smoothing curves and conducted threshold effect analysis. Furthermore, we conducted a two-sample Mendelian randomization (MR) study, leveraging genetic data from genome-wide association studies (GWAS) on circulating lipids, blood glucose, and SNHL. The primary analytical method for the MR study was the application of the inverse-variance-weighted (IVW) approach.

RESULTS

In our multivariate linear regression analysis, a substantial positive correlation emerged between the TyG index and SNHL [2.10 (1.80-2.44), p < 0.0001]. Furthermore, using a two-segment linear regression model, we found an L-shaped relationship between TyG index, fasting blood glucose and SNHL with an inflection point of 9.07 and 94 mg/dL, respectively. Specifically, TyG index [3.60, (1.42-9.14)] and blood glucose [1.01, (1.00-1.01)] concentration higher than the threshold values was positively associated with SNHL risk. Genetically determined triglyceride levels demonstrated a causal impact on SNHL (OR = 1.092, p = 8.006 × 10). In addition, blood glucose was found to have a protective effect on SNHL (OR = 0.886, p = 1.012 × 10).

CONCLUSIONS

An L-shaped association was identified among the TyG index, fasting blood glucose, and SNHL in the American population. TyG index of more than 9.07 and blood glucose of more than 94 mg/dL were significantly and positively associated with SNHL risk, respectively.