Obesity-related indicators and tuberculosis: A Mendelian randomization study

Abstract

Purpose: Obesity is a strong risk factor for many diseases, with controversy regarding the cause(s) of tuberculosis (TB) reflected by contradictory findings. Therefore, a larger sample population is required to determine the relationship between obesity and TB, which may further inform treatment.

Methods: Obesity-related indicators and TB mutation data were obtained from a genome-wide association study database, while representative instrumental variables (IVs) were obtained by screening and merging. Causal relationships between exposure factors and outcomes were determined using two-sample Mendelian randomization (MR) analysis. Three tests were used to determine the representativeness and stability of the IVs, supported by sensitivity analysis.

Results: Initially, 191 single nucleotide polymorphisms were designated as IVs by screening, followed by two-sample MR analysis, which revealed the causal relationship between waist circumference [odds ratio (OR): 2.13 (95% confidence interval (CI): 1.19-3.80); p = 0.011] and TB. Sensitivity analysis verified the credibility of the IVs, none of which were heterogeneous or horizontally pleiotropic.

Conclusion: The present study determined the causal effect between waist circumference and TB by two-sample MR analysis and found both to be likely to be potential risk factors.

Fig 1. Flow chart of the experimental design of this study.

SNPs for obesity-related indicators were identified as genetic instrumental variables. Assumption 1: The genetic variations are strongly associated with exposure; Assumption 2: The genetic variations are not associated with either known or unknown confounders; Assumption 3: SNPs should influence risk of the outcome through the exposure, not through other pathways.

Fig 2

(a, c): Scatter plot of SNPs associated with waist circumference and risk of TB. The plot shows the SNP effects on waist circumference (x-axis, SD units) as well as TB (y-axis, OR) with 95% CI. The MR regression slopes of the lines represent the causal estimates using five approaches (IVW, MR-Egger, weighted median, simple mode, and weighted mode; (b, d): Funnel diagram corresponding to waist and hip circumference.

Fig 3. Leave-one-out plots for causality analysis of waist circumference, hip circumference, waist-to-hip ratio and body mass index on TB.