Causal relationship between BMI-mediated different modes of transportation and risk of osteoarthritis: a Mendelian randomization and mediation analysis.

OBJECTIVE

The importance of Transport-related Physical Activity in chronic disease management is increasingly recognized. Previous studies have reported relationships between different modes of transportation and the risk of osteoarthritis (OA), but the conclusions across studies have been inconsistent. Body Mass Index (BMI) has been shown to be directly associated with different transportation modes and is also a significant risk factor for OA. Therefore, our study aims to investigate the causal relationship between different modes of transportation and OA, mediated by BMI, using Mendelian randomization (MR) and mediation analysis.

METHODS

Genome-wide association study (GWAS) data for different transportation modes were extracted from the IEU openGWAS database as exposure data. GWAS summary data for BMI were used as potential mediator variables, and OA data from the IEU openGWAS database were used as outcome data. A two-sample, two-step MR analysis was performed using the Inverse-Variance Weighted (IVW) method to investigate the causal relationships between different modes of transportation and OA, as well as the mediating role of BMI in these associations. Supplementary analyses were conducted using the weighted median method and MR-Egger regression. Cochran's Q test was used to assess heterogeneity among the SNPs, and the MR-Egger intercept test was applied to examine horizontal pleiotropy.

RESULTS

The MR analysis revealed significant associations between transportation modes and the risk of OA. Cycling (OR = 0.568, 95% CI = 0.325-0.991, P = 0.047), public transportation (OR = 0.584, 95% CI = 0.426-0.802, P = 0.0009), and walking (OR = 0.497, 95% CI = 0.376-0.657, P < 0.0001) were negatively associated with OA risk, whereas driving (OR = 1.19, 95% CI = 1.018-1.385, P = 0.029) was positively associated with OA risk. Mediation analysis indicated that BMI partially mediated the causal relationship between cycling, public transportation, walking, driving, and OA risk, with mediation proportions of 69.6%, 31.9%, 25.0%, and 39.8%, respectively. Cochran's Q test indicated some heterogeneity among the SNPs included in the analysis, and the MR-Egger intercept test suggested no evidence of genetic pleiotropy.

CONCLUSION

Genetically predicted walking, cycling, or public transportation use was negatively associated with the risk of OA, whereas driving increased the risk of OA. BMI mediated over 25% of the causal relationship between different transportation modes and OA.

CLINICAL TRIAL NUMBER

Not applicable.