Metabolites and Charcot Foot: A Comprehensive Analysis Through Mendelian Randomization.

Abstract

Background: Multiple studies have shown metabolites may have potential effects on Charcot foot. However, the Mendelian randomization method has not yet explored the relationship between metabolites and Charcot foot.

Methods: We selected genetic variants from the publicly available Genome-wide Association Studies (GWAS) summary database to represent 1400 metabolites described in recent research. Mendelian randomization (MR) analysis was carried out to examine the relationships between these metabolites and Charcot foot. Significant single nucleotide polymorphism (SNP) data associated with exposure were screened out through association analysis. Valid instrumental variables (IVs) were then selected, excluding SNPs with F-statistic values below 10. The MR analyses primarily employed the inverse variance weighted (IVW) method. Bayesian weighted Mendelian randomization (BWMR), constrained maximum likelihood(cML), contamination mixture(Conmix), robust adjusted profile score(RAPS), and debiased inverse-variance weighted(deIVW) method were used to enhance the results. Colocalization analysis was performed to identify shared causal genetic variants associated with the resulting phenotypes. Sensitivity analyses, including assessments of Cochrane's Q test, egger intercept, and MR PRESSO test were conducted to confirm the robustness of the results.

Results: After preliminary MR exploration, the IVW results exhibited positive causal relationships between hexadecenedioate (C16:1-DC) levels (OR = 0.698, 95%CI: 0.586 to 0.831, P_FDR= 0.040), octadecadienedioate (C18:2-DC) levels (OR = 0.665, 95%CI: 0.552 to 0.800, P_FDR= 0.021), octadecanedioylcarnitine (C18-DC) levels (OR = 0.676, 95%CI: 0.553 to 0.827, P_FDR= 0.067) and Charcot foot. Colocalization analysis indicated that the above three metabolites share a common causal variant at the same genomic location with Charcot foot. Sixty-four metabolites with suggestive causal relationships with Charcot foot were also identified, among which 25 kinds of metabolites were positively correlated with Charcot foot, and 33 metabolites were negatively associated with Charcot foot. The BWMR, cML, Conmix, RAPS, and deIVW results supported our preliminary MR results. In several results, sensitivity analyses showed heterogeneity and horizontal pleiotropy, while the causal relationships obtained through FDR correction did not show any significant heterogeneity and horizontal pleiotropy. No reverse causal association was detected.

Conclusion: We detected protective and risk metabolites in Charcot foot. Controlling metabolites may decrease Charcot foot risk and serve as a novel therapeutic biomarker for the therapy.