Inflammation and keratoconus: A comprehensive bidirectional Mendelian randomization analysis

Abstract

An increasing body of evidence supports the involvement of inflammation and immune responses in the occurrence and development of keratoconus (KC). However, the causal relationship between inflammatory factors and KC remains unclear. We employed a 2-way Mendelian randomization (MR) approach to investigate the interaction between KC and inflammatory factors. Instrumental variables for 41 circulating inflammatory regulators and 12 matrix metalloproteinases (MMPs) were selected from genome-wide association studies of European ancestry. Summary statistics for KC were obtained from a genome-wide association study comprising 2116 cases and 24,626 controls of European ancestry. The primary analytical method for assessing causality was the inverse-variance weighted method. Two additional MR methods (MR-Egger and weighted median) were employed to complement the inverse-variance weighted results. In addition, several sensitivity analyses were conducted to evaluate heterogeneity, horizontal pleiotropy, and stability. Our findings indicated that genetically predicted higher levels of macrophage inflammatory protein-1β (odds ratio = 1.126, 95% confidence interval: 1.029–1.232, P = .01) and MMP-13 (odds ratio = 1.211, 95% confidence interval: 1.070–1.371, P = .003) were positively associated with an elevated risk of KC. Conversely, genetically predicted KC was associated with increased levels of interferon-gamma, interleukin-4, and MMP-1. Our current study provided suggestive evidence supporting causal associations of macrophage inflammatory protein-1β and MMP-13 with the risk of KC. In addition, KC appeared to affect the expression of interferon-gamma, interleukin-4, and MMP-1.