A Mendelian randomization study of the entire phenome to explore the causal links between epilepsy.

Abstract

The causes and triggering factors of epilepsy are still unknown. The results of genome-wide association studies can be utilized for a phenome-wide association study using Mendelian randomization (MR) to identify potential risk factors for epilepsy. This study utilizes two-sample MR analysis to investigate whether 316 phenotypes, including lifestyle, environmental factors, blood biomarker, and more, are causally associated with the occurrence of epilepsy. The primary analysis employed the inverse variance weighted (IVW) model, while complementary MR analysis methods (MR Egger, Wald ratio) were also employed. Sensitivity analyses were also conducted to evaluate heterogeneity and pleiotropy. There was no evidence of a statistically significant causal association between the examined phenotypes and epilepsy following Bonferroni correction (p<1.58 × 10-4) or false discovery rate correction. The results of the MR analysis indicate that the frequency of tiredness or lethargy in the last 2weeks (p=0.042), blood uridine (p=0.003), blood propionylcarnitine (p=0.041), and free cholesterol (p=0.044) are suggestive causal risks for epilepsy. Lifestyle choices, such as sleep duration and alcohol consumption, as well as biomarkers including steroid hormone levels, hippocampal volume, and amygdala volume were not identified as causal factors for developing epilepsy (p>0.05). Our study provides additional insights into the underlying causes of epilepsy, which will serve as evidence for the prevention and control of epilepsy. The associations observed in epidemiological studies may be partially attributed to shared biological factors or lifestyle confounders.