Causal effects of COVID-19 on structural changes in specific brain regions: a Mendelian randomization study
Shaojiong Zhou,Tao Wei,Xiaoduo Liu,Yufei Liu,Weiyi Song,Xinwei Que,Yi Xing,Zhibin Wang,Yi Tang
DOI: https://doi.org/10.1186/s12916-023-02952-1
IF: 9.3
2023-07-19
BMC Medicine
Abstract:Abstract Background Previous studies have found a correlation between coronavirus disease 2019 (COVID-19) and changes in brain structure and cognitive function, but it remains unclear whether COVID-19 causes brain structural changes and which specific brain regions are affected. Herein, we conducted a Mendelian randomization (MR) study to investigate this causal relationship and to identify specific brain regions vulnerable to COVID-19. Methods Genome-wide association study (GWAS) data for COVID-19 phenotypes (28,900 COVID-19 cases and 3,251,161 controls) were selected as exposures, and GWAS data for brain structural traits (cortical thickness and surface area from 51,665 participants and volume of subcortical structures from 30,717 participants) were selected as outcomes. Inverse-variance weighted method was used as the main estimate method. The weighted median, MR-Egger, MR-PRESSO global test, and Cochran’s Q statistic were used to detect heterogeneity and pleiotropy. Results The genetically predicted COVID-19 infection phenotype was nominally associated with reduced cortical thickness in the caudal middle frontal gyrus ( β = − 0.0044, p = 0.0412). The hospitalized COVID-19 phenotype was nominally associated with reduced cortical thickness in the lateral orbitofrontal gyrus ( β = − 0.0049, p = 0.0328) and rostral middle frontal gyrus ( β = − 0.0022, p = 0.0032) as well as with reduced cortical surface area of the middle temporal gyrus ( β = − 10.8855, p = 0.0266). These causal relationships were also identified in the severe COVID-19 phenotype. Additionally, the severe COVID-19 phenotype was nominally associated with reduced cortical thickness in the cuneus ( β = − 0.0024, p = 0.0168); reduced cortical surface area of the pericalcarine ( β = − 2.6628, p = 0.0492), superior parietal gyrus ( β = − 5.6310, p = 0.0408), and parahippocampal gyrus ( β = − 0.1473, p = 0.0297); and reduced volume in the hippocampus ( β = − 15.9130, p = 0.0024). Conclusions Our study indicates a suggestively significant association between genetic predisposition to COVID-19 and atrophy in specific functional regions of the human brain. Patients with COVID-19 and cognitive impairment should be actively managed to alleviate neurocognitive symptoms and minimize long-term effects.
medicine, general & internal