Brain volume and network changes in pediatric MOGAD: A retrospective cohort study based on non-acute brain MRI
Shangru Li, Zhihao Li, Yun Qin, Xiang Ding, Xuting Chang, Jie Zhang, Yao Sui, Ye WuBackground
Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD), as the most common type of pediatric central nervous system idiopathic inflammatory demyelinating diseases, has an unclear impact on brain volume and brain networks in children.
Objective
To investigate differences in brain volume and network in children with MOGAD and controls, and analyze their related factors and relationship with neurological function in MOGAD.
Design
This was a single-center retrospective study from Peking University First Hospital.
Methods
Children with MOGAD and controls were enrolled. For MOGAD children, clinical data and non-acute phase brain MRI (≥3 months post-attacks) before the last follow-up were collected. We analyzed brain morphological features and networks using the Morphometric Inverse Divergence (MIND) method. Group comparisons and associations with clinical factors/neurological function were analyzed.
Results
We analyzed 73 brain MRI scans from 73 children with MOGAD and 401 brain MRI scans from 401 age- and sex-matched control children. We found: (1) Children with MOGAD showed a statistically significant reduction in total brain volume (TBV) compared with controls, with cortical gray matter volume (cGMV) being significantly reduced. (2) The cortical thinning regions in children with MOGAD was concentrated in the frontal-parietal junction and temporal areas. (3) The MIND brain network connectivity in children with MOGAD was weakened in the salience network, default mode network (DMN), frontoparietal control network, auditory and language network. (4) Onset age negatively correlated with brain volumes. And EDSS score was negatively correlated with TBV, while average academic performance was positively correlated with TBV.
Conclusions
Pediatric MOGAD involves reduced brain volume, with cortical thinning concentrated in the frontoparietal junction and temporal lobe, and brain network abnormalities in the salience network, DMN, frontoparietal control network, auditory and language network.