Longitudinal behavioural and brain morphological changes before and after hemispherotomy
Presented During:
Thursday, June 27, 2024: 11:30 AM - 12:45 PM
COEX
Room:
Grand Ballroom 101-102
Poster No:
370
Submission Type:
Abstract Submission
Authors:
Ziyu Bao1, Hao Yu2, Yijun Chen1, Lixin Cai2, Gaolang Gong1
Institutions:
1State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Resea, Beijing, China, 2Pediatric Epilepsy Center, Peking University First Hospital, Beijing, China
First Author:
Ziyu Bao
State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Resea
Beijing, China
State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Resea
Beijing, China
Co-Author(s):
Yijun Chen
State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Resea
Beijing, China
State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Resea
Beijing, China
Gaolang Gong
State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Resea
Beijing, China
State Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Resea
Beijing, China
Introduction:
Hemispherotomy is an effective surgery for treating refractory epilepsy from diffuse unihemispheric lesions[1], [2]. To date, however, postsurgery neuroplastic changes following pediatric hemispherotomy remain unclear. In the present study, we aim to systematically investigate longitudinal changes in gray matter volume (GMV) before and after surgery in two groups of pediatric patients with left and right hemispherotomy.
Methods:
Pediatric epilepsy patients undergoing left or right hemispherotomy at the pediatric epilepsy center and having high-quality pre- and postoperative structural MRI were included (29 left hemispherotomy patients, age of surgery: 3.5±2.5 years; 28 right hemispherotomy patients, age of surgery: 4.6±2.5 years).
Longitudinal voxel-based analyses were used to determine the voxelwise GMV within the unaffected brain regions. To control for the dramatic developmental effect, age-adjusted GMV within unaffected brain regions was derived voxel by voxel using a normative modeling approach with an age-matched reference cohort of 2115 healthy children(Fig. 1). [3]
For each patient, we manually outlined the mask of the cerebral hemisphere that underwent surgery on both the pre- and postoperative T1-weighted images. We then carried out VBM analyses to ensure unbiased comparisons between the two hemispheres, a symmetric T1 template in MNI space was constructed using 2115 healthy children.
To evaluate how GMV changes, we performed a voxelwise linear mixed-effects model (LMEM) analysis within the GM mask of the unaffected regions, with the onset of epilepsy, age at surgery, and etiology as covariates. Multiple comparisons were corrected using the random field theory (RFT) method (uncorrected p < .001), and clusters with a corrected p < .05/2 (2 patient groups) were considered significant.
Longitudinal voxel-based analyses were used to determine the voxelwise GMV within the unaffected brain regions. To control for the dramatic developmental effect, age-adjusted GMV within unaffected brain regions was derived voxel by voxel using a normative modeling approach with an age-matched reference cohort of 2115 healthy children(Fig. 1). [3]
For each patient, we manually outlined the mask of the cerebral hemisphere that underwent surgery on both the pre- and postoperative T1-weighted images. We then carried out VBM analyses to ensure unbiased comparisons between the two hemispheres, a symmetric T1 template in MNI space was constructed using 2115 healthy children.
To evaluate how GMV changes, we performed a voxelwise linear mixed-effects model (LMEM) analysis within the GM mask of the unaffected regions, with the onset of epilepsy, age at surgery, and etiology as covariates. Multiple comparisons were corrected using the random field theory (RFT) method (uncorrected p < .001), and clusters with a corrected p < .05/2 (2 patient groups) were considered significant.
·Figure 1. Schematic of MRI image processing.
Results:
The age-adjusted GMV values represent the patient's deviation from the age- and sex-matched norm: negative and positive values indicate a trend of GM shrinkage and expansion, respectively. In most patients in the two groups, both preoperative and postoperative age-adjusted GMV values were negative across the vast majority of the unaffected regions, suggesting an overall shrinking pattern and underdevelopment of GM compared with the healthy children.
After correcting for multiple comparisons, we observed 4 clusters showing significant age-adjusted GMV changes for the left hemispherotomy group (Fig. 2A) and 3 significant clusters for the right hemispherotomy group (Fig. 2B).
In both groups, the largest cluster covered almost the entire contralateral cerebrum and exhibited significantly increased GMV (left hemispherotomy: t = 6.92, p < 0.001; right hemispherotomy: t = 7.44, p < 0.001).
The second largest cluster was located around the entire ipsilateral cerebellum and exhibited significantly increased age-adjusted GMV (left hemispherotomy: t = 6.95, p < 0.001; right hemispherotomy: t = 8.53, p < 0.001).
In contrast to the two clusters, the cluster in the contralateral cerebellum consistently showed significantly decreased age-adjusted GMV in both groups (left hemispherotomy: t = -8.32, p < 0.001; right hemispherotomy: t = -7.73, p < 0.001).
Finally, there was one significantly decreased age-adjusted GMV cluster around the contralateral cingulate gyrus in the left hemispherotomy group (t =-5.23, p < 0.001) but not in the right hemispherotomy group.
After correcting for multiple comparisons, we observed 4 clusters showing significant age-adjusted GMV changes for the left hemispherotomy group (Fig. 2A) and 3 significant clusters for the right hemispherotomy group (Fig. 2B).
In both groups, the largest cluster covered almost the entire contralateral cerebrum and exhibited significantly increased GMV (left hemispherotomy: t = 6.92, p < 0.001; right hemispherotomy: t = 7.44, p < 0.001).
The second largest cluster was located around the entire ipsilateral cerebellum and exhibited significantly increased age-adjusted GMV (left hemispherotomy: t = 6.95, p < 0.001; right hemispherotomy: t = 8.53, p < 0.001).
In contrast to the two clusters, the cluster in the contralateral cerebellum consistently showed significantly decreased age-adjusted GMV in both groups (left hemispherotomy: t = -8.32, p < 0.001; right hemispherotomy: t = -7.73, p < 0.001).
Finally, there was one significantly decreased age-adjusted GMV cluster around the contralateral cingulate gyrus in the left hemispherotomy group (t =-5.23, p < 0.001) but not in the right hemispherotomy group.
·Figure 2. Longitudinal change in age-adjusted GMV before and after surgery for the two patient groups.
Conclusions:
Both left and right hemispherotomy patients showed widespread GMV increases in the contralateral cerebrum and ipsilateral cerebellum but GMV decreases in the contralateral cerebellum. With normative modeling, the neurodevelopment-induced and hemispherotomy-induced increases in GMV can be well differentiated. This widespread greater GM development could be considered a part of neuroplastic changes induced by the hemispherotomy.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism) 1
Modeling and Analysis Methods:
Multivariate Approaches
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Neuroanatomy Other
Novel Imaging Acquisition Methods:
Anatomical MRI 2
Keywords:
Epilepsy
Plasticity
STRUCTURAL MRI
1|2Indicates the priority used for review
Provide references using author date format
[2] Moosa, A. N. V. et al. (2013), 'Longitudinal seizure outcome and prognostic predictors after hemispherectomy in 170 children', Neurology, vol. 80, no. 3, pp. 253–260.
[3] Rutherford, S. et al. (2022), 'The normative modeling framework for computational psychiatry', Nature Protocols, vol. 17, no. 7, pp. 1711–1734.