Elsevier

Neuroscience Letters

Volume 728, 29 May 2020, 134949
Neuroscience Letters

Research article
Longitudinal Changes of Cerebellar Thickness in Autism Spectrum Disorder

https://doi.org/10.1016/j.neulet.2020.134949Get rights and content

Highlights

  • The cerebellar lobule thickness decreased in the right Crus II and the Crus II asymmetry was reduced in individuals with ASD.

  • The reduction in lobular thickness and the asymmetry in Crus II were associated with the severity of stereotyped behavior symptoms.

Abstract

Many studies have reported abnormal cerebellar volume in patients with autism spectrum disorder (ASD) and that this abnormal volume can change with age. In the present study, we used CERES, an automated and reliable quantitative analysis tool, and adopted a longitudinal design to examine developmental changes in the cerebellar lobular thickness in ASD and quantified the relationship between cerebellar thickness development and clinical symptoms. Nineteen individuals with ASD (16 males; age, 12.53 ± 2.34 years at baseline, interval: 2.33 years) and 14 typically developing controls (TD; 12 males; age, 13.50 ± 1.77 years at baseline, interval: 2.31 years) underwent T1-weighted magnetic resonance imaging at two time points. To explore the relationship between cerebellar lobular thickness and the symptoms of ASD, the correlation of Autism Diagnostic Observation Schedule (ADOS) score with lobular thickness data was calculated. The cerebellar lobule thickness decreased in the right Crus II and the Crus II asymmetry was reduced in individuals with ASD. The reduction in lobular thickness and the asymmetry in Crus II were associated with the severity of stereotyped behavior symptoms. Structural differences and behavioral correlations were concentrated in the right cerebellar Crus II. These results emphasize the importance of the potential functional effect of structural differences in cerebellar subregions on ASD and suggest that the changes of thickness in the right cerebellar Crus II are related to the core profile of ASD.

Introduction

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by impaired social and communication abilities, stereotyped behavior, and reduced interest [1]. Previous neuroimaging studies have reported a wide range of regional brain abnormalities in ASD, such as in the frontal, parietal, and limbic regions, the basal ganglia, and the cerebellum [2]. Over the years, many reports have shown that cerebellar abnormalities were associated with the onset of ASD [3]. There have been reports of cerebellar structural and functional differences in ASD, which suggests that cerebellar dysfunction may be important in the etiology of the disease [4]. In support of this, the first case of abnormal cerebellar anatomy with autism was published in 1980 [5] and described a decrease in the number of Purkinje cells in the cerebellar vermis of an autism patient. Since then, almost all postmortem analyses of ASD individuals have shown that the size and number of Purkinje cells was decreased regardless of age, sex and cognitive ability [6]. In 1987, using in vivo magnetic resonance imaging (MRI), researchers first reported the observation of an abnormality of the cerebellum in an autism patient. Since then, various studies have found significant differences in the volume of cerebellar subregions between ASD individuals and typically developing (TD) controls. Although meta-analysis-based neuroimaging studies have found that several cerebellar subregions are affected in ASD [7,8], there was no study to examine the structural differences in the brains of individuals with ASD using lobular thickness while also assessing the relationship between these subregions of the cerebellum and core ASD symptoms.

The cerebellum shows a highly regular arrangement of neurons and connections and is supposed to support a large number of parallel computing capabilities [9]. Anatomically, the cerebellum consists of twelve lobules (lobules I-X, Crus I and Crus II). There is also growing evidence supporting the idea that cerebellar subregions have functionally distinct roles in movement and in cognitive and affective functions [10]. For example, lobule VIII is found to include somatomotor networks, connected with the sensorimotor region in the cerebral cortex and involved in sensorimotor tasks [10]. Lobules VI and VII are subdivided into Crus I, Crus II and VIIB, and they receive inputs from the prefrontal cortex and parietal lobe associative regions and are involved in cognitive tasks [11]. A functional connectivity study suggested that the cerebellum is connected to association networks involved in cognitive and affective processes rather than somatomotor networks [12]. Increasingly, the cerebellum is thought to be related to nonmotor functional networks [11], and it is functionally connected to the resting-state network of the whole brain, involving higher-order cognitive networks such as the frontoparietal network (FPN) [12]. Several meta-analysis reports mapping the cerebellar activation clusters in different cognitive domains have described their involvement in a large number of tasks including attention, working memory, executive function, and other behaviors [10,13]. Clinical studies have suggested that lesions in the posterior regions of the cerebellum could result in difficulties in executive function, language, memory and emotion, while damage to the anterior cerebellum can lead to motor impairments and minimal cognitive effects [14]. Based on these findings, the role of the human cerebellum has been extended to include higher-order cognitive and affective processes [11,15]. The unique connectivity patterns of different cerebellar subregions lead to a different functional topography, from which different regions manage different types of information [11]. Taking into account the location of cerebellar subregions and the functional differences in ASD, this topography is important and may help to explain the cerebellar findings in ASD.

Various neuroimaging studies have found abnormalities in several cerebellar subregions in ASD compared with control subjects. Previous structural studies have observed hypoplasia in the posterior vermis with ASD and found that gray matter (GM) decreased in the right Crus I, lobule VIII, and lobule IX [7,8]. Fewer studies have observed decreased GM in the left Crus I, and some researchers have observed an overall increase in cerebellar GM [7,8]. Functional MRI studies show that cerebellar activation in ASD is reduced in social, language, and motor tasks. For example, compared with the TD group, Crus I was underactivated during facial and vocal stimuli processing [16] and executive functioning tasks [17]; lobule VII was abnormally activated during semantic processing [18]; and lobule IV/V was not engaged in motor tasks [19]. Clinical studies about ASD have found that deformities of the cerebellar vermis are related to social and affective disorders and these abnormalities in the cerebellar hemisphere are associated with defects in language expression and in motor and executive functions. These symptoms are linked to ASD [20]. A study showed a 40-fold increase in ASD incidence in premature infants suffering cerebellar damage compared with controls [21]. In addition, patients with tuberous sclerosis (TSC) have a high incidence of ASD symptoms [22], and TSC is specifically associated with tubers within the cerebellum [23].

Cerebellar abnormalities play an important role in ASD [24]. However, few studies have explored the role of subregions in ASD, and few studies have further examined the relationship between ASD symptoms and cerebellar subregions [25,26]. To date, most studies have examined differences at the hemispheric level, while regional differences are often not located in specific cerebellar lobules. In view of the emerging cerebellar functional topographic maps and various cerebellar regions involved in the pathophysiology of autism, it is important to study more discrete subregions in the cerebellum and consider their functional correlation. The present study investigates the thickness of cerebellar subregions in ASD and links the structural findings to the core symptoms of the disorder. In addition, because it is not easy to measure the thickness of the cerebellar cortex, no quantitative changes in thickness have been studied. In this study, we used an automated and reliable quantitative analysis tool to accurately quantify cerebellar thickness. Finally, cross-sectional studies couldn’t make a conclusive distinction between causes and consequences and there is a lack of longitudinal neuroimaging studies. Therefore, we used longitudinal neuroimaging design. The goal of this study was to longitudinally characterize developmental changes in the thickness of cerebellar subregions in ASD and to assess correlations between volume changes and clinical measures. To our knowledge, this is the first study to examine the cerebellar subregions in ASD for lobular thickness analysis. Moreover, this is the first study to correlate cerebellar lobular thickness with autistic symptoms.

Section snippets

Participants

For all analyses, the publicly available longitudinal MRI datasets from the ABIDE program (http://fcon_1000.projects.nitrc.org/indi/abide/) were used. The participants were 19 individuals with ASD (16 males; age 12.53 ± 2.34 years at baseline) and 14 TD controls (12 males; age 13.50 ± 1.77 years at baseline). Two scans were available for each participant, with a mean interval between 2.33 ± 0.70 years for ASD and 2.31 ± 0.65 years for TDC. Participants were recruited and scanned at one of two

Demographics

The demographic characteristics of the sample are summarized in Table 1. There were no differences in interscan interval or baseline age, performance IQ, verbal IQ, or full-scale IQ across groups (all p > 0.05). The χ2 test was used to analysis the gender differences on ASD and TDC, results showed there was no gender difference (χ2 = 0.014, p =  0.905).

Group differences in the development of cerebellar structures

Descriptive statistics for cerebellar thickness analysis are provided in Table 2 and Fig. 2. There was no significant effect of time point on

Discussion

In the present study, we investigated the structural differences of cerebellar subregions in ASD and the relationship between the developmental change in cerebellar lobular thickness and core autistic symptoms. To our knowledge, this is the first study to examine the lobular cortical thickness of the cerebellum in individuals with ASD. Lobular thickness was analyzed using CERES. The right lobule VIIB and VIIIA thicknesses were observed in the main group, and they showed a larger thickness in

Limitations

The strengths of the current study include the longitudinal design and the uniform segmentation approach. However, the sample size was small, and the age distribution was not centralized, therefore it is unknown if our findings will generalize to other groups. Although scanner site was included as a covariate in our analysis, conclusions may be strengthened if data were collected from the same site.

Conclusion

In this longitudinal study, we observed that decreased right Crus II and reduced right asymmetry are important biomarkers in ASD, and their developmental changes were associated with the ADOS stereotyped behavior score, offering an explanation for the behavior deficits in ASD. Future research will focus on exploring the specific contribution of this area to ASD symptomology.

Credit Author Statement

Yanpei Wang analyzed the data and wrote the draft of the paper. LeiHao and Qinfang Xu amend and proofread the draft of the paper. Qinfang Xu, Chenyi Zuo, Liying Zhao and LeiHao participated in the discussion and offered some good ideas. All authors reviewed the manuscript.

Declaration of Competing Interest

No potential conflict of interest was reported by the authors.

Acknowledgements

This research was supported by Humanities & Social Science Program of Ministry of Education in China (No. 18YJA90018) and The National Natural Science Foundation of China (No. 31662083).

References (39)

  • Y. Wang et al.

    Gender differences in anomalous subcortical morphology for children with ADHD

    Neurosci Lett

    (2018)
  • M. Assaf et al.

    Abnormal functional connectivity of default mode sub-networks in autism spectrum disorder patients

    Neuroimage

    (2010)
  • E.B. Becker et al.
    (2013)
  • A.M. D’Mello et al.

    Cerebellar gray matter and lobular volumes correlate with core autism symptoms

    NeuroImage. Clinical

    (2015)
  • J.G. Levitt et al.

    Cerebellar vermis lobules VIII-X in autism

    Progress in neuro-psychopharmacology & biological psychiatry

    (1999)
  • C.H. Salmond et al.

    Heterogeneity in the patterns of neural abnormality in autistic spectrum disorders: evidence from ERP and MRI

    Cortex; a journal devoted to the study of the nervous system and behavior

    (2007)
  • S.H. Chen et al.

    Temporal dynamics of cerebro-cerebellar network recruitment during a cognitive task

    Neuropsychologia

    (2005)
  • A.P. Association

    Diagnostic and statistical manual of mental disorders

    BMC Med

    (2013)
  • R.S. Williams et al.

    Autism and mental retardation: neuropathologic studies performed in four retarded persons with autistic behavior

    Arch Neurol

    (1980)
  • Cited by (11)

    View all citing articles on Scopus
    View full text