__________________________________________________________________________________________________

Michelle Hartley-McAndrew, MD;* Arie Weinstock, MD

This review focuses on developments and ongoing advances in neuroimaging research that help to characterize the structural and neuronal phenotype  of autism. Autism is a disorder in which early diagnosis is paramount in order to intervene  and initiate therapies early. While there are limitations to current diagnostic modalities,  a variety of neuroimaging techniques are emerging as possible objective means in providing  an early bio-marker for the diagnosis of autism in those at risk. 

Key Words: autism, neuroimaging,  diagnostic phenotype autism, neuroimaging, diagnostic phenotype 

INTRODUCTION 

Autism spectrum disorders, as currently defined  by the DSM-IV (Diagnostics and Statistics Manual of Mental Disorders 4th edition)¹ criteria are a group of neuro-developmental disorders that impact a child’s social and communication  skills with repetitive mannerisms, interests and often restricted adherence to routines.  Advances in this heterogenous disorder has been challenging due to the inability to link  the onset of the disorder to any one gene or environmental trigger. The CDC (Center for Disease Control) now estimates a 1 in 88 prevalence of autism spectrum disorders. It is unclear now  how this number will be affected with the upcoming change in the diagnostic criteria in 2013  as outlined by the DSM V. 

The single known best tool in altering the course  of this disorder is early intervention and initiation of therapies. It follows that early  diagnosis is imperative to improved long-term outcome. The total annual societal per capita  cost of caring for and treating a person with autism in the US has been estimated to be about  $3.2 million and approximately $35 billion for an entire birth cohort of people with autism.  The total lifetime cost could range from $13 billion to $76 billion. Literature shows that  people with autism spend twice as much as the typical American over their lifetimes and spend  approximately 60% of those incremental direct medical costs after 21 years of age.² These costs will continue to rise with the increase in prevalence. 

In 1987 Lovaas showed that the earlier the intervention,  the greater the impact on cognitive and social functioning which has been confirmed in later  studies and randomized control trials.^3,4,5 The positive effect of early intensive behavioral intervention as well as the difference  in early presentation of autistic feature onset leads to the understanding of autism spectrum  disorders as a group of disorders marked by plasticity and heterogeneity.^4,6 Atypical markers for autism such as impairments in visual tracking, eye contact,  social interest, orienting to name, etc. are present at as early as 12 months of age.^4,7,8,9 However, clinical testing of early onset of autistic features are currently experimental  in children less than 18 months of age.^8,10,11 In addition to decreased sensitivity and specificity of clinical tests in early infancy,  there is also instability of the diagnosis in children younger than 30 months of age.⁸ In the search for an objective diagnostic phenotype, potential advances in neuroimaging  may in the future be utilized to aid in the earlier diagnosis of autism when concerning developmental  indicators arise. 

STRUCTURAL MRI 

Structural MRI studies have contributed to the  current understanding of the neuroanatomical phenotype of autism. Reports have indicated an  onset of increased white and gray matter cortical volume in autistic subjects 2-4 years of  age that may be tied to the onset of the core features of autism that occur around that age.^12,13,14 Volume differences have been seen with age related enlargement of the caudate nucleus¹⁵ the amygdala^12,16 reduced size of the corpus callosum^17,18,19 with variable and inconsistent findings in volume size of the cerebellum, brainstem and  hippocampus.^{12,14,18,20,21}

FUNCTIONAL MRI 

While the early increase in grey and white matter  volume provides supporting evidence for abnormalities in connectivity pathways, it is functional  neuroimaging that has added insight to the differences in connections between cortical regions  that exist in autism.^22,23 Functional MRI (fMRI) provides information regarding the functional organization of cognitive  changes and has shown changes in this organization in children with autism. FMRI techniques  have been used to examine the neural bases of social cognition and facial processing.  For example, differences in activation during theory of mind tasks including decreased activation  of frontal cortical components, decreased amygdala and increased superior temporal gyrus activation  are seen in individuals with autism. These differences suggest that dysfunction of the amygdala  results in failure of emotional and mental state processing in autism and a reduction in frontal  activation results in deficits of executive function.²⁴ Facial processing in individuals with autism is also effected as elucidated by fMRI studies.  During facial recognition in typical subjects, activation has been seen in the right fusiform  gyrus.^25,26 Schultz reproduced this finding in controls showing that levels of activity in the right fusiform gyrus were increased during facial recognition while there was increased activity  observed in the inferior temporal gyri during object processing. In the study group with autism  and Asperger’s syndrome, there was significantly less right hemisphere fusiform gyrus activation  and instead, more right inferior temporal gyrus activation in response to facial recognition.²⁷

Executive functioning tasks have shown less robust  activation in brain regions in autistic individuals compared to controls. In one study by  Ring et al. that performed fMRI during an embedded figure task showed activation of areas  of the mesial temporal gyrus and inferior temporal gyrus, the supramarginal gyrus and precuneus,  the inferior frontal gyrus and the middle occipital gryus in both the study and control groups.  The control group however demonstrated greater activation in the right dorsolateral prefrontal  cortex and bilateral parietal cortex than the autistic study group. The authors therefore  postulated that subjects with autism rely on visuospatial feature analysis rather than working  memory, while controls rely more on working memory when performing the embedded figure task.²⁸ In a planning and execution motor task by Muller analysis found activation in the contralateral  perirolandic cortex, basal ganglia and thalamus, bilateral supplementary motor area and ipsilateral  cerebellum for both control and autism groups. Activation was less pronounced however in the  autistic study group. These brain imaging studies have shown that people with autism show  less activation in regions related to higher level cognition and more activation in the regions  associated with lower level cognition when compared to controls.²⁹

Table 1. Summary of Task-based fMRI techniques in autism.

Functional MRI in a study by Just et al.  during language and sentence comprehension tasks further supported the growing view that  autism was a disorder of functional connectivity with decreased synchronization in specific  cortical areas compared to controls. The authors conclude that there is a disruption in the  coordination or integration of cortical circuitry making particular tasks such as social interaction,  which requires high demands on information integration, deficient in individuals with autism.²³Again, patterns found in fMRI studies have supported that subjects with autism prefer visually  based processing styles with more activation in the posterior brain regions while there is  less activation in the anterior cortical regions during memory and language tasks.³⁰ There appears to be greater processing of information in the right hemisphere compared to  the left indicating use of nonverbal strategies in information processing in the autistic  population.³⁰ Task Based functional MRI techniques (Table 1), while valuable, require active participation  by the subject and are less suited for studies in infants and toddlers.¹⁴

Resting fMRI studies have demonstrated differences  in the default neuronal circuitry when the brain is at rest. The fMRI studies in autism previously  described were performed in the context of a cognitive social or non-social task.  However, when the brain is not engaged in any task, the brain is still active. The regions  involved in the default network include the posterior cingulate cortex, retrosplenial,  lateral parietal/angular gyrus, medial prefrontal cortex, superior frontal gyrus,  regions of the temporal lobe and the parahippocampal gyrus.³¹ It is unclear what the exact function of this activation is, however it is speculated that  it may be involved in the homeostasis of excitatory and inhibitory neuronal function. 

Results of resting fMRI studies in adults with  autism have been variable. Underconnectivity was found between anterior and posterior regions  along with medial structures including the anterior cingulate and posterior cingulate cortices  in a study by Cherkassky et al. involving 57 adults with high functioning autism compared  with age and IQ matched controls.³² In a more recent study in adults with ASD (autism spectrum disorder), it was found that  compared to controls, the ASD group showed weaker connectivity between the posterior cingulate  cortex and superior frontal gyrus and a stronger connectivity between the posterior cingulate  and the right temporal lobe as well as right parahippocampal gyrus. Interestingly,  in this study the poorer the social functioning, the weaker the connectivity between the  posterior cingulate cortex and the right superior frontal gyrus. Repetitive and restricted  behaviors correlated with stronger connectivity between the posterior cingulate cortex and  the right parahippocampal gyrus.³³ In examining adolescents with ASD, Weng et al. found decreased connectivity between the  posterior cingulate cortex and the superior frontal gyri, the temporal lobes and the parahippocampal  gyri. This weaker connectivity also correlated to social impairment. The repetitive behaviors,  in contrast to findings in adults with autism correlated with weaker functional connectivity  between the posterior cingulate cortex, the prefrontal gyri, the temporal lobes and the parahippocampal gyri. They attributed this difference to a dramatic change in the relationship between the  default network and the repetitive behaviors.³¹

In further investigating the striatum during  the resting state in school-aged children with Autism, Asperger’s Disorder and Pervasive Developmental  Disorder Not Otherwise Specified, it was found that there was excessive as well as aberrant  functional connectivity in striatal-cortical circuitry compared to typically developing controls.³⁴ In toddlers with autism during natural sleep, resting fMRI has shown significantly weaker  inter-hemispheric correlations in the inferior frontal gyrus and superior temporal gyrus regions  compared to typically developing and language delayed controls; again, specifically areas  which are known to be associated with language production and comprehension. It appears that  synchronization and functional connectivity are disrupted early in autism development as supported  by fMRI studies.³⁵

PET AND SPECT 

Functional neuroimaging such as positron emission  tomography (PET), and single photon emission computed tomography (SPECT) give insight to cortical  activation during rest or during a performance task. Studies involving both SPECT and PET  scans have detected bitemporal hypoperfusion in the superior temporal gyrus in autistic subjects.²² During functional activation studies, in both PET and SPECT scans, there is more activation  of the right hemisphere than the left during verbal auditory stimulation again supporting  the idea of the use of aberrant neural circuitry in the processing of verbal information in  autistic individuals.^22,29

MEG

Recently, MEG (magnetoencephalography), which  has typically been used in identifying epileptogenic foci in epilepsy, has been introduced  in studying possible neural markers for autism. Similar to an EEG (electroencephalogram) MEG  measures neural electrical activity; however, it uses MRI technique to measure the magnetic  fields produced by the electrical activity. Using MEG to measure auditory evoked responses,  studies have shown delayed latency of M100 in response to a tone stimulus in subjects with autism.^36,37,38 This finding appears to be specific to autism and not just a feature of language delay and  impairment.³⁹

DTI

Another neuroimaging technique, diffusion tensor  imaging (DTI), has also given insight into the connectivity of neural pathways in autism.  Specifically, white matter provides the physical foundation for corticocortical and cortico-subcortical  connectivity. DTI is a visualization of white matter tract integrity in vivo. Areas of white  matter disruption have been shown in autism with significantly reduced fractional anisotropy  (FA), a measure of structural integrity. In a study by Barnea-Goraly et al. reduced FA values  were seen adjacent to regions of the brain which are activated during social and emotional  processes such as facial recognition including the fusiform gyrus, the superior temporal sulcus,  anterior cingulate, amygdala and ventromedial prefrontal cortex. Also, disrupted white matter  integrity was seen in regions required for theory of mind tasks such as the ventromedial prefrontal  cortex, anterior cingulate, temporoparietal junction and its connection to the extrastriate  region, the superior temporal sulcus and amygdala. In addition, decreased FA was found in  the corpus callosum, which is postulated to impair executive functioning, and sensorimotor  processing.⁴⁰ Further, these structural differences are seen over time when compared with age matched  controls. Reduced FA has been seen in areas crucial in interhemispheric connectivity including  the posterior midbody/isthmus of corpus callosum, left and right anterior corona radiata near  the genu of the corpus callosum, right anterior corona radiata/forceps minor and right retrolenticular  portion of the internal capsule over an age range of 10-35 years where FA, although increasing  over time does not reach that of controls.⁴¹ This confirms there is an abnormal neuronal developmental trajectory in autism. 

While abnormal white matter integrity in the  autism population has been demonstrated in the literature, the application of tract-based  spatial statistics has given further understanding of the aberrant long range neural connections.  In a study by Shukla, reduced fractional anisotropy and increased mean diffusivity were seen  in the corpus callosum, anterior and posterior limbs of the internal capsule, inferior longitudinal  fasciulus, inferior fronto-occipital fasciculus, superior longitudinal fasciculus,  cingulum, anterior thalamic radiation and corticospinal tract in the autism group compared  with controls.⁴² In the further aim for determining the presence of a specific autistic neural phenotype,  Jou et al. suggest there are several long-range fiber tracts that are affected in autism;  specifically the inferior fronto-occipital fasciculus and superior longitudinal fasciculus  are most severely affected.⁴³ The inferior fronto-occipital fasciculus has been known to connect directly to the fusiform  gyrus as well as give connections between the frontal, temporal, parietal and occipital lobes.  Disruption of this connection results in impairments in the ability to properly identify  and recognize facial emotions.⁴⁴ The superior longitudinal fasciculus connects areas of the temporo-parietal cortex to areas  of the frontal lobe and functions to regulate motor behavior, aid in working memory and auditory information processing. It is specifically these long-range fiber tracts that are used in  social information processing and language and communication skills that are impaired in autism.⁴³

The above has illustrated the aberrant development  of white matter and connectivity in the autistic population through neuroimaging and describes  the abnormal developmental trajectory that corresponds with autistic symptoms. However are  there any indicators prior to onset of symptoms when it is so crucial to diagnose early?  More recently a prospective study that performed DTI sequences on high-risk siblings of individuals  with autism at 6, 12 and 24 months showed statistically significant higher fractional anisotropy  at 6 months of age in those who developed autism followed by a more attenuated developmental trajectory resulting in a lower fractional anisotropy by 24 months compared to those high  risk individuals that did not go on to develop autism.⁴⁵ Fiber tractography indicated while lower at 6 months, the rate of change in FA from 6 to  24 months was significantly greater in the ASD-negative group in the corpus callosum,  bilateral fornix, association tracts including the inferior longitudinal fasciulus and uncinate  as well as projection tracts such as the left anterior thalamic radiation and internal capsule compared to the ASD positive group. This blunted developmental trajectory in white matter  integrity is suggested to begin prior to the onset of clinical features in the autistic population.  While it is the abnormal trajectory demonstrated over time, which is more difficult to illustrate  in a single study, it may be a potential area of monitoring in high-risk siblings.  This study implicates that the aberrant development of neural connectivity as seen by DTI  can precede the manifestations of autistic symptoms in the first year of life. 

Table 2. Neuroimaging Techniques used in Autism.

CONCLUSION 

The promising developments and advances in newer  neuroimaging techniques (Table 2) reinforce the conception of autism as a disorder of neural  connectivity that is pervasive among multiple networks and not localized to any one specific  brain region. The heterogeneity of this disorder makes diagnosis a challenge. Attempts to  give objective diagnostic determinations are limited by the age at which these tests are valid.  This is particularly problematic when early diagnosis is crucial to improved developmental outcome. Current neuroimaging studies shed light on evidence of potential critical time periods  of abnormal white matter pathway development, which may give insight to the pathogenesis of  autism prior to the onset of symptoms. Further goals to depict the onset of neural abnormalities  may also help in illuminating the mechanisms underlying them as well as give optimal opportunity  for intervention. 

Michelle Hartley-McAndrew, MD;* Arie Weinstock,  MD

State University of New York at Buffalo School  of Medicine and Biomedical Sciences, Buffalo, NY 

*Corresponding Author: Clinical  Assistant Professor, Dept of Child Neurology, State University of New York at Buffalo School  of Medicine and Biomedical Sciences, Medical Director, Children’s Guild Autism Spectrum Disorder  Center, Women and Children’s Hospital of Buffalo, 219 Bryant Street Buffalo, NY 14222.  Assistant Professor, D’Youville College. Tel: 716-878-7600. (Email: mhartley-mcandrew@kaleidahealth.org) 

CONFLICT OF INTEREST 
None. 

REFERENCES

1.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, edn. 4, text revision. Washington DC. 2000. American Psychiatric Press.

2.Ganz ML. The Lifetime Distribution of the Incremental Societal Costs of Autism. Arch Pediatr Adolesc Med. 2007;161(4):343-349.

3.Beglinger L, Smith T. Concurrent Validity of Social Subtype and IQ after Early Intensive Behavioral Intervention in Children with Autism: A Preliminary Investigation. J Autism Dev Disord. 2005;35(3):295-303.

4.Landa RJ, Holman KC, Garrett-Mayer E. Social and Communication Development in Toddlers with Earlier and Later Diagnosis of Autism Spectrum Disorders. Arch Gen Psychiatry. 2007;64(7):853-864.

5.Dawson G, Rogers S, Munsons J. et al. Randomized Controlled Trial of an Intervention for Toddlers with Autism: The Early Start Denver Model. Pediatrics. 2010;125(1):e17-e23.

6.Warren Z, McPheeters ML, Sathe N, Foss-Feig JH, Glasser A, Veinstra-VanderWeele J. A Systematic Review of Early Intensive Intervention for Autism Spectrum Disorders, Pediatrics. 2011;127(5):e1303-1311.

7.Zwaigenbaum L, Bryson S, Rogers T, Roberts W, Brian J, Szatmari P. Behavioral Manifestations of Autism in the First Year of Life. Int J Devl Neurosci. 2005;23:143-152.

8.Landa RJ. Diagnosis of Autism Spectrum Disorders in the First Three Years of Life. Nat Clin Pract Neurol. 2008;4(3):138-147.

9.Filipek PA, Accardo PJ, Baranek GT et al. The Screening and Diagnosis of Autism Spectrum Disorders. J Autism Dev Disord. 1999;29(6):439-484.

10.Lord C, Storoschuk S, Rutter M, Pickles A. Using the ADI-R to diagnose autism in preschool children. Infant Mental Health Journal. 2006;14(3):171-254.

11.Lord C, Rutter M, DiLavore P, Risi S. Autism Diagnostic Observation Schedule. Los Angeles, CA-Western Psychological Services. 1995.

12.Sparks BF, Friedman SD, Shaw DW, Aylward EH, Echelard D, Artue AA et al. Brain Structural Abnormalities in Young Children with Autism Spectrum Disorder. Neurology. 2002;59(2):184-192.

13.Courchesne E, Carper R, Akshoomoff N. Evidence of Brain Overgrowth in the First Year of Life in Autism. JAMA. 2003;290(3):337-344.

14.Mosconi M, Zwaigenbaum L, Pivin J. Structural MRI in Autism: Findings and Future Directions. Clin Neurosci Res. 2006;6(3):135-144.

15.Sears LL, Vest C, Mohomed S, Bailey J, Ranson BJ, Piven J. An MRI study of the basal ganglia in autism. Prog Neuropsychopharmacol Biol Psychiatry. 1999;23(4):613-624.

16.Nordahl CW, Scholz R, Yang X, Buoncore MH, et al. Increased Rate of Amygdala Growth in Children Aged 2 to 4 Years with Autism Spectrum Disorders. Arch Gen Psychiatry. 2012; 69(1): 53-61.

17.Egaas B, Courchesne E, Saitoh O. Reduced Size of the Corpus Callosum in Autism. Arch Neurol. 1995;52:794-801.

18.Piven J, Bailey J, Ransom B, Arndt S. An MRI Study of the Corpus Callosum in Autism. Am J Psychiatry 1997;154:1051-1056.

19.Frazier TW, Kashavan MS, Minshew NJ, Hardan AY. A Two Year Longitudinal MRI study of the Corpus Callosum in Autism. J Autism. Dev Disord. 2012; Feb 18. [Epub ahead of print]

20.Cody H, Pelphrey K, Piven J. Structural and Functional Magnetic Resonance Imaging of Autism. Int J Dev Neurosci. 2002;20(3-5):421-438.

21.Hazlett HC, Poe M, Gerig G, et al. Magnetic Resonance Imaging and Head Circumference Study of Brain Size in Autism: Birth through age 2 years. Arch Gen Psychiatry. 2005; 62(12): 1366-76.

22.Boddaert N, Zilbovicius M. Functional Neuroimaging and Childhood Autism. Pediatr Radiol. 2002; 32(1): 1-7.

23.Just MA, Cherkassky VL, Keller TA, Minshew NJ. Cortical Activitation and Synchronization During Sentence Comprehension in High Functioning Autism: Evidence of Underconnectivity. Brain. 2004;127(Pt 8):1811-1821.

24.Baron-Cohen S, Ring HA, Wheelwright S, et al. Social Intelligence in the normal and autistic brain: an fMRI study. Eur J Neurosci. 1999;11(6):1891-1898.

25.Kanwisher N, McDermott J, Chun MM. The Fusiform Face Area: A Module in Human Extrastriate Cortex Specialized for Face Perception. J Neurosci. 1997;17(11):4302-4311.

26.McCarthy G, Puce A, Gore JC, Allison T. Face Specific Processing in the Human Fusiform Gyrus. J Cogn Neurosci. 1997;9(5):605-610.

27.Schultz RT, Gantheir I, Klin A, Fulbright RK, et al. Abnormal Ventral Temporal Cortical Activity during face discrimination among autism and Asperger’s Syndrome. Arch Gen Psychiatry. 2000;57(4):331-340.

28.Ring H, Baron-Cohen S, Wheelwright S, et al. Cerebral Correlates of Preserved Cognitive Skills in Autism: A Functional MRI Study of the Embedded Figures Task Performance. Brain. 1999;122(Pt 7):1305-1315.

29.Muller RA, Pierce K, Ambrose JB, Allen G, Courchesne E. Aytpical Patterns of Cerebral Motor Activation in Autism: A Functional Magnetic Resonance Study. Biol Psychiatry. 2001;49(8):665-676.

30.Koshino H, Carpenter PA, Minshew NJ, Cherkassky VL, Keller TA, Just MA. Functional Connectivity in an fMRI Working Memory Task in High-Functioning Autism. Neuroimage. 2005;24(3):810-821.

31.Weng S, Wiggins JL, Peltier SJ, et al. Alterations of Resting State Functional Connectivity in the Default Network in Adolescents with Autism Spectrum Disorders. Brain Res. 2010;1313(202):1-27.

32.Cherkassky VL, Kana RK, Keller TA, Just MA. Functional Connectivity in a Baseline Resting-state Network in Autism. Neuroreport. 2006;17(16):1687-1690.

33.Monk CS, Petlier SJ, Wiggins JL, et al. Abnormalities of Intrinsic Functional Connectivity in Autism Spectrum Disorders. Neuroimage. 2009;47(2):764-772.

34.Di Martino A, Kelley C, Grzadzinski R, et al. Aberrant Striatal Functional Connectivity in Children with Autism. Biol Psychiatry. 2011;69(9):847-856.

35.Dinstein I, Pierce K, Eyler L, et al. Disrupted Neural Synchronization in Toddlers with Autism. Neuron. 2011;70(6):1218-1225.

36.Gage NM, Siegel B, Roberts TP. Cortical Auditory System Maturational Abnormalities in Children in Autism Disorder: An MEG Investigation. Dev Brain Res. 2003t;144(2):201-209.

37.Roberts TP, Schmidt GL, Egeth M, et al. Electrophysiological Signatures: Magnetoencephalographic Studies of the Neural Correlates of Language Impairment in Autism Spectrum Disorders. Int J Psychophysiol. 2008;68(2):149-160.

38.Roberts TP, Khan SY, Rey M, et al. MEG Detection of Delayed Auditory Evoked Responses in Autism Spectrum Disorder: Towards an Imaging Biomarker for Autism. Autism Res. 2010;3(1):8-18.

39.Roberts TP, Heiken K, Kahn SY, et al. Delayed Magnetic Mismatch Negativity Field, but not Auditory M100 response in Specific Language Impairment. Neuroreport. 2012;23(8):463-468.

40.Barnea-Goraly N, Kwon H, Menon V, Eliez S, Lotspeich L, Reiss AL. White Matter Structure in Autism: Preliminary Evidence from Diffusion Tensor Imaging. Biol Psychiatry. 2004;55(3):323-326.

41.Keller TA, Kana RK, Just MA. A Developmental Study of the Structural Integrity of White Matter in Autism. Neuroreport. 2007;18(1):23-27.

42.Shukla DK, Keehn B, Muller RA. Tract-specific analyses of diffusion tensor imaging show widespread white matter compromise in autism spectrum disorder. J Child Psychol Psychiatry. 2011;52(3):286-295.

43.Jou RJ, Mateljevic N, Kaiser MD, Sugrue DR, Volkmar FR, Pelphrey KA. Structural Neural Phenotype of Autism: Preliminary Evidence from a Diffusion Tensor Imaging Study Using Tract-Based Spatial Statistics. AM J Neuroradiol. 2011;32(9):1607-1613.

44.Philippi CL, Mehta S, Grabowski T, Adolphs F, Rudrauf D. Damage to Association Fiber Tracts Impairs Recognition of the Facial Expression of Emotion. J Neurosci. 2009;29(48):15089-15099.

45.Wolff JJ, Gu H, Gerig G. et al. Differences in White Matter Fiber Tract Development Present from 6 to 24 months in Infants with Autism. Am J Psychiatry. 2012; 169(6):589-600.