Neurobiological Abnormalities in the First Few Years of Life in Individuals Later Diagnosed with Autism Spectrum Disorder: A Review of Recent Data

Table 1

Studies investigating neurological differences in individuals with ASD: study characteristics and findings.

Author	Age of sample	Sample size	Aim of the study	Findings
Schumann et al. [56]	1–5 years (mean age 3 years). Each child returned at approx. 5 years for final clinical evaluation.	41 autism, 9 PDD-NOS, and 39 TD controls.	Amygdala volume	Toddlers with ASD diagnosis (32 boys, 9 girls) had a larger right and left amygdala compared with typically developing toddlers (28 boys, 11 girls).
Nordahl et al. [57]	Time 1—mean age 37 months. Time 2—one years later.	At time 1—132 boys (85 with ASD and 47 controls with TD). At time 2—70 boys (45 with ASD and 25 TD controls).	Amygdala volumes and total cerebral volumes (TCV).	Despite no difference in TCV growth (although the TCV was significantly enlarged at both time points in the ASD group), at both time points, growth rate and amygdala volume were greater in children with ASD, with enlargement found to be greater at time two.
Mosconi et al. [58]	18–35 months (2 years). Followed-up at 42–59 months (4 years).	50 ASD and 33 controls (11 DD and 22 TD).	Investigated associations between specific ASD behaviours (joint attention) and amygdala volume.	Bilateral enlargement of amygdala volume was found in children with ASD. There was a disproportionate right amygdala volume enlargement compared to total tissue volume. Amygdala enlargement was associated with increased JA at age four years.
Hazlett et al. [59]	18–42 months	52 fragile x syndrome (FXS); 63 autism, 19 DD, and 31 TD.	Caudate nucleus volume and amygdala volume.	Children with FXS and autism disorder had substantially enlarged caudate volume and smaller amygdala volume. Children with ASD without FXS (i.e., idiopathic autism) had only modest enlargement in their caudate nucleus volumes while enlargement of their amygdala volumes were more pronounced.
Webb et al. [60]	3-4 years	45 children with ASD, 14 children with DD, and 26 children with TD.	Cerebellar vermal structures and their association with severity of symptoms and cognitive functioning in children with ASD.	Reduced total vermis volumes (vermis lobe VI-VII area) in the ASD children. No correlation was found between cerebellar measurements and severity of ASD symptoms nor verbal, nonverbal, or full scale IQ.
Courchesne et al. [61]	2–16 years	7 ASD and 6 control male children.	Postmortem prefrontal tissue	Children with ASD had 67% more neurons in the PFC compared to controls, including 79% more in dorsalateral-PFC and 29% more in medial-PFC.
Holmboe et al. [62]	9 to 10 month old	31 siblings of children who have been diagnosed with ASD and 33 low-risk control infants.	Frontal cortex functioning using a task exploring attention and inhibition.	A subset of sibs-ASD infants had difficulty disengaging attention from a centrally presented stimulus in order to orient to a peripheral stimulus indicating atypical frontal cortex functioning in the infant broader autism phenotype.
Santos et al. [63]	4–14 years	Postmortem brains of 4 young patients with ASD and 3 aged-matched controls.	von Economo neurons in the frontoinsular cortex using postmortem brain tissue.	A significantly higher ratio of VENs to pyramidal neurons was found in the sample of ASD patients.
Eyler et al. [64]	12–48 months	40 with ASD and 40 TD.	Lateralised abnormalities of temporal cortex processing of language.	Deficient left hemisphere response to speech sounds and exhibited abnormally right-lateralised temporal cortex response to language was found in at-risk toddlers who later received a diagnosis of ASD. Difference becomes greater with age.
Hazlett et al.; [65]	About two years of age. MRI was carried out again approximately 24 months later (when aged 4-5 years; 38 children with ASD; 21 controls).	59 children with ASD and 38 control children. Thirty-eight children with ASD and 21 comparison cases were examined at the follow-up visit.	Early growth trajectories in brain volume (cerebral gray and white matter) and cortical thickness.	Generalised cerebral cortical enlargement in individuals with ASD at both two and four to five years (being 9% larger in ASD group). Despite no difference in cortical thickness, children with ASD had enlargement in both grey and white matter volume for all cortical lobes (temporal, frontal and parieto-occipital lobes). Disproportionate enlargement in temporal lobe white matter only was found in the ASD group after controlling for total brain volume.
Hoeft et al. [66]	FXS group—mean age 2.9 years. idiopathic autism (iAUT)—mean age 2.77 years. Typical developing—mean age 2.55 years. Idiopathic developmentally delayed controls—mean age 2.96 years.	52 males diagnosed with FXS. 63 with Idiopathic autism (iAUT). 31 TD. 19 idiopathic DD controls.	Whole-brain morphometric patterns.	Greater volume was evident in iAUT compared with controls, who in turn had greater volume than FXS. Frontal and temporal grey and white matter regions often implicated in social cognition, including the medial prefrontal cortex, orbitofrontal cortex, superior temporal region, temporal pole, amygdala, insula, and dorsal cingulum were abnormal in FXS and iAUT.
Schumann et al. [67]	1.5 years–5 years of age. Mean, 30 months plus or minus 10 months).	41 toddlers who received a confirmed diagnosis of autism disorder at 48 months of age and 44 TD controls.	Cerebral gray and white matter growth.	Cerebral grey and white matter growth abnormalities in individuals with ASD. Within cortex, the most significant differences in volume, and age-related change took place in anterior regions of the brain (frontal grey, temporal grey, and cingulate grey cortices). Posterior cerebral regions less affected. Abnormal growth most pronounced in temporal grey matter volumes.
Rommelse et al. [68]	0–19 months	129 children with ASD and 59 children with non-ASD psychiatric disorders.	Head circumference, height, and weight.	Similar abnormal patterns of growth compared to population norms were found in both groups. Abnormal HC growth may actually be common to psychiatric disorders, rather than ASD specifically. However, the most apparent difference was that the children with ASD showed an increased HC relative to height up to two months of age, an increase not found in the PC group at this age.
Muratori et al. [69]	Birth (TO), 1-2 months (T1), 3–5 months (T2), and 6–12 months (T3).	50 with ASD and 100 TD.	Head circumference, body height, and body weight.	Weight was significantly less in ASD subjects compared to healthy infants from 1-2 months onwards. After controlling for weight and height, excessive rate of HC growth from birth was found in the individuals with ASD.
Fukumoto et al. [70]	3–12 months	280 children with ASD.	Head circumference, body height, and body weight.	Increases in HC growth from 3–12 months, in height from 3–9 months, and in body weight from 3 to 6 and 12 months were found in the males with ASD. Increases in HC, body height and body weight were only observed at three months in the females with ASD. Only HC in the male ASD group was significantly increased from 6–9 months after birth.
Chawarska et al. [71]	0–24 months	Autism disorder (𝑛=64), PDD-NOS (𝑛=34), global developmental delay (𝑛=13), and other developmental problems (𝑛=18), and TD boys (𝑛=55).	Head circumference growth in ASD, height, and weight growth. Investigate association between head circumference growth from birth to 24 months and measures of cognitive functioning.	Boys with ASD were found to be significantly longer by 4.8 months, had greater HC by age 9.5 months and weighed more by age 11.4 months, compared to the typically developing boys. No other clinical groups displayed an overgrowth. Boys with ASD in the top 10% of overall physical size in infancy displayed more severe social deficits and lower adaptive functioning at 2 years.
Nordahl et al. [72]	2–4 years	Boys and girls with ASD (𝑛=53, no regression (nREG); 𝑛=61, regression (REG)) and a control group of age-matched typically developing controls (𝑛=66).	Total brain volume (rapid head growth).	Abnormal brain enlargement was most common in boys with regressive autism. Brain size in boys without regression was similar to controls. Head circumference in boys with regressive autism was normal at birth but deviated from normal growth trajectories (other groups) around the age of 4–6 months. No brain size differences in girls with autism (𝑛=22, ASD; 𝑛=24, controls).
Courchesne et al. [37]	12 months and 50 years of age for the typical group and 2–50 years for the ASD group.	259 ASD subjects and 327 TD subjects.	Brain size based on the analyses of 586 longitudinal and cross-sectional MRI scans.	Evidence of overgrowth throughout infancy and the toddlerhood in both boys and girls with ASD which was subsequently followed by an accelerated rate of decline in size.
Barnard-Brak; et al. [73]	3 time points (9, 24 and 36 months)	About 9,000 children.	Head circumference growth trajectory.	No difference in HC at any of the 3 time points in the children with ASDs.
Whitehouse et al. [74]	18 weeks gestation and also at birth	14 children with ASD were matched with four control participants (𝑛=56).	Head circumference was measured using ultrasonography at about 18 weeks gestation and also at birth using a paper tape-measure. Overall body size was indexed by foetal femur-length and birth length.	No difference in head circumference at either time-point between the groups.
Constantino et al. [75]	15 time points starting from birth to 36 months of age	48 sibling pairs in which one (𝑛=28) or both (𝑛=20) sibs were affected by an ASD and 85 control male sibling pairs	Serial head orbitofrontal circumference measurements.	Significant acceleration of head growth in individuals with ASD compared to controls. The study also showed that infant HG trajectory may be endophenotypic but was not a reliable indicator of risk of ASD among siblings of ASD in this study.
Gray et al.  [76]	Birth and 18.5 months.	Children with autism (𝑛=86) and children with DD without autism (𝑛=40).	Head circumference at birth and rate of change in head circumference.	No differences between the group of children with both ASD and developmental delay compared with the group with developmental delay alone. However, when compared with normative data, children with ASD had significantly smaller HCs at birth and significantly larger HC at 18.5 months of age.
Mraz et al. [77]	0–25 months	24 children who maintained their diagnoses, 15 children who lost their diagnoses and 37 TD controls.	Head circumference, length and weight growth.	Compared to controls, HC and weight growth were significantly larger in both ASD groups and there were no significant differences between ASD groups.
McCleery et al. [78]	10 months	20 high-risk infants (siblings of an older sibling diagnosed with ASD) and 20 low-risk control subjects.	Cortical responses to face/object processing using event-related potentials.	The low-risk group displayed faster responses to faces compared to object stimuli (P400) which was not observed in the high-risk group. Conversely, faster responses to objects than faces in high risk but not low-risk infants (N290). Right hemisphere advantage (greater hemispheric asymmetry) in the typical that was not found in the ASD group.
Luyster et al. [79]	12 months	32 infants at high-risk of ASD and 24 low-risk control infants.	Atypical neural responses to social stimuli.	No significant group differences in the neural response to faces. Trend for the low-risk group to exhibit more marked differential response to familiar and unfamiliar faces (in the anticipated direction) compared with high-risk infants.
Elsabbagh et al. [80]	Mean age—10 months	Nineteen infant siblings of children diagnosed with ASD and 17 control infants with no family history of ASD.	Neural correlates of direct and averted gaze.	Prolonged latency of the occipital P400 event-related potentials component in response to direct gaze was exhibited in the sib-ASD group compared to control infant. No difference between the groups in the P400 latency for Averted gaze.
Elsabbagh et al. [8]	6–10 month. About 18 to 30 months later, these children were clinically assessed for ASD.	Infants at high familial risk for ASD (𝑁=54) and infants at low risk (𝑛=50).	Neural sensitivity to eye gaze.	Characteristics of ERP components evoked in response to dynamic eye gaze shifts during infancy were associated with ASD diagnosis at 36 months.
Key and Stone [81]	Approximately nine months 15 days old	20 typical infants and 15 infant siblings of children diagnosed with ASD.	Speed of processing of novel versus familiar faces using event related potentials and eye tracking.	Both infant groups demonstrated the ability to differentiate between mothers and strangers, as shown in the amplitude modulations of posterior N290/P400 and frontal/central Nc responses. However there was a delayed ERP response to the stranger face (as evidenced by the latency of the P400 response) in the typical infants only.
Key and Stone [82]	Mean age of High-Risk group was 9.01 (0.34).	35 infants (20 average-risk typical infants, 15 high-risk siblings of children with ASD).	To investigate whether infants at high risk for ASDs process facial features (eyes, month) differently. Also, whether this is associated with social and communicative skills.	All infants detected eye and mouth changes. However, different brain mechanisms were used. Facial feature changes were related to changes in activity of the face perception mechanisms (N290) for the average-risk group only.
Chawarska and Shic [83]	Testing at 2 years and then again at 4 years. Chronological age (months) for ASD at time 1 26.9 (6.2) and time 2 46.4 (6.4). For control group mean age at time 1 26.3 (6.5) and time 2 46.3 (4.3).	44 children with ASD and 30 TD controls.	Eye tracking—atypical face scanning.	Toddlers with ASD looked increasingly away from faces with age and atypically attended to key features within the face. They also demonstrated at both ages impairment in the ability to recognise faces.
Bosl et al. [84]	6–24 months	46 high risk for ASD, defined on the basis of having an older sibling with a confirmed diagnosis of ASD and 33 controls.	Modified multiscale entropy (mMSE) computed on the basis of resting state EEG data.	Multiscale entropy appears to go through a different developmental trajectory in infants at high risk for ASD than it does in typically developing controls with differences being most marked at ages 9–12 months.
Webb et al. [85]	ASD group—18–47 months. Typically dev.—12–30 months.	24 children with ASD and 32 TD children.	Neural responses to familiar and unfamiliar faces.	Delayed development in the individuals with ASD was indicated since neural responses to faces in this group of children resembled those observed in younger typically developing children.
Dinstein et al. [86]	1–3.5 years	72 toddlers in total. Broken down in study: All toddlers (12–46 months)—Autism (𝑛=29), Control (𝑛=30) and language delay (𝑛=13). Young toddlers (12–24 months)—Autism (𝑛=12), Control (𝑛=16) and Language Delay (𝑛=11).	Spontaneous cortical activity of naturally sleeping toddlers with autism. fMRI data.	Toddlers with autism exhibited significantly weaker interhemispheric synchronization (i.e., weak ‘‘functional connectivity’’ across the two hemispheres) in putative language areas.
Stahl et al. [87]	10 month olds	10 infant at risk of ASD.	Aim was to discuss the use of machine learning and discrimination methods and their possible application to the analysis of infant event-related potential (ERP) data.	Classification methods (regularised discriminant function analyses and support vector machines) can increase the discriminative power of ERP measurements. Using cross-validation, both methods successfully discriminated at above chance levels between groups of infants at high and low risk of a later diagnosis of autism. However, infants could only be discriminated in the direct gaze condition, not in the averted gaze condition.
Wolff et al. [88]	6 to 24 months in high-risk infants.	92 high-risk infant siblings from an ongoing imaging study of ASD.	The authors prospectively examined white matter fiber tract organisation from 6 to 24 months in high-risk infants who developed ASD by 24 months.	The fractional anisotropy trajectories for 12 of 15 fiber tracts were significantly different between the infants who developed ASDs compared to those who did not.
Hazlett et al. [89]	68 boys with idiopathic autism (ASD). 18 to 42 months of age.	53 boys with fragile X syndrome (FXS), 68 boys with idiopathic autism (ASD), and a comparison group of 50 typically developing and developmentally delayed controls.	Total brain volumes and regional (lobar) tissue volumes were also examined.	Children with idiopathic autism were found to have a generalised cortical lobe enlargement.
Calderoni et al. [90]	Female children with ASD (ASDf)—2–7 years.	ASDf (𝑛=38) compared to 38 female age and non verbal IQ matched controls.	Aim to investigate the neuroanatomical phenotype of female children with ASD.	The between-group whole-brain and brain-segment volume comparison revealed a total intracranial volume (TIV) enlargement of approximately 5% in female children with ASD. The conventional VBM analysis showed evidence of an increased GM volume in a specific region of the left superior frontal gyrus of ASDf. The implementation of the SVM analysis on the GM segments obtained in the VBM-DARTEL pre-processing highlighted a more complex circuitry of increased cortical volume in ASDf, involving bilaterally the SFG and the right temporo-parietal junction (TPJ).
Zeegers et al. [91]	Between 2–7 years.	34 children with ASD and 13 developmentally delayed children without ASD, (matched on age and developmental level).	To investigate volumes of cranium, total brain, cerebellum, grey and white matter, ventricles, hippocampus, and amygdale.	No significant differences in volumes of intracranium, total brain, ventricles, cerebellum, grey or white matter or amygdala and hippocampus between the ASD group and the developmentally delayed group were found.
Hazlett et al. [92]	6 month-old infants at high risk for ASD	Infants at high risk for ASD (𝑛=98) compared to infants without family members with ASD (𝑛=36).	MRI study examining head circumference, brain volume and radiologic abnormalities.	No group differences.
Duffy and Als [93]	Between 2–4 years	The 2- to 12-year-old subsample consisted of 430 ASD- and 554 C-group subjects (𝑛=984).	The current study attempts to answer the as yet open question of coherence differences between children with ASD and neuro-typical healthy controls. EEG coherence data was evaluated in a large sample of children with ASD and compared to a large neurotypical, medically healthy, normal, age-comparable control group.	A stable pattern of EEG spectral coherence was found to distinguish children with ASD from neurotypical controls.