From impasse to insight in autism research: from behavioral symptoms to biological explanations.

The incomplete interface between remediation-oriented research and basic science research has hampered progress toward gaining insight into the etiologies of autism, despite the availability of abundant research data. Investigators of these two research domains differ in their background training and primary goals, which necessarily affect their missions, perspectives, research questions posed, methodologies selected, and interpretation of data from the research. Miscommunication between the two types of researchers has brought about disagreement on nearly every aspect of the research process. We discuss both sides of the impasse: a traditional clinical practice perspective based on the requirement for finding immediate answers to the remediation question and the basic science perspective with the goal of delineating the sequence of biological changes from the initial cause(s) of abnormal development to behavioral outcome. Although remediation-oriented research aims at alleviation of symptoms for today's patients, we propose that a basic science perspective seeks insight into the triggering causes and pathogenesis of the disorder from which better diagnosis and remediation may be devised for patients in the future. We suggest that research in autism can progress beyond the impasse of disagreement and competition toward information integration and insight by means of dialogue, data exchange, discussion, collaboration, and cooperation.

Neurologic abnormalities in infantile autism.

Neuroanatomic, pathologic, and neurobehavioral studies point to a cerebellar and parietal abnormality in autism. We used a standardized protocol to examine neurologic function in 28 pediatric autistic subjects and 24 pediatric normal healthy volunteer controls. As a group, the autistic subjects had quantitative measures from magnetic resonance imaging suggesting hypoplasia or hyperplasia of the cerebellar vermis, as well as measurements of posterior corpus callosum suggesting abnormalities of posterior cortex. In groups of tests that reflect cerebellar and parietal function, the neurologic abnormalities detectable by clinical examination were significantly greater for autistic subjects than for normal controls. These studies confirm that the structural and behavioral deficit in autism does lead to abnormalities that can be detected on the clinical neurologic examination.

Impairment in shifting attention in autistic and cerebellar patients.

MRI and autopsy evidence of early maldevelopment of cerebellar vermis and hemispheres in autism raise the question of how cerebellar maldevelopment contributes to the cognitive and social deficits characteristic of autism. Compared with normal controls, autistic patients and patients with acquired cerebellar lesions were similarly impaired in a task requiring rapid and accurate shifts of attention between auditory and visual stimuli. Neurophysiologic and behavioral evidence rules out motor dysfunction as the cause of this deficit. These findings are consistent with the proposal that in autism cerebellar maldevelopment may contribute to an inability to execute rapid attention shifts, which in turn undermines social and cognitive development, and also with the proposal that the human cerebellum is involved in the coordination of rapid attention shifts in a fashion analogous to its role in the coordination of movement.

Abnormality of cerebellar vermian lobules VI and VII in patients with infantile autism: identification of hypoplastic and hyperplastic subgroups with MR imaging.

OBJECTIVE: Infantile autism is a neurobehavioral disorder that is widely believed to have etiologically distinct subtypes, including subtypes with a genetic basis, but no neuroanatomic evidence firmly supports this belief. To date, only one type of cerebellar abnormality has been identified in patients with autism: hypoplasia of the vermis and hemispheres. By using a large sample of autistic patients and healthy volunteers along with precise MR imaging and quantitative procedures, we sought to replicate previous reports of cerebellar vermian hypoplasia in autism and to identify additional subtypes of cerebellar abnormality. MATERIALS AND METHODS: Using MR technology, we imaged and measured posterior and anterior vermian regions in 50 autistic patients (2-40 years old) and 53 healthy control subjects (3-37 years old). The autistic patients had social, language, cognitive, behavioral, and medical history characteristics that were typical of the general autistic population. By using precise procedures for positioning and aligning MR slices, we obtained comparable MR images within and across subject groups. RESULTS: Statistical analyses showed two subgroups of autistic patients, one (86% of the patients) with findings consistent with vermian hypoplasia and another (12% of the patients) with evidence of vermian hyperplasia. The hypoplasia subgroup included 43 patients whose mean midsagittal area for vermian lobules VI and VII was 237 +/- 38 mm2, and the hyperplasia subgroup included six patients whose mean area was 377 +/- 12 mm2. Thus, the area of lobules VI and VII in the hypoplasia subgroup was 16% smaller than the mean area in the control subjects (282 +/- 42 mm2) (p < .0001), whereas that in the hyperplasia subgroup was 34% larger (p < .0001). Analyses showed that these two subtypes of vermian abnormalities were present across all ages of autistic patients studied. CONCLUSION: Two different subtypes of autistic patients can be identified on the basis of the presence of vermian hypoplasia or hyperplasia as seen on MR images. Possible origins for vermian hypoplasia include environmental trauma and genetic factors.

Parietal lobe abnormalities detected with MR in patients with infantile autism.

OBJECTIVE: Infantile autism is a neurologic disorder that severely disrupts the development of many higher cognitive functions. The most consistent abnormal neuroanatomic findings in autism are loss of Purkinje neurons in the posterior cerebellum as detected by autopsy studies and hypoplasia of the posterior cerebellar vermis and hemispheres as detected by in vivo neuroimaging. Evidence of developmental arrest has also been detected in limbic structures in autopsy studies of autistic patients with mental retardation. Neither in vivo neuroimaging nor autopsy studies of autistic persons have reported abnormalities in the cerebrum. Because the cerebrum mediates many higher cognitive functions, such as social communication, language, abstract reasoning, planning, and organization, that are known to be deficient in patients with autism, a closer examination of the neuroanatomy of the cerebrum in infantile autism is warranted. MATERIALS AND METHODS: MR images of 21 healthy autistic patients (6-32 years old) were mixed with MR images of control subjects and reviewed on four separate occasions by a neuroradiologist for any neuroanatomic abnormalities. Autism was diagnosed on the basis of criteria for autism as defined by the Diagnostic and Statistical Manual of Mental Disorders, and the autistic patients did not have any other concurrent neurologic disorders. To control for systematic bias in judging the type and location of abnormalities in the autistic population, three control groups were used: a normal control group of 12 subjects, a control group of 23 nonautistic patients with a variety of brain abnormalities for the first review, and another control group of 17 nonautistic patients for the second review. Control patients with brain abnormalities were selected from patients' files on the basis of MR findings of a variety of brain abnormalities. All MR images were coded for anonymity, randomly mixed, and examined by a neuroradiologist blinded to the purpose of the study and to the group membership of each subject. All normal and abnormal findings seen on the MR images of each subject were described on a standard form listing all major brain structures to ensure an examination of each structure in turn. To test for reliability, three subsequent reviews were performed by the same neuroradiologist. RESULTS: Parietal lobes were abnormal in appearance in 43% (9/21) of autistic patients. Cortical volume loss in the parietal lobes was seen in seven autistic patients; in four of these cases, cortical volume loss extended either into the adjacent superior frontal or occipital lobe. Additional abnormalities detected with MR in these nine patients included white matter volume loss in the parietal lobes (three patients) and thinning of the corpus callosum, especially along the posterior body (two patients). Abnormalities were bilateral. The mesial, lateral, and orbital regions of the frontal lobes; temporal lobes; limbic structures; basal ganglia; diencephalon; and brainstem were normal in all autistic patients. No abnormalities were found in the 12 normal control subjects. The control subjects with neurologic abnormalities had various abnormal findings consistent with their medical conditions. CONCLUSION: Our results indicate that the parietal lobes are reduced in volume in a portion of the autistic population. Possible origins for this localized cerebral abnormality include early-onset altered development and late-onset progressive atrophy.

Cerebellar and cerebral abnormalities in Rett syndrome: a quantitative MR analysis.

Rett syndrome is a neurodegenerative disease of young girls that begins in early childhood with autismlike behavior and loss of language skills, and progresses with marked deterioration of the motor system in the second decade of life. The purpose of this study was to determine if neuroanatomic changes detected with MR imaging could help to explain the clinical presentation and progression of signs and symptoms in these patients. Accordingly, computer-assisted planimetry was used to measure various dimensions of cerebral, cerebellar, and brainstem structures on sagittal and transverse MR images of 13 patients with Rett syndrome and 10 healthy volunteers. Dimensions of the cerebrum, basal ganglia, cerebellum, and brainstem were measured on transverse images. Areas of cerebellar vermian lobules, the fourth ventricle, the pituitary gland, and the corpus callosum were measured on sagittal images. Fourteen dimensions and areas were measured in each patient and each control subject; according to two-tailed Student's t tests, all but two values were significantly smaller in the patients with Rett syndrome than in control subjects. Graphing the measurements against age by using simple linear regression revealed progressive cerebellar atrophy without evidence of atrophy of the brainstem or cerebrum. Our results indicate that patients with Rett syndrome have global hypoplasia of the brain and progressive cerebellar atrophy increasing with age. Cerebellar atrophy with age may contribute to the deterioration of the motor system seen in older patients with Rett syndrome.

Absence of magnetic resonance imaging evidence of pontine abnormality in infantile autism.

In vivo studies involving magnetic resonance imaging and studies of neuropathologic specimens have shown that autism is most consistently associated with developmental hypoplasia of the neocerebellum. We investigated whether the cerebellar hypoplasia was accompanied by gross structural abnormalities in the major input (cerebrocerebellar) and output (cerebrorubral) pathways to the cerebellum by measuring the area of the ventral pons (including the pontine nuclei and the transverse fibers) and the midbrain on midsagittal magnetic resonance images in 34 autistic and 44 subjects. The area of the entire pons and several regions of interest within the midbrain (including the superior and inferior colliculi) were also determined with midsagittal magnetic resonance images. We found no significant difference between measurements of the pons and midbrain in autistic and control subjects. Our data show no evidence of gross anatomic abnormalities in the input and output pathways to the cerebellum in autism, a finding that is consistent with previous studies of neuropathologic specimens; rather, the reduced size of the neocerebellum in autism appears to be the result of maldevelopment within the cerebellum itself.

Brainstem auditory evoked potentials in receptive developmental language disorder.

It has been hypothesized that receptive developmental language disorder (RDLD) may be explained by an auditory processing deficit. The neuroanatomical locus of this deficit is unknown. Brainstem auditory evoked potentials (BAEPs) reflect the functioning of the auditory nerve and auditory brainstem pathways to high-frequency acoustical stimulation in humans and reflect the first stages of auditory processing. These were studied in 12 subjects with RDLD (four females and eight males, ages 12 to 19) and twelve control subjects (three females and nine males, ages 14 to 24). Click intensity and rate of stimulation were varied. The BAEPs for the RDLD group were comparable to the control group as well as to hospital norms across intensity levels and stimulation rates. The evidence obtained suggests that a disorder in the neurophysiological systems underlying the BAEPs and reflecting initial stages of auditory processing is not essential for RDLD.

Reduced cerebellar hemisphere size and its relationship to vermal hypoplasia in autism.

Cerebellar hemisphere size was calculated in 10 autistic and 8 normal control subjects by summing the cross-sectional areas of cerebellar hemisphere tissue measured on paramidline sagittal magnetic resonance images. The areas of two cerebellar vermal regions (lobules I through V and lobules VI through VII) were also measured using the midsagittal image. Our cumulative slice area measure of cerebellar hemisphere size was significantly smaller in the autistic subjects than in the control group. The cumulative slice area correlated positively with the area of vermal lobules VI through VII only in the autistic subjects. Our results indicated that the decreased size of the cerebellar hemispheres and vermal lobules VI through VII was associated with autism.

Pathophysiologic findings in nonretarded autism and receptive developmental language disorder.

In nonretarded autistic, receptive developmental language disordered, and normal subject groups, we recorded in auditory and visual target detection tasks two neurophysiological components of the event-related brain potential, Nc and P3b. Existent research shows that, in normals, Nc and P3b appear early in development, are associated with attention and memory processes, and are endogenous which means that they are triggered by internal, consciously initiated attentional and cognitive mechanisms and that they can be triggered even by the omission of sensory stimulation so long as it has meaning or importance for the subject. In this report, Nc and P3b were recorded in response to auditory and visual stimulation and to the omission of auditory and visual stimulation. Consistent with the hypothesis that non-retarded autism involves abnormal attentional and cognitive responses to important information, P3b was found to be smaller than normal and Nc was small and often absent in the nonretarded autistic group even under the condition when no auditory language or sensory processing was required. Receptive developmental language disorder has been linked with difficulties in processing sequences of auditory stimuli, and in this study P3b was found to be somewhat enlarged in this group even under the conditions when P3b was elicited by stimuli separated by 1 sec and also when P3b was elicited by the omission of stimulation.

Hypoplasia of cerebellar vermal lobules VI and VII in autism.

Autism is a neurologic disorder that severely impairs social, language, and cognitive development. Whether autism involves maldevelopment of neuroanatomical structures is not known. The size of the cerebellar vermis in patients with autism was measured on magnetic resonance scans and compared with its size in controls. The neocerebellar vermal lobules VI and VII were found to be significantly smaller in the patients. This appeared to be a result of developmental hypoplasia rather than shrinkage or deterioration after full development had been achieved. In contrast, the adjacent vermal lobules I to V, which are ontogenetically, developmentally, and anatomically distinct from lobules VI and VII, were found to be of normal size. Maldevelopment of the vermal neocerebellum had occurred in both retarded and nonretarded patients with autism. This localized maldevelopment may serve as a temporal marker to identify the events that damage the brain in autism, as well as other neural structures that may be concomitantly damaged. Our findings suggest that in patients with autism, neocerebellar abnormality may directly impair cognitive functions that some investigators have attributed to the neocerebellum; may indirectly affect, through its connections to the brain stem, hypothalamus, and thalamus, the development and functioning of one or more systems involved in cognitive, sensory, autonomic, and motor activities; or may occur concomitantly with damage to other neural sites whose dysfunction directly underlies the cognitive deficits in autism.

Abnormal neuroanatomy in a nonretarded person with autism. Unusual findings with magnetic resonance imaging.

Recent studies of infantile autism using computed tomographic scanning emphasized the importance of studying cases of classic autism (Kanner's syndrome) without complicating conditions such as mental retardation. Computed tomographic scan studies of such patients reported no evidence of anatomical abnormalities of cerebral hemispheres or of subcortical structures, which are defined by landmarks such as the lateral ventricles and lentiform nuclei. Examination of the cerebellum was not mentioned. The most recent postmortem neuropathologic study reported significant cerebellar abnormality, but the study was of a severely retarded autistic individual. Using magnetic resonance imaging, we have found in vivo evidence of a significant and unusual cerebellar malformation in a person with the classic form of autism uncomplicated by mental retardation (current nonverbal IQ = 112), epilepsy, history of drug use, postnatal trauma, or disease. The finding showed hypoplasia of the declive, folium, and tuber in posterior vermis, but not of the anterior vermis, and hypoplasia of only the medial aspect of each cerebellar hemisphere. The right posterior cerebral hemisphere also showed pathologic findings.