Affiliations: The Neuro, McGill, Canada
Journal reference: https://doi.org/10.1097/HRP.0000000000000072
Summary: Autism Spectrum Disorders (ASD) are sometimes difficult to diagnose. This article summarizes a systematic review examining measures of connectivity of different brain areas and how these connections might change in a person with ASD.
Autism spectrum disorder (ASD) is a collection of developmental conditions which affects individuals in different ways. Though a variety of behavioural symptoms might be considered within the spectrum, ASD is most often characterised by social impairment and repetitive behaviours or interests. Because of its variable nature, ASD is difficult to diagnose and study with precision, as many competing factors may have an impact on how the disorder manifests. With improving research methods and medical understanding of the disorder, researchers hope to discover the biological causes of ASD, approaching the disorder from different angles of study. A systematic review by Rane and colleagues (2015) assessed studies which focused on measures of connectivity –how the brain is functionally and anatomically connected, and how these connections might change in a person with ASD.
Connectivity can be measured in the brain in numerous ways, including looking at cell-to-cell connections or at the brain as a whole. The former presents detailed measures of how brain cells are connected directly, but requires invasive methods such as slicing a brain after a subject has died. With brain scanners using magnetic resonance imaging (MRI), we can use non-invasive methods to look at a picture of the whole brain. The compromise with MRI scans, however, is a much lower resolution; we cannot scan at a single cell resolution across the whole brain but we can make estimations and inferences about bigger connections. The two main methods used to study brain connectivity are diffusion tensor imaging (DTI) and functional MRI (fMRI). DTI uses water diffusion in the brain to estimate the structural (ie anatomical) connections within the cortex, while fMRI measures blood flow to indicate what areas are active in coherence with each other, thereby suggesting a functional connection.
Rane et al. looked at the previous decade’s studies using DTI and fMRI to research changes in connectivity in individuals with ASD. The main purpose of clinical research assessing differences between subjects with and without a disorder is to demonstrate whether a pattern can be found which distinguishes the two groups. Researchers also look for patterns that are consistent within the group with a specific disorder. Defining these patterns helps clinicians look for clear signs of a disorder in their patients and identify appropriate interventions. ASD presents a challenge in this respect because it is expressed in a highly individualised manner: not all people with ASD have the same symptoms or biological make-up. Rane et al. hoped to combine information from different studies and better illustrate whether a specific pattern of brain connectivity was emerging in ASD research.
Across 33 resting-state fMRI and 36 DTI studies, Rane et al. found a range of results, implicating many different brain regions and patterns of connectivity. The included studies used different age ranges, different gender groups and multiple imaging protocols. The metrics they used to define connectivity also varied, as well as the behavioural tests employed to assess social or language ability. Despite this, some commonalities did emerge, namely that ASD subjects exhibit decreased connectivity, both structurally and functionally. This may mean that ASD subjects’ brains had lower estimations in the number of physical connections from one cortical region to another, or that the directionality of these connections was weaker as determined by decreased fractional anisotropy (FA, a measure of how one-directional a connection is). Using resting state fMRI scans a reduction in connectivity is indicated by weaker coherence between regions or fewer functional “hubs” depicted across the brain. Depending on the region or type of connection assessed, studies showed both increases and decreases in the same measures across the brain, as opposed to a consistent uni-directional pattern in every area.
Most studies found decreases especially in long-range connections which transmit information between areas that are not physically near each other. These long-range connections are the super highways of the brain, integrating and consolidating information from many different inputs. Most of these affected connections included those emerging from the prefrontal cortex (PFC), a region at the front of the brain used in higher order cognitive processing. Related to the PFC and also frequently noted here for reduced long-range connections are the anterior and posterior cingulate cortices (ACC and PCC), regions that are involved in cognitive and inhibitory control. It is suggested that this decrease in long-range connectivity could be due to over-pruning. Pruning is a developmental process which streamlines unnecessary brain connections and makes them more efficient. Studies have shown that children with ASD may first have an overgrowth of brain connections and then overcompensate with too much pruning, resulting in fewer connections than normal. This process may be regionally specific as certain other shorter or local connections do not appear to be reduced in ASD. In fact, local connections may even show some increased connectivity.
How these changes relate to behaviour is less clear. Many studies showed a negative correlation between reduced connectivity and social impairment, meaning that the bigger the reduction in connectivity, the more severe the deficit in behavioural ability. These findings, however, were not consistent with some studies showing increased expressive and oral reading ability or no strong correlation between brain changes and behaviour at all. As such, direct relationships between brain connectivity and behaviour cannot be inferred conclusively. To establish definitive correlations between physical brain patterns and behavioural symptoms requires rigorous testing, with much larger sample sizes and results that hold across groups and circumstances. As it stands, the literature still cannot illustrate a consistent, direct cause between brain connectivity patterns and ASD symptoms.
The use of MRI techniques to assess ASD is crucially helpful in understanding the disorder in development, and its effects on the brain as a whole. While many of the studies reviewed here indicated an emerging trend of reduced connectivity and its relationship to social impairment, more data are necessary to reach reliable conclusions. With better tool development and increased study sizes, researchers will gain stronger insight into the biological patterns of ASD, brain connectivity and how these interact to influence behaviour.