Elsevier

Clinical Neurophysiology

Volume 126, Issue 2, February 2015, Pages 297-303
Clinical Neurophysiology

Development of the posterior basic rhythm in children with autism

https://doi.org/10.1016/j.clinph.2014.04.022Get rights and content

Highlights

  • We quantitatively modeled development of the posterior basic rhythm in children.

  • Compared to normal children, a subset of autistic children has an earlier maturation of the posterior basic rhythm, especially in the 2- to 4-year old age range.

  • Autistic children also show a distributed decrease in band-limited coherence of the posterior basic rhythm.

Abstract

Objective

Early detection of autism is critical for effective intervention, but currently, no simple screening tests are available. Furthermore, little is known about the development of brain dynamics in young children. We examine the early neurophysiological manifestations of autism by retrospectively analyzing EEG. In particular, we focus on maturation of the posterior basic rhythm (PBR), which is one of the most characteristic features of the normal EEG, and comprises a discrete functional state.

Methods

Subjects with a diagnosis of autism (n = 74), as well as normal (n = 134) and epileptic (n = 108) controls, were extracted retrospectively from our digital EEG database. Segments with clear PBR were extracted, and standard signal analysis methods were used to calculate peak PBR frequency, power, and coherence.

Results

In our cohort, a subset of autistic children show accelerated development of the PBR, with early maturation especially in the 2- to 4-year old range. The overall coherence of PBR-specific activity is also lower in autistic children in our cohort.

Conclusions

These findings provide evidence that autism is associated with accelerated development of the PBR.

Significance

These findings generate a clinical hypothesis for future prospective studies on the efficacy of these simple measures as a diagnostic or screening tool.

Introduction

The prevalence of autism and autistic spectrum disorders is thought to approach 1% (Autism and Developmental Disabilities Monitoring Network Surveillance Year, 2006 Principal Investigators, 2009, Baron-Cohen et al., 2009), but the underlying pathophysiology is yet unclear. Recently, attention has been directed to the “connectivity hypothesis,” whereby the autistic brain is thought to have aberrant white matter tracts. This hypothesis was inspired by observations of abnormal anatomical development of head circumference (Kanner, 1968) and white matter (Herbert et al., 2004) in autistic children, and posits that underconnectivity leads to less integration of information, and more posterior autonomy in sensory processing (Just et al., 2012). Functional results suggestive of this hypothesis have been obtained from resting state fMRI, MEG, resting state EEG and event related potential (ERP) recordings.

In order to investigate developmental dynamics and connectivity of the EEG, we focus here on the posterior basic rhythm (posterior dominant rhythm, PBR) (Berger, 1929). The PBR is one of the most characteristic features of the normal EEG. It is most prominent in the occiput, and is strongest during quiet wakefulness, with eyes closed. It appears stably at around 3 months at around 3–4 Hz, and accelerates to the adult range of around 10 Hz by approximately 10 years old (Kellaway, 1990). We focus on PBR for three main reasons: Firstly, the developmental time course of PBR coincides with the time during which signature pathological events of autism are thought to occur. Secondly, it is a discrete resting state which is robustly present in most EEG records with high signal-to-noise ratio. Thirdly, EEG results from adult and aging patients suggest that PBR is associated with cognitive performance (Klimesch, 1999, Sauseng et al., 2009).

Section snippets

EEG database and data acquisition

Digital EEGs from 93 autistic individuals were identified by retrospectively searching the EEG database at Georgetown University Hospital. Only children 0–16 years old at the time of acquisition were included. Included studies were conducted between March 2003 and May 2012. As controls, 134 normal children and 108 epileptic children were also identified. Diagnoses were corroborated by review of clinical charts and referral materials. Nineteen subjects were both autistic and epileptic according

Development of PBR frequency

We confirmed the well-known trend of PBR frequency acceleration (Berger, 1932, Kellaway, 1990) in our normal subject group (Fig. 1A), and modeled the trend with a nonlinear sigmoidal fit to a hyperbolic tangent: 1.07088×106Tanh0.114486x+6.45642-1.07087×106. This fit was plotted (solid line), along with 95% single sample prediction intervals (dotted lines) and raw data points (dots). In Fig. 1B, autistic (green) and epileptic (red) subjects are plotted. Subjects who were both autistic and

Discussion

We analyzed retrospective data from children with a diagnosis of autism and those without, and focused on characteristics of the PBR in order to investigate the development of resting-state cortical dynamics in autistic children. Our main findings are that a statistically significant subset of children with autism show earlier PBR maturation than would be expected from normal development. Furthermore, the global coherency of PBR activity is decreased in autistic children.

Various recent studies

Conclusions

In our retrospective study, we find statistically significant population trends in the development of autistic cortical dynamics. Our main finding is that our autistic cohort shows accelerated maturation of the PBR, especially in the age range of 2–4 years old. Our statistically significant findings for such a simple metric is notable given the variability of our clinical sample. Whether the simple measures presented here could potentially differentiate between autism subtypes remains to be

Acknowledgements

This study was supported by the Alexander von Humboldt Stiftung/Foundation. We thank Drs. Tim Wanger and Iain Dewitt for scientific comments on the manuscript.

References (35)

Cited by (7)

View all citing articles on Scopus
View full text