▪ Mitochondrial Disorders [Weissman et al.,2008]

Biochemical evidence consistent with possible mitochondrial dysfunction in ASD (increased blood lactate levels) has been reported for more than two decades, with more recent direct interest in mitochondrial disorders. One population-based study reported that 7% of individuals with ASD also met clinical criteria for a mitochondrial disorder [Oliveira et al., 2005]. Neurological findings were notably absent.

The current study examined a group of 25 patients with ASD who also had clinical and biopsy (23 muscle biopsy, 11 skin biopsy and 1 liver biopsy) confirmed mitochondrial disorders. The male:female ratio was 13:12, whereas there is typically a strong male preponderance in ASD. In contrast to Oliveira et al. (2005), the authors found that 15 of the 25 affected individuals had at least one neurological finding uncommon in ASD (e.g. ptosis, microcephaly). Seven of the 25 individuals also had a cardiovascular abnormality and 24 had a major clinical finding uncommon in ASD. Excessive fatiguability was common. Although regression affects about one third of individuals with ASD, regression in individuals with a mitochondrial disorder showed atypical features: multiple episodes of regression, loss of motor skills and regression after the age of 3 years.

Two patients had mitochondrial DNA sequence variants of probable pathogenicity. Another four had mitochondrial DNA sequence variants of unclear pathogenicity. It is of note that most “whole genome” studies do not include investigation of the mitochondrial genome. However, there are many nuclear genes that code for mitochondrial proteins. Of particular interest is the replicated association at the mitochondrial aspartate/glutamate transporter, SLC25A12, in ASD. Variation at this gene has not been directly connected to mitochondrial disorders.

Many of the clinical features of the patients reported in this article are consistent with syndromal autism associated with chromosomal disorders. Patients with a duplication of chromosome 15q11–q13 of maternal origin have been found to have mitochondrial dysfunction [Filipek et al., 2003]. It will be useful to combine examination of mitochondrial function with advanced chromosomal analysis capable of detecting copy number variation that may be related to mitochondrial disorders.

This article assists in our understanding of clinical features associated with mitochondrial disorders in ASD, and in description of the mitochondrial findings. The article is not able to answer population-level questions, such as whether the patients with biopsy confirmed mitochondrial disorders are representative of the larger group of patients with ASD with mitochondrial dysfunction. Ascertainment bias of patients with ASD and associated clinical features of mitochondrial disorders seems difficult to circumvent, given the need for relatively invasive diagnostic confirmation by muscle biopsy.

▪ Restricted and Repetitive Behavior Factor Structure: Circumscribed Interests [Lam, Bodfish, & Piven,2008]

The Autism Diagnostic Interview (ADI-R) and Autism Diagnostic Observation Schedule have advanced reliability of diagnosis across research teams. The initial purpose of these instruments was categorical diagnosis of autism, not to identify clinical heterogeneity in ASD. However, extensive data have been collected using these instruments on existing patient samples and these are instruments for which item level reliability has been a hallmark of research administration. Therefore, it is not surprising that the factor structure of the ADI-R is of considerable interest.

In a previous literature review, an article discussing the factor structure of the full ADI-R was reviewed [Boomsma et al., 2008]. The article by Lam et al., reviewed here, is one of a series of articles in which the focus is upon the restricted and repetitive behavior domain of the ADI-R. The authors used a sample independent of previous samples. A mixture of individuals with ASD from simplex and multiplex families was included.

The authors confirmed the previous factors of repetitive motor behaviors and insistence on sameness. Repetitive motor behaviors were found to be negatively correlated with chronological age and verbal IQ, as in previous studies. Both repetitive motor behaviors and insistence on sameness were weakly positively correlated with social and verbal communication domain scores.

A novel finding was the identification of a third factor, circumscribed interests. This factor did not have significant correlation with chronological age, verbal IQ, social domain or communication domain scores. None of the factors were correlated with each other. The circumscribed interests factor included the items, circumscribed interests, unusual preoccupations and unusual attachments to objects. The authors state that the factor may be unique to autism, but acknowledge this hypothesis will require careful study in other disorders. For example, there may be overlap in the circumscribed interests factor in ASD with the recently highlighted hoarding factor in obsessive–compulsive disorder.

The authors used the multiplex sample to determine that there was evidence of familiality of the insistence on sameness and circumscribed interests factors, but not the repetitive motor behaviors factor. The authors initially performed the factor analysis with ADI current scores, but found a similar solution with ADI ever scores, providing some indication of temporal stability of the factor structure, although longitudinal samples would be necessary to confirm stability over time. Most groups have established item level reliability only if an ADI-R severity code of 3 is downcoded to 2 on each item (as occurs in the algorithm). The authors appear to have used 3 scores that were not downcoded. This approach may provide a more full range of item scores, but if the scores are not reliable, more noise than signal may be introduced. In the same way that the authors had a primary and secondary analysis of current and ever scores, it may be productive to reanalyze with a more reliable downcoded structure. This is less a critique of the current study and more a suggestion for further investigation in the area. The authors provided an excellent summary of work to be done in the area, including work looking at the ADI-R in relationship to the Repetitive Behaviors Scale—Revised.