Monday, 22 July 2013

Epilepsy in autism revisited

I've said it before and I'll say it again, a diagnosis of autism or autism spectrum disorder is seemingly protective of nothing when it comes to comorbidity. Indeed, in these days of common ground and overlap (think ESSENCE) one of the more positive changes in thinking about autism is the move away from viewing the condition as some sort of homogenous, stand-alone diagnosis ripe for over-arching generalised theories. It's not.

Stealing a kiss @ Wikipedia  
Of that growing list of comorbidity which can and does sometimes follow autism, one condition or set of conditions in particular, has the ability to severely impact on the health and wellbeing of a person, in some unfortunate cases leading to the most severe outcomes: epilepsy.

I've covered the topic of epilepsy and seizure disorders previously on this blog (see here). The main conclusion from that post was that epilepsy represents one of the more common comorbidities which can present alongside autism and that there is no positive side to being diagnosed with epilepsy in terms of how it can affect quality of life and how it is, in some cases, linked to early mortality.

With these facts in mind, I was therefore interested to read the paper by Emma Viscidi and colleagues* (open-access) which looked at a number of issues around the autism-epilepsy link in terms of the prevalence of epilepsy and some of the factors which might influence the start and presentation of epilepsy in cases of autism.

The paper is open-access so there's no real need for me to get the fine-toothed comb out on it. I will however summarise some key points:

  • Based on participant data from four separate initiatives - AGRE, Simons Simplex collection, Autism Consortium, NSCH 2007 - authors collated data on nearly 6000 mainly young people diagnosed with an autism spectrum condition.
  • There were some differences among the various datasets in terms of how the diagnostic coding of autism was arrived at which needs to be kept in mind with this paper, but on the whole most diagnoses were confirmed by gold-standard assessment tools.
  • In terms of the presence of epilepsy, these were "based on parent report in all of the samples" without independent, professional verification of the diagnosis.
  • Several other facets of presentation were [variably] collected / ascertained from the available datasets including things like cognitive ability, history of developmental regression (see here) and the severity of autism presentation (at least at the time of initial presentation to the various initiatives).
  • Results: well, bearing in mind some variability between the datasets and various demographic slants such as participants being predominantly male and white, the frequency of reported epilepsy occurring in cases of autism ranged from 2.9% - 6.7% and the prevalence (based on the NSCH 2007 data) was 12.5%.
  • When stratifying the data across age groups (6 and below, 7-9, 10-12, 13 and above years) it appeared that older chronological age was associated with an increased frequency of reports of epilepsy; from the NSCH data, upwards of 25% of 13 year olds and older with autism were reported to present with epilepsy.
  • Similarly other recorded factors also seemed to be linked to the appearance of epilepsy including: the presence of developmental regression, those with poorer language skills, lower cognitive ability, etc.
  • The authors conclude that "epilepsy is a common co-morbid condition in individuals with ASD" and that "low IQ is the best clinical predictor of epilepsy in children with ASD".

As I've mentioned, there are some pretty big holes in this research; not least the lack of independent verification when it comes to the presence of epilepsy or not. Actually, I'm going to rephrase that last sentence to say that this is not a 'hole' in the research but rather something that needs to be investigated further in light of my previous posts talking about parents actually being quite good at spotting clinical signs (see here) and other comorbidity (see here) when it comes to autism. A shocker I know.

That aside, there are several potentially important lessons which can be learned from this data, not least when it comes to 'predicting' who with autism, might eventually go on to develop epilepsy and hence where screening and clinical management for this particular comorbidity perhaps needs to be focused in these times of limited resources and austerity. The accompanying paper by Jedrzejczyk-Goral and colleagues** looking at what types of epilepsy might be more commonly associated with autism is also helpful.

The related issue of intellectual disability (ID) in cases of autism seemingly being a magnet for epilepsy is also worthy of greater follow-up. One wonders for example, whether this finding might overlap with the other data looking at things like genetics in relation to autism and how ID seems to shoulder the larger burden when it comes to CNV load for example (see here). Indeed, whether some of these CNVs or other point mutations (SNPs et al) might actually be more relevant to the onset of epilepsy than the presence of autism itself? And then what about that inborn error of the branched-chain amino acids work discussed not so long ago on this blog (see here) where epilepsy and autism was presented?

Finally(!) my thoughts also wander back to the recent chatter about ketogenic diets and autism (see here and here). I'm not for one minute giving anything like medical advice on this issue - don't mess with epilepsy or indeed, antiepileptics it seems (more to follow on this). I'm merely suggesting that when potentially knowing which children/young adults with autism are more likely to develop epilepsy, and given the increasing interest in managing certain types of epilepsy through the ketogenic approach, whether further controlled research might wish to have a closer look at whether there are important links (or not) between these areas.

To close, Radio 2 listeners here in the UK might recognise a version of this song... enjoy yourself (it's later than you think).


* Viscidi EW. et al. Clinical Characteristics of Children with Autism Spectrum Disorder and Co-Occurring Epilepsy. PLoS ONE 2013;  8(7): e67797. doi:10.1371/journal.pone.0067797

** Jedrzejczyk-Goral B. et al. 1925 – The frequency of different types of seizures in children suffered from autism and epilepsy. European Psychiatry. 2013; 28: supplement (Abstracts of the 21th European Congress of Psychiatry).

---------- Viscidi EW (2013). Clinical Characteristics of Children with Autism Spectrum Disorder and Co-Occurring Epilepsy PLoS ONE DOI: 10.1371/journal.pone.0067797