Objective measures of sleep in autism meta-analysed

"Children with ASD [autism spectrum disorder] have small but measurable objective differences in their sleep parameters that are consistent with subjective reporting."

That was the main conclusion reached in the meta-analysis from Marilisa Elrod and Bradley Hood [1] who looked at the collected peer-reviewed data "that used objective measures such as actigraphy or polysomnography (PSG) to describe the sleep parameters of TST [total sleep time], SL [sleep latency], and SE [sleep efficiency] in children with ASD compared with children with TD [typical development]."

Just in case you didn't want to click on the links to some of those explanations, actigraphy is all about measuring movements and so useful in analysing the 'rest/activity' cycle; PSG is a slightly move comprehensive way of measuring various biophysical parameters during sleep; sleep latency is basically the amount of time it takes for a person to fall asleep, and sleep efficiency refers to how much sleep a person gets from time of lying down to getting up from bed (usually expressed as a percentage).

Ten studies met the authors' inclusion criteria covering nearly 350 children diagnosed with an ASD and 221 asymptomatic controls. Researchers reported that children with autism spent on average half an hour less per day TST (total sleep time) and took about 10 minutes longer to fall asleep than controls. Sleep efficiency was also marginally reduced for the autism group. As one might have expected, there was some "notable heterogeneity" across the various study results. That and the fact that comorbidity (if I can still call it that) such as intellectual (learning) disability also seemed to play some hand in the results obtained: "those with ASD and intellectual disability (ID) had a significant decrease in TST as compared with TD peers."

There are a few points to make about this research. First is the continuing idea that 'subjective reporting' when it comes to autism is actually not a bad indicator of what might be going on or have been going on. We've seen evidence of this quite a few times now in the peer-reviewed literature: maternal recall vs. medical records (see here), the reported presence of gastrointestinal (GI) symptoms (see here), first concerns about autism (see here) and now possibly with sleep in mind. Obviously this doesn't mean that every single person questioned about a topic area of autism is going to be providing an authoritative history allowing research to 'do away' with more objective measures. But it does mean that parent or caregiver report might be a very good place to start.

Next, allowing for the text "small but measurable", Elrod & Hood further add to the quite voluminous literature indicating that sleep issues can be a real point of contention for quite a few people on the autism spectrum. I've talked about sleep a few times on this blog (see here and see here) including the idea that melatonin might be a medication of consideration for some [2]. More recently I've also discussed the idea that other factors might play some role in sleep issues in cases of autism as per some research on pain predicting sleeping problems (see here) and/or behavioural sleep intervention being indicated (see here) (particularly when certain comorbidity might be present). Whatever the reasons/intervention suggested, sleep or rather a lack of sleep (quality sleep), is probably not a great thing for anyone.

And with that, some music and a song about a sidewinder sleeping...

----------

[1] Elrod MG. & Hood BS. Sleep Differences Among Children With Autism Spectrum Disorders and Typically Developing Peers: A Meta-Analysis. J Dev Behav Pediatr. 2015 Feb 18.

[2] Rossignol DA. & Frye RE. Melatonin in autism spectrum disorders. Curr Clin Pharmacol. 2014;9(4):326-34.

----------

Elrod, M., & Hood, B. (2015). Sleep Differences Among Children With Autism Spectrum Disorders and Typically Developing Peers Journal of Developmental & Behavioral Pediatrics DOI: 10.1097/DBP.0000000000000140

The label of autism rarely exists in a diagnostic vacuum

"Most young ASD [autism spectrum disorder] children met the criteria for additional psychopathology." That was the primary conclusion reported by Fernando Salazar and colleagues [1].

At the risk of sounding like a broken record going on and on about how the diagnosis/label of autism very rarely exists in a diagnostic vacuum when it comes to comorbidity, I did think it important that the findings of Salazar et al were [briefly] brought to your attention. I've talked a few times on this blog about the ESSENCE around autism (see here) and autism plus (see here) as putting some flesh on the scientific bones that the presentation of autism is often only one part of behaviour in those diagnosed as being on the autism spectrum. That and the plurality of autism...

"Most common diagnoses were: generalized anxiety disorder (66.5 %), specific phobias (52.7 %) and attention deficit hyperactivity disorder (59.1 %)." The sorts of comorbid diagnoses listed most frequently by Salazar and colleagues are no strangers to autism comorbidity research as per some of my other musings on anxiety (see here) and ADHD (attention deficit hyperactivity disorder) (see here).

Anxiety in particular, has been something in receipt of quite a bit of research attention, whether it be through discussions on possible reasons why (see here) or associations made with more physical features also known to be over-represented following a diagnosis of autism (see here). I don't doubt that the hows and whys connecting autism and anxiety are likely to be complicated and quite individual. Indeed, from Salazar et al: "Higher IQ was associated with anxiety disorders" which kinda taps into some other work on the effect of insight on the presentation of anxiety (see here) for example.

There's little more for me to say about the Salazar paper aside from pointing out the impressive authorship roll call included on the paper, including those who've also looked at more somatic comorbidity alongside the diagnosis of autism (see here) and horror of horrors, gut barrier issues appearing in some children on the spectrum (see here). Oh, and that comorbidity occurring alongside a diagnosis of autism doesn't always have to be psychological/behavioural is an important point to raise too...

Music to close: Golden Brown by The Stranglers.

----------

[1] Salazar F. et al. Co-occurring Psychiatric Disorders in Preschool and Elementary School-Aged Children with Autism Spectrum Disorder. J Autism Dev Disord. 2015. March 4.

----------

Salazar, F., Baird, G., Chandler, S., Tseng, E., O’sullivan, T., Howlin, P., Pickles, A., & Simonoff, E. (2015). Co-occurring Psychiatric Disorders in Preschool and Elementary School-Aged Children with Autism Spectrum Disorder Journal of Autism and Developmental Disorders DOI: 10.1007/s10803-015-2361-5

Psychotic symptoms managed by a gluten-free diet?

The case report detailed by William Eaton and colleagues [1] illustrating how a gluten-free diet might not just be the treatment of choice for the autoimmune condition coeliac (celiac) disease is served up for your consumption today.

The authors report the story of 'Chris' an 8-year old boy who experienced various symptoms including "intermittent auditory and visual hallucinations" then moving later in his life to being hospitalised and eventually diagnosed with "major depressive disorder with psychotic features". Accompanying his behavioural presentation, we learn that Chris had a blood test showing "the presence of antinuclear antibodies (ANAs)" and later IgE antigluten antibodies. ANAs are generally reported in the context of autoimmunity and specifically related to conditions such as systemic lupus erythematosus (SLE). Chris however "had no symptoms of any autoimmune disease". IgE antibodies to gluten is reflective of an allergy / type 1 hypersensitivity to gluten. This is probably also why gluten was removed from his diet at the behest of his mother who just happened to be a dietitian.

"After the dietary change, the intensity of Chris's auditory hallucinations declined dramatically and the violent element diminished". It was also after this point that he was discharged from hospital. The authors discuss how over the course of 2 years on a gluten-free diet, Chris's symptoms continued to abate, further suggesting some relationship between gluten and psychiatry. This change also had an impact on the prescription of antipsychotic medication and his eventual withdrawal from the use of risperidone.

There is also another twist to this tale. Those ANAs which has previously been reported eventually led him to receiving a diagnosis of "autoimmune inner ear disease", something that had been kinda seen in one of his extended family members. Indeed, some further inspection of his medical family tree suggested that both autoimmune disease and psychiatric conditions were not an uncommon finding.

As per other case studies talked about on this blog, one has to be slightly careful in how the experiences of Chris are discussed so as not to over-generalise. I am not for example, saying that every case of major depressive disorder with psychotic features is a 'diet thing' nor that autoimmunity might 'correlate' with such behavioural features despite other suggestions (see here). That being said, the sorts of testing undertaken on/for Chris - conducting an MRI scan and looking at EEGs onwards to looking for some of the potential tell-tale signatures of issues with gluten (including coeliac disease) - might be something that clinicians charged with the care of people presenting like Chris would like to emulate. I would definitely like to see more research in this area too (see here).

This is not however the first time that the potential extra-intestinal effects of gluten have discussed with psychiatry and behaviour in mind on this blog. I'll draw your attention to the entry talking about another case report highlighting how gluten consumption might have had something to do with a case of OCD and tics (see here). This follows other work venturing into the realms of schizophrenia (see here) and even cases of autism (see here).

With the introduction of schizophrenia into discussions, I also note an interesting addition to the authorship of the Eaton paper (who himself has some research form in this area) with the name 'F. Curtis Dohan Jr'. Those of you who have heard about the late Dohan Sr (his father) and his insightful work looking at a possible relationship between gluten and psychiatry (see here) might know that Dr Dohan Jr has kinda followed in his father's footsteps. I actually had the good fortune to email converse with Dr Dohan Jr and briefly discuss the family research legacy a few years back. Let's just say that after that conversation, I remained an avid fan of this research topic.

Just before I let you go, outside of the idea that receipt of a psychiatric / behavioural label should be a starting point to further inquiry not an endpoint (where have I heard that before??) I do think we need to be asking quite a few more questions about why there might be a link between behaviour, immune function and food. The opinion piece from Moises Velasquez-Manoff (see here) covers quite a bit of the work done in this area including how coeliac disease might not always be confined to a gastrointestinal (GI) presentation. "Rather than celiac disease driving autoimmune brain problems, he thinks, distinct autoimmune diseases are likely to cluster in the same individual. Avoiding gluten won’t entirely mend these patients" is one of the quotes used, which nicely brings me back to the another 'old friend' of this blog, gut hyperpermeability (leaky gut) as a potentially important bridge and an important part of the triad that includes gut bacteria and mucosal immunity. Oh, and the idea that outside of coeliac disease, there might be a whole range of [non-coeliac] gluten-related conditions to consider...

To close: Pavement and Cut Your Hair.

----------

[1] Eaton WW. et al. Improvement in Psychotic Symptoms After a Gluten-Free Diet in a Boy With Complex Autoimmune Illness. Am J Psychiatry. 2015; 172: 219-221.

----------

Eaton, W., Chen, L., Dohan, F., Kelly, D., & Cascella, N. (2015). Improvement in Psychotic Symptoms After a Gluten-Free Diet in a Boy With Complex Autoimmune Illness American Journal of Psychiatry, 172 (3), 219-221 DOI: 10.1176/appi.ajp.2014.14040550

Decreased plasma levels of lipoxin A4 in autism

The paper by Chun-Lin Yan and colleagues [1] talking about significantly lower plasma levels of lipoxin A4 (LXA4) "a mediator involved in the resolution of inflammation" in cases of childhood autism is the point of discussions today.

Continuing an important theme of immune system involvement in at least some cases of autism, Yan et al focused on a less well-trodden path looking at lipoxins that seem to be involved in something of a yin and yang relationship with another set of eicosanoids (signalling molecules formed through oxidation of certain fatty acids), the leukotrienes. If leukotrienes are thought of as the promoters of inflammation, the lipoxins might be seen as promoting the resolution of inflammation. That all being said, don't be fooled in to thinking that their relationship is anything but complex.

So:

• Seventy-five children - "confirmed ASD [autism spectrum disorder] cases" - were included for study alongside 75 age- and sex-matched asymptomatic controls. 
• Plasma levels of LXA4 were analysed alongside the severity of autism via the use of the Childhood Autism Rating Scale (CARS).
• Results: mean levels of LXA4 were significantly lower in the group with autism compared to controls. Further, something of a relationship was found between CARS scores and plasma levels of LXA4.
• As per other autism researchers (see here), ROC analysis was also undertaken in an attempt to see if LXA4 *might* be "an indicator for an auxiliary diagnosis of ASD". A specific cut-off point for LXA4 levels of 81.5 pg/ml was thought to be potentially predictive (albeit with specificity stats which aren't great, 76%). 
• The authors conclude that with more to do: "autistic children had lower plasma LXA4 levels, suggesting an increased susceptibility to recurring inflammation in these samples."

Before heading a little further into these findings, I note that China is starting to come around to autism research in quite a big way in recent years. My recent coverage of the paper by Zhang and colleagues [2] talking about thioredoxin levels in children with autism (see here) is testament to that, including utilising the formula of potential biomarker analysis + CARS + ROC analysis. I'm assuming that Yan et al had either read the Zhang paper or had contact with the authors given that some of their terminology is pretty similar (such as the term 'auxiliary diagnosis' which I'm yet to understand thoroughly). Similar terminology was also used in other papers too [3] (see here).

The lower mean levels of LXA4 reported by authors, potentially describing a reduction in the ability to put out the 'inflammatory fire', are interesting. I've talked a few times on the blog on how looking at inflammation and autism is all-well-and-good, but quite a bit more focus is needed on the counter-balance processes in place which keep inflammation in check (see here). It seem that the Yan paper fits this bill. Insofar as other work on LXA4 with autism in mind, well, a quick PubMed search only reveals one other paper by Das [4] which seems to be more of a review/opinion paper over the currency of hard data.

This is an interesting area of autism research that really needs quite a bit more investigation; not least replicating the Yan findings in other geographical populations. I hold back from saying anything further about the specificity of these findings to just autism given that other work looking at severe asthma, for example, has reported reduced levels of LXA4 too [5]. This is perhaps all the more interesting in light of some researchers making potential connections between autism and asthma (see here) which might also need to be included in any future schedules of work.

Music. Waiting Room by Fugazi.

----------

[1] Yan CL. et al. Decreased plasma levels of lipoxin A4 in children with autism spectrum disorders. Neuroreport. 2015 Feb 24.

[2] Zhang QB. et al. Thioredoxin: a novel, independent diagnosis marker in children with autism. Int J Dev Neurosci. 2015 Feb;40:92-6.

[3] Gong ZL. et al. Serum 25-hydroxyvitamin D levels in Chinese children with autism spectrum disorders. Neuroreport. 2014 Jan 8;25(1):23-7.

[4] Das UN. Autism as a disorder of deficiency of brain-derived neurotrophic factor and altered metabolism of polyunsaturated fatty acids. Nutrition. 2013 Oct;29(10):1175-85.

[5] Celik GE. et al. Lipoxin A4 levels in asthma: relation with disease severity and aspirin sensitivity. Clin Exp Allergy. 2007 Oct;37(10):1494-501.

----------

Yan CL, Zhang J, & Hou Y (2015). Decreased plasma levels of lipoxin A4 in children with autism spectrum disorders. Neuroreport PMID: 25714424

Boiling down ADOS for autism detection (again)

Today I want to direct your attention to the paper by Kosmicki and colleagues [1] (open-access) reporting that the use of "machine learning algorithms" could help "streamline ASD [autism spectrum disorder] risk detection and screening."

Regular readers of this blog might have already cottoned on to the fact that any talk about applying "computational and statistical methods" to autism screening and/or diagnosis can really mean only one person and research group: Dennis Wall from Stanford University. To quote from his institutional website on this area of research, the aim is to "evaluate the degree of redundancy of the ADOS and ADIR and if so determine whether a reduced set of uncorrelated features could correctly classify individuals with the same accuracy as the gold-standard diagnostic tests." ADOS and ADI-R by the way, are some, if not the, gold-standard schedules when it comes to the assessment of autism. The idea is that boiling down these respective schedules might save both time and resources when it comes to identifying those where a diagnosis of ASD is indicated. In case you'd like some history about this line of work, look no further than here...

The latest paper from the Wall group continues the research journey looking this time at modules 2 and 3 of the ADOS where previous work looked at module 1 (see here). In case you're not familiar with the concept of modules in ADOS, it's all about selecting the correct module according to verbal fluency (see here) where module 1 is for those who have very little or inconsistent phrase speech and modules 2 and 3 represent increasing phrase speech with also a little more focus on the use of age-appropriate props.

The results? Based on the development of 'classifiers' for each module, several machine learning algorithms were developed and tested (see here). One of the algorithms, ADTree, is by the way, the same classifier used in the previous module 1 ADOS work [2]. But ADTree did not perform best on this occasion: "The logistic regression classifier based on analysis of archival records from ADOS module 2 consisted of nine items, 67.86% fewer than the complete ADOS module 2, and performed with 98.81% sensitivity and 89.39% specificity in independent testing." Further: "The SVM module 3 classifier based on analysis of archived ADOS module 3 records consisted of 12 items, 57.14% fewer than the complete ADOS module 3, and performed with more than 97% sensitivity and specificity in testing."

The authors conclude: "These results support the notion that fewer behaviors when measured using machine learning tools can achieve high levels of accuracy in autism risk prediction."

Anyone who has either professional or personal experience of undertaking an ADOS will know that this is a highly specialised assessment schedule which often requires some time to complete. It's nothing like as time-consuming as the ADI but still, significant efforts and resources are needed to carry out the assessment and do so with skill and reliability (and maintain those all-important reliability stats). Wall et al have really started to shake the establishment when it comes to ADOS (and ADI) when asking just how much of the schedule is really needed to assess for autism/ASD. This on top of their other work talking about assessment 'triage' via YouTube videos using, horror of horrors, non-clinical raters (see here). I'm not saying that these approaches are ready for clinical practice; quite a bit more replicative work is required [3] including crossing geographical boundaries. But something like the idea that "mobile health approaches that ultimately enable individuals to receive more expedient care than is possible under the current paradigms" is a tantalising prospect.

So: Start by The Jam.

----------

[1] Kosmicki JA. et al. Searching for a minimal set of behaviors for autism detection through feature selection-based machine learning. Transl Psychiatry. 2015 Feb 24;5:e514.

[2] Wall DP. et al. Use of machine learning to shorten observation-based screening and diagnosis of autism. Transl Psychiatry. 2012 Apr 10;2:e100.

[3] Bone D. et al. Applying Machine Learning to Facilitate Autism Diagnostics: Pitfalls and Promises. J Autism Dev Disord. 2014 Oct 8.

----------

Kosmicki JA, Sochat V, Duda M, & Wall DP (2015). Searching for a minimal set of behaviors for autism detection through feature selection-based machine learning. Translational psychiatry, 5 PMID: 25710120

Individualised medicine and autism: a brave new world

Pharmacogenetics: "the study of inherited genetic differences in drug metabolic pathways which can affect individual responses to drugs, both in terms of therapeutic effect as well as adverse effects."

Having recently watched a rather interesting documentary on the BBC titled 'Can you cure my cancer?' illustrating how the era of personalised medicine is here and now (see here) in at least one aspect of medicine, I was really quite interested in the science of how our genes might affect our response (or not) to things like medication. Pharmacogenetics is not necessarily a new concept (see here) but only in recent years with the increasingly inexpensive ways and means that the genome can be mapped, have we seen something of a translation of the idea from laboratory to bedside on a broader scale.

Now it seems that autism research and practice might also be benefiting from this brave new world as per the paper from Teri Smith and colleagues [1] (open-access available here). This is a case report of an adolescent diagnosed with an autism spectrum disorder (ASD) among other things, who presented with several genetic 'issues' that may well have impacted on the metabolism of certain types of medicines used to manage specific symptoms. The genetic issues identified included some pertinent to the workings of cytochrome p450 enzymes which may have had important effects on the metabolism of certain pharmaceutics. Interestingly too was the finding that this young man also "showed a MTHFR C/T gene variation that suggested reduced enzymatic activity associated with a reduced conversion of folic acid to methylfolate." MTHFR controlling the production of methylenetetrahydrofolate reductase has something of a history with autism in mind (see here) as it might have in a few other diagnoses too (see here).

"It is clear in this clinical case that certain recommendations for care could be made due to
pharmacogenetic findings." That was one of the concluding sentiments from the authors, alongside how said genomic testing was all done via a spit sample (so being a pretty non-invasive method of collecting DNA).

I'm impressed with this paper and the precedent it may well set. As per my other musings on the use of pharmaceutics when it comes to autism (see here for example) I would firmly place myself in the 'buyer beware' category when it comes to medicating in cases of autism but do understand that there is a place for some pharmaceutics to tackle certain issues associated with autism under certain circumstances. That is, alongside good medicines management. If the science of pharmacogenetics can aid the process of medicating when it comes to the label of autism, I'm sure most people would be happy to see it as an addition particularly when talking about the more plural 'autisms'.

If I had to quibble at all with the idea of pharmacogenetics it might however be to say two things:

• (i) the structural genome represents only one part of what we call gene function. As per the rise and rise of the science of epigenomics (see the Nature special on this), it is becoming more and more apparent that there are lots of issues potentially impacting on the how the genome 'works' not necessarily just tied down to structural mutation(s). Looking for genetic mutations is all well and good but might not necessarily give you all the data on specific gene functions. And speaking of gene expression, I'll draw your attention to the paper from Carolyn Ch'ng and colleagues [2] noting that following their meta-analysis of gene expression in cases of autism: "A subset of the highly ranked genes is suggestive of effects on mitochondrial function." Mitochondrial function and autism eh?
• (ii) The name Jeremy Nicholson has appeared a few times on this blog (see here for example) based on some research looking at potential biomarkers for autism [3]. Aside from such research forays, Prof. Nicholson can perhaps also be credited with "the principle of pharmacometabonomics" [4], that is that those trillions of wee beasties which inhabit our gastrointestinal (GI) tract - the gut microbiome - might also have the ability to influence our response to certain types of medicines. I've talked about this issue elsewhere (see here). That and the fact that certain GI issues might also potentially affect things like drug absorption (see here) and the recipe starts to get a little more complicated.

Still, I'll keep my eye open for more on this topic, and also when some brave soul decides that genomic / epigenomic / microbiomic analysis might even help with gauging response to other interventions put forward for some on the autism spectrum...

To close: The Ballad of Bilbo Baggins by the great, late Leonard Nimoy.

----------

[1] Smith T. et al. Pharmacogenetics Informed Decision Making in Adolescent Psychiatric Treatment: A Clinical Case Report. Int J Mol Sci. 2015 Feb 20;16(3):4416-4428.

[2] Ch'ng C. et al. Meta-Analysis of Gene Expression in Autism Spectrum Disorder. Autism Res. 2015; Feb 26.

[4] Yap IK. et al. Urinary metabolic phenotyping differentiates children with autism from their unaffected siblings and age-matched controls. J Proteome Res. 2010 Jun 4;9(6):2996-3004.

[5] Clayton TA. et al. Pharmacometabonomic identification of a significant host-microbiome metabolic interaction affecting human drug metabolism. Proc Natl Acad Sci U S A. 2009 Aug 25;106(34):14728-33.

----------

Smith T, Sharp S, Manzardo AM, & Butler MG (2015). Pharmacogenetics Informed Decision Making in Adolescent Psychiatric Treatment: A Clinical Case Report. International journal of molecular sciences, 16 (3), 4416-4428 PMID: 25710722