Bone health, casein-free diet and autism

The Handy Man can @ Paul Whiteley

Bones, aside from being the nickname of the good Doctor McCoy (whose name has been mentioned in a Lancet article very recently), have already featured on this blog and here we are again with a rather long entry it has to be said.

Of all the studies so far done which have included some element of autism and the gluten- and casein-free diet, one has stuck in my mind in particular. The study in question was by Hediger and colleagues* and looked at bone cortical thickness in boys with an autism spectrum condition. They reported that bone thickness seemed to be reduced in cases of autism compared to reference values and this reduction was particularly marked for those on a casein-free diet.

Casein-free diet = removal of mammalian dairy produce from the diet = important source of calcium.

As someone who has put quite a bit of time into looking at the potential efficacy (and safety) of restriction diets for some cases of autism, these were quite worrying results given the potential long-term implications of following a casein-free diet and balancing potential gains in behavioural outcomes with potential losses to long-term physical health.

At the time I remember talking to quite a few people in the know about autism, calcium and bone health and where this could all eventually lead. Some said that the study whilst interesting needed a lot more follow-up in terms of participant numbers, suitable control groups (rather than reliance on population reference ranges), looking at bone mineral density not just thickness and the various other factors the authors themselves listed as potential confounders such as the presence of gastrointestinal (GI) disorders, vitamin D levels and limited physical activity. Others said that outside of a known food allergy, this study was evidence that dietary changes should not be made for people with autism and to close the book.

A few subsequent papers have attracted my attention since. This paper by Mouridsen and colleagues** reports on the numbers of fractures reported in autism compared with not-autism. I apologise that I am only able to provide the paper abstract (having myself taken advantage of the Springer open-access window last year to get the full-text). The authors specifically set out to look at fractures on the basis of the Hediger findings, looking at adults diagnosed with autism as children (n=118) compared to matched controls (n=336) to see if the risk of fractures was higher in autism potentially as a result of the low bone thickness described. The results were interesting in that the autism group actually presented with a lower level of fracture than the comparison group (11.9% vs. 24.7%). There were some differences in the positioning of fractures between the groups where participants with autism were at greater likelihood of fracturing their forearm than controls (5.9% vs. 3.2%) but less likely to fracture their hand or wrist (2.5% vs. 8.0% p=0.05). This last result on the hand and wrist was particularly relevant to the Hediger findings given that they relied on measurements from the "second metacarpal on the hand-wrist radiographs" (one of the bones linking your finger to your hand).

Whilst it was slightly comforting to read these results, Mouridsen et al make some interesting comments about other factors potentially affecting their results including the more protective environment that participants with autism were likely to live in, less alcohol abuse, lower rates of sporting activities and tool use, etc. They also point out that recording hospital visits because of fracture might not necessarily mean fewer fractures; other less severe fractures might have occurred but not been known about or recorded. My mind also wanders back to issues like pain perception and autism which in some people has been suggested to be slightly different.

A second paper by Mary Hediger and colleagues (open-access)*** also caught my eye. Interestingly Hediger et al were following up one of their confounding variables, a role for vitamin D, to potentially account for some of their earlier results. The results were mixed. On the one hand , no significant group differences were noted in circulating vitamin D levels as reported through plasma levels of 25(OH)D between children with autism (on and off a casein-free diet) and a comparison group. On the other hand, 54/89 (61%) of the total children involved in the study were vitamin D deficient (below the AAP 20ng/ml threshold). Whether or not the fact that control participants were having tonsillectomies at the time of sampling affected the results, I don't know but there does seem to be some suggestion that recurrent tonsillitis might be linked to lower vitamin D levels anyway.

Whilst not wishing to belittle the original findings from Hediger and colleagues on bone thickness and a potential effect from a casein-free diet, I do wonder if more authoritative work is required in this area before too many conclusions are drawn. To some extent this paper by Goodarzi and Hemayattalab**** has tried to cover the calcium / physical activity aspects but there still remains much to do. Not so long ago I discussed vitamin D and gut barrier integrity and the various inferences that "could" be made from this research. Intestinal conditions such as coeliac (celiac) disease have long been linked to bone health including some interesting data on bone thickness. More recently studies have suggested bone health can be [positively] affected by use of a gluten-free diet in cases of coeliac disease.  I'm not necessarily saying that these same conditions are relevant to autism but certainly one does wonder about those cases where bowel issues and hyperpermeability of the gut have been reported as comorbid.

Now.. if I said the word Chumbawumba would you perhaps partake in a bit of tub-thumping with me?

* Hediger M. et al. Reduced bone cortical thickness in boys with autism or autism spectrum disorder. JADD. 38: 2008

** Mouridsen SE. et al. Fractures in individuals with and without a history of infantile autism. A Danish register study based on hospital discharge diagnoses. JADD. 2011

*** Molloy C. et al. Plasma 25(OH)D concentration in children with autism spectrum disorder. Dev Med Child Neurol. 52: 2010

**** Goodarzi M. & Hemayattalab R. Bone mineral density accrual in students with autismspectrum disorders: Effects of calcium intake and physical training. Res Autism Spectrum Disorders. November 2011.