Metal detecting and autism

As many teenagers do, I went through a period during young adulthood when my whole world seemed to revolve around music, and in particular the various 'bad boys' of heavy metal. We are talking Iron Maiden, Guns 'n Roses and a few bands from the wayback like the Zepplin. I didn't quite grow a mullet and never really had the physique for spandex (I had a part-time job at a well-known supermarket which perhaps would not have taken too well to such a fashion sense) but I was there in spirit. Heavy metal rocked (sort of). Today such memories are quite distant. Aside from catching the odd tune on the car radio, the closest I seem to get to heavy metal is reading papers about metals in relation to autism and lots of other things. Ah yes, my youth is lost.

I digress. I talked about iron and autism a few days back. A role for metals, particularly the heavy variety, has been quite a consistent feature of autism research and discussion over the years. The topic has been approached from lots of different angles and once again, is related to quite a lot of often vigorous debate.

The angle I specifically want to take in this post relates to a recently published piece of research on the analysis of metals in hair samples taken from children with autism. There are a few other areas that this leads into, which I may or may not pursue further in this post.

The paper in question reported the results of an analysis based on ICP Mass Spectrometry comparing samples from children with autism against 'unbalanced' controls. ICP-MS (Inductively coupled plasma - mass spectrometry) is a rather handy method for looking for metals. Without giving a lecture on MS, the ICP part of the method is a way of ionising a sample sufficiently to conduct analysis. It is the method of choice for metal analysis and tends to be very accurate down to quite low concentrations of compound (parts per billion in some cases).

Anyway, the authors primarily reported higher concentrations of compounds such as lithium, selenium and molybdenum from the hair samples from children with autism. Looking at their results, they also found several other interesting things such as low levels of iron (Fe) in the children with autism vs. controls (p=0.036) and higher levels of mercury (Hg) (p=0.042) all from hair analysis (these weren't considered 'significant' given that the study probability value, p, was set at <0.01 = 1%).

They also suggested that the use of chelation and other vitamin/mineral supplementation did not seem to significantly affect the amounts of metals detected in hair (although median values were lower in the supplemented group for aluminium, cadmium, copper, molybdenum, lead and selenium and higher for zinc).
I think there are a few details related to these findings that perhaps require further comment.

Hair analysis is a bit like looking at rock formations and determining previous weather patterns. It can provide a retrospective look at various things related to health over a period of time rather than just being 'here and now' snapshot like blood or urine. It is dependent on small details such as a person having some hair and also, having hair which is longer than a few days growth to provide such information (tip: if you carry out random drug analysis using hair and one of your employees decides to shave off all his hair and eyebrows, ask a few questions).

Hair analysis for metals/non-metals has been reported in autism previously using similar methods. Without trying to cherry-pick from the results so far:

• this paper for example, reported an opposing result in relation to lithium in hair from children with autism.
• This paper by contrast reported higher levels of lead, mercury and uranium in autism - bearing in mind that it was performed on children with autism in Kuwait and their fairly recent history of warfare associated with that region.
• Finally this study from India reported raised copper, lead and mercury levels in autism alongside lower magnesium and selenium.

There are other studies, but I have picked these out to show the 'heterogeneity' of results obtained and suggest how local factors related to where people live, what they eat and drink and factors contributory to environmental 'exposure events' might influence the results obtained.

Of course whilst detecting (or not detecting) such metals is an interesting finding, without any accompanying functional results (from blood, plasma or tissues) the results don't say very much. An entirely different post is perhaps required to examine the potential relationship between (heavy) metals and the presentation of autism.

Rock on!