The study Aluminium in brain tissue in autism is problematic from the start. As Orac and The Blood-Brain Barrier Scientist note, Christopher Exley sits on the editorial board of the “Journal of Trace Elements in Medicine and Biology“ where the study was published. The study also had an extremely rapid submission to acceptance time frame which was 26 October for submission, 21 November for a submitted revision, 23 November for acceptance and 26 November for online publication. The sole funding source for this study was the Child Medical Safety Research Institute (CMSRI) which is a Dwoskin family foundation. CMSRI is the funding source for the duo of Shaw and Tomljenovic to the tune of $900,000.00. Clare Dwoskin is very very anti-vaccine which is evidenced by her statement to journalist John Stossel:
What his daughter went through is NOTHING compared to what the families of autistic children go through every day of their lives. No disease can match this record of human devastation. Vaccines are a holocaust of poison on our children’s brains and immune systems. Shame on you all.More on what kind of "research" the Dwoskins finance can be found here and here. The Dwoskins' latest study (prior to this one) with Christopher Shaw as the corresponding author was retracted due to image manipulation and dodgy methods.
While these issues individually don't disqualify the merits of a study, taken together they do cast serious doubt on the scientific rigour, review and impartiality of the study. More importantly, the methods and results are the issues which ultimately render this study completely meaningless.
The thrust of this study is laid out in the abstract and introduction that state:
Human exposure to the environmental toxin aluminium has been linked, if tentatively, to autism spectrum disorder.And
Paediatric vaccines that include an aluminium adjuvant are an indirect measure of infant exposure to aluminium and their burgeoning use has been directly correlated with increasing prevalence of ASD [11].No it hasn't and is based upon some cherry-picked, weak studies. The authors obtained five samples of brain tissue from people with confirmed autism spectrum disorder (ASD) for aluminium measurement. They also obtained ten paraffin-embedded brain tissue samples from ASD individuals for fluorescent microscopy.
There were no controls. Yes you read that correctly. Not a single control for any aspect of this study; I'll elaborate. The authors state:
Total aluminium was measured in each sample by transversely heated graphite furnace atomic absorption spectrometry (TH GFAAS) using matrix-matched standards and an established analytical programme alongside previously validated quality assurance data [13].The "established analytical programme alongside previously validated quality assurance data" is their own protocol which is from these two studies, Brain burdens of aluminum, iron, and copper and their relationships with amyloid-β pathology in 60 human brains and The Identification of Aluminum in Human Brain Tissue Using Lumogallion and Fluorescence Microscopy. I don't think validation means what they think it means. This was internally-validated and not a recognised method for TH GFAAS by having been actually validated independently.
In the recent study two technical things (among many) stand out for the recently published "Aluminium in brain tissue in autism", first there was no mention of blanks used. Granted it was an abbreviated description given the protracted version was explained in "Brain burdens of aluminum, iron and copper and their relationships with amyloid-β pathology in 60 human brains" (House, Exley et al. 2012), however there is no mention of an aluminium blank used to correct for contamination in either the body of the text nor in the description of Table 1 (Mold, Exley et al. 2017) where the values for aluminium content appear. But this does appear in the cited study in both the body of the text and the description for it's Table 3 (House, Exley et al. 2012) aluminium content of brain samples. It's worth mentioning that in the cited study (House, Exley et al. 2012), several samples have negative values and a high degree of intra-sample variation. This is a problem with with their validation and they weakened this further by not including blanks in the autism study (Mold, Exley et al. 2017) as evidenced by the higher degree of intra-sample variation.
The second technical issue was the use of mean in the aluminium autism study. Exley stated in his own cited brain aluminium study that due to non-normal distribution of intra-sample values, median with standard deviation was calculated. This is correct but this was not done for the autism aluminium study, instead they used mean with standard deviation which skewed the means much higher because the outlier values were captured. These values are further meaningless since the standard deviations are higher than the mean values for 3 of the four lobes sampled.
The mean (SD) aluminium content across all 5 individuals for each lobe were 3.82(5.42), 2.30(2.00), 2.79(4.05) and 3.82(5.17) μg/g dry wt. for the occipital, frontal, temporal and parietal lobes respectively.Another huge problem for this study is both lack of controls and their own scale of pathology:
Previous measurements of brain aluminium, including our 60 brain study [15], have allowed us to define loose categories of brain aluminium content beginning with ≤1.00 μg/g dry wt. as pathologically benign (as opposed to ‘normal’). Approximately 40% of tissues (24/59) had an aluminium content considered as pathologically-concerning (≥2.00 μg/g dry wt.) while approximately 67% of these tissues had an aluminium content considered as pathologically- significant (≥3.00 μg/g dry wt.). The brains of all 5 individuals had at least one tissue with a pathologically-significant content of aluminium.They cited The Identification of Aluminum in Human Brain Tissue Using Lumogallion and Fluorescence Microscopy (Mirza, Exley 2016) but it was not developed from samples of 60 brains; it was developed from multiple samples of one brain. They rated the wildly variable observed aluminium levels in the Mold, Exley et al. 2017 study using a scale they developed in Mirza, Exley 2016 based on one single brain donation. The authors must have a lot of contempt for their target audience and benefactors or are woefully inept to try and pass this stratification off as valid. But more importantly, the results are meaningless because they did not include any controls, nor did they include any information on medical and dietary history nor causes of deaths of the individuals whose samples were donated.
The microscopy isn't any better. There is no validation of the stain specificity nor control for contamination. The Blood-Brain Barrier Scientist discusses the problems with the microscopy in detail. Futhermore, they use Photoshop for visualising and enhancing images:
The subsequent merging of fluorescence and bright-field channels was achieved using Photoshop (Adobe Systems Inc. US).This is not standard practice, in fact it's rather amateurish. There are dedicated software packages available for fluorescent microscopes that track image capture and enhancements. There is also no randomisation of microscopy sections nor blinding of the reader. Presumably (or should be) there are several images taken and randomly selected for results discussion. This was not done so for all we know, images are cherry-picked for maximum effect. The fact that the microscopy suddenly had ten different samples (instead of the five previously noted) begs the question of IRB approval not to mention their Table S1 does not appear in the published study.
They make very far-reaching conclusions about their findings but fail to qualify any clinical relevance. No histology was done so actual pathology of tissues samples is lacking. The discussion is loaded with supposition with no evidence to support their claims.
A limitation of our study is the small number of cases that were available to study and the limited availability of tissue. Regarding the latter, having access to only 1g of frozen tissue and just 3 serial sections of fixed tissue per lobe would normally be perceived as a significant limitation. Certainly if we had not identified any significant deposits of aluminium in such a small (the average brain weighs between 1500 and 2000g) sample of brain tissue then such a finding would be equivocal. However, the fact that we found aluminium in every sample of brain tissue, frozen or fixed, does suggest very strongly that individuals with a diagnosis of ASD have extraordinarily high levels of aluminium in their brain tissue and that this aluminium is pre-eminently associated with non-neuronal cells including microglia and other inflammatory monocytes.No the actual limitations of this study are the utter lack of controls, validation and rigour at every step of the way. Non ASD people and with no pathology are going to also have aluminium deposits in their brain tissue. Brain tissue banks are precious resources for neuroscience investigators. It's a gross violation of the scientific method to be wasted on rubbish studies such as this to satisfy a rich matron with a hatred for vaccines.
ETA 12.18.17: Buzzfeed just reported that scientists they spoke with, who have relevant experience have rejected this study due to the problems with the methodology listed here and on other blogs. Additionally, Christopher Exley has a substantial conflict of interest which he failed to report. It turns out he is a benefactor and shill for SilicaWaters.com who sell ACILIS by Spritzer, a product that is touted by Exley to remove aluminium from our bodies. SilicaWaters reports that 10% of their sales is donated to Exley's research into the benefits of drinking silica-rich water.
