Thanks to a commenter for pointing me to a newish study (open access): Infantile zinc deficiency: Association with autism spectrum disorders.
I see Paul Whiteley already covered it here. If you are interested in the autism spectrum disorders, I very highly recommend his blog. Paul is very sober and clear-sighted about the limitations of the research information but has an open mind about what might be clues to the pathophysiology of these disorders. Since it is a long paper, and Paul and I have quite different styles, I think there is plenty of room for my own thoughts.
Atlas Genius: Trojans (right click to open in new tab)
In this study, nearly 2000 children with autistic disorders had hair tested for zinc deficiency. 584 (29.6%) subjects had levels lower than two standard deviations below the mean of the reference range. 43.5% of the male children ages 0-3 and 52.5% of the female children 0-3 in the sample were zinc deficient. Older children with autism in the study had a lower incidence of zinc deficiency, which continued to decrease with the age of the subject sample, to the point where autistic children over the age of 10 had normal zinc levels. Normal hair zinc in healthy children appears to be around 130 ppm, and one autistic 2 year old in the study had a level of 10.7 ppm.
However you slice it (and let's take into account the limitations of a strictly observational study), that is an impressive finding. The authors speculate that epigenetics could be a factor in the pathophysiology of autism. Epigenetics is the alteration of gene expression by environmental influences. Mineral deficiencies could certainly alter gene expression and perhaps cause an epigenetic disorder. The other side of the argument is that active inflammation and stress causes us to waste zinc (and other minerals) as part of the inflammatory process. Since we know autistic spectrum disorders are related to inflammation, it would make perfect sense that babies and toddlers with autism would have low zinc levels. Low serum zinc also seems to be associated with heavy metal toxicity, which is also associated with autistic spectrum disorders (more often cadmium and aluminum, according to the study authors, than mercury).
But measuring a mineral deficiency is more tricky than it might appear. Zinc can be sequestered in the bones or liver and bound up by protein in the serum. Zinc levels can vary considerably based on infection, stress, low protein levels in the blood, and even time of day. Hair levels could perhaps be considered a more reliable indicator of full-body zinc stores, so may be a more useful measure for these sorts of studies.
Zinc deficiency could very plausibly be part of what causes the symptoms of autism. Zinc plays an important role in protein synthesis, cell growth and repair, and levels need to be super topped off in pregnant women and infants where all these processes are occurring at a more rapid rate than any other time in one's life. Also, we know that a "leaky gut" has been associated with autism in a rather well-designed study, and with leaky guts there is malabsorption of nutrients and minerals. Many children with autism have signs from birth (when everyone has a leaky gut, up through 6 months of age when there is no argument that exclusive breastmilk is best, period), but some children seem to develop normally and do not exhibit symptoms until later. Could leaky guts leading to toxin exposure, and micronutrient deficiency have something to do with these cases? Seems a plausible theory.
In addition, in the setting of zinc deficiency, the intestinal zinc importer (the awesomely named Zip4) is upregulated, sensibly enough. Increasing Zip4 can increase the absorption of toxic heavy metals such as cadmium and lead. Therefore an early zinc deficiency could plausibly lead to more absorption of toxic metals. (Creepy fact from the article: soy-based infant formulas have 6 times higher cadmium levels than cow's milk formulas, and cereal-based formulas 4-21 times higher levels (1).)
I don't think there is any argument that we all need sufficient zinc, particularly pregnant women and babies. As far as I know, there are no controlled studies of zinc supplementation in pregnant women and babies and autism risk, so we really can't make any conclusion or recommend any sort of increased zinc supplementation in babies at this point. There is more data in ADHD and I should probably pull those papers myself (some of them are too old to pull easily) and do a post on it. Super-high zinc intake has killed older people who used zinc-based denture cream by causing copper deficiency and heart arrhythmia, and babies and little children (like the elderly) will be more vulnerable to these sorts of insults.
More data please! Researchers, rescue us from the rocky shoals of the unknown.
(Hmmm. My favorite diaper cream is zinc-based. I wonder how much is absorbed and if that has any bearing on zinc levels, zinc deficiency, etc.)