Nutritional and Metabolic Status of Children With Autism vs. Neurotypical Children, and the Association With Autism Severity

James B Adams; Tapan Audhya; Sharon McDonough-Means; Robert A Rubin; David Quig; Elizabeth Geis; Eva Gehn; Melissa Loresto; Jessica Mitchell; Sharon Atwood; Suzanne Barnhouse; Wondra Lee


Nutr Metab. 2011;8(41) 

In This Article



The general agreement of the present neurotypical reference ranges of many vitamins, minerals, and primary amino acids with published reference ranges from standard sources provides validation of the methodology used. Children with autism have mean levels of vitamins, minerals, and most primary amino acids that generally lie within published reference ranges. In the care of children with autism, the practitioner is therefore unable to discern emerging metabolic dysfunction or utilize measured values of these standard analytes as the basis for clinical decision-making regarding supplementation in most cases. More reliable and of greater clinical significance are levels of sulfation, SAM, uridine, glutathione, oxidative stress, and ATP/NADH/NADPH which are very likely to be abnormal.

In the sections below we provide a detailed discussion of each of the categories of measurements. However, some of the results are inter-related, so we wish to first discuss some of the major results.

Sulfation and ATP Children with autism had significantly lower levels of plasma sulfate, including both free and total sulfate, consistent with several previous studies. ATP is required in the kidney to resorb sulphate (recycling of sulphate is important because sulphate is poorly absorbed from the gut, and conversion from cysteine is slow). This study found a significant correlation of ATP with free and total plasma sulphate (r = 0.32 and 0.44, respectively), suggesting that decreased ATP is a significant contributor to decreased sulphate levels in children with autism.

Methylation and ATP Children with autism had significantly impaired methylation, as evidenced by low levels of SAM (the primary methyl donor) and high levels of plasma uridine (which requires methylation to be converted to thymidine). SAM is formed from methionine by methionine adenyosyl transferase, which requires ATP. Methionine levels were similar in the autism and neurotypical groups, but ATP levels were very significantly lower in the autism group, suggesting that low levels of ATP are at least part of the reason for decreased levels of SAM. Methionine in the body comes partially from the diet (it is an essential amino acid) and partially by the recycling of homocysteine to methionine (via methionine synthase or methyl transferase). Methionine synthase requires methyl-B12 and 5-methyl-tetrahydrofolate, a derivative of folic acid. Both vitamin B12 and folic acid were similar in the autism and control groups, consistent with normal levels of methionine. Normal levels of methionine are consistent with one previous study[10] which involved the use of age-matched controls with no intake of vitamins/minerals; previous studies of methionine levels had reported significant differences, but those studies involved either children with autism taking a vitamin/mineral supplement[8] or significant age differences.[9]

Oxidative Stress Children with autism had significantly elevated oxidative stress, as indicated by increased GSSG/GSH ratio (glutathione is the primary anti-oxidant in the body), and increased plasma nitrotyrosine. GSSG is reduced to GSH by glutathione reductase, which requires NADPH. NADPH levels were substantially lower in children with autism, which would explain why they also had a decreased GSH/GSSG ratio. These results are consistent with several previous studies discussed in the Introduction.


The data show that, on average, children with autism have lower levels of biotin, and trend toward lower levels of vitamin B5, vitamin E, and total carotenoids. Those nutrients were also more likely to be below the reference range of the neurotypical group. The autism group had somewhat higher levels of vitamin C.

The broad distribution of vitamin B6 (measured as the active form, P5P, inside the RBC) is very interesting. Two previous studies[45,46] found high levels of total B6 in the plasma, and in those papers it was hypothesized that it might be due to an impaired conversion of B6 to the active form, P5P. That hypothesis may be valid for the subset of children with low RBC P5P, but there also appears to be a subset with high RBC P5P. Overall, the broad distribution of RBC P5P suggests that there is a subset of children who need more vitamin B6, and a subset who have high levels of B6. This is consistent with 11 double-blind, placebo-controlled studies which mostly found that about half of children or adults with autism benefit from mega-doses of vitamin B6 (500–1000 mg, or 250–500× the RDA).[45] This study suggests that there may be a subgroup (those with low RBC P5P) who would likely benefit from B6 supplementation. However, since B6 levels are not dramatically low, it is possible that high doses (20–40× RDA) instead of mega-doses (100–250× the RDA) might be sufficient, unless P5P-dependent enzymes are highly defective and require very high levels of P5P to function normally.

The higher level of total choline in the children with autism is interesting, and may suggest an impairment in conversion of choline to acetylcholine.

There is current interest in the role of Vitamin D in not only bone metabolism but also immune system function.[47] D3, synthesized in the skin on exposure to sunlight, and D2, supplied in plant and fungi based foods, are both prohormones for 25-OH-D which was measured in this study. The role of vitamin D in autism[48,49] has begun to receive attention. In this current study there was no difference in plasma vitamin D levels in children with autism versus controls. Measured levels for the neurotypical children were consistent with the results of the NHANES US population study. In both study groups there is a subgroup, which is borderline or below published RR, even in a geographic location (Arizona) with potential for high sun exposure. The cause for the decreased vitamin D levels in the general population is not well understood; it is likely multifactorial. One known factor is the change in modern lifestyles which have resulted in less exposure to direct sunlight. Thus, the "average" level of the general population may not be "optimal." So, it may be that most children, including children with ASD, may benefit from more vitamin D. The present results somewhat differ with a study in Egypt,[19] which found that young children with autism (5.3 +/- 2.8 yr) had significantly lower levels of 25 (OH)D (28.5 +/- 16.4 mcg/l) than did controls (40.1 +/- 11.8 mcg/l) of similar age. The levels for the autism group for the Egyptian and the present study are in close agreement; however the Egyptian normative control levels are much higher than for the present study. To determine if the difference with the present results was partially due to differences in study age, the present data for children ages 5–10 years was analyzed, but found no difference between the autism and control groups. So, the difference appears to be due to differences in the control groups and may reflect differences in entire population reference ranges, perhaps due to a variety of factors, such as sun exposure, dietary intake or factors interfering in vitamin D metabolism.

It is interesting to compare the results of this study with other studies. The following findings for children with autism relative to controls are consistent with other studies: lower vitamin E;[15,20] normal levels of vitamin B12,[11,50] serum folate[11,50] and vitamin A.[20] The present finding of increased vitamin C is consistent with one study,[20] but inconsistent with another study that found normal levels of vitamin C.[15] The present finding of slightly lower levels of total carotenes is in contrast to a study[15] in Saudi Arabia which found greatly increased levels of beta carotene; the reason for that difference is unclear (typically 80% of total carotene is beta-carotene, but may be due to differences in which nation the study was conducted in, due to differences in ethnicity, diet, and other factors. Fasting status is also important, as food intake rapidly affects levels of most vitamins.

Functional Tests

FIGLU is an intermediate in the deamination of histidine. Conversion of FIGLU to glutamic acid is the 4th step in this process and requires the enzyme formiminotransferase and the co-factor folic acid. The elevated urinary FIGLU suggests deficiency in either the enzyme or in folic acid. Plasma folic acid was slightly lower in the autistic group but the difference was not significant. Plasma folic acid did not significantly correlate with FIGLU levels. As a water soluble vitamin, folic acid varies daily with intake. FIGLU excretion may be representative of a longer metabolic period. FIGLU may be a more sensitive indicator of need for folic acid than measurements of plasma folic acid.

N-methyl-nicotinamide is a metabolite of vitamin B3 (niacin). The finding of higher levels of urinary n-methyl-nicotinamide in the autism group (possibly significant) is consistent with another study[51] which found elevated urinary n-methyl-nicotinamide in children with autism compared to neurotypical children. Whole blood niacin levels were very similar in the autism and neurotypical groups, and whole blood niacin levels did not significantly correlate with n-methyl-nicotinamide levels, which suggests that n-methyl-nicotinamide is a more sensitive assay.

Children with autism had normal levels of kryptopyroles in urine, although a few outliers raised their standard deviation. The measurements of kryptopyroles should not be confused with measurements of the "mauve factor" which was once thought to be kryptopyrole but is actually hydroxyhemopyrrolin-2-one (HPL).[52]


The much lower level of WB lithium is consistent with a previous study,[21] which found lower levels of lithium in younger children with autism and their mothers, but not in the older children. Those measurements were done in hair, which appears to be a more sensitive measure than blood, based on a study of lithium-deficient animals.[53] Animals fed a lithium-deficient diet had decreased immune function and suffered from more chronic infections, which may partially explain why several studies have found that children with autism have significantly more ear infections than typical children.[54–57] Lithium-deficient animals also had decreased activity of monoamine oxidase, which can greatly affect neurotransmitter levels. Low levels of lithium are associated with a wide range of psychiatric disorders.[58] The primary biochemical effect of lithium is that it regulates two enzymes, glycogen synthase kinase 3-beta and phosphatidylinositol phosphatase, both of which have been linked to autism.[59,60] Both enzymes are involved in the growth factor signaling pathway (PI3-kinase). Low levels of lithium are of concern and these results suggest that low-level lithium supplementation may be beneficial for mood stabilization in this group. Along with other studied nutrients, the role lithium plays in the etiology of autism warrants further study.

Serum iron and serum ferritin were similar in the autism and neurotypical groups. A previous study[22] found that 8 of 96 American children with ASD were anemic (haemoglobin < 110 g/l). In that study, the age range of the general group was 3–13 yr, but 7 of the 8 autism cases were in children under age 5. Another study[23] found that 16% of 96 Canadian children with ASD ages 1–10 yr had low serum ferritin (< 10–12 mcg/L), with little effect of age. The present study of older children with ASD (ages 5–16 yr) found only 2% of the children had serum ferritin levels below 12 mcg/L, which is roughly consistent with the results for older children in the study by Latif et al,[22] but somewhat lower than the rate found in the study by Dosman et al 2006.[23] Combining the results of all three studies, anemia seems to be a common problem in young children with autism (below age 5), but perhaps less common in older children with autism, likely consistent with the general population.

There was a slightly higher, but statistically significant, RBC iron in the autism group. The clinical significance of elevated RBC iron is unclear, but it did correlate with autism severity (for the PDD-BI), so elevated RBC iron may be problematic. This finding did not correlate with oxidative stress in this study; further research is warranted.

Most other minerals were only slightly different in the autism group compared to the control group. The slightly higher level of copper (WB and RBC) suggests most children with autism do not need copper supplementation. The slightly lower (possibly significant) level of magnesium in serum and WB suggests that modest magnesium supplementation might be beneficial, but RBC magnesium was normal, so at most this suggests only a minor need. The possibly significant slightly higher levels of RBC potassium, RBC phosphorus, and RBC boron seem to be minor fluctuations, and may be a statistical artifact.

The autism group had an average level of urinary iodine that was similar to the average level of the neurotypical group, but 25% of the autism group had urinary iodine levels that were below the 10th percentile of the neurotypical group. 14% of the autism group and 11% of the control group had insufficient iodine (< 100 ng/ml) based on the WHO 2007 criteria, although this needs to be interpreted cautiously since those children also had lower creatinine levels (more dilute urine) and the WHO criteria is based on volume, not normalized by creatinine, since they are focused on population analyses, not individual analyses. A previous study[21] found significantly lower levels of iodine in hair of children with autism compared to controls. It is unknown if hair levels are indicative of iodine status, whereas urinary levels are accepted as the best measure of iodine status, as most iodine is excreted in the urine.[61] Overall, it does appear that there is a significant subset of children with autism with low levels of iodine, which is one of the leading causes of mental retardation worldwide,[61] and we believe that further investigation of iodine status and thyroid status is warranted.

It is interesting to compare our results with a smaller study of RBC minerals in young Canadian children with autism (n = 20) compared to neurotypical children (n = 15), both with average ages of 3.9 years.[24] That study found lower levels of RBC selenium (p = 0.0006), higher levels of RBC molybdenum (p = 0.01), and possibly lower levels of RBC zinc (p = 0.08). They found similar levels of chromium, copper, magnesium, and manganese in RBC in the two groups. This smaller study partially agrees with the present study (normal levels of RBC chromium, magnesium, manganese), but partially disagrees in that the present study did not find a difference in RBC selenium, molybdenum, or zinc, and possibly found slightly higher RBC copper. Part of the difference in study results may be due to differences in ages (averages of 4 yr vs. 10 yr), geographic/dietary issues (Canada vs. Arizona), and study size.

A study[25] of copper and zinc levels of 20 British older children with autism compared to 30 British older neurotypical children did not find any significant difference in zinc or copper levels in plasma, although the autism group did have a 10% higher level of copper, similar to the present results for WB zinc (no difference) and WB copper (7% higher, p = 0.02).

A study of zinc levels in 45 children with ASD age 4–12 years compared to 41 neurotypical controls of similar age in Turkey found lower levels of zinc in WB and RBC.[26] This disagrees with the British study[25] and the present US study, possibly due to difference in culture/diet/geographical area.

Sulfation, Methylation, Glutathione and Oxidative Stress

Sulfate is the third most abundant essential mineral in the body. Roughly 80% of sulfate is produced in vivo by oxidation of methionine or cysteine, both sulphur-containing amino acids which are provided from dietary proteins. Sulfation is important for many reactions including detoxification, inactivation of catecholamines, synthesis of brain tissue, sulfation of mucin proteins which line the gastrointestinal tract, and more. The measurement of total plasma sulfate involves many substances in the plasma, including neurotransmitters, steroids, glycosaminoglycans, phenols, amino acids, peptides, and other molecules. The findings of very low free and total plasma sulfate are consistent with two studies, which found decreased plasma sulfate compared to neurotypical controls.[62,63] Low sulfate in the plasma is also consistent with four studies[62,64–66] which found that children with ASD had a significantly decreased sulfation capacity compared to controls, based on low levels of the paracetamol-sulfate/paracetamol-glucuronide (PS/PG) ratio in urine following administration of paracetamol (acetaminophen, trade name of Tylenol). The finding of low plasma sulfate is also consistent with a large study that found high sulfate in the urine of children with autism,[67] as sulfate wasting in the urine partly explains low levels in the plasma. ATP is required for the kidneys to resorb sulfate, and in this study ATP was correlated with levels of free and total plasma sulfate (r = 0.32 and 0.44, respectively), so this suggests that low levels of ATP are a significant contributor to decreased sulfate in children with autism. That paper also reported high levels of urinary sulfite, suggesting that there was a problem of converting sulfite to sulfate in the mitochondria. In 38% of cases (14/38) urinary sulfite and sulfate levels improved by giving 50 mcg of molybdenum, presumably since the enzyme for converting sulfite to sulfate (sulfite oxidase) contains molybdenum. In this study there was a weak correlation (r = 0.24) of RBC molybdenum with plasma free sulfate, but no significant correlation with plasma total sulfate.

The findings of decreased reduced glutathione (GSH), increased oxidized glutathione (GSSH), and increased ratio of oxidized:reduced glutathione are consistent with four previous studies.[8–10,63] The finding of decreased SAM in RBC is consistent with three studies of SAM in plasma.[8–10] Our finding of an unusually broad distribution of SAH in RBC only slightly higher (+5%, n.s.) than that of controls is similar to that of one study[10] which also found a slightly higher level of SAH in plasma (+6%, n.s.), but different from two other studies which found higher levels of SAH in plasma.[8,9] It should be noted that we measured SAM and SAH in RBC, whereas the other studies[8–10] measured SAM and SAH in plasma; RBC levels are far greater than plasma levels, but seem to provide similar results for children with autism vs. controls.

SAM is converted to SAH by the transfer of a methyl group, so the ratio of SAM/SAM is a measure of the body's methylation capacity. The finding of a slightly decreased SAM/SAH ratio (-10%, p = 0.006) is similar to one study[10] which found a slightly decreased level (-17%, n.s.), and similar to two other studies[8,9] which also found decreased levels of SAM/SAH; however, those two studies found a larger decrease (-44% and -27%, respectively) in the SAM/SAH ratio than in this study, due to lower levels of SAM.

The finding of very elevated plasma uridine suggests impaired methylation (methylation is required for conversion of uridine to thymidine), consistent with the decreased level of SAM.

The finding of elevated plasma adenosine in 33% of the autism group is consistent with two previous studies[8,9] and suggests an impairment in adenosine deaminase since adenosine levels are normal. Adenosine binds to the active site of SAH hydrolase and inactivates it[68,69] so elevated adenosine would block conversion of SAH to homocysteine, and hence may partially explain the decreased SAM/SAH ratio.

The finding of increased nitrotyrosine shows that this is also a good marker for oxidative stress in children with autism, and is consistent with the other measurements of oxidative stress (oxidized glutathione and ratio of oxidized:reduced glutathione).

Overall, the problems with SAM, glutathione, and oxidative stress suggest that children with autism need increased anti-oxidant support, folinic acid (not folic) and vitamin B12 to support the methionine cycle. (Folic acid is not sufficient in most cases. In the James et al 2004[8] study 16 of the 20 children with autism were taking a multivitamin and mineral supplement containing 400 ug folic acid and 3 ug vitamin B-12 but still had abnormal methylation; folinic acid, not folic acid, was needed to normalize methylation).


The finding of decreased plasma ATP, NADH, and NADPH, but normal levels of their precursor (vitamin B3- niacin) may suggest an impairment in the formation of NADH from niacin. The clinical significance of decreased levels of ATP in the plasma is unclear, since most ATP is intracellular. Impaired transport of ATP throughout the body is suggested. ATP is the primary energy source for many metabolic reactions, and decreased levels of plasma ATP may be related to decreased muscle tone and decreased endurance commonly observed in children with autism[21,70] NADH is mainly involved in catabolic reactions (energy metabolism and mitochondrial function) whereas NADPH is involved in anabolic reactions (antioxidation and reductive biosynthesis).[71] Decreased plasma ATP and NADH may also relate to impaired mitochondrial function, which has been reported in some children with autism.[72–75]

The reason for the decreased ATP, NADH, and NADPH is unclear. ATP, NADH, and NADPH were all strongly correlated with one another (r = 0.67 = 0.69), and all had a significant correlation with ratio of GSSH:GSG (r = -0.34 to -0.52), and plasma nitrotyrosine (r = -0.24 to -0.51) indicating that increased oxidative stress is associated with low ATP, NADH, and NADPH. Increased oxidatative stress would result in a significant conversion of NADH to NAD+ (which was not measured in this study). This could also be true to a limited extent for NADPH, but usually NADP+ is only a very small fraction of NADPH. It also may be due to impaired function and/or amount of translocator protein in the mitochondrial membrane, which would impair transport of ATP from the mitochondria into the cytoplasm[76] and into the plasma.

Primary Amino Acids

One significant abnormality in the plasma amino acids was elevated glutamate in the ASD group. Glutamate is the most prominent neurotransmitter. It is ubiquitous throughout the central nervous system where it modulates synaptic plasticity, vital to memory, learning and regulation and modulates gene expression, functioning in post-synaptic excitation as well as some inhibition. Overstimulation leads to excitotoxicity, creating oxidative stress, mitochondrial damage and ultimately may play a role in neurodegeneration. Peripherally it plays a role in taste, skin pain sensation and pancreatic exocrine function. Increased glutamate may be linked to some of the behavioral problems and features commonly associated with autism. It may indicate an increased need for vitamin B6, needed for conversion of glutamate to glutamine.

Another significant abnormality was decreased tryptophan in the ASD group. This may be due to decreased protein intake, and/or impaired digestion of protein into amino acids; the latter seems more likely since most of the essential amino acids were normal. Tryptophan is a precursor to the synthesis of serotonin, so decreased tryptophan is likely to impair serotonin synthesis (important in neurogenesis and neurotransmission). The importance of low tryptophan in autism was demonstrated by a randomized, double-blind, placebo-controlled crossover study that found that a 24-hour-diet low in tryptophan followed by a tryptophan-deficient amino acid drink led to a significant worsening in behavior in adults with autism.[77]

There were several other differences which were only possibly significant (p < 0.05). The ASD group also had slightly increased serine, and slightly decreased phenylalanine and tyrosine. The low tyrosine is presumably due to the low phenylalanine, since tyrosine is derived from phenylalanine, and they were highly correlated with one another (r = 0.74). The low phenylalanine could be due to decreased protein intake and/or impaired digestion of protein into those amino acids.

Our results suggest that some children with autism (those with low tryptophan and phenylalanine) would benefit from either increased protein intake, use of digestive enzymes containing proteases, and/or supplements of tryptophan (or 5-HTP) and phenylalanine.

Two previous studies by James et al[8,9] found decreased plasma methionine in children with autism vs. controls, but their most recent study[10] did not; the latter is consistent with the present study, which did not find a significant difference in plasma methionine.

Although there have been several other studies of plasma amino acids in children with autism, they suffered from one or more limitations, including small sample size, no fasting requirement, lack of neurotypical controls, and/or age differences between the autism and control groups, and thus have not surprisingly yielded contradictory results to one another, and to the present results.

Overall, the present data on plasma amino acids appears to be more reliable than most of the other studies of plasma amino acids, due to its larger size, overnight fast, use of neurotypical controls, use of controls of similar age, and simultaneous testing of autism and control groups. The present results suggest that children with autism, on average, have lower levels of plasma tryptophan, increased plasma glutamate, and possibly lower levels of several other amino acids.

Secondary Amino Acids

The only significant abnormality in the secondary amino acids was increased beta-aminoisobutyrate (also known as 3-aminoisobutyric acid) in the autism group. Beta-aminoisobutyrate is a product formed by the catabolism of thymine, one of the four nucleobases in the nucleic acid of DNA. So, elevated beta-aminoisobutyrate may indicate an increased rate of DNA turnover, and/or it may indicate an inhibition of the conversion of beta-aminoisobutyrate into the intermediates that eventually lead to the citric acid cycle.

There was also a possibly significant lower level of taurine, which could be due to an impairment of the conversion of methionine to cysteine to taurine, but it could also be due in part to increased wasting of taurine in the urine, which has also been reported.[78] This is consistent with another study,[63] which also found low plasma taurine in children with autism, although that study found a greater difference.

For "homocystine + homocysteine", it was undetectable in most cases, but it was detectable in a higher fraction of the autism group (29%) vs. the control group (11%), suggesting a possible abnormality in metabolism of homocystine and/or homocysteine in some children, but due to measurement limitations this requires investigation with more sensitive techniques that do not oxidize the homocysteine.

Potential Confounding Factors: Medication, Gastrointestinal and Dietary Issues

45% of the autism group were taking medications, and some of those can affect nutritional status, metabolism, and gastrointestinal function. However, a t-test comparison of the two groups revealed only two small differences between the autism groups taking and not taking medications, so medication use appeared to have little effect on the results reported here. Many of the autism group had gastrointestinal problems, either separate from or secondary to medications being taken, which may have affected nutritional intake. As shown in Table 1, 18% of the autism group had moderate or severe diarrhea, and 41% had moderate or severe constipation (9% of the total participants had both moderate or severe diarrhea and constipation, apparently alternating between them). Also, 53% of the children had moderate or severe problems with self-restricted diets (only interested in eating a few foods, often resulting in poor nutritional intake). Finally, 16% of the children with autism were on special diets (see Table 1), which could result in some differences in nutritional status.

Correlation Analysis

In most cases the sign of the correlation analysis is consistent with our understanding. For example, low amounts of an essential mineral (sulfate) or essential amino acid (phenylalanine) correlated with more severe autism. The positive correlation of RBC iron with autism severity suggests that excess RBC iron requires further research. Excess iron could increase oxidative stress through the Fenton reaction, in which iron catalyzes the conversion of hydrogen peroxide into two hydroxyl radicals and water.

Regression Analysis

It is very interesting that almost all of the biomarker groups have significant associations with variation in autism severity. Vitamins (especially vitamin B6, vitamin C, and n-methyl-nicotinamide) had a very high adjusted R2 for the PDD-BI (0.57) (see Table 13). Minerals (especially calcium, iron, and zinc) had moderate adjusted R2 (0.22–0.38), as did the primary amino acids (adjusted R2 of 0.22–0.38) (see Table 13). Secondary amino acids and the group of Sulfation/Methylation/Glutathione/Oxidative Stress had significant but only modest adjusted R2 (0.15–0.26) (see Table 14 and Table 15). The group of ATP/NADH/NADPH/CoQ10 had a significant association for only one of the three severity measures (see Table 14). So, overall, it seems that certain vitamins, minerals, amino acids, and (to a lesser extent) other biomarkers have significant associations with variations in the severity of autism (see Table 16).

The Overall analysis yielded very significant associations with the variation in autism severity, especially for the PDD-BI, followed by the SAS and then the ATEC. The most significant biomarkers were generally different for each autism severity instrument, but this is not surprising since some biomarkers have significant correlations with one another, and small differences can lead to one biomarker replacing another in the analysis. Rather than focusing on a particular biomarker, the Overall analysis is best interpreted as demonstrating that variations in the severity of autism are strongly associated with the nutritional and metabolic biomarkers investigated in this study. Of course, other factors not addressed in this study also contribute to the severity of autism, and all of these factors are worthy of further research.

These significant associations may offer clues to the etiology of autism. Vitamins, minerals, and essential amino acids are, by definition, essential for human life, so low levels of those critical nutrients will impair metabolic pathways and may contribute to the developmental delays which occur in autism. We hypothesize that the nutritional and metabolic problems reported in this study may be associated with the etiology of autism. We also hypothesize that correction of these problems may help reduce some of the symptoms and co-morbidities of autism, although much more research is needed to investigate these hypotheses.

Comparison With Other Studies

There are two other general studies that are also interesting to compare with our results. One study[18] found inadequate intake of folic acid, vitamin B6, calcium, vitamin A, vitamin C, and zinc in most children with autism in China. However, they only assessed dietary intake, and compared dietary intake based on established literature values (no control group); they did not measure blood levels.

It is also interesting to compare our results with a study by Mehl-Madrona et al 2010.[79] The present results suggest that micronutrient therapy with vitamins and minerals would be beneficial, and that is consistent with their study, which found that micronutrient supplementation was comparable with pharmaceuticals in terms of improvements in the Childhood Autism Rating Scale, and resulted in better improvements on the Aberrant Behavior Checklist (p < 0.0001), Clinical Global Impressions (p = 0.003), and Self-Injurious Behavior (p = 0.005). Similarly, a small randomized, double-blind, placebo-controlled study of a multi-vitamin/mineral supplement found that it was beneficial to children with autism, primarily in reducing gastrointestinal problems and sleep problems.[45]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: