Re: studies if Hg in normal vs autistic children

[Guest guest]

I think that Dr Amy Holmes did a study. It may be on the ARI website

karen

[Guest guest]

This is the only one I know.

http://factsformedia.com/factsformedia/BradstreetStudy703.pdf

Carsten

[ ] studies if Hg in normal vs autistic children

I'm trying to track down studies that have been done where autistic

children and normal developing children were challenged with DMSA (or

any other chelator) and then had their Hg levels determined. Does

anyone know of any such studies?

Thanks

Barry

bhicks@...

=======================================================

[Guest guest]

Here is Amy Holmes Study!

Sorry for the long e-mail but I didn't know if a pdf. format attachment would

make it to the list.

I attached it anyway " Amy Holmes final.article.pdf (128 KB) "

Carsten

Reduced Levels of Mercury in First Baby Haircuts

of Autistic Children

Amy S. Holmes,1 Mark F. Blaxill,2 and Boyd E. Haley3

1Baton Rouge, Louisiana, USA

2SafeMinds, Cambridge, Massachusetts, USA

3Chemistry Department, University of Kentucky, Lexington, Kentucky, USA

Reported rates of autism have increased sharply in the United

States and the United Kingdom. One possible factor underlying

these increases is increased exposure to mercury through

thimerosal-containing vaccines, but vaccine exposures need to be

evaluated in the context of cumulative exposures during gestation

and early infancy. Differential rates of postnatal mercury elimination

may explain why similar gestational and infant exposures produce

variable neurological effects. First baby haircut samples were

obtained from 94 children diagnosed with autism using Diagnostic

and Statistical Manual of Mental Disorders, 4th edition (DSM IV)

criteria and 45 age- and gender-matched controls. Information on

diet, dental amalgam fillings, vaccine history, Rho D immunoglobulin

administration, and autism symptom severity was collected

through a maternal survey questionnaire and clinical observation.

Hair mercury levels in the autistic group were 0.47 ppm versus

3.63 ppm in controls, a significant difference. The mothers in the

autistic group had significantly higher levels of mercury exposure

through Rho D immunoglobulin injections and amalgam fillings

than control mothers.Within the autistic group, hair mercury levels

varied significantly across mildly, moderately, and severely autistic

children, with mean group levels of 0.79, 0.46, and 0.21 ppm, respectively.

Hair mercury levels among controls were significantly

correlated with the number of the mothers' amalgam fillings and

their fish consumption as well as exposure to mercury through

childhood vaccines, correlations that were absent in the autistic

group. Hair excretion patterns among autistic infants were significantly

reduced relative to control. These data cast doubt on the

efficacy of traditional hair analysis as a measure of total mercury

exposure in a subset of the population. In light of the biological

plausibility of mercury's role in neurodevelopmental disorders, the

present study provides further insight into one possible mechanism

by which early mercury exposures could increase the risk of autism.

Keywords Amalgam, Autism, Hair, Mercury, Thimerosal

Received 8 October 2002; accepted 14 March 2003.

The Autism Research Institute and The Wallace Foundation provided

funding. The authors' thanks also extend to Quig of Doctor's

Data Inc. for hair analysis support and to Barker for quantitative

analysis support.

Address correspondence to Mark F. Blaxill, 22 Fayerweather Street,

Cambridge MA 02138, USA. E-mail: Blaxill.mark@...

Autism has been defined by symptoms rather than causes

since itwas first characterized by Kanner in the 1940s (Eisenberg

and Kanner 1956). Since Rutter's (Rutter 1978) further elaboration

of diagnostic standards in 1976, the prevailing standards for

diagnosis (Diagnostic and Statistical Manual of Mental Disorders,

3rd edition [DSM III] 1980; 3rd edition-revised [DSMIII-

R] 1987; 4th edition [DSM IV] 1994) have included impairment

in three domains: social relatedness, communication, and

behavior. In a small number of cases, either genetic (Wahlstrom

et al. 1986; Bolton et al. 2002; Steffenburg et al. 1996) or environmental

(Stromland et al. 1994; and Hersh 1997;

Aronson, Hagberg, and Gillberg 1997) causes have been

established, but the vast majority of cases remain idiopathic.

The need to account for the relative contribution of genetic

and environmental causes has taken on increased importance

in light of possible sharp increases in the incidence of autism.

Early prevalence studies in the United States (Burd, Fisher, and

Kerbeshian 1987;Treffert 1970; Ritvo et al. 1989) and the United

Kingdom (Lotter 1966; Wing and Gould 1979; Deb and Prasad

1994) reported low rates of autism-generally less than 5 per

10,000-among children born before 1990. Studies of populations

born in the 1990s, however, show far higher (Bertrand

et al. 2001; Baird et al. 2000) and increasing (Department of Developmental

Services 1999; Kaye, del Melero-Montes, and Jick

2001; et al. 1999) rates of autism and autism spectrum

disorders (ASDs), in some cohorts as high as 55 per 10,000 for

autism and 80 per 10,000 for ASDs.

These increases clearly point to the rising importance of environmental

factors and raise the possibility of an etiological role

for toxic exposures: either prenatal, postnatal, or in some cumulative

pattern that combines the effect of maternal, gestational,

and infant exposures. One group (Bernard et al. 2001) has hypothesized

a causal connection between mercury exposure and

the symptoms of autism.

Until recently, thimerosal, a preservative containing 49.6%

ethyl mercury, was used in three childhood vaccines: hepatitis

B, Haemophilus influenzae B (Hib), and diphtheria-pertussistetanus

(DPT). Hib and hepatitis B were introduced to the U.S.

International Journal of Toxicology, 22:277-285, 2003

Copyright c

American College of Toxicology

ISSN: 1091-5818 print / 1092-874X online

DOI: 10.1080/10915810390220054 277

278 A. S. HOLMES ET AL.

infant vaccination schedule in October 1990 and November

1991, respectively. In addition, most varieties of Rho D immunoglobulin

injections, administered to Rh-negative mothers

during pregnancy, contained thimerosal until late in the 1990s.

The Institute of Medicine has investigated the connection

(Stratton, Gable, and McCormick 2001) between mercury exposure

from thimerosal-containing vaccines and neurodevelopmental

disorders, including autism, and found insufficient evidence

to accept or reject a causal connection, but concluded that

such a connection was biologically plausible and recommended

a comprehensive research program.

In addition to ethyl mercury, other possible sources of early

mercury exposure include fetal exposures to inorganic mercury

inhaled by the mother from dental amalgam fillings (Drasch

et al. 1994; Vimy et al. 1997) and to methyl mercury intestinally

absorbed as a consequence of maternal fish consumption.

Little is known about the specific patterns of mercury absorption,

distribution, metabolism, or excretion in human infants.

The large majority of infants immunized with the full

complement of thimerosal-containing vaccines have not been

diagnosed with an adverse effect, such as neurodevelopmental

delay. Nevertheless, ecological analysis of the timing of the

increases in autism incidence and the increased exposure to mercury

in thimerosal-containing vaccines fails to exclude a causal

relationship between the two trends of rising autism incidence

and rising mercury exposure (Blaxill 2001).

Fully prospective studies of the role of mercury exposure

in autism have not yet been designed and even retrospective

studies are highly constrained by availability of relevant biological

samples. Many families do, however, retain locks of infant

hair, especially the first baby haircut. These samples provide an

opportunity for analysis when other opportunities have passed.

Although hair mercury levels provide only a partial insight into

the excretion patterns of autistic infants, they offer substantial

availability advantages and can provide a useful test of the plausibility

of the autism treatment hypothesis.

In a clinical practice, one of the study authors (ASH) submitted

hair samples from autistic patients for commercial laboratory

testing for toxic metal exposure. Most of these mercury hair

levels were found to be low, contrary to a first-order hypothesis

of heavy metal toxicity in autism. She then asked patients to

submit first baby haircut samples for analysis, thereby testing

a sample that would more accurately reflect early exposures.

With two exceptions (these coming from a different commercial

laboratory than her preferred source and the source used in the

current study), these samples yielded hair mercury levels that

were consistently close to zero.

Based on this observation, and on the possibility that impaired

mercury excretion might be an important susceptibility factor

underlying recent increases in autism, she expanded her investigation.

She increased the sample of autistic first baby haircut

samples and collected a set of age- and gender-matched control

baby haircut samples. Notably, the control samples were

collected under the condition that the child received all their

childhood vaccinations on schedule, in order that they would

show comparable postnatal exposure levels. Consequently, this

study does not attempt to examine the role of childhood vaccine

exposures in autism. Although there are limits to the design, we

believe that our study effectively examines the null hypothesis of

no differential excretion rates in the hair of infants subsequently

diagnosed with autism.

MATERIALS AND METHODS

Patient Recruitment and Profile (Table 1)

All autistic patients were referred to the clinical practice of

ASH with a confirmed diagnosis of DSM IV autism by either a

pediatric neurologist or developmental pediatrician. The mother

of each autistic child was interviewed for exposure information

using a structured survey questionnaire. The autistic children

were between the ages of 2 and 15 at the time of interview, with

a median age of 7. Although the location of primary residence

was slightly skewed to the Midwest and Southeast, in part due

to proximity to the clinical practice, the autistic patients provided

a good cross-section of the different regions of the United

States, with an additional 6% coming from England, Canada,

and Mexico. Boys outnumbered girls, with a male:female ratio

of 3.5:1, consistent with the typical population prevalence in

autism studies (Fombonne 1999).

Conditions of Autistic Baby Hair Collection

Autistic patients sent their baby hair samples directly to Doctor's

Data Inc. (DDI,West Chicago, IL) following DDI's instructions

for the hair minerals test. First baby haircut samples had

TABLE 1

Study group profiles

Autistic group Control group

Number of males/ 73/21 (3.5:1) 34/11 (3.1:1)

females (ratio)

Median year of birth 1994 (1985-1999) 1994 (1990-1999)

(range)

Median months at baby 17.7 (11-24) 17.8 (12-24)

haircut timing (range)

Residencea

Northeast 15% 22%

Midwest 28% 22%

Southeast 25% 22%

Mountain/Plains/ 14% 20%

South Central

West 12 % 13%

International 6% 0%

aNortheast: CT, MA, VT, NY, NJ, PA. Midwest: IL, OH, MI, WI,

MN, MO. Southeast: FL, GA, NC, SC, VA, LA, MS, AL, AR, KY, TN.

Mountain/Plains/South Central: CO, KS, ND, SD, NE, UT, ID, TX,

OK, AZ.West: CA,WA, NV. International: Canada, Mexico, England.

AUTISTIC INFANTS SHOW REDUCED MERCURY HAIR LEVELS 279

been collected by the parents between 11 and 24 months of

age, with a mean age at haircut of 17.7 months. The minimum

sample amount was 0.25 g. Before deciding to standardize on

a single testing source, a minority of hair samples (20 in all)

were collected in the course of clinical treatment of additional

autistic patients and sent to a separate commercial laboratory-

not DDI-that performs similar testing. These results were reviewed

by the clinic but excluded from this study. Results from

this other laboratory were similar to those from DDI, showing

low hair mercury levels in autistic patients. Two of the 20 excluded

test results included mercury levels that were higher than

the reported levels for any of the autistic subjects in the present

study.

Controls Recruitment and Baby Hair Collection

Normal controls were recruited through an appeal to autism

parent groups and through autism newsletters. None of these

control children or parents were interviewed in person, but each

of the mothers was interviewed over the telephone. Hair collection

procedures were the same as for the autistic patients. The

inclusion criteria for controls included the following: no developmental

disabilities or chronic illness of any kind, no siblings

on the autistic spectrum, and completion of the recommended

childhood vaccinations on schedule. These controls were recruited

with the objective of matching the autistic patients in

terms of gender and age profile (see Table 1). Although not a

condition of recruitment, the state of residence was quite similar

to the autistic sample, minimizing the possibility of regional

exposure bias.

Hair Analysis Methods

Laboratory testing was conducted using Inductively Coupled

Plasma Mass Spectroscopy (ICP-MS) in blinded fashion to the

clinical status of the hair provider. Hair specimens were collected

based on retained samples of a minimum required amount

of 0.25 g of hair from the child's first haircut. In the laboratory,

the hair specimens were further cut and washed using a

modified method developed by the International Atomic Energy

Agency (IAEA) (Ryabukin 1998). Aliquots (about 0.2 g) of the

washed hair samples were digested with nitric acid in a CEM

(CEM Corporation, s, NC, USA) microwave oven with

temperature feedback control. All element determinations were

made on an ICP-MS (Elan 5000; Perkin-Elmer, Norwalk, CT,

USA) using a flow injection sample uptake system (FIAS 400;

Perkin-Elmer). Accuracy was assessed and verified using a hair

standard reference material (SRM) from China (GBW 09101,

30 elements). Puchyr et al. (1998) have described this method

and analytical performance in detail.

Clinical Observation of Subjects Based on Clinic Visits

All previously diagnosed autistic patients were also observed

by ASH in a clinical setting. A repetitive diagnostic interview

was not conducted, but an overall assessment of each child was

performed and each was assigned an autism severity level: severe,

moderate, or mild. The definitions for these categories were

as follows: (a) severe-no expressive language at all, very little

evidence of receptive language, constant divergent gaze in

presence of clinician or parent, no toy play, " people treated as

objects " ; ( mild-some expressive language, including short

phrase speech and ability to communicate wants and needs,

responsive to commands indicating functional receptive language,

some eye contact, some appropriate toy play, obvious

connection with parents and/or other family members; and

© moderate-subjects not meeting criteria for either the severe

or mild groups.

Data Collection for Other Maternal Exposures

ASH interviewed the mothers of autistic and normal children

to obtain information on mercury exposure during and after

gestation. Exposure measures were developed from survey

questions in four categories: (a) maternal amalgams during pregnancy

were estimated by direct observation by the mother (either

using a mirror, or counted by her husband) of amalgam surfaces

at time of interview less new fillings since the gestation period;

( exposures through Rho D immunoglobulin injections during

pregnancy were self-reported by the mother; © childhood

vaccinations, including the timing of exposures to hepatitis B,

DPT, and Hib vaccines, were obtained based on a joint review

of the child's pediatrician's records; and (d) fish consumption

during pregnancy in four categories was estimated using a fourlevel

scale. The four levels estimated were based on the relative

frequency of meals in which fish was consumed: " heavy " was

once a week or more, " moderate " was less than weekly and more

than monthly, " little " was less than once a month, with the final

category being " none. "

Statistical Analysis

Data were analyzed using Microsoft Excel version 9.0.3821

SR-1. Comparison of distributions of autistics and controls was

made using the two-tailed test. Multiple-regression analysis on

the normal hair sample was performed with the hair mercury

level as the dependent variable and three independent variables.

For amalgam fillings, we used the square of the number of fillings.

We chose an exponential curve for several reasons: mothers

with more fillings would be likely to have larger fillings

and larger exposed surface areas per filling; multiple amalgams

would be more likely to react with each other and release higher

levels of mercury due to galvanism or abrasion; higher levels

of amalgam exposure could lead to greater retention in maternal

tissue; and mothers with large numbers of fillingswould be more

likely to have had recent dental work done. In addition, the exponential

relationship produced consistently superior statistical

correlations. For fish diets, we assigned a monthly number of fish

meals for each response level: 5 per month for " heavy " consumption,

2.5 per month for " moderate " consumption, 0.5 for " little "

consumption, and 0 for no consumption. For vaccine exposure,

280 A. S. HOLMES ET AL.

FIGURE 1

A plot of the birth hair mercury levels of nonautistic versus

autistic children. Solid circles represent individual female

subjects and open circles represent individual male subjects.

we used the total number of micrograms received through all

thimerosal-containing vaccines.

ANALYSIS OF DATA AND RESULTS

Hair Mercury: Autistics Versus Controls

Figure 1 shows the results of the analysis comparing the excretion

of mercury in baby hair of autistic children and normal

controls. The hair levels of mercury in autistic children were significantly

lower than in controls (p < .0000004), with a mean of

0.47 ppm compared to a mean of 3.63 ppm in the control group.

Rho D Immunoglobulin and Amalgam Exposure: Autistics

Versus Controls

Despite the lower levels of mercury excretion in baby hair,

the mothers of autistic children had higher exposures to mercury

when pregnant with their autistic child than did normal

controls. Table 2 shows that the number of Rho D immunoglobulin

injections received by mothers in the autistic group was

significantly higher than the mothers of controls (0.52 versus

0.09; p < .0000004). Forty-six percent of the autistic mother

received Rho D immunoglobulin injections as compared to 9%

of the control mothers. In addition, the number of amalgam

fillings present in the mouths of mothers of autistic children

exceeded the number of fillings for mothers of controls (8.35

versus 6.60; p < .01). Thirty-four percent of the mothers in the

autistic sample had 10 or more amalgam fillings as compared to

18% of the controls.

Hair Mercury Levels Within Autistic Population: Mild

Versus Moderate Versus Severe

Within the autistic sample, the level of mercury in hairwas inversely

correlated with the symptom severity level. Figure 2 displays

the distribution of hair levels across the three subgroups of

autistic children. The mean hair mercury content was 0.21 ppm

TABLE 2

Exposure differences in autistic group as compared to controls

Autistic group Control group

(N = 94) (N = 45)

Mercury levels in first baby 0.47 (±0.28)a 3.63 (±3.56)

haircut (ppm, mean ± SD)

Rho D immunoglobulin shots 0.53 (±0.67)b 0.09 (±0.29)

during pregnancy (number

per mother, mean ± SD)

Amalgam fillings during 8.35 (±3.43)c 6.60 (±3.55)

pregnancy (number per

mother, mean ± SD)

aStatistically different from control group (p < .0000004).

bStatistically different from control group (p < .0000004).

cStatistically different from control group (p < .01).

in the severe group, 0.46 ppm in the moderate group (severe

versus moderate: p < .000002), and 0.71 ppm in the mild

group (moderate versus mild: p < .0004; severe versus mild:

p < .00000003). Even these stark differences were somewhat

moderated by a clear trend toward reduced mercury levels in

female hair within the mild group.

Differences Within Autistic Population: Gender

and Developmental Patterns

Table 3 provides information on other differences between

the three subgroups within the autistic sample. In addition to the

differences in hair mercury levels, the gender distribution varied

substantially across the three subgroups. The mild group had

the highest percentage of females, at 56%, whereas the moderate

and severe groups had 14% and 4% females, respectively. In

Figure 2, it is also apparent that the female children in the mild

FIGURE 2

A plot of the birth hair mercury levels in autistic children based

on the clinical severity of the disease. Solid circles represent

individual female subjects and open circles represent

individual male subjects.

AUTISTIC INFANTS SHOW REDUCED MERCURY HAIR LEVELS 281

TABLE 3

Differences within autistic population

Mild group Moderate group Severe group

(N = 27) (N = 43) (N = 24)

Mercury levels in first baby 0.71 ppm (±0.3) 0.46 ppm (±0.19)a 0.21 ppm

(±0.18)a,b

haircut (ppm, mean ± SD)

Males: females 12:15 37:6 23:1

Percent regressive 100 93 21

aStatistically different from mildly autistic group (p < .0004).

bStatistically different from mildly autistic group (p < .000000003).

cStatistically different from moderately autistic group (p < .0000002).

group made up over90%of the children belowthe mean and only

21% of the children above the mean. In addition, the developmental

patterns varied strongly across the three subgroups. The

severe group was the most likely to have demonstrated consistency

in symptoms from birth, only 21% displayed any pattern of

developmental regression. By contrast, the vast majority of the

mild and moderate groups reported some kind of developmental

regression.

Correlation Between Exposure and Hair Levels

In the control sample, the levels of mercury in baby hair

were significantly explained by gestational mercury exposures.

Figure 3 demonstrates that a single exposure variable, maternal

amalgam fillings, was strongly correlated with mercury hair

levels in control children, but not in autistic children. Several different

regression models were applied-including one, two, and

FIGURE 3

A plot of the birth hair mercury levels of nonautistic (control)

children versus autistics compared to the grouped numbers of

dental amalgams of the birth mothers. N equals the number of

subjects and the control-to-autistic ratio for each subset is

presented.

three independent variable regressions-and the three-variable

equation shown in Figure 4 provided the best statistical fit. Maternal

amalgam fillings were significantly correlated with mercury

levels in all regressions and on their own explain over

60% of the difference in normal hair levels. Reported maternal

fish consumption during pregnancy is an additional and signifi-

cant contributor to hair mercury levels. Vaccine exposure from

all childhood immunizations also reached significance at the

95% confidence level. By contrast, similar regressions for autistic

mercury hair levels (not shown) fail to reach significance

for any exposure variable. Moreover, applying the exposure coefficients

from the control group to the autistic group yields a

sharply higher rate of predicted excretion levels than the actual

results, whereas the predicted results based on known exposures

in the control group are remarkably close to the actual results (see

Figure 4). This reflects the high explanatory power (R2 = .79)

of the multiple regression model in the control sample.

DISCUSSION

Past Studies Using Hair as a Marker in Autism

Although this is the first analysis of the first baby haircut of

autistic infants, a number of previous studies have measured the

hair contents of autistic subjects. The earliest studies (Wecker

et al. 1985; Shearer et al. 1982; Gentile et al. 1983) analyzed hair

from subjects born before 1981 and only one of these (Wecker

et al. 1985) measured mercury. All of these studies found some

significant differences between autistic and control groups. A

group of autistic subjects averaging 5.7 years of age (Wecker

et al. 1985) showed low hair levels of calcium, magnesium,

copper, manganese, chromium, and lithium, but similar levels of

mercury compared to controls. A group averaging 8 years of age

(Shearer et al. 1982) showed low levels of cadmium excretion.

A third group of unspecified age (Gentile et al. 1983) showed

elevated levels of magnesium and potassium.

A more recent study (Holloway et al. 2001) of 50 autistic

families (no agewas specified) investigated both heavy metal exposures

and hair levels for the purpose of testing the hypothesis

of different metabolism and/or exposure levels between autistic

282 A. S. HOLMES ET AL.

FIGURE 4

Actual hair levels in autistics and controls are compared to

a predicted value. The predicted value is obtained using

the regression equation for controls: Birth hair mercury

level = (5.60) + 0.04 (amalgam volume(1)) + 1.15 (fish

consumption(2)) + 0.03 (vaccine(3)) R2 = .79. Perfect

prediction of actual hair levels by the regression model is

represented by the dashed line. Filled diamonds

represent individual nonautistic subjects and open circles

represent autistics.

subjects and controls. This study used current, not baby hair,

samples and found slightly reduced levels of mercury and lead

in autistic hair relative to controls and significant differences in

exposure, including increased maternal intake of fish and an increased

rate of ear infection in infancy and associated exposure

to antibiotic treatments.

Hair analysis has frequently been used as a measure of mercury

exposure. In particular, it has been common practice

(Grandjean et al. 1997, 1998) to measure maternal hair levels as

a marker for mother-to-fetus exposures that could affect subsequent

brain development. Hair mercury analysis has also been

criticized as a diagnostic tool for treatment (Kales and Goldman

2002), and hair minerals test results from commercial laboratories

have been criticized as inconsistent and unreliable. A recent

critical review (Seidel et al. 2001) included the testing source

for this study.

In our view, this recent review offered criticisms that were

applied in an undifferentiated fashion to a group of laboratories

and made no attempt to distinguish between proper and improper

practices within the group. Much of the criticism was justified

in the case of other facilities, because many of the laboratories

examined made use of outdated technologies and exploited their

testing results for commercial purposes, including promotion of

nutritional supplements. Close reading of the analysis shows that

the laboratory used in our analysis used none of the questionable

practices deployed by most other laboratories and was one

of only two laboratories that employ the most advanced (ICPMS)

testing equipment.We believe that the exemplary practices

of DDI, the advantages of ICP-MS, the specificity of the timing

of our sample collection, the fact that the laboratory was blind

to the clinical status of the samples, and our use of a single, consistently

calibrated laboratory protocol all combine to mitigate

any concerns over the reliability of our results.

Infant Mercury Exposure and Autism

The mercury exposure levels in infants from the recommended

U.S. childhood immunization schedule exceeded the threshold

set by the U.S. Environmental Protection Agency (EPA) for most

of the 1990s. This increased level of exposure came about as a

consequence of the addition of two new thimerosal-containing

vaccines, hepatitis B in November 1991 and Hib in October

1990, to the U.S. infant vaccination schedule. For example, a

typical 2-month-old infant would have received an average of

0.25 µg/kg/day if immunized on schedule, as compared with the

EPA threshold safety level of 0.1 µg/kg/day. This average exposure

may well understate the severity of these exposures. On the

days of vaccination, the bolus dose of mercury was many times

the threshold. This excess exposure went undetected until the

summer of 1999 when an Food and Drug Administration (FDA)

review identified the problem. Subsequent to this discovery, the

American Academy of Pediatrics (AAP) and the Centers for Disease

Control and Prevention (CDC) (1999) issued a joint statement

suspending the birth dose of hepatitisBvaccine and recommending

the phasing out of thimerosal from all infant vaccines.

The special case of direct human infant exposure to thimerosal

has never been studied. Yet the hypothesized (Bernard et al.

2001) relationship between mercury exposure and autism is

supported by a number of ecological connections: increases in

autism rates in the United States have accompanied increasing

exposure to thimerosal-containing vaccines (Blaxill 2001); increases

in autism rates in England (Lotter 1966,Wing and Gould

1979; Deb and Prasad 1994; Baird et al. 2000; Kaye et al. 2001;

et al. 1999) followed closely on the heels of a change in

the recommended immunization schedule and physician incentive

structure (Salisbury and Dittman 1999) that increased the

level of thimerosal exposure in the first four months of life; and

rates of autism rose in Japan's Fukushima prefecture (Hoshino

et al. 1982) immediately after the 1965 incident in neighboring

Niigata prefecture when an chemical factory released large

amounts of mercury into a local river.

Iatrogenic exposure to mercury has been shown to cause

childhood disease. Mercury used in teething powder preparations

in the first half of the 20th century was identified as the

cause of acrodynia, a serious disease of young children that

puzzled medical professionals for decades. Resistance to the evidence

of mercury poisoning delayed full acceptance of the evidence

for many years (Dally 1997). There are numerous parallels

between the symptoms of acrodynia and autism, including loss

of sociability and communication skills (Bernard et al. 2001).

Animal Models of Infant Mercury Excretion

Despite the absence of direct observation of mercury toxicokinetics

in human infants, there are a number of animal models

AUTISTIC INFANTS SHOW REDUCED MERCURY HAIR LEVELS 283

that provide insight into the differences in the kinetics of mercury

between infants and adults. Independent of the effects on

the brain, these models demonstrate that several factors unique

to infants can contribute to reduced excretion capacity.

Mercury is excreted mostly through the feces and is heavily

dependent on the biliary secretion of inorganic mercury transported

into bile by glutathione. Studies in neonatal rats (Ballatori

and son 1984) have demonstrated reduced bile flow and

glutathione excretion that together contribute to a reduced ability

in infants to excrete methyl mercury. In a range of studies,

research has demonstrated that both milk diets and antibiotic

administration reduce the excretion of mercury (Kostial et al.

1979, 1981; Rowland, , and Doherty 1984), effects

that are both common and amplified (Kostial et al. 1978) in

suckling animals. The presence of healthy gut flora is critical

(Rowland, , and Grass 1978; Rowland, , and

Doherty 1984) to the excretion of mercury and early exposure

to mercury has even been shown to alter the composition of gut

flora and promote both antibiotic-resistant and mercury-resistant

strains (Summers et al. 1993). All of these mechanisms suggest

that the infant period is one of both high potential variation in

mercury excretion capacity and increased risk of unintended

consequences arising from mercury exposure.

Mercury Excretion in Human Infants

Who Became Autistic

Our findings are consistent with the hypothesis connecting

mercury exposure and autism. Autistic infants released dramatically

lower levels of mercury into hair than control infants. In

our autistic group, this reduced level was not associated with

lower levels of overall exposure, quite the contrary. In many,

though not all, exposure categories, autistic infants experienced

higher levels of mercury exposure. As a matter of design, we

did not attempt to assess the impact of differences in vaccine

exposures, because we only included controls who received a

full exposure to mercury through the thimerosal in vaccines.

Autistic infants in our sample experienced increased exposure

levels through maternal Rho D immunoglobulin injections

(the large majority of licensed preparations sold during the study

period used thimerosal as a preservative). Forty-three out of 94,

or 46%, of the children in our sample were exposed to mercury

through these injections, as compared to 4 out of 45, or

9%, of controls. Several of the autistic mothers received multiple

injections, which resulted in a mean number of Rho D

immunoglobulin injections in the autistic group of 0.52 injections

per child, as compared to 0.09 among the controls. This

observation is supported by a similar finding of elevated Rh incompatibility

in mothers of autistic children in a previous study

(Juul-Dam, Townsend, and Courchesne 2001). The level of Rh

incompatibility we observed in our sample, however, is signifi-

cantly greater than the rate observed in mothers in the previous

study, 46% versus 12%. (The prevalence of Rh incompatibility

in our controls was also higher, with a rate of 9% as compared

to 3% in the previous study.) The rate of antenatal prophylaxis

in our sample seems high and may be a result of increased treatment

rates of Rh incompatibility as well as increasing frequency

of the practice of antenatal prophylaxis, a relatively recent development.

Because our study is the second report of elevated

rates of autism in children born to Rh-negative mothers, this is

a finding that deserves further investigation.

Increased numbers of amalgam fillings in the mother has been

associated with increased fetal mercury exposure in several studies

(Drasch et al. 1994; Vimy et al. 1997). Our results suggest

that autistic children received increased exposure through outgassing

of amalgam fillings than controls. The average level of

amalgam fillings among mothers of autistic children was significantly

greater than controls, with 8.35 fillings per mother in the

autistic group ands 6.6 fillings among control mothers. Mothers

of autistic children were far more likely to have received extensive

dental work, with 35 of 94 mothers, or 37%, having 10 or

more amalgam fillings as compared to 8 of 45, 18%, of controls.

Maternal dietary consumption of fish was not significantly

associated with autism (data not shown).

Within the autistic group there were also strong differences in

hair mercury levels. Lower hair mercury levels were significantly

associated with the severity of the autistic behavior observed in

the clinic. Adjusting for gender differences, these results were

even stronger, because the " mildly autistic " group was disproportionately

female. Within the mildly autistic group, female

hair levels were almost uniformly lower than the male levels.

This suggests that factors related to gender might offer a level

of protection to female infants who might otherwise demonstrate

more severe symptoms. By contrast, boys who displayed

symptoms of similar severity nevertheless successfully released

larger amounts of mercury, suggesting that boys might require

high levels of mercury elimination to develop at similar rates.

The increased male risk of autism has been extensively documented

(Fombonne 1999; Gillberg and Wing 1999).

The control group showed a very strong correlation between

measurable mercury exposure and the amount released into hair.

This suggests that normal children have an ability to defend

themselves against potentially toxic exposures and may demonstrate

little negative effect despite exposures that were relatively

large. By contrast, autistic infants who experienced comparable

exposure to mercury were completely incapable of excreting

mercury through hair at the levels that might have been predicted

based on the excretion patterns of the control infants.

Possible Consequences of Low Hair Mercury Levels

In past reviews of potential risk from mercury in vaccines

(Stratton, Gable, and McCormick 2001), the possibility of neurological

damage due to exposure to thimerosal has been minimized

as a " theoretical, but unproven " risk. A core concern

among reviewers has been the absence of evidence that " lowdose "

exposures such as those administered through vaccines

have the ability to cause any detectable harm.

Our study suggests two reasons why " lowdose " (where " low "

is relative to demonstrably harmful or even fatal doses and not

284 A. S. HOLMES ET AL.

the modeled EPA standard) exposures might raise the risk of

developmental damage. First, vaccine exposures do not occur in

isolation, but rather represent one among several pathways of

exposure through which the fetal and infant brain might accumulate

toxic levels of mercury. These pathways must therefore

be evaluated in the context of cumulative exposures, any one of

which might be harmless on its own but when combined with

other sources might contribute to harmful overall levels. Both

the autistic and the control children in our study showed increased

mercury risk based on multiple sources of exposure: in

the autistic group, both Rho D immunoglobulin and amalgam

fillings in the mother were elevated relative to controls; in the

control group, hair mercury levels were significantly correlated

with maternal amalgam fillings and fish consumption as well as

vaccine thimerosal exposure.

Second, the risk of any exposure will be greater if a larger

fraction of the toxin is retained in tissue and not excreted quickly.

Although hair is a minor pathway for mercury excretion and is far

less important than feces and urine, the low levels of mercury in

the hair of autistic infants support a hypothesis that these infants

were retaining mercury in tissue at a higher rate than control

infants. The lack of mercury in the hair of autistics may be due to

a decrease in blood mercury levels feeding the hair follicles. This

decrease is likely caused by the retention of the mercury inside

the cells where it most likely causes its major biological damage.

When mercury is not available to the hair follicle, it is less

likely to be available to the primary detoxification and excretory

pathways and retained in tissue. If we presume that a portion

of the tissue mercury retention is sequestered in the central nervous

system and is available to cause neurological damage at

sensitive points in brain development, then it is plausible that

mercury-associated damage might be a meaningful element in

the pathological process that leads to an outcome of autism.

Limitations of the Current Study

We recognize that there are limitations to the current study.

The study was not the result of a fully prospective design, recruitment

of autistic study subjects was influenced by medical

care-seeking behavior, the testing facilities were not under the

direct control of the investigators, and the resultant population

distributions may not be representative of the autism population

as a whole. Additional research is necessary both to replicate

these findings in autism and to elaborate on the impact of all the

major risk factors associated with toxic exposures to mercury.

CONCLUSIONS

The reduced levels of mercury in the first baby haircut of

autistic infants raise clear questions about the detoxification capacity

of a subset of infants. Despite hair levels suggesting low

exposure, these infants had measured exposures at least equal to

a control population, suggesting that control infants were able

to eliminate mercury more effectively. In the case of autistic

infants, those in our sample were exposed to higher levels of

mercury during gestation, through dental amalgams or Rho D

immunoglobulin injections in the mother. The addition of multiple

postnatal exposures to mercury in childhood vaccines would

have more severe consequences in infants whose detoxification

capacity is reduced or who may be closer to a dangerous threshold

exposure. In the case of control infants, mercury hair levels

were strongly affected by exposure levels, suggesting that

detoxification and excretion played an important role in ensuring

normal development in children with elevate toxic exposure

relative to peers. If reduced overall mercury elimination

is related to hair elimination, then autistic infants will retain

significantly higher levels of mercury in tissue, including the

brain, than normal infants. In light of the biological plausibility

of mercury's role in neurodevelopmental disorders, our study

provides further insight into one possible mechanism by which

early mercury exposures could increase the risk of autism.

REFERENCES

Aronson, M., B. Hagberg, and C. Gillberg. 1997. Attention deficits and autism

spectrum disorders in children exposed to alcohol during gestation: A followup

study. Dev. Med. Child Neurol. 39:58-587.

Baird, G.,T. Charman, S. Baron-Cohen, A. , J. Swettenham, S. Wheelwright,

and A. Drew. 2000. A screening instrument for autism at 18 months of age: A

6-year follow-up study. J. Am. Acad. Child Adolesc. Psychiatry 39:694-702.

Ballatori, N., and T. W. son. 1984. Dependence of biliary secretion of

inorganic

mercury on the biliary transport of glutathione. Biochem. Pharmacol.

33:1093-1098.

Bernard, S., A. Enayati, L. Redwood, H. , and T. Binstock. 2001. Autism:

A novel form of mercury poisoning. Med. Hypotheses 56:462-471.

Bertrand, J., A. Mars, C. Boyle, F. Bove, M. Yeargin-Allsopp, and P. Decoufle.

2001. Prevalence of Autism in a United States Population: The Brick Township,

New Jersey, Investigation. Pediatrics 108:1155-1161.

Blaxill, M. F. 2001. Rising incidence of autism: Association with thimerosal.

Presentation to Immunization Safety Review, Cambridge, MA. Available at:

http://www.iom.edu/iom/iomhome.nsf/WFiles/Blaxill/$file/Blaxill.PDF

Bolton, P. F., R. J. Park, J. N. Higgins, P. D. Griffiths, and A. Pickles. 2002.

Neuro-epileptic determinants of autism spectrum disorders in tuberous sclerosis

complex. Brain 125(Pt 6):1247-1255.

Burd, L., W. Fisher, and J. Kerbeshian. 1987. A prevalence study of pervasive

developmental disorders in North Dakota. J. Am. Acad. Child Adolesc.

Psychiatry 26:700-703.

Centers for Disease Control and Prevention. 1999. Thimerosal in vaccines: A

joint statement of the American Academy of Pediatrics and the Public Health

Service. MMWR Morb. Mortal. Wkly Rep. 48:563-565.

Dally, A. 1997. The rise and fall of pink disease. Soc. Hist. Med. 10:291-304.

Deb, S., and K. B. Prasad. 1994. The prevalence of autistic disorder among

children with a learning disability. Br. J. Psychiatry 165:395-399.

Department of Developmental Services. 1999. Changes in the population

of persons with autism and pervasive developmental disorders in California's

Developmental Services System: 1987-1998. A Report to the

Legislature. Sacramento, CA. Available at: http://www.dds.ca.gov/autism/

pdf/autism report 1999.pdf

Diagnostic and statistical manual of mental disorders, 3rd edition (DSM-III),

1980, 87-90. Washington, DC: American Psychiatric Society.

Diagnostic and statistical manual of mental disorders, 3rd edition-revised

(DSM-III-R), 1987, 38-39. Washington DC: American Psychiatric Society.

Diagnostic and statistical manual of mental disorders, 4th edition (DSM-IV),

1994, 66-71. Washington, DC: American Psychiatric Society.

Drasch, G., I. Schupp, H. Hofl, R. Reinke, and G. Roider. 1994. Mercury burden

of human fetal and infant tissues. Eur. J. Pediatr. 153:607-610.

AUTISTIC INFANTS SHOW REDUCED MERCURY HAIR LEVELS 285

Eisenberg, L., and L. Kanner. 1956. Early Infantile Autism. Am. J.

Orthopsychiatr.

26:556-566.

Fombonne, E. 1999. The epidemiology of autism: A review. Psychol. Med.

29:769-786.

Gentile, P. S., M. J. Trentalange, W. Zamichek, and M. . 1983. Brief

report: Trace elements in the hair of autistic and control children. J. Autism

Dev. Disord. 13:205-206.

Gillberg, C., and L. Wing. 1999. Autism: Not an extremely rare disorder. Acta

Psychiatr. Scand. 99:399-406.

Grandjean, P., P.Weihe, R. F. White, and F. Debes. 1998. Cognitive performance

of children prenatally exposed to " safe " levels of methylmercury. Environ.

Res. 77:165-172.

Grandjean, P., P. Weihe, R. F. White, F. Debes, S. Araki, K. Yokoyama, K.

Murata, N. Sorensen, R. Dahl, and P. J. nsen. 1997. Cognitive deficit in

7-year-old children with prenatal exposure to methylmercury. Neurotoxicol.

Teratol. 19:417-428.

Holloway, C., J. B. , F. Castro, M. Kerr, and M. Margolis. 2001.

Investigation

of heavy metal toxicity in people with autism. Presentation to the

International Meeting for Autism Research (IMFAR), San Diego, CA.

Hoshino, Y., H. Kumashiro, Y. Yashima, R. Tachibana, and M.Watanabe. 1982.

The epidemiological study of autism in Fukushima-ken. Folia Psychiatr. Neurol.

Jpn. 36:115-124.

Kales, S. N., and R. H. Goldman. 2002. Mercury exposure: Current concepts,

controversies, and a clinic's experience. J. Occup. Environ. Med. 44:143-154.

Kaye, J. A., M. del Melero-Montes, and H. Jick. 2001. Mumps, measles, and

rubella vaccine and the incidence of autism recorded by general practitioners:

A time trend analysis. BMJ 322:460-463.

Juul-Dam, N., J. Townsend, and E. Courchesne. 2001. Prenatal, perinatal, and

neonatal factors in autism, pervasive developmental disorder-not otherwise

specified, and the general population. Pediatrics 107:E63.

Kostial, K., M. Blanusa, I. Rabar, and I. Simonovic. 1981. More data on mercury

absorption in relation to dietary treatment in rats. Toxicol. Lett. 7:201-205.

Kostial, K., D. Kello, S. Jogo, I. Rabar, and T. Maljkovic. 1978. Influence of

age

on metal metabolism and toxicity. Environ. Health Perspect. 25:81-86.

Kostial, K., I. Rabar, M. Ciganovic, and I. Simonovic. 1979. Effect of milk on

mercury absorption and gut retention in rats. Bull. Environ. Contam. Toxicol.

23:566-571.

Lotter, V. 1966. Epidemiology of autistic conditions in young children. I:

Prevalence.

Soc. Psychiatry 1:124-137.

Puchyr, R. F., D. A. Bass, R. Gajewski, M. Calvin,W. Marquardt, K. Urek, M. E.

Druyan, and D. Quig. 1998. Preparation of hair for measurement of elements

by inductively coupled plasma-mass spectrometry (ICP-MS). Bio. Trace El.

Res. 62:167-182.

Ritvo, E. R., B. J. Freeman, C. Pingree, A. Mason-Brothers, L. Jorde, W. R.

Jenson, W. M. McMahon, P. B. sen, A. Mo, and A. Ritvo. 1989. The

UCLA-University of Utah epidemiologic survey of autism: Prevalence. Am.

J. Psychiatry 146:194-199.

Rowland, I. R., M. J. Davies, and P. Grasso. 1978. Metabolism of methylmercuric

chloride by the gastro-intestinal flora of the rat. Xenobiotica 8:37-43.

Rowland, I. R., R. D. , and R. A. Doherty. 1984. Effects of diet on

mercury metabolism and excretion in mice given methylmercury: Role of gut

flora. Arch. Environ. Health 39:401-408.

Rutter,M. 1978. Diagnosis and Definition. In Autism: A reappraisal of concepts

and treatments, ed. M. Rutter and E. Schopler, 1-25. NewYork: Plenum Press.

Ryabukin, Y. S. 1978. Activation analysis of hair as an indicator of

contamination

of man by environmental trace element pollutants. International Atomic

Energy Agency report, IAEA/RL/50, Vienna.

Salisbury, D. M., and S. Dittman. 1999. Immunization in Europe. In Vaccines,

3rd ed., ed. S. A. Plotkin and W. A. Orenstein, 1033-1046. New York: W.B.

Saunders.

Seidel, S., R. Kreutzer,D. , S. McNeel, and D. Gilliss. 2001. Assessment of

commercial laboratories performing hair mineral analysis. JAMA 285:67-72.

Shearer, T. R., K. Larson, J. Neuschwander, and B. Gedney. 1982. Minerals in the

hair and nutrient intake of autistic children. J. Autism Dev. Disord. 12:25-34.

Steffenburg, S., C. L. Gillberg, U. Steffenburg, and M. Kyllerman. 1996. Autism

in Angelman syndrome: A population-based study. Pediatr. Neurol. 14:131-

136.

Stratton, K., A. Gable, and M. McCormick, eds. 2001. Immunization safety

review: Thimerosal containing vaccines and neurodevelopmental disorders.

Institute of Medicine, Washington, DC: National Academy Press.

Stromland, K., V. Nordin, M. , B. Akerstrom, and C. Gillberg. 1994.

Autism in thalidomide embryopathy: A population study. Dev. Med. Child

Neurol. 36:351-356.

Summers, A. O., J. Wireman, M. J. Vimy, F. L. Lorscheider, B. Marshall, S. B.

Levy, S. , and L. Billard. 1993. Mercury released from dental " silver "

fillings provokes an increase in mercury- and antibiotic-resistant bacteria in

oral and intestinal floras of primates. Antimicrob. Agents Chemother. 37:825-

834.

, B., E. , C. P. Farrington, M. C. Petropoulos, I. Favot-Mayaud,

J. Li, and P. A. Waight. 1999. Autism and measles, mumps, and rubella

vaccine: No epidemiological evidence for a causal association. Lancet 353:

2026-2029.

Treffert, D. A. 1970. Epidemiology of infantile autism. Arch. Gen. Psychiatry

22:431-438.

Vimy, M. J., D. E. Hooper, W. W. King, and F. L. Lorscheider. 1997. Mercury

from maternal " silver " tooth fillings in sheep and human breast milk.Asource

of neonatal exposure. Biol. Trace Elem. Res. 56:143-152.

Wahlstrom, J., C. Gillberg, K. H. Gustavson, and G. Holmgren. 1986. Infantile

autism and the fragile X. A Swedish multicenter study. Am. J. Med. Genet.

23:403-408.

Wecker, L., S. B. , S. R. Cochran, D. L. Dugger, andW. D. . 1985.

Trace element concentrations in hair from autistic children. J. Ment. Defic.

Res. 29(Pt 1):15-22.

, P. G., and J. H. Hersh. 1997. A male with fetal valproate syndrome

and autism. Dev. Med. Child Neurol. 39:632-634.

Wing, L., and J. Gould. 1979. Severe impairments of social interaction and

associated

abnormalities in children: Epidemiology and classification . J. Autism

Dev. Disord. 9:11-29.

Re: [ ] studies if Hg in normal vs autistic children

I think that Dr Amy Holmes did a study. It may be on the ARI website

karen