Mercury Toxicity: Technical Overview


Several review articles have been published on health effects of mercury (for example, see Clarkson et al., 1988; Goyer, 1991; ATSDR, 1992; WHO, 1976, 1989, 1990 and 1991). The following section is not meant to be a duplication of these extensive technical reviews; rather, it is a brief summary of the effects observed after human exposures to mercury in its three most prevalent forms:

The mode of entry of these forms of mercury into the body, their distribution within the body, and the conversion of one form to another by metabolic processes in the body all influence the type and extent of toxicity observed. These factors are discussed below.

Fate of Mercury after Contact

The toxicity of a chemical is determined by the dose or amount taken into the body. The specific effects further depend on the amount or concentration that reaches specific organs, such as the brain or kidneys, that are sensitive to poisoning by the chemical. Factors that affect the amount of mercury reaching an organ are the rate at which it enters the bloodstream (its absorption efficiency) through the skin, the lungs or the gastrointestinal system; the rate at which it is distributed to the different body organs; and changes in its chemical structure that may occur in the different organs due to metabolism.

Mercury can exist in several different forms: metallic mercury, the type found in many thermometers, has no electrical charge (it is neutral); inorganic mercury is positively charged at a level of either +1 or +2; organic mercury is a complex of mercury with carbon containing compounds. Both the charge and chemical form of mercury affect how it is absorbed and transported in the body. Uncharged mercury can move into cells readily; mercury that has a charge is largely prevented from passing across barrier membranes such as the blood brain barrier and the placenta, unless it is carried through as part of another molecule. Organic mercury compounds can accumulate in living organisms such as fish.

The distribution and toxicity of mercury in the body is complex since any one of the three chemical forms can be changed to all of the others.

In the body, conversion to the charged, inorganic form predominates but other transformations can occur.

Each chemical form of mercury produces a specific set of toxic symptoms. Complex patterns of effects may be observed, however, due to conversions of the initial form into the others in the body. For example, inorganic mercury (positively charged form) is highly toxic to the kidney. Since it is charged it does not readily pass through the blood brain barrier and thus is less toxic to the brain. Inorganic mercury is itself not readily transformed into the uncharged forms in the body. In contrast, metallic and organic mercury can more readily cause brain damage since they can pass through the protective blood-brain barrier. At high exposure levels the favored conversion of these forms of mercury to the inorganic form cannot sufficiently minimize the toxic accumulation of mercury in the brain. These compounds can also cause kidney toxicity in part because they are readily transformed to inorganic mercury in the body. Thus, exposure to all three forms of mercury can result in kidney toxicity while brain toxicity is not commonly seen following exposures to inorganic mercury.

Tables 1 and 2 summarize the absorption, distribution, and metabolism of these three forms of mercury; a brief discussion follows.

Table 1. Extent of absorption of mercury after contact by different routes

Form of Mercury Extent of Absorption by Route of Contact
IngestionDermal ContactInhalation
Metallic (e.g. the form in thermometers) Very low for liquid form Moderate for vaporized form High for vaporized form
Inorganic (e.g. sometimes usedin health and beauty products) Low to moderate (higher in infants and children) Low to moderate Low to moderate
Organic (e.g. methyl Hg; the predominant form found in fish) High Low to moderate High

Table 2. Fate of mercury in the body

Form of Mercury Mercury Retained in: Transformed to other forms: Whole body half-life (months) Excreted primarily in:
Metallic Kidney
1 to 2 Feces (most)
Inorganic Kidney (most)
(Brain if high intake results in large amounts being transformed)
Metallic and Organic
(generally less favored)
1.5 to 2 Urine (most)
Organic (methyl compounds) Kidney (most)
(generally favored)
2 or 4 Feces (most)

Metallic mercury

Exposure to metallic mercury is primarily to vaporized mercury in industrial settings with some exposure to vaporized dental amalgam and to liquid mercury from spillage in the home. Spillage at home may occur if a mercury containing thermometer or thermostat is broken; the silvery metallic mercury will evaporate. As a vapor, it is well-absorbed into the blood and highly toxic when either inhaled or in contact with skin.

In the blood, metallic mercury may remain in plasma where it can be transported to organs such as the brain. It may also enter red blood cells, where it is readily transformed to the inorganic form. Inorganic mercury can either return to the blood plasma and combine with carrier proteins there or remain in the red blood cell.

Inorganic mercury does not readily enter or pass out of the brain nor do appreciable amounts pass between a pregnant woman's blood and blood of the fetus. Thus, metallic mercury that is transformed to inorganic mercury in either the brain or the fetus accumulates there. Mercury may also accumulate in the kidney as a result of its binding to sensitive tissue sites.

Other chemicals in the body can alter the rate of transformation from metallic mercury to inorganic mercury and the distribution to different body organs. Ethanol inhibits the conversion and seems to be protective against the accumulation of inorganic mercury in the brain. Results of studies on exposed human populations which use alcohol cannot be used to predict tolerable exposure levels for populations which have little or no alcohol intake (such as young children).

After absorption, the vaporized metallic mercury is excreted in the breath with trace amounts going to urine and feces. Once transformed to inorganic mercury, excretion is through urine and feces. After it is absorbed into the body the amount of metallic mercury present is reduced by half every 1-2 months (half-life). Larger amounts of mercury in the body (body burdens) take longer to be removed than smaller amounts. Different organs release accumulated mercury at different rates; brain and kidney have been found to retain mercury for a lifetime.

Inorganic mercury

Inorganic mercury can occur at two charge levels: mercuric (Hg+2) and mercurous (Hg+1). Both are toxic to humans, but effects of the doubly charged form tend to be more severe.

Absorption after ingestion is appreciable for both forms; ingestion and entry through the skin are the main ways inorganic mercury enters the body. Accidental poisoning from ingestion and skin application has been reported as a result of long-term use of mercurous compounds sometimes found in health and beauty preparations including some laxatives, skin-lightening creams, and baby teething powders.

Once absorbed into the bloodstream, inorganic mercury combines with proteins in the plasma or enters the red blood cells. It does not readily pass into the brain or fetus but may enter into other body organs. The liver is a major site of metabolism for mercury, and all mercury absorbed from the stomach and intestine is carried in blood directly to the liver. Inorganic mercury is transformed at a relatively low rate to both metallic and organic forms, allowing for the possibility of toxic effects from these forms following high level exposures to inorganic mercury. Some of the inorganic mercury may also be combined with other chemicals in liver bile; it is then carried in bile to the intestine and excreted in feces. If it leaves the liver in the bloodstream, it can then go to other organs, including the kidney. In the kidney, much of the plasma mercury is quickly absorbed into the kidney and excreted in urine. However, some may bind to cells in the kidney, persisting there for even a lifetime. Mercury in blood may also be transferred to breast milk or deposited in hair.

The concentration of inorganic mercury in the kidney is directly related to the amount taken in. The concentration of mercury in urine is usually measured to estimate the extent of a recent exposure.

While as much as 90% of ingested inorganic mercury can be unabsorbed and thus excreted within a few days of exposure, the half-life of the portion that is absorbed is approximately 2 months.

Organic mercury

Organic forms of mercury used as pesticides have been previous sources of widespread exposure of both humans and wildlife. Presently, human exposure is primarily due to eating foods containing methyl mercury, such as fish and shellfish. Human consumption of fish eating predators (some bird species, raccoons, and aquatic mammals such as seals and whales) results in a higher exposure rate since mercury is concentrated in their tissue after they eat contaminated fish and shellfish.

Methyl mercury in food is almost completely absorbed into blood. Once absorbed, most methyl mercury is transferred to the red blood cells; the rest is bound by carrier molecules in the plasma. This distribution is relatively stable so the concentration in blood is a useful indicator of the extent of a recent or ongoing exposure.

Because of the retention in red blood cells, methyl mercury in blood is slowly transferred to other organs; this transfer continues even after ingestion of contaminated food ends. Mercury absorbed into the bloodstream from the stomach and intestine goes to the liver, where it may be metabolized to inorganic mercury (and subsequently excreted as described above); combined with bile chemicals directly and excreted; or, combined with bile chemicals and reabsorbed from the intestine. This recirculation between intestine and liver continues until the organic mercury is excreted or released from the liver into the bloodstream.

Methyl mercury in the bloodstream can enter the brain and cross the placenta. Once in these and other organs, methyl mercury can be metabolized to other inorganic forms that become concentrated in the brain or fetus. Thus, even when blood mercury levels are decreasing, concentrations in the brain and fetus may still be high or even be increasing. Methyl mercury also persists in muscle tissue; because of this, ingestion of animals which have taken in methyl mercury can result in methyl mercury poisoning.

Methyl mercury is also transferred from blood to milk and hair. The concentration of mercury in milk is lower than in the mother's plasma, and most of the mercury in milk is in the inorganic form. In contrast, mercury is concentrated in hair, the ratio of hair methyl mercury to blood methyl mercury ranging from 200 to 1 to 300 to 1. The hair concentration can be used as an indicator of previous mercury exposures, for example during pregnancy, even if there were no obvious signs of exposure at the time of occurrence.

For most people, a period of approximately two months is required to clear half of an absorbed quantity of methyl mercury from the body. However, this may require about 4 months in some people, possibly because of genetic differences. This prolonged retention could put these people and their fetuses at greater risk of toxic effects.

Treatment of mercury poisoning by modifying its fate in the body

The chemical form, extent of exposure and entry into the body, the rate of change from one form to another, and the rate of removal from the body all affect the type and severity of symptoms seen after mercury poisoning. In general, medical treatment immediately after exposure involves the removal of mercury from blood by the administration of substances which will bind mercury and carry it out of the body. This form of treatment is know as chelation therapy. If treatment is delayed until mercury is concentrated or trapped in sensitive organs (e.g., the brain and the fetus), attempts to remove it will not be successful.

This type of treatment may not be protective in those who are exposed to high concentrations of mercury since it could cause release of mercury from less sensitive organs to blood; once in blood, there is a risk of mercury being transported to more sensitive tissues such as the brain.

Other therapeutic approaches utilize substances that bind mercury in the intestine so that it does not enter (or re-enter) the blood. This may be useful in speeding the removal of methyl mercury which is cycling between the liver and the intestine. Mercury in the brain is not appreciably removed by either approach.

Toxic Effects

Mercury can be toxic when inhaled, eaten, or when placed on the skin. At low concentrations, it may seem to have no effect but signs of toxicity may develop later or become noticeable with continued exposure. Toxicity in humans is evidenced by loss of feeling or a burning sensation in arms and legs, psychological effects, loss of memory, loss of vision, loss of hearing, paralysis, congenital malformations, kidney toxicity, and death. Prenatal toxicity can result in a child with normal appearance at birth but who later exhibits a developmental delay in the ability to walk and/or talk. Because of the long latent period for observable effects, the need for treatment may be recognized too late.

With respect to the potential for mercury compounds to cause cancer considerable uncertainty exists. In spite of the large numbers of people exposed to mercury epidemiological studies addressing the carcinogenicity of mercury are relatively few and are limited in their ability to detect an effect due the small numbers of people actually studied in any one investigation; possible exposures to other carcinogens and poor mercury exposure information also limit the confidence in these studies. Overall these human studies have not demonstrated a clear association between mercury and cancer. Additional research in this area is clearly needed. In addition to a sparse data base on humans, relatively few cancer studies have been performed under standard laboratory procedures. Those studies which have been completed suggest that exposure to inorganic forms of mercury might increase kidney, forestomach, thyroid, and lymphoid tissue tumors in some rodents. Animal studies also suggest that organic mercury compounds may cause kidney tumors at high levels of exposure where significant kidney toxicity occurs.

At this time the USEPA has proposed to classify methyl mercury and inorganic mercury as EPA Group C compounds (possible human carcinogens). Metallic mercury was deemed to be not classifiable due to insufficient data (EPA Group D). These classifications are currently under review.

Limited evidence suggests that mercury may decrease the body's defenses against cancer cells and infectious agents by depressing the immune system. Other studies have demonstrated the ability of mercury to cause chromosomal effects, an outcome that is frequently associated with transformation of normal cells to cancer cells. Further work on this aspect of mercury toxicity is needed.

When exposure is limited to one form of mercury, a characteristic set of toxic effects usually appears. However, as each chemical form can be metabolized to the others, a subset of clinical signs related to other forms can appear if high enough concentrations in the body are reached.

Table 3 lists the body organs and functions most affected by either the inhalation, ingestion, or dermal contact routes of exposure. These effects have been noted in both humans and animals except as indicated. Areas of the Table for which there are inadequate data to form a conclusion are left blank.

Metallic mercury

Metallic mercury toxicity is most usually a result of exposure to the vaporized form. A brief exposure to a high concentration in air results in toxicity to the lung--chest pain, bronchitis, pneumonitis. If the air concentration is lower, there may be no early signs of toxic effects because the vaporized mercury is cleared from the lungs to the blood or by exhaling. Poisoning from inhaled metallic mercury can also occur after a chronic low level exposure. Three cardinal signs of this type of exposure are excitability (erethism), tremors, and gingivitis.

Excitability and tremors are results of the deposition of mercury in the nervous system. There is a rapid transfer of the vaporized form from blood to the brain; transformation of metallic mercury to the inorganic form in the brain results in accumulation. Both forms may be toxic while in the brain. Unsteadiness and tremor when trying to move or to hold objects (intention tremor) and various manifestations of excitability can develop after a long latent period.

Table 3. Organs and functions affected after exposure to various forms of mercury

Toxic effects Vaporized Metallic Mercury Inorganic Mercury Organic Mercury (Methylated)
Prenatal exposure effects in nervous system Yes (limited data from animals) Yes (animals) Yes
Postnatal exposure effects:
Nervous systemYesYesYes
Cardiovascular systemYesYes (animals)Yes (animals))
Gastrointestinal systemYes YesYes
MuscleYes Yes
Blood cell countYes
Skin and eyesYesYesYes
Immune systemYesYes (animals)Yes (animals)
PancreasYes (extensive data from Japan populations)
ThyroidYes (limited data)
CancerYes (limited data from animals)Yes (limited data from animals)

The unsteadiness is seen most dramatically when the patient is asked in the clinic to hold both arms out to the side for three minutes. The patient is unable to do so, and will begin to flap the arms to relieve the stress (seagull sign). The psychological signs include insomnia, loss of appetite, shyness, emotional instability, and memory loss. Some reversal of these effects may occur upon removal from contact with mercury. With continued exposure, more severe tremor and muscle spasms as well as death may result.

Literature reports of incidents of mercury vapor toxicity include another type seen mainly in children--acrodynia (also known as Swift's disease or pink disease). In this disease, which occurs infrequently even among children exposed there is weight loss, loss of appetite, irritability, muscle weakness, learning disorders, and redness (hence "pink") and peeling of skin on fingers and toes. Children have most frequently shown these symptoms when calomel (a substance containing inorganic mercury) was used as a soothing agent on their teething rings. The same symptoms have been seen in children exposed to mercury vapor from contaminated floors or carpeting. Since the symptoms from inhaled mercury were accompanied by high urinary excretion rates and were the same symptoms as seen after calomel exposure, it may be that they were related to the transformation of metallic mercury to the inorganic form. It is thought that these signs may be the result of an autoimmune reaction against tissue containing mercury.

Studies of workers exposed to mercury have found that tremors and an abnormal walking gait occurred after chronic (1-5 years) exposure to 0.076 mg vaporized mercury per cubic meter of air. Mild tremors occurred at 0.026 mg/cu.m. Immune deficiency occurred in those exposed to as little as 0.106 mg/cu.m. (effects summarized by ATSDR). These numbers indicate that the toxic effects of inhaled mercury can occur at low concentrations.

A current epidemic of metallic mercury poisoning is going on now in the Amazon Rain Forest (described in Branches et al., 1993) among native people employed as gold miners. In Brazil alone, over a million miners are directly exposed to mercury vapors in the gold extraction process. Many others are exposed in the refinement and working of gold contaminated with mercury. Both sets of workers display signs of metallic mercury toxicity and excrete mercury in urine, but the gold shop workers have higher blood levels. Medical investigators studying these workers suggest that they may suffer increased levels of exposure resulting from vaporization of mercury as the contaminated gold is heated indoors.

The exposed miners and gold workers studied to date have all been adult men. No instances of exposed pregnant women have been described. In one case an individual studied, who did not work with gold at all, was found to have had a high blood mercury level. It was subsequently discovered that he lived above a gold shop and was almost certainly poisoned by mercury vapors from that source. Since these residents are also part of the exposed population and could include pregnant women, future investigations may extend to these families.

Inorganic mercury

Inorganic mercury toxicity can result from ingestion or direct skin contact with inorganic mercury or it can occur as a result of transformation of metallic mercury to inorganic mercury in the body. Poisoning has also resulted in the past when mercury containing calomel was used on teething rings; when mercury soaps and creams were applied as skin lighteners; or when laxatives containing inorganic mercury were taken chronically. Somewhat different signs of toxicity result depending on whether the mercury is in the mercuric (+2) or mercurous (+1) form.

Taken in a high dose (over 10 percent in water), mercuric chloride produces severe abdominal cramps, bloody diarrhea, and suppression of urine. Death of important tubule cells in the kidney also occurs after exposure to this form of mercury. Loss of these cells results in kidney malfunction including release of essential plasma proteins into urine (albuminuria) and excessive retention of water in the body tissues (edema). Death can result from shock and kidney failure within 24 hours, but if the patient is otherwise stabilized and placed on dialysis, the kidney may eventually repair itself using the surviving cells.

Ingestion of lower concentrations of mercuric chloride in water or food can result in an autoimmune reaction to kidney cells altered by exposure to mercury. The first signs are an inflammation of the glomerulus (the location where plasma fluids are filtered to the urinary tract); the body then further reacts immunologically to the degraded cells, causing further damage.

Mercurous compounds are less toxic than mercuric compounds. Calomel (mercurous chloride) was used in medicine; placed on gums of teething children to reduce pain; and was used as a skin lotion. Adverse responses to this form of mercury is thought to result from an immune reaction in the skin. Symptoms include a reddish skin and rash (leading to the common name of "pink disease"), fever, swollen lymph nodes and spleen, and peeling hands and feet. Mercurous compounds have also been used in the treatment of syphilis, as purgatives, and as both internal and external disinfectants. Toxicity and even death, generally as a result of kidney failure, have resulted from long-term use or misuse of these substances. Current regulations on prescription drugs and consumer products have decreased this type of exposure.

Organic mercury

Poisonings by organic mercury have occurred primarily as the result of contamination of food with methyl mercury. Extensive descriptions and analyses of symptoms have been described in reports on several widespread poisoning episodes where foods became inadvertently contaminated with high levels of methyl mercury. Studies have also been made of people exposed to more modest levels of methyl mercury in food. Some of the more extensive documentation of mercury effects in people include studies on populations in the following regions:

In all cases, the severity of symptoms was increased when the food was either more highly contaminated or eaten in larger quantities. In adults, the first signs of toxicity included abnormal sensation (tingling or numbness) in arms and legs. This effect was correlated with a cumulative intake of 25-40 mg methyl mercury and 5 ug of mercury in a gram of hair (hair to blood ratio of approximately 250 to 1). An average daily intake of 3-7 ug methyl mercury per kilogram body weight could be expected to produce such effects. Other early effects included blurred vision and a general feeling of malaise.

At higher mercury exposure levels and correspondingly higher body burdens, additional symptoms appeared. These included: loss of coordination of gait (ataxia); slurred speech (dysarthria); loss of peripheral vision; loss of hearing; coma; kidney failure; loss of memory; abnormal blood sugar; and quadriplegia. Symptoms were due to toxic effects on the brain, peripheral nerves, pancreas, immune system, and kidneys. In addition, in some people, genetic changes were observed in lymphocytes, suggesting that such changes could also occur in other tissues, including the reproductive organs.

The evidence from numerous epidemiological studies indicates that the fetus is very sensitive to mercury. The children of women exposed to methyl mercury during pregnancy may show signs of toxic effects either at birth or later in childhood. Some mothers who had a hair concentration of 6 ug of mercury per gram of hair (6 ug Hg/g hair) or higher during pregnancy had children who, compared to those not so exposed to mercury, started walking and talking later in life and who scored lower in tests designed to measure other physical and mental development. Children whose mothers had even higher maternal exposure levels (hair mercury concentrations of up to 400 ug/g), were affected with a greater frequency and suffered more severe symptoms. These included mental retardation, cerebral palsy and a high degree of irritability and sensitivity to touch.

Comparison of the doses needed for adult toxicity and fetal toxicity is difficult since the fetus preferentially accumulates methyl mercury; the ratio of mercury in fetal blood to maternal blood is about 5:1. Thus, the fetus is exposed to a greater overall concentration of methyl mercury than the mother. Additionally, there may also be a greater rate of transfer of mercury to the brain in the fetus. Pregnant women may therefor show little if any adverse effect following mercury exposure but still have an affected child.

Because methyl mercury is secreted into breast milk, nursing infants of mothers exposed to mercury only after pregnancy can also be exposed to methyl mercury. Children exposed in this way have been shown to have methyl mercury in their blood; since few children were observed in these studies, and none followed through full development to adults, it is not possible to determine the effects of this type of exposure. Available data, however, suggest that effects of exposure after birth are less severe than effects from a prenatal exposure.

Adverse effects have been found to be persistent in survivors of all major epidemics of methyl mercury poisoning. In the Iraq epidemic and in the United States family exposed by eating pork, follow-up studies showed that serious effects (quadriplegia, mental defect, loss of vision, etc.) persisted for the duration of follow-up or until death; mercury remained in the brain over this period of time as well. In both situations, methyl mercury had been ingested for as little as 3 months (at high levels); medical attention, including chelation therapy, had been provided to the family in the United States.

Because of the seriousness of the effects associated with methyl mercury poisoning, their insidious onset, and the persistence of symptoms, environmental and public health professionals have focused their efforts on preventing exposures, especially of the fetus. As early as 1976, the World Health Organization (WHO, 1976) recommended that no more than 0.3 mg total methyl mercury be ingested per week. Other agencies have recommended limits for allowable daily intakes of mercury in its various forms or have set limits for concentrations in air, water, food, and other environmental media. A recent evaluation of data on methyl mercury resulted in the suggestion that the reference dose (the daily dose likely to be without significant adverse effects) for a chronic (long-term) exposure to this organo-metal should be somewhat lower than the previous value recommended by the USEPA (Stern, 1993). This and other data on mercury intake were evaluated by USEPA which recently lowered its recommended reference dose to 0.1 ug/kg/day from its earlier value of 0.3 ug/kg/day.

It is important to note, however, that the hazard of low doses of mercury, especially attributable to fish containing methyl mercury, is a matter of considerable controversy. Two recent studies on fairly large numbers of children exposed to mercury in utero are especially relevant. Results, which have not yet been published, from a study conducted on children living in the Faroe Islands, report an association between mercury exposure and developmental effects in the children studied (Science Scope, p. 10045, Science [271], 1996). In contrast, a second study, discussed below, did not detect any clearly adverse effects among children living on the Seychelles Islands. This extensive study, although not conclusive, is reassuring with respect to consumption of fish containing low levels of mercury. However, additional analysis of data derived from further follow-up of these children remains to be completed, making it premature to draw final conclusions regarding this investigation. Because of the quality and potential significance of this study, it is more extensively discussed below.

Seychelles Island Study

Previous reports by others (e.g., Grandjean, et al., 1992; Kjellstrom et al., 1986, 1989) have suggested that children born to women who eat fish or whale meat contaminated with organic mercury during pregnancy run a risk of delayed neurological development. Other studies showed that the average daily intake of organic mercury could be estimated by analyzing the amount of mercury deposited in the growing hair shaft during the exposure period (e.g., Cox et al., 1989). Data from an epidemic of organomercury poisoning caused by ingestion of pesticide-treated grain was used to determine that a 5% risk of developmental defects was associated with maternal hair concentrations as low as 10-20 ug/g (WHO, 1990).

The results of these studies were confounded because exposure to mercury from other sources could affect the outcome. Scientists at Rochester University conducted an international search for a population that was free of exposure to mercury from industrial sources, in which women were not exposed extensively to other factors influencing rate of fetal development (such as alcohol and tobacco), which had accessible quality medical institutions, which had a high rate of schooling and tracking of the children, which was stable and amenable to the study, and which had a high rate of local fish consumption. In addition, the investigators looked for a population that had methylmercury in hair below 20 ug/g.

The Seychelles Islands was chosen as the optimal site for the study. An extensive epidemiological study on the relationship between mercury content of mother's hair and fetal health was conducted in the Seychelles Islands by an international team of scientists. The details of the design of a pilot and full study; results after 66 months of observation in the pilot study; and results after 29 months in the full study are reported in several articles of one issue of the journal NeuroToxicology (1995; Issue 16(4)).

At the Seychelles Islands, fish from reefs are eaten daily as a source of protein. These fish have relatively low mercury content, many species being below 0.1 ug Hg per g fish (wet weight). However, most women in this study ate 10-14 fish meals per week a considerably higher rate than seen with most US citizens. No estimates of the average daily dose of mercury were provided in these reports. In the pilot study on 789 mother-infant pairs, the methylmercury deposited in mother's hair during pregnancy ranged from 0.59 ppm to 36.4 ppm, with a median concentration of 6.6 ppm.

International neonatal physical developmental indices were used, including birth weight and other physical measurements, Apgar scores, and gestational age to investigate potential mercury related effects. The Revised Denver Developmental Screening Test (DDST-R) was given to children between the ages of 5 and 109 weeks of age; this measures motor, perceptual, and cognitive development at an early age. Results of the DDST-R are scored as normal, questionable and abnormal. If the questionable category is grouped with the abnormal scores, a positive association between mercury levels and developmental effects is observed. If grouped with the normal scores the study is negative for mercury effect.

Two hundred and seventeen children whose mothers had a median hair mercury content of 7.1 ppm were again evaluated at 66 months using additional psychological testing procedures for that age. Standard assessment procedures (Endnote 1) on cognitive, sensory, language, and comprehension abilities were adapted to avoid cultural and language bias. Physical examinations were conducted periodically as usual during childhood and the medical records were available for inclusion in the study.

The pilot study provided suggestive evidence of an association between mercury content of hair and fetal development. A slight negative correlation was found between hair mercury and these scores; it was not statistically significant. Five children had General Cognitive Index scores less than 60; 3 had clinical deficiencies in fine motor coordination, hearing, and language similar to what was seen in other children exposed to methylmercury prenatally when the mothers ate contaminated fish (Minamata study). The authors considered the possibility that a examination of a larger population might give different results.

The main study on 740 children over 66 months is still in progress. In this group, the mothers' hair mercury ranged from 0.5 to 26.7 ppm, with a median of 5.9 ppm, during pregnancy. Developmental tests with additional endpoints were added to those previously administered in order to expand the sensitivity of the testing. (Endnote 2) Observations at 6.5, 19 and 29 months were analyzed and are discussed in the currently published reports.

There was no association between mother's hair mercury content and childhood development scores at any of the observation times to date. At 19 and 29 months, one endpoint-- activity level of the male children-- was decreased as maternal hair mercury level increased. Other factors influencing this measurement, such as parental attention, other children in the family, etc., are hard to control and the authors are cautious in assigning weight to this in the absence of other findings. A highly significant association was, however, recently reported for this endpoint among certain women in the study (Philip Davidson, Boston Risk Assessment Group seminar, May 8, 1996).

Other evidence on mercury exposure of the infants in the Seychelles comes from autopsy data. Brain tissue has been routinely preserved for mercury analysis. Although the neonatal deaths are not strictly from the study group, they represent children of the same general population. Brain tissues from autopsies on infants in Rochester were also analyzed as controls. Neither the Rochester nor the Seychelles children had displayed symptoms of neurological deficiencies before death.

Tissue from 32 Seychelles children was examined for content of mercury. The data ranged from 50 to 250 ppb, representing mainly methylmercury. Brains from 12 children from Rochester were similarly analyzed and found to contain less than 50 ppb mercury (with one unexplainable exception). Histological observations on the Seychelles tissues showed no abnormality in cerebral or cerebella cortical organization; other changes were not indicative of mercury effect since the same features were present in the Rochester control tissue. These results, although limited in extent, suggest that the islanders were in fact exposed to greater overall amounts of mercury compared to US citizens. It is not clear, however, that the metabolism and distribution of methylmercury in this population would be the same. Toxic effects may relate to the dose rate and timing of exposure rather than the total dose received.

In conclusion, the Seychelles Island study has not detected any clearly significant association between mercury and developmental effects in a large population of children exposed via consumption of fish bearing low levels of this metal. Follow-up of these children continues and the investigators will be reporting additional results in the future. The fact that the study is not yet completed combined with questions over the applicability of the study to episodic exposures to fish more heavily contaminated with mercury suggest that it is premature to draw any firm conclusions based on this work at this time.


  1. McCarthy Genral Cognitive Index and Perceptual Performance subscale; The Preschool Language Scale Total Language and Auditory Comprehension subscale.

  2. Fagan test of visual recognition memory (6.5 months) and Bayley Scales of Infant Development (19 and 29 months).

Where to Next:

Questions or Comments?
C. Mark Smith
MA DEP Office of Research and Standards