|
|
 |
|
|
|
Letters - March 10, 1999
Effects of Methylmercury Exposure on Neurodevelopment
To the Editor: The article by Dr Davidson and colleagues[1] presents an important public health issue but is uninformative for several reasons. The mercury (Hg) content of the mother's hair was used as an indicator of fetal exposure to methylmercury (MeHg). However, a study in the Faroe Islands[2] showed that this biomarker does not corrlate closely with the Hg that has reached the fetal circulation. As expected, the cord-blood concentration is a better overall predictor of neurobehavioral risks.[3]
Exposures other than Hg may also have affected the neurobehavioral performance in the Seychelles. The authors considered only polychlorinated biphenyls in some postnatal serum samples. More important is prenatal exposures to pesticide residues, eg, from tropical fruit and vegetables, especially in mothers who ate less fish during pregnancy.
At a joint meeting of Hg researchers in 1994, we presented current criteria for selection of neurobehavioral tests as they pertain to MeHg.[4] Unfortunately, the test battery used by Davidson et al[1] satisfies these criteria only partially. Thus, the sensitivity to neurotoxicants is likely to be lost when using American omnibus tests, translated into Creole, administered in another culture by nurses, and then adjusted according to American norms. One appropriate neuropsychological test, the Bender Visual Motor Gestalt Test, was not scored by the Göttingen method recommended by the World Health Organization and used in the Faroe Islands study.[3] For comparison, children exposed to lead performed slightly better than controls on the crude Bender Gestalt score used in the Seychelles study,[1] but showed a significant deficit when scored according to the Göttingen protocol.[5]
The Seychelles may also differ in several ways from most fishing communities in industrialized countries. Details are not given on family structure, maternal literacy and education, or paternal employment, and the authors instead refer to "caregivers." English and French are known to be official languages in addition to Creole, and bilingualism may affect results on verbal tests. Also, while the tests ideally should be administered by a single neuropsychologist, no information is offered on the uniformity of test administration procedures among the various nurses giving the tests.
Imprecise assessment of fetal MeHg exposure, involvement of other neurotoxicant exposures, inappropriate scoring and insufficient sensitivity of neurobehavioral tests, and the presence of social cofounders may result in loss of statistical power and, perhaps, biased results. Unless the authors can clarify and correct these problems in future efforts, their study may not be as useful for risk-assessment purposes as they claim.
Philippe Grandjean, MD, PhD
Roberta F. White, PhD
Boston University Schools of Medicine and Public Health Boston, Mass
1. Davidson PW, Myers GJ, Cox C, et al. Effects of prenatal and postnatal methylmercury exposure from fish consumption on neurodevelopment: outcomes at 66 months of age in the Seychelles Child Development Study. JAMA. 1998;280:701-707.2. Grandjean P, Weihe P, Jørgensen PJ, Clarkson T, Cernichiari E, Viderø T. Impact of maternal seafood diet on fetal exposure to mercury, selenium, and lead. Arch Environ Health. 1992;47:185-195.
3. Grandjean P, Weihe P, White RF, et al. Cognitive deficits in 7-year-old children with prenatal exposure to methylmercury. Neurotoxicol Teratol. 1997;19:417-428.
4. White RF, Debes F, Dahl R, Grandjean P. Development and field testing of a neuropsychological test battery to assess the effects of methylmercury exposure in the Faroe Islands. In: Proceedings of the International Symposium on Assessment of Environmental Pollution and Health Effects of Methylmercury. Kumamoto, Japan: Kumamoto University; 1994:127-140.
5. Trillingsgaard A, Hansen ON, Beese I. The Bender-Gestalt test as a neurobehavioural measure of preclinical visual-motor integration deficits in children with low-level lead exposure. In: Neurobehavioural Methods in Occupational and Environmental Health (Environmental Health No. 3). Copenhagen, Denmark: World Health Organization; 1985:189-193.
(JAMA. 1999;281:896)
To the Editor: Dr Davidson and colleagues[1] reported no significant association between neurological developmental deficits and gestational exposure to MeHg in their Seychelles study. They largely dismiss the relevance to US consumers of positive associations observed in a similar study in the Faroe Islands[2] and suggest that fish consumption advisories for ocean fish in US markets are not warranted. They postulate different exposure patterns as a likely explanation for contradictory findings, with Faroese episodically consuming pilot whale with high concentrations of MeHg (1.6 ppm) and Seychellois regularly consuming fish with lower concentrations. We call attention to several issues regarding exposure patterns and metrics.
Although Davidson et al refer specifically to ocean fish in US markets, it is important to note that US freshwater fish frequently have Hg concentrations exceeding 1.0 ppm.[3] Consumption of such fish by subsistence fishers, vacationing anglers, and others may result in the episodic, high Hg exposure pattern presumed for the Faroese. Moreover, the presumed Seychelles exposure pattern is based on median concentrations for 25 species of common fish, some of which are large predators presumably with high levels of Hg. Their data provide no information on relative contribution to the diet by species and thus do not permit determination of whether episodic consumption of fish with high concentrations of Hg may occur in the Seychelles as well as (presumably) the Faroes. They previously acknowledged that no dietary survey of this cohort was conducted and that data necessary to calculate daily Hg intake (or its variability) are not available.[4]
In contrasting the Seychelles and Faroes findings, Davidson et al give insufficient attention to the choice of exposure metric. The Seychelles study relies on maternal hair Hg concentration, whereas the Faroes study used cord blood and maternal hair Hg. An earlier Seychelles postmortem study[5] showed that neonatal blood and maternal hair Hg concentrations had similar correlations with neonatal brain Hg, but estimation of brain Hg from neonatal blood had a much smaller statistical error. In the Faroes study, cord blood Hg was generally a stronger predictor of developmental effects than maternal hair.[2] These different exposure metrics thus provide different information on MeHg exposure during development, and the choice of metric may therefore be critical to the detection of subtle developmental effects due to MeHg. Failure to detect an association between MeHg exposure and developmental effects in the Seychelles study using maternal hair as the only exposure metric should not, therefore, be viewed as definitive.
Alan H. Stern, DrPH, DABT
Michael Gochfeld, MD, PhD
UMDNJ-Robert Wood Johnson Medical School Piscataway, NJ
1. Davidson PW, Myers GJ, Cox C, et al. Effects of prenatal and postnatal methylmercury exposure from fish consumption on neurodevelopment: outcomes at 66 months of age in the Seychelles Child Development Study. JAMA. 1998;280:701-707.2. Grandjean P, Weihe P, White RF, et al. Cognitive deficits in 7-year-old children with prenatal exposure to methylmercury. Neurotoxicol Teratol. 1997;19:417-428.
3. Northeast States for Coordinated Air Use Management (NESCAUM). Northeast states and Eastern Canadian provinces mercury study-a framework for action. In: Mercury in Northeastern Freshwater Fish: Current Levels and Ecological Impacts. Northeast States for Coordinated Air Use Management; Boston, Mass: February 1998: chapter 4.
4. Shamlaye CF, Marsh DO, Myers GJ, et al. The Seychelles Child Development Study on neurodevelopmental outcomes in children following in utero exposure to methylmercury from a maternal fish diet: background and demographics. Neurotoxicology. 1995;16:597-612.
5. Cernichiari E, Brewer R, Myers GJ, et al. Monitoring methylmercury during pregnancy: maternal hair predicts fetal brain exposure. Neurotoxicology. 1995;16:705-710.
(JAMA. 1999;281:896-897)
In Reply: Drs Grandjean and White and Drs Stern and Gochfeld question the significance of our study of MeHg exposure from fish consumption in the Seychelles Islands. We believe the issues they raise are opinions not based on scientific facts.
First, hair is well established and widely used as a biological marker for MeHg exposure and was recommended as the measure of choice by the World Health Organization. Hair correlates with brain levels and has a smaller coefficient of variation (percent error) than blood.[1] Longitudinal analysis of hair allows the recapitulation of exposure during the entire period of pregnancy.
Second, there is no conclusive scientific evidence that pesticides cause developmental disabilities and no basis for believing pesticide exposure affected our analysis. Even so, pesticide use in the Seychelles is limited and monitoring over the past 10 years by the Seychelles Public Health Laboratory has shown levels well within accepted limits.[2]
Our evaluation battery was composed of tests used in previous studies to successfully identify neurotoxic effects and recommended by the Agency for Toxic Substances and Disease Registry.[3] The argument that the test battery was not appropriate because no adverse associations with MeHg were found is specious. Test characteristics have been carefully monitored in the Seychelles study and only those tests performing as expected were used.[4] Tests were given by Seychellois nurses with extensive pediatric experience who were specially trained to administer the test battery and then monitored with reliability between testers and a PhD pediatric psychologist, between the testers themselves, and across time. These reliability measurements have been consistently high as reported.[4]
The Koppitz scoring method we used for the Bender Gestalt Test is used worldwide for children as young as 5 years. The Göttinger Formreproduktions Test (G-F-T) scoring system has a lower age limit of 6 years.[5]
Data on family life and social status have been published.[1,5,6] The Hollingshead measure of socioeconomic status includes parental education and employment.
A number of different human studies have examined the association between neurodevelopment and MeHg exposure from fish consumption, sometimes with divergent findings. Research and public policy interests would be best served by scientifically based critiques to explain the divergence.
Philip W. Davidson, PhD
Gary J. Myers, MD
Christopher Cox, PhD
Elsa Cernichiari, MS
Thomas W. Clarkson, PhD
University of Rochester School of Medicine and Dentistry Rochester, NY
Conrad Shamlaye, MB, ChB
Ministry of Health, Republic of Seychelles Victoria, Seychelles
1. Cernichiari E, Brewer R, Myers G, et al. Monitoring methylmercury during pregnancy: maternal hair predicts fetal brain exposure. Neurotoxicology. 1995;16:705-710.2. Seychelles Public Health Laboratory Report. Republic of Seychelles: Ministry of Health; October 1998.
3. Ambler R, Gilbertini M. Pediatric Environmental Neurobehavioral Test Battery. Atlanta, Ga: Agency for Toxic Substances and Disease Registry; 1996.
4. Davidson PW, Myers GJ, Cox C, et al. Neurodevelopmental test selection, administration, and performance in the main Seychelles Child Development Study. Neurotoxicology. 1995;16:665-676.
5. Schlange H, Stein B, von Boetticher I, Taneli Gottinger S. Formreproduktions-Test. Göttingen, Germany: Verlag für Psychologie.
6. Shamlaye CF, Marsh DO, Myers GJ, et al. The Seychelles Child Development Study on neurodevelopmental outcomes in children following in utero exposure to methylmercury from a maternal fish diet: background and demographics. Neurotoxicology. 1995;16:597-612.
(JAMA. 1999; 281:897)