Human health implications of environmental contaminants in Arctic

Review

Human health implications of environmental contaminants in

Arctic Canada: A review

J. Van Oostdam

, S.G. Donaldson

a,b

, M. Feeley

, D. Arnold

, P. Ayotte

G. Bondy

, L. Chan

. Dewaily

, C.M. Furgal

, H. Kuhnlein

, E. Loring

G. Muckle

, E. Myles

, O. Receveur

, B. Tracy

, U. Gill

, S. Kalhok

Environmental Contaminants Bureau, Safe Environments Program, Health Canada, Rm 4-046, BMO Building, 269 Laurier Avenue W.,

AL4904B, Ottawa, ON, Canada K1A 0K9

Carleton University, 1125 Coloney By Drive, Ottawa, ON, Canada K1S 5B6

Health Canada, Food Directorate, Bureau of Chemical Safety, Banting Research Center, Tunney’s Pasture, Ottawa, ON, Canada K1A 0L2

Unite´ de recherchen sante´ publique (Centre hospitalier universitaire de Que´bec - Centre hospitalier de l’Universite´ Laval),

Universite´ Laval, 945 Ave Wolfe, Ste. Foy, Que´bec, Canada G1V 5B3

Centre for Indigenous Peoples’ Nutrition and Environment, Macdonald Campus of McGill University, 21,111 Lakeshore Road,

Ste.-Anne-de-Bellevue, Quebec, Canada H9X 3V9

De´partment Science Politique et Unite´ de recherche en sante´ publique (Centre hospitalier universitaire de Que´bec - Centre hospitalier de

l’Universite´ Laval), Universite´ Laval, 945 Ave Wolfe, Ste. Foy, Que´bec, Canada G1V 5B3

Environmental Contaminants Research Division, Inuit Tapiriit Kanatami, 170 Laurier Avenue West, 12th Floor, Ottawa, ON,

Canada, K1P 5V5

cole de psychologie et Unite´ de recherche en sante´ publique (Centre hospitalier universitaire de Que´bec - Centre hospitalier de l’Universite´

Laval), Universite´ Laval, Ste Foy, Que´bec, Canada G1K 7P4

AXYS Environmental Consulting Ltd., Suite 300, 805 8th Ave SW, Calgary, Alberta, Canada T2P 1H7

Faculte´ de Medicin, Nutrition, Universite´ de Montreal, CP6128, Succursale Centre Ville Montreal, QC, Canada H3C 3J7

Health Canada, Environmental Health Directorate, Radiation Protection Bureau, 775 Brookfield Road, AL 6302D1, Ottawa, ON,

Canada K1A 0L2

Health Canada, Health Products and Food Branch, 2nd Floor, Qualicum Twr A, 2936 Baseline, AL 3302C Nepean, ON, Canada K1A 0K9

Indian and Northern Affairs, Northern Science and Contaminants Research Directorate, 10 Wellington Street, Gatineau, Quebec,

Canada K1A 0H4

Accepted 30 March 2005

Available online 16 November 2005

Abstract

The objectives of this paper are to: assess the impact of exposure to current levels of environmental contaminants in the

Canadian Arctic on human health; identify the data and knowledge gaps that need to be filled by future human health research

and monitoring; examine how these issues have changed since our first assessment [Van Oostdam, J., Gilman, A., Dewailly, E

Usher, P., Wheatley, B., Kuhnlein, H. et al., 1999. Human health implications of environmental contaminants in Arctic Canada:

doi:10.1016/j.scitotenv.2005.03.034

* Corresponding author. Tel.: +1 613 941 3570.

E-mail address: jay

van

[email protected] (J. Van Oostdam).

Science of the Total Environment 351–352 (2005) 165 – 246

www.elsevier.com/locate/scitotenv

a review. Sci Total Environ 230, 1–82]. The primary exposure pathway for contaminants for various organochlorines (OCs) and

toxic metals is through the traditional northern diet. Exposures tend to be higher in the eastern than the western Canadian Arctic.

In recent dietary surveys among five Inuit regions, mean intakes by 20- to 40-year-old adults in Baffin, Kivalliq and Inuvialuit

communities exceeded the provisional tolerable daily intakes (pTDIs) for the OCs, chlordane and toxaphene. The most recent

findings in NWT and Nunavut indicate that almost half of the blood samples from Inuit mothers exceeded the level of concern

value of 5 Ag/L for PCBs, but none exceeded the action level of 100 Ag/L. For Dene/Me´tis and Caucasians of the Northwest

Territories exposure to OCs are mostly below this level of concern. Based on the exceedances of the pTDI and of various blood

guidelines, mercury and to a lesser extent lead (from the use of lead shot in hunting game) are also concerns among Arctic

peoples. The developing foetus is likely to be more sensitive to the effects of OCs and metals than adults, and is the age groups

of greatest risk in the Arctic. Studies of infant development in Nunavik have linked deficits in immune function, an increase in

childhood respiratory infections and birth weight to prenatal exposure to OCs. Balancing the risks and benefits of a diet of

country foods is very difficult. The nutritional benefits of country food and its contribution to the total diet are substantial.

Country food contributes significantly more protein, iron and zinc to the diets of consumers than southern/market foods. The

increase in obesity, diabetes and cardiovascular disease has been linked to a shift away from a country food diet and a less active

lifestyle. These foods are an integral component of good health among Aboriginal peoples. The social, cultural, spiritual,

nutritional and economic benefits of these foods must be considered in concert with the risks of exposure to environmental

contaminants through their exposure. Consequently, the contamination of country food raises problems which go far beyond the

usual confines of public health and cannot be resolved simply by risk-based health advisories or food substitutions alone. All

decisions should involve the community and consider many aspects of socio-cultural stability to arrive at a decision that will be

the most protective and least detrimental to the communities.

Keywords: Arctic regions; Environmental monitoring; PCBs; Organochlorines; Mercury; Maternal; Infant; Monitoring environmental pollution;

Northern populations; Public health; Risk factors; Risk-benefit management

Contents

1. Introduction .................................................... 168

1.1. Aboriginal peoples of Canada ....................................... 168

1.2. Aboriginal perspectives on food and health ................................ 170

1.3. Factors that contribute to Aboriginal Northerners’ exposure to country food contamination........ 170

1.4. Evaluation of research in CACAR and application to benefit and risk

assessment/management .......................................... 171

1.5. Research ethics ............................................... 172

2. Exposure assessment ............................................... 172

2.1. Country food consumption in the Arctic .................................. 173

2.2. Contaminant levels in people and their relationship to traditional food diets ............... 174

2.2.1. Tissue levels of contaminant results ................................ 175

2.2.2. Levels of mercury in hair and blood ............................... 183

2.2.3. Population groups and studies ................................... 184

2.2.4. Maternal hair ........................................... 185

2.2.5. Maternal/cord blood ........................................ 186

2.2.6. Levels of selenium in maternal blood ............................... 187

2.2.7. Levels of lead in maternal blood ................................. 187

2.2.8. Levels of cadmium in maternal blood ............................... 187

2.2.9. Radionuclide exposure....................................... 187

2.3. Trends in traditional/country food dietary intakes and contaminant exposures ............... 189

3. Toxicology. .................................................... 190

3.1. Priority contaminants ............................................ 190

3.1.1. Toxaphene ............................................. 190

3.1.2. Chlordane ............................................. 192

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246166

3.2. Toxicological effects induced by exposure to food-chain contaminant mixtures ............. 193

3.2.1. Discussion ............................................ 196

3.3. Contaminant and dietary nutrient interactions ............................... 196

3.3.1. Discussion ............................................ 197

4. Epidemiology and human biomarkers ...................................... 198

4.1. Immune system function.......................................... 199

4.1.1. Clinical outcomes ......................................... 200

4.1.2. Biomarkers ............................................ 201

4.2. Neurodevelopment ............................................. 202

4.2.1. Clinical outcomes ......................................... 202

4.2.2. Biomarkers of developmental effects ............................... 204

4.3. Sex hormone disruption .......................................... 206

4.3.1. Clinical outcomes ......................................... 207

4.3.2. Hormonal biomarkers ....................................... 208

4.4. Oxidative stress .............................................. 208

5. Risk-benefit characterization, assessment and advice ............................... 209

5.1. Contaminant exposure risks ........................................ 210

5.1.1. Contaminant intakes ....................................... 210

5.2. Special considerations for risk management in Arctic communities . . .................. 217

5.2.1. Nutritional benefits ........................................ 217

5.3. Social, cultural, spiritual and economic benefits of country food . . . .................. 220

5.4. Assessment of perceptions of risks, benefits and safety of country foods ................. 222

5.4.1. Perceptions of risks in the north ................................. 222

5.4.2. Research on the perceptions of food-chain contamination in the north .............. 222

5.4.3. Impacts of these perceptions ................................... 224

5.5. Risk-benefit characterization........................................ 224

5.5.1. Risk management frameworks .................................. 224

5.5.2. Problem identification and context ............................... 225

5.5.3. Risk and benefit assessment ................................... 225

5.5.4. Risk characterization....................................... 227

5.5.5. Assumption/uncertainties of concern .............................. 228

5.5.6. Weighing benefits and risks–challenges in practice ....................... 228

5.5.7. Option analysis/evaluation .................................... 228

5.5.8. Selecting a risk management option ............................... 229

5.5.9. Implementation ......................................... 229

5.5.10. Monitoring and evaluating the decision ............................. 229

5.6. Risk and benefit communication...................................... 229

6. Conclusions .................................................... 231

6.1. Aboriginal perspectives on food and health and interpretation of research results............. 231

6.2. Exposure assessment............................................ 232

6.3. Toxicology................................................. 232

6.4. Epidemiology and biomarkers ....................................... 233

6.5. Risk and benefit characterization, assessment and advice ......................... 233

7. Knowledge gaps ................................................. 235

7.1. Exposure assessment............................................ 235

7.2. Toxicology................................................. 235

7.3. Epidemiology ............................................... 236

7.4. Risk and benefit characterization, assessment and advice ......................... 236

Acknowledgements .................................................. 236

References ....................................................... 237

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 167

1. Introduction

This paper reviews what is known about the human

health implications of contaminants (i.e., organochlor-

ines, heavy metals and radionuclides) in the Canadian

Arctic, identifies the important knowledge gaps and

examines how these issues have changed since the first

assessment (Van Oostdam et al., 1999). Specifically, it

discusses the knowledge of human exposure to and

possible health effects of current levels of environ-

mental contaminants in the Canadian Arctic, and iden-

tifies the data and knowledge gaps that need to be

filled by future human health research and monitoring.

This paper draws on existing knowledge and incor-

porates new information from the research and com-

munication activities funded under the human health

and communication subprogrammes of the Canadian

federal Northern Contaminants Program (NCP), which

began in 1991. The NCP was established in response to

concerns about human exposure to elevated levels of

contaminants in fish and wildlife species that are impor-

tant to the diets of many northern Aboriginal peoples.

Early studies (Wong, 1986) indicated that there was a

wide spectrum of substances such as persistent organic

pollutants, heavy metals and radionuclides at unexpect-

edly high levels in the Arctic ecosystem.

Under the first phase of NCP (NCP-I; 1991–1997)

research focussed on gathering the data required to

determine the levels, spatial patterns, and sources of

contaminants in the northern ecosystem (including its

people), and the probable duration of the problem.

Results generated through NCP-I were synthesised

and published in 1997 in the first Canadian Arctic

Contaminants Assessment Report (CACAR-I)

(CACAR, 1997). In 1998, the NCP began its second

phase (NCP-II). NCP-II focuses upon questions about

the impacts and risks to human health that may result

from current levels of contamination in key Arctic

food species as well as determining the temporal

trends of contaminants of concern in people, key

Arctic indicator species and the environment.

The objective of this paper is to address concerns

about possible adverse human health effects in people

exposed to the present levels of environmental con-

taminants in the Arctic. These concerns are especially

significant for many northern Aboriginal peoples

because of the high proportion of country foods in

their diet. These food sources include marine animals

(e.g., whales, walrus, seals, and fish) and terrestrial

game. Clearly, contaminants are only some of many

factors affecting human health and well-being in the

Arctic. Factors such as lifestyle (e.g., alcohol consump-

tion, smoking tobacco product s and narcotic use), diet-

ary choices, socio-economic factors, and genetic

predisposition are important health determinants that

need to be considered when evaluating the results

described herein.

1.1. Aboriginal peoples of Canada

The Aboriginal peoples of Canada as defined in

s.35 of the Constitution Act, 1982, include the Inuit,

the Me´tis and the First Nations, who in the Arctic

include the Dene and Yukon First Nations. All of

these peoples are present in Canada’s North. For sta-

tistical purposes, this section will describe the Inuit of

Northwest Territories (NWT), Nunavut, Nunavik

(northern Que´bec) and northern Labrador; the Dene

and Me´tis of the NWT; and the 14 First Nations of the

Yukon Territory. Me´tis are present in the Mackenzie

Valley and southern Yukon (Fig. 1.1.1).

About 56,000 (or 7.5%) of Canada’s total Abori-

ginal population of 743,000 live in the country’s

Arctic region, where they comprise just over half

(53%) of the combined population (Table 1.1.1). In

Nunavut, the Aborigi nal population accounts for

approximately 84% of the total population of

24,665. A p opulation break-down by ethnicity in

Nunavut shows the Inuit make up 83% of the popula-

tion. The general pattern in Nunavik is similar, with

about 88% of the population being Inuit and 11% non-

Aboriginal. The corresponding figures for the NWT

are about 28% First Nations, 10% Inuit, 9% Me´tis,

and 52% non-Aboriginal, of the total NWT population

of 39,455. Overall, an average 87% of people living in

those areas defined as Inuit regions of Canada identi-

fied themselves as Inuit. In the Yukon, Aboriginal

peoples comprise about 20% of the total population

and consist mainly of First Nations and Me´tis peoples.

Additional 1996 demographic data for Canada’s

Arctic regions reveals high proportions of the total

population between the ages of 0 and 14 years (34%),

and between the ages of 25 to 44 years (33%) (Fig.

1.1.2), while the 65 years-and-over age group repre-

sents only 3% of the total population of Arctic Canada

(Statistics Canada, 2002 (based on 1996 data)). In

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246168

comparison, the overall population of Canada has an

older age structure with smaller proportions in the

younger age groups and greater proportions in the

older age groups.

In Arctic Canada the Inuit population has the highest

percentage of people in the 0–14 age group (42%), and

the non-A boriginal population the lowest (22%).

Approximately 19% of the Inuit population in Arctic

Canada are between ages 15 and 24, vs. roughly 13% of

the non-Aboriginal population . Again, a similar pattern

Table 1.1.1

Aboriginal peoples: population size and proportion of the total

population in each region of Arctic Canada, 1996

Region Total

population

% Aboriginal

peoples

Nunavut 24,665 84 (20,695)

Northwest Territories 39,455 48 (19,010)

Nunavik (Northern Quebec) 8700 89 (7765)

Labrador (arctic) 2435 88 (2145)

Yukon 30,650 20 (6170)

Total Canadian Arctic Regions 105,905 53 (55,785)

Source: Statistics Canada (2001).

Data presented in this table are for those who identify with one

or more Aboriginal groups (Metis, Inuit, or North American Indian).

Fig. 1.1.1. General locations of Arctic cultural groups (adapted from Van Oostdam et al., 1999).

Percentage of each age group

100

120

Inuit* Métis* First Nations

Arctic Canada Canada

Non-aboriginal* All ethnicities

65+ 45–64 25–44 15–24 0–14

Fig. 1.1.2. Age distribution of Canadian Arctic population by

ethnicity.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 169

by ethnicity holds true in the rest of Canada. There are

lower proportions of Inuit in the 25–44 and 45–64 age

groups in Arctic Canada compared to the total Can a-

dian population (33% and 22%, respec tively).

In comparison to Arctic Canada, the total Canadian

population shows an even distribution, with a smaller

population of young people (21% aged 0–14) and a

larger population aged 65 years and over (12%),

accounted for by the large non-Aboriginal population.

Both the Aboriginal and non-Aboriginal populations

of Arctic Can ada have a lower percentage of people in

the 65-and-over age group (range of 3–6% for the four

ethnic groups of interest).

1.2. Aboriginal perspectives on food and health

In Aboriginal populations, food is perceived as an

integral component of being healthy. While this sec-

tion does not discuss this issue in full detail, it attempts

to illustrate how Aboriginal perspectives on food and

health differ from those in most Western societies.

Aboriginal peoples describe their food quite speci-

fically as Inuit food, or Dene food, as the case may be.

The term bco untry foodQ refers to mammals, fish,

plants, berries, and waterfowl/seabirds harvested

from the local environment. Imported foods refer to

all other foods which are brought mainly from the

store and general ly imported from other regions of the

country/world. Country foods play a critical role in

the social, cultural, spiritual, economic and nutritional

well-being in many Aboriginal communities. Within

many northern Aboriginal societies exchanging coun-

try food involves a complex set of social and cultural

rules and procedures that relate to the structure and

organization of these societies.

For Inuit, country food are directly associated with

physical health and well-being. At Sanikiluaq, for

example, people regard certain foods, such as seal, as

being capable of generating bodily warmth and

strength in a way that imported food cannot. Such

foods are therefore essential in the diet of Inuit activ-

ities such as hunting (Usher et al., 1995). However, as

Egede (1995) notes, country foods represent not only

health in the physical sense but also form an essential

basis of personal and community well-being for Inuit

and Dene.

Among Inuit, individual life is itself seen as a

synthesis of two elements that can be roughly trans-

lated as: the body (the physical being and function-

ality of the human body) and the soul (spirit, mind,

immediate emotional state, or even the expression of

consciousness) (Borre´, 1994). Country food is impor-

tant in contributing to the way individuals frame their

own well-being (Ridington, 1988). Am ong Inuit, this

integration is accomplished through capturing, sharing

and consuming country food. The cultural aspects and

knowledge transfer related to the harvesting of coun-

try foods such as sharing and communal processing of

food are important to individual and community

health.

These cultural aspects have been examined in sur-

vey-type studies in Arctic Canada. For example, infor-

mation collected in interviews in three major dietary

surveys demonstrated a similar agreement of selected

cultural attributes of the harvesting and use of country

food. Data for the five Inuit areas (n = 1721 individual

interviews) show that more than 80% of respondents

agreed that harvesting and using country food by the

family gives a wide range of social, cultural, spiritual,

economic and nutritional benefits ( Kuhnlein et al.,

2000).

1.3. Factors that contribute to Aboriginal Norther-

ners’ exposure to country food contamination

Food is a significant vector for contaminant expo-

sure in all populations. A substantial proportion of the

Aboriginal diet consists of country food; therefore,

Aboriginal peoples have a higher risk of contaminant

exposure if contaminant levels are higher in traditional

foods than the non-Aboriginal populations in the

North specifically, or in the rest of Canada.

Aboriginal peoples are especially likely to be

exposed to environmental contaminants such as mer-

cury or polychlorinated biphenyls (PCBs) through the

consumption of fish and mammalian organ meats or

marine mammal fats, respectively (Kinloch et al.,

1992; Kuhnlein et al., 1995c). Health impact assess-

ments must consider both exposure to contaminants

through country foods and the many benefits of these

foods. Health impact assessments must also be con-

sidered in terms of the distinctive significance of

country food to Aboriginal peoples. Vulnerability to

a hazard is a function not only of the intrinsic nature

of the hazard, but also of the geographical location

and socio-economic conditions of peop le exposed to

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246170

it, as well their cultural perspective of the hazard

(Hewitt, 1983; Douglas, 1992).

Dewailly et al. (2001c) conducted a study to better

understand the socio-demographic factors that influ-

ence Nunavimmiut exposure to contaminants and the

intake of certain nutrients found to be defic ient among

some sub-groups in this Nunavik population. These

researchers re-analyzed 1992 Sante´ Que´bec data and

identified a number of relationships between contami-

nant exposures, nutritional deficiencies, and various

socio-demographic factors among residents in the

region. Inuit couples appeared to have higher levels of

country food consumption than those reporting to be

divorced, widowed or single. Mean contaminant intake

increased with age, with the geometric mean age-

adjusted intake being higher among Inuit with fewer

years of formal education completed, among those liv-

ing in couples, and among those living in households of

more than six individuals. For several nutrients, mean

age-adjusted intake was higher among those residing in

communities along the Hudson coast. Some nutritional

deficiencies were more frequently associated with indi-

viduals less often empl oyed in full-time work.

The results of the Dewailly et al. (2001a) study

reflect the effects of some traditional socio-demo-

graphic factors (occupation, community of residence,

marital status and household structure, age, gender,

and level of education) on contaminant and nutrient

intakes through country food consumption. These

results identify sociological characteristics of indivi-

dual and their potential effects on such things as their

geographical, social, and economic access to both

country and imported foods in Nunavik. The findings

are helpful to health officials, as they support the

identification of potential bat-riskQ groups for some

contaminant exposures and nutritional deficiencies

based on existing community variables that are com-

monly available. Further, it provides information as to

the potential bwhyQ question relating to individuals’

exposure or lack of access to these foodstuffs. In

general, this information supports the development

and design of more effective promotion strategies

for country food and health.

Economic factors also affect exposure of Aborigi-

nal peoples to contaminants in country food. Country

food is an economic necessity for most Aboriginal

peoples. In many northern commu nities, employment

and incomes are low, and country food makes a sig-

nificant addition to effective household income. The

cost of a standard basket of imported food to provide a

nutritionally adequate diet is prohibitively high in

Arctic communities. For example, previous calcula-

tions of the cost to purchase equivalent amounts of

imported meats in local stores have resulted in esti-

mated value of country food production in the NWT

of about $55 million, or well over $10,000 per Abori-

ginal household per year (Usher and Wenzel, 1989).

Although this estimate was published in 1989, and

applied to the NWT when it still included Nunavut,

the key message still holds today. Moreover such

estimates apply only to the product itself, as a food

commodity, and do not include any other values that

Aboriginal peoples commonly attach to country foods

such as the activity of hunting itself, or the collective

cultural and social values of sharing and teaching.

Finally, there are health costs associated with not

eating country food, which are borne both by

Aboriginal peoples themselves and by public health

programs. Fo r example, reduced country food con-

sumption in northern Aboriginal populations,

coupled with decreased physical activity, has been

associated with obesity, dental caries, anemia, low-

ered resistance to infection, and diabetes (Szathm-

ary et al., 1987; Thouez et al., 1989).

1.4. Evaluation of research in CACAR and applica-

tion to benefit and risk asses sment/management

The Northern Contaminants Program (NCP) has

generated a considerable body of knowledge on the

human health implications of contaminants. This

assessment report reviews this information and pre-

sents research that has been conducted in or is relevant

to Arctic contaminants issues. As with any scientific

review caution shoul d be used when interpreting the

results of any single study. In assessing causal asso-

ciations between findings in any epidemiological or

toxicological study the following factors must be

considered: the strength and magnitude of the associa-

tion, consistency of the association, dose–response

relationship, temporally correct association, and bio-

logical plausibility of the association. Several studies

are needed to draw any conclusions about factors such

as consistency of the association. Many toxicological

and epidemiological studies examining the issue from

several perspe ctives may have to be revie wed to draw

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 171

causal associations between environmen tal contami-

nants and human health effects and may also point out

new directions for research and evaluation.

The results from 5 to 10 years of NCP research have

been gathered from a variety of studies. For example,

these studies have ranged from animal experiments on

toxaphene which found immune effects at low doses,

to labor atory tests of components and metabolites of

chlordane which indicate that levels of oxychlordane

and trans-nonachlor in highly exposed individuals

may approach levels where physiological changes

are seen in rodents. Animal experiments based on

the mixture of contaminants found in Arctic marine

mammals demonstrated effects on immune systems.

Epidemiological studies in the Canadian Arctic have

documented higher exposures to environmental con-

taminants among a number of Arctic populations.

Initial results from ongoing cohort study among the

Inuit population of Nunavik have shown a decrease in

birth size possibly related to increasing PCB concen-

trations. Ongoing studies related to this birth cohort

study have also found a possible link between con-

taminants and immune deficits in Inuit infants. Similar

findings have also been seen on birth size and immu-

nological effects in the Michigan fish-eater studies and

the Netherlands populations, respectively.

As discussed, these results need to be interpreted

with caution. Some of the studies are the first analyses

of data and have not yet been published and critiqued

by the scientific community. Once published and sub-

jected to full peer review, and available for comment in

the scientific literature, the importance and validity of

data and its interpretation will be confirmed. Other

factors such as lifestyle (e.g., alcohol and cigarette

consumption), diet, socio-economic status, and genetic

predisposition are important determinants of health. In

fact, contaminants may only play a modest role in

determining many important health outcomes (e.g.,

birth size). All these factors need to be considered

when evaluating the results described herein.

1.5. Research ethics

A number of considerations must be taken when

undertaking research in the Arctic. Communities must

consent to the research within their jurisdiction and,

when seeking informed consent, researchers should

explain the potential benefits or harmful effects the

research may have on individuals, communities and/or

the environment. No undo pressure should be applied

to obtain this consent and greater emphasis should be

placed on the risks to cultural values than on the

potential contributions of the research to knowledge.

Communication and participation planning have

become integral components of the research proposal

development process for projects funded through the

NCP. Methodologies and guidelines have been devel-

oped through the NCP to ensure an ethical and

responsible approach to contaminants research in the

Arctic. The purpose of these guidelines is to ensure

researchers respect and appreciate their responsibil-

ities to the communities in all aspects of their research

from research needs assestment to project design and

implementation to communication of the final results

and conclusions.

2. Exposure assessment

Key aspects of exposure to environmental contami-

nants include evaluating sources of exposure and

measuring human tissue levels of specific contami-

nants. In the context of Arctic Canada, consumption

of contaminated traditional foods is a major vector of

exposure to organochlorines, heavy met als, and radio-

nuclides. Since Van Oostdam et al. (1999), documen-

tation of the Canadian Arctic food systems has vastly

improved. The importance of traditional food in the

diet of Nunavik Inuit (Blanchet et al., 2000) and the

diet of pregnant women in the Inuvik region (Toffle-

mire, 2000) have been better documented. The safety

of traditional food has been the object of surveillance

(Lawn and Hill, 1998; Lawn and Harvey, 2001), and

three major studies of dietary intake in Arctic com-

munities have been completed by the Centre for Indi-

genous Peoples’ Nutrition and Environment (CINE)

(Kuhnlein et al., 2001a,b).

Fig. 2.1.1 shows the communities included in the

three CINE surveys, identified as Yukon First Nations,

Dene and Me´tis, and Inuit. In Van Oostdam et al.

(1999), partial results from the Dene and Me´tis survey

were known, and they Yukon survey was underway.

Data collection in Inuit communities ended in April

1999. Final reports from these three major d ietary

surveys are now available (Receveur et al., 1996,

1998; Kuhnlein et al., 2000), and additional peer-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246172

reviewed publications are forthcoming. Data from the

Dene and Me´tis communities have led to several pub-

lications developing the points already reported in Van

Oostdam et al. (1999): how traditional food improves

the nutritional quality of the diet (Receveur et al.,

1997); what levels of exposure are associated with

traditional food consumption for organochlorines and

heavy metals (Berti et al., 1998a; Kim et al., 1998) and

radioactivity (Berti et al., 1998b ); and what are some of

the social, cultural benefits (Receveur et al., 1998a,b)

and determinants (Simoneau and Receveur, 2000)of

traditional food use. Throughout this large resear ch

effort, the objectives were to understand the patterns

of traditional food use as well as the benefits and risks

of using this food in the context of the total diet.

Comparable methods were used in all of three surveys

whereby several types of interviews were conducted in

two major seasons with individuals in randomly

selected households. Food samples were collected for

analysis of several co ntaminants and nutrients. In all,

24-h recall interviews were conducted with 1012 Dene

and Me´tis, 802 Yukon First Nation residents, and 1875

Inuit, and more than 700 food samples were contrib-

uted to the research during the study years.

2.1. Country food consumption in the Arctic

In Table 2.1.1, average weekly frequency of con-

sumption of main traditional food items during late

winter and fall in the major geographical areas is

given. The most frequently mentioned traditional

food items are caribou, moose, salmon, whitefish,

grayling, trout, coney, scoter duck, cisco, walleye,

spruce hen, pike, ptarmigan, Arctic char, Canada

goose, muskox, eider duck, crowberry, beluga muktuk,

ringed seal, narwhal muktuk, partridge, and cloud-

berry. More than 250 different species of wildlife,

plants and animals were identified in workshops by

community residents as forming the rich framework of

the traditional food systems of Arctic peoples. Regio-

nal differences in species used most frequently are due

to ecosystem variety and cultural preferences.

Igluligaarjuk

Kangigliniq

Qamanittuaq

Qikiqtarjuaq

Kimmirut

Nunainguk

Aqvituq

Marruuvik

Kikiak

Mittimatalik

Qausuittuq

Iqaluktuuttiaq

Ulukhaqtuuq

Kugluktuk

Déline

Tthedzeh

Kóé

Bèhcho

Kó/Edzo

S’òmbak ’ é

Lutsel K’e

Deninu kue

Hay River

K’áágee

Tthebacha

Dewé

Yahti

Líidli

Kóé

Tthedzehk’

édli Kóé

Paulatuuq

Tuktoyaktuk

Iglulik

Arvilikjuaq

Aklavik

Teetl’it Zheh

Tsiigehtchic

K’ahbamitúé

Rádeli Kó

Dawson

City

Mayo

Carmacks

Ross River

Carcross

Atlin

Watson

Lake

Lower

Post

Beaver

Creek

Burwash

Landing

First Nations communities in theYukon

Metis and Dene communities in the NWT

Inuit communities in NWT, Nunavut and Labrador

Fig. 2.1.1. Communities participating in CINE dietary assessments.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 173

Results of 24-h recall dietary interviews in the three

areas (Yukon, Dene and Me´tis, and Inuit) demonstrated

the extent of traditional food use. Fig. 2.1.2 shows that

country food use as a percent of total dietary energy in

the three areas, varied from a low of 6% in communities

close to urban centres, to a high of 40% in the more

remote areas. In terms of gram s per person per day

(amount per capita calculated from averaging all 24-h

recall data for fall and late winter combined), traditional

food represents: for women 20–40 years of age,

approximately 106 g in the Yukon, 144 g in Dene

and Me´tis communities, and 194 g in Inuit commu-

nities; for men of the same age group the corresponding

values were 169 g, 235 g, and 245 g for each region,

respectively. Traditional food consumption increases

with age so that in the age group 40 years and over, per

capita daily consumption for women was: 193 g, 335 g,

and 341 g in the Yukon, Dene and Me´tis, and Inuit

communities, and 236 g, 343 g, 440 g, respectively, for

the men in each region ( Kuhnlein and Receveur, 2001).

2.2. Contaminant levels in people and their relation-

ship to traditional food diets

A number of maternal/cord blood contaminant stu-

dies were done in the NWT, Nunavut and Nunavik

between 1994 and 2000 (Fig. 2.2.1 ), and these studies

have given an assessment of the spatial variation in

contaminant levels. All of the studies in Nunavut took

place before devolution and so all of these population

groups are included in NWT/Nunavut. In the NWT

and Nunavut there are four ethnic groups: Caucasians,

Dene and Me´tis, Inuit, and Other (East Asian and

African). Since Caucasians in the NWT consume

mostly imported foods their contaminant patterns are

very similar to southern Canadian values, they pro-

vide a good comparison for northern peoples who

consume northern traditional foods. The number of

mothers in each region and ethnic group varied due to

number of mothers available and the time of sam-

pling. Populations sampled ranged from 175 Inuit

mothers in Nunavik to 145 Inuit, 134 Caucasian, 93

Dene and Me´tis and 13 Other mothers in the NWT/

Nunavut (Tables 2.2.1–2.2.7). The Caucasian and

Table 2.1.1

Five country/traditional food items most often consumed (yearly average of days per week)

Region Food item and yearly average of days per week consumed

12 3 4 5

Yukon Moose 1.6 Caribou 0.7 Salmon 0.6 Grayling 0.4 Trout 0.1

Dene and Metis

Gwich’in Caribou 3.2 Whitefish 1.3 Coney 0.5 Moose 0.3 Scoter 0.2

Sahtu Caribou 2.5 Moose 1.0 Trout 0.8 Whitefish 0.7 Cisco 0.3

Dogrib Caribou 3.9 Whitefish 1.2 Trout 0.2 Moose 0.2 Walleye 0.2

Deh-cho Moose 2.7 Whitefish 0.9 Caribou 0.8 Spruce Hen 0.4 Pike 0.3

South Slave Moose 2.2 Caribou 1.9 Whitefish 1.8 Trout 0.4 Ptarmigan 0.2

Inuit

Inuvialuit Caribou 1.8 Char 0.5 Goose 0.2 Whitefish 0.2 Muskox 0.1

Kitikmeot Caribou 1.2 Char 0.9 Muskox 0.3 Trout 0.3 Eider Duck 0.2

Kivalliq Caribou 1.9 Char 0.4 Crowberry 0.2 Beluga Muktuk 0.2 Trout 0.1

Baffin Caribou 1.3 Seal 1.0 Char 0.9 Narwal Muktuk 0.2 Beluga Muktuk 0.1

Labrador Caribou 1.3 Trout 0.5 Partridge 0.3 Cloudberry 0.3 Char 0.2

Source: Kuhnlein (2002).

Yukon Dene/Métis

Inuit

Percent energy

Source: Kuhnlein and Receveur, 2001

Fig. 2.1.2. Percentage of energy from traditional/country foods in

the Yukon, Dene and Me´tis, and Inuit communities.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246174

Dene and Me´tis mothers were from more than one

region in the NWT/Nunavut but there were no sig-

nificant differences in contaminant levels between

these populations, so they were combined into their

respective groups. Inuit came from four regions in

NWT/Nunavut and there were significant differences

between various groups, and so they are retained as

separate groups. In Nuna vik only 159 Inuit mothers

were included in the population sampled.

2.2.1. Tissue levels of contaminant results

The results presented here for the NWT and Nuna-

vut were supplied by Butler Walker et al. (2003,

2005), who are preparing the NWT/Nunavut Environ-

mental Contaminants Exposure Baseline report. The

results for Nunavik were supplied by Muckle et al.

(2001a,b) in an upcoming paper and in some specific

analyses for this report.

Residues of a number of organochlorine pesti -

cides have been attributed to long-range transport

to the Arctic, and approximately 11 of these pesti-

cides were included in the maternal/cord blood mon-

itoring programs. The levels of eight of the major

pesticides found most commonly in Canadian Arctic

maternal blood samples are presented in Table 2.2.1.

Oxychlordane and trans-nonachlor are components/

metabolites of technical chlordane, an older genera-

tion insecticide no longer used in most of the devel-

oped world. It can be seen that Inuit have levels that

are six to ten times higher than those seen in other

population groups such as Caucasians, Dene and

Me´tis, or Other (Fig. 2.2.2). When the Inuit groups

in NWT/Nunavut/Nunavik are examined, the Inuit

from Baffin have the highest oxychlordane and

trans-nonachlor levels. Similar patterns are seen for

HCB, mirex and toxaphene, as the Inuit mothers

have markedly higher levels of these pesticides

than those seen in rest of the groups (see Figs.

2.2.3 and 2.2.4 for HCB and toxaphene levels,

respectively).

Fig. 2.2.1. Contaminant studies in the Northwest Territories, Nunavut, and Nunavik.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 175

Table 2.2.1

Mean levels of organochlorine pesticides in maternal blood, by region and ethnic group (geometric means, range, Ag/L plasma)

OC contaminant Ethnicity/Region

Caucasian

(n = 134) Metis/Dene

(n = 93) Other

(n = 13) Inuit

Baffin

(n = 30) Inuvik

(n = 31) Kitikmeot

(n = 63) Kivalliq

(n = 17) Nunavik

n =159)

Years of study 1994–1999 1994–1999 1994–1995 1996 1998–1999 1994–1995 1996–1997 1995–2000

Oxychlordane 0.05 (nd–0.22) 0.04 (nd–0.23) 0.04 (nd–0.21) 0.58 (0.09–2.4) 0.15 (0.03–1.1) 0.29 (nd–2.9) 0.36 (nd–6.2) 0.30 (0.01–3.9)

Trans-Nonachlor 0.06 (0.02–0.26) 0.06 (nd–0.37) 0.07 (0.02–0.30) 0.64 (0.16–2.5) 0.28 (0.05–1.8) 0.31 (nd–3.0) 0.44 (0.03–3.7) 0.46 (0.01–4.6)

p,pV-DDT 0.05 (nd–0.19) 0.03 (nd–0.13) 0.22 (nd–3.2) 0.14 (0.04–0.47) 0.07 (nd–0.45) 0.08 (nd–0.33) 0.09 (nd–0.35) 0.09 (0.02–1.1)

p,pV-DDE 0.91 (0.22–11.2) 0.69 (0.15–5.3) 4.0 (0.51–34) 2.1 (0.55–6.0) 1.1 (0.40–3.8) 1.3 (0.12–7.8) 1.7 (0.21–7.2) 2.2 (0.14–18)

DDE: DDT 18 (nd–75) 18 (nd–89) 15 (nd–31) 15 (7.1–43) 13 (nd–51) 15 (nd–53) 19 (nd–52) 23 (2.8–209)

HCB 0.12 (0.04–0.61) 0.18 (0.02–1.7) 0.11 (0.02–0.40) 0.53 (0.14–1.5) 0.31 (0.06–1.2) 0.56 (0.05–4.5) 0.46 (0.07–1.8) 0.31 (0.05–2.8)

h-HCH 0.09 (nd–0.55) 0.04 (nd–0.13) 0.48 (0.04–39) 0.11 (nd–0.44) 0.08 (nd–0.25) 0.09 (nd–0.44) 0.09 (nd–0.30) 0.04 (0.02–0.25)

Mirex 0.02 (nd–0.14) 0.02 (nd–0.21) 0.01 (nd–0.07) 0.06 (nd–0.19) 0.03 (nd–0.11) 0.05 (nd–0.38) 0.06 (nd–0.80) 0.07 (0.01–0.60)

Toxaphene (total) 0.05

(nd–0.50) 0.07

(nd–0.81) NA 0.59 (nd–6.4) 0.43 (nd–3.6) 0.686

0.74 (nd–5.2) NA

Parlar 26 0.01

(nd–0.04) 0.01

(nd–0.06) NA 0.10 (0.02–0.57) 0.05 (nd–0.36) 0.096

0.08 (nd–0.43) NA

Parlar 50 0.01

(nd–0.05) 0.01

(nd–0.07) NA 0.13 (0.03–0.66) 0.06 (nd–0.43) 0.146

0.10 (nd–0.57) NA

NA= Not available; nd = not detected.

Source: Butler Walker et al. (2003).

Source: Muckle (2000) and Muckle et al. (2001b).

N = 25.

N = 42.

Four composites (n = 12, 12, 12 and 14; Seddon, 1996).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246176

Table 2.2.2

Mean levels of PCBs in maternal blood, by region and ethnic group (geometric means, range, Ag/L plasma)

OC contaminant Ethnicity/Region

Caucasian

(n = 134) Metis/Dene

(n = 93) Other

(n = 13) Inuit

Baffin

(n = 30) Inuvik

(n = 31) Kitikmeot

(n = 63) Kivalliq

(n = 17) Nunavik

(n = 159)

Years of Study 1994–99 1994–99 1994–95 1996 1998–99 1994–95 1996–97 1995–2000

Aroclor 1260

1.3 (0.24–5.7) 1.3 (0.26–14) 1.1 (0.31–3.7) 8.0 (2.0–27) 2.4 (0.62–7.9) 4.5 (0.20–27) 5.6 (0.41–60) 6.0 (0.10–48)

PCB 28 0.02 (nd–0.10) 0.01 (nd–0.06) 0.02 (nd–0.04) 0.02 (nd–0.05) 0.01 (nd–0.09) 0.01 (nd–0.06) 0.01 (nd–0.07) NA

PCB 52 0.01 (nd–0.07) 0.01 (nd–0.04) 0.01 (nd–0.02) 0.03 (nd–0.08) 0.02 (nd–0.06) 0.02 (nd–0.11) 0.02 (nd–0.09) NA

PCB 99 0.04 (nd–0.19) 0.03 (nd–0.28) 0.03 (nd–0.09) 0.19 (0.04–0.73) 0.08 (0.02–0.28) 0.12 (nd–0.81) 0.13 (nd–1.3) NA

PCB 101 0.01 (nd–0.03) 0.01 (nd–0.04) 0.01 (nd–0.02) 0.02 (nd–0.06) 0.02 (nd–0.07) 0.02 (nd–0.07) 0.02 (nd–0.04) NA

PCB 105 0.01 (nd–0.05) 0.01 (nd–0.06) 0.01 (nd–0.02) 0.04 (nd–0.15) 0.02 (nd–0.07) 0.02 (nd–0.09) 0.03 (nd–0.22) NA

PCB 118 0.04 (nd–0.27) 0.04 (nd–0.26) 0.03 (nd–0.09) 0.14 (0.03–0.50) 0.07 (0.02–0.32) 0.09 (nd–0.40) 0.09 (nd–0.66) 0.10 (0.01–0.84)

PCB 128 nd nd nd 0.01 (nd–0.05) 0.01 (nd–0.03) 0.01 (nd–0.06) 0.01 (nd–0.02) NA

PCB 138 0.11 (0.02–0.48) 0.10 (0.02–0.98) 0.10 (0.03–0.29) 0.51 (0.12–1.5) 0.19 (0.05–0.67) 0.30 (0.02–1.6) 0.37 (0.03–3.3) 0.42 (0.01–3.1)

PCB 153 0.14 (0.03–0.61) 0.16 (0.03–1.8) 0.12 (0.03–0.41) 1.0 (0.25–3.9) 0.26 (0.06–0.88) 0.56 (0.02–3.6) 0.70 (0.05–8.3) 0.75 (0.03–6.1)

PCB 156 0.02 (nd–0.11) 0.02 (nd–0.22) 0.02 (nd–0.07) 0.06 (0.02–0.30) 0.02 (nd–0.08) 0.05 (nd–0.31) 0.05 (nd–0.49) NA

PCB 170 0.03 (nd–0.16) 0.03 (nd–0.39) 0.03 (nd–0.10) 0.18 (0.04–0.95) 0.04 (nd–0.13) 0.01 (nd–0.67) 0.11 (nd–2.3) NA

PCB 180 0.09 (nd–0.50) 0.08 (nd–0.12) 0.07 (0.02–0.29) 0.40 (0.07–1.8) 0.08 (0.02–0.30) 0.27 (0.02–1.7) 0.28 (0.03–4.2) 0.32 (0.02–2.3)

PCB 183 0.01 (nd–0.05) 0.02 (nd–0.14) 0.01 (nd–0.03) 0.05 (nd–0.16) 0.02 (nd–0.07) 0.03 (nd–0.11) 0.05 (nd–0.44) NA

PCB 187 0.03 (nd–0.13) 0.05 (nd–0.52) 0.03 (nd–0.13) 0.18 (0.05–0.53) 0.06 (0.02–0.25) 0.01 (0.01–0.54) 0.13 (0.02–1.1) NA

Sum of 14 PCBs 0.52 (0.11–2.2) 0.52 (0.12–5.5) 0.43 (0.13–1.4) 2.7 (0.70–9.4) 0.82 (0.23–2.7) 1.6 (0.12–9.4) 1.9 (0.17–22) 2.3 (0.17–16)

NA= Not available; nd = not detected.

Source: Butler Walker et al. (2003).

Source: Muckle (2000) and Muckle et al. (2001a,b).

Aroclor 1260 = 5.2 (PCB 153 + 138) (Weber, 2002).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 177

In contrast to the above pattern, the patterns

demonstrated by h-HCH (beta-hexach lorocyclohex-

ane) and DDE (dichlorodiphenyl dichloroethylene),

a metabolite of the pesticide DDT, are quite different.

Levels of h-HCH are 5 to 12 times higher among

individuals in the Other group than levels seen in

Inuit, Caucasians, or Dene and Me´tis (Fig. 2.2.5).

The levels of DDE in the Other group are also roughly

three to six times higher than the levels observed in

Inuit, Caucasian, or Dene and Me´tis (Fig. 2.2.6). The

Other group is made up of people with African and

East Asian ancestry, and their exposure to these con-

taminants may have taken place in Africa or East

Asia, or in foods imported from these regions,

where h-HCH and DDE are still widely used.

Polychlorinated biphenyls (PCBs) are a group of

industrial compounds that were widely used in the

1950s, 1960s, and 1970s, and are now banned and

being phased out but are still widespread in the envir-

onment. PCBs have been implicated in deleterious

effects on the learning abilities of young children,

and are a major contaminant of concern in the Arctic.

Inuit mothers have the highest levels of PCBs mea-

sured as Aroclor 1260 compared to Caucasians, Dene

and Me´tis, and the Other mothers (Table 2.2.2 and

Fig. 2.2.7). When the Inuit from all regions are com-

pared, it can be seen the Baffin Inuit have the highest

levels of PCBs (8.0 Ag/L).

When PCBs are examined on a congener-specific

basis, the more highly chlorinated congeners (PCBs

128–187) exhibit a similar pattern to that seen for

PCBs as Aroclor 1260. Concentrations of PCB con-

geners are found in the following order: PCB

153 N 138 N 180 N 187, etc. There are no marked differ-

ences in congener patterns among the various ethnic

groups. The Inuit have a slightly greater proportion of

the most predominant PCB 153 when compared to the

Caucasian and Dene and Me´tis groups, and also have

slightly smaller proportions of PCBs 28, 52, 101 and

128 (see Table 2.2.2).

A number of studies have shown that some PCBs

have dioxin-like activity. Twelve of the 209 PCBs

share certain toxicological properties with dioxins

and furans and have been assigned dioxin-like toxi-

city equivalency factors (TEFs) by the World Health

Organization (Van den Berg et al., 1998). The con-

centration of these bdioxin-likeQ PCBs can be multi-

plied by their TEF to calculate a PCB toxic

equivalency (TCDD–TEQ) which is then added to

the dioxin/furan component for a total TCDD–TEQ.

There have not been many dioxin, furan and dioxin-

like PCB analyses conducted under the Canadian

Northern Contaminants Program maternal blood con-

taminant monitoring progra m due to the expense and

large sample volumes needed for these analyses.

Recently some analyses of composite samples were

undertaken to assess the relative contributions of

dioxins, furans and dioxin-like PCBs in Arctic Cana-

Table 2.2.3

Dioxins and furans and PCBs in maternal blood

Ethnicity 2,3,7,8, TCDD–TEQs

(ND =0) (pg/L plasma)

PCBs D+F

D+F+PCBs % PCBs

Inuit 117.8 29.8 147.7 80%

Caucasian 42.0 115.0 157.0 27%

Dene 54.0 67.0 121.0 45%

TEQs= toxic equivalents.

D+F=dioxins and furans.

Table 2.2.4

Cord and maternal contaminants (lipid weight basis)

Contaminant N

Cord

Maternal

Ratio

Correlation

p value

B-HCH 116 68 59 1.4 0.87 0.0001

Hexachlorobenzene 335 37 40 1.1 0.95 0.0001

Oxychlordane 143 32 54 0.63 0.96 0.0001

PCBs (Aroclor 1260) 303 360 454 0.84 0.96 0.0001

Source: Van Oostdam (2001).

Abbreviations: B-HCH, beta-hexachlorocyclohexane; PCBs, polychlorinated biphenyls.

Sample size: cord–maternal pairs.

Concentration (Ag/kg lipid, arithmetic mean).

Cord/maternal blood (paired data only).

Pearson’s correlations.

Statistical significance of cord/maternal Pearson’s correlations.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246178

dian populations (Van Oostdam and Feeley, 2002). In

Table 2.2.3 the results are outlined for the 11 com-

posite analyses for maternal blood from some of the

mothers in the same groups previously described for

organochlorines and PCBs. The Caucasian mothers

had slightly higher levels of total TCDD–TEQs

(dioxins and furans a nd dioxin-like PCBs 2,3,7,8-

TCDD–TEQs) than Inuit or Dene mothers (157 vs.

148 and 121 pg/L, respectively) but the relative

contributions of dioxin-like PCBs are very different

in the three groups. Among Inuit mothers 80% of the

total TCDD–TEQs are accounted for by dioxin-like

PCBs while Caucasian mothers have only 27% of

the total TCDD–TEQs coming from PCBs. The

Dene mothers have a more equal contribution of

dioxins and furans and dioxin-like PCBs in the over-

all total TCDD–TEQs. The main source of exposure

for dioxins and furans, PCBs and many other orga-

nochlorine contaminants is through diet. Due to the

expense and more difficult analyses there are few

data on levels of dioxins, furans and dioxin-like

PCBs in Arctic traditional/country foods. In spite

of this, it seems reasonable to speculate that Cauca-

sian mothers obtained their higher levels of dioxins

and furans in their diet from imported foods while

Inuit mothers likely obtained their higher levels of

dioxin-like PCBs from traditional/country foods such

as marine mammals whi ch contain higher levels of

PCBs (well documented).

Comparisons can be also made based on contami-

nant levels found in cord blood. A similar pattern to

that seen in the maternal data emerges for various

contaminants, although the magnitude does vary in

maternal and cord blood for contaminant data from

the NWT and Nunavut (Van Oostdam et al., 2001).

Since lipid levels vary markedly between maternal

and cord blood, cord-to-maternal comparisons are

best made on a lipid weight (lw) basis. The strong

correlation between cord and maternal blood levels

can be seen in Table 2.2.4; correlations range from

0.87 to 0.96 and all are highly significant. These

correlations indicate that either cord or maternal

blood levels may be used in health impact studies of

these contaminants. Some organochlorines (lipid

Table 2.2.5

Worldwide comparisons of maternal blood levels of PCBs (Aroclor 1260) and h-HCH (geometric means, Ag/L plasma)

Contaminant NWT/Nunavut/Nunavik Inuit

a,b

NWT other

Greenland

Inuit

c,d

Arctic Russia

(non-indigenous)

e,f,g

India

(indigenous)

PCBs (Aroclor 1260) 2.4–8.0 1.1 6.4–36 3.3–4.3

Beta-HCH 0.04–0.11 0.48 0.07–0.24 0.31–3.1 127

Source: Butler Walker et al. (2003).

Source: Muckle (2000) and Muckle et al. (2001a,b).

Source: Deutch (2001).

Source: Deutch and Hansen (2000).

Source: Klopov et al. (1998).

Source: Klopov (2000), Klopov and Shepovalnikov (2000), and Klopov and Tchachchine (2001).

Source: Odland (2001).

Source: Sharma and Bhatnagar (1996).

Table 2.2.6

Current and historic levels

of mercury in maternal hair (Ag/g)

Community Year(s) tested Age group (years) No. tested Range Geometric mean GSD

Min Max

Inuvik+region 1976–1979 15–45 70 0.5 41.4 4.16 2.91

Inuvik+region 1998–1999 15–45 85 b LOD 11.5 0.49 2.74

Tuktoyaktuk 1976–1979 15–45 48 0.5 41.2 4.62 2.76

Tuktoyakuk 1998–1999 15–45 7 b LOD 11.5 2.29 5.69

Source: Snider and Gill (2001).

LOD: below analytical method detection limits (0.4 Ag/g).

Peak exposure levels reported as parts per million (ppm) in hair.

GSD: Geometric mean standard deviation.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 179

weight basis) such as PCBs, oxychlordane, and DDE/

DDT are found at lower levels in the cord blood than

the maternal blood, thus resulting in cord-to-maternal

blood ratios of 0.6 to 0.9. PCB congeners in cord

blood were also 0.7 to 0.9 times the mat ernal concen-

tration. Some organochlorines such as h-HCH, hexa-

chlorobenzene, and heptachlorepoxide showed the

opposite pattern, with 1.4 to 3.2 times higher levels

being found in the cord blood compared to maternal

blood. Similar patterns for a number of contaminants

have been reported by Bjerregaard and Hansen (2000)

for Greenland Inuit. Although many researchers have

taken the ratio between the cord and maternal blood

concentrations to be a static relationship, recent work

by Van Oostdam et al. (2001) has indicated that this

ratio does vary by contaminant concentration, with the

highest cord-to-maternal ratios being found at the

lowest concentrations.

Comparisons to other populations in the world

can easily be made. No specific comparisons were

made to the southern Canadian population, as the

Caucasian population included in the NWT/Nunavut

sampling has a similar diet and resulting similar

contaminant levels as the southern Canadian popula-

tion, which can serve as a useful benchmark for

southern levels of contaminants. In comparison to

Canadian Inuit mothers from Nunavik and Nunavut/

NWT, Inuit from Greenland have markedly higher

levels of PCBs as Aroclor 1260, i.e., 2.4–8.0 Ag/L

vs. 6.4–36 Ag/L, respectively (Table 2.2.5). This

difference may resul t from consumption of more

marine mammals in the traditional Greenland diet.

Compared to Inuit mothers, low er levels of PCBs are

seen in non-Indigenous people living in Arctic Rus-

sia. In contrast, levels of h-HCH are markedly

higher in the mothers from India, followed by non-

Indigenous mothers of Arctic Russia and then

mothers in the Other group in the NWT/Nunavut

(Table 2.2.5). This indicates that DDT must still be

used in the Indian and Russian commercial food

supply or used for insect control in the local envir-

onment. The markedly higher levels of h-HCH in

mothers in India (127 Ag/L) (Sharma and Bhatnagar,

1996) are likely due to the more extensive present-

day use of h-HCH in the tropics to protect against

insect vectors in the home and on crops.

Human environmental monitoring for organo-

chlorines does not have a long history in the Cana-

Table 2.2.7

Mean concentrations of metals in maternal blood, by ethnicity and region (geometric mean (range), Ag/L whole blood)

Metal Ethnicity/region

Caucasian

(n = 134) Metis/Dene

(n = 92) Other

(n = 13) Inuit

Baffin

(n = 31) Inuvik

(n = 31) Kitikmeot

(n = 63) Kivalliq

(n = 17) Nunavik

(n = 162)

Lead 21.2 (2.1–58) 31.1 (5.0–112) 22 (5.0–44) 42 (5.0–120) 19 (2.1–102) 36.1 (6.2–178) 29 (12–64) 50 (5.2–259)

Mercury (total) 0.9 (nd–4.2) 1.4 (nd–6.0) 1.3 (0.20–3.4) 6.7 (nd–34) 2.1 (0.60–24) 3.4 (nd–13) 3.7 (0.60–12) 10.4 (2.6–44)

Mercury (organic) 0.69 (0.0–3.6) 0.80 (0.0–4.0) 1.2 (0.0–3.0) 6.0 (0.0–29) 1.8 (0.0–21) 2.9 (0.0–11) 2.7 (0.40–9.7) NA

Selenium 123 (80–184) 117 (67–160) 128 (97–156) 118 (99–152) 118 (88–151) 122 (86–171) 106 (77–156) 318 (150–1232)

Cadmium 0.43 (nd–8.5) 0.65 (nd–5.9) 0.36 (nd–3.2) 1.7 (0.03–6.2) 1.0 (nd–7.1) 1.9 (0.01–7.8) 1.4 (0.11–7.7) NA

NA= Not available; nd = not detected.

Source: Butler Walker et al. (2005).

Source: Muckle et al. (2001a,b).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246180

dian Arctic. In most cases there are only one or a few

data points to assess contaminant trends—the late

1980s to the mid-1990s. Methodological differences

in sampling strategies, and sampling high consumers

versus general mat ernal samples make it difficult to

compare the specific results of these two time points.

Fig. 2.2.2. Maternal contaminant levels in Arctic Canada: oxychlordane (Ag/L plasma).

Fig. 2.2.3. Maternal contaminant levels in Arctic Canada: hexachlorobenzene (Ag/L plasma).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 181

Recently Dallaire et al. (2003a) has analyz ed the cord

blood contaminant data by year collected by Muckle

et al., (2001a,b) and has found that there is a sig-

nificant downward trend in cord blood contaminant

levels over 1996–2000 (Fig. 2.2.8). A decrease in

traditional food consumption or a decrease in con-

Fig. 2.2.4. Maternal contaminant levels in Arctic Canada: total toxaphene (Ag/L plasma).

Fig. 2.2.5. Maternal contaminant levels in Arctic Canada: h-hexachlorocyclohexane (Ag/L plasma).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246182

taminant levels in these traditional foods have been

hypothesized as the cause of this decrease (Dallaire et

al., 2003a).

2.2.2. Levels of mercury in hair and blood

Traditional/country foods, such as fish and sea

mammals, are considered the primary environmental

Fig. 2.2.7. Maternal contaminant levels in Arctic Canada: PCBs (as Aroclor 1260) (Ag/L plasma).

Fig. 2.2.6. Maternal contaminant levels in Arctic Canada: p,pV-DDE (Ag/L plasma).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 183

pathway for mercury for Canadians living in the

Arctic. Environmental and health concerns about mer-

cury levels in human tissues in Canada were first

identified in the early 1970s in two northern Ontario

First Nations communities following contamination of

fish due to effluents from chloralkali plants (Sherbin,

1979; Health Canada, 1979). Later on, these same

concerns were raised in other Aboriginal communities

across Canada (e.g., Nelson House, Manitoba; James

Bay, Que´bec) affected by hydroelectric developments

(McKeown-Eyssen and Ruedy, 1983 ).

Following these developments, the Medical Ser-

vices Branch of Health Canada determined human

tissue concentrations and regional distribution of mer-

cury in Aboriginal peoples of Canada. These mercury

determinations (total and organic mercury) have been

done for adults, pregnant and reproductive-age

women, and children including newborns. Up to

1998, the First Nations and Inuit Health Branch

(FNIHB), formerly the Medical Services Branch of

Health Canada, carried out community health risk

assessments by analyzing hair and blood samples

obtained from more than 40,000 residents in more

than 500 First Nations and Inuit communities across

Canada (Van Oostdam et al., 1999). Since the mid-

1980s, the mercury monitoring program among Cree

First Nations of Que´bec has been continued by the

Cree Boa rd of Health and Social Services. A similar

program has been undertaken by the Governme nt of

the Northwest Territories (GNWT) Department of

Health and Social Services.

2.2.3. Population groups and studies

Specific population groups with exposure to mer-

cury through consumption of traditional/country foods

in NWT, Nunavut and northern Que´bec (Nunavik)

include four ethnic groups: Caucasian, Dene and

Me´tis, Inuit, and Other (East Asian, African). There

are extensive data sets on patterns of exposure, includ-

ing seasonal and temporal patterns, measured by hair

and blood mercury levels as biomarkers and by eating

pattern surveys, for First Nations and Inuit across

PCBs

HCB

DDE

Oxychlordane

100

150

200

250

300

350

400

450

500

1993 1995 1997 1999 2001

Year of birth

1993 1995 1997 1999 2001

Year of birth

1993 1995 1997 1999 2001

Year of birth

100

150

200

250

300

350

400

1993 1995 1997 1999 2001

Year of birth

Fig. 2.2.8. Adjusted mean organochlorine (OC) concentrations according of the year of birth: (a) PCBs; (b) DDE; (c) HCB; (d) oxychlordane

(Dallaire et al., 2003a).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246184

Canada covering a 30-year period (Wheatley and

Paradis, 1998; Health Canada, 1999). These data

sets are maintained at Health Canada, the Cree

Board of Health and Social Services of James Bay

(for the James Bay Cree since 1988) (Dumont et al.,

1998), and the Department of Indian Affairs and

Northern Development for data from the NCP over

the past 10 years (Van Oostdam et al., 1999). The

research and data on mercury from human tissue

monitoring programs by the above-mentioned sources

have led to one of the largest environmental contami-

nant research projects in Canada.

It is being carried out in northern Canada by

Dewailly and colleagues to examine detrimental

effects on children exposed to dietary mercury and

other contaminants. Recent maternal hair and mater-

nal/cord blood contaminant studies in the NWT have

given an assessment of the spatial and temporal

variation in mercury levels (Tables 2.2.6, 2.2.7 and

2.2.8)(Snider and Gill, 2001; Dewailly et al., 1999;

Butler Walker et al., 2005). The number of mothers

in each region and ethnic group varie d due to the

number of mothers eligible to be included in the

sampling procedure.

2.2.4. Maternal hair

The Inuvik region is located in the Mackenzie

Delta and Sahtu regions of the NWT. The mercury

exposure as measured in hair of women of reproduc-

tive age during 1976–1979 and 1999–1999 are sum-

marized in Table 2.2.6. Clearly there are time trends

and regional differences. Mercury levels in hair were

lower in both Inuvik and region and Tuktoyatuk

1998–1999 than those seen in 1976–1979. Levels in

1998 were approximately four times higher in Tuk-

Table 2.2.8

Worldwide comparisons of maternal blood mercury levels (Ag/L

whole blood) for women living in arctic regions

Site Year(s) collected N GM

(Ag/L) GSD

Greenland

Disko Bay 1994–1996

180 12.8

Iceland 1994–1996

40 2.9 1.3

Norway

Kirkenes 1994

40 3.4 1.2

Bergen 1994

50 3.4 1.1

Hammerfest 1994

57 2.5 0.8

Sweden 1994–1996

23 1.6 1.2

GM: geometric mean.

GSD: geometric standard deviation.

Source: Bjerregaard and Hansen (2000).

Source: AMAP (1998).

Fig. 2.2.9. Maternal contaminant levels in Arctic Canada: total mercury (Ag/L plasma).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 185

toyatuk women than Inuvik region women. Mercury

levels may now be lower in hair of women from some

NWT communities but differences in sampling strate-

gies may effect any conclusion on time trend. Some of

the results reported by Wheatley and Paradis (1998)

may be effected by resampli ng communities/residents

with previously high mercury values.

Similarly, Dewailly et al. (1999) in their studies in

the Salluit region reported that the mean (geometric)

mercury concentration in hair samples collected in

1999 was 5.7 Ag/g, a value 2.5-fold lower than the

15.1 Ag/g determined in a 1978 survey conducted by

Health Canada. The mean mercury concentrations

measured during 1999 in the 18- to 39- and the 40-

to 59-year-old age groups were also five and four times

lower ( p b 0.001), respectively, than those documented

during the 1978 survey. Dewailly et al. (2000a) has re-

evaluated these data and reported that mercury levels

in the Salluit residents’ hair were five- to six-fold

lower in 1979 and 1980 compa red to 1978. It was

pointed out that 1978 also had an exceptional harvest

of beluga and other marine species. Higher consump-

tion related to a large traditional/country food harvest

could explain why high mercury concentrations were

found in hair samples in the 1978. Dewailly et al.

(2000a) concluded that mercury levels in Salluit v ary

markedly from year to year due to variation in avail-

ability of traditional/country foods, but recent tissue

levels are similar to the levels found in the late 1970s.

2.2.5. Maternal/cord blood

The Caucasian, and Dene and Me´tis mothers were

from more than one region in the NWT/Nunavut, but

there were no significant differences in contaminant

levels between these pop ulations, so they were com-

bined into their respective groups. Inuit came from

four regions in NWT/Nunavut, and there were signifi-

cant differences among various groups, so they were

retained as separate groups. In this data set, signifi-

cantly higher levels of mercury were found in maternal

blood from Inuit women compared to Caucasian, Dene

and Me´tis, or mothers in the Other group (see Table

2.2.7 and Fig. 2.2.9). In terms of geographic represen-

tation, among individuals tested in the program Inuit in

the Baffin region had higher levels of mercury in

maternal/cord blood than those seen in the Kitikmeot,

Kivalliq and Inuvik regions. Markedly higher levels of

mercury were observed in Inuit from Nunavik. Mer-

cury has been noted to concentrate on the fetal side of

the placental circulation, and cord blood mercury con-

centrations were 1.5 and 1.8 times greater than mater-

nal concentrations in Nunavik and Nunavut,

respectively. Similar patterns have been reported by

Bjerregaard and Hansen (2000) for Greenland Inuit.

Dallaire et al. (2003a) found that a significant reduc-

tion of lead and mercury concentratio ns in umbilical

cord blood of Inuit infants born in Nunavik (Que´bec,

Canada) between 1994 and 2001 (Fig. 2.2.10).

Some of the recent data from Arctic countries on

mercury levels in maternal blood are shown in Table

2.2.8. No specific comparisons were made to the

southern Canadian population, as the Caucasian popu-

lation included in the NWT/Nunavut sampling has a

similar diet and therefore will have similar contami-

nant levels as the southern Canadian population, which

Dichotomous model

F = 20.60

;

P < 0.0001

Exponential model

F = 16.02 ; P < 0.0001

0,00

0,05

0,10

0,15

0,20

0,25

0,30

1993 1995 1997 1999 2001

Year of birth

100

120

1993 1995 1997 1999 2001

Year of birth

Fig. 2.2.10. Adjusted mean metal concentrations according to the

year of birth for (a) lead and (b) mercury (Dallaire et al., 2003a).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246186

serves as a useful benchmark for southern levels of

contaminants. Maternal and cord blood mercury levels

were low, wer e similar in Norway, Iceland and Swe-

den, and were similar to levels in Caucasians, and

Dene and Me´tis mothers in the NWT (Table 2.2.7).

Mercury levels in Greenland Inuit mothers were higher

than those seen in Inuit mothers from Nunavik and

Nunavut/NWT ( Table 2.2.7). These higher levels are

likely due to the consumption of more marine mam-

mals in the traditional Greenland diet.

2.2.6. Levels of selenium in maternal blood

Selenium levels in maternal serum are very similar

among Caucasian, Dene and Me´tis, Other, and Inuit

(Baffin, Inuvik, Kitikmeot, and Kivalliq) peoples from

the NWT/Nunavut (range of 117–128 Ag/L) (see

Table 2.2.7). Seleni um levels were elevated (318 Ag/

L) only among Nunavik Inuit who had the highest

levels of mercury. High concentrations of selenium

are generally found in the traditional/country food diet

of the Inuit of Canada (and Greenland), particularly

marine mammals (e.g., whale skin). This element is a

component of the glutathione peroxidases and it is

believed that it may act as an antagonist to methyl-

mercury, thereby offering some protection against

potential adverse health effects due to methylmercury

exposure; however, there is still much controversy

over the role of selenium in methylmercury toxicity

(AMAP, 1998).

2.2.7. Levels of lead in maternal blood

Table 2.2.7 shows that lead levels in maternal

blood are moderately elevated among some of the

Inuit groups (Baffin, Kitikmeot, Kivalliq, and Nuna-

vik) and the Dene and Me´tis (range of 29–50 Ag/L),

while Caucasians, Other, and Inuvi k Inuit have lower

levels (range of 19–22 Ag/L) (Butler Walker et al.,

2005). Research on lead isotope signatures has indi-

cated that these elevations of blood lead levels are

likely due to the use of lead shot in the hunting of

traditional/country foods, and thus its presence in the

wild game consumed (Dewailly et al., 2000b).

In an international context, lead levels are moder-

ately elevated among some of the Inuit mothers from

Greenland (range of 31–50 Ag/L), and less so among

women from Siberian Russia (range of 21–32 Ag/L)

compared to other circumpolar countries or regions

(AMAP, 1998).

2.2.8. Levels of cadmium in maternal blood

Table 2.2.7 shows that the Inuit have the highest

mean levels of cadmium in maternal blood, ranging

from 1.0 Ag/L in Inuvik to 1.9 Ag/L in Kitikmeot.

Baffin and Kivalliq Inuit have levels of 1.7 Ag/L and

1.4 Ag/L, respectively. These values are roughly 1.5–5

times higher than blood cadmium levels of Dene and

Me´tis (0.65), Caucasians (0.43), and Other (0.36).

Note that of these three latter groups, blood cadmium

levels were highest among the Dene and Me´tis, but

still lower than any of the Inuit groups. This difference

is likely due to the high rate of smoking among Inuit

mothers and the high cadmium content in Canadian

tobacco (Benedetti et al., 1994). It may also result

from the consumption of marine and terrestrial mam-

mal liver and kidney. Although liver and kidney can

contain significant amounts of cadmium, Elinder

(1992) has shown that cadmium absorption from

food is only 3% to 7% while the absorption of inhaled

cadmium is from 10% to 60%.

2.2.9. Radionuclide exposure

Van Oostdam et al. (1999) identified the lichen

caribou human food chain as the most significant radia-

tion exposure pathway for Arctic residents. Caribou are

a major source of meat in many northern communities.

These animals graze over wide areas on slowly grow-

ing lichens, which can accumulate deposited radionu-

clides and other contaminants over decades . Most of

the radiation exposure received through this food chain

resulted from anthropogenic radiocesium isotopes and

from the naturally occurring lead-210 polonium-210

decay chain (Van Oostdam et al., 1999).

2.2.9.1. Radiocesium. The radiocesium isotopes—

cesium-137 (

137

Cs) and small traces of cesium-134

(

134

Cs)—entered the environment as a result of atmo-

spheric testing of nuclear weapons. The most intense

period of testing occurred in the late 1950s and early

1960s. At that time

137

Cs was contributing a radiation

dose as high as 5 mSv (millisieverts) per year to some

northern residents (Tracy et al., 1997), in addition to a

normal radiation background dose of about 2.4 mSv/

year (UNSCEAR, 1993). After the cessation of atmo-

spheric testing by the major powers in 1963, levels of

137

Cs in the northern environment decreased drama-

tically but have persisted at lower concentrations right

up to the present. This is due to the long half-life (30

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 187

years) of

137

Cs and to its slow turnover rate in Arctic

ecosystems. Also, until 1980, there were small inje c-

tions of radiocesium into the atmosphere from con-

tinued testing by non-signatories of the 1963 Limited

Test Ban Treaty. On September 10, 1996 the General

Assembly of the United Nations adopted a Compre-

hensive Test Ban Treaty banning the testing of all

nuclear weapons in all environments for all time

(Preparatory Commission, 1996). A number of

nations have yet to ratify this Treaty and in 1998

India and Pakistan conducted several underground

nuclear tests, although none of these tests were

observed to have released any measurable radioactiv-

ity to the environment.

Large quantities of radiocesium (

137

Cs and

134

Cs in

a ratio of about 2 to 1) were released from the Cher-

nobyl nuclea r reactor accident in 1986. Although fall-

out from this accident had a major impact on reindeer

herds in Scandinavia (Xhman and Xhman, 1994), the

effect on Canadian caribou herds was to increase

radiocesium levels by about only 20% above residual

fallout levels (Marshall and Tracy, 1989; MacDonald,

2002). Because of its shorter half-life (2 years),

134

disappears from the environment much more quickly

than

137

Cs.

Tracy et al. (1997) gave details on whole-body

measurements of radiocesium carried out by Health

Canada on northern peoples in 1989–1990. The aver-

age radiation doses for northern communities at that

time were all less than 0.1 mSv/year and the highest

single dose was 0.4 mSv/year. These are insignificant

compared to normal background radiation doses of 2 to

3 mSv/year. Since that report, recent measurements of

radiocesium levels in Canadian caribou herds have

shown that levels are decreasing further, with an eco-

logical halftime of about 10 years (MacDonald et al.,

1996).

2.2.9.2. Lead-210 and polonium-210. Naturally

occurring uranium in rocks and soils releases radon

gas into the atmosphere. Van Oostdam et al. (1999)

showed that radon gas itself is not a particular problem

in the North. However, the decay of radon in air gives

rise to the long-lived lead-210 (

210

Pb; half-life = 22.3

years), which in turn decays to the alpha-emitting

radionuclide polonium-210 (

210

Po; half-life = 138

days). These radionuclides eventually settle onto

lichens and other vegetation in more or less equal

concentrations. When the lichens are eaten by caribou,

210

Pb is concentrated in bone and

210

Po accumulates in

the meat and organs of the animal, which are in turn

eaten by northern peoples. Van Oostdam et al. (1999)

indicated

210

Po doses as high as 10 mSv/year for some

northern communities consuming large amounts of

caribou. The report recommended that more research

be carried out to better characterize this source of

radiation exposure.

The radionuclides

210

Pb and

210

Po are naturally

occurring and have been present in the Arctic environ-

ment for thousands of years at more or less the same

concentrations as found today. Local enhancements

may have resulted from uranium mining, milling, and

ore transport operations, although there is no evidence

that any such contamination is widespread.

Since the publication of Van Oos tdam et al. (1999),

extensive dietary surveys have been carried out by

CINE (see Section 2.1) which more accurately reflect

the amount of polonium-210 ingested through caribou

meat and other traditional/country foods. For exam-

ple, in a 24-h recall survey, Berti et al. (1998b)

obtained a caribou consumption of 224 g/day for

men and 178 g/day for women living in Gwich’in,

NWT, who were over 40 years of age. These values

have been corroborated independently by whole body

surveys of

137

Cs in northern residents and in the meat

they have been consuming (Tracy and Kramer, 2000).

Concentrations of

210

Po in Canadian caribou meat are

known to vary from 10 to 40 Bq/kg, with an average

of about 20 Bq/kg (Thomas, 1994). Until recently,

questions remained as to how much polonium is

absorbed in the human inte stine and how long it

remains in the body. A study by Thomas et al.

(2001) shows that intestinal absorption of polonium

is indeed more than 50% and that the retention time in

the human body is significantly longer than the pre-

viously accepted value of 50 days. All of these factors

combined would give a radiation dose of 3 to 4 mSv/

year to a high consumer of caribou meat, over and

above normal background radiation.

2.2.9.3. Summary of radionuclide exposures. Infor-

mation for the two most important radionucli-

des—

137

Cs and

210

Po—contributing to the radiation

exposure of northern residents is summarized in Table

2.2.9. The table gives concentrations of the two radio-

nuclides in caribou meat, concentrations in human

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246188

tissues where available, and the radiation doses to

humans consuming a traditional diet of caribou and

other traditional/country foods. All of these doses

must be interpreted in the context of a worldwide

average exposur e to natural background radiation of

2.4 mSv/year (UNSCEAR, 1993). The radiocesium

doses have decreased to insignificant levels. Barring

future inputs from another nuclear reactor accident or

from a resumption of nuclear weapons testing, the

radiocesium levels in northern food chains should be

completely undetectable in a few years. The

210

doses remain somewhat elevated over and above nor-

mal background radiation (2.4 mSv/year).

2.3. Trends in traditional/country food dietary intakes

and contaminant exposures

Interview data from Inuit communities in 1999–

2000 included questions about how individuals com-

pared their intake of traditional/country food at that

time to intake 5 years earlier. More than 50% of the

entire sample from all Inuit communities reported that

they were consuming the same or more than they did

previously. When asked why more traditional/country

food was not consumed by the family, most men-

tioned the expenses of hunting and other priorities

for their tim e. Many individuals (42–68% of respon-

dents in the five regions) expressed their wish to be

able to purchase traditional/country food in the com-

mercial food stores in their communities because they

did not have time to harvest these foods themselves

(Kuhnlein et al., 2000).

In addition to the studies conducted by CINE, a

review of circumpolar Inuit dietary studies was con-

ducted by Dewailly et al. (2000d) . The study collected

and reviewed all available dietary studies from the

circumpolar Inuit regions to conduct an analysis of

data coverage, and temporal and geographic trends in

consumption, contaminant level s and nutrient status.

Research conducted in Canada covered in this review

included: the Sante´ Que´bec study in Nunavik (1995);

Lawn and Langer (1994) survey in Labrador and the

NWT; a survey in Broughton Island by Kuhnlein

(1989); research conducted by Wein et al. (1998) in

the Belcher Islands, NWT; and a study by Wein and

Freeman (1992) in Aklavik, NWT.

Results obtained from recent studies for the Cana-

dian Nor th were compared to the Nutrition Canada

Eskimo Survey conducted in 1971. The analysis

shows that in 1971, daily traditional/country food

consumption varied between 191 and 219 g/day

among women, while amounts today vary between

164 and 448 g/day. Among men, traditional/country

food intakes varied between 277 and 646 g/day in

1971 and are now between 171 and 615 g/day, as

reported in the more recent studies. In Canada, intakes

of traditional/country food (by weight) do not seem to

have changed significantly in the last 20 years.

Changes in the Inuit diet are primarily related to an

increase in the amount of imported foods consumed.

More cereals, meats, fruits, vegetables and sugars are

included today than before. Mean energy intakes

among Inuit were lower in 1971 than today, and the

contribution of protein to total energy intake has

decreased in the last 30 years while the total lipid

intake has increased on average. According to the

Sante´ Que´bec study this is attributable in Nunavik

to a higher consumption of market meats and prepared

meals. This increase is also related to the degree of

urbanization of Inuit communities (Schaefer and

Steckle, 1980). Despite the fact that higher nutrient

intakes are observed today compared to the last 20–30

years, several recent studies still show low intakes of

Table 2.2.9

Radionuclide levels in caribou meat and people in the Canadian

Arctic, and resulting radiation doses to people

Levels/Doses Radionuclide

Cesium-137 Polonium-210

Levels in caribou meat (Bq/kg)

Mid-1960s 500–3000 10–40

Late 1980s Up to 700 10–40

Late 1990s Up to 300 10–40

Levels in northern residents (Bq/kg)

Mid-1960s Up to 1500

Late 1980s Up to 110

Doses to northern residents (mSv/year)

Mid-1960s Up to 5

9–12

Late 1980s Up to 0.4

3–4

Late 1990s Up to 0.2

3–4

Source: Tracy and Kramer (2000).

Doses based on measured whole-body concentrations of

137

(Tracy et al., 1997).

Doses based on estimated caribou consumption in a typical

northern diet and on human metabolic parameters. The higher

210

Po doses in the 1960s is not based on any changes in environ-

mental levels of

210

Po but on an estimated higher consumption of

caribou meat at that time (Tracy and Kramer, 2000).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 189

vitamins A and C, and folic acid and calcium among

Inuit populations (Dewailly et al., 2000d).

Similar dietary intake measurements in Qikiqtar-

juaq (Broughton Island), Nunavut were taken in

1987–1988 and 1999. The earlier study, however,

was more intensive in that data were collected

bimonthly for the entire year, whereas in the latter

study data were collected during two periods repre-

senting the estimated highest and lowest intake sea-

sons. In both studies, measurements of mercury,

chlordane, PCBs and toxaphene were made in foods

eaten, for development of the food database. It was

shown that mercury exposure data were similar in the

two surveys, thus implying little change in dietary

pattern or mercury concentration in food. However,

reported intakes of the organochlorines were some-

what higher in the latter survey, particularly among

the highest users (95th percentile). This was due to a

higher reported consumption of narwhal muktuk and

blubber in the community (Kuhnlein et al., 1995c).

3. Toxicology

3.1. Priority contaminants

3.1.1. Toxaphene

Toxaphene, or chlorinated bornane, was one of the

most extensively applied insecticides in North Amer-

ica during the 1960s and 1970s. Encouraged as a

replacement pesticide for DDT, total global use esti-

mates for toxaphene are on the order of up to 1.2

billion kg (McDonald and Hites, 2003). Due to its

environmental persistence (soil half-life of up to 14

years) and ability to be trans ported in the atmosphere,

toxaphene continues to be a contaminant of concern

for Arctic populations who consume country or tra-

ditional food items. A number of dietary surveys

conducted in the Canadian Arctic have provided

intake estimations which suggest high level consu-

mers of marine mammal fatty tissues and organs

(blubber, liver, brain, and kidneys) may exceed the

current tolerable daily intake (TDI) guideline for

toxaphene, in certain cases by over two orders of

magnitude (Section 5.1.1). This higher intake of tox-

aphene from the diet can then result in increased

concentrations of residues being detected in human

tissue samples. For example, total toxaphene residues

detected in maternal blood samples from Caucasians

living in the Arctic were found to be over 10-fold

lower than Inuit (0.05 ppb vs. 0.74 ppb, respectively)

(Section 2.2).

Toxicological issues related to toxaphene, which

in part led to its deregistration as a pesticide,

included it being classified as a suspected carci nogen

and endocrine disruptor. To further address these

aspects, with an ultimate goal of possibly revising

the current Health Canada TDI of 0.2 Ag/kg body

weight (bw)/day, a number of experimental studies

were devised.

Toxaphene’s ability to interact with DNA, a critical

event in the development of certain cancers, was

tested in a series of in vitro studies. In a standard

mutagenicity assay, technical toxaphene was tested in

five strains of Salmonella typhimurium for its ability

to induce point mutations (Schrader et al., 1998). This

test is commonly employed as an initial screen for

chemical genotoxic activity, with the premise being

that DNA point mutations are implicated in many

human genetic diseases, including tum our formation.

Results indicated that while toxaph ene was able to

induce mutations in bacteria, the addition of a meta-

bolic activation system significantly decreased its

activity. In addition, toxaphene was tested for its

ability to induce sister chromatid exchanges (SCE)

in hamster lung fibroblasts (V79 cells). Although

not related to mutation frequency, an increase in

SCE beyond a background value indicates a chemical

has caused genetic (DNA) damage, such as adducts,

which is correlated to recombinational DNA repair.

Even at cytotoxic concentrations, toxaphene was only

able to induce nonsignificant increases in SCE.

Technical toxaphene is a complex mixture of

chlorinated bornanes, theoretically comprised of up

to 32,000 different congeners. As with PCBs, various

abiotic and biotic processes results in selective decom-

position of the technical formulation to the extent

where only 100–200 congeners are detected in envir-

onmental samples (Gouteux et al., 2002; MacLeod et

al., 2002). To address possibly toxicological differ-

ences that may exist between technical toxaphen e and

benvironmentalQ toxaphene, technical toxaphene was

fed to cod for 4 months and then toxpahene residues

(biotransformed) were extracted and purified from cod

liver (Besselink et al., 2000). Following an initial

exposure to a known liver carcinogen, N-nitroso-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246190

diethylamine, female rats were treated for 20 weeks

with 0.46–12.5 mg cod liver-extracted toxaphene

(biotransformed)/kg bw/week by subcutaneous injec-

tion. One week following the last dose, rat livers were

removed and assessed histopathologically for altered

hepatic foci, a preneoplastic lesion associated with the

development of liver cancer. Unlike the positive con-

trol, TCDD, neither technical toxaphene or toxaphene

extracted from cod liver significantly increased the

mean foci area or percent area fraction of foci in the

liver. Overall, these preliminary results would suggest

that biotransformed toxaphene does not have strong

tumour promoting activity.

Following a preliminary study where nonhuman

primates were dosed with technical toxaphene for 52

weeks (1.0 mg/kg bw/day), an additional longer term

study was initiated. Groups of 10 female cynomolo-

gus monkeys were given daily oral doses of 0, 0.1, 0.4

or 0.8 mg technical toxaphene/kg bw/day for at least

75 weeks. After 75 weeks, treated females were mated

with control males, with continued toxaphene expo-

sure for the additional 26-week mating and gestation

period. Prior to the mating phase of the experiment,

immunological assessment of the treated females was

conducted between weeks 33 and 70 (Tryphonas et

al., 2001).

No parameters related to fertility and gestation

were affected by toxaphene dosing. One of the more

prominent offspring observations was that male neo-

nates from the toxaphene treated dams weighed less at

birth and grew at a slower rate during lactation than

the controls, whereas the opposite effect was seen

with fema le offspring. At the highest dose, male new-

borns weighed about 10% less than controls while on

average females from all treated dams weighed

approximately 24% more than the controls.

During immunologic assessment of the treated

females, effects related mainly to humoral immunity

(antibody mediated) were observed. The primary and

secondary response (IgM and IgG) to T-cell-depen-

dent antigens (sheep erythrocytes and tetanus toxoid)

was decreased in the animals dosed with 0.4 and/or

0.8 mg toxaphene/kg bw/day. These results suggest

that the animals could possibly be more susceptible to

bacterial infections.

3.1.1.1. Discussion. Although questions remain as

to the significance from a human health perspective of

the tumours induced in experimental animals by tech-

nical toxaphene, short term in vitro and in vivo studies

suggest toxaphene would be classified as a weak

mutagen and an equivocal tumour promoter. Asso-

ciated with toxaphene-induced liver cancer in rodents

were observations of hepatic toxicity (liver weight

increase, fatty cell degeneration, focal necrosis). In a

study designed to investigate a possible link between

liver toxicity in humans and toxaphene, 104 autopsy

samples from Greenland Inuit, collected from 1992 to

1994, were assessed for histopathological findings as

well as concentrations of contaminants (Dewailly et

al., 1995). The average sum (mean) of chlordane-

related chemicals in a representative sample of omen-

tal fat (n = 41) was 2.80 ppm (range of 0.57–9.61

ppm) while toxaphene (detected in 100% of samples)

was found at lower levels (mean = 0.57 ppm; range of

0.09–1.85 ppm). No differences were seen in the

levels of contaminants in samples for which liver

pathologies (fatty degeneration, p ortal fibrosis, and

inflammation) were or were not present. While not

designed to provide conclusive evidence, the results

are at least suggestive that the toxaphene concentra-

tions found are not associated with detectable symp-

toms of liver toxicity in humans.

From the main reproductive study in no n-human

primates, ingestion of up to 0.8 mg toxaphene/kg bw/

day had no immediate effect on fertility. In compar-

ison, the 95th percentile estimated daily intake of

toxaphene by Inuit in Qikiqtarjuaq from a 1998–99

dietary survey was 26 Ag/kg bw/day, or approximately

30-fold lower (Secti on 5.1.1). A similar comparison

can be made with respect to body residues accumu-

lated through the diet. For example, body burdens

(total bioaccumulated) as estimated from the non-

human primate feeding study would be in the order

of 2.9, 13.6 and 32.6 mg for the 0.1, 0.4 and 0.8 mg/

kg bw/day doses, respectively (Arnold et al., 2001;

Arnold, 2003). Average toxaphene body burdens

found in Baffin and Kivalliq women who participated

in a maternal blood study would be 1.1 and 9.1 mg,

respectively based on total lipid, or within the same

range where immunologic effects and infan t weight

changes were seen ( Butler Walker et al., 2003). This

would suggest that, for certain consumers of country

or traditi onal food, exposure to toxaphene may be

sufficiently high as to warrant concerns related to

the immune system and infant physical development.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 191

3.1.2. Chl ordane

North American production of chlordane began in

1947 and peaked at approximately 5000 tons/year by

1974. Chlordane was first registered in Canada as a

pesticide in 1949 with subsequent sale and/or use

prohibited after 1995. One of the major applications

of chlordane was as a termiticide in the U.S. where its

estimated there have been over 30 million houses

treated (Kilburn and Thornto n, 1995).

Technical chlordane actually consists of up to 120

similar compounds with 5–6 major isomers predomi-

nating (Table 3.1.1). Of interest is that up to 10% of

chlordane can be the related insecticide heptachlor

while up to 20% of commercial formulations of the

latter can be chlordane.

As with other structurally-similar chlorinated insec-

ticides, chlordane was identified as a contaminant of

concern in certain Arctic communities based on esti-

mated daily intakes exceeding recommended guide-

lines as established by various health agencies. Mean

intake of chlordane by the Inuit community in Qikiq-

tarjuaq (Broughton Island) in 1998 was estimated at

0.62 Ag/kg bw/day. High consumers of country foods

(top 5%) were thought to be ingesting up to 5 Ag/kg

bw/day of chlordane residues. Consumption of marine

mammal blubber (beluga, narwhal, walrus) was

responsible for up to 84% of the total chlordane intake

although these three food items contributed less than

4% of the traditional diet (Kuhnlein and Receveur,

2001). In contrast, a conservative estimation is that

North Americans are only exposed from food to up to

0.01 Ag/kg bw/day of chlordane (Doughert y et al.,

2000). These intake figures are in comparison to

btolerableQ guidelines, as established by Health

Canada and US EPA, of 0.05–0.5 Ag/kg bw/day.

Following an evaluation of chlordane residues in

marine mammals by Health Canada, it became appar-

ent that there were major differences between a ctual

human exposure through consumption of these foods

and the technical chlordane mixtures used in the

safety assessments (Table 3.1.1). In particular, trans-

nonachlor (TNC), a minor component of technical

chlordane, and oxychlodane (OXY), a stable metabo-

lite of chlordanes and nonachlors, can contribute over

75% to the chlordane-related residues found in marine

mammal blubber (Muir et al., 2000). As important,

trans-nonachlor and oxychlordane account for more

than 90% of the chlordane residues found in human

milk samples ( Newsome and Ryan, 1999).

As the majority of toxicology data used to establish

tolerable intake values was generated using technical

chlordane, a major goal of NCP was to address the

lack of information on chlordane-related residues

commonly found in country foods and humans.

For the initial series of experiments, rats were

orally dosed for 28 days with cis- and trans-nonachlor

or oxychlordane and the results compared to technical

chlordane. At the highest dose tested, 25 mg/kg bw/

day, no animals treated with either cis-nonachlor or

technical chlordane died while there was significant

morbidity/mortality with trans-nonachlor and espe-

cially oxychlordane at lower doses (Table 3.1.2).

Residue analysis conducted at study termination

and after the animals had been given an addit ional

28 or 56 days following dosing indicated that oxy-

chlordane was approximately 2–3 times more bioac-

cumulative than trans-nonachlor and 19–26 times

greater than trans-chlordane. While both male and

female animals effectively cleared trans-chlordane

Table 3.1.1

Relative percent contribution to chlordane total

Isomer Technical

chlordane

(%)

bBiologicalQ

chlordane

(%)

cis-Chlordane 19 1.7

trans-Chlordane 24 1.4

cis-Nonachlor 2.7 8.3

trans-Nonachlor 9.7 54.9

Heptachlor 10 6.8

Oxychlordane 0 27

Buchert et al. (1989).

Marine mammal blubber average, 1993–1994.

As heptachlor epoxide.

Table 3.1.2

Chlordane-induced mortality in sub-acute studies

Chemical Dose

(mg/kg

bw/day)

Time to

morbidity/mortality

(days)

Mortality

incidence

(%)

Technical chlordane 25 N 28 0

cis-Nonachlor 25 N 28 0

trans-Nonachlor 2.5 N 28 0

25 17–20 43

Oxychlordane 10 1 100

2.5 26 4

1 N 28 0

Bondy et al. (2000, 2003).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246192

residues by 28 days after dosing, oxychlordane and

trans-nonachlor were significantly more persistent

especially in females, with up to 90% of oxychlordane

still present after 56 days following dosing (Bondy et

al., 2003). Histopathological indications of organ

effects (mainly liver, kidney and thyroid) were appar-

ent at the highest dose (25 mg/kg bw/d ay; 10 mg/kg

bw/day OXY) for all chemicals, but also at lower

doses for TNC and OXY (0.25 and 2.5 mg/kg bw/

day, respectively) (Bondy et al., 2003). Residue ana-

lysis also supported the concept that there is a link

between total amount of chlordane-related chemicals

bioaccumulated and indications of toxi city.

Additional experimental investigations focused on

endpoints related to immunology as previous studies

had suggested prenatal exposure to various OC pesti-

cides was associated with an increased risk of recurrent

otitis media in Inuit children. Unlike toxaphene, treat-

ment with certain of the chlordane compounds, mainly

nonachlors, resulted in a stimulatory response in terms

of Ig production. However, in female rats, there was a

dose-related decreased resistance to Listeria infectivity

with TNC, starting at doses as low as 0.25 mg/kg bw/

day (Tryphonas et al., 2003). The sex-related differ-

ence, in terms of female animals being more suscep-

tible to bacterial infections, was once again thought to

be related to amount of accumulated chlordane resi-

dues (higher and more persistent in females).

3.1.2.1. Discussion. At sufficiently high levels of

exposure, technical chlordane is capabl e of producing

a variety of toxic effects in both experimental animals

and humans. The lowest adverse effect dose in experi-

mental animals was estimated to be approximately 50

Ag/kg bw/day, based on histopathological lesions pro-

duced in liver. Long-term exposure to doses of tech-

nical chlordane as high as 250 Ag/kg bw/day produced

liver weight increases and hepatocellular swelling in

rats. In comparison, high consumers of country foods

are likely ingesting up to 5 Ag/kg bw/day of chlor-

dane-related chemicals.

Results from these series of experiments provided

indications that the more persistent and bioaccumula-

tive chlordane-related chemicals, nonachlors and oxy-

chlordane, are also more toxic than the technical

mixture. For examp le, exposure for short durations

to cis- and trans-nonachlor produced significant

effects in various immunology parameters, including

a decreased ability to mount an effective challenge

against bacterial infection. A similar response was not

observed with techni cal chlordane, even at doses 100-

fold higher. As toxicity of chlordanes appears to be

related to the extent of accumu lated residues, compar-

ison of tissue or organ concentrations between experi-

mental animals and humans may be a reasonable

approach to refining the risk assessment. At the lowest

dose tested which caused effects, 0.25 mg/kg bw/day,

total chlordane residues in experimental animals were

approximately 0.8 mg/kg bw. In comparison, the

average body burden of sum of chlordanes from a

recent Arctic survey of maternal blood was approxi-

mately 319 Ag/kg bw, or almost 25-fold lower

(Muckle et al., 2001b). However, values from this

survey ranged up to 212 Ag/kg bw, or only 4-fold

lower. This would imply that certain Arctic consumers

of countr y foods might accumulate sufficient quanti-

ties of chlordane-related chemicals to be within the

range where immunological effects are seen in experi-

mental animals.

3.2. Toxicological effects induced by exposure to food-

chain contaminant mixtures

While the majority of toxicology studies focus on

individual chemicals, from the perspective of food-

borne contaminants, exposure is almost always to

complex mixtures. These mixtures are comprised of

not only chemicals with different physicochemical

properties but also possible metabolic/degradation pro-

ducts of the original chemicals, often more toxic than

the original compounds. Methodologies have been

suggested to address the issue of risk assessment for

complex mixtures (US EPA, 2000; Groten et al., 2001).

These can involve either reconstitution or extraction of

the mixture from the matrix in question (soil, air, or

water) or, in the case of certain foods, actual incorpora-

tion of the food commodity in question into experi-

mental diets. Applicable for Arctic residents are studies

by Lapierre and colleagues in which marine mammal

fat was fed to rodents and immunological endpoints

assessed (Lapierre et al., 1999).

To further address the issue of complex mixtures of

environmental contaminants found in country foods, a

contaminant mixture was reconstituted to approximate

what is found in ringed seal blubber and fed to fema le

pigs from 4 months of age until weaning of their first

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 193

litter (Table 3.2.1)(Bilrha et al., 2004). Following

weaning, the piglets were assessed for a variety of

developmental and immunological parameters at

intervals up to 8 months of age.

There were no signs of maternal toxi city although

female pigs in the high dose group were exposed to

contaminant levels that exceeded tolerable daily

intakes by up to 2 orders of magnitude. There was,

however, a dose-dependent decrease in the length at

birth for female piglets (7% decrease at the highest

contaminant intake). Following contaminant assess-

ment in the piglets at 1 month of age, various immu-

nologic tests were conducted up to the age of 8

months. Minimal effects were seen with respect to

physical development, although there was a slight

decrease in testis weight for all animals (not dose-

dependent) and sperm motility for the high dose

animals. For immunological endpoints, one of the

more significant findings was a reduction in antibody

response to Mycoplasm a hypopneumoniae vaccina-

tion seen in the piglets exposed perinatally to the

highest dose from the age of 5 months onwards. At

8 months of age, only 10% of piglets in the high dose

group had what would be considered as a positive

response to this antigen, compared to 50% in the

controls (Bilrha et al., 2004). Additional findings

included slightly enhanced proliferative responses of

lymphocytes and enhanced phagocytic activity of leu-

kocytes, mainly evident in the mid and high dose

animals. As the findings became more prevalent as

the animals aged, it was suggested that innate or

natural immune defense was less affected as compared

to acquired immunity.

When assessed at delivery, the pigs showed a dose-

dependent increase in plasma concentrations for most

of the OCs found in the complex mixture. For exam-

ple, low, mid and high dose group plasma concentra-

tions of PCBs (sum of congeners) were 153, 1425 and

11,485 ppb, respectively (controls not detected). In

comparison, the mean concentration of PCBs in

maternal plasma samples from an Inuit cohort in

Nunavik was 397 ppb lipid, with a range up to 1951

ppb lipid, or only 6-fold lower than the high dose

group (Muckle et al., 2001b).

The same complex mixture of contaminants was

tested in vitro for its ability to affect the maturation

and embryonic development of porcine oocytes.

While doses of contaminants almost 1000-times

higher than those seen in plasma of Inuit women of

reproductive age had no effect on the fertilization of

oocytes, there was a significant decrease at the highest

dose tested of polyspermy (Campagna et al., 2001).

However, there were indications that the in vitro

maturation process and subsequent developmental

Table 3.2.1

Composition of the organochlorine mixture

Chemical Weight percent in mixture (wt.%)

PCB mixture

32.6

Technical chlordane 21.3

p,pV-DDE 19.3

p,pV-DDT 6.8

Technical toxaphene 6.5

a-HCH 6.2

Aldrin 2.5

Dieldrin 2.1

1,2,4,5-Tetrachlorobenzene 0.9

p,pV-DDD 0.5

h-HCH 0.5

Hexachlorobenzene 0.4

Mirex 0.2

g-HCH 0.2

Pentachlorobenzene 0.2

Source: Ayotte (2001).

Mixture containing 2,4,4V-trichlorobiphenyl (320 mg), 2,2V,4,4V-

tetrachlorobiphenyl (256 mg), 3,3V,4,4V-tetrachlorobiphenyl (1.4

mg), 3,3V,4,4V,5-pentachlorobiphenyl (6.7 mg), Aroclor 1254 (12.8

g), and Aroclor 1260 (19.2 g).

Table 3.2.2

Composition of complex mixture based on human blood residues

Chemical Amount in 5 mg/kg bw dose (mg)

PCB mixture

1.100

Aldrin 0.005

h-HCH 0.075

cis-Nonachlor 0.052

p,pV-DDE 0.919

p,pV-DDT 0.057

Dieldrin 0.022

Hexachlorobenzene 0.296

Heptachlor epoxide 0.023

Mirex 0.029

Oxychlordane 0.136

Toxaphene 0.07

trans-Nonachlor 0.220

A OCs 1.900

Methylmercury chloride 2.000

Source: Bowers et al. (2003).

Containing PCBs 28, 52, 99, 101, 105, 118, 128, 138, 153, 156,

170, 180, 183, 187.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246194

competency of oocytes was negatively affected at

lower doses. When similar endpoints were investi-

gated with both oocytes and sperm being treated in

vitro, an increase in polyspermic oocytes oc curred at

lower doses while a decrease was evident at higher

doses (Campagna et al., 2002).

In a similar experimental design, pregnant rats were

exposed from gestation day 1 to the end of lactation to

a complex mixture of contaminants based on what has

been found in blood from Arctic populations known to

consume country/traditional food items (Table 3.2.2).

Pregnant rats were treated orally (food supplement)

with 0, 0.05, 0.5 or 5.0 mg/kg bw/day of the mixture

from gestation day 1 to the end of lactation (postnatal

day 23). Developmental landmarks were assessed in

the pups as well as a number of neurobehaviour end-

points up to maturity (postnatal day 60). A 15 mg/ kg

bw/day dose of Aroclor 1254 served as a positive

control (Bowers et al., 2003).

The contaminant mixture had no significant effect

on maternal gestational weight gain, litter size, pup

sex ratio or offspring mortality however maternal

body weight in the high dose mixture group decreased

by 6–10% during lactation compared to the controls.

Pup weight gain was decreased throughout lactation in

the high dose and positive control groups, with the

difference in weight increasing as the pups matured

(20% by PND 23). The decreased weight gain

observed during lactation also persisted into adult-

hood (PND 80). Maternal blood samples collected at

5–7 days after the final dose exhibited mercury con-

centrations ranging from 0.5 to 46.4 ppm for the 3

complex mixture dose groups, compared to maximum

values of 44 ppb in the previously mentioned Nunavik

cohort (Muckle et al., 2001b). Offspring blood mer-

cury values ranged from 13.0 to 250 ppb for the 3

mixture groups compared to an average cord blood

concentration of 22.7 ppb from the Nunavik cohort.

Various organ weights at PND 35 were also affected

(liver and brain increased; spleen an d thymus

decreased) in mainly the same two dose groups. Cer-

tain organ weight effects also persisted when mea-

sured at PND 75 (brain and spleen).

In young rats (17 days old), the highest mixture

dose produced time-dependent hyperactivity and

decreased rearing in a novel open-field, both sugges-

tive of altered reactivity to the environment. In con-

trast, 17-day-old pups from the Aroclor positive

control group exhibited hypoactivity and decreased

rearing, suggestive of reduced motor activity or

decreased interaction with the novel environment. At

30 days of age, hypoactivity and decreased rearing was

still evident in Aroclor-treated pups compared to the

controls, but to a lesser extent. Acoustic startle and

prepulse inhibition of acoustic startle were evaluated in

the pups at PND 21 and 44. Acoustic startle is a basic

defensive reflex response to sudden noise while pre-

pulse inhibition (PPI) of acoustic startle is a measure of

the integration of sensory information (i.e., informa-

tion processing) in the brain and reflects basic neural

functions. Offspring from the positive control group

(Aroclor 1254) consistently exhibited a decrease in the

startle response at PND 21 and 44 while the mixture

had little impact on the startle response at PND 21. The

lowest mixture dose however decreased PPI at PND 21

but not at PND 44. Gender differences in the effects of

the mixture and the positive control treatment emerged

as animals matured. For instance, at PND 44 the two

highest mixture doses and the positive control

decreased startle in males but produced small increases

in startle responses in females. Similarly, while PPI

was unaffected in fema les, Aroclor and the two highest

mixture doses produced small increases in PPI. in

males. When visual discrimination, a measure of learn-

ing, was tested in the offspring at PND 100, only

females from the high dose mixture group showed a

decreased response.

In a further attempt to define the relationship

between exposure to complex mixtures of contami-

nants and potential adverse effects, physi ological

based toxicokinetics (PBTK) modeling was applied

Table 3.2.3

Comparison of PBTK-derived TDIs to estimated contaminant

intakes

OC contaminant Average

Inuit intake

(Ag/kg

bw/day)

Current

TDI

(Ag/kg

bw/day)

Estimated TDI

from modeling

rat data

(Ag/kg bw/day)

Hexachlorobenzene 0.12 0.27 2.7

Chlordanes 0.43 0.05 0.02

HCHs (a, h, g) 0.03 0.3 4.4

Total DDT 0.53 20 18

Mirex 0.01 0.07 0.21

PCBs (sum 14 congeners) 0.66 1 17

Dieldrin ND 0.1 18.5

Source: Chan et al. (1997, 2000). ND—not determined.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 195

to residue data obtained following a subacute dietary

exposure period. Female rats were dosed (gavage) for

28 days with an organochlorine mixture based on

average intakes from a Qikiqtarjuaq dietary survey

conducted in 1988 (Chan et al., 1997). At the end of

the dosing period, various tissues were collected for

both toxicological assessmen t and residue analysis

(Chan et al., 2000). Comparison of tissue residues

found in the rats at the highest dose tested (observed

adverse effect level) to estimated equivalent human

residues from modeling suggested that the average

dietary intake of chlordane by Inuit was approxi-

mately 20 times greater than a suggested safe dose.

For all other OCs, except DDT, the tolerable intake

from the modeling exercise suggested that actual diet-

ary intakes would be less than doses capable of indu-

cing toxic effects (Table 3.2.3).

Physiologically based toxicokinetic (PBTK) mod-

els have been shown to be useful tools in predicting

the target tissue dose of chemicals in a mixture and in

extrapolating effects between species (Haddad et al.,

2000). In agreement with the chlordane toxicological

studies, these results suggest that consumers of coun-

try foods may be accumulating chlordane residues at

concentrations where adverse effects are observed in

experimental animals.

3.2.1. Discussion

Toxicological studies with complex mixtures that

mimic real life situations (exposure or body burdens)

have the added advantage of identifying any potential

interactions that may occur between the different

contaminants. For example, preli minary experimental

results suggest that PCBs and methylmercury could

act synergistically to induce adverse neurological

effects (Bemis and Seegal, 1999). Also, non-cyto-

toxic doses of toxaphene have been shown to sig-

nificantly enhance the genotoxic response of a

carcinogenic PAH in cultured human liver cells

(Wu et al., 2003).

Two series of experiments with dosing based on

slightly different complex mixtures has provided evi-

dence that: 1) organochlorines can cause effects

related to fetal development (decreased birth weight),

immunosuppression (deceased antibody production)

and possibly oocyte maturation; and 2) organochlor-

ines combined with methylmercury produced beha-

vioural alterations both similar and unique to those

seen with PCBs alone. Physical developmental delays

of offspring from the combined in utero/lactational

exposure also persist into adulthood. There were indi-

cations that certain effects from both experiments

were occurring at contaminant concentrations seen

in recent blood surveys from Arctic residents. This

would imply that these endpoints would be relevant to

assess in human studies investigating associations

between adverse health outcomes and exposure to

environmental contaminants.

3.3. Contaminant and dietary nutrient interactions

The ability of diet to affect the prevalence of

various adverse health outcomes in humans has been

clearly established. For example, its been estimated

that 20–60% of all human cancers are related to diet

(Doll, 1992). Diets rich in various antioxidants,

including vitamins and essential trace elements, are

hypothesized to have a beneficial influence on risks

associated with other major human diseases, including

diabetes, osteoporosis, kidney and cardiovascular dis-

eases (IOM, 2000). Dietary long chain polyunsatu-

rated fatty acids, such as eicosapentaenoic acid (EPA)

and docosahexaenoic acid (DHA), play an essential

role in human growth and neurodevelopment while

the same fatty acids have recently been associated

with a reduced risk of impaired cognitive function

in adults (Das, 2003; Kalmijn et al., 2002).

From an Arctic perspective, results from experi-

mental research have suggested certain diet compo-

nents can either protect against or attenuate adverse

effects caused by contaminants of concern. For exam-

ple, protein, fibre and trace elements (calcium, iron,

and zinc) can all influence the extent of cadmium

bioavailability and toxicity (WHO, 2000). Earlier stu-

dies have suggested replacing casein with fish protein

can decrease the toxicity of methylmercury in mice

(Ohi et al., 1976). Although the mechanism is unclear,

selenium has been shown to be protective against

mercury-induced toxicity, possibly through complex-

ing reactions (Watanabe, 2002; Whanger, 2001).

Developmental exposure to both a selenium deficient

diet and methylmercury can result in synergistic

effects on the neurobe havioural functions in mice

(Watanabe et al., 1999).

To further define the benefi ts associated with con-

suming country foods, a series of experiments were

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246196

devised to investigate potential interactions between

dietary components and contaminants. In a first

experiment, rats were exposed to diets containing

different types of protein (casein, fish or whey) for

4 weeks prior to dosing with methylmercury for 14

days. At the end of the methylmercury treatment

period, various organs were analyzed for mercury

content and the presence of thio barbiturate reactive

substances (TBARS), as an indication of oxidative

stress (lipid peroxidation). No mercury pharmacoki-

netic differences were seen and the type of dietary

protein had no effect on organ TBARS (Chan et al.,

2004). In a similar experimental design, the lipid or

fat component of diets was modified so rats ingested

20% by weight either seal oil, fish oil, DHA, soya oil

or beef lard for the same 4-week period followed by

14 days of oral dosing with methylmercury. Only in

rats consuming the soya-containing diet was there no

increase in TBARS concentrations in brain tis sue

suggesting dietary oleic and linoleic acids (essential

fatty acids) are important in preventing methylmer-

cury-induced oxidation damage (Chan et al., 2004).

Related studies have shown that selenium did not

prevent increases in mercuric chloride-induced

TBARS in rat liver homogenates isolated from ani-

mals maintained on a soya oil-supplemented diet

(Farina et al., 2003).

Selenium has been recognized as an essential trace

element for both animals and humans. In parti cular,

selenium is thought to be important for maintaining

membrane integrity and reducing oxidative damage to

cellular macromolecules, including lipids and DNA

(Rayman, 2000). Experimental evidence to date indi-

cates that selenium not only can decrease the extent of

methylmercury neurotoxicity but if a selenium defi-

ciency condition exists, mercury toxicity can be

enhanced. The chemical form of selenium appears to

be an important factor in its ability to influence methyl-

mercury toxicity. For example, sodium selenite

(Na

SeO

), an animal feed supplement, forms com-

plexes with mercury to a greater extent than selenium

isolated from liver (Magos et al., 1983). Both selenium

and vitamin E are known to be capable of reducing the

toxic effects induced by acute or chronic exposure to

methylmercury (Ganther, 1978). Hamsters exposed to

methylmercury concurrently with vitamin E exhibited

no signs of intoxication whereas those exposed to

methylmercury alone develop severe symptoms of poi-

soning (Chang et al., 1978). As with selenium, a diet

enhanced with vitamin E is protective against mercury-

induced oxidative damage while a deficient diet

enhances methylmercury hepatic lipid peroxidation

(Andersen and Andersen, 1993). In an experiment to

further address the relationship between methylmer-

cury and antioxidants, female rats were maintained

on diets supplemented with selenium, vitamin E or

both for 8 weeks prior to mating and during gestation.

Beginning 4 weeks before mating, all rats were dosed

with 1.25 mg methylmercury/kg bw/day and allowed to

deliver their litters. Selenium alone did not prevent

mercury-induced pup mortality while vitamin E alone

and in combination with selenium did (Chan, 2002).

To further characterize the potential toxicity of mer-

cury found in country foods, rats were fed diets in

which seal liver, a major source of methylmercury

exposure for Inuit, was added. The seal liver, contain-

ing either 20 ppm (low) or 110 ppm (high) mercury,

comprised 20% of the diet and was fed to rats for 14

days before absorption; organ distribution rates for

mercury were then determined. Whereas 16% of mer-

cury from the low dose seal liver diet was absorbed,

only 3% in the high dose diet bioaccumulated which

indicates the chemical form of mercury in seal liver is

substantially less bioavailable than methylmercury in

fish tissue (usually 90%+) (Chan, 2000a,b). However,

2% of both the seal liver mercury doses was distributed

to brain compared to 0% in rats fed the control diet (veal

liver with 110 ppm inorganic HgCl

). This supports

analytical findings that on average only 2% of total

mercury in ringed seal liver was in the form of methyl-

mercury where as up to 53% of mercury was estimated

to be insoluble mercuric selenide (HgSe) (Wagemann et

al., 2000). Previous experimental results have indicated

a total mercury dose of 158 mg from seal liver did not

induce toxic effects in cats (Eaton et al., 1980).

3.3.1. Discussion

A recent review has concluded there is a link

between exposure to chemicals from the environment

and over 200 different human diseases (Janssen et al.,

2004). Certain essential dietary nutrients, including

vitamins, minerals and fatty acids, found in tradi-

tional/country foods have been suggested as being

capable of reducing or possibly preventing adverse

effects caused by contaminants. Results from preli-

minary experiments have indicated that omega-6 fatty

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 197

acids in soya oil were more effective at preventing

lipid peroxidation in brain s of methylmercury treated

mice than oils rich in omega-3 fatty acids (DHA and

EPA). In addition, mercury in seal liver is less bioa-

vailable when it occurs at high concentrati ons, possi-

bly due to the insoluble complexes it forms with

selenium. Selenium, an essential trace element for

humans, did not prevent methylmercury-ind uced pup

mortality while another antioxidant, vitamin E, alone

or in combination with selenium did. From a human

perspective, to date there has been little epidemiologic

evidence that selenium supplementation has a thera-

peutic benefit with envir onmentally associated health

disorders (Lacour et al., 2004). Furthe r studies dealing

with interactions betw een dietary components and

contaminants will have to consider both the mechan-

istic implications as well as the relevance.

4. Epidemiology and human biomarkers

Over the past decade most research efforts in Arc-

tic Canada were focussed on the characterization of

exposure of northerners. Epidemiological studies are

now in place or in preparation. In order to detect early

biological changes preceding overt diseases, biologi-

cal markers of effects were also validated and intro-

duced in epidemiological studies.

Van Oostdam et al. (1999) identified areas with

knowledge gaps where research programs should be

oriented. Under epidemiology, the following priorities

were listed:

1) Evaluate the developmental status (neurobeha-

vioural, immunological, reproductive, etc.) of new-

borns in the Arctic;

2) Evaluate fetal effects of merc ury; and

3) Evaluate kidney function among people exposed to

cadmium (via liver and kidney consumption, or

smoking).

In this section, the discussion is limited to methyl-

mercury and POPs. Other contaminants such as cad-

mium and lead are now considered of low priority

since major sources are local point sources (sm oking

for cadmium and lead shot for lead) , and information

on their toxicity is already available, validated, and

widely used by regional/national/international health

organizations. Furthermore, most health risk uncer-

tainty related to the presence of contaminants in the

Arctic food chain is du e to methylmercury and POPs.

Most of this section is devoted to prenatal exposure

and adverse developmental effects resulting in altered

immune and nervous system function early in life. In

addition, since oxidation is one of the possible

mechanisms of mercury toxicity, biomarkers of oxi-

dative stress are being investigated as biomarkers of

effect in the adult Inuit population.

Since Inuit people constitute the most exposed

group in the Canadian Arctic, this assessment is

focussed on this group.

Conducting epidemiological studies in the Arctic is

difficult and should take into account the following

specific considerations.

Mixtures of contaminants found in the Arctic.

Multi-chemical interactions of ecological ly rele-

vant mixtures (at relevant concentrations) can

have an effect different from the sum of the effects

of each component of the mixture. Arctic seafood

contains a mixture of contaminants which may

differ from those found in other parts of the

world. How important these differences are is not

known. The PCB congener profile found in human

tissues in the Arctic, however, is similar to that

found in southern Canada. Effects of mixtures need

to be better understood.

Population size. The small size of the popula tions

living in the Arctic limits the use of epidemiology.

The number of disease cases is small and case

control studies are extremely difficult to conduct.

Only cross-sectional and cohort designs are possi-

ble. Health outcomes most amenable to epidemio-

logical research in the Arctic are those with a range

of stages from normal to abnormal, and which can

be measured in all individuals (e.g., neurodevelop-

ment, immune parameters, bone density, and ferti-

lity parameters), and are not based on a stochastic

distribution (having the disease or not; e.g., can-

cer). It is probably possible to avoid this limitation

by implementing circumpolar studies.

Toxicant–nutrient interactions. The possibility that

nutrients present in seafood could modify or coun-

teract the toxicity of contaminants is highly probable

and specific to fish-eating populations. These inter-

actions need to be better understood.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246198

Genetic factors. It is important to recognize that

genetic variability may affect the susceptibility of

individuals or populations to the effects of pollu-

tants. Gene/environment interactions could also

explain why some populations or individuals are

more susceptible than others. Since few genetic

studies have been conducted in the Arctic, and con-

sidering that Aboriginal peoples have their own

genetic background, the impact of genetic poly-

morphisms that are involved in xenobiotics metabo-

lism and toxicity in the Arctic should be evaluated.

Confounders. Many health endpoints have multi-

factorial causes and environmental stressors can

contribute in varying degrees to the etiology of

these diseases. Compared to the role that lifestyle

and genetic factors play in the etiology of most

diseases, contaminants likely play a modest role.

However, this contribution is highly prevent able.

4.1. Immune system function

In children and young adults accidentally exposed to

large doses of PCBs and polychlorinated dibenzofurans

(PCDFs) in Taiwan (bYu-Cheng diseaseQ), serum IgA

and IgM concentrations as well as percentages of total

T-cells, active T-cells and suppressor T-cells were

decreased compared to values of age- and sex-matched

controls (Chang et al., 1981). The investig ation of

delayed-type hypersensitivity responses furthe r indi-

cated that cell-mediated immune system dysfunction

was more frequent among patients than controls.

Infants born to Yu-Cheng mothers had more episodes

of bronchitis or pneum onia during their first 6 months

of life than unexposed infants from the same neighbor-

hoods (Rogan et al., 1988). The authors speculated that

the increased frequency of pulmon ary diseases could

result from a generalized immune disorder induced by

transplacental or breast milk exposure to dioxin-like

compounds, more likely PCDFs (Rogan et al., 1988).

8- to 14-year-old children born to Yu-Cheng mothers

were shown to be more prone to middle-ear diseases

than were matched controls (Chao et al., 1997).

In Dutch preschool children the effects of perina-

tal background exposure to PCBs and dioxins per-

sisted into childhood and was associated with a

greater susceptibility to infectious diseases. Levels

of PCB exposure, based on serum PCB congener

153 values, of this cohort from two Dutch cities

were compa rable with those found among Canadian,

northern Que´bec Inuit (Fig. 4.1.1). Common infec-

tions acquired early in life may prevent the develop-

ment of allergy, so PCB exposure mig ht be

associated with the lower prevalence of allergic dis-

eases found in this study (Weisglas-Kuperus et al.,

2000). It should be also be noted that Canadian Inuit

from northern Que´bec have higher levels of PCB

153 than modern US populations (1990s) studied,

but similar to historical levels in the US (1960s and

1970s) and lower levels than those seen in the Faroe

Islands (Fig. 4.1.1).

U.S./11 Cities (1959–1965)

U.S./California (1964–1967)

U.S./North Carolina (1978–1982)

U.S./Michigan (1980–1981)

Netherlands/2 Cities (1990–1992)

U.S./New York (1991–1994)

Germany/Düsseldorf (1993–1995)

U.S./Massachusetts (1993–1998)

Denmark/Faroe Islands (1994–1995)

Canada/Northern Quebec (1995–1998)

0.01 0.025 0.05 0.10 0.20 0.40 0.80 1.60 3.20

g/g serum lipid

Fig. 4.1.1. Distribution of PCB 153 concentration in cord serum or plasma, 10 studies (Longnecker et al., 2003).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 199

Some of the data presented in Section 2 on human

exposure has relied on summary measures such as

PCBs as Aroclor 1260. It should be noted that PCB

153 on average was found at the highest concentra-

tions in all populations sampled (Table 2.2.2) and is

one of the key PCBs that determines overall concen-

tration of PCBs measured as Aroclor 1260.

Organic or inorganic mercury has cytotoxic activ-

ities for cellular components of the immune system in

several species of rodent s. Methylmercury, a form of

organic mercury, can alter non-specific defense

mechanisms, such as inhibition of natural killer

(NK) cell’s activity in rats and mice. It also decreases

the expression of certain activation markers of T-cells

(HLA-Dr and IL-2R) (NRC, 1992). Moreover, it has

been well demonstrated that methylmercury can affect

the function of B-cells and therefore reduce the

humoral-mediated response (Daum, 1993).

4.1.1. Clinical outcomes

In Nunavik, an epidemiological study conducted

during 1989–1991 investigated whether organochlor-

ine exposure is associated with the incidence of infec-

tious diseases and immune dysfunction in Inuit infants

(Dewailly et al., 2000c). The number of infectious

disease episodes during the first year of life of 98

breast-fed and 73 bottle-fed infants was determined.

Concentrations of organochlorines (OCs) were mea-

sured in early breast milk samples and used as surro-

gates to prenatal exposure levels. Otitis media

(middle-ear infection) was the most frequent disease,

with 80% of breast-fed and 81% of bottle-fed infants

experiencing at least one episode during the first year

of life. During the second follow-up period, the risk of

otitis media increased with prenatal exposure to p,pV-

DDE, HCB and dieldrin. The relative risk (RR) for 4-

to 7-month-old infants in the highest tertile of p,pV-

DDE exposure compared to infants in the lowest was

1.87 [95% confidence interval (CI), 1.07–3.26]. The

lowest tertile had a breast milk concentration of p,pV-

DDE of b 730 vs. N 1320 lg/kg lipid in the third or

highest tertile. The relative risk of otitis media over

the entire first year of life also increased with prenatal

exposure to p,pV-DDE (RR, 1.52; 95% CI, 1.05–2.22)

and HCB (RR, 1.49; 95% CI, 1.10–2.03). Further-

more, the relative risk of recurrent otitis media (z 3

episodes) increased with prenatal exposure to these

compounds. No clinically relevant differences were

noted between breast-fed and bottle-fed infants with

regard to biomarkers of immune function, and prena-

tal OC exposure was not associated with these bio-

markers. In this study, the potential confounders were

not fully controlled i.e., omega-3 fatt y acids, vitamin

A and smoking. For this reason, studies were designed

and initiated to address these issues.

In Nunavik, there are two ongoing studies on the

effect of OCs on immune function and infectious

disease incidence. The first one is a component of

the cohort study, where infectious diseases are mon-

itored during the first year of life (Dallaire et al., 2004).

Results showed that that prenatal exposure to PCBs

and DDE was associated with a higher incidence rate

of acute infections durin g the first 6 months of life.

Although the associations were not always statistically

significant because of limited statistical power, infants

in the highest quartiles of PCBs and DDE exposure

had systematically more episodes of infections than

their counterparts in the first quartile of exposure. This

was mostly observed during the first 6 months of life,

and the association was much weaker when infections

during the first 12 months of life were considered.

The second study is a review of all the medical

files of 350 preschool children who participated in the

Nunavik umbilical cord blood monitoring program in

1993–1996 (Dallaire et al., in preparation(b)). Preli-

minary results showed that durin g the first 5 years of

life, children in the higher quartiles of PCB prenatal

exposure had a significantly higher incidence rate of

outpatient visits for otitis medi a and LRTIs, but not

for URTIs. The association between PCB exposure

and otitis adopted a clear dose–response pattern. This

study confirms the associations previously observed

in the same population. Furthermore, it shows that the

relation between OCs and respiratory infection seems

to persist passed the first months of life.

It should also be noted, however, that some studies

have not found a link between pre or post-natal PCB

or dioxin exposure in infants and respiratory tract

symptoms but did find shifts in various types of

white blood cells related to contaminant levels (Weis-

glas-Kuperus et al., 1995). Ot her factors such as

vitamin A and omega-3 fatty acids can also modulate

immune function and vitamin A has been found to be

deficient in some Canadian Inuit populations (Blan-

chet et al., 2000). It will be important to see the full

results of the Nunavik studies when it will be possible

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246200

to address a number of the important determinants of

immune status.

4.1.2. Biomarkers

In 1997, an international symposium was held in

Bilthoven (Netherlands) to discuss the most appropri-

ate biomarkers of effect that could be used in epide-

miological studies investigating the effects of

contaminants on immune function (Van Loveren et

al., 1999). One of the stronger conclusions was to use

antibody responses to vaccination with an antigen

having no prior exposure.

4.1.2.1. Lymphocyte subsets and immunoglobulins.

In the course of the 1990 cohort conducted in Nuna-

vik, biomarkers of immune system function (lympho-

cyte subsets and plasma immunoglobulins) were

determined in venous blood samples collected from

breast-fed and bottle-fed infants at 3, 7 and 12 months

of age. Results showed that at age 3 months, concen-

trations of white blood cells (lymphocytes and more

specifically those of the CD4 subtype), were lower

( p V 0.05) in blood samples from breast-fed babies

when compared to those in the bottle-fed group. At

7 and 12 months of age, IgA concentrations were

lower ( p V 0.05) in breast-fed infants than in bottle-

fed infants. This also appeared to be the case for CD4/

CD8 ratios, although differences were not statistically

significant. None of the immunological parameters

was associated with prenatal OC exposure (Dewailly

et al., 2000c).

In the scope of the ongoing cohort study on neu-

rodevelopmental effects of Arctic contaminants

(Muckle et al., 2001a), an immune component was

added in 1998. The following immune function bio-

markers have been selected: antibody response fol-

lowing vaccination; complement system; cytokine

production by Th1/Th2 cells; and vitamin A status.

4.1.2.2. Antibody response following vaccination. An-

tibody response to vaccination is an intermediate mar-

ker of the competence of the adaptive immunity to

infections. Vaccination programs include essentially

three types of products: 1) killed vaccine (influenza,

whole-cell pertussis, inactivated polio); 2) protein-con-

jugated or protein-based vaccine (Haemophilus influ-

enzae type b, diphtheria, tetanus, acellular pertussis

vaccine, hepatitis B); and 3) attenuated live virus

(measles–mumps–rubella, varicella, BCG). Antibody

response to conjugated Haemophilus influenzae type b

(Hib) is of great interest. This vaccine is important in

Inuit children because, prior to immunization, Hib was

the most frequent cause of bacterial meningitis in Inuit

children, which was 5–10 times more frequent than in

caucasian children (Ward et al., 1986).

4.1.2.3. Complement system. The complement (CV)

system plays an important role in natural immunity

against infectious agents. It is particularly essential in

young children for whom the acquired immune system

is not yet fully developed. Deficiency of many of the

CV components is associated with increased suscept-

ibility to infections, generally of the upper respiratory

tract. In a murine model, exposure to OC compounds

increased the susceptibility to Streptococcus pneumo-

niae infections, decreased C3 levels, and lowered total

CV hemolytic activity (White et al., 1986).

4.1.2.4. Cytokine production by Th1/Th2 Cells. Or-

ganochlorines and heavy metals could modulate the

production of Th1/Th2-type cytokines. Along with

their effects on Th1/Th2-type cytokines, OCs and

metal ions are known to alter B-cell activity and to

impair host resistance to several bacterial and viral

infections (Heo et al., 1996). High levels of OCs and

metal ions in blood and tissues are frequently related to

fish consumption. Fish oil-supplemented diets (rich in

omega-3 fatty acids) have generally been show n to

reduce plasma levels of some cytokines. Most human

studies have shown decreased plasma levels or dimin-

ished production of IL-1 and TNF. Both contaminants

and omega-3 fatty acids alter the balance between Th1-

and Th2-type cytokines, and could impair host resis-

tance to infections. One of the principal limitations of

using cytokines is their high variability due to minor

infections (usually non-detected). The extremely high

incidence of minor infections in the Arctic strongly

limits the use of cytokines in epidemiological studies.

4.1.2.5. Vitamin A status. Vitamin A influences the

expression of over 300 genes and thus plays a major

role in cellular differentiation, including that of cells

related to immune response (Sommer and West, 1996;

Semba, 1994). Results from different anim al and

human studies vary; however, almost all studies

revealed that lymphopoiesis and/or maturation of lym-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 201

phocytes are altered (generally reduced) with vitamin

A deficiency (Sommer and West, 1996; Olson, 1994;

Semba, 1994). Vitamin A deficiency could increase

the frequency, severity, and duration of infections.

Lower respiratory disease was associated with vitamin

A deficiency in many cross-sectional clinical and

population-based studies. Also, otitis media was

among the first infections to be associated with vitamin

A deficiency in humans (Sommer and West, 1996;

Bloem et al., 1990; Semba, 1994).

Vitamin A clinical deficiency has never been docu-

mented in Canadian Arctic populations. However,

numerous studies have docum ented inadequate

intakes of vitamin A (Receveur et al., 1998a). A

more recent report suggests that the daily vitamin A

intake in Nunavik falls below the recommended

intake (Blanchet et al., 2000). Furthermore, persistent

organic pollutants such as OCs have been shown to

alter vitamin A homeostasis in many species, includ-

ing primates (reviewed by Zile, 1992 ). It is important,

therefore, to gain a better understanding of the rela-

tionships between vitamin A, OC levels, and infec-

tious disease incidence in Arctic populations.

In a recent pilot study, plasma concentrations of

retinol were measured in cord blood samples to assess

the vitamin A status of 135 Inuit newborns from

Arctic Que´bec and 22 newborns from the general

population of southern Que´bec. Mean retinol concen-

trations were 148.2 ng/mL and 242.8 ng/mL, respec-

tively (Dallaire et al., 2003b). Vitamin A levels of

more than 200 ng/mL have been considered as normal

vitamin A status, those between 100 and 200 ng/mL

as low, and those lower than 100 ng/mL as deficient

(Sommer and West, 1996). The difficulty of using

vitamin A as an effect biomarker of PCB exposure

is related to 1) the variability of vitamin A intake

among individuals and 2) non-systematic supplemen-

tation programs in infants. Vitamin A status is being

measured in the preschool and the newborn studies

currently underway in Nunavik.

4.2. Neurodevelopment

4.2.1. Clinical outcomes

4.2.1.1. Polychlorinated biphenyls (PCBs). Effects

of prenatal exposure to background levels of PCBs and

other OCs from environmental sources have been stu-

died since the 1980s in prospective longitudinal studies

in Michigan, North Carolina, the Netherlands and

Oswego, New York. The principal source of PCB

exposure was Great Lakes fish consumption in both

the Michigan (Schwartz et al., 1983) and the Oswego

(Stewart et al., 1999) studies, and consumption of dairy

products in the Netherlands (Koopman-Esseboom et

al., 1994a). Newborns from the North Carolina cohort

were exposed to background levels of PCBs, and there

was no specific source of exposure (Rogan et al.,

1986b). Comparison of exposure levels between

these different cohort studies is presented in Fig. 4.1.1.

Growth effects. The effects of prenatal exposure

to PCBs and other OC compounds from environmen -

tal sources on birth size and duration of pregnancy

have been investigated in the Michigan, North Car-

olina and Netherlands studies In Michigan, higher

cord serum PCB concentrations were associated at

birth with lower weight, smaller head circumference,

and shorter gestation (Fein et al., 1984; Jacobson et

al., 1990b). Similar effects were observed in the Neth-

erlands study up to 3 months of age (Patandin et al.,

1999). Prenatal PCB exposure was not associated with

birth size or growth at 1 year of age in North Carolina,

the cohort with the lowest PCB exposure (Rogan et

al., 1986a).

A recent study on infant development was con-

ducted in northern Que´bec. Prenatal PCB exposure in

this cohort is 2–3 times higher than that observed in

general populations in southern Que´bec and in Massa-

chusetts (USA); is similar to that found in Michigan and

the Netherlands; and is about 2–3 times lower than

found in Greenland and the Faroe Islands. After con-

trolling for potential confounders with the studied out-

comes, higher cord plasma PCB 153 concentrations

were associated with lower weight and height at birth,

and to shorter gestation (Muckle et al., in preparation).

These findings are consistent with the results of pre-

vious epidemiological studies conducted in popula-

tions exposed to PCBs through consumption of PCB-

contaminated food. The hypothesis of protective effects

of omega-3 fatty acids against the negative effects of

PCB exposure on birth outcomes was examined for the

first time in this study. The results indicate that the

negative effects of prenatal PCB exposure on human

growth and gestation remained significant despite the

significant beneficial effects of omega-3 fatty acids on

these same endpoints (Muckle et al., 2004).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246202

Neurobehavioural effects. Exposure to PCBs was

associated with less optimal newborn behavioural

function (e.g., reflexes, tonicity and activity levels)

in three of the four studies previously described (Huis-

man et al., 1995a; Rogan et al., 1986a; Stewart et al.,

2000). Adverse neurological effects of exposure to

PCBs have been found in infants up to 18 months

of age in the Netherlands study (Huisman et al.,

1995b). In Michigan, prenatal PCB exposure was

associated with poorer visual recognition memory in

infancy (Jacobson et al., 1985, 1990a, 1992), an effect

that was recently confirmed in the Oswego study

(Darvill et al., 2000). In North Carolina, deficits in

psychomotor development up to 24 months wer e seen

in the most highly exposed children (Gladen et al.,

1988; Rogan and Gladen, 1991). In Michigan, the

Netherlands, Oswego and Germany, prenatal PCB

exposure was linked to poorer intellectual function

during childhood (Jacobson et al., 1990a; Jacobson

and Jacobson, 1996; Patandin et al., 1999; Stewart et

al., 2003; Vreugdenbil et al., 2001; Walkowiak et al.,

2001). Virtually all the adverse neurobe havioural

effects reported to date were linked to prenatal expo-

sure, indicating that the embryo and fetus are particu-

larly vulnerable to these substances. Moreo ver, recent

analyses from the Michigan and the Netherlands

cohorts underline the greater vulnerability to prenatal

PCB exposure in non breast fed infants (Jacobson and

Jacobson, 2004; Vreudenhil et al, in press, cited in

Jacobson and Jacobson, 2004).

One prospective, longitudinal study, performed in

the Faroe Islands, examined the effects of prenatal

exposure to low doses of methylmercury resulting

from fish and pilot whale consumption (Grandjean

et al., 1992, 1997). Because pilot whale tissues

contain other neurotoxicants, this cohort was also

exposed to PCBs and, as such, is the only cohort

studied to date where the main source of PCB

exposure was consumption of marine mammals, as

is the case for the Inuit. Further analysis provides

some evidence that the effects of methylmercury

identified in this study are not the consequence of

exposure to PCBs. Again, PCB concentrations in

cord serum in these different studies are presented

in Fig. 4.1.1.

4.2.1.2. Methylmercury. Three well-designed, pro-

spective, longitudinal studies that examined the neu-

robehavioural effects of prenatal exposure to low

doses of methylmercury were performed in New Zeal-

and, the Faroe Islands, and the Seychelles Islands.

High methylmercury exposure in the Faroese popula-

tion results from fish and pilot whale consumption

(Grandjean et al., 1992); deep-sea and reef fish con-

sumption is the source of exposure for the Seychelles

population (Myers et al., 1995); and fish consumption

is the source of exposure for the New Zealand popu-

lation (Kjellstro¨m et al., 1986).

Comparisons of mercury concentrations in Inuit

populations with those cohorts are shown in Table

4.2.1. Greenlandic Inuits have the highest levels of

maternal and cord blood mercury. High and quite

similar levels of mercury are also seen in the Faroe

Islands, Seychelles and New Zealand and Nunavik

studies and lower levels are found in the southern

Que´bec and US populations cited (Muckle et al.,

2001b).

The Faroe Islands study reported associations

between maternal hair mercury concentrations cor-

responding to the pregnancy period and children’s

performance on neurobehavioural tests, particularly

in the domains of fine motor funct ion, attention,

language, visual–spatial abilities, and verbal mem-

ory (Grandjean et al., 1997). Those effects were

also found to be associated with cord blood mer-

cury concentration (Grandjean et al., 1999). In the

New Zealand study, higher hair mercury levels

were associated with poorer neurodevelopmental

test scores in similar domains to those observed

in the Faroe study (Crump et al., 1998; Kjellstro¨m

et al., 1986). In contrast, prenatal methylmercury

exposure was not related to neurobehavioural

effects in the Seychelles Islands study (Davidson

et al., 1995, 1998; Myers et al., 1995). Recently,

the authors of the Seychelles Islands study reported

that two of the 21 neurobehavioural end points

assessed at 9 years of age were associated with

prenatal methylmercury exposure, but their conclu-

sion was that these associations were probably due

to chance as a result of multiple analyses (Myers et

al., 2003).

Differences in the marker of mercury exposure,

the end points assessed, age at testing, source and

pattern (stable vs. episodic) of exposure between the

Faroe and the Seychelles Islands studies have been

suggested to account for the differences in the find-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 203

ings (NRC, 2000). The New Zealand study, in which

the exposure and research design were similar to the

Seychelles Islands study, also found neurobeha-

vioural effects, as did the pilot study conducted in

the Seychelles Islands population.

Since 1997, Nunavik mothers and infants have

been observed in a prospective longitudinal study.

Exposure data were published (Muckle et al.,

2001a,b). Statistical analysis of the data is ongoing

and results on growth and neurobehavioural effects of

perinatal exposur e to PCBs, other OC compounds,

and methylmercury will be published soon.

In addition, a sub-sample of the newborns who

participated in the Nunavik cord blood monitoring

program in 1993–1996 were evaluated at preschool

age in 2000 and 2001 using a battery of sensitive tests

designed to assess fine motor, gross motor, neurolo-

gical, and neurophysiological functions. Results will

be available in the near future.

4.2.2. Biomarkers of developmental effects

4.2.2.1. Cytochrome P4501A1 induction and DNA

adduct formation. Overexpression and increased

activity of cytochrome P4501A1 (CYP1A1) in pla-

centa have been linked to reduced birth weight (Pelk-

onen et al., 1979; Lucier et al., 1987). This enzyme

can bioactivate several xenobiotics to reactive inter-

mediates that can bind DNA.

CYP1A1 induction and DNA adducts were tested

as possible markers of early develo pmental effects

related to OC exposure in Inuit women from Nunavik

(Lagueux et al., 1999). CYP1A1-dependent ethoxyr-

esorufin-O-diethylase activity (EROD) and DNA

adducts were measured in placenta samp les obtained

from 22 Inuit women from Nunavik. These biomar-

kers were also assessed in 30 women from a Que´bec

urban centre (Sept- I

les) who served as a reference

group. Prenatal OC exposure was determined by mea-

Table 4.2.1

Comparison of mercury (total) concentrations in Nunavik with those observed in other cohorts

Cohort (reference) Medium Years N Geometric mean Range Interquartile range

Canada

Nunavik Inuit Cord blood (Ag/L) 1996–2000 95 18.5 2.8–97.0 12.0–27.2

Maternal blood (Ag/L) 1993–1995 130 10.4 2.6–44.2 6.6–17.0

Maternal hair (Ag/g) 1992 123 3.7 0.3–14.0 2.5–6.2

Southern Quebec Cord blood (Ag/L)

1977–1978 1108 1.0 0.9–1.0

James Bay Cree Women hair, not pregnant (Ag/g)

1981 70 2.5 Max=19.0

Northern Quebec Cree Maternal hair (Ag/g) 215 6.0

5.2

USA Women hair, not pregnant (Ag/g) 1981 1274 0.36

0.14–0.90

1546 0.24

0.09–0.62

Faroe Islands

First cohort Cord blood (Ag/L) 1986–1987 894 22.9 13.4–41.3

Maternal hair (Ag/g) 1994–1995 914 4.3 2.6–7.7

Second cohort Cord blood (Ag/L) 163 20.4 1.9–102.0 11.8–40.0

Maternal hair (Ag/g) 144 4.1 0.4–16.3 2.5–7.4

Seychelles Island

Main study Maternal hair (Ag/g) 1989–1990 740 5.9 0–25 6.0

Pilot study Maternal hair (Ag/g) 789 6.6 0.6–36.4 6.1

New Zealand Maternal hair (Ag/g) 1978–1984 935 8.3

6.0–86.0

Greenland, Disko Bay Cord blood (Ag/L) 1994–1996 178 25.3 2.4–181.0

Maternal blood (Ag/L) 1994–1996 180 12.8 1.9–75.6

Source: Muckle et al. (2001b).

The average Hg concentration was reported in nmol/L, this concentrations was divided by 5 to transform to Ag/L.

95% confidence interval.

Women aged between 15 and 39 years old.

Arithmetic mean.

Standard deviation.

Among seafood consumers.

Among non-seafood consumers.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246204

suring these compounds in umbilical cord plasma.

Placental EROD activity and the amount of DNA

adducts thought to be induced by OC exposur e were

significantly higher in the Nunavik group than in the

reference group. For both biomarkers, smoking was

found to be an important confounding factor, but OC

exposure was significantly associated with EROD

activity and DNA-adduct levels when controlling for

self-declared smoking status. It was then concluded

that CYP1A1 induction and DNA adducts in placental

tissue could const itute useful biomarkers of early

effects induced by environmental exposure to OCs

(Lagueux et al., 1999). In the latter study, however,

there were very few Inuit women who did not smoke

during pregnancy, the smoker and non-smoker groups

were not balanced, and the smoking status was not

ascertained with a biomarker. Therefore, a second

study was conducted to determine if environmental

exposure to PCBs induces placental CYP1A1 in Inuit

women. This more recent study was designed to con-

trol the confounding effects of smoking better. The

use of cotinine concentration in meconium and of

cadmium concentrations in placenta as markers of

prenatal exposure to tobacco smoke was previously

validated (Pereg et al., 2001). Placenta, cord blood

and meconium samples were obtained from 35 Inuit

women in Nunavik and 30 women in Sept-I

les (refer-

ence population). Efforts were made to sample more

smokers in the Sept-I

les population and more non-

smokers in the Nunavik population in order to balance

the smokers and non-smokers groups. Smoking status

was ascertained with the use of cotinine concentration

in the meconium and, when necessary, individuals

were reassigned to the proper smoking category

based on this marker.

PCB concent rations were measured in cord plasma

and CYP1A1 activity (EROD) was assessed in placen-

tal tissue. Despite the higher PCB exposure of the Inuit

population, both groups showed similar EROD activ-

ities when data were stratified according to the smok-

ing status ascertained by the cotinine concentration. In

the Nunavik popula tion, EROD activity was correlated

with 2,2V,4,4V,5,5V-hexachlorobiphenyl (PCB 153)

plasma concentration (a marker of exposure to the

environmental PCB mixture). However, cotinine con-

centrations in meconium were also significantly corre-

lated to PCB 153 plasma concentrations, and

multivariate analyses failed to demonstrate a signifi-

cant contribution of PCB exposure to placental

CYP1A1 activity when tobacco smoking (as estimated

with cotinine concentration in the meconium) was

included in the analysis. Results from this study, there-

fore, showed that low-level environmental PCB expo-

sure does not induce any increase in CYP1A1 activity

in the placenta, leaving tobacco smoking as the major

modulating factor (Pereg et al., 2002). Exposure of

Inuit from Nunavik to PCBs was much lower than the

PCBs/PCDFs exposure in Taiwanese children. This

could explain these contradictory results.

4.2.2.2. Thyroid hormones. Although many theories

exist on how PCBs affect neurodevelopment, the main

hypothesis involves the effect of PCBs on thyroid

hormone homeostasis (Porterfield and Hendry,

1998). Thyroid hormones regulate neuronal prolifera-

tion, cell migration and differentiation, including con-

trol of when differentiation begins and when cell

proliferation ends (Hamburgh, 1969).

One hundred and eighty-two singleton term births

were evaluated in the Faroe Islands, where marine

food includes pilot whale (Steuerwald et al., 2000).

Maternal serum, hair, milk and umbilical cord blood

were analyzed for contaminants. Levels of essent ial

fatty acids, selenium, and thyroid hormones were

determined in cord blood. Each infant’s neurologic

optimality score was determined at two weeks of age,

adjusted for gestational age, and predictors were

assessed by regression analysis. Thyroid function

was normal and not associated with PCB exposure.

In the Netherlands study, 418 mother–infant pairs

were enrolled. Thyroid hormone levels were in the

normal range, but higher dioxin and PCB–TEQ levels

in human milk were significantly correlated with

lower thyroid T3 (triiodothyronine) and T4 (thyrox-

ine) levels and with higher levels of thyroid-stimulat-

ing hormone (TSH) in the infants’ plasma at the age of

2 weeks and 3 months. Thyroid hormone level altera-

tions detected in this study, however, were not directly

associated with neurological dysfunction (Koopman-

Esseboom et al., 1994b).

In Nunavik, 466 measurements were taken of thyr-

oid hormones in Inuit newborns’ umbilical cord blood

samples in the cord blood monitoring program that

took place between 1993 and 1996. Free T4 (thyrox-

ine), total T3 (triiodothyronine), TBG (thyroxin-bind-

ing globulin) and TSH were measured. Hydroxylated

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 205

metabolites of PCBs (OH-PCBs) and other phenolic

compounds were also measured in a subsample

(n = 10). As expected, birth weight was positively

associated with thyroid hormones (thyroxine, TBG).

For this reason, further analyses were adjusted for

birth weight. After adjustment, TBG and TSH levels

were significantly and negatively associated with PCB

congener levels (Dewailly et al., 1998)

The main transport mechanism of thyroid hor-

mones to the brain requires passing through the

blood/brain barrier via a thyroid hormone transport

protein called transthyretin (TTR) (Chanoine and

Braverman, 1992). Although PCBs show some bind-

ing affinity for TTR (Chauhan et al., 2000), OH-PCBs

have much higher in vitro binding affinities—as high

as 12 times the binding affinity of the natural ligand,

thyroxine (T4) (Lans et al., 1994; Cheek et al., 1999;

Brouwer, 1991). Binding to TTR is not limited to OH-

PCBs—other chlorinated phenolic compounds such

as pentachlorophenol (PCP), halogenated phenols,

and brominated flame retardants also have strong

affinities for TTR (Van den Berg et al., 1991; Van

den Berg, 1990; Meerts et al., 2001). Recently, PCP

was found to be the dominant phenolic compound

determined in Inuit whole blood (Sandau et al.,

2000). Other halogenated phenolic compounds may

also be important contaminants in plasma, as they

have been found to exhibit toxicological properties

similar to those of OH-PCBs (Van den Berg et al.,

1991; Schuur et al., 1998).

PCBs have been previously measured in umbilical

cord plasma; however, few studies have examined

levels of hydroxylated metabolites in blood, especially

in humans. OH-PCBs have been quantified recently in

Canadian Inuit whole blood (Sa ndau et al., 2000) and

in Swedish and Latvian fish-eaters (Sjo¨din et al.,

2000). Sandau et al. (2000) examined chlorinated

phenolic compounds in umbilical cord plasma of new-

borns in three populations with different PCB expo-

sures, including the Inuit population. Retinol and

thyroid hormone status [triiodothyronine (T3), free

thyroxine (T4), thyroid-stimulating hormone (TSH),

and thyroxin-binding globulin (TBG)] were deter-

mined in most samples. The authors found an inverse

association (r =  0.47; p = 0.01) between log-normal-

ized free thyroxine and log-normalized total phenolic

compounds (sum PCP and OH-PCBs). Total chlori-

nated phenolic compounds were also negatively asso-

ciated with T3 (r =  0.48, p = 0.03) (Sandau et al.,

2002).

4.3. Sex hormone disruption

The development and maintenance of reproductive

tissues are to a large extent controlled by steroid

hormones. Some environmental chemicals mimic,

while others antagonize natural hormone activity

when tested with in vitro assays or in whole animal

models. Studies dating back to the late sixties identi-

fied o,pV-DDT, a minor constituent of technical DDT,

as a weak estrogenic compound capable of causing an

increase in rat uterine weight in the classic immature

female rat model (Bitman and Cecil, 1970). This com-

pound and a few others sharing estr ogenic properties

have been implicated in abnorm al sexual development

in birds (Fry and Toone, 1981), and in feminized

responses in male fish (Jobling et al., 1995).

Certain male reproductive tract disorders (cryp-

torchidism, hypospadiasis, and testicular cancer)

have been reported to be increasing in parallel with

the introduction of xenoestrogens such as DDT into

the environment. Reduced semen quality was also

reported in certain regions of the world during the

last half of the 20th century (Carlsen et al., 1992;

Auger et al., 1995). Although these alterations are

thought to be mediated by the estrogen receptor,

they are also consistent with inhibition of androgen

receptor-mediated events. Kelce et al. (1995) identi-

fied the major and persi stent DDT metabolite, p,pV-

DDE, as a potent anti-androgenic agent in male rats.

In addition to inhibiting androgen binding to the

androgen receptor, this compound, when administered

to pregnant dams, also induced characteristic anti-

androgenic effects in male pups (reduced ano-genital

distance; presence of thoracic nipples). Treatment

with p,pV-DDE at wean ing delayed the onset of pub-

erty, while treatment of adult rats resulted in reduced

seminal vesicle and ventral prostate weights.

2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is

yet another OC which has been shown to alter sexual

development in male rats (Mably et al., 1992).

Decreases in epididymis and caudal epididymis

weights, and in daily sperm production and caudal

epididymal sperm number were observed at day 120

and at most earlier times, when a dose as little as 64

ng/kg was administered to dams on day 15 of gesta-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246206

tion. A number of compounds structurally related to

TCDD, including other 2,3,7,8-chloro-substituted

dibenzo-p-dioxins, dibenzofurans, as well as non-

ortho and mono-ortho substituted PCB congeners,

bind to the Ah receptor and displ ay similar toxicolo-

gical properties.

Typical OC mixtures found in highly exposed

human populations contain a large variety of OC

compounds, including substances with estrogenic,

anti-estrogenic, or anti-androgenic capacities. It may

therefore be anticipated that complex real life mix-

tures, composed of numerous compounds that can

interact with different receptors involved in cell dif-

ferentiation and growth, could affect reproduction and

development, and be involved in the pathogenesis of

hormonally responsive cancers.

4.3.1. Clinical outcomes

4.3.1.1. Sexual maturation of newborn males. DDE

was recently found to inhibit binding of androgen to

its receptor and to block androgen action in rodents.

Normal development of male genitalia in mammals

depends on androgen action. Recently, Longnecker et

al. (2002) used stored serum samples to examine the

relationship between maternal DDE levels during

pregnancy and adjusted odds of cryptorchidism

(n = 219), hypospadias (n = 199), and polythelia

(extra nipples) (n = 167) among male offspring,

using a nested case-control design with one control

group (n = 552). Subjects were selected from a United

States birth cohort study begun in 1959–1966, when

DDE levels were much higher than they are at present.

Compared with boys whose mother’s recovery-

adjusted serum DDE level was less than 21.4 lg/L,

boys with maternal levels greater than or equal to 85.6

lg/L had adjusted odds ratios of 1.3 (95% confidence

interval (CI): 0.7, 2.4) for crypt-orchidism, 1.2 (95%

CI: 0.6, 2.4) for hypospadias, and 1.9 (95% CI: 0.9,

4.0) for polythelia. For cryptorchidism and polythelia,

the results were consistent with a modest-to-moderate

association, but in no instance was the estimate very

precise. The results were inconclusive. In this cohort,

DDE concentrations in umbilical cord serum were

much higher than in Nunavik (geometric mean

 2.2, range of 0.14 to 18 Ag/L, Table 2.2.1).

Sexual maturation of newborn males is examined

within the on-going cohort study conducted in

Nunavik, and ano-genital distance and penis length

are recorded. In adults, no study on hormone-as so-

ciated diseases (breast cancer, endometriosis, and

male fertility) has been conducted in the Canadian

Arctic.

4.3.1.2. Environmental risk factors for osteoporo-

sis. POPs have recently been associated with an

increased risk of osteoporosis in humans. The rela-

tionship between DDE and bone mineral density was

recently examined in 68 sedentary Australian women

who reported adequate dietary intake of calcium

(Beard et al., 2000). Reduced bone mineral density

was correlated significantly with age (r =  0.36,

p = 0.004), as well as with increases in the log of

DDE levels in serum (r =  0.27, p = 0.03). The

authors also used multiple-regression analysis to

examine the influence of other predictor variables

on the relationship between log DDE and bone

mineral density. The strongest model ( p = 0.002)

included log DDE ( p = 0.018), age ( p = 0.002), and

years on hormone replacement therapy ( p = 0.10) as

predictor variables, and this model afforded a pre-

diction of 21% of bone mineral density variation.

These results sugges t that past community exposures

to DDT may be associated with reduced bone

mineral density in women. As a potent androgen

receptor antagonist, DDE may reduce the inhibitory

effect on cytokines and result in the inappropriate

turnover of osteoclasts or inadequate production of

osteoblasts within bone marrow, thus leading to

reduced bone density (Beard et al., 2000). These

are interesting results, but they must not be over-

interpreted as this was a small study and the meno-

pausal status and time since menopause, two

important determinants of osteoporosis, were not

included in the analyses.

A study was conducted in Nuuk, Greenland,

in September 2000 to evaluate the prevalence

of risk factors for osteoporosis fracture and more

particularly environmental factors and their asso-

ciation with bone mass in menopausal women

(Coˆte´ et al., 2005). The risk of osteoporosis frac-

ture was assessed using an ultrasound bone den-

sitometer. All three ultrasound parameters adjusted

for age were lower in this Inuit population com-

pared with the women of Que´bec (Coˆte´ et al.,

2005).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 207

The study found that 19% of the Inuit women had a

high risk of osteoporosis fracture compared to 7.2%

for the women of southern of Que´bec. To identify

which risk factors were associated with ultrasound

parameters, a multiple linear regression model invol-

ving the stepwise removal of non-significant indepen-

dent variables was used. The independent predictors

of the bone density were age ( p = 0.018), BMI

( p = 0.018), former users of oral contraceptives

( p = 0.003), current hormone replacement therapy

users ( p = 0.005), and the sum of log of mono-ortho

PCB TEQs ( p = 0.004). These variables accounted for

36% of the variance of the bone density (r

= 0.359)

(Coˆte´ et al., 2005). Mono-ortho substit uted congeners

(IUPAC numbers 105, 118 and 156) share some

structural similarities with TCDD and can bind to

the aryl hydrocarbon receptor (AhR). The conse-

quences of activation of the AhR pathway on osteo-

porosis, however, are not clear. One possible

explanation is that dioxin -like compounds elicit a

broad spectrum of anti-estrogenic activities and may

reduce bone density through this mechanism. In sum-

mary, mono-ortho PCB concentrations were asso-

ciated with a low bone density, and this association

remained significant after controlling for potential

confounding factors (Coˆte´ et al., 2005).

4.3.2. Hormonal biomarkers

4.3.2.1. Hormone profiles. To obtain a steroid pro-

file of androgens for both precursors and metabolites

of dihydrotestosterone (DHT), a series of steroids,

including DHT, could be measured: dehydroepian-

drosterone (DHEA), androst-5-ene-3 eta,17-diol,

androstenedione, testosterone, DHT, estrone, estradiol,

DHEA sulfate, androstane-3,17-diol glucuronide, and

androsterone glucuronide. These steroid levels help

explain alteration in steroidogenic enzymes in classical

steroidogenic tissues such as adrenals and the testis,

and for steroidogenic transforming enzymes localized

in peripheral tissues (prostate and skin).

In the Arctic, no study has been conducted to

determine hormone profiles. A pilot study was

recently performed in Greenland (n = 48 males) and

the following male hormones were measured: DHEA,

delta5-diol, delta4, testosterone, DHT, E1 and E2. A

study on male fertility is also on going in Greenland

(Toft et al., 2003).

4.4. Oxidative stress

Although oxidative stress has been associated

with many chronic diseases (e.g., cancer, cardiovas-

cular disease (CVD), neuro-degenerative disorders,

etc.), no specific epidemiological study of these

outcomes has been conducted in the Arctic to assess

the role of oxidants and antioxidants. Results

reported by Salonen et al. (1995) suggest that the

high CVD mortality observed among fish-eaters

from Finland could be explained by the high mer-

cury c ontent of the fish (mainly non-fatty freshwater

species), which could counteract the beneficial

effect of fish consumption. This group noted a

significant association between mercury concentra-

tion in the hair of eastern Finnish men and the risk

of coronary heart disease (CHD). Mercury can pro-

mote the peroxidation of lipids, resulting in more

oxidized low-density lipoprotein (LDL), which has

been implicated as an initiator of atherosclerosis. In

the same population, Salonen et al. (1982) pre-

viously observed an enhanced risk of CHD death

in subjects with low serum selenium concentrations,

an antioxidant which may possibly block the mer-

cury-induced lipid peroxidation.

That both mercury and selenium can modulate

CHD risk is also suggested by observations in fish-

eating coastal populations such as Inuit living in

Arctic regions. Inuit consume large amounts of fish

and marine mammals, and consequently receive large

doses of mercury. Contrary to the situation in eastern

Finland, however, the mortality rate from CHD in

Inuit is low. Although, it was reported that omega-3

fatty acids are strong protective factors for cardiovas-

cular diseases among Inuit (Dewailly et al., 2001a),

the protection could also result from a high intake of

selenium, (Be´langer et al., 2003), through the con-

sumption of traditional/country foods such as muktuk

(beluga and narwhal skin) and sea mammal liver

which are rich in selenium.

Methylmercury is a highly toxic environmental

neurotoxin that can cause irreparable damage to the

central nervous system (Choi, 1989; Clarkson, 1993,

1997). Althoug h the underlying biochemical and

molecular mechanisms that lead to impaired cell

function and nerve cell degeneration are not well

understood, there is abundant evidence supporting

the hypothesis that a major mechanism of methyl-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246208

mercury neurotoxicity invol ves oxidative stress (Sar-

afian and Verity, 1991; Yee and Choi, 1996). Mer-

cury increases production of reactive oxygen species

(ROS) via deregulation of mitochondrial electron

transport, as wel l as through glutathione (GSH) deple-

tion (Lund et al., 1993). The oxidative stress hypoth-

esis is clearly supported by the finding that

methylmercury neurotoxicity can be inhibited by var-

ious antioxidants, including selenium (Park et al.,

1996) and N-acetyl-l-cysteine, a precursor of GSH

(Ornaghi et al., 1993).

Glutathione peroxidase (GSHPx) and glutathione

reductase (GSHRd) activities were measured in blood

samples from 142 residents of Salluit, Nunavik

(Dewailly et al., 2001b). Activities of enzymes

involved in detoxication of free radic als were mea-

sured to investigate the relationships between mer-

cury, selenium and oxidative stress. Mercury was

found to be negatively correlated with GSHRd activ-

ity, an NADPH-dependent enzyme that regenerates

glutathione from glutathione disulfi de. In contrast,

plasma selenium concentration was positive ly corre-

lated to GSHPx activity, a selenoenzyme that cata-

lyzes the conversion of hydrogen peroxide to water.

Mercury exposure may, therefore, diminish defense

mechanisms against oxidative stress by limiting the

availability of glutathione, while selenium may afford

protection by favouring the destruction of hydrogen

peroxide.

A biochemical assessment of the oxidative stress in

adult residents of Nunavik is underway using the

following three other indices:

1) The first is the ratio of coenzyme Q10-reduced

form (ubiquinol-10) to oxidized coenzyme Q10

(ubiquinone-10) in plasma, which is now consid-

ered on e of the most reliable and sensitive indices

of an oxidative stress in vivo (Yamashita, 1977). In

contrast to the total level of coenzyme Q10, which

is reported to be associated with multiple factors

including gender, age, cholesterol and triglycerides

levels (Kaikkonen et al., 1999), the ubiquinol-10/

ubiquinone-10 ratio index is apparently indepen-

dent of these variables and thus represents an

oxidative stress index of choice.

2) Increased levels of specific F2-isoprostanes (direct

oxidation metabolites of arachidonic acid) in plasma

and/or urine are another index recently used to

demonstrate oxidative stress in several pathological

conditions involving oxygen free radical formation

(Pratico, 1999; Patrono and FitzGerald, 1997). The

most easily measurable and frequently used F2-iso-

prostane species as a marker of oxidative stress in

vivo is 8-isoprostaglandin F2-alpha (Pratico, 1999;

Patrono and FitzGerald, 1997).

3) The level of plasm atic LDL oxidation could also be

assessed as a potential marker of oxidative stress.

Preliminary results indicate that oxidized LDL was

significantly lower in Inuit subjects than normal

Caucasian population (1.6



, p b 0.0001), support-

ing the previous observation that omega-3 fatty

acids and selenium could be strong protective fac-

tors for cardiovascular diseases among Inuit

(Be´langer et al., 2003).

These results will become available in 2006. This

study also includes a neurological assessment and

should produce some interesting results on another

measure of oxidative stress and its possible relation-

ship to neurological outcomes.

5. Risk-benefit character ization, assessment and

advice

The contamination of wildlife in the Arctic ecosys-

tem poses a complex problem. Weighing the benefits

of country food consumption against the risk of con-

taminant exposure is very difficult, and cuts across the

disciplines of nutrition, toxicology, envir onmental pol-

icy, sociology, and public health practice; and each

discipline has its unique perspectiv e. A comprehensive

risk-benefit assessment/management framework is

needed to coordinate the various perspectives and to

respond to the challenging task of weighin g the risks

and benefits of country food consumption in northern

Canada.

The NCP has adopted such a framework for risk

assessment and management, one whi ch involves a

cooperative multi-agency approach, where the pro-

blem is considered in its ecological and public health

context, and those who are affected by the risk

management decisions are involved in the decision-

making process. This collaborative approac h is

required under the NCP as territories and provinces

are responsible for delivering health advice to their

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 209

residents but the broad range of issues requires input

from all NCP partners (federal and territorial depart-

ments, Aboriginal groups, local communities and

other interested parties).

This section explains the process of risk-benefit

characterization. It focuses on risk-benefit character-

ization in Arctic communities and the special consid-

erations that are required. An important consideration

in Arctic communities is the role that risk perception

plays in communication. Final ly, an evaluation of

current risks is provided.

5.1. Contaminant exposure risks

5.1.1. Contaminant intakes

5.1.1.1. Persistent organic pollutants. The contami-

nants having the greatest exceedances of dietary

guidelines were chlordane and toxaphene. Fig. 5.1.1

shows the overall mean inta kes of chlordane, toxa-

phene and mercury for various Aboriginal groups,

based on data collected in the mid to the late 1990s.

Mean intakes of Dene and Me´tis of the NWT and

Yukon First Nations were below the provisional Tol-

erable Daily Intakes (pTDIs) for these contaminants,

whereas the mean intakes in Inuit communities

exceeded the pTDIs for chlordane and toxaphene

(Chan, 2002).

Among the five major Inuit regions, Baffin, Kival-

liq and Inuvialuit communities mean inta kes of 20- to

40-year-old adults exceeded the pTDIs for chlordane

and toxaphene while the intakes for Kitikmeot and

Labrador did not exceed the pTDIs (Fig. 5.1.2). When

intake data in the Baffin region were separated by age

group, it is seen in Fig. 5.1.3 that mean intakes of all

age groups exceeded the pTDI for toxaphene, and that

the three adult age-groups exceeded the pTDI for

chlordane. The intake of both contaminants increases

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

PTDIYukon

n = 802

Dene

n = 1012

Inuit

n = 1875

Mean intake (µg/kg/day)

Toxaphene

Mercury

Chlordane

Fig. 5.1.1. Mean intakes of chlordane, toxaphene, and mercury in

northern Canada (Ag/kg/day) (Kuhnlein et al., 2001b).

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.2

0.4

LabradorKivalliqKitikmeotInuvialuitBaffin

Mean intake (µg/kg/day)

Toxaphene

Chlordane

PTDI CHL

PTDI TOX

Fig. 5.1.2. Mean intakes of toxaphene and chlordane in different

regions (ages 20–40 years) (Chan et al., in preparation(a)).

61+41– 6020 – 4015–19

Mean intake (µg/kg/day)

ToxapheneChlordane

PTDI CHL

PTDI TOX

Age group (years)

Fig. 5.1.3. Mean intakes of toxaphene and chlordane among differ-

ent age groups in Baffin (Chan et al., in preparation(a)).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246210

with each age group. This increasing level of exposure

with age is associated with the corresponding age-

related increased intake of country food as was docu-

mented in Van Oostdam et al. (1999).

Individual intake of contaminants depends primar-

ily on the amount of each country food consumed by

the individual and the concentration of contaminants

in that food. In this research contaminants contained

in purchased market foods wer e not considered. Table

5.1.1 shows the sources of chlordane, toxaphene, and

PCBs in the dietary intake profiles of the Baffin

region (n = 522 interviews) . It can be seen that the

three most important food species, by their propor-

tional contribution to total weight of food consumed,

are caribou, ringed seal and Arctic char. However, the

three species representing the largest proportions of

chlordane and PCB exposure were the blubbers of

narwhal, walrus, and beluga. For toxaphene, the

greatest proportional contributions wer e made by wal-

rus blubber, ringed seal flesh, and beluga blubber.

Species which only make up a small proportion of

the diet but high proportions of the contaminant

exposure have markedly higher concentrations of

these contaminants (Kuhnlein et al., 2000). Propor-

tionate contributions to chlordane and toxaphene

exposures are also shown in Table 5.1.2 for all five

Inuit regions. In regions such as Baffin and Kivalliq

which had intakes higher than the pTDI for chlordane

and toxaphene (Fig. 5.1.2) it can be seen that the

major sources of exposure are marine mammal blub-

ber (beluga, walrus or narwhal). In regions such as

Labrador where the mean intakes did not exceed the

pTDI, caribou meat and various fish species were the

Table 5.1.1

Sources of organochlorines in the Baffin Region (percent contribution)

Species Part Weight percent

contribution

(wt.%)

Chlordane percent

contribution

(%)

PCBs percent

contribution

(%)

Toxaphene percent

contribution

(%)

Caribou Flesh 38.2 0.9 1.3 0.1

Ringed seal Flesh 18.7 0.8 2.4 8.9

Arctic char Flesh 15.6 2.2 1.5 3

Narwhal Muktuk 5 1.8 7 0.1

Walrus Flesh 3.2 1.7 0.6 0.1

Ringed seal Broth 2.9 0.2 1.1 0.1

Polar bear Flesh 2.8 1.5 3.1 0.1

Narwhal Blubber 1.9 37.9 44.5 35.6

Ptarmigan Flesh 1.3 0 0 0

Beluga Muktuk 1.2 1.7 1.7 0.9

Walrus Blubber 1.2 34.9 22.2 43.1

Beluga Blubber 0.4 11.1 8.5 6.3

Ringed seal Blubber 0.3 1.9 1.3 0.3

Polar bear Fat 0.1 2.3 1.6 0.5

Total % 92.8 98.9 96.8 99.1

Source: Kuhnlein and Receveur (2001).

Percent by weight of each species contributing to the traditional diet.

Percent of each contaminant contributed by each food.

Table 5.1.2

Proportionate contributions of three main food sources of chlordane

and toxaphene, in five Inuit regions, by food item

Inuit Region Food item and

proportionate

contribution

of chlordane (%)

Food item and

proportionate

contribution

of toxaphene (%)

Inuvialuit Beluga blubber (84%) Beluga blubber (81%)

Caribou meat (5%) Beluga muktuk (4%)

Beluga muktuk (5%) Arctic char flesh (3%)

Kitikmeot Beluga blubber (59%) Beluga blubber (52%)

Arctic char flesh (13%) Arctic char flesh (26%)

Caribou meat (7%) Ringed seal flesh (9%)

Kivalliq Beluga blubber (62%) Beluga blubber (53%)

Beluga muktuk (15%) Walrus blubber (15%)

Walrus blubber ( 8%) Arctic char flesh (10%)

Baffin Narwhal blubber (38%) Walrus blubber (43%)

Walrus blubber (35%) Narwhal blubber (36%)

Beluga blubber (11%) Ringed seal flesh (9%)

Labrador Caribou meat (30%) Salmon flesh (41%)

Lake trout flesh (20%) Lake trout flesh (29%)

Salmon flesh (12%) Arctic char flesh (9%)

Source: Kuhnlein and Receveur (2001).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 211

major source of exposure, indicating lower concentra-

tion in these species.

The population distributions of seven organochlor-

ines are shown in Table 5.1.3 for Qikiqtarjuaq—a

community selected because of high country food

use. Mean intakes exceeded the organochlorine

pTDIs for chlordane, PCBs and toxaphene, while

median intakes exceeded the pTDI only for toxa-

phene. The 95th percentile indicates the level of

intake by the highest 5% of consumers. When the

95th percentile/pTDI was compu ted it was noted that

high consumers are exceeding the pTDIs by more

than 100-fold for toxaphene and chlordane, and 15-

fold for PCBs (Kuhnlein et al., 2000, 2001a,b; Batal,

2001). It is understood from dietary survey results

that any one individual will not likely eat the same

amount of high contaminant food items on several

consecutive days. If more days of intake data from

individuals were available, a tighter distribution and

therefore a lower prevalence of high levels of expo-

sure would result. The risk associated with usual

levels of exposure (that is, levels, lower than the

ones characterized by the 95th percentile/pTDI but

higher than the pTDI) need to be characterized, since

they appear prevalent in the Baffin region. These

results (Kuhnlein et al., 2001a,b, 2003; Batal, 2001)

confirm the earlier studies indicating higher expo-

sures to various POPs in Inuit communities and

those that were reviewed in Van Oostdam et al.

(1999).

Possible changes in dietary intake of contaminants

are a significant concern in the Canadian Arctic.

Insight into the change in contaminant intake over

the past 12 years in Qikiqtarjuaq (Broughton Island),

Nunavut, a community known to have high country

food consumption levels, can be achieved by compar-

ing data collected in 1987–1988 (Kuhnlein et al.,

1995a; Chan et al., 1997) to data collected in 1998–

1999 (Kuhnlein et al., 2000). Intakes of organochlor-

ines, including PCBs, chlordane, and toxaphene

(Table 5.1.4), were higher in the more recent study,

particularly among the high-end consumers (95th per-

centile). While the organochlorine concentrations

used for the estimation of the two surveys were simi-

lar (data not shown), the amount of narwhal muktuk

and blubber consumed was significantly higher in

1998–1999. The major sources of organochlorines

were narwhal blubber and muktuk, walrus blubber,

and ringed seal blubber reported in 1987–1988 com-

pared to narwhal blubber and muktuk in 1998–1999

(Kuhnlein et al., 1995c, 2000).

PCB tissue levels and guidelines. While chlor-

dane and toxaphene are of significant concern due to

exceedances of dietary guidelines, there are only lim-

ited human or animal toxicology data available for

these contaminants and no human tissue guidelines

have been developed. More information is available

on human tissue levels and on the animal and human

Table 5.1.3

Population distribution of organochlorine intake in Qikiqtarjuaq (Ag/kg bw/day)

Organochlorine N PTDI (Ag/kg/day) N N PTDI Mean Median 95th percentile 95th/PTDI

Chlordane 110 0.05 34 0.62 0.02 5.18 104

HCB 110 0.27 14 0.23 0.02 1.7 6

DDT 110 20 9 1.58 0.04 13.1 0.7

HCH 110 0.3 11 0.06 0 0.36 1

Mirex 110 0.07 5 0.01 0 0.04 0.6

PCBs 110 1 16 1.9 0.05 15.3 15

Toxaphene 110 0.2 52 3.34 0.26 26.2 131

Source: Kuhnlein and Receveur (2001).

Table 5.1.4

Comparison of daily intake of selected contaminants in Qikiqtarjuaq

in 1987–1988 and 1998–1999

Contaminant Daily intake

(Ag/kg bw/day)

in 1987–1988

Daily intake

(Ag/kg bw/day)

in 1998–1999

Median 95th

percentile

Median 95th

percentile

Mercury (total) 1 7 0.6 6.4

PCBs 0.3 3.6 0.05 15

Chlordane 0.06 2.1 0.02 5.2

Toxaphene 0.05 8.1 0.26 26

Source: Kuhnlein et al. (1995a) and Chan et al. (1997).

Source: Kuhnlein et al. (2000).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246212

toxicity of PCBs, which has allowed the development

of human tissue PCB guidelines. This section compares

PCB tissue levels in various ethnic groups to the PCB

guidelines developed by Health Canada.

Blood contaminant levels in the NWT, Nunavut

and Nunavik, are presented and compared in Sec-

tion 2.2.1. For PCBs (as Aroclor 1260), Health Cana-

da’s maternal blood guidelines have a level of concern

where concentrations are N 5 Ag/L, and an action level

where concentrations are z 100 Ag/L. The most recent

findings (Fig. 5.1.4) indicate that, on average, 43% of

the blood samples from Inuit mothers in NWT/

Nunavut exceeded the level of concern; of these,

87% were b 20 Ag/L, and none exceeded the Action

level of 100 Ag/L (Butler Walker et al., 2003). The

corresponding values for Dene and Me´tis, and Cau-

casians exceeding the level of concern wer e 3.2%

and 0.7%, respectively (Butler Walker et al., 2003).

Fig. 5.1.4 also shows that exceedances of 5 Ag/L

PCB blood guideline vary markedly among five

Inuit regions, with greater exceedances in Baffin

(73%), Kivalliq (59%), and Nunavik (59%), where

the greater levels of PCBs were observed (Butler

Walker et al., 2003; Ayotte, 2001). Inuvik had the

lowest percentage exceedance (16%) among the Inuit

regions. This information on blood guideline excee-

dances supports the previous assessment of dietary

intakes, as it is the Inuit groups that exceed the TDIs

for dietary contaminants intakes as well as the tissue

guidelines.

5.1.1.2. Metals—mercury, cadmium, and lead. Toxic

metals such as mercury, cadmi um, and lead can

bioaccumulate and this was well documented by

Van Oostdam et al. (1999). Concentrations of these

metals vary markedly between tissues and species of

animals. Metals accumulate mostly markedly in

organs such as liver and kidney (Chan et al.,

1995). The dietary surveys outlined in Section 3.1

measured both dietary intakes for all country foods as

well as contaminant levels and allowed estimates of

exposures to be calcul ated for specific population

groups. Fig. 5.1.1 shows that the Inuit have the high-

est intakes of mercury and their mean intakes

approach the provisional tolerable daily intake

(pTDI) for mercury of 0.71 lg/kg/day (WHO, 1978 )

while the mean intakes of mercury among First

Nations peoples in the Yukon and NWT are well

below the pTDI. The pTDI of 0.71 lg/kg/day is

based on total mercury; the pTDI for methylmercury

is 0.47 lg/kg/day. Much of the present data on dietary

levels of mercury in food are only available for total

mercury so the comparisons made in Fig. 5.1.1 are

reasonable. Certain species, such as marine mammals,

contain mostly inorganic mercury while other species

such as fish contain mostly organic mercury ( Wage-

mann et al., 1997), so more accurate risk assessments

could be undertaken if methylm ercury was speci fi-

cally measured in country foods. Recently Health

Canada has developed a new pTDI for methylmer-

cury for children and women of childbearing age of

0.2 lg/kg/day (Health Canada, 1998). It would not be

reasonable to use this pTDI in Fig. 5.1.1 illustrates the

results for all men and women sampled in these

communities, but it does indicate that children and

women of childbearing age could be exceeding the

pTDI.

Percent of population

100

Caucasian

Métis/Dene

Other

Inuit–Baffin

Inuit–Inuvik

Inuit–Kitikmeot

Inuit–Kivalliq

Inuit–Nunavik

Ethnicity–Regional

% at Level of Concern (>5 and <100 µg/L)

0.7

3.2

59 59

Fig. 5.1.4. Maternal blood guideline exceedances for PCBs as

Aroclor 1260 in Arctic Canada, by region and ethnicity (Van

Oostdam, 2001).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 213

Table 5.1.5 shows that for the Inuit regions some

food items with low concentrations of mercury, such

as caribou meat, contribute a small amount but sig-

nificant proportion of mercury exposure due to the

high intake of this country food. Other country foods

only rarely eaten, such as ringed seal kidney, contri-

bute a significant proportion of the contaminant expo-

sure due to the high concentration of mercury therein.

There is also marked regional variability in the coun-

try foods contributing to exposure. In the Baffin

region, where the highest levels of mercury in

human tis sues were found, the most significant

sources of exposure were marine mammals, ringed

seals and walrus. This necessitates dietary recommen-

dations that are specific to local populations and that

reflect the particular species contributing the most

significant exposure.

Exposure to mercury also varies markedly among

the various Inuit groups as was the case for the

various organochlorines. Fig. 5.1.5 indicates that the

Inuit from Baffin have the highest exposure to mer-

cury and that the older age groups have higher expo-

sures to mercury. As was noted for the intake of POPs,

Section 5.1.1, this increased exposure with increasing

age is associated with the corresponding age-related

increased intake of country food.

To further examine metal exposures, data are pre-

sented for Qikiqtarjuaq (Broughton Island), Nunavut,

as these data exemplify a wide range of exposures to

toxic metals (arsenic, cadmium, mercury and lead).

Table 5.1.6 shows that mean intakes exceed the

pTDIs only for mercury. The 95th percentile indicates

the level of intake by the highest 5% of consumers. In

computing the 95th percentile/pTDI ratio, it was

noted that high consumers are exceeding the pTDIs

by nine-fold for mercury and ratios for all other

metals are less than three. This indicates that mercury

is of greatest concern. As noted earlier in the case of

organochlorines, it is unders tood from dietary survey

results that an individual will not likely eat the same

level of high contaminant food items on several con-

secutive days. If more days of intake data from indi-

viduals were available, a tighter distribution and

therefore a lower prevalence of high levels of expo-

sure would likely result. Nevertheless, it is advisable

to discourage high intakes by individuals whenever

possible and, since regular levels of exposure (i.e.,

Table 5.1.5

Proportionate contributions of three main food sources of total

mercury, and total mercury concentrations by food item in five

Inuit regions

Inuit region Food item Proportionate

contribution

of Hg (%)

concentration

(mg/kg)

Inuvialuit Caribou meat 52 0.06

Beluga muktuk 11 0.73

Lake trout 11 0.85

Kitikmeot Caribou meat 27 0.06

Arctic char flesh 15 0.10

Caribou ribs 7 0.14

Kivalliq Caribou meat 38 0.06

Beluga muktuk 24 0.73

Lake trout flesh 12 0.85

Baffin Ringed seal meat 36 0.40

Narwhal muktuk 14 0.56

Caribou meat 12 0.06

Labrador Caribou meat 37 0.06

Lake trout flesh 36 0.85

Ringed seal kidney 6 2.84

Source: Kuhnlein and Receveur (2001).

2.5

1.5

0.5

Labrador

Kivalliq

Kitikmeot

Inuvialuit

Baffin

61+ 41–60 20–40 15–19

TDI

Total Mercury Intake (µg/kg/day)

Fig. 5.1.5. Mean intake of total mercury in different regions (Ag/kg/

day) (Chan et al., in preparation(b)).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246214

levels lower than the ones characterized by the 95th

percentile/pTDI but higher than the pTDI) appear

widespread in the Baffin region, the risk associated

with this exposure needs to be characterized. These

results (Kuhnlein et al., 2000, 2001a,b; Batal, 2001)

confirm the earlier studies reviewed by Van Oostdam

et al. (1999).

To see whether there was a temporal change in

mercury intake in Qikiqtarjuaq (Broughton Island),

data on the daily intakes of mercury collected in

1998–1999 (Kuhnlein et al., 2000) were compared

to those collected in 1987–1988 (Kuhnlein et al.,

1995c; Chan et al., 1997). Intakes of mercury were

similar in the two surveys (Table 5.1.4), indicating

there has been little change in dietary pattern and/or

mercury concentrations in the foods. The major

sources of mercury in 1998–1999 (Kuhnlein et al.,

2000) compared to those collected in 1998–1999 are

ringed seal meat, narwhal muktuk and polar bear

flesh. In 1987–1988, the major sources were also

ringed seal meat, narwhal muktuk and polar bear

(Chan et al., 1995).

5.1.1.3. Con taminant tissue levels and guidelines.

Mercury. In the 1970s Heal th Canada developed

blood guidelines for methylmercury. Blood levels

below 20 Ag/L are classified as being in the acceptable

range, between 20 and 100 Ag/L as bincreasing riskQ

and levels greater than 100 Ag/L in blood as bat-riskQ

(Health Canada, 1979). The United States undertook a

re-evaluation of methylmercury and developed a

benchmark dose level of 58 Ag/L for this metal

(NRC, 2000). Applying the United States’ suggested

10-fold safety factor to the benchmark dose level

allows the development of a maternal blood guideline

of 5.8 Ag/L.

Table 5.1.6

Population distribution of heavy metal intake in Qikiqtarjuaq (Ag/kg bw/day)

Metal N PTDI (Ag/kg/day) N N PTDI N N 0 Mean Median 95th percentile 95th/PTDI

Arsenic 110 2 30 110 1.3 0.3 5.6 2.8

Cadmium 110 1 27 110 0.6 0.2 2.3 2.3

Total mercury 110 0.71 47 110 1.6 0.6 6.4 9

Lead 110 3.57 49 110 3.6 2.7 10.4 2.9

Source: Kuhnlein and Receveur (2001).

Table 5.1.7

Blood guideline exceedances for methyl mercury, lead, and cadmium in Arctic Canada, by region and ethnicity

Region/Ethnic group Country/Ethnic

Group/Region

N Methyl mercury Lead Cadmium

Percent of samples

exceeding 5.8 Ag/L

(%)

Percent of samples

exceeding or equal

to 20 Ag/L

(%)

Percent of samples

exceeding

100 Ag/L (%)

Percent of samples

exceeding 5 Ag/L

(%)

NWT/Nunavut

Caucasian

(1994–1999) 134 0 0 0 6

Metis/Dene

(1994–1999) 92 1 0 2 3

Other

(1995) 13 0 0 0 0

Inuit

(1994–2000)

Baffin

(1996) 31 68 10 10 13

Inuvik

(1998–1999) 31 16 3 3 7

Kitikmeot

(1994–1995) 63 27 0 2 10

Kivalliq

(1996–1997) 17 35 0 0 18

Nunavik

(1995–2000) 162 79

16 12 NA

NA= Not available.

Based on US EPA 1999 re-evaluation of methyl mercury.

Increasing risk range is 20–100 Ag/L, Health Canada.

Guideline value of 5 Ag/L is for occupational exposure.

Source: Butler Walker et al. (2005).

Source: Ayotte (2001).

z 5.8 Ag/L value.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 215

Butler Walker et al. (2005) and Ayotte (2002) show

that among the mothers from NWT/Nunavut and

Nunavik, only Inuit exceeded the Canadian level of

concern of 20 Ag/L and almost no Caucasian or Dene/

Me´tis mothers exceeded the lower guideline of 5.8 Ag/

L based on the United States evaluation. These differ-

ences among the three ethnic groups are almost cer-

tainly due to eating patterns. Among five Inuit

regions, percentage exceedances of the Canadian

guideline level of concern of 20 Ag/L for organic

mercury were highest in Nunavik (16%) and Baffin

(10%) (Table 5.1.7 and Fig. 5.1.6). The percentage

exceedance of the 5.8 Ag/L United States blood guide-

line for organic mercury among Canadian Inuit

women ranged from a low of 16% in Inuvik to

highs of 79% in Nunavik and 68% in Baffin. No

mothers were in the bat-riskQ range of N 100 Ag/L.

The groups that have the greatest exceedance of

these blood guidelines are also the same groups that

most often exceed the tolerable daily dietary intakes

discussed previously.

The proportion of Inuit mothers who are in the

bincreasing riskQ range for mercury has decreased

quite markedly over the years. Wheatley and Paradis

(1996) reported that 56% of mothers included in their

1972–1989 monitoring study were in the bincreasing

riskQ range compared to 3–16% of Inuit mothers in the

1990s (Table 5.1.7). However care must be taken in

concluding trends based on some of the historical

mercury data as this data set may be biased due to

re-sampling communities and individuals with high

mercury values ( Van Oostdam et al., 1999). Based on

the new United States evaluation, a much greater

proportion of Inuit women exceed the guideline indi-

cated that although levels of exposure may have

decreased, concerns for health associated with expo-

sure still exist.

Lead. Canada and the United States use 100 Ag/L

as the action level for lead. In NWT/Nunavut, all

average lead levels in women’s blood samples were

markedly lower than the 100 Ag/L action level, but

3.4% and 2.2% of the blood samples from the Inuit,

and Dene and Me´tis women, respectively, exceeded

the action level (Butler Walker et al., 2005) (see Table

5.1.7). Among Inuit women, percentage exceedances

ranged from a low of 0% in Kivalliq to a high of 12%

in Nunavik. Historical data have shown that levels

among many populations have decreased in parallel to

the phase out of lead in gasoline around the world

(AMAP, 1998), indicating that the lead issue in the

Arctic may be due to local use of lead. Recent

research by Dewailly et al. (2000b) using lead isotope

tracers has indicated that most of the current lead

exposure among Nunavik Inuit results from the use

of lead shot.

Cadmium. The guideline of 5 Ag/L for cadmium

is an occupational level, and, in the absence of any

other guideline, is offered as a general guidance. In

NWT/Nunavut, cadmium levels in women’s blood

exceeded this guideline in 10% of Inuit, 6% of

Caucasians, 3% of Dene and Me´tis, and 0% of

women in the Other group. Among Inuit, the per-

centage exceedances ranged from a low of 7% in the

Inuvik region to a high of 18% in the Kivalliq region

(Butler Walker et al., 2005) (see Table 5.1.7). A

number of studies have shown that there are mark-

edly higher smoking rates among north ern Aborigi-

0000000

Percent of population

100

Caucasian

Métis/Dene

Other

Inuit–Baffin

Inuit–Inuvik

Inuit–Kitikmeot

Inuit–Kivalliq

Inuit–Nunavik

Ethnicity–Re

ional

% >20 µg/L and <100 µg/L (1) % >5.8 µg/L (2)

Fig. 5.1.6. Maternal blood guideline exceedances for organic mer-

cury in Arctic Canada, by region and ethnicity (Van Oostdam, 2001).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246216

nal peoples (Benedetti et al., 1994), and other studies

have found that smoking and not country food con-

sumption contributes most cadmi um exposure (But-

ler Walker et al., 2005).

Radionuclides. The debate continues in interna-

tional scientific circles as to whether there is a dose

threshold for radiation-induced cancer. Radiation pro-

tection authorities have always assumed a linear, no-

threshold hypothesis in setting exposure limits. This

means that all radiation exposures, no matter how

small, carry a risk of producing cancer which is

proportional to the dose received. However there is

mounting evidence that a threshold may exist for

radiation. The US-based Health Physics Society has

recommended against quantitative estimation of

health risk below an individual dose of 50 mSv (milli-

sieverts) in 1 year or a lifetime dose of 100 mSv over

and above normal background radiation (Health Phy-

sics Society, 1996). The issue is of significance to

northerners, who were exposed in the past to nuclear

fallout, or who have slightly higher exposures from

caribou meat consumption.

Even if one assumes that the linear no-threshold

hypotheses are correct, the risk of continued con-

sumption of caribou meat is very small. Table 2.2.9

indicates a maximum dose rate (at present) of about

3 to 4 mSv/year. The International Commission on

Radiological Protection (ICRP) gives a lifetime risk

coefficient of 7.5.



 5

per mSv of exposure for

cancer and other serious illnesses (ICRP, 1991). If a

person were to receive a lifetime radiation dose of:

3:5 mSv=year  70 years ¼ 245 mSv

245 mSv  7:5  10

5

=mSv

¼ lifetime risk of 1:8% from average carib ou

meat consumption:

A normal background radiation of 2.4 mSv/year

would give a lifetime risk of 1.3%. The normal

incidence of cancer in the general population is

approximately 25%. The only way to reduce these

doses further would be to restrict caribou meat

consumption, which would deny northerners the

many benefits of a diet of fish and wildlife. Other

lifestyle choices such as smoking and alcohol con-

sumption have a much greater influence on rates of

cancer.

Recent evidence from A-bomb survivors (Shimizu

et al., 1999) and Chernobyl emergency workers (Iva-

nov et al., 2001) indicates that high dose radiation

may be a factor in increased mortality from cardio-

vascular disease. This link needs to be studied further

in groups exposed to high levels of radiation. The

relevancy of this to northerners who have lower expo-

sures is uncertain at this time.

5.2. Special considerations for risk management in

Arctic communities

5.2.1. Nutri tional benefits

The nutritional benefits of country food and its

contribution to the total diet are substantial, although

only 6–40% of total dietary energy may be derived

from this food source. These foods are important

sources of lipids, vitamins, minerals and protein

and in many cases are the primary source of many

important nutrients (Van Oostdam et al., 1999).

Research has found that days bwithQ traditional/coun-

try food have significantly less fat, saturated fat,

protein, sucrose and total carbohydrate than do

days bwithoutQ country food (Table 5.2.1). These

findings are consistent across the Canadian Arctic

Table 5.2.1

Percent energy from macronutrients on days with or without tradi-

tional/country food (least square meansF S.E.M.)

Nutrient Region With TF (%) Without TF (%)

Carbohydrate Yukon 37.2F 0.7

(n = 410)

41.3 F 0.8

(n = 387)

Dene/Me´tis 34.5 F 0.6

(n = 661)

41.3 F 0.8*

(n = 346)

Inuit 35.2 F 0.5

(n = 968)

46.6 F 0.7*

(n = 632)

Sucrose Yukon 9.1 F 0.4 11.2 F 0.5*

Dene/Me´tis 9.1F 0.4 11.8F 0.5*

Inuit 12.3 F 0.4 16.7 F 0.5*

Protein Yukon 31.1 F 0.5 19.0 F 0.5*

Dene/Me´tis 30.7 F 0.4 20.1F 0.6*

Inuit 34.2 F 0.4 17.9 F 0.6*

Fat Yukon 31.6 F 0.6 40.3 F 0.6*

Dene/Me´tis 34.2 F 0.5 38.8F 0.7*

Inuit 30.1 F 0.4 35.9 F 0.6*

Saturated

fatty acids

Yukon 10.6 F 0.3 14.4F 0.3*

Dene/Me´tis 11.6F 0.2 13.6F 0.3*

Inuit 9.1 F 0.2 12.1 F 0.2*

Source: Kuhnlein et al. (2004).

* Different from with TF, p b 0.01.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 217

and confirm that decreasing country food in the diet

is likely to have negative health consequences in part

through the corresponding increase in total fat, satu-

rated fat and sucrose consumption above recom-

mended levels, a lower intake of the vitamins A,

D, and E, riboflavin and B6, as well as a decreased

use of the important minerals iron, zinc, copper,

magnesium, manganese, phosphorus, potassium,

and selenium (Kuhnlein et al., 2004). Although

some nutrients, such as vitamin C, were provided

mainly by fortified market foods, one should not

undermine the great potential that some traditional

foods could contribute to the diets of indigenous

peoples, if they were better utilized. Fediuk et al.

(2002) listed a variety of rich sources of vitamin C

from animal and plant traditional food such as raw

fish eggs, raw whale skin, the livers of caribou and

ringed seal, and blueberries. With the contemporary

higher levels of fat intake, an increasing number of

cardiovascular deaths can be expected (NRC, 1989);

similarly, high saturated fat intake and high sucrose

intake have been associated with increased risk of

colorectal cancers (World Cancer Research Fund,

1997). The rising levels of excess weight and obesity

documented in Yukon (Receveur et al., 1998a) and

Inuit communities (Kuhnlein et al., 2004) need to be

further considered when the balance between tradi-

tional/country foods and imported foods is evaluated:

18% of women 20–40 years of age, and 33% of

women 40–60 years of age in the five Inuit regions

surveyed by CINE in 1998–1999 had a body mass

index (BMI) over 30. An increase in mean BMI over

a 5-year period (1992–1997) h as also been reported

for women in the Inuit communities of Repulse Bay

and Pond Inlet (Lawn and Harvey, 2001). This

increase in the proportion of energy derived from

fat observed on days when no country food is con-

sumed could be associated with the growing preva-

lence of exce ss weight observed in Yukon and Inuit

communities. Although usually providing smaller

proportions of daily energy and total dry weight in

diets than imported food, country food was shown to

contribute significantly more protein, iron and zinc

to the diets of Baffin Inuit children (Berti et al.,

1999).

In Table 5.2.2, the top three sources of several

nutrients in the total daily diet as reported consumed

in the five Inuit regions show that at least one

country food is mentioned for each nutrient with

the exception of calcium. It is also clear that iron

and zinc are almost entirely contributed by country

food, whereas the top three calcium sources are

Table 5.2.2

Top three sources of selected nutrients from 24-h recalls (fall and late winter combined) in five Inuit regions

Nutrient Inuit region

Inuvialuit (n = 387) Kitikmeot (n = 300) Kivalliq (n = 341) Baffin (n = 522) Labrador (n =417)

Vitamin A Caribou liver Caribou liver Carrots Walrus liver Carrots

Carrots Carrots Beluga blubber Ringed seal liver Margarine

Beef liver Mixed frozen vegetables Caribou meat Narwhal blubber Ringed seal kidney

Omega-3 fatty acids Margarine Arctic char flesh Arctic char flesh Arctic char flesh Margarine

Arctic char flesh Caribou meat Caribou meat Walrus blubber Salmon flesh

Beluga blubber Margarine Beluga blubber Ringed seal broth Salad dressing

Vitamin E Margarine Caribou meat Caribou meat Caribou meat Margarine

Potato chips Bannock Potato chips Bannock Potato chips

Caribou meat Margarine Bannock Narwhal blubber Caribou meat

Calcium White bread Bannock Bannock Bannock Evaporated milk

2% milk Pizza Pizza 2% milk 2% milk

Pizza White bread White bread Pizza White bread

Zinc Caribou meat Caribou meat Caribou meat Caribou meat Caribou meat

Ground beef Ground beef Ground beef Ringed seal meat Ground beef

Caribou dried meat Caribou dried meat Beluga muktuk Narwhal muktuk Chicken

Iron Caribou meat Caribou meat Caribou meat Ringed seal meat Caribou meat

White bread Ringed seal meat Beard. seal meat Caribou meat Partridge meat

Ground beef Caribou dried meat Ringed seal meat Walrus meat White bread

Source: Kuhnlein and Receveur (2001).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246218

entirely from imported food (Kuhnlein et al.,

2001b,c; Blanchet et al., 2000).

Since CACAR-I, new analytical results for nutri-

ents in country foods have shown some remarkable

nutrient properties. For example, exceptional sources

of vitamin C (N 20 mg/100 g fresh weight) in Inuit

country foods have been shown to be beluga muk-

tuk, narwhal muktuk, seal liver, caribou liver, cisco

eggs and blueberries. Excellent sources of omega

fatty acids, vitamin A, vitamin D and vitamin E

have been shown to be beluga blubber and oil,

narwhal blubber, Arctic char, burbot eggs, beluga

muktuk and narwhal muktuk. As well, several nutri-

tionally essential minerals have recently bee n

reported in high levels in Arctic foods, including

iron, zinc, selenium, copper, magnesium and man-

ganese (Blanchet et al., 2000; Kuhnlein et al.,

2001a).

Fish are an important part of the country food diet

across the Canadian North. The health benefits of con-

suming 1–2 servings of fish per week (17–34 g/day) have

been well documented (Simopoulos, 1991; Kromhout et

al., 1995; Weisburger, 2000) and some studies support

recommendations for great er amounts, up to 69 g/day

(Kris-Etherton et al., 2000). Health benefits would

accrue most likely for preventing heart disease (Shmidt

et al., 2000), but a lso possibly for certain cancers

(World Cancer Research Fund, 1997) and type-2 dia-

betes (Feskens et al., 1991, 1995; Ekblond et al., 2000).

Fish intake by Aboriginal peoples is summarized in

Table 5.2.3 and suggests that they are likely to fully

benefit at the current levels of consumption.

Limiting fish consumption in any way based on

levels of organic mercury exposure would be further

unwarranted in Inuit communities, given that the main

source of exposure in these communities was caribou,

representing 52%, 27%, 38% and 27% of total expo-

sure in Inuvialuit, Kitikmeot, Kivallik, and Labrador,

respectively (Kuhnlein et al., 2000). In Baffin, seal

meat was the single main contributor of organic mer-

cury (36% of the total exposure level). Seal meat and

caribou are not highly contaminated with organic

mercury, but their frequent consumption makes them

the main sources of organic mercury intake in Baffin.

In contrast to fish, the health benefits of consuming

seal and caribou, above and beyond what can be

inferred from their specific nutritional values, have

not been studied and remain unknown.

Table 5.2.3

Reported daily fish consumption by gender and age group in three recent dietary surveys among Canadian Arctic indigenous peoples

Indigenous peoples region (main fish species) Women (age 20–40) Women (age 41+) Men (age 20–40) Men (age 41+)

g/person/day

(no. of 24-h recalls)

g/person/day

(no. of 24-h recalls)

g/person/day

(no. of 24-h recalls)

g/person/day

(no. of 24-h recalls)

Yukon first nations (salmon, trout, grayling,

whitefish)

48 (n = 253) 76 (n = 169) 69 (n = 221) 91 (n = 159)

Dene and Metis

Gwichin (whitefish, loche, inconnu, arctic char,

lake trout)

65 (n = 47) 66 (n =54) 35 (n = 51) 158 (n = 43)

Sahtu (whitefish, lake trout, cisco, loche) 13 (n = 49) 154 (n = 42) 37 (n = 46) 105 (n = 43)

Dogrib (whitefish, pike, trout, loche) 8 (n = 28) 82 (n = 29) 32 (n = 21) 120 (n = 31)

Deh-Cho (whitefish, lake trout, loche) 10 (n = 72) 39(n = 38) 37 (n = 51) 13 (n = 56)

Akaitcho (whitefish, lake trout, loche) 21 (n = 85) 34 (n = 67) 44 (n = 79) 30 (n = 80)

Inuit

Inuvialuit (arctic char, whitefish, cisco, herring,

lake trout)

54 (n = 121) 111 (n = 74) 12 (n = 100) 124 (n = 45)

Kitikmeot (arctic char, lake trout, whitefish) 34 (n = 113) 137 (n = 49) 36 (n = 76) 136 (n = 41)

Kivallik (arctic char, lake trout) 21 (n = 112) 109 (n = 68) 65 (n = 87) 169 (n = 48)

Baffin (arctic char, arctic cod, cisco) 39 (n = 160) 72 (n =105) 47 (n = 112) 73 (n = 102)

Labrador (lake trout, salmon, rock cod, arctic char) 24 (n = 112) 50 (n =97) 29 (n =80) 60 (n = 89)

Data adapted from Receveur et al. (1996, 1998) and Kuhnlein et al. (2000).

Estimates obtained by averaging food intake over all 24-h recalls collected in two seasons (Sep–Nov and Feb–Apr).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 219

Other researchers have also noted that the shift

away from country food diets may also be linked to

the rise in obesity, diabete s and cardiovascular disease

among Aboriginal peoples in the United States and

Canada (Young, 1993; Young et al., 1993). Increased

saturated fat, sucrose, and alcohol in diets has led to

higher incidences of gall bladder disease, tooth decay,

alcoholism and its complications, and fetal alcohol

syndrome. Poor diet quality has also been associated

with higher incidences of anemia, otitis media,a

variety of infections, and to some kinds of cancer

(Kuhnlein and Receveur, 1996).

5.3. Social, cult ural, spiritual and economic benefits

of countr y food

Historically, circumpolar northern Indigenous

populations have relied on a variety of Arctic flora

and fauna for survival. The way of life for Aboriginal

peoples in Arctic Canada has been and still is very

much defined by their relationship with the environ-

ment. Today country foods are still central to cultural,

social and spiritual well-being in many regions. They

are essential to the social and cultural health of indi-

viduals and communities (e.g., Van Oostdam et al.,

1999; Receveur et al., 1996; Sante´ Que´bec, 1995;

Condon et al., 1995 ). Country foods, and related

activities (hunting, fishing, collecting, distribution,

preparation, and consumption), still play critical

roles in everyday life in many northern communities

for reasons that include their social and cultural

importance, formal and informal economic value,

and contributions to physical, mental and spiritual

well-being (Dewailly et al., 2000d).

Each food, and its related activities, provide spe-

cific benefits which must be considered when balan-

cing the benefits and risks of exposure to

environmental contaminants through country food

consumption. Arctic Canada is undergoing signifi-

cant economic, political, social, cultural and environ-

mental chang e as a result of influences and pressures

both from within and outside of the region. The

maintenance of the link between the people and

the land is therefore important, further stressing the

importance of the collection, sharing, and consump-

tion of wild foods in the North. Aboriginal peoples

report that country foods define, maintain and

increase aspects of the cultural, social and spiritual

identity and well-being (Dewailly et al., 2000d;

Kuhnlein et al., 2000; Bjerregaard and Young,

1998; Egeland et al., 1998). Culture is strongly

tied to language and participation in traditional activ-

ities and hunting and gathering various traditional

resources provide the opportunity for these aspects

of culture to be expressed. As the Inuit Circumpolar

Conference stated, hunting is an integral social activ-

ity to Inuit communities.

bHunting ... is crucial to sustaining, reproducing, and

expressing Inuit social relations. Where disruptions in

primary cooperative subsistence activities occur,

social relations at the community level suffer. Sealing

... is ... d... the patterned acquisition and use of (an

animal) in such a way as to enhance the social rela-

tionships existing in a community.T ... Inuit have

always depended [sic] animals and other Inuit to

maintain their culture and perpetuate their society.

...sealing continues to establish and reaffirm produc-

tive and cooperative social relationships that are so

crucial to Inuit survival through the hunting and sha-

ring of seals. For many Inuit sealing is ... a dset of

culturally established responsibilities, rights, and

obligations.TQ (Inuit Circum polar Conference, 1996)

Hunting, fishing and gathering of wild resources

and the subsequent sharing of those items with indi-

viduals throughout the community are social activ-

ities bringing together individuals, families and

generations, and are often the focus of celebrations

and festivities. In this way, they form and maintain

an important social fabric among individuals which

supports community health and well-being. For

example, not only is muktuk nutritionally and psy-

chologically beneficial, but its widespread sharing

among relatives and between communities creates

and sustains the bonds that remain the basis of

Inuit social, cultural and economic relationships

today (Freeman et al., 1998). Further, these activities

are opportunities for the transfer of knowledge

between generations and the maintenance of lan-

guage, as they necessitate and use traditional knowl-

edge and components of Abor iginal language, thus

passing on information about hunting techniques,

places and local history while on the land. Similarly,

the preparation of country foods provides opportu-

nities for coming together, learning and sharing

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246220

among individuals in communities. The link between

Aboriginal people and the land, seas and waterways

they travel and use is also regarded in a spiritual

sense as an offering or gift. This spiritual relationship

with the environment is expressed, maintained and

strengthened by the ongoing practices of hunting,

fishing and gathering of resources in a respectful

way. Specifically, in regards to social and mental

health, country foods and related activities are

reported to define and maintain aspects of identity

and Abor iginal culture.

bInuit foods may have contaminants, they also have

much more. They have sustained us, nourished us,

brought us together, and given us the sense of who we

areQ (Egede, 1995).

To the people who have traditionally hunted on

particular traplines, these areas are their primary

world. Their roots and their ideas of who they are,

are tied to their lands. The ties of particular people to

the land are strongest to their own traplines since

much of their own personal history and their family

history is knit into the cultural and spiritual landscape

of the North (Weinstein, 1976). Berkes and Farkas

(1978) state that the retention and rejuvenation of

Aboriginal practices (traditional food habits, such as

eating of fish and game) depend to some extent on

maintaining a sense of identity and pride.

Northerners state:

bInuit foods give us health, well-being and identity.

Inuit foods are our way of life... Total health includes

spiritual well-being. For us to be fully healthy, we

must have our foods, recognising the benefits they

bring. Contaminants do not affect our souls. Avoiding

food from fear does.Q (Egede, 1995).

bFor the old people, the food for them is real food—

strong food like igunaq, seal and all kinds of country

food, and when they do not have it for a long time,

they star t to crave it...it’s their way of life.Q (Quaqtaq

resident, Nunavik; as in O’neil et al., 1997)

bThey are so important to me because they are who

we are. They are part of being Inuit. I have always

eaten them and always will.Q (Labrador Inuk; as in

Furgal 1999)

There are also economic realities which influence

how much Aboriginal peoples rely on traditional/

country food. Employment and income are often

low or uncertain compared to southern Canada,

fluctuating with development projects an d seasonal

opportunities. In many cases and for many people,

the costs of nutritious imported food are prohibi-

tively high in northern communities. For example,

in 1989 the cost to purchase equivalent amounts of

imported meat in local stores was estimated to be

over $10 000 per Aboriginal household per year

(Usher and Wenzel, 1989). Country food, therefore,

is an economic necessity for many Aboriginal

northerners.

In a recent study, Kuhnlein et al. (2000) documen-

ted the cultural, socia l and economic benefits of coun-

try food among adult Inuit in five regions of the

Canadian Arctic. Results indicated that z 90% of

1721 respondents believed that harvesting and using

country food by the family:

1. contributes to physical fitness and good health

2. provides people with healthy food

3. favours sharing in the community

4. is an essential part of the culture

5. is an occasion for adults to display responsibility

for their children

6. provides education on the natural environment

7. contributes to children’s educat ion

8. provides skills in survival

9. provides skills in food preparation at home

In addition, the survey reported that interviewees

recognize country food as being significantly more

healthy for children, healthy for pregnant/breast feed-

ing women, tasty and more important to community

life compared to market food. It is clear from the

research initiative by Kuhnlein et al. (2000) that

country food provides multiple social and cultu ral

benefits to Inuit. The effects of losing these benefits

as a result of decreasing use of country food are of

concern in several ways, including increased inci-

dences of diabetes and heart disease, and effects on

mental health (Van Oostdam et al., 1999).

When country food is compromised by contami-

nants, more than Aboriginal peoples’ health is

affected; their economy, culture, spiritual well-being

and way of life are also threatened. The importance

of country food is further stressed by the lack of

healthy, accessible and economically viable nutrition

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 221

alternatives in many communities and for many

individuals. Many market foods are expensive, of

lower nutritional value and deprive Aboriginal peo-

ple of the cultural and social significance and other

benefits of hunting and consuming country food.

Consequently, the contamination of country food

raises problems which go far beyond the usual con-

fines of public health and cannot be resolved simply

by risk-based health advisories or food substitutions

alone.

5.4. Assessment of perceptions of risks, benefits and

safety of country foods

It is widely understood that the public does not see

risk in the same way as technical bexpertsQ (Kasper-

son, 1986; Erikson, 1990; Douglas, 1992) and that

even among experts there are differences in their

judgments of certain risks (Kraus et al., 1992). Indi-

vidual factors such as age, gender, education, occupa-

tion, language, world view, and culture may all

influence the individuals’ perceptions and concerns

about certain hazards (Flynn et al., 1994; Slovic and

Peters, 1995). Also, certain attributes of hazards have

been identified as being strongly influential on the

perception and acceptance of risks among the public.

These factors include: involuntary exposures, uncer-

tainty about probabilities or consequences of expo-

sure, lack of personal experience with the risk (fear of

the unknown), effects of exposure delayed in time,

genetic effects (effects on the next generation), acci-

dents related to anthropogenic activity, unequal dis-

tributed risks and benefits, and the ease of perception

of the associated benefits (Slovic, 1987; Douglas,

1986; Pidgeon et al., 1992).

For the individual, perception is a combination of

personal and collective (i.e., social) factors that influ-

ence the way in which one understands issues, and to

which one reacts and takes acti on at the individual

level (i.e., personal behaviours). It is recogni zed that

the individuals’ basic conceptualizations for risk are

much richer than that of experts and reflect legitimate

concerns that are often omitted from formal risk

assessments (Slovic, 1987). It is for these reasons

that the perceptions of those involved and affected

by a hazard must be considered as they directly

influence the effectiveness of any risk management

decision and action (including communications) taken

to minimize risks and to maximize the benefits of, in

this case, consuming country foods in the North. It is

critical to assess and understand the northern public’s

perception of food-chain contamination and the way

in which this issue is being addressed in order for

decision-makers and risk communicators to design

successful activities to support the reduction of

risks. Better and more effective risk management

processes, decisions and communications help mini-

mize anxiety, build trust, and avoid negative repercus-

sions (e.g., confusion misperception, rejection of

advice, mistrust, and introduction of exposure to indir-

ect risks).

Consideration of public perception in these activ-

ities supports the development of more effective risk

reduction strategies.

5.4.1. Perceptions of risks in the north

With the issue of food-chain contamination by

various organochlorines, heavy metals and radionu-

clides, through primarily long-range transport,

northern people are faced with involuntary exposure

to a different type of hazard, as it is not easily

observed or detected by conventional means.

Furthermore, because of the central importance of

country foods in Aboriginal communities in the

North, even today, understanding and considering

this context is critical to effectively communicating

and managing these issues. Due to the significant

qualitative and symbolic benefits attributed to coun-

try foods in Aboriginal cultures, the perceptions and

concerns related to food contamination are not sim-

ply proportiona l to the level of harvesting and

consumption activities in communities. Reports of

contamination undermine confidence in the environ-

ment as well as in harvesting activities as sources

of individual and collective well-being. These and

other characteristics complicate the management of

this issue, further emphasizing the need to under-

stand public perceptions of this issue and how they

are addressed.

5.4.2. Research on the perceptions of food-chain con-

tamination in the north

Van Oostdam et al. (1999) detailed efforts made in

assessing and collecting both qualitative and quanti-

tative data on the perceptions of environmental con-

taminants among northe rn residents, and related

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246222

activities during Phase I of the NCP. The discussion of

results presented here summarizes this information

and presents any further knowledge gained in this

area to date.

The Sante´ Que´bec (1994) survey assessed percep-

tions of Nunavik residents towards contaminants in

country foods and revealed the high level of aware-

ness (62%) and the desire to know more (87%) about

this issue. Overall, 55% of Nunavik residents consid-

ered country foods to be more healthful and nutritious

than commercial foods, and 21% believed that com-

mercial foods were of higher quality either because

they were better, more modern, or because country

goods were contaminated. These perceptions varied

with age, with the most favourable attitudes towards

country foods being among individuals in the 25- to

45-year-old age group, and the least favourable atti-

tudes among the younger participants. The survey also

indicated that 14% of people reported having changed

their habits upon becoming aware of food-chain con-

tamination. About 11% reduced country food, while

3% discontinued consumption altogether (Dewailly et

al., 1994 ). The survey, however, did not assess the

duration of these changes. Work done by O’Neil et al.

(1997) and Poirier and Brooke (1997) in Nunavik

communities further documented these positive per-

ceptions of country foods. More recent work by

Dewailly et al. (1999) in Salluit found that most

individuals in this community (71%) believe country

food to be bvery good for their healthQ, as was found

by Furgal et al. (1999) among most (91%) of the

Labrador Inuit surveyed in the community of Nain,

who reported their belief in the safety of country

foods. As Nunavik residents state:

bCountry food is preventing you from diseases.

Therefore it is a medicine. When you are sick and

you are trying to gain back your strength, you eat

country food, it’s your medicine.Q (Quaqtaq resident;

O’Neil et al., 1997)

bIn the winter we feel that wild meat is better to

keep warm and for the body to feel good. We even

bring frozen meat with us to eat while we are out

because it is better for maintaining warmth and

vitality out on the land. It is different in this way

from Qallunat food, like chicken and other fry-pan

type foods, because when we eat this we get cold

more easily and become hungry again rather

quickly.Q (Inuit health official; Poirier and Brooke,

1997)

Despite the strong belief in the health and safety of

country foods, concern is raised in some regions a bout

the presence of and potential effects of contaminants

on animal and human health, as well as the activities

used to investigate these issues in wildlife (Furgal et

al., 1999; O’Neil et al., 1997). Northerners are con-

cerned a bout the health and well-being of wildlife

(Usher et al., 1995; O’Neil et al., 1997; LIA, 1997;

Poirier and Brooke, 1999; Furgal et al., 1999). As one

Nunavimmiut hunter stated:

bI even discarded a beluga, a fine health, young adult

male at his best becau se I thought it had mercury in

him. He had this noticeable but tiny hole in his skin.

I’d hear somewhere that there were people who

injected mercury into beluga and I was afraid that

this one had been contaminated.Q (Nunavimmiut hun-

ter; Poirier and Brooke, 1997)

In a review of 13 contaminant cases in Aboriginal

communities from Alaska to Nunavik, Usher et al.

(1995) found that a lack of attention and understand-

ing of the perception of contaminants among Abori-

ginal peoples in communication efforts had significant

impacts on the receptiveness and effect of messages

that were delivered to address public concerns. In

most cases, communities were alerted to the possibi-

lity of food-chain contamination only from reports of

animal tissues laboratory results, which were some-

times obtained for purposes not even directly related

to local food safety concerns. With the exception of

cases where there were visible pollution events (e.g.,

Exxon Valdez oil spill in Alaska) it was only after the

reports of the presence of the contaminant(s) and the

possible toxic effects that residents started to suspect

that previously unexplained events, particularly

abnormalities in animals, fish or humans, might be

related to contaminants (Van Oostdam et al., 1999).

Almost all cases reviewed gave rise to local uncer-

tainty and anxiety, potentially related to the balance of

clear and credible information on toxicity (especially

in cases of chronic but low-level exposure) or bsafeQ

levels. Such concepts, standards, safety factors, and

chronic exposures are all aspects of risk messages that

are difficult and confusing, yet critical to explain in

risk communication exercises. In cases where there

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 223

has been fairly strong evidence of low risk (e.g.,

cadmium in caribou liver and kidneys) or of benefits

outweighing the risks (e.g., organochlorines in marine

mammal fat), and this is clear ly communicated, the

problem of uncertainty and anxiety appears to have

been minimized (Usher et al., 1995).

The connection between anxiety about potential

health effects from country food consumption and

actual consumption behaviours is complex. Signifi-

cant reductions in country food consumption can

occur when people are informed of binvisible contam-

inantsQ in the local food supply. Yet, invisibility alone

is not the problem, as Aboriginal people have for

years dealt with problems of trichinosis, botulism,

and hypervitaminosis using their experience and tra-

ditional knowledge.

Northerners also regularly report knowing when an

animal is unfit for consumption. These statements are

often based on knowledge of visible abnormalities in

wildlife such as parasitic or other infections. Confu-

sion among these concepts has been reported by

residents in some communities. Participants in a

study conducted in Salluit, Nunavik stated a belief

that organic mercury was caused by a bug, bacteria,

germ or worm (Dewailly et al., 1999). The association

could be made here with a significant amount of work

done in this and other Nunavik communities on tri-

chinosis in walrus. The combination of factors on this

issue means that community residents have only

scientists to trust in confirming the presence of con-

taminants in country food. There is often uncertainty

as to how much to respond to the hazard. They must

rely on individuals using different modes of under-

standing, communication and inquiry, and there are

often competing messages about the nature and extent

of the risks by different experts (Elias and O’Neil,

1995). Works by Usher et al. (1995), O’Neil et al.

(1997) and Poirier and Brooke (1999) show the poten-

tial distrust resulting from this confusion, and the

effects they have on reception of later explanation or

clarification of the situation.

5.4.3. Impacts of these perceptions

Regardless of the knowledge of pollution or con-

taminants, many northerners report that they would

continue to eat country foods despite the advice given

by health officials, because country foods are part of

their culture, lifestyle and health (O’Nei l et al., 1997;

Furgal et al., 1999). Nunavik residents participating in

recent work investigatin g the factors influencing indi-

viduals’ intent to eat country foods (Furgal et al., 2001)

report various advantages to eating country foods,

including: the health and nutritional benefits; the eco-

nomic/cost advantages; as well as aspects of taste and

freshness. These participants and women in the western

Canadian Arctic (Kuhnlein and Dickson, 2001), did not

identify the concern of contaminants or food safety

issues as a factor strongly influencing their choice of

whether or not to eat country foods.

Further work on the potential effects of differing

perceptions and on the public understanding of con-

taminants on individuals’ behaviours conducted in

one Baffin Island community (Clyde River), shows

that some adjustment in hunting activities has

occurred as a result of concern for contaminant in

the local area (Wenzel and Qillaq, 2000). Ongoing

research in Nunavik will shed light on the social

processes involved in risk perception and the effects

of perception on individual health-related behaviours

(Bruneau et al., 2001).

5.5. Risk-benefit characterization

The need for a collaborative, multi-agency

approach in risk assessment and management was

identified in the late 1980s, and has now evolved

into a broad comprehensive framework that is detailed

in this section.

5.5.1. Risk management frameworks

The main objective of the NCP is: to reduce and,

whenever possible, eliminate contaminants in country

foods while providing information that assists

informed decision-making by individuals and commu-

nities in their food use (Van Oostdam et al., 1999). This

could be considered the overriding risk management

goal for all a ctivities under NCP.

The Presidential/Congressional commission on

Risk Assessment and Management presented a very

useful Framework for Environmental Risk Manage-

ment (Fig. 5.5.1) in its final report in 1997. In this

report, principles for risk management decision-mak-

ing were outlined. Among them were:

1. Identifying and clearly characterising the problem

in its public health and ecological context;

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246224

2. Establishing a process that elicits the views of

those affected by the decision, so that different

values and perceptions are considered;

3. Careful analysis of the weight of scientific evi-

dence concerning potential effects to human health

and to the environment; and

4. Considering a range of risk management options.

Health Canada’s updated (1997) risk management

framework descri bed in the document Health Canada

Decision-making Framework for Identifying, Asses-

sing and Managing Risks (Health Canada, 2000), is

similar to the Framework for Environmental Health

Risk Management developed by the Presidential/

Congressional Commission on Risk Assessment

and Risk Management. Guiding principals for risk

management presented in Health Canada’s frame-

work included:

1. maintaining and improving health as a primary

objective;

2. involving interested and affected parties;

3. taking a broad perspective;

4. using a collaborative and integrated approach;

5. tailori ng the process to the issue and its context;

6. using the precautionary approach;

7. clearly defining roles, responsibilities and account-

abilities; and

8. striving to make the process trans parent.

Common to both risk management frameworks are

the importance of engaging stakeholders and affected

parties during each phase of the process; the impor-

tance of appropriate and timely risk communication;

and ensuring that the health risks are considered within

the appropriate ecological and pu blic health context.

5.5.2. Problem identification and context

The problem identification and context stage of the

environmental health risk management framework is

extremely important in identifying, assessing and

managing health risks. In this stage the problem is

identified, put into its ecological and public health

context, risk management goals are determined and

stakeholders are identified and engaged. The NCP has

extensive processes in place to en sure that project

proposals are reviewed by all stakeholders and that

goals and objectives are relevant for the NCP and

follow principles set out by the NCP and its partners.

Project proposals are reviewed by Territorial Con-

taminant Committees (TCCs) in the Yukon, NWT,

Nunavut and Nunavik/Labrador. These committees

have representation from local, territorial and federal

governments, community representatives and Abori-

ginal groups. Committee members rank the proposal

in terms of relevance, communication and capacity

building, among other crit eria. Proposals then go to

review committees for human health, socio-cultural

relevance, education and communication, and biolo-

gical monitoring, where more detailed technical

reviews occur. The proposals undergo a final revie w

at the level of the NCP Management Committee.

Stakeholders have the opportunity at several levels

to be engaged in the problem/context stage of the

NCP risk management framework.

A small amount of funding is also available for

locally identified contaminant concerns, such as a

potential contaminated site or wildlife species that

may pose a risk to the health and well-being of

local community members. This is one example

under NCP where those who are affected by the risk

management decisions are directly involved in identi-

fying and characterizing the problem, and in each

subsequent step of the risk management process.

5.5.3. Risk and benefit assessment

Benefit and risk assessment provides the scientific

foundation for risk management decision-making. It

Fig. 5.5.1. Framework for environmental health risk management

(Presidential/Congressional Commission on Risk Assessment and

Risk Management, 1997a,b).

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 225

involves the careful objective analysis of the weight of

evidence and scientific information on risks and ben-

efits, and thorough consideration of the qualitative

values associated with risk perception, and social,

cultural and spiritual identity.

Benefit and risk assessment includes the identifi-

cation and develo pment of options to manage the

health risk. Under the NCP framework, the assess-

ment of risks and benefits is undertaken in a multi-

disciplinary fashion with participation from research-

ers, severa l levels of government, and community

representatives. This process evolves as new scientific

information on health outcomes, nutritional, socio-

cultural and economic benefits, and risk perception

becomes available.

Under the NCP, benefit and risk assessmen t is

undertaken on a case-by-case basis and involves

both scientific evidence and good judgement. Risks

are assessed by Health Canada using their risk assess-

ment process (Health Canada, 1993, 1994, 2000), and

in recent years also by the Centre for Indigenous

Peoples’ Nutrition and Environment (CINE). The

risk assessment process used for assessing the

human health implications of country food consum p-

tion in Arctic comm unities was described in detail in

the CACAR-I (Van Oostdam et al., 1999), and is

summarized in the following text.

5.5.3.1. Risk assessment. There is no universal defi-

nition of health risk. It has been defined as: the prob-

ability that a substance or event will produce harm

under specific conditions (US EPA, 1997). Risk arises

from exposure to a harmful substance or event and

depends on whether that substance or situation can

cause harm. Risk is determined by the careful con-

sideration of the nature, likelihood, and severity of

adverse effects on human health and the environment.

What must be understood is that risk is complex and

multifaceted. It does not mean the same thing to all

people, and perceptions of risk are important in both

risk assessment and management, and in related

decision-making.

Risk assessment is the process wherein information

is gathered and analyzed to determine the likelihood

that a specific adverse health effect will occur in an

individual or populatio n following exposure to a hazar-

dous agent. This process often establishes a best esti-

mate of the risk as there are often gaps in information.

The Health Canada risk assessment process con-

sists of four steps: 1) hazard identification; 2) hazard

characterization; 3) exposure assessment; and 4) risk

characterization.

Hazard identification and characterization are based

on toxicological studies conducted in laboratories and/

or epidemiological studies to establish dose–response

relationships for contaminants. This invol ves gather-

ing information on what type of adverse health effects

might be expected as a result of exposure to the agent,

and how quickly these effects might be experienced.

The no-obs erved-adverse effect-level or NOAEL, is

the dose at which no biologically adverse effects are

observed in the study population as compared to the

control population.

After the identification of the appropriate NOAELs

and uncertainty factors, exposure guidelines (accepta-

ble or tolerable daily intakes, provisional tolerable

daily or weekly intakes: ADI/TDI/pTDI/pTWI) can

be calculated by dividing the NOAEL by the uncer-

tainty factor. The uncertainty factors adjust for the

inter- and intra-species variation between and within

animal and human populations. These intakes when

averaged over the course of the entire human life

span are thought to represent the exposur e for humans

that is without appreciable risk of adverse health

effects, based on the available information (Van Oost-

dam et al., 1999). These guidelines are reviewed when

new toxicological and epidemiolo gical information

becomes available, as was the case for mercury and

for the pTDI for women of childbearing age and for

children established in 1998 (Health Canada, 1998).

To estimate risk, contaminant levels in country

foods are used to estimate the probable daily intake

(PDI). The PDI is calculated by multiplying the aver-

age contaminant concentration in a food by the esti-

mated daily intake of the food and this number is then

divided by the average body weight of the consumer.

The PDI is then compared to the exposure guideline

value for the particular contaminant in question to

estimate risk. If the PDI exceeds the TDI, then advice

to limit consumption may be issued, and risk manage-

ment decisions need to be considered. Risk character-

ization, benefit assessment, and the identification and

evaluation of risk management options are completed

by the Territorial Health agencies and their partners.

These processes often involve the multi-stakeholder

contaminant committees established under NCP in

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246226

each territory and in the northern provincial regions of

Nunavik (Que´bec) and Labrador.

Different risk assessment procedures are used for

substances determined to be genoto xic carcinogens. A

more conservative approach is taken where no expo-

sure level is considered to be free of risk. Animal and/or

human study data are used to mathematically predict

the exposure effect on various cancer risks, with risks

usually assessed as one additional cancer per 100,000

population (10

 5

risk) or one additional cancer per

million (10

 6

risk) (Van Oostdam et al., 1999).

In recent years, the Centre for Indigenous Peoples’

Nutrition and Environment (CINE) at McGill Uni-

versity has been active in assessing risks and benefits

associated with country food consumption. For

example, a risk characterization of arsenic in berries

in the Akaitcho Territory, as was toxaphene in fish

from Fort McPherson, NWT (Stephens and Chan,

2001) was completed. CINE assesses risks using a

process similar to that previously described, but with

a significant difference, in that they use information

from dietary studies they have conducted under the

NCP across northern Canada. These dietary studies

(Receveur et al., 1996, 1998; Kuhnlein et al., 2000)

provide consumption informat ion by community, by

sex and age group, and by season. In addition, these

studies present information on the nutritional benefits

of country foods being consumed, on the cultural and

social significance of country foods and on risk

perceptions.

5.5.3.2. Benefit assessment. Comparing the quanti-

tative risks with primarily qualitative benefits asso-

ciated with country food is very challenging. Attempts

have been made to assess the nutritional benefits of

country food and to qualify the social/cultural bene-

fits. However, a formal quantitative methodology to

weigh and compare the many benefits associated with

the health risks of a diet of fish and wildlife has not

been established. For example, it has been well

demonstrated that fish consumption has significant

health benefits in reducing cardiovascular disease

among northern populations. Benefits are assessed

and relative risks related to the benefits are consid-

ered, as are perceptions of risk.

Benefits are often assessed at the territorial or local

level. Territorial Health Departments and TCCs (or

sub-committees of these groups) come together to

discuss benefits, and the other determinants involved

in the risk characterization. Information on the bene-

fits of country food consumption gained from research

activities is considered, and local Aboriginal represen-

tatives are included in the meeting to ensure that their

knowledge and perspectives are included. Local repre-

sentatives are best positioned to consider the socio-

cultural and economic importance of specific country

food items in the region of concern.

5.5.4. Risk characterization

Risk characterization provides a key source of

information for risk management decision-making.

It is a process where the uncertainty and assump-

tions that were used in the risk analysis are evalu-

ated and the strength of scien tific evidence for health

effects associated with the contam inant is consid-

ered. Risk characterization involves the use of reli-

able scientific information from a range of

disciplines (e.g., biology, physiology, economics,

and social sciences) and shoul d provide opportu-

nities for discussion and deliberation. Scientific

uncertainties, related assumptions, potential impacts

on decision-making and health-related needs are

carefully considered. The success of the risk char-

acterization process depends on the participation of

interested and affected parties in discussions, so that

they may understand and participate effectively in

the risk management process.

Within the NCP risk management framework,

risk characterization is discu ssed by risk assessors,

risk managers, health authorities, Aboriginal groups,

community or regional representatives, territorial and

federal governments and researchers. Risks are char-

acterized and discussed within the same forum in

which health benefits and risk management options

are discussed. Risk assessors and scientists provide

information on the uncertainties and assumptions

that are associated with the TDIs and ADIs for

various contaminants and on the strength of evi-

dence for health effects associated with a particular

contaminant. Researchers provide information on

covariables for studies to assess all factors that

may have had an influence on outcomes of a

study. Territorial government and community repre-

sentatives, local health author ities, Abor iginal groups

and other interested parties help to ensure that all

health-related questions of concern are answered and

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 227

that the risk characterization has put the risk in the

proper perspective.

5.5.5. Assumptions/uncertainties of concern

The effects of chronic low-dose exposure in

human popula tions remain controversial because of

the inconsistency of findings among different studies

and populations. Moreover, multiple factors such as

smoking, substance abuse, nutrition and pre-existing

health status can potentially affect the development

of health effects associated with a particular contami-

nant. Addition al considerations must include the fact

that most country foods contain many toxic metal or

organochlorine contaminants (Chan et al., 1995,

1997; Kuhnlein et al., 1995c; Berti et al., 1998a).

Risk characterization of multiple chemical exposures

poses a major challenge for risk managers, as most

risk assessments address only single chemicals.

Information on toxicologic interactions of environ-

mental contaminants is limited. However, there is

ample evidence sugges tive of supra-additive and

infra-additive interactions among environmental con-

taminants (Krishnan and Brodeur, 1994). Research is

being conducted under the NCP to gain a better

understanding of toxicologic interactions between

contaminants and between nutrients and contami-

nants. However, research has not progressed far

enough to allow this to be included in the present

risk characterizations.

5.5.6. Weighing benefits and risks—challenges in

practice

After the risks are characterized and the nature and

extent of uncertainty and assumptions are understood,

the characterized risk must be weighed and compared

with the known and potential benefits. Risks and

benefits are discussed and weighed in a multi-agency

forum, with participation from risk assessors, risk

managers, health authorities, Aboriginal governments,

and community or regional representatives, territorial

and federal governments and researchers. It is impor-

tant to have as much technical and non-technical

information as possible on which to base the compar-

ison of risks and benefits, and to consider the per-

spectives and interests of all affected parties. These

working groups or committees must take on the chal-

lenging task of considering the uncertainty associated

with the health risks, the difficulty in comparing

quantitative risks with qualitative benefits, and the

consideration of other associated health risks and

issues in the region.

5.5.7. Option analysis/evaluation

Once risks have been characterized, risk manage-

ment options can then be identified and evaluated. At

this stage, the specific nature of regional issues must

be considered, and public perceptions, uncertainties,

and the social, economic and cultural consequences of

the options are to be taken into careful consideration.

Under the NCP, the identification and evaluation of

risk management options occur at the level of the

Territorial Contaminants Committees, or a sub-set of

these in the form of a Heal th Advisory Working

Group. Those invol ved in identifying and analyzing

risk management options include: territorial or pro-

vincial health authorities, Aboriginal groups, scien-

tists, risk assessors (Health Canada or CINE)

community or regional repres entatives, environmental

health officers, nutritionists and local and /or federal

government representatives. It is within these groups

that benefits are identified, assessed and compared to

risks; assumptions and uncertainties associated with

the risks are discussed; and local perceptions of risks

are considered.

It is important that options are identified with the

wide rage of interested and affected parties, because

the responsibility for managing the risk is often

shared, and various organizations may be involved

in implementing the selected strategy. Interested and

affected parties can help to identify criteria for ana-

lyzing options by providing requi red information,

participating in the analyses, and providing advice

as to whether the results of the analysis are acceptable

and help to redefine risk management goals as

required.

When analyzing potential risk management strate-

gies, the following are considered: risks vs. benefits;

expected costs of implementing the plan; available

resources; unintended consequ ences (creation of new

risks, or unwanted social, cultural, ethical, environ-

mental and other indirect health impacts); and the

perceptions, concerns and values of interested and

affected parties. Risk management options relevant to

the issue of contaminants in country food can include:

regulatory measures (at local or international level); the

release of consumption advice or identification of alter-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246228

native food sources; education and communication

strategies to help people make informed decisions; or

not taking action when action is not required.

5.5.8. Selecting a risk management option

The results of the optio n analysis are used to select

the most appropriate risk management option or

options. Selecting an appropriate risk management

strategy is always a challeng e because complete infor-

mation is often not available. Uncertainty factors relat-

ing to the intake values (TDI/ADI) may be large,

making it difficult to accurately characterize the risk

for particular contaminants. Data on dietary exposures

may be limited and actual exposures may be lower than

estimated. Good information on risk perceptions may

also not be available.

There have been several cases for the Canadian

Arctic wher e recommendations to continue consuming

country foods have been made despite the potential risk

of adverse effects from exposure to contaminants. The

benefits of continued consumption are great, and there

is uncertaint y associated with the assessment of risk.

All individuals are exposed to mixtures of contami-

nants, not sole contaminants as health risk assessments

assume; there is evidence of interaction between nutri-

ents and contaminants. In addition, the assessments are

based on daily consumption levels over a lifetime,

which are not consistent with the largely seasonal diet

of northern Aboriginal peoples.

5.5.9. Implementation

Once the preferred risk management option is

selected, a strategy for its implementation is developed.

All affected parties need to participate in developing

the management strategy. A protocol is established for

the release of information related to contaminants in a

prompt and understandable way. The communication

of information must take into consideration cultural

differences, language differences, and other potential

barriers to comprehension such as individual percep-

tion of risk. Personal perception of risk can have a

significant impact on personal and community reaction

to the information presented (Van Oostdam et al.,

1999).

5.5.10. Monitoring and evaluating the decision

Once a risk management decision has been made, its

implementation is carefully monitored and evaluated to

see if it is appropriate and effective. If implementation

problems occur, or if knowledge of the hazard or the

risk changes, the decision should be revie wed. This

step is extremely important and illustrates how the

process of risk management is an evolving process

and is subject to change as new information about the

situation is learned and assessed. This approach must

be continually modified to suit each situation and each

community (Van Oostdam et al., 1999).

5.6. Risk and benefit communication

Effective communication is fundamental to the aims

of the NCP. Northerners must have appropriate infor-

mation on the contaminant risks, and nutritional bene-

fits to make informed decisions about the harvesting

and consumption of country food. Communicating risk

and benefit information in a balanced, clear, accessible

and meaningful way to northern communities requires

sound and accurate technical information as well as

respect for and understanding of the knowledge, per-

ceptions, concerns and priorities of northerners. Thus,

benefit and risk communication of the potentially sen-

sitive environmental health information generated

under the NCP presents its own unique set of consid-

erations and challenges, and has been a focus of

resources, efforts and research during NCP-II.

Communicating about contaminants is typically a

challenging process in the North, due to part nature of

the subject matter. Stiles and Usher (1998) acknowl-

edge that messages describing risk are the most diffi-

cult ones to generate. There is a fundamental

difference between the quantitative way in which

risk assessment results are characterized and the qua-

litative terms with which the general public thinks

about risks. While health professionals are accus-

tomed to risk calculations given as probabilities with

built-in uncertainty factors, people receiving the mes-

sage generally want definitive, clear-cut answers to

the question of bWhat does it mean to me?Q or bIs my

food safe to eat?Q. This requires that technical termi-

nology be put into plain language, whether be it

English, French or the Aboriginal language under-

stood by the audience; that visual representations of

risk avoid the use of complex data graphs and figures;

and that the messages be developed and delivered

with consideration for the cultures and knowledge

systems of Aboriginal peoples.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 229

The contextual, technical, social and procedural

challenges for communicators associated with these

differences were reported in detail in CACAR-I (Va n

Oostdam et al., 1999) and by Usher et al. (1995), and

reiterated in more recent work by Lampe et al. (2000)

and Furgal et al. (2005). These reports emphasize the

fact that within the cultures of northern Aboriginal

peoples, the environment, including the foods that it

offers, is a defining element of identity and overall

well-being. Country foods are more than a source of

nutrition. It is an important source of economic, cul-

tural, social and spiritual well-being. Practical alter-

natives to country food are also not readily accessible

to many northerners due to the associated costs and

availability. Communicators must thus be sensitive

and aware that any information that leads to a disrup-

tion of country food production, sharing and con-

sumption patterns of northern comm unities has the

potential to affect the northern way of life and the

social fabric that keeps communities healthy, active

and sustainable.

Effective risk and benefit communication is a pro-

cess, not simply a product (Usher et al., 1995). Ideally,

it is a dialogue: an ongoing two-way or multi-direc -

tional exchange of information, perspectives, ideas

and feelings among individuals that creates and fos-

ters opportunities for feedback and mutual learning.

Effective risk and benefit communication use various

media (e.g., reports, pamphlets, posters and audio–

video materials), not to substitute face-to-face discus-

sion, but to complement, enhance and facilitate dia-

logue and exchange. Ideally, too, it will portray a

situation accurately such that recipients of the infor-

mation will have a sound and shared appreciation of

the issue at hand and its solutions, and will feel

empowered to take action or make informed choices

(Van Oostdam et al., 1999). Practical guidelines for

developing communications materials specific to var-

ious media and delivering presentations to northern

audiences have been developed by Stiles and Usher

(1998) and Lampe et al. (2000), and in-depth assess-

ment of NCP communications materials and methods

can be found in the Knowledge in Action publication

(see Furgal et al., 2003 for more details).

An appreciation o f the profound importance of

effective communication was gained through the

experiences of early communications efforts that fell

short of the ideal. These instances made it evident that

misguided communications efforts, however well-

intended, can and have led to significant impacts,

both direct and indirect, intentional and unintentional,

among the affected populations. The implications of

poor risk communication of contaminants in country

food and recommendations for improving communi-

cation efforts are becoming well documented (Wheat-

ley and Wheatley, 1981; Usher et al., 1995; Powell

and Leiss, 1997; Van Oostdam et al., 1999; Lampe et

al., 2000; Furgal et al., 2005). In their revie w of 13

instances since the 1970s of dealing with contam inant

issues in Aboriginal communities, Usher et al. (1995)

reported that deficient communications efforts on

environmental contaminant had left communities feel-

ing anxious, angry and confused and disrupted social

and cultural aspects of life relat ed to harvesting and

consumption of country food.

Usher et al. (1995) reported that 3 years following

the completion of follow-up studies, residents were

still anxious about the health risk posed by PCBs, and

attributed many of their health problems, including

cancer, suicides and premature births, to these con-

taminants. Hunters in one community had also shifted

their seal hunting areas more distant from a perceived

area of contamination (Wenzel and Qillaq, 2000). In

this case, the communication process had begun in

earnest by the researchers, through seeking appropri-

ate permission to conduct the study and informing

residents by co mmunity radio. Several factors, how-

ever, compounded to create the communication crisis.

This may have been averted through a more careful

consideration of the particula r risk and benefit trade-

offs of country food consumption, and by a thorough

communication plan developed in cooperation with

the community using funds allocated for communica-

tion activities (Usher et al., 1995; Powell and Leiss,

1997).

The practice of risk communication related to con-

taminants in northern country food has changed over

the past two decades, building upon lessons learned

along the way. This parallels loose ly what Powell and

Leiss (1997) describe as three distinct stages in the

evolution of risk communication:

1. The early days focussed on the quantitative and

comparative expressions of risk estimates;

2. The next phase adopted additional aspects of

enhancing the credibility of the source and mes-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246230

sage by building trust through public discussion

and developing messages based on unde rstanding

the information needs, perceptions and concerns of

the audience;

3. The current efforts further emphasized the social

context of risk management by building effective

stakeholder relationships through long-term orga-

nizational commitments to practising responsible

risk communication.

In terms of the NCP, this has meant a transition from

risk communication to risk-benefit communication;

that is, from delivery of risk assessment results to

ongoing two-way dialogue on the benefits and risks

associated with the issue. Risks are placed within the

context of the nutritional, social, cultural, economic

and spiritual benefits of consuming country food, and

within a greater public health context relating to general

health and well-being. In place of one-way models of

message dissemination, NCP-II strove toward shared

decision-making through partnerships involving terri-

torial and regional contaminants/health comm unities

and representatives of Aboriginal organisations at all

stages of the risk management and risk communication

process. As a result of these changes, there has also

been a shift away from health authorities issuing advi-

sories to providing general advice which is discu ssed

and delivered by individuals who are trusted in the

community, using various formats and materials.

To support the shared decision-making process for

benefit and risk communication, the NCP established

various protocols and supports networks in all north-

ern regions. In the Yukon, NWT and Nunavut, the

Territorial Contaminants Committees, whose mem-

bers included representatives of Aboriginal organiza-

tions and Territorial Health Departments, determine

the applicability of Health Canada recommendations

to northerners and devise an appropriate communica-

tion plan for the release of the information. Depending

on the significance of the risk assessment results, the

information may be released through a number of

methods, including a press release, radio phone-in

shows, and various forms of print media. The Terri-

torial Health Departments, in association with the

regional Aboriginal organizations, take the lead in

ensuring that people receive the appropriate informa-

tion. Similar processes are in place in Nunavik and

Labrador, involving, respec tively, the Nunavik Re-

gional Board of Health and Social Services in con-

sultation with the Nunavik Nutrition and Health

Committee, and the Labrador Inuit Association and

Labrador Inuit Health Commission in consultation

with the relevant health authority.

6. Conclusions

The key objectives of this paper are to assess the

impact of exposure to current levels of environmental

contaminants in the Canadian Arctic on human health,

and to identify the data and knowledge gaps that need

to be filled by future human health research and

monitoring.

6.1. Aboriginal perspectives on food and health and

interpretation of research results

Traditional/country foods are an integral compo-

nent of good health among Aboriginal peoples, and

their social, cultural, spiritual, nutritional and eco-

nomic benefits must be considered in concert with

the risks of exposure to environmental contaminants

via consumption of traditional/country foods.

Persistent environmental contaminants such as

PCBs, toxaphene, DDT and mercury biomagnify and

bioaccumulate in the food chain. Because a substantial

proportion of the Aboriginal diet consists of tradi-

tional/country food, Aboriginal peoples have a greater

contaminant exposure than non-Aboriginal peoples or

people in southern Canada. Traditional/country food is

an economic necessity for many Aboriginal peoples,

and provides additional benefits beyond the activity of

hunting itself, as it is the tie that binds Aboriginal

peoples together and allows the sharing and teaching

of their traditional social and cultural values.

In assessing causal associations between findings

in any epidemiological or toxicological study the

following factors must be considered: strength and

magnitude of the association, consistency of the asso-

ciation, dose–response relationship, temporally cor-

rect association, and biological plausibility of the

association. Determining the adverse human health

effects due to the presence of contaminants in tradi-

tional/country foods is very difficult as many factors

contribute to the health of an individual and in fact

contaminants may only play a modest role in deter-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 231

mining an individual’s health status. Factors such as

lifestyle (e.g., alcohol consumption, smok ing, and

substance abuse), diet, socioeconomic status and

genetic predisposition need to be considered when

evaluating the results in this report.

6.2. Exposure assessment

A recent assessment of the average weekly fre-

quency of consumption of main traditional/country

food items in Dene and Me´tis, Yukon, and Inuit

communities has shown a very wide range of these

foods, including more than 250 different species of

wildlife, plants and animals. Regional differences in

species used most frequently are due to ecosystem

variety and cultural preferences Traditional/coun try

food use for women and men 20–40 years of age is

highest in Inuit communities, followed by Dene and

Me´tis of the NWT and then First Nations people of

the Yukon. It was also noted that for Can adian

Inuit, intakes of traditional/country food do not

seem to have significantly changed in the last 20

years.

Inuit mothers have oxychlordane and trans-nona-

chlor levels that are 6–12 times higher than those in

Caucasians, Dene and Me´tis or Other mothers, with

Baffin Inuit mothers having the highest levels. Simi-

lar patterns were observ ed for PCBs, HCB, mirex

and toxaphene. Levels of h-HCH are 5–12 times

higher among the Other mothers than in Inuit, Cau-

casians or Dene and Me´tis mothers, while DDE

levels in the Other mothers are also roughly 3–6

times higher. This higher exposure may be due to

the fact that the Other mothers consists of people

whose exposure may be related to their possible

African or East Asian country of origin or foods

imported from these regions. PCBs are a major con-

taminant of concern in the Arctic. Inuit mothers have

the highest levels of PCBs compared to Caucasians,

Dene and Me´tis and the Other mothers, with Baffin

Inuit having the highest levels. Caucasians mothers

from the NWT have slightly higher levels of total

TCDD–TEQs (dioxins and furans and dioxin-like

PCBs) than Inuit or Dene mothers. The relative

contributions of dioxins and furans and dioxin-like

PCBs to the total TCDD–TEQs are very different in

the three groups. Among Inuit mothers dioxin-like

PCBs account for 80% of the total TCDD–TEQs,

while for Caucasian mothers its only 27%; Dene

mothers have a relatively equal contribution from

both sources.

Recent research has revealed significantly higher

levels of organic mercury in maternal blood of Inuit

women compared to Caucasian, Dene and Me´tis, or the

Other group. Nunavik and Baffin Inuit have higher

levels in maternal/cord blood than those seen in the

Kitikmeot, Kivalliq and Inuvik regions. Mercury levels

in Salluit resident’s vary markedly from year to year

due to variation in availability of traditional/country

foods, but recent tissue levels are similar to the levels

found in the late 1970s.

Research on lead isotope signatures has shown that

the slightly elevated blood lead levels among some of

the Inuit groups and the Dene and Me´tis are likely due

to the use of lead shot in hunting of traditional/country

foods, and thus its presence in consumed wild game.

Blood cadmium levels are roughly 1.5 to 5 times higher

among Inuit than in Dene and Me´tis, Caucasians, and

Other. This difference is likely due to the high rate of

smoking among Inuit mothers and the high cadmium

content in Canadian tobacco.

Radiocesium entered the atmosphere as a result of

atmospheric testing of nuclear weapons. Since the

CACAR-I (Van Oostdam et al., 1999), measurements

of radiocesium levels in Canadian caribou herds

have shown that levels are now considered insignif-

icant and are continuing to decrease, with an ecolo-

gical half-life of about 10 years. The radionuclides

lead-210 and polonium-210 are naturally occurring

and supply most of the radiation dose to consumers

of traditional food. These radionuclides have been

present in the Arctic envir onment for thousands of

years at more or less the same concentrations as

today and accumulate in the lichen–caribou–human

food chain. Current estimates of polonium exposure

would give a radiation dose of 3 to 4 mSv/year to a

high consumer of caribou meat over and above

normal background radiation exposure of 2 to 3

mSv/year.

6.3. Toxicology

In experimental studies, to clarify the potential

toxicity of toxaphene, a no-observed-effect-level

(NOEL) of 0.1 mg toxaphene/kg bw/day was found

for non-human primates regarding immunological

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246232

effects. These findings support those seen in rodents

where doses were 10 times higher (no NOEL was

found in these studies). In certain Arctic populations,

toxaphene residues are within the same range as the

experimental studies where effects on immune func-

tion and infant size have been observed. While more

information is needed on the carcinogenic and endo-

crine-disruptive capacity of toxaph ene, new and forth-

coming information will be useful in a re-assessment

of the toxicity and provisional tolerable daily intake

(pTDI) of this chemical.

The pattern of chlordane-related residue accumu-

lation in humans parallels that seen in the highest

levels of the Arctic marine food chain. Estimated

chlordane intake values for Arctic residents vary

from levels below the Health Canada pTDI to levels

approaching or exceeding the higher US EPA TDI,

depending on marine mammal fat consumption.

More research is required on to clarify the risk

that chlordane poses to human health. Based on

the latest rodent studies at Health Canada, a thor-

ough risk assessment with the goal of re-assessing

the TDI should take into account that the major

chlordane metabolite oxychlordane is almost 10-

fold more toxic than the parent chlordanes and

nonachlors, and that trans-nonachlor and oxychlor-

dane are among the more bioaccumulative chlordane

contaminants.

There is some evidence that dietary nutrients

such as polyunsaturated fatty acids, fish protein,

and selenium may interact with methylmercury toxi-

city at least on the mechanistic level. The two

former nutrients may be protective against methyl-

mercury neurotoxicity; however, there is no conclu-

sive evidence of such an effect. Selenium has been

the main nutritional factor considered by epidemio-

logical and clinical studies to date. No epidemiolo-

gical studies, however, have ever shown a

correlation between selenium intake and the occur-

rence or absence of symptoms for methylmercury

intoxication. There are also inconsistencies in the

protective effects seen in animal studies, and the

role of selenium remains to be confirmed. In animal

studies, selenium alone did not provide protection

against fetal mortality caused by in utero methyl-

mercury treatment, while vitamin E alone and the

combination of vitamin E and selenium treatments

did.

6.4. Epidemiology and biomarkers

Neurobehavioural effects of perinatal exposure to

environmental contaminants are best studied in pro-

spective longitudinal cohort studies starting during

pregnancy because there are numerous other determi-

nants of infant and child development that need to be

documented and are best assessed during pregnancy or

at birth. Examples of these determinants are prenat al

exposure to alcohol, drugs and tobacco; exposure to

other environmental contaminants such as lead; and

intake of nutrients such as selenium and omega-3 poly-

unsaturated fatty acids (PUFA) during pregnancy.

For methylmercury, the Faroe Islands study appears

to be the most useful to extrapolate results to the

Canadian Arctic [exposure to PCBs and mercury, simi-

lar sources (e.g., marine mammals), intake of omega-3

polyunsaturated fatty acids]. Differences remain, such

as the genetic profile, a higher intake of selenium

among Inuit, and a different pattern of exposure

(peaks). Detailed analyses provide some evidence

that the neurobehavioural deficits identified are the

effects of methylmercury and not the consequence of

exposure to PCBs.

In Nunavik, otitis media (middle ear infection)

during the first year of life was associated with pre-

natal exposure to p,pV-DDE, HCB and dieldrin.

Furthermore, the relative risk of recurrent otitis

media increased with prenatal exposure to these com-

pounds. To date, the ongoing Nunavik study supports

the hypothesis that the high incidence of infections

observed in Inuit children (mostly respiratory infec-

tions) maybe due in part to high prenatal exposure to

POPs, although definitive conclusions need to await

final adjustment for important possible confounders

(e.g., parental smoking, and vitamin A).

The strength of the negative association of PCBs

on birth weight for newborns in northern Que´bec was

comparable to the associations with prenatal exposure

to alcohol and smoking during pregnancy. The nega-

tive effects of prenatal PCB exposure on birth weight

and duration of pregnancy could still be demonstrated

despite the beneficial effects of omega-3 fatty acids

from the traditional/country diet. The ongoing Nuna-

vik cohort study should help to shed more light on

contaminant exposure and potential related health

effects in the areas of neurodevelopment and immune

function.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 233

6.5. Risk and benefit characterization, assessment and

advice

The primary contaminants of concern in the con-

text of traditional/country food consumption in Arctic

Canada are the persistent organic pollutants PCBs,

chlordane and toxaphene, the toxic metal mercury

and naturally occurring radionuclides.

In recent dietary surveys among five Inuit regions

(Baffin, Inuvialuit, Kitikmeot, Kivalliq and Labra-

dor), mean intakes by 20- to 40-year-old adults in

Baffin, Kivalliq and Inuvialuit communities exceeded

the pTDIs for chlordane and toxaphene. High con-

sumers (95th percentile) of traditional/country foods

are exceeding the pTDIs by many-fold for toxaphene,

chlordane, and PCBs. The Inuit populations that had

the greatest exceedance of the pTDIs also had the

greatest exceedance of the PCB maternal blood

guideline.

In one Inuit community, Qikiqtarjuaq (Broughton

Island), Nuna vut intakes of organochlorines including

PCBs, chlordane, and toxaphene were higher in 1998–

1999 than in 19 87–1988, particularly among the high-

end consumers (95th percentil e). While the organo-

chlorine concentrations used for the estimations were

similar, variations in the amounts of certain tradi-

tional/country foods consumed likely account for

this difference.

Mercury is the toxic metal of greatest concern in

the Canadian Arctic. Dietary intakes of mercury were

similar in the two surveys conducted in 1987–1988

and 1998–1999, indicating that there has been little

change in dietary pattern and/or mercury concentra-

tions in traditional/country foods. Based on the excee-

dances of the provisional tolerable daily intake (pTDI)

and of various blood guidelines, mercury, and to a

lesser extent lead (from use of lead shot in hunting of

game), may be a significant conce rn among Arctic

peoples. It may be easier to change peoples’ use of

lead shot, as the users of the shot are also the con-

sumers of the wild game. For mercury, the connection

between those affected in the North and sources of

pollution in the south is not as direct and it may be

more difficult to institute appropriate pollution abate-

ment measures.

There is mounting evidence that a threshold may

exist for radiation-induced cancer and this threshold

may be equal to or greater than 100 mSv. The issue is

of significance to northerners, who were exposed in

the past to nuclear weapons fallout, or who have

slightly higher exposures due to caribou meat con-

sumption. Even if one assumes that the linear no-

threshold hypothesis is correct, the risk of continued

consumption of caribou meat is very small. A max-

imum radiation dose rate at presen t is estimated to be

about 3 to 4 mSv/ year. The only way to reduce these

radiation doses further would be to restrict caribou

meat consumption, which woul d deny northerners the

nutritional, social and cultural benefits of this impor-

tant traditional/country food.

The nutritional benefits of traditional/country food

and its contribution to the total diet are substantial,

although only 6–40% of total dietary energy may be

from this food source. Research findings are consis-

tent across the Canadian Arctic and confirm that

decreasing traditional/country food in the diet is likely

to have negative health consequences, in part through

the corresponding increase in tota l fat, saturated fat

and sucros e above recommended levels. Traditional/

country food also contributes significantly more pro-

tein, iron and zinc to the diets of Inuit children than

imported foods.

Effective communication is fundamental to the

aims of the NCP. Communication efforts incorporate

Aboriginal knowledge on the nutritional benefits and

risks associated with the consumption of traditional/

country foods and of imported foods, and on the

importance of a traditional lifestyle to overall health

and well-being. These efforts provide northerners with

the appropriate information to make more informed

decisions about harvesting and consumption of both

traditional/country and imported foods, and also to

avoid some of the mistakes of earlier communication

efforts related to environmental contaminants.

No one method of communication is best, but what

is needed is a variety of methods and materials appro-

priate to communities in different regions and specific

to groups within each community, all aimed at p rovid-

ing opportunities for two-way exchanges of informa-

tion and allowing an understanding of the issue to

develop. This builds a collective comprehension of

benefits and risks, as well as the limitati ons of the

scientific assessment.

The practice of risk communication has changed

over the past two decades, building upon lessons

learned along the way. In place of one-way models of

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246234

message dissemination, NCP-II strove toward shared

decision-making through partnerships involving Ter-

ritorial Health Departments and regional contami-

nants/health committees, community representatives

and representatives of Aboriginal organizations at

all stages of the risk management and risk commu-

nication process.

7. Knowledge gaps

7.1. Exposure assessment

Continued monitoring and assessment of the fre-

quency of consumption of traditional/country foods

in a limited number Dene and Me´tis, First Nations,

and Inuit communities is needed to identify any

significant changes in consumption patterns. Focus

needs to be placed on communities with the highest

exposure but must include an assessment of regional

variations.

The monitoring of organochlorines in human tis-

sues does not have a long history in the Canadian

Arctic, and continued monitoring of tissue levels is

needed to determine contaminant trends. Special

attention should be given to oxychlordane, trans-

nonachlor, HCB, mirex, toxaphene, PCBs, and mer-

cury, especially among the Inuit of Nunavut and

Nunavik. There may be systematic differences

among subgroups in these regions, and more data

are needed.

Although there is a longer history of Arctic envir-

onmental monitoring of mercury, than of organo-

chlorines differences in sampling strategies have

made any conclusions on trends very tenuous. Reg-

ular monitoring of tissue levels among communities

with a range of exposures for mercury and selenium

is advisable.

Ongoing concern about the effects of naturally

occurring polonium-210 in northern food chains

has led to a research project looking at the evi-

dence of damage from this radionuclide in cul-

tured human and animal cells using advanced

techniques in molecular biology. This work needs

to be completed, as the detection of these effects

could serve as an early warning, long before any

overt health effects occur or alleviate unfounded

concerns.

Focussed monitoring of traditional/country food

consumption in specific Arctic communities through

surveys of dietary intakes of environmental contami-

nants, particularly of organochlorines (e.g., PCBs,

chlordane, toxaphene) will help to provide more

spatial and temporal data on changes in dietary

patterns or contaminant concentrations in food. Com-

munities with high traditional/country food consump-

tion and a range of contaminant exposures need to be

included.

Surveys of dietary intake in one Inuit community

have shown that mercury exposures were similar in

1987–1988 and 1999, thus implying little change in

dietary pattern or mercury concent ration in food.

However, monitoring in a range of communities

would provide additional data points on which to

base firmer conclusions.

7.2. Toxicology

More research is needed on the bioaccumul ation

and/or metabolism of toxaphene congeners by aquatic

and terrestrial organisms, including human, in order to

ascertain the number of congeners present in

btoxapheneQ and in the environment; to clarify their

specific toxicity (e.g., immunological effects, infant

size, mutagenicity and genotoxicity, and endocrine

effects); and the appropriate TDI for these congeners.

The most relevant toxi cological issues to be

addressed in future research on chlordane contami-

nants are the relationship between tissue levels in

humans and in experimental animals, and the occur-

rence of specific functional changes, including but not

limited to immune suppression. Since trans-nonachlor

and oxychlordane appear to be present at higher levels

in marine mammal tissues than are other chlordane-

related contaminants, it would be useful if future

human exposure assessments focussed on trans-non-

achlor and oxychlordane tissue levels.

Additional experimental animal and in vitro studies

on the toxicological effects of complex organochlor-

ine mixtures should also include assessment of the

interactions between substances that may result in

antagonism, additivity, or synergism and identification

of compounds responsible for toxic effects. Particular

attention needs to be paid to in utero and lactational

exposure and developmental, reproductive effects,

and immunological effects.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 235

It is clear that there is a need for more studies

designed specifically to address the role of nutrition

in the metab olism and detoxification of methylmer-

cury. It is also important to collect more detailed

dietary information in future epidemiological studies

of methylmercury exposure. A controlled human

study on effects of various nutrients such as omega-

6 fatty acids, selenium and vitamin E on methylmer-

cury toxicokinetics will be useful to confi rm the

results obtained from the animal experiments.

7.3. Epidemiology

The Nunavik cohort study should allow differentia-

tion of the specific deficits attributable to PCBs and

those associated with mercury, since the cord blood

PCB-mercury intercorrelation is low. With the estab-

lishment of a prospective Arctic birth-cohort, there will

be opportunities to look at other contaminants which

may have neurodevelopmental effects, as well as to

study long-term effects that can only be documented at

school age or later in the course of development.

The only recent epidemiological study of contami-

nant effects conducted in the Canadian Arctic is the

one in Nunavik. This study suggested an association

between in utero exposure to POPs and susceptibility

to infections during the first year of life. Although

efforts have been made to also use effect biomarkers

(vitamin A and cytokines), as noted earlier, the low

predictivity of these markers limits their use. The

usefulness of new markers (antibodies post-immuni-

zation and complement system) needs to be assessed

within the cohort study in Nunavik and elsewhere in

the Arctic.

Epidemiological studies on stochastic diseases

(e.g., breast cancer and other hormonal cancers) are

extremely difficult to conduct in the Arctic because

the number of cases is low. This is due to 1) the size of

the population and 2) the fact that many chronic

diseases (cancer) have a low incidence. To assess

these health endpoints, long-term circumpolar studies

need to be used to increase the sample size and

increase the likelihood of detection of any real effects.

Epidemiological studies in the Canadian Arctic should

be restricted to diseases or conditions that: 1) have

high-incidence rates; 2) have gradients of severity

(from normality to overt abnormalities); and 3) are

easy to diagnose.

7.4. Risk and benefit characterization, assessment and

advice

Focussed monitoring and assessment of organo-

chlorine intake, especially by high-end consumers of

traditional/country foods in specific Inuit commu-

nities, are strongly warranted. The risks associated

with usual levels of exposure to organic mercury via

consumption of contaminated traditional/country

foods also need to be characterized since they too

appear prevalent in the Baffin and Nunavik regions.

The substantial nutritional benefits of traditional/

country food and its contribution to the total diet have

been documented; however, further research is needed

on the negative health consequences of not consuming

traditional/country food.

Few projects have been conducted under the NCP

to specifical ly investigate public perceptions of con-

taminants in traditional/country foods and related

risks, understandings and misunderstandings of the

issue, and the potential impacts they may have on

individual behaviour and activities. There is a conti-

nuing need to document and understand public per-

ceptions of contaminants and how the issue is being

addressed in the North.

There is a need to further clarify messages for

Aboriginal people s, taking into account different com-

munity-of-origin dialects that can significantly affect

the reading comprehension of printed material, and

improving the effectiveness of some of the methods

used to convey environmental health information.

Research is also needed on ways that women of child-

bearing age use information to make informed deci-

sions regarding consumption of PCB- and mercury-

contaminated traditional/country food items.

Acknowledgements

The list of people who have been involved in the

environmental contaminant, dietary and health effects

research studies discussed in this paper is extremely

long, from researchers to contaminants coordinators

in each region, to community health workers, to

dietitians, to nurses, to mothers and community resi-

dents. We are very grateful to them. Without their

efforts and contributions of new and ground breaking

research and cooperation we would have very little to

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246236

present in this paper. We would also like to thank F.

Dallaire, M. Mirault, P. Julien, J. Bailey, S. Bernier,

H. Bilrha, S. Bruneau, S. Jordan, D. Pereg, J. Butler

Walker and R. Roy who worked closely with the

authors of this paper. Special thanks to P. Bjerre-

gaard, D. Villeneuve and G. Egeland who reviewed

an earlier version of this paper.

References

AMAP. AMAP assessment report: arctic pollution issues. Oslo,

Norway7 Arctic Monitoring and Assessment Programme;

1998. 859 pp.

Andersen HR, Andersen O. Effects of dietary alpha-tocopherol

and beta-carotene on lipid peroxidation induced by methyl

mercuric chloride in mice. Pharmacol Toxicol 1993;73(4):

192 – 201.

Arnold D. Personal communication, Health Canada, Ottawa,

Ontario, Canada; 2003.

Arnold DL, Bryce F, Baccanale C, Hayward S, Tanner JR, MacLel-

lan E, et al. Toxicological consequences of toxaphene ingestion

by cynomologus (Macaca fascicularis) monkeys: Part 1 Pre-

mating phase. Food Chem Toxicol 2001;39:467 – 76.

Auger J, Kunstmann JM, Czyglik F, Jouannet JP. Decline in semen

quality among fertile men in Paris during the past 20 years. N

Engl J Med 1995;332:281 – 5.

Ayotte P. Personal communication. Laval University, Ste-Foy, Que´-

bec. Canada; 2001.

Ayotte P. Personal communication. Laval University, Beauport,

Que´bec, Canada; 2002.

Batal M. Sociocultural determinants of traditional food intake across

indigenous communities in the Yukon and Denendeh. PhD

thesis. Dietetics and Human Nutrition, and CINE (Centre for

Indigenous Peoples’ Nutrition and Environment), McGill Uni-

versity, Montreal, Que´bec, Canada; 2001.

Beard J, Marshall S, Jong K, Newton R, Triplett-McBride T,

Humphries B, et al. 1,1,1-trichloro-2,2-bis ( p-chlorophenyl)-

ethane (DDT) and reduced bone mineral density. Arch Environ

Health 2000;55(3):177 – 80.

Be´langer MC, Julien P, Berthiaume L, Noe¨l M, Mirault ME,

Dewailly E. Mercury, fish consumption and the risk of cardio-

vascular diseases in Inuit of Nunavik. Can J Cardiol

2003;19(Suppl. A):133 – 4.

Bemis JC, Seegal RF. Polychlorinated biphenyls and methylmer-

cury act synergistically to reduce rat brain dopamine content in

vitro. Environ Health Perspect 1999;107(11):879 – 85.

Benedetti JL, Dewailly E

, Turcotte F, Lefebvre M. Unusually high

blood cadmium associated with cigarette smoking among three

subgroups of the general population, Que´bec, Canada. Sci Total

Environ 1994;152:161 – 7.

Berkes F, Farkas CS. Eastern James Bay Cree Indians: changing

patterns of wild food use and nutrition. Ecol Food Nutr

1978;7:155– 72.

Berti PR, Receveur O, Chan HM, Kuhnlein HV. Dietary exposure to

chemical contaminants from traditional food among adult Dene/

Me´tis in the western Northwest Territories, Canada. Environ Res

1998;76(2):131– 42.

Berti PR, Chan HM, Receveur O, MacDonald CR, Kuhnlein HV.

Population exposure to radioactivity from consumption of car-

ibou among the Dene/Me´tis of Denendeh (western Northwest

Territories, Canada). J Expo Anal Environ Epidemiol

1998;8(2):145– 58.

Berti PR, Hamilton SE, Receveur O, Kuhnlein HV. Food use and

nutrient adequacy in Baffin Inuit children and adolescents. J

Diet Pract Res 1999;60(2):63 – 70.

Besselink HT, Nixon E, McHugh B, KlungsUyr J, Brouwer A. In

vitro and in vivo tumor promoting potency of technical toxa-

phene, UV-irradiated toxaphene and biotransformed toxaphene.

Organohalog Compd 2000;47:113 – 6.

Bilrha H, Roy R, Wagner E, Belles-Isles M, Bailey JL, Ayotte P.

Effects of gestational and lactational exposure to organochlorine

compounds on cellular, humoral, and innate immunity in swine.

Toxicol Sci 2004;77:41 – 50.

Bitman J, Cecil HC. Estrogenic activity of DDT analogs and

polychlorinated biphenyls. J Agric Food Chem 1970;18(6):

1108– 12.

Bjerregaard P, Hansen JC. Organochlorines and heavy metals in

pregnant women from the Disko Bay area in Greenland. Sci

Total Environ 2000;245(1–3):195– 202.

Bjerregaard P, Young TK. The circumpolar inuit: health of a popu-

lation in transition. Copenhagen7 Munksgaard; 1998.

Blanchet C, Dewailly E

, Ayotte P, Bruneau S, Receveur O, Holub

BJ. Contribution of selected traditional and market foods to the

diet of Nunavik Inuit women. Can J Diet Pract Res 2000;

61(2):50 – 9.

Bloem M, Wedel M, Egger RJ, Speek AJ, Schrijver J, Saowa-

kontha S, et al. Mild vitamin A deficiency and risk of

respiratory tract diseases and diarrhea in preschool children

in Northeastern Thailand. Am J Epidemiol 1990;131(2):

3320 – 39.

Bondy GS, Newsome WH, Armstrong CL, Suzuki CAM, Doucet J,

Fernie S, et al. trans-Nonachlor and cis-nonachlor toxicity in

Sprague–Dawley rats: comparison with technical chlordane.

Toxicol Sci 2000;58:386 – 98.

Bondy G, Armstrong C, Coady L, Doucet J, Robertson P, Feeley M,

et al. Toxicity of the chlordane metabolite oxychlordane in

female rats: clinical and histopathological changes. Food

Chem Toxicol 2003;41(2):291 – 301.

Borre´ KS. The healing power of the seal: the meaning of Inuit

health practice and belief. Arct Anthropol 1994;31(1):1– 15.

Bowers WJ. Neurobehavioural effects of in utero and lactational

exposure to a chemical mixture based on blood profiles in

Canadian Arctic populations. Canadian Arctic Contaminants

Assessment Symposium. Oral Presentation. Ottawa, Ontario,

Canada; 2003.

Brouwer A. Role of biotransformation in PCB-induced alterations

in vitamin A and thyroid hormone metabolism in laboratory and

wildlife species. Biochem Soc Trans 1991;19:731 – 7.

Bruneau S, Furgal C, Dewailly E

, Grey M. Incorporation of scien-

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 237

tific knowledge into Inuit knowledge in Nunavik. In: Kalhok S,

editor. Synopsis of research conducted under the 2000–2001

Northern Contaminants Program. Ottawa, Ontario, Canada7

Department of Indian Affairs and Northern Development;

2001. p. 291 – 3.

Buchert H, Class T, Ballschmiter K. High resolution gas chroma-

tography of technical chlordane with electron capture- and

mass selective detection. Fresenius Z Anal Chem 1989;

333(3):211 – 7.

Butler Walker J, Seddon L, McMullen E, Houseman J, Tofflemire

K, Corriveau A, et al. Organochlorine levels in maternal and

umbilical cord blood plasma in Arctic Canada. Sci Total

Environ 2003;302(1-3):27 – 52.

Butler Walker J, Houseman J, Seddon L, Tofflemire K, Mills C,

Corriveau A, et al. Maternal and umbilical cord levels of

mercury, lead, cadmium and essential trace elements in Arctic

Canada. Environ Res 2005 [In press].

CACAR, 1997. In: Jensen J, Adare L, Shearer R, editors. Canadian

Arctic Contaminants Assessment Report. Ottawa, Ontario,

Canada7 Department of Indian Affairs and Northern Develop-

ment; 1997.

Campagna C, Sirard MA, Ayotte P, Bailey JL. Impaired maturation,

fertilization, and embryonic development of porcine oocytes

following exposure to an environmentally relevant organochlor-

ine mixture. Biol Reprod 2001;65(2):554 – 60.

Campagna C, Guillemette C, Paradis R, Sirard MA, Ayotte P, Bailey

JL. An environmentally relevant organochlorine mixture impairs

sperm function and embryo development in the porcine model.

Biol Reprod 2002;67(1):80 – 7.

Carlsen E, Giwercman A, Keiding N, Skakkebaek NE. Evidence for

decreasing quality of semen during past 50 years. Br Med J

1992;305:609–13.

Chan HM. Dietary effects on methylmercury toxicity in rats. In:

Kalhok S, editor. Synopsis of research conducted under the

1999–2000 Northern Contaminants Program. Ottawa, Ontario,

Canada7 Department of Indian Affairs and Northern Develop-

ment; 2000. p. 11– 5.

Chan HM. Toxicology of mercury and selenium in ringed seal

tissues. In: Kalhok S, editor. Synopsis of research conducted

under the 1999–2000 Northern Contaminants Program. Ottawa,

Ontario, Canada7 Department of Indian Affairs and Northern

Development; 2000. p. 16 – 9.

Chan HM. Personal communication. Centre for Indigenous Peoples’

Nutrition and Environment (CINE), Ste-Anne-de-Bellevue,

Que´bec; 2002.

Chan HM, Kim C, Khoday K, Receveur O, Kuhnlein HV.

Assessment of dietary exposure to trace metals in Baffin

Inuit food samples. Environ Health Perspect 1995;103(7–8):

740– 6.

Chan HM, Berti PR, Receveur O, Kuhnlein HV. Evaluation of the

population distribution of dietary contaminant intakes in an

Arctic population using Monte Carlo statistics. Environ Health

Perspect 1997;105(3):316 – 21.

Chan HM, Krishnan K, Iverson F, Suzuki C. Use of physiologically

based toxicokinetic (PBTK) model for risk assessment of expo-

sure to mixtures of organochlorines in traditional food. In:

Kalhok S, editor. Synopsis of research conducted under the

1999–2000 Northern Contaminants Program. Ottawa, Ontario,

Canada7 Department of Indian Affairs and Northern Develop-

ment; 2000. p. 20 – 7.

Chan HM, Mehta R, Lok E, Gill S, Jin D. Effects of selenium,

vitamin E and phytate on methylmercury toxicity in male Spra-

gue–Dawley rats. In: Smith S, Stow J, Carrillo F, editors.

Synopsis of research conducted under the 2003–2004 Northern

Contaminants Program. Ottawa, Ontario, Canada7 Department

of Indian Affairs and Northern Development; 2004. p. 29 – 32.

Chan HM, Receveur O, Kuhnlein HV. Estimation of dietary expo-

sure to organochlorines in Inuit communities in Northern

Canada. Environ Health Perspect in preparation(a).

Chan HM, Receveur O, Kuhnlein HV. Dietary mercury intakes in

Inuit communities in Northern Canada. Sci Total Environ in

preparation(b).

Chang LW, Gilbert M, Sprecher J. Modification of methylmercury

neurotoxicity by vitamin E. Environ Res 1978;17(3):356 – 66.

Chang K, Hsieh K, Lee T, Tang S, Tung T. Immunologic evaluation

of patients with polychlorinated biphenyl poisoning: determina-

tion of lymphocyte subpopulations. Toxicol Appl Pharmacol

1981;61:58– 63.

Chanoine JP, Braverman LE. The role of transthyretin in the trans-

port of thyroid hormone to cerebrospinal fluid and brain. Acta

Med Austriaca 1992;19(Suppl. 1):25 – 8.

Chao W, Hsu C, Guo YL. Middle-ear disease in children exposed

prenatally to polychlorinated biphenyls and polychlorinated

dibenzofurans. Arch Environ Health 1997;52(4):257–62.

Chauhan KR, Kodavanti PRS, McKinney JD. Assessing the role of

ortho-substitution on polychlorinated biphenyl binding to trans-

thyretin, a thyroxine transport protein. Toxicol Appl Pharmacol

2000;162:10– 21.

Cheek AO, Kow K, Chen J, McLachlan JA. Potential mechan-

isms of thyroid disruption in humans: interaction of organo-

chlorine compounds with thyroid receptor, transthyretin, and

thyroid-binding globulin. Environ Health Perspect 1999;

107(4):273 – 8.

Choi BH. The effects of methylmercury on the developing brain.

Prog Neurobiol 1989;32:447 – 70.

Clarkson TW. Mercury: major issues in environmental health.

Environ Health Perspect 1993;100:31 – 8.

Clarkson TW. The toxicology of mercury. Crit Rev Clin Lab Sci

1997;34(3):369– 403.

Condon RG, Collings P, Wenzel G. The best part of life: subsistence

hunting, and economic adaptation among young adult Inuit

males. Arctic 1995;48(1):31 – 46.

Coˆte´ S, Dewailly E, Dodin S, Blanchet C, Ayotte P, Mulvad G, et al.

PCB exposure and risk of osteoporosis in Greenlandic meno-

pausal women. 2005. In preparation.

Crump KS, Kjellstro¨ m T, Shipp AM, Silvers A, Stewart A. Influ-

ence of prenatal mercury exposure upon scholastic and psycho-

logical test performance: benchmark analysis of a New Zealand

cohort. Risk Anal 1998;18(6):701 – 13.

Dallaire F, Dewailly E, Muckle G, Ayotte P. Time trends of

persistent organic pollutants and heavy metals in umbilical

cord blood of Inuit infants born in Nunavik (Quebec, Canada)

between 1994 and 2001. Environ Health Perspect 2003a;

111(13):1660 – 4.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246238

Dallaire F, Dewailly E

, Shademani R, Laliberte´ C, Bruneau S,

Blanchet C, et al. Vitamin A status in umbilical cord blood of

infants from three separate regions of the province of Que´bec,

Canada. Can J Public Health 2003b;94(5):386– 90.

Dallaire F, Dewailly E

, Muckle G, Ve´zina C, Jacobson S,

Jacobson J, Ayotte P. Acute infections and environmental

exposure to organochlorines in Inuit infants from Nunavik.

Environ Health Perspect 2004;112(14):1359– 65.

Dallaire F, Dewailly E, Ve´zina C, Ayotte P, Muckle G, Bruneau S.

Effect of prenatal exposure to organochlorines on incidence of

acute respiratory infections in preschool Inuit children in pre-

paration(b).

Darvill T, Lonky E, Reihman J, Stewart P, Pagano J. Prenatal

exposure to PCBs and infant performance on the Fagan test of

infant intelligence. Neurotoxicology 2000;21(6):1029– 38.

Das UN. Long-chain polyunsaturated fatty acids in the growth and

development of the brain and memory. Nutrition 2003;19(1):

62 – 5.

Daum JR. Immunotoxicology of cadmium and mercury on B-lym-

phocytes: 1 Effects on lymphocyte function. Int J Immunophar-

macol 1993;15(3):383 – 94.

Davidson PW, Myers GJ, Cox C, Shamlaye C, Marsh DO, Tanner

MA, et al. Longitudinal neurodevelopmental study of Seychel-

lois children following in utero exposure to methylmercury

from maternal fish ingestion: outcomes at 19 and 29 months.

Neurotoxicology 1995;16(4):677 – 88.

Davidson PW, Myers GJ, Cox C, Axtell C, Shamlaye C, Sloane-

Reeves J, et al. Effects of prenatal and postnatal methylmercury

exposure from fish consumption on neurodevelopment. JAMA

1998;280(8):701–7.

Deutch B. Personal communication. Centre of Arctic Environmental

Medicine, Aarhus University, Aarhus, Denmark, 2001.

Deutch B, Hansen JC. High human plasma levels of organochlorine

compounds in Greenland Regional differences and lifestyle

effects. Dan Med Bull 2000;47:132– 7.

Dewailly E

, Bruneau S, Laliberte C, Lebel G, Gingras S, Grondin J,

et al. Contaminants. In: Jette´ JA health profile of the Inuit: report

of the Sante Que´bec health survey among the Inuit of Nunavik,

1992. Montre´al7 Montreal, Ministe´re de la Sante´ et des Services

Sociaux, Governement du Que´bec; 1994. p. 73 – 107.

Dewailly E, Hansen JC, Pedersen HS, Mulvad G, Ayotte P, Weber

JP, et al. Concentration of PCBs in various tissues from autop-

sies in Greenland. Organohalog Compd 1995;26:175 – 80.

Dewailly E

, Bruneau S, Ayotte P, Lebel G. Evaluation of prenatal

exposure to organochlorines and heavy metals in Nunavik new-

borns 1993–1996. Environ Stud 1998;74:293 – 312.

Dewailly E

, Bruneau S, Ayotte P. Mercury in Salluit: temporal trend

and interaction with selenium. In: Kalhok S, editor. Synopsis of

research conducted under the 1998–1999 Northern Contami-

nants Program. Ottawa, Ontario, Canada7 Department of Indian

Affairs and Northern Development; 1999. p. 265 – 71.

Dewailly E

, Ayotte P, Bruneau S, Schwartz H, Grey M, Weber J, et

al. Mercury in Salluit: temporal trend and interaction with

selenium. In: Kalhok S, editor. Synopsis of research conducted

under the 2003–2004 Northern Contaminants Program. Ottawa,

Ontario, Canada7 Department of Indian Affairs and Northern

Development; 2000a. p. 32 – 6.

Dewailly E

, Levesque B, Duchesne J, Dumas P, Scheuhammer A,

Gariepy C, et al. Lead shot as a source of lead poisoning in the

Canadian Arctic. Epidemiology 2000b;11(4):S146.

Dewailly E

, Ayotte P, Bruneau S, Gingras S, Belles-Isles M, Roy R.

Susceptibility to infections and immune status in Inuit infants

exposed to organochlorines. Environ Health Perspect

2000c;108(3):205–11.

Dewailly E

, Blanchet C, Chaumette P, Receveur O, Lawn J, Nob-

mann E, et al. Diet profile of circumpolar Inuit. Que´bec,

Canada7 GE

TIC, Universite´ Laval; 2000d.

Dewailly E

, Blanchet C, Lemiuex S, Sauve´ L, Gingras S, Ayotte P,

et al. -3 fatty acids and cardiovascular disease risk factors

among the Inuit of Nunavik. Am J Clin Nutr 2001a;

4(4):464 – 73.

Dewailly E

, Blanchet C, Gingras S, Furgal C, Bruneau S. Socio-

demographic factors influencing nutrition and contaminant

exposure in Nunavik. In: Kalhok S, editor. Synopsis of research

conducted under the 2000–2001 Northern Contaminants Pro-

gram. Ottawa, Ontario, Canada7 Department of Indian Affairs

and Northern Development; 2001b. p. 26 – 35.

Dewailly E

, Beuter A, Ayotte P, Bruneau S, Mirault ME

, Grey M, et

al. Mercury in Salluit: effects of mercury on oxidative status and

sensorimotor functions. In: Kalhok S, editor. Synopsis of

research conducted under the 2000–2001 Northern Contami-

nants Program. Ottawa, Ontario, Canada7 Department of Indian

Affairs and Northern Development; 2001c. p. 22 – 5.

Doll R. The lessons of life: keynote address to the nutrition and

cancer conference. Cancer Res 1992;52(7 Suppl.):2024s – 9s.

Dougherty CP, Holtz SH, Reinert JC, Panyacosit L, Axelrad DA,

Woodruff TJ. Dietary exposures to food contaminants across the

United States. Environ Res A 2000;84:170 – 85.

Douglas M. Risk acceptability according to the social sciences. New

York, USA7 Basic Books; 1986.

Douglas M. Risk and blame—essays in cultural theory. London,

England7 Routledge; 1992.

Dumont C, Girard M, Bellavance F, Noel F. Mercury levels in the

Cree population of James Bay, Que´bec, from 1988 to 1993/94.

Can Med Assoc J 1998;158(11):1439– 45.

Eaton RD, Secord DC, Hewitt P. An experimental assessment of the

toxic potential of mercury in ringed seal liver for adult labora-

tory cats. Toxicol Appl Pharmacol 1980;55(3):514 – 21.

Egede I. Inuit food and Inuit health: contaminants in perspective.

Presentation to inuit circumpolar conference, seventh general

assembly, Nome, Alaska, July, 1995; 1995.

Egeland GM, Feyk LA, Middaugh JP. The use of traditional foods

in a healthy diet in Alaska: risks in perspective. Anchorage,

Alaska7 Department of Health and Social Services, Alaska

Division of Public Health; 1998. 140 pp.

Ekblond A, Mellemkjaer A, Tjonneland L, Suntum M, Stripp C,

Overvad K, et al. A cross-sectional study of dietary habits and

urinary glucose excretion: a predictor of non-insulindependent

diabetes mellitus. Eur J Clin Nutr 2000;54:343 – 9.

Elias B, O’Neil J. A study into the social, cultural and disci-

plinary understanding of risk perception and risk acceptability

of contaminants in the Canadian Arctic. Report prepared for

the Arctic Environmental Strategy Contaminants Program.

Northern Health Research Unit, Department of Community

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 239

Health Sciences, University of Manitoba, Manitoba, Canada;

1995.

Elinder CG. Cadmium as an environmental hazard. IARC Sci Publ

1992;118:123 – 32.

Erikson K. Toxic reckoning: business faces a new kind of fear.

Harvard Bus Rev 1990;90(1):118– 26.

Farina M, Soares FA, Feoli A, Roehring C, Brusque AM, Rotta L,

et al. In vitro effects of selenite and mercuric chloride on liver

thiobarbituric acid–reactive substances (TBARS) and non-pro-

tein thiols from rats: influences of dietary cholesterol and

polyunsaturated and saturated fatty acids. Nutrition 2003;19:

531– 5.

Fediuk K, Hidiroglou N, MadCre R, Kuhnlein HV. Vitamin C in

Inuit traditional food and women’s diets. J Food Compos Anal

2002;15:221– 35.

Fein GG, Jacobson JL, Jacobson SW, Schwartz PM, Dowler JK.

Prenatal exposure to polychlorinated biphenyls: effects on birth

size and gestational age. J Pediatr 1984;105:315 – 20.

Feskens EJM, Bowles CH, Kromhout D. Inverse association

between fish intake and risk of glucose intolerance in normo-

glycemic elderly men and women. Diabetes Care 1991;14(11):

935– 41.

Feskens EJM, Virtanen SM, Ra¨sa¨nen L, Tuomilehto J, Stengard J,

Pekkanen J, et al. Dietary factors determining diabetes and

impaired glucose tolerance. Diabetes Care 1995;18(8):1104–12.

Flynn J, Slovic P, Mertz CK. Gender, race, and perception of

environmental health risks. Risk Anal 1994;14(6):1101–8.

Freeman MMR, Bogoslovskaya L, Caulfield RA, Egede I, Krupnik

MG, Stevenson MG, editors. Inuit, whaling, and sustainability.

Walnut Creek, CA7 AltaMira Press; 1998 208 pp.

Fry DM, Toone CK. DDT-induced feminization of gull embryos.

Science 1981;213:922 – 4.

Furgal CM. Addressing Northern decision making capacity: the

case of health advisories and the Labrador Inuit. PhD thesis.

Faculty of Environmental Studies, School of Planning, Univer-

sity of Waterloo, Waterloo, Ontario, Canada; 1999.

Furgal CM, Bernier S, Godin G, Gingras S, Grondin J, Dewailly E

Decision making and diet in the North: Balancing the physical,

economic and social components. In: Kalhok S, editor. Synopsis

of research conducted under the 2000–2001 Northern Contami-

nants Program. Ottawa, Ontario, Canada7 Department of Indian

Affairs and Northern Development; 2001. p. 42 – 3.

Furgal C, Kalhok S, Loring E, Smith S. Knowledge in action:

northern contaminants program structures, processes and pro-

ducts. Canadian arctic contaminants assestment report II.

Ottawa, Ontario, Canada7 Indian and Northern Affairs Canada;

2003 90 pp.

Furgal CM, Powell S, Myers H. Digesting the message about

contaminants in the Canadian North: a review and recommenda-

tions for future research and action. Arctic 2005;58(2):103–14.

Ganther HE. Modification of methylmercury toxicity and metabo-

lism by selenium and vitamin E: possible mechanisms. Environ

Health Perspect 1978;25:71 – 6.

Gladen BC, Rogan WJ, Hardy P, Thullen J, Tingelstad J, Tully M.

Development after exposure to polychlorinated biphenyls and

dichlorodiphenyl dichloroethene transplacentally and through

human milk. J Pediatr 1988;113(9):991– 5.

Gouteux B, Lebeuf M, Trottier S, Gagne JP. Analysis of six relevant

toxaphene congeners in biological samples using ion trap MS/

MS. Chemosphere 2002;49(2):183 – 91.

Grandjean P, Weihe P, Jorgensen PJ, Clarkson T, Cernichiari E,

Videro T. Impact of maternal seafood diet on fetal exposure to

mercury, selenium, and lead. Arch Environ Health 1992;47(3):

185 – 95.

Grandjean P, Weihe P, White RF, Debes F, Araki S, Yokoyama K,

et al. Cognitive deficit in 7-year old children with prenatal

exposure to methylmercury. Neurotoxicol Teratol 1997;19:

417 – 28.

Grandjean P, Budtz-Jorgensen E, White RF, Jorgensen PJ, Weihe P,

Debes F, et al. Methylmercury exposure biomarkers as indica-

tors of neurotoxicity in children aged 7 years. Am J Epidemiol

1999;150(3):301–5.

Groten JP, Feron VJ, Suhnel J. Toxicology of simple and complex

mixtures. Trends Pharmacol Sci 2001;22(6):316 – 22.

Haddad S, Charest-Tardif G, Tardif R, Krishnan K. Validation of a

physiological modeling framework for simulating the toxicoki-

netics of chemicals in mixtures. Toxicol Appl Pharmacol

2000;167(3):199–209.

Hamburgh M. The role of thyroid and growth hormones in neuro-

genesis. In: Moscona A, Monroy A, editors. Current topics in

developmental biology. New York, USA7 Academic Press;

1969. p. 109 – 48.

Health Canada. Methylmercury in Canada: exposure of First

Nations and Inuit residents to methylmercury in the Canadian

environment. Ottawa, Ontario, Canada7 Medical Services

Branch, Department of National Health and Welfare; 1979.

200 pp.

Health Canada. Health risk determination: the challenges of health

protection. Ottawa, Canada7 Minister of Supply and Services

Canada. Cat. No. H49-40/1993E; 1993.

Health Canada. Human health risk assessment for priority sub-

stances. Ottawa, Canada7 Canada Communication Group;

1994. 36 pp.

Health Canada. Food directorate memo on re-evaluation of mercury

TDI. Ottawa, Ontario, Canada7 Health Canada Report; 1998.

Health Canada. Methylmercury in Canada Exposure of First

Nations and Inuit residents to methylmercury in the Canadian

environment. Ottawa, Ontario, Canada7 Medical Services

Branch, Health Canada; 1999. 150 pp.

Health Canada. Health Canada decision-making framework for

identifying, assessing, and managing health risks. Ottawa,

Ontario, Canada7 Minister of Public Works and Government

Services Canada; 2000. 75 pp.

Health Physics Society Newsletter, 1996. Volume, X.X.I.V., Num-

ber 3, March, p. 3.

Heo Y, Parsons PL, Lawrence DA. Lead differentially modifies

cytokine production in vitro and in vivo. Toxicol Appl Pharma-

col 1996;138:149 – 57.

Hewitt K. The idea of calamity in a technocratic age. In: Hewitt

K, editor. Interpretations of calamity from the viewpoint of

human ecology. Boston, MA, USA7 Allen and Unwin; 1983.

p. 3 – 32.

Huisman M, Koopman-Esseboom C, Fidler V, Hadders-Algra M,

Van der Paauw CG, Tuinstra LGMT, et al. Perinatal exposure

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246240

to polychlorinated biphenyls and dioxins and its effect on

neonatal neurological development. Early Hum Dev 1995;41:

111 – 27.

Huisman M, Koopman-Esseboom C, Lanting CI, Van der Paauw

CG, Tuinstra LGMT, Fidler V, et al. Neurological condition in

18-month-old children perinatally exposed to polychlorinated

biphenyls and dioxins. Early Hum Dev 1995;43:165 – 76.

ICRP (International Commission on Radiological Protection). 1990

Recommendations of the international commission on radiolo-

gical protection. ICRP Publication. Oxford, England7 Pergamon

Press; 1990.

Institute of Medicine (IOM). Dietary reference intakes for vitamin,

C, vitamin, E, selenium, and carotenoids. Washington, DC7

National Academy Press; 2000.

Inuit Circumpolar Conference. The arctic sealing industry: a retro-

spective analysis of its collapse and options for sustainable

development: Part I. Retrospective analysis. Arctic Environmen-

tal Protection Strategy; 1996, 83 pp.

Ivanov VK, Gorski AI, Maksioutov MA, Tsyb AF, Souchkevitch

GN. Mortality among the Chernobyl emergency workers: esti-

mation of radiation risks (preliminary analysis). Health Phys

2001;81:514– 21.

Jacobson JL, Jacobson SW. Intellectual impairment in children

exposed to polychlorinated biphenyls in utero. N Engl J Med

1996;335(11):783 – 9.

Jacobson JL, Jacobson SW. Prenatal exposure to polychlorinated

biphenyls and attention at school age. Obstet Gynecol Surv

2004;59(6):412–3.

Jacobson SW, Fein GG, Jacobson JL, Schwartz PM, Dowler JK.

The effects of intrauterine PCB exposure in visual recognition

memory. Child Dev 1985;56:853 – 60.

Jacobson JL, Jacobson SW, Humphrey HEB. Effects of in utero

exposure to polychlorinated biphenyls and related contaminants

on cognitive functioning in young children. J Pediatr 1990;

116:38– 45.

Jacobson JL, Jacobson SW, Humphrey HEB. Effects of exposure to

PCBs and related compounds on growth and activity in children.

Neurotoxicol Teratol 1990;12(4):319 – 26.

Jacobson JL, Jacobson SW, Padgett RJ, Brumitt GA, Billings RL.

Effects of prenatal PCB exposure on cognitive processing effi-

ciency and sustained attention. Dev Psychol 1992;28:297–306.

Janssen S, Solomon G, Schettler T, 2004. Chemical Contaminants

and Human Disease: A Summary of Evidence. (http://www.

protectingourhealth.org/corethemes/links/2004-0203spreadsheet.

htm).

Jobling S, Reynolds T, White R, Parker MG, Sumpter JP. A variety

of environmentally persistent chemicals, including some phtha-

late plasticizers, are weakly estrogenic. Environ Health Perspect

1995;103:582–7.

Kaikkonen J, Nyyssonen K, Tuomainen T, Ristonmaa U, Salonen

JT. Determinants of plasma coenzyme Q10 in humans. FEBS

Lett 1999;443:163 – 6.

Kalmijn S, van Boxtel MPJ, Ocke´ M, Verschuren WMM, Kromh-

out D, Launer LJ. Dietary intake of fatty acids and fish in

relation to cognitive performance at middle age. Neurology

2002;62:275– 80.

Kasperson RE. Six propositions on public participation and

their relevance for risk communication. Risk Anal 1986;

6(3):275 – 81.

Kelce WR, Stone CR, Laws SC, Gray LE, Kemppainen JA, Wilson

EM. Persistent DDT metabolite p,p-DDE is a potent androgen

receptor agonist. Nature 1995;375:581 – 5.

Kilburn KH, Thornton JC. Protracted neurotoxicity from chlordane

sprayed to kill termites. Environ Health Perspect 1995;103(7–8):

690 – 4.

Kim C, Chan HM, Receveur O. Risk assessment of cadmium

exposure in Fort Resolution, Northwest Territories, Canada.

Food Addit Contam 1998;15(3):307 – 17.

Kinloch D, Kunhlein H, Muir DCG. Inuit foods and diet: a pre-

liminary assessment of benefits and risks. Sci Total Environ

1992;122:247–78.

Kjellstro¨ m T, Kennedy P, Wallis S, Mantell C. Physical and

mental development of children with prenatal exposure to

mercury from fish: Stage 1 Preliminary Test at Age 4.

Solna, Sweden7 National Swedish Environmental Protection

Board; 1986.

Klopov VP. Hygienic assessment of environmental pollution in the

north of Russia and its effect on population health. Doctoral

thesis. St. Petersburg; 2000, 46 pp. (in Russian).

Klopov VP, Shepovalnikov VN. The levels of exposure to key POPs

among adult and children populations in Yamal Peninsula,

Arctic Russia. Proceedings of Workshop on Persistent Organic

Pollutants (POPs) in the Arctic: Human Health and Environ-

mental Concerns. Rovaniemi, 18–20 January 2000. University

of the Arctic: Rovaniemi, 2000. Abstract 19; 2000. 4 pp.

Klopov VP, Tchachchine VP. Environmental pollution and risk

assessment for population in the Russian North. St. Petersburg,

Russia7 State Medical Academy; 2001. 225 pp. (in Russian).

Klopov V, Odland JO, Burkow IC. Persistent organic pollutants in

maternal blood plasma and breast milk from Russian arctic

populations. Int J Circumpolar Health 1998;57:4239 – 48.

Koopman-Esseboom C, Huisman M, Weisglas-Kuperus N,

Boersma ER, de Ridder MAJ, Van der Paauw CG, et al.

Dioxin and PCB levels in blood and human milk in relation

to living areas in the Netherlands. Chemosphere 1994;29(9–

11):2327– 38.

Koopman-Esseboom C, Huisman M, Weisglas-Kuperus N, Van der

Paauw CG, Tuinstra LGMT, Boersma ER, et al. PCB and dioxin

levels in plasma and human milk of 418 Dutch women and their

infants Predictive value of PCB congener levels in maternal

plasma for fetal and infant’s exposure to PCBs and dioxins.

Chemosphere 1994;28(9):1721 – 32.

Kraus N, Malmfors T, Slovic P. Intuitive toxicology expert and lay

judgements of chemical risks. Risk Anal 1992;12(2):215 – 31.

Kris-Etherton PM, Shaffer Taylor D, Yu-Poth S, Huth P, Moriarty

K, Fishell V, et al. Polyunsaturated fatty acids in the food chain

in the United States. Am J Clin Nutr 2000;71:179S – 88S

[Suppl.].

Krishnan K, Brodeur J. Toxic interactions among environmental

pollutants: corroborating laboratory observations with human

experience. Environ Health Perspect 1994;102(Suppl. 9):11– 7.

Kromhout D, Feskens EJM, Bowles CH. The protective effect of a

small amount of fish on coronary heart disease mortality in an

elderly population. Int J Epidemiol 1995;24(2):340– 4.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 241

Kuhnlein HV. Nutritional and toxicological components of Inuit

diets in Broughton Island, NWT School of dietetics and human

nutrition. Montreal, Que´bec, Canada7 McGill University; 1989.

100 pp.

Kuhnlein HV. Personal communication. Centre for Indigenous Peo-

ples’ Nutrition and Environment (CINE), Ste-Anne-de-Belle-

vue, Que´bec; 2002.

Kuhnlein HV, Dickson C. Food choice decisions by Western

Arctic Aboriginal women and children, Year 1. In: Kalhok

S, editor. Synopsis of research conducted under the 2000–2001

Northern Contaminants Program. Ottawa, Ontario, Canada7

Department of Indian Affairs and Northern Development;

2001. p. 54– 5.

Kuhnlein HV, Receveur O. Dietary change and traditional food

systems of indigenous peoples. Annu Rev Nutr 1996;16:

417– 42.

Kuhnlein HV, Receveur O. Personal communication. Centre for

Indigenous Peoples’ Nutrition and Environment (CINE), Ste-

Anne-de-Bellevue, Que´bec; 2001.

Kuhnlein HV, Soueida R, Receveur O. Baffin Inuit food use by age,

gender and season. J Can Diet Assoc 1995;6(4):175 – 83.

Kuhnlein HV, Receveur O, Muir DCG, Chan HM, Soueida R.

Arctic indigenous women consume greater than acceptable

levels of organochlorines. J Nutr 1995;125(10):2501 – 10.

Kuhnlein HV, Receveur O, Chan HM, Loring E. Assessment of

dietary benefit/risk in Inuit communities. Ste-Anne-de-Bellevue,

Que´bec7 Centre for Indigenous Peoples’ Nutrition and Environ-

ment (CINE); 2000.

Kuhnlein HV, Receveur O, Chan HM. Nutrient benefits of

Arctic traditional/country food. In: Kalhok S, editor. Synop-

sis of research conducted under the 2000–2001 Northern

Contaminants Program. Ottawa, Ontario, Canada7 Depart-

ment of Indian Affairs and Northern Development; 2001.

p. 56–64.

Kuhnlein HV, Receveur O, Chan HM. Traditional food systems

research with Canadian indigenous peoples. Int J Circumpolar

Health 2001;60(2):112– 22.

Kuhnlein HV, Receveur O, Ing A. Energy, fat and calcium in

bannok consumed by Canadian Inuit. J Am Diet Assoc

2001;101(5):580–1.

Kuhnlein HV, Chan HM, Receveur O, Egeland G. Canadian arctic

indigenous peoples, traditional food systems and POPs. In:

Downie DL, Fenge T, editors. Northern lights against POPs:

combating toxic threats in the arctic. Montreal7 McGill-Queens

University Press; 2003. p. 22 – 40.

Kuhnlein HV, Receveur O, Soueida R, Egeland GM. Arctic indi-

genous peoples experience the nutrition transition with changing

dietary patterns and obesity. J Nutr 2004;134:1447 – 53.

Lacour M, Zunder T, Restle A, Schwarzer G. No evidence for an

impact of selenium supplementation on environment associated

health disorders—a systematic review. Int J Hyg Environ Health

2004;207(1):1–13.

Lagueux J, Pereg D, Ayotte P, Dewailly E

, Poirier GG. Cytochrome

P450 CYP1A1 enzyme activity and DNA adducts in placenta

from women environmentally exposed to organochlorines.

Environ Res 1999;80(4):369 – 82.

Lampe J, Murphy F, Furgal C, Craig L. Country food nutrition and

health: developing effective communication strategies in Labra-

dor (Year 2). In: Kalhok S, editor. Synopsis of research con-

ducted under the 2003–2004 Northern Contaminants Program.

Ottawa, Ontario, Canada7 Department of Indian Affairs and

Northern Development; 2000. p. 271 – 80.

Lans MC, Spiertz C, Brouwer A, Koeman JH. Different competition

of thyroxine binding to transthyretin and thyroxine-binding

globulin by hydroxy-PCBs, PCDDs and PCDFs. Eur J Pharma-

col Environ Toxicol 1994;270:129 – 36.

Lapierre P, De Guise S, Muir DC, Norstrom R, Beland P, Fournier

M. Immune functions in the Fisher rat fed beluga whale (Del-

phinapterus leucas) blubber from the contaminated St Lawrence

estuary. Environ Res 1999;80(2 Pt. 2):S104 – 12.

Lawn J, Harvey D. Change in nutrition and food security in two

Inuit communities, 1992 to 1997. Ottawa, Ontario, Canada7

Department of Indian Affairs and Northern Development;

2001. 108 pp.

Lawn J, Hill F. Alternative Northern food baskets. Ottawa, Ontario,

Canada7 Department of Indian Affairs and Northern Develop-

ment; 1998. 47 pp.

Lawn J, Langer N. Air stage subsidy monitoring program. Final

report. Food consumption survey, Vol. 1. Department of Indian

Affairs and Northern Development, Ottawa, Ontario, Canada;

1994, 64 pp.

LIA (Labrador Inuit Association). Environmental health study, Final

report. Eco-research project. Labrador Inuit Association, Nain,

Labrador, Canada; 1997, 63 pp.

Longnecker MP, Klebanoff MA, Brock JW, Zhou H, Gray KA,

Needham LL, et al. Maternal serum level of 1,1-dichloro-2,2-

bis( p-chlorophenyl)ethylene and risk of cryptorchidism,

hypospadias, and polythelia among male offspring. Am J Epi-

demiol 2002;155(4):313 – 22.

Longnecker MP, Wolff MS, Gladen BC, Brock JW, Grandjean P,

Jacobson JL, et al. Comparison of polychlorinated biphenyl

(PCB) levels across studies of human neurodevelopment.

Environ Health Perspect 2003;111(1):65 – 70.

Lucier G, Nelson K, Everson R, Wong T, Philpot R, Tiernan T, et al.

Placental markers of human exposure to polychlorinated biphe-

nyls and polychlorinated dibenzofurans. Environ Health Per-

spect 1987;76:79 – 87.

Lund BO, Miller DM, Woods JS. Studies on Hg(II)-induced H

formation and oxidative stress in vivo and in vitro in rat kidney

mitochondria. Biochem Pharmacol 1993;45(10):2017–24.

Mably TA, Bjerke DL, Moore RW, Gendron-Fitzpatrick A, Peterson

RE. In utero and lactational exposure of male rats to 2,3,7,8-

tetrachlorodibenzo-p-dioxin: 3 Effects on spermatogenesis and

reproductive capability. Toxicol Appl Pharmacol 1992;114:

118– 26.

MacDonald CR. Personal communication. Northern Environmental

Consulting. Pinawa, Manitoba; 2002.

MacDonald CR, Ewing LL, Elkin BT, Wiewel AM. Regional

variation in radionuclide concentrations and radiation doses in

caribou (Rangifer tarandus) in the Canadian Arctic. Sci Total

Environ 1996;182:53 – 73.

MacLeod M, Woodfine D, Brimacombe J, Toose L, Mackay D. A

dynamic mass budget for toxaphene in North America. Environ

Toxicol Chem 2002;21(8):1628– 37.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246242

Magos L, Clarkson TW, Hudson AR. Differences in the effects of

selenite and biological selenium on the chemical form and

distribution of mercury after the subcutaneous administration

of HgCl

and selenium to rats. J Pharmacol Exp Ther

1983;228:478–83.

Marshall H, Tracy BL. Radiocesium in the Caribou of Northern

Canada Internal report available from the Radiation Protection

Bureau 6302D1. Ottawa, Ontario, Canada7 Health Canada;

1989.

McDonald JG, Hites RA. Radial dilution model for the distribution

of toxaphene in the United States and Canada on the basis of

measured concentrations in tree bark. Environ Sci Technol

(37):475– 81.

McKeown-Eyssen GE, Ruedy J. Prevalence of neurological

abnormalities in Cree Indians exposed to methylmercury in

Northern Que´bec. Clin Invest Med 1983;6(3):161–9.

Meerts IA, Letcher RJ, Hoving S, Marsh G, Bergman A, Lemmen

JG, et al. In vitro estrogenicity of polybrominated diphenyl

ethers, hydroxylated PDBes, and polybrominated bisphenol A

compounds. Environ Health Perspect 2001;109(4):399 – 407.

Muckle G. Personal communication. Laval University, eauport,

Que´bec; 2000.

Muckle G, Ayotte P, Dewailly E

, Jacobson SW, Jacobson JL.

Determinants of polychlorinated biphenyls and methylmercury

exposure in Inuit women of childbearing age. Environ Health

Perspect 2001a;109(9):957 – 63.

Muckle G, Ayotte P, Dewailly E

, Jacobson SW, Jacobson JL.

Prenatal exposure of the northern Que´bec Inuit infants to envir-

onmental contaminants. Environ Health Perspect 2001b;

109(12):1291–9.

Muckle G, Dewailly E

, Ayotte P, Jacobson SW, Jacobson JL.

Contributions of PCBs, pesticides, MeHg and N-3 fatty acids

to fetal growth and motor development in inuit infants in Arctic

Quebec. Twenty-first International Neurotoxicology Confer-

ence. Conference Proceedings. Honolulu, Hawaii; 2004.

Muckle G, Ayotte P, Dewailly E

, Jacobson SW, Jacobson JL.

Prenatal exposure to PCBs and birth size in Northern Que´bec

Inuit, Canada in preparation.

Muir DCG, Born EW, Koczansky K, Stern GA. Temporal and

spatial trends of persistent organochlorines in Greenland walrus

(Odobenus rosmarus romarus). Sci Total Environ 2000;245:

73 – 86.

Myers GJ, Marsh DO, Davidson PW, Cox C, Shamlaye CF, Tanner

MA, et al. Main neurodevelopmental study of Seychellois chil-

dren following in utero exposure to methylmercury from a

maternal fish diet: outcome at six months. Neurotoxicology

1995;16(4):653–64.

Myers GJ, Davidson PW, Cox C, Shamlaye CF, Palumbo D,

Cernichiari E, et al. Prenatal methylmercury exposure from

ocean fish consumption in the Seychelles child development

study. Lancet 2003;361:1686–92.

Newsome WH, Ryan JJ. Toxaphene and other chlorinated com-

pounds in human milk from northern and southern Canada.

Chemosphere 1999;39(3):519 – 26.

NRC (National Research Council). Diet and Health: implications for

reducing chronic disease risk. Washington, DC7 National Aca-

demic Press; 1989.

NRC (National Research Council). Biological markers in immuno-

toxicology. Washington, DC7 National Academic Press; 1992.

NRC. Toxicological effects of methylmercury. Washington, DC7

National Academy Press; 2000.

Odland JO. Personal communication. Institute of Community Med-

icine, University of TromsU, TromsU, Norway; 2001.

Ohi G, Nishigaki S, Seki H, Tamura Y, Maki T, Konno H, et al.

Efficacy of selenium in tuna and selenite in modifying methyl-

mercury intoxication. Environ Res 1976;12:49 – 58.

Olson JA. Hypovitaminosis A: contemporary scientific issues.

Symposium: Clinical nutrition in developing countries:

towards the application of contemporary concepts and technol-

ogy. Ames, Iowa, USA7 American Institute of Nutrition; 1994.

p. 461S – 1466S.

O’Neil JD, Elias B, Yassi A. Poisoned food: cultural resistance to

the contaminants discourse in Nunavik. Arct Anthropol

1997;34(1):29– 40.

Ornaghi F, Ferrini S, Prati M, Giavini E. The protective effects of N-

acetyl-l-cysteine against methyl mercury embryotoxicity in

mice. Fundam Appl Toxicol 1993;20:437 – 45.

Park ST, Lim KT, Chung YT, Kim SU. Methylmercury-induced

neurotoxicity in cerebral neuron culture is blocked by antiox-

idants and NMDA receptor antagonists. Neurotoxicology

1996;17(1):37– 45.

Patandin S, Lanting CL, Mulder PGH, Boersma ER, Sauer PJ,

Weisglas-Kuperus N. Effects of environmental exposure to

polychlorinated biphenyls and dioxins on cognitive abilities

in Dutch children at 42 months of age. J Pediatr 1999;

134(1):33 – 41.

Patrono C, FitzGerald GA. Isoprostanes: potential markers of oxi-

dant stress in atherothrombotic disease. Arterioscler Thromb

Vasc Biol 1997;17:2309 – 15.

Pelkonen O, Ka¨rki N, Koivisto M, Tuimala R, Kauppila A. Mater-

nal cigarette smoking, placental aryl hydrocarbon hydroxylase

and neonatal size. Toxicol Lett 1979;3:331 – 5.

Pereg D, Lagueux J, Dewailly E

, Poirier GG, Ayotte P. Cigarette

smoking during pregnancy: comparison of biomarkers for

inclusion in epidemiological studies. Biomarkers 2001;6:

161 – 73.

Pereg D, Dewailly E

, Poirier GG, Ayotte P. Placental CYP1A1

activity in Inuit women environmentally exposed to poly-

chlorinated biphenyls. Environ Health Perspect 2002;110(6):

1–6.

Pidgeon N, Hood C, Jones D, Turner B, Gibson R. Risk: analysis,

perception, management Chapter 5 risk perception Report of a

royal society study group. London, England7 The Royal Society;

1992. 201 pp.

Poirier S, Brooke L. Inuit perceptions of contaminants, environ-

mental knowledge and land use in Nunavik: the case of Salliut.

Preliminary research report, Eco-research program, CHUL

Research Centre, Beauport, Que´bec, Canada; 1997.

Poirier S, Brooke L. Inuit perceptions of contaminants, environ-

mental knowledge and land use in Nunavik. In: Eco-research

Avativut/Ilusivut research program, Final report. Craig L (ed.).

Unite´ de recherche en sante´ publique, CHUQ-Pavillon,

C.H.U.L., Beauport, Que´bec, Canada, 43–53; 1999.

Porterfield SP, Hendry LB. Impact of PCBs on thyroid hormone

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 243

directed brain development. Toxicol Ind Health 1998;14(1/

2):103 – 20.

Powell D, Leiss W. Mad cows and mother’s milk. The perils of poor

risk communication. Montreal, Que´bec, Canada7 McGill-

Queen’s University Press; 1997.

Pratico D. F2-isoprostanes: sensitive and specific non-invasive

indices of lipid peroxidation in vivo. Atherosclerosis 1999;

147:1– 10.

Preparatory Commission. Comprehensive-Nuclear-Test-Ban Treaty

(CTBT). Preparatory Commission for the CTBT Organization,

Provisional Technical Secretariat, Vienna, Italy; 1996.

Presidential/Congressional Commission on Risk Assessment and

Risk Management. Framework for Environmental Health Risk

Management. Washington, DC; 1997a.

Presidential/Congressional Commission on Risk Assessment and

Risk Management. Symposium on a public health approach to

environmental protection. Presidential/Congressional Commis-

sion on Risk Assessment and Risk Management, Washington,

DC, USA; 1997b, 17 pp.

Rayman MP. The importance of selenium to human health. Lancet

2000;356:233–41.

Receveur O, Boulay M, Mills C, Carpenter W, Kuhnlein H. Var-

iance in food use in Dene/Me´tis communities Centre for indi-

genous peoples’ nutrition and environment (CINE). Montreal,

Que´bec, Canada7 McGill University; 1996. October.

Receveur O, Boulay M, Kuhnlein HV. Decreasing traditional food

use affects diet quality for adult Dene/Me´tis in 16 communities

of the Canadian Northwest Territories. J Nutr 1997;127(11):

2179 – 86.

Receveur O, Kassi N, Chan HM, Berti PR, Kuhnlein HV. Yukon

First Nations assessment of dietary benefit/risk Centre for indi-

genous peoples’ nutrition and environment (CINE). Montreal,

Que´bec, Canada7 McGill University; 1998a. 160 pp.

Receveur O, Kuhnlein HVDene Nation, Me´tis Nation (NWT).

Benefits of traditional food in Dene/Me´tis communities. Int J

Circumpolar Health 1998b;57:219 – 21 [Suppl.].

Ridington R. Trail to heaven: knowledge and narrative in a northern

native community. Vancouver7 Douglas and McIntyre; 1988.

Rogan W, Gladen BC. PCBs, DDE, and child development at 18

and 24 months. Ann Epidemiol 1991;1:409 – 13.

Rogan W, Gladen BC, McKinney JD, Carreras N, Hardy P, Thullen

J, et al. Neonatal effects of transplacental exposure to PCBs and

DDE. J Pediatr 1986;109(2):335 – 41.

Rogan W, Gladen BC, McKinney JD, Carreras N, Hardy P, Thullen

J, et al. Polychlorinated biphenyls (PCBs) and dichlorodiphenyl

dichloroethene (DDE) in human milk: effects of maternal fac-

tors and previous lactation. Am J Public Health 1986;76(2):

172– 7.

Rogan W, Gladen BC, Hung K, Koong S, Shih L. Congenital

poisoning by polychlorinated biphenyls and their contaminants

in Taiwan. Science 1988;241:334 – 6.

Salonen JT, Alfthan G, Huttunen JK, Pikkarainen J, Puska P.

Association between cardiovascular death and myocardial

infarction and serum selenium in a matched-pair longitudinal

study. Lancet 1982;2(8291):175–9.

Salonen JT, Seppanen K, Nyyssonen K, Korpela H, Kauhanen J,

Kantola M, et al. Intake of mercury from fish, lipid peroxida-

tion, and the risk of myocardial infarction and coronary, cardi-

ovascular and any death in eastern Finnish men. Circulation

1995;91:645– 55.

Sandau CD, Ayotte P, Dewailly E

, Duffe J, Norstrom RJ. Analysis

of hydroxylated metabolites of PCBs (OH-PCBs) and other

chlorinated phenolic compounds in whole blood from Canadian

Inuit. Environ Health Perspect 2000;108:611 – 6.

Sandau CD, Ayotte P, Dewailly E

, Duffe J, Norstrom RJ. PCP and

hydroxylated-PCB metabolites in umbilical cord plasma of

neonates from coastal populations in Que´bec. Environ Health

Perspect 2002;110:411 – 7.

Sante´ Que´bec. Sante´ Que´bec: a health profile of the Inuit. In: Jette´J,

editor. Ministe´re de la Sante´ et des Services Sociaux. Montreal,

Que´bec, Canada7 Gouvernement du Que´bec; 1994 271 pp.

Sante´ Que´bec. Report on the Sante Que´bec Health Survey among

the Inuit of Nunavik, 1992 Diet a health determining factor.

Ministe´re de la Sante´ et des Services Sociaux. Gouvernement du

Que´bec, Montreal, Que´bec, Canada; 1995.

Sarafian T, Verity MA. Oxidative mechanisms underlying methyl

mercury neurotoxicity. Int J Dev Neurosci 1991;9(2):147 – 53.

Schaefer O, Steckle J. Dietary habits and nutritional base of native

populations of the Northwest Territories. Science Advisory Board

of the Northwest Territories. Government of the Northwest Ter-

ritories, Yellowknife, Northwest Territories, Canada; 1980,

37 pp.

Schrader TJ, Boyes BG, Matula TI, Heroux-Metcalf C, Langlois I,

Downie RH. In vitro investigation of toxaphene genotoxicity in S

typhimurium and Chinese hamster V79 lung fibroblasts. Mutat

Res/Genet Toxicol Environ Mutagen 1998;413(2):159 – 68.

Schuur AG, Legger FF, van Meeteren ME, Moonen MJ, van

Leeuwen-Bol I, Bergman A, et al. In vitro inhibition of

thyroid hormone sulfation by hydroxylated metabolites of

halogenated aromatic hydrocarbons. Chem Res Toxicol

1998;11:1075 – 81.

Schwartz PM, Jacobson SW, Fein G, Jacobson JL, Price HA. Lake

Michigan fish consumption as a source of polychlorinated

biphenyls in human cord serum, maternal serum, and milk.

Am J Public Health 1983;73(3):293 – 6.

Seddon L. Maternal and Cord Blood Monitoring for Environmental

Contaminants: Final report for Kitikmeot Health Board and

Mackenzie Regional Health Service. Yellowknife, NWT,

Canada7 Department of Health and Social Services; 1996.

200 pp.

Semba RD. Vitamin A, immunity, and infection. Clin Infect Dis

1994;19:489– 99.

Sharma M, Bhatnagar P. Organochlorine pesticides and preterm

labour in human beings. Curr Sci 1996;71(8):628– 31.

Sherbin IG. Mercury in the Canadian Environment Economic and

technical review report EPS 3-EC-79-6. Ottawa, Ontario,

Canada7 Environmental Protection Service, Environment

Canada; 1979.

Shimizu Y, Pierce DA, Preston DL, Machuchi K. Studies of mor-

tality of atomic bomb survivors Report 12: Part II Non-cancer

mortality: 1950–1990. Radiat Res 1999;152:374– 89.

Shmidt EB, Skou HA, Christensen JH, Dyerberg J. N-3 fatty acids

from fish and coronary artery disease: implications for public

health. Public Health Nutr 2000;3(1):91 – 8.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246244

Simoneau N, Receveur O. Attributes of vitamin A and calcium-rich

food items consumed in K’asho Got’ine, Northwest Territories,

Canada. J Nutr Educ 2000;32:84 – 93.

Simopoulos AP. Omega-3 fatty acids in health and disease and in

growth in development. Am J Clin Nutr 1991;54:438 – 63.

Sjo¨din A, Hagmar L, Klasson-Wehler E, Bjo¨ rk J, Bergman A.

Neutral and phenolic organohalogen contaminants in human

plasma related to intake of Baltic fish. Environ Health Perspect

2000;108(11):1035 – 41.

Slovic P. Perception of risk. Science 1987;236:280 – 5.

Slovic P, Peters E. The role of affect and world views as orienting

dispositions in the perception and acceptance of nuclear power.

Eugene, Oregon, USA7 Decision Research; 1995.

Snider J, Gill U. Unpublished results. Health Canada, Ottawa,

Ontario, Canada; 2001.

Sommer A, West KP. Vitamin A deficiency: health, survival, and

vision. New York, USA7 Oxford University Press; 1996.

Statistics Canada. Profile of Canada’s Aboriginal population. Cus-

tom Database (Adapted from 1996 data). Ottawa, Ontario,

Canada; 2001.

Stephens G, Chan L. Risk characterization of arsenic exposure from

consumption of berries in the Akaitcho territory. In: Kalhok S,

editor. Synopsis of research conducted under the 2000–2001

Northern Contaminants Program. Ontario, Canada7 Department

of Indian Affairs and Northern Development. Ottawa; 2001.

p. 271 – 3.

Steuerwald U, Weihe P, Jorgensen PJ, Bjerve K, Brock J,

Heinzow B, et al. Maternal seafood diet, methylmercury

exposure, and neonatal neurologic function. J Pediatr

2000;136(5):599–605.

Stewart P, Darvill T, Lonky E, Reihman J, Pagano J, Bush B.

Assessment of prenatal exposure to PCBs from maternal con-

sumption of Great Lakes fish: an analysis of PCB pattern and

concentration. Environ Res 1999;80(2 Pt2):S87 – 96.

Stewart P, Reihman J, Lonky E, Darvill T, Pagano J. Prenatal PCB

exposure and neonatal behavioural assessment scale (NBAS)

performance. Neurotoxicol Teratol 2000;22:21 – 9.

Stewart PW, Reihman J, Lonky EI, Darvill TJ, Pagano J. Cognitive

development in preschool children prenatally exposed to PCBs

and MeHg. Neurotoxicol Teratol 2003;25:1 – 12.

Stiles M, Usher PJ. A guide to making presentations in northern

communities for the Northern Contaminants Program. Ottawa,

Ontario, Canada7 Minister of Indian Affairs and Northern Devel-

opment; 1998.

Szathmary EJE, Ritenbaugh C, Goodby CSM. Dietary change and

plasma glucose levels in an Amerindian population undergoing

cultural transition. Soc Sci Med 1987;24(10):791– 804.

Thomas PA. Dosimetry of

210

Po in caribou, wolves and humans in

northern Canada. Health Phys 1994;66(6):678– 90.

Thomas PA, Fisenne I, Chorney D, Baweja AS, Tracy BL. Human

absorption and retention of polonium-210 from caribou meat.

Radiat Prot Dosimetry 2001;97:241 – 50.

Thouez JP, Rannou A, Foggin P. The other face of development:

native population, health status, and indicators of malnutrition—

the case of the Cree and Inuit of northern Que´bec. Soc Sci Med

1989;29(8):965–74.

Tofflemire K. Inuvik Regional Human Contaminants Monitoring

Program: Regional Report. IRHSSB, Inuvik. July; 2000,

38 pp.

Toft G, Pedersen HS, Bonde JP. Semen quality in Greenland.

International Congress for Cirucumpolar Health 12, Conference

Abstract, Nuuk; 2003.

Tracy BL, Kramer GH. A method for estimating caribou consump-

tion by northern Canadians. Arctic 2000;53(1):42– 52.

Tracy BL, Kramer GH, Zielinski JM, Jiang H. Radiocesium body

burdens in residents of northern Canada from 1963–1990.

Health Phys 1997;72(3):431 – 42.

Tryphonas H, Arnold DL, Bryce F, Huang J, Hodgen M, Ladouceur

DT, et al. Effects of toxaphene on the immune system of

cynomologus (Macaca fascicularis) monkeys. Food Chem Tox-

icol 2001;39(9):947–58.

Tryphonas H, Bondy G, Hodgen M, Coady L, Parenteau M, Arm-

strong C, et al. Effects of cis-nonachlor, trans-nonachlor and

chlordane on the immune system of Sprague–Dawley rats fol-

lowing a 28-day oral (gavage) treatment. Food Chem Toxicol

2003;41(1):107– 18.

UNSCEAR (United Nations Scientific Committee on the Effects of

Atomic Radiation). Sources and effects of ionizing radiation.

UNSCEAR. United Nations, New York, USA; 1993.

US EPA (United States Environmental Protection Agency). Gui-

dance for cumulative risk assessment: Part 1. Planning and

scoping. Washington, DC. (available at http://www.epa.gov/

brownfields/pdf/cumrisk2.pdf); 1997.

US EPA. Supplementary guidance for conducting health risk assess-

ment of chemical mixtures. EPA/630/R-00/002; 2000.

Usher PJ, Wenzel G. Socio-economic aspects of harvesting. In:

Ames R, Axford D, Usher PJ, Weick E, Wenzel G, editors.

Keeping on the land: a study of the feasibility of a comprehen-

sive wildlife support programme in the northwest territories.

Ottawa, Ontario, Canada7 Canadian Arctic Resources Commit-

tee; 1989.

Usher P, Baikie M, Demmer M, Nakashima D, Stevenson MG, Stiles

M. Communicating about contaminants in country food:

the experience in aboriginal communities. Ottawa, Ontario,

Canada7 Research Department, Inuit Tapirisat of Canada; 1995.

January.

Van den Berg KJ. Interaction of chlorinated phenols with thyroxine

binding sites of human transthyretin, albumin and thyroid bind-

ing globulin. Chem Biol Interact 1990;76:63 – 75.

Van den Berg KJ, van Raaij JA, Bragt PC, Notten WR. Interactions

of halogenated industrial chemicals with transthyretin and

effects on thyroid hormone levels in vivo. Arch Toxicol

1991;65:15 – 9.

Van den Berg M, Birnbaum L, Bosveld ATC, Brunstro¨m B, Cook P,

Feeley M, et al. Toxic equivalency factors (TEFS) for PCBs,

PCDDs, PCDFs for humans and wildlife. Environ Health Per-

spect 1998;106(12):775 – 92.

Van Loveren H, Germolec D, Koren HS, Luster MI, Nolan C,

Repetto R. Report of the Bilthoven symposium: advancement

of epidemiological studies in assessing the human health effects

of immunotoxic agents in the environment and the workplace.

Biomarkers 1999;4(2):135 – 57.

Van Oostdam J. Personal communication. Health Canada, Ottawa,

Ontario, Canada; 2001.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246 245

Van Oostdam J, Feeley M. Personal communication. Health

Canada, Ottawa, Ontario, Canada; 2002.

Van Oostdam J, Gilman A, Dewailly E

, Usher P, Wheatley B,

Kuhnlein H, et al. human health implications of environmental

contaminants in Arctic Canada: a review. Sci Total Environ

1999;230:1–282.

Van Oostdam J, Tremblay N, Walker J, McMullen E, Smith S.

Contaminants in cord blood and relationship to maternal

blood levels in the Northwest Territories and Nunavut, Canada:

1994–1999. Epidemiology 2001;12(4):S80 [ISEE 428].

Vreugdenbil HI, Lanting CI, Mulderm PGH, Boersma ER, Weis-

glas-Kuperus N. Effects of prenatal PCB and dioxin background

exposure on cognitive and motor abilities in Dutch children at

school age. J Pediatr 2001;140:48 – 56.

Wagemann R, Trebacz E, Hunt R, Boila G. Percent methylmercury

and organic mercury in tissues of marine mammals and fish

using different experimental and calculation methods. Environ

Toxicol Chem 1997;16(9):1859–66.

Wagemann R, Trebacz E, Boila G, Lockhart WL. Mercury species

in the liver of ringed seals. Sci Total Environ 1997;261(1–3):

21 – 32.

Walkowiak J, Wiener J, Fastabend A, Heinzow B, Kra¨mer U,

Schmidt E. Environmental exposure to polychlorinated biphe-

nyls and quality of the home environment: effects on psycho-

development in early childhood. Lancet 2001;358:1602 – 7.

Ward JI, Lum MK, Hall DB, Silimperi DR, Bender TR. Haemo-

philus influenzae type b disease in Alaska: background epide-

miology for a vaccine efficacy trial. J Infect Dis 1986;153(1):

17 – 26.

Watanabe C. Modification of mercury toxicity by selenium: prac-

tical importance? Tohoku J Exp Med 2002;196(2):71 – 7.

Watanabe C, Yin K, Kasanuma Y, Satoh H. In utero exposure to

methylmercury and Se deficiency converge on the neurobeha-

vioral outcome in mice. Neurotoxicol Teratol 1999;21:83 – 9.

Weber JP. The AMAP Ring Test: a laboratory intercomparison

program for POPs in human serum. Presented at the AMAP

Conference and Workshop: Impacts of POPs and Mercury on

Arctic Environments and Humans, TromsU, Norway; 2002.

January.

Wein EE, Freeman MMR. Inuvialuit food use and food preferences

in Aklavik, NWT, Canada. Arctic Med Res 1992;51(4):159 – 72.

Wein EE, Freeman MM, Makus JC. Preliminary assessment of

nutrients in daily diets of a sample of Belcher Island Inuit adults.

Int J Circumpolar Health 1998;57(1):205 – 10.

Weinstein MS. What the land provides: an examination of the Fort

George subsistence economy and the possible consequences on

it of the James Bay hydroelectric project eds. Grand Council of

the Crees (eds.). Montreal, Que´bec, Canada; 1976. 255 pp.

Weisburger JH. Approaches for chronic disease prevention based on

current understanding of underlying mechanisms. Am J Clin

Nutr 2000;71:1710S – 4S [Suppl.].

Weisglas-Kuperus N, Sas TC, Koopman-Esseboom C, van der

Zwan CW, De Ridder MA, Beishuizen A, et al. Immunologic

effects of background prenatal and postnatal exposure to dioxins

and polychlorinated biphenyls in Dutch infants. Pediatr Res

1995;38(3):404–10.

Weisglas-Kuperus N, Patandin S, Berbers GAM, Sas TCJ, Mulder

PGH, Sauer PJJ, et al. Immunologic effects of background

exposure to polychlorinated biphenyls and dioxins in Dutch

preschool children. Environ Health Perspect 2000;108(12):

1203 – 7.

Wenzel GW, Qillaq E. Inuit response to environmental contaminants

information: the Cape Christian-Clyde River case. In: Kalhok S,

editor. Synopsis of research conducted under the 2003–2004

Northern Contaminants Program. Ottawa, Ontario, Canada7

Department of Indian Affairs and Northern Development;

2000. p. 301 – 7.

Whanger PD. Selenium and the brain: a review. Nutr Neurosci

2001;4(2):81– 97.

Wheatley B, Paradis S. Balancing human exposure, risk and reality:

questions raised by the Canadian Aboriginal methylmercury

program. Neurotoxicology 1996;17(1):241 – 9.

Wheatley B, Paradis S. Northern exposure: further analysis of the

results of the Canadian Aboriginal methylmercury program. Int

J Circumpolar Health 1998;57(Suppl. 1):586 – 90.

Wheatley MA, Wheatley B. The effect of eating habits on mercury

levels among Inuit residents of Sugluk, P.Q. Etudes/Inuit Studies

1981,5(1):27:43.

White KLJ, Lysy HH, McKay JA. Modulation of serum comple-

ment levels following exposure to polychlorinated dibenzo-p-

dioxins. Toxicol Appl Pharmacol 1986;84:209 – 19.

WHO (World Health Organization). Summary of toxicological data

of certain food additives and contaminants. Twenty-second

Report of the Joint FAO/WHO Expert Committee on Food

Additives, Geneva. WHO Food Additive Series No 13, 1978.

WHO (World Health Organization). Food additive series 46: cad-

mium. Geneva7 Safety Evaluation of Certain Food Additives

and Contaminants; 2000.

Wong M. Chemical residues in fish and wildlife harvested in North-

ern Canada. Environmental Studies. Ottawa, Ontario, Canada7

Department of Indian Affairs and Northern Development; 1986.

World Cancer Research Fund. American institute for cancer

research, food, nutrition and the prevention of cancer: a global

perspective. Washington, DC, USA7 American Institute for

Cancer Research; 1997. 670 pp.

Wu XJ, Lu WQ, Mersch-Sundermann V. Benzo(a)pyrene induced

micronucleus formation was modulated by persistent organic

pollutants (POPs) in metabolically competent human HepG2

cells. Toxicol Lett 2003;144(2):143–50.

Xhman B, Xhman G. Radiocesium in Swedish reindeer after the

Chernobyl fallout: seasonal variations and long-term decline.

Health Phys 1994;66(5):503 – 12.

Yamashita F. Clinical features of polychlorobiphenyl (PCB)-

induced fetopathy. Paediatrician 1977;6:20 – 7.

Yee S, Choi BH. Oxidative stress in neurotoxic effects of methyl-

mercury poisoning. Neurotoxicology 1996;17:17 – 26.

Young TK. Diabetes mellitus among native Americans in Canada

and United States: an epidemiological review. Am J Human Biol

1993;5:399– 413.

Young TK, Moffat MEK, O’Neil JD. Cardiovascular diseases in a

Canadian Arctic population. Am J Public Health 1993;83:

881 – 7.

Zile MH. Vitamin A homeostasis endangered by environmental

pollutants. Proc Soc Exp Biol Med 1992;201:141– 53.

J. Van Oostdam et al. / Science of the Total Environment 351–352 (2005) 165–246246