sign in sign up
<<
Home , Inuit women, Nunavik

This article was downloaded by: [University of Santa Cruz] On: 19 2012, At: 22:03 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Archives of Environmental Health: An International Journal Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/vzeh20 Exposure of the Population of ( Québec) to Lead and Mercury Éric Dewailly a b , Pierre Ayotte a b , Suzanne Bruneau a , Germain Lebel a , Patrick Levallois a b & Jean Philippe Weber c a Unité de Recherche en Santé Publique, Centre Hospitalier Universitaire de Québec (CHUQ), Beauport, Québec, b Département de Médecine Sociale et Préventive, Université Laval, Ste-Foy, Québec, Canada c Centre de Toxicologie du Québec, Institut National de Santé Publique, Ste-Foy, Québec, Canada Version of record first published: 05 Apr 2010.

To cite this article: Éric Dewailly , Pierre Ayotte , Suzanne Bruneau , Germain Lebel , Patrick Levallois & Jean Philippe Weber (2001): Exposure of the Inuit Population of Nunavik (Arctic Québec) to Lead and Mercury, Archives of Environmental Health: An International Journal, 56:4, 350-357 To link to this article: http://dx.doi.org/10.1080/00039890109604467

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. Exposure of the Inuit Population of Nunavik (Arctic ) to Lead and Mercury

ERIC DEWAILLY PATRICK LEVALLOIS PIERRE AYOTTE Unite de Recherche en SantC Publique Unit4 de Recherche en SantC Publique Centre Hospitalier Universitaire de Quebec (CHUQ) Centre Hospitalier Universitaire de QuCbec (CHUQ) Beauport, Quebec, Canada Beauport, Quebec, Canada and and DCpartement de MCdecine Sociale et Preventive DCpartement de MCdecine Sociale et Preventive UniversitC Lava1 UniversitC Lava1 Ste-Foy, QuCbec, Canada Ste-Foy, QuCbec, Canada JEAN PHILIPPE WEBER SUZANNE BRUNEAU Centre de Toxicologie du QuCbec CERMAIN LEBEL lnstitut National de SantC Publique Unite de Recherche en SantC Publique Ste-Foy, Quebec, Canada Centre Hospitalier Universitaire de QuCbec (CHUQ) Beauport, QuCbec, Canada

ABSTRACT. The authors conducted a survey during 1992 to evaluate blood levels of lead and mercury in Inuit adults of Nunavik (Arctic Quebec, Canada). Blood samples obtained from 492 participants (209 males and 283 females; mean age = 35 yr) were analyzed for lead and total mercury; mean (geometric) concentrations were 0.42 pnol/l (range = 0.04-2.28 pmol/l) and 79.6 nmol/l (range = 4-560 nmol/l), respectively. Concentrations of omega-3 fatty acid in plasma phospholipids-a biomarker of marine food consumption-were corre- lated with mercury (r = .56! p < .001) and, to a lesser extent, with blood lead levels (r = .31, p < .001). Analyses of variance further revealed that smoking, age, and consumption of waterfowl were associated with lead concentrations (8= .30, p < .001), whereas age and consumption of seal and beluga were related to total mercury levels (8 = .30, p < .001). A significant proportion of reproductive-age women had lead and mercury concen- trations that exceeded those that have been reportedly associated with subtle neurodevel- Downloaded by [University of California Santa Cruz] at 22:03 19 December 2012 opmental deficits in other populations.

MERCURY AND LEAD are widespread environmen- Native people who rely on seafood for subsistence tal contaminants that originate from both anthropo- can receive unusually high doses of heavy metals genic (e.g., mining, smelting, fossil fuel burning, waste because they are located at a high trophic level of the incineration) and natural (e.g., local geology, volca- aquatic food chain, along which biomagnification of noes, degassing in aquatic environments) sources. In persistent contaminants occurs. Few studies have docu- addition to the contribution of local sources, these mented lead exposure in Arctic populations. Results heavy metals can be transported from distant sources to from a 1987 survey conducted among Northern Green- the Arctic by oceanic and atmospheric transport. Lead land Inuit hunters and their families revealed median and mercury have been found in all components of the blood lead concentrations of 0.46 pmol/l (range = Arctic ecosystem.'T2 0.1 5-1.1 6 pmol/l) for 35 men and 0.27 pmol/l (range =

350 Archives of Environmental Health Rlollso~ion:Mar~e-FranceGagnon. Unit6 de iecherche en rant6 publique de Qudbec.

Fig. 1. Localization of Nunavik.

0.04-1.57 pmol/l) for 32 women.3 Seafood consump- mals. In (Eastern Canadian Arctic, formerly tion (i.e., , , , and beluga part of the ), the mean mercury whale) was the likely source of exposure, and age was concentration documented in 286 between 1972 Downloaded by [University of California Santa Cruz] at 22:03 19 December 2012 associated positively with blood lead concentration. In and 1989 was 97 nmol/l (range = 5-1,000 nmol/l). Mer- the Faroe Islands, 52 adult women who consumed fish cury concentrations determined in 1982 and 1983 for and pilot whale meat had a mean blood lead level of 76 Den& from the western part of the Northwest Terri- 0.10 pmol/l (range = 0.04-0.1 7 prn~l/l).~ tories averaged 53 nmol/l (range = 7-332 nm~l/l).~ Mercury is mainly present as methylmercury in fish More recently, the survey conducted by Grandjean et and marine mammals, and their consumption consti- al.4 among 53 women from the Faroe Islands revealed a tutes an important source of exposure, especially in sus- mean blood mercury concentration of 60 nmol/l (range tenance populations such as the Inuits. Various surveys = 13-249 nmol/l). In general populations that consume have documented the level of mercury exposure in Arc- little or no fish or marine mammals, blood mercury val- tic populations. Analysis of blood samples collected ues generally exist around 10 nmol/l. For example, a between 1977 and 1982 from 142 Inuits residing in recent survey revealed that blood concentrations in Nunavik (Arctic Quebec) revealed a mean blood mer- 1,127 healthy American men averaged 10 nmol/l.6 cury concentration of 240 nmol/l (range = 21-1,269 In view of the relatively high exposure of the Inuit nm~l/l).~Mercury concentrations were associated population of Nunavik to mercury documented in the strongly with consumption of fish and marine mam- late 1970s and early 1980s and the lack of data on lead

July/August 2001 [vol. 56 (No. 4)] 351 exposure, we deemed that a new survey was necessary (Pharmacia Mercury monitor [Piscataway, New Jersey]). to (1) obtain data on lead exposure in this population; Samples were microwave-digested with nitric acid, and (2) compare mercury exposure 10 yr later; (3) obtain an aliquot was used for the analysis. We added a body burden data in various subgroups of this popula- stannous chloride/cadmium chloride mixture to the tion, including women of reproductive age; and (4) sample to reduce both methyl and inorganic mercury to study in detail the factors modulating this exposure. In elemental mercury. Inorganic mercury was measured in this study, we report on the biological exposure to lead 18 samples for which the total mercury concentration and total mercury in 492 Inuit adults from Nunavik and was greater than 70 nmol/l. Reduction with stannous on the associated dietary, life-style, and sociodemo- chloride alone allowed for selective reduction of graphic factors. inorganic mercury. We used reference material from the Qubbec Toxi- Material and Method cology Center interlaboratory program (blood samples containing known concentrations of mercury or lead) to Potential participants in this health survey were verify the accuracy and the precision of analytical adults (i.e., 2 18 yr of age) who inhabited 400 house- methods.8 Duplicates were run every 10 samples. holds that were selected randomly from the 1,378 Moreover, 10% of participants provided 2 blood sam- households located in the 14 villages of Nunavik. The ples, and blind analyses were performed by the labora- total population at the time of the survey was 7,078. The tory. Matrix-matched (lead) or matrix-free (mercury) cal- 14 villages are scattered along the 1,500-km shoreline ibration was performed daily and reagent blanks were of , , and , and dis- run accordingly. lnterlaboratory quality control was tances between neighboring settlements vary between secured through Qubbec Toxicology Center’s own pro- 80 and 275 km (Fig. 1). We performed systematic sam- gram and that of the Centers for Disease Control and pling after sorting the survey base by household address Prevention (scores of 100% in 1995, 1996, and 1997, to favor a more complete coverage of the territory and respectively). to avoid the selection of next-door neighbors. Further- Detection limits were 0.05 pmol/l for lead and 1 more, so that each village would be represented, we nmol/l for mercury. The laboratory used 4 reference stratified the sample by village, with quasi-proportional specimens of 2.8 pmol/l, 2.1 pmol/l, 1.2 pmol/l, and representation of the number of households in each 0.3 pmol/l to calibrate the analytical method for lead, stratum. and the corresponding coefficients of variation were Data collection for this survey was achieved between 2.4%, 2.9%, 2.3%, and 5.0%, whereas relative biases September 17, 1992, and December 1, 1992, by 6 were +2.0%, +2.2%, -O.8%, and -0.2%, respectively teams, each of which included a nurse and 2-4 Inuit (n= 10). The coefficient of variation was 2%, and the interviewerhnterpreters. Informed consent was - relative bias was -5.5% for the inorganic mercury refer- tained from all individuals who participated in the sur- ence specimen (45 nmol/l; n = 10). The coefficient of vey. After touring each village to inform the population variation was 3.2%, and the relative bias was -6.4% for that a survey would be conducted, the interviewers vis- the total mercury reference specimen (90 nmol/l; n = ited each household and asked to speak with a person 10). Sixty-one subjects had 2 blood samples analyzed 18 yr of age or older to complete the identification chart for lead, and Pearson’s correlation coefficient between of the survey. With the information thereby obtained, a the 2 measurements was .996 (p < .OOl). For mercury, main respondent was designated, who then completed results from 2 separate blood samples were available for a household questionnaire. We also administered an 47 participants, and Pearson’s r was .995 (p < .001). individual questionnaire to each participant in the We used the concentration of omega-3 polyunsatu- household regarding various topics, including life-style rated fatty acids (N-3 PUFAs) in plasma phospholipids habits, diet, and health problems. Later, we asked par- as a biological marker for seafood cons~mption,~~’~and ticipants to attend a clinical visit at the local health cen- they were measured by the Lipid Analytical Laboratory ter, during which a registered nurse conducted anthro- at the University of Guelph (Dr. Bruce J. Holub). Plasma Downloaded by [University of California Santa Cruz] at 22:03 19 December 2012 pometric measurements and collected blood samples. samples were extracted with a chloroform/methanol During a subsequent visit at home, a nurse adminis- mixture, and the resulting lipid extracts were applied tered a 24-hr recall to all participants and a food-fre- onto thin-layer chromatography plates for isolation of quency questionnaire only to women who were neither the phospholipid fraction. Following transmethylation, pregnant nor breast-feeding. Daily intakes of various the fatty acid composition of plasma phospholipids was food items were estimated; participants identified serv- determined by capi Ilary gas-I iquid chromatography. ing sizes, which were presented as plastic models of Statistical analysis. Concentrations of heavy metals in various sizes that represented traditional food servings.’ blood and fatty acids in plasma phospholipids followed laboratory procedures. We used state-of-the-art a log-normal distribution. Therefore, we conducted sta- instrumentation to determine blood lead concentrations tistical tests on log-transformed values, and geometric with graphite furnace atomic absorption spectrometry means were presented in the descriptive display statis- (Perkin Elmer, model ZL 41 00 [Shelton, Connecticut]). tics. Arithmetic means were also displayed; that display Samples were diluted and injected directly into the facilitated comparisons with results obtained in other instrument. We determined blood mercury concen- surveys. We used weighted values and crude sample trations by cold-vapor atomic absorption spectrometry sizes to compute means and confidence intervals. We

352 Archives of Environmental Health Table 1 .-Lead Concentrations in Blood Samples Collected '*OSfrom 492 Inuit Adults Residing in Nunavik, by Gender and Age

Age group Gender (yr) n AM GM 95% CI Range

Male 209 0.54 0.48 0.45, 0.51 0.08-1.67 18-24 40 0.40 0.37 0.32, 0.42 0.16-1.08 25-44 102 0.51 0.46 0.42, 0.51 0.08-1.31 45-75 67 0.65 0.60 0.54, 0.66 0.15-1.67 Female 283 0.47 0.38 0.36, 0.41 0.04-2.28 18-24 67 0.31 0.26 0.23, 0.31 0.04-0.83 25-44 131 0.46 0.39 0.36, 0.43 0.08-2.07 45-75 85 0.61 0.52 0.46, 0.59 0.12-2.28

Notes: AM = arithmetic mean, GM = gec.metric mean, and CI = confidence interval. Blood lead concentration (pmol/l) I Fig. 2. Frequency distribution of lead concentrations in blood sam- Table 2.-Analysis of Variance of Blood lead Concentrations ples of 492 Inuits from Nunavik. (log Values) among 197 , by Age, Smoking Status, and Waterfowl Consumption

Independent variable F df p

Age group (1 8-24 yr, 25-44 yr, 45-75 yr) 24.02 2 < .001 Daily consumption of Canada goose and ducks (gm) 21.36 1 < .001 Smoking status (never-, ever-, current-smokers) 4.04 2 ,019 Model (rZ= .30) 16.46 5 < ,001

Note: df = degrees of freedom. I..- .. A r = .31, p < .001 associations between heavy metals blood levels and M 5 -1.5 various sociodemographic, life-style, or dietary vari- 0 10 20 30 40 ables. These analyses were restricted to women because daily consumption of traditional foods was quantified Concentration of N-3 PUFA (96 plasma phospholipids) for this subgroup only. All personal characteristics that Fig. 3. Correlation between concentrations of lead in blood were associated significantly with heavy metal concen- (log-transformed values) and omega-3 polyunsaturated fatty acids trations in blood (p I .lo) were considered in the (N-3 PUFAs) in plasma phospholipids of 491 Inuits from Nunavik. model. To be retained in the final model, a variable had to show a statistically significant association (i.e., p I .05) with lead or mercury concentrations in blood. designed the weighting scheme to compensate for any Downloaded by [University of California Santa Cruz] at 22:03 19 December 2012 distortion introduced by the sampling procedure with Results regard to the probability of an individual being selected, considering his or her age, sex, and of Of the 382 eligible households, 305 (79.8%) had an residence (Ungava Bay or Hudson Bayhtrait) and the eligible respondent who agreed to fill out the identifi- distribution of these variables in the entire Nunavik cation chart and the household questionnaire; there- population. fore, 766 persons aged 18-74 yr became admissible to We used Student's t test to test for differences in the physical examination and biological analyses. heavy metals concentrations according to dichotomous Among this group of 766 individuals, 518 (67.6%) independent variables (gender or region of residence). agreed to participate in this component of the survey, We performed one-way analyses of variance to test dif- and blood samples were available for 492 (64.4%) of ferences in categorical variables (age, smoking). We the eligible participants. The mean age of these partici- used simple correlation analyses (Pearson's correlation pants was 35.7 yr (95Y0 confidence interval [CI] = 33.6, coefficients) to assess relationships between dependent 37.7) for the 209 men and was 35.0 yr (95% CI = 33.5, variables and continuous independent variables. We 36.4) for the 283 women. performed analyses of variance to assess multivariate Lead was detected in all 492 blood samples, and the

July/August2001 Wol. 56 (No. 4)] 353 Table 3.-Mercury Concentrations in Blood Samples Collected from 492 Inuit Adults Residing in Nunavik, by Gender and Age

Age group Gender (yr) n AM CM 95°/~CI Range

Male 209 113.5 75.0 65.2, 86.2 4.0-482.0 18-24 40 66.5 47.6 36.5, 62.1 6.0-287.0 25-44 102 89.4 62.0 51.6, 74.4 4.0-478.0 45-75 67 178.5 136.6 109.9, 169.7 4.0-482.0 Female 283 106.3 83.2 76.5, 90.5 10.0-560.0 18-24 67 67.4 53.9 45.9, 63.2 10.0-336.0 25-44 131 95.1 77.7 69.5, 86.7 13.0-397.0 45-75 85 156.7 135.2 120.2, 152.1 45.0-560.0

Notes: AM = arithmetic mean, GM = geometric mean, and CI = confidence interval. Blood mercury concentration (pmol/l)

Fig. 4. Frequency distribution of total mercury concentration in blood samples of 492 Inuits from Nunavik. Table 4.-Analysis of Variance of Blood Mercury Concentrations (log Values) among 21 3 Inuit Women, by Age and Sea Mammal Consumption

3.0 Independent variable F df p

Age group (1 8-24 yr, 25-44 yr, 45-75 yr) 28.98 2 < .001 Daily consumption of seal and beluga whale* (gm) 24.53 1 < ,001 Model (r2 = .30) 29.61 3 c .001

Note: df = degrees of freedom. * Delphinapterus leucas.

f = .56, p < ,001 R ... 6 0.40 pmol/l, and 0.43 pmol/l for never-, ever-, and cur- B Y rent-smokers, respectively (p = .005). on 5 One objective of this survey was to investigate the relationship between heavy metal concentrations and Concentration of N-3 PUFA (% plasma phospholipids) dietary habits. In Nunavik, approximately one-half of the Fig. 5. Correlation between concentrations of total mercury in population resides in 7 settlements along the Ungava blood (log-transformedvalues) and omega3 polyunsaturatedfatty Bay, whereas the other half lives in 7 settlements along acids (N-3 PUFAs) in plasma phospholipids of 490 Inuits from the Hudson Bay shore line. People in Hudson Bay com- Nunavik. munities generally eat more seafood-especially sea mammals and waterfowl-whereas the Ungava Bay population relies more on terrestrial game (e.g., cari-

Downloaded by [University of California Santa Cruz] at 22:03 19 December 2012 frequency distribution is shown in Figure 2. The mean bou). The mean lead concentration in the 291 partici- (geometric) concentration for the whole group was 0.42 pants from Hudson Bay settlements was 0.48 pmol/l, pmol/l (arithmetic mean = 0.49 pmol/l), with values compared with 0.35 pmol/l for the 201 people from ranging from 0.04 to 2.28 pmol/l. Mean blood lead Ungava Bay communities (p < .OOl). Furthermore, con- concentrations by gender and age group are shown in centrations of N-3 PUFAs in plasma phospholipids-a Table 1. The mean lead concentration in men (i.e., 0.48 biomarker of seafood consumption-were correlated pmol/l) was 26% higher (p < .05) than that in women weakly with log-transformed blood lead concentrations (0.38 pmol/l). Lead concentration was associated with (r = .31, p < .001) (Fig. 3). We restricted further investi- age, as was evidenced by the increasing trend in mean gation of dietary sources of lead exposure to women concentrations with age groups seen for both sexes (p < inasmuch as men did not complete the 24-hr recall. .OW). Inuits in the 45-75-yr age group had a mean Results of dietary surveys are presented elsewhere.’ blood mercury concentration 1.8 times greater (p < Daily consumption of waterfowl (Canada goose and .001) than that of younger individuals (i.e., 18-24-yr ducks) was the dietary item that was most highly corre- age group). Smoking was also associated with blood lated with blood lead levels (r= .31, p < .001 [n = 2141). lead levels: mean concentrations were 0.30 pmol/l, To further pinpoint the main factors associated with

354 Archives of Environmental Health lead exposure in this population, we performed analy- our results should be representative of the whole Inuit ses of variance, using log-transformed blood lead con- adult population from this region of the Arctic. Further- centrations as the dependent variable and the indepen- more, analyses of mercury and lead in blood samples of dent variables that were found to be associated with participants were the object of strict internal quality, as lead levels in the univariate analysis. A model that well as interlaboratory controls, thereby providing a included age (categorical), smoking (categorical), and high degree of confidence in body burden data docu- waterfowl (goose and duck) consumption (continuous) mented in the present study. as independent variables explained 30% of the varia- The median concentration of lead in blood samples tion of blood lead concentration in Inuit women (p < from Nunavik adults observed in the present study (i.e., .OOl). All 3 independent variables were associated sig- 0.43 pmol/l) was similar to that reported by Milman et nificantly (p< .05) with blood lead levels (Table 2). aL3 for 67 adult Greenlanders in 1987 (i.e., 0.39 pmol/l). All 492 samples were analyzed for total mercury, and The mean (arithmetic) concentration was nearly 5-fold they contained detectable amounts. The frequency dis- greater than the concentrations noted in 53 women from tribution of total mercury concentrations is shown in the Faroe Islands4 and in the general U.S. population Figure 4. The mean (geometric) concentration was 79.6 during the same time period (i.e., 1991-1994).” More- nmol/l (arithmetic mean = 109.3 nmol/l) for the entire over, comparisons with concentrations determined in sample (range = 4-560 nmol/l). Inorganic mercury con- fish eaters from both Canada and the United States dur- centration was determined in a subsample of 18 indi- ing the early 1990s also reveal a relatively high exposure viduals with high total mercury concentrations. The to lead in the Inuit population. Kearney et a1.12 reported mean inorganic concentration was 48.7 nmol/l, which a mean blood lead concentration of 4.3 pg/dl among represented 18% of the total mercury concentration in 155 Lake fish eaters-a level 2.4 times lower this subsample (264.9 nmol/l). The mean concentrations than that noted for the 492 Inuit adults in our study (0.49 of total mercury in blood of Inuit participants, by gen- pmol/l [10.2 pg/dl]). Similarly, Lake Michigan fish eaters der and age group, are shown in Table 3. Concentra- also had a lower mean blood lead concentration (3.8 tions of total mercury in men (75.0 nmol/l) were not dif- pg/dl) than that of the Inuit p0pu1ation.l~ ferent from those in women (83.2 nmol/l). Mean total Blood lead concentrations in the Inuit population mercury concentrations increased with age in both increased with age and were higher among men than sexes (p< .001).The mean value for the older age group among women-a result reported previously by others for (45-74 yr of age) was 2.7-fold greater (p < .001) than various adult populations elsewhere in the ~orld.~,’”’~ that among young adults (18-24 yr of age). Smoking Lead has a very long half-life and, consequently, the high- was not associated with blood mercury concentrations er concentrations observed in older individuals likely (data not shown). reflect its accumulation with age. Differences in eating The mean mercury concentration for Ungava Bay res- habits between age groups may also explain, in part, the idents was lower than that found in the more tradition- association of blood lead levels with age. al Hudson Bay residents (54.6 nmol/l vs. 93.0 nmol/l, Mercury in the form of the organic compound respectively; p < .OOl). The 34 non-Inuit participants methylmercury is a major contaminant of aquatic food had a mean mercury concentration of 18.6 nmol/l, chains. The mean (arithmetic) concentration of total compared with 83.1 nmol/l among Inuits-again mercury noted in blood samples of the 492 Inuits (1 09 reflecting the close link between Inuit life-style and nmol/l) was 2.2-fold lower than that of 240 nmol/l mea- exposure to mercury. N-3 PUFA concentrations were sured in 142 Inuits from Nunavik between 1977 and correlated moderately (r = .56, p < .001) with log-trans- 1982.5 Therefore, mercury exposure appears to have formed blood mercury concentrations (Fig. 5). Among decreased substantially over a 10-yr period in this pop- women who completed the 24-hr dietary recall, daily ulation. Given that the data on levels of mercury in fish consumption of seal and (meat and fat) and sea mammals do not suggest decreasing temporal was the variable most highly correlated with blood mer- trend^,^ the most likely explanation for the decrease in cury concentrations (r = .33, p < .001 [n= 2131). An mercury exposure is a change in the diet (i.e., away Downloaded by [University of California Santa Cruz] at 22:03 19 December 2012 analysis of variance further revealed that both age and from the traditional Inuit diet) of the new generations. sea mammal tissue consumption were associated with Indeed, the mean (geometric) N-3 PUFA concentration mercury concentrations in Inuit women (p< .001), and in plasma phospholipids of Inuit in the 18-24-yr age the resulting model explained 30% (p < .001) of the group documented in the present survey (6.4% [95% CI variance (Table 4). = 5.9, 6.91) was nearly two times lower (p < .001) than that noted among Inuits in the 45-75-yr age group Discussion (12.2% [95% CI = 11.5, 13.0]).” Nevertheless, mercury exposure in this population In this study, we aimed to assess mercury and lead remains higher than those encountered during the exposure in the Inuit population of Nunavik and to early 1990s in populations that reside elsewhere in the identify factors modulating this exposure. Given the Arcti~.~,~For example, the mean concentration ob- large sample size (N = 492), which represented rough- served in women from the Faroe Islands was 60 ly 10% of the entire adult population in Nunavik, the nm~l/l.~Differences are even larger in populations fairly high participation rate (67%), as well as the that live at more southern latitudes. Comparative data weighted sampling scheme used in the present survey, obtained during a 1991 survey that took place in the

July/August2001 [Vol. 56 (No. 4)l 355 Qubbec City region (analyses performed by the same There are no official recommendations in Canada laboratory that we used in the present study) indicated regarding the acceptable concentration of lead in blood that 39 (78%) of the 50 adult participants had total of adults. According to American authorities, blood lead mercury concentrations in blood below the detection concentrations should not exceed 1 or 1.2 limit of the analytical method (10 nmol/l).ls Hence, Furthermore, to prevent the occurrence of adverse effects mean concentrations in the Inuit population were at on fetal development, pregnant women should present least 10-fold greater than those found in this popula- lead concentrations below 0.5 pm01/1.2~,~~Blood lead tion sample from southern Quebec. concentrations observed in the course of this survey were Various lines of evidence indicate that the traditional on average below these values. Only 5% of the adult diet, which comprises several species from the Arctic Nunavik population displayed lead concentrations equal aquatic food chain, is partly responsible for the high body to or greater than 1 pmol/l, and 2% had concentrations burden of lead and mercury in the Inuit population. First, above or equal to 1.2 pmol/l. However, 26% of females blood concentrations were higher among the more tradi- in the 1W4-yr age group had blood lead concentrations tional Inuits who live in communities along the shore of equal to or greater than 0.5 pmol/l. Nearly 3% had lead Hudson Bay, compared with individuals who reside concentrations equal to or exceeding 1 pmol/l. along the Ungava Bay and whose diet includes increas- According to recommendations formulated by the ing amounts of store-bought food. Second, N-3 PUFA World Health Organization (WHO), no more than 5% concentrations in plasma phospholipids, a biomarker of of individuals in a population should display a methyl- marine food consumption, were correlated weakly to mercury concentration that exceeds 1,000 nmo1/1.26 blood lead concentrations (Fig. 3) and were correlated Concentrations of total mercury noted in the present moderately to blood mercury levels (Fig. 5). Third, daily study did not exceed 560 nmol/l. In highly exposed consumption of marine species was correlated to blood individuals, 82% of total mercury was organic mercury concentrations of heavy metals in Inuit women. (likely methylmercury for the most part). WHO issued A multivariate analysis further revealed that con- more stringent recommendations for pregnant women, sumption of waterfowl and cigarette smoking were the stating that not more than 5% of this subgroup should main factors that influenced blood lead levels (Table 2). exhibit methylmercury concentrations above 400 When we included age in addition to these factors, nmo1/1.26 In our survey, no women of child-bearing age nearly one-third of the variance in blood lead levels was exhibited concentrations of this magnitude (maximum explained. Smoking was previously reported as a deter- concentration in the 18-44-yr age group: 397 nmol/l). minant of blood lead concentration^.'^^^^ However, the Nonetheless, even at concentrations below 400 nmol/l, association noted between blood lead concentrations WHO officials warn that subtle neurotoxic effects could and the consumption of waterfowl was not reported be encountered in newborns, and they suggest that previously in the literature. Lead is neither bioaccumu- research efforts should be stepped up in this area. lated nor biomagnified in the marine food web to the Recent data from the Faroe Islands suggest that the neu- same extent as is mercury. Lead has been measured in rologic status of children can be affected by low-level low concentrations (typically 0.05 pg/gm [wet weight1 prenatal exposure to mercury. A modest but statistically and, in almost all cases, below 0.4 pg/gm) in various significant difference in mean cord blood mercury con- species of fish, marine mammals, and ~aterfowl.~High centration (2.1 pg/l 110.5 nmol/ll) was observed be- lead concentrations have been reported occasionally in tween 7-yr-old children who performed suboptimally breast tissue of seabirds when they were killed by lead on a finger-opposition test and 7-yr-old children whose shot.5 Additional studies in which measurements of performance was deemed Considering the lead-stable isotopes ratios were used suggest that the level of exposure to mercury in Inuit women of repro- ingestion of lead shot residues present in birds con- ductive age, this study raises the possibility that subtle tributes significantly to lead exposure in Inuits from neurological deficits could be occurring in Inuit chil- Nunavik. Other sources of lead exposure common in dren as a result of their prenatal exposure to mercury. southern (e.g., lead paint, contaminated drink- There are, however, major differences between the Downloaded by [University of California Santa Cruz] at 22:03 19 December 2012 ing water) are not encountered in Nunavik.21 diet of Faroese and the diet of Inuits, and care must be The best model describing the variation of blood mer- exerted before one concludes that Inuit children are at cury levels (30% of the variance explained) included sea risk. Seafood is a good source of selenium (an essential mammal consumption (seal and beluga whale) and age trace element), which provides some protection against as independent variables (Table 4). Several authors report- methylmercury-induced neurotoxicity in various experi- ed an association between sea mammal meat consump- mental systems.26,28In the Faroe Islands, the median tion and mercury exposure in Arctic population^.^,^,^^ concentration of selenium in umbilical cord blood was Hansen et a1.22 noted a mean blood level of 62.5 pg/1 1.40 pmol/l, and a weak but statistically significant asso- (312 nmol/l) among Inuit Greenlanders who consumed ciation was noted between selenium concentrations and more than 6 seal meals per week, whereas in the group the number of fish meals consumed per week.27Similar that ate 1 seal meal or less per week, the mean value was median cord blood selenium concentrations of 1.51 22.2 pg/l (1 11 nmol/l). Mercury concentrations increase pmol/l and 1.14 pmol/l were documented in Norway with trophic levels in the Arctic marine food web,23 and (Saami) and Russia (Kola peninsula), re~pectively.~How- they reach levels that frequently exceed 0.5 pg/gm (wet ever, the median concentration in Nunavik cord blood weight) in muscle of ringed seal and cetaceans2 samples was 3.67 pmol/15-a level 2.4-fold higher than

356 Archives of Environmental Health that measured in the Faroe Islands. This difference in 6. Kingman A, Albertini T, Brown LJ. Mercury concentrations in urine and whole blood associated with amalgam exposure in a selenium status likely resulted from mattak (beluga US military population. J Dental Res 1998; 77:461-71. whale skin) consumption, which is part of the tradition- 7. Blanchet C, Dewailly t, Ayotte P, et al. Contribution of selected al diet in Nunavik. Mattak generally contains between 4 traditional and market food to Nunavik Inuit women’s diet. Can J and 10 pg/gm of selenium (wet eight)^ and is the most Diet Prac Res 2000; 61 :50-59. important source of selenium among all traditional foods 8. Weber J-P. Accuracy and precision of trace metal determinations in biological fluids-an interlaboratorycomparison program. In: Sub- in Nunavik.’The mean daily selenium intake from tradi- ramanian KS, lyengar GV, Okamoto K (Eds). Biological Trace Ele- tional foods was estimated at 0.1 3 mg in Nunavik Inuit ment Research: Multidisciplinary Perspectives, ACS Symposium women between 18 and 39 yr of age.’ This unusually Series 445. Washington, DC: American Chemical Society, 1991. high selenium intake may afford protection against 9. Brown AJ, Pang E, Roberts DCK. Erythrocyte eicosapentaenoic acid versus docosahexaenoic acid as a marker for fish and fish oil methylmercury-induced neurotoxicity. consumption. Prostaglandins Leukot Essent Fatty Acids 1991; In summary, results from this survey indicate that lead 44~103-06. exposure may constitute a public health problem in 10. Silverman DI, Reis GI, Sacks FM, et al. Usefulness of plasma Nunavik women of reproductive age. Studies are under phospholipid N-3 fatty acid levels in predicting dietary fish way and should better identify sources of lead exposure in intake in patients with coronary artery disease. Am J Cardiol 1990; 662360-62. this population. Preliminary results indicate that replacing 11. Pirkle JL. Exposure of the US population to lead, 1991-1994. lead shots by steel shots could substantially reduce lead Environ Health Perspect 1998; 106:745-50. exposure in this population. With respect to methylmer- 12 Kearney J, Cole DC, Haines D. Report on the Great Lakes anglers cury, a large proportion of women of reproductive age pilot exposure assessment study. Ottawa, Canada: Health Cana- may be at risk as a result of possible fetal toxicity. How- da, 1995. 13. Hovinga ME, Sowers M, Humphrey HEB. Environmental expo- ever, in view of the high selenium intake, which may sure and life-style predictors of lead, cadmium, PCB and DDT counteract methylmercury-induced toxicity, local public levels in Great Lakes fish eaters. Arch Environ Health 1993; health authorities did not recommend reducing seafood 48:98-104. consumption. A cohort study, which was initiated in 14. Ducoffre G, Claeys F, Bruaux P. Lowering time trend of blood lead levels in Belgium since 1978. Environ Res 1990; 51:25-34. 1996, evaluates the potential neurodevelopmental effects 15. Berode M, WietlisbachV, Rickenbach M, et al. Lifestyle and envi- associated with methylmercury exposure in this popula- ronmental factors as determinant of blood lead levels in a Swiss tion, while taking into account exposure to other food- population. Environ Res 1991; 55:l-17. chain contaminants (e.g., polychlorinated biphenyls) and 16. Chu NF, Liou SH, WuTN, et al. Reappraisal of the relation between nutrients (e.g., selenium, omega-3 fatty acids). blood lead concentration and blood pressure among the general population in Taiwan. Occup Environ Med 1999; 56:30-33. 17. SantC Quebec. A Health Profile of the Inuit: Report of the SantP *******at* Quebec Health Survey among the Inuit of Nunavik, 1992. Mon- treal, Quebec, Canada: Ministere de la SantC et des Services This study was made possible through the financial participation of Sociaux, Gouvernement du Quebec, 1998. the Department of Indian and Northern Affairs Canada (Northern 18. Laliberte C, Dewailly t, Gingras S, et al. Mercury contamination Contaminants Program) and Hydro-Quebec. in fishermen of the Lower North Shore of the Gulf of St-Lawrence We are indebted to Mireille Jetteand Louise Guyon, both of whom (Quebec, Canada). In: Vernet JP (Ed). Impact of Heavy Metals on are from Sante Quebec, who coordinated the Sante Quebec Health the Environment. Amsterdam, the Netherlands: Elsevier Press, Survey among the Inuits of Nunavik. Many thanks to Dr. Bruce Holub 1992; pp 15-28. from the University of Guelph who performed omega-3 fatty acids 19. Muldoon SB, Cauley JA, Kuller LH, et al. Lifestyle and sociode- analysis and to Alain Leblanc from the Quebec Toxicology Center for mographic factors as determinants of blood lead levels in elder- mercury and lead analysis. ly women. Am J Epidemiol 1994; 139:599-608. Submitted for publication February 21, 2000; revised; accepted for 20. Weyermann M, Brenner H. Alcohol consumption and smoking publication September 19, 2000. habits as determinants of blood lead levels in a national popula- Requests for reprints should be submitted to Dr. hic Dewailly, tion sample from Germany. Arch Environ Health 1997; 52: Unit6 de Recherche en Sante Publique, CHUQ (Pavillon CHUL), 233-39. 2400 d’Estimauville, Beauport, Quebec, Canada G1 E 7G9. 21. Levesque B, Duchasne JF, Gariepy C, et al. Investigation of sources of lead exposure among Inuit newborns. Sci Total Envi- ron (submitted for publication). ********** 22. Hansen JC, Wulf HC, Kromann N, et al. Human exposure to

Downloaded by [University of California Santa Cruz] at 22:03 19 December 2012 heavy metals in East . I. Mercury. Sci Total Environ 1983; 26:233-43. References 23. Dietz R, Riget F, Johansen P. Lead, cadmium, mercury and sele- nium in Greenland marine animals. Sci Total Environ 1996; 1. Barrie LA, Gregor D, Hargrave B, et al. Arctic contaminants. 186:67-93. Source, occurrence and pathways. Sci Total Environ 1992; 24. Landrigan PJ. Lead in the modern work place. Am J Public Health 122~1-74. 1990; 80:907-08. 2. Muir DCG, Wagemann R, Hargrave BT, et al. Arctic marine 25. National Institute of Occupational Safety and Health. Lead poi- ecosystem contamination. Sci Total Environ 1992; 122:75-134. soning among battery reclamation workers-Alabama, 1991. 3. Milman N, Mathiassen 6, Hansen JC, et al. Blood levels of lead, MMWR Morbid Mortal Wkly Rep 1992; 41 :301-04. cadmium and mercury in a hunter population 26. World Health Organization (WHO). Environmental Health Crite- from the Thule district. Trace Elem Electrolytes 1994; 1 1:3-8. ria 101: Methylmercury. Geneva, Switzerland: WHO, 1990. 4. Grandjean P, Weihe P, Jorgensen PI, et al. Impact of maternal 27. Grandjean P, Weihe P, White RF, et al. Cognitive deficit in 7-year- seafood diet on fetal exposure to mercury, selenium and lead. old children with prenatal exposure to mercury. Neurotoxicol Arch Environ Health 1992; 47:185-95. Teratol 1997; 19:417-28. 5. Arctic Monitoring and Assessment Programme (AMAP). AMAP 28. Whanger PD. Selenium in the treatment of heavy metal poison- Assessment Report: Arctic Pollution Issues. Oslo, Norway: ing and chemical carcinogenesis. J Trace Elem Electrolytes AMAP, 1998. Health Dis 1992; 6:209-21.

July/August 2001 [vol. 56 (No. 4)] 357