ent 372 (2007) 554–561www.elsevier.com/locate/scitotenv

Science of the Total Environm

Skull pathology in East Greenland and Svalbard polar bears(Ursus maritimus) during 1892 to 2002 in relation to

organochlorine pollution

Christian Sonne a,b,⁎, Frank F. Rigét a, Rune Dietz a, Øystein Wiig c,Maja Kirkegaard a, Erik W. Born d

a National Environmental Research Institute, Department of Arctic Environment, Frederiksborgvej 399, DK-4000 Roskilde, Denmarkb Department of Veterinary Basic Sciences, Royal Veterinary and Agricultural University, Bülowsvej 17, DK-1870 Frederiksberg C, Denmark

c Natural History Museum, University of Oslo, PO Box 1172 Blindern, N-0562 Oslo, Norwayd Greenland Institute of Natural Resources, PO Box 570, DK-3900 Nuuk, Greenland, Denmark

Received 14 August 2006; received in revised form 2 October 2006; accepted 13 October 2006Available online 5 December 2006

Abstract

East Greenland and Svalbard polar bears (Ursus maritimus) are heavily polluted with long-range transported organochlorinessuch as PCBs (polychlorinated biphenyls). To investigate the negative health impacts, a time-trend study of skull pathology wasconducted on 269 East Greenland and 241 Svalbard polar bears. The skulls were sampled during 1892–2002 and 1964–1992,respectively. Seven different pathological changes were found: adonti, displacement of teeth, caries, osseous proliferations,exostosis, tooth wear and periodontitis. Only tooth wear and periodontitis was in a prevalence that allowed statistical treatment. Themost severe cases of tooth wear and periodontitis were accompanied by a substantial loss of alveolar bone structure. The prevalenceof tooth wear and periodontitis increased significantly with age ( pb0.001) with incisor wear being more severe than in canines,premolars and molars ( pb0.001). No sex difference was found for tooth wear ( p=0.22) while a significant difference betweensexes was found for periodontitis ( p=0.01) with males having higher prevalence than females (odds ratio of 2.5 for males:females). In East Greenland, the prevalence of tooth wear was significantly higher in polar bears collected in the pre pollutionperiod (b1960) than in bears sampled during polluted periods (1960–1980 and 1981–2002) ( pb0.001). Regarding periodontitis,the prevalence was not significantly different between pre-pollution and pollution periods ( p=0.309). Polar bears from Svalbardhad significantly higher prevalence of tooth wear ( pb0.001) and periodontitis ( p=0.02) than polar bears from East Greenland.The tooth wear and periodontitis odds ratios for Svalbard:East Greenland were 135 and 2.6, respectively. Hence, we found a clearage/sex link and geographical difference but no evidence for an association between skull pathology and exposure toorganochlorines in East Greenland and Svalbard polar bears.© 2006 Elsevier B.V. All rights reserved.

Keywords: East Greenland; Organochlorines; Periodontitis; Polar bear; Svalbard; Tooth wear; Ursus maritimus

⁎ Corresponding author. Wildlife Veterinarian and Toxicologist, Ministry of Environment, National Environmental Research Institute, Departmentof Arctic Environment, Frederiksborgvej 399; PO Box 358, DK-4000 Roskilde, Denmark. Tel.: +45 46 30 19 54; fax: +45 46 30 19 14.

E-mail address: csh@dmu.dk (C. Sonne).

0048-9697/$ - see front matter © 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.scitotenv.2006.10.024

Table 1Description of categories of toothwear andperiodontitis used for analysesof 510 skulls of polar bears from East Greenland and Svalbard collectedduring 1892–2002 and 1964–1992, respectively

Pathology Category

0 1 2 3

Tooth wear No wear Crownflattened1–2 mm

Crown flattened2–5 mm; pulpcavity exposed

Toothworn morethan 2/3

Periodontitis Noperiodontitis

Wideningof toothalveoli

Wideningof toothalveoli;

Wideningof toothalveoli;

Partiallyloss(=“holes”)of alveolarbone tissue

Totalloss ofalveolarbone tissue

555C. Sonne et al. / Science of the Total Environment 372 (2007) 554–561

1. Introduction

Polar bears (Ursus maritimus) from East Greenland,Svalbard and the Kara Sea are highly polluted withorganochlorines (Lie et al., 2003; Dietz et al., 2004;Verreault et al., 2005). Recently, it has been suggested thathigh levels of organochlorines may induce endocrinedisruption (Braathen et al., 2004; Haave et al., 2003;Oskam et al., 2003, 2004; Skaare et al., 2001) andimmunosuppression (Bernhoft et al., 2000; Lie et al.,2004, 2005) in Svalbard polar bears. Furthermore,organochlorines are believed to reduce bone mineraldensity (BMD) and size of sexual organs and to be a co-factor in the development of renal lesions, liver and lymphorgan changes in East Greenland polar bears (Sonne et al.,2004, 2005a, 2006a,b; Kirkegaard et al., 2005).

Laboratory studies have shown that organochlorinesinduce periodontitis in mink (Mustela vison) (Renderet al., 2000a,b, 2001) and in humans PCB seems tointerfere with normal teeth outbreak (Gladen et al., 1990;Rogan, 1979; Miller, 1985). In various studies of wildlifemarine mammals, relationships between exposure toorganochlorines and exostosis, periodontitis, osteoporosisand widening of canine alveoli have been documented(Zakharov and Yablokov, 1990; Bergman et al., 1992;Mortensen et al., 1992; De Guise et al., 1995; Schandorff,1997a,b). In the present study, we examine pathology inpolar bear skulls collected during 1892–2002 in EastGreenland and during 1964–1992 in Svalbard. Theprevalence of tooth wear and periodontitis is evaluatedin relation to age, sex and sampling sites. Based on timetrend organochlorine analyses by Dietz et al. (2004) andAMAP (2004) – and the fact that organochlorines maysuppress the immune system of polar bears – the impactfrom pollution on prevalence of tooth wear and period-ontitis is analyzed over time.

2. Materials and methods

2.1. Sampling, preparation and age estimation

Skulls from East Greenland (69°00′N to 74°00′N)were collected by scientific expeditions and huntersduring 1892–2002 (n=269). They are now in the pos-session of Zoological Museum (Copenhagen, Den-mark). Skulls from Svalbard held at the Museum ofNatural History (University of Oslo, Norway) werecollected during 1964–1992 (n=241). Skulls from1964–1973 were collected by hunters, while skullsfrom 1974–1992 were collected from bears killed in selfdefence, for protection of properties or others reasons.Age determination was carried out by counting the

cementum Growth Layer Groups (GLGs) of the lower I3tooth after decalcification, thin sectioning (14 μm) andstaining (toluidine blue) using the method described byDietz et al. (1991) and Hensel and Sorensen (1980).

2.2. Periodontitis and tooth wear

The skulls were examined for pathology. During theexamination, 7 different pathological changes werefound (tooth wear, periodontitis including alveolar boneloss, adonti, displacement of tooth, caries, osseousproliferations and exostosis). The prevalence of toothbreakage was close to zero and therefore included aswear (it was not possible to separate wear and breakage).However, only tooth wear and periodontitis prevalenceallowed for quantitatively data analyses. The skulls werecategorized into four groups according to the degree oftooth wear and periodontitis (Table 1).

2.3. Statistical analyses

All statistical analyses were performed with the freesoftware R version 2.1.0 and the significance level was setto p=0.05. The number of skulls included in the differentstatistical analyses varied due to missing information insome cases on sex, age and collection year.

Wilcoxon rank sum test with continuity correctionwas used to test for a difference in the prevalence oftooth wear and periodontitis between left and right andamong canines, molars, premolars and incisors. In thefurther statistical analyses, all data were transformedinto only two categories (absence and presence) becausethe frequencies in the categories higher than 0 were toolow for further sub-division to allow for statistical

556 C. Sonne et al. / Science of the Total Environment 372 (2007) 554–561

analyses. Finally, Pearson's Chi-squared test with Yates'continuity correction (2×2 Tables) was applied to testfor differences in the prevalence in tooth wear andperiodontitis between upper and lower tooth row andbetween males and females.

We investigated the effect of age, sex and collectionyear on the prevalence of tooth wear and periodontitisby logistic regression analyses. The model selection wasbased on the Chi-squared test of the deviance reductionwhen successive adding the factors; age, sex andcollection year (Model 1). The data from East Greenlandwere used for the time trend (collection year) analyses,because only these data covered a long time period. Agewas divided into 2-year age-classes [0–2], [2–4], [4–6],[6–8], [8–10], [10–12] and N12 years old and treated asa continuous variable. Initial plots of logit transformedprobabilities of having tooth wear and periodontitisshowed that the assumption of linearity with age usingthese age-classes was reasonable. In case of collectionyear, these were divided into 3 year-classes: b1960,1960–1980 and N1980 assuming that bears livingbefore 1960 were not exposed to organochlorines,bears living during 1960–ca. 1980 were relativelyhighly polluted with organochlorines and that bearsliving ca. 1980–2002 were relatively lower pollutedwith organochlorines based on AMAP (2004) and Dietzet al. (2004) (Fig. 1). The main purpose of this analysiswas to test whether there was an influence fromcollection year (i.e., degree of exposure to organochlor-ines) on the prevalence of tooth wear and periodontitis.The relationship between age, area (East Greenland andSvalbard) and sex, respectively, and prevalence of toothwear and periodontitis was also tested by logisticregression analyses (Model 2) using the same modelselection and age-classes criteria as described above. We

Fig. 1. Mean values of organochlorines (ng/g lipid weight) in subcutaneous a(left) and the procentual decrease during 1990–2001 (right). Modified from

did not test for interactions between factors as theprevalence of tooth wear/periodontitis in age-class [0–2], [2–4], [4–6] and [6–8] were too low which increasedthe risk for significant and spurious interactions. Thishigh risk is often the case in largely exploratory analysesas pointed out by Anderson et al. (2001).

The difference in the prevalence of tooth wear betweenEastGreenland and Svalbard polar bears was tested takingage and sex into account. As the tooth wear prevalence inEast Greenland polar bears collected before 1960 wasfound to be significantly different from polar bearscollected after 1960, the East Greenland data before 1960were omitted from the analyses. Furthermore, no datawere available from Svalbard in the period before 1960 (atotal of 341 skull were included in the analyses).

The estimated effects from the factors on theprevalence of tooth wear and periodontitis were finallyexpressed by odds ratio (OR=exp [parameter estimate])and its standard error (SEOR=exp [SE]). Wald test wasused to test for significance of parameter coefficientsderived from the logistic regression analyses andthereby also the significance of the OR.

3. Results

The initial analyses revealed no difference in wearand periodontitis between left and right side for any tooth(Wilcoxon rank sum tests, pN0.05) or between upperand lower tooth row (Pearson's Chi-squared test,pN0.05). However, the prevalence of tooth wear wassignificantly higher in incisors than in canines, molarsand premolars (Pearson's Chi-squared test, pb0.001),while such a difference was not found for periodontitis(Pearson's Chi-squared test, p=0.20) (Table 2). Further-more, there was no difference between sexes in tooth

dipose tissue from 92 East Greenland polar bears sampled 1999–2001Dietz et al. (2004).

Table 3Results from the logistic regression based on test of deviance reductionfor successive adding of the factors age, sex and collection year (EastGreenland skulls; Model 1) and the factors age, sex and collection area(Model 2) on tooth wear and periodontitis, respectively

Model (n) Parameter df Deviance P (N|Chi|)

Tooth wear 1 (225) Age 1 123.6 b0.001Age+Year 2 18.4 b0.001Age+Year+Sex

1 15.8 0.085

Tooth wear 2 (341) Age 1 64.4 b0.001Age+Area 1 180.6 b0.001Age+Area+Sex

1 1.48 0.220

Periodontitis 1 (223) Age 1 37.9 b0.001Age+Year 2 2.35 0.309Age+Year+Sex

1 2.20 0.138

Periodontitis 2 (415) Age 1 62.7 b0.001Age+Area 1 9.17 0.02Age+Area+Sex

1 7.53 0.01

557C. Sonne et al. / Science of the Total Environment 372 (2007) 554–561

wear and periodontitis in any of the teeth (Pearson's Chi-squared test, all: pN0.05) except for incisor where theprevalence of periodontitis was highest in males(Pearson's Chi-squared test, p=0.002).

Regarding levels of organochlorines in the presentEast Greenland polar bears and reasons for dividing timeperiods into b1960, 1960–1980 and 1981–2002 theseare given in AMAP (2004), Dietz et al. (2004) and Fig. 1.

3.1. Tooth wear

The influence over time on tooth wear was testedon East Greenland skulls, separately, taking sex and ageinto account. Age and collection year were significantwhen added to the model ( pb0.001), while no signif-icant deviance reduction was found when adding sex( p=0.085) (Table 3, Model 1). The odds ratio (OR)influence from age on tooth wear was estimated to 2.51(SEOR=1.12) while the OR for b1960:1981–2002 and1960–1980:1981–2002 was 6.79 (SEOR=1.61) and 1.54(SEOR=1.66), respectively. The odds ratio estimate forpolar bears collected before 1960 (6.79) was significantlydifferent when compared to polar bears collected in theperiod 1981–2002 ( pb0.05). Hence, skulls sampled inthe pre-pollution period (b1960) seemed to exhibit largerprevalence of tooth wear than skulls from the supposedpollution periods. No significant change in prevalencewas detected after 1960 ( pN0.05).

Table 3 shows the deviance reduction when succes-sively adding factors to the logistic regression model.Age and area were highly significant ( pb0.001) whenadded to the model, while no significant reduction of thedeviance was found when adding sex ( p=0.22) (Table 3,

Table 2Prevalence of periodontitis and tooth wear [n (%)] in various teeth ofpolar bears from East Greenland (n=269) and Svalbard (n=241)collected during 1892–2002 and 1964–1992, respectively

Incisors Canines Premolars Molars

Tooth wear0 215 (44) 385 (81) 486 (97) 486 (97)1 274 (56) 91 (19) 14 (3) 14 (3)Sum 489 (100) 476 (100) 500 (100) 500 (100)

Periodontitis0 476 (96) 480 (97) 467 (94) 475 (95)1 22 (4) 17 (3) 32 (6) 24 (5)Sum 498 (100) 497 (100) 499 (100) 499 (100)

0: No tooth wear/periodontitis; 1: presence of tooth wear/periodontitis.Tooth wear increased significantly in the order incisorNcaninesNpremolars/molars ( pb0.001). No such difference was found forperiodontitis.Due to missing teeth in some cases, the number of investigated skullsvaries between groups.

Model 2). Based on the logistic regression modelincluding the significant factors (age and area), theodds ratio for increasing in age-classes was 3.26(SEOR=1.15). The odds ratios for polar bears inSvalbard were estimated to 135 (SEOR=1.73) whencompared to polar bears in East Greenland. The observedand estimated probabilities of having tooth wear in thetwo areas are shown in Fig. 2.

3.2. Periodontitis

To some extent, periodontitis seemed to be secondaryto toothwear (Table 2). First, the influence from collectionyear on periodontitis was tested on East Greenland skulls,

Fig. 2. The observed proportions of polar bears having wear in EastGreenland (squares) and at Svalbard (circles) including predictedprobabilities derived from logistic regression models (East Green-land=solid line; Svalbard=broken line).

Fig. 3. The observed proportions of male (left) and female (right) polar bears having periodontitis in East Greenland (squares) and at Svalbard (circles)including predicted probabilities derived from logistic regression models (East Greenland=solid line; Svalbard=broken line).

558 C. Sonne et al. / Science of the Total Environment 372 (2007) 554–561

separately, taking sex and age into account (Model 1).Agewas significant ( pb0.001)when added to themodel,while no significant reduction of the deviance was foundwhen adding collection year ( p=0.309) and sex( p=0.138). The odds ratio for increasing in age-classeswas 2.01 (SEOR=1.15).

The influence from age, sex and sampling area on theprevalence of periodontitis was tested. As no EastGreenland periodontitis time trend was found, all skullswere included in the analysis. Age ( pb0.001), area( p=0.02) and sex (p=0.01) were all significant whenadded to the model (Table 3, Model 2). The odds ratiofor increasing in age-classes was estimated to 2.0(SEOR=1.10) and the odds ratio was 2.5 (SEOR=1.41)for male polar bears compared to female polar bears.Polar bears from Svalbard had an odds ratio of 2.6(SEOR=1.42) compared to polar bears from EastGreenland. The observed and estimated probabilitiesof having periodontitis divided on area for males andfemales, respectively, are given in Fig. 3.

4. Discussion

4.1. Tooth wear and periodontitis

Tooth wear was most frequently found in the incisorsand canines. Periodontitis was found in only 3–6% ofthe skulls. Tooth wear in carnivores and secondarily tothis periodontitis, have been associated with age andbehavioural patterns (Van Valkenburgh, 1988a,b; Stir-ling, 1969; Fenton et al., 1998; Patterson et al., 2003;Persson et al., 2004), whereas immunosuppressions,heavy metals and organochlorines/fluorides are sus-pected to have influence as well (Render et al., 2000a,b,2001; Zakharov and Yablokov, 1990; Bergman et al.,1992; Mortensen et al., 1992; Schandorff, 1997a,b;deGuise et al., 1995).

Van Valkenburgh (1996) studied incidence of toothbreakage among large predatory mammals and foundthat the most comment broken teeth were the car-nassials (i.e., 4th premolar above and 1st premolarbelow) while the incisors were least commonly broken.The same general pattern was found by Fenton et al.(1998) studying tooth breakage in bats and carnivores.These results are different from our findings in polarbears where the incisors were the most worn. Due to thelow prevalence of breakage (close to zero) we com-bined breakage into the categorization wear degreewhich make the data less comparable. Van Valkenburgh(1996) studied feeding behaviour in large Africancarnivores and found that incisors were used exten-sively to pull muscles from the carcass. We have oftenobserved how polar bears use their incisors whentearing slices of blubber or mussels off seal carcassesand believe that this feeding behaviour explain the highprevalence of wear in incisors in relation to the othertooth types.

To some extent, the periodontitis observed in thepresent study was a secondary inflammatory reactiondue to tooth wear opening of the pulp cavitysubsequently increasing the risk for infection andabscessing (Stirling, 1969; Clarke and Hirsch, 1990).Potentially, abscessing may increase the risk ofmortality as discussed below.

4.2. Age and sex patterns

Van Valkenburgh (1988a,b) found that tooth break-age increased with age in some species of largecarnivores. In one of four species examined, theincidence of broken teeth differed between sexes.Female lions (Panthera leo) had more broken teeththan males. This could be due to differences in huntingtactics or just be result of biased ages in the samples

559C. Sonne et al. / Science of the Total Environment 372 (2007) 554–561

(Van Valkenburgh, 1988a,b). Fenton et al. (1998) sug-gested that inter sexual aggression could account fordifferences in damaged teeth between sexes. However,in the present study the incidence of broken canines inpolar bears was not influenced by sex. Derocher et al. (inpreparation) found that canine breakage in adult malepolar bears at Svalbard was higher than in adult femalesand that the breakage in males increased particularlywith age after 14 years. The pattern was believed to beassociated with intense male to male competition foraccess to females as suggested by Ramsay and Stirling(1986).

Usually, there is no difference in the frequency ofperiodontitis between men and women when normal-ising for environmental factors and social behaviour(Persson et al., 2004). In the present study we did notfind a difference between sexes which probably can beascribed to the fact that we also did not find a sex-relateddifference in tooth wear.

4.3. Spatial patterns

Geographical differences in tooth wear are knownfrom e.g. different ungulate species (Sokolov et al.,1996) and such changes are ascribed to a difference infood composition and genetics. Investigations of largeAfrican carnivores showed that bone consumption wasthe main factor in developing tooth wear (VanValkenburgh, 1996). That kind of tooth wear in canineshas been described as micro wear which is the preformof canine breakage (Anyonge, 1996). Therefore, thereason for the obvious difference in tooth wear andperiodontitis between Svalbard and East Greenlandcould be due to a difference in bone consumption (foodcompetition) from population density and food avail-ability as described in large carnivorous by Anyonge(1996).

4.4. Tooth wear and periodontitis as mortality factor?

Stirling (1969) investigated tooth wear as a mortalityfactor in the Antarctic Weddell seal (Leptonychotesweddelli). Stirling (1969) found that the amount of toothwear and necrosis increased markedly after ca. 9 yearsof age and that this may contributed markedly to themortality in adult individuals of this seal species. Wecannot evaluate the potential mortality from the presentlesions in polar bears, but it cannot be excluded that themost severe lesions could contribute to increasedmortality from starvation due to abscessing and necrosissubsequently avoiding large prey species or evenanorexia.

4.5. Tooth wear, periodontitis and organochlorines

Based on time trend studies and chemical analyses,environmental levels of various organochlorines havebeen linked to periodontitis and osteopenia in Baltic andKattegat seals (Zakharov and Yablokov, 1990; Bergmanet al., 1992; Mortensen et al., 1992; Schandorff, 1997a,b;Lind et al., 2003) and polar bears (Sonne et al., 2004).Similar pathology have in controlled laboratory studies ofmink (M. vison) been induced by CB-126 and 2,3,7,8-TCDD (Render et al., 2000a,b, 2001). Furthermore,endocrine disrupting organochlorines such as PCB havebeen associated with abnormal teeth outbreak in children(Gladen et al., 1990; Rogan, 1979; Miller, 1985) andenvironmental stress (e.g., organochlorines) can also leadto reduction of tooth calcium (Siegel et al., 1992).

Dietz et al. (2004, in press) and Sonne et al. (2005b)analyzed the levels of organochlorines (PCBs andpesticides), brominated flame retardants (PBDEs) andmono-ortho PCBs, PCDDs, PCDFs (and their corre-sponding WHO-TEQ) in subcutaneous adipose tissuefrom the East Greenland polar bears collected during1999–2001. Sonne (2004) showed that these organo-chlorine and TEQs levels were significantly lowercompared to the studies from seals and exposed minkwhich was also the case for Svalbard (AMAP, 2004).The fact that adipose levels of organochlorines weresignificantly lower compared to the seal and minkstudies may explain the lack of a contaminant-relatedtime trend in polar bear skull periodontitis and theperiodontitis complex may also not be a sensible pa-rameter for exposure to organochlorines below certainthreshold levels. The fact that East Greenland skullssamples before the onset of pollution (b1960) had aneven higher frequency of tooth wear and periodontitisthan skulls sampled in the polluted period (N1960)indicates that organochlorines do not have a significantimpact on the development of skull pathology in polarbears from East Greenland and Svalbard. The reasonscould be a difference in other environmental factorssuch as access to food, climate changes and genetics(bottleneck) as discussed by Sonne et al. (2005c) in theirskull study of fluctuating asymmetry in East Greenlandpolar bears during the same period of 1892–2002.

5. Conclusions

Our results clearly showed that the prevalence oftooth wear and periodontitis increased with age and thatthe frequency was highest in males and most severe forincisors. In East Greenland, the prevalence of tooth wearwas significantly higher in polar bears collected before

560 C. Sonne et al. / Science of the Total Environment 372 (2007) 554–561

1960 while this was not the case for periodontitis. EastGreenland polar bears sampled after 1959 had signifi-cantly lower prevalence of tooth wear and periodontitisthan polar bears from Svalbard. Hence, we found a clearage/sex link and geographical difference but no evi-dence for an association between skull pathology andexposure to organochlorines in East Greenland andSvalbard polar bears.

Acknowledgements

Danish Cooperation for Environment in the Arctic,The Commission for Scientific Research in Greenlandand Lundbeck Foundation are acknowledged for finan-cial support. Jonas Brønlund, who gathered the samplesthrough local hunters, Hanne Tuborg Sandell and BirgerSandell who helped with local contacts to hunters andfinally Jeppe Møhl, Mogens Andersen, Abdi Hedayatand Hans Baagøe at the Zoological Museum ofCopenhagen, who provided the historical skull collec-tion for analyses and facilities for maceration and pre-paration of recently collected skulls. Steen Andersen(www.huntersmagazine.com)made the instruction videofor the polar bear hunters, for which Lars Åby alsosupplied footage.

References

AMAP. AMAP assessment 2002: persistent organic pollutants in theArctic. Oslo: ArcticMonitoring andAssessment Programme; 2004.

Anderson DR, Burnham KP, Gould WR, Cherry S. Concern aboutfinding effects that re acutally spurious. Wildl Soc Bull 2001;29(1):311–6.

Anyonge W. Microwear on canines and killing behaviour in largecarnivores: saber function in smilodon fatalis. J Mammal 1996;77(4):1059–67.

Bergman A, OlssonM, Reiland S. Skull-bone lesions in the Baltic greyseal Halichoerus grypus. Ambio 1992;21:517–9.

Bernhoft A, Skaare JU, Wiig Ø, Derocher AE, Larsen HJS. Possibleimmunotoxic effects of organochlorines in polar bears (Ursusmaritimus) at Svalbard. J Toxicol Environ Health A 2000;57(7):561–74.

Braathen M, Derocher AE, Wiig Ø, Sørmo EG, Lie E, Skaare JU, et al.Relationships between PCBs and thyroid hormones and retinol infemale and male polar bears. Environ Health Perspect 2004;112:826–33.

ClarkeNG,HirschRS. Periodontitis and angular alveolar lesions: a criticaldistinction. Oral Surg Oral Med Oral Pathol 1990;69(5):564–71.

De Guise S, Lagace A, Beland P, Girard C, Higgins R. Nonneoplasticlesions in beluga whales (Delphinapterus leucas) and other marinemammals from the St. Lawrence estuary. J Comp Pathol 1995;112(3):257–71.

Derocher AE, Andersen M, Wiig Ø. in preparation. Mating ecology ofpolar bears at Svalbard. Behavioral Ecol.

Dietz R, Heide-Jørgensen MP, Härkönen T, Teilmann J, Valentin N.Age determination of European harbour seal (Phoca vitulina L.).Sarsia 1991;76:17–21.

Dietz R, Riget FF, Sonne C, Letcher RJ, Born EW, Muir DCG.Seasonal and temporal trends in polychlorinated biphenyls andorganochlorine pesticides in East Greenland polar bears (Ursusmaritimus), 1990–2001. Sci Total Environ 2004;331:107–24.

Dietz R, Riget FF, Sonne C, Muir DCG, Backus S, Born EW, et al. Ageand Seasonal Variability of Polybrominated Diphenyl Ethers inFree-Ranging East Greenland Polar Bears (Ursus maritimus).Environ Pollut in press.

Fenton MB, Waterman JM, Roth JD, Lopez E, Fienberg SE. Toothbreakage and diet: a comparison of bats and carnivores. J ZoolLond 1998;246:83–8.

Gladen BC, Taylor JS, Wu YC, Ragan NB, Rogan WJ, Hsu CC.Dermatological findings in children exposed transplacentally toheat-degraded polychlorinated biphenyls in Taiwan. Br J Dermatol1990;122(6):799–808.

Haave M, Ropstad E, Derocher AE, Lie E, Dahl E, Wiig Ø, et al.Polychlorinated biphenyls and reproductive hormones in femalepolar bears at Svalbard. EnvironHealth Perspect 2003;111(4):431–6.

Hensel RJ, Sorensen FE. Age determination of live polar bears.International Conf Bear Res Manage 1980;4:93-100.

Kirkegaard M, Sonne C, Leifsson PS, Dietz R, Born EW, Letcher RJ,et al. Histology of selected immunological organs in polar bear(Ursus maritimus) from East Greenland in relation to levels oforganohalogens. Sci Total Environ 2005;341:119–32.

Lie E, Bernhoft A, Riget FF, Belikov SE, Boltunov AN, Derocher AE,et al. Geographical distribution of organochlorine pesticides(OCPs) in polar bears (Ursus maritimus) in the Norwegian andRussian Arctic. Sci Total Environ 2003;306:159–70.

Lie E, Larsen HJS, Larsen S, Johansen GM, Derocher AE, Lunn NJ,et al. Does high organochlorine (OC) exposure impair the resis-tance to infection in polar bears (Ursus maritimus)? Part I. Effectof OCs on the humoral immunity? J Toxicol Environ Health 2004;A(67):555–82.

Lie E, LarsenHJS, Larsen S, JohansenGM,DerocherAE, LunnNJ, et al.Does high organochlorine (OC) exposure impair the resistance toinfection in polar bears (Ursus maritimus)? Part II. Effect of OCs onmitogen and antigen induced lymphocyte proliferation? J ToxicolEnviron Health 2005;68:457–84.

Lind PM, Bergman A, Olsson M, Örberg J. Bone mineral density inmale Baltic grey seal. Ambio 2003;32(6):385–8.

Miller RW. Congenital PCB poisoning: a re-evaluation. Environ HealthPerspect 1985;60:211–4.

Mortensen PÅ, Bergman A, Bignert A, Hansen HJ, Härkönen T,Olsson M. Prevalence of skull lesions in harbour seals (phocavitulina) in Swedish and Danish museum collections: 1835–1988.Ambio 1992;21:520–4.

Oskam IC, Ropstad E, Dahl E, Lie E, Derocher AE, Wiig Ø, et al.Organochlorines affect the major androgenic hormone, testoster-one, in male polar bears (Ursus maritimus) at Svalbard. J ToxicolEnviron Health A 2003;66(22):2119–39.

Oskam IC, Ropstad E, Lie E, Derocher AE, Wiig Ø, Dahl E, et al.Organochlorines affect the steroid hormone cortisol in polar bears(Ursus maritimus) at Svalbard, Norway. J Toxicol Environ HealthA 2004;67:959–77.

PattersonBD,Neiburger EJ, Kasiki SM. Tooth breakage and dental diseaseas causes of carnivore–human conflicts. JMammal 2003;84(1):190–6.

Persson GR, Persson RE, Hollender LG, Kiyak. The impact ofethnicity, gender and marital status on periodontal and systemichealth of older subjects in the Trials to Enhance Elders' Teetch andOral Health (TEETH). J Periodontol 2004;75(6):817–23.

Ramsay MA, Stirling I. On the mating system of polar bears. Can JZool 1986;64:2142–51.

561C. Sonne et al. / Science of the Total Environment 372 (2007) 554–561

Render JA, Hochstein JR, Aulerich RJ, Bursian SJ. Proliferation ofperiodontal squamous epithelium in mink fed 2,3,7,8-tetrachlor-odibenzo-p-dioxin (TCDD). Vet Hum Toxicol 2000a;42(2):86.

Render JA, Aulerich RJ, Bursian SJ, Nachreiner RF. Proliferation ofmaxillary and mandibular periodontal squamous cells in mink fed3,3′4,4′,5-pentachlorobiphenyl (PCB 126). J Vet Diagn Invest2000b;12(5):477–9.

Render JA, Bursian SJ, Rosenstein DS, Aulerich RJ. Squamousepithelial proliferation in the jaws of mink fed diets containing 3,3′4,4′,5-pentachlorobiphenyl (PCB 126) or 2,3,7,8-tetrachlorodi-benzo-p-dioxin (TCDD). Vet Hum Toxicol 2001;43(1):22–6.

Rogan WJ. PCBs and cola-colored babies: Japan, 1968, and Taiwan,1979. Teratology 1979;26:259–61.

Schandorff S. Developmental stability and skull lesions in the harbourseal (Phoca vitulina) in the 19th and 20th centuries. Ann Zool Fenn1997a;34:151–66.

Schandorff S. Developmental stability and the harbour seal epizooticin 1998. Ann Zool Fenn 1997b;34:167–75.

Siegel MI, Mooney MP, Taylor AB. Dental and skeletal reduction as aconsequence of environmental stress. Acta Zool Fenn 1992;191:145–9.

Skaare JU, Bernhoft A, Wiig Ø, Norum KR, Haug E, Eide DM, et al.Relationship between plasma levels of organochlorines, retinol andthyroid hormones from polar bears (Ursus maritimus) at Svalbard.J Toxicol Environ Health A 2001;62:227–41.

Sokolov AA, Gruzdev AR, Pronyaev AV. Patterns of cheek toothwearing in reindeer (Rangifer tarandus) from different popula-tions. Zool 1996;75(10):1573–80 [Zool Zh 75 (10): 1573–1580].

Sonne C. 2004. Organohalogen concentrations and a gross andhistologic assessment of multiple organ systems in East Greenlandpolar bears (Ursus maritimus). PhD thesis. National Environmen-tal Research Institute, Denmark. 200 pp.

Sonne C, Dietz R, Born EW, Riget FF, Kirkegaard M, Hyldstrup L, et al.Is bone mineral composition disrupted by organochlorines in EastGreenland polar bears (Ursus maritimus)? Environ Health Perspect2004;112(17):1711–6.

SonneC, Dietz R, Leifsson PS, Born EW,KirkegaardM,Riget FF, et al.Do organohalogen contaminants contribute to liver histopathologyin East Greenland polar bears (Ursus maritimus)? Environ HealthPerspect 2005a;113(11):1569–74.

SonneC,Leifsson PS,DietzR,BornEW,Letcher RJ,KirkegaardM, et al.Enlarged clitoris in wild polar bears (Ursus maritimus) can bemisdiagnosed as pseudohermaphroditism. Sci Total Environ2005b;337:45–58.

Sonne C, Riget FF, Dietz R, Kirkegaard M, Born EW, Letcher RJ, et al.Trends in fluctuating asymmetry in East Greenland polar bears(Ursus maritimus) from 1892 to 2002 in relation to organohalogenpollution. Sci Total Environ 2005c;341:81–96.

SonneC,DietzR,Leifsson PS,BornEW,KirkegaardM,LetcherRJ, et al.Are organohalogen contaminants a co-factor in the development ofrenal lesions in East Greenland polar bears (Ursus maritimus)?Environ Toxicol Chem 2006a;25(6):1551–7.

Sonne C, Leifsson PL, Dietz R, Born EW, Letcher RJ, Hyldstrup L, et al.Xenoendocrine pollutants may reduce size of sexual organs in EastGreenland polar bears (Ursus maritimus). Environ Sci Technol2006b;40:5668–74.

Stirling I. Tooth wear as a mortality factor in the Weddell seal(Leptonychotes weddelli). J Mammal 1969;50:559–65.

Verreault J, Muir DCG, Norstrom RJ, Stirling I, Fisk AT, GabrielsenGW, et al. Chlorinated hydrocarbon contaminants and metabolitesin polar bears (Ursus maritimus) from Svalbard, East Greenland,Alaska and the Canadian arctic during 1999−2002. Sci TotalEnviron 2005;351–352:369–90.

Van Valkenburgh B. Incidence of tooth breakage among large,predatory mammals. Am Nat 1988a;131(2):291–302.

Van Valkenburgh B. Dental micro wear and dietary differences in lingand fossil carnivores. Am Zool 1988b;28(4):A175.

Van Valkenburgh B. Feeding behaviour in free-ranging, large Africancarnivores. J Mammal 1996;77(1):240–54.

Zakharov MZ, Yablokov AV. Skull asymmetry in the baltic grey seal:effects of environmental pollution. Ambio 1990;19(5):266–9.