- Email: [email protected]
Associations between vitamin D status and atherosclerosis among Inuit in Greenland
Accepted Manuscript Associations between vitamin D status and atherosclerosis among Inuit in Greenland Camilla U. Gjødesen, Marit E. Jørgensen, Peter Bjerregaar, Inger K. Dahl-Petersen, Christina V.L. Larsen, Martin Noël, Mads Melbye, Arieh S. Cohen, Marika Lundqvist, David M. Hougaard, Jørn W. Helge, Nina O. Nielsen PII:
S0021-9150(17)31413-2
DOI:
10.1016/j.atherosclerosis.2017.11.028
Reference:
ATH 15282
To appear in:
Atherosclerosis
Received Date: 25 August 2017 Revised Date:
10 November 2017
Accepted Date: 28 November 2017
Please cite this article as: Gjødesen CU, Jørgensen ME, Bjerregaar P, Dahl-Petersen IK, Larsen CVL, Noël M, Melbye M, Cohen AS, Lundqvist M, Hougaard DM, Helge JøW, Nielsen NO, Associations between vitamin D status and atherosclerosis among Inuit in Greenland, Atherosclerosis (2017), doi: 10.1016/j.atherosclerosis.2017.11.028. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Associations between vitamin D status and atherosclerosis among Inuit in Greenland
Camilla U. Gjødesena, Marit E. Jørgensena,b,c, Peter Bjerregaara,c, Inger K. Dahl-Petersena, Christina
RI PT
V. L. Larsena, Martin Noël d, Mads Melbyee,f,g, Arieh S. Cohenh, Marika Lundqvisth, David M. Hougaardh, Jørn W. Helgei, Nina O. Nielsena,j
b
c
Steno Diabetes Centre Copenhagen, Gentofte, Denmark.
SC
National Institute of Public Health, University of Southern Denmark, Copenhagen, Denmark.
M AN U
a
Greeenland Centre for Health Research, University of Greenland, Nuuk, Greenland.
d
Université Laval, Quebec, Canada.
e
Department of Epidemiology Research, Statens Serum Institut, Copenhagen, Denmark.
f
Department of Clinical Medicine, University of Copenhagen, Denmark
g
h
i
TE D
Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA Department for Congenital Disorders, Statens Serum Institut, Copenhagen, Denmark.
Center of Healthy Aging, Department of Biomedical Science, University of Copenhagen,
Centre for Nutrition and Rehabilitation, University College Absalon, Næstved, Denmark (current
address).
AC C
j
EP
Copenhagen, Denmark
*Correspondence to: [email protected] (C. U. Gjødesen)
1
ACCEPTED MANUSCRIPT Abstract Background and aims: Low levels of vitamin D are suspected to be a risk factor for cardiovascular disease and atherosclerosis. The aim of this study was to assess the prevalence of subclinical
RI PT
atherosclerosis among Inuit in Greenland, and to evaluate the association with vitamin D status. We hypothesized that low vitamin D status could be associated with higher carotid intima-media thickness (IMT) as a marker of atherosclerosis.
SC
Methods: 756 adults from the Inuit Health in Transition (IHIT) study carried out in Greenland in the period 2005-2010 were included. A blood sample donated in 1987 was available for a sub-sample
M AN U
of 102 individuals. Serum 25(OH)D3 from the IHIT study and the 1987 survey was used as a measure of vitamin D status. IMT measurements were conducted by ultrasound scanning. The prevalence of atherosclerosis was estimated, and the association between serum 25(OH)D3 and IMT measurements was examined by linear regression.
TE D
Results: The overall prevalence of subclinical atherosclerosis was 20.1% (n=152). The linear regression analyses indicated a weak positive association between serum 25(OH)D3 level and IMT measurements from the IHIT study, though not statistically significant after adjustment for potential
EP
confounders (β=0.35% per 10 nmol/l 25(OH)D3, p=0.06). Linear regression analyses of the association between serum 25(OH)D3 level in the 1987 survey and IMT measurements also
AC C
indicated a positive, though not statistically significant, association after adjustment (β=0.07% per 10 nmol/l 25(OH)D3, p=0.86). Conclusions: Our findings did not support the hypothesis of an association between low vitamin D levels and risk of atherosclerosis.
2
ACCEPTED MANUSCRIPT 1. Introduction Vitamin D is an essential fat-soluble vitamin with great importance for human health. There is an international consensus that vitamin D is important for maintenance of calcium homeostasis, bone
RI PT
health and prevention of fractures (1). Recent studies have identified low vitamin D as a new potential risk factor for atherosclerosis and cardiovascular disease in general (2, 3). Carotid
atherosclerosis can be used as a clinical marker of systemic atherosclerosis due to early onset,
SC
location and easy detection by ultrasound (4). Additionally, the presence of carotid plaque and measurement of the carotid intima-media thickness (IMT) can be used as a surrogate marker of
M AN U
subclinical atherosclerosis (3).
In the Arctic, dermal production of vitamin D is low due to the large solar zenith angle (SZA), long periods with low sun exposure and the continuous need for protective outdoor clothing (5, 6). Additionally, the dark skin pigmentation in the Inuit population may reduce vitamin D
TE D
synthesis in the skin (7). Previously, the high intake of traditional vitamin D rich diet containing fatty fish, seal and whale, probably compensated for the poor skin synthesis of vitamin D (5-7). During the last 50-60 years, the traditional diet of the Inuit has largely been replaced by imported
EP
foods (8), along with an observed lower vitamin D status in Arctic populations (6, 9, 10). A new study confirms that vitamin D deficiency is increasing in Greenland (10).
AC C
Formerly, it was a common notion that ischaemic heart disease is rare among Inuit in Greenland due to their marine diet with a high content of omega-3-fatty acids. However, the scientific evidence for this is weak since it is not based on systematic studies, but rather uncertain mortality statistics (11, 12). So far, only two studies have reported on atherosclerosis among Inuit in Greenland (13, 14). One of these reported that indigenous Greenlanders develop atherosclerosis to the same extent as Europeans (13), and the other concluded that the extent of advanced atherosclerosis in Greenlanders was similar to that reported among Alaska natives but less than
3
ACCEPTED MANUSCRIPT reported for Alaska non-natives (14). To our knowledge, no studies have investigated the association between vitamin D status and atherosclerosis in Arctic populations. In this study, we hypothesized that low vitamin D levels were associated with higher
RI PT
risk of atherosclerosis among Inuit. We used serum 25-hydroxyvitamin D3 (25(OH)D3) concentration as a measure of vitamin D status and included blood samples collected in 2005-2010 and in 1987. Carotid intima-media thickness (IMT) was used as a surrogate marker of subclinical
SC
atherosclerosis. Subclinical atherosclerosis is hereafter referred to as atherosclerosis.
M AN U
2. Materials and methods 2.1 Study population
The Inuit Health in Transition (IHIT) study, a general health study in Greenland and Canada that aimed to investigate health and diseases, lifestyle and life conditions in the Arctic regions, was
TE D
carried out between 2005 and 2010 (15). One purpose of the IHIT study was to assess risk factors of cardiovascular disease to obtain a better understanding of the health effects of the transition from a traditional to a modern and industrialized lifestyle in the Arctic regions (15). Participants in the
EP
IHIT study in Greenland were selected through a stratified random sample of the adult population (+ 18 years) born in Greenland or Denmark. Greenland was divided into 10 regions based on
AC C
geography and community size. From each region, one or more towns and villages were selected. The towns were selected as representative of the region in terms of living conditions (15). Villages were chosen randomly in the regions, and all adults living there were invited to participate (15). The total study sample consisted of 3108 Inuit (9.1% of the total adult population in 2005). The IHIT study population represented all age groups (+18 years), all geographical areas and all community sizes of Greenland. A valid serum 25(OH)D3 measurement was obtained from 2,877 individuals (93%) (10). 1,043 of the participants had an IMT measurement conducted (only a random selection
4
ACCEPTED MANUSCRIPT of individuals aged 40 years or above were examined), and 793 of these measurements were valid (individual mean IMT was based on a minimum of 6 segment measurements as described in section 2.4). Individuals with valid serum 25(OH)D3 and valid IMT measurements were included, which
RI PT
lead to the final number of 756 participants in the present study. Thus, due to invalid (less than 6 segment measurements available) data on IMT (n=250), missing data on IMT (due to random selection and cut-off at ≥ 40 years, n=1914), missing data on serum 25(OH)D3 (n=37) and ethnicity
SC
other than Greenlander (n=151), a total of 2352 individuals in the IHIT study were excluded from this study.
M AN U
In 1987, due to a syphilis epidemic in western and southern Greenland, the health authorities in Greenland conducted a population-based serological examination. Among those that participated in the survey in 1987, we identified 102 who had a blood sample available for measurement of serum 25(OH)D3 and also had an IMT measurement conducted as part of the IHIT
TE D
study in 2005-2010. The above described measurements allowed for an assessment of a crosssectional association between 25(OH)D3 status and IMT in 2005-2010 and an association between
EP
25(OH)D3 status in 1987 and development of IMT later in life (2005-2010) (fig. 1).
2.2 Collection of blood samples
AC C
Blood samples from 1987 were drawn from May to June, whereas samples from 2005-2010 were collected during all months of the year in the period from April 2005 to October 2010, except for July, November and December. Blood samples were handled and stored as described elsewhere (10).
2.3 Measurement of vitamin D
5
ACCEPTED MANUSCRIPT Serum 25(OH)D3 was used as a measure of vitamin D status. Samples were analysed after storage at -80 °C for 3-7 years (IHIT samples) and 26 years (1987-samples), respectively (10). Analyses were conducted by liquid chromatography-tandem mass spectrometry (LC-MSMS) using the
RI PT
“MSMS vitamin D” kit from Perkin Elmer (Waltham, MA) as described previously (16). This method measured both serum 25(OH)D2 and 25(OH)D3, but since concentrations of 25(OH)D2
2.4 Measurement of IMT by carotid ultrasonography
SC
were negligible, only 25(OH)D3 was included in the analysis.
M AN U
Measurements of the intimal to medial arterial wall thickness of the carotid arteries were performed using a high resolution B-mode ultrasound portable device (Model LogiqBook, GE Medical System, Milwaukee, WI, USA) with a linear 4–10 MHz probe (Model 10LB-Rs, GE Medical System, Milwaukee, WI, USA) by two experienced sonographers who conducted the ultrasound
TE D
scanning according to a standard protocol (17, 18). A transverse scan was performed prior to a longitudinal scan in order to detect any hemodynamically relevant stenosis or presence of significant plaque defined by a focal echogenic structure protruded into the lumen vessel. Segments
EP
of 1 cm of the near and far walls of the Common carotid (1 cm below the flow divider), the Bifurcation and the Internal carotid (1 cm distal to the flow divider) arteries were digitally recorded
AC C
and used for evaluation of the IMT. The segments had to be free of atherosclerotic plaque. All images were saved in digital imaging and communication in medicine (DICOM) database format. A single reader blinded to the clinical data performed off-line measurements. Analyses of 3 nonconsecutive frames were completed using a dedicated semi-automatic edge detection software (Carotid analyser for research v5.5.6, Medical Imaging Application LLC, Coraville, IA, USA). When the analyser determined a region of interest (ROI) of 1 cm in length, the program automatically located the two interfaces defining walls (lumen-intima and media adventitia) using
6
ACCEPTED MANUSCRIPT algorithms based on grey level density and tissue recognition and drew two parallel lines. In case of visual discrepancies, the analyser had the possibility of editing the lines by fine-tuning the detection algorithms. IMT was measured every 10 µm of the length of the ROI (n = 100 successive local
RI PT
measures) for each frame. The measurements of each frame of the Common, Bifurcation and Internal carotid segments of the near and far wall on carotid artery in both right and left side of the neck were averaged and referenced as a mean IMT (17, 19). This study will refer to a mean IMT
SC
from each participant calculated as an average of available measured segments. However, to be a valid IMT, a minimum of 6 segment measurements were included in the mean IMT with at least
M AN U
one segment measurement from each frame (Common, Bifurcation and Internal) of both right and left carotid arteries. Atherosclerosis was categorized with the following cut-off values according to Riccioni et al. (20): No evidence of atherosclerosis (IMT < 0.8 mm), some atherosclerosis (0.8 mm
2.5 Potential confounders
TE D
≤ IMT ≤ 1.2 mm) and extensive atherosclerosis (IMT > 1.2 mm).
Information on age, sex and smoking was collected by interviews as part of the IHIT study. Hepatic insulin resistance (HOMA-IR) was calculated using the formula: fasting plasma glucose (mmol/L)
EP
x fasting plasma insulin (mU/L)/22.5 (21). Systolic BP was measured three times with an automatic
AC C
measuring device (Kivex UA-779) with an appropriately sized cuff. The last two measurements were averaged and used for the analysis. Physical activity (PAEE) was recorded by the International Physical Activity Questionnaire (long version IPAQ), modified and validated for the living conditions of the Greenlandic population (22). Anthropometric measurements were used to calculate body mass index (BMI) as: weight in kg/(height in m2). HDL and triglycerides were measured in serum by automated enzymatic colorimetic methods. LDL cholesterol was calculated with the Friedewald formula. Methyl mercury was measured by inductively coupled mass spectromy (ICP-MS) at the Centre de Toxicologie, Institut National de Santé Public, Québec,
7
ACCEPTED MANUSCRIPT Canada, as described previously (23). Information about genetic admixture was estimated from blood tests where the admixture proportion of Inuit and European ancestry was quantified. The analyses have accurately quantified the extent of European and Inuit ancestry in the participants
RI PT
included in the present study (24).
2.6 Ethics
SC
Written informed consent was obtained from each participant included in the IHIT study, the study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki and t has been
M AN U
reviewed and approved by the Ethical Review Committee for Greenland.
2.7 Statistical methods
All statistical analyses were performed using the statistical software STATA version 14 (StataCorp.
TE D
2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP). The associations between serum 25(OH)D3 and IMT as a measure of atherosclerosis were explored by linear regression analysis with serum 25(OH)D3 as the explanatory variable and IMT measures as the
EP
outcome variable. IMT measures did not follow a normal distribution and were therefore logtransformed to achieve normal distribution before entered into the linear regression model. β-
AC C
estimates from linear regression models were back-transformed and presented as a percentage increase per 10 nmol/L increase in serum 25(OH)D3. The cross sectional association between IMT and serum 25(OH)D3 measured in the IHIT study was assessed in three linear regression models and adjusted for an increasing number of confounders. Model 1: Age, sex and genetic admixture. Model 2: As model 1 + BMI, smoking, HDL, triglycerides, systolic BP, HOMA-IR and PAEE. Model 3: As model 2 + methyl mercury. The association between serum 25(OH)D3 measured in the 1987 survey and IMT measured in IHIT was assessed in two linear regression models: Model 1:
8
ACCEPTED MANUSCRIPT Unadjusted. Model 2: Adjusted for age, sex and genetic admixture. In all models, a quadratic term of age was included in order to take into account the non-linear effect of age on the association
3. Results 3.1 Characteristics of the study population
RI PT
between serum 25(OH)D3 and IMT.
SC
In the IHIT study, the median age was 46 years (interquartile range, IQR, 40-56 years), the median serum 25(OH)D3 concentration was 54 nmol/L (IQR 35-75 nmol/L) and the median IMT was 0.69
M AN U
mm (IQR 0.61-0.77mm). A sub-group of 485 participants from the IHIT study had previously participated in the 1987 survey, and among them, data on IMT and serum 25(OH)D3 was available from 102 individuals. Of the 102 individuals in the 1987 sample, 52 (51%) were men. The median age was 31 years (IQR 23-39) in 1987, median serum 25(OH)D3 concentration was 72 nmol/L
TE D
(IQR 49-109 nmol/L) and median IMT was 0.70 mm (IQR 0.62-0.80 mm). Table 1 gives characteristics of the study population in 1987 and 2005-2010 by sex. The prevalence of smokers was 69 % among men and 72% among women. The degree of genetic admixture was
EP
82% for both men and women. Median PAEE was higher among men than women (56 kj/kg/day vs.
AC C
46 kj/kg/day). IFG was more prevalent among men than women (23% vs. 16%).
3.2 Prevalence of atherosclerosis using IMT as a surrogate marker Based on IMT measurements, three (0.4%) of the 756 individuals had extensive atherosclerosis and 149 (19.7%) had some atherosclerosis. Thus, according to IMT, the overall prevalence of atherosclerosis was 20.1% in the study population.
3.3 Associations between IMT and serum 25(OH)D3 in the IHIT sample
9
ACCEPTED MANUSCRIPT Table 2 shows the association between serum 25(OH)D3 level and IMT measurements in the IHIT sample. In model 1, the estimate indicated no association between serum 25(OH)D3 level and IMT measurements after adjustment for age, sex and genetic admixture. After additional adjustment for
RI PT
BMI, smoking, HDL, triglycerides, systolic BP, HOMA-IR and PAEE (model 2), the estimate indicated a positive association between serum 25(OH)D3 and IMT. Thus, increasing serum 25(OH)D3, expressed as 10 nmol/L, was associated with increasing IMT (0.41%, 95% confidence
SC
interval (CI): 0.07% – 0.75%, p=0.02). However, after additional adjustment for methyl mercury
M AN U
(model 3), the association disappeared (0.35%, CI: -0.01% – 0.71%, p=0.06).
3.4 Associations between IMT and serum 25(OH)D3 in the 1987 sample A linear unadjusted regression analysis (n=102) of the association between serum 25(OH)D3 level in the 1987 sample and IMT measurements in the IHIT sample showed a positive association between serum 25(OH)D3 level and IMT (1.1%, CI: 0.4% -1.9%, p=0.003). However, after
4. Discussion
EP
p=0.86).
TE D
adjustment for age, sex and admixture, the association disappeared (0.07%, CI: -0.68% – 0.82%,
The overall prevalence of atherosclerosis was 20.1% according to the applied definitions (IMT ≥
AC C
0.8). A Canadian study among Inuit in Nunavik found that 11,5 % of the study population had an IMT greater than 1 mm (19). That study referred to a higher risk of cardiovascular disease if IMT was greater than 1 mm, however, not linking IMT directly to atherosclerosis. The difference in cutoff between the Nunavik study and our study may at least in part explain the difference in the prevalence of atherosclerosis based on IMT. A study among Alaskan Eskimos reported that carotid atherosclerosis assessed by ultrasound was more commonly detected in that population compared to American populations (25), but the specific prevalence was not reported. A Canadian study
10
ACCEPTED MANUSCRIPT examining a multi-ethnic low-risk population found an overall prevalence of atherosclerosis of 23% (26). In that study, subclinical atherosclerosis was defined according to measures of mean maximum cIMT≥75th percentile and is therefore difficult to compare directly to the present study.
RI PT
The Manfredonia Study, which examined an Italian population and used the same cut-off as used in the present study, found an atherosclerosis prevalence of 54% (27). However, a higher mean age in the Manfredonia Study may explain the higher prevalence. Most results are obviously not in line
SC
with the notion that Inuit may have lower risk of cardiovascular disease due to their intake of traditional Arctic diet with a high content of omega 3-fatty acids. As indicated above, there are
M AN U
challenges by comparing the prevalence of atherosclerosis among different populations due to methodological inconsistencies in the current literature.
The present study is, to our knowledge, the first to report on the association between serum 25(OH)D3 and atherosclerosis in an Inuit population. In contrast to our hypothesis, we found
TE D
a weak positive association between 25(OH)D3 and IMT measures, though not statistically significant after adjustment for age, sex, genetic admixture, BMI, smoking, HDL, triglycerides, systolic BP, HOMA-IR, PAEE and methyl mercury. Previous studies in other populations
EP
evaluating the association between 25(OH)D levels and atherosclerosis assessed by IMT have shown conflicting results. In The Multi-Ethnic Study of Atherosclerosis including 3251 low risk
AC C
participants, 25(OH)D were not significantly associated with IMT (28). Furthermore, in a multicenter study examining the association between serum 25(OH)D concentrations and IMT measures in 3430 middle-aged and elderly subjects with high CVD risk found no independent relationship between 25(OH)D and segment-specific or composite IMT measures (29). In contrast, several studies have found that low 25(OH)D levels were associated with higher IMT (3, 30-32) and low vitamin D seems also associated with adverse CV outcome (33). One study showed a positive association between serum 25(OH)D and IMT among patients with sufficient 25(OH)D levels (≥50
11
ACCEPTED MANUSCRIPT nmol/L) (34). Overall, most studies conducted in non-Inuit populations found that low vitamin D was associated with atherosclerosis. The present study indicates, however, that 25(OH)D3 plays a minor, if any, role in the development of atherosclerosis in an Arctic population. It is reasonable to
RI PT
consider if the inverse association found between 25(OH)D and atherosclerosis in some studies may be explained by residual confounding, e.g. predisposing cardiovascular risk factors, genetic factors, sun exposure and dietary factors. The lack of an association is in agreement with findings from
SC
Mendelian Randomization studies and a meta-analysis of eight RCTs suggesting that associations between 25(OH)D and atherosclerosis in epidemiological studies result from reverse causation or
M AN U
residual confounding (35, 36).
In the present study, methyl mercury was perceived as a classic potential confounder due to its association with both vitamin D and IMT. Environmental pollutants and methyl mercury have been shown to be associated with cardiovascular outcomes (23, 37-39) and IMT (40, 41). Of
TE D
particular interest to Arctic populations, some studies suggest that Inuit who eat a traditional diet with a high intake of fatty fish, seal and whale, and therefore have a high intake of vitamin D (10), are at the same time highly exposed to environmental pollutants including methyl mercury (23, 42).
EP
Environmental pollutants and mercury bioaccumulate in marine mammals and fish and are transferred to humans through the diet. Mercury is present in the blood and organs of Greenlanders
AC C
in concentrations well above international guidelines (39, 43). In the present study, the positive association between serum 25(OH)D3 and IMT was eliminated after adjustment for methyl mercury. A possible explanation could be that the association was “carried” by methyl mercury. Other environmental contaminants and heavy metals specific for the arctic area, such as polychlorinated biphenyls and cadmium (44), are reasonable to consider as factors with possible impacts on the findings.
12
ACCEPTED MANUSCRIPT The Inuit population is highly genetically differentiated from other populations, because of being isolated and affected by bottlenecks in population size (45), and diet and climate adaptation in the Arctic has favored genetic variants associated with severe insulin resistance and
RI PT
increased susceptibility to cardiometabolic disease (46, 47). In the present study, the population structure was taken into account by adjusting the association between 25(OH)D3 and IMT for genetic admixture, with the aim of minimizing the risk of confounding by this factor.
SC
There was a positive association between serum 25(OH)D3 level in the 1987 sample and IMT measurements from the IHIT sample, though not statistically significant after adjustment
M AN U
for age, sex and genetic admixture. Thus, the longitudinal findings confirmed the absence of an association in the cross-sectional analysis. According to the literature review, only the Multi-Ethnic Study of Atherosclerosis, with 3251 low-risk participants, examining serum 25(OH)D in relation to IMT, had a longitudinal design (10 years follow-up) (28), and the finding of no association is
TE D
consistent with the findings in the present study. However, a limitation in the present study is the lack of information on IMT from the 1987 sample. Furthermore, adjustments for confounders were limited to age, sex and genetic admixture, which may have resulted in residual confounding.
EP
Additionally, the sample size of 102 participants in this part of the study may not provide sufficient power to detect a possible association. However, the finding of a similar association pattern with
AC C
IMT when using 25(OH)D3 measurements from the 1987 sample and from the IHIT sample indicates some robustness of the findings. Strengths of the study include the use of serum 25(OH)D3 measurements from a large
population based and geographically representative study of Inuit in Greenland, who also had a carotid ultrasonography. Vitamin D status (serum 25(OH)D3) was measured in blood samples and not estimated from questionnaires, which minimizes the risk of misclassification. Another major strength is that detailed ultrasound measurements of IMT from both right and left carotid arteries
13
ACCEPTED MANUSCRIPT were performed and 6-12 measurements were included. The American Heart Association has endorsed the use of IMT in cardiovascular risk assessment (17) and B-mode ultrasound IMT measurements are seen as the best current method to provide objective information of the degree of
RI PT
subclinical atherosclerosis (26, 48). Thus, both exposure and outcome are objectively measured and not based on self-reports. Furthermore, methyl mercury, HDL, triglycerides, genetic admixture and HOMA-IR levels were obtained from measurements in serum samples, and detailed information on
SC
physical activity was available from the IPAQ validated questionnaire and used for adjustment of the analyses.
M AN U
In conclusion, this study is the first to examine the association between vitamin D and atherosclerosis among Inuit in Greenland. No longitudinal or cross sectional association was found between serum 25(OH)D3 and IMT in the IHIT sample and the 1987 sample, and based on these results it seems reasonable to conclude that vitamin D has no impact on the development of
TE D
atherosclerosis among Inuit in Greenland. Genetic status is likely overruling a possible role of vitamin D in the development of atherosclerosis in this population. Furthermore, the contribution of lifestyle related risk factors such as smoking, physical inactivity and a diet containing high levels of
AC C
vitamin D status.
EP
saturated fat and sugar might affect the development of atherosclerosis substantially more than
Conflict of interest
The authors declared they do not have anything to disclose regarding conflict of interest with respect to this manuscript.
14
ACCEPTED MANUSCRIPT Financial support The study was supported by the Government of Greenland’s health department (2011-047012/2012074239), Aase og Ejnar Danielsen’s Foundation (10-000444), Dagmar Marshalls Foundation,
RI PT
Karen Elise Jensens Foundation and Nuna Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
SC
Author contributions
C.U.G., N.O.N., M.E.J and J.H. conceived and designed the study. C.U.G. analyzed the data and
M AN U
wrote the manuscript. N.O.N. assisted with data analysis. P.B., I.D.P, C.V.L.L, M.E.J. and M.N. collected the IHIT data. M.M., A.S.C., M.L. and D.M.H. contributed with study materials and analysis tools. All authors reviewed the manuscript and contributed with critical comments to the content. All authors approved the submitted version of the manuscript. C.U.G. is the guarantor of
TE D
this work, and, as such, had full access to all data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
EP
Acknowledgements
The authors thank the study participants for participating in the study. Mrs. Ingelise Olesen is
AC C
thanked for her efforts in the practical organization of the IHIT study and participation in data collection, and the interviewers are thanked for their valuable assistance in data collection. The late Eric Dewailly is thanked for his contribution to the collection of the ultrasound measurements. Colleagues from the laboratories of the Steno Diabetes Centre Copenhagen, Gentofte, Denmark, in particular Mrs. Maja Lis Dybdahl Halkjær, are thanked for indispensable assistance with handling of the blood samples.
15
SC
RI PT
ACCEPTED MANUSCRIPT
References
M AN U
1. Palacios C, Gonzalez L. Is vitamin D deficiency a major global public health problem? J Steroid Biochem Mol Biol. 2014; 144 Pt A:138-45. 2. Kalkan GY, Gür M, Koyunsever NY, Şeker T, Gözükara MY, et al. Serum 25Hydroxyvitamin D Level and Aortic Intima-Media Thickness in Patients Without Clinical Manifestation of Atherosclerotic Cardiovascular Disease. J Clin Lab Anal. 2015; 29(4):30511.
TE D
3. Hao Y, Ma X, Luo Y, Ni J, Dou J, et al. Additional role of serum 25-hydroxyvitamin D3 levels in atherosclerosis in Chinese middle-aged and elderly men. Clin Exp Pharmacol Physiol. 2014; 41(3):174-9. 4. Grobbee DE, Bots ML. Carotid artery intima-media thickness as an indicator of generalized atherosclerosis. J. Intern Med. 1994; 236:567-73.
EP
5. Webb AR. Who, what, where and when-influences on cutaneous vitamin D synthesis. Prog Biophys Mol Biol. 2006; 92(1):17-25.
AC C
6. Andersen S, Jacobsen A, Laurberg P. Vitamin D status in North Greenland is influenced by diet and season: indicators of dermal 25-hydroxy vitamin D production north of the Arctic Circle. Br J Nutr. 2012; 110:50-57. doi:10.1017/S0007114512004709. 7. Fares JEH, Weiler HA. Implications of the nutrition transition for vitamin D intake and status in aboriginal groups in the Canadian Arctic. Nutr Rev. 2016; 74 (9): 571-583. doi: https://doi.org/10.1093/nutrit/nuw020 8. Jeppesen C, Bjerregaard P. Consumption of traditional food and adherence to nutrition recommendations in Greenland. Scand J Public Health. 2012;40(5):475-81. 9. Rejnmark L, Jørgensen ME, Pedersen MB, Hansen JC, Heickendorff L, et al. Vitamin D insufficiency in Greenlanders on a westernized fare: ethnic differences in calcitropic hormones between Greenlanders and Danes. Calcif Tissue Int. 2004; 74(3):255-63.
16
ACCEPTED MANUSCRIPT
10. Nielsen NO, Jørgensen ME, Friis H, Melbye M, Soborg B, et al. Decrease in vitamin D status in the Greenlandic adult population from 1987-2010. PLoS One. 2014;9(12):e112949. 11. Jørgensen ME, Bjerregaard P, Kjaergaard JJ, Borch-Johnsen K. High prevalence of markers of coronary heart disease among Greenland Inuit. Atherosclerosis. 2008; 196(2):772-8.
RI PT
12. Bjerregaard P, Young TK, Hegele RA. Low incidence of cardiovascular disease among the Inuit - what is the evidence? Atherosclerosis. 2003; 166 (2):351-7. 13. Hansen JP, Hancke S, Møller-Petersen J. Atherosclerosis in Greenland: an ultrasonographic investigation. Arctic Med Res. 1991; Suppl:400-3.
SC
14. Pedersen HS, Mulvad G, Newman WP, Boudreau DA. Atherosclerosis in coronary arteries and aorta among Greenlanders: an autopsy study. Atherosclerosis. 2003; 170(1):93-103.
M AN U
15. Bjerregaard P. Inuit Health in Transition - Greenland survey 2005-2010. Population sample and survey methods. 2nd edition. 2011;SIF Writings on Greenland vol. 19:1-26. Link: http://www.si-folkesundhed.dk/upload/inuit_health_in_transition_greenland_methods _5_2nd_revision.pdf 16. Nielsen NO, Strøm M, Boyd HA, Andersen EW, Wohlfahrt J, Lundquist M, et al. Vitamin D status during pregnancy and the risk of subsequent postpartum depression: A case-control study. PLOS ONE. 2013; 8: e80686. doi:10.1371/journal.pone.0080686 PMID: 24312237.
TE D
17. Oliveira GM, De Lorenzo A, de Souza EG, Santos B. Association between Carotid Intimamedia Thickness and Early-onset Coronary Artery Disease: Does the site of Sonographic Evaluation Matter? Heart Lung Circ. 2015; 24(9):932-5.
EP
18. Stein JH, Korcarz CE, Hurst RT, Lonn E, Kendall CB, et al. Use of carotid ultrasound to identify subclinical vascular disease and evaluate cardiovascular disease risk: a consensus statement from the American Society of Echocardiography Carotid Intima-Media Thickness Task Force. Endorsed by the Society for Vascular Medicine. J AM Soc Echocardiogr. 2008; 21(2):93-111.
AC C
19. Noël M, Dewailly E, Chateau-Degat ML, Counil É, Laouan-Sidi EA, et al. Cardiovascular risk factors and subclinical atherosclerosis among Nunavik Inuit. Atherosclerosis. 2012; 221(2):558-64. 20. Riccioni G, D'Orazio N, Palumbo N, Bucciarelli V, Ilio E, et al. Relationship between plasma antioxidant concentrations and carotid intima-media thickness: the Asymptomatic Carotid Atherosclerotic Disease In Manfredonia Study. Eur J Cardiovasc Prev Rehabil. 2009;16(3):351-7. 21. Johnson-Down L, Labonte ME, Martin ID, Tsuji LJ, Nieboer E, et al. Quality of diet is associated with insulin resistance in the Cree (Eeyouch) indigenous population of northern Québec. Nutr Metab Cardiovasc Dis. 2015; 25(1):85-92.
17
ACCEPTED MANUSCRIPT 22. Dahl-Petersen IK. Physical activity pattern and its relation to glucose metabolism in Greenland [PhD thesis]. Centre for Health Research in Greenland: University of Southern Denmark; 2013.
RI PT
23. Nielsen AB, Davidsen M, Bjerregaard P. The association between blood pressure and whole blood methylmercury in a cross-sectional study among Inuit in Greenland. Environ Health Perspect. 2012; 11: 44. doi: 10.1186/1476-1069X-1111-1144. 24. Moltke I, Fumagalli M, Korneliussen TS, Crawford JE, Bjerregaard P, et al. Uncovering the genetic history of the present-day Greenlandic population. Am J Hum Genet. 2015; 96(1):54-69.
SC
25. Cutchins A, Roman MJ, Devereux RB, Ebbesson SO, Umans JG, et al. Prevalence and correlates of subclinical atherosclerosis in Alaska Eskimos: the GOCADAN study. Stroke. 2008; 39(11):3079-82.
M AN U
26. Grewal J, Anand S, Islam S, Lonn E, Investigators SaS-A. Prevalence and predictors of subclinical atherosclerosis among asymptomatic "low risk" individuals in a multiethnic population. Atherosclerosis. 2008; 197(1):435-42. 27. Riccioni G, D'Orazio N, Palumbo N, Bucciarelli V, Ilio E, et al. Relationship between plasma antioxidant concentrations and carotid intima-media thickness: the Asymptomatic Carotid Atherosclerotic Disease In Manfredonia Study. Eur J Cardiovasc Prev Rehabil. 2009; 16(3):351-7.
TE D
28. Blondon M, Sachs M, Hoofnagle AN, Ix JH, Michos ED, et al. 25-Hydroxyvitamin D and parathyroid hormone are not associated with carotid intima-media thickness or plaque in the multi-ethnic study of atherosclerosis. Arterioscler Thromb Vasc Biol. 2013; 33(11):2639-45.
EP
29. Deleskog A, Piksasova O, Silveira A, Gertow K, Baldassarre D, et al. Serum 25hydroxyvitamin D concentration in subclinical carotid atherosclerosis. Arterioscler Thromb Vasc Biol. 2013; 33(11):2633-8.
AC C
30. Carrelli AL, Walker MD, Lowe H, McMahon DJ, Rundek T, et al. Vitamin D deficiency is associated with subclinical carotid atherosclerosis: the Northern Manhattan study. Stroke. 2011; 42(8):2240-5. 31. van de Luijtgaarden KM, Voûte MT, Hoeks SE, Bakker EJ, Chonchol M, et al. Vitamin D deficiency may be an independent risk factor for arterial disease. Eur J Vasc Endovasc Surg. 2012; 44(3):301-6. 32. Reis JP, von Mühlen D, Michos ED, Miller ER, Appel LJ, et al. Serum vitamin D, parathyroid hormone levels, and carotid atherosclerosis. Atherosclerosis. 2009; 207(2):58590. 33. Tuńón J, Cristóbal C, Tarín N, Aceńa A, González-Casaus ML, et al. Coexistence of Low Vitamin D and High Fibroblast Growth Factor-23 Plasma Levels Predicts an Adverse Outcome in Patients with Coronary Artery Disease. 2014; 9, e95402.
18
ACCEPTED MANUSCRIPT
34. van Dijk SC, Sohl E, Oudshoorn C, Enneman AW, Ham AC, et al. Non-linear associations between serum 25-OH vitamin D and indices of arterial stiffness and arteriosclerosis in an older population. Age Ageing. 2015; 44(1):136-42.
RI PT
35. Manousaki D, Mokry LE, Ross S, Goltzman D, Richards JB. Mendelian Randomization Studies Do Not Support a Role for Vitamin D in Coronary Artery Disease. Circ Cardiovasc Genet. 2016 Aug; 9(4):349-56. doi: 10.1161/CIRCGENETICS.116.001396. Epub 2016 Jul 14.
SC
36. Wang L, Manson JE, Song Y, Sesso HD. Systematic review: Vitamin D and calcium supplementation in prevention of cardiovascular events. Ann Intern Med. 2010 Mar 2; 152(5):315-23. doi: 10.7326/0003-4819-152-5-201003020-00010.
M AN U
37. Salonen JT, Seppänen K, Lakka TA, Salonen R, Kaplan GA. Mercury accumulation and accelerated progression of carotid atherosclerosis: a population-based prospective 4-year follow-up study in men in eastern Finland. Atherosclerosis. 2000; 148(2):265-73. 38. Valera B, Jørgensen ME, Jeppesen C, Bjerregaard P. Exposure to persistent organic pollutants and risk of hypertension among Inuit from Greenland. Environ Res. 2013; 122:65-73.
TE D
39. Bjerregaard P, Mulvad G. The best of two worlds: how the Greenland Board of Nutrition has handled conflicting evidence about diet and health. Int J Circumpolar Health. 2012; 71:18588.
EP
40. Skoczynska A, Poreba R, Steinmentz-Beck A, Martynowicz H, Affelska-Jercha A, et al. The dependence between urinary mercury concentration and carotid arterial intima-media thickness in workers occupationally exposed to mercury vapour. Int J Occup Med Environ Health. 2009; 22:135-142.
AC C
41. Salonen. JT, Seppänen K, Lakka TA, Salonen R, Kaplan GA. Mercury accumulation and accelerated progression of carotid atherosclerosis: a population-based prospective 4-year follow-up study in men in eastern Finland. Atherosclerosis. 2000; 148:265-273. 42. Jeppesen C, Jørgensen ME, Bjerregaard P. Assessment of consumption of marine food in Greenland by a food frequency questionnaire and biomarkers. Int J Circumpolar Health. 2012; 71:18361. 43. Hansen JC, Van Oostdam J. AMAP Assessment 2009: Human Health in the Arctic. 2009. 44. Fagerberg B, Kjelldahl J, Sallsten G, Barregard L, Forsgard N, et al. Cadmium exposure as measured in blood in relation to macrophage density in symptomatic atherosclerotic plaques from human carotid artery. 2016; 249: 209-214.
19
ACCEPTED MANUSCRIPT 45. Tvermosegaard M, Dahl-Petersen IK, Nielsen NO, Bjerregaard P, Jørgensen ME. Cardiovascular Disease Susceptibility and Resistance in Circumpolar Inuit Populations. Can J Cardiol. 2015; 31(9):1116-23.
RI PT
46. Fumagalli M, Moltke I, Grarup N, Racimo F, Bjerregaard P, et al. Greenlandic Inuit show genetic signatures of diet and climate adaptation. Science. 2015 Sep 18; 349(6254):1343-7. doi: 10.1126/science.aab2319. 47. Moltke I, Grarup N, Jørgensen ME, Bjerregaard P, Treebak JT, et al. A common Greenlandic TBC1D4 variant confers muscle insulin resistance and type 2 diabetes. Nature. 2014 Aug 14; 512(7513):190-3. doi: 10.1038/nature13425. Epub 2014 Jun 18.
AC C
EP
TE D
M AN U
SC
48. van den Oord SC, Sijbrands EJ, ten Kate GL, van Klaveren D, van Domburg RT, et al. Carotid intima-media thickness for cardiovascular risk assessment: systematic review and meta-analysis. Atherosclerosis. 2013;228(1):1-11.
20
RI PT
ACCEPTED MANUSCRIPT
Attending 1987 survey (n=7609) Excluded Not attending IHIT study (n=7124) Attending both 1987 survey and IHIT study 2005-2010 (n=485) Excluded No excess material(n=155)
Excluded No IMT measurements (n=228)
TE D
Blood sample available for serum 25(OH)D3 measurement (n=330)
Attending IHIT study 2005-2010 (n=3108)
M AN U
Flow chart of the IHIT study (2005-2010) and 1987 survey.
SC
Fig. 1. Selection of study participants.
Attending IHIT study 2005-2010 with valid IMT measurement (n=793)
Excluded No IMT measurements conducted (n=1914) Invalid IMT measurements (n=250) Ethnicity other than Greenlander (n=151)
Excluded No data on serum 25(OH)D3 (n=37)
IHIT sample with valid IMT and serum 25(OH)D3 measurements (n=756)
AC C
EP
1987 sample Attending both IHIT study and 1987 survey (and have IMT measurements) (n=102)
Study population IHIT study (n=756) 1987 sample (n=102); a sub-sample of the IHIT study
21
ACCEPTED MANUSCRIPT
Table 1. Characteristics of the study populations in 1987 (n=102) and 2005-2010 (n=756) according to sex.
n
Men
n
Women
Age (years)
52
33 (25-40)
50
27 (21-36)
25(OH)D3 (nmol/L)
52
77 (48-120)
IMT (mm)
52
0.72 (0.65-0.86)
n
Men
Age (years)
376
47 (40-57)
380
45 (40-59)
BMI (kg/m2)
373
24 (22-28)
379
25 (22-29)
Serum 25(OH)D3 (nmol/L)
376
55 (35-78)
380
53 (34-72)
IMT (mm)
376
0.71 (0.62-0.81)
380
0.67 (0.60-0.74)
(kJ/kg/day)
375
56 (23-104)
380
46 (24-73)
HDL (mmol/L)
376
1.6 (1.3-2.0)
380
1.7 (1.4-2.1)
376
3.5 (2.9-4.1)
380
3.5 (2.9-4.2)
376
0.9 (0.7-1.3)
380
1.0 (0.8-1.3)
376
5.8 (5.4-6.2)
380
5.6 (5.2-6)
2h plasma glucose (mmol/L)
369
5.2 (4.1-6.5)
376
5.7 (4.7-6.8)
Fasting insulin (pmol/L)
376
28 (20-44.5)
380
40.5 (27-57)
369
63 (32-142)
375
139 (77-231)
376
7.3 (4.9-12.1)
380
10.3 (6.6-14.2)
Systolic BP (mmHg)
376
130 (121-145)
380
122 (111-136)
Diastolic BP (mmHg)
376
79 (70-89)
380
75 (68-83)
Mercury (µg/L)
376
24 (10-49)
379
19 (9-37)
Genetic admixture (%)
374
82 (63-99)
379
82(66-99)
Smokers (%)
375
69
380
72
IFG (%)
314
23
332
16
LDL (mmol/L) Triglycerides (mmol/L)
HOMA-IR
AC C
2h insulin (pmol/L)
EP
Fasting plasma glucose (mmol/L)
TE D
Physical activity energy expenditure
50
66 (50-92)
50
0.66 (0.59-0.77)
n
Women
SC
M AN U
IHIT sample (2005-2010)
RI PT
1987 sample
22
ACCEPTED MANUSCRIPT IGT (%)
252
4
294
8
T2DM (%)
376
11
380
6
AC C
EP
TE D
M AN U
SC
RI PT
Median with interquartile range and proportions are presented.
23
ACCEPTED MANUSCRIPT Table 2. Linear regression analysis of the cross sectional associations between serum 25(OH)D3 level and IMT measurements in the IHIT sample (n=753).
Model 1
Model 2
Adjusted for age, sex and genetic
Model 3
Additionally adjusted for BMI, smoking, HDL, triglycerides, systolic
mercury
RI PT
admixture
Additionally adjusted for methyl
BP, HOMA-IR and physical activity n
Change (95% CI)
p-
n
value
0.51% (-0.39-1.43)
p=0.27
n
Change (95% CI)
665
0.41% (0.07-0.75)
p=0.0
664
pvalue
0.35% (-0.01-0.71)
SC
753
p-
value
Percentage change in IMT per 10 nmol/L increase in serum 25(OH)D3
Change (95% CI)
AC C
EP
TE D
M AN U
2
24
p=0.06
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Highlights
The prevalence of atherosclerosis based on intima media thickness (IMT) was 20.1%.
•
No association was found between serum vitamin D (25(OH)D3) and IMT.
•
Genetic status and lifestyle factors may overrule the role of vitamin D.
AC C
EP
TE D
M AN U
SC
RI PT
•
S0021-9150(17)31413-2
DOI:
10.1016/j.atherosclerosis.2017.11.028
Reference:
ATH 15282
To appear in:
Atherosclerosis
Received Date: 25 August 2017 Revised Date:
10 November 2017
Accepted Date: 28 November 2017
Please cite this article as: Gjødesen CU, Jørgensen ME, Bjerregaar P, Dahl-Petersen IK, Larsen CVL, Noël M, Melbye M, Cohen AS, Lundqvist M, Hougaard DM, Helge JøW, Nielsen NO, Associations between vitamin D status and atherosclerosis among Inuit in Greenland, Atherosclerosis (2017), doi: 10.1016/j.atherosclerosis.2017.11.028. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Associations between vitamin D status and atherosclerosis among Inuit in Greenland
Camilla U. Gjødesena, Marit E. Jørgensena,b,c, Peter Bjerregaara,c, Inger K. Dahl-Petersena, Christina
RI PT
V. L. Larsena, Martin Noël d, Mads Melbyee,f,g, Arieh S. Cohenh, Marika Lundqvisth, David M. Hougaardh, Jørn W. Helgei, Nina O. Nielsena,j
b
c
Steno Diabetes Centre Copenhagen, Gentofte, Denmark.
SC
National Institute of Public Health, University of Southern Denmark, Copenhagen, Denmark.
M AN U
a
Greeenland Centre for Health Research, University of Greenland, Nuuk, Greenland.
d
Université Laval, Quebec, Canada.
e
Department of Epidemiology Research, Statens Serum Institut, Copenhagen, Denmark.
f
Department of Clinical Medicine, University of Copenhagen, Denmark
g
h
i
TE D
Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA Department for Congenital Disorders, Statens Serum Institut, Copenhagen, Denmark.
Center of Healthy Aging, Department of Biomedical Science, University of Copenhagen,
Centre for Nutrition and Rehabilitation, University College Absalon, Næstved, Denmark (current
address).
AC C
j
EP
Copenhagen, Denmark
*Correspondence to: [email protected] (C. U. Gjødesen)
1
ACCEPTED MANUSCRIPT Abstract Background and aims: Low levels of vitamin D are suspected to be a risk factor for cardiovascular disease and atherosclerosis. The aim of this study was to assess the prevalence of subclinical
RI PT
atherosclerosis among Inuit in Greenland, and to evaluate the association with vitamin D status. We hypothesized that low vitamin D status could be associated with higher carotid intima-media thickness (IMT) as a marker of atherosclerosis.
SC
Methods: 756 adults from the Inuit Health in Transition (IHIT) study carried out in Greenland in the period 2005-2010 were included. A blood sample donated in 1987 was available for a sub-sample
M AN U
of 102 individuals. Serum 25(OH)D3 from the IHIT study and the 1987 survey was used as a measure of vitamin D status. IMT measurements were conducted by ultrasound scanning. The prevalence of atherosclerosis was estimated, and the association between serum 25(OH)D3 and IMT measurements was examined by linear regression.
TE D
Results: The overall prevalence of subclinical atherosclerosis was 20.1% (n=152). The linear regression analyses indicated a weak positive association between serum 25(OH)D3 level and IMT measurements from the IHIT study, though not statistically significant after adjustment for potential
EP
confounders (β=0.35% per 10 nmol/l 25(OH)D3, p=0.06). Linear regression analyses of the association between serum 25(OH)D3 level in the 1987 survey and IMT measurements also
AC C
indicated a positive, though not statistically significant, association after adjustment (β=0.07% per 10 nmol/l 25(OH)D3, p=0.86). Conclusions: Our findings did not support the hypothesis of an association between low vitamin D levels and risk of atherosclerosis.
2
ACCEPTED MANUSCRIPT 1. Introduction Vitamin D is an essential fat-soluble vitamin with great importance for human health. There is an international consensus that vitamin D is important for maintenance of calcium homeostasis, bone
RI PT
health and prevention of fractures (1). Recent studies have identified low vitamin D as a new potential risk factor for atherosclerosis and cardiovascular disease in general (2, 3). Carotid
atherosclerosis can be used as a clinical marker of systemic atherosclerosis due to early onset,
SC
location and easy detection by ultrasound (4). Additionally, the presence of carotid plaque and measurement of the carotid intima-media thickness (IMT) can be used as a surrogate marker of
M AN U
subclinical atherosclerosis (3).
In the Arctic, dermal production of vitamin D is low due to the large solar zenith angle (SZA), long periods with low sun exposure and the continuous need for protective outdoor clothing (5, 6). Additionally, the dark skin pigmentation in the Inuit population may reduce vitamin D
TE D
synthesis in the skin (7). Previously, the high intake of traditional vitamin D rich diet containing fatty fish, seal and whale, probably compensated for the poor skin synthesis of vitamin D (5-7). During the last 50-60 years, the traditional diet of the Inuit has largely been replaced by imported
EP
foods (8), along with an observed lower vitamin D status in Arctic populations (6, 9, 10). A new study confirms that vitamin D deficiency is increasing in Greenland (10).
AC C
Formerly, it was a common notion that ischaemic heart disease is rare among Inuit in Greenland due to their marine diet with a high content of omega-3-fatty acids. However, the scientific evidence for this is weak since it is not based on systematic studies, but rather uncertain mortality statistics (11, 12). So far, only two studies have reported on atherosclerosis among Inuit in Greenland (13, 14). One of these reported that indigenous Greenlanders develop atherosclerosis to the same extent as Europeans (13), and the other concluded that the extent of advanced atherosclerosis in Greenlanders was similar to that reported among Alaska natives but less than
3
ACCEPTED MANUSCRIPT reported for Alaska non-natives (14). To our knowledge, no studies have investigated the association between vitamin D status and atherosclerosis in Arctic populations. In this study, we hypothesized that low vitamin D levels were associated with higher
RI PT
risk of atherosclerosis among Inuit. We used serum 25-hydroxyvitamin D3 (25(OH)D3) concentration as a measure of vitamin D status and included blood samples collected in 2005-2010 and in 1987. Carotid intima-media thickness (IMT) was used as a surrogate marker of subclinical
SC
atherosclerosis. Subclinical atherosclerosis is hereafter referred to as atherosclerosis.
M AN U
2. Materials and methods 2.1 Study population
The Inuit Health in Transition (IHIT) study, a general health study in Greenland and Canada that aimed to investigate health and diseases, lifestyle and life conditions in the Arctic regions, was
TE D
carried out between 2005 and 2010 (15). One purpose of the IHIT study was to assess risk factors of cardiovascular disease to obtain a better understanding of the health effects of the transition from a traditional to a modern and industrialized lifestyle in the Arctic regions (15). Participants in the
EP
IHIT study in Greenland were selected through a stratified random sample of the adult population (+ 18 years) born in Greenland or Denmark. Greenland was divided into 10 regions based on
AC C
geography and community size. From each region, one or more towns and villages were selected. The towns were selected as representative of the region in terms of living conditions (15). Villages were chosen randomly in the regions, and all adults living there were invited to participate (15). The total study sample consisted of 3108 Inuit (9.1% of the total adult population in 2005). The IHIT study population represented all age groups (+18 years), all geographical areas and all community sizes of Greenland. A valid serum 25(OH)D3 measurement was obtained from 2,877 individuals (93%) (10). 1,043 of the participants had an IMT measurement conducted (only a random selection
4
ACCEPTED MANUSCRIPT of individuals aged 40 years or above were examined), and 793 of these measurements were valid (individual mean IMT was based on a minimum of 6 segment measurements as described in section 2.4). Individuals with valid serum 25(OH)D3 and valid IMT measurements were included, which
RI PT
lead to the final number of 756 participants in the present study. Thus, due to invalid (less than 6 segment measurements available) data on IMT (n=250), missing data on IMT (due to random selection and cut-off at ≥ 40 years, n=1914), missing data on serum 25(OH)D3 (n=37) and ethnicity
SC
other than Greenlander (n=151), a total of 2352 individuals in the IHIT study were excluded from this study.
M AN U
In 1987, due to a syphilis epidemic in western and southern Greenland, the health authorities in Greenland conducted a population-based serological examination. Among those that participated in the survey in 1987, we identified 102 who had a blood sample available for measurement of serum 25(OH)D3 and also had an IMT measurement conducted as part of the IHIT
TE D
study in 2005-2010. The above described measurements allowed for an assessment of a crosssectional association between 25(OH)D3 status and IMT in 2005-2010 and an association between
EP
25(OH)D3 status in 1987 and development of IMT later in life (2005-2010) (fig. 1).
2.2 Collection of blood samples
AC C
Blood samples from 1987 were drawn from May to June, whereas samples from 2005-2010 were collected during all months of the year in the period from April 2005 to October 2010, except for July, November and December. Blood samples were handled and stored as described elsewhere (10).
2.3 Measurement of vitamin D
5
ACCEPTED MANUSCRIPT Serum 25(OH)D3 was used as a measure of vitamin D status. Samples were analysed after storage at -80 °C for 3-7 years (IHIT samples) and 26 years (1987-samples), respectively (10). Analyses were conducted by liquid chromatography-tandem mass spectrometry (LC-MSMS) using the
RI PT
“MSMS vitamin D” kit from Perkin Elmer (Waltham, MA) as described previously (16). This method measured both serum 25(OH)D2 and 25(OH)D3, but since concentrations of 25(OH)D2
2.4 Measurement of IMT by carotid ultrasonography
SC
were negligible, only 25(OH)D3 was included in the analysis.
M AN U
Measurements of the intimal to medial arterial wall thickness of the carotid arteries were performed using a high resolution B-mode ultrasound portable device (Model LogiqBook, GE Medical System, Milwaukee, WI, USA) with a linear 4–10 MHz probe (Model 10LB-Rs, GE Medical System, Milwaukee, WI, USA) by two experienced sonographers who conducted the ultrasound
TE D
scanning according to a standard protocol (17, 18). A transverse scan was performed prior to a longitudinal scan in order to detect any hemodynamically relevant stenosis or presence of significant plaque defined by a focal echogenic structure protruded into the lumen vessel. Segments
EP
of 1 cm of the near and far walls of the Common carotid (1 cm below the flow divider), the Bifurcation and the Internal carotid (1 cm distal to the flow divider) arteries were digitally recorded
AC C
and used for evaluation of the IMT. The segments had to be free of atherosclerotic plaque. All images were saved in digital imaging and communication in medicine (DICOM) database format. A single reader blinded to the clinical data performed off-line measurements. Analyses of 3 nonconsecutive frames were completed using a dedicated semi-automatic edge detection software (Carotid analyser for research v5.5.6, Medical Imaging Application LLC, Coraville, IA, USA). When the analyser determined a region of interest (ROI) of 1 cm in length, the program automatically located the two interfaces defining walls (lumen-intima and media adventitia) using
6
ACCEPTED MANUSCRIPT algorithms based on grey level density and tissue recognition and drew two parallel lines. In case of visual discrepancies, the analyser had the possibility of editing the lines by fine-tuning the detection algorithms. IMT was measured every 10 µm of the length of the ROI (n = 100 successive local
RI PT
measures) for each frame. The measurements of each frame of the Common, Bifurcation and Internal carotid segments of the near and far wall on carotid artery in both right and left side of the neck were averaged and referenced as a mean IMT (17, 19). This study will refer to a mean IMT
SC
from each participant calculated as an average of available measured segments. However, to be a valid IMT, a minimum of 6 segment measurements were included in the mean IMT with at least
M AN U
one segment measurement from each frame (Common, Bifurcation and Internal) of both right and left carotid arteries. Atherosclerosis was categorized with the following cut-off values according to Riccioni et al. (20): No evidence of atherosclerosis (IMT < 0.8 mm), some atherosclerosis (0.8 mm
2.5 Potential confounders
TE D
≤ IMT ≤ 1.2 mm) and extensive atherosclerosis (IMT > 1.2 mm).
Information on age, sex and smoking was collected by interviews as part of the IHIT study. Hepatic insulin resistance (HOMA-IR) was calculated using the formula: fasting plasma glucose (mmol/L)
EP
x fasting plasma insulin (mU/L)/22.5 (21). Systolic BP was measured three times with an automatic
AC C
measuring device (Kivex UA-779) with an appropriately sized cuff. The last two measurements were averaged and used for the analysis. Physical activity (PAEE) was recorded by the International Physical Activity Questionnaire (long version IPAQ), modified and validated for the living conditions of the Greenlandic population (22). Anthropometric measurements were used to calculate body mass index (BMI) as: weight in kg/(height in m2). HDL and triglycerides were measured in serum by automated enzymatic colorimetic methods. LDL cholesterol was calculated with the Friedewald formula. Methyl mercury was measured by inductively coupled mass spectromy (ICP-MS) at the Centre de Toxicologie, Institut National de Santé Public, Québec,
7
ACCEPTED MANUSCRIPT Canada, as described previously (23). Information about genetic admixture was estimated from blood tests where the admixture proportion of Inuit and European ancestry was quantified. The analyses have accurately quantified the extent of European and Inuit ancestry in the participants
RI PT
included in the present study (24).
2.6 Ethics
SC
Written informed consent was obtained from each participant included in the IHIT study, the study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki and t has been
M AN U
reviewed and approved by the Ethical Review Committee for Greenland.
2.7 Statistical methods
All statistical analyses were performed using the statistical software STATA version 14 (StataCorp.
TE D
2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP). The associations between serum 25(OH)D3 and IMT as a measure of atherosclerosis were explored by linear regression analysis with serum 25(OH)D3 as the explanatory variable and IMT measures as the
EP
outcome variable. IMT measures did not follow a normal distribution and were therefore logtransformed to achieve normal distribution before entered into the linear regression model. β-
AC C
estimates from linear regression models were back-transformed and presented as a percentage increase per 10 nmol/L increase in serum 25(OH)D3. The cross sectional association between IMT and serum 25(OH)D3 measured in the IHIT study was assessed in three linear regression models and adjusted for an increasing number of confounders. Model 1: Age, sex and genetic admixture. Model 2: As model 1 + BMI, smoking, HDL, triglycerides, systolic BP, HOMA-IR and PAEE. Model 3: As model 2 + methyl mercury. The association between serum 25(OH)D3 measured in the 1987 survey and IMT measured in IHIT was assessed in two linear regression models: Model 1:
8
ACCEPTED MANUSCRIPT Unadjusted. Model 2: Adjusted for age, sex and genetic admixture. In all models, a quadratic term of age was included in order to take into account the non-linear effect of age on the association
3. Results 3.1 Characteristics of the study population
RI PT
between serum 25(OH)D3 and IMT.
SC
In the IHIT study, the median age was 46 years (interquartile range, IQR, 40-56 years), the median serum 25(OH)D3 concentration was 54 nmol/L (IQR 35-75 nmol/L) and the median IMT was 0.69
M AN U
mm (IQR 0.61-0.77mm). A sub-group of 485 participants from the IHIT study had previously participated in the 1987 survey, and among them, data on IMT and serum 25(OH)D3 was available from 102 individuals. Of the 102 individuals in the 1987 sample, 52 (51%) were men. The median age was 31 years (IQR 23-39) in 1987, median serum 25(OH)D3 concentration was 72 nmol/L
TE D
(IQR 49-109 nmol/L) and median IMT was 0.70 mm (IQR 0.62-0.80 mm). Table 1 gives characteristics of the study population in 1987 and 2005-2010 by sex. The prevalence of smokers was 69 % among men and 72% among women. The degree of genetic admixture was
EP
82% for both men and women. Median PAEE was higher among men than women (56 kj/kg/day vs.
AC C
46 kj/kg/day). IFG was more prevalent among men than women (23% vs. 16%).
3.2 Prevalence of atherosclerosis using IMT as a surrogate marker Based on IMT measurements, three (0.4%) of the 756 individuals had extensive atherosclerosis and 149 (19.7%) had some atherosclerosis. Thus, according to IMT, the overall prevalence of atherosclerosis was 20.1% in the study population.
3.3 Associations between IMT and serum 25(OH)D3 in the IHIT sample
9
ACCEPTED MANUSCRIPT Table 2 shows the association between serum 25(OH)D3 level and IMT measurements in the IHIT sample. In model 1, the estimate indicated no association between serum 25(OH)D3 level and IMT measurements after adjustment for age, sex and genetic admixture. After additional adjustment for
RI PT
BMI, smoking, HDL, triglycerides, systolic BP, HOMA-IR and PAEE (model 2), the estimate indicated a positive association between serum 25(OH)D3 and IMT. Thus, increasing serum 25(OH)D3, expressed as 10 nmol/L, was associated with increasing IMT (0.41%, 95% confidence
SC
interval (CI): 0.07% – 0.75%, p=0.02). However, after additional adjustment for methyl mercury
M AN U
(model 3), the association disappeared (0.35%, CI: -0.01% – 0.71%, p=0.06).
3.4 Associations between IMT and serum 25(OH)D3 in the 1987 sample A linear unadjusted regression analysis (n=102) of the association between serum 25(OH)D3 level in the 1987 sample and IMT measurements in the IHIT sample showed a positive association between serum 25(OH)D3 level and IMT (1.1%, CI: 0.4% -1.9%, p=0.003). However, after
4. Discussion
EP
p=0.86).
TE D
adjustment for age, sex and admixture, the association disappeared (0.07%, CI: -0.68% – 0.82%,
The overall prevalence of atherosclerosis was 20.1% according to the applied definitions (IMT ≥
AC C
0.8). A Canadian study among Inuit in Nunavik found that 11,5 % of the study population had an IMT greater than 1 mm (19). That study referred to a higher risk of cardiovascular disease if IMT was greater than 1 mm, however, not linking IMT directly to atherosclerosis. The difference in cutoff between the Nunavik study and our study may at least in part explain the difference in the prevalence of atherosclerosis based on IMT. A study among Alaskan Eskimos reported that carotid atherosclerosis assessed by ultrasound was more commonly detected in that population compared to American populations (25), but the specific prevalence was not reported. A Canadian study
10
ACCEPTED MANUSCRIPT examining a multi-ethnic low-risk population found an overall prevalence of atherosclerosis of 23% (26). In that study, subclinical atherosclerosis was defined according to measures of mean maximum cIMT≥75th percentile and is therefore difficult to compare directly to the present study.
RI PT
The Manfredonia Study, which examined an Italian population and used the same cut-off as used in the present study, found an atherosclerosis prevalence of 54% (27). However, a higher mean age in the Manfredonia Study may explain the higher prevalence. Most results are obviously not in line
SC
with the notion that Inuit may have lower risk of cardiovascular disease due to their intake of traditional Arctic diet with a high content of omega 3-fatty acids. As indicated above, there are
M AN U
challenges by comparing the prevalence of atherosclerosis among different populations due to methodological inconsistencies in the current literature.
The present study is, to our knowledge, the first to report on the association between serum 25(OH)D3 and atherosclerosis in an Inuit population. In contrast to our hypothesis, we found
TE D
a weak positive association between 25(OH)D3 and IMT measures, though not statistically significant after adjustment for age, sex, genetic admixture, BMI, smoking, HDL, triglycerides, systolic BP, HOMA-IR, PAEE and methyl mercury. Previous studies in other populations
EP
evaluating the association between 25(OH)D levels and atherosclerosis assessed by IMT have shown conflicting results. In The Multi-Ethnic Study of Atherosclerosis including 3251 low risk
AC C
participants, 25(OH)D were not significantly associated with IMT (28). Furthermore, in a multicenter study examining the association between serum 25(OH)D concentrations and IMT measures in 3430 middle-aged and elderly subjects with high CVD risk found no independent relationship between 25(OH)D and segment-specific or composite IMT measures (29). In contrast, several studies have found that low 25(OH)D levels were associated with higher IMT (3, 30-32) and low vitamin D seems also associated with adverse CV outcome (33). One study showed a positive association between serum 25(OH)D and IMT among patients with sufficient 25(OH)D levels (≥50
11
ACCEPTED MANUSCRIPT nmol/L) (34). Overall, most studies conducted in non-Inuit populations found that low vitamin D was associated with atherosclerosis. The present study indicates, however, that 25(OH)D3 plays a minor, if any, role in the development of atherosclerosis in an Arctic population. It is reasonable to
RI PT
consider if the inverse association found between 25(OH)D and atherosclerosis in some studies may be explained by residual confounding, e.g. predisposing cardiovascular risk factors, genetic factors, sun exposure and dietary factors. The lack of an association is in agreement with findings from
SC
Mendelian Randomization studies and a meta-analysis of eight RCTs suggesting that associations between 25(OH)D and atherosclerosis in epidemiological studies result from reverse causation or
M AN U
residual confounding (35, 36).
In the present study, methyl mercury was perceived as a classic potential confounder due to its association with both vitamin D and IMT. Environmental pollutants and methyl mercury have been shown to be associated with cardiovascular outcomes (23, 37-39) and IMT (40, 41). Of
TE D
particular interest to Arctic populations, some studies suggest that Inuit who eat a traditional diet with a high intake of fatty fish, seal and whale, and therefore have a high intake of vitamin D (10), are at the same time highly exposed to environmental pollutants including methyl mercury (23, 42).
EP
Environmental pollutants and mercury bioaccumulate in marine mammals and fish and are transferred to humans through the diet. Mercury is present in the blood and organs of Greenlanders
AC C
in concentrations well above international guidelines (39, 43). In the present study, the positive association between serum 25(OH)D3 and IMT was eliminated after adjustment for methyl mercury. A possible explanation could be that the association was “carried” by methyl mercury. Other environmental contaminants and heavy metals specific for the arctic area, such as polychlorinated biphenyls and cadmium (44), are reasonable to consider as factors with possible impacts on the findings.
12
ACCEPTED MANUSCRIPT The Inuit population is highly genetically differentiated from other populations, because of being isolated and affected by bottlenecks in population size (45), and diet and climate adaptation in the Arctic has favored genetic variants associated with severe insulin resistance and
RI PT
increased susceptibility to cardiometabolic disease (46, 47). In the present study, the population structure was taken into account by adjusting the association between 25(OH)D3 and IMT for genetic admixture, with the aim of minimizing the risk of confounding by this factor.
SC
There was a positive association between serum 25(OH)D3 level in the 1987 sample and IMT measurements from the IHIT sample, though not statistically significant after adjustment
M AN U
for age, sex and genetic admixture. Thus, the longitudinal findings confirmed the absence of an association in the cross-sectional analysis. According to the literature review, only the Multi-Ethnic Study of Atherosclerosis, with 3251 low-risk participants, examining serum 25(OH)D in relation to IMT, had a longitudinal design (10 years follow-up) (28), and the finding of no association is
TE D
consistent with the findings in the present study. However, a limitation in the present study is the lack of information on IMT from the 1987 sample. Furthermore, adjustments for confounders were limited to age, sex and genetic admixture, which may have resulted in residual confounding.
EP
Additionally, the sample size of 102 participants in this part of the study may not provide sufficient power to detect a possible association. However, the finding of a similar association pattern with
AC C
IMT when using 25(OH)D3 measurements from the 1987 sample and from the IHIT sample indicates some robustness of the findings. Strengths of the study include the use of serum 25(OH)D3 measurements from a large
population based and geographically representative study of Inuit in Greenland, who also had a carotid ultrasonography. Vitamin D status (serum 25(OH)D3) was measured in blood samples and not estimated from questionnaires, which minimizes the risk of misclassification. Another major strength is that detailed ultrasound measurements of IMT from both right and left carotid arteries
13
ACCEPTED MANUSCRIPT were performed and 6-12 measurements were included. The American Heart Association has endorsed the use of IMT in cardiovascular risk assessment (17) and B-mode ultrasound IMT measurements are seen as the best current method to provide objective information of the degree of
RI PT
subclinical atherosclerosis (26, 48). Thus, both exposure and outcome are objectively measured and not based on self-reports. Furthermore, methyl mercury, HDL, triglycerides, genetic admixture and HOMA-IR levels were obtained from measurements in serum samples, and detailed information on
SC
physical activity was available from the IPAQ validated questionnaire and used for adjustment of the analyses.
M AN U
In conclusion, this study is the first to examine the association between vitamin D and atherosclerosis among Inuit in Greenland. No longitudinal or cross sectional association was found between serum 25(OH)D3 and IMT in the IHIT sample and the 1987 sample, and based on these results it seems reasonable to conclude that vitamin D has no impact on the development of
TE D
atherosclerosis among Inuit in Greenland. Genetic status is likely overruling a possible role of vitamin D in the development of atherosclerosis in this population. Furthermore, the contribution of lifestyle related risk factors such as smoking, physical inactivity and a diet containing high levels of
AC C
vitamin D status.
EP
saturated fat and sugar might affect the development of atherosclerosis substantially more than
Conflict of interest
The authors declared they do not have anything to disclose regarding conflict of interest with respect to this manuscript.
14
ACCEPTED MANUSCRIPT Financial support The study was supported by the Government of Greenland’s health department (2011-047012/2012074239), Aase og Ejnar Danielsen’s Foundation (10-000444), Dagmar Marshalls Foundation,
RI PT
Karen Elise Jensens Foundation and Nuna Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
SC
Author contributions
C.U.G., N.O.N., M.E.J and J.H. conceived and designed the study. C.U.G. analyzed the data and
M AN U
wrote the manuscript. N.O.N. assisted with data analysis. P.B., I.D.P, C.V.L.L, M.E.J. and M.N. collected the IHIT data. M.M., A.S.C., M.L. and D.M.H. contributed with study materials and analysis tools. All authors reviewed the manuscript and contributed with critical comments to the content. All authors approved the submitted version of the manuscript. C.U.G. is the guarantor of
TE D
this work, and, as such, had full access to all data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
EP
Acknowledgements
The authors thank the study participants for participating in the study. Mrs. Ingelise Olesen is
AC C
thanked for her efforts in the practical organization of the IHIT study and participation in data collection, and the interviewers are thanked for their valuable assistance in data collection. The late Eric Dewailly is thanked for his contribution to the collection of the ultrasound measurements. Colleagues from the laboratories of the Steno Diabetes Centre Copenhagen, Gentofte, Denmark, in particular Mrs. Maja Lis Dybdahl Halkjær, are thanked for indispensable assistance with handling of the blood samples.
15
SC
RI PT
ACCEPTED MANUSCRIPT
References
M AN U
1. Palacios C, Gonzalez L. Is vitamin D deficiency a major global public health problem? J Steroid Biochem Mol Biol. 2014; 144 Pt A:138-45. 2. Kalkan GY, Gür M, Koyunsever NY, Şeker T, Gözükara MY, et al. Serum 25Hydroxyvitamin D Level and Aortic Intima-Media Thickness in Patients Without Clinical Manifestation of Atherosclerotic Cardiovascular Disease. J Clin Lab Anal. 2015; 29(4):30511.
TE D
3. Hao Y, Ma X, Luo Y, Ni J, Dou J, et al. Additional role of serum 25-hydroxyvitamin D3 levels in atherosclerosis in Chinese middle-aged and elderly men. Clin Exp Pharmacol Physiol. 2014; 41(3):174-9. 4. Grobbee DE, Bots ML. Carotid artery intima-media thickness as an indicator of generalized atherosclerosis. J. Intern Med. 1994; 236:567-73.
EP
5. Webb AR. Who, what, where and when-influences on cutaneous vitamin D synthesis. Prog Biophys Mol Biol. 2006; 92(1):17-25.
AC C
6. Andersen S, Jacobsen A, Laurberg P. Vitamin D status in North Greenland is influenced by diet and season: indicators of dermal 25-hydroxy vitamin D production north of the Arctic Circle. Br J Nutr. 2012; 110:50-57. doi:10.1017/S0007114512004709. 7. Fares JEH, Weiler HA. Implications of the nutrition transition for vitamin D intake and status in aboriginal groups in the Canadian Arctic. Nutr Rev. 2016; 74 (9): 571-583. doi: https://doi.org/10.1093/nutrit/nuw020 8. Jeppesen C, Bjerregaard P. Consumption of traditional food and adherence to nutrition recommendations in Greenland. Scand J Public Health. 2012;40(5):475-81. 9. Rejnmark L, Jørgensen ME, Pedersen MB, Hansen JC, Heickendorff L, et al. Vitamin D insufficiency in Greenlanders on a westernized fare: ethnic differences in calcitropic hormones between Greenlanders and Danes. Calcif Tissue Int. 2004; 74(3):255-63.
16
ACCEPTED MANUSCRIPT
10. Nielsen NO, Jørgensen ME, Friis H, Melbye M, Soborg B, et al. Decrease in vitamin D status in the Greenlandic adult population from 1987-2010. PLoS One. 2014;9(12):e112949. 11. Jørgensen ME, Bjerregaard P, Kjaergaard JJ, Borch-Johnsen K. High prevalence of markers of coronary heart disease among Greenland Inuit. Atherosclerosis. 2008; 196(2):772-8.
RI PT
12. Bjerregaard P, Young TK, Hegele RA. Low incidence of cardiovascular disease among the Inuit - what is the evidence? Atherosclerosis. 2003; 166 (2):351-7. 13. Hansen JP, Hancke S, Møller-Petersen J. Atherosclerosis in Greenland: an ultrasonographic investigation. Arctic Med Res. 1991; Suppl:400-3.
SC
14. Pedersen HS, Mulvad G, Newman WP, Boudreau DA. Atherosclerosis in coronary arteries and aorta among Greenlanders: an autopsy study. Atherosclerosis. 2003; 170(1):93-103.
M AN U
15. Bjerregaard P. Inuit Health in Transition - Greenland survey 2005-2010. Population sample and survey methods. 2nd edition. 2011;SIF Writings on Greenland vol. 19:1-26. Link: http://www.si-folkesundhed.dk/upload/inuit_health_in_transition_greenland_methods _5_2nd_revision.pdf 16. Nielsen NO, Strøm M, Boyd HA, Andersen EW, Wohlfahrt J, Lundquist M, et al. Vitamin D status during pregnancy and the risk of subsequent postpartum depression: A case-control study. PLOS ONE. 2013; 8: e80686. doi:10.1371/journal.pone.0080686 PMID: 24312237.
TE D
17. Oliveira GM, De Lorenzo A, de Souza EG, Santos B. Association between Carotid Intimamedia Thickness and Early-onset Coronary Artery Disease: Does the site of Sonographic Evaluation Matter? Heart Lung Circ. 2015; 24(9):932-5.
EP
18. Stein JH, Korcarz CE, Hurst RT, Lonn E, Kendall CB, et al. Use of carotid ultrasound to identify subclinical vascular disease and evaluate cardiovascular disease risk: a consensus statement from the American Society of Echocardiography Carotid Intima-Media Thickness Task Force. Endorsed by the Society for Vascular Medicine. J AM Soc Echocardiogr. 2008; 21(2):93-111.
AC C
19. Noël M, Dewailly E, Chateau-Degat ML, Counil É, Laouan-Sidi EA, et al. Cardiovascular risk factors and subclinical atherosclerosis among Nunavik Inuit. Atherosclerosis. 2012; 221(2):558-64. 20. Riccioni G, D'Orazio N, Palumbo N, Bucciarelli V, Ilio E, et al. Relationship between plasma antioxidant concentrations and carotid intima-media thickness: the Asymptomatic Carotid Atherosclerotic Disease In Manfredonia Study. Eur J Cardiovasc Prev Rehabil. 2009;16(3):351-7. 21. Johnson-Down L, Labonte ME, Martin ID, Tsuji LJ, Nieboer E, et al. Quality of diet is associated with insulin resistance in the Cree (Eeyouch) indigenous population of northern Québec. Nutr Metab Cardiovasc Dis. 2015; 25(1):85-92.
17
ACCEPTED MANUSCRIPT 22. Dahl-Petersen IK. Physical activity pattern and its relation to glucose metabolism in Greenland [PhD thesis]. Centre for Health Research in Greenland: University of Southern Denmark; 2013.
RI PT
23. Nielsen AB, Davidsen M, Bjerregaard P. The association between blood pressure and whole blood methylmercury in a cross-sectional study among Inuit in Greenland. Environ Health Perspect. 2012; 11: 44. doi: 10.1186/1476-1069X-1111-1144. 24. Moltke I, Fumagalli M, Korneliussen TS, Crawford JE, Bjerregaard P, et al. Uncovering the genetic history of the present-day Greenlandic population. Am J Hum Genet. 2015; 96(1):54-69.
SC
25. Cutchins A, Roman MJ, Devereux RB, Ebbesson SO, Umans JG, et al. Prevalence and correlates of subclinical atherosclerosis in Alaska Eskimos: the GOCADAN study. Stroke. 2008; 39(11):3079-82.
M AN U
26. Grewal J, Anand S, Islam S, Lonn E, Investigators SaS-A. Prevalence and predictors of subclinical atherosclerosis among asymptomatic "low risk" individuals in a multiethnic population. Atherosclerosis. 2008; 197(1):435-42. 27. Riccioni G, D'Orazio N, Palumbo N, Bucciarelli V, Ilio E, et al. Relationship between plasma antioxidant concentrations and carotid intima-media thickness: the Asymptomatic Carotid Atherosclerotic Disease In Manfredonia Study. Eur J Cardiovasc Prev Rehabil. 2009; 16(3):351-7.
TE D
28. Blondon M, Sachs M, Hoofnagle AN, Ix JH, Michos ED, et al. 25-Hydroxyvitamin D and parathyroid hormone are not associated with carotid intima-media thickness or plaque in the multi-ethnic study of atherosclerosis. Arterioscler Thromb Vasc Biol. 2013; 33(11):2639-45.
EP
29. Deleskog A, Piksasova O, Silveira A, Gertow K, Baldassarre D, et al. Serum 25hydroxyvitamin D concentration in subclinical carotid atherosclerosis. Arterioscler Thromb Vasc Biol. 2013; 33(11):2633-8.
AC C
30. Carrelli AL, Walker MD, Lowe H, McMahon DJ, Rundek T, et al. Vitamin D deficiency is associated with subclinical carotid atherosclerosis: the Northern Manhattan study. Stroke. 2011; 42(8):2240-5. 31. van de Luijtgaarden KM, Voûte MT, Hoeks SE, Bakker EJ, Chonchol M, et al. Vitamin D deficiency may be an independent risk factor for arterial disease. Eur J Vasc Endovasc Surg. 2012; 44(3):301-6. 32. Reis JP, von Mühlen D, Michos ED, Miller ER, Appel LJ, et al. Serum vitamin D, parathyroid hormone levels, and carotid atherosclerosis. Atherosclerosis. 2009; 207(2):58590. 33. Tuńón J, Cristóbal C, Tarín N, Aceńa A, González-Casaus ML, et al. Coexistence of Low Vitamin D and High Fibroblast Growth Factor-23 Plasma Levels Predicts an Adverse Outcome in Patients with Coronary Artery Disease. 2014; 9, e95402.
18
ACCEPTED MANUSCRIPT
34. van Dijk SC, Sohl E, Oudshoorn C, Enneman AW, Ham AC, et al. Non-linear associations between serum 25-OH vitamin D and indices of arterial stiffness and arteriosclerosis in an older population. Age Ageing. 2015; 44(1):136-42.
RI PT
35. Manousaki D, Mokry LE, Ross S, Goltzman D, Richards JB. Mendelian Randomization Studies Do Not Support a Role for Vitamin D in Coronary Artery Disease. Circ Cardiovasc Genet. 2016 Aug; 9(4):349-56. doi: 10.1161/CIRCGENETICS.116.001396. Epub 2016 Jul 14.
SC
36. Wang L, Manson JE, Song Y, Sesso HD. Systematic review: Vitamin D and calcium supplementation in prevention of cardiovascular events. Ann Intern Med. 2010 Mar 2; 152(5):315-23. doi: 10.7326/0003-4819-152-5-201003020-00010.
M AN U
37. Salonen JT, Seppänen K, Lakka TA, Salonen R, Kaplan GA. Mercury accumulation and accelerated progression of carotid atherosclerosis: a population-based prospective 4-year follow-up study in men in eastern Finland. Atherosclerosis. 2000; 148(2):265-73. 38. Valera B, Jørgensen ME, Jeppesen C, Bjerregaard P. Exposure to persistent organic pollutants and risk of hypertension among Inuit from Greenland. Environ Res. 2013; 122:65-73.
TE D
39. Bjerregaard P, Mulvad G. The best of two worlds: how the Greenland Board of Nutrition has handled conflicting evidence about diet and health. Int J Circumpolar Health. 2012; 71:18588.
EP
40. Skoczynska A, Poreba R, Steinmentz-Beck A, Martynowicz H, Affelska-Jercha A, et al. The dependence between urinary mercury concentration and carotid arterial intima-media thickness in workers occupationally exposed to mercury vapour. Int J Occup Med Environ Health. 2009; 22:135-142.
AC C
41. Salonen. JT, Seppänen K, Lakka TA, Salonen R, Kaplan GA. Mercury accumulation and accelerated progression of carotid atherosclerosis: a population-based prospective 4-year follow-up study in men in eastern Finland. Atherosclerosis. 2000; 148:265-273. 42. Jeppesen C, Jørgensen ME, Bjerregaard P. Assessment of consumption of marine food in Greenland by a food frequency questionnaire and biomarkers. Int J Circumpolar Health. 2012; 71:18361. 43. Hansen JC, Van Oostdam J. AMAP Assessment 2009: Human Health in the Arctic. 2009. 44. Fagerberg B, Kjelldahl J, Sallsten G, Barregard L, Forsgard N, et al. Cadmium exposure as measured in blood in relation to macrophage density in symptomatic atherosclerotic plaques from human carotid artery. 2016; 249: 209-214.
19
ACCEPTED MANUSCRIPT 45. Tvermosegaard M, Dahl-Petersen IK, Nielsen NO, Bjerregaard P, Jørgensen ME. Cardiovascular Disease Susceptibility and Resistance in Circumpolar Inuit Populations. Can J Cardiol. 2015; 31(9):1116-23.
RI PT
46. Fumagalli M, Moltke I, Grarup N, Racimo F, Bjerregaard P, et al. Greenlandic Inuit show genetic signatures of diet and climate adaptation. Science. 2015 Sep 18; 349(6254):1343-7. doi: 10.1126/science.aab2319. 47. Moltke I, Grarup N, Jørgensen ME, Bjerregaard P, Treebak JT, et al. A common Greenlandic TBC1D4 variant confers muscle insulin resistance and type 2 diabetes. Nature. 2014 Aug 14; 512(7513):190-3. doi: 10.1038/nature13425. Epub 2014 Jun 18.
AC C
EP
TE D
M AN U
SC
48. van den Oord SC, Sijbrands EJ, ten Kate GL, van Klaveren D, van Domburg RT, et al. Carotid intima-media thickness for cardiovascular risk assessment: systematic review and meta-analysis. Atherosclerosis. 2013;228(1):1-11.
20
RI PT
ACCEPTED MANUSCRIPT
Attending 1987 survey (n=7609) Excluded Not attending IHIT study (n=7124) Attending both 1987 survey and IHIT study 2005-2010 (n=485) Excluded No excess material(n=155)
Excluded No IMT measurements (n=228)
TE D
Blood sample available for serum 25(OH)D3 measurement (n=330)
Attending IHIT study 2005-2010 (n=3108)
M AN U
Flow chart of the IHIT study (2005-2010) and 1987 survey.
SC
Fig. 1. Selection of study participants.
Attending IHIT study 2005-2010 with valid IMT measurement (n=793)
Excluded No IMT measurements conducted (n=1914) Invalid IMT measurements (n=250) Ethnicity other than Greenlander (n=151)
Excluded No data on serum 25(OH)D3 (n=37)
IHIT sample with valid IMT and serum 25(OH)D3 measurements (n=756)
AC C
EP
1987 sample Attending both IHIT study and 1987 survey (and have IMT measurements) (n=102)
Study population IHIT study (n=756) 1987 sample (n=102); a sub-sample of the IHIT study
21
ACCEPTED MANUSCRIPT
Table 1. Characteristics of the study populations in 1987 (n=102) and 2005-2010 (n=756) according to sex.
n
Men
n
Women
Age (years)
52
33 (25-40)
50
27 (21-36)
25(OH)D3 (nmol/L)
52
77 (48-120)
IMT (mm)
52
0.72 (0.65-0.86)
n
Men
Age (years)
376
47 (40-57)
380
45 (40-59)
BMI (kg/m2)
373
24 (22-28)
379
25 (22-29)
Serum 25(OH)D3 (nmol/L)
376
55 (35-78)
380
53 (34-72)
IMT (mm)
376
0.71 (0.62-0.81)
380
0.67 (0.60-0.74)
(kJ/kg/day)
375
56 (23-104)
380
46 (24-73)
HDL (mmol/L)
376
1.6 (1.3-2.0)
380
1.7 (1.4-2.1)
376
3.5 (2.9-4.1)
380
3.5 (2.9-4.2)
376
0.9 (0.7-1.3)
380
1.0 (0.8-1.3)
376
5.8 (5.4-6.2)
380
5.6 (5.2-6)
2h plasma glucose (mmol/L)
369
5.2 (4.1-6.5)
376
5.7 (4.7-6.8)
Fasting insulin (pmol/L)
376
28 (20-44.5)
380
40.5 (27-57)
369
63 (32-142)
375
139 (77-231)
376
7.3 (4.9-12.1)
380
10.3 (6.6-14.2)
Systolic BP (mmHg)
376
130 (121-145)
380
122 (111-136)
Diastolic BP (mmHg)
376
79 (70-89)
380
75 (68-83)
Mercury (µg/L)
376
24 (10-49)
379
19 (9-37)
Genetic admixture (%)
374
82 (63-99)
379
82(66-99)
Smokers (%)
375
69
380
72
IFG (%)
314
23
332
16
LDL (mmol/L) Triglycerides (mmol/L)
HOMA-IR
AC C
2h insulin (pmol/L)
EP
Fasting plasma glucose (mmol/L)
TE D
Physical activity energy expenditure
50
66 (50-92)
50
0.66 (0.59-0.77)
n
Women
SC
M AN U
IHIT sample (2005-2010)
RI PT
1987 sample
22
ACCEPTED MANUSCRIPT IGT (%)
252
4
294
8
T2DM (%)
376
11
380
6
AC C
EP
TE D
M AN U
SC
RI PT
Median with interquartile range and proportions are presented.
23
ACCEPTED MANUSCRIPT Table 2. Linear regression analysis of the cross sectional associations between serum 25(OH)D3 level and IMT measurements in the IHIT sample (n=753).
Model 1
Model 2
Adjusted for age, sex and genetic
Model 3
Additionally adjusted for BMI, smoking, HDL, triglycerides, systolic
mercury
RI PT
admixture
Additionally adjusted for methyl
BP, HOMA-IR and physical activity n
Change (95% CI)
p-
n
value
0.51% (-0.39-1.43)
p=0.27
n
Change (95% CI)
665
0.41% (0.07-0.75)
p=0.0
664
pvalue
0.35% (-0.01-0.71)
SC
753
p-
value
Percentage change in IMT per 10 nmol/L increase in serum 25(OH)D3
Change (95% CI)
AC C
EP
TE D
M AN U
2
24
p=0.06
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Highlights
The prevalence of atherosclerosis based on intima media thickness (IMT) was 20.1%.
•
No association was found between serum vitamin D (25(OH)D3) and IMT.
•
Genetic status and lifestyle factors may overrule the role of vitamin D.
AC C
EP
TE D
M AN U
SC
RI PT
•