SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Design and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Objective

During recent decades, the prevalence of metabolic morbidity has increased rapidly in adult Greenlandic Inuit. To what extent this is also reflected in the juvenile Inuit population is unknown. The objective was, therefore, in the comparison with Danish children, to evaluate metabolic profiles in Greenlandic Inuit children from the capital in the southern and from the northern most villages

Design and Methods

187 Inuit and 132 Danish children were examined with anthropometrics, pubertal staging, fasting blood samples, and a maximal aerobic test.

Results

Both Inuit children living in Nuuk and the northern villages had significantly higher glucose, total cholesterol, apolipoprotein A1 levels, and diastolic blood pressure compared with Danish children after adjustment for differences in adiposity and aerobic fitness levels. The Inuit children living in Nuuk had significantly higher BMI, body fat %, HbA1c, and significantly lower aerobic fitness and ApoA1 levels than northern living Inuit children.

Conclusions

Greenlandic Inuit children had adverse metabolic health profile compared to the Danish children, the differences where more pronounced in Inuit children living in Nuuk. The tendencies toward higher prevalence of diabetes and metabolic morbidity in the adult Greenlandic Inuit population may also be present in the Inuit children population.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Design and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Prior to western modernization, a low incidence of cardiovascular disease (CVD) was reported in the Greenlandic Inuit population [1]. However, during recent decades, the prevalence of CVD and type 2 diabetes has increased rapidly now reaching a higher prevalence than most European populations [2]. In addition, the high ratio of impaired glucose tolerance (IGT) to type 2 diabetes may indicate that the prevalence of type 2 diabetes and its cardiovascular comorbidities will continue to increase in the adult Greenland Inuit population [2].

Type 2 diabetes and CVD are rare in childhood. However, atherosclerogenesis is a gradual process starting in childhood and progresses throughout life. In accordance, obesity and adverse metabolic risk profiles in childhood predict a higher degree of coronary atherosclerosis in young adulthood [3]. The long latency period from the onset of these metabolic adverse risk factors to manifest CVD makes it difficult to establish thresholds that can reliably be used to predict and identify children at risk [4, 5]. Therefore, adequate definitions and therapeutic intervention strategies in children are less well established than in adults. Obesity and physical inactivity are two major causal factors in the development of type 2 diabetes and CVD, both of which have its roots in childhood and often continue into adulthood [6].

No data on cardiovascular fitness and metabolic risk factors in Greenlandic Inuit children and adolescents are available. The aim of this study was therefore to describe cardiovascular fitness, obesity, and metabolic risk profiles in two groups of Inuit children living in Nuuk, the capital of Greenland, and in the most northern part of Greenland and to compare these two groups with an age-matched group of Danish children and adolescents. We hypothesized that the observed increased prevalence of metabolic-related cardiovascular morbidity in the adult Inuit population would also be present in Inuit children.

Design and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Design and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Participants

A total of 319 Inuit and Danish children and adolescents were enrolled in the study. The Danish participants were recruited from the COPENHAGEN Puberty Study [7, 8]. One hundred and thirty-two healthy Danish children (boys; n = 62), aged 8.5-16.1 years, volunteered from three schools in the Copenhagen area. This group has been described in detail previously [9].

The 187 Inuit children and adolescents aged 5.7-17.1 years were recruited from public schools in Greenland: from the capital, Nuuk (55 boys and 47 girls), from Qaanaaq (31 boys and 40 girls), the northern most located village, and from Siorapaluk, the northern most located all-year inhabited village settlement in Greenland (12 boys and 2 girls, which included all the children in Siorapaluk who could perform a satisfactory maximal oxygen uptake test on cycle ergometer were recruited). In Nuuk all the children were recruited from the same school in the center of the town. There are approximately 1900 children in four public schools in Nuuk. All schools have children aged from 5 to 17. The number of children in each school is 450-550. In Qaanaaq there is approximately 160 children aged 5-17 in one single school. In Siorapaluk there are approximately 17 children aged 5-17 in one school.

All tests in Greenland were carried out in August. The Danish participants were tested from August to November. The same protocols, equipment, analytical methods, and examiners were involved in testing both the Danish and the Inuit children. The only girl reporting use of hormonal contraception was not excluded. None of the children reported any medical conditions that could affect the measurements.

All children arrived in the morning after 10 h of fasting. Blood samples were withdrawn and clinical examination including pubertal staging, anthropometrics, and blood pressure measurements were carried out. The children were offered a light meal, which was consumed at least 90 min before the maximal test.

Pubertal development

Pubertal development was evaluated according to Tanners Classification [10]. Testicular volume was estimated by palpation using Praders Orchidometer. The date of last menstrual bleeding was recorded in postmenarch girls. A trained pediatrician performed the pubertal classification.

Body composition

Body composition was evaluated by height, weight, BMI, skin fold thickness, and bioimpedance analysis (BIA) (Holtain Ltd., Crymych, UK) in all children. BMI Z-scores were calculated according to the CDC 2000 reference [11]. Children were classified as overweight or obese if the BMI Z-score was above the 85th percentile for age. In addition, the Danish children were whole body DEXA scanned (Hologic CDR 1000/W densitometer; Hologic Inc., Bedford, MA) with estimations of fat-free mass (FFM) and fat mass (FM). Sex-specific formulas for estimation of FFM were developed based on the Danish children with valid data from both DEXA and BIA. The impedance factor (ZI) was calculated as height squared divided by the impedance. Using linear regression analysis FFM was estimated by the following equations: boys: FFM = −0.79 + ZI × 0.44 + weight × 0.36 (r = 0.99). Girls: FFM = 0.27 + ZI × 0.39 + weight × 0.38 (r = 0.99). From estimations of FFM, FM (weight − FFM) and BF% (FM/weight) were calculated. These formulas were subsequently applied to all Danish and Inuit children. Skin fold thickness was measured with skin fold caliper (Harpenden, West Sussex, England) at the suprailiac, subscapular, triceps, and biceps location.

Blood pressure

Blood pressure was measured once in the left arm after a 10-min rest in supine position with a standard sphygmomanometer (Heine Gamma G5; Heine Optotechnik, Herrsching am Ammersee, Germany).

Blood samples

In the fasted state, venous blood was drawn from an antecubital vein and was subsequently centrifuged at 3000 rpm for 10 min and plasma was isolated and stored at −20°C.

Aerobic fitness

Maximal oxygen uptake (aerobic fitness was assessed during a progressive cycle ergometer test using an electronically braked cycle ergometer [Ergomedic 839; Monark, Varberg, Sweden]) calibrated on every test day. Oxygen uptake was measured directly using an online pulmonary gas analyzer system (Quark CPET; Cosmed, Rome, Italy). Heart rate was recorded continuously throughout the test using a heart rate monitor (Polar Electro, Oulu, Finland). The two different protocols used and criteria for a satisfactory maximal effort have been described previously [12, 13]. In brief, criterion for a satisfactory maximal effort using the online gas analyzer was a respiratory exchange ratio above 1.0.

The online gas analyzer was not brought to Siorapaluk, because the transport to the settlement only could be arranged with a whole-day trip in a small fishing boat. To include the Inuit (n = 35) and Danish children (n = 9) with a satisfactory maximal test, but without valid online gas analysis data, aerobic fitness was estimated from maximal power output as previously described in detail for the Danish children [9]. These formulas were applied to all the children in the study. In those Inuit children with valid direct measurements of aerobic fitness, the estimated values based on the abovementioned formulas [9] were correlated to the direct measurements in both Inuit boys (r = 0.749) and girls (r = 0.878) (both P < 0.001), respectively.

Analyses

Fasting blood samples were analyzed for glucose, insulin, total cholesterol (TC), high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol (LDL), triglyceride (TG), apolipoprotein A (ApoA), apolipoprotein B (ApoB), and hypersensitive C-reactive protein (HsCRP) and were all determined on automated Roche Modular Analytics (SWA) modules (Roche Diagnostics, Mannheim, Germany) by conventional assays (Elecsys insulin, GLU, CHOL, CFAS HDL/LDL-C plus, TG GPO-PAP, APO A-1, and APO B, version 2; Roche, Mannheim, Germany). This is described in a separate study on the Danish children [9]. Glycosylated hemoglobin (HbA1c) levels on Inuit and Danish children were analyzed using fingertip capillary blood with a Bayer DCA 2000+ (Bayer Healthcare, Elkhart, IN). Individuals with a HsCRP level above 2 mmol l−1 were excluded. All the samples from the Inuit and Danish children were analyzed using the same methods in the same laboratory. The blood and plasma samples from Greenland were stored and subsequently transported on dry ice to Copenhagen within a few days where they were stored at −80°C until analyzed.

Statistics

All statistical analyses were done using PSAW 18.0 for windows XP. Baseline characteristics are presented as means ± SD. Simple group comparisons were done by Student's t-tests. Pearson's correlations were used for simple correlations between the main outcome variables. Differences in prevalence of overweight and obesity between groups were evaluated by Fisher's exact tests. Differences in metabolic risk factors between northern and southern living Inuit children and Danish children were evaluated using general linear models with the individual metabolic risk factors included as dependent variables and the three groups (Danish, southern living Inuit, and northern living Inuit) included as an independent group variable. Analyses in girls and boys were done separately. In the first set of analyses, the associations between metabolic risk factors and place of living were adjusted for pubertal stage (group), age (continuous), as well as the interaction between the two (age × pubertal stage). The interaction between age and puberty was included to account for the differential effect of age on some metabolic parameters in different pubertal stages. For example, insulin is increasing with age in Tanner stage 1 and decreasing with age in Tanner stage 5. In the second set, additional adjustments were made for adiposity and aerobic fitness levels (both continuous variables). Insulin and TG levels needed log transformation to obtain approximate normal Gaussian distributions of the residuals as well as to obtain a residual variance that did not depend on the level. To present all variables on the same scale (relative instead of absolute differences [Figure 1]), the remaining dependent variables were also log transformed. This transformation did not change any of the statistical results and did not have any effect on the validity of the models.

image

Figure 1. Metabolic risk factors in Inuit and Danish children. All data are log transformed to show relative differences in percentage between southern and northern living Inuit and Danish children. Danish children are reference. Results are median and error bars are 95% confidence intervals. Boys: Danish, n = 62; South, n = 55; North, n = 43. Girls: Danish, n = 70; South, n = 47; North, n = 41.

Download figure to PowerPoint

Ethics

The study was performed in accordance with the Declaration of Helsinki II. The study protocols were approved by the ethics committee of the capital region (J.nr. KF 01 282214, KF 11 2006-2033) and the commission for scientific research in Greenland (J.nr 505-117). All children and their parents gave their informed written consent.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Design and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Differences in metabolic risk factors between Danish and Inuit children

Basic characteristic and unadjusted group comparisons are shown in Table 1. The prevalence of overweight and obesity was significantly higher in the southern living Inuit girls (31.9%), but not in northern living Inuit girls (7.9%) compared with the Danish girls (11.4%). Compared with Danish boys (12.9%), no significant differences in prevalence of overweight or obesity were found in southern (20.0%) or northern (18.6%) living Inuit boys.

Table 1. Characteristics of south and north living Inuit and Danish boys and girls
 BoysGirls
Danish, n = 62South, n = 55North, n = 43Danish, n = 70South, n = 47North, n = 42
  1. Unadjusted data and comparisons of anthropometric, blood pressure, and metabolic blood values between Danish and southern and northern living Inuit children. Results are means ± SD.

  2. a

    P < 0.05 Danish vs. south.

  3. b

    P < 0.05 Danish vs. north.

  4. c

    P < 0.05 north vs. south.

Pubertal stage (I–V) (%)34/24/11/11/1944/20/13/18/549/21/14/2/1414/11/19/43/1330/21/21/21/637/12/17/22/12
Age (years)12.4 ± 2.0a, b11.7 ± 2.010.9 ± 3.012.3 ± 2.1a, b11.3 ± 2.111.2 ± 2.7
Height (cm)157.5 ± 14.6a, b149.2 ± 12.9c140.5 ± 16.7155.3 ± 13.4a, b146.5 ± 12.4142.8 ± 13.4
Weight (kg)47.6 ± 13.6b43.3 ± 10.2c37.8 ± 15.345.7 ± 12.1b43.5 ± 13.2c38.2 ± 11.3
BMI (kg m−2)18.8 ± 2.719.2 ± 2.518.3 ± 3.018.6 ± 2.819.8 ± 3.4c18.3 ± 2.7
Sum of SF (mm)38.8 ± 16.8b38.5 ± 19.4c28.3 ± 11.538.7 ± 14.6a51.0 ± 25.0c35.8 ± 11.3
BF (%)20.1 ± 4.8b19.6 ± 5.518.2 ± 4.323.2 ± 4.124.1 ± 4.6c22.2 ± 4.0
VO2max (ml kg−1 min−1)46.2 ± 7.2b46.7 ± 6.0c49.4 ± 6.040.1 ± 5.7b39.2 ± 6.6c42.8 ± 5.3
Systolic BP (mm Hg)112 ± 11112 ± 12109 ± 12110 ± 9111 ± 11112 ± 10
Diastolic BP (mm Hg)64 ± 10a68 ± 765 ± 863 ± 9a, b69 ± 1069 ± 7
Glucose (mmol l−1)4.8 ± 0.5a, b5.3 ± 0.5c5.1 ± 0.44.7 ± 0.5a, b5.0 ± 0.45.0 ± 0.4
Insulin (pmol l−1)45 ± 21a55 ± 3044 ± 2957 ± 2865 ± 39c51 ± 25
HbA1c (mmol l−1)5.3 ± 0.2a5.4 ± 0.2c5.2 ± 0.25.2 ± 0.2a, b5.4 ± 0.3c5.1 ± 0.2
TC (mmol l−1)3.52 ± 0.49a, b4.04 ± 0.644.15 ± 0.673.78 ± 0.61a, b4.21 ± 0.714.15 ± 0.46
HDL (mmol l−1)1.48 ± 0.261.51 ± 0.301.39 ± 0.361.49 ± 0.33a1.37 ± 0.301.41 ± 0.30
LDL (mmol l−1)2.09 ± 0.442.25 ± 0.582.18 ± 0.552.24 ± 0.57a2.49 ± 0.67c2.15 ± 0.41
TG (mmol l−1)0.71 ± 0.34b0.74 ± 0.26c0.90 ± 0.410.87 ± 0.360.91 ± 0.320.84 ± 0.28
ApoA (μmol l−1)49.9 ± 6.0a, b59.6 ± 9.063.3 ± 11.550.7 ± 7.0a, b56.6 ± 9.8c65.3 ± 11.4
ApoB (μmol l−1)2.1 ± 0.4b2.3 ± 0.52.5 ± 0.62.3 ± 0.5a, b2.5 ± 0.62.5 ± 0.4
HsCRP (mg l−1)0.48 ± 0.450.40 ± 0.31c0.34 ± 0.30.41 ± 0.360.49 ± 0.480.55 ± 0.44

Because of the uneven distribution in age and pubertal stages between the three groups of children, differences in the metabolic risk factors between groups were adjusted for pubertal stage and age in general linear models, and presented as relative differences (Figure 1).

We evaluated all metabolic risk factors after additional adjustment for BF% and aerobic fitness. With the exception of the differences in plasma insulin and LDL levels between southern living Inuit and Danish girls as well as the differences in HDL levels between northern living Inuit and Danish boys, all the above mentioned findings (Figure 1) remained significantly different between groups after adjustment for BF% and aerobic fitness. Replacing the BIA-estimated BF% with total sum of skin folds in the models did not change any of the statistical outcomes.

Associations between the different metabolic parameters and BF%, and aerobic fitness were similar in the three groups for both boys and girls. The fluctuations in aerobic fitness, BF%, and metabolic parameters during the different pubertal stages did not vary significantly between the different groups.

Differences in metabolic risk factors between southern and northern living Inuit children

Unadjusted comparisons between the northern and southern living Inuit boys and girls, respectively, are shown in Table 1. Adjusted for pubertal stage and age, the southern living Inuit boys had significantly higher BMI, BF%, glucose, insulin, and HbA1c levels and significantly lower aerobic fitness, TG, and ApoA1 levels compared with northern living Inuit boys. The differences in HbA1c, TG, and ApoA1 levels remained significant after adjustment for total BF% and aerobic fitness. In girls, southern living Inuit had significantly higher levels of BMI, BF%, HbA1c, and LDL and significantly lower aerobic fitness and ApoA1 levels than northern living Inuit adjusted for age and pubertal stage. These differences remained significant after adjustment for total BF% and aerobic fitness. The prevalence of overweight and obesity was significantly higher in the southern living girls compared with northern living Inuit girls (31.9 vs. 7.3%). No significant differences in prevalence of overweight and obesity were found between southern and northern living boys (20.0 vs. 18.6%).

Inuit children from the north had only Inuit parents and grandparents, whereas 3 girls and 14 boys from the south had one parent or one grandparent of Danish decent. Excluding these from the analysis did not change any of the findings. Having more than one non-Inuit parent or grandparent was exclusion criteria.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Design and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

In this, the first study on metabolic risk factors in Greenlandic Inuit children, we found more adverse metabolic risk profiles in Inuit children compared with Danish children. The most adverse metabolic profiles were found in the southern living Inuit children, some of which were associated with a higher fatness and lower fitness in these children. However, even after adjustment for differences in aerobic fitness and fatness, fasting glucose, total cholesterol, apolipoproteinA1 levels, and diastolic blood pressure were still higher in both the southern and northern living Inuit children compared with Danish children.

A substantial decrease in work-related physical activity and an increase in the import and availability of western food products have been evident in Greenland during the last decades [14] and adaptation to a more westernized lifestyle has been associated with adverse metabolic risk profiles [15]. Modernization in developing economies has shown increasing prevalence of diabetes with urbanization [16]. However, studies from Greenland among adult Inuit have not been able to demonstrate a clear association between metabolic morbidity and urbanization [17, 18]. The observed differences in aerobic fitness and BF% could only partly explain the differences found between Inuit and Danish children. Thus, we cannot exclude that differences in habitual diet and genetics provide some explanation for the differences observed.

The comparison between the Inuit children showed that the southern living Inuit boys had higher BMI, BF%, glucose, insulin, and HbA1c and lower aerobic fitness and ApoA1 levels than the northern living Inuit boys. Similarly, the Inuit girls from the south had higher levels of BMI, BF%, HbA1c, and LDL and lower aerobic fitness and ApoA1 compared with the northern living Inuit girls. Although some of these differences, especially in boys, were associated with a higher BF% and a lower aerobic fitness, the intra-Inuit differences in metabolic factors did not seem to be solely explained by these factors. Genetic migration studies have shown that Inuit living in the most northern part descend from later migration waves from North America than the Inuit now living in the south [19]. Therefore, we cannot exclude that the differences seen between the northern and the southern living Inuit children may be partly influenced by genetic differences. Differences in the consumption of traditional Inuit marine and westernized diet may have affected the intra-Inuit differences.

The prevalence of overweight and obesity has increased, and the age at onset of obesity has decreased during recent decades in Inuit children living in Nuuk [20]. In this study, we used BMI classification defined for non-Inuit populations and found a high prevalence of overweight and obesity especially in southern living Inuit (31.9% among girls and 20.0% among boys) compared with Danish children (girls 11.4% and boys 12.9%). A recent study indicated that adult Inuit had a more favorable metabolic risk profile for a given level of BMI and waist circumference (WC) compared with a Danish reference population [23]. These results may indicate greater tolerance to overweight in adult Inuit; however, the relationship between adiposity and BMI and WC, respectively, may differ between different ethnic groups making comparisons based on these adiposity estimates difficult to interpret. In this study, there were no indications that the influence of adiposity on metabolic risk factors should be different between Danish and Inuit children.

The risk of metabolic morbidity increases markedly in the most obese children [24], and adverse metabolic risk in childhood seems to predict later occurrence of the metabolic syndrome [24, 25]. A study of Danish children demonstrated that boys had lower physical fitness and were more adipose in 1997-1998 than in 1985-1986. In addition, the investigators found that the difference between the fit and the unfit and the difference between the lean and the overweight children were greater in 1997-1998 than in 1985-1986 for both boys and girl [13]. Similar data are not available for Inuit children, but this trend may also occur in Inuit children in Greenland. Thus, identification and intervention for the group of overweight children are important, especially considering the observed changes in metabolic health in the Greenlandic Inuit population.

Strength and limitations

Some metabolic risk parameters change markedly during puberty. A major strength of the study is the thorough pubertal classification performed by trained paediatricians. Even though we used an online system to directly measure oxygen uptake, all the comparison analyses between the groups were done based on the maximal power output in order to include all Inuit children from the northern villages. Inuit are not accustomed to ride bikes. Therefore, underestimation of aerobic fitness in these children may have occurred. The equations used to estimate BF% were validated by DXA in Danish children. The single-frequency BIA equipment used in this study evaluates total body impedance, which is dependent on diameter and length of the body cylinders (extremities and truncus). Ethnic differences in anthropometrics may have resulted in incorrect estimates of FFM and FM in Inuit. Skin fold thickness as an estimation of body fat may also differ between Inuit and Danish children; however, the BIA-derived BF% was highly correlated with sum of skin fold thickness, and substituting the BF% with sum of skin fold thickness did not change any of the reported results. No data on socioeconomic status were available for the Greenlandic Inuit children. Therefore, differences in socioeconomic status may have played a role in the differences seen in this study. In addition, residual confounding in the adjustment for fitness and fatness should also been taken into consideration.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Design and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

In this study, we found a more adverse metabolic health profile in Greenlandic Inuit children compared with Danish children. The differences were more pronounced in the southern living Inuit. These findings may be linked to the adverse trend in health observed in the adult Greenlandic Inuit population.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Design and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References

We thank the Medical Center in Qaanaaq for housing and the use of their equipment. We are grateful to the participating children, schools, and the teachers in Nuuk, Qaanaaq, and Siorapaluk. We also thank Mogens Morgen for helping out with the planning of the project and Assoc Prof. Jim Cotter for assisting with the experiments in Nuuk.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Design and Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgments
  9. References