- Email: [email protected]
Prevalence of the metabolic syndrome among extremely obese adolescents in Italy and Germany
diabetes research and clinical practice 88 (2010) 14–21
Contents lists available at ScienceDirect
Diabetes Research and Clinical Practice jou rna l hom ep ag e: w ww.e lse v ier .com/ loca te /d iab res
Prevalence of the metabolic syndrome among extremely obese adolescents in Italy and Germany Claudio L. Lafortuna a,*, Fulvio Adorni b, Fiorenza Agosti c, Alessandra De Col c, Kolja Sievert e, Wolfgang Siegfried e, Alessandro Sartorio c,d a
Istituto di Bioimmagini e Fisiologia Molecolare, Consiglio Nazionale delle Ricerche, via Cervi, 93, I-20090 Segrate, Milano, Italy Istituto di Tecnologie Biomediche, Consiglio Nazionale delle Ricerche, Segrate, Milano, Italy c Istituto Auxologico Italiano, IRCCS, Laboratorio Sperimentale di Ricerche Auxo-endocrinologiche, Milano, Italy d Istituto Auxologico Italiano, IRCCS, Divisione di Auxologia, Piancavallo (VB), Italy e Adipositas Zentrum Insula, Bischofswiesen, Germany b
article info
abstract
Article history:
Juvenile metabolic syndrome (MetS) is a growing major medical problem in industrialised
Received 22 October 2009
countries. We estimated its prevalence among two similar clinic-based sequentially
Received in revised form
recruited cohorts of extremely obese adolescents (age: 12–18 years) from Italy (N = 665,
22 December 2009
males = 271, females = 394) and Germany (N = 661, males = 261, females = 400) using the
Accepted 4 January 2010
recent IDF paediatric criteria. The prevalence of the MetS was 23.3% among the Italians and
Published on line 21 January 2010
40.4% among the Germans. A multivariate logistic regression revealed an increased risk related to age (adjusted odd ratio (AOR): 2.24; 95% confidence interval (CI): 1.59–3.16;
Keywords:
p < 0.001), BMI SDS (AOR: 3.61; 95% CI: 2.33–5.60; p < 0.001), male gender (AOR: 2.36; 95%
Adolescent obesity
CI: 1.80–3.10; p < 0.001), and in German adolescents (AOR: 2.56; 95% CI: 1.98–3.31; p < 0.001).
Metabolic syndrome
Among Italian adolescents having the MetS, 83% had 3 abnormalities, 16% had 4 abnormal-
Multivariate binary logistic
ities while less than 1% had all the 5 abnormalities. In the German cohort, 67%, 28% and 5%
regression
of affected individuals had 3, 4 and 5 abnormalities, respectively. These results indicate that
European Union
MetS is highly prevalent among extremely obese adolescents, and suggest that (besides age, obesity and gender) national sociocultural factors, as alimentary trends, could be important. Further tools should be developed to understand international epidemiological differences concerning obesity and its comorbidities in relation to lifestyles in the countries of European Union. # 2010 Elsevier Ireland Ltd. All rights reserved.
1.
Introduction
Juvenile obesity is a major threat for health and an uncontrolled world-wide epidemic, considered an alarming key predictor for obesity in adulthood, whose increasing prevalence and severity have resulted in higher prevalence of comorbid conditions in children and adolescents [1]. In fact, as in adults, also in childhood and adolescence [2,3], obesity plays a central role in the development of the metabolic syndrome (MetS), an important clustering of metabolic abnormalities
and anthropometric characteristics entailing an increased risk for mortality from cardiovascular and all causes in adults [4,5], as well as an increase in type-2 diabetes and early cardiovascular disease in juvenile age [6]. The mechanisms underlying the relations between the physiological functions regulating the involved abnormalities (which include also high blood pressure, low levels of highdensity lipoprotein cholesterol, high triglycerides levels, high plasma glucose concentration) are not fully known. This prompted to a proliferation of clinical definitions of the MetS,
* Corresponding author. Tel.: +39 02 21717207; fax: +39 02 21717558. E-mail address: [email protected] (C.L. Lafortuna). 0168-8227/$ – see front matter # 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2010.01.008
diabetes research and clinical practice 88 (2010) 14–21
especially in children and adolescents, which ultimately led to the release by the International Diabetes Federation (IDF) of the consensus guidelines for defining and diagnosing the MetS in adults [7] and in the paediatric age [8], with the purpose to rationalise the existing definitions and conflicting opinions and to enable the estimate of global prevalence of the syndrome. Thus, this recent IDF consensus report for the definition of the MetS in children and adolescents [8] may be used in populations of this age group as a useful tool in facilitating surveillance for the syndrome and international comparisons of prevalence estimate, as well as in helping to unravel pathogenic mechanisms. Therefore, the objective of the present study was to employ this recent IDF paediatric definition of the MetS [8] in order to: (1) evaluate the prevalence of the MetS in two large samples of extremely obese treatment-seeking Caucasian adolescents from two European nations; (2) to assess the effect of age, obesity degree and gender on this prevalence among the two samples; and (3) to assess the prevalence of each component of the MetS and their predominant combinations.
2.
Patients and methods
2.1.
Subjects
Six-hundred-sixty-five Caucasian pubertal extremely obese patients (271 males and 394 females) admitted to the Auxology Division at Istituto Auxologico Italiano, Piancavallo, Italy, and 611 (261 males and 400 females) admitted at Adipositas Zentrum Insula, Bischofswiesen, Germany, were sequentially enrolled in the study, throughout a period of about 40 months. Both institutions include a medical health care division specifically devoted to the treatment of juvenile obesity, and patients were referred to by primary care practitioners in response to a request of medical help for obesity. The inclusion criteria into the study, agreed by the two institutions were: (a) Tanner stage > 3 [9]; (b) age 18 years; (c) diagnosis of extreme obesity (BMI 97th centile for age). No specific exclusion criterion was defined. Moreover, at the moment of the enrollment, no adolescent was following any controlled dietary regimen or structured protocol of physical activity. The Ethical Committee of both institutions approved the study and the nature of the investigation was explained to the patients and their relatives, who gave the written informed consent.
2.2.
15
standard tubes; triglycerides (TG) and high-density-lipoprotein cholesterol (HDL) as well as fasting plasma glucose (FPG) were immediately measured with enzymatic-colorimetric methods, after appropriate processing. At the time of their admission to the study, 1% of adolescents of the Italian cohort and 2% of the German one were non-insulin dependent diabetics while no Italian and 0.5% Germans were diabetics taking insulin; 1% were on a regimen of treatment for arterial hypertension among the Italians and 2% among the Germans. Table 1 shows mean and median values of the main characteristics of the subjects taking part to the study. According to the IDF criteria for MetS diagnosis in children and adolescents [8], the youths of the study were considered to have the MetS if they had abdominal obesity, AO (WC 90th percentile for ages < 16 years, and 94 cm for males and 80 cm for female for ages > 16 years) plus two or more of the following factors: (1) raised TG level: 150 mg/dL (1.7 mmol/L) for ages < 16 years and the same cutoff or specific treatment for this lipid abnormality for ages > 16 years; (2) reduced HDLcholesterol: <40 mg/dL (1.03 mmol/L) for males and females for ages <16 years, and <40 mg/dL for males and <50 mg/dL (1.29 mmol/L) for females, or specific treatment for this lipid abnormality for ages >16 years; (3) raised blood pressure (BP): systolic blood pressure 130 mmHg or diastolic blood pressure 85 mmHg for ages <16 years, and same cutoff or treatment of previously diagnosed hypertension for ages > 16 years; (4) raised FPG concentration: 100 mg/dL (5.6 mmol/L) or previously diagnosed type-2 diabetes for all ages.
2.3.
Statistical methods
All values are given as means SD. The effects of age, gender and nationality on the degree of obesity were evaluated within each cohort and in the pooled cohorts by means of two- and three-factor analysis of variance (ANOVA), as appropriate. The independent effects of age, degree of obesity, gender and nationality on the prevalence of the MetS were evaluated by means of a multivariate binary logistic regression analysis. To this purpose, the subjects were stratified into three groups for age (<14, 14–16, 16 years) and three groups for the degree of obesity (<3, 3–4, >4 BMI SDS). p-Values less than 0.05 were considered statistically significant. All the analyses were performed by means of the statistical software package SPSS 17 (SPSS Inc., Chicago, IL) for Windows.
Evaluation of subjects
3. At hospital admission, height and weight were measured respectively with a Harpenden stadiometer (Holtain Ltd., UK) and an electronic scale (Selus, UK), and BMI was calculated and expressed as standard deviation score (BMI SDS) for the subjects of each cohort using age and sex matched standards for the respective national population [10,11]. Waist circumference (WC) was measured according with international standards [12]. Two blood pressure (BP) determinations were performed to obtain the systolic and diastolic values using a conventional manometer after the patient had been sitting at least 15 min, and the mean values were used for analyses. Blood samples were collected after an overnight fast in
Results
Among the adolescents of both the Italian and German cohorts, the degree of obesity, as evaluated by the SDS of BMI, increased with age ( p < 0.001) and was higher ( p < 0.001) in girls than in boys (two-factor ANOVA). No significant effect was detected for the interaction of factors age, gender and nationality, when the two cohorts were pooled (three-factor ANOVA). Using the criteria of the new IDF definition of the MetS in children and adolescents [8], the overall prevalence among the obese adolescents of the Italian cohort was 23.3% and 40.4% in the German cohort. The prevalence estimated within the different strata of age and obesity degree is presented in
16
diabetes research and clinical practice 88 (2010) 14–21
Table 1 – The average characteristics of the obese boys and girls analysed in the study. Mean
SD
M
F
A
M
F
A
Median
Range
A
A
Italians (N = 665, males = 271, females = 394) Age (years) 15.2 Height (m) 1.69 Body weight (kg) 110.8 38.5 BMI (kg/m2) BMI (SDS) 3.14 Waist circumference (cm) 117.9 Triglycerides (mg/dL) 103.8 HDL-cholesterol (mg/dL) 43.5 Fasting plasma glucose (mg/dL) 77.3 Systolic blood pressure (mmHg) 128.4 Diastolic blood pressure (mmHg) 78.2
15.4 1.61 96.5 37.1 3.89 108.2 91.7 48.3 75.0 123.1 77.0
15.3 1.64 102.3 37.6 3.58 112.2 96.7 46.3 75.9 125.3 77.5
1.6 0.09 22.7 6.1 0.59 13.8 45.4 11.2 7.6 12.9 8.1
1.6 0.07 17.7 5.8 0.67 14.3 42.9 10.6 7.8 10.8 7.2
1.6 0.09 21.1 6.0 0.74 14.9 44.3 11.1 7.8 12.0 7.6
15.4 1.64 99.0 36.7 3.56 111.0 85.0 45.0 76.0 120.0 80.0
12.0–18.0 1.37–1.99 58.1–217.8 26.6–66.5 2.08–5.70 73.0–170.0 21–345 24–113 51 120 90–170 50–110
Germans (N = 661, males = 261, females = 400) Age (years) 15.2 Height (m) 1.76 Body weight (kg) 126.7 BMI (kg/m2) 40.8 BMI (SDS) 3.48 Waist circumference (cm) 127.8 Triglycerides (mg/dL) 134.0 HDL-cholesterol (mg/dL) 43.4 Fasting plasma glucose (mg/dL) 86.3 Systolic blood pressure (mmHg) 135.1 Diastolic blood pressure (mmHg) 82.4
15.2 1.67 113.0 40.2 3.76 119.1 121.1 46.8 87.1 131.2 80.0
15.2 1.71*** 118.4*** 40.5** 3.65 122.6*** 126.2*** 45.5 86.8*** 132.8*** 81.0***
1.5 0.09 25.3 7.1 0.49 15.9 69.8 10.1 10.8 16.9 13.2
1.5 0.07 21.7 7.1 0.56 15.1 58.6 11.2 17.4 16.4 12.6
1.5 0.09 24.1 7.1 0.55 16.0 63.5 10.9 15.1 16.7 12.9
15.2 1.70 114.6 39.4 3.62 121.0 113.0 44.0 85.0 130.0 80.0
12.0–18.0 1.50–2.01 71.2–213.2 26.5–73.2 2.11–5.43 84.0–180.0 31–578 23–106 60–292 88–180 41–124
M = males, F = females; and A = all. Student’s t-test to assess significance of the difference between Italian and German cohort. p < 0.001. *** p < 0.001. **
Table 2 for the two genders in the two cohorts. The independent risk for the MetS in the different strata of age, obesity degree and gender evaluated with a multivariate binary logistic regression is presented in Table 3 for the two
Table 2 – Prevalence (%) of the MetS for the different strata of age and obesity degree in boys and girls from the two investigated cohorts. Overall N Italians Total Age (years) <14 14–15.9 >16 BMI (SDS) <3 3–3.99 >4 Germans Total Age (years) <14 14–15.9 >16 BMI (SDS) <3 3–3.99 >4
Boys %
N
Girls %
N
%
665
23.3
271
29.5
394
19.0
157 262 246
9.6 22.5 32.9
68 114 89
17.8 34.4 41.6
89 148 157
12.1 15.0 28.0
157 302 206
20.4 17.9 33.5
116 132 23
25.0 28.0 60.9
41 170 183
7.3 10.0 30.1
661
40.4
261
45.6
400
37.0
172 278 211
36.0 32.7 54.0
70 109 82
44.3 39.4 54.9
102 169 129
30.4 28.4 53.5
79 423 159
20.3 38.5 55.3
42 185 34
23.8 47.0 64.7
37 238 125
16.2 31.9 52.8
cohorts separately, and in Table 4 for the two pooled cohorts, where the effect of nationality is also shown. All individuals participating in this study from both countries fulfilled the criterion for AO, whereas the other single components of the syndrome were present to a different degree. Raised BP was the most prevalent and raised FPG was the least prevalent abnormality observed in both samples. For all abnormalities, Germans had higher prevalence, comparatively with Italians (BP: 66.1% vs. 44.7%; HDL: 39.5% vs. 37.4%; TG: 24.5% vs. 10.8%; FPG: 7.6% vs. 0.6%). The prevalence of each metabolic abnormality in the different strata of age and BMI SDS is shown in Fig. 1 for both cohorts. Among the youths with the MetS, those with 3 metabolic abnormalities (19.4% of the Italians and 27.2% of the Germans) presented as leading combinations: (a) AO, high BP and low HDL in 60.6% of the Italians and in 39.3% of the Germans, (b) AO, high BP and raised TG, in 5.2% of the Italians and 15.4% of the Germans, and (c) AO, low HDL and raised TG, in 17.4% of the Italians and 7.1% of the Germans. Among those with 4 abnormalities (3.8% of the Italians and 11.2% of the Germans), the prevailing combination was AO, high BP, low HDL and raised TG in both Italians (14.2%) and Germans (21.0%), whereas AO, high BP, low HDL and raised FPG were consistently detected in Germans only (3.7%), being presented by only 0.6% of Italians; the remaining part of the adolescents with the MetS had all the 5 abnormalities (0.6% of the Italians and 4.9% of the Germans). Among the adolescents without the MetS, the predominant pattern was the two-factor combination of abdominal adiposity and high BP (33% of Italians and 49.2% of Germans).
17
diabetes research and clinical practice 88 (2010) 14–21
Table 3 – Effect of age, obesity degree and gender on the prevalence of the metabolic syndrome among obese adolescents of the Italian and German cohorts, assessed with multivariate binary logistic regression. N Italians Total Age (years) <14 14–16 >16 BMI (SDS) <3 3–4 >4 Gender Girls Boys Germans Total Age (years) <14 14–16 >16 BMI (SDS) <3 3–4 >4 Gender Girls Boys
AOR
CI (95% low)
CI (95% high)
p
665 157 262 246
1 2.777 5.344
1.411 2.745
5.464 10.405
0.003 <0.001
157 302 206
1 0.931 3.651
0.546 1.942
1.587 6.862
0.793 <0.001
394 271
1 3.750
2.307
6.095
<0.001
172 278 211
1 0.764 1.549
0.506 0.998
1.155 2.403
0.202 0.051
79 423 159
1 2.784 5.227
1.507 2.620
5.143 10.427
0.001 <0.001
400 261
1 1.806
1.282
2.546
0.001
661
N: number of individuals; AOR: adjusted odd ratio; and CI: confidence interval.
4.
Discussion
The present study estimated that, using the new definition for MetS in children and adolescents proposed by the IDF [8], approximately 23% of the extremely obese adolescents from Italy and 40% from Germany were classified as having the MetS, independent contributory effects being exerted by age, degree of obesity and male gender.
Due to several methodological factors, among which a considerable variability in criteria used to define MetS, and heterogeneity of age and obesity degree in evaluated samples, it is difficult to compare these results obtained in extremely obese adolescents in Italy and Germany with previous data available from these countries. In Italy, Calcaterra et al. [3], using the MetS definition by Weiss et al. [2], reported a prevalence of 24.6% in a cohort of
Table 4 – Effect of age, obesity degree, gender and nationality on the prevalence of the metabolic syndrome among obese adolescents of the Italian and German cohorts, assessed with multivariate binary logistic regression. N Overall Total Age (years) 14 14–16 >16 BMI (SDS) <3 3–4 >4 Gender Girls Boys Nationality Italy Germany
AOR
CI (95% low)
CI (95% high)
p
1326 329 540 457
1 1.164 2.244
0.831 1.594
1.630 3.158
0.378 <0.001
236 725 365
1 1.515 3.613
1.032 2.333
2.223 5.595
0.034 <0.001
794 532
1 2.361
1.796
3.104
<0.001
665 661
1 2.564
1.984
3.313
<0.001
N: number of individuals; AOR: adjusted odd ratio; and CI: confidence interval.
18
diabetes research and clinical practice 88 (2010) 14–21
Fig. 1 – Prevalence of individual metabolic abnormalities stratified by age (panels A and B) and by BMI SDS (panels C and D), among the two cohorts of obese adolescents. Bp: high blood pressure or treatment for diagnosed hypertension, HDL: low HDL-cholesterol or specific treatment for this abnormality, TG: high triglycerides or specific treatment for this abnormality, FPG: high fasting plasma glucose or diagnosed type-2 diabetes.
obese children and adolescents with BMI SDS above 2, and no MetS in another small cohort of normal weight control youths. Similarly, Invitti et al. [13] reported a prevalence of 23.3% (using the WHO modified definition) in a clinic-based sample of 6–16-year-old Italian obese boys and girls above the 97th percentile for age and sex, but they did not give any indication on MetS prevalence inside their group of nonobese controls. Analogous figures (12.5% for girls and 23.6% for males) were observed by Caranti et al. [14] in a cohort of Italian obese adolescents hospitalized for obesity treatment. Noto et al. [15] investigated the entire population of male children and adolescents aged 7–14 in a rural district of southern Italy and detected an overall prevalence of MetS of 0.6% using the IDF criteria and 2.2% using the Goodman criteria [16], but obesity prevalence was not estimated in that study. On the other side, in Germany, Reinehr et al. [17], explored the prevalence of the MetS in a clinic-based sample of overweight (including 80% of obese) 4–16-year-old Caucasian children and adolescents and reported a prevalence of 29%, 21% and 14% using the definition by Weiss et al. [2], Cook et al. [18] and IDF for adults [19], respectively, but it is not possible to evaluate the degree of obesity of the sample, since BMI SDS was not given in that study. None of the normal weight control children fulfilled the criteria for MetS. Bosy-Westphal et al. recently studied MetS traits in 90 German families including at
least one overweight or obese member over three generations and detected a MetS prevalence of 7% in boys and 8.1% in girls [20]. Thus, while the figure of prevalence we estimated in Italian obese adolescents matches previous findings [3,13,14], such a high MetS prevalence in German obese youths was never reported before. From these results, it also appeared a significant effect of the obesity degree on the prevalence of the MetS, in both national samples. Similarly, Weiss et al. [2] reported a significant obesitydependent trend of several metabolic characteristics, including also MetS components, in a cohort of U.S. obese children and adolescents. In fact, obesity has been described as the central causative factor in the development of the MetS due to the production by adipose cells of bioactive substances directly influencing insulin sensitivity and vascular injury [21,22]. The degree of obesity, evaluated through BMI percentile, has been found to contribute in a dramatic way to the development of the MetS in a large U.S. population study concerning adolescents aged 12–19 [18], prevalence rate ranging from 0.1% in individuals below the 85th percentile to 28.7% in those above the 95th percentile. In the cited study [18], no effect on prevalence was reported for age or pubertal stage, but the MetS was about 3 times more prevalent in males. On the contrary, among the extremely obese adolescents of the present study, and
diabetes research and clinical practice 88 (2010) 14–21
especially in Italy, an independent effect of age was also detected and further analysis seems required to explain this trend among obese individuals. The present investigation also revealed wide variations in overall prevalence of the single metabolic abnormalities which determine the MetS, as well as in their combinations in both nations. The combinations including high BP among other abnormalities are responsible for the MetS in 82% of Italians and in 87% of Germans, apparently due to the high prevalence of hypertension in these cohorts. In a recent study on a similar sample of Italian obese adolescents [23], we revealed by means of factor analysis that systolic and diastolic pressures are segregated together with obesity variables (BMI SDS, WC and waist-to-hip ratio) in males, and as a separate factor in females, without association with insulin resistance, hyperglycaemia and dyslipidemia, a similar finding being also reported in extremely obese women [24]. This should indicate that blood pressure regulation is chiefly mediated by other separate physiological processes [25] with different relevances among genders, despite the highly significant contribution of hypertension to the MetS. Remarkably, in spite of the extremely elevated degree of obesity displayed by the adolescents of both the cohorts investigated in the present study, hyperglycaemia was the abnormality with the lowest prevalence, although over 10 times more prevalent in Germans (7.6%) than in Italians (0.6%). Thus, 14.6% of the Germans and 2.6% of the Italians with the MetS presented combinations of abnormalities including hyperglycaemia. By contrast, we previously observed that 58% of obese adult women aged 18–83 with the MetS had combinations of abnormalities including hyperglycaemia occurring in significantly older patients with a prevalence increasing progressively in subsequent strata of age [26]. No age dependent prevalence of any combination of abnormalities was instead detected in the adolescents of the present study (results not shown). The mechanisms contributing to the combination of the different abnormalities of the MetS in the juvenile age, as in the adulthood, still appear unclear and require further analysis. Although the MetS-related risks for type-2 diabetes and early cardiovascular disease are now being defined also in the juvenile age and recognised to increase the future risk in adulthood [6], it is not known if, among individuals with the MetS, different combinations of abnormalities are necessarily associated also with different risks. In fact, at present, it is not possible to establish, among the single combinations, which set of abnormalities will evolve in future time toward more serious sequelae of the MetS or, conversely, which one is potentially more reversible on treatment. Both issues may have considerable importance in the strategy of treatment and prevention of the MetS in juvenile obesity. In fact, in spite of the high prevalence of the MetS observed in samples of obese youths from different countries [3,4,14,15,18], favourable changes in single components and reduction of the MetS prevalence have been reported in juvenile age in response to interventions by means of a lifestyle modification [27,28]. The multivariate analysis performed on the data of the present study actually takes into account possible differences in distribution within the multiple strata of age, obesity degree and gender among the subjects from the two nations. In spite
19
of this, some caution should be used in interpreting the differences observed between the two cohorts. Although both samples were from similar clinic-based series of adolescents seeking medical care for obesity, and had comparable degree of obesity, mean age and sex composition, these cohorts belong to populations with different anthropometric characteristics. In fact, according to the respective national reference charts, in the age range of this study, German boys and girls at the 50th percentile appear to be taller and to have higher BMI than their Italian counterpart [10,11,21], and it is difficult to ascertain the impact of these factors on the urge to seek medical help in relation to the actual clinical status. It should be also recalled that no survey performed with comparable methodology on national populations in order to assess obesity and MetS prevalence in youths is presently available in Italy and Germany, and an international comparison should be taken with caution. As a further element prompting to some circumspection in the interpretation of the national differences emerged from the study is the lack of information about the socio-economic composition of the two different participating cohorts. It has been in fact shown in a study on 97,721 adolescents from different European and North American countries that a lower socio-economic status is associated with less healthy behaviours in relation with physical activity and alimentary trends, and a poorer health perception [29]. Nonetheless, some factors may be potentially responsible for a different prevalence of the MetS among the obese adolescents from these two nations. Although genetic and ethnic factors have not been explored in the present study and may have a substantial influence, alimentary habits could play an important role. Different trends in dietary intake have been reported among children and adolescents in Europe [30], especially concerning north–south differences in patterns of fat assumption. Youths in Southern European countries, compared to those in Central and Northern countries, tended to report lower intake of saturated fat and higher consumption of monounsaturated fatty acids, whose major source is olive oil, a trend specifically confirmed in German boys and girls [31]. This pattern of low saturated/unsaturated fat ratio, is indeed one the central features of the Mediterranean-style dietary pattern, for which mounting epidemiological evidence indicates a favourable effect on obesity, diabetes and MetS, possibly due to the increase in antioxidant and anti-inflammatory capacity [32,33]. Thus, although no information was gained about specific dietary trends in the present study, one could speculate that difference in sociocultural factors may have affected also the alimentary habits toward differently healthy patterns among the two samples of adolescents from Italy and Germany. In conclusion, a high prevalence of MetS among extremely obese adolescents from a clinical-based context of two different European countries was detected. The remarkably higher prevalence observed among German boys and girls seems to be due to the increased prevalence of practically all the abnormalities concurring to the definition of the syndrome, but differences in the combinations of abnormalities have also been observed between the two nations, and further experimental research is needed to understand the pathophysiology of the different combinations. Differences in
20
diabetes research and clinical practice 88 (2010) 14–21
alimentary trends between the adolescents from the two countries could be hypothesised, but systematic information concerning the prevalence of obesity and MetS in relation to age and dietary habits in the different nations should be still gained and it is advisable that common strategies by national governments in the European Union should be coded and developed to this aim.
Conflict of interest There are no conflict of interest.
Funding This work was partially supported by Italian government funding for Progetti di Ricerca Corrente, from Ministero del Lavoro, della Salute e delle Politiche Sociali.
Acknowledgements The authors wish to thank Mrs. A. Seddone (head nurse) and the nursing staff at the Division of Auxology, Istituto Auxologico Italiano for having supported the study with their professional skilfulness.
references
[1] S.R. Daniels, M.S. Jacobson, B.W. McCrindle, R.H. Eckel, B.M. Sanner, American Heart Association Childhood Obesity Research Summit: executive summary, Circulation 119 (2009) 2114–2123. [2] R. Weiss, J. Dziura, T.S. Burgert, W.V. Tamborlane, S.E. Taksali, C.W. Yeckel, et al., Obesity and the metabolic syndrome in children and adolescents, N Engl J Med 350 (2004) 2362–2374. [3] V. Calcaterra, C. Klersy, T. Muratori, S. Telli, C. Caramagna, F. Scaglia, et al., Prevalence of metabolic syndrome (MS) in children and adolescents with varying degrees of obesity, Clin Endocrinol (Oxf.) 68 (2008) 868–872. [4] E.S. Ford, W.H. Giles, A.H. Mokdad, Increasing prevalence of the metabolic syndrome among U.S. adults, Diab Care 27 (2004) 2444–2449. [5] E.S. Ford, Risks for all-cause mortality, cardiovascular disease, and diabetes associated with the metabolic syndrome: a summary of the evidence, Diab Care 28 (2005) 1769–1778. [6] B.M. Nathan, A. Moran, Metabolic complications of obesity in childhood and adolescence: more than just diabetes, Curr Opin Endocrinol Diab Obes 15 (2008) 21–29. [7] K.G. Alberti, P. Zimmet, J. Shaw, Metabolic syndrome—a new world-wide definition: a Consensus Statement from the International Diabetes Federation, Diab Med 23 (2006) 469–480. [8] P. Zimmet, K.G. Alberti, F. Kaufman, N. Tajima, M. Silink, S. Arslanian, et al.,IDF Consensus Group, The metabolic syndrome in children and adolescents—an IDF consensus report, Pediatr Diab 8 (2007) 299–306. [9] J.M. Tanner, Growth at Adolescence, 2nd ed., Blackwell Scientific Publications, Oxford, 1962.
[10] E. Cacciari, S. Milani, A. Balsamo, F. Dammacco, F. De Luca, F. Chiarelli, et al., Italian cross-section charts for height, weight and BMI (6–20 y), Eur J Clin Nutr 56 (2002) 171–180. [11] F. Schaefer, M. Georgi, E. Wu¨hl, K. Scha¨rer, Body mass index and percentage fat mass in healthy German schoolchildren and adolescents, Int J Obes Relat Metab Disord 22 (1998) 461–469. [12] World Health Organization, 1987. Measuring obesity— Classification and description of anthropometric data. Report on a WHO consultation on the epidemiology of obesity, EUR/ICP/NUT 125. WHO regional office for Europe, Warsaw. [13] C. Invitti, C. Maffeis, L. Gilardini, B. Pontiggia, G. Mazzilli, A. Girola, et al., Metabolic syndrome in obese Caucasian children: prevalence using WHO-derived criteria and association with nontraditional cardiovascular risk factors, Int J Obes (Lond.) 30 (2006) 627–633. [14] D.A. Caranti, S. Lazzer, A.R. Daˆmaso, F. Agosti, R. Zennaro, M.T. de Mello, et al., Prevalence and risk factors of metabolic syndrome in Brazilian and Italian obese adolescents: a comparison study, Int J Clin Pract 62 (2008) 1526–1532. [15] D. Noto, T. Niglio, A.B. Cefalu`, E. Martino, F. Fayer, M. Mina, et al., Obesity and the metabolic syndrome in a student cohort from Southern Italy, Nutr Metab Cardiovasc Dis 19 (2009) 620–625. [16] E. Goodman, S.R. Daniels, J.B. Meigs, L.M. Dolan, Instability in the diagnosis of metabolic syndrome in adolescents, Circulation 115 (2007) 2316–2322. [17] T. Reinehr, G. de Sousa, A.M. Toschke, W. Andler, Comparison of metabolic syndrome prevalence using eight different definitions: a critical approach, Arch Dis Child 92 (2007) 1067–1072. [18] S. Cook, M. Weitzman, P. Auinger, M. Nguyen, W.H. Dietz, Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 1988–1994, Arch Pediatr Adolesc Med 157 (2003) 821–827. [19] K.G. Alberti, P. Zimmet, J. Shaw, IDF Epidemiology Task Force Consensus Group. The metabolic syndrome—a new worldwide definition, Lancet 366 (2005) 1059–1062. [20] A. Bosy-Westphal, S. Onur, C. Geisler, A. Wolf, O. Korth, M. Pfeuffer, et al., Common familial influences on clustering of metabolic syndrome traits with central obesity and insulin resistance: the Kiel obesity prevention study, Int J Obes (Lond.) 31 (2007) 784–790. [21] S. Furukawa, T. Fujita, M. Shimabukuro, M. Iwaki, Y. Yamada, Y. Nakajima, et al., Increased oxidative stress in obesity and its impact on metabolic syndrome, J Clin Invest 114 (2004) 1752–1761. [22] L. Hutley, J.B. Prins, Fat as an endocrine organ: relationship to the metabolic syndrome, Am J Med Sci 330 (2005) 280– 289. [23] C.L. Lafortuna, F. Adorni, F. Agosti, A. De Col, R. Zennaro, D. Caranti, et al., Factor analysis of metabolic syndrome components in severely obese girls and boys, J Endocrinol Invest 32 (2009) 552–558. [24] C.L. Lafortuna, F. Adorni, F. Agosti, A. Sartorio, Factor analysis of metabolic syndrome components in obese women, Nutr Metab Cardiovasc Dis 18 (2008) 233–241. [25] G.M. Reaven, H. Lithell, L. Landsberg, Hypertension and associated metabolic abnormalities: the role of insulin resistance and the sympathoadrenal system, N Engl J Med 334 (1996) 374–381. [26] A. Sartorio, F. Agosti, F. Adorni, F. Pera, C.L. Lafortuna, Effect of age, degree and distribution of adiposity on the prevalence of the metabolic syndrome in a cohort of obese Italian women, Diab Res Clin Pract 78 (2007) 225–233.
diabetes research and clinical practice 88 (2010) 14–21
[27] T. Reinehr, G. de Sousa, M. Wabitsch, Changes of cardiovascular risk factors in obese children effects of inpatient and outpatient interventions, J Pediatr Gastroenterol Nutr 43 (2006) 506–511. [28] T. Reinehr, M. Kleber, A.M. Toschke, Lifestyle intervention in obese children is associated with a decrease of the metabolic syndrome prevalence, Atherosclerosis 207 (2009) 174–180. [29] M. Richter, M. Erhart, C.A. Vereecken, A. Zambon, W. Boyce, S. Nic Gabhainn, The role of behavioural factors in explaining socio-economic differences in adolescent health: a multilevel study in 33 countries, Soc Sci Med 69 (2009) 396–403.
21
[30] J. Lambert, C. Agostoni, I. Elmadfa, K. Hulshof, E. Krause, B. Livingstone, et al., Dietary intake and nutritional status of children and adolescents in Europe, Br J Nutr 92 (Suppl. 2) (2004) S147–211. [31] M.F. Rolland-Cachera, F. Bellisle, M. Deheeger, Nutritional status and food intake in adolescents living in Western Europe, Eur J Clin Nutr 54 (Suppl. 1) (2000) S41–46. [32] D. Giugliano, K. Esposito, Mediterranean diet and metabolic diseases, Curr Opin Lipidol 19 (2008) 63–68. [33] L. Serra-Majem, B. Roman, R. Estruch, Scientific evidence of interventions using the Mediterranean diet: a systematic review, Nutr Rev 64 (Suppl. 1) (2006) S27–47.
Contents lists available at ScienceDirect
Diabetes Research and Clinical Practice jou rna l hom ep ag e: w ww.e lse v ier .com/ loca te /d iab res
Prevalence of the metabolic syndrome among extremely obese adolescents in Italy and Germany Claudio L. Lafortuna a,*, Fulvio Adorni b, Fiorenza Agosti c, Alessandra De Col c, Kolja Sievert e, Wolfgang Siegfried e, Alessandro Sartorio c,d a
Istituto di Bioimmagini e Fisiologia Molecolare, Consiglio Nazionale delle Ricerche, via Cervi, 93, I-20090 Segrate, Milano, Italy Istituto di Tecnologie Biomediche, Consiglio Nazionale delle Ricerche, Segrate, Milano, Italy c Istituto Auxologico Italiano, IRCCS, Laboratorio Sperimentale di Ricerche Auxo-endocrinologiche, Milano, Italy d Istituto Auxologico Italiano, IRCCS, Divisione di Auxologia, Piancavallo (VB), Italy e Adipositas Zentrum Insula, Bischofswiesen, Germany b
article info
abstract
Article history:
Juvenile metabolic syndrome (MetS) is a growing major medical problem in industrialised
Received 22 October 2009
countries. We estimated its prevalence among two similar clinic-based sequentially
Received in revised form
recruited cohorts of extremely obese adolescents (age: 12–18 years) from Italy (N = 665,
22 December 2009
males = 271, females = 394) and Germany (N = 661, males = 261, females = 400) using the
Accepted 4 January 2010
recent IDF paediatric criteria. The prevalence of the MetS was 23.3% among the Italians and
Published on line 21 January 2010
40.4% among the Germans. A multivariate logistic regression revealed an increased risk related to age (adjusted odd ratio (AOR): 2.24; 95% confidence interval (CI): 1.59–3.16;
Keywords:
p < 0.001), BMI SDS (AOR: 3.61; 95% CI: 2.33–5.60; p < 0.001), male gender (AOR: 2.36; 95%
Adolescent obesity
CI: 1.80–3.10; p < 0.001), and in German adolescents (AOR: 2.56; 95% CI: 1.98–3.31; p < 0.001).
Metabolic syndrome
Among Italian adolescents having the MetS, 83% had 3 abnormalities, 16% had 4 abnormal-
Multivariate binary logistic
ities while less than 1% had all the 5 abnormalities. In the German cohort, 67%, 28% and 5%
regression
of affected individuals had 3, 4 and 5 abnormalities, respectively. These results indicate that
European Union
MetS is highly prevalent among extremely obese adolescents, and suggest that (besides age, obesity and gender) national sociocultural factors, as alimentary trends, could be important. Further tools should be developed to understand international epidemiological differences concerning obesity and its comorbidities in relation to lifestyles in the countries of European Union. # 2010 Elsevier Ireland Ltd. All rights reserved.
1.
Introduction
Juvenile obesity is a major threat for health and an uncontrolled world-wide epidemic, considered an alarming key predictor for obesity in adulthood, whose increasing prevalence and severity have resulted in higher prevalence of comorbid conditions in children and adolescents [1]. In fact, as in adults, also in childhood and adolescence [2,3], obesity plays a central role in the development of the metabolic syndrome (MetS), an important clustering of metabolic abnormalities
and anthropometric characteristics entailing an increased risk for mortality from cardiovascular and all causes in adults [4,5], as well as an increase in type-2 diabetes and early cardiovascular disease in juvenile age [6]. The mechanisms underlying the relations between the physiological functions regulating the involved abnormalities (which include also high blood pressure, low levels of highdensity lipoprotein cholesterol, high triglycerides levels, high plasma glucose concentration) are not fully known. This prompted to a proliferation of clinical definitions of the MetS,
* Corresponding author. Tel.: +39 02 21717207; fax: +39 02 21717558. E-mail address: [email protected] (C.L. Lafortuna). 0168-8227/$ – see front matter # 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.diabres.2010.01.008
diabetes research and clinical practice 88 (2010) 14–21
especially in children and adolescents, which ultimately led to the release by the International Diabetes Federation (IDF) of the consensus guidelines for defining and diagnosing the MetS in adults [7] and in the paediatric age [8], with the purpose to rationalise the existing definitions and conflicting opinions and to enable the estimate of global prevalence of the syndrome. Thus, this recent IDF consensus report for the definition of the MetS in children and adolescents [8] may be used in populations of this age group as a useful tool in facilitating surveillance for the syndrome and international comparisons of prevalence estimate, as well as in helping to unravel pathogenic mechanisms. Therefore, the objective of the present study was to employ this recent IDF paediatric definition of the MetS [8] in order to: (1) evaluate the prevalence of the MetS in two large samples of extremely obese treatment-seeking Caucasian adolescents from two European nations; (2) to assess the effect of age, obesity degree and gender on this prevalence among the two samples; and (3) to assess the prevalence of each component of the MetS and their predominant combinations.
2.
Patients and methods
2.1.
Subjects
Six-hundred-sixty-five Caucasian pubertal extremely obese patients (271 males and 394 females) admitted to the Auxology Division at Istituto Auxologico Italiano, Piancavallo, Italy, and 611 (261 males and 400 females) admitted at Adipositas Zentrum Insula, Bischofswiesen, Germany, were sequentially enrolled in the study, throughout a period of about 40 months. Both institutions include a medical health care division specifically devoted to the treatment of juvenile obesity, and patients were referred to by primary care practitioners in response to a request of medical help for obesity. The inclusion criteria into the study, agreed by the two institutions were: (a) Tanner stage > 3 [9]; (b) age 18 years; (c) diagnosis of extreme obesity (BMI 97th centile for age). No specific exclusion criterion was defined. Moreover, at the moment of the enrollment, no adolescent was following any controlled dietary regimen or structured protocol of physical activity. The Ethical Committee of both institutions approved the study and the nature of the investigation was explained to the patients and their relatives, who gave the written informed consent.
2.2.
15
standard tubes; triglycerides (TG) and high-density-lipoprotein cholesterol (HDL) as well as fasting plasma glucose (FPG) were immediately measured with enzymatic-colorimetric methods, after appropriate processing. At the time of their admission to the study, 1% of adolescents of the Italian cohort and 2% of the German one were non-insulin dependent diabetics while no Italian and 0.5% Germans were diabetics taking insulin; 1% were on a regimen of treatment for arterial hypertension among the Italians and 2% among the Germans. Table 1 shows mean and median values of the main characteristics of the subjects taking part to the study. According to the IDF criteria for MetS diagnosis in children and adolescents [8], the youths of the study were considered to have the MetS if they had abdominal obesity, AO (WC 90th percentile for ages < 16 years, and 94 cm for males and 80 cm for female for ages > 16 years) plus two or more of the following factors: (1) raised TG level: 150 mg/dL (1.7 mmol/L) for ages < 16 years and the same cutoff or specific treatment for this lipid abnormality for ages > 16 years; (2) reduced HDLcholesterol: <40 mg/dL (1.03 mmol/L) for males and females for ages <16 years, and <40 mg/dL for males and <50 mg/dL (1.29 mmol/L) for females, or specific treatment for this lipid abnormality for ages >16 years; (3) raised blood pressure (BP): systolic blood pressure 130 mmHg or diastolic blood pressure 85 mmHg for ages <16 years, and same cutoff or treatment of previously diagnosed hypertension for ages > 16 years; (4) raised FPG concentration: 100 mg/dL (5.6 mmol/L) or previously diagnosed type-2 diabetes for all ages.
2.3.
Statistical methods
All values are given as means SD. The effects of age, gender and nationality on the degree of obesity were evaluated within each cohort and in the pooled cohorts by means of two- and three-factor analysis of variance (ANOVA), as appropriate. The independent effects of age, degree of obesity, gender and nationality on the prevalence of the MetS were evaluated by means of a multivariate binary logistic regression analysis. To this purpose, the subjects were stratified into three groups for age (<14, 14–16, 16 years) and three groups for the degree of obesity (<3, 3–4, >4 BMI SDS). p-Values less than 0.05 were considered statistically significant. All the analyses were performed by means of the statistical software package SPSS 17 (SPSS Inc., Chicago, IL) for Windows.
Evaluation of subjects
3. At hospital admission, height and weight were measured respectively with a Harpenden stadiometer (Holtain Ltd., UK) and an electronic scale (Selus, UK), and BMI was calculated and expressed as standard deviation score (BMI SDS) for the subjects of each cohort using age and sex matched standards for the respective national population [10,11]. Waist circumference (WC) was measured according with international standards [12]. Two blood pressure (BP) determinations were performed to obtain the systolic and diastolic values using a conventional manometer after the patient had been sitting at least 15 min, and the mean values were used for analyses. Blood samples were collected after an overnight fast in
Results
Among the adolescents of both the Italian and German cohorts, the degree of obesity, as evaluated by the SDS of BMI, increased with age ( p < 0.001) and was higher ( p < 0.001) in girls than in boys (two-factor ANOVA). No significant effect was detected for the interaction of factors age, gender and nationality, when the two cohorts were pooled (three-factor ANOVA). Using the criteria of the new IDF definition of the MetS in children and adolescents [8], the overall prevalence among the obese adolescents of the Italian cohort was 23.3% and 40.4% in the German cohort. The prevalence estimated within the different strata of age and obesity degree is presented in
16
diabetes research and clinical practice 88 (2010) 14–21
Table 1 – The average characteristics of the obese boys and girls analysed in the study. Mean
SD
M
F
A
M
F
A
Median
Range
A
A
Italians (N = 665, males = 271, females = 394) Age (years) 15.2 Height (m) 1.69 Body weight (kg) 110.8 38.5 BMI (kg/m2) BMI (SDS) 3.14 Waist circumference (cm) 117.9 Triglycerides (mg/dL) 103.8 HDL-cholesterol (mg/dL) 43.5 Fasting plasma glucose (mg/dL) 77.3 Systolic blood pressure (mmHg) 128.4 Diastolic blood pressure (mmHg) 78.2
15.4 1.61 96.5 37.1 3.89 108.2 91.7 48.3 75.0 123.1 77.0
15.3 1.64 102.3 37.6 3.58 112.2 96.7 46.3 75.9 125.3 77.5
1.6 0.09 22.7 6.1 0.59 13.8 45.4 11.2 7.6 12.9 8.1
1.6 0.07 17.7 5.8 0.67 14.3 42.9 10.6 7.8 10.8 7.2
1.6 0.09 21.1 6.0 0.74 14.9 44.3 11.1 7.8 12.0 7.6
15.4 1.64 99.0 36.7 3.56 111.0 85.0 45.0 76.0 120.0 80.0
12.0–18.0 1.37–1.99 58.1–217.8 26.6–66.5 2.08–5.70 73.0–170.0 21–345 24–113 51 120 90–170 50–110
Germans (N = 661, males = 261, females = 400) Age (years) 15.2 Height (m) 1.76 Body weight (kg) 126.7 BMI (kg/m2) 40.8 BMI (SDS) 3.48 Waist circumference (cm) 127.8 Triglycerides (mg/dL) 134.0 HDL-cholesterol (mg/dL) 43.4 Fasting plasma glucose (mg/dL) 86.3 Systolic blood pressure (mmHg) 135.1 Diastolic blood pressure (mmHg) 82.4
15.2 1.67 113.0 40.2 3.76 119.1 121.1 46.8 87.1 131.2 80.0
15.2 1.71*** 118.4*** 40.5** 3.65 122.6*** 126.2*** 45.5 86.8*** 132.8*** 81.0***
1.5 0.09 25.3 7.1 0.49 15.9 69.8 10.1 10.8 16.9 13.2
1.5 0.07 21.7 7.1 0.56 15.1 58.6 11.2 17.4 16.4 12.6
1.5 0.09 24.1 7.1 0.55 16.0 63.5 10.9 15.1 16.7 12.9
15.2 1.70 114.6 39.4 3.62 121.0 113.0 44.0 85.0 130.0 80.0
12.0–18.0 1.50–2.01 71.2–213.2 26.5–73.2 2.11–5.43 84.0–180.0 31–578 23–106 60–292 88–180 41–124
M = males, F = females; and A = all. Student’s t-test to assess significance of the difference between Italian and German cohort. p < 0.001. *** p < 0.001. **
Table 2 for the two genders in the two cohorts. The independent risk for the MetS in the different strata of age, obesity degree and gender evaluated with a multivariate binary logistic regression is presented in Table 3 for the two
Table 2 – Prevalence (%) of the MetS for the different strata of age and obesity degree in boys and girls from the two investigated cohorts. Overall N Italians Total Age (years) <14 14–15.9 >16 BMI (SDS) <3 3–3.99 >4 Germans Total Age (years) <14 14–15.9 >16 BMI (SDS) <3 3–3.99 >4
Boys %
N
Girls %
N
%
665
23.3
271
29.5
394
19.0
157 262 246
9.6 22.5 32.9
68 114 89
17.8 34.4 41.6
89 148 157
12.1 15.0 28.0
157 302 206
20.4 17.9 33.5
116 132 23
25.0 28.0 60.9
41 170 183
7.3 10.0 30.1
661
40.4
261
45.6
400
37.0
172 278 211
36.0 32.7 54.0
70 109 82
44.3 39.4 54.9
102 169 129
30.4 28.4 53.5
79 423 159
20.3 38.5 55.3
42 185 34
23.8 47.0 64.7
37 238 125
16.2 31.9 52.8
cohorts separately, and in Table 4 for the two pooled cohorts, where the effect of nationality is also shown. All individuals participating in this study from both countries fulfilled the criterion for AO, whereas the other single components of the syndrome were present to a different degree. Raised BP was the most prevalent and raised FPG was the least prevalent abnormality observed in both samples. For all abnormalities, Germans had higher prevalence, comparatively with Italians (BP: 66.1% vs. 44.7%; HDL: 39.5% vs. 37.4%; TG: 24.5% vs. 10.8%; FPG: 7.6% vs. 0.6%). The prevalence of each metabolic abnormality in the different strata of age and BMI SDS is shown in Fig. 1 for both cohorts. Among the youths with the MetS, those with 3 metabolic abnormalities (19.4% of the Italians and 27.2% of the Germans) presented as leading combinations: (a) AO, high BP and low HDL in 60.6% of the Italians and in 39.3% of the Germans, (b) AO, high BP and raised TG, in 5.2% of the Italians and 15.4% of the Germans, and (c) AO, low HDL and raised TG, in 17.4% of the Italians and 7.1% of the Germans. Among those with 4 abnormalities (3.8% of the Italians and 11.2% of the Germans), the prevailing combination was AO, high BP, low HDL and raised TG in both Italians (14.2%) and Germans (21.0%), whereas AO, high BP, low HDL and raised FPG were consistently detected in Germans only (3.7%), being presented by only 0.6% of Italians; the remaining part of the adolescents with the MetS had all the 5 abnormalities (0.6% of the Italians and 4.9% of the Germans). Among the adolescents without the MetS, the predominant pattern was the two-factor combination of abdominal adiposity and high BP (33% of Italians and 49.2% of Germans).
17
diabetes research and clinical practice 88 (2010) 14–21
Table 3 – Effect of age, obesity degree and gender on the prevalence of the metabolic syndrome among obese adolescents of the Italian and German cohorts, assessed with multivariate binary logistic regression. N Italians Total Age (years) <14 14–16 >16 BMI (SDS) <3 3–4 >4 Gender Girls Boys Germans Total Age (years) <14 14–16 >16 BMI (SDS) <3 3–4 >4 Gender Girls Boys
AOR
CI (95% low)
CI (95% high)
p
665 157 262 246
1 2.777 5.344
1.411 2.745
5.464 10.405
0.003 <0.001
157 302 206
1 0.931 3.651
0.546 1.942
1.587 6.862
0.793 <0.001
394 271
1 3.750
2.307
6.095
<0.001
172 278 211
1 0.764 1.549
0.506 0.998
1.155 2.403
0.202 0.051
79 423 159
1 2.784 5.227
1.507 2.620
5.143 10.427
0.001 <0.001
400 261
1 1.806
1.282
2.546
0.001
661
N: number of individuals; AOR: adjusted odd ratio; and CI: confidence interval.
4.
Discussion
The present study estimated that, using the new definition for MetS in children and adolescents proposed by the IDF [8], approximately 23% of the extremely obese adolescents from Italy and 40% from Germany were classified as having the MetS, independent contributory effects being exerted by age, degree of obesity and male gender.
Due to several methodological factors, among which a considerable variability in criteria used to define MetS, and heterogeneity of age and obesity degree in evaluated samples, it is difficult to compare these results obtained in extremely obese adolescents in Italy and Germany with previous data available from these countries. In Italy, Calcaterra et al. [3], using the MetS definition by Weiss et al. [2], reported a prevalence of 24.6% in a cohort of
Table 4 – Effect of age, obesity degree, gender and nationality on the prevalence of the metabolic syndrome among obese adolescents of the Italian and German cohorts, assessed with multivariate binary logistic regression. N Overall Total Age (years) 14 14–16 >16 BMI (SDS) <3 3–4 >4 Gender Girls Boys Nationality Italy Germany
AOR
CI (95% low)
CI (95% high)
p
1326 329 540 457
1 1.164 2.244
0.831 1.594
1.630 3.158
0.378 <0.001
236 725 365
1 1.515 3.613
1.032 2.333
2.223 5.595
0.034 <0.001
794 532
1 2.361
1.796
3.104
<0.001
665 661
1 2.564
1.984
3.313
<0.001
N: number of individuals; AOR: adjusted odd ratio; and CI: confidence interval.
18
diabetes research and clinical practice 88 (2010) 14–21
Fig. 1 – Prevalence of individual metabolic abnormalities stratified by age (panels A and B) and by BMI SDS (panels C and D), among the two cohorts of obese adolescents. Bp: high blood pressure or treatment for diagnosed hypertension, HDL: low HDL-cholesterol or specific treatment for this abnormality, TG: high triglycerides or specific treatment for this abnormality, FPG: high fasting plasma glucose or diagnosed type-2 diabetes.
obese children and adolescents with BMI SDS above 2, and no MetS in another small cohort of normal weight control youths. Similarly, Invitti et al. [13] reported a prevalence of 23.3% (using the WHO modified definition) in a clinic-based sample of 6–16-year-old Italian obese boys and girls above the 97th percentile for age and sex, but they did not give any indication on MetS prevalence inside their group of nonobese controls. Analogous figures (12.5% for girls and 23.6% for males) were observed by Caranti et al. [14] in a cohort of Italian obese adolescents hospitalized for obesity treatment. Noto et al. [15] investigated the entire population of male children and adolescents aged 7–14 in a rural district of southern Italy and detected an overall prevalence of MetS of 0.6% using the IDF criteria and 2.2% using the Goodman criteria [16], but obesity prevalence was not estimated in that study. On the other side, in Germany, Reinehr et al. [17], explored the prevalence of the MetS in a clinic-based sample of overweight (including 80% of obese) 4–16-year-old Caucasian children and adolescents and reported a prevalence of 29%, 21% and 14% using the definition by Weiss et al. [2], Cook et al. [18] and IDF for adults [19], respectively, but it is not possible to evaluate the degree of obesity of the sample, since BMI SDS was not given in that study. None of the normal weight control children fulfilled the criteria for MetS. Bosy-Westphal et al. recently studied MetS traits in 90 German families including at
least one overweight or obese member over three generations and detected a MetS prevalence of 7% in boys and 8.1% in girls [20]. Thus, while the figure of prevalence we estimated in Italian obese adolescents matches previous findings [3,13,14], such a high MetS prevalence in German obese youths was never reported before. From these results, it also appeared a significant effect of the obesity degree on the prevalence of the MetS, in both national samples. Similarly, Weiss et al. [2] reported a significant obesitydependent trend of several metabolic characteristics, including also MetS components, in a cohort of U.S. obese children and adolescents. In fact, obesity has been described as the central causative factor in the development of the MetS due to the production by adipose cells of bioactive substances directly influencing insulin sensitivity and vascular injury [21,22]. The degree of obesity, evaluated through BMI percentile, has been found to contribute in a dramatic way to the development of the MetS in a large U.S. population study concerning adolescents aged 12–19 [18], prevalence rate ranging from 0.1% in individuals below the 85th percentile to 28.7% in those above the 95th percentile. In the cited study [18], no effect on prevalence was reported for age or pubertal stage, but the MetS was about 3 times more prevalent in males. On the contrary, among the extremely obese adolescents of the present study, and
diabetes research and clinical practice 88 (2010) 14–21
especially in Italy, an independent effect of age was also detected and further analysis seems required to explain this trend among obese individuals. The present investigation also revealed wide variations in overall prevalence of the single metabolic abnormalities which determine the MetS, as well as in their combinations in both nations. The combinations including high BP among other abnormalities are responsible for the MetS in 82% of Italians and in 87% of Germans, apparently due to the high prevalence of hypertension in these cohorts. In a recent study on a similar sample of Italian obese adolescents [23], we revealed by means of factor analysis that systolic and diastolic pressures are segregated together with obesity variables (BMI SDS, WC and waist-to-hip ratio) in males, and as a separate factor in females, without association with insulin resistance, hyperglycaemia and dyslipidemia, a similar finding being also reported in extremely obese women [24]. This should indicate that blood pressure regulation is chiefly mediated by other separate physiological processes [25] with different relevances among genders, despite the highly significant contribution of hypertension to the MetS. Remarkably, in spite of the extremely elevated degree of obesity displayed by the adolescents of both the cohorts investigated in the present study, hyperglycaemia was the abnormality with the lowest prevalence, although over 10 times more prevalent in Germans (7.6%) than in Italians (0.6%). Thus, 14.6% of the Germans and 2.6% of the Italians with the MetS presented combinations of abnormalities including hyperglycaemia. By contrast, we previously observed that 58% of obese adult women aged 18–83 with the MetS had combinations of abnormalities including hyperglycaemia occurring in significantly older patients with a prevalence increasing progressively in subsequent strata of age [26]. No age dependent prevalence of any combination of abnormalities was instead detected in the adolescents of the present study (results not shown). The mechanisms contributing to the combination of the different abnormalities of the MetS in the juvenile age, as in the adulthood, still appear unclear and require further analysis. Although the MetS-related risks for type-2 diabetes and early cardiovascular disease are now being defined also in the juvenile age and recognised to increase the future risk in adulthood [6], it is not known if, among individuals with the MetS, different combinations of abnormalities are necessarily associated also with different risks. In fact, at present, it is not possible to establish, among the single combinations, which set of abnormalities will evolve in future time toward more serious sequelae of the MetS or, conversely, which one is potentially more reversible on treatment. Both issues may have considerable importance in the strategy of treatment and prevention of the MetS in juvenile obesity. In fact, in spite of the high prevalence of the MetS observed in samples of obese youths from different countries [3,4,14,15,18], favourable changes in single components and reduction of the MetS prevalence have been reported in juvenile age in response to interventions by means of a lifestyle modification [27,28]. The multivariate analysis performed on the data of the present study actually takes into account possible differences in distribution within the multiple strata of age, obesity degree and gender among the subjects from the two nations. In spite
19
of this, some caution should be used in interpreting the differences observed between the two cohorts. Although both samples were from similar clinic-based series of adolescents seeking medical care for obesity, and had comparable degree of obesity, mean age and sex composition, these cohorts belong to populations with different anthropometric characteristics. In fact, according to the respective national reference charts, in the age range of this study, German boys and girls at the 50th percentile appear to be taller and to have higher BMI than their Italian counterpart [10,11,21], and it is difficult to ascertain the impact of these factors on the urge to seek medical help in relation to the actual clinical status. It should be also recalled that no survey performed with comparable methodology on national populations in order to assess obesity and MetS prevalence in youths is presently available in Italy and Germany, and an international comparison should be taken with caution. As a further element prompting to some circumspection in the interpretation of the national differences emerged from the study is the lack of information about the socio-economic composition of the two different participating cohorts. It has been in fact shown in a study on 97,721 adolescents from different European and North American countries that a lower socio-economic status is associated with less healthy behaviours in relation with physical activity and alimentary trends, and a poorer health perception [29]. Nonetheless, some factors may be potentially responsible for a different prevalence of the MetS among the obese adolescents from these two nations. Although genetic and ethnic factors have not been explored in the present study and may have a substantial influence, alimentary habits could play an important role. Different trends in dietary intake have been reported among children and adolescents in Europe [30], especially concerning north–south differences in patterns of fat assumption. Youths in Southern European countries, compared to those in Central and Northern countries, tended to report lower intake of saturated fat and higher consumption of monounsaturated fatty acids, whose major source is olive oil, a trend specifically confirmed in German boys and girls [31]. This pattern of low saturated/unsaturated fat ratio, is indeed one the central features of the Mediterranean-style dietary pattern, for which mounting epidemiological evidence indicates a favourable effect on obesity, diabetes and MetS, possibly due to the increase in antioxidant and anti-inflammatory capacity [32,33]. Thus, although no information was gained about specific dietary trends in the present study, one could speculate that difference in sociocultural factors may have affected also the alimentary habits toward differently healthy patterns among the two samples of adolescents from Italy and Germany. In conclusion, a high prevalence of MetS among extremely obese adolescents from a clinical-based context of two different European countries was detected. The remarkably higher prevalence observed among German boys and girls seems to be due to the increased prevalence of practically all the abnormalities concurring to the definition of the syndrome, but differences in the combinations of abnormalities have also been observed between the two nations, and further experimental research is needed to understand the pathophysiology of the different combinations. Differences in
20
diabetes research and clinical practice 88 (2010) 14–21
alimentary trends between the adolescents from the two countries could be hypothesised, but systematic information concerning the prevalence of obesity and MetS in relation to age and dietary habits in the different nations should be still gained and it is advisable that common strategies by national governments in the European Union should be coded and developed to this aim.
Conflict of interest There are no conflict of interest.
Funding This work was partially supported by Italian government funding for Progetti di Ricerca Corrente, from Ministero del Lavoro, della Salute e delle Politiche Sociali.
Acknowledgements The authors wish to thank Mrs. A. Seddone (head nurse) and the nursing staff at the Division of Auxology, Istituto Auxologico Italiano for having supported the study with their professional skilfulness.
references
[1] S.R. Daniels, M.S. Jacobson, B.W. McCrindle, R.H. Eckel, B.M. Sanner, American Heart Association Childhood Obesity Research Summit: executive summary, Circulation 119 (2009) 2114–2123. [2] R. Weiss, J. Dziura, T.S. Burgert, W.V. Tamborlane, S.E. Taksali, C.W. Yeckel, et al., Obesity and the metabolic syndrome in children and adolescents, N Engl J Med 350 (2004) 2362–2374. [3] V. Calcaterra, C. Klersy, T. Muratori, S. Telli, C. Caramagna, F. Scaglia, et al., Prevalence of metabolic syndrome (MS) in children and adolescents with varying degrees of obesity, Clin Endocrinol (Oxf.) 68 (2008) 868–872. [4] E.S. Ford, W.H. Giles, A.H. Mokdad, Increasing prevalence of the metabolic syndrome among U.S. adults, Diab Care 27 (2004) 2444–2449. [5] E.S. Ford, Risks for all-cause mortality, cardiovascular disease, and diabetes associated with the metabolic syndrome: a summary of the evidence, Diab Care 28 (2005) 1769–1778. [6] B.M. Nathan, A. Moran, Metabolic complications of obesity in childhood and adolescence: more than just diabetes, Curr Opin Endocrinol Diab Obes 15 (2008) 21–29. [7] K.G. Alberti, P. Zimmet, J. Shaw, Metabolic syndrome—a new world-wide definition: a Consensus Statement from the International Diabetes Federation, Diab Med 23 (2006) 469–480. [8] P. Zimmet, K.G. Alberti, F. Kaufman, N. Tajima, M. Silink, S. Arslanian, et al.,IDF Consensus Group, The metabolic syndrome in children and adolescents—an IDF consensus report, Pediatr Diab 8 (2007) 299–306. [9] J.M. Tanner, Growth at Adolescence, 2nd ed., Blackwell Scientific Publications, Oxford, 1962.
[10] E. Cacciari, S. Milani, A. Balsamo, F. Dammacco, F. De Luca, F. Chiarelli, et al., Italian cross-section charts for height, weight and BMI (6–20 y), Eur J Clin Nutr 56 (2002) 171–180. [11] F. Schaefer, M. Georgi, E. Wu¨hl, K. Scha¨rer, Body mass index and percentage fat mass in healthy German schoolchildren and adolescents, Int J Obes Relat Metab Disord 22 (1998) 461–469. [12] World Health Organization, 1987. Measuring obesity— Classification and description of anthropometric data. Report on a WHO consultation on the epidemiology of obesity, EUR/ICP/NUT 125. WHO regional office for Europe, Warsaw. [13] C. Invitti, C. Maffeis, L. Gilardini, B. Pontiggia, G. Mazzilli, A. Girola, et al., Metabolic syndrome in obese Caucasian children: prevalence using WHO-derived criteria and association with nontraditional cardiovascular risk factors, Int J Obes (Lond.) 30 (2006) 627–633. [14] D.A. Caranti, S. Lazzer, A.R. Daˆmaso, F. Agosti, R. Zennaro, M.T. de Mello, et al., Prevalence and risk factors of metabolic syndrome in Brazilian and Italian obese adolescents: a comparison study, Int J Clin Pract 62 (2008) 1526–1532. [15] D. Noto, T. Niglio, A.B. Cefalu`, E. Martino, F. Fayer, M. Mina, et al., Obesity and the metabolic syndrome in a student cohort from Southern Italy, Nutr Metab Cardiovasc Dis 19 (2009) 620–625. [16] E. Goodman, S.R. Daniels, J.B. Meigs, L.M. Dolan, Instability in the diagnosis of metabolic syndrome in adolescents, Circulation 115 (2007) 2316–2322. [17] T. Reinehr, G. de Sousa, A.M. Toschke, W. Andler, Comparison of metabolic syndrome prevalence using eight different definitions: a critical approach, Arch Dis Child 92 (2007) 1067–1072. [18] S. Cook, M. Weitzman, P. Auinger, M. Nguyen, W.H. Dietz, Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 1988–1994, Arch Pediatr Adolesc Med 157 (2003) 821–827. [19] K.G. Alberti, P. Zimmet, J. Shaw, IDF Epidemiology Task Force Consensus Group. The metabolic syndrome—a new worldwide definition, Lancet 366 (2005) 1059–1062. [20] A. Bosy-Westphal, S. Onur, C. Geisler, A. Wolf, O. Korth, M. Pfeuffer, et al., Common familial influences on clustering of metabolic syndrome traits with central obesity and insulin resistance: the Kiel obesity prevention study, Int J Obes (Lond.) 31 (2007) 784–790. [21] S. Furukawa, T. Fujita, M. Shimabukuro, M. Iwaki, Y. Yamada, Y. Nakajima, et al., Increased oxidative stress in obesity and its impact on metabolic syndrome, J Clin Invest 114 (2004) 1752–1761. [22] L. Hutley, J.B. Prins, Fat as an endocrine organ: relationship to the metabolic syndrome, Am J Med Sci 330 (2005) 280– 289. [23] C.L. Lafortuna, F. Adorni, F. Agosti, A. De Col, R. Zennaro, D. Caranti, et al., Factor analysis of metabolic syndrome components in severely obese girls and boys, J Endocrinol Invest 32 (2009) 552–558. [24] C.L. Lafortuna, F. Adorni, F. Agosti, A. Sartorio, Factor analysis of metabolic syndrome components in obese women, Nutr Metab Cardiovasc Dis 18 (2008) 233–241. [25] G.M. Reaven, H. Lithell, L. Landsberg, Hypertension and associated metabolic abnormalities: the role of insulin resistance and the sympathoadrenal system, N Engl J Med 334 (1996) 374–381. [26] A. Sartorio, F. Agosti, F. Adorni, F. Pera, C.L. Lafortuna, Effect of age, degree and distribution of adiposity on the prevalence of the metabolic syndrome in a cohort of obese Italian women, Diab Res Clin Pract 78 (2007) 225–233.
diabetes research and clinical practice 88 (2010) 14–21
[27] T. Reinehr, G. de Sousa, M. Wabitsch, Changes of cardiovascular risk factors in obese children effects of inpatient and outpatient interventions, J Pediatr Gastroenterol Nutr 43 (2006) 506–511. [28] T. Reinehr, M. Kleber, A.M. Toschke, Lifestyle intervention in obese children is associated with a decrease of the metabolic syndrome prevalence, Atherosclerosis 207 (2009) 174–180. [29] M. Richter, M. Erhart, C.A. Vereecken, A. Zambon, W. Boyce, S. Nic Gabhainn, The role of behavioural factors in explaining socio-economic differences in adolescent health: a multilevel study in 33 countries, Soc Sci Med 69 (2009) 396–403.
21
[30] J. Lambert, C. Agostoni, I. Elmadfa, K. Hulshof, E. Krause, B. Livingstone, et al., Dietary intake and nutritional status of children and adolescents in Europe, Br J Nutr 92 (Suppl. 2) (2004) S147–211. [31] M.F. Rolland-Cachera, F. Bellisle, M. Deheeger, Nutritional status and food intake in adolescents living in Western Europe, Eur J Clin Nutr 54 (Suppl. 1) (2000) S41–46. [32] D. Giugliano, K. Esposito, Mediterranean diet and metabolic diseases, Curr Opin Lipidol 19 (2008) 63–68. [33] L. Serra-Majem, B. Roman, R. Estruch, Scientific evidence of interventions using the Mediterranean diet: a systematic review, Nutr Rev 64 (Suppl. 1) (2006) S27–47.