Metabolic syndrome in childhood and increased arterial stiffness in adulthood — The Cardiovascular Risk in Young Finns Study

Abstract

Objective. We conducted the present study to examine the associations of two different paediatric metabolic syndrome (MetS) definitions and recovery from childhood MetS with arterial pulse wave velocity (PWV), an index of arterial stiffness, measured in adulthood.

Methods. A total of 945 subjects participated in the base-line study in 1986 (then aged 9–18 years) and the adult follow-up in 2007 (then aged 30–39 years). Cardiovascular risk factor data were available at both base-line and follow-up. In the follow-up study, arterial PWV was measured using a whole-body impedance cardiography device.

Results. Subjects suffering from MetS in childhood (prevalence 11.1%–14.1%) had higher arterial PWV after 21-year follow-up when compared with those not afflicted by the syndrome in childhood (P < 0.007). An increasing number of the MetS components in childhood were associated with increased PWV in adulthood (P for trend = 0.005). Subjects who recovered from the MetS during the 21-year follow-up period had lower PWV than those with persistent MetS (P < 0.001).

Conclusion. MetS in childhood predicted increased arterial stiffness in adulthood, and recovery from childhood MetS was associated with decreased arterial PWV in adulthood. The current results emphasize the importance of the prevention and controlling of MetS risk factors both in childhood and adulthood.

Key words::

• Arterial stiffness
• epidemiology
• metabolic syndrome

Abbreviations
BMI	=	body mass index
CVD	=	cardiovascular disease
HDL	=	high-density lipoprotein
IMT	=	carotid intima-media thickness
LDL	=	low-density lipoprotein
MetS	=	metabolic syndrome
NCEP	=	National Cholesterol Education Program Adult Treatment Panel III guideline
Ped1MetS	=	the first paediatric metabolic syndrome definition
Ped2MetS	=	the second paediatric metabolic syndrome definition
PWV	=	pulse wave velocity

Key messages

• Childhood metabolic syndrome predicted increased arterial pulse wave velocity, a measure of increased arterial stiffness, in adulthood.

• Recovery from childhood metabolic syndrome was associated with decreased arterial stiffness in adulthood.

Introduction

Metabolic syndrome (MetS) is a constellation of several cardiovascular risk factors, including hypertension, obesity, glucose intolerance, and dyslipidaemia. MetS has been associated with an increased risk for cardiovascular disease (CVD) and type 2 diabetes mellitus (1). Previous studies have also shown that MetS in childhood predicts CVD in adulthood (2) and increases the risk of adult MetS (3).

High pulse wave velocity (PWV), as an index for arterial stiffness, has proven an independent predictor of all-cause and cardiovascular mortality for several patient groups (4–6). High PWV has also been associated with cardiovascular mortality and coronary heart disease among generally healthy older adults (7). In addition, PWV has been found to be increased in young adults with MetS and an increasing number of MetS components (8). Moreover, it has been previously shown that the increase in PWV with age is greater in middle-aged subjects with MetS than in those without MetS (9,10), and that subjects with persistent MetS have higher PWV rates than those without MetS or a regression of MetS (11). Although the relationships between MetS and PWV have been widely studied in adult population samples, limited data are available concerning the associations between childhood MetS, its components, and arterial stiffness (12). Iannuzzi et al. (13) showed in a cross-sectional setting that obese children with MetS have increased arterial stiffness, and Whincup et al. (14) reported a strong inverse association between the number of metabolic syndrome components and arterial distensibility in apparently healthy adolescents. In addition, we have shown with the present adult population that carotid artery elasticity decreases and PWV increases with the increase in the number of cardiovascular risk factors measured in childhood (15,16). However, the relationship between childhood MetS and adulthood PWV is, to the best of our knowledge, unknown.

Therefore, the present study was undertaken to examine the relationships between childhood MetS and adulthood arterial PWV among 945 subjects participating in the Cardiovascular Risk in Young Finns Study. We also determined whether recovery from childhood MetS over a 21-year follow-up has favourable effects on arterial stiffness in adulthood.

Materials and methods

Subjects

The first cross-sectional survey was conducted in 1980 and included 3,596 participants (aged 3–18 years) who were randomly selected from the national (population) register (17). Thereafter, several follow-up studies have been conducted. The study cohort for the present analysis included those subjects who participated in the 1986 survey at the ages of 9, 12, 15, or 18 years as well as the adult follow-up in 2007 (then aged 30–39 years), and for whom complete risk factor data were available in 1986, in addition to risk factor and pulse wave velocity data in 2007 (n = 959). After excluding those female participants who were pregnant at the time of the follow-up (n = 14), a total of 945 subjects were included in the present analysis. Subjects participating in the 1986 survey were selected to comprise the base-line study sample in the present analysis, since fasting glucose was not measured in the 1980 survey. Informed written consent was obtained from all subjects, and the study was approved by local ethics committees.

Biochemical analyses and clinical characteristics

Venous blood samples were collected after a 12-hour fast. Standard methods were used for high-density lipoprotein (HDL) cholesterol, triglycerides, insulin, and plasma glucose concentration measurements. Details of all of the methods have been described previously (15,18,19). Waist circumference (only in 2007), height, and weight were measured, and body mass index (BMI; kg/m²) was calculated. Blood pressure was measured from the brachial artery with standard methods, as described previously (15). The mean of three measurements was used in the analysis.

Definition of metabolic syndrome

According to the updated National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III) definition, adult subjects were categorized as having MetS (in 2007) if they met at least three of the following conditions (20): waist circumference ≥ 102 cm for men and ≥ 88 cm for women; triglycerides ≥ 1.7 mmol/L or relevant drug treatment; HDL cholesterol < 1.03 mmol/L for men and < 1.3 mmol/L for women, or relevant drug treatment; systolic blood pressure ≥ 130 mmHg or diastolic blood pressure ≥ 85 mmHg, or relevant drug treatment; fasting glucose ≥ 5.6 mmol/L or relevant drug treatment.

Because there is no universally accepted definition for paediatric MetS (12), we created two paediatric MetS definitions similar to the previous study by Lambert et al. (21). Values of age- and sex-specific cut-off points used to define risk factors were estimated from the study population. Overweight was defined as BMI ≥ 85th percentile (22). High triglycerides, high systolic blood pressure, hyperinsulinaemia, and low HDL cholesterol were defined as values in the respective extreme quartiles (triglycerides ≥ 75th percentile, systolic blood pressure ≥ 75th percentile, fasting insulin ≥ 75th percentile, HDL cholesterol ≤ 25th percentile). These cut-off points are also similar to those used in the Bogalusa Heart Study (23). Hyperglycaemia was defined as fasting blood glucose ≥ 6.1 mmol/L (24). The first paediatric MetS (Ped1MetS) definition required the presence of any three of these six risk factors. The second paediatric MetS (Ped2MetS) definition required the presence of hyperinsulinaemia and any two of the other five risk factors.

The study population was classified further into four different groups according to MetS status in 1986 and 2007: control group (no MetS in 1986 or 2007), recovery group (MetS in 1986 but not in 2007), incident group (no MetS in 1986 but MetS in 2007), and the persistent group (MetS in both 1986 and 2007). These groups were comprised separately for the two paediatric MetS definitions (Ped1MetS and Ped2MetS).

Arterial pulse wave velocity measurement

PWV was determined by using a commercially available whole-body impedance cardiography monitor, the CircMon B202 (CircMon™; JR Medical Ltd, Tallinn, Estonia). A pair of electrically connected current electrodes (Blue Sensor type R-00-S; Medicotest A/S, Ölstykke, Denmark) were placed on the distal parts of the extremities just proximal to the wrists and the ankles. Voltage electrodes were placed proximal to the current electrodes, with a distance of 5 cm between the centres of the electrodes. The distal impedance plethysmogram was recorded from a popliteal artery at knee joint level. The active electrode was placed on the lateral side of the knee joint and the reference electrode on the calf, the distance between the electrodes being about 20 cm. Alternating electrical current was applied to current electrodes and change in whole-body impedance was measured from voltage electrodes. The whole-body impedance decreases when the pulse pressure enters the aortic arch and changes the diameter of the aorta. The CircMon software measures the time difference between the onset of the decrease in impedance in the whole-body impedance signal caused by pulse wave in the aortic arch and, subsequently, in the popliteal artery signal. The PWV can be determined from the distance and the time difference between the two recording sites. A more detailed description of the method (16,25) and the validation study (25) has been published previously.

Statistical analyses

All statistical analyses were performed with SPSS for Windows (version 16.0; SPSS Inc., Chicago, IL, USA). Comparison of base-line and follow-up characteristics between subjects with and without MetS was performed using the t test for continuous variables and the chi-square for sex as a categorical variable. The skewed distributions of triglycerides and insulin were corrected logarithmically before statistical analyses. The univariate relationships between PWV and MetS components were studied by means of regression analysis. Multivariable regression models, including age and sex, were constructed to examine the independent effects of the MetS components on PWV. In regression analysis we used heart rate-specific z scores for PWV because heart rate may be a confounding factor (26). Variation in risk variables during the 21-year follow-up was studied by subtracting the base-line value from the follow-up value. Age- and sex-adjusted differences in risk variable changes, as well as heart rate-, sex-, and age-adjusted mean PWV values, were analysed using general linear models. There were no interactions between sex, MetS, and PWV, and analyses were therefore performed with the sexes combined. Statistical significance was determined as two-tailed P < 0.05.

Results

Base-line and follow-up characteristics of the study subjects are shown in Table I. The prevalences of MetS in the base-line paediatric population (ages 9–18 years in 1986) were 14.1% for the Ped1MetS definition and 11.1% for the Ped2MetS definition. Furthermore, the prevalence of MetS in the follow-up adult population (ages 30–39 years in 2007) was 18.1%. Subjects with MetS were older (in the adult population) and had a higher body mass index, systolic and diastolic blood pressure, triglycerides, fasting glucose, fasting insulin, and waist circumference (only in the adult population), as well as lower HDL cholesterol, when compared with subjects without MetS (P < 0.05 for all).

Table I. Base-line and follow-up characteristics of study subjects.

	1986			2007
Variable	No MetS	Ped1MetS	Ped2MetS	No MetS	Adult MetS
Number of subjects	812	133	105	774	171
Age (years)	13.6 ± 3.3	13.5 ± 3.4	13.5 ± 3.3	34.4 ± 3.3	35.3 ± 3.4^b
Sex (%, females)	55.5	48.9	51.4	56.8	44.4^a
Body mass index (kg/m^2)	18.9 ± 2.7	22.2 ± 3.6^b	22.4 ± 3.8^b	24.5 ± 3.7	31.2 ± 4.3^b
Systolic blood pressure (mmHg)	110.0 ± 11.4	119.7 ± 12.4^b	118.9 ± 12.8^b	117.6 ± 12.7	127.5 ± 15.0^b
Diastolic blood pressure (mmHg)	62.5 ± 9.2	65.3 ± 10.3^a	64.9 ± 10.1^a	72.8 ± 10.4	81.5 ± 11.6^b
HDL cholesterol (mmol/L)	1.6 ± 0.3	1.3 ± 0.2^b	1.3 ± 0.2^b	1.4 ± 0.3	1.1 ± 0.3^b
Triglycerides (mmol/L)	0.8 (0.6–1.0)	1.3 (1.0–1.5)^b	1.2 (1.0–1.5)^b	1.0 (0.8–1.4)	2.0 (1.7–2.7)^b
Insulin (mmol/L)	8.4 (6.5–12.0)	14.3 (11.0–18.0)^b	16.5 (13.0–19.3)^b	5.7 (3.9–8.8)	13.2 (9.4–18.1)^b
Glucose (mmol/L)	4.7 ± 0.5	4.9 ± 0.5^b	4.9 ± 0.5^b	5.2 ± 0.4	5.7 ± 0.9^b
Waist circumference (cm)	–	–	–	84.3 ± 10.8	103.0 ± 11.3^b

MetS = metabolic syndrome; Ped1MetS = the first paediatric MetS definition; Ped2MetS = the second paediatric MetS definition; 1986 No MetS = no MetS according to Ped1MetS or Ped2MetS.

Values are presented as unadjusted mean ± standard deviation or geometric mean (25th–75th percentile) or percentages of subjects. Waist circumference was not measured in 1986.

^aP < 0.05; ^bP < 0.001 in pairwise comparison No MetS versus MetS.

In univariate regression analysis, hyperinsulinaemia (P = 0.005) and hypertension (P < 0.024) as individual components of paediatric MetS were directly associated with PWV measured in adulthood (Table II). In multivariable regression analysis, age (P < 0.001), sex (P < 0.001), and hyperinsulinaemia (P = 0.021) were directly and independently associated with PWV in adulthood (Table II). In univariate regression analysis for the adult population, hypertension (β ± SE 0.631 ± 0.068), obesity (0.361 ± 0.071), high triglycerides (0.502 ± 0.077), and hyperglycaemia (0.407 ± 0.078) as individual components of adult MetS were directly associated with PWV (P < 0.001 for all). In multivariable regression analysis for the adult population, age (0.050 ± 0.009; P < 0.001), sex (0.515 ± 0.061; P < 0.001), hypertension (0.406 ± 0.067; P < 0.001), obesity (0.277 ± 0.070; P < 0.001), and high triglycerides (0.169 ± 0.078; P = 0.031) as individual components of adult MetS were independently associated with PWV. When using continuous paediatric risk variables in multivariable regression analysis, sex, age, systolic blood pressure, and triglycerides were directly and independently associated with PWV in adulthood (P < 0.02 for all) (data not shown).

Table II. Univariate and multivariable relationships between components of paediatric MetS (ages 9–18 years in 1986) and adult PWV (in 2007) (n = 945).

	PWV
	β ± SE	P
Univariate relations
Hyperinsulinaemia	0.203 ± 0.071	0.005
Hypertension	0.165 ± 0.073	0.024
Overweight	0.149 ± 0.090	0.099
Low HDL cholesterol	0.022 ± 0.074	0.764
High triglycerides	0.117 ± 0.073	0.110
Hyperglycaemia	0.002 ± 0.564	0.998
Multivariable relations
Age	0.059 ± 0.009	< 0.001
Sex	0.608 ± 0.060	< 0.001
Hyperinsulinaemia	0.164 ± 0.071	0.021
Hypertension	0.126 ± 0.069	0.068
Overweight	0.039 ± 0.089	0.661
Low HDL cholesterol	−0.036 ± 0.072	0.613
High triglycerides	0.082 ± 0.073	0.260
Hyperglycaemia	−0.200 ± 0.526	0.703

PWV = pulse wave velocity; MetS = metabolic syndrome; β = regression coefficient; SE = standard error.

Heart rate-specific z scores were used for PWV.

Adulthood PWV adjusted for age, sex, and heart rate was higher in subjects with paediatric MetS (P < 0.007 for both paediatric MetS definitions) and adult MetS (P < 0.001) than in those not suffering from MetS in childhood or adulthood (Figure 1A). There was also an increasing trend for adult PWV with the increase in the number of paediatric MetS components (P for trend = 0.005) (Figure 1B). Subjects who had persistent MetS over the 21-year follow-up had higher PWV than those without MetS at base-line and follow-up (P < 0.001) (Figure 1C). In addition, subjects who recovered from MetS during the 21-year follow-up had lower PWV than those with persistent MetS (P < 0.001) (Figure 1C). Furthermore, we observed increasing trends (age- and sex-adjusted) for BMI (P < 0.001), systolic blood pressure (P < 0.002), diastolic blood pressure (P < 0.05), triglycerides (P < 0.001), fasting insulin (P < 0.001), and fasting glucose (P < 0.001), as well as a decreasing trend for HDL cholesterol (P < 0.003), over the 21-year follow-up in the MetS persistent group when compared to the MetS recovery group, with both paediatric MetS definitions.

Figure 1. A: Comparison of adult PWV between subjects with and without MetS according to the paediatric MetS definitions (Ped1MetS and Ped2MetS) (ages 9–18 years in 1986) and the adult MetS definition (ages 30–39 years in 2007). B: Adult PWV by the number of paediatric MetS components (ages 9–18 years in 1986). C: Comparison of adult PWV between the 21-year changes in MetS status. PWV values are age-, sex-, and heart rate-adjusted means and standard errors. Values under columns indicate the number of subjects in each group. *For both paediatric MetS definitions (Ped1MetS and Ped2MetS). (PWV = pulse wave velocity; MetS = metabolic syndrome; Ped1MetS = the first paediatric MetS definition requiring any three of six risk factors (hyperinsulinaemia, high triglycerides, high systolic blood pressure, overweight, hyperglycaemia, low HDL cholesterol); Ped2MetS = the second paediatric MetS definition requiring the presence of hyperinsulinaemia and any two of the other five risk factors; Control = no MetS at 1986 according to Ped1MetS or Ped2MetS and no MetS at 2007 according to the adult MetS definition; Recovery = MetS at 1986 according to Ped1MetS or Ped2MetS but not at 2007 according to the adult MetS definition; Incident = no MetS at 1986 according to Ped1MetS or Ped2MetS but MetS at 2007 according to the adult MetS definition; Persistent = MetS both at 1986 according to Ped1MetS or Ped2MetS and 2007 according the adult MetS definition.)

All analyses were repeated after excluding subjects on antihypertensive (n = 46), lipid-lowering (n = 11), or antidiabetic (n = 6) medications, with essentially similar results. In addition, all analyses were repeated using a recently published harmonized MetS definition in adulthood (27), with essentially similar results. Moreover, all analyses were repeated without heart rate correction for PWV, with essentially similar findings (data not shown).

Discussion

The present study shows that MetS and the accumulation of the number of MetS components in childhood have an adverse effect on arterial stiffness in adulthood. Children and adolescents with MetS at base-line had higher PWV after a 21-year follow-up in adulthood, when compared to their peers who did not suffer from MetS at base-line. In addition, persistent MetS over the 21-year follow-up associated with increased arterial stiffness. Subjects with persistent MetS had higher PWV than those who recovered or were free from of MetS during the 21-year follow-up.

To the best of our knowledge, this is the first study to report an association between increased adult PWV in subjects with MetS and increasing number of MetS components in childhood. Previously, elevated cardiovascular risk levels in childhood have been shown to predict increased carotid intima-media thickness (IMT), a marker of early atherosclerosis, in adulthood (28). In addition, decreased carotid artery elasticity and increased PWV have been found in adults with increasing number of cardiovascular risk factors measured in childhood (15,16). However, childhood MetS and arterial stiffness have received little academic interest to date. Iannuzzi et al. (13) reported in a cross-sectional setting that arterial stiffness was increased in obese children with MetS, and Whincup et al. (14) found the arterial distensibility in adolescents to decrease with the increase in the number of MetS components.

In our cohort, hyperinsulinaemia and hypertension as individual components of paediatric MetS were associated with adult PWV in the univariate regression analysis, and childhood hyperinsulinaemia was an independent predictor of increased PWV in adulthood in the multivariable analysis. Furthermore, we found a border-line significant association (P = 0.068) between childhood hypertension and adulthood PWV in the multivariable model. These findings are consistent with the notion that childhood systolic blood pressure is directly associated with PWV in young adulthood (16,29). In addition, blood pressure measured in childhood and adolescence has been found to predict decreased carotid artery elasticity in adulthood (15). Furthermore, insulin levels in childhood have been shown to correlate inversely with carotid artery compliance and directly with Young's elastic modulus in adulthood (15). Taken together, the current findings provide further insight into the associations of childhood MetS, the components of childhood MetS, and adulthood arterial stiffness.

Our findings supported the view that PWV is increased in young adults with MetS when compared to subjects not afflicted by the syndrome (8). Previously, hypertension and obesity as individual components of MetS have been found to be independent predictors of PWV, and triglyceride concentration has shown a direct association with PWV (30). In line with this, we found the independent adult determinants of increased PWV to be hypertension, obesity, and high triglycerides as individual components of adult MetS. Therefore, our observations regarding adult MetS and PWV fall in line with the previous studies.

Tomiyama et al. (11) have previously shown with a middle-aged Japanese male population that the annual rate of increase in PWV is higher in subjects with persistent MetS than in those with a regression of or no MetS. In addition, Nakanishi et al. (9) and Safar et al. (10) have reported with middle-aged populations that the increase in PWV escalates with age in subjects with MetS, as compared with those without the condition. In line with these studies, we found that subjects who had persistent MetS over the 21-year follow-up had a 1.2–1.3 m/s (depending on the paediatric MetS definition used) higher PWV than subjects who recovered from MetS, and a 1.2 m/s higher PWV compared to subjects without MetS over the 21-year follow-up. This difference in PWV is noteworthy, since Vlachopoulos et al. reported a 14%–15% increase in cardiovascular events, cardiovascular mortality, and all-cause mortality for each 1 m/s increase in aortic PWV (31).

One plausible explanation for the increased PWV in the MetS persistent group as compared to the MetS recovery group could be provided by the increasing trends for BMI, systolic and diastolic blood pressure, triglycerides, fasting insulin, and fasting glucose, as well as the decreasing trend for HDL cholesterol, in subjects with persistent MetS when compared to those who recovered from MetS over the 21-year follow-up. This is in accordance with previous studies that have demonstrated a decrease in carotid-femoral PWV after weight reduction in middle-aged men (32), a reduction in carotid-femoral PWV after long-term trandolapril treatment in older adults (33), and a decrease in femoral-ankle PWV after atorvastatin treatment in patients with type 2 diabetes mellitus (34). We have also recently shown that recovery from MetS in young adults during a 6-year follow-up period was associated with the reversibility of IMT progression and with a decreased rate of carotid artery distensibility regression, suggesting that arterial structure and function may be restored in young adults with MetS by improving the metabolic risk factor profile and by weight reduction (35). However, as reviewed by Zieman et al. (36) and Stehouwer et al. (37), the mechanism by which arterial stiffness increases in subjects with MetS is a complex and partly unknown process. Nevertheless, it has been suggested that chronic hyperglycaemia and hyperinsulinaemia increase the local activity of the renin-angiotensin-aldosterone system and the expression of angiotensin type 1 receptor in vascular tissue, promoting the development of wall hypertrophy and fibrosis. Furthermore, impaired glucose tolerance enhances the glycation of proteins with the cross-linking of collagen, which may lead to the loss of collagen elasticity. In addition, endothelial dysfunction caused by high low-density lipoprotein (LDL) cholesterol, free fatty acids, endothelin-1, inadequate vasodilatory effects of insulin, or decreased adiponectin may explain, at least in part, the increased arterial stiffness (36,37). Further studies are clearly needed on the pathophysiological mechanisms behind the adverse effects of MetS on arterial stiffness and those affecting the arteries when recovering from MetS.

A potential limitation of the present study is the whole-body impedance cardiography method, which is not yet widely used in epidemiology settings to measure PWV, apparently limiting comparability of the present findings with the observations from other cohorts. However, PWV values measured between the aortic arch and popliteal artery using the CircMon are highly comparable to those measured by Doppler ultrasound method (25). In addition, reference values (38), as well as good repeatability and reproducibility indexes (99% and 87%, respectively) for PWV measured by CircMon have been published previously (39). Moreover, the whole-body impedance cardiography is a handy and reliable tool in epidemiological studies, because the method is operator-independent and inexpensive, and the observed variability of PWV is mainly physiological (25,39).

Our study has some other limitations. Firstly, cut-off point-based definitions of MetS in children and adolescents have been shown to have marked short-term instability (40,41). Secondly, MetS was defined at base-line and at follow-up, but PWV was measured only at follow-up. Therefore, the current study does not allow the evaluation of longitudinal changes in PWV in relation to MetS status. Thirdly, overweight at base-line was determined by using age- and sex-specific cut-off points for BMI, which might be a less sensitive method to estimate adiposity. However, in determining MetS, BMI is considered an alternative if waist circumference data are not available (42). Finally, lack of a standard paediatric MetS definition and the ethnically homogeneous study sample may limit the generalizability of the present results. A potential strength of the current study is the use of age- and sex-specific cut-off points for risk variables at base-line, which may help to avoid misclassification due to age and gender differences in risk factor levels in childhood and adolescence.

In conclusion, findings in the present study suggest that MetS in childhood is associated with subclinical vascular damage in adulthood, and that recovery from childhood MetS during a 21-year follow-up period might have positive effects on arterial stiffness. Since MetS and increased PWV have been shown to be strong predictors of CVD, the current results emphasize the importance of the prevention and controlling of MetS risk factors both in childhood and adulthood.

Declaration of interest: This study was financially supported by the Academy of Finland (grants no. 77841, 117832, 201888, 121584, and 126925); the Social Insurance Institution of Finland; the Turku University Foundation; the Kuopio, Tampere and Turku University Hospital Medical Funds; the Emil Aaltonen Foundation (TL); the Juha Vainio Foundation; the Finnish Foundation of Cardiovascular Research; and the Finnish Cultural Foundation. The authors declare no conflict of interest.