Prevalence of the metabolic syndrome and its relationship with diabetes mellitus by aging

Abstract

Introduction. It is important to make a prompt diagnosis of metabolic syndrome (MetS) in order to prevent the development of cardio-/cerebro- vascular diseases and diabetes mellitus (DM). The authors estimated the risk of development of DM by the presence/absence of MetS and age groups.

Methods. A cross-sectional study of subjects undergoing intensive health examination was conducted (3149 men aged 30–69 years). Diagnosis of MetS was based on the criteria of the National Cholesterol Education Program Expert Panel (NCEP).

Results. The prevalence of DM occurring in association with MetS increased with age; it was 11.9% in subjects with MetS in their 30s, it was 19.8% in subjects with MetS in their 60s. The prevalence of DM among subjects who had one or two components of MetS also increased with age. There was a significant progressive increase of the odds ratio in subjects in their 30s, 40s, 50s and 60s who were judged as having MetS; significant increase of the odds ratio was seen in subjects in their 60s, even in those who were not judged as having MetS.

Conclusions. Subjects with MetS show a high prevalence of DM, and the prevalence increased with age in the subjects.

Keywords:

• Diabetes mellitus
• metabolic syndrome
• aging

Introduction

Metabolic syndrome (MetS) is associated with an increase in the risk of development of cardiovascular diseases (CVD) and diabetes mellitus (DM) [1–3]. The risk of CVD associated with MetS is independent of other related factors, such as smoking, psychological stress and alcohol consumption [1]. The risk of CVD associated with MetS is manifested across a wide spectrum of glyemia [4–6]. Increase in the overall and CVD mortality has been reported in subjects with MetS, with the risk of death increasing with increasing number of components of MetS [7].

The prevalence of MetS has been reported to increase with age [8–10]. The estimated prevalences in Japan are 8.5%, 16.7%, 25% and 36.4% in male subjects in their 30s, 40s, 50s and 60s in Japan [11]. However, the prevalence of MetS in the above survey was calculated using different criteria because of the limitation of the blood sampling time: as fasting venous blood samples could not be collected, the serum triglyceride and plasma glucose levels could not be determined; instead of fasting plasma glucose, hemoglobin A1c was used to determine the glycemic status, and dyslipidemia as a component of MetS was judged based solely on the serum level of high-density lipoprotein cholesterol.

Miura et al. reported that the prevalence of MetS in chronic heart failure patients is more than double compared with the general population in Japan, suggesting that the metabolic components may have a substantial effect on the development of both ischemic and non-ischemic chronic heart failure [12]. In their report, there are also some descriptions on the prevalence of MetS in Japanese general population. But the prevalence of MetS changes by using different definition of MetS. Although the MetS has a substantial effect on the development of type 2 diabetes [13], there is a significant relationship between lifestyle and the number of MetS components in a general Japanese population [14]. To clarify the relationship among MetS, DM and aging, we investigated the prevalence of the MetS according to the international criteria in combination with the prevalence of DM. In other words, our scientific contribution is the fine analysis on the associations among MetS, DM prevalence and aging, controlling for the lifestyle confounders such as alcohol intake, smoking habits and regular exercise.

Subjects and methods

A total of 3149 male subjects aged 30–69 years who attended an intensive health checkup programme were enrolled in this study. Subjects who were receiving treatment for coronary heart disease, cerebrovascular disease, hypertension, DM and/or dyslipidemia were included. At the baseline, all participants were subjected to a screening protocol, including determination of the anthropometric and biochemical parameters. Informed consent was obtained from each of the participants. This study was conducted with the approval of the Institutional Review Board of Ota General Hospital, Gunma Prefecture, Japan (17 July 2010).

Health examination data

Venous blood samples were collected after the patients had fasted overnight. Serum levels of triglycerides, high-density lipoprotein cholesterol (HDL) and fasting blood glucose were determined (AU2700, Olympus Co. Ltd., Japan). The blood pressure was measured twice in all the participants (Nippon COLIN BP-103iII, Japan) and the values from the second measurement were adopted for the analysis. Waist circumference was measured at the level midway between the iliac crest and the 12th rib. A 75-g oral glucose tolerance test was performed, and the diagnosis of DM was made based on the following criteria [15].

• 1. HbA1c ≥6.5%, determined according to the National Glycohemoglobin Standardization Program (NGSP), was an essential criterion.

• 2. One of the following three criteria for plasma glucose; (1) fasting plasma glucose ≥126 mg/dl (7 mmol/l), (2) 2 h plasma glucose ≥200 mg/dl (11.1 mmol/l), (3) casual plasma glucose ≥200 mg/dl (11.1 mmol/l).

If the subject was under medication for DM, he was counted as having DM.

Definitions of lifestyle-related variables and MetS

Self-reported information was collected on lifestyle-related variables. The smoking habit was categorised as ‘current smoking’ or ‘no smoking’. The drinking habit was categorised as ‘daily drinking’ or ‘no daily drinking’ (including ‘no drinking’). Physical activity was categorised as presence of the exercise habit (exercise ≥2 times per week for more than one year) or absence of the exercise habit.

In accordance with the criteria of the US National Cholesterol Education Program Adult Treatment Panel III (NCEP-ATPIII) [16], MetS was diagnosed when three or more of the following criteria were fulfilled: fasting blood glucose ≥100 mg/dl (5.6 mmol/l) or receiving treatment for DM, blood pressure ≥130/85 mm Hg (either value) or receiving antihypertensive drug treatment, plasma triglycerides ≥150 mg/dl (1.7 mmol/l), plasma HDL <40 mg/dl (1.0mmol/l) in men, or receiving treatment for dyslipidemia and waist circumference ≥85 cm in men [17]. Each of the five components of MetS was assigned a value of 1 when it was judged as being present, and 0 when it was judged as being absent. Then, the number of components of MetS was also calculated.

Statistical analysis

For the analysis of the data, the study population was divided into four age groups (30–39 years, 40–49 years, 50–59 years and 60–69 years). The subjects were also divided into three groups based on the number of components of MetS; 0, 1–2 and 3–5 (MetS). The analysis was carried out with the SPSS 16.0 software package (SPSS Japan, Tokyo). Analysis of variance, Cochran–Armitage test and logistic regression analysis were used as the statistical tests. A two-tailed p-value <0.05 was considered as denoting significance.

Results

The mean age (±standard deviation) of the subjects was 51.3 (±9.0) years. The prevalence of MetS in the subjects stratified by the age group is shown in Table I. The overall prevalence of MetS in the male subjects was 33.2%. The prevalence increased with age; it was estimated to be 19.9%, 27.8%, 37.5% and 39.8% in the subjects in their 30s, 40s, 50s and 60s, respectively (p < 0.001). About half (55%) of the subjects with MetS had three components of the MetS, 31.3% had four and 13.7% had all the five components.

Table I. The prevalences of medication for DM, hypertension and dyslipidemia, and those of metabolic syndrome and its components stratified by the age group (n=3149).

	30–39 years, n=337	40–49 years, n=951	50–59 years, n=1216	60–69 years, n=645	p for trend
DM treatment	1.8%	2.3%	5.1%*	5.4%*	<0.001
HT treatment	2.7%	8.3%*	25.7%*	34.6%*	<0.001
DL treatment	2.1%	4.7%	11.6%*	14.3%*	<0.001
MetS	19.9%	27.8%*	37.5%*	39.8%*	<0.001
WC	43.9%	48.5%	51.1%	49.5%	ns
Glucose	32.9%	36.7%	48.4%*	58.3%*	<0.001
Blood pressure	24.9%	34.6%*	46.7%*	53.8%*	<0.001
TG	29.1%	35.3%	38.9%*	35.3%	<0.05
HDLC	13.4%	14.0%	20.0%*	23.1%*	<0.001
DM diagnosis	11.9%	14.0%	19.7%*	19.8%*	<0.001

DM, diabetes mellitus; HT, hypertension; DL, dyslipidemia; MetS, metabolic syndrome; WC, waist circumference; TG, triglyceride; HDLC, high-density lipoprotein cholesterol; p for trend was checked by Cochran–Armitage test.

*p<0.05 against prevalence in 30s by Dunnett's multiple comparisons.

The prevalence of DM also increased with age; it was estimated to be 11.9%, 14%, 19.7% and 19.8% in the subjects with MetS in their 30s, 40s, 50s and 60s, respectively (p < 0.001). The prevalence of DM among the subjects who had one or two components of MetS also increased with age, being estimated to be 3.6%, 2.5%, 5.6% and 10.7% in such subjects in their 30s, 40s, 50s and 60s, respectively (p < 0.05).

The odds ratios for DM in subjects in their 30s with no MetS as a control group are shown in Table II. There were significant increase of the odds ratios in subjects in their 30s, 40s, 50s and 60s with MetS after adjustment for the exercise habit, drinking habit and smoking habit, being 5.83 (95% confidence interval (CI): 1.94–17.48), 7.62 (95% CI: 3.15–18.42), 12.14 (95% CI: 5.21–28.29) and 12.24 (95% CI: 5.10–29.39), respectively, in these age groups. In addition, there was a significant increase of the odds ratio in subjects in their 60s who did not have MetS, being 4.96 (95% CI: 2.03–12.12) in this group. Among the lifestyle factors, the odds ratios (95% CI) for DM in subjects who did not smoke, drink daily and exercised regularly were 0.68 (0.52–0.90), 1.60 (1.22–2.09) and 1.17 (0.88–1.56), respectively.

Table II. Multivariate odds ratios for diabetes mellitus determined by logistic regression analysis, according to the age group and presence/absence of metabolic syndrome, adjusted for the smoking habit, drinking habit and exercise habit.

Age	MetS	Odds ratio	95% CI	Significance
30–39	No	1
	Yes	5.83	1.94, 17.48	p <0.01
40–49	No	0.80	0.30, 2.17	ns
	Yes	7.62	3.15, 18.42	p <0.001
50–59	No	2.12	0.88, 5.15	ns
	Yes	12.14	5.21, 28.29	p <0.001
60–69	No	4.96	2.03, 12.12	p <0.001
	Yes	12.24	5.10, 29.39	p <0.001

ns, not significant; CI, confidence interval.

Significant differences against 30s with no MetS were checked.

Discussion

The prevalence of the MetS increased with age in this study, consistent with previous reports [8–10 ,18]. This trend was also observed for each of the components of MetS, except the waist circumference.

Use of different criteria for the diagnoses of MetS results in different estimates of the prevalence of MetS and its components [12]. In addition, the cutoff value of waist circumference as a component of MetS differs among in different races and different ethnicities [16,17], which would influence the trends of the prevalence of MetS. Furthermore, the absolute risks (incidence and mortality) of cardio/cerebrovascular diseases are also related to factors other than components of the MetS [2,3,18].

The prevalence of MetS in Japan has been reported to plateau in men in the 60s [11], which is in agreement with our data (not presented) and also other previous reports [8,18]. The authors speculate that the decline in the prevalence of MetS in subjects in their 70s could be explained in part by the healthy-inhabitant effect of the target population. But there are some reports that testosterone secretion in older men are also linked to modifiable risk factors like obesity, smoking and lifestyles [19,20]. Obese men had a greater decline of free and total testosterone and sex hormone binding globulin (SHBG) levels, compared to men who were never classified as obese, and men with low total testosterone and low SHBG have an increased risk of diabetes [21]. There is a space to conduct the advanced study on the association between testosterone level and MetS prevalence.

In this study, the prevalence of DM increased with age in subjects with MetS. The percentage of DM subjects on medication was 46.2% in all, which did not increase with age. As progression of vascular damage is accelerated by DM [22], immediate commencement of medication is required in patients with MetS and DM in order to reduce the risk of CVD [23].

Lifestyle modification should also be recommended in subjects diagnosed as having MetS [24]. In this study, ‘no smoking’ was significantly associated with a lower odds ratio for DM, while habitual exercise did not significantly influence the odds ratio for DM. Furthermore, ‘no daily drinking’ was significantly associated with a higher odds ratio for DM. We cannot conclude a cause–effect relationship between DM and lifestyle factors or MetS from the study results, because of the cross-sectional study design. Further study is needed, because there is a possibility that the risk of DM is reduced with quitting the habit of daily drinking. Precise monitoring of the exercise and dietary habits is also required.

Acknowledgements

The authors wish to express their appreciation to the study participants. This work was supported partly by Grant-in-Aid for Scientific Research (C) (20590616).

Declaration of interest:

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.