Disclosure: The authors declared no conflict of interest.
Effect of longitudinal changes in visceral fat area on incidence of metabolic risk factors: The hitachi health study
Article first published online: 29 MAY 2013
Copyright © 2013 The Obesity Society
Volume 21, Issue 10, pages 2126–2129, October 2013
How to Cite
Matsushita, Y., Nakagawa, T., Yamamoto, S., Takahashi, Y., Yokoyama, T., Mizoue, T. and Noda, M. (2013), Effect of longitudinal changes in visceral fat area on incidence of metabolic risk factors: The hitachi health study. Obesity, 21: 2126–2129. doi: 10.1002/oby.20347
Relevant conflicts of interest/financial disclosures: Nothing to report. Full financial disclosures and author notes may be found in the online version of this article.
Funding agencies: This research was supported by a grant from the Ministry of Health, Labour and Welfare of Japan.
- Issue published online: 5 OCT 2013
- Article first published online: 29 MAY 2013
- Accepted manuscript online: 14 FEB 2013 07:18AM EST
- Manuscript Accepted: 6 DEC 2012
- Manuscript Received: 29 MAY 2012
To examine the incidences of metabolic risk factors according to changes in visceral fat area (VFA) in a large Japanese population.
Design and Methods
The subjects were 973 men who received a computed tomography (CT) examination in health checkups twice (2004-2005 and 2007-2008), and not having two or more of metabolic risk factors (except for the waist circumference) in 2004-2005. VFA was measured using CT. To assess the potential influence of changes in VFA for the 3-year incidences of each metabolic risk factor and clustering metabolic risk factors, logistic regression analyses were used.
A significant association was observed between the change in VFA and the components of the metabolic risk factors. Incidences of the components of the metabolic risk factors were significantly higher among subjects with a larger increase in VFA and were significantly lower among subjects with a larger decrease in VFA (trend P < 0.001). Significant increases in the odds ratios for the incidences of high triglycerides and low high-density lipoprotein cholesterol level were observed among subjects with ≥50 cm2 VFA increase.
The adoption of a lifestyle that does not increase the VFA is important for preventing metabolic syndrome.
Cardiovascular disease (CVD) is one of the leading causes of morbidity and mortality in the world . Previous reports have shown that obesity plays a significant role in increasing cardiovascular risk . Some indicators of obesity, such as the visceral fat area (VFA), or visceral adipose tissue, are more strongly associated with the risk of CVD than other indicators of obesity, such as waist circumference, body mass index (BMI) [3, 4], or subcutaneous fat area (SFA) [2, 3]. A larger VFA is strongly related to a higher prevalence of impaired fasting glucose levels , diabetes , insulin resistance , hypertension , abnormality of lipid metabolism , and metabolic risk factors [7, 8]. Several cross sectional and other types of studies have examined the association between VFA and metabolic risk factors. For instance, in some prospective studies, VFA was measured at baseline and its correlation between risk factors for diabetes (plasma insulin level, homeostasis model assessment for insulin resistance, etc.) was evaluated [9-11]. However, no study has measured the longitudinal change in VFA among the same subjects twice, at baseline and at follow-up, to examine its relation to the incidences of metabolic risk factors in a large cohort study. Thus, whether an increase in VFA leads to an increase in the risk of CVD or whether a decrease in VFA leads to a decrease in the risk of CVD remain uncertain. Clarifying this point could be useful for preventing CVD in clinical settings.
Therefore, we examined the incidences of the clustering of metabolic risk factors and its components among subjects who did not have each of the risk factors or the clustering of metabolic risk factors at baseline according to the aforementioned changes in VFA.
Methods and Procedures
Overall, a total of 13,965 male employees and their spouses who, after more than 12 h of fasting, underwent health checkups during the baseline survey performed in 2004 and 2005 at the Hitachi Health Care Center in Ibaraki Prefecture. Of them, 3,119 men received a computed tomography (CT) scan. We studied 1,106 men who also participated in a 3-year follow-up survey performed in 2007 and 2008 (aged 30-72 years in 2004 and 2005) in the final analysis. Subjects without each metabolic risk factor (except for the waist circumference) or clustering of risk factors were included into the analysis of the incidence. Numbers of subjects in each group are shown in Table 1. Informed consent was obtained from each examinee regarding the use of his or her data for research purposes. The present study was approved by the ethics review committee of the National Center for Global Health and Medicine.
|Subjects with 0–1 metabolic risk factor||Subjects with ≥2 metabolic risk factors|
|Age (years)||52.7a (8.4)||55.1a (7.8)|
|Height (cm)||168.9a (6.2)||168.2a (5.9)|
|Weight (kg)||66.7a (9.1)||70.4a (9.2)|
|BMI (kg/m2)||23.3a (2.6)||24.8a (2.7)|
|VFA (cm2)||113.2a (52.1)||149.1a (50.5)|
|SFA (cm2)||118.2a (51.9)||136.8a (50.4)|
|VFA/SFA||1.0a (0.4)||1.2a (0.6)|
|Waist circumference (cm)||84.6a (7.8)||89.2a (8.3)|
|SBP (mmHg)||120.1a (11.7)||130.6a (10.0)|
|DBP (mmHg)||76.1a (8.3)||82.7a (7.2)|
|Triglycerides (mg/dL)||113.4a (62.3)||193.6a (133.5)|
|HDL-cholesterol (mg/dL)||59.1a (13.6)||51.8a (14.7)|
|Fasting glucose (mg/dL)||103.2a (16.1)||121.2a (27.9)|
|High blood pressure||261b (26.8)||520b (78.9)|
|Hyperglycemia||132b (13.6)||458b (69.5)|
|High triglycerides||154b (15.8)||451b (68.4)|
|Low HDL cholesterol||17b (1.7)||142b (21.5)|
|VFA ≥ 100cm2||584b (60.0)||549b (83.3)|
In this study, subjects with two or more of the four risk factors (high blood pressure, high triglyceride, low high-density lipoprotein (HDL) cholesterol, and hyperglycemia) defined in the criteria of the National Cholesterol Education Program's Adult Treatment Panel III guidelines in 2005 , except for waist circumference, were defined as having the clustering of metabolic risk factors. Subjects currently receiving treatment for hyperlipidemia, hypertension, or diabetes were deemed as having the respective risk factors, regardless of the biochemical values. Subjects with metabolic risk factor and/or who were screened for diabetes, hypertension, hyperlipidemia, and obesity at the health check received brief recommendation on life style modification verbally and/or were advised to consult a doctor, if necessary.
The 3-year change in the participants' VFA (designated as ΔVFA) was categorized into seven groups (Group 1, ≤ −50 cm2; Group 2, > −50 cm2 and ≤ −30 cm2; Group 3, > −30 cm2 and ≤ −10 cm2; Group 4, > −10 cm2 and <10 cm2; Group 5, ≥ 10 cm2 and < 30 cm2; Group 6, ≥ 30 cm2 and < 50 cm2; Group 7, ≥ 50 cm2). Odds ratios (95% confidence intervals (CI)) of the 3-year incidence of each metabolic risk factor and/or clustering of metabolic risk factors according to the seven groups of ΔVFA were estimated with the use of a multiple logistic regression analysis, where adjustments were made for the following potential confounders: age, VFA, and each parameter of the metabolic risk factors (high blood pressure, hyperglycemia, high triglycerides, or low HDL cholesterol) at baseline. P value for trend across the seven groups was calculated by applying consecutive integers to the categories in the logistic regression model. Similarly, odds ratios for incidences of four risk factors for metabolic syndrome were calculated. All analyses were performed using SPSS (Version 15.0, SPSS, IL).
Characteristics of subjects at baseline are shown in Table 1. The mean values of age, the waist circumference, VFA, SFA, and BMI of the subjects who were without clustering of metabolic risk factors at baseline were 52.7 ± 8.4 years, 84.6 ± 7.8 cm, 113.2 ± 52.1 cm2, 118.2 ± 51.9 cm2, and 23.3 ± 2.6 kg/m2, respectively. The incidences of the metabolic risk factors at the 3-year follow-up survey are shown in Figure 1. The odds ratios for the incidences of the clustering of metabolic risk factors according to the ΔVFA groups were 0.45, 0.63, 0.74, 1.00 (ref.), 0.70, 1.04, and 3.91, respectively (trend P < 0.001). These results did not change after further adjustments for lifestyle factors (smoking and alcohol drinking) or lifestyle factors and ΔSFA. The odds ratios (95% CI) of the incidences of hyperglycemia, high triglycerides, and low HDL cholesterol for the group with the largest increase in ΔVFA (50 cm2 or more) were 2.98 (1.33-6.68), 4.88 (2.51-9.47), and 3.36 (1.26-8.91), respectively. The odds ratios (95% CI) of the incidences of high blood pressure, hyperglycemia, high triglycerides, and low HDL cholesterol for the group with the largest decrease in ΔVFA (−50 cm2 or less) were 0.83 (0.34-2.04), 1.16 (0.42-3.15), 0.19 (0.06-0.59), and 0.42 (0.13-1.39), respectively. In subjects with a decrease in ΔVFA, the incidences were low for high triglycerides and low HDL cholesterol, but did not change for high blood pressure and hyperglycemia. Of the components of the metabolic syndrome, especially the incidences of high triglycerides and low HDL-cholesterol paralleled with ΔVFA (trend P < 0.05).
This study investigated the change in metabolic risk factors between a baseline examination and a 3-year follow-up examination among subject groups divided according to the change in VFA, as measured using CT. We calculated the odds ratios for the incidences of metabolic risk factors and found that the risks of hyperglycemia, high triglycerides, low HDL cholesterol, and clustering of metabolic risk factors were significantly greater for the ΔVFA ≥ 50 cm2 group, compared with the stable ΔVFA groups. For subjects with a decrease in ΔVFA, the incidences of high triglycerides and low HDL cholesterol levels were low, but the incidences of high blood pressure and hyperglycemia did not change.
In previous cross-sectional studies, significant and positive correlations between VFA measured using CT and systolic blood pressure, serum triglycerides were shown . In previous longitudinal studies, the effects of changes in VFA , BMI , weight , and waist circumference [16-18] on the metabolic risk factors have been examined. One of those reports studied 1,106 men and showed that changes in the VFA were associated with changes in metabolic risk factors . Another report showed a strong linear trend between increasing BMI and a worsening of various variables of metabolic risk factors, including blood pressure and lipid profiles . Similarly, another report showed that weight changes were linearly related to all measurements of each component of metabolic risk factors . Other studies have shown that a reduction in waist circumference achieved through lifestyle modifications is closely linked to an improvement in metabolic risk factors [17, 18]. These studies suggest that the changes in VFA, BMI, weight, and waist circumference are related to risk factors for CVD. However, the impact of the change in VFA (measured twice in the same person), which is regarded as the strongest indicator of a risk of CVD, has remained uncertain. To our knowledge, this is the first study to analyze the relationship between the incidences of metabolic risk factors and changes in VFA by measuring VFA twice.
Our findings clearly showed that changes in VFA were significantly associated with the incidences of metabolic risk factors.
When the odds ratios of the incidences of the metabolic risk factors for the ΔVFA ≥ 50 cm2 were compared with those for the stable ΔVFA group (ΔVFA > −10 cm2, < 10 cm2) as a reference group, the highest odds ratio was seen for a high triglycerides level. The odds ratios for the incidences of high triglycerides and low HDL cholesterol levels, which are indicators of hyperlipidemia, were significantly higher among subjects with a larger increase in VFA and were significantly lower among subjects with a larger decrease in VFA (trend P < 0.05). In previous studies , the mean value of each component of metabolic risk factors or the prevalence of metabolic risk factors and its components were compared according to changes in BMI, body weight, and waist circumference. To our knowledge, however, no other studies have compared the strength of the association among metabolic risk factors to the change in VFA. Our study is the first to clarify the risk of each disease by calculating comparable odds ratios.
In conclusion, this study of Japanese men showed that changes in VFA were associated with the incidences of metabolic risk factors, with a significant increase in the odds ratios observed with a ΔVFA ≥ 50 cm2. This association was most pronounced for the risk of high triglycerides and low HDL cholesterol levels. The adoption of a lifestyle that does not result in an increase in VFA is important for preventing metabolic syndrome.
This study was supported by a grant from the Ministry of Health, Labour and Welfare of Japan.
- 12Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults.Executive summary of the third report of the national cholesterol education program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel). JAMA 2001;285:2486–2497.