Association between physical activity and metabolic syndrome in middle-aged Japanese: a cross-sectional study

The participants in this cross-sectional study were 521 healthy middle-aged Japanese volunteers without diabetes, cardiovascular disease, and musculoskeletal diseases recruited from local community newspapers in Tsukuba, Ibaraki. We excluded individuals who were missing data on physical activity (n = 11), MetS components (n = 18) or dietary intake (n = 9). After excluding these individuals, a total of 179 men and 304 women were included in this analysis [age, 30-64 years; body mass index (BMI), 17.2-45.5 kg/m²]. This study was approved by the Ethical Committee of the Institute of Health and Sport Sciences and the Institute of Clinical Medicine at the University of Tsukuba. Written informed consent was obtained from all participants. Participant characteristics (n = 483) are presented in Table 1.

Body weight was measured to the nearest 0.1 kg using a digital scale (TBF-551; Omron Healthcare Co., Ltd., Kyoto, Japan), and height to the nearest 0.1 cm using a wall-mounted stadiometer. BMI was calculated as body weight/height² (kg/m²). Waist circumference (WC) was measured three times to the nearest 0.1 cm at the midpoint between the lower costal margin and the iliac crest using a calibrated measuring tape. Body fat percentage (% body fat) and muscle mass percentage (% muscle mass) were calculated from the impedance value (HBF-352, Omron Healthcare Co., Ltd., Kyoto, Japan) [20].

A triaxial accelerometer (HJA-350IT, Active style Pro, Omron Healthcare Co., Ltd.) was used to measure physical activity and sedentary time [21]. To ensure a valid reflection of long-term day-to-day activities, the accelerometer was required to be worn by participants for 7 days while the subjects were continuing their regular daily activities [22]. The participants were instructed to wear the accelerometer on an elastic waist band on the right hip throughout the day from the time they woke up in the morning until they went to bed at night, except during showers and bathing. Participants who did not record 10 hours/day of activity for 7 days were excluded from further analyses [22]. The mean wear time was 11.8 ± 1.6 hours/day. Accelerometer data were calculated as the monitoring time spent in each of three different intensity levels for 1 week, using software provided by the manufacturer (HMS-HJA-IC01J; Omron Healthcare Co., Ltd.) as follows: sedentary (1.1-1.4 METs), low (1.5-2.9 METs), and moderate-to-vigorous (≥ 3 METs). To compare the impact of different levels of MVPA on the risk of MetS, we defined two categories of MVPA based on the physical activity reference recommended in the EPAR2006: < 23 METs h/week and ≥ 23 METs h/week.

Blood pressure was measured twice using an electronic digital blood pressure monitor (SM-100; Omron Healthcare Co., Ltd.) with subjects in a seated position after a 10-min rest period. If the two readings differed by more than 5 mmHg, a third reading was taken and the lowest was used. Approximately 10 mL of blood was drawn from each subject after an overnight fast. Fresh samples were used for enzymatic analysis of triglyceride levels, while fasting plasma glucose levels were assayed using the glucose oxidase method. Serum high-density lipoprotein cholesterol (HDL-C) was measured using heparin-manganese precipitation.

The Japanese criteria for MetS were used to evaluate the prevalence of MetS and pre-MetS in this study [23, 24]. Based on these criteria, to diagnose MetS, the subject must present with abdominal obesity, which is a mandatory criterion because it is a strong predictor of cardiovascular risk factors, in addition to two or more of the other components. For the diagnosis of pre-MetS, the subject must have abdominal obesity and one of the other components, including (1) WC ≥ 85 cm for men and ≥ 90 cm for women, which is considered an essential component; (2) dyslipidemia (triglyceride ≥ 150 mg per dL and/or HDL-C level < 40 mg per dL, or specific treatment for these lipid abnormalities); (3) hypertension (systolic blood pressure ≥ 130 mmHg and/or diastolic blood pressure ≥ 85 mmHg, or treatment of previously diagnosed hypertension); or (4) hyperglycemia (fasting plasma glucose ≥ 110 mg/dL). In this study, we analyzed physical activity for MetS subjects and pre-MetS subjects in a single group (MetS/pre-MetS).

A nutritionist assessed the daily nutritional intake of each subject, including carbohydrate, fat and protein, as well as the total calorie intake per day, by evaluating the patients' 3-day weight and dietary records. The food data of the dietary records were converted to energy and nutrient data by the dietician and analyzed using Excel Eiyo-kun software (Ver. 4.0; Kenpaku Co., Ltd., Tokyo, Japan). Participants also completed a general health questionnaire, which recorded smoking and menopausal status.

Logistic regression analysis was used to predict MetS from the levels of physical activity based on the EPAR2006 reference (≥ 23 METs h/week or < 23 METs h/week); this was done separately for men and women. In all logistic regression models, age, level of low intensity activity, sedentary time, smoking, calorie intake and BMI were included as covariates. The total cohort model was also adjusted for sex. The women's model was adjusted for menopause. Multicollinearity between sedentary time, level of low intensity activity and MVPA was tested but was not significant.

Classification and regression tree analysis (CART) [25, 26] was used to establish the cutoff value for required physical activity and the best predictor for the prevalence of MetS. CART analysis is constructed by repeatedly splitting subsets of the data using all explanatory variables to create two subgroups. Explanatory variables with the best improvement are selected to split the data based on the improvement score, and are evaluated to identify the optimal cutoff point (continuous variable) or groups (nominal variables). We calculated P values at each branch using χ² tests. Values are expressed as means ± standard error (SE) or as percentages. Values of P < 0.05 were considered statistically significant. All data were analyzed using SPSS version 17 or R version 2.10.1 [27].

This present study is the first, to our knowledge, to examine the association between the EPAR2006-recommended physical activity levels and MetS. The analysis showed that physically active men (≥ 23 METs h/week) were at less risk of MetS than their physically inactive counterparts (< 23 METs h/week). In contrast, the association between MVPA and MetS in women was less clear. Our physical activity findings were similar to those reported by other similarly designed studies [6‐8]. Contemporary analyses have shown that leisure-time physical activity is effective in preventing and treating MetS [7, 29], and most studies have shown that MetS is inversely associated with physical activity or physical fitness in men [6, 30, 31]. In contrast, few studies have observed inverse associations between physical activity and MetS in women [32].

Our study confirms the results of other studies showing an inverse association between physical activity and MetS, which applies particularly to men. Zhu et al. [33] reported a significant association between physical activity and MetS risk factors in men, but not in women, after considering covariates like age, race, income level and education. Moreover, the prevalence of MetS was 31% lower in physically active men than in their inactive counterparts, whereas the corresponding percentage among women was 17% [33].

Another study has noted differences between men and women in the association between physical activity or fitness and MetS [34]. Brien et al. [6] reported a significant association between physical activity and MetS in men, but not in women, after adjusting for age, smoking, alcohol consumption and income level. The authors attributed these sex differences to hormonal differences, varying patterns of fat distribution, differences in the definition of MetS, and differences in the types and intensities of physical activity [8, 34]. Therefore, the results of the present study and those of other investigations may represent a true difference between sexes in terms of the association between physical activity and MetS.

This study also aimed to establish a cutoff value for the required amount of physical activity that would prevent MetS in middle-aged Japanese individuals. Although many studies have included cross-sectional analyses [6, 9, 13, 30, 31], our study was the first, to our knowledge, to examine the optimal cutoff value of physical activity required to prevent the onset of MetS. CART analysis revealed that the optimal cutoff value of 26.5 METs h/week represented the level of physical activity required to prevent MetS, and was slightly higher than the value reported by EPAR2006 (23 METs h/week) to prevent lifestyle-related disease [14, 15].

Furthermore, applying the EPAR2006 physical activity recommendations resulted in a MetS prevalence of 41.0% (data not shown), which was much higher than that determined in our study (25.4%) when we applied the new cutoff value for physical activity (Figure 1). These results suggest that the EPAR2006 recommendations for physical activity should be increased to reduce the prevalence of MetS among middle-aged Japanese individuals. Our results do not negate the current EPAR2006 physical activity reference value, and the discrepancy between the physical cutoff values can be explained. First, the EPAR2006 physical activity reference was established using a systematic review that primarily summarized the evidence on the effects of physical activity on lifestyle-related disease, rather than focusing on MetS. Second, this systematic review included studies performed in Western countries, and relatively few studies in Japanese individuals were included. Third, the techniques used to measure physical activity may have differed. However, the present study established an association between objectively measured physical activity and MetS in Japanese individuals. For these reasons, the findings of this study might have an important influence on effective promoting physical activity focused on preventing MetS among middle-aged Japanese individuals.

The primary strength of our study was that we used a triaxial accelerometer to objectively measure physical activity and sedentary times. However, there are several limitations to be discussed. First, it is not possible to infer causality or specify the direction of the observed effects because of the cross-sectional study design. Second, prevalence of MetS/pre-MetS was higher in this study subjects (men: 53.1%, women: 34.2%) compared with general Japanese adults in the National Health and Nutrition Examination Survey in 2008 (men: 45.3%, women: 18.6%). Therefore, the results in this study may not generalize to the Japanese population. Although we controlled for several covariates, there may be other unexplored residual confounding variables such as genetic variation and sociocultural factors that may partly explain our findings. Nevertheless, we considered factors directly related to MetS prevalence, including sedentary time, low intensity physical activity, calorie intake, smoking, menopause status and BMI [35‐38], in this study. The study sample size may also be a limitation, but it was sufficiently large to allow us to perform CART analyses to determine the optimal cutoff value for physical activity necessary to prevent MetS. Despite these limitations, our findings help us to understand the association between physical activity and MetS in middle-aged Japanese individuals.