Mid-upper arm circumference, central obesity and metabolic syndrome in middle-aged and elderly Chinese: the REACTION study

Background The mid-upper arm circumference (MUAC) is a proxy for upper body subcutaneous fat and a reliable screening measure for identification of individuals with abnormal local fat distribution. The purpose of present study was to evaluate the association between MUAC and metabolic syndrome (MetS) as well as other metabolic phenotype in the middle-aged and elderly population. Methods We measured the MUAC in a cross-sectional sample with a total of 9787 subjects aged 40 years and older in Shanghai, China. The measurement of MUAC is performed on the right arm using a non‐elastic tape held midway between the acromion and the olecranon processes in duplicate, with arm hanging loosely at the side of the body. The MetS was defined according to the Joint Statement of the International Diabetes Federation Task Force on Epidemiology and Prevention. The association of MUAC with MetS was tested in logistic regression analyses and reported as odds ratio (OR) with 95% confidence interval (CI). Results prevalence of MetS, overweight, and central obesity raised sharply from the smaller MUAC groups to the larger MUAC groups both in men and women. MUAC was positively correlated with waist circumference, BMI, fasting insulin, HOMA-IR, triglycerides, systolic and diastolic blood pressure, and inversely correlated with adiponectin and HDL-cholesterol after adjusting for age and gender. As the MetS components accumulated, the MUAC increased concomitantly. The risk of MetS in the highest MUAC quartile was significant higher (odds ratio 1.77; 95% confidence interval 1.51-2.09) than the lowest quartile after adjustment of potential confounders. Conclusion Our results indicated that large MUAC is an independent risk factor of MetS in Chinese individuals.


Abstract
Background The mid-upper arm circumference (MUAC) is a proxy for upper body subcutaneous fat and a reliable screening measure for identification of individuals with abnormal local fat distribution. The purpose of present study was to evaluate the association between MUAC and metabolic syndrome (MetS) as well as other metabolic phenotype in the middle-aged and elderly population.
Methods We measured the MUAC in a cross-sectional sample with a total of 9787 subjects aged 40 years and older in Shanghai, China. The measurement of MUAC is performed on the right arm using a non-elastic tape held midway between the acromion and the olecranon processes in duplicate, with arm hanging loosely at the side of the body. The

MetS was defined according to the Joint Statement of the International Diabetes
Federation Task Force on Epidemiology and Prevention. The association of MUAC with MetS was tested in logistic regression analyses and reported as odds ratio (OR) with 95% confidence interval (CI).
Results T he prevalence of MetS, overweight, and central obesity raised sharply from the smaller MUAC groups to the larger MUAC groups both in men and women. MUAC was positively correlated with waist circumference, BMI, fasting insulin, HOMA-IR, triglycerides, systolic and diastolic blood pressure, and inversely correlated with adiponectin and HDLcholesterol after adjusting for age and gender. As the MetS components accumulated, the MUAC increased concomitantly. The risk of MetS in the highest MUAC quartile was significant higher (odds ratio 1.77; 95% confidence interval 1.51-2.09) than the lowest quartile after adjustment of potential confounders. Conclusion Our results indicated that large MUAC is an independent risk factor of MetS in Chinese individuals. Background 4 Metabolic syndrome (MetS), which contains a cluster of metabolic abnormalities including central obesity, hypertension, dysglycemia, and dyslipidemia that together culminate in the increased risk of cardiovascular disease (CVD) and diabetes, is a major global public health problem [1]. Of note, the prevalence of MetS is up to 30% in middle-aged and elderly people in China [2,3]. Given the continuous increase of aging population in China, high prevalence of MetS is a concerning observation which should be put on the agenda.
It is well established that upper body obesity is a high risk for hypertension, hyperlipidemia, and dysglycemia [4]. One predominant mechanism linking upper body obesity with these components of MetS is abnormally elevated free fatty acids (FFAs) flux [5]. Notably, instead of visceral fat, it is upper body subcutaneous adipose tissue as main source of excess FFA in upper body obese individuals [5,6]

Data collection
A standardized questionnaire was applied by trained physicians to collect essential information, including sex, age, lifestyle factors, education status, physical activity and previous medical history. Anthropometric measurements were collected by certified physicians using standard protocols in duplicate base on the NHANES Anthropometry Procedures Manual (https://wwwn.cdc.gov/nchs/data/nhanes/2015-2016/manuals/2016_Anthropometry_Procedures_Manual.pdf). The measurement of MUAC is performed on the right arm using a non-elastic tape held midway between the acromion and the olecranon processes, with arm hanging loosely at the side of the body. Waist circumference was measured with a non-elastic tape held midway between the lower rib margin and the iliac crest at the end of a gentle expiration. Blood pressure (BP) was measured with an automated electronic device (OMRON Model1 Plus; Omron Company, Kyoto, Japan). Overweight was defined as body mass index (BMI) ≥24 kg/m 2 , central obesity was defined as WC ≥85 cm for men and ≥80 cm for women.

Biochemical measurements
Peripheral venous blood samples were collected after an overnight fast for at least 10 hours. The plasma glucose level was measured by glucose oxidase method (ADVIA-1650 Chemistry System, Bayer, Leverkusen, Germany). Total cholesterol, triglycerides, low-

Statistical analysis
Normally distributed data were reported as means ± SD, whereas variables with a skewed distribution were expressed as median (interquartile range) and log transformed to approximate normality before analysis. Categorical variables were depicted by frequency and percentage. Analysis of covariance for continuous variables and multivariate logistic regression analysis for categorical variables were applied for the comparison according to MUAC quartiles. Correlation coefficients between MUAC and metabolic features were calculated by partial correlation analysis after adjusted age and gender. Multivariate logistic regression models were used to estimate the odds ratios (ORs) for MetS and its components. Potential confounding variables including age, gender, smoking, alcohol drinking, physical activity, educational status, self-reported CVD, C-reactive protein (CRP), adiponectin, BMI, and HOMA-IR were controlled in the regression models. Data management and statistical analysis were performed with SPSS (version 23.0). P<0.05 was considered statistically significant.

Characteristics of participants according to MUAC quartiles
The mean of MUAC were 29.24 cm for male and 28.41 cm for female (P <0.001), respectively. The individuals were divided into four groups based on the quartiles of MUAC. When analyzed by quartiles of MUAC levels, as summarized in Table 1, the subjects with larger MUAC were more likely to be smokers, alcohol drinkers, and with comorbidities including hypertension, hyperlipidemia, diabetes (all P <0.05). With regard to metabolic parameters, the individuals in the higher MUAC quartiles showed higher levels of SBP, DBP, BMI, waist circumference or waist-to-hip ratio, fasting glucose, insulin, HOMA-IR, CRP, triglycerides (all P <0.001), and LDL cholesterol (P = 0.003). Conversely, the subjects with larger MUAC displayed lower plasma adiponectin and HDL cholesterol levels (both P <0.001). However, elevated MUAC exhibited no association with the levels of total cholesterol in this study.

Association between MUAC and MetS
Partial correlation analysis shown that MUAC has the strongest correlation with waist circumference after adjusted for age and gender (Table 2). Additionally, MUAC was also strongly correlated with BMI and insulin.
As presented in Figure1, the prevalence of MetS, overweight, and central obesity raised sharply from the smaller MUAC groups to the larger MUAC groups both in men and women ( Fig. 1). In addition, women seem to have higher prevalence of MetS and central obesity compared with men. From 1st to 4th quartile (Q1 to Q4), the prevalence of MetS were 32.0%; 45.6%;59.3%; 72.4% (P = 0.007) in men, and 35.4%; 49.4%; 64.5%; 75.1% (P <0.001) in women, respectively. In the largest MUAC subgroups (Q4), the prevalence of overweight was 83.6% in men and 83.3% in women while the prevalence of central obesity was 84.7% in men and 88.4% in women.
Furthermore, with the accumulation of MetS components, MUAC was gradually increased both in male and female (Fig. 2). It is worth noting that as the components of MetS increased one by one, the relationship between MUAC and MetS was statistically significant, both in subjects with Mets (number of MetS components ≥3) and the subjects without MetS (number of MetS components <3). On this ground, MUAC may be directly or indirectly related to each MetS component.

Discussion
We observed a significant and independent association between large MUAC and the increased risk of MetS and its key components in a large-scale population study.
Furthermore, we found that MUAC is strongly correlated with BMI, waist circumference, and insulin, which indicated that large MUAC is a potential screening tool for identifying overweight, central obesity, and insulin resistance.
BMI and waist circumference are commonly screening tool for identifying individuals with abnormal distribution of body fat. Nevertheless, BMI cannot provide accurate information about the local distribution of body fat and it is difficult to obtain height and weight for patients who cannot stand. As for waist circumference, the deficiency of daily application lies in the big difference of preprandial and postprandial measurements. In view of above reasons, MUAC began to show diagnostic value for assessing nutritional status. Compared to other anthropometric measurements, MUAC is not only easier to obtain, but also has other advantages such as being more accurate, convenient and low-cost. Small MUAC has shown excellent performance in assessing malnutrition and predicting mortality both in children [12] and older individuals [13,14]. More recently, large MUAC has been recognized as a valid tool for detecting overweight and obesity in children and adolescents [15,16].
However, the study about whether MUAC is associated with obesity-related metabolic abnormality, such as MetS, is scare. Currently, we demonstrate that large MUAC, as a proxy of upper-body subcutaneous adipose, is a risk factor of MetS. Moreover, consistent with previous studies, we found large MUAC also tightly correlated with overweight and central obesity among Chinese middle-aged and older subjects.
Obesity, characterized by the expansion of adipose tissue, is a key causative factor in the development of MetS. The abnormal accumulation of fat affects adipose tissue metabolic capacities, endocrine, and immune function and leads to altered production of lipid mediators, adipokines, pro-or anti-inflammatory cytokines, and impaired signalling pathways that contribute to obesity related metabolic abnormality [17]. Moreover, the adipose tissue is not only a depot of excess energy but also a highly active metabolic endocrine organ that secretes numerous biologically active molecules, which are collectively termed adipokines [31]. When adipose tissue expands, the capacity of adipocytes to function as endocrine cells and secrete various adipokines is altered in individuals with obesity and MetS [27,32]. These abnormal levels of adipokine, including adiponectin, leptin, and retinol-binding protein 4, are linked to insulin resistance, impaired triglyceride storage and increased fatty acids in circulation [33]. Furthermore, as fat accumulation, substantial infiltration of immune cells occurs, and there is a specific crown-like disposition of macrophages around single necrotic adipocytes in obese people [34] and subjects with MetS[32]. Subsequently, proinflammatory pathways were activated, and certain proinflammatory cytokines and chemokines ( i.e. TNF-α, IL-1β, IL-6 and MCP-1) were overflowed that result in low-grade inflammation and insulin resistance [17,32]. In line with previous studies, our findings also indicated that MUAC is positive correlated with CRP and negative correlated with adiponectin, which is a welldocumented adipokine with potent anti-inflammatory activity. Overall, adipose dysfunctions, inflammation, and stress linking mid-upper arm obesity to insulin resistance and MetS.
To our knowledge, this is the first study to evaluate the association between MUAC and Mets among large-scale middle-aged and older people. The major strength of this study is the analysis based on a large sample. Potential covariates were strictly controlled in the analysis, so as to eliminated the possibility of residual confounding effects. Nevertheless, there are several limitations in this study. For one thing, we did not measure tissue composition of the mid-upper arm. Due to limitations of epidemiological screening conditions, we could not quantify the adipose accumulation by more accurate radiographic measures. Therefore, the amount and size of subcutaneous adipocyte and muscle fat are not clear. For another, due to the present study is a cross-sectional analysis, we cannot draw the causality from our findings. Additionally, it is still unclear whether our findings in middle-aged/older Chinese subjects can be generalized to younger populations or individuals of other ethnicities.
In brief, our study demonstrated that large MUAC is correlated with an elevated risk of having MetS even after adjustment for potential covariates. These findings provide a novel insight in the association between upper body obesity and MetS, and a potential screening  c These variables were log transformed before analysis.
d Self-reported CVD including stroke and coronary heart disease. All correlation coefficients were calculated after adjustment for age and gender.