Locomotive syndrome is associated with body composition and cardiometabolic disorders in elderly Japanese women

This study was conducted in a rural area (Tanabe city, Wakayama Prefecture, Japan) between January 2013 and March 2015. The study inclusion criteria were as follows: 1) Japanese women, age > 60 years; 2) ability to walk independently; and 3) living at home and being capable of self-care. All participants initially underwent body composition measurements, in the order of height, weight, and body fat percentage, followed by an evaluation of LS status at a public hall where a “Lecture meeting and checkup for health” was held with support from the local government. Afterwards, 198 women were asked to complete a self-report questionnaire at home regarding comorbid conditions and then to return the questionnaire by mail. A stamped envelope was provided to encourage the return of the questionnaire. A total of 165 women underwent the measurements and returned the questionnaire (mean age ± standard deviation, 68.8 ± 6.1 years; range, 60–83 years).

Body weight and body fat percentage were measured using the KaradaScan Body Composition Monitor with Scale (HBF-362; Omron Co., Kyoto, Japan) while participants were wearing normal indoor clothing. The procedure was performed with the participants standing barefoot on a metal surface in accordance with the manufacturer’s instructions. BMI was calculated as weight divided by the square of height, and obesity was defined as BMI ≥25 kg/m² in accordance with the guidelines of the Japan Society for the Study of Obesity [27]. Bone status was assessed using speed of sound (SOS) measured using a quantitative ultrasound (QUS) device (Canon Life Care Solutions Inc., CM-200, Osaka, Japan) at the calcaneus of the dominant foot while the participants were barefoot and seated. QUS, which has a number of advantages, including portability, low cost, and a lack of exposure to radiation [28], enables the evaluation of bone quality, especially the microarchitecture at the calcaneus, and is useful for predicting risk for future fracture [29]. Bone status was shown as the percent of Young Adult Mean of the SOS (%YAM) [30]. It has been reported that calcaneal QUS parameters reflect the characteristics of the trochanteric area of the proximal hip, although these values are not specifically reflective of values of the femoral neck or shaft [31].

LS status (presence and degree) was evaluated based on GLFS-25 scores (Additional file 1). The GLFS-25 is a self-report questionnaire composed of the following 25 items focusing on the month before completing the measure: four questions on pain; 16 on activities of daily living; three on social function; and two on mental health status [7]. These 25 items are scored from 0 (no impairment) to 4 (severe impairment), with a total score range from 0 to 100. Higher scores indicate worse locomotive function. The cutoff score for LS, as determined by receiver operating characteristic (ROC) analysis, is 16 points [7].

Comorbid conditions of hypertension, hyperlipidemia, and diabetes were assessed using the following question on the self-report questionnaire: “Do you presently take medication for hypertension, diabetes, or hyperlipidemia?”

Participants were classified as LS (≥16) or non-LS (<16) based on GLFS-25 scores, and then independent variables were compared between groups. For numerical variables, normality of distribution and homogeneity of variance were tested before across-group comparisons. When the assumptions of normal distribution and homogeneity of variance were met in both groups, we performed the student t-test, and when the assumption of normal distribution was met, but not the assumption of homogeneity of variance, we performed Welch’s t-test. When the data were non-normally distributed, the Wilcoxon signed-rank test was used.

ROC analysis was used to evaluate the threshold of each body composition measure (BMI, body fat percentage, and %YAM) in order to discriminate the LS from the non-LS group. An area under the ROC (AUC-ROC) curve of 1.00 was taken to indicate perfect discrimination, whereas an AUC-ROC of 0.50 was taken to indicate the complete absence of discrimination.

Multiple logistic regression analysis was performed to evaluate the age-adjusted significance of the prevalence of LS. The chi-square test was used for comparison of prevalence or number of cardiometabolic disorders between non-LS and LS. The Wilcoxon signed-rank test was used for comparison of GLFS-25 scores classified by with and without cardiometabolic disorders, as well as by the number of present disorders. Statistical analysis was conducted using JMP 11 (SAS Institute, Cary, NC). All statistical tests were 2-tailed, and a significance level of 0.05 was used.

LS was proposed by the JOA in 2007 in order to identify individuals at high risk of requiring nursing care owing to problems associated with the locomotive system [2]. The GLFS-25 was subsequently developed to measure the presence and degree of LS in Japanese individuals [7]. However, since its implementation, the GLFS-25 cutoff value for identifying individuals with LS has been determined in accordance with health-related quality of life [7, 11]; therefore, information on the association between GLFS-25 scores and body composition is limited. Therefore, the primary purpose of this study was to determine the association between LS as defined by GLFS-25 scores and body composition measures in elderly Japanese women.

Our results showed that participants with LS were shorter, had a higher body fat percentage, a higher BMI, and lower bone status than participants without LS. Previous studies have reported similar results in middle-aged and elderly Japanese women [5, 14]. Muramoto et al. found that GLFS-25 scores had a significant positive correlation with body fat percentage and BMI, a negative correlation with body height and BMD, and no correlation with body weight according to correlation analysis [5].

Based on comparative analysis, participants with LS have been shown to have significantly greater BMI and body fat percentage and lower height than those without LS, whereas no significant difference has been observed in body weight or BMD [14].

In the present study, we found that participants with LS were shorter than those without LS. Shorter height has been reported to be significantly associated with fear of falling in elderly Japanese individuals [32]. Shorter height may be caused by an age-related change in the curvature of the spine or atrophy of trunk extension muscles, which can decrease postural control. A reduction in postural control can cause fear of falling or a decline in the amount of physical activity [33]. Therefore, we propose that the LS group included more participants that had lost height due to a change in the curvature of the spine or atrophy of trunk extension muscles than the non-LS group, and therefore had less postural control and engaged in fewer activities in daily life, which increased their risk for developing LS.

The present results showed that participants with LS had a higher body fat percentage than those without LS. Increased body fat causes more mechanical stress in weight-bearing joints and promotes the degeneration of joint tissue through the production and release of adipokines [34]. Adipokines are derived from adipocytes and may upregulate receptor activators of nuclear kappa B ligand, leading to increased bone resorption and reduced BMD [35]. Participants with a higher body fat percentage may have secreted more adipocytes, and this may have had a negative influence on the movement of the joints, thereby increasing the risk of LS.

The present study showed that a BMI ≥23.5 k/m² was significantly associated with LS, with an OR of 3.78 as identified by multiple logistic regression analysis. The Japan Society for the Study of Obesity defines the cutoff for obesity as a BMI 25 kg/m² [8]. In the present study, the mean BMI of the participants with LS was ≥25 kg/m². Furthermore, GLFS-25 scores were higher in obese than in non-obese participants (Fig. 2d). LS is closely associated with age-related skeletal disorders such as osteoporosis, OA, lumbar spinal stenosis (LSS), degenerative spinal disease and sarcopenia [4]. Furthermore, obesity is a risk factor for these disorders because mechanical overload on weight-bearing joints can activate chondrocytes, accelerate the degeneration of cartilage, and increase static compressive loading and pressures associated with postures that damage disc integrity [36‐38]. Moreover, it has been proposed that metabolic factors, including inflamed adipose tissue, dyslipidemia, oxidative stress, endothelial dysfunction and leptin dysregulation, as well as the clustering of these factors in metabolic syndrome, may play a crucial role in obesity-induced OA [39‐41]. These findings support the present results regarding the association between obesity and LS.

Obesity, hypertension, diabetes, and dyslipidemia are known as the “deadly quartet” [24]. Numerous studies have reported that BMD is associated with these disorders [17‐21]. The second purpose of the present study was to determine the association between LS and cardiometabolic disorders. Our results showed that LS is associated with hypertension, diabetes, and overweight, as well as with higher BMI. Furthermore, GLFS-25 scores significantly increased with the number of present cardiometabolic disorders.

There are some reports on the association between cardiometabolic disorders and OA. Some evidence suggests that metabolic factors such as type 2 diabetes mellitus and elevated glucose concentration are associated with the development and progression of OA [40, 42]. In particular, the advanced glycation end products in cartilage collagen seem to be associated with both the senescent cartilage matrix and reduced chondrocyte function [43]. The presence of advanced glycation end products associated with the expression of advanced glycation end-product receptors in the cartilage collagen results in the increased production of matrix metalloproteinase and the modulation of the chondrocyte phenotype to hypertrophy and OA [44, 45].

OA and hypertension have been shown to frequently coexist [46]. The proposed mechanism of the development of OA with hypertension is as follows: narrow and/or constricted vessels restrict blood flow to subchondral bone, impairing circulation and nutritional supply to overlying articular cartilage, which ultimately contributes to the deterioration of cartilage in OA [47].

Mutual relations exist between the occurrence and presence of musculoskeletal diseases, particularly knee OA and cardiometabolic disorders [48]. Yoshimura et al. suggested that metabolic risk factors such as overweight, hypertension, hyperlipidemia, and impaired glucose tolerance increase the risk of occurrence and progression of knee OA [49, 50]. Recent reports have indicated that waist circumference, back muscle strength, and spinal inclination angle are important risk factors for LS [22]. In the present study, we demonstrated that LS is associated with hypertension and obesity, as well as a higher BMI. Furthermore, GLFS-25 scores significantly increased with the number of present cardiometabolic disorders. These findings suggest a close relationship between the locomotive system and cardiometabolic organs.

The proportion of adults with BMI >25 kg/m² has significantly increased worldwide [51]. The present findings contribute to the identification of factors that may prevent locomotive disorder and metabolic syndrome, particular in Western societies, in which many patients have metabolic syndrome. Although the concept of LS is currently used only in Japan, we believe it will become more common worldwide as the population continues to age.

The results of the present study suggest that BMI might be a useful measure for the simple detection of LS. Furthermore, hypertension and diabetes were found to be associated with LS. Weight management and prevention of these disorders may help protect against LS in elderly women. Elderly men should be included in future studies.

This study did have several limitations. First, the sample size of 165 was small; this number only represents about 1.2 % of all women aged between 60 and 83 years in Tanabe city. Furthermore, no significant relationship was found between LS and dyslipidemia; this may have been due in part to a lack of statistical power. Second, because the participants in this study were all Japanese women, care should be taken in generalizing the results to men or other ethnic groups. Third, data from a cross-sectional study are not sufficient to determine whether a causal relationship exists between BMI, LS, and cardiometabolic disorders. LS may cause obesity or hypertension and diabetes because it limits physical activity. Conversely, these cardiometabolic disorders may lead to the development of LS. It is therefore crucial to perform longitudinal studies to clarify the causal relationships among these factors. Fourth, comorbid conditions were only assessed using self-report questionnaires; therefore, blood pressure, blood glucose concentration, and blood lipid concentration measurements were not controlled. Thus, untreated participants with comorbid conditions may have been excluded from analysis; however, this possibility is low because participants attending the “Lecture meeting” would have been expected to have relatively high health awareness. Fifth, further research in larger-sized studies should measure lean mass because it is an important component of BMI. It is possible that BMI underestimates body fat percentage in clinical populations [52].