Advertisement

Preserved Lower Limb Muscle Mass Prevents Insulin Resistance Development in Nondiabetic Older Adults

Open AccessPublished:December 30, 2022DOI:https://doi.org/10.1016/j.jamda.2022.12.002

      Abstract

      Objective

      To investigate roles of sarcopenia indexes in prediction of development of insulin resistance in nondiabetic older adults.

      Design

      A 2-year follow-up cohort.

      Setting and participants

      The Tanno-Sobetsu study, a prospective observational cohort, included 194 community-dwelling nondiabetic older adults during 2017-2019.

      Methods

      Lower limb, upper limb, appendicular, and trunk muscle masses by a bioelectrical impedance analysis, grip strength, knee extension torque, and walking speed were measured in study participants aged ≥65 years (79 men and 115 women) at baseline. Muscle mass and strength were divided by the weight, and then multiplied by 100 to calculate the weight ratio (%). Insulin resistance was assessed by homeostasis model (HOMA-IR) at baseline, and the study participants whose HOMA-IR was less than 1.73 at baseline were followed for a maximum of 2 years. The study endpoint was development of insulin resistance defined as HOMA-IR ≥1.73. The adjusted hazard ratio (HR) of each sarcopenia component for development of insulin resistance was calculated.

      Results

      Lower limb muscle mass (HR 0.88, 95% CI 0.79-0.98) and appendicular muscle mass (HR 0.89, 95% CI 0.81-0.99), but not other sarcopenia components, were associated with the development of insulin resistance, independently of sex and age, HOMA-IR, and waist circumference at baseline.

      Conclusions and Implications

      The loss of lower limb muscle mass is a significant risk factor for development of insulin resistance independently of obesity in nondiabetic older adults. The lower limb muscle mass may be a novel target of interventions for the prevention of diabetes in older adults.

      Keywords

      The population of Japan is rapidly aging, with the percentage of the population aged ≥65 years at 28.8% in 2021, and is estimated to reach 33.3% by 2036.

      Cabinet Office of Japan. The aging society white paper in 2021. Accessed December 24, 2021. ​https://www8.cao.go.jp/kourei/whitepaper/w-2021/zenbun/pdf/1s1s_01.pdf

      One of the most common diseases that increases with ageing is type 2 diabetes.
      • Wild S.
      • Roglic G.
      • Green A.
      • et al.
      Global prevalence of diabetes: estimates for the year 2000 and projections for 2030.
      ,
      • Charvat H.
      • Goto A.
      • Goto M.
      • et al.
      Impact of population aging on trends in diabetes prevalence: a meta-regression analysis of 160,000 Japanese adults.
      Because type 2 diabetes in older adults is a strong risk factor for atherosclerotic cardiovascular events and death,
      • Rockwood K.
      • Awalt E.
      • MacKnight C.
      • et al.
      Incidence and outcomes of diabetes mellitus in elderly people: report from the Canadian Study of Health and Aging.
      ,
      • Huang E.S.
      • Laiteerapong N.
      • Liu J.Y.
      • et al.
      Rates of complications and mortality in older patients with diabetes mellitus: the diabetes and aging study.
      the numbers of diabetes and diabetes-related deaths are predicted to further increase in the next decades. Insulin resistance is a major pathophysiology underlying prediabetic glucose intolerance and type 2 diabetes. It has long been known that the accumulation of visceral fat is closely related to the development of insulin resistance.
      • Fujioka S.
      • Matsuzawa Y.
      • Tokunaga K.
      • et al.
      Contribution of intra-abdominal fat accumulation to the impairment of glucose and lipid metabolism in human obesity.
      However, it has been reported that older adults often have increased insulin resistance without obesity, unlike the middle aged.
      • Fujii M.
      • Ohnishi H.
      • Saitho S.
      • et al.
      Comparison of the effect of abdominal obesity on new onset of type 2 diabetes in a general Japanese elderly population with that in a non-elderly population-The Tanno and Sobetsu study.
      In fact, 75% of patients with type 2 diabetes aged ≥65 years are nonobese in Japan.
      • Yoshimura Y.
      • Kamada C.
      • Takahashi K.
      • et al.
      Relations of nutritional intake to age, sex and body mass index in Japanese elderly patients with type 2 diabetes: the Japanese Elderly Diabetes Intervention Trial.
      Therefore, interventions other than those to reduce visceral fat mass might be important for prevention of diabetes in older adults.
      Sarcopenia is an age-related decline in skeletal muscle mass, strength and function, which is associated with increased disability in older adults.
      • Chen L.K.
      • Woo J.
      • Assantachai P.
      • et al.
      Asian working group for sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment.
      ,
      • Marzetti E.
      • Calvani R.
      • Tosato M.
      • et al.
      Sarcopenia: an overview.
      Because skeletal muscle is a major target organ for insulin, sarcopenia is thought to influence risk of development of insulin resistance. In fact, several cross-sectional studies have shown an association between sarcopenia and increased insulin resistance.
      • Srikanthan P.
      • Hevener A.L.
      • Karlamangla A.S.
      Sarcopenia exacerbates obesity-associated insulin resistance and dysglycemia: findings from the National Health and Nutrition Examination Survey III.
      ,
      • Kim T.N.
      • Park M.S.
      • Lim K.I.
      • et al.
      Relationships between sarcopenic obesity and insulin resistance, inflammation, and vitamin D status: the Korean Sarcopenic Obesity Study.
      In our recent cross-sectional study, among the components of sarcopenia, loss of lower limb muscle mass was found to be strongly associated with insulin resistance in nondiabetic older adults.
      • Seko T.
      • Akasaka H.
      • Koyama M.
      • et al.
      Lower limb muscle mass is associated with insulin resistance more than lower limb muscle strength in non-diabetic older adults.
      However, there are few longitudinal studies on the relationship between sarcopenia and insulin sensitivity, and results of a recent study suggests that sarcopenia, which was defined by low muscle mass, low muscle strength and/or walking speed has little impact on risk of insulin resistance.
      • Scott D.
      • Cumming R.
      • Naganathan V.
      • et al.
      Associations of sarcopenic obesity with the metabolic syndrome and insulin resistance over five years in older men: The Concord Health and Ageing in Men Project.
      To our knowledge, no study has examined relationships between multiple indices of sarcopenia and temporal change in insulin sensitivity in older adults. This issue was examined in the present study by use of data from a cohort that has been prospectively followed for decades.

      Methods

      Participants

      This longitudinal study used data from the Tanno-Sobetsu study, a prospective community-based cohort study aimed at elucidation of cardiovascular risk factors.
      • Matsumoto T.
      • Ohnishi H.
      • Sato T.
      • et al.
      Insulin resistance is associated with longitudinal changes of cardiac repolarization heterogeneity in apparently healthy subjects.
      Of the 605 residents who underwent a medical examination in Sobetsu Town in 2017, a total of 342 individuals aged ≥65 years also underwent sarcopenia-related testing. The inclusion criterion of the age for sarcopenia-related tests was ≥65 years according to the definition of an older adult because sarcopenia is more frequent and a larger social issue in older adults than in younger ones.
      • Chen L.K.
      • Woo J.
      • Assantachai P.
      • et al.
      Asian working group for sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment.
      As we previously reported,
      • Oimatsu H.
      • Saitoh S.
      • Ura N.
      • et al.
      A practical index for evaluation of insulin resistance.
      ,
      • Ohnishi H.
      • Saitoh S.
      • Ura N.
      • et al.
      Relationship between insulin resistance and accumulation of coronary risk factors.
      insulin resistance was defined as the homeostasis model assessment of insulin resistance (HOMA-IR) ≥1.73. The sensitivity and specificity of HOMA-IR ≥1.73 were 64.3% and 78.9%, respectively, when insulin resistance determined by euglycemic hyperinsulinemic glucose clamp technique was used as a gold standard.
      • Oimatsu H.
      • Saitoh S.
      • Ura N.
      • et al.
      A practical index for evaluation of insulin resistance.
      ,
      • Ohnishi H.
      • Saitoh S.
      • Ura N.
      • et al.
      Relationship between insulin resistance and accumulation of coronary risk factors.
      From the 342 residents, 130 individuals who already had insulin resistance (HOMA-IR ≥ 1.73) or type 2 diabetes [fasting glucose ≥ 126 mg/dL, hemoglobin A1c (HbA1c) ≥ 6.5% or with treatment for diabetes] at baseline and 18 individuals who could not be followed up or had missing data were excluded. The remaining 194 individuals contributed to analyses in the present study (Supplementary Figure 1).
      This study conformed to the principles outlined in the Declaration of Helsinki and was performed with the approval of the institutional ethical committee of Sapporo Medical University (approval number: 24-2-21). Written informed consent was received from all of the participants.

      Anthropometrics and Sociodemographic Data

      Participants received a medical examination in the early morning after overnight fasting. Height and weight were measured using a digital scale (TANITA Co, Ltd); body mass index was calculated as weight (kg) / height (m) squared. Waist circumference was measured around the height of the navel when viewed from the front after exhaling. Information regarding regular medication for type 2 diabetes and smoking habit (current/former/never) was collected by public health nurses in an interview form.

      Muscle Mass Measurements

      Muscle mass (kg) was measured using a bioelectrical impedance analysis (BIA; In Body 470; In Body Japan Co), and lower limb, upper limb, and trunk muscle masses were also determined at baseline. In addition, appendicular muscle mass was calculated as a sum of the lower and upper limb muscle masses. Each muscle mass was divided by the weight and multiplied by 100 to calculate the weight ratio (%).
      • Lim S.
      • Kim J.H.
      • Yoon J.W.
      • et al.
      Sarcopenic obesity: prevalence and association with metabolic syndrome in the Korean Longitudinal Study on Health and Aging (KLoSHA).
      This system applies electricity at frequencies of 5, 50, 250, and 500 kHz through the body. It has been shown that the coefficients of determination (R2) between the muscle mass measured by the BIA and the muscle mass measured by the dual-energy X-ray absorptiometry method, which is the most reliable method for determination of body composition, are high (ie, R2 = 0.88 in men and 0.83 in women).
      • Kim M.
      • Shinkai S.
      • Murayama H.
      • et al.
      Comparison of segmental multifrequency bioelectrical impedance analysis with dual-energy X-ray absorptiometry for the assessment of body composition in a community-dwelling older population.

      Muscle Strength and Muscle Function Measurements

      As indexes of muscle strength, grip strength, and isometric knee extension torque were measured at baseline. The grip strength (kg) of both hands was measured twice using a Smedley-type grip strength meter (Grip D; Takei Scientific Instruments Co, Ltd) when the participants were standing with full elbow extension.
      • Wong S.L.
      Grip strength reference values for Canadians aged 6 to 79: Canadian Health Measures Survey, 2007 to 2013.
      The knee extension torque (Nm) of both sides was measured twice using a handheld dynamometer (mobie MT-100; SAKAI Med Co, Ltd). The participants sat on a chair with 90o knee flexion, and the force sensor was fixed to the distal side of the leg by a belt.
      • Bohannon R.W.
      • Bubela D.J.
      • Magasi S.R.
      • et al.
      Relative reliability of three objective tests of limb muscle strength.
      The knee extension torque was calculated by multiplying the maximal isometric knee extension force by the lower leg length. The handgrip strength and the knee extension torque were adopted as the average value on both sides, divided by weight, and multiplied by 100 to calculate the weight ratio (%). Muscle function was assessed by comfortable walking speed at baseline. The comfortable walking speed was measured by 4-m walking test, which was set at an acceleration road and a deceleration road of 1 m as in a previous report.
      • Maggio M.
      • Ceda G.P.
      • Ticinesi A.
      • et al.
      Instrumental and non-instrumental evaluation of 4-meter walking speed in older individuals.

      Serum Measurements

      The serum biochemical parameters measured were fasting glucose, fasting insulin, HbA1c, triglycerides, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), total cholesterol, highly sensitive C-reacting protein, and albumin. HOMA-IR was calculated using the formula: fasting glucose (mg/dL) × fasting insulin (μU/mL) / 405.

      Follow-up and Study Endpoint

      The endpoint of this study was development of insulin resistance (ie, HOMA-IR ≥ 1.73). The study participants were followed until the annual examination in 2019. Individuals who did not receive health examinations in 2018 or 2019 were defined as censored cases at the time of the last health checkup. The mean of the follow-up was 1.8 years, median 2 years, range 1-2 years.

      Statistical Analysis

      All numerical values are expressed as mean ± SD or medians and range. The Student t test, Mann-Whitney U test, and chi-square method were used to compare baseline characteristics between groups, if appropriate. The participants were divided into 2 groups by HOMA-IR at the end of the follow-up: participants with HOMA-IR ≥1.73 and those with HOMA-IR <1.73 at the end of the follow-up were grouped into an insulin resistance group and a non–insulin resistance group, respectively.
      The hazard ratio (HR) and 95% CI of each muscle mass, muscle strength, and walking speed at baseline for development of insulin resistance were calculated using the Cox proportional hazards model. Age and sex were selected as confounding factors in model 1. In model 2, the baseline HOMA-IR was added to the confounding factors used in model 1, and the baseline waist circumference together with the confounding factors in model 2 was incorporated in model 3. Finally, model 4 added baseline triglycerides and HDL-C to the confounding factors used in model 3. Similar Cox proportional hazard model analyses were conducted for males and females separately. IBM SPSS Statistics version 22 (IBM Inc) was used for statistical analysis. The significance level in all analyses was set at P < .05.

      Results

      The characteristics of the study participants at baseline are shown in Table 1. Mean height, weight, lower limb muscle mass, upper limb muscle mass, appendicular muscle mass, trunk muscle mass, handgrip strength, knee extension torque, fasting glucose, and percentages of participants with smoking were significantly higher in men than in women. Total cholesterol, HDL-C, and LDL-C were significantly higher in women than in men. There was no significant difference in age, body mass index, waist circumference, comfortable walking speed, fasting insulin, HOMA-IR, HbA1c, triglycerides, highly sensitive C-reacting protein, or serum albumin between men and women.
      Table 1Baseline Characteristics in Men and Women
      Total (N = 194)Men (n = 79)Women (n = 115)P
      Age, y75.1 (6.6)75.6 (6.2)74.7 (6.8).39
      Height, cm154.9 (8.9)162.7 (6.0)149.5 (6.0)< .001
      Weight, kg53.5 (9.7)58.9 (9.4)49.8 (8.1)< .001
      BMI22.2 (3.1)22.2 (3.1)22.2 (3.0).96
      Waist circumference, cm82.6 (9.6)81.8 (9.4)83.2 (9.8).33
      Skeletal muscle mass, kg weight ratio (%)
       Lower limb muscle mass25.8 (3.2)27.9 (2.6)24.3 (2.7)< .001
       Upper limb muscle mass6.99 (1.10)8.09 (0.72)6.24 (0.54)< .001
       Appendicular muscle mass32.8 (4.1)36.0 (3.1)30.6 (3.2)< .001
       Trunk muscle mass37.7 (4.5)40.4 (3.7)35.8 (3.9)< .001
      Muscle strength
       Hand grip strength, kg weight ratio (%)54.2 (13.2)61.6 (13.4)49.2 (10.3)< .001
       Knee extension torque, Nm weight ratio (%)127.5 (42.1)137.0 (44.7)121.7 (39.6).020
      Walking speed: comfortable speed, m/s1.18 (0.23)1.14 (0.23)1.20 (0.23).11
      Biochemical
       Fasting glucose, mg/dL94.5 (9.6)96.4 (9.7)93.2 (9.4).020
       Fasting insulin, μU/mL4.13 (1.50)3.89 (1.34)4.30 (1.58).05
       HOMA-IR0.97 (0.36)0.93 (0.34)0.99 (0.38).27
       HbA1c, %5.58 (0.30)5.54 (0.28)5.60 (0.32).19
       Total cholesterol, mg/dL209.7 (36.1)195.3 (32.1)219.6 (35.5)< .001
       HDL-C, mg/dL68.4 (20.1)61.6 (13.9)73.0 (22.3)< .001
       LDL-C, mg/dL121.9 (27.8)114.1 (27.0)127.2 (27.1).001
       Triglycerides, mg/dL84.5 (67.0-109.0)82.0 (67.0-112.0)88.0 (67.0-108.0).67
       Highly sensitive C-reacting protein, mg/dL0.04 (0.02-0.09)0.04 (0.02-0.10)0.04 (0.02-0.08).21
       Albumin, g/dL4.44 (0.22)4.41 (0.24)4.46 (0.21).14
      Smoking: current/never/former, n (%)13 (7) / 137(71) / 44 (23)11 (14) / 27 (34) / 41 (52)2 (2) / 110 (96) / 3 (3)< .001
      BMI, body mass index.
      Data are presented as means (SDs), medians (interquartile ranges) or percentages.
      Differences in baseline characteristics between the non–insulin resistance group and insulin resistance group are presented in Table 2. Mean weight, body mass index, waist circumference, fasting glucose, fasting insulin, HOMA-IR, and triglycerides were significantly higher in the insulin resistance group than in the non–insulin resistance group. On the other hand, lower limb muscle mass, upper limb muscle mass, appendicular muscle mass, trunk muscle mass, handgrip strength, and knee extension torque were significantly higher in the non–insulin resistance group than in the insulin resistance group. There was no significant difference in mean age, percentage of men, height, walking speed, HbA1c, total cholesterol, HDL-C, LDL-C, highly sensitive C-reacting protein, albumin, or percentages of smokers between the 2 groups.
      Table 2Baseline Characteristics of the Participants With or Without Insulin Resistance
      Total (N = 194)Non–Insulin Resistance Group (HOMA-IR < 1.73)

      (n = 132)
      Insulin Resistance Group (HOMA-IR ≥ 1.73)

      (n = 62)
      P
      Age, y75.1 (6.6)75.0 (6.6)75.3 (6.6).78
      Men, n (%)56 (42)23 (37).48
      Height, cm154.9 (8.9)154.7 (8.7)155.0 (9.2).88
      Weight, kg53.5 (9.7)51.9 (9.3)57.0 (9.6).001
      BMI22.2 (3.1)21.5 (2.9)23.6 (2.9)< .001
      Waist circumference, cm82.6 (9.6)80.8 (9.5)86.5 (8.8)< .001
      Skeletal muscle mass, kg weight ratio (%)
       Lower limb muscle mass25.8 (3.2)26.4 (3.1)24.5 (3.1)< .001
       Upper limb muscle mass6.99 (1.10)7.12 (1.13)6.71 (1.00).015
       Appendicular muscle mass32.8 (4.1)33.5 (4.0)31.2 (3.9)< .001
       Trunk muscle mass37.7 (4.5)38.6 (4.3)35.8 (4.2)< .001
      Muscle strength
       Hand grip strength, kg weight ratio (%)54.2 (13.2)55.8 (13.5)50.9 (11.9).015
       Knee extension torque, Nm weight ratio (%)127.5 (42.1)132.9 (39.5)117.2 (45.4).021
      Walking speed: Comfortable speed, m/s1.18 (0.23)1.20 (0.22)1.13 (0.24).07
      Biochemical
       Fasting glucose, mg/dL94.5 (9.6)93.5 (9.6)96.8 (9.4).020
       Fasting insulin, μU/mL4.13 (1.50)3.66 (1.19)5.12 (1.58)< .001
       HOMA-IR0.97 (0.36)0.84 (0.28)1.23 (0.38)< .001
       HbA1c, %5.58 (0.30)5.56 (0.31)5.61 (0.30).33
       Total cholesterol, mg/dL209.7 (36.1)211.2 (34.0)206.5 (40.3).39
       HDL-C, mg/dL68.4 (20.1)69.6 (17.6)65.7 (24.4).20
       LDL-C, mg/dL121.9 (27.8)123.2 (25.8)119.2 (31.6).36
       Triglycerides, mg/dL84.5 (67.0-109.0)80.5 (64.0-103.5)97.5 (74.7-133.2).001
       Highly sensitive C-reacting protein, mg/dL0.04 (0.02-0.09)0.04 (0.02-0.09)0.04 (0.02-0.08).46
       Albumin, g/dL4.44 (0.22)4.43 (0.20)4.46 (0.25).42
      Smoking: current/never/former, n (%)13 (7)/ 137 (71)/ 44 (23)10 (7)/ 91 (69)/ 31 (23)3 (5)/ 46 (74)/ 13 (21).68
      BMI, body mass index.
      Data are presented as means (SDs), medians (interquartile ranges), or percentages.
      Results of Cox proportional hazards model analyses for development of insulin resistance are shown in Table 3. HRs of lower limb muscle mass, upper limb muscle mass, appendicular mass, trunk muscle mass, grip strength, and knee extension torque for development of insulin resistance were significantly low when adjusted for sex and age (model 1): 0.81, 0.50, 0.83, 0.86, 0.90, and 0.49, respectively. HRs of lower limb muscle mass and appendicular muscle mass for development of insulin resistance were significantly low after additional adjustment with baseline HOMA-IR (model 2), with baseline HOMA-IR plus baseline waist circumference (model 3) or with baseline HOMA-IR and waist circumference plus baseline triglycerides and HDL-C (model 4).
      Table 3Hazard Ratio for Development of Insulin Resistance, by Baseline Muscle Mass, Muscle Strength, and Walking Speed
      Model 1Model 2Model 3Model 4
      HR (95% CI)PHR (95% CI)PHR (95% CI)PHR (95% CI)P
      Muscle mass, weight ratio
       Lower limb muscle mass0.81 (0.73-0.90)< .0010.88 (0.79-0.97).0170.88 (0.79-0.98).0310.89 (0.79-0.98).044
       Upper limb muscle mass0.50 (0.32-0.78).0030.66 (0.42-1.05).070.70 (0.42-1.16).160.68 (0.41-1.14).14
       Appendicular muscle mass0.83 (0.76-0.91)< .0010.89 (0.81-0.97).0160.89 (0.81-0.99).0300.90 (0.81-0.99).039
       Trunk muscle mass0.86 (0.81-0.93)< .0010.89 (0.85-1.01).060.91 (0.81-1.03).140.90 (0.79-1.02).11
      Muscle strength, weight ratio
       Hand grip strength0.90 (0.10-0.81).0300.55 (0.04-7.27).651.06 (0.06-18.2).960.92 (0.05-16.0).95
       Knee extension torque0.49 (0.24-0.97).0400.76 (0.36-1.57).450.83 (0.39-1.75).630.71 (0.33-1.52).38
      Walking speed: comfortable speed, m/s0.34 (0.10-1.09).070.39 (0.12-1.27).120.43 (0.31-1.38).150.33 (0.10-1.11).07
      Model 1: adjusted for sex and age at baseline; model 2: model 1 + HOMA-IR at baseline; model 3: model 2 + waist circumference at baseline; model 4: model 3 + triglycerides and HDL-C at baseline.
      Table 4 shows the HR and 95% CI of each sarcopenia-related index for development of insulin resistance in women and men groups of participants. In women, age-adjusted HR (model 1) of lower limb muscle mass, upper limb muscle mass, appendicular mass, trunk muscle mass, and grip strength for development of insulin resistance were significantly low: 0.75, 0.41, 0.79, 0.84, and 0.02, respectively. When additionally adjusted for baseline HOMA-IR (model 2), the statistical significance of lower limb muscle mass, appendicular muscle mass, and trunk muscle mass remained, and the association of lower limb muscle mass and appendicular muscle mass with change in HRs remained significant when baseline waist circumference was added to the adjustment (model 3) or when triglycerides and HDL-C were added to the adjustment (model 4). In contrast to those in women, association between development of insulin resistance and indices of muscle mass, muscle strength, or walking speed was not detected by HRs and 95% CI in men. The point estimates of HRs of the muscle mass indices in men were not much different from those in women, but 95% CIs of the HRs were larger in men.
      Table 4Hazard Ratio for Development of Insulin Resistance, by Baseline Muscle Mass, Muscle Strength, and Walking Speed, by Gender
      Model 1Model 2Model 3Model 4
      HR (95% CI)PHR (95% CI)PHR (95% CI)PHR (95% CI)P
      Muscle mass (weight ratio)
       Lower limb muscle mass
      Women0.75 (0.65-0.87)< .0010.82 (0.71-0.96).0140.80 (0.66-0.97).0250.80 (0.66-0.97).025
      Men0.86 (0.73-1.03).110.93 (0.78-1.10).330.93 (0.78-1.10).410.96 (0.79-1.15).68
       Upper limb muscle mass
      Women0.41 (0.21-0.81).0100.57 (0.28-1.16).120.61 (0.28-1.34).220.59 (0.27-1.33).20
      Men0.56 (0.31-1.02).060.74 (0.41-1.36).340.77 (0.38-1.56).480.79 (0.38-1.62).52
       Appendicular muscle mass
      Women0.79 (0.69-0.89)< .0010.85 (0.74-0.97).0170.83 (0.70-0.98).0200.83 (0.70-0.98).029
      Men0.87 (0.75-1.01).070.93 (0.80-1.07).330.93 (0.80-1.08).380.96 (0.81-1.12).61
       Trunk muscle mass
      Women0.84 (0.77-0.92)< .0010.90 (0.81-0.99).0390.87 (0.75-1.01).060.86 (0.74-1.01).06
      Men0.90 (0.80-1.01).070.97 (0.85-1.10).701.05 (0.82-1.34).691.00 (0.78-1.29).96
      Muscle strength, weight ratio
       Hand grip strength
      Women0.02 (0.01-0.56).0220.17 (0.01-9.12).390.38 (0.01-33.4).670.41 (0.01-36.9).69
      Men0.29 (0.01-6.55).431.45 (0.04-47.2).832.78 (0.05-129.9).601.71 (0.03-85.3).78
       Knee extension torque
      Women0.98 (0.97-1.01).210.99 (0.97-1.01).430.99 (0.97-1.01).410.46 (0.15-1.37).16
      Men1.01 (0.98-1.01).521.01 (0.98-1.02).661.01 (0.98-1.02).691.08 (0.38-3.06).87
      Walking speed: Comfortable speed, m/s
       Women0.36 (0.11-1.18).090.42 (0.13-1.38).150.48 (0.14-1.59).230.39 (0.11-1.37).14
       Men1.03 (0.13-7.71).971.55 (0.19-12.7).671.52 (0.18-12.6).691.35 (0.14-12.7).79
      Women (n = 115); men (n = 79).
      Model 1, adjusted for age at baseline; model 2, model 1 + HOMA-IR at baseline; model 3, model 2 + waist circumference at baseline; model 4, model 3 + triglycerides and HDL-C at baseline.

      Discussion

      The results of this longitudinal study indicate that lower limb muscle mass, but not muscle strength (grip strength, knee extensor torque), or walking speed, was associated with significant increase in risk for development of insulin resistance independently of obesity in older Japanese. The findings suggest that loss of skeletal muscle mass, rather than decrease in muscle strength and muscle function, predisposes older adults to subsequent reduction of insulin sensitivity.
      Skeletal muscle is the target tissue of insulin and responsible for about 70% of glucose clearance.
      • DeFronzo R.A.
      Lilly lecture 1987. The triumvirate: beta-cell, muscle, liver. A collusion responsible for NIDDM.
      Therefore, the loss of muscle mass is expected to result in a reduction in glucose disposal, leading to increase in HOMA-IR. However, metabolic alterations associated with skeletal muscle changes appears to contribute to insulin resistance.
      • Kalyani R.R.
      • Corriere M.
      • Ferrucci L.
      Age-related and disease-related muscle loss: the effect of diabetes, obesity, and other diseases.
      • Cleasby M.E.
      • Jamieson P.M.
      • Atherton P.J.
      Insulin resistance and sarcopenia: mechanistic links between common co-morbidities.
      • Merz K.E.
      • Thurmond D.C.
      Role of skeletal muscle in insulin resistance and glucose uptake.
      Age-related progressive loss of muscle mass is associated with increase in intermuscular fat and intracellular lipids.
      • Morino K.
      • Petersen K.F.
      • Shulman G.I.
      Molecular mechanisms of insulin resistance in humans and their potential links with mitochondrial dysfunction.
      • Delmonico M.J.
      • Harris T.B.
      • Visser M.
      • et al.
      Longitudinal study of muscle strength, quality, and adipose tissue infiltration.
      • Tardif N.
      • Salles J.
      • Guillet C.
      • et al.
      Muscle ectopic fat deposition contributes to anabolic resistance in obese sarcopenic old rats through eIF2α activation.
      Although triglyceride is the main storage form of lipid, the derivatives such as ceramide and diacylglycerol are thought to contribute to insulin resistance by impairing insulin receptor-mediated signaling in skeletal muscle.
      • Morino K.
      • Petersen K.F.
      • Shulman G.I.
      Molecular mechanisms of insulin resistance in humans and their potential links with mitochondrial dysfunction.
      The impaired insulin signaling not only suppresses glucose uptake and its intracellular metabolism but also induces negative protein balance by reduced protein synthesis and increased proteolysis.
      • Kalyani R.R.
      • Corriere M.
      • Ferrucci L.
      Age-related and disease-related muscle loss: the effect of diabetes, obesity, and other diseases.
      • Cleasby M.E.
      • Jamieson P.M.
      • Atherton P.J.
      Insulin resistance and sarcopenia: mechanistic links between common co-morbidities.
      • Merz K.E.
      • Thurmond D.C.
      Role of skeletal muscle in insulin resistance and glucose uptake.
      The impairment of insulin signaling is unlikely to be the only mechanism of insulin resistance in skeletal muscle. Involvement of mitochondrial dysfunction, inflammation, and upregulated myostatin in development of both insulin resistance and sarcopenia has also been proposed.
      • Kalyani R.R.
      • Corriere M.
      • Ferrucci L.
      Age-related and disease-related muscle loss: the effect of diabetes, obesity, and other diseases.
      ,
      • Cleasby M.E.
      • Jamieson P.M.
      • Atherton P.J.
      Insulin resistance and sarcopenia: mechanistic links between common co-morbidities.
      Epidemiologic studies also support a causal role of insulin resistance in progression of sarcopenia.
      • Lee C.G.
      • Boyko E.J.
      • Strotmeyer E.S.
      • et al.
      Association between insulin resistance and lean mass loss and fat mass gain in older men without diabetes mellitus.
      • Park S.W.
      • Goodpaster B.H.
      • Strotmeyer E.S.
      • et al.
      Accelerated loss of skeletal muscle strength in older adults with type 2 diabetes: the health, aging, and body composition study.
      • Park S.W.
      • Goodpaster B.H.
      • Lee J.S.
      • et al.
      Excessive loss of skeletal muscle mass in older adults with type 2 diabetes.
      An earlier longitudinal study that enrolled ambulatory men aged ≥65 years has shown that greater lean mass loss occurred during 4.6-year follow-up in nondiabetic insulin-resistant men than in insulin-sensitive men.
      • Lee C.G.
      • Boyko E.J.
      • Strotmeyer E.S.
      • et al.
      Association between insulin resistance and lean mass loss and fat mass gain in older men without diabetes mellitus.
      Type 2 diabetes in older adults (aged 70-79 years) has been reported to be associated with accelerated loss of leg muscle mass and strength during 3-6 years of follow-up.
      • Park S.W.
      • Goodpaster B.H.
      • Strotmeyer E.S.
      • et al.
      Accelerated loss of skeletal muscle strength in older adults with type 2 diabetes: the health, aging, and body composition study.
      ,
      • Park S.W.
      • Goodpaster B.H.
      • Lee J.S.
      • et al.
      Excessive loss of skeletal muscle mass in older adults with type 2 diabetes.
      Taken together, the findings in basic and population studies suggest that mechanisms of sarcopenia and insulin resistance can form a vicious cycle, leading to simultaneous progression of sarcopenia and glucose intolerance.
      The results of the present study suggest that the lower limb muscle mass exerts more influence on the development of insulin resistance than the upper limb muscle mass and trunk muscle mass. The difference in the contribution in muscle mass regions might be attributable to absolute size of skeletal muscle mass. The lower limb muscle mass was approximately 26% of total muscle mass, and the upper limb muscle mass was approximately 7%. The trunk muscle was approximately 38%, but it included smooth muscles of internal organs since bioelectrical impedance analysis was used for determination of muscle mass. It is notable that the loss of lower limb muscle mass is associated with age-related decreased physical activity and is more remarkable than the loss of upper limb muscle mass.
      • Seino S.
      • Shinkai S.
      • Iijima K.
      • et al.
      Reference values and age differences in body composition of community-dwelling older Japanese men and women: a pooled analysis of four cohort studies.
      ,
      • Tanimoto Y.
      • Watanabe M.
      • Kono R.
      • et al.
      Aging changes in muscle mass of Japanese.
      Therefore, lower limb skeletal muscle mass is likely to be an important predictor of insulin resistance and also a target for preventive strategies in older adults.
      Although results of the analysis of total participant data showed statistical significance in the decreased HR of lower limb muscle mass for development of insulin resistance (Table 3), significant decrease in the HR was observed in women but not in men in subgroup analyses (Table 4). Women, regardless of age, are known to have less muscle mass than men,
      • Tanimoto Y.
      • Watanabe M.
      • Kono R.
      • et al.
      Aging changes in muscle mass of Japanese.
      ,
      • Gallagher D.
      • Visser M.
      • De Meersman R.E.
      • et al.
      Appendicular skeletal muscle mass: effects of age, gender, and ethnicity.
      and in this study, women had less muscle mass than men, corrected for body weight to account for physique at baseline. Thus, the loss of muscle mass in women may sensitively reflect the development of insulin resistance. However, it is difficult to conclude that there is a sex difference in a role of lower limb muscle mass as a risk factor of insulin resistance. Point estimates of HR for lower limb muscle mass in men showed a similar trend to women, and the sample size of male participants was relatively small, leaving a possibility of type 2 error in the statistical analysis of HR. This issue needs to be addressed in the future by enrollment of a large number of cases.
      Scott et al
      • Scott D.
      • Cumming R.
      • Naganathan V.
      • et al.
      Associations of sarcopenic obesity with the metabolic syndrome and insulin resistance over five years in older men: The Concord Health and Ageing in Men Project.
      reported that sarcopenic obesity and sarcopenic nonobesity did not appear to confer greater risk for incident metabolic syndrome or insulin resistance than obesity alone during a 5-year follow-up in community-dwelling older men. They argue that avoiding age-related increases in body fat is more important than preventing the sarcopenia in order to reduce the incidence of metabolic syndrome and insulin resistance in older adults. The findings by Scott et al are apparently contradictory to the results of the present study. However, it is difficult to sort out reasons for the apparent difference because there were major differences in study methods. Scott et al determined HOMA-IR only at 5 years of follow-up, and its change during the follow-up was unclear. They defined sarcopenia as low appendicular muscle mass, which is combined with low hand grip strength and/or low gait speed according to the European Working Group on Sarcopenia (EWGSOP) criteria. Although the participants in both of the studies were community-dwelling older adults, the inclusion criteria of age (≥70 years), sex (only men), and ethnicity of the study participants (Caucasians), and exclusion criteria (only living in a residential aged care facility) in a study by Scott et al were different from those in the present study. In addition, a proportion of participants with obesity (42.6%) was different between the 2 studies. Nevertheless, roles of muscle mass loss in each body region, functional capacity of the skeletal muscle, increase in adipose tissue, and preexisting change in insulin sensitivity in development of insulin resistance and diabetes need to be further investigated by use of standardized indexes of muscle mass and functions.
      There are limitations in this study. First, the sample size of the present study was relatively small (ie, 194), because this cohort included those who voluntarily underwent a specific health checkup in Sobetsu Town. Nevertheless, we were still able to detect significant and relevant differences between groups (ie, non–insulin resistance or insulin resistance groups). In the future, this issue should be addressed by enrolling more participants to allow for improved statistical power, additional analysis of more confounding factors, and calculation of cutoff values for lower limb muscle mass. Second, because the follow-up period for the target was as short as 2 years, it possibly led to underestimation of the impact of sarcopenia on the development of insulin resistance. Third, multivariate analysis of model 4 adjusted for the inclusion of HDL-C may partially explain the confounding factor physical activity,
      • Monda K.L.
      • Ballantyne C.M.
      • North K.E.
      Longitudinal impact of physical activity on lipid profiles in middle-aged adults: the Atherosclerosis Risk in Communities Study.
      but does not reflect actual measurements of physical activity. Fourth, we used HOMA-IR, an index that has been used in numerous epidemiologic studies. HOMA-IR is reasonably correlated with insulin sensitivity determined with hyperinsulinemic euglycemic glucose clamp (r = 0.51-0.65),
      • Gutch M.
      • Kumar S.
      • Razi S.M.
      • et al.
      Assessment of insulin sensitivity/resistance.
      but we cannot exclude the possibility of some underestimation of insulin resistance of the skeletal muscle.
      • Furugen M.
      • Saitoh S.
      • Ohnishi H.
      • et al.
      Matsuda-DeFronzo insulin sensitivity index is a better predictor than HOMA-IR of hypertension in Japanese: the Tanno-Sobetsu study.
      Finally, because this study enrolled only subjects who voluntarily received health checkups, we cannot exclude self-selection bias. Although there are these limitations, the research findings would contribute to the prevention of diabetes mellitus and arteriosclerotic disease in older adults, promoting the identification of risk for individuals on the development of insulin resistance and enhancing the development of efficient preventive exercise programs.

      Conclusions and Implications

      The results of the present longitudinal study suggest that loss of lower limb muscle mass is a risk factor for development of insulin resistance independent of obesity in nondiabetic older Japanese adults although such a role of risk factor is not shared by upper limb muscle mass, trunk muscle mass, muscle strength, or muscle function. Whether strategies to preserve lower limb muscle mass can prevent diabetes in older adults remains to be further investigated. This association needs to be confirmed in further large-scale longitudinal studies.

      Acknowledgments

      This study was supported by a grant from the Japanese Society for Promotion of Science (grant no. JP20K10534). We thank the public health nurses and other staff in the Sobetsu-Town for their cooperation in collecting and managing the study data, and we acknowledge the cohort study participants.

      Supplementary Data

      Figure thumbnail fx1
      Supplementary Fig. 1A flowchart of participants in the study cohort.

      References

      1. Cabinet Office of Japan. The aging society white paper in 2021. Accessed December 24, 2021. ​https://www8.cao.go.jp/kourei/whitepaper/w-2021/zenbun/pdf/1s1s_01.pdf

        • Wild S.
        • Roglic G.
        • Green A.
        • et al.
        Global prevalence of diabetes: estimates for the year 2000 and projections for 2030.
        Diabetes Care. 2004; 27: 1047-1053
        • Charvat H.
        • Goto A.
        • Goto M.
        • et al.
        Impact of population aging on trends in diabetes prevalence: a meta-regression analysis of 160,000 Japanese adults.
        J Diabetes Investig. 2015; 6: 533-542
        • Rockwood K.
        • Awalt E.
        • MacKnight C.
        • et al.
        Incidence and outcomes of diabetes mellitus in elderly people: report from the Canadian Study of Health and Aging.
        CMAJ. 2000; 162: 769-772
        • Huang E.S.
        • Laiteerapong N.
        • Liu J.Y.
        • et al.
        Rates of complications and mortality in older patients with diabetes mellitus: the diabetes and aging study.
        JAMA Intern Med. 2014; 174: 251-258
        • Fujioka S.
        • Matsuzawa Y.
        • Tokunaga K.
        • et al.
        Contribution of intra-abdominal fat accumulation to the impairment of glucose and lipid metabolism in human obesity.
        Metabolism. 1987; 36: 54-59
        • Fujii M.
        • Ohnishi H.
        • Saitho S.
        • et al.
        Comparison of the effect of abdominal obesity on new onset of type 2 diabetes in a general Japanese elderly population with that in a non-elderly population-The Tanno and Sobetsu study.
        Nippon Ronen Igakkai Zasshi. 2011; 48: 71-77
        • Yoshimura Y.
        • Kamada C.
        • Takahashi K.
        • et al.
        Relations of nutritional intake to age, sex and body mass index in Japanese elderly patients with type 2 diabetes: the Japanese Elderly Diabetes Intervention Trial.
        Geriatr Gerontol Int. 2012; 12: 29-40
        • Chen L.K.
        • Woo J.
        • Assantachai P.
        • et al.
        Asian working group for sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment.
        J Am Med Dir Assoc. 2020; 21: 300-307
        • Marzetti E.
        • Calvani R.
        • Tosato M.
        • et al.
        Sarcopenia: an overview.
        Aging Clin Exp Res. 2017; 29: 11-17
        • Srikanthan P.
        • Hevener A.L.
        • Karlamangla A.S.
        Sarcopenia exacerbates obesity-associated insulin resistance and dysglycemia: findings from the National Health and Nutrition Examination Survey III.
        PLoS One. 2010; 5: e10805
        • Kim T.N.
        • Park M.S.
        • Lim K.I.
        • et al.
        Relationships between sarcopenic obesity and insulin resistance, inflammation, and vitamin D status: the Korean Sarcopenic Obesity Study.
        Clin Endocrinol (Oxf). 2013; 78: 525-532
        • Seko T.
        • Akasaka H.
        • Koyama M.
        • et al.
        Lower limb muscle mass is associated with insulin resistance more than lower limb muscle strength in non-diabetic older adults.
        Geriatr Gerontol Int. 2019; 12: 1254-1259
        • Scott D.
        • Cumming R.
        • Naganathan V.
        • et al.
        Associations of sarcopenic obesity with the metabolic syndrome and insulin resistance over five years in older men: The Concord Health and Ageing in Men Project.
        Exp Gerontol. 2018; 108: 99-105
        • Matsumoto T.
        • Ohnishi H.
        • Sato T.
        • et al.
        Insulin resistance is associated with longitudinal changes of cardiac repolarization heterogeneity in apparently healthy subjects.
        Cardiol Ther. 2019; 8: 239-251
        • Oimatsu H.
        • Saitoh S.
        • Ura N.
        • et al.
        A practical index for evaluation of insulin resistance.
        J Japan Diab Soc. 2000; 43: 205-213
        • Ohnishi H.
        • Saitoh S.
        • Ura N.
        • et al.
        Relationship between insulin resistance and accumulation of coronary risk factors.
        Diabetes Obes Metab. 2002; 4: 388-393
        • Lim S.
        • Kim J.H.
        • Yoon J.W.
        • et al.
        Sarcopenic obesity: prevalence and association with metabolic syndrome in the Korean Longitudinal Study on Health and Aging (KLoSHA).
        Diabetes Care. 2010; 33: 1652-1654
        • Kim M.
        • Shinkai S.
        • Murayama H.
        • et al.
        Comparison of segmental multifrequency bioelectrical impedance analysis with dual-energy X-ray absorptiometry for the assessment of body composition in a community-dwelling older population.
        Geriatr Gerontol Int. 2015; 15: 1013-1022
        • Wong S.L.
        Grip strength reference values for Canadians aged 6 to 79: Canadian Health Measures Survey, 2007 to 2013.
        Health Rep. 2016; 27: 3-10
        • Bohannon R.W.
        • Bubela D.J.
        • Magasi S.R.
        • et al.
        Relative reliability of three objective tests of limb muscle strength.
        Isokinet Exerc Sci. 2011; 19
        • Maggio M.
        • Ceda G.P.
        • Ticinesi A.
        • et al.
        Instrumental and non-instrumental evaluation of 4-meter walking speed in older individuals.
        PLoS One. 2016; 11: e0153583
        • DeFronzo R.A.
        Lilly lecture 1987. The triumvirate: beta-cell, muscle, liver. A collusion responsible for NIDDM.
        Diabetes. 1988; 37: 667-687
        • Kalyani R.R.
        • Corriere M.
        • Ferrucci L.
        Age-related and disease-related muscle loss: the effect of diabetes, obesity, and other diseases.
        Lancet Diabetes Endocrinol. 2014; 2: 819-829
        • Cleasby M.E.
        • Jamieson P.M.
        • Atherton P.J.
        Insulin resistance and sarcopenia: mechanistic links between common co-morbidities.
        J Endocrinol. 2016; 229: R67-81
        • Merz K.E.
        • Thurmond D.C.
        Role of skeletal muscle in insulin resistance and glucose uptake.
        Compr Physiol. 2020; 10: 785-809
        • Morino K.
        • Petersen K.F.
        • Shulman G.I.
        Molecular mechanisms of insulin resistance in humans and their potential links with mitochondrial dysfunction.
        Diabetes. 2006; 55: S9-s15
        • Delmonico M.J.
        • Harris T.B.
        • Visser M.
        • et al.
        Longitudinal study of muscle strength, quality, and adipose tissue infiltration.
        Am J Clin Nutr. 2009; 90: 1579-1585
        • Tardif N.
        • Salles J.
        • Guillet C.
        • et al.
        Muscle ectopic fat deposition contributes to anabolic resistance in obese sarcopenic old rats through eIF2α activation.
        Aging Cell. 2014; 13: 1001-1011
        • Lee C.G.
        • Boyko E.J.
        • Strotmeyer E.S.
        • et al.
        Association between insulin resistance and lean mass loss and fat mass gain in older men without diabetes mellitus.
        J Am Geriatr Soc. 2011; 59: 1217-1224
        • Park S.W.
        • Goodpaster B.H.
        • Strotmeyer E.S.
        • et al.
        Accelerated loss of skeletal muscle strength in older adults with type 2 diabetes: the health, aging, and body composition study.
        Diabetes Care. 2007; 30: 1507-1512
        • Park S.W.
        • Goodpaster B.H.
        • Lee J.S.
        • et al.
        Excessive loss of skeletal muscle mass in older adults with type 2 diabetes.
        Diabetes Care. 2009; 32: 1993-1997
        • Seino S.
        • Shinkai S.
        • Iijima K.
        • et al.
        Reference values and age differences in body composition of community-dwelling older Japanese men and women: a pooled analysis of four cohort studies.
        PLoS One. 2015; 10: e0131975
        • Tanimoto Y.
        • Watanabe M.
        • Kono R.
        • et al.
        Aging changes in muscle mass of Japanese.
        Nippon Ronen Igakkai Zasshi. 2010; 47: 52-57
        • Gallagher D.
        • Visser M.
        • De Meersman R.E.
        • et al.
        Appendicular skeletal muscle mass: effects of age, gender, and ethnicity.
        J Appl Physiol (1985). 1997; 83: 229-239
        • Monda K.L.
        • Ballantyne C.M.
        • North K.E.
        Longitudinal impact of physical activity on lipid profiles in middle-aged adults: the Atherosclerosis Risk in Communities Study.
        J Lipid Res. 2009; 50: 1685-1691
        • Gutch M.
        • Kumar S.
        • Razi S.M.
        • et al.
        Assessment of insulin sensitivity/resistance.
        Indian J Endocrinol Metab. 2015; 19: 160-164
        • Furugen M.
        • Saitoh S.
        • Ohnishi H.
        • et al.
        Matsuda-DeFronzo insulin sensitivity index is a better predictor than HOMA-IR of hypertension in Japanese: the Tanno-Sobetsu study.
        J Hum Hypertens. 2012; 26: 325-333