Advertisement

Standing is associated with insulin sensitivity in adults with metabolic syndrome

Open AccessPublished:August 14, 2021DOI:https://doi.org/10.1016/j.jsams.2021.08.009

      Abstract

      Objectives

      To determine how components of accelerometer-measured sedentary behavior (SB) and physical activity (PA), and fitness are associated with insulin sensitivity in adults with metabolic syndrome.

      Design

      Cross-sectional.

      Methods

      Target population was middle-aged (40–65 years) sedentary adults with metabolic syndrome. SB, breaks in SB, standing, and PA were measured for four weeks with hip-worn accelerometers. VO2max (ml/min/kg) was measured with maximal cycle ergometry. Insulin sensitivity was determined by hyperinsulinaemic-euglycaemic clamp (M-value) and fasting blood sampling (HOMA-IR, insulin). Multivariable regression was used for analyses.

      Results

      Sixty-four participants (37 women; 58.3 [SD 6.8] years) were included. Participants spent 10.0 (1.0) h sedentary, 1.8 (0.6) h standing, and 2.7 (0.6) h in PA and took 5149 (1825) steps and 29 (8) breaks daily. In sex-, age- and accelerometer wear time-adjusted model SB, standing, steps and VO2max were associated with M-value (β = −0.384; β = 0.400; β = 0.350; β = 0.609, respectively), HOMA-IR (β = 0.420; β = −0.548; β = −0.252; β = −0.449), and insulin (β = 0.433; β = −0.541; β = −0.252; β = −0.453); all p-values < 0.05. Breaks associated only with M-value (β = 0.277). When further adjusted for body fat %, only standing remained significantly associated with HOMA-IR (β = −0.381) and insulin (β = −0.366); significance was maintained even when further adjusted for SB, PA and fitness. Light and moderate-to-vigorous PA were not associated with insulin sensitivity.

      Conclusions

      Standing is associated with insulin sensitivity markers. The association with HOMA-IR and insulin is independent of adiposity, PA, SB and fitness. Further studies are warranted, but these findings encourage replacing sitting with standing for potential improvements in insulin sensitivity in adults at increased type 2 diabetes risk.

      Abbreviations:

      APE (angle for posture estimation), BP (blood pressure), FFM (fat free mass), HOMA-IR (homeostatic model assessment of insulin resistance), LPA (light physical activity), MAD (mean amplitude deviation), MET (metabolic equivalent), MetS (metabolic syndrome), MVPA (moderate-to-vigorous physical activity), NEFA (non-esterified fatty acids), PA (physical activity), SB (sedentary behavior), T2D (type 2 diabetes), WC (waist circumference)

      Keywords

      Practical implications

      • Standing is associated with insulin sensitivity in inactive sedentary adults at increased risk of developing type 2 diabetes.
      • Adiposity appears to partially mediate the relationship between sedentary behavior, fitness and insulin sensitivity, further highlighting the importance of healthy body composition for metabolic health.
      • These findings can provide additional insights to aid in development of interventions, policies and guidelines for type 2 diabetes prevention in inactive high-risk populations.

      1. Introduction

      Although health benefits of regular physical activity (PA) are well-known, physical inactivity is common globally.
      • Guthold R.
      • Stevens G.A.
      • Riley L.M.
      • et al.
      Worldwide trends in insufficient physical activity from 2001 to 2016: a pooled analysis of 358 population-based surveys with 1.9 million participants.
      Sedentary behavior (SB) has been identified as an important health risk, particularly for type 2 diabetes (T2D).
      • Ekelund U.
      • Tarp J.
      • Fagerland M.W.
      • et al.
      Joint associations of accelerometer measured physical activity and sedentary time with all-cause mortality: a harmonised meta-analysis in more than 44 000 middle-aged and older individuals.
      ,
      • Biswas A.
      • Oh P.I.
      • Faulkner G.E.
      • et al.
      Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults: a systematic review and meta-analysis.
      Low cardiorespiratory fitness is also a risk factor for insulin resistance
      • Huth C.
      • Pigeon É.
      • Riou M.
      • et al.
      Fitness, adiposopathy, and adiposity are independent predictors of insulin sensitivity in middle-aged men without diabetes.
      (predictor of T2D), and low fitness and high SB together increase the odds of metabolic syndrome (MetS) and T2D.
      • van der Velde J.H.P.M.
      • Schaper N.C.
      • Stehouwer C.D.A.
      • et al.
      Which is more important for cardiometabolic health: sedentary time, higher intensity physical activity or cardiorespiratory fitness? The Maastricht study.
      Accelerometer-measured PA has been consistently reported as major determinant of insulin sensitivity,
      • Balkau B.
      • Mhamdi L.
      • Oppert J.M.
      • et al.
      Physical activity and insulin sensitivity: the RISC study.
      ,
      • Ekelund U.
      • Griffin S.J.
      • Wareham N.J.
      Physical activity and metabolic risk in individuals with a family history of type 2 diabetes.
      whereas the role of SB and patterns of SB (i.e., breaks in sitting) is more unclear.
      • Healy G.N.
      • Matthews C.E.
      • Dunstan D.W.
      • et al.
      Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003-06.
      • Ekelund U.
      • Brage S.
      • Griffin S.J.
      • et al.
      Objectively measured moderate- and vigorous-intensity physical activity but not sedentary time predicts insulin resistance in high-risk individuals.
      • Cooper A.R.
      • Sebire S.
      • Montgomery A.A.
      • et al.
      Sedentary time, breaks in sedentary time and metabolic variables in people with newly diagnosed type 2 diabetes.
      Furthermore, studies investigating associations between standing and insulin sensitivity are lacking. Only few studies have accounted for fitness, and most have used only surrogate markers of insulin resistance (e.g., homeostatic model of insulin resistance [HOMA-IR], fasting insulin).
      To our knowledge, only one study has investigated associations of accelerometer-measured SB and PA with insulin sensitivity assessed with the gold standard hyperinsulinaemic-euglycaemic clamp,
      • Balkau B.
      • Mhamdi L.
      • Oppert J.M.
      • et al.
      Physical activity and insulin sensitivity: the RISC study.
      whereas associations between breaks and clamp-measured insulin sensitivity have not been studied. Additionally, associations of SB, PA and fitness with insulin sensitivity have not been studied in sedentary adults with MetS. As SB is associated with increased cardiometabolic risk and MetS strongly predicts T2D, understanding the relationship between SB and PA habits and insulin sensitivity in this population is important from T2D prevention perspective. Therefore, we investigated the associations of accelerometer-measured SB, breaks, standing, and PA, and fitness with insulin sensitivity – measured by hyperinsulinaemic-euglycaemic clamp and surrogate measures – in inactive sedentary adults with MetS.

      2. Methods

      Data consists of baseline data of an intervention (Clinicaltrials.gov NCT03101228) collected 2017–2019 at Turku PET Centre (Turku, Finland). A four-week accelerometer-measurement was carried out during intervention screening phase, after which hyperinsulinaemic-euglycaemic clamp, maximal cycle ergometry, fasting blood sampling and measurements of blood pressure (BP) and body composition were performed to determine baseline values. All participants gave written informed consent. The study was approved by the Ethics Committee of the Hospital District of Southwest Finland (16/1810/2017), and good clinical practice and Declaration of Helsinki were followed.
      Participants were recruited from the local community by newspaper advertisements. The inclusion criteria were age 40–65 years; physical inactivity (< 120 min/week of self-reported moderate-to-vigorous PA [MVPA]); accelerometer-measured sitting time ≥ 10 h/day or 60 % of accelerometer wear time/day; BMI 25–40 kg/m2; BP < 160/100 mm Hg; fasting glucose < 7.0 mmol/l, and fulfilment of MetS criteria including three of the following: waist circumference (WC) ≥ 94 cm for men/≥ 80 cm for women, triglycerides ≥ 1.7 mmol/l, HDL < 1.0 mmol/l for men/< 1.3 mmol/l for women, systolic BP ≥ 130 and/or diastolic BP ≥ 85 mm Hg, or fasting glucose > 5.6 mmol/l. The exclusion criteria were previous cardiac event; diagnosed diabetes; abundant alcohol consumption (according to national guidelines); use of narcotics, cigarette or snuff tobacco; depressive or bipolar disorder; and any chronic disease/condition that could endanger participant safety or study procedures, or interfere with interpretation of results.
      SB, breaks in SB, standing, and PA were assessed by a tri-axial accelerometer (UKK AM30, UKK Institute, Tampere, Finland) with digital acceleration sensor (ADXL345, Analog Devices, Norwood, MA, USA) attached to a hip-mounted belt, as reported previously.
      • Sjöros T.
      • Vähä-Ypyä H.
      • Laine S.
      • et al.
      Both sedentary time and physical activity are associated with cardiometabolic health in overweight adults in a 1 month accelerometer measurement.
      Accelerometers were worn during waking hours for four consecutive weeks (except during water-based activities). Wear time of 10–19 h/day and ≥ four days of measurement was considered valid. Data was analyzed in six-second epochs by validated mean amplitude deviation (MAD) method, and MAD values were converted to metabolic equivalents (METs).
      • Vähä-Ypyä H.
      • Vasankari T.
      • Husu P.
      • et al.
      Validation of cut-points for evaluating the intensity of physical activity with accelerometry-based mean amplitude deviation (MAD).
      SB and standing were defined as ≤ 1.5 METs, light-intensity physical activity (LPA) as 1.5–2.9 METs, and MVPA as ≥ 3.0 METs. Moderate and vigorous activity are combined as MVPA, as the time spent in vigorous activity was negligible. In order to differentiate between SB and standing, body posture was assessed by validated angle for posture estimation (APE) method, which identifies postures with 90 % accuracy in free-living conditions.
      • Vähä-Ypyä H.
      • Husu P.
      • Suni J.
      • et al.
      Reliable recognition of lying, sitting, and standing with a hip-worn accelerometer.
      SB was defined as APE ≥ 11.6° and standing as < 11.6°. Breaks were determined as SB periods with one-minute exponential moving average < 1.5 METs ending in vertical acceleration and subsequent standing posture or movement.
      • Vähä-Ypyä H.
      • Husu P.
      • Suni J.
      • et al.
      Reliable recognition of lying, sitting, and standing with a hip-worn accelerometer.
      Number of steps/day was determined by an algorithm splitting acceleration into vertical and horizontal components. The algorithm requires walking speed of ~3 km/h to detect every step.
      • Vähä-Ypyä H.
      • Husu P.
      • Suni J.
      • et al.
      Reliable recognition of lying, sitting, and standing with a hip-worn accelerometer.
      Periods with acceleration of each axis within 187.5 mg range for ≥ 30 min were considered non-wear time.
      Whole-body insulin-stimulated glucose uptake (M-value) was measured with hyperinsulinaemic-euglycaemic clamp after fasting overnight. Insulin (Actrapid, 100 U/ml, Novo Nordisk, Bagsvaerd, Denmark) was infused at 40 mU/m2 body surface area/min rate during the first 4 min, at 20 mU/m2/min during minutes 4–7, and thereafter at 10 mU/m2/min. Four minutes after starting insulin infusion, 20 % glucose infusion was started. The rate was adjusted according to plasma glucose concentration measured every 5–10 min. M-value (μmol/kg/min) was calculated in 20-min intervals from steady-state glucose values. Fasting insulin and HOMA-IR were determined as surrogate markers of insulin sensitivity as described below.
      Venous blood samples were drawn after fasting ≥ 10 h and analyzed at the Turku University Hospital Laboratory. Plasma insulin was measured by electrochemiluminescence immunoassay (Cobas 8000 e801), plasma glucose by enzymatic reference method with hexokinase GLUC3, and cholesterol (total, LDL, HDL), triglycerides and non-esterified fatty acids (NEFA) by enzymatic colorimetric tests (Cobas 8000 c702). HbA1c was measured by turbidimetric inhibition immunoassay (Cobas 6000 c501); all analyzers by Roche Diagnostics GmbH, Mannheim, Germany. HOMA-IR was calculated with formula (insulin x glucose/22.5).
      Fitness was assessed by maximal cycle ergometry (eBike EL Ergometer + CASE v6.7, GE Medical Systems Information Technologies, Inc. Milwaukee, WI, USA) with direct respiratory gas measurements (Vyntus CPX, CareFusion, Yorba Linda, CA, USA). Cycling pace was ~65 rpm, and intensity started at 25 W. Every 3 min the intensity increased by 25 W until volitional exhaustion, and perceived exertion on Borg scale and BP were measured. VO2max was determined if ≥ 1 criterion was met: respiratory exchange ratio > 1.0, plateau in VO2, or heart rate within ±10 bpm of age-predicted maximum. VO2max was defined as the highest one-minute average in ml/min/kg. VO2max per fat free mass (FFM) (ml/min/kgFFM), and maximal load (Wmax) were also determined. Fitness data is available for 58 participants, as the test was stopped for other reasons (e.g., knee pain) for five participants before reaching volitional exhaustion and the abovementioned criteria, and results of one participant were lost due to technical difficulties.
      Weight, body fat % and FFM were estimated with air displacement plethysmography (Bod Pod, COSMED USA, Inc., Concord, CA, USA) after fasting ≥ 4 h. Height was measured with stadiometer. BMI was calculated from weight and height (kg/m2). WC was measured midway between the iliac crest and the lowest rib. BP was measured by a digital monitor (Apteq AE701f, Rossmax International LtD, Taipei, Taiwan) after ≥ 10 min of sitting.
      Descriptive statistics (means [SD]) were calculated, and differences between sexes were tested with unpaired t-test. Normal distribution was evaluated visually and by Shapiro-Wilk test, and log10 transformation was performed as needed. Associations were examined with Pearson partial correlation analysis and multivariable linear regression. Regression model always included one insulin sensitivity marker as dependent, and one activity/fitness parameter as independent variable. Model 1 was adjusted for sex, age, and accelerometer wear time, and Model 2 additionally for body fat %. To further analyze independence of associations between standing and insulin sensitivity, SB, total PA and VO2max (ml/min/kg) were included in Model 3. Variance inflation factors < 5 were considered not to have multicollinearity issues. Missing data was handled by pairwise deletion. Statistical significance was set at p < 0.05 (two-tailed). For illustrative purposes data are stratified by tertiles of standing time and log10 transformed insulin sensitivity markers are presented in original scale. Analyses were performed with IBM SPSS Statistics 27.0 (IBM Corp., Armonk, NY, USA) and JMP Pro 15.1.0 (SAS Institute Inc., Cary, NC, USA).

      3. Results

      Mean age was 58.3 (SD 6.8) years and BMI 31.6 (4.3) kg/m2. Accelerometers were worn for 26 (4) days, and 14.54 (0.97) h/day. Participants spent 10.0 (1.0) h sedentary, 1.8 (0.6) h standing, and 2.7 (0.6) h in PA, and took 5149 (1825) steps and 29 (8) breaks daily (Table A.1 in supplementary file online).
      Correlation analyses between adiposity, lipid profile, and insulin sensitivity showed that weight, BMI, body fat % and WC correlated with M-value (r = −0.54; −0.59; −0.61; and − 0.67, respectively), HOMA-IR (r = 0.59; 0.62; 0.61; 0.59), and fasting insulin (r = 0.58; 0.61; 0.62; 0.58). HDL also correlated with M-value (r = 0.26), HOMA-IR (r = −0.27) and insulin (r = −0.30), while triglycerides correlated with M-value (r = −0.31) and insulin (r = 0.26). NEFA correlated only with M-value (r = −0.38); all p-values < 0.05 (Table A.2).
      In correlation analyses between activity/fitness parameters and adiposity, SB, MVPA, steps and VO2max (ml/min/kg) correlated with weight (r = 0.40; −0.42; −0.51; and − 0.61, respectively), BMI (r = 0.32; −0.36; −0.45; −0.64), body fat % (r = 0.33; −0.35; −0.43; −0.64), and WC (r = 0.35; −0.42; −0.49; −0.61). Breaks correlated with weight (r = −0.30), BMI (r = −0.34) and WC (r = −0.34), while standing correlated with body fat % (r = −0.31). Total PA correlated with weight (r = −0.37) and WC (r = −0.29); all p-values < 0.05 (Table A.3).
      Multivariable regression model was used to examine associations between activity/fitness parameters and insulin sensitivity. In sex-, age- and wear time-adjusted model SB was detrimentally, and standing, steps, and VO2max (ml/min/kg) beneficially associated with M-value, HOMA-IR and insulin (Table 1). Breaks associated only with M-value. Neither LPA, MVPA nor total PA was associated with any insulin sensitivity markers. All associations between fitness and insulin sensitivity turned non-significant when fitness was expressed as VO2max (ml/min/kgFFM) (Table 1).
      Table 1Associations between activity/fitness parameters and insulin sensitivity estimated by multivariable linear regression model.
      M-value (μmol/kg/min)HOMA-IR
      Data available for 63 participants.
      Fasting insulin
      Data available for 63 participants.
      (pmol/l)
      Model 1Model 2Model 1Model 2Model 1Model 2
      βpβpβpβpβpβp
      Sedentary, h/day− 0.384

      (−0.671, −0.099)
      0.01− 0.171

      (−0.424, 0.082)
      0.180.420

      (0.141, 0.700)
      0.0040.214

      (−0.034, 0.463)
      0.090.433

      (0.152, 0.715)
      0.0030.221

      (−0.027, 0.47)
      0.08
      Standing, h/day0.400

      (0.134, 0.667)
      0.0040.212

      (−0.022, 0.447)
      0.08− 0.548

      (−0.792, −0.304)
      <0.001− 0.381

      (−0.598, −0.165)
      <0.001− 0.541

      (−0.789, −0.292)
      <0.001− 0.366

      (−0.585, −0.148)
      0.001
      LPA, h/day0.060

      (−0.215, 0.335)
      0.670.052

      (−0.168, 0.272)
      0.64− 0.001

      (−0.274, 0.271)
      0.990.010

      (−0.209, 0.228)
      0.93− 0.031

      (−0.307, 0.244)
      0.82− 0.020

      (−0.238, 0.199)
      0.86
      MVPA, h/day0.219

      (−0.043, 0.482)
      0.10− 0.002

      (−0.232, 0.228)
      0.99− 0.142

      (−0.405, 0.120)
      0.260.084

      (−0.142, 0.310)
      0.46− 0.147

      (−0.412, 0.118)
      0.270.086

      (−0.140, 0.312)
      0.45
      Total PA, h/day0.157

      (−0.110, 0.425)
      0.240.036

      (−0.184, 0.257)
      0.74− 0.077

      (−0.342, 0.189)
      0.570.048

      (−0.170, 0.265)
      0.66− 0.099

      (−0.367, 0.169)
      0.460.028

      (−0.19, 0.245)
      0.80
      Breaks in sedentary time/day0.277

      (0.015, 0.539)
      0.040.125

      (−0.097, 0.348)
      0.26− 0.206

      (−0.471, 0.056)
      0.12− 0.048

      (−0.271, 0.175)
      0.67− 0.217

      (−0.484, 0.049)
      0.11− 0.054

      (−0.277, 0.170)
      0.63
      Steps/day0.350

      (0.105, 0.594)
      0.010.107

      (−0.121, 0.337)
      0.35− 0.252

      (−0.501, −0.003)
      0.0470.005

      (−0.224, 0.234)
      0.96− 0.252

      (−0.504, −0.001)
      0.0490.014

      (−0.214, 0.243)
      0.90
      VO2max, ml/min/kg
      Data available for 58 participants.
      0.609

      (0.351, 0.866)
      <0.0010.281

      (−0.015, 0.577)
      0.06− 0.449

      (−0.720, −0.178)
      0.002− 0.050

      (−0.349, 0.249)
      0.74− 0.453

      (−0.727, −0.179)
      0.002− 0.031

      (−0.327, 0.265)
      0.83
      VO2max, ml/min/kgFFM
      Data available for 58 participants.
      0.238

      (−0.047, 0.524)
      0.100.217

      (−0.006, 0.439)
      0.06− 0.055

      (−0.341, 0.230)
      0.70− 0.033

      (−0.259, 0.192)
      0.77− 0.051

      (−0.339, 0.237)
      0.72− 0.028

      (−0.252, 0.195)
      0.80
      Wmax
      Data available for 58 participants.
      0.338

      (−0.012, 0.689)
      0.060.099

      (−0.199, 0.398)
      0.51− 0.042

      (−0.396, 0.312)
      0.810.218

      (−0.068, 0.504)
      0.13− 0.048

      (−0.406, 0.310)
      0.790.222

      (−0.061, 0.505)
      0.12
      HOMA-IR = homeostatic model of insulin resistance; LPA = light-intensity physical activity; MVPA = moderate-to-vigorous physical activity; PA = physical activity; FFM = fat free mass.
      Values expressed as standardized β coefficients (95% CI); all values on log scale. Bold p-values indicate statistical significance (p < 0.05).
      Model 1 adjusted for sex, age, and accelerometer wear time.
      Model 2 adjusted for sex, age, body fat %, and accelerometer wear time.
      a Data available for 63 participants.
      b Data available for 58 participants.
      Adding body fat % to the model turned all associations with M-value non-significant (Table 1). All associations with HOMA-IR and insulin also turned non-significant, except for standing. The association between standing and M-value was near-significant (p = 0.08) also, and remained significant when adjusted for BMI instead of body fat % (data not shown). The associations of standing with HOMA-IR and insulin remained significant when SB, total PA and VO2max (ml/min/kg) were entered into the model one at a time along with sex, age, wear time and body fat %, as well as when SB, PA and fitness were all included in the same model (Table 2). When adjusted for sex, age, body fat %, SB, PA, and fitness, also the association between standing and M-value was near-significant (p = 0.07). Fig. 1 shows the unadjusted associations of standing with M-value and HOMA-IR. Dose-response associations were observed between tertiles of standing time (< 1.5, 1.5–2.0, and > 2.0 h/day) and both markers.
      Table 2Associations of standing with insulin sensitivity markers when adjusted for SB, total PA, and fitness.
      M-value (μmol/kg/min)HOMA-IR
      Data available for 63 participants.
      Fasting insulin
      Data available for 63 participants.
      (pmol/l)
      βpβpβp
      Standing (h/day) adjusted for:
      Sex, age, body fat %, wear time, SB (h/day)0.195

      (−0.136, 0.525)
      0.24− 0.474

      (−0.780, −0.170)
      0.003− 0.437

      (−0.745, −0.128)
      0.006
      Sex, age, body fat %, wear time, total PA (h/day)0.210

      (−0.028, 0.449)
      0.08− 0.392

      (−0.610, −0.173)
      <0.001− 0.374

      (−0.595, −0.154)
      0.001
      Sex, age, body fat %, wear time, VO2max (ml/min/kg)
      Data available for 58 participants.
      0.215

      (−0.031, 0.462)
      0.09− 0.352

      (−0.590, −0.115)
      0.005− 0.325

      (−0.563, −0.087)
      0.009
      Sex, age, body fat %, SB (h/day), total PA (h/day), VO2max (ml/min/kg)
      Data available for 58 participants.
      0.266

      (−0.018, 0.550)
      0.07− 0.345

      (−0.615, −0.074)
      0.01− 0.293

      (−0.565, −0.022)
      0.04
      HOMA-IR = homeostatic model of insulin resistance; SB = sedentary behavior; PA = physical activity.
      Values expressed as standardized β coefficients (95% CI); all values on log scale. Bold p-values indicate statistical significance (p < 0.05).
      a Data available for 63 participants.
      b Data available for 58 participants.
      Fig. 1
      Fig. 1Dose-response association between standing time and insulin sensitivity.
      Insulin sensitivity presented as a) M-value and b) HOMA-IR stratified by tertiles of accelerometer-measured standing time (h/day; means [SD]) in sedentary adults with metabolic syndrome (n = 64). HOMA-IR = homeostatic model assessment of insulin resistance. ** = p < 0.01, *** = p < 0.001.
      None of the SB, PA or fitness variables were associated with fasting glucose or HbA1c (Table A.4). All analyses were also run with lean mass M-value (μmol/kgFFM/min) as the dependent variable instead of M-value (μmol/kg/min), but the results were unchanged (data not shown). Additional analyses including the interaction between sex and each SB/PA variable in Model 1 were run, and stronger association of standing with HOMA-IR and insulin in men was the only statistically significant difference (men: β = −0.753 and − 0.763, respectively; women: β = −0.401 and − 0.377; p-values < 0.05).

      4. Discussion

      Our results indicate that standing is favorably associated with whole-body insulin sensitivity (M-value), HOMA-IR and fasting insulin. The association with HOMA-IR and insulin remains significant even when adjusted for adiposity, SB, PA, and fitness. Thus it seems that the more time is spent standing, the better insulin sensitivity is in sedentary adults at increased T2D risk. Adjustment for adiposity eliminated associations of SB and fitness with insulin sensitivity, suggesting that adiposity likely mediates these associations. This further highlights the importance of healthy body composition for metabolic health in individuals with MetS.
      Novel finding was the positive association between standing and insulin sensitivity. This is likely explained by muscle contraction required for standing, which promotes translocation of GLUT4 and thus increases glucose uptake and insulin sensitivity.
      • Bergouignan A.
      • Latouche C.
      • Heywood S.
      • et al.
      Frequent interruptions of sedentary time modulates contraction- and insulin-stimulated glucose uptake pathways in muscle: ancillary analysis from randomized clinical trials.
      We observed an association – although eliminated by adjustment for BMI – between standing and HOMA-IR also in our study of 144 overweight/obese adults with and without MetS.
      • Sjöros T.
      • Vähä-Ypyä H.
      • Laine S.
      • et al.
      Both sedentary time and physical activity are associated with cardiometabolic health in overweight adults in a 1 month accelerometer measurement.
      In contrast, Biddle et al.
      • Biddle G.J.H.
      • Edwardson C.L.
      • Henson J.
      • et al.
      Associations of physical behaviours and behavioural reallocations with markers of metabolic health: a compositional data analysis.
      found no association between standing and insulin sensitivity in compositional data analysis in individuals at T2D risk. To our knowledge, no other studies have investigated associations between standing and insulin sensitivity. However, isotemporal substitution modeling and experimental evidence suggests that insulin metabolism in high-risk individuals may be improved by replacing sitting with standing
      • Edwardson C.L.
      • Henson J.
      • Bodicoat D.H.
      • et al.
      Associations of reallocating sitting time into standing or stepping with glucose, insulin and insulin sensitivity: a cross-sectional analysis of adults at risk of type 2 diabetes.
      or interrupting sitting with standing bouts.
      • Henson J.
      • Davies M.J.
      • Bodicoat D.H.
      • et al.
      Breaking up prolonged sitting with standing or walking attenuates the postprandial metabolic response in postmenopausal women: a randomized acute study.
      In healthy populations, however, interrupting or replacing sitting with standing have not affected insulin action.
      • Saeidifard F.
      • Medina-Inojosa J.R.
      • Supervia M.
      • et al.
      The effect of replacing sitting with standing on cardiovascular risk factors: a systematic review and meta-analysis.
      ,
      • Pulsford R.M.
      • Blackwell J.
      • Hillsdon M.
      • et al.
      Intermittent walking, but not standing, improves postprandial insulin and glucose relative to sustained sitting: a randomised cross-over study in inactive middle-aged men.
      It may be that increased standing and frequent standing breaks could benefit sedentary populations with metabolic dysfunctions, whereas in general population standing might not be enough to improve insulin metabolism. However, in healthy adults standing has an inverse dose-response association with mortality,
      • Jain P.
      • Bellettiere J.
      • Glass N.
      • et al.
      The relationship of accelerometer-assessed standing time with and without ambulation and mortality: the WHI OPACH study.
      and increased energy expenditure in comparison to sitting.
      • Saeidifard F.
      • Medina-Inojosa J.R.
      • Supervia M.
      • et al.
      Differences of energy expenditure while sitting versus standing: a systematic review and meta-analysis.
      Therefore, standing is proposed as a feasible and generally safe alternative to sitting, and our results extend the evidence base.
      This is the first study investigating associations between accelerometer-measured breaks in SB and insulin sensitivity measured by hyperinsulinaemic-euglycaemic clamp. Previous evidence from studies using surrogate markers is inconsistent.
      • Healy G.N.
      • Matthews C.E.
      • Dunstan D.W.
      • et al.
      Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003-06.
      ,
      • Cooper A.R.
      • Sebire S.
      • Montgomery A.A.
      • et al.
      Sedentary time, breaks in sedentary time and metabolic variables in people with newly diagnosed type 2 diabetes.
      ,
      • Carson V.
      • Wong S.L.
      • Winkler E.
      • et al.
      Patterns of sedentary time and cardiometabolic risk among Canadian adults.
      We found that breaks were associated with clamp-measured whole-body insulin sensitivity but not surrogates, when adjusted for sex, age and accelerometer wear time. Breaking up sitting requires contraction of muscles which contribute largely to whole-body glucose disposal, whereas surrogate measures primarily reflect hepatic insulin sensitivity,
      • Muniyappa R.
      • Lee S.
      • Chen H.
      • et al.
      Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage.
      possibly explaining the association only with M-value. However, adjustment for adiposity eliminated the association.
      The offsetting effect of adiposity on the associations of breaks, SB and fitness with insulin sensitivity suggests that adiposity mediates to some extent these associations, which is biologically plausible. Obesity increases plasma concentration of NEFA, which then increases glucose and triglyceride production and other lipid/lipoprotein abnormalities, and reduces insulin-mediated glucose uptake and glycogen synthesis; all of which contribute to development of insulin resistance.
      • Eckel R.H.
      • Grundy S.M.
      • Zimmet P.Z.
      The metabolic syndrome.
      Our finding of body fat % and WC, as well as NEFA, triglycerides and HDL, correlating with M-value supports this mechanism. It is supported also by previous studies in which adiposity explained 65 % of the variance in clamp-measured insulin sensitivity,
      • Holt H.
      • Wild S.
      • Wareham N.
      • et al.
      Differential effects of fatness, fitness and physical activity energy expenditure on whole-body, liver and fat insulin sensitivity.
      and mediated the associations of MVPA and fitness with insulin sensitivity.
      • García-Hermoso A.
      • Martínez-Vizcaíno V.
      • Recio-Rodriguez J.I.
      • et al.
      Abdominal obesity as a mediator of the influence of physical activity on insulin resistance in Spanish adults.
      ,
      • Chartrand D.J.
      • Larose E.
      • Poirier P.
      • et al.
      Visceral adiposity and liver fat as mediators of the association between cardiorespiratory fitness and plasma glucose-insulin homeostasis.
      Additionally, we observed inverse correlations between adiposity measures, PA and fitness, and adjustment for adiposity turned the association between fitness and insulin sensitivity non-significant. These data together suggest that PA and fitness likely affect insulin sensitivity through effects leading to improved body composition. It is worth noting, however, that health status may also influence activity habits; e.g., it is possible that obesity leads to increased SB and not the other way around.
      • Pedisic Z.
      • Grunseit A.
      • Ding D.
      • et al.
      High sitting time or obesity: which came first? Bidirectional association in a longitudinal study of 31,787 Australian adults.
      Our finding that SB is not associated with insulin sensitivity when adjusted for adiposity is in line with results in overweight/obese adults with newly diagnosed T2D,
      • Cooper A.R.
      • Sebire S.
      • Montgomery A.A.
      • et al.
      Sedentary time, breaks in sedentary time and metabolic variables in people with newly diagnosed type 2 diabetes.
      whereas in adults with family history of T2D SB was not associated with insulin sensitivity independent of adiposity.
      • Ekelund U.
      • Griffin S.J.
      • Wareham N.J.
      Physical activity and metabolic risk in individuals with a family history of type 2 diabetes.
      We also found that neither LPA, MVPA nor total PA was associated with insulin sensitivity. Similar results have been reported previously regarding LPA,
      • Ekelund U.
      • Griffin S.J.
      • Wareham N.J.
      Physical activity and metabolic risk in individuals with a family history of type 2 diabetes.
      ,
      • Shalev-Goldman E.
      • McGuire K.A.
      • Ross R.
      Waist circumference and cardiorespiratory fitness are independently associated with glucose tolerance and insulin resistance in obese women.
      although in our recent study including overweight/obese adults with and without MetS, LPA was associated with HOMA-IR.
      • Sjöros T.
      • Vähä-Ypyä H.
      • Laine S.
      • et al.
      Both sedentary time and physical activity are associated with cardiometabolic health in overweight adults in a 1 month accelerometer measurement.
      Also contrary to our current findings, other studies show an association between MVPA and insulin sensitivity.
      • Ekelund U.
      • Brage S.
      • Griffin S.J.
      • et al.
      Objectively measured moderate- and vigorous-intensity physical activity but not sedentary time predicts insulin resistance in high-risk individuals.
      ,
      • Shalev-Goldman E.
      • McGuire K.A.
      • Ross R.
      Waist circumference and cardiorespiratory fitness are independently associated with glucose tolerance and insulin resistance in obese women.
      Findings regarding total PA, however, are more conflicting.
      • Ekelund U.
      • Griffin S.J.
      • Wareham N.J.
      Physical activity and metabolic risk in individuals with a family history of type 2 diabetes.
      ,
      • Sjöros T.
      • Vähä-Ypyä H.
      • Laine S.
      • et al.
      Both sedentary time and physical activity are associated with cardiometabolic health in overweight adults in a 1 month accelerometer measurement.
      ,
      • Shalev-Goldman E.
      • McGuire K.A.
      • Ross R.
      Waist circumference and cardiorespiratory fitness are independently associated with glucose tolerance and insulin resistance in obese women.
      It is possible that the amount of PA in this specific, homogenous population – highly sedentary adults with MetS – is not enough and has too little variation to detect significant associations, or the association is blunted by adiposity and other risk factors.
      Similar to previous studies in healthy and high-risk populations,
      • Healy G.N.
      • Matthews C.E.
      • Dunstan D.W.
      • et al.
      Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003-06.
      ,
      • Shalev-Goldman E.
      • McGuire K.A.
      • Ross R.
      Waist circumference and cardiorespiratory fitness are independently associated with glucose tolerance and insulin resistance in obese women.
      ,
      • Henson J.
      • Yates T.
      • Biddle S.J.
      • et al.
      Associations of objectively measured sedentary behaviour and physical activity with markers of cardiometabolic health.
      neither fasting glucose nor HbA1c was associated with SB/PA variables. Discrepancies between our study and others could be due to differences in populations and methodology. Our participants were sedentary adults with MetS, and thus at great likelihood of developing T2D. Although most abovementioned studies were also conducted in high-risk populations, participants were slightly healthier overall, and without as many metabolic abnormalities as in our study. Others have not specifically recruited sedentary/inactive participants, which has possibly led to greater PA variation and thus enabled detection of significant associations between PA and insulin sensitivity. Other reasons for discrepancies could be methodological issues relating to e.g., assessments of PA and insulin sensitivity. Previous accelerometer-assessments have generally lasted only ≤ 7 days, and insulin sensitivity assessment methods have varied. Discrepancies could also be affected by variation in confounding factors and analysis methods.
      A major strength of our study is the insulin sensitivity assessment with hyperinsulinaemic-euglycaemic clamp in addition to surrogate markers. M-value directly measures the amount of metabolized glucose under steady-state conditions, thus providing a reliable measure of whole-body insulin sensitivity. It is a costly and labor-intensive method however, which is why surrogates are often used. Another strength is the accelerometer-measurement of SB and PA with validated analysis algorithms for both the intensity of PA and body posture. Additionally, the 4-week measurement in our study likely represents the participants' long-term behaviors more truthfully than the ≤ 7- day measurement period used in most accelerometer studies.
      Our study also has limitations. The cross-sectional setting limits interpretation of causality, and although accelerometers provide more accurate measures of SB and PA than self-report and insights into patterns of SB (i.e., breaks), they cannot account for all activities (e.g., water-based activities). Furthermore, the step detection algorithm may not be able to count all sporadic steps taken at slow speeds; however, they still accumulate LPA. Our results might be generalizable only to inactive, sedentary, white populations with MetS, but given the prevalence of sedentary lifestyles and overweight/obesity, these results are likely generalizable to a larger proportion of Western populations. The considerable number of outcomes and analyses is also a potential limitation, although the main results remain even with lower statistical significance level (p ≤ 0.01). The relatively small sample size may also have limited detection of some associations. Thus further studies with larger samples are needed to confirm our findings, and to study causal relations between standing, as well as other SB/PA habits, and insulin sensitivity.

      5. Conclusion

      Our results indicate that standing is associated with insulin sensitivity. Furthermore, the association of standing with HOMA-IR and fasting insulin is independent of adiposity, SB, PA, and fitness. This suggests that simply standing more might be a potential way to improve insulin sensitivity in sedentary high-risk populations, warranting further (intervention) studies. Adiposity likely mediates the associations of SB and fitness with insulin sensitivity, which further highlights the importance of healthy body composition to prevent development of T2D in individuals at increased risk.

      Data availability

      The datasets generated during the current study are available from the corresponding author on reasonable request.

      Funding

      This work was supported by Academy of Finland; Finnish Cultural Foundation; Finnish Diabetes Research Foundation; Hospital District of Southwest Finland; and Juho Vainio Foundation. The study funders were not involved in the design of the study; the collection, analysis, and interpretation of data; writing the report; and did not impose any restrictions regarding the publication of the report.

      Declaration of interest statement

      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      Confirmation of ethical compliance

      All participants gave written informed consent. The study was approved by the Ethics Committee of the Hospital District of Southwest Finland (16/1810/2017), and good clinical practice and Declaration of Helsinki were followed.

      Acknowledgments

      This study was conducted within the Centre of Excellence in Cardiovascular and Metabolic Research, supported by the Academy of Finland, University of Turku, Turku University Hospital, and Åbo Akademi University.
      We thank the staff of Turku PET Centre, University of Turku, and the laboratory personnel in the Turku University Hospital Laboratory for their assistance.

      Appendix A. Supplementary data

      References

        • Guthold R.
        • Stevens G.A.
        • Riley L.M.
        • et al.
        Worldwide trends in insufficient physical activity from 2001 to 2016: a pooled analysis of 358 population-based surveys with 1.9 million participants.
        Lancet Glob Health. 2018; 6: e1077-e1086
        • Ekelund U.
        • Tarp J.
        • Fagerland M.W.
        • et al.
        Joint associations of accelerometer measured physical activity and sedentary time with all-cause mortality: a harmonised meta-analysis in more than 44 000 middle-aged and older individuals.
        Br J Sports Med. 2020; 54: 1499-1506
        • Biswas A.
        • Oh P.I.
        • Faulkner G.E.
        • et al.
        Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults: a systematic review and meta-analysis.
        Ann Intern Med. 2015; 162: 123-132
        • Huth C.
        • Pigeon É.
        • Riou M.
        • et al.
        Fitness, adiposopathy, and adiposity are independent predictors of insulin sensitivity in middle-aged men without diabetes.
        J Physiol Biochem. 2016; 72: 435-444
        • van der Velde J.H.P.M.
        • Schaper N.C.
        • Stehouwer C.D.A.
        • et al.
        Which is more important for cardiometabolic health: sedentary time, higher intensity physical activity or cardiorespiratory fitness? The Maastricht study.
        Diabetologia. 2018; 61: 2561-2569
        • Balkau B.
        • Mhamdi L.
        • Oppert J.M.
        • et al.
        Physical activity and insulin sensitivity: the RISC study.
        Diabetes. 2008; 57: 2613-2618
        • Ekelund U.
        • Griffin S.J.
        • Wareham N.J.
        Physical activity and metabolic risk in individuals with a family history of type 2 diabetes.
        Diabetes Care. 2007; 30: 337-342
        • Healy G.N.
        • Matthews C.E.
        • Dunstan D.W.
        • et al.
        Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003-06.
        Eur Heart J. 2011; 32: 590-597
        • Ekelund U.
        • Brage S.
        • Griffin S.J.
        • et al.
        Objectively measured moderate- and vigorous-intensity physical activity but not sedentary time predicts insulin resistance in high-risk individuals.
        Diabetes Care. 2009; 32: 1081-1086
        • Cooper A.R.
        • Sebire S.
        • Montgomery A.A.
        • et al.
        Sedentary time, breaks in sedentary time and metabolic variables in people with newly diagnosed type 2 diabetes.
        Diabetologia. 2011; 55: 589-599
        • Sjöros T.
        • Vähä-Ypyä H.
        • Laine S.
        • et al.
        Both sedentary time and physical activity are associated with cardiometabolic health in overweight adults in a 1 month accelerometer measurement.
        Sci Rep. 2020; 10: 20578
        • Vähä-Ypyä H.
        • Vasankari T.
        • Husu P.
        • et al.
        Validation of cut-points for evaluating the intensity of physical activity with accelerometry-based mean amplitude deviation (MAD).
        PLoS One. 2015; 10e0134813
        • Vähä-Ypyä H.
        • Husu P.
        • Suni J.
        • et al.
        Reliable recognition of lying, sitting, and standing with a hip-worn accelerometer.
        Scand J Med Sci Sports. 2018; 28: 1092-1102
        • Bergouignan A.
        • Latouche C.
        • Heywood S.
        • et al.
        Frequent interruptions of sedentary time modulates contraction- and insulin-stimulated glucose uptake pathways in muscle: ancillary analysis from randomized clinical trials.
        Sci Rep. 2016; 6: 32044
        • Biddle G.J.H.
        • Edwardson C.L.
        • Henson J.
        • et al.
        Associations of physical behaviours and behavioural reallocations with markers of metabolic health: a compositional data analysis.
        Int J Environ Res Public Health. 2018; 15: 10
        • Edwardson C.L.
        • Henson J.
        • Bodicoat D.H.
        • et al.
        Associations of reallocating sitting time into standing or stepping with glucose, insulin and insulin sensitivity: a cross-sectional analysis of adults at risk of type 2 diabetes.
        BMJ Open. 2017; 7e014267
        • Henson J.
        • Davies M.J.
        • Bodicoat D.H.
        • et al.
        Breaking up prolonged sitting with standing or walking attenuates the postprandial metabolic response in postmenopausal women: a randomized acute study.
        Diabetes Care. 2016; 39: 130-138
        • Saeidifard F.
        • Medina-Inojosa J.R.
        • Supervia M.
        • et al.
        The effect of replacing sitting with standing on cardiovascular risk factors: a systematic review and meta-analysis.
        Mayo Clin Proc Innova Qual Outcomes. 2020; 4: 611-626
        • Pulsford R.M.
        • Blackwell J.
        • Hillsdon M.
        • et al.
        Intermittent walking, but not standing, improves postprandial insulin and glucose relative to sustained sitting: a randomised cross-over study in inactive middle-aged men.
        J Sci Med Sport. 2017; 20: 278-283
        • Jain P.
        • Bellettiere J.
        • Glass N.
        • et al.
        The relationship of accelerometer-assessed standing time with and without ambulation and mortality: the WHI OPACH study.
        J Gerontol A Biol Sci Med Sci. 2021; 76: 77-84
        • Saeidifard F.
        • Medina-Inojosa J.R.
        • Supervia M.
        • et al.
        Differences of energy expenditure while sitting versus standing: a systematic review and meta-analysis.
        Eur J Prev Cardiol. 2018; 25: 522-538
        • Carson V.
        • Wong S.L.
        • Winkler E.
        • et al.
        Patterns of sedentary time and cardiometabolic risk among Canadian adults.
        Prev Med. 2014; 65: 23-27
        • Muniyappa R.
        • Lee S.
        • Chen H.
        • et al.
        Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage.
        Am J Physiol Endocrinol Metab. 2008; 294: 15-26
        • Eckel R.H.
        • Grundy S.M.
        • Zimmet P.Z.
        The metabolic syndrome.
        Lancet. 2005; 365: 1415-1428
        • Holt H.
        • Wild S.
        • Wareham N.
        • et al.
        Differential effects of fatness, fitness and physical activity energy expenditure on whole-body, liver and fat insulin sensitivity.
        Diabetologia. 2007; 50: 1698-1706
        • García-Hermoso A.
        • Martínez-Vizcaíno V.
        • Recio-Rodriguez J.I.
        • et al.
        Abdominal obesity as a mediator of the influence of physical activity on insulin resistance in Spanish adults.
        Prev Med. 2016; 82: 59-64
        • Chartrand D.J.
        • Larose E.
        • Poirier P.
        • et al.
        Visceral adiposity and liver fat as mediators of the association between cardiorespiratory fitness and plasma glucose-insulin homeostasis.
        Am J Physiol Endocrinol Metab. 2020; 319: E548-E556
        • Pedisic Z.
        • Grunseit A.
        • Ding D.
        • et al.
        High sitting time or obesity: which came first? Bidirectional association in a longitudinal study of 31,787 Australian adults.
        Obesity. 2014; 22: 2126-2130
        • Shalev-Goldman E.
        • McGuire K.A.
        • Ross R.
        Waist circumference and cardiorespiratory fitness are independently associated with glucose tolerance and insulin resistance in obese women.
        Appl Physiol Nutr Metab. 2014; 39: 358-362
        • Henson J.
        • Yates T.
        • Biddle S.J.
        • et al.
        Associations of objectively measured sedentary behaviour and physical activity with markers of cardiometabolic health.
        Diabetologia. 2013; 56: 1012-1020

      Linked Article