THE RELATIONSHIP BETWEEN INSULIN RESISTANCE, BONE MINERAL DENSITY AND FRACTURE RISK IN POSTMENOPAUSAL WOMEN

<jats:p>nema</jats:p>


INTRODUCTION
Skeletal muscle is a vital tissue that supports body posture and is also the primary glucose uptake site after a meal. Skeletal muscle is significantly related to insulin resistance 1 . The connection between bone strength or mineral bone density and insulin resistance is very complex 2 . According to the study, when the inflammatory response is inadequate, as in the case of aging muscles, acellular fat droplets and adipocytes tend to accumulate, so the development of insulin resistance may be the inflammatory response of the muscles 3 . This results in the secretion of different cytokines, chemokines, and adipocytes, which affects insulin resistance 1,3 .
The World Health Organization has defined natural menopause as the least twelve consecutive months of amenorrhea, not physiologic and pathological causes. According to statistics, the mean age of natural menopause is 51 years in industrialized nations, compared to 48 years in low and non-industrialized nations 1 . With the average life span extended to 70 years, most women will spend more than one-third of their life beyond the menopausal transition. Besides, the proportion of menopausal women is rising since the aging population is expanding rapidly.
A significant number of studies, on the other hand, studies discuss the impact of reduced muscle mass on bone mineral density and the consequences that result from them, in the first place, a higher incidence of osteoporotic fractures 4 . It is known that muscle mass and osteoporosis, and metabolic disorders are closely related. However, data on the association of muscle properties, bone mass, and insulin resistance are lacking 5 . Besides, few of them talk about the association of bone mineral density, muscle mass, and insulin resistance in postmenopausal women, especially in women in Serbia.
The key goal of this study was to determine the effects of insulin resistance on bone mineral density and fracture risk and evaluate the relationship between muscle properties (muscle mass, muscle strength, and physical performance) and bone mineral density and insulin resistance in postmenopausal women in Serbia.

Ethical concerns
The protocol, as well as the study procedures, were approved by the ethics committee, the Clinical Center in Kragujevac (number 01/17-3765), and the Faculty of Medical science, University of Kragujevac (number 01-15581/3-6) from November 2017.
to June 2018.

Study design
The study was conducted at the Clinical Center Kragujevac, the reference healthcare institution for osteodensitometry in the region of central Serbia. The study was designed as a clinical, non-interventional, observational, cross-sectional study and included 66 women over 65 years of age who were selected through random sampling. Participants were divided into two groups and based on the HOMA-IR limit values as used in the study by Nikolic et al. 6. The cut-off value for participants from the group "Low HOMA-IR" was <2, and for those from the group "High HOMA-IR," the values of insulin resistance were > 2. Among the study participants were 44 women with osteoporosis (T score <2.5) and 22 women with normal bone mineral density or osteopenia (T score ≥-2.5, without fracture data).

Inclusion and exclusion criteria
The inclusion criteria were confirmed as a menopause of at least five years based on no menstruation. None of the participants had the following diseases that affect bone mineral density such as hyperthyroidism, hyperparathyroidism, renal failure, malabsorption syndrome, chronic colitis, multiple myeloma, leukemia, chronic arthritis, DM, or previous use of therapy that interfere with bone metabolism (e.g., glucocorticoids, heparin, warfarin, thyroxin, and estrogen). Also, the exclusion criteria were cigarette smoking, alcohol intake, BMI> 30kg/m 2 , and <19kg/m 2 , respectively. Before joining the study, all participants confirmed their participation in it with their own signatures.

The insulin resistance
Index expressed as HOMA-IR was calculated using the following equation, as described by Matthews et al.: HOMA-IR=[glucose (mg/dL)×insulin (μU/mL)]/405 for each participants 7 . Due to its simplicity and calculation, the most commonly used technique in clinical practice but also in epidemiological studies for the assessment of insulin resistance was the homeostatic test (HOMA-IR) 7 .

Osteodensitometric, anthropometric, and body composition measurements
Bone Mass Density (BMD; g/cm 2 ) was measured on the lumbar spine (LS) in the region L1-L4 and total hip in all participants. The measurement was done with a densitometer with X-ray energy absorption (DXA) (QDR 4500, Hologic Model Discoveri Inc., Waltham, MA) 8 . Participants did not wear metal items (for example, clips, belts, brassieres, jewelry) and shoes. There were instructed to be motionless during the scan.
Daily standardized quality control of DXA instruments was performed using the manufacturer's phantom spine before the start of the study. The definition of osteopenia and osteoporosis was made using the World Health Organization (WHO) -2.5 < T-score < -1 and T-score -2.5, respectively. Body weight and height were obtained from the mean of three measurements. The accurate and precise values of these body composition parameters were also estimated from the DXA scan of the total body, which included BMC (bone mineral content), LM (lean mass), and FM (fat mass).
Following the manufacturer's guidelines, all scans were obtained and analyzed by the same experienced operator 9 . Muscle strength was measured using the handgrip test (HGT) dynamometer and is closely related to the muscle strength of the lower extremities 10 . In our study, the Jamar dynamometer was used, which is small, portable, and easy to handle. It was considered a reduced muscle strength HGT <16kg 11 . To measure physical performance, we use the walk's speed test at a distance of 4 m (gait speed-GS).
Physical ability is considered to be reduced when the gait speed is <0.8m / s for 4m 12 .

Fracture risk
FRAX algorithm was used to calculate the probability of major osteoporotic fractures and hip fracture ((www.Sheffield.ac.Uk/FRAX/), using data specific to our country 13,14 . FRAX Index 1 represented the probability of a major osteoporotic fracture (clinical fracture of the spine, forearm, hip, or shoulder), while FRAX Index 2 represents the probability of hip fracture.

Statistical analyses
IBM SPSS version 20 (IBM Company, Armonk, NY) was used for all statistical analyses. The outcome variables used were BMD of the whole body and at skeletal sites.
The sample size was calculated using G*Power software version 3.1, and 66 subjects were required for a 90% power and 5% for the t-test. The cases and controls were distributed between two groups 15 . The values of all variables for the whole body and regional sites were presented as mean (M)standard deviation (SD). Comparison of mean values between two groups of subjects, those with osteoporosis and those with normal bone mineral density/ osteopenia, are classified according to their spine, bones, and BMD of the entire body, as well as weight, height, BMI, LM, FM, total weight and body fat.
Correlation analyses of the whole body, regional sites BMD, and T-scores with the independent variables such as weight, LM, FM, and BMD were performed to obtain the Pearson's correlations. We use stepwise multiple linear regression analysis to obtain determinants/predictors for the outcome variables. All p-values were reported significant at 0.05 or less 16 .

HOMA-IR Groups
The average values and standard deviations/standard errors of means of the examined parameters according to the level of HOMA-IR are shown in Table 1. Body mass index, waist size, Lean Mass, Fat mass, Total mass, and Lean+BMC were significantly different in Low-HOMA-IR and High-HOMA-IR groups (p<0.05). Other tested parameters were not significantly different in these groups (Table 1).

Baseline anthropometric and demographic characteristics of the study population
The average age of participants was 71.20±4.72 years, with the range being 65 to 83. For women with normal bone mass/osteopenia, the mean age was (±SD) 70.91 ± 4.97 years, while the mean age for women with osteoporosis was 71.34 ± 5.09 (Table 2).

Parameters, Bone Mineral Density (BMD)
In the study population, regarding their BMD values, LM was shown to have a higher degree of positive correlation with BMD on the lumbar spine (ꞵ 0.418, p<0.001) but also had a significant effect on the hip (ꞵ 0.416, p<0.01). In contrast, FM showed a high degree of positive correlation with both BMD at the hip (ꞵ 0.473, p<0.001) and the lumbar spine (ꞵ 0.480, p<0.001).
In this study, the results showed a significant degree of a positive correlation between HGT and bone mineral density of the hip (ꞵ 0.331, p<0.01) and spine (ꞵ 0.243, p<0.05), whereas GS was only correlated with bone mineral density on the hip (ꞵ 0.268, p<0.05). (Table 3).
A positive correlation was confirmed between HOMA-IR and fat mass (ꞵ=0.322, p˂0.05) and total mass (ꞵ=0.287, p˂0.05). However, there was no correlation between HOMA-IR and lean mass (ꞵ 0.163, p >0.05). (Table 4). A significant degree of positive correlation was obtained between HOMA-IR and body mass index (ꞵ=0.381, p˂0.01) and waist circumference (ꞵ=0.405, p 0.001). A high degree of positive correlation was also observed between HOMA-IR and bone mineral density on the spine (ꞵ=0.362, p 0.01) and the T score of the spine (ꞵ=0.359, p 0.01). Besides, a correlation was also shown between HOMA-IR and hip bone mineral density (ꞵ=0.264, p˂0.05) and T score hip (ꞵ=0.305, p˂0.05). In the study, no correlation was confirmed between insulin resistance and muscle strength, measured by handgrip, and physical performance measured by gait speed (Table   4).

Univariate linear regression model of analysis between HOMA-IR and BMD parameters, DXA scores, and Blood markers in postmenopausal women
Univariate regression analysis with HOMA-IR as the dependent variable showed marginally significant associations between HOMA-IR and lumbar spine BMD (p = 0.055, stand. beta coefficient = 0.311), which means that an increase of HOMA-IR will lead to an increase of values of BMD of the spine (Table 5). Also, univariate regression analysis confirmed the association between HOMA-IR and changes of T-score of the spine (p = 0.009, stand. beta coefficient = 0.387) ( Table 6). Regarding to the significance of blood markers as a predictors, we statistically confirmed the significance of Insulin level (p = 0.000, stand. beta coefficient = 0.241), Glucose level (p = 0.000, stand. beta coefficient = 0.350) and inversed association and marginally significance of fT4 levels (p = 0.071, stand. beta coefficient =-0.027) ( Table 7).
Considering that the association between other tested parameters and HOMA-IR appears to be statistically insignificant in univariate linear regression analysis, other variables are not recognized as a predictor of changing HOMA-IR values in postmenopausal women (Tables 5-7).

Comparison of Anthropometric parameters between women with osteoporosis and women with normal to osteopenic bone mass
No significant difference was observed (p = 0.935) in age between women with osteoporosis (M = 69.5) and women in group with normal bone mass/osteopenic (M = 70).
The group of women with normal BMD/osteopenia has an 18% higher body mass (р<0.001), 13% more LM (р<0.001), and even 30% more FM (р<0.001). This group of women has about 9% more LMI (р=0.003) and about 25% more FMI (р=0.001). HOMA-IR had a mean of 1.92 in the subjects and ranged from 0.2 to 6.7. HOMA-IR values were 1.56 in the subjects with osteoporosis and 2.68 in the group with normal BMD / osteopenia (Table 3).

DISCUSSION
The association between muscle properties, bone mineral density, and insulin resistance in this study was evaluated based on body composition parameters, muscle strength, and physical performance 17 . Based on these parameters, we provide clinical evidence that body composition changes, muscle strength, and physical performance are associated with decreased bone mineral density. Also, adipose tissue accumulation and an increase in total mass are closely related to insulin resistance. Finally, we confirmed the association between bone mineral density and insulin resistance in postmenopausal women in Serbia.
Fat mass is a significant source of proinflammatory cytokines that mediate bone metabolism, and postmenopausal women tend to accumulate visceral fat. Some authors reported in their studies the independent effect of fat mass on BMD over estrogens, insulin, and leptin 16,18 . It was also observed that the relative contribution of body composition parameters to BMD depends on gender, ethnicity, and age 19 .
Ho-Pam et al. 8  In the present study, lean mass, muscle strength, and physical performance were not associated with insulin resistance. In contrast, adipose tissue and bone mineral density on the hip, and especially on the spine, were significantly associated with insulin resistance: women with higher adipose tissue showed higher insulin resistance levels. Therefore, our study implies that the reduction of lean mass accompanied by its damage and the This is consistent with our results, which indicate that although it has been found that there is a positive correlation between insulin resistance and BMD, there is no correlation between insulin resistance and fracture risk. This means that women with higher insulin resistance levels and higher bone mineral density do not have a lower fracture risk.
Several studies have investigated the correlation between muscle parameters, fat accumulation, and insulin resistance 23 . With aging, muscle mass is lost, muscle damage and fat accumulation occur. Specifically, the infiltration of muscle tissue by fat leads to the activation of apoptotic cells and the release of inflammatory cytokines and adipokines, leading to the development of insulin resistance 24 . On this basis, the idea that local inflammation in the muscle followed by the accumulation of fat by secretion of cytokines and adipokines instead of a decrease in muscle strength and physical performance may have a more important role in the production of insulin resistance 25 . In this regard, our results may provide clinical evidence to support the results of other studies.

CONCLUSION
These results suggest that lean mass and fat mass significantly affect BMD and muscle strength, and physical performance. Besides, a decrease in lean mass, muscle damage, and fat buildup has been associated with a higher incidence of insulin resistance in postmenopausal women. Finally, bone mineral density on the hip, and especially on the spine, was associated with the onset of insulin resistance. However, there is no correlation between insulin resistance and fracture risk. These results significantly contribute to understanding the changes that occur in the body with aging in these women.

Basic Anthropometric Characteristics of study population divided into Low-HOMA-
IR and High-HOMA-IR groups. The cut-off value for HOMA-IR was 2; Independent T-Test confirmed statistical differences for normally distributed data with the level of significance of 0.05; HOMA-IR -Homeostatic model assessment -insulin resistance; BMD LH -Bone mineral density left hip; BMC -Bone mineral content.

Comparison of Anthropometric parameters and Dual/Energy X/Ray absorptiometry
(DXA) in a woman with normal BMD/osteopenia and a woman with osteoporosis.   β-standardized β-coefficient; Sig-significance level; B-coefficient of the model.

Univariate linear regression analysis between HOMA-IR and Blood markers
βstandardized β-coefficient; Sig -significance level; B-coefficient of the model.