- Email: [email protected]
Positive Association of Obesity and Insulin Resistance With Bone Mineral Density in Tunisian Postmenopausal Women
ARTICLE IN PRESS Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health, vol. ■, no. ■, 1–9, 2017 © 2017 The International Society for Clinical Densitometry. 1094-6950/■:1–9/$36.00 http://dx.doi.org/10.1016/j.jocd.2017.05.015
Original Article
Positive Association of Obesity and Insulin Resistance With Bone Mineral Density in Tunisian Postmenopausal Women Rim Cherif,*,1,2 Feten Mahjoub,3 Hela Sahli,4 Elhem Cheour,4 Laurence Vico,2 Mohsen Sakly,1 and Nebil Attia1 1
UR11ES33, Research Unit ’Integrated Physiology, Laboratory of Biochemistry-Human Nutrition, Faculty of Sciences of Bizerte, Carthage University, Tunis, Tunisia; 2INSERM U1059, Integrative Biology of Bone Tissue Laboratory, Lyon University, St-Etienne, France; 3Department of Diabetology, National Institute of Nutrition, Tunis, Tunisia; and 4ImmunoRheumatology Laboratory, Rabta Hospital, Faculty of Medicine of Tunis, Tunis El Manar University, Tunis, Tunisia
Abstract The association of bone mineral density (BMD) with obesity and insulin resistance remains unclear. This study aimed to explore these associations in Tunisian menopausal women. Eighty-one postmenopausal women were recruited. Data were analyzed for obese (N = 57) and non-obese women (N = 24) and for insulinresistant (N = 43) and non insulin-resistant women (N = 36). Anthropometric and biochemical parameters were recorded. BMD in different sites and body composition were measured using dual-energy X-ray absorptiometry. Higher BMD was observed in obese women than those non-obese in the left femur (p = 0.0067), right femur (p = 0.0108), total hip (p = 0.0077), and the whole body (p = 0.0276). Also BMD was significantly greater in insulin-resistant women than in non–insulin-resistant women when measured in the left femur and total hip. Positive correlations were recorded between BMD and anthropometric parameters, body composition parameters, and glycemia (r = 0.249, p < 0.05). Multiple linear regression analysis shows that only trunk fat (p < 0.05) and lean mass (p < 0.05) were independently and positively related to BMD, and the waist circumference was the only anthropometric parameter independently and negatively associated to BMD. BMD is improved in obese and insulin-resistant women. Also, trunk fat and lean mass are likely to be key positive independent factors for BMD. Key Words: Bone mineral density; fat; insulin resistance; lean mass; obesity.
Introduction
studies have demonstrated that obese subjects present higher BMD than normal-weight subjects (4,5) and increased body weight or body mass index (BMI) is related to higher BMD and reduced fracture risk (6). In postmenopausal women, obesity has been considered a protective factor for bone loss and osteoporosis, likely for increases in the number of adipocytes, which are important sources of estrogen derived from aromatization, resulting in increased BMD (7,8). Although there is a little doubt that body weight has an effect on bone, whether it is the effect of fat mass or lean mass that influences BMD is disputed. Studies in postmenopausal women have suggested that fat mass plays a key role (9,10), and both fat and lean mass have been reported to be significant predictors of BMD (11,12). However, there is also
Osteoporosis is a major public health problem characterized by a decrease in bone mineral density (BMD) and an alteration of bone quality (1). Obesity, another common disease, has been demonstrated to be closely related with osteoporosis (2,3). Despite their sedentary lifestyle, several
Received 10/22/16; Revised 05/22/17; Accepted 05/23/17. Disclosure of interest: The authors report no conflicts of interests in this work. *Address correspondence to: Rim Cherif, PhD, UR11ES33 Integrated Physiology, Laboratory of Biochemistry and Human Nutrition, Faculty of Sciences of Bizerte, Carthage University, Jarzouna, 7021 Bizerte, Tunisia. E-mail: [email protected]
1
ARTICLE IN PRESS 2 disagreement in the literature on the positive effect of lean and fat mass on BMD (10,13). In addition to obesity, there have been conflicting reports about the relation between insulin-resistance and BMD. Some authors have reported elevated BMD in the presence of insulin resistance (14,15), whereas some have reported decreased BMD (16,17). Because of the discrepancies in the results, the association between insulin resistance and bone mass remains unclear, and further studies are needed to explain this relationship. Prevalence of osteoporosis and metabolic syndrome greatly increases in the Tunisian population. It was estimated that 31.6% of women develop the metabolic syndrome (18) and 23.4% are osteoporotic (19). Based on the conflicting studies results and the high prevalence of these diseases in Tunisia, we evaluate the relationship between BMD and obesity at different bone sites in a cohort of Tunisian postmenopausal women. We also investigate other factors related to body composition (leg fat, trunk fat, total fat, lean mass) in order to identify those potentially associated with BMD. We also sought to determine the association between insulin resistance and bone mass in normal-weight women.
Patients and Methods Patients A total of 81 postmenopausal women (age range: 50– 83 years) were recruited consecutively between June 2014 and April 2015 from the Endocrinology Department of the National Institute of Nutrition of Tunis, Tunisia, during visits for routine checkups. The study was approved by the local ethic committee of National Institute of Nutrition of Tunis and written or oral consent was obtained from all the patients before the study. Written consents were not possible in all cases because the majority of eligible subjects in our study were unable to read or write. The consent includes the agreement of patients to participate in the clinical study and to undergo free densitometry examination. Original inclusion criteria were age >50, postmenopausal (defined as cessation of menstruation for at least 1 year), and no previous osteoporotic fracture or known diagnosis of osteoporosis. Women with liver or renal disease, inflammatory disease, parathyroid and thyroid, chronic inflammatory rheumatism, early natural menopause before age 40, and receiving medicine known to influence bone metabolism, such as corticosteroids, heparin, anticonvulsants, vitamin D or calcium supplementations, and bisphosphonates, were not included. Participants were questioned about their age, age at the onset of menopause, education, occupation, family history of osteoporosis, history of peripheral traumatic fractures, weight history, smoking habits, level of physical activity, and current medication use. Smoking habit was categorized as non-smoker and current smoker. Physical activity was selfgraded by participants and categorized as sedentary, moderate, and important.
Cherif et al. Anthropometric Measurements Body weight was measured to the nearest 0.1 kg and height was determined to the nearest centimeter. BMI was calculated as the weight (kg) divided by the square of height (m2). Patients were considered as non-obese if BMI < 30 kg/ m2 and as obese if BMI ≥ 30 kg/m2 according to the World Health Organization (WHO) definition (20). Waist circumference was measured at the narrowest part of the abdomen, that is, at the natural indentation between the 10th rib and the iliac crest (minimum waist).
Biochemical Analysis Fasting blood samples were taken after fasting for at least 12 h. Venous blood samples were taken from an antecubital vein and placed into heparinized or non-heparinized tubes. Tubes were centrifuged at 3000 × g for 10 min. Serum fasting blood glucose, glycated hemoglobin (HbA1c), lipid levels (total cholesterol, triglycerides, and high-density lipoprotein cholesterol) were determined by well-validated laboratory routine methods. Serum lowdensity lipoprotein cholesterol values were estimated using the Friedwald formula (21). Fasting blood samples were collected in non-heparinized tube for the serum insulin concentrations measurement.The tube was left for 30 min at room temperature (to allow clot formation) then centrifuged at 1000 × g for 15 min. The serum was separated and stored in microtubes of 250 µL at 80°C until the day of manipulation. After adequate thawing, insulin was measured by enzyme-linked immunosorbent assay (ALPCO kit, Salem, NH). Insulin resistance was measured according to the following formula: HOMA-IR (homeostasis model assessment of insulin resistance) = fasting insulin (µU/mL) × fasting glucose (mmol/ L)/22.5 (22). Patients were considered as insulin resistant when HOMA-IR ≥ 2.6 and non-insulin resistant when HOMA-IR < 2.6.
Dual-energy X-ray Absorptiometry (DXA) Measurements Densitometry examination of patients was realized at the Rheumatology Department of the Rabta Hospital of Tunis. Lumbar (anteroposterior projection at L1–L4), femur, femoral neck, total hip, and whole body BMD as well as fat (leg, trunk, and total) and total lean mass were measured by DXA using GE-Lunar PRODIGYTM device. Quality control procedures were carried out in accordance with the manufacturer’s recommendations. Instrument variation was determined by a daily calibration procedure using a phantom supplied by manufacturer. The precision error was <2.0% for each measured sites at standard speed based on repeated scans in a random sample of 30 subjects. BMD values were expressed in grams per square centimeter (g/cm2), and fat and lean mass values were expressed in grams.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS Association of Obesity and Insulin Resistance With BMD Statistical Analysis The characteristics of the patients are described with means ± standard deviations (SD). Differences between groups were compared by an independent 2-sample t test. Pearson’s correlations were performed to examine the association between bone-related parameters and biochemical and anthropometric parameters. Multiple linear regression analysis was used to reveal statistically independent relationships between variables. In all cases, the level of statistical significance was set at p < 0.05. All statistics were performed using the StatView package (Version 5, SAS Institute, Inc., Cary, NC).
Results The clinical and laboratory characteristics of the 81 postmenopausal women are listed in Table 1. The mean age of Table 1 Descriptive Characteristics of the Study Population Characteristics Age (yr) Years since menopause (yr) Weight (kg) Height (cm) Waist circumference (cm) BMI (kg/m2) Glycemia (mmol/L) HbA1c (%) Serum insulin (µL U/mL) HOMA-IR Total cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglycerides (mmol/L) L-BMD (g/cm2) LF-BMD (g/cm2) RF-BMD (g/cm2) TH-BMD (g/cm2) LFN-BMD (g/cm2) RFN-BMD (g/cm2) WB-BMD (g/cm2) Leg fat mass (g) Trunk fat mass (g) Total body fat mass (g) Total lean mass (g)
N = 81 58.40 ± 6.08 10.79 ± 7.02 81.54 ± 13.49 156.45 ± 6.10 110.72 ± 12.94 33.36 ± 5.42 10.89 ± 4.23 8.70 ± 1.94 9.10 ± 11.14 4.09 ± 4.39 5.00 ± 1.40 1.23 ± 0.28 3.04 ± 1.03 1.62 ± 0.88 1.08 ± 0.16 1.02 ± 0.14 1.00 ± 0.13 1.01 ± 0.13 0.93 ± 0.14 0.93 ± 0.12 1.15 ± 0.09 12084.77 ± 4040.97 19052.77 ± 4062.34 37326.48 ± 8615.18 40805.76 ± 5766.78
Data are mean ± SD. Abbr: BMD, bone mineral density; BMI, body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; HOMA-IR, homeostasis model assessment of insulin resistance; L-BMD, lumbar spine BMD; LDL, low-density lipoprotein; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD.
3
the participants was 58.40 ± 6.08 years and the years since menopause was 10.79 ± 7.02. The mean weight (±SD) and BMI was 81.54 ± 13.49 kg and 33.36 ± 5.42 kg/m2, respectively. In this population, the average of glycemia was 10.89 ± 4.23 mmol/L and HOMA-IR index was 4.09 ± 4.39. The BMD and body composition at all the studied sites are shown in detail in Table 1. In order to analyze the effect of obesity on BMD, participants were classified according to their BMI as obese (N = 57) and non-obese (N = 24). Mean weight and BMI were 68.04 ± 6.83 kg and 27.67 ± 2.12 kg/m2, respectively, in non-obese women group and 87.22 ± 11.39 kg (p < 0.0001) and 35.75 ± 4.53 kg/m2 (p < 0.0001), respectively, in obese women group. Comparison between obese and nonobese patients showed that there was no difference between the groups for age, age of menopause, glycemia, and HOMA-IR index. However, we found that obese women had significantly greater BMD at the left femur (p = 0.006), right femur (p = 0.010), total hip (p = 0.007), and the whole body (p = 0.027) compared with non-obese women. Also, there were statistically significant differences between groups in the body composition. In fact, leg, trunk, and total body fat mass was higher in obese women than in non-obese women group (p < 0.0001 for each parameters) as well as total lean mass (p < 0.0001) (Table 2). To investigate the effect of insulin resistance on BMD, the studied population was divided into insulin-resistant (N = 43) and non insulin-resistant (N = 36) groups according to the HOMA-IR index. The weight, waist circumference, and BMI were not significantly different between groups. The mean glycemia and serum insulin were 11.80 ± 3.88 mmol/L and 12.82 ± 14.02 µl U/mL, respectively, in insulin-resistant group and 9.45 ± 4.17 mmol/L (p = 0.011) and 4.67 ± 1.92 µL U/mL (p = 0.0009) in non insulin-resistant group. The HOMA-IR index was highly different between the 2 groups (p < 0.0001). Statistically significant differences between groups in the DXA measurements were observed in only 2 sites: left femur (p = 0.02) and total hip (p = 0.04), with greater BMD values in the insulin-resistant group in both sites (Table 3). Univariate regression analysis was evaluated for the whole population (Table 4). This analysis revealed that both age and age of menopause were significantly and negatively related to all measured sites BMD except for lumbar BMD. However, body weight was positively related to femur, total hip, and the whole body BMD. A positive correlation was noted between BMI and right femur BMD (r = 0.259, p < 0.05), total hip BMD (r = 0.258, p < 0.05), and whole body BMD (r = 0.282, p < 0.05). Glycemia was positively correlated to the whole body BMD (r = 0.249, p < 0.05) but not to other sites. We did not find any relation between BMD measures and lipid variables at any sites. For body composition measurements, trunk fat, total body fat, and total lean mass correlate positively and significantly to BMD measurements (Table 4). Multiple linear regression analysis was performed to evaluate independent predictor of BMD measurements. In
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS 4
Cherif et al. Table 2 Clinical Characteristics According to the Presence or Absence of Obesity
Age (yr) Years since menopause (yr) Weight (kg) Height (cm) Waist circumference (cm) BMI (kg/m2) Glycemia (mmol/L) HbA1c (%) Serum insulin (µL U/mL) HOMA-IR Total cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglyceride (mmol/L) L-BMD (g/cm2) LF-BMD (g/cm2) RF-BMD (g/cm2) TH-BMD (g/cm2) LFN-BMD (g/cm2) RFN-BMD (g/cm2) WB-BMD (g/cm2) Leg fat mass (g) Trunk fat mass (g) Total body fat mass (g) Total lean mass (g)
Non obese women (n = 24)
Obese women (n = 57)
p Value
58.70 ± 5.46 11.58 ± 6.93 68.04 ± 6.83 156.79 ± 6.91 100.20 ± 7.55 27.67 ± 2.12 11.96 ± 4.60 9.04 ± 2.10 6.43 ± 4.52 3.14 ± 1.97 5.67 ± 1.65 1.25 ± 0.18 3.54 ± 1.13 1.82 ± 1.31 1.05 ± 0.15 0.96 ± 0.13 0.94 ± 0.12 0.95 ± 0.13 0.91 ± 0.12 0.90 ± 0.12 1.11 ± 0.09 9074.04 ± 2125.98 15205.70 ± 3129.60 28899.75 ± 4297.60 36587.37 ± 2820.78
58.28 ± 6.37 10.44 ± 7.09 87.22 ± 11.39 156.31 ± 5.79 115.15 ± 12.18 35.75 ± 4.53 10.44 ± 4.03 8.56 ± 1.87 10.20 ± 12.79 4.49 ± 5.03 4.70 ± 1.17 1.22 ± 0.32 2.86 ± 0.94 1.54 ± 0.64 1.09 ± 0.16 1.05 ± 0.13 1.02 ± 0.13 1.04 ± 0.13 0.94 ± 0.15 0.95 ± 0.12 1.17 ± 0.09 13352.45 ± 3993.87 20672.59 ± 3246.38 40874.57 ± 7429.55 42581.93 ± 5779.51
0.774 0.512 <0.0001 0.751 <0.0001 <0.0001 0.140 0.317 0.173 0.216 0.009 0.736 0.028 0.244 0.291 0.0067 0.0108 0.0077 0.301 0.103 0.0276 <0.0001 <0.0001 <0.0001 <0.0001
Data are mean ± SD. Significant differences are shown in bold font. Abbr: BMD, bone mineral density; BMI, body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; HOMAIR, homeostasis model assessment of insulin resistance; L-BMD, lumbar spine BMD; LDL, low-density lipoprotein; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD.
this analysis, we only included parameters which were significantly correlated to BMD in the univariate analysis. As shown in Table 5, age of menopause was independently and negatively associated to BMD measures in most sites. In the anthropometric parameters, only waist circumference was a significant predictor BMD in the lumbar spine (p < 0.05) and left femoral neck (p < 0.05). Also, trunk fat and total lean mass remain related to BMD, and we observed independent and positive relation between these parameters and BMD in the majority of the studied sites (Table 5).
Discussion In this study, we examined whether or not obesity and fat-related parameters (waist circumference, trunk fat, leg fat, and total body fat) as well as lean mass might influence BMD in a cohort of 81 Tunisian postmenopausal
women. We also evaluated the effect of type 2 diabetes parameters (glycemia, HbA1c, insulin level) especially the insulin resistance on BMD. Our results clearly show that obese patients display higher BMD than non-obese subjects. Furthermore, independently from obesity, we found that patients suffering from insulin resistance present elevated BMD. This incremental effect of obesity and insulin resistance on BMD may partly explain the positive correlations that we found between BMD measures and weight, waist circumference, BMI, glycemia, trunk fat, total body fat, and total lean mass. However, our analysis showed that only the age of menopause and waist circumference are independently and negatively related to BMD, and only trunk fat and lean mass are independently and positively associated to BMD. Thus, of the body composition parameters, trunk fat and lean mass may increase BMD especially in the right femur, total hip, and right femoral neck. However, waist circumference seems to decrease BMD in the lumbar site.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS Association of Obesity and Insulin Resistance With BMD
5
Table 3 Clinical Characteristics According to the Presence or Absence of Insulin Resistance
Age (yr) Years since menopause (yr) Weight (kg) Height (cm) Waist circumference (cm) BMI (kg/m2) Glycemia (mmol/L) HbA1c (%) Serum insulin (µL U/mL) HOMA-IR Total cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglyceride (mmol/L) L-BMD (g/cm2) LF-BMD (g/cm2) RF-BMD (g/cm2) TH-BMD (g/cm2) LFN-BMD (g/cm2) RFN-BMD (g/cm2) WB-BMD (g/cm2) Leg fat mass (g) Trunk fat mass (g) Total body fat mass (g) Total lean mass (g)
Non insulin-resistant (n = 36)
Insulin-resistant (n = 43)
p Value
58.72 ± 5.71 12.05 ± 6.78 78.91 ± 12.36 155.97 ± 7.22 108.47 ± 13.82 32.54 ± 5.35 9.45 ± 4.17 8.02 ± 2.05 4.67 ± 1.92 1.76 ± 0.56 5.12 ± 1.36 1.28 ± 0.28 3.01 ± 0.99 1.66 ± 1.02 1.06 ± 0.15 0.98 ± 0.12 0.97 ± 0.10 0.97 ± 0.11 0.90 ± 0.12 0.91 ± 0.11 1.13 ± 0.08 12040 ± 4504.88 18367.97 ± 3515.27 36464 ± 8727.33 38937.41 ± 4378.74
58.39 ± 6.44 10.09 ± 7.15 83.75 ± 14.37 156.81 ± 5.16 112.62 ± 12.31 34.07 ± 5.50 11.80 ± 3.88 9.17 ± 1.68 12.82 ± 14.02 6.04 ± 5.20 4.78 ± 1.37 1.18 ± 0.29 3.03 ± 1.08 1.55 ± 0.68 1.09 ± 0.17 1.05 ± 0.15 1.02 ± 0.15 1.04 ± 0.15 0.95 ± 0.15 0.95 ± 0.13 1.16 ± 0.10 12114.09 ± 3743.27 19576.39 ± 4515.67 37992.95 ± 8738.93 42483.90 ± 6380.43
0.813 0.227 0.116 0.548 0.161 0.218 0.0110 0.0076 0.0009 <0.0001 0.322 0.175 0.957 0.597 0.434 0.020 0.084 0.0402 0.095 0.120 0.192 0.936 0.195 0.440 0.0060
Data are mean ± SD. Significant differences are shown in bold font. Abbr: BMD, bone mineral density; BMI, body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; HOMAIR, homeostasis model assessment of insulin resistance; L-BMD, lumbar spine BMD; LDL, low-density lipoprotein; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD.
In agreement with our results, other researchers also reported the protective effect of obesity on BMD. Maïmoun et al found that after multivariable adjustments, BMD was higher among obese subjects than those with normalweight at weight-bearing and non–weight-bearing sites in a cohort of French population (23). Also, studies have demonstrated that obesity improves BMD and have observed higher BMD in obese patients than non-obese control subjects (4,5) and a decrease in fracture risk (6). The mechanism for this is that greater weight provides a greater mechanical loading on the bone, prompting the osteocytes to send a signal which increases the bone development or decreases the bone resorption (24). In our study, in addition to body weight and BMI, other body composition parameters were associated to BMD. Indeed, we found that each waist circumference, trunk fat, total body fat, and total lean mass positively correlated with BMD. However, the positive correlation between BMD and waist circumference becomes negative after multiple re-
gression analysis, and only trunk fat and lean mass were independently and positively related to BMD.Thus, our data support that body composition plays an important role in bone health, with evidence for a role of lean and fat mass for high BMD. These results are consistent with those of previous studies in menopausal women. Cui et al examined the relative contribution of body composition to BMD at various sites and reported that fat mass was the only determinant of BMD at the lumbar, distal forearm, and calcaneus sites in postmenopausal women (25). Recently, it was also demonstrated that fat mass seems to be a significant predictor of spine, total hip BMD (12), and proximal femur (26). However, in our current study, we found that trunk fat had significant effect only in the right femur, total hip, and right femoral neck. Thus, our results partly differ from other investigations which recorded a significant positive effect of trunk fat on BMD at the arm, trunk, pelvis (27), and lumbar spine (28). Although these data and our results indicate that fat exerts a protective effect on bone,
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS 6
Cherif et al. Table 4 Univariate Regression Analysis for Association Between Bone Mineral Density and Measured Parameters
Age Years since menopause Weight Height Waist circumference BMI Glycemia HbA1c Serum insulin HOMA-IR Total cholesterol HDL cholesterol LDL cholesterol Triglyceride Leg fat mass Trunk fat mass Total body fat mass Total lean mass
L-BMD
LF-BMD
RF-BMD
TH-BMD
LFN-BMD
RFN-BMD
WB-BMD
−0.172 −0.219 0.205 0.435*** −0.047 −1.5.10−4 0.079 0.034 0.045 0.020 0.063 −0.094 0.134 −0.003 0.049 0.162 0.126 0.282*
−0.352** −0.397** 0.390** 0.298* 0.193 0.251 0.146 0.118 0.149 0.168 −0.026 −0.127 −0.073 0.075 0.097 0.381** 0.292** 0.476***
−0.379** −0.419*** 0.362** 0.225* 0.202 0.259* 0.135 0.103 0.076 0.107 −0.006 −0.124 −0.084 0.119 0.067 0.395** 0.286** 0.444*
−0.369** −0.413*** 0.381*** 0.265* 0.200 0.258* 0.142 0.112 0.115 0.140 −0.016 −0.127 −0.080 0.098 0.083 0.392** 0.293** 0.466***
−0.356** −0.421*** 0.232* 0.382** 0.002 0.051 0.124 0.119 0.056 0.074 0.001 −0.006 −0.101 0.124 0.017 0.243* 0.144 0.339**
−0.404** −0.491*** 0.260* 0.320* 0.073 0.110 0.087 0.035 0.025 0.028 0.039 −0.041 −0.099 0.210 0.009 0.284* 0.171 0.363**
−0.304* −0.226* 0.450*** 0.365** 0.263* 0.282* 0.249* 0.138 0.006 0.059 −0.119 −0.108 −0.056 −0.094 0.143 0.347** 0.302** 0.608***
Data are represented by the Pearson’s correlations coefficients. Abbr: BMD, bone mineral density; BMI, body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; HOMAIR, homeostasis model assessment of insulin resistance; L-BMD, lumbar spine BMD; LDL, low-density lipoprotein; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD. *p < 0.05; **p < 0.001; ***p < 0.0001.
other studies have described a negative relation between fat and bone (29–31). It is probable that body composition plays an important role in bone health. In terms of lean mass, it is prob-
able that if an individual has a higher proportion of lean mass, he or she has a significant load-bearing activity, which stimulates bone remodeling (32). An interesting finding of our study is that lean mass correlated positively with BMD,
Table 5 Multiple Linear Regression Analysis for Association Between Bone Mineral Density and the Measured Parameters
Age Years since menopause Weight Height Waist circumference BMI Glycemia Trunk fat mass Total body fat mass Total lean mass
L-BMD
LF-BMD
RF-BMD
TH-BMD
LFN-BMD
RFN-BMD
WB-BMD
0.145 −0.178 0.281 0.315 −0.415* −0.012 0.034 0.220 −0.088 0.213
0.059 −0.358* 0.853 −0.269 −0.204 −0.651 0.093 0.439 −0.426 0.490
0.011 −0.379* 0.534 −0.235 −0.135 −0.421 0.088 0.561* −0.062 0.615*
0.036 −0.373* 0.707 −0.256 −0.172 −0.546 0.092 0.505* −0.138 0.411*
0.049 −0.364* 0.409 0.046 −0.387* −0.274 0.061 0.482 −0.350 0.306
0.014 −0.438** 0.438 −0.102 −0.197 −0.399 0.024 0.497* −0.395 0.361
0.019 −0.143 0.045 0.132 −0.181 0.054 0.176 0.211 −0.153 0.579*
Data are represented by the standardized coefficient β. Abbr: BMD, bone mineral density; BMI, body mass index; L-BMD, lumbar spine BMD; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD. *p < 0.05; **p < 0.001. Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS Association of Obesity and Insulin Resistance With BMD and this relation remains significant after multiple linear regression in the right femur, total hip, and the whole body. This suggests that the higher amount of lean mass found in obese patients have a protective role for bone. A similar finding indicates that lean mass was positively related to BMD at the femoral neck and lumbar spine and has an important role in good bone health (33–36). Diabetes itself affects bone turnover and increased risk of fracture (37). However, BMD is known to decrease in type 1 diabetics and increase in type 2 diabetics. This incoherence of BMD in diabetics is attributed to hyperinsulinemia (38). In the present study, we found that BMD was higher in patients with insulin resistance than patients without insulin resistance. In addition, our data indicated that the positive effect of insulin resistance on BMD was independent from obesity because BMI and fat mass were similar between the 2 groups. Interestingly, a positive correlation was found between glycemia and the whole body BMD. However, this relation disappears after multiple linear regression. In addition, we did not find any correlation between BMD and HbA1c, hyperinsulinemia, and HOMA-IR. Furthermore, it is important to note that this protective effect of insulin resistance may be related to lean mass which is different between the 2 groups. This result indicated that increased lean mass in type 2 diabetes have a beneficial effect on BMD measurements, especially those from the left femur and total hip. Our finding is consistent with previous studies (14,15) that found an increase in BMD in insulin-resistant subjects. These authors suggested that the presence of insulin resistance may protect, in part, against bone loss due to estrogen deficiency or aging in postmenopausal women, and may contribute to higher BMD by exercising anabolic effects on bone structures especially when the hyperinsulinemia is prominent.Yamaguchi et al reported that insulin resistance may increase BMD, especially at the femoral neck, and reduce the risk of vertebral fractures in patients with type 2 diabetes (39). Despite the significant difference in the BMD values between insulin-resistant and non insulin-resistant women, we have found no correlation of the serum insulin level and HOMAIR with BMD. Contrary to our results, studies have shown a significant positive association between circulating serum levels (40–42), HOMA-IR (43), and BMD. These conflicting results may be due to the difficulty in interpretation of HOMA-IR in subjects on antidiabetic medications. Contrary to these studies and our data, a recent study by Srikanthan et al showed that insulin resistance was associated with lower femoral neck strength (44). These conflicting results may be due to the heterogeneity of type 2 diabetes or the different study methodologies. Although our study provides finding, it has some limitations; that is, the transversal design implicated that the causality of our results cannot be demonstrated, and therefore, further prospective studies are necessary. In particular, the exact mechanism responsible for insulin’s role in bone homeostasis are still unclear. Thus, further experimental studies at the bone tissue and cellular levels are
7
needed. Also, the sample size was not large enough to make definite conclusion. Finally, all participants were Tunisian postmenopausal women, and our results may not be generalized to other age range, genders, and ethnicities. However, our relatively homogeneous study population (in terms of gender, age, age of menopause) may have strengthened our finding because many of these factors are known to profoundly impact the evaluation of body fatness, insulin-resistance, and BMD (45,46). Also, a major advantage of this study is that no participant had taken a drug known to interfere with bone, fat, or lean tissue metabolism. Also, all patients enrolled in this study do not take calcium supplementation or vitamin D, which might influence results. Moreover, our study investigates for the first time the effect of obesity, fat mass, lean mass, and insulin resistance on BMD in a population of Tunisian postmenopausal women. In conclusion, our finding suggests that patients suffering from obesity present elevated BMD in the femur, total hip, and the whole body. We have also observed a higher BMD values in insulin-resistant group than in non–insulinresistant subjects independently from obesity. Positive correlations were found between the anthropometric parameters, body fat, lean mass, glycemia, and BMD, whereas only trunk fat and lean mass seem to be a strong independent predictor of BMD in this population. Further research is required to explore the bone metabolism and the mechanisms underlying these finding. Consequently, it is necessary to analyze bone remodeling markers that could play a role in bone tissue adaptation to obesity and insulin resistance and that may account for the differences in BMD.
Acknowledgments This work was funded by the “Ministry of Higher Education and Scientific Research of Tunisia.” The authors would like to express their gratitude to the patients for their participation.
References 1. Greco EA, Lenzi A, Migliaccio S. 2015 The obesity of bone. Ther Adv Endocrinol Metab 6(6):273–286. 2. Tóth E, Ferenc V, Mészáros S, et al. 2005 [Effects of body mass index on bone mineral density in men]. Orv Hetil 146(28):1489–1493. 3. Zhao L-J, Liu Y-J, Liu P-Y, et al. 2007 Relationship of obesity with osteoporosis. J Clin Endocrinol Metab 92(5):1640– 1646. 4. Beck TJ, Petit MA, Wu G, et al. 2009 Does obesity really make the femur stronger? BMD, geometry, and fracture incidence in the Women’s Health Initiative Observational Study. J Bone Miner Res 24(8):1369–1379. 5. Andersen S, Frederiksen KD, Hansen S, et al. 2014 Bone structure and estimated bone strength in obese patients evaluated by high-resolution peripheral quantitative computed tomography. Calcif Tissue Int 95(1):19–28. 6. Nielson CM, Srikanth P, Orwoll ES. 2012 Obesity and fracture in men and women: an epidemiologic perspective. J Bone Miner Res 27(1):1–10.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS 8 7. Douchi T, Yamamoto S, Oki T, et al. 2000 Difference in the effect of adiposity on bone density between pre- and postmenopausal women. Maturitas 34(3):261–266. 8. Filip R, Raszewski G. 2009 Bone mineral density and bone turnover in relation to serum leptin, alpha-ketoglutarate and sex steroids in overweight and obese postmenopausal women. Clin Endocrinol (Oxf) 70(2):214–220. 9. Reid IR, Ames R, Evans MC, et al. 1992 Determinants of total body and regional bone mineral density in normal postmenopausal women—a key role for fat mass. J Clin Endocrinol Metab 75(1):45–51. 10. Reid IR, Ames RW, Evans MC, et al. 1994 Determinants of the rate of bone loss in normal postmenopausal women. J Clin Endocrinol Metab 79(4):950–954. 11. Gnudi S, Sitta E, Fiumi N. 2007 Relationship between body composition and bone mineral density in women with and without osteoporosis: relative contribution of lean and fat mass. J Bone Miner Metab 25(5):326–332. 12. Zhu K, Hunter M, James A, et al. 2015 Associations between body mass index, lean and fat body mass and bone mineral density in middle-aged Australians: the Busselton Healthy Ageing Study. Bone 74:146–152. 13. Greco EA, Francomano D, Fornari R, et al. 2013 Negative association between trunk fat, insulin resistance and skeleton in obese women. World J Diabetes 4(2):31–39. 14. Melton LJ, Riggs BL, Leibson CL, et al. 2008 A bone structural basis for fracture risk in diabetes. J Clin Endocrinol Metab 93(12):4804–4809. 15. Ma L, Oei L, Jiang L, et al. 2012 Association between bone mineral density and type 2 diabetes mellitus: a meta-analysis of observational studies. Eur J Epidemiol 27(5):319–332. 16. Arikan S, Tuzcu A, Bahceci M, et al. 2012 Insulin resistance in type 2 diabetes mellitus may be related to bone mineral density. J Clin Densitom 15(2):186–190. 17. Shin D, Kim S, Kim KH, et al. 2014 Association between insulin resistance and bone mass in men. J Clin Endocrinol Metab 99(3):988–995. 18. Belfki H, Ben Ali S, Aounallah-Skhiri H, et al. 2013 Prevalence and determinants of the metabolic syndrome among Tunisian adults: results of the Transition and Health Impact in North Africa (TAHINA) project. Public Health Nutr 16:582– 590. 19. Sahli H, Testouri N, Chihaoui MB, et al. 2009 Bone mineral density in healthy Tunisian women. Maturitas 63:227–232. 20. WHO. 2017 Obesity and overweight [Internet]. WHO. Available at: http://www.who.int/mediacentre/factsheets/fs311/en/. Accessed: September 21, 2016. 21. Friedewald WT, Levy RI, Fredrickson DS. 1972 Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 18(6):499–502. 22. Matthews DR, Hosker JP, Rudenski AS, et al. 1985 Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28(7):412–419. 23. Maïmoun L, Mura T, Leprieur E, et al. 2016 Impact of obesity on bone mass throughout adult life: influence of gender and severity of obesity. Bone 90:23–30. 24. Seeman E, Delmas PD. 2006 Bone quality—the material and structural basis of bone strength and fragility. N Engl J Med 354(21):2250–2261. 25. Cui L-H, Shin M-H, Kweon S-S, et al. 2007 Relative contribution of body composition to bone mineral density at different sites in men and women of South Korea. J Bone Miner Metab 25(3):165–171.
Cherif et al. 26. Nur H, Toraman NF, Arica Z, et al. 2013 The relationship between body composition and bone mineral density in postmenopausal Turkish women. Rheumatol Int 33(3):607– 612. 27. Zhang J, Jin Y, Xu S, et al. 2015 Associations of fat mass and fat distribution with bone mineral density in Chinese obese population. J Clin Densitom 18(1):44–49. 28. Saarelainen J, Honkanen R, Kröger H, et al. 2011 Body fat distribution is associated with lumbar spine bone density independently of body weight in postmenopausal women. Maturitas 69(1):86–90. 29. Fu X, Ma X, Lu H, et al. 2011 Associations of fat mass and fat distribution with bone mineral density in pre- and postmenopausal Chinese women. Osteoporos Int 22(1):113– 119. 30. Blaauw R, Albertse EC, Hough S. 1996 Body fat distribution as a risk factor for osteoporosis. S Afr Med J 86(9):1081– 1084. 31. Kim K-C, Shin D-H, Lee S-Y, et al. 2010 Relation between obesity and bone mineral density and vertebral fractures in Korean postmenopausal women. Yonsei Med J 51(6):857– 863. 32. Weeks BK, Beck BR. 2008 The BPAQ: a bone-specific physical activity assessment instrument. Osteoporos Int 19(11):1567– 1577. 33. Harris S, Dallal GE, Dawson-Hughes B. 1992 Influence of body weight on rates of change in bone density of the spine, hip, and radius in postmenopausal women. Calcif Tissue Int 50(1):19–23. 34. Casale M, von Hurst PR, Beck KL, et al. 2016 Lean mass and body fat percentage are contradictory predictors of bone mineral density in pre-menopausal Pacific Island women. Nutrients 8(8). 35. Kang EK, Park HW, Baek S, Lim JY. 2016 The association between trunk body composition and spinal bone mineral density in Korean males versus females: a farmers’ cohort for agricultural work-related musculoskeletal disorders (FARM) study. J Korean Med Sci 31(10):1595–1603. 36. Sinaki M, Itoi E, Wahner HW, et al. 2002 Stronger back muscles reduce the incidence of vertebral fractures: a prospective 10 year follow-up of postmenopausal women. Bone 30(6):836–841. 37. Vestergaard P. 2007 Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2 diabetes—a meta-analysis. Osteoporos Int 18(4):427–444. 38. Thrailkill KM, Lumpkin CK, Bunn RC, et al. 2005 Is insulin an anabolic agent in bone? Dissecting the diabetic bone for clues. Am J Physiol Endocrinol Metab 289(5):E735–E745. 39. Yamaguchi T, Kanazawa I, Yamamoto M, et al. 2009 Associations between components of the metabolic syndrome versus bone mineral density and vertebral fractures in patients with type 2 diabetes. Bone 45(2):174–179. 40. Haffner SM, Bauer RL. 1993 The association of obesity and glucose and insulin concentrations with bone density in premenopausal and postmenopausal women. Metab Clin Exp 42(6):735–738. 41. Barrett-Connor E, Kritz-Silverstein D. 1996 Does hyperinsulinemia preserve bone? Diabetes Care 19(12):1388– 1392. 42. Stolk RP, Van Daele PL, Pols HA, et al. 1996 Hyperinsulinemia and bone mineral density in an elderly population: the Rotterdam Study. Bone 18(6):545–549. 43. Dennison EM, Syddall HE, Aihie Sayer A, et al. 2004 Type 2 diabetes mellitus is associated with increased axial bone density in men and women from the Hertfordshire Cohort
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS Association of Obesity and Insulin Resistance With BMD Study: evidence for an indirect effect of insulin resistance? Diabetologia 47(11):1963–1968. 44. Srikanthan P, Crandall CJ, Miller-Martinez D, et al. 2014 Insulin resistance and bone strength: findings from the study of midlife in the United States. J Bone Miner Res 29(4):796– 803.
9
45. Reid IR. 2010 Fat and bone. Arch Biochem Biophys 503(1):20– 27. 46. Dimitri P, Bishop N, Walsh JS, Eastell R. 2012 Obesity is a risk factor for fracture in children but is protective against fracture in adults: a paradox. Bone 50(2):457– 466.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
Original Article
Positive Association of Obesity and Insulin Resistance With Bone Mineral Density in Tunisian Postmenopausal Women Rim Cherif,*,1,2 Feten Mahjoub,3 Hela Sahli,4 Elhem Cheour,4 Laurence Vico,2 Mohsen Sakly,1 and Nebil Attia1 1
UR11ES33, Research Unit ’Integrated Physiology, Laboratory of Biochemistry-Human Nutrition, Faculty of Sciences of Bizerte, Carthage University, Tunis, Tunisia; 2INSERM U1059, Integrative Biology of Bone Tissue Laboratory, Lyon University, St-Etienne, France; 3Department of Diabetology, National Institute of Nutrition, Tunis, Tunisia; and 4ImmunoRheumatology Laboratory, Rabta Hospital, Faculty of Medicine of Tunis, Tunis El Manar University, Tunis, Tunisia
Abstract The association of bone mineral density (BMD) with obesity and insulin resistance remains unclear. This study aimed to explore these associations in Tunisian menopausal women. Eighty-one postmenopausal women were recruited. Data were analyzed for obese (N = 57) and non-obese women (N = 24) and for insulinresistant (N = 43) and non insulin-resistant women (N = 36). Anthropometric and biochemical parameters were recorded. BMD in different sites and body composition were measured using dual-energy X-ray absorptiometry. Higher BMD was observed in obese women than those non-obese in the left femur (p = 0.0067), right femur (p = 0.0108), total hip (p = 0.0077), and the whole body (p = 0.0276). Also BMD was significantly greater in insulin-resistant women than in non–insulin-resistant women when measured in the left femur and total hip. Positive correlations were recorded between BMD and anthropometric parameters, body composition parameters, and glycemia (r = 0.249, p < 0.05). Multiple linear regression analysis shows that only trunk fat (p < 0.05) and lean mass (p < 0.05) were independently and positively related to BMD, and the waist circumference was the only anthropometric parameter independently and negatively associated to BMD. BMD is improved in obese and insulin-resistant women. Also, trunk fat and lean mass are likely to be key positive independent factors for BMD. Key Words: Bone mineral density; fat; insulin resistance; lean mass; obesity.
Introduction
studies have demonstrated that obese subjects present higher BMD than normal-weight subjects (4,5) and increased body weight or body mass index (BMI) is related to higher BMD and reduced fracture risk (6). In postmenopausal women, obesity has been considered a protective factor for bone loss and osteoporosis, likely for increases in the number of adipocytes, which are important sources of estrogen derived from aromatization, resulting in increased BMD (7,8). Although there is a little doubt that body weight has an effect on bone, whether it is the effect of fat mass or lean mass that influences BMD is disputed. Studies in postmenopausal women have suggested that fat mass plays a key role (9,10), and both fat and lean mass have been reported to be significant predictors of BMD (11,12). However, there is also
Osteoporosis is a major public health problem characterized by a decrease in bone mineral density (BMD) and an alteration of bone quality (1). Obesity, another common disease, has been demonstrated to be closely related with osteoporosis (2,3). Despite their sedentary lifestyle, several
Received 10/22/16; Revised 05/22/17; Accepted 05/23/17. Disclosure of interest: The authors report no conflicts of interests in this work. *Address correspondence to: Rim Cherif, PhD, UR11ES33 Integrated Physiology, Laboratory of Biochemistry and Human Nutrition, Faculty of Sciences of Bizerte, Carthage University, Jarzouna, 7021 Bizerte, Tunisia. E-mail: [email protected]
1
ARTICLE IN PRESS 2 disagreement in the literature on the positive effect of lean and fat mass on BMD (10,13). In addition to obesity, there have been conflicting reports about the relation between insulin-resistance and BMD. Some authors have reported elevated BMD in the presence of insulin resistance (14,15), whereas some have reported decreased BMD (16,17). Because of the discrepancies in the results, the association between insulin resistance and bone mass remains unclear, and further studies are needed to explain this relationship. Prevalence of osteoporosis and metabolic syndrome greatly increases in the Tunisian population. It was estimated that 31.6% of women develop the metabolic syndrome (18) and 23.4% are osteoporotic (19). Based on the conflicting studies results and the high prevalence of these diseases in Tunisia, we evaluate the relationship between BMD and obesity at different bone sites in a cohort of Tunisian postmenopausal women. We also investigate other factors related to body composition (leg fat, trunk fat, total fat, lean mass) in order to identify those potentially associated with BMD. We also sought to determine the association between insulin resistance and bone mass in normal-weight women.
Patients and Methods Patients A total of 81 postmenopausal women (age range: 50– 83 years) were recruited consecutively between June 2014 and April 2015 from the Endocrinology Department of the National Institute of Nutrition of Tunis, Tunisia, during visits for routine checkups. The study was approved by the local ethic committee of National Institute of Nutrition of Tunis and written or oral consent was obtained from all the patients before the study. Written consents were not possible in all cases because the majority of eligible subjects in our study were unable to read or write. The consent includes the agreement of patients to participate in the clinical study and to undergo free densitometry examination. Original inclusion criteria were age >50, postmenopausal (defined as cessation of menstruation for at least 1 year), and no previous osteoporotic fracture or known diagnosis of osteoporosis. Women with liver or renal disease, inflammatory disease, parathyroid and thyroid, chronic inflammatory rheumatism, early natural menopause before age 40, and receiving medicine known to influence bone metabolism, such as corticosteroids, heparin, anticonvulsants, vitamin D or calcium supplementations, and bisphosphonates, were not included. Participants were questioned about their age, age at the onset of menopause, education, occupation, family history of osteoporosis, history of peripheral traumatic fractures, weight history, smoking habits, level of physical activity, and current medication use. Smoking habit was categorized as non-smoker and current smoker. Physical activity was selfgraded by participants and categorized as sedentary, moderate, and important.
Cherif et al. Anthropometric Measurements Body weight was measured to the nearest 0.1 kg and height was determined to the nearest centimeter. BMI was calculated as the weight (kg) divided by the square of height (m2). Patients were considered as non-obese if BMI < 30 kg/ m2 and as obese if BMI ≥ 30 kg/m2 according to the World Health Organization (WHO) definition (20). Waist circumference was measured at the narrowest part of the abdomen, that is, at the natural indentation between the 10th rib and the iliac crest (minimum waist).
Biochemical Analysis Fasting blood samples were taken after fasting for at least 12 h. Venous blood samples were taken from an antecubital vein and placed into heparinized or non-heparinized tubes. Tubes were centrifuged at 3000 × g for 10 min. Serum fasting blood glucose, glycated hemoglobin (HbA1c), lipid levels (total cholesterol, triglycerides, and high-density lipoprotein cholesterol) were determined by well-validated laboratory routine methods. Serum lowdensity lipoprotein cholesterol values were estimated using the Friedwald formula (21). Fasting blood samples were collected in non-heparinized tube for the serum insulin concentrations measurement.The tube was left for 30 min at room temperature (to allow clot formation) then centrifuged at 1000 × g for 15 min. The serum was separated and stored in microtubes of 250 µL at 80°C until the day of manipulation. After adequate thawing, insulin was measured by enzyme-linked immunosorbent assay (ALPCO kit, Salem, NH). Insulin resistance was measured according to the following formula: HOMA-IR (homeostasis model assessment of insulin resistance) = fasting insulin (µU/mL) × fasting glucose (mmol/ L)/22.5 (22). Patients were considered as insulin resistant when HOMA-IR ≥ 2.6 and non-insulin resistant when HOMA-IR < 2.6.
Dual-energy X-ray Absorptiometry (DXA) Measurements Densitometry examination of patients was realized at the Rheumatology Department of the Rabta Hospital of Tunis. Lumbar (anteroposterior projection at L1–L4), femur, femoral neck, total hip, and whole body BMD as well as fat (leg, trunk, and total) and total lean mass were measured by DXA using GE-Lunar PRODIGYTM device. Quality control procedures were carried out in accordance with the manufacturer’s recommendations. Instrument variation was determined by a daily calibration procedure using a phantom supplied by manufacturer. The precision error was <2.0% for each measured sites at standard speed based on repeated scans in a random sample of 30 subjects. BMD values were expressed in grams per square centimeter (g/cm2), and fat and lean mass values were expressed in grams.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS Association of Obesity and Insulin Resistance With BMD Statistical Analysis The characteristics of the patients are described with means ± standard deviations (SD). Differences between groups were compared by an independent 2-sample t test. Pearson’s correlations were performed to examine the association between bone-related parameters and biochemical and anthropometric parameters. Multiple linear regression analysis was used to reveal statistically independent relationships between variables. In all cases, the level of statistical significance was set at p < 0.05. All statistics were performed using the StatView package (Version 5, SAS Institute, Inc., Cary, NC).
Results The clinical and laboratory characteristics of the 81 postmenopausal women are listed in Table 1. The mean age of Table 1 Descriptive Characteristics of the Study Population Characteristics Age (yr) Years since menopause (yr) Weight (kg) Height (cm) Waist circumference (cm) BMI (kg/m2) Glycemia (mmol/L) HbA1c (%) Serum insulin (µL U/mL) HOMA-IR Total cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglycerides (mmol/L) L-BMD (g/cm2) LF-BMD (g/cm2) RF-BMD (g/cm2) TH-BMD (g/cm2) LFN-BMD (g/cm2) RFN-BMD (g/cm2) WB-BMD (g/cm2) Leg fat mass (g) Trunk fat mass (g) Total body fat mass (g) Total lean mass (g)
N = 81 58.40 ± 6.08 10.79 ± 7.02 81.54 ± 13.49 156.45 ± 6.10 110.72 ± 12.94 33.36 ± 5.42 10.89 ± 4.23 8.70 ± 1.94 9.10 ± 11.14 4.09 ± 4.39 5.00 ± 1.40 1.23 ± 0.28 3.04 ± 1.03 1.62 ± 0.88 1.08 ± 0.16 1.02 ± 0.14 1.00 ± 0.13 1.01 ± 0.13 0.93 ± 0.14 0.93 ± 0.12 1.15 ± 0.09 12084.77 ± 4040.97 19052.77 ± 4062.34 37326.48 ± 8615.18 40805.76 ± 5766.78
Data are mean ± SD. Abbr: BMD, bone mineral density; BMI, body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; HOMA-IR, homeostasis model assessment of insulin resistance; L-BMD, lumbar spine BMD; LDL, low-density lipoprotein; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD.
3
the participants was 58.40 ± 6.08 years and the years since menopause was 10.79 ± 7.02. The mean weight (±SD) and BMI was 81.54 ± 13.49 kg and 33.36 ± 5.42 kg/m2, respectively. In this population, the average of glycemia was 10.89 ± 4.23 mmol/L and HOMA-IR index was 4.09 ± 4.39. The BMD and body composition at all the studied sites are shown in detail in Table 1. In order to analyze the effect of obesity on BMD, participants were classified according to their BMI as obese (N = 57) and non-obese (N = 24). Mean weight and BMI were 68.04 ± 6.83 kg and 27.67 ± 2.12 kg/m2, respectively, in non-obese women group and 87.22 ± 11.39 kg (p < 0.0001) and 35.75 ± 4.53 kg/m2 (p < 0.0001), respectively, in obese women group. Comparison between obese and nonobese patients showed that there was no difference between the groups for age, age of menopause, glycemia, and HOMA-IR index. However, we found that obese women had significantly greater BMD at the left femur (p = 0.006), right femur (p = 0.010), total hip (p = 0.007), and the whole body (p = 0.027) compared with non-obese women. Also, there were statistically significant differences between groups in the body composition. In fact, leg, trunk, and total body fat mass was higher in obese women than in non-obese women group (p < 0.0001 for each parameters) as well as total lean mass (p < 0.0001) (Table 2). To investigate the effect of insulin resistance on BMD, the studied population was divided into insulin-resistant (N = 43) and non insulin-resistant (N = 36) groups according to the HOMA-IR index. The weight, waist circumference, and BMI were not significantly different between groups. The mean glycemia and serum insulin were 11.80 ± 3.88 mmol/L and 12.82 ± 14.02 µl U/mL, respectively, in insulin-resistant group and 9.45 ± 4.17 mmol/L (p = 0.011) and 4.67 ± 1.92 µL U/mL (p = 0.0009) in non insulin-resistant group. The HOMA-IR index was highly different between the 2 groups (p < 0.0001). Statistically significant differences between groups in the DXA measurements were observed in only 2 sites: left femur (p = 0.02) and total hip (p = 0.04), with greater BMD values in the insulin-resistant group in both sites (Table 3). Univariate regression analysis was evaluated for the whole population (Table 4). This analysis revealed that both age and age of menopause were significantly and negatively related to all measured sites BMD except for lumbar BMD. However, body weight was positively related to femur, total hip, and the whole body BMD. A positive correlation was noted between BMI and right femur BMD (r = 0.259, p < 0.05), total hip BMD (r = 0.258, p < 0.05), and whole body BMD (r = 0.282, p < 0.05). Glycemia was positively correlated to the whole body BMD (r = 0.249, p < 0.05) but not to other sites. We did not find any relation between BMD measures and lipid variables at any sites. For body composition measurements, trunk fat, total body fat, and total lean mass correlate positively and significantly to BMD measurements (Table 4). Multiple linear regression analysis was performed to evaluate independent predictor of BMD measurements. In
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS 4
Cherif et al. Table 2 Clinical Characteristics According to the Presence or Absence of Obesity
Age (yr) Years since menopause (yr) Weight (kg) Height (cm) Waist circumference (cm) BMI (kg/m2) Glycemia (mmol/L) HbA1c (%) Serum insulin (µL U/mL) HOMA-IR Total cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglyceride (mmol/L) L-BMD (g/cm2) LF-BMD (g/cm2) RF-BMD (g/cm2) TH-BMD (g/cm2) LFN-BMD (g/cm2) RFN-BMD (g/cm2) WB-BMD (g/cm2) Leg fat mass (g) Trunk fat mass (g) Total body fat mass (g) Total lean mass (g)
Non obese women (n = 24)
Obese women (n = 57)
p Value
58.70 ± 5.46 11.58 ± 6.93 68.04 ± 6.83 156.79 ± 6.91 100.20 ± 7.55 27.67 ± 2.12 11.96 ± 4.60 9.04 ± 2.10 6.43 ± 4.52 3.14 ± 1.97 5.67 ± 1.65 1.25 ± 0.18 3.54 ± 1.13 1.82 ± 1.31 1.05 ± 0.15 0.96 ± 0.13 0.94 ± 0.12 0.95 ± 0.13 0.91 ± 0.12 0.90 ± 0.12 1.11 ± 0.09 9074.04 ± 2125.98 15205.70 ± 3129.60 28899.75 ± 4297.60 36587.37 ± 2820.78
58.28 ± 6.37 10.44 ± 7.09 87.22 ± 11.39 156.31 ± 5.79 115.15 ± 12.18 35.75 ± 4.53 10.44 ± 4.03 8.56 ± 1.87 10.20 ± 12.79 4.49 ± 5.03 4.70 ± 1.17 1.22 ± 0.32 2.86 ± 0.94 1.54 ± 0.64 1.09 ± 0.16 1.05 ± 0.13 1.02 ± 0.13 1.04 ± 0.13 0.94 ± 0.15 0.95 ± 0.12 1.17 ± 0.09 13352.45 ± 3993.87 20672.59 ± 3246.38 40874.57 ± 7429.55 42581.93 ± 5779.51
0.774 0.512 <0.0001 0.751 <0.0001 <0.0001 0.140 0.317 0.173 0.216 0.009 0.736 0.028 0.244 0.291 0.0067 0.0108 0.0077 0.301 0.103 0.0276 <0.0001 <0.0001 <0.0001 <0.0001
Data are mean ± SD. Significant differences are shown in bold font. Abbr: BMD, bone mineral density; BMI, body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; HOMAIR, homeostasis model assessment of insulin resistance; L-BMD, lumbar spine BMD; LDL, low-density lipoprotein; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD.
this analysis, we only included parameters which were significantly correlated to BMD in the univariate analysis. As shown in Table 5, age of menopause was independently and negatively associated to BMD measures in most sites. In the anthropometric parameters, only waist circumference was a significant predictor BMD in the lumbar spine (p < 0.05) and left femoral neck (p < 0.05). Also, trunk fat and total lean mass remain related to BMD, and we observed independent and positive relation between these parameters and BMD in the majority of the studied sites (Table 5).
Discussion In this study, we examined whether or not obesity and fat-related parameters (waist circumference, trunk fat, leg fat, and total body fat) as well as lean mass might influence BMD in a cohort of 81 Tunisian postmenopausal
women. We also evaluated the effect of type 2 diabetes parameters (glycemia, HbA1c, insulin level) especially the insulin resistance on BMD. Our results clearly show that obese patients display higher BMD than non-obese subjects. Furthermore, independently from obesity, we found that patients suffering from insulin resistance present elevated BMD. This incremental effect of obesity and insulin resistance on BMD may partly explain the positive correlations that we found between BMD measures and weight, waist circumference, BMI, glycemia, trunk fat, total body fat, and total lean mass. However, our analysis showed that only the age of menopause and waist circumference are independently and negatively related to BMD, and only trunk fat and lean mass are independently and positively associated to BMD. Thus, of the body composition parameters, trunk fat and lean mass may increase BMD especially in the right femur, total hip, and right femoral neck. However, waist circumference seems to decrease BMD in the lumbar site.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS Association of Obesity and Insulin Resistance With BMD
5
Table 3 Clinical Characteristics According to the Presence or Absence of Insulin Resistance
Age (yr) Years since menopause (yr) Weight (kg) Height (cm) Waist circumference (cm) BMI (kg/m2) Glycemia (mmol/L) HbA1c (%) Serum insulin (µL U/mL) HOMA-IR Total cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglyceride (mmol/L) L-BMD (g/cm2) LF-BMD (g/cm2) RF-BMD (g/cm2) TH-BMD (g/cm2) LFN-BMD (g/cm2) RFN-BMD (g/cm2) WB-BMD (g/cm2) Leg fat mass (g) Trunk fat mass (g) Total body fat mass (g) Total lean mass (g)
Non insulin-resistant (n = 36)
Insulin-resistant (n = 43)
p Value
58.72 ± 5.71 12.05 ± 6.78 78.91 ± 12.36 155.97 ± 7.22 108.47 ± 13.82 32.54 ± 5.35 9.45 ± 4.17 8.02 ± 2.05 4.67 ± 1.92 1.76 ± 0.56 5.12 ± 1.36 1.28 ± 0.28 3.01 ± 0.99 1.66 ± 1.02 1.06 ± 0.15 0.98 ± 0.12 0.97 ± 0.10 0.97 ± 0.11 0.90 ± 0.12 0.91 ± 0.11 1.13 ± 0.08 12040 ± 4504.88 18367.97 ± 3515.27 36464 ± 8727.33 38937.41 ± 4378.74
58.39 ± 6.44 10.09 ± 7.15 83.75 ± 14.37 156.81 ± 5.16 112.62 ± 12.31 34.07 ± 5.50 11.80 ± 3.88 9.17 ± 1.68 12.82 ± 14.02 6.04 ± 5.20 4.78 ± 1.37 1.18 ± 0.29 3.03 ± 1.08 1.55 ± 0.68 1.09 ± 0.17 1.05 ± 0.15 1.02 ± 0.15 1.04 ± 0.15 0.95 ± 0.15 0.95 ± 0.13 1.16 ± 0.10 12114.09 ± 3743.27 19576.39 ± 4515.67 37992.95 ± 8738.93 42483.90 ± 6380.43
0.813 0.227 0.116 0.548 0.161 0.218 0.0110 0.0076 0.0009 <0.0001 0.322 0.175 0.957 0.597 0.434 0.020 0.084 0.0402 0.095 0.120 0.192 0.936 0.195 0.440 0.0060
Data are mean ± SD. Significant differences are shown in bold font. Abbr: BMD, bone mineral density; BMI, body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; HOMAIR, homeostasis model assessment of insulin resistance; L-BMD, lumbar spine BMD; LDL, low-density lipoprotein; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD.
In agreement with our results, other researchers also reported the protective effect of obesity on BMD. Maïmoun et al found that after multivariable adjustments, BMD was higher among obese subjects than those with normalweight at weight-bearing and non–weight-bearing sites in a cohort of French population (23). Also, studies have demonstrated that obesity improves BMD and have observed higher BMD in obese patients than non-obese control subjects (4,5) and a decrease in fracture risk (6). The mechanism for this is that greater weight provides a greater mechanical loading on the bone, prompting the osteocytes to send a signal which increases the bone development or decreases the bone resorption (24). In our study, in addition to body weight and BMI, other body composition parameters were associated to BMD. Indeed, we found that each waist circumference, trunk fat, total body fat, and total lean mass positively correlated with BMD. However, the positive correlation between BMD and waist circumference becomes negative after multiple re-
gression analysis, and only trunk fat and lean mass were independently and positively related to BMD.Thus, our data support that body composition plays an important role in bone health, with evidence for a role of lean and fat mass for high BMD. These results are consistent with those of previous studies in menopausal women. Cui et al examined the relative contribution of body composition to BMD at various sites and reported that fat mass was the only determinant of BMD at the lumbar, distal forearm, and calcaneus sites in postmenopausal women (25). Recently, it was also demonstrated that fat mass seems to be a significant predictor of spine, total hip BMD (12), and proximal femur (26). However, in our current study, we found that trunk fat had significant effect only in the right femur, total hip, and right femoral neck. Thus, our results partly differ from other investigations which recorded a significant positive effect of trunk fat on BMD at the arm, trunk, pelvis (27), and lumbar spine (28). Although these data and our results indicate that fat exerts a protective effect on bone,
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS 6
Cherif et al. Table 4 Univariate Regression Analysis for Association Between Bone Mineral Density and Measured Parameters
Age Years since menopause Weight Height Waist circumference BMI Glycemia HbA1c Serum insulin HOMA-IR Total cholesterol HDL cholesterol LDL cholesterol Triglyceride Leg fat mass Trunk fat mass Total body fat mass Total lean mass
L-BMD
LF-BMD
RF-BMD
TH-BMD
LFN-BMD
RFN-BMD
WB-BMD
−0.172 −0.219 0.205 0.435*** −0.047 −1.5.10−4 0.079 0.034 0.045 0.020 0.063 −0.094 0.134 −0.003 0.049 0.162 0.126 0.282*
−0.352** −0.397** 0.390** 0.298* 0.193 0.251 0.146 0.118 0.149 0.168 −0.026 −0.127 −0.073 0.075 0.097 0.381** 0.292** 0.476***
−0.379** −0.419*** 0.362** 0.225* 0.202 0.259* 0.135 0.103 0.076 0.107 −0.006 −0.124 −0.084 0.119 0.067 0.395** 0.286** 0.444*
−0.369** −0.413*** 0.381*** 0.265* 0.200 0.258* 0.142 0.112 0.115 0.140 −0.016 −0.127 −0.080 0.098 0.083 0.392** 0.293** 0.466***
−0.356** −0.421*** 0.232* 0.382** 0.002 0.051 0.124 0.119 0.056 0.074 0.001 −0.006 −0.101 0.124 0.017 0.243* 0.144 0.339**
−0.404** −0.491*** 0.260* 0.320* 0.073 0.110 0.087 0.035 0.025 0.028 0.039 −0.041 −0.099 0.210 0.009 0.284* 0.171 0.363**
−0.304* −0.226* 0.450*** 0.365** 0.263* 0.282* 0.249* 0.138 0.006 0.059 −0.119 −0.108 −0.056 −0.094 0.143 0.347** 0.302** 0.608***
Data are represented by the Pearson’s correlations coefficients. Abbr: BMD, bone mineral density; BMI, body mass index; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; HOMAIR, homeostasis model assessment of insulin resistance; L-BMD, lumbar spine BMD; LDL, low-density lipoprotein; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD. *p < 0.05; **p < 0.001; ***p < 0.0001.
other studies have described a negative relation between fat and bone (29–31). It is probable that body composition plays an important role in bone health. In terms of lean mass, it is prob-
able that if an individual has a higher proportion of lean mass, he or she has a significant load-bearing activity, which stimulates bone remodeling (32). An interesting finding of our study is that lean mass correlated positively with BMD,
Table 5 Multiple Linear Regression Analysis for Association Between Bone Mineral Density and the Measured Parameters
Age Years since menopause Weight Height Waist circumference BMI Glycemia Trunk fat mass Total body fat mass Total lean mass
L-BMD
LF-BMD
RF-BMD
TH-BMD
LFN-BMD
RFN-BMD
WB-BMD
0.145 −0.178 0.281 0.315 −0.415* −0.012 0.034 0.220 −0.088 0.213
0.059 −0.358* 0.853 −0.269 −0.204 −0.651 0.093 0.439 −0.426 0.490
0.011 −0.379* 0.534 −0.235 −0.135 −0.421 0.088 0.561* −0.062 0.615*
0.036 −0.373* 0.707 −0.256 −0.172 −0.546 0.092 0.505* −0.138 0.411*
0.049 −0.364* 0.409 0.046 −0.387* −0.274 0.061 0.482 −0.350 0.306
0.014 −0.438** 0.438 −0.102 −0.197 −0.399 0.024 0.497* −0.395 0.361
0.019 −0.143 0.045 0.132 −0.181 0.054 0.176 0.211 −0.153 0.579*
Data are represented by the standardized coefficient β. Abbr: BMD, bone mineral density; BMI, body mass index; L-BMD, lumbar spine BMD; LF-BMD, left femur BMD; LFN-BMD, left femoral neck BMD; RF-BMD, right femur BMD; RFN-BMD, right femoral neck BMD; TH-BMD, total hip BMD; WB-BMD, whole body BMD. *p < 0.05; **p < 0.001. Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS Association of Obesity and Insulin Resistance With BMD and this relation remains significant after multiple linear regression in the right femur, total hip, and the whole body. This suggests that the higher amount of lean mass found in obese patients have a protective role for bone. A similar finding indicates that lean mass was positively related to BMD at the femoral neck and lumbar spine and has an important role in good bone health (33–36). Diabetes itself affects bone turnover and increased risk of fracture (37). However, BMD is known to decrease in type 1 diabetics and increase in type 2 diabetics. This incoherence of BMD in diabetics is attributed to hyperinsulinemia (38). In the present study, we found that BMD was higher in patients with insulin resistance than patients without insulin resistance. In addition, our data indicated that the positive effect of insulin resistance on BMD was independent from obesity because BMI and fat mass were similar between the 2 groups. Interestingly, a positive correlation was found between glycemia and the whole body BMD. However, this relation disappears after multiple linear regression. In addition, we did not find any correlation between BMD and HbA1c, hyperinsulinemia, and HOMA-IR. Furthermore, it is important to note that this protective effect of insulin resistance may be related to lean mass which is different between the 2 groups. This result indicated that increased lean mass in type 2 diabetes have a beneficial effect on BMD measurements, especially those from the left femur and total hip. Our finding is consistent with previous studies (14,15) that found an increase in BMD in insulin-resistant subjects. These authors suggested that the presence of insulin resistance may protect, in part, against bone loss due to estrogen deficiency or aging in postmenopausal women, and may contribute to higher BMD by exercising anabolic effects on bone structures especially when the hyperinsulinemia is prominent.Yamaguchi et al reported that insulin resistance may increase BMD, especially at the femoral neck, and reduce the risk of vertebral fractures in patients with type 2 diabetes (39). Despite the significant difference in the BMD values between insulin-resistant and non insulin-resistant women, we have found no correlation of the serum insulin level and HOMAIR with BMD. Contrary to our results, studies have shown a significant positive association between circulating serum levels (40–42), HOMA-IR (43), and BMD. These conflicting results may be due to the difficulty in interpretation of HOMA-IR in subjects on antidiabetic medications. Contrary to these studies and our data, a recent study by Srikanthan et al showed that insulin resistance was associated with lower femoral neck strength (44). These conflicting results may be due to the heterogeneity of type 2 diabetes or the different study methodologies. Although our study provides finding, it has some limitations; that is, the transversal design implicated that the causality of our results cannot be demonstrated, and therefore, further prospective studies are necessary. In particular, the exact mechanism responsible for insulin’s role in bone homeostasis are still unclear. Thus, further experimental studies at the bone tissue and cellular levels are
7
needed. Also, the sample size was not large enough to make definite conclusion. Finally, all participants were Tunisian postmenopausal women, and our results may not be generalized to other age range, genders, and ethnicities. However, our relatively homogeneous study population (in terms of gender, age, age of menopause) may have strengthened our finding because many of these factors are known to profoundly impact the evaluation of body fatness, insulin-resistance, and BMD (45,46). Also, a major advantage of this study is that no participant had taken a drug known to interfere with bone, fat, or lean tissue metabolism. Also, all patients enrolled in this study do not take calcium supplementation or vitamin D, which might influence results. Moreover, our study investigates for the first time the effect of obesity, fat mass, lean mass, and insulin resistance on BMD in a population of Tunisian postmenopausal women. In conclusion, our finding suggests that patients suffering from obesity present elevated BMD in the femur, total hip, and the whole body. We have also observed a higher BMD values in insulin-resistant group than in non–insulinresistant subjects independently from obesity. Positive correlations were found between the anthropometric parameters, body fat, lean mass, glycemia, and BMD, whereas only trunk fat and lean mass seem to be a strong independent predictor of BMD in this population. Further research is required to explore the bone metabolism and the mechanisms underlying these finding. Consequently, it is necessary to analyze bone remodeling markers that could play a role in bone tissue adaptation to obesity and insulin resistance and that may account for the differences in BMD.
Acknowledgments This work was funded by the “Ministry of Higher Education and Scientific Research of Tunisia.” The authors would like to express their gratitude to the patients for their participation.
References 1. Greco EA, Lenzi A, Migliaccio S. 2015 The obesity of bone. Ther Adv Endocrinol Metab 6(6):273–286. 2. Tóth E, Ferenc V, Mészáros S, et al. 2005 [Effects of body mass index on bone mineral density in men]. Orv Hetil 146(28):1489–1493. 3. Zhao L-J, Liu Y-J, Liu P-Y, et al. 2007 Relationship of obesity with osteoporosis. J Clin Endocrinol Metab 92(5):1640– 1646. 4. Beck TJ, Petit MA, Wu G, et al. 2009 Does obesity really make the femur stronger? BMD, geometry, and fracture incidence in the Women’s Health Initiative Observational Study. J Bone Miner Res 24(8):1369–1379. 5. Andersen S, Frederiksen KD, Hansen S, et al. 2014 Bone structure and estimated bone strength in obese patients evaluated by high-resolution peripheral quantitative computed tomography. Calcif Tissue Int 95(1):19–28. 6. Nielson CM, Srikanth P, Orwoll ES. 2012 Obesity and fracture in men and women: an epidemiologic perspective. J Bone Miner Res 27(1):1–10.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS 8 7. Douchi T, Yamamoto S, Oki T, et al. 2000 Difference in the effect of adiposity on bone density between pre- and postmenopausal women. Maturitas 34(3):261–266. 8. Filip R, Raszewski G. 2009 Bone mineral density and bone turnover in relation to serum leptin, alpha-ketoglutarate and sex steroids in overweight and obese postmenopausal women. Clin Endocrinol (Oxf) 70(2):214–220. 9. Reid IR, Ames R, Evans MC, et al. 1992 Determinants of total body and regional bone mineral density in normal postmenopausal women—a key role for fat mass. J Clin Endocrinol Metab 75(1):45–51. 10. Reid IR, Ames RW, Evans MC, et al. 1994 Determinants of the rate of bone loss in normal postmenopausal women. J Clin Endocrinol Metab 79(4):950–954. 11. Gnudi S, Sitta E, Fiumi N. 2007 Relationship between body composition and bone mineral density in women with and without osteoporosis: relative contribution of lean and fat mass. J Bone Miner Metab 25(5):326–332. 12. Zhu K, Hunter M, James A, et al. 2015 Associations between body mass index, lean and fat body mass and bone mineral density in middle-aged Australians: the Busselton Healthy Ageing Study. Bone 74:146–152. 13. Greco EA, Francomano D, Fornari R, et al. 2013 Negative association between trunk fat, insulin resistance and skeleton in obese women. World J Diabetes 4(2):31–39. 14. Melton LJ, Riggs BL, Leibson CL, et al. 2008 A bone structural basis for fracture risk in diabetes. J Clin Endocrinol Metab 93(12):4804–4809. 15. Ma L, Oei L, Jiang L, et al. 2012 Association between bone mineral density and type 2 diabetes mellitus: a meta-analysis of observational studies. Eur J Epidemiol 27(5):319–332. 16. Arikan S, Tuzcu A, Bahceci M, et al. 2012 Insulin resistance in type 2 diabetes mellitus may be related to bone mineral density. J Clin Densitom 15(2):186–190. 17. Shin D, Kim S, Kim KH, et al. 2014 Association between insulin resistance and bone mass in men. J Clin Endocrinol Metab 99(3):988–995. 18. Belfki H, Ben Ali S, Aounallah-Skhiri H, et al. 2013 Prevalence and determinants of the metabolic syndrome among Tunisian adults: results of the Transition and Health Impact in North Africa (TAHINA) project. Public Health Nutr 16:582– 590. 19. Sahli H, Testouri N, Chihaoui MB, et al. 2009 Bone mineral density in healthy Tunisian women. Maturitas 63:227–232. 20. WHO. 2017 Obesity and overweight [Internet]. WHO. Available at: http://www.who.int/mediacentre/factsheets/fs311/en/. Accessed: September 21, 2016. 21. Friedewald WT, Levy RI, Fredrickson DS. 1972 Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 18(6):499–502. 22. Matthews DR, Hosker JP, Rudenski AS, et al. 1985 Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28(7):412–419. 23. Maïmoun L, Mura T, Leprieur E, et al. 2016 Impact of obesity on bone mass throughout adult life: influence of gender and severity of obesity. Bone 90:23–30. 24. Seeman E, Delmas PD. 2006 Bone quality—the material and structural basis of bone strength and fragility. N Engl J Med 354(21):2250–2261. 25. Cui L-H, Shin M-H, Kweon S-S, et al. 2007 Relative contribution of body composition to bone mineral density at different sites in men and women of South Korea. J Bone Miner Metab 25(3):165–171.
Cherif et al. 26. Nur H, Toraman NF, Arica Z, et al. 2013 The relationship between body composition and bone mineral density in postmenopausal Turkish women. Rheumatol Int 33(3):607– 612. 27. Zhang J, Jin Y, Xu S, et al. 2015 Associations of fat mass and fat distribution with bone mineral density in Chinese obese population. J Clin Densitom 18(1):44–49. 28. Saarelainen J, Honkanen R, Kröger H, et al. 2011 Body fat distribution is associated with lumbar spine bone density independently of body weight in postmenopausal women. Maturitas 69(1):86–90. 29. Fu X, Ma X, Lu H, et al. 2011 Associations of fat mass and fat distribution with bone mineral density in pre- and postmenopausal Chinese women. Osteoporos Int 22(1):113– 119. 30. Blaauw R, Albertse EC, Hough S. 1996 Body fat distribution as a risk factor for osteoporosis. S Afr Med J 86(9):1081– 1084. 31. Kim K-C, Shin D-H, Lee S-Y, et al. 2010 Relation between obesity and bone mineral density and vertebral fractures in Korean postmenopausal women. Yonsei Med J 51(6):857– 863. 32. Weeks BK, Beck BR. 2008 The BPAQ: a bone-specific physical activity assessment instrument. Osteoporos Int 19(11):1567– 1577. 33. Harris S, Dallal GE, Dawson-Hughes B. 1992 Influence of body weight on rates of change in bone density of the spine, hip, and radius in postmenopausal women. Calcif Tissue Int 50(1):19–23. 34. Casale M, von Hurst PR, Beck KL, et al. 2016 Lean mass and body fat percentage are contradictory predictors of bone mineral density in pre-menopausal Pacific Island women. Nutrients 8(8). 35. Kang EK, Park HW, Baek S, Lim JY. 2016 The association between trunk body composition and spinal bone mineral density in Korean males versus females: a farmers’ cohort for agricultural work-related musculoskeletal disorders (FARM) study. J Korean Med Sci 31(10):1595–1603. 36. Sinaki M, Itoi E, Wahner HW, et al. 2002 Stronger back muscles reduce the incidence of vertebral fractures: a prospective 10 year follow-up of postmenopausal women. Bone 30(6):836–841. 37. Vestergaard P. 2007 Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2 diabetes—a meta-analysis. Osteoporos Int 18(4):427–444. 38. Thrailkill KM, Lumpkin CK, Bunn RC, et al. 2005 Is insulin an anabolic agent in bone? Dissecting the diabetic bone for clues. Am J Physiol Endocrinol Metab 289(5):E735–E745. 39. Yamaguchi T, Kanazawa I, Yamamoto M, et al. 2009 Associations between components of the metabolic syndrome versus bone mineral density and vertebral fractures in patients with type 2 diabetes. Bone 45(2):174–179. 40. Haffner SM, Bauer RL. 1993 The association of obesity and glucose and insulin concentrations with bone density in premenopausal and postmenopausal women. Metab Clin Exp 42(6):735–738. 41. Barrett-Connor E, Kritz-Silverstein D. 1996 Does hyperinsulinemia preserve bone? Diabetes Care 19(12):1388– 1392. 42. Stolk RP, Van Daele PL, Pols HA, et al. 1996 Hyperinsulinemia and bone mineral density in an elderly population: the Rotterdam Study. Bone 18(6):545–549. 43. Dennison EM, Syddall HE, Aihie Sayer A, et al. 2004 Type 2 diabetes mellitus is associated with increased axial bone density in men and women from the Hertfordshire Cohort
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017
ARTICLE IN PRESS Association of Obesity and Insulin Resistance With BMD Study: evidence for an indirect effect of insulin resistance? Diabetologia 47(11):1963–1968. 44. Srikanthan P, Crandall CJ, Miller-Martinez D, et al. 2014 Insulin resistance and bone strength: findings from the study of midlife in the United States. J Bone Miner Res 29(4):796– 803.
9
45. Reid IR. 2010 Fat and bone. Arch Biochem Biophys 503(1):20– 27. 46. Dimitri P, Bishop N, Walsh JS, Eastell R. 2012 Obesity is a risk factor for fracture in children but is protective against fracture in adults: a paradox. Bone 50(2):457– 466.
Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health
Volume ■, 2017