The aim of this study is to investigate the circulating OPG levels in postmenopausal women with diabetes and prediabetes and explore the relationships between serum OPG and insulin resistance. A total of 271 unrelated Chinese postmenopausal women were recruited in this study. The subjects were divided into type 2 diabetes mellitus (T2DM) group (
Osteoprotegerin (OPG) is a secreted glycoprotein that belongs to the tumor necrosis factor receptor superfamily. OPG is known to act as a decoy receptor for the receptor activator of nuclear factor-
It has been well recognized that OPG could regulate bone metabolism through essential roles in the formation, activation, and survival of osteoclasts [
Elevated concentrations of OPG have been reported in diabetic individuals and were independently associated with the diabetic microvascular complications [
A total of 271 unrelated southern Han Chinese postmenopausal women were recruited from the database of our previous female osteoporosis study [
Diabetes was defined using the World Health Organization 1999 criteria: (1) fasting plasma glucose (FPG) ≥ 7.0 mmol/L or (2) 2 h postprandial plasma glucose (2hPG) in oral glucose tolerance test ≥ 11.1 mmol/L or (3) typical symptoms with random plasma glucose ≥ 11.1 mmol/L. Impaired glucose regulation, also termed as prediabetes, was defined as follows: 6.1 mmol/L ≤ FPG < 7.0 mmol/L and/or 7.8 mmol/L ≤ 2hPG < 11.1 mmol/L. The criteria for normal glucose regulation were defined as follows: FPG < 6.1 mmol/L and 2hPG < 7.8 mmol/L.
Exclusion criteria included subjects who suffered from diseases associated with disordered glucose metabolism, such as chronic liver disease, severe kidney disease, hypothyroidism, and pituitary or adrenal diseases. Participants who had taken glucocorticosteroid or bisphosphonates within the past 3 months and had a fracture within 1 year were also excluded. In the study, 12 participants had taken calcium and/or vitamin D. In the diabetes group, the average duration of diabetes was 4.0 ± 4.5 years, 55 individuals were newly diagnosed with diabetes without any medication, 2 patients were treated with insulin, and 36 patients were treated with oral hypoglycemic agents, such as metformin, sulfonylureas, and acarbose, but no thiazolidinediones.
Demographic information was collected by an interview through a questionnaire, such as age, smoking history, alcohol intake, and a detailed medical history, especially the history of diabetes. The study was approved by the ethics committee of the Third Hospital of Nanchang, and the subjects provided written informed consent before participation in the study.
Height and weight were measured to the nearest 1 cm and 0.1 kg, respectively. Body mass index (BMI) was calculated [formula: BMI = body weight (kg)/height2 (m2)]. Waist circumference and hip circumference (to the nearest 1 cm) were measured, and waist-to-hip ratio (WHR) was calculated [formula: WHR = waist circumference (cm)/hip circumference (cm)]. Blood pressure and pulse rates were measured in the sitting position using electronic sphygmomanometer Omron, HEM-7200 (Omron Inc., Tokyo, Japan). Body fat percentage was detected using electronic body fat measuring instrument Omron, HBF-375 (Omron Inc., Tokyo, Japan). Bone mineral densities at the lumbar spine (L2–L4) and femoral neck were measured by dual X-ray absorptiometry (DXA, GE-Lunar Prodigy, Waukesha, WI, USA).
Blood samples were taken after an overnight fast. Measurements of serum biochemical parameters, such as FPG, 2hPG, alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), blood urea nitrogen (BUN), creatinine (Cr), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), total cholesterol (TC), triglycerides (TG), and blood uric acid (URCA), were performed using commercially available kits (Siemens Inc., Erlangen, Germany) on the day of the collection. Subsequently, serum was separated, stored at −80°C, until OPG analysis. Serum insulin was measured using an electrochemiluminescence immunoassay (Roche Diagnostics, Indianapolis, USA), while glycated hemoglobin (HbA1c) was detected by high-performance liquid chromatography (Bio-Rad Inc., CA, USA). Serum OPG levels were measured by enzyme-linked immunosorbent assay kit (RayBiotech, Norcross, GA, USA). All of the samples were assayed blindly, in duplicate, and in random order. The sensitivity of the kit was minimally 1 pg/mL and the mean intra- and interassay coefficients of variation were below 10%.
Insulin resistance was determined by homeostasis model assessment (HOMA-IR), and HOMA-IR index was calculated [formula: fasting plasma glucose (mmol/L) × fasting serum insulin (mU/mL)/22.5]. According to previous studies, HOMA-IR index > 1.8 has been suggested as a cut-off point to indicate increased insulin resistance [
Data are reported as means ± standard deviation or median (25th–75th percentile). SPSS 19.0 (SPSS Inc., Chicago, IL, USA) was used to perform the statistical analysis. Normality was tested for via the Kolmogorov-Smirnov test and if violated, a natural logarithmic transformation (e.g., HOMA-IR) was used where possible. Statistical comparisons between groups were made using one-way ANOVA with post hoc comparisons for normally distributed variables [
The basic characteristics of T2DM group (
Clinical and biochemical characteristics of the study subjects.
Variable | NGR group |
IGR group |
T2DM group |
---|---|---|---|
Age (year) | 62.83 ± 8.57 | 62.38 ± 9.02 | 63.81 ± 8.69 |
SBP (mmHg) | 138.40 ± 20.76 | 141.80 ± 19.54 | 147.62 ± 21.06a |
DBP (mmHg) | 80.06 ± 13.61 | 80.9 ± 12.43 | 80.97 ± 12.93 |
Pulse rates (beats/min) | 87 ± 13 | 88 ± 13 | 87 ± 11 |
Height (cm) | 153.10 ± 5.50 | 154.03 ± 6.28 | 153.71 ± 6.23 |
Weight (kg) | 57.24 ± 9.25 | 60.36 ± 10.41a | 60.80 ± 8.59a |
BMI (kg/m2) | 24.40 ± 3.56 | 25.38 ± 3.67 | 25.70 ± 3.06a |
Waist circumference (cm) | 77.33 ± 9.16 | 79.84 ± 9.38 | 81.95 ± 8.86a |
Hip circumference (cm) | 91.98 ± 7.78 | 92.53 ± 8.99 | 93.51 ± 7.84 |
Waist-to-hip ratio | 0.84 ± 0.06 | 0.86 ± 0.06a | 0.88 ± 0.06a |
Body fat percentage (%) | 37.34 ± 3.66 | 37.69 ± 4.11 | 37.81 ± 3.84 |
ALT (U/L) | 22.0 (17.0–22.8) | 28.0 (21.0–41.3)a | 27.0 (17.5–42.0)a |
AST (U/L) | 28.0 (24.0–35.5) | 30.0 (23.8–34.0) | 26.0 (21.0–34.5) |
ALP (U/L) | 77.0 (64.8–92.8) | 82.0 (65.8–96.5) | 82.0 (70.0–96.0) |
BUN (mmol/L) | 4.76 (4.03–5.81) | 4.67 (3.82–5.41) | 5.03 (4.02–5.98) |
Cr ( |
62.0 (69.0–77.0) | 70.0 (61.8–77.0) | 67.0 (60.0–78.0) |
URCA ( |
282.0 (243.8–334.0) | 316.5 (283.8–376.5)a | 312.0 (263.5–363.5)a |
TC (mmol/L) | 5.38 ± 1.13 | 5.62 ± 1.14 | 5.81 ± 1.52a |
TG (mmol/L) | 1.07 (0.85–1.34) | 1.34 (1.08–1.85)a | 1.40 (1.05–1.93)a |
HDL-C (mmol/L) | 1.58 (1.38–1.91) | 1.47 (1.32–1.73) | 1.41 (1.26–1.63)a |
LDL-C (mmol/L) | 2.94 (2.64–3.48) | 3.24 (2.81–3.75) | 3.22 (2.71–3.70) |
FPG (mmol/L) | 4.60 (4.18–4.90) | 5.19 (4.99–5.89)a | 6.16 (6.89–8.98)a, b |
2hPG (mmol/L) | 5.62 (4.82–6.22) | 8.52 (7.98–9.53)a | 13.46 (11.61–16.77)a, b |
HbA1c (%) | 5.5 (5.4–5.7) | 6.0 (5.6–6.3)a | 7.4 (6.1–8.0)a, b |
FINS (mIU/L) | 8.00 (4.94–11.18) | 11.55 (7.25–15.48)a | 10.99 (7.39–15.34)a |
HOMA-IR index | 1.54 (1.02–2.22) | 2.39 (1.80–3.78)a | 3.48 (2.33–5.06)a, b |
HOMA-B | 155.60 (86.03–253.09) | 126.30 (76.02–222.09) | 64.27 (40.22–107.40)a, b |
QUICKI | 0.362 ± 0.032 | 0.334 ± 0.034a | 0.319 ± 0.027a, b |
Lumbar BMD (mg/cm2) | 0.866 ± 0.173 | 0.903 ± 0.168 | 0.937 ± 0.146a |
Femoral neck BMD (mg/cm2) | 1.006 ± 0.174 | 1.053 ± 0.186 | 1.074 ± 0.155a |
Serum OPG (pg/mL) | 151.00 ± 45.72 | 169.28 ± 64.91a | 183.20 ± 56.53a |
eGFR | 81.98 ± 17.61 | 80.96 ± 18.84 | 81.72 ± 18.47 |
Smoking history (%) | 1 (1.14) | 2 (2.22) | 2 (2.15) |
Alcohol intake (%) | 3 (3.41) | 3 (3.33) | 2 (2.15) |
Calcium and vitamin D supplements (%) | 3 (3.41) | 5 (5.56) | 4 (4.30) |
Data are presented as means ± SD or median (25th–75th percentile). a
2hPG, 2 h postprandial plasma glucose; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMD, bone mineral density; BMI, body mass index; BUN, blood urea nitrogen; Cr, creatinine; eGFR, estimate glomerular filtration rate; FINS, fasting insulin; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; HDL-C, high-density lipoprotein cholesterol; HOMA-B, homeostasis model assessment of
Comparsion of serum osteoprotegerin levels with different glucose regulation.
Model | Serum osteoprotegerin (pg/mL) |
|
|
|
||
---|---|---|---|---|---|---|
NGR group |
IGR group |
T2DM group |
||||
Model 1a | 151.00 ± 4.87 | 169.28 ± 6.84 | 183.20 ± 5.86 | 0.031 | <0.001 | 0.096 |
Model 2b | 153.69 ± 5.33 | 170.75 ± 5.19 | 179.23 ± 5.16 | 0.023 | 0.001 | 0.247 |
Model 3c | 154.54 ± 5.24 | 170.61 ± 5.05 | 178.56 ± 4.98 | 0.031 | 0.001 | 0.261 |
Serum OPG data is presented as means ± SE. IGR, impaired glucose regulation; NGR, normal glucose regulation; T2DM, type 2 diabetes.
aUnadjusted; badjusted for age, body mass index, waist-to-hip ratio, and body fat percentage; cadjusted for age, body mass index, waist-to-hip ratio, body fat percentage, eGFR, blood uric acid, aspartate aminotransferase, and alkaline phosphatase; dNGT group versus IGR group; eNGT group versus T2DM group; fIGR group versus T2DM group.
Bivariate correlation analysis of serum OPG levels with clinical parameters was performed (Table
Bivariate correlation analysis between study parameters and serum osteoprotegerin levels.
Parameters | Bivariate correlation | |
---|---|---|
|
| |
Age | 0.453 | <0.001 |
Height | −0.115 | 0.058 |
Weight | −0.082 | 0.178 |
BMI | −0.036 | 0.552 |
Waist circumference | 0.078 | 0.203 |
Hip circumferences | −0.005 | 0.793 |
Waist-to-hip ratio | 0.125 | 0.039 |
Body fat percentage | 0.213 | <0.001 |
ALT | −0.014 | 0.823 |
AST | 0.086 | 0.157 |
ALP | 0.175 | 0.004 |
BUN | 0.025 | 0.688 |
Cr | 0.193 | 0.001 |
Calcium | −0.001 | 0.981 |
URCA | 0.046 | 0.451 |
TC | −0.006 | 0.924 |
TG | 0.027 | 0.660 |
HDL-C | 0.012 | 0.847 |
LDL-C | −0.077 | 0.207 |
FPG | 0.093 | 0.133 |
2hPG | 0.270 | <0.001 |
HbA1c | 0.214 | <0.001 |
FINS | 0.106 | 0.080 |
Ln HOMA-IR | 0.134 | 0.027 |
HOMA-B | −0.037 | 0.546 |
QUICKI | −0.194 | 0.001 |
eGFR | −0.331 | <0.001 |
Lumbar BMD | −0.189 | 0.002 |
Femoral neck BMD | −0.130 | 0.032 |
2hPG, 2 h postprandial plasma glucose; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMD, bone mineral density; BMI, body mass index; BUN, blood urea nitrogen; Cr, creatinine; eGFR, estimate glomerular filtration rate; FINS, fasting insulin; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; HDL-C, high-density lipoprotein cholesterol; HOMA-B, homeostasis model assessment of
In stepwise multiple linear regression analysis with OPG as a dependent variable, age, body fat percentage, waist-to-hip ratio, BMI, SBP, DBP, ALT, AST, ALP, eGFR, TC, TG, HDL-C, LDL-C, 2hPG, HbA1c, and HOMA-IR were added to the model. Finally, HOMA-IR, age, 2hPG, AST, ALP, and eGFR were found to be independent predictors of serum OPG (all
Multiple linear regression analysis with osteoprotegerin as a dependent variable.
Parameters |
|
|
---|---|---|
Age | 2.747 | <0.001 |
HOMA-IR | 2.591 | 0.024 |
eGFR | −0.457 | 0.010 |
AST | 0.575 | 0.001 |
ALP | 0.362 | 0.003 |
2hPG | 2.289 | 0.001 |
2hPG, 2 h postprandial plasma glucose; ALP, alkaline phosphatase; AST, aspartate aminotransferase; eGFR, estimate glomerular filtration rate; HOMA-IR, homeostasis model of assessment for insulin resistance.
In logistic regression analyses with insulin resistance as the dependent variable, OPG, BMI, ALT, URCA, HDL-C, LDL-C, and HbA1c were significantly associated with insulin resistance (all
Stepwise logistic regression analysis with insulin resistance as a dependent variable.
Parameters | Odds ratio | 95% confidence interval |
|
---|---|---|---|
OPG | 1.009 | 1.003–1.015 | 0.006 |
BMI | 1.238 | 1.112–1.378 | <0.001 |
URCA | 1.005 | 1.001–1.010 | 0.020 |
HDL-C | 0.351 | 0.132–0.930 | 0.035 |
LDL-C | 1.724 | 1.162–2.558 | 0.007 |
HbA1c | 2.048 | 1.368–3.065 | 0.001 |
ALT | 1.024 | 1.005–1.043 | 0.012 |
ALT, alanine aminotransferase; BMI, body mass index; HbA1c, glycated hemoglobin; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoproteincholesterol; OPG, osteoprotegerin; URCA, blood uric acid.
In the present study, we demonstrated that circulating OPG concentrations were increased in Chinese postmenopausal women with diabetes and prediabetes. Moreover, serum OPG levels showed significant correlation with insulin resistance.
HOMA-IR was used as a surrogate measure of insulin resistance in our study. Although HOMA-IR was not the gold standard for assessment of insulin sensitivity, it was a clinically useful index used in many studies [
Previous reports had reported similar results about the association between serum OPG levels and HOMA-IR in subjects with type 2 diabetes mellitus [
OPG could be produced by a variety of cells of the cardiovascular system, including vascular smooth muscle cells and endothelial cells, and OPG represents a protective factor for vascular system [
In our study, serum OPG levels showed significant association with insulin resistance, but the mechanisms underlying the association are currently unclear. It is thought that inflammation may link OPG to insulin resistance. Insulin resistance is a hallmark of type 2 diabetes mellitus and regarded as a chronic low-grade systemic inflammation [
It has been recognized that the OPG is a cytokine that increases the mineral density and volume of bone tissue by decreasing the number of active osteoclasts. In our study, bivariate analysis showed an inverse correlation between OPG levels and BMD in postmenopausal women. The result was teleologically interpreted as a counter-regulatory mechanism in order to prevent further bone loss. A significantly negative correlation between circulating levels of OPG and BMD was also reported in two studies in patients with type 1 and 2 diabetes mellitus [
Additionally, we found serum OPG levels were positively correlated with age, 2hPG, HbA1c, AST, ALP, and eGFR through linear regression analysis. These findings were consistent with evidences from previous studies that found serum OPG levels were associated with several clinical and biochemical parameters [
The main limitation of our study is small sample size; our current findings need to be confirmed in further large population studies. In addition, serum OPG concentrations are gender-specific; for example, women have higher circulating OPG levels than men. We conducted a study in a select group of Chinese postmenopausal women, and the subjects were recruited from the database of our previous female osteoporosis study. Thus, future researches in randomly selected and different gender population are needed. Furthermore, the exact mechanisms underlying the observed associations in this study remain to be determined.
The current study demonstrates that serum OPG level is elevated in postmenopausal women with diabetes and prediabetes and significantly associated with insulin resistance. These findings suggest that OPG might be implicated in the pathogenesis of diabetes and is a potential biomarker of insulin resistance in subjects with diabetes and prediabetes.
The authors have declared no competing interests.
Peng Duan and Min Yang contributed equally to this work.
This work was supported by the National Natural Science Foundation of China (no. 81260133) and the Jiangxi Province Science Foundation for Youths (no. 20151BAB215012).