Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Oct 10.
Published in final edited form as: Endocr Pract. 2015 Jan;21(1):14–22. doi: 10.4158/EP14229.OR

Osteoprotegerin is a Better Serum Biomarker of Coronary Artery Calcification than Osteocalcin in Type 2 Diabetes

Raelene E Maser 1,2, M James Lenhard 2,3, Michael B Sneider 4, Ryan T Pohlig 5
PMCID: PMC5634139  NIHMSID: NIHMS908908  PMID: 25100392

Abstract

Objective

Coronary artery calcification (CAC) is a prominent feature of atherosclerosis and is associated with cardiovascular events. In vitro studies have suggested that osteoprotegerin (OPG) and osteocalcin (OC) exert anticalcification potential in the vessel wall. The objective of this study was to investigate the association of CAC and serum bone biomarkers in persons with type 2 diabetes.

Methods

We examined 50 individuals with type 2 diabetes. CAC imaging was performed by multidetector computed tomography. CAC scores ≥10, expressed in Agatston units, were considered abnormal. OC, undercarboxylated OC (ucOC), and OPG levels were determined by enzyme-linked immunosorbent assay.

Results

Abnormal CAC scores were found for 64% of the study cohort. OPG levels were significantly elevated (5.5 ± 2.0 pmol/L vs. 4.2 ± 1.7 pmol/L; P = .026) for those with abnormal CAC scores. No univariate differences were found for OC or ucOC. Logistic regression analyses revealed that an increase in serum OPG level was significantly associated with an increase in CAC score (odds ratio, 3.324; 95% confidence interval, 1.321 to 8.359; P = .011). Longer duration of diabetes was a significant covariate (P = .026), whereas nonsignificant covariates in the final model were age, gender, systolic blood pressure, body mass index, insulin resistance determined by the homeostasis model assessment for insulin resistance, leptin, adiponectin, and glycemic control. The Nagelkerke R2 for the model was 0.66. Neither OC nor ucOC were significantly associated with elevated CAC scores.

Conclusion

Our results suggest that OPG is a more useful serum biomarker than OC or ucOC for identifying those at increased risk of arterial calcification in type 2 diabetes.

INTRODUCTION

Calcification of the coronary arteries is highly correlated with atherosclerosis. Coronary artery calcification (CAC) is characterized by the deposition of calcium in coronary arteries and CAC scores are predictive of cardiac events in type 2 diabetes mellitus (1, 2).

Osteoprotegerin (OPG), a secreted glycoprotein, belongs to the tumor necrosis factor–receptor superfamily (3). OPG is known to inhibit osteoclastogenesis by acting as a decoy receptor for the receptor activator of nuclear factor-κB ligand (RANKL). The coordinated synthesis of RANKL and OPG is an essential feature of bone remodeling. The OPG/RANKL signaling pathway also plays a role in regulating bone matrix proteins in vascular smooth muscle cells (4), with an imbalance in the OPG/RANKL signaling pathway potentially central to the process of vascular calcification (5). In vascular smooth muscle cells, RANKL appears to promote calcification, whereas the action of OPG may be protective (6). OPG circulates in the blood but at lower levels than in the tissues (5). Previous studies have shown that elevated serum OPG levels are associated with the presence and severity of coronary artery disease (CAD) determined by coronary angiography (7) and with CAC scores in asymptomatic individuals with type 2 diabetes (8).

Osteocalcin (OC) is a hydroxyapatite-binding protein synthesized by osteoblasts and contains 3 residues of gamma-carboxyglutamic acid that are responsible for the protein’s calcium-binding properties (9). A fraction of newly synthesized OC is released into the circulation, providing a noninvasive marker of osteoblastic activity (10). OC functions as a continuous inhibitor of calcification in vessel walls and has been demonstrated in nondiseased aortas as well as in all stages of atherosclerosis (11). Recently, an increased risk of coronary heart disease was shown to be associated with serum OC levels in Chinese patients (31% diagnosed with type 2 diabetes) who underwent coronary angiography, where the number of stenotic coronary arteries was associated with decreasing serum OC levels (12).

Posttranslational carboxylation of OC by a vitamin K–dependent carboxylase yields both carboxylated and undercarboxylated (uc) OC molecules in serum (13). When OC is undercarboxylated, its avidity for hydroxyapatite is lower, providing a means for ucOC to more easily enter the circulation (14). OC appears to be not only involved in bone metabolism but may also be involved in glucose homeostasis and adipocyte secretion of adipokines (e.g., adiponectin) (15), with in vitro studies having suggested that ucOC may be the isoform that mediates metabolic functions (16). Some human studies (17, 18) (but not all [19]) have shown an inverse association between OC and insulin resistance measured by the homeostasis model assessment for insulin resistance (HOMA-IR). An inverse relationship between ucOC and fasting plasma glucose and glycated hemoglobin (HbA1c) levels was reported for Japanese men with type 2 diabetes (20). OC and ucOC levels were also recently shown to be inversely associated with abdominal aortic calcification in men with type 2 diabetes (21). It should be noted that this association was independent of HbA1c and HOMA-IR (21). In patients with essential hypertension, ucOC was associated with carotid calcification (10). There is, however, a paucity of studies that have examined the relationship between OPG, OC, ucOC, and CAC in type 2 diabetes. There is a need for serum biomarkers to help identify patients at risk for future cardiovascular events. Thus, the objective in this study was to investigate the association between CAC and serum bone biomarkers in persons with type 2 diabetes without a history of CAD or ischemic cardiac events.

METHODS

Subjects

A total of 51 participants volunteered to participate in this study and were evaluated at the Diabetes and Metabolic Research Center, Christiana Care Health System, Newark, Delaware. Individuals with type 2 diabetes mellitus were eligible for the study if they were ≥18 years old. Exclusion criteria included: (1) history of a myocardial infarction, percutaneous coronary interventions, coronary artery bypass graft surgery, acute coronary syndromes, recent/ ongoing atrial fibrillation, or acute myocardial ischemia; (2) dose changes for antihypertensive and antidiabetes medications 2 months prior to enrollment; and (3) chronic kidney disease ≥ stage 3b. It should be noted that 1 individual that was enrolled in the study had stage 3b chronic kidney disease and thus after exclusion of this individual the results for 50 individuals were utilized. This study had approval of the Institutional Review Board of Christiana Care Corporation, and each participant gave written informed consent before participating in the study.

Clinical Measurements and CAC Scoring

Weight and height were measured using a stadiometer. Body mass index (BMI) was calculated as body weight divided by height squared (kg/m2). Blood pressure (BP) was monitored electronically in the supine posture using an oscillometric automatic recorder. CAC scores were determined via a 64-slice multidetector computed tomography scanner (Somatom Definition, Siemens Medical Solutions, Forchheim, Germany). Threshold calcium was set using a density of 130 Hounsfield units, and approximately 80 scans were obtained in 3-mm contiguous sections of the heart. CAC scores were calculated on U.S. Food and Drug Administration–approved postprocessing workstations according to the method of Agatston et al (22). CAC scores <10 were considered normal, whereas scores ≥10 were considered abnormal (23).

Blood Analytes

OPG and OC (intact) levels were measured using enzyme-linked immunosorbent assay (ELISA) kits (ALPCO, Salem, NH). An ELISA kit was also used to determine ucOC concentrations (Takara Bio, Inc, Otsu, Shiga, Japan). An electrochemiluminescence immunoassay was used to measure insulin, C-peptide, parathyroid hormone, and 25-hydroxyvitamin D levels. Specifically, insulin and C-peptide concentrations were determined on a Roche Cobas e601 chemistry analyzer; parathyroid hormone was measured using a Roche Elecsys 2010 analyzer; and 25-hydroxyvitamin D levels were assayed via a Roche Cobas e411 analyzer (Roche Diagnostics, Indianapolis, IN). Glucose measurements were performed using an enzymatic method, and serum creatinine levels were determined using an enzymatic colorimetric assay. Insulin resistance was calculated using the HOMAIR online calculator downloaded from http://www.dtu.ox.ac.uk (24). HbA1c was measured by high-performance liquid chromatography (HPLC) using a Tosoh G7 automated HPLC analyzer (Tosoh Bioscience, Inc, South San Francisco, CA). Calcium levels were determined using the Vitros Ca slide method on the Vitros 5,1 FS chemistry system (Ortho-Clinical Diagnostics, Rochester, NY). Leptin and total adiponectin levels were measured by radioimmunoassay (Millipore Corporation, Billerica, MA).

Statistical Analyses

Comparisons of demographic and metabolic parameters between those with normal versus abnormal CAC scores were made with unpaired t tests for continuous data and contingency table (chi-square) analysis for categorical variables. Pearson and Spearman rank correlation coefficients, where appropriate, were used to evaluate associations between demographic and metabolic parameters and continuous CAC scores and serum bone biomarkers. Logistic regression was used to assess the association between normal versus abnormal CAC scores and serum bone biomarkers (i.e., OPG, OC, ucOC) and other potential independent variables (i.e., age, duration of diabetes, HOMA-IR, BMI, gender, systolic BP, HbA1c, leptin, and adiponectin). Normality was tested and if violated, a natural logarithmic transformation (e.g., ucOC) was used where possible.

RESULTS

Table 1 provides participants’ physical demographic and metabolic parameters. Individuals with abnormal CAC scores (i.e., ≥10) had increased OPG and lower total adiponectin concentrations. No participants had known CAD or a history of ischemic cardiac events.

Table 1.

Participant Demographics and Metabolic Parametersa

CAC scores <10
(n = 18)		 CAC scores ≥10
(n = 32)		 P value
Age (years) 60 ± 13 65 ± 7 .134
Duration (years) 11 ± 6 14 ± 9 .284
Male/female (n) 5/13 16/16 .149
Systolic blood pressure (mm Hg) 121 ± 12 127 ± 14 .082
Diastolic blood pressure (mm Hg) 73 ± 6 74 ± 8 .877
HbA1c (%) 7.4 ± 1.6 7.2 ± 0.9 --
Ln HbA1c 1.99 ± 0.19 1.97 ± 0.12 .944
Body mass index (kg/m2) 32.8 ± 6.4 32.6 ± 4.7 .912
HOMA-IR 1.8 ± 0.8 2.3 ± 1.3 --
Ln HOMA-IR 0.48 ± 0.47 0.65 ± 0.62 .319
Leptin (ng/mL) 28.6 ± 20.3 19.6 ± 14.5 --
Ln leptin 3.0 ± 0.88 2.7 ± 0.77 .167
Total adiponectin (mg/L) 8.6 ± 6.8 6.0 ± 4.3 --
Ln total adiponectin 1.97 ± 0.56 1.62 ± 0.59 .042
25-hydroxyvitamin D (ng/mL) 29.9 ± 11.5 30.0 ± 10.6 .979
Parathyroid hormone (pg/mL) 40.4 ± 17.6 33.6 ± 13.9 .137
Calcium (mg/dL) 9.5 ± 0.5 9.5 ± 0.4 .808
Serum creatinine (mg/dL) 0.81 ± 0.21 0.89 ± 0.27 .290
Osteoprotegerin (pmol/L) 4.2 ± 1.7 5.5 ± 2.0 .026
Osteocalcin (ng/mL) 7.2 ± 3.0 7.2 ± 2.2 .929
Undercarboxylated osteocalcin (ng/mL) 3.5 ± 2.2 6.4 ± 5.9 --
Ln undercarboxylated osteocalcin 0.99 ± 0.77 1.4 ± 1.0 .128
Open in a new tab

Abbreviations: CAC = coronary artery calcification; HbA1c = glycated hemoglobin; HOMA-IR = homeostasis model assessment for insulin resistance; Ln = natural logarithmic transformation.

a

Data are presented as mean ± SD.

Table 2 shows correlation coefficients for demographic and metabolic parameters with continuous CAC scores and serum bone biomarkers. CAC scores were only positively associated with one serum bone biomarker (i.e., OPG [r = 0.38; P < .01]). It should be noted that neither HOMA-IR nor HbA1c was associated with any of the serum bone biomarkers.

Table 2.

Correlation Analyses

Variable CAC scores
rs a 		OPG
rb 		Osteocalcin
r 		Ln ucOC
r
Age (years) 0.29c 0.22 0.12 −0.15
Duration (years) 0.10 0.06 −0.05 −0.13
Gender −0.28c 0.40d 0.19 −0.07
Systolic blood pressure (mm Hg) 0.10 0.25 0.22 0.07
Diastolic blood pressure (mm Hg) 0.01 −0.02 0.15 0.37d
HbA1c (%) 0.02 −0.07 −0.14 −0.21
Body mass index (kg/m2) −0.01 0.02 −0.13 0.09
HOMA-IR 0.24 −0.10 0.06 −0.09
Leptin (ng/mL) −0.11 0.30c 0.05 0.04
Total adiponectin (mg/L) −0.29c 0.13 0.09 0.04
25-hydroxyvitamin D (ng/mL) −0.02 0.09 −0.25 −0.05
Parathyroid hormone (pg/mL) −0.02 0.17 −0.02 −0.11
Calcium (mg/dL) 0.01 0.33c −0.06 0.03
Serum creatinine (mg/dL) 0.15 −0.24 −0.08 0.10
Osteocalcin (ng/mL) −0.01 0.13 -- --
ucOC (ng/mL) 0.15 0.00 0.13 --
OPG (pmol/L) 0.38d -- -- --
Open in a new tab

Abbreviations: CAC = coronary artery calcification; HbA1c = glycated hemoglobin; HOMA-IR = homeostasis model assessment for insulin resistance; OPG = osteoprotegerin; r = Pearson correlation coefficient; rs = Spearman rank correlation coefficient; ucOC = undercarboxylated osteocalcin.

a

Presented for continuous CAC scores that were not normally distributed.

b

Presented for OPG, osteocalcin, and the natural logarithmic transformation of undercarboxylated osteocalcin.

c

P<.05.

d

P<.01.

Table 3 shows a comparison of serum bone biomarkers for those with good glycemic control (i.e., HbA1c < 7%) versus those with poorer control. Although all biomarkers were higher for those with good glycemic control, there were no statistically significant differences between the groups. Likewise, we also compared serum bone biomarkers based on amount of insulin resistance (Table 4). HOMA-IR >1.8 has been suggested as a cut-off value to indicate increased insulin resistance (25). There were no statistically significant differences between the groups with respect to OPG, OC, or ucOC.

Table 3.

Serum Bone Biomarkers by Glycemic Control Levela

Measure HbA1c <7 %
(n = 24)		 HbA1c ≥7%
(n = 26)		 P value
Osteoprotegerin (pmol/L) 5.3 ± 2.3 4.7 ± 1.6 .284
Osteocalcin (ng/mL) 7.7 ± 3.0 6.8 ± 1.9 .208
Undercarboxylated osteocalcin (ng/mL) 6.9 ± 6.2 3.8 ± 3.2 --
Ln undercarboxylated osteocalcin 1.5 ± 1.0 1.0 ± 0.8 .072
Open in a new tab

Abbreviations: HbA1c = glycated hemoglobin; Ln = natural logarithmic transformation.

a

Data are presented as mean ± SD.

Table 4.

Serum Bone Biomarkers by Insulin Resistancea

Measure HOMA-IR ≤1.8
(n = 27)		 HOMA-IR >1.8
(n = 23)		 P value
Osteoprotegerin (pmol/L) 5.1 ± 1.9 4.8 ± 2.1 .608
Osteocalcin (ng/mL) 7.3 ± 2.6 7.1 ± 2.4 .698
Undercarboxylated osteocalcin (ng/mL) 5.6 ± 5.8 5.0 ± 4.1 --
Ln undercarboxylated osteocalcin 1.3 ± 1.0 1.2 ± 0.9 .870
Open in a new tab

Abbreviations: HOMA-IR = homeostasis model assessment for insulin resistance; Ln = natural logarithmic transformation.

a

Data are presented as mean ± SD.

Table 5 provides results of logistic regression analyses, with normal versus abnormal CAC scores as the dependent variable. A series of regression models were used to evaluate the relationship between CAC scores and the 3 serum bone biomarkers, individually and jointly. These models included adjusting for the covariates mentioned above. Individually, OPG was significantly associated with an abnormal CAC score. Neither OC nor ucOC were significantly associated with CAC scores. In a model including all 3 of the serum bone biomarkers, OPG remained the only significant biomarker. The fact that OPG was significant regardless of the model suggests that increased OPG levels are robustly associated with elevated CAC scores. Based on these results, the final model presented here included OPG as the serum bone biomarker. Longer duration of diabetes was the only significant covariate associated with elevated CAC scores. Nonsignificant covariates in the model were age, gender, systolic BP, BMI, HOMA-IR, leptin, adiponectin, and HbA1c. The model Nagelkerke R2 was 0.66, indicating that approximately 66% of the variation in CAC scores could be explained by this model.

Table 5.

Full Logistic Regression Model with CAC Scores as the Dependent Variable

Parameter Odds Ratio 95% Confidence Interval P value
Osteoprotegerin (pmol/L) 3.324 1.321–8.359 .011
Duration of diabetes (years) 1.175 1.020–1.355 .026
Age (years) 1.083 0.951–1.234 .230
Systolic blood pressure (mm Hg) 1.109 0.974–1.262 .120
Gender 0.082 0.003–2.614 .157
Body mass index (kg/m2) 1.159 0.858–1.565 .337
HOMA-IR 3.056 0.681–13.726 .145
Leptin (ng/mL) 0.919 0.831–1.016 .097
Adiponectin (mg/L) 0.990 0.781–1.254 .931
HbA1c (%) 0.741 0.283–1.943 .542
Open in a new tab

Abbreviations: CAC = coronary artery calcification; HbA1c = glycated hemoglobin; HOMA-IR = homeostasis model assessment for insulin resistance.

We were concerned with the number of covariates, given our sample size; thus, an automated logistic regression procedure was employed. A backward, stepwise selection with removal testing based on the likelihood-ratio statistic from the maximum partial likelihood estimates was used. The initial model contained all 3 serum bone biomarkers and the 9 covariates. After backwards elimination, the final model included 4 effects (i.e., OPG, duration, systolic BP, and gender) with increased OPG levels, longer duration of diabetes, higher systolic BP, and male gender being associated with elevated CAC scores (Table 6). It should be noted that although HOMA-IR remained in the final model, it was not statistically significant (P = .054). The Nagelkerke R2 was 0.59 and accounted for almost as much variance as the model above.

Table 6.

Final Backward Selected Logistic Regression Model With CAC Score as the Dependent Variable

Parameter Odds Ratio 95% Confidence Interval P value
Osteoprotegerin (pmol/L) 2.823 1.356–5.878 .006
Duration of diabetes (years) 1.157 1.021–1.311 .022
Systolic blood pressure (mm Hg) 1.125 1.016–1.245 .024
Gender 0.021 0.001–0.297 .004
HOMA-IR 2.903 0.982–8.582 .054
Open in a new tab

Abbreviations: CAC = coronary artery calcification; HOMA-IR = homeostasis model assessment for insulin resistance.

DISCUSSION

In this study, we investigated whether serum bone biomarkers (i.e., OPG, OC, and ucOC) are associated with arterial calcification in the coronary arteries in persons with type 2 diabetes who did not have known CAD or a history of ischemic cardiac events. The results of this study demonstrate that serum OPG levels, not OC or ucOC levels, are associated with the presence of abnormal CAC scores. Our results support the results of an earlier study (8) in which increased serum OPG levels were associated with elevated CAC scores in asymptomatic individuals with type 2 diabetes. Our results are in agreement with several other studies that have shown an increase in serum OPG in association with vascular injury assessed by various methods in patients with and without type 2 diabetes (7, 26–28). The mechanisms for the increased serum OPG levels are, however, not clear.

OPG is widely expressed in human tissues, including bone and the vasculature (7). OPG is a regulator of bone resorption and prevents osteoclast differentiation. Growing evidence suggests that OPG/RANKL proteins may be important in vascular calcification (6). Insulin resistance is a key feature of type 2 diabetes. It is possible that elevated serum OPG may reflect this underlying condition. Indeed, one study involving a group of newly diagnosed individuals with diabetes and normal controls found that OPG and HOMA-IR are positively correlated (29). Our results, however, in a group of persons with type 2 diabetes with a longer duration of diabetes were in contrast to these findings but in agreement with another study of individuals with abnormal glucose tolerance in which no association between HOMA-IR and OPG was found (30). It is also possible that poor glycemic control may affect OPG levels. This was suggested in a study of 166 individuals with type 2 diabetes showing that those with higher HbA1c levels have increased OPG concentrations (31). Our results are in contrast, not revealing an association between OPG and glycemic control. Our results suggest that the elevation in serum OPG in our study cohort was due to factors other than insulin resistance and glycemic control. Guzel et al (5) suggested that increased OPG levels may function as a compensatory mechanism to limit vascular damage and to keep the activation of inflammatory pathways in check. Thus, it is possible that a rise in serum OPG levels may reflect silent vascular calcification, with OPG’s compensatory process occurring in a self-defensive response to the progression of disease. Clearly, additional investigations are needed to determine the mechanism responsible for increased serum OPG levels, but it appears that serum OPG is a promising biomarker to identify patients at risk.

In vitro studies have shown that OC is a continuous inhibitor of calcification in the vessel wall (11). Studies of patients with and without CAD have revealed conflicting results. Two studies (12, 32) showed that OC is lower in those with CAD, with decreasing serum OC associated with an increasing number of diseased vessels. In contrast, Poungvarin et al (33) found no difference in OC between controls and those with CAD. Our results are in more agreement with the latter study (33), as we failed to show an association between OC and abnormal CAC scores. Potential mechanisms that have been hypothesized to explain an underlying association between OC and CAD involve beneficial effects of OC on glucose metabolism via enhancement of insulin sensitivity (32). Recent clinical studies in humans have demonstrated some conflicting associations between OC and glucose metabolism and insulin resistance. For example, using multiple regression analysis adjusted for age, BMI, and creatinine, Kanazawa et al (34) showed that OC levels are negatively associated with fasting glucose, HbA1c, and HOMA-IR in persons with type 2 diabetes. Likewise, Zhang et al (12) found an inverse relationship between OC and fasting glucose and HbA1c after adjusting for CAD status. These results would seem to indicate that individuals with poorer glycemic control are more likely to have lower OC levels. Diaz-Lopez et al (18), studying a group of individuals newly diagnosed with diabetes, showed that OC concentrations are inversely associated with HOMA-IR after adjustment for gender, age, BMI, and other confounding factors. Wang et al (19), however, showed conflicting results in 66 cases with type 2 diabetes. Although OC was inversely correlated with HbA1c in that study, the authors found no relationship between OC and HOMA-IR. Likewise, Gower et al (35) did not show an association between OC and HOMA-IR in 63 overweight/obese adults. These investigators did, however, observe a positive association between OC and insulin sensitivity as assessed with an intravenous glucose tolerance test. These observations suggest that the potential effect of OC on insulin sensitivity may be associated with skeletal muscle and not the liver. It is tempting to speculate that OC’s function in glucose metabolism may be a more localized (rather than systemic) effect for mechanical support of new bone formation via OC’s role in bone growth. Nonetheless, the results of the current study failed to show any associations between OC and measures of glucose metabolism or CAC.

Studies that have examined the association between ucOC levels and arterial disease are limited. Okura et al (10) showed that in patients with essential hypertension, circulating levels of ucOC are higher for patients with carotid calcification compared to those without calcification. In addition, logistic regression analysis showed that ucOC is an independent determinant, suggesting that ucOC is a circulating biomarker of carotid artery calcification (10). To the best of our knowledge, our study is the first to address ucOC and CAC. We did not, however, find any significant difference in ucOC levels between those with and without abnormal CAC scores. Results from animal studies initially suggested that ucOC is the isoform that mediates metabolic functions (16). Studies in humans have produced conflicting results. For example, Iki et al (36) showed that ucOC is inversely associated with HOMA-IR, fasting glucose, and HbA1c in elderly Japanese men with and without type 2 diabetes. In contrast, Mori et al (37) found no association between ucOC and insulin resistance in patients with type 2 diabetes, where insulin resistance was assessed using the euglycemic hyperinsulinemic clamp technique. Likewise, in this study, we did not find any association with ucOC in terms of insulin resistance or glycemic control. Clearly, additional research is needed to determine the isoform of OC that is associated with indices of glucose metabolism in humans.

We also investigated two other potential surrogate markers (leptin and adiponectin) of coronary calcification, as these biomarkers of adipose function may add value to predicting vascular disease and due to their relationship with bone biomarkers. Adiponectin appears to exert cardioprotective effects with low levels of adiponectin shown to be associated with elevated CAC scores (38). We also demonstrated that individuals with elevated CAC scores had lower adiponectin levels. However, in logistic analysis, adiponectin was not an independent determinant of CAC scores. The association between leptin and CAC has produced conflicting reports (39, 40). Although our results do not support an independent association between leptin or adiponectin and vascular calcification, it is possible that the relationship between these adipocytes and CAC is driven by covariation with other related risk factors.

Our study has several strengths, including simultaneous assessment of OPG, OC, and ucOC in the same cohort along with the measurement of adipokines. This study also has some potential limitations that deserve mention. This was a cross-sectional study and thus cause-effect relationships cannot be determined. It is known that with increasing age, the risk of having calcium deposits in the coronary arteries increases. This was confirmed in this study via a positive correlation between age and CAC scores as a continuous variable. When CAC scores were dichotomized to form two groups, age was not significantly different between the groups, and the final backward stepwise logistic regression model selected longer duration of diabetes and higher systolic BP as significant effects. It should be noted, however, that both effects coincide with increasing age. In addition, other investigators (7) showed that age is not significantly associated with CAD in multivariate logistic regression analyses in which OPG levels were evaluated. We used HOMA-IR as a surrogate measure of insulin resistance. Although HOMA-IR is not the gold standard, it is a clinically useful index used in many studies. We measured only intact molecules of OC, levels of which would be lower than total OC levels. Our study did not include a control group (i.e., nondiabetic individuals of a similar age range), nor did we have normal reference ranges from our local population for the serum bone biomarkers. It should be noted, however, that our results for these biomarkers appear comparable to other studies that have included small control groups (5, 18). Some participants may have been using medications that could have influenced coronary calcification and/or serum bone biomarkers. It should be noted, however, that there were no dose changes for antihypertensive and antidiabetes medications 2 months prior to enrollment. No invasive measures to determine the presence/absence or extent of atherosclerotic disease were used in this study, but the determination of CAC scores does provide a valid marker of atherosclerotic burden.

CONCLUSION

In summary, the present study measured 3 serum biomarkers of bone metabolism in patients with type 2 diabetes. We confirm that in patients with abnormal CAC scores, elevated serum OPG levels are present. Because cardiovascular disease morbidity and mortality are high in persons with type 2 diabetes, it is important to identify biochemical markers of increased risk. The identification of such biomarkers could be useful in the clinical setting for determination of subclinical cardiovascular disease. This would permit the development of strategies designed to help in the reduction of cardiovascular events. Further studies will be required, however, to determine exactly what information elevated serum OPG levels are indicating with regard to underlying changes that are occurring in the vascular wall.

Acknowledgments

Funding for this study was provided by the Clinical Research Committee, Department of Medicine, Christiana Care Health System, Newark, Delaware. Parts of this study were presented as a poster presentation at the American Diabetes Association 74th Annual Scientific Session, San Francisco, California, June 13–17, 2014.

Abbreviations

BMI

body mass index

BP

blood pressure

CAC

coronary artery calcification

CAD

coronary artery disease

HbA1c

glycated hemoglobin

HOMA-IR

homeostasis model assessment for insulin resistance

OC

osteocalcin

OPG

osteoprotegerin

RANKL

receptor activator of nuclear factor-κB ligand

ucOC

undercarboxylated osteocalcin

Footnotes

DISCLOSURE

The authors have no multiplicity of interest to disclose.

References

  • 1.Elkeles RS, Godsland IF, Feher MD, et al. Coronary calcium measurement improves prediction of cardiovascular events in asymptomatic patients with type 2 diabetes: the PREDICT study. Eur Heart J. 2008;29:2244–2251. doi: 10.1093/eurheartj/ehn279. [DOI] [PubMed] [Google Scholar]
  • 2.Agarwal S, Morgan T, Herrington DM, et al. Coronary calcium score and prediction of all-cause mortality in diabetes: the Diabetes Heart study. Diabetes Care. 2011;34:1219–1224. doi: 10.2337/dc11-0008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Simonet WS, Lacey DL, Dunstan CR, et al. Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell. 1997;89:309–319. doi: 10.1016/s0092-8674(00)80209-3. [DOI] [PubMed] [Google Scholar]
  • 4.Jeffcoate W. Vascular calcification and osteolysis in diabetic neuropathy-is RANK-L the missing link? Diabetologia. 2004;47:1488–1492. doi: 10.1007/s00125-004-1477-5. [DOI] [PubMed] [Google Scholar]
  • 5.Guzel S, Seven A, Kocaoglu A, et al. Osteoprotegerin, leptin and IL-6: association with silent myocardial ischemia in type 2 diabetes mellitus. Diab Vasc Dis Res. 2013;10:25–31. doi: 10.1177/1479164112440815. [DOI] [PubMed] [Google Scholar]
  • 6.Wu M, Rementer C, Giachelli CM. Vascular calcification: an update on mechanisms and challenges in treatment. Calcif Tissue Int. 2013;93:365–373. doi: 10.1007/s00223-013-9712-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Jono S, Ikari Y, Shioi A, et al. Serum osteoprotegerin levels are associated with the presence and severity of coronary artery disease. Circulation. 2002;106:1192–1194. doi: 10.1161/01.cir.0000031524.49139.29. [DOI] [PubMed] [Google Scholar]
  • 8.Anand DV, Lahiri A, Lim E, Hopkins D, Corder R. The relationship between plasma osteoprotegerin levels and coronary artery calcification in uncomplicated type 2 diabetic subjects. J Am Coll Cardiol. 2006;47:1850–1857. doi: 10.1016/j.jacc.2005.12.054. [DOI] [PubMed] [Google Scholar]
  • 9.Delmas PD, Eastell R, Garnero P, Seibel MJ, Stepan J. The use of biochemical markers of bone turnover in osteoporosis. Committee of Scientific Advisors of the International Osteoporosis Foundation. Osteoporos Int. 2000;11(suppl 6):S2–S17. doi: 10.1007/s001980070002. [DOI] [PubMed] [Google Scholar]
  • 10.Okura T, Kurata M, Enomoto D, et al. Undercarboxylated osteocalcin is a biomarker of carotid calcification in patients with essential hypertension. Kidney Blood Press Res. 2010;33:66–71. doi: 10.1159/000289575. [DOI] [PubMed] [Google Scholar]
  • 11.Dhore CR, Cleutjens JP, Lutgens E, et al. Differential expression of bone matrix regulatory proteins in human atherosclerotic plaques. Arterioscler Thromb Vasc Biol. 2001;21:1998–2003. doi: 10.1161/hq1201.100229. [DOI] [PubMed] [Google Scholar]
  • 12.Zhang Y, Qi L, Gu W, et al. Relation of serum osteocalcin level to risk of coronary heart disease in Chinese adults. Am J Cardiol. 2010;106:1461–1465. doi: 10.1016/j.amjcard.2010.07.013. [DOI] [PubMed] [Google Scholar]
  • 13.Bulló M, Moreno-Navarrete JM, Fernández-Real JM, Salas-Salvadó J. Total and undercarboxylated osteocalcin predict changes in insulin sensitivity and beta cell function in elderly men at high cardiovascular risk. Am J Clin Nutr. 2012;95:249–255. doi: 10.3945/ajcn.111.016642. [DOI] [PubMed] [Google Scholar]
  • 14.Clemens TL, Karsenty G. The osteoblast: an insulin target cell controlling glucose homeostasis. J Bone Miner Res. 2011;26:677–680. doi: 10.1002/jbmr.321. [DOI] [PubMed] [Google Scholar]
  • 15.Kim YS, Paik IY, Rhie YJ, Suh SH. Integrative physiology: defined novel metabolic roles of osteocalcin. J Korean Med Sci. 2010;25:985–991. doi: 10.3346/jkms.2010.25.7.985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Lee NK, Sowa H, Hinoi E, et al. Endocrine regulation of energy metabolism by the skeleton. Cell. 2007;130:456–469. doi: 10.1016/j.cell.2007.05.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Saleem U, Mosley TH, Jr, Kullo IJ. Serum osteocalcin is associated with measures of insulin resistance, adipokine levels, and the presence of metabolic syndrome. Arterioscler Thromb Vasc Biol. 2010;30:1474–1478. doi: 10.1161/ATVBAHA.110.204859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Díaz-López A, Bulló M, Juanola-Falgarona M, et al. Reduced serum concentrations of carboxylated and undercarboxylated osteocalcin are associated with risk of developing type 2 diabetes mellitus in a high cardiovascular risk population: A nested case-control study. J Clin Endocrinol Metab. 2013;98:4524–4531. doi: 10.1210/jc.2013-2472. [DOI] [PubMed] [Google Scholar]
  • 19.Wang Q, Zhang B, Xu Y, Xu H, Zhang N. The relationship between serum osteocalcin concentration and glucose metabolism in patients with type 2 diabetes mellitus. Int J Endocrinol. 2013;2013:842598. doi: 10.1155/2013/842598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kanazawa I, Yamaguchi T, Yamauchi M, et al. Serum undercarboxylated osteocalcin was inversely associated with plasma glucose level and fat mass in type 2 diabetes mellitus. Osteoporos Int. 2011;22:187–194. doi: 10.1007/s00198-010-1184-7. [DOI] [PubMed] [Google Scholar]
  • 21.Ogawa-Furuya N, Yamaguchi T, Yamamoto M, Kanazawa I, Sugimoto T. Serum osteocalcin levels are inversely associated with abdominal aortic calcification in men with type 2 diabetes mellitus. Osteoporos Int. 2013;24:2223–2230. doi: 10.1007/s00198-013-2289-6. [DOI] [PubMed] [Google Scholar]
  • 22.Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Jr, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15:827–832. doi: 10.1016/0735-1097(90)90282-t. [DOI] [PubMed] [Google Scholar]
  • 23.Kramer CK, Zinman B, Gross JL, et al. Coronary artery calcium score prediction of all cause mortality and cardiovascular events in people with type 2 diabetes: systematic review and meta-analysis. BMJ. 2013;346:f1654. doi: 10.1136/bmj.f1654. [DOI] [PubMed] [Google Scholar]
  • 24.Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care. 1998;21:2191–2192. doi: 10.2337/diacare.21.12.2191. [DOI] [PubMed] [Google Scholar]
  • 25.Geloneze B, Vasques AC, Stabe CF, et al. HOMA1-IR and HOMA2-IR indexes in identifying insulin resistance and metabolic syndrome: Brazilian Metabolic Syndrome Study (BRAMS) Arq Bras Endocrinol Metabol. 2009;53:281–287. doi: 10.1590/s0004-27302009000200020. [DOI] [PubMed] [Google Scholar]
  • 26.Aoki A, Murata M, Asano T, et al. Association of serum osteoprotegerin with vascular calcification in patients with type 2 diabetes. Cardiovasc Diabetol. 2013;12:11. doi: 10.1186/1475-2840-12-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Jung CH, Lee WY, Kim SY, et al. The relationship between coronary artery calcification score, plasma osteoprotegerin level and arterial stiffness in asymptomatic type 2 DM. Acta Diabetol. 2010;47(suppl 1):145–152. doi: 10.1007/s00592-009-0154-z. [DOI] [PubMed] [Google Scholar]
  • 28.Stepień E, Fedak D, Klimeczek P, et al. Osteoprotegerin, but not osteopontin, as a potential predictor of vascular calcification in normotensive subjects. Hypertens Res. 2012;35:531–538. doi: 10.1038/hr.2011.231. [DOI] [PubMed] [Google Scholar]
  • 29.Kim SM, Lee J, Ryu OH, et al. Serum osteoprotegerin levels are associated with inflammation and pulse wave velocity. Clin Endocrinol (Oxf) 2005;63:594–598. doi: 10.1111/j.1365-2265.2005.02390.x. [DOI] [PubMed] [Google Scholar]
  • 30.O’Sullivan EP, Ashley DT, Davenport C, et al. A comparison of osteoprotegerin with adiponectin and high-sensitivity C-reactive protein (hsCRP) as a marker for insulin resistance. Metabolism. 2013;62:34–38. doi: 10.1016/j.metabol.2012.06.005. [DOI] [PubMed] [Google Scholar]
  • 31.Altinova AE, Toruner F, Akturk M, et al. Relationship between serum osteoprotegerin, glycemic control, renal function and markers of atherosclerosis in type 2 diabetes. Scand J Clin Lab Invest. 2011;71:340–343. doi: 10.3109/00365513.2011.570868. [DOI] [PubMed] [Google Scholar]
  • 32.Bao Y, Zhou M, Lu Z, et al. Serum levels of osteocalcin are inversely associated with the metabolic syndrome and the severity of coronary artery disease in Chinese men. Clin Endocrinol (Oxf) 2011;75:196–201. doi: 10.1111/j.1365-2265.2011.04065.x. [DOI] [PubMed] [Google Scholar]
  • 33.Poungvarin N, Leowattana W, Mahanonda N, Bhuripanyo K, Pokum S, Worawattananon P. Biochemical markers of bone turnover in angiographically-demonstrated coronary artery disease patients and healthy Thais. J Med Assoc Thai. 2000;83(suppl 2):S13–S18. [PubMed] [Google Scholar]
  • 34.Kanazawa I, Yamaguchi T, Tada Y, Yamauchi M, Yano S, Sugimoto T. Serum osteocalcin level is positively associated with insulin sensitivity and secretion in patients with type 2 diabetes. Bone. 2011;48:720–725. doi: 10.1016/j.bone.2010.12.020. [DOI] [PubMed] [Google Scholar]
  • 35.Gower BA, Pollock NK, Casazza K, Clemens TL, Goree LL, Granger WM. Associations of total and undercarboxylated osteocalcin with peripheral and hepatic insulin sensitivity and beta-cell function in overweight adults. J Clin Endocrinol Metab. 2013;98:E1173–E1180. doi: 10.1210/jc.2013-1203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Iki M, Tamaki J, Fujita Y, et al. Serum undercarboxylated osteocalcin levels are inversely associated with glycemic status and insulin resistance in an elderly Japanese male population: Fujiwara-kyo Osteoporosis Risk in Men (FORMEN) study. Osteoporos Int. 2012;23:761–770. doi: 10.1007/s00198-011-1600-7. [DOI] [PubMed] [Google Scholar]
  • 37.Mori K, Emoto M, Motoyama K, et al. Undercarboxylated osteocalcin does not correlate with insulin resistance as assessed by euglycemic hyperinsulinemic clamp technique in patients with type 2 diabetes mellitus. Diabetol Metab Syndr. 2012;4:53. doi: 10.1186/1758-5996-4-53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Jover E, Hernández-Romero D, Hurtado JA, Romero-Aniorte AI, Casas T, Valdés M. Serum adiponectin level as a biomarker of coronary artery calcification and severe coronary lesions [in English and Spanish] Rev Esp Cardiol (Engl Ed) 2012;65:1060–1061. doi: 10.1016/j.recesp.2012.03.031. [DOI] [PubMed] [Google Scholar]
  • 39.Reilly MP, Iqbal N, Schutta M, et al. Plasma leptin levels are associated with coronary atherosclerosis in type 2 diabetes. J Clin Endocrinol Metab. 2004;89:3872–3878. doi: 10.1210/jc.2003-031676. [DOI] [PubMed] [Google Scholar]
  • 40.Iribarren C, Husson G, Go AS, et al. Plasma leptin levels and coronary artery calcification in older adults. J Clin Endocrinol Metab. 2007;92:729–732. doi: 10.1210/jc.2006-1138. [DOI] [PubMed] [Google Scholar]

RESOURCES