Osteoporos Int DOI 10.1007/s00198-014-2704-7
SHORT COMMUNICATION
Bone quality assessment in type 2 diabetes mellitus R. Dhaliwal & D. Cibula & C. Ghosh & R. S. Weinstock & A. M. Moses
Received: 20 December 2013 / Accepted: 26 March 2014 # International Osteoporosis Foundation and National Osteoporosis Foundation 2014
Abstract Summary The increased risk for fractures in type 2 diabetes mellitus (T2DM) despite higher average bone density is unexplained. This study assessed trabecular bone quality in T2DM using the trabecular bone score (TBS). The salient findings are that TBS is decreased in T2DM and low TBS associates with worse glycemic control. Introduction Type 2 diabetes mellitus is a risk factor for osteoporotic fractures despite high average bone mineral density (BMD). The aim of this study was to compare BMD with a noninvasive assessment of trabecular microarchitecture, TBS, in women with T2DM. M e t h o d s I n a c r o s s - s e c t i o n a l s t u d y, t r a b e c u l a r microarchitecture was examined in 57 women with T2DM and 43 women without diabetes, ages 30 to 90 years. Lumbar spine BMD was measured by dual-emission x-ray absorptiometry (DXA), and TBS was calculated by examining pixel variations within the DXA images utilizing TBS iNsight software. Results Mean TBS was lower in T2DM (1.228±0.140 vs. 1.298±0.132, p=0.013), irrespective of age. Mean BMD was higher in T2DM (1.150±0.172 vs. 1.051±0.125, p= 0.001). Within the T2DM group, TBS was higher (1.254± 0.148) in subjects with good glycemic control (A1c≤7.5 %) R. Dhaliwal (*) : R. S. Weinstock : A. M. Moses Endocrinology, Diabetes and Metabolism, Department of Medicine, SUNY Upstate Medical University, 750 East Adams Street, Syracuse, NY 13210, USA e-mail: [email protected] D. Cibula : C. Ghosh Department of Public Health and Preventive Medicine, SUNY Upstate Medical University, Syracuse, NY, USA C. Ghosh Department of Mathematics, SUNY College at Buffalo, Buffalo, NY, USA
compared to those (1.166±0.094; p=0.01) with poor glycemic control (A1c>7.5 %). Conclusion In T2DM, TBS is lower and associated with poor glycemic control. Abnormal trabecular microarchitecture may help explain the paradox of increased fractures at a higher BMD in T2DM. Further studies are needed to better understand the relationship between glycemic control and trabecular bone quality. Keywords Osteoporosis . Trabecular bone . Type 2 diabetes
Introduction Osteoporosis is a public health concern with substantial economic burden attributable to pain, low functionality, institutionalization, impaired quality of life, and mortality associated with fragility fractures [1]. Traditionally, bone mineral density (BMD) has been used as a major determinant of bone strength. In routine clinical practice, dual-energy x-ray absorptiometry (DXA) is used as the “gold standard” to evaluate BMD [2]. Fracture risk, however, cannot be predicted by BMD in all individuals. There is a considerable overlap in BMD values between individuals who sustain fragility fractures and those who do not [3, 4]. Factors contributing to bone strength and resistance to fracture including bone mineralization, bone turnover, cortical macrogeometry, and trabecular microarchitecture are not fully captured by BMD [5, 6]. Diabetes mellitus is a risk factor for osteoporotic fractures [7, 8]. Individuals with type 2 diabetes mellitus (T2DM) often have a higher BMD compared to those with type 1 diabetes or without diabetes [9, 10]. BMD and body mass index (BMI) are positively correlated in T2DM. Despite higher average BMD, individuals with T2DM have a greater risk of fragility fractures [10, 11]. It is possible that determinants of bone
Osteoporos Int
strength other than BMD may be contributing to this increased risk of fractures in individuals with T2DM. Trabecular microarchitecture is an important component of bone quality that is not measured by conventional DXA. Trabecular bone score (TBS) evaluates 3D characteristics of bone microarchitecture in the spine and is an indicator of variation in mean bone thickness [12–14]. TBS utilizes an experimental variogram of 2D projection images of existing DXA analysis and when used in addition to BMD at the same site, helps predict bone quality and osteoporotic fractures [14, 15]. Higher TBS value reflects denser bone with higher fracture resistance [16]. Unlike DXA, TBS values are not significantly affected by osteoarthritic changes, making it potentially of even greater value when evaluating bone quality of the spine [17]. The present study was conducted to assess and compare trabecular microarchitecture utilizing TBS and BMD measurements from conventional DXA, in adult women with and without T2DM. We hypothesized that TBS would be similar in women with and without poorly controlled T2DM.
Methods and materials Study design This was a retrospective, cross-sectional study of adult women who had outpatient DXA imaging at Upstate Medical University in Syracuse, NY. Data were collected on Caucasian women, ages 30–90 years, with and without T2DM. Exclusion criteria included the following: (1) history of spinal surgery, (2) evidence of arthritis or arthrosis in the lumbar spine, (3) two or more nonobservable lumbar vertebrae on DXA, or (4) scoliosis of lumbar spine. Final analysis included 57 women with T2DM and 43 women without diabetes.
The study was approved by the Institutional Review Board for the Protection of Human Subjects of SUNY Upstate Medical University at Syracuse, New York. Anonymity of each subject was ensured and maintained by using subject specific numeric codes on all records, including DXA examination files and final dataset. For each subject, the following parameters were determined from medical records and results of DXA exam: age, weight, height, hemoglobin A1c (A1c), BMI, BMD, and projected area, for each vertebra, L1 through L4. Imaging Densitometry All subjects had total lumbar spine (BMD) measured on one densitometer (GE Lunar iDXA) by the same technician. Mean of the individual measurements for L1–L4, excluding any fractured and/or arthrosed vertebrae was recorded. Trabecular microarchitecture assessment Trabecular bone score computation was performed in the same regions of measurement as those used for BMD via TBS software [TBS iNsight V1.0 (Med-Imaps)] installed on GE Lunar iDXA machine. TBS was calculated as the mean value of the individual measurements for vertebrae L1–L4, based on gray-level analysis of DXA images. Any fractured and/or arthrosed vertebrae were excluded from computation. Statistical analysis Statistical analyses were performed using the IBM SPSS, Windows version 22. Descriptive statistics were expressed as mean±SD. Comparison of continuous variables between groups, women with and without T2DM, was performed using independent sample t tests. Linear regression using maximum likelihood estimates for the model parameters was
Table 1 BMD and TBS in Women with and without T2DM Characteristic
T2DM (n=57)
NDM (n=43)
p value
Agea (years) Heighta (cm) BMIa (kg/m2) L spine BMDa(g/cm2) TBSa L spine BMDa,b (g/cm2) TBSa,b TBSa,b,c
65.82 (11.24, 62.84 to 68.81) 159.13 (5.67, 157.62 to 160.63) 35.58 (9.06, 33.17 to 37.98) 1.150 (0.172, 1.10 to 1.20) 1.228 (0.140, 1.19 to 1.27) 1.139 (0.179, 1.09 to 1.19) 1.221 (0.146, 1.18 to 1.26) 1.216 (0.142, 1.18 to 1.25)
64.09 (12.22, 60.33 to 67.85) 158.13 (6.58, 156.10 to 160.16) 27.35 (8.71, 24.67 to 30.03) 1.051 (0.125, 1.01 to 1.09) 1.298 (0.132, 1.26 to 1.34) 1.066 (0.121, 1.03 to 1.10) 1.307 (0.129, 1.27 to 1.35) 1.314 (0.127, 1.28 to 1.35)
0.463 0.418 0.000 0.001 0.013 0.022 0.003 0.001
T2DM type 2 diabetes mellitus, NDM nondiabetes mellitus a
Mean (SD, 95 % CI); p value (t test)
b
Adjusted for age and BMI (linear regression)
c
Adjusted for age, BMI, and BMD (linear regression)
Osteoporos Int Table 2 BMD and TBS in Women with T2DM by glycemic control Characteristic
A1c≤7.5 % (n=40)
A1c>7.5 % (n=17)
p value
Agea (years) Heighta (cm) BMIa (kg/m2) L spine BMDa (g/cm2) TBSa L spine BMDa,b (g/cm2) TBSa,b TBSa,b,c
67.37 (12.16, 63.48 to 71.26) 159.39 (5.04, 157.66 to 161.00) 34.84 (9.64, 31.75 to 37.92) 1.153 (0.165, 1.10 to 1.21) 1.254 (0.148, 1.21 to 1.30) 1.151 (0.169, 1.10 to 1.20) 1.257 (0.141, 1.21 to 1.30) 1.257 (0.134, 1.21 to 1.30)
62.18 (7.86, 58.14 to 66.22) 158.53 (7.07, 154.89 to 162.16) 37.32(7.48, 33.47 to 41.16) 1.143 (0.191, 1.05 to 1.24) 1.166 (0.094, 1.12 to 1.21) 1.148 (0.189, 1.06 to 1.24) 1.160 (0.089, 1.12 to 1.20) 1.160 (0.096, 1.11 to 1.21)
0.062 0.605 0.348 0.855 0.010 0.966 0.002 0.003
a
Mean (SD, 95 % CI); p value (t test)
b
Adjusted for age and BMI (linear regression)
c
Adjusted for age, BMI, and BMD (linear regression)
used to calculate adjusted mean BMD and TBS and to assess the association of factors (A1c, BMD, BMI, and age) and TBS in the T2DM group. These analyses were performed using a priori p value of 0.05, with two-tailed hypothesis tests and 95 % confidence intervals.
Results Analyses included 57 women with T2DM and 43 women without diabetes (Tables 1 and 2). Mean (SD) age was similar in the two groups: 66±11 years in T2DM compared to 64± 12 years in the nondiabetes group (p=0.46). The T2DM group had a higher BMI (35.58±9.06 kg/m2 vs. 27.35±8.71 kg/m2; p7.5 %:n= 17). Although unadjusted and adjusted BMD did not differ significantly between the A1c subgroups, unadjusted and adjusted TBS were significantly higher in subjects with good glycemic control as compared to those with poor glycemic control [1.257±0.134, 95 % CI 1.21 to 1.30 vs. 1.170±0.096; 95 % CI 1.12 to 1.21; p=0.010] (Table 2). Further analyses of the T2DM group using linear regression revealed that TBS is negatively related to A1c; the model predicts decrease in TBS by 0.024 (95 % CI −0.044 to −0.005) for each unit increase in A1c, with all other modeled factors constant (age, BMI, and BMD) (Table 3). TBS is positively related to BMI in the T2DM group; TBS is predicted to increase by 0.004 (95 % CI 0.001 to 0.008) per unit increase in BMI, with other modeled factors constant. Age and lumbar spine BMD did not explain variations in TBS.
Discussion Several studies have documented an increase in fracture incidence in individuals with T2DM despite higher BMD [7, 8]. BMD assessment is therefore of limited utility in predicting fracture risk in T2DM. Structural and material changes in
Model
Coefficients (B)
Wald chi-square
p value
95 % CI
Constant Age BMI L spine BMD A1C
1.068 0.001 0.004 0.121 −0.024
30.14 0.150 5.321 1.423 −5.979
0.000 0.699 0.021 0.233 0.014
0.678, 1.450 −0.003, 0.004 0.001, 0.008 −0.078, 0.319 −0.044, −0.005
Osteoporos Int
bone, including an increase in cortical porosity, are considered contributory factors to this increased fracture risk [18, 19]. While hyperglycemia has been implicated in poor skeletal health, the effect of glycemic control on trabecular bone quality is unknown. In this study, we examined the usefulness of TBS in T2DM and its association with glycemic control. TBS, a noninvasive assessment, was significantly lower in T2DM women (in unadjusted and adjusted models) compared to those without diabetes. These data suggest that trabecular microarchitecture is altered in T2DM and may help explain the increased fracture risk in this population. Our findings are consistent with the results of a large retrospective cohort study from the Manitoba Bone Density Program that compared TBS and DXA in individuals with and without diabetes [20]. BMD was higher and TBS was lower in women with diabetes compared to those without diabetes [20]. This study also reported that a larger portion of the diabetes-associated fracture risk was explained by TBS than BMD, but did not distinguish between type 1 and type 2 diabetes. The current study focused only on women with T2DM. After adjusting for potential confounders, TBS was reduced especially in the presence of poor glycemic control. In contrast, BMD appeared unaffected by glycemic control. Age and lumbar spine BMD showed no association with TBS. This is the first study to suggest that A1c, in addition to BMI, is an important determinant of trabecular bone quality in T2DM. As noted in a meta-analysis by Ma et al. [21], there is evidence that BMI and A1c are positively correlated with BMD. The lack of association between A1c and BMD in our study may be a consequence of limited sample size. Hyperglycemia and insulin resistance could contribute to poor skeletal health in T2DM through several mechanisms, including the production of advanced glycation end products (AGEs) and oxidative stress. Adverse effects on trabecular microarchitecture and cortical porosity may relate to the inability of standard BMD measures to explain the higher fracture risk in T2DM. TBS, a simple noninvasive test, may be of use in more accurately assessing bone health in T2DM when used in conjunction with DXA. These findings need to be further explored in relationship to other factors including diabetes-related complications. Future studies could also assess whether TBS can improve with better glycemic control. Limitations of this study include the small sample size and lack of fracture data. In addition, it is possible that medications, including glycemic control drugs, could independently influence bone quality. This could not be analyzed in the present study due to the relatively small number of participants and large variety of anti-diabetes medications. Further limitations are lack of information on other potential factors associated with bone quality and fractures (duration of diabetes, age of onset of diabetes, complications of diabetes, falls). This pilot study was not designed to measure fracture risk. The
findings of the current study also may not be generalizable to type 1 diabetes, men, other ethnic groups, or populations. The results of this exploratory investigation indicate that TBS may be a useful tool in assessing bone health and understanding the paradox of increased fractures at higher BMD in T2DM. It suggests a negative association between TBS and A1c levels highlighting a potential detrimental effect of hyperglycemia on trabecular microarchitecture. Further investigations are needed to confirm these findings, examine the potential use of TBS to accurately predict fracture risk, and to better understand factors affecting bone quality in diabetes. Acknowledgments Funding for the TBS software was provided by the Division of Endocrinology, Diabetes and Metabolism, SUNY Upstate Medical University, Syracuse, New York. The authors thank Joseph A. Spadaro, PhD for his contribution to preliminary analysis of the study. Conflicts of interest None.
References 1. Lips P, van Schoor NM (2005) Quality of life in patients with osteoporosis. Osteoporos Int 16(5):447–455 2. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. report of a WHO study group. World Health Organ Tech Rep Ser 1994; 843:1–129 3. Cranney A, Jamal SA, Tsang JF, Josse RG, Leslie WD (2007) Low bone mineral density and fracture burden in postmenopausal women. CMAJ 177(6):575–580 4. Stone KL, Seeley DG, Lui LY et al (2003) BMD at multiple sites and risk of fracture of multiple types: long-term results from the study of osteoporotic fractures. J Bone Miner Res 18(11):1947–1954 5. Rice JC, Cowin SC, Bowman JA (1988) On the dependence of the elasticity and strength of cancellous bone on apparent density. J Biomech 21(2):155–168 6. DalleCarbonare L, Giannini S (2004) Bone microarchitecture as an important determinant of bone strength. J Endocrinol Invest 27(1): 99–105 7. Janghorbani M, Feskanich D, Willett WC, Hu F (2006) Prospective study of diabetes and risk of hip fracture: the nurses’ health study. Diabetes Care 29(7):1573–1578 8. Strotmeyer ES, Cauley JA, Schwartz AV et al (2005) Nontraumatic fracture risk with diabetes mellitus and impaired fasting glucose in older white and black adults: the health, aging, and body composition study. Arch Intern Med 165(14):1612–1617 9. Melton LJ 3rd, Riggs BL, Leibson CL et al (2008) A bone structural basis for fracture risk in diabetes. J Clin Endocrinol Metab 93(12): 4804–4809 10. de Liefde II, van der Klift M, de Laet CE, van Daele PL, Hofman A, Pols HA (2005) Bone mineral density and fracture risk in type-2 diabetes mellitus: the Rotterdam study. Osteoporos Int 16(12):1713– 1720 11. Giangregorio LM, Leslie WD, Lix LM et al (2012) FRAX underestimates fracture risk in patients with diabetes. J Bone Miner Res 27(2):301–308 12. Bousson V, Bergot C, Sutter B, Levitz P, Cortet B (2012) Scientific Committee of the Groupe de Recherche et d’Informationsur les Osteoporoses. Trabecular bone score (TBS): available knowledge,
Osteoporos Int clinical relevance, and future prospects. Osteoporos Int 23(5):1489– 1501 13. Roux JP, Wegrzyn J, Boutroy S, Bouxsein ML, Hans D, Chapurlat R (2013) The predictive value of trabecular bone score (TBS) on whole lumbar vertebrae mechanics: an ex vivo study. Osteoporos Int 24(9): 2455–2460 14. Pothuaud L, Carceller P, Hans D (2008) Correlations between greylevel variations in 2D projection images (TBS) and 3D microarchitecture: applications in the study of human trabecular bone microarchitecture. Bone 42(4):775–787 15. Hans D, Goertzen AL, Krieg MA, Leslie WD (2011) Bone microarchitecture assessed by TBS predicts osteoporotic fractures independent of bone density: the Manitoba study. J Bone Miner Res 26(11):2762–2769 16. Breban S, Briot K, Kolta S et al (2012) Identification of rheumatoid arthritis patients with vertebral fractures using bone mineral density and trabecular bone score. J Clin Densitom 15(3):260–266
17. Winzenrieth R, Dufour R, Pothuaud L, Hans D (2010) A retrospective case-control study assessing the role of trabecular bone score in postmenopausal Caucasian women with osteopenia: analyzing the odds of vertebral fracture. Calcif Tissue Int 86(2):104–109 18. Leslie WD, Rubin MR, Schwartz AV, Kanis JA (2012) Type 2 diabetes and bone. J Bone Miner Res 27(11):2231–2237 19. Burghardt AJ, Issever AS, Schwartz AV et al (2010) Highresolution peripheral quantitative computed tomographic imaging of cortical and trabecular bone microarchitecture in patients with type 2 diabetes mellitus. J Clin Endocrinol Metab 95(11):5045–5055 20. Leslie WD, Aubry-Rozier B, Lamy O, Hans D (2013) Manitoba Bone Density Program. TBS (trabecular bone score) and diabetesrelated fracture risk. J Clin Endocrinol Metab 98(2):602–609 21. Ma L, Oei L, Jiang L, Estrada K, Chen H, Wang Z (2012) Association between bone mineral density and type 2 diabetes mellitus: a metaanalysis of observational studies. Eur J Epidemiol 27(5):319–332
SHORT COMMUNICATION
Bone quality assessment in type 2 diabetes mellitus R. Dhaliwal & D. Cibula & C. Ghosh & R. S. Weinstock & A. M. Moses
Received: 20 December 2013 / Accepted: 26 March 2014 # International Osteoporosis Foundation and National Osteoporosis Foundation 2014
Abstract Summary The increased risk for fractures in type 2 diabetes mellitus (T2DM) despite higher average bone density is unexplained. This study assessed trabecular bone quality in T2DM using the trabecular bone score (TBS). The salient findings are that TBS is decreased in T2DM and low TBS associates with worse glycemic control. Introduction Type 2 diabetes mellitus is a risk factor for osteoporotic fractures despite high average bone mineral density (BMD). The aim of this study was to compare BMD with a noninvasive assessment of trabecular microarchitecture, TBS, in women with T2DM. M e t h o d s I n a c r o s s - s e c t i o n a l s t u d y, t r a b e c u l a r microarchitecture was examined in 57 women with T2DM and 43 women without diabetes, ages 30 to 90 years. Lumbar spine BMD was measured by dual-emission x-ray absorptiometry (DXA), and TBS was calculated by examining pixel variations within the DXA images utilizing TBS iNsight software. Results Mean TBS was lower in T2DM (1.228±0.140 vs. 1.298±0.132, p=0.013), irrespective of age. Mean BMD was higher in T2DM (1.150±0.172 vs. 1.051±0.125, p= 0.001). Within the T2DM group, TBS was higher (1.254± 0.148) in subjects with good glycemic control (A1c≤7.5 %) R. Dhaliwal (*) : R. S. Weinstock : A. M. Moses Endocrinology, Diabetes and Metabolism, Department of Medicine, SUNY Upstate Medical University, 750 East Adams Street, Syracuse, NY 13210, USA e-mail: [email protected] D. Cibula : C. Ghosh Department of Public Health and Preventive Medicine, SUNY Upstate Medical University, Syracuse, NY, USA C. Ghosh Department of Mathematics, SUNY College at Buffalo, Buffalo, NY, USA
compared to those (1.166±0.094; p=0.01) with poor glycemic control (A1c>7.5 %). Conclusion In T2DM, TBS is lower and associated with poor glycemic control. Abnormal trabecular microarchitecture may help explain the paradox of increased fractures at a higher BMD in T2DM. Further studies are needed to better understand the relationship between glycemic control and trabecular bone quality. Keywords Osteoporosis . Trabecular bone . Type 2 diabetes
Introduction Osteoporosis is a public health concern with substantial economic burden attributable to pain, low functionality, institutionalization, impaired quality of life, and mortality associated with fragility fractures [1]. Traditionally, bone mineral density (BMD) has been used as a major determinant of bone strength. In routine clinical practice, dual-energy x-ray absorptiometry (DXA) is used as the “gold standard” to evaluate BMD [2]. Fracture risk, however, cannot be predicted by BMD in all individuals. There is a considerable overlap in BMD values between individuals who sustain fragility fractures and those who do not [3, 4]. Factors contributing to bone strength and resistance to fracture including bone mineralization, bone turnover, cortical macrogeometry, and trabecular microarchitecture are not fully captured by BMD [5, 6]. Diabetes mellitus is a risk factor for osteoporotic fractures [7, 8]. Individuals with type 2 diabetes mellitus (T2DM) often have a higher BMD compared to those with type 1 diabetes or without diabetes [9, 10]. BMD and body mass index (BMI) are positively correlated in T2DM. Despite higher average BMD, individuals with T2DM have a greater risk of fragility fractures [10, 11]. It is possible that determinants of bone
Osteoporos Int
strength other than BMD may be contributing to this increased risk of fractures in individuals with T2DM. Trabecular microarchitecture is an important component of bone quality that is not measured by conventional DXA. Trabecular bone score (TBS) evaluates 3D characteristics of bone microarchitecture in the spine and is an indicator of variation in mean bone thickness [12–14]. TBS utilizes an experimental variogram of 2D projection images of existing DXA analysis and when used in addition to BMD at the same site, helps predict bone quality and osteoporotic fractures [14, 15]. Higher TBS value reflects denser bone with higher fracture resistance [16]. Unlike DXA, TBS values are not significantly affected by osteoarthritic changes, making it potentially of even greater value when evaluating bone quality of the spine [17]. The present study was conducted to assess and compare trabecular microarchitecture utilizing TBS and BMD measurements from conventional DXA, in adult women with and without T2DM. We hypothesized that TBS would be similar in women with and without poorly controlled T2DM.
Methods and materials Study design This was a retrospective, cross-sectional study of adult women who had outpatient DXA imaging at Upstate Medical University in Syracuse, NY. Data were collected on Caucasian women, ages 30–90 years, with and without T2DM. Exclusion criteria included the following: (1) history of spinal surgery, (2) evidence of arthritis or arthrosis in the lumbar spine, (3) two or more nonobservable lumbar vertebrae on DXA, or (4) scoliosis of lumbar spine. Final analysis included 57 women with T2DM and 43 women without diabetes.
The study was approved by the Institutional Review Board for the Protection of Human Subjects of SUNY Upstate Medical University at Syracuse, New York. Anonymity of each subject was ensured and maintained by using subject specific numeric codes on all records, including DXA examination files and final dataset. For each subject, the following parameters were determined from medical records and results of DXA exam: age, weight, height, hemoglobin A1c (A1c), BMI, BMD, and projected area, for each vertebra, L1 through L4. Imaging Densitometry All subjects had total lumbar spine (BMD) measured on one densitometer (GE Lunar iDXA) by the same technician. Mean of the individual measurements for L1–L4, excluding any fractured and/or arthrosed vertebrae was recorded. Trabecular microarchitecture assessment Trabecular bone score computation was performed in the same regions of measurement as those used for BMD via TBS software [TBS iNsight V1.0 (Med-Imaps)] installed on GE Lunar iDXA machine. TBS was calculated as the mean value of the individual measurements for vertebrae L1–L4, based on gray-level analysis of DXA images. Any fractured and/or arthrosed vertebrae were excluded from computation. Statistical analysis Statistical analyses were performed using the IBM SPSS, Windows version 22. Descriptive statistics were expressed as mean±SD. Comparison of continuous variables between groups, women with and without T2DM, was performed using independent sample t tests. Linear regression using maximum likelihood estimates for the model parameters was
Table 1 BMD and TBS in Women with and without T2DM Characteristic
T2DM (n=57)
NDM (n=43)
p value
Agea (years) Heighta (cm) BMIa (kg/m2) L spine BMDa(g/cm2) TBSa L spine BMDa,b (g/cm2) TBSa,b TBSa,b,c
65.82 (11.24, 62.84 to 68.81) 159.13 (5.67, 157.62 to 160.63) 35.58 (9.06, 33.17 to 37.98) 1.150 (0.172, 1.10 to 1.20) 1.228 (0.140, 1.19 to 1.27) 1.139 (0.179, 1.09 to 1.19) 1.221 (0.146, 1.18 to 1.26) 1.216 (0.142, 1.18 to 1.25)
64.09 (12.22, 60.33 to 67.85) 158.13 (6.58, 156.10 to 160.16) 27.35 (8.71, 24.67 to 30.03) 1.051 (0.125, 1.01 to 1.09) 1.298 (0.132, 1.26 to 1.34) 1.066 (0.121, 1.03 to 1.10) 1.307 (0.129, 1.27 to 1.35) 1.314 (0.127, 1.28 to 1.35)
0.463 0.418 0.000 0.001 0.013 0.022 0.003 0.001
T2DM type 2 diabetes mellitus, NDM nondiabetes mellitus a
Mean (SD, 95 % CI); p value (t test)
b
Adjusted for age and BMI (linear regression)
c
Adjusted for age, BMI, and BMD (linear regression)
Osteoporos Int Table 2 BMD and TBS in Women with T2DM by glycemic control Characteristic
A1c≤7.5 % (n=40)
A1c>7.5 % (n=17)
p value
Agea (years) Heighta (cm) BMIa (kg/m2) L spine BMDa (g/cm2) TBSa L spine BMDa,b (g/cm2) TBSa,b TBSa,b,c
67.37 (12.16, 63.48 to 71.26) 159.39 (5.04, 157.66 to 161.00) 34.84 (9.64, 31.75 to 37.92) 1.153 (0.165, 1.10 to 1.21) 1.254 (0.148, 1.21 to 1.30) 1.151 (0.169, 1.10 to 1.20) 1.257 (0.141, 1.21 to 1.30) 1.257 (0.134, 1.21 to 1.30)
62.18 (7.86, 58.14 to 66.22) 158.53 (7.07, 154.89 to 162.16) 37.32(7.48, 33.47 to 41.16) 1.143 (0.191, 1.05 to 1.24) 1.166 (0.094, 1.12 to 1.21) 1.148 (0.189, 1.06 to 1.24) 1.160 (0.089, 1.12 to 1.20) 1.160 (0.096, 1.11 to 1.21)
0.062 0.605 0.348 0.855 0.010 0.966 0.002 0.003
a
Mean (SD, 95 % CI); p value (t test)
b
Adjusted for age and BMI (linear regression)
c
Adjusted for age, BMI, and BMD (linear regression)
used to calculate adjusted mean BMD and TBS and to assess the association of factors (A1c, BMD, BMI, and age) and TBS in the T2DM group. These analyses were performed using a priori p value of 0.05, with two-tailed hypothesis tests and 95 % confidence intervals.
Results Analyses included 57 women with T2DM and 43 women without diabetes (Tables 1 and 2). Mean (SD) age was similar in the two groups: 66±11 years in T2DM compared to 64± 12 years in the nondiabetes group (p=0.46). The T2DM group had a higher BMI (35.58±9.06 kg/m2 vs. 27.35±8.71 kg/m2; p7.5 %:n= 17). Although unadjusted and adjusted BMD did not differ significantly between the A1c subgroups, unadjusted and adjusted TBS were significantly higher in subjects with good glycemic control as compared to those with poor glycemic control [1.257±0.134, 95 % CI 1.21 to 1.30 vs. 1.170±0.096; 95 % CI 1.12 to 1.21; p=0.010] (Table 2). Further analyses of the T2DM group using linear regression revealed that TBS is negatively related to A1c; the model predicts decrease in TBS by 0.024 (95 % CI −0.044 to −0.005) for each unit increase in A1c, with all other modeled factors constant (age, BMI, and BMD) (Table 3). TBS is positively related to BMI in the T2DM group; TBS is predicted to increase by 0.004 (95 % CI 0.001 to 0.008) per unit increase in BMI, with other modeled factors constant. Age and lumbar spine BMD did not explain variations in TBS.
Discussion Several studies have documented an increase in fracture incidence in individuals with T2DM despite higher BMD [7, 8]. BMD assessment is therefore of limited utility in predicting fracture risk in T2DM. Structural and material changes in
Model
Coefficients (B)
Wald chi-square
p value
95 % CI
Constant Age BMI L spine BMD A1C
1.068 0.001 0.004 0.121 −0.024
30.14 0.150 5.321 1.423 −5.979
0.000 0.699 0.021 0.233 0.014
0.678, 1.450 −0.003, 0.004 0.001, 0.008 −0.078, 0.319 −0.044, −0.005
Osteoporos Int
bone, including an increase in cortical porosity, are considered contributory factors to this increased fracture risk [18, 19]. While hyperglycemia has been implicated in poor skeletal health, the effect of glycemic control on trabecular bone quality is unknown. In this study, we examined the usefulness of TBS in T2DM and its association with glycemic control. TBS, a noninvasive assessment, was significantly lower in T2DM women (in unadjusted and adjusted models) compared to those without diabetes. These data suggest that trabecular microarchitecture is altered in T2DM and may help explain the increased fracture risk in this population. Our findings are consistent with the results of a large retrospective cohort study from the Manitoba Bone Density Program that compared TBS and DXA in individuals with and without diabetes [20]. BMD was higher and TBS was lower in women with diabetes compared to those without diabetes [20]. This study also reported that a larger portion of the diabetes-associated fracture risk was explained by TBS than BMD, but did not distinguish between type 1 and type 2 diabetes. The current study focused only on women with T2DM. After adjusting for potential confounders, TBS was reduced especially in the presence of poor glycemic control. In contrast, BMD appeared unaffected by glycemic control. Age and lumbar spine BMD showed no association with TBS. This is the first study to suggest that A1c, in addition to BMI, is an important determinant of trabecular bone quality in T2DM. As noted in a meta-analysis by Ma et al. [21], there is evidence that BMI and A1c are positively correlated with BMD. The lack of association between A1c and BMD in our study may be a consequence of limited sample size. Hyperglycemia and insulin resistance could contribute to poor skeletal health in T2DM through several mechanisms, including the production of advanced glycation end products (AGEs) and oxidative stress. Adverse effects on trabecular microarchitecture and cortical porosity may relate to the inability of standard BMD measures to explain the higher fracture risk in T2DM. TBS, a simple noninvasive test, may be of use in more accurately assessing bone health in T2DM when used in conjunction with DXA. These findings need to be further explored in relationship to other factors including diabetes-related complications. Future studies could also assess whether TBS can improve with better glycemic control. Limitations of this study include the small sample size and lack of fracture data. In addition, it is possible that medications, including glycemic control drugs, could independently influence bone quality. This could not be analyzed in the present study due to the relatively small number of participants and large variety of anti-diabetes medications. Further limitations are lack of information on other potential factors associated with bone quality and fractures (duration of diabetes, age of onset of diabetes, complications of diabetes, falls). This pilot study was not designed to measure fracture risk. The
findings of the current study also may not be generalizable to type 1 diabetes, men, other ethnic groups, or populations. The results of this exploratory investigation indicate that TBS may be a useful tool in assessing bone health and understanding the paradox of increased fractures at higher BMD in T2DM. It suggests a negative association between TBS and A1c levels highlighting a potential detrimental effect of hyperglycemia on trabecular microarchitecture. Further investigations are needed to confirm these findings, examine the potential use of TBS to accurately predict fracture risk, and to better understand factors affecting bone quality in diabetes. Acknowledgments Funding for the TBS software was provided by the Division of Endocrinology, Diabetes and Metabolism, SUNY Upstate Medical University, Syracuse, New York. The authors thank Joseph A. Spadaro, PhD for his contribution to preliminary analysis of the study. Conflicts of interest None.
References 1. Lips P, van Schoor NM (2005) Quality of life in patients with osteoporosis. Osteoporos Int 16(5):447–455 2. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. report of a WHO study group. World Health Organ Tech Rep Ser 1994; 843:1–129 3. Cranney A, Jamal SA, Tsang JF, Josse RG, Leslie WD (2007) Low bone mineral density and fracture burden in postmenopausal women. CMAJ 177(6):575–580 4. Stone KL, Seeley DG, Lui LY et al (2003) BMD at multiple sites and risk of fracture of multiple types: long-term results from the study of osteoporotic fractures. J Bone Miner Res 18(11):1947–1954 5. Rice JC, Cowin SC, Bowman JA (1988) On the dependence of the elasticity and strength of cancellous bone on apparent density. J Biomech 21(2):155–168 6. DalleCarbonare L, Giannini S (2004) Bone microarchitecture as an important determinant of bone strength. J Endocrinol Invest 27(1): 99–105 7. Janghorbani M, Feskanich D, Willett WC, Hu F (2006) Prospective study of diabetes and risk of hip fracture: the nurses’ health study. Diabetes Care 29(7):1573–1578 8. Strotmeyer ES, Cauley JA, Schwartz AV et al (2005) Nontraumatic fracture risk with diabetes mellitus and impaired fasting glucose in older white and black adults: the health, aging, and body composition study. Arch Intern Med 165(14):1612–1617 9. Melton LJ 3rd, Riggs BL, Leibson CL et al (2008) A bone structural basis for fracture risk in diabetes. J Clin Endocrinol Metab 93(12): 4804–4809 10. de Liefde II, van der Klift M, de Laet CE, van Daele PL, Hofman A, Pols HA (2005) Bone mineral density and fracture risk in type-2 diabetes mellitus: the Rotterdam study. Osteoporos Int 16(12):1713– 1720 11. Giangregorio LM, Leslie WD, Lix LM et al (2012) FRAX underestimates fracture risk in patients with diabetes. J Bone Miner Res 27(2):301–308 12. Bousson V, Bergot C, Sutter B, Levitz P, Cortet B (2012) Scientific Committee of the Groupe de Recherche et d’Informationsur les Osteoporoses. Trabecular bone score (TBS): available knowledge,
Osteoporos Int clinical relevance, and future prospects. Osteoporos Int 23(5):1489– 1501 13. Roux JP, Wegrzyn J, Boutroy S, Bouxsein ML, Hans D, Chapurlat R (2013) The predictive value of trabecular bone score (TBS) on whole lumbar vertebrae mechanics: an ex vivo study. Osteoporos Int 24(9): 2455–2460 14. Pothuaud L, Carceller P, Hans D (2008) Correlations between greylevel variations in 2D projection images (TBS) and 3D microarchitecture: applications in the study of human trabecular bone microarchitecture. Bone 42(4):775–787 15. Hans D, Goertzen AL, Krieg MA, Leslie WD (2011) Bone microarchitecture assessed by TBS predicts osteoporotic fractures independent of bone density: the Manitoba study. J Bone Miner Res 26(11):2762–2769 16. Breban S, Briot K, Kolta S et al (2012) Identification of rheumatoid arthritis patients with vertebral fractures using bone mineral density and trabecular bone score. J Clin Densitom 15(3):260–266
17. Winzenrieth R, Dufour R, Pothuaud L, Hans D (2010) A retrospective case-control study assessing the role of trabecular bone score in postmenopausal Caucasian women with osteopenia: analyzing the odds of vertebral fracture. Calcif Tissue Int 86(2):104–109 18. Leslie WD, Rubin MR, Schwartz AV, Kanis JA (2012) Type 2 diabetes and bone. J Bone Miner Res 27(11):2231–2237 19. Burghardt AJ, Issever AS, Schwartz AV et al (2010) Highresolution peripheral quantitative computed tomographic imaging of cortical and trabecular bone microarchitecture in patients with type 2 diabetes mellitus. J Clin Endocrinol Metab 95(11):5045–5055 20. Leslie WD, Aubry-Rozier B, Lamy O, Hans D (2013) Manitoba Bone Density Program. TBS (trabecular bone score) and diabetesrelated fracture risk. J Clin Endocrinol Metab 98(2):602–609 21. Ma L, Oei L, Jiang L, Estrada K, Chen H, Wang Z (2012) Association between bone mineral density and type 2 diabetes mellitus: a metaanalysis of observational studies. Eur J Epidemiol 27(5):319–332
