Bone mineral density and trabecular bone score in elderly type 2 diabetes Southeast Asian patients with severe osteoporotic hip fractures

Linsey U Gani; Kundan R Saripalli; Karen Fernandes; Suet F Leong; Koh T Tsai; Pei T Tan; Le R Chong; Thomas F J King

doi:10.1371/journal.pone.0241616

. 2020 Nov 19;15(11):e0241616. doi: 10.1371/journal.pone.0241616

Bone mineral density and trabecular bone score in elderly type 2 diabetes Southeast Asian patients with severe osteoporotic hip fractures

Linsey U Gani ^1,^*, Kundan R Saripalli ¹, Karen Fernandes ², Suet F Leong ², Koh T Tsai ², Pei T Tan ³, Le R Chong ², Thomas F J King ¹

Editor: Robert Daniel Blank⁴

PMCID: PMC7676677 PMID: 33211723

Abstract

Introduction

Studies show trabecular bone score (TBS) may provide information regarding bone quality independent of bone mineral density (BMD) in type 2 diabetes (DM2) patients. We analyzed our Southeast Asian severe osteoporotic hip fracture patients to study these differences.

Methods

We conducted a retrospective cross-sectional analysis of subjects admitted to Changi General Hospital, Singapore with severe osteoporotic hip fractures from 2014–2017 who had BMD performed. Electronic records were reviewed and subjects were classified as having diabetes according to the WHO 2019 criteria. DM2 patients were classified according to their HbA1c into well controlled (HbA1c < 7%) and poorly controlled (HbA1c ≥ 7%) DM2.

Results

Elderly patients with hip fractures present with average femur neck T scores at the osteoporotic range, however those with DM2 had higher BMD and TBS values compared to non DM2 patients. These differences were statistically significant in elderly women—poorly controlled elderly DM2 women with hip fracture had the highest total hip T-score (-2.57 ± 0.86) vs (-2.76 ± 0.96) in well controlled DM2 and (-3.09 ± 1.01) in non DM2 women with hip fracture, p < 0.001. In contrast, TBS scores were lower in poorly controlled DM2 women with hip fracture compared to well controlled DM2 women with hip fracture (1.22 ± 0.11) vs (1.24 ± 0.09), but these were still significantly higher compared to non DM2 women with hip fracture (1.19 ± 0.10), p < 0.001. In elderly men with hip fractures, univariate analysis showed no statistically significant differences in TBS or hip or LS BMD between those with poorly controlled DM2, well controlled DM2 and non DM2. The differences in TBS and BMD remained significant in all DM2 women with hip fractures even after adjustments for potential confounders. Differences in TBS and BMD in poorly controlled DM2 men with hip fractures only became significant after accounting for potential confounders. However, upon inclusion of LS BMD into the multivariate model these differences were attenuated and remained significant only between elderly women with well controlled DM2 and non DM2 women with hip fractures.

Conclusions

Elderly patients with DM2 and severe osteoporosis present with hip fractures at a higher BMD and TBS values compared to non DM2 patients. These differences were significant after adjustment for confounders in all DM2 women and poorly controlled DM2 men with hip fractures, TBS differences were attenuated with the inclusion LS BMD. Further studies are needed to ascertain differences in BMD and TBS in older Southeast Asian DM2 patients with variable glycemic control and severe osteoporosis.

Introduction

Diabetes and osteoporosis are both major health challenges. The global prevalence of diabetes among adults over 18 years has risen from 4.7% in 1980 to 8.5% in 2014 [1]. Worldwide, 1 in 3 women as well as 1 in 5 men over the age of 50 years old will experience osteoporotic fractures [2]. Asians, especially South Asians are predisposed toward DM2 to a greater extent than Caucasians [3]. Singapore has a prevalence of DM2 at 10.5% which is higher than the world average of 8.8%, with estimates of prevalence rising to 15% in 2050 [4]. It is also projected that more than 50% of all osteoporotic fractures will occur in Asia by the year 2050 [5]. Studies have shown that patients with DM2 have a higher risk of fragility fracture, including a 40% to 70% increased hip fracture risk [6,7]. Taken together this implies a burgeoning epidemic of diabetes and fragility fractures, especially in Asia.

Fracture assessment in DM2 patients is complex. DM2 patients have a high fracture risk despite higher bone mineral density (BMD) results [8–11]. These have been attributed to wide ranging factors from types of medication use, presence of DM2 complications and disease duration [12,13]. Furthermore, studies have also shown that there are ethnic differences in the relationship between DM2 and fracture risk [14].

Trabecular bone score (TBS) is a grey–level textural metric that is obtained from lumbar spine dual energy X-ray absorptiometry (DXA) images. Decreased TBS has been found to be associated with an elevated risk for osteoporotic fractures independent of BMD in cohort studies. These results were confirmed by a recent meta-analysis of prospective cohort data [15] and adopted as evidence in position papers [16,17].

Various studies have looked into differences in TBS between DM2 and non-DM2 patients. The majority of studies in ethnic Caucasian populations have demonstrated the use of TBS independently of BMD in predicting lower bone quality in DM2 patients. However, studies in different ethnic groups have shown varying results particularly with respect to gender and age [18,19]. In particular, TBS differences were found in studies in those younger than 65 years old [20,21] and were not seen in the older population [22].

The variable performance of TBS in different population could be explained by the differential effects of clinical variables such as gender, age and osteoporotic hip T-score which has the greatest impact on TBS score in study subjects with DM2 [23]. Other studies have also noted potential impact of body mass index (BMI) and body composition on TBS and lumbar spine (LS) BMD results in older men [24–27]. Furthermore, previous studies have also alluded to the impact of fasting plasma glucose, fasting insulin and HOMA-IR to TBS scores which points to the complexity of bone quality measurements in DM2 subjects [20,22].

Given the variable findings of the relationships of TBS and DM2 in different ethnicities and clinical groups, we sought to validate its performance in elderly Southeast Asian patients with severe osteoporosis. We hypothesized that severe osteoporotic elderly patients with DM2 will have lower bone trabecular scores despite having higher BMD values. In this study, we analyzed an older Southeast Asian cohort with severe osteoporosis presenting with fragility hip fractures in a regional hospital setting in Singapore to study the differences in BMD and TBS in patients with and without DM2.

Subjects and methods

Study population

We conducted a retrospective cross-sectional analysis of all patients admitted to Changi General Hospital with acute fragility hip fractures from 2014–2017. Clinical and demographic data were extracted from the electronic medical records. There were 1378 patients who were admitted for hip fractures during this period. We excluded 66 admissions for recurrent hip fractures and 174 patients who did not have any BMD performed. We also excluded patients with TBS or BMD reports with at least one lumbar level that were not included due to degeneration, instrumentation or previous fractures to minimize the impact of these conditions on LS BMD and TBS performance. Patients with previous exposure to bisphosphonates were also excluded from the analyses to reduce the impact of these drugs on BMD and TBS results. The final study population was 753 subjects. The cohort was divided into different gender and ethnicity according to the recorded ethnicity group and gender documented in the electronic medical records. The diagnosis of DM2 was established using the World Health Organization (WHO) 2019 criteria [28] on the basis of having an HbA1c of 6.5% or greater, or current treatment with oral antidiabetic drugs or insulin. We excluded patients with type 1 diabetes on the basis of the documented medical history of ketoacidosis and age of onset of diabetes before 25 years. DM2 patients were classified according to their HbA1c into well controlled (HbA1c < 7%) and poorly controlled (HbA1c ≥ 7%) DM2. The study was approved by the SingHealth Centralised Institutional Review Board for the waiver of informed consent in this retrospective data analysis.

Clinical evaluation

Baseline clinical data was obtained on all patients presenting for fragility hip fracture. Information regarding medical history, including previous fragility fractures, parental history of fragility fractures, history of rheumatoid arthritis, steroid use of >5 mg/d for >3 months, presence of dementia, and previous amputation were collected from the electronic medical records. Medication data collected includes use of oral antidiabetic drugs, insulin doses (unit/kg), calcium and vitamin D supplementation, oral anti-resorptives and other anti-osteoporosis medications. We further recorded data on smoking status and alcohol intake (> 3 units/d). Anthropometric data was extracted from the electronic medical record and BMI was calculated as weight divided by height squared (kg/m²).

Biochemical evaluation

HbA1c level (%) was determined by immunoturbidimetric assay (Cobas 8000, Roche Diagnostics, Switzerland). HbA1C values within 6 months of admitted date were included in this analysis, there were 9 DM2 women and 1 DM2 man without a recent HbA1C performed that was excluded from the DM2 subgroup comparison and regression analysis. Serum creatinine (umol/L) was measured using indirect ion-specific electrode (Roche Diagnostics, Switzerland) and estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology (CKD-EPI) equation. 25-hydroxyvitamin D was measured by radioimmunoassay (Roche Diagnostics, Switzerland), TSH and FT4 levels were also measured by immunoassay (Abbot affinity, Chicago, USA). Microalbuminuria was determined by an Albumin Creatinine Ratio (ACR) of 30–300 mg/g according to the American Diabetes Association guidelines. Patients with eGFR < 60 ml/min were classified to those with chronic kidney disease (CKD) for the purpose of this analysis.

Bone mineral density and trabecular bone score measurements

All BMD scans were performed on a single densitometer (Hologic QDR Discovery Wi, USA). The region of interest (ROI) was set as the total hip (non-fractured hip site), femoral neck and first to fourth lumbar vertebrae. We excluded vertebrae with fractures or degeneration causing >1 standard deviation greater areal BMD from the immediately adjacent vertebrae in accordance with the International Society for Clinical Densitometry guidelines for individual vertebrae exclusion. The BMD precision error (percentage of coefficient variation) was 1% for the total hip with a least significant change of 0.034 g/cm², 2.3% for the femoral neck with a least significant change of 0.041 g/cm² and 1% for the lumbar spine with a least significant change of 0.022g/cm².

TBS was analysed with iNsight software (Version 3.0.2.0 Medimaps, France) for the same ROI used for BMD measurements by two of the authors (LRC and KF) who were blinded to the patient clinical data. TBS values were calibrated to standard values using the TBS calibration phantom (17 cm thickness and 25% fat mass equivalent), and were adjusted for BMI between 15–37 kg/m². No patients had a BMI above 35 in this study. The least significant change (LSC) for TBS was 4.24%. The short-term precision of TBS calculation was 1.53% (CV) from the same set of DXA scans used to evaluate the precision of the BMD measurements. TBS value of ≥ 1.35 is considered normal, between 1.20 to 1.35 intermediate, and ≤1.20 to be degraded.

Statistical analysis

All analysis was performed using the Statistical Package for Social Science (SPSS version 21.0, Chicago IL, USA). Data were expressed as mean ± standard deviation (SD) for numerical data or frequency (percentage) for categorical data. For missing data, listwise deletion was applied. Sensitivity analysis was conducted to evaluate the robustness of the results in the presence of variables which had more than 20% of missing data. We further imputed data of patients with 3 available levels of LS spine for analysis to assess changes in our results and found consistency in the trend of the results. Interaction term analysis for gender and DM2 status were performed to ascertain the presence of gender differences in TBS and BMD differences. We included the interaction term analysis impact on TBS in S6 Table which found a significant interaction effect between gender and DM2 status.

In univariate analysis, numerical variables were compared with 2-sample T-test and categorical variables were examined with chi-square test or Fisher’s exact test between DM2 and non-DM2 patients. We subdivided the DM2 groups into well controlled (HbA1C < 7%) and poorly controlled (HbA1C ≥ 7%) in comparing them to the non-DM2 patients. Multivariate analysis was used to assess the significant of differences in TBS and BMD measurements between well controlled and poorly controlled DM2 patients. In the first adjusted model, we used analysis of covariance (ANCOVA) models adjusted for age and BMI (Model 1), further adjustments to this model was made to incorporate significant variables such as race, amputation, presence of CKD, 25(OH)D level (Model 2). A third model for analysis of differences in TBS was incorporated with addition of LS BMD to assess if TBS differences were independent of LS BMD value. A p-value of < 0.05 was considered statistically significant.

In a subgroup analysis of DM2 only patients, we analyzed the association between TBS and DXA measurements with treatment variables in well controlled and poorly controlled DM2 patients. The association of TBS and BMD were analyzed for associations with metformin, sulfonylurea, insulin use, presence of microvascular complication (defined by a history of microalbuminuria, renal impairment greater than CKD 3 and a previous history of amputation) and degree of DM2 control. Linear regression was performed to determine the relationship between diabetes status and TBS with other variables to be adjusted. A two-tailed, p-value of <0.05 was considered statistically significant.

Results

Cohort characteristics and demographics

There were 1138 patients admitted with fragility hip fractures from 2014 to 2017 who had BMD performed. After exclusion of patients with incomplete lumbar spine or TBS scores due to degeneration, unavailability of hip BMD due to previous instrumentation to contralateral hip and previous antiresorptive treatments, the final cohort analysed consisted of 753 patients (Fig 1). All patients had their BMD performed within 6 months of the fracture date with a median duration of 14–19 days from their hip fracture (S1 Table), there were no significant differences in duration between hip fracture date and BMD performed in DM2 and non DM2 patients. Of the hip fracture patients studied, 68.8% (n = 518) were women and 31.2% (n = 235) were men. 32.4% (n = 168) of the women had DM2 and 31.1% (n = 73) of the men had DM2. Median duration of diabetes was 8 years in DM2 women and 7 years in DM2 men. Average HbA1C was 7.10% in DM2 women and 7.40% in DM2 men. The majority of the patients in the cohort were of Chinese ethnicity, which is consistent with the general population demographic in Singapore.

Table 1 summarises the baseline characteristic differences of the patients with respect to gender and diabetes status. Women with DM2 who were admitted with fragility hip fractures were younger (76.9 years) compared to the non-DM2 (78.3 years) patients although this was not statistically significant. There was no statistically significant difference in age of men with hip fractures between DM2 and non-DM2 patients. Women with hip fractures who had DM2 had slightly higher BMI compared to non-DM2 patients (22.91 vs 21.75 kg/m²). There was no significant difference in BMI in men between DM and non-DM patients. There were no significant differences in the rate of smoking, alcohol consumption, history of previous fractures, rheumatoid arthritis, steroid exposure, history of secondary osteoporosis and dementia between DM2 and non-DM2 patients. There was a higher rate of amputation in women and men with DM2. Average eGFR in DM2 patients were lower in both women and men, however the presence of CKD (as defined by eGFR <60) was statistically significantly higher only in DM2 women. 25-OH-D level was significantly lower in men with DM2 compared to non-DM2, there were no significant difference in the vitamin D status in women. There were higher rates of calcium and vitamin D supplementation use in DM2 women compared to non-DM2 women. There were no differences in length of stay and mortality between DM2 and non-DM2 women and men.

Table 1. Baseline demographic and clinical variables, trabecular bone score (TBS), bone mineral density (BMD) in patients who are non-diabetic (ND) and those with diabetes mellitus type 2 (DM2).

Patients with DM2 and non-DM2 (ND) are stratified by gender.

Variable	Women (n = 518)			Men (n = 235)
Diabetes Status	ND (n = 350)	DM2 (n = 168)	p-value	ND (n = 162)	DM2 (n = 73)	p-value
Age (years)	78.27 ± 10.39	76.90 ± 8.40	0.139	74.01 ± 13.27	74.16 ± 10.33	0.931
Height (cm)	150.26 ± 6.16	150.72 ± 5.95	0.422	160.37 ± 8.67	160.97 ± 7.98	0.613
Weight (kg)	49.16 ± 11.14	52.16 ± 10.36	0.004	57.02 ± 10.69	57.59 ± 11.25	0.710
BMI (kg/m²)	21.75 ± 4.62	22.91 ± 4.09	0.006	22.24 ± 4.24	22.32 ± 4.61	0.894
BMI group (Asian)
Underweight (< 18.5)	91 (26.1%)	19 (11.3%)	0.002	29 (17.9%)	14 (19.2%)	0.994
Normal (18.5–22.9)	126 (36.1%)	74 (44.0%)		67 (41.4%)	30 (41.1%)
Overweight (23–24.9)	65 (18.6%)	33 (19.6%)		33 (20.4%)	14 (19.2%)
Obese (≥ 25)	67 (19.2%)	42 (25.0%)		33 (20.4%)	15 (20.5%)
Race
Chinese	278 (79.4%)	110 (65.5%)	0.002	131 (80.9%)	47 (64.4%)	0.028
Malay	50 (14.3%)	34 (20.2%)		13 (8.0%)	15 (20.5%)
Indian	9 (2.6%)	14 (8.3%)		8 (4.9%)	5 (6.8%)
Others	13 (3.7%)	10 (6.0%)		10 (6.2%)	6 (8.2%)
Current smoker	6 (1.7%)	0 (0.0%)	0.184	21 (13.0%)	10 (13.7%)	0.877
Alcohol > 3 units per day	3 (0.9%)	0 (0.0%)	0.544	9 (5.6)	2 (2.7)	0.510
Previous fracture	52 (14.9%)	19 (11.3%)	0.272	18 (11.1%)	6 (8.2%)	0.498
Rheumatoid arthritis	2 (0.6%)	0 (0.0%)	1.000	0 (0.0%)	0 (0.0%)	NA
Steroids (> 3 months)	3 (0.9%)	1 (0.6%)	1.000	2 (1.2%)	0 (0.0%)	1.000
Dementia	39 (11.1%)	26 (15.5%)	0.163	20 (12.3%)	9 (12.3%)	0.997
Amputation	1 (0.3%)	8 (4.8%)	0.001	0 (0.0%)	2 (2.7%)	0.096
eGFR	68.29 ± 22.05	59.61 ± 23.98	<0.001	70.70 ± 23.52	61.77 ± 24.57	0.008
CKD (eGFR < 60)	131 (37.4%)	83 (49.4%)	0.010	48 (29.6%)	29 (39.7%)	0.127
25(OH)D (ug/L)	21.97 ± 10.47	20.78 ± 10.97	0.241	25.15 ± 11.93	21.58 ± 9.67	0.026
Calcium & Vitamin D supplementation	96 (27.4%)	64 (38.1%)	0.014	21 (13.0%)	16 (21.9%)	0.081
HbA1c (%)	5.79 ± 0.87	7.10 ± 1.66	<0.001	5.63 ± 0.77	7.40 ± 1.73	<0.001
Length of stay	9 (6, 13)	9 (7, 14)	0.245	9 (7, 13)	10 (7, 17)	0.101
Inpatient mortality	0 (0.0%)	2 (1.2%)	0.105	3 (1.9%)	0 (0.0%)	0.554
TBS	1.19 ± 0.10	1.23 ± 0.10	<0.001	1.31 ± 0.09	1.32 ± 0.10	0.229
BMD Lumbar Spine(g/cm²)	0.72 ± 0.16	0.83 ± 0.15	<0.001	0.87 ± 0.18	0.93 ± 0.18	0.126
T score Lumbar Spine	-2.44 ± 1.33	-1.47 ± 1.34	<0.001	-0.96 ± 1.55	-0.57 ± 1.58	0.082
BMD Total Hip(g/cm²)	0.56 ± 0.12	0.60 ± 0.11	<0.001	0.69 ± 0.13	070 ± 0.13	0.402
T score Total Hip	-3.09 ± 1.01	-2.69 ± 0.92	<0.001	-2.43 ± 0.97	-2.27 ± 1.02	0.251
BMD Femur Neck(g/cm²)	0.47 ± 0.11	0.51 ± 0.10	<0.001	0.57 ± 0.12	0.58 ± 0.13	0.812
T score Femur Neck	-3.22 ± 0.89	-2.83 ± 0.92	<0.001	-2.73 ± 0.90	-2.69 ± 0.97	0.731

Open in a new tab

Numeric data was presented in mean ± SD, except LOS was presented in median (interquartile range).

TBS and BMD differences between women and men with well controlled (HbA1C < 7%) or poorly controlled DM2 (HbA1C ≥ 7%) compared to non DM2 patients with hip fractures

Table 2 summarizes TBS and BMD differences between women and men with well controlled or poorly controlled DM2 compared to non DM2 patients with hip fractures. Poorly controlled DM2 women present with hip fractures at younger age (73.7 ± 8.3 years) compared to those with well controlled DM2 (79.4 ± 7.8 years) and non DM2 women (78.3 ± 10.4 years) with hip fractures. This was similar in men with poorly controlled DM2 men presenting with hip fractures at younger age (71.4 ± 11.1 years) compared to well controlled DM2 (77.0 ± 9.0 years) and non DM2 men (74.1 ± 13.3 years) with hip fractures. Mean HbA1C in the well-controlled DM2 were 6.06 ± 0.65% and 6.18 ± 0.54% in elderly women and men respectively. In the poorly-controlled DM2 elderly women and men, average HbA1C were 8.29 ± 1.67% and 8.62 ± 1.74% respectively. Overall, there was a trend to higher BMD and T-score values in all sites as DM2 control worsened. The differences were significant in all sites in women, but was only significant in the total hip site in men.

Table 2. TBS and BMD values in DM2 and non DM2 women and men stratified to well controlled (HbA1c < 7%) and poorly controlled (HbA1c ≥ 7%).

Variable		Women (n = 509)			Men (n = 234)
Diabetes Status	Non-DM (n = 350)	Hba1c < 7% (n = 85)	Hba1c ≥ 7% (n = 74)	p-value	Non-DM (n = 162)	HbA1c <7% (n = 36)	HbA1c ≥ 7% (n = 36)	p-value
Age	78.27 ± 10.39	79.44 ± 7.77	73.64 ± 8.32	<0.001	74.01 ± 13.27	76.94 ± 8.98	71.36 ± 11.08	0.163
BMI	21.75 ± 4.62	23.19 ± 4.00	22.70 ± 4.28	0.015	22.24 ± 4.24	22.10 ± 3.81	22.65 ± 5.34	0.844
HbA1C (%)		6.06 ± 0.65	8.29 ± 1.67			6.18 ± 0.54	8.62 ± 1.74
TBS	1.19 ± 0.10	1.24 ± 0.09	1.22 ± 0.11	<0.001	1.31 ± 0.09	1.31 ± 0.10	1.34 ± 0.10	0.143
BMD L-spine (g/cm²)	0.72 ± 0.16	0.84 ± 0.15	0.82 ± 0.15	<0.001	0.89 ± 0.18	0.90 ± 0.16	0.95 ± 0.20	0.188
BMD total Hip (g/cm²)	0.56 ± 0.12	0.59 ± 0.11	0.62 ± 0.10	<0.001	0.69 ± 0.13	0.68 ± 0.14	0.73 ± 0.12	0.099
BMD F-neck (g/cm²)	0.47 ± 0.11	0.50 ± 0.10	0.52 ± 0.10	<0.001	0.57 ± 0.12	0.56 ± 0.13	0.60 ± 0.12	0.252
T score Lumbar spine	-2.44 ± 1.33	-1.36 ± 1.29	-1.59 ± 1.35	<0.001	-0.96 ± 1.55	-0.79 ± 1.41	-0.33 ± 1.75	0.106
T score Total Hip	-3.09 ± 1.01	-2.76 ± 0.96	-2.57 ± 0.86	<0.001	-2.43 ± 0.97	-2.48 ± 1.08	-2.01 ± 0.88	0.051
T score F-neck	-3.22 ± 0.89	-2.91 ± 0.92	-2.68 ± 0.89	<0.001	-2.73 ± 0.90	-2.89 ± 1.00	-2.46 ± 0.90	0.129

Open in a new tab

TBS values were higher in well controlled DM2 women with hip fracture (1.22 ± 0.11) compared to poorly controlled DM2 women (1.24 ± 0.09), although both are in the category of intermediate degradation. This was also reflected in the lower LS BMD (0.82 ± 0.15 g/cm²) vs (0.84 ± 0.15 g/cm²) and T-score (-1.59 ± 1.35) vs (-1.36 ± 1.29) compared to those with well controlled DM2, with all being in the same osteopenic category. Non DM2 women with hip fracture had the lowest TBS in the degraded category (1.19 ± 0.10) and LS BMD (0.72 ± 0.15 g/cm²) with T-score at (-2.49 ± 1.35) in the osteoporotic category.

In men with hip fractures, those with poorly controlled DM2 had significantly higher total hip BMD and T-score (0.73 ± 0.12 g/cm², T-score -2.01 ± 0.88) compared with non DM2 men (0.69 ± 0.13 g/cm², T-score -2.43 ± 0.97).

We performed multivariate analysis to assess if differences in TBS and BMD in DM2 patients with well or poorly controlled DM2 versus non DM2 patients with hip fractures could be confounded by other potential variables. Covariates included in the analysis were variables that were found to be significantly different between DM2 and non DM2 patients which included age, BMI category, race, history of amputation, vitamin D status and presence of CKD. The results are shown in Table 3a for well controlled DM2 patients and Table 3b for poorly controlled DM2 patients.

Table 3.

a. Mean Differences in Trabecular Bone Score (TBS) and Bone Mineral Density (BMD) at the Lumbar Spine and Hip between subjects with well-controlled DM2 (HbA1c <7.0%) and non-DM2. b. Mean Differences in Trabecular Bone Score (TBS) and Bone Mineral Density (BMD) at the Lumbar Spine and Hip between subjects with poorly controlled DM2 (HbA1c ≥ 7.0%) and non-DM2.

	Women			Men
	Mean Difference	95% CI	p-value	Mean Difference	95% CI	p-value
TBS
Age and BMI category adjusted	0.05	0.03, 0.07	<0.001	0.002	-0.03, 0.04	0.925
Multivariate adjusted^†	0.05	0.03, 0.07	<0.001	-0.003	-0.04, 0.03	0.857
Multivariate adjusted^{^}	0.02	0.001, 0.05	0.044	-0.003	-0.03, 0.03	0.834
BMD Lumbar Spine
Age and BMI category adjusted	0.10	0.07, 0.13	0.017	0.02	-0.04, 0.08	0.543
Multivariate adjusted^†	0.09	0.06, 0.13	<0.001	-0.0004	-0.06, 0.06	0.990
BMD Total Hip
Age and BMI category adjusted	0.03	0.001, 0.05	0.042	-0.003	-0.05, 0.04	0.880
Multivariate adjusted^†	0.03	0.002, 0.06	0.038	-0.02	-0.06, 0.03	0.514
BMD Femur Neck
Age and BMI category adjusted	0.02	0.001, 0.05	0.042	-0.01	-0.05, 0.03	0.680
Multivariate adjusted^†	0.03	0.002, 0.05	0.033	-0.02	-0.06, 0.02	0.377
TBS
Age and BMI category adjusted	0.02	-0.002, 0.05	0.065	0.03	-0.002, 0.07	0.063
Multivariate adjusted^†	0.03	0.0002, 0.05	0.048	0.04	0.002, 0.07	0.040
Multivariate adjusted^{^}	0.01	-0.02, 0.03	0.678	0.01	-0.02, 0.04	0.364
BMD Lumbar Spine
Age and BMI category adjusted	0.07	0.04, 0.11	<0.001	0.06	-0.001, 0.13	0.055
Multivariate adjusted^†	0.07	0.03, 0.11	<0.001	0.08	0.10, 0.15	0.028
BMD Total Hip
Age and BMI category adjusted	0.04	0.01, 0.06	0.011	0.04	0.0001, 0.09	0.049
Multivariate adjusted^†	0.04	0.01, 0.06	0.014	0.05	0.004, 0.10	0.034
BMD Femur Neck
Age and BMI category adjusted	0.03	0.01, 0.05	0.012	0.02	-0.02, 0.06	0.266
Multivariate adjusted^†	0.03	0.004, 0.05	0.025	0.02	-0.03, 0.06	0.481

Open in a new tab

^†Model was adjusted for age, BMI category, race, amputation, eGFR < 60, 25(OH)D.

^ Model was adjusted for age, BMI category, race, amputation, eGFR < 60, 25(OH)D, BMD LS.

Differences in TBS and BMD L spine remained statistically significant despite adjustments with covariates in well and poorly controlled DM2 women with hip fractures. TBS differences became attenuated upon inclusion of LS BMD into the model and remained only significantly different in well controlled DM2 women with hip fractures (Table 3a).

Differences in TBS and BMD at all sites were only significantly different in poorly controlled DM2 men with hip fractures after adjustments with covariates (Table 3b). However, these TBS differences became attenuated upon inclusion of LS BMD into the model and became non-significant in poorly controlled DM2 men with hip fractures.

Association of TBS with LS BMD and BMI

S1 Table shows that there is a significant correlation between BMD and TBS values with the strongest correlation between BMD LS and TBS values. These were equally significant in both elderly men and women with or without DM2 who had hip fractures. These correlations remained significant with adjustments of age, BMI, vitamin D status and presence of CKD.

TBS results were also negatively correlated to BMI, although this did not reach statistical significance in both men and women. Conversely BMD was significantly positively correlated with BMI at all sites (p<0.05) for both men and women (S2 Table). Differences in TBS were significant in the BMI category of underweight (< 18.5 kg/m²) for both men and women with DM2. In the normal BMI category (18.5–22.9 kg/m²), only DM2 women showed significantly higher TBS values compared to non-DM2 women. These differences were not seen in the overweight and obese categories (S1 Fig).

Subgroup analysis in DM2 patients between well controlled vs poorly controlled DM2 patients with hip fractures

We analysed differences between well controlled and poorly controlled DM2 patients to elucidate potential contributors to differences in their BMD and TBS. There were significantly higher percentage of poorly controlled DM2 women on metformin (83.8%) vs those with well controlled DM2 (68.2%). There were also higher percentage of insulin use among poorly controlled DM2 women and men compared to those who are well controlled, however there were no significant difference in insulin unit per kg administered. We assessed for associations between the use of insulin, metformin or sulphonylurea, presence of microvascular complications, duration of DM2 and well controlled vs poorly controlled DM2 to TBS scores or BMD results in both women and men with DM2. Our analysis showed in DM2 women with hip fractures, poorer DM2 control was associated with lower TBS, whereas in men with hip fractures, poorer DM2 control was associated with higher TBS scores. We found that DM2 women with insulin use had significantly higher TBS score, lumbar spine and femoral neck BMD after adjustment for confounding variables. Of note the total number of insulin users in our cohort were small at n = 29 (S5a Table). When we included LS BMD into analysis of TBS associations with DM2 variables, only LS BMD remained significantly associated with TBS implying that TBS associations with LS BMD was stronger than any DM2 variables (S5b Table). A separate analysis with HbA1c as a continuous variable did not show any significant association of HbA1C with TBS and BMD values. To reduce the probability of a skewed distribution of HbA1C affecting this result, we repeated the analysis with a log transformation of the HbA1c to reduce its variability, however the results were similar (S7 Table).

Discussion

Our study adds to the growing body of literature on the characteristics of BMD and TBS in DM2 patients, especially in older patients with severe osteoporotic hip fractures of different ethnic backgrounds with well controlled or poorly controlled DM2. In this study, we looked at older Southeast Asian patients with severe osteoporosis who have sustained fragility hip fractures to assess differences in BMD and TBS in DM and non-DM2 patients. Elderly patients over 70 years of age are a major source of more than 70% of health care cost and contributes 45% to 60% of all major fractures [29]. In this regard our findings may be most applicable to the oldest and most severe of osteoporosis patients. The DM2 prevalence in this study of around 30% is consistent with our national data for an older population [30]. DM2 women on average had hip fractures at a younger age (76.9 years) compared to non-DM2 (78.3 years) despite higher BMD and TBS values. In our cohort, men present on average at a younger age (74.0 years) with hip fractures compared to women. Inpatient mortality and length of stay did not differ between DM2 and non-DM2 patients. Our results underscore current understanding in the limitation of BMD and TBS use in assessing bone quality in elderly patients with osteoporosis and DM2. Although all patients with hip fractures present with average femur neck (FN) T-scores at osteoporotic range, patients with DM2 and hip fractures had consistently higher BMD and TBS values compared to non DM2 patients with hip fractures. These differences were statistically significant in between non DM2 elderly women and well controlled or poorly controlled DM2 elderly women despite adjustment for potential significant confounders. In elderly men with hip fractures, these differences were only significant between non DM2 and poorly controlled DM2 after adjustments for microvascular complications and vitamin D status. Although other studies demonstrated that TBS could be a useful adjunct in assessing bone quality of DM2 patients, we found that in our cohort of older severe osteoporotic patients that differences in TBS values showed similar trend to BMD LS Spine values. Differences in TBS were not independent of LS BMD values in men and women regardless of the degree of DM2 control. Within DM2 patients, elderly women with poorly controlled DM2 had lower TBS values compared to those that were well controlled but this trend were reversed in elderly DM2 men. It is interesting to note that when we looked at HbA1C as a continuous variable, differences in TBS and BMD did not exhibit significant associations.

There may be a few explanations for the differences of the results observed in our study compared to previously reported studies in the literature. Ours is a much older population with severe osteoporosis with fragility hip fractures, significantly lower BMI and higher percentages of poorly controlled DM2 patients. The average age in the study population was 79 years for women and 74 years for men. Previous studies have shown an accelerated loss of BMD and TBS with age, with the rate of TBS decline after 65 years increasing by 50% [31]. As there are no current normative TBS data for healthy adults in Singapore, if we were to reference a geographically close South East Asian nation (Thailand), our TBS results are within the range of their normative data for population between 70–80 years old [25]. Normative TBS values in the 70–80 y/o female were recorded at 1.218. The TBS value in this population of severe osteoporotic elderly female non DM2 patient was recorded at 1.19 and DM2 was 1.23. The same normative data in elderly (70–80 years) Thai male documented a TBS of 1.302, in our study of severe osteoporotic elderly males TBS were 1.31 in non DM2 and 1.32 DM2 men. Further larger and longitudinal studies should focus on impact of DM2 in this elderly population to understand these differences better.

In addition, in this study we found that HbA1C as a continuous variable did not exhibit any impact on TBS and BMD values. However, separation of patient analysis based on HbA1C of 7% into well controlled and poorly controlled DM2 found that TBS were significantly lower in women with poorly controlled DM2 but was persistently higher in men with poorly controlled DM2. Future studies could consider assessing different levels of glycemic control threshold and its impact on BMD and TBS measurements in women and men with DM2. It is also worth noting that differences between TBS, LS BMD and total hip BMD of poorly controlled DM2 men and non DM2 men became statistically significant only after accounting for these important clinical variables. These also further imply the need for future study of BMD and TBS differences in DM2 population to take into account the contribution of DM2 disease control and complications. These factors may also explain our differing results from a previous study in elderly Japanese men with average of 72.9 years which compared DM2 patient with average HbA1C of 6.5% to non DM2 patients and did not find any significant differences in TBS results [22]. This same study also further demonstrated the potential contributor of impaired glycemic control and insulin resistance Other potential explanation for the differences in the results of our study is the lower BMI of our elderly population (21.8 kg/m² in non-DM and 22.9 kg/m² in DM women, 22.2 kg/m² and 22.3 kg/m² in non-DM2 and DM2 men respectively). Other studies predominantly looked at relatively younger DM2 populations of average age around 65 years with higher BMI. The Manitoba [32] study cohort had BMI of DM2 and non-DM2 patients around 29.7 vs 26.7 kg/m², while another study from Korea [20] had average cohort ages between 62–66 years old with BMI of in women of 25.3 kg/m² (DM) vs 24.5 kg/m² (non-DM) and in men of 24.4 km/m² (DM) vs 23.5 kg/m² (non-DM). In the Vietnamese study the cohort average age was 60 years old in DM2 women with an average BMI of 25.0 kg/m² and 56 years old in DM2 men with an average BMI of 25.7 kg/m² [21]. Recent analyses have indicated that TBS is inversely related to BMI and abdominal fat, it may well be that our DM2 population of relatively lower BMI may not exhibit these effects in lowering TBS. This was also evident in our results that DM2 women and men with underweight BMI and DM2 women with normal BMI had significantly higher TBS scores than non-DM2, with no differences found in the overweight and obese categories.

Previous studies looking at the effects of insulin on BMD have shown differing results with some studies finding that exogenous insulin increases [33], reduces [34] or has a neutral effect on BMD [35,36] (S3 Table). Interestingly, although in our univariate analysis there were no demonstrable differences between TBS and BMD of DM2 women who are on insulin and non-insulin users, there appears to a statistically significant association of higher LS and femoral neck BMD after adjustments for age, BMI, disease duration, HbA1C, microvascular complications and other oral hypoglycaemia agents in the multivariate model. However, as our study is cross-sectional analysis with small number of insulin users, further observational follow up study will be needed to understand the relationship between the use of insulin in DM2 and BMD in the elderly DM2 women.

Taken together, bone quality assessment in DM2 patients remains a complex issue. Differences in DM2 disease control and complication, together with age and BMI may contribute differently to the results of current bone quality measurement differences. These differences were most significant in elderly DM2 hip fracture patients with poorly controlled DM2 and would be important for clinicians who are looking after elderly DM2 patients to be aware of.

Our study has a few strengths in particular our large number of diabetic patients and completeness of data with regards to DM2 duration, medications use and rates of documented microalbuminuria and history of amputation. Our study cohort also consisted of a uniquely older patients who presented with fragility fractures and in this regard our results would be applicable to patients at highest risk of fracture. We were also able to rule out other potential causes of secondary osteoporosis in our study and were also able to document 25(OH)D levels and data regarding previous treatment with anti-resorptives and supplementation with calcium and vitamin D. We excluded those with history of vertebral fractures, degeneration and surgical instrumentation from falsely elevating the BMD and TBS results. To further reduce the potential interference of previous anti-resorptive use in the BMD, we excluded these patients from our analysis as well.

The limitations of this study include the recruitment of only elderly hip fracture patients in our study. These patients have severe osteoporosis and as such our results may not be applicable to younger patients without severe osteoporosis. All BMD scans were performed on a single densitometer (Hologic QDR Discovery Wi, USA) and TBS was analysed with iNsight software (Version 3.0.2.0 Medimaps, France), previous studies have shown that this first version of TBS may be more significantly impacted by BMI compared to the newer version [37]. Hence whether our results may be applicable to the newer version of TBS is not known. There was also a lack of age matched control and local normative TBS data in the elderly population to allow further inference of the effect of DM2 and hip fracture on TBS and BMD results. We also had a lack of retinopathy data that would be important to be included for microvascular complications; however studies have shown that microalbuminuria is also a reliable indicator of diabetic retinopathy [38–40]. Our study also did not address the issue of falls and sarcopenia, both of which are important factors in the contribution of fracture risks in DM2 patients [41]. The cross-sectional nature of the study also cannot account for the potential dynamic changes in bone that occurs with progression of DM2. The maximum duration of DM2 in our patients was 9 years which did not allow us to study the effect of a longer duration of DM2 (>10 years) and its impact on TBS. We also did not have the glycemic status of the non-DM2 patients and would be unable to rule out patients with impaired fasting glucose or undiagnosed DM2 in the non-DM2 cohort.

In conclusion elderly patients with DM2 and severe osteoporosis present with hip fractures at a higher BMD and TBS values compared to non DM2 patients. These differences were significant after adjustment for confounders in all DM2 women and poorly controlled DM2 men with hip fractures, TBS differences were attenuated with the inclusion LS BMD. Further studies are needed to ascertain differences in BMD and TBS in older Southeast Asian DM2 patients with variable glycemic control and severe osteoporosis.

Supporting information

S1 Fig. Relationship of trabecular bone score (TBS) and body mass index (BMI) in women and men with and without DM2.

(TIF)

Click here for additional data file.^{(333.2KB, tif)}

S1 Table. Duration (days) from date of admission to BMD analysis stratified by gender and DM2 status.

(DOCX)

Click here for additional data file.^{(21.4KB, docx)}

S2 Table. Correlation between TBS and BMD in DM2 and non DM2 patients stratified by gender.

(DOCX)

Click here for additional data file.^{(14KB, docx)}

S3 Table. Correlation between BMI with TBS and BMD in DM2 and nonDM2 patients stratified by gender.

(DOCX)

Click here for additional data file.^{(13.5KB, docx)}

S4 Table. Demographic and clinical variables of DM2 patients subdivided into gender and DM2 control.

(DOCX)

Click here for additional data file.^{(17.5KB, docx)}

S5 Table

a: Relationship of trabecular bone score with diabetes mellitus (DM2) medications, complications and glycaemia control (β Coefficient). b: Relationship of Trabecular Bone Score adjusted for Lumbar Spine BMD with Diabetes Mellitus (DM2) Medication, Complication and Glycaemia Control (β Coefficient).

(DOCX)

Click here for additional data file.^{(16.1KB, docx)}

S6 Table. Interaction term analysis between TBS and DM2 status with gender.

(DOCX)

Click here for additional data file.^{(12.9KB, docx)}

S7 Table. Relationship of trabecular bone score with diabetes mellitus (DM2) medications, complications and glycaemia control (β Coefficient).

(DOCX)

Click here for additional data file.^{(15KB, docx)}

Acknowledgments

We thank the help of the Valued Care Hip Fracture Program in Changi General Hospital, Singapore for their help in providing us with patient data.

Data Availability

All relevant data are within the manuscript and its Supporting information files. Data cannot be shared publicly because of potentially identifying and patient sensitive information. Restriction of data access to study team as approved by the SingHealth Institutional Review Board - IRB (CIRB Ref 2017/2563) Data are available from the Singhealth IRB (contact for request of access: irb@singhealth.com.sg).

Funding Statement

The authors received no specific funding for this work.

References

1.Sarwar N, Gao P, Seshasai SR, Gobin R, Kaptoge S, Di Angelantonio E, et al. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet (London, England). 2010;375(9733):2215–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Kanis JA, Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: Synopsis of a WHO report. Osteoporosis International. 1994;4(6):368–81. 10.1007/BF01622200 [DOI] [PubMed] [Google Scholar]
3.Chan JC, Malik V, Jia W, Kadowaki T, Yajnik CS, Yoon KH, et al. Diabetes in Asia: epidemiology, risk factors, and pathophysiology. Jama. 2009;301(20):2129–40. 10.1001/jama.2009.726 [DOI] [PubMed] [Google Scholar]
4.Phan TP, Alkema L, Tai ES, Tan KHX, Yang Q, Lim W-Y, et al. Forecasting the burden of type 2 diabetes in Singapore using a demographic epidemiological model of Singapore. BMJ Open Diabetes Research & Care. 2014;2(1):e000012 10.1136/bmjdrc-2013-000012 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Cooper C, Campion G, Melton LJ 3rd. Hip fractures in the elderly: a world-wide projection. Osteoporos Int. 1992;2(6):285–9. 10.1007/BF01623184 [DOI] [PubMed] [Google Scholar]
6.Janghorbani M, Van Dam RM, Willett WC, Hu FB. Systematic review of type 1 and type 2 diabetes mellitus and risk of fracture. Am J Epidemiol. 2007;166(5):495–505. 10.1093/aje/kwm106 [DOI] [PubMed] [Google Scholar]
7.Janghorbani M, Feskanich D, Willett WC, Hu F. Prospective study of diabetes and risk of hip fracture: the Nurses’ Health Study. Diabetes Care. 2006;29(7):1573–8. 10.2337/dc06-0440 [DOI] [PubMed] [Google Scholar]
8.Bonds DE, Larson JC, Schwartz AV, Strotmeyer ES, Robbins J, Rodriguez BL, et al. Risk of fracture in women with type 2 diabetes: the Women’s Health Initiative Observational Study. J Clin Endocrinol Metab. 2006;91(9):3404–10. 10.1210/jc.2006-0614 [DOI] [PubMed] [Google Scholar]
9.de L II, van der Klift M, de Laet CE, van Daele PL, Hofman A, Pols HA. Bone mineral density and fracture risk in type-2 diabetes mellitus: the Rotterdam Study. Osteoporos Int. 2005;16(12):1713–20. 10.1007/s00198-005-1909-1 [DOI] [PubMed] [Google Scholar]
10.Hanley D, Brown J, Tenenhouse A, Olszynski W, Ioannidis G, Berger C, et al. Associations Among Disease Conditions, Bone Mineral Density, and Prevalent Vertebral Deformities in Men and Women 50 Years of Age and Older: Cross-Sectional Results From the Canadian Multicentre Osteoporosis Study. Journal of Bone and Mineral Research. 2003;18(4):784–90. 10.1359/jbmr.2003.18.4.784 [DOI] [PubMed] [Google Scholar]
11.Melton LJ 3rd, Riggs BL, Leibson CL, Achenbach SJ, Camp JJ, Bouxsein ML, et al. A bone structural basis for fracture risk in diabetes. J Clin Endocrinol Metab. 2008;93(12):4804–9. 10.1210/jc.2008-0639 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Chen H, Li X, Yue R, Ren X, Zhang X, Ni A. The effects of diabetes mellitus and diabetic nephropathy on bone and mineral metabolism in T2DM patients. Diabetes research and clinical practice. 2013;100(2):272–6. 10.1016/j.diabres.2013.03.007 [DOI] [PubMed] [Google Scholar]
13.Krakauer JC, McKenna MJ, Buderer NF, Rao DS, Whitehouse FW, Parfitt AM. Bone loss and bone turnover in diabetes. Diabetes. 1995;44(7):775–82. 10.2337/diab.44.7.775 [DOI] [PubMed] [Google Scholar]
14.Looker AC, Eberhardt MS, Saydah SH. Diabetes and fracture risk in older U.S. adults. Bone. 2016;82:9–15. 10.1016/j.bone.2014.12.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.McCloskey EV, Oden A, Harvey NC, Leslie WD, Hans D, Johansson H, et al. A Meta-Analysis of Trabecular Bone Score in Fracture Risk Prediction and Its Relationship to FRAX. J Bone Miner Res. 2016;31(5):940–8. 10.1002/jbmr.2734 [DOI] [PubMed] [Google Scholar]
16.Harvey NC, Gluer CC, Binkley N, McCloskey EV, Brandi ML, Cooper C, et al. Trabecular bone score (TBS) as a new complementary approach for osteoporosis evaluation in clinical practice. Bone. 2015;78:216–24. 10.1016/j.bone.2015.05.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Silva BC, Broy SB, Boutroy S, Schousboe JT, Shepherd JA, Leslie WD. Fracture Risk Prediction by Non-BMD DXA Measures: the 2015 ISCD Official Positions Part 2: Trabecular Bone Score. J Clin Densitom. 2015;18(3):309–30. 10.1016/j.jocd.2015.06.008 [DOI] [PubMed] [Google Scholar]
18.Rianon N, Ambrose CG, Buni M, Watt G, Reyes-Ortiz C, Lee M, et al. Trabecular Bone Score Is a Valuable Addition to Bone Mineral Density for Bone Quality Assessment in Older Mexican American Women With Type 2 Diabetes. J Clin Densitom. 2018;21(3):355–9. 10.1016/j.jocd.2018.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Jain RK, Narang DK, Hans D, Vokes TJ. Ethnic Differences in Trabecular Bone Score. J Clin Densitom. 2017;20(2):172–9. 10.1016/j.jocd.2016.04.003 [DOI] [PubMed] [Google Scholar]
20.Kim JH, Choi HJ, Ku EJ, Kim KM, Kim SW, Cho NH, et al. Trabecular bone score as an indicator for skeletal deterioration in diabetes. J Clin Endocrinol Metab. 2015;100(2):475–82. 10.1210/jc.2014-2047 [DOI] [PubMed] [Google Scholar]
21.Ho-Pham LT, Tran B, Do AT, Nguyen TV. Association between pre-diabetes, type 2 diabetes and trabecular bone score: The Vietnam Osteoporosis Study. Diabetes research and clinical practice. 2019;155:107790 10.1016/j.diabres.2019.107790 [DOI] [PubMed] [Google Scholar]
22.Iki M, Fujita Y, Kouda K, Yura A, Tachiki T, Tamaki J, et al. Hyperglycemia is associated with increased bone mineral density and decreased trabecular bone score in elderly Japanese men: The Fujiwara-kyo osteoporosis risk in men (FORMEN) study. Bone. 2017;105:18–25. 10.1016/j.bone.2017.08.007 [DOI] [PubMed] [Google Scholar]
23.Martineau P, Leslie WD, Johansson H, Harvey NC, McCloskey EV, Hans D, et al. In which patients does lumbar spine trabecular bone score (TBS) have the largest effect? Bone. 2018;113:161–8. 10.1016/j.bone.2018.05.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Leslie WD, Krieg MA, Hans D. Clinical factors associated with trabecular bone score. Journal of clinical densitometry: the official journal of the International Society for Clinical Densitometry. 2013;16(3):374–9. 10.1016/j.jocd.2013.01.006 [DOI] [PubMed] [Google Scholar]
25.Sritara C, Thakkinstian A, Ongphiphadhanakul B, Amnuaywattakorn S, Utamakul C, Akrawichien T, et al. Age-Adjusted Dual X-ray Absorptiometry-Derived Trabecular Bone Score Curve for the Lumbar Spine in Thai Females and Males. J Clin Densitom. 2016;19(4):494–501. 10.1016/j.jocd.2015.05.068 [DOI] [PubMed] [Google Scholar]
26.Evans AL, Paggiosi MA, Eastell R, Walsh JS. Bone Density, Microstructure and Strength in Obese and Normal Weight Men and Women in Younger and Older Adulthood. Journal of Bone and Mineral Research. 2015;30(5):920–8. 10.1002/jbmr.2407 [DOI] [PubMed] [Google Scholar]
27.Langsetmo L, Vo TN, Ensrud KE, Taylor BC, Cawthon PM, Schwartz AV, et al. The Association Between Trabecular Bone Score and Lumbar Spine Volumetric BMD Is Attenuated Among Older Men With High Body Mass Index. J Bone Miner Res. 2016;31(10):1820–6. 10.1002/jbmr.2867 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.WHO. Classification of Diabetes Mellitus 2019 WHO. 2019.
29.Tatangelo G, Watts J, Lim K, Connaughton C, Abimanyi-Ochom J, Borgström F, et al. The Cost of Osteoporosis, Osteopenia, and Associated Fractures in Australia in 2017. J Bone Miner Res. 2019;34(4):616–25. 10.1002/jbmr.3640 [DOI] [PubMed] [Google Scholar]
30.Phan TP, Alkema L, Tai ES, Tan KHX, Yang Q, Lim W-Y, et al. Forecasting the burden of type 2 diabetes in Singapore using a demographic epidemiological model of Singapore. BMJ Open Diabetes Research & Care. 2014;2(1):e000012 10.1136/bmjdrc-2013-000012 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Dufour R, Winzenrieth R, Heraud A, Hans D, Mehsen N. Generation and validation of a normative, age-specific reference curve for lumbar spine trabecular bone score (TBS) in French women. Osteoporosis International. 2013;24(11):2837–46. 10.1007/s00198-013-2384-8 [DOI] [PubMed] [Google Scholar]
32.Leslie WD, Aubry-Rozier B, Lamy O, Hans D, Manitoba Bone Density P. TBS (trabecular bone score) and diabetes-related fracture risk. J Clin Endocrinol Metab. 2013;98(2):602–9. 10.1210/jc.2012-3118 [DOI] [PubMed] [Google Scholar]
33.Thrailkill KM, Charles K. Lumpkin J, Bunn RC, Kemp SF, Fowlkes JL. Is insulin an anabolic agent in bone? Dissecting the diabetic bone for clues. American Journal of Physiology-Endocrinology and Metabolism. 2005;289(5):E735–E45. 10.1152/ajpendo.00159.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Ruppert K, Cauley J, Lian Y, Zgibor JC, Derby C, Solomon DH. The effect of insulin on bone mineral density among women with type 2 diabetes: a SWAN Pharmacoepidemiology study. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 2018;29(2):347–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.de Waard EAC, Koster A, Melai T, van Geel TA, Henry RMA, Schram MT, et al. The association between glucose metabolism status, diabetes severity and a history of fractures and recent falls in participants of 50 years and older-the Maastricht Study. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 2016;27(11):3207–16. 10.1007/s00198-016-3645-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Starup-Linde J, Gregersen S, Vestergaard P. Associations with fracture in patients with diabetes: a nested case–control study. BMJ Open. 2016;6(2):e009686 10.1136/bmjopen-2015-009686 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Schacter GI, Leslie WD, Majumdar SR, Morin SN, Lix LM, Hans D. Clinical performance of an updated trabecular bone score (TBS) algorithm in men and women: the Manitoba BMD cohort. Osteoporos Int. 2017;28(11):3199–203. 10.1007/s00198-017-4166-1 [DOI] [PubMed] [Google Scholar]
38.Kim SM, Moon YW, Lim SJ, Yoon BK, Min YK, Lee DY, et al. Prediction of survival, second fracture, and functional recovery following the first hip fracture surgery in elderly patients. Bone. 2012;50(6):1343–50. 10.1016/j.bone.2012.02.633 [DOI] [PubMed] [Google Scholar]
39.Lee WJ, Sobrin L, Lee MJ, Kang MH, Seong M, Cho H. The Relationship Between Diabetic Retinopathy and Diabetic Nephropathy in a Population-Based Study in Korea (KNHANES V-2, 3)Relationship Between DR and DN in Korea. Investigative Ophthalmology & Visual Science. 2014;55(10):6547–53. [DOI] [PubMed] [Google Scholar]
40.Lee M-K, Han K-D, Lee J-H, Sohn S-Y, Hong O-K, Jeong J-S, et al. Normal-to-mildly increased albuminuria predicts the risk for diabetic retinopathy in patients with type 2 diabetes. Scientific Reports. 2017;7(1):11757 10.1038/s41598-017-11906-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Compston J. Type 2 diabetes mellitus and bone. Journal of Internal Medicine. 2018;283(2):140–53. 10.1111/joim.12725 [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0241616.r001

Decision Letter 0

Robert Daniel Blank

16 Jun 2020

PONE-D-20-14006

Bone Mineral Density and Trabecular Bone Score in Older Type 2 Diabetes Southeast Asian Patients with Osteoporotic Hip Fractures

PLOS ONE

Dear Dr. Gani,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Reviewers are particularly concerned about notable features of your study population and the differences between them and other cohorts of patients with DM2 who have been studied regarding low trauma fracture. In a revised MS, please be explicit about the unique features of your cohort and what conclusions, if any, might be applicable more generally. How will reading this MS help either clinicians or investigators think about fracture risk in patients with DM?

Please submit your revised manuscript by Jul 31 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Robert Daniel Blank, MD, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records used in your retrospective study. Specifically, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.

3. Please note that according to our submission guidelines (http://journals.plos.org/plosone/s/submission-guidelines), outmoded terms and potentially stigmatizing labels should be changed to more current, acceptable terminology. For example: “Caucasian” should be changed to “white” or “of [Western] European descent” (as appropriate).

4. Please include your tables as part of your main manuscript and remove the individual files. Please note that supplementary tables should remain as separate "supporting information" files.

5. Please include a caption for each figure.

6. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study presents a detailed analysis of TBS, and lumbar spine, femoral neck BMD in a cohort of women and men of Asian origin with Diabetes compared to those without diabetes. The study is well conducted. However, the main limitation of the study is that the cohort analysed included only people with fracture. Thus, the contribution of TBS to fracture risk could not be assessed.

Comments:

Introduction is very long and has some very detailed descriptions for ethnic differences which are more suited for the Discussion section of the manuscript. I suggest reducing the Introduction and only presenting a succinct summary of the ethnic differences to build the hypothesis.

Methods: Some details related to study design are missing. For example, it is not clear whether hip fracture patients were invited for a clinical evaluation or their records were simply extracted from examination of medical records.

When was BMD measurement performed in relation to hip fracture (prior, after) and at what interval?

How was the duration of diabetes established?

Minor:

Figure 1 - should include the number of people with DM for all stages.

Reviewer #2: Bone mineral density (BMD) and trabecular bone score (TBS) assessed from DXA each provide independent information regarding skeletal status and fracture risk. Type 2 diabetes is increasingly recognized as a risk factor for fracture despite higher BMD, though the mechanism for this is complex. Reduced TBS in women with type 2 diabetes has been noted in many studies including an unreferenced meta-analysis (PMID 31214749) and may explain some of the excess fracture risk. The same meta-analysis did not detect altered TBS in men.

The current retrospective cross-sectional analysis was performed in women and men admitted with hip fracture who subsequently underwent DXA. Type 2 diabetes mellitus and covariates were identified from review of medical records. The authors found that TBS was higher (not lower) in DM2 women with hip fracture compared to those without DM2, and could not identify a TBS difference in men. BMD was increased at all skeletal sites in women with DM2 versus those without DM2; again no differences were observed in DM2 men.

General comments:

Overall, the report is well written and the analytical approach is appropriate. Increased BMD in DM2 is already widely known and this study contributes relatively little in that regard. The authors speculate on why their results differ from those reported elsewhere, and suggest this may relate to the “Asian diabetes phenotype” with a different pattern of low BMI and visceral adiposity, though this does not explain why their results differ from other Asian populations none of which showed higher TBS in DM2 women. Is it not possible that this reflects unique characteristics in the hip fracture population? The TBS implications of the study are therefore uncertain. In the absence of clinically relevant outcomes (fractures) the authors can do little more than conclude “Further studies are needed to ascertain the differences in BMD and TBS in Southeast Asian DM2 patients”.

Specific comments:

1. The number of individuals with hip fracture who did not undergo BMD testing is uncertain. This is important to report since if the included patients differ from those who were excluded this might bias results. Also the interval between hip fracture and DXA is not stated and needs to be clarified, since rapid BMD loss (especially from the hip) is known to occur following hip fracture. Finally, the reference data used for T-score reporting needs to be stated (including whether these are gender-neutral or gender-matched).

2. The study population excluded “patients with TBS or BMD reports with at least one lumbar level that were not included due to degeneration, instrumentation or previous fractures”. I am not sure if this is correctly worded or not, but exclusion of cases based upon a single vertebral level showing structural artifact would be very restrictive, especially in an elderly population with hip fractures where some degree of degenerative/structural change would be almost universal. Please clarify.

3. The authors state that they excluded individuals with previous bisphosphonate exposure. However bisphosphonates are widely used as treatment following hip fracture. This suggests either that treatment rates were extremely low (indeed only 18 individuals exclude due to antiresorptive treatment) or that the DXA testing was performed very shortly after hip fracture before treatment was initiated. Please clarify.

4. Approximately 1/3 of the hip fracture patients had DM2. Was this by design or does it reflect the true prevalence of diabetes in this population?

5. The authors do not adequately discuss the confounding effect that abdominal tissue thickness has on TBS measurements. “Raw” TBS decreases in relation to increasing abdominal tissue thickness and therefore the software algorithm includes a correction based upon BMI. This BMI adjustment may or may not be optimal for individuals with DM2, may not be applicable to populations that have a different pattern of abdominal adiposity, and differs for Hologic and GE DXA scanners (Hologic scanners are particularly sensitive to the effects of BMI.). Therefore, it is uncertain whether the reported TBS findings are a true reflection of skeletal properties or limitations in the TBS algorithm. A future version of the TBS algorithm that directly corrects for tissue thickness may be helpful, though this is not currently available for use.

6. The authors highlight differences between women and men in terms of how DM2 affects the BMD and TBS measurements. Significant differences were seen in women but not in men, this also reflects the relatively larger number of women versus men. Indeed, the 95% CIs for men in Table 2 (mean differences for TBS and BMD) appear to include the point estimates for women. It is therefore uncertain whether the gender difference is real or simply reduced power in men. A formal interaction analysis (gender x DM2) would clarify this.

7. Figure 2 is not very insightful and could be moved to the Supplementary section.

8. Please provide units for the coefficients reported in Supplementary Table 3.

9. Some references are repeated (#21 and #34, maybe others).

Reviewer #3: The manuscript is well written, topical and unique in the sense they have study severe OP patient (By hip fracture). It is however a pity that the study does not include non-T2DM-non-severe-OP matched for age controls. If you can add such controls, it will really add values to your study. If you cannot, then at least I would be comment on the value of TBS in severe OP with or without T2DM. Indeed, in both case, for example the Spine BMD corresponds to the osteopenic category while TBS corresponds to the degraded classification. Normal population at that age would have been in the partially degraded categories. Not giving such information is misleading as it give the wrong impression that TBS in such population (including T2DM) is normal while it is not. Along the same line, be careful in comparing your results with other studies as very few of them have Severe OP (by Hip fracture) and cannot be compared directly. Also to avoid any confusion you may add systematically “severe osteoporotic” e.g. non-DM2 vs DM2 severe osteoporotic patients.

Another very important point: While the outcomes contribute to a better understanding of the relationship between T2DM and TBS, the authors could go several steps further in the analysis to take into account current knowledge. Indeed, it has been repeatedly reported that TBS is lower in pre-T2DM patients or in uncontrolled T2DM patients as compared to controlled T2DM. Based on your supplementary table 2 you have a high number of patients with HbA1c patient above 7%. I would then repeat the analysis (and corresponding adjustment – tables 1&2) with three groups: non-DM2 vs DM2(HbA1c<7%) vs DM2(HbA1c>7%) severe osteoporotic patients.

Minor comments:

The mention of the study in HK population (ref 23) is coming unexpectedly as it is not on T2DM patient. What is the message you want to pass over here?

It might be worth mention in the introduction as stated above the impact of pre-T2DM and non-controlled T2DM on TBS. These results can better explained “variable performance of TBS in different population” than BMI as in most of the study BMI has been used as co-founding adjustment variable.

You are excluding patient with previous exposure to bisphosphonate. Is IT not the case for the other OP treatments? Does it mean that all the other patients with severe OP are not treated? Please explain.

Precision assessment for both BMD and TBS ae based on which population? Which age and number? If your CV for TBS is effectively 1% then your LSC should be 1.96 x root-mean-squared 2 x CV = 2.77% and not 4.24%. Can you explain the discrepancy?

The authors are performing multiple adjustment. Are you use that you are not over adjusting as many of the cofounding variables are not significantly different between groups …Wouldn’t you prefer to be more clinically strategic in the choice of the adjustment variable?

In table 2, for TBS it would also make an adjustment for Age, BMI and BMD lumbar spine, as we want to investigate the independence of TBS association between DM2 and non-DM2 from density.

Results / discussion:

See some of my major comments above.

I would still set the context where both non-DM2 and DM2 severe OP patients by hip fracture have degraded structure (TBS =< 1.23) while the spine BMD belongs to osteopenia category. Such results should be compared to expected BMD and TBS for that age (normal condition – reference curve).

You make a substantial paragraph on the impact of BMI on TBS while this one is barely 0.2% (while 17.6% for BMD). It seems to be that your results do not support the hypothesis that BMI would be a role here… You may reduce this one and focus on outcomes from new analysis to be performed (adjustment by spine BMD, < > HbA1c etc…)

Your report significant relationship between TBS and insulin for women in suppl. Table 3. But it is not mentioned at all in the text. On purpose?

In your limitations, I would clearly states that there is no age matched controls.

In conclusion, … I would add something along these lines: TBS and BMD in older DMs… are higher than non-DM2… in severe OP despite multiple adjustment. However overall values for hip BMD and spine TBS corresponds to osteoporosis and degraded structure respectively while spine BMD is osteopenic…

At the end, you can’t see that TBS is not useful in elderlies… you can only say that it may not be useful when you already have severe OP patients by Hip fracture…(although let’s see you results after adjustment for spine BMD and the category > HbA1c)

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2020 Nov 19;15(11):e0241616. doi: 10.1371/journal.pone.0241616.r002

Author response to Decision Letter 0

24 Jul 2020

Dear PLOS ONE editor

We thank you for the opportunity to revise our manuscript and detail here our replies to the reviewers. We thank our reviewers for their time in reviewing this revision and hope this would be sufficient to answer their queries .

Comments:

We thank the reviewer for the comment and have amended our introduction and discussion sections accordingly.

We thank the reviewer for the questions . This was a retrospective study conducted on hip fracture patients presenting to our institution between 2014-2017 , clinical and demographic details were extracted from their electronic medical records. We have clarified this in the descriptions of our methods

When was BMD measurement performed in relation to hip fracture (prior, after) and at what interval?

We thank the reviewer for this question. BMDs were performed during the same admission for hip fracture or within 6 months of the fracture , we have added this information into the manuscript as a supplementary table 1. The median is 14 to 19 days from date of admission from hip fracture to BMD analysis.

How was the duration of diabetes established?

We thank the reviewer for the question. The diagnosis of DM2 was established using the World Health Organization (WHO) 2019 criteria (29) on the basis of having an HbA1c of 6.5% or greater, or current treatment with oral antidiabetic drugs or insulin. Duration of DM2 was established from the first documented diagnoses as established by the WHO definition to the date of hip fracture. These values were recorded according to the number of months, however for ease of reading these values were presented in terms of years in the analysis. We have added this information to further clarify this to the reader.

Minor:

Figure 1 - should include the number of people with DM for all stages.

We thank the reviewer for the suggestion , this has been added to figure 1 .

General comments:

We thank the reviewer for his comments, we agree that the differences observed in our study is likely a reflection of the severe osteoporosis of this elderly hip fracture population. We further clarify in our discussion as suggested by reviewer #3 that although the TBS are higher in this study they still all belonged to the intermediate - degraded category which is consistent with LS BMD results obtained. We have also substantially added to the analysis in this revision by incorporating the differences in the well-controlled and poorly controlled DM2 patients as suggested by reviewer #3 and found that these differences were significant in all DM2 women and poorly controlled DM2 men. In particular the differences in poorly controlled DM2 men became significant after adjustments for microvascular complications and vitamin D status. We also further found that these differences were not independent of LS BMD results in this population. As such , we humbly believe that the finding of this study would further add to the current available literature and further inform clinicians in parts of the world looking after older DM2 patients with poorly controlled DM2 from a non-Caucasian ethnicity group on potential differences in TBS and BMD findings. We also suggest that future studies in DM2 patients should consider accounting for DM2 complications and vitamin D status

Specific comments:

We thank the reviewer for this question , there were 174 patients who did not have BMD performed and were not included in this analysis. BMDs were performed during the admission or within 6 months of the hip fracture date. We included this information as supplementary table 1, median duration (days) from hip fracture date to BMD analysis was between 14-19 days and there were no significant differences between DM2 and non DM2 patients on the length of time from hip fracture and BMD analysis .

We thank the reviewer for the question , we excluded patients with incomplete BMD L spine and TBS scores to ensure that artefactual interferences from conditions such as osteoarthritis, severe scoliosis and degeneration would not interfere with the results of the analysis. We would like to assure the reviewer that when we performed the repeat analysis with inclusion of patients with at least 3 viable levels of LS Spine for BMD and TBS analysis , our results and analysis showed similar patterns to our current analysis. We have added this information into the statistical methods and sensitivity analysis to clarify this for our readers.

We thank the reviewer for the comment. Indeed osteoporosis treatment rate in the elderly in our community is very low , we have previously published osteoporosis treatment rates even after a hip fracture at 1 year was between 10-31% . In this study population about 10% of the patients have had a previous history of fracture. We hope to be able to improve these rates of treatment with the establishment of our osteoporosis liaison service and further implement primary fracture prevention treatments prior to the occurrence of a hip fracture.

4. Approximately 1/3 of the hip fracture patients had DM2. Was this by design or does it reflect the true prevalence of diabetes in this population?

Singapore has one of the highest incidence of DM2 in the world and the DM2 prevalence in the elderly has been documented in the national statistics to be around 30% which is in keeping with this study population ( National Health Survey 1998, 2004, 2010, Ministry of Health, Singapore ), we have included this reference to clarify this point (Phan TP, Alkema L, Tai ES, Tan KHX, Yang Q, Lim W-Y, et al. Forecasting the burden of type 2 diabetes in Singapore using a demographic epidemiological model of Singapore. BMJ Open Diabetes Research & Care. 2014;2 )

We thank the reviewer for the comments, we agree that further study on the impact of body composition especially abdominal tissue thickness on TBS is important . And we agree that in keeping with previous published studies the current Hologic TBS version are particularly sensitive for BMI. We have put this as a limitation of the study in the discussion section to clarify this point

We thank the reviewers for the comment .Gender is known to have an effect on BMD and TBS as demonstrated by previous studies and hence we have separated the analysis of this study into different gender. We have added the information on gender and DM2 analysis into the supplementary section ( Supplementary table 6 ) of the manuscript. Our analysis have found that there is an interaction effect between gender and DM2 . With using Male with DM2 as reference group , female without DM2 were found to have significantly lower TBS . Female with DM2 also had statistically significant lower TBS compared to male with DM2 and there is no significant difference between male without DM2 and male with DM2. This is consistent with the results of the current analysis and multivariate model. Differences were only found in elderly males with poorly controlled DM2 and after adjustments with potentially confounding variables as related in the manuscript.

7. Figure 2 is not very insightful and could be moved to the Supplementary section.

We thank the reviewer for the comment and have moved this figure into the supplementary section as supplementary figure 1

8. Please provide units for the coefficients reported in Supplementary Table 3.

We thank the reviewer for the comment . Supplementary table 3 is now supplementary table 5 and units for the covariate insulin (unit/kg ) , DM2 duration ( years ) and HbA1C ( % ) has been provided.

9. Some references are repeated (#21 and #34, maybe others).

We thank the reviewer for the comment and have update the references appropriately to remove any duplicates

We thank the reviewer for this very insightful comment. We agree that our population studied belong to the “severe osteoporotic” group and we have clarified this further in our manuscript to avoid misleading the readers that the TBS values are normal. We have further inserted information regarding TBS degradation severity to clarify this point within the manuscript. We followed the software generated TBS degradation score and classification which was TBS value of � 1.35 is considered normal, 1.20 to 1.35 is considered to intermediate and � 1.20 to be degraded.

We thank the reviewer for the insightful comments and suggestions , we have repeated the analysis with the well- controlled and poorly controlled DM2 patients and comparing them to the non DM2 patients. We found that differences in TBS and BMD were significant in all DM2 women and persisted after adjustment for potential significant confounders. TBS and BMD differences became significant in elderly poorly controlled DM2 men after adjusting for microvascular complications and vitamin D status. Differences in TBS became attenuated after inclusion of BMD LS but remained significant only in elderly well controlled DM2 women. We hope that these results will help better inform future clinicians looking after elderly DM2 patients in interpreting TBS and BMD findings. We further suggest the importance of including details on DM2 microvascular complications and vitamin D status for future analysis as this may confound BMD and TBS results as shown in our elderly men with poorer DM2 control.

Minor comments:

The mention of the study in HK population (ref 23) is coming unexpectedly as it is not on T2DM patient. What is the message you want to pass over here?

We thank the reviewer for this comment and agree that this reference may not provide additional information for the purpose of this study and have removed this paragraph and reference from the manuscript

We thank the reviewer for this comment , we have added the references to studies which alluded to the role of impaired fasting glucose , insulin resistance and HOMA IR to TBS in addition to BMI in the introduction section.

We thank the reviewer for this comment. Indeed osteoporosis treatment rate in the elderly in our community is very low , we have previously published osteoporosis treatment rates even after a hip fracture at 1 year was between 10-31% . In this study population about 10% of the patients have had a previous history of fracture. We hope to be able to improve these rates of treatment with the establishment of our osteoporosis liaison service and better implement primary fracture prevention treatments prior to the occurrence of a hip fracture.

We thank the reviewer for the question. We apologise for the error Our TBS CV are 1.53% with an LSC of 4.24% for this Hologic devices . This was a value calculated together with the assistance of our vendor. This has been amended appropriately in the manuscript. The BMD T score is based on our normative local population . TBS scores are not graded with a T score as we don’t currently have a local normative data to compare this to. We hope that future studies in our local population will be able to provide further information on the normative TBS data.

- Shall we adjust with only significant variables on top of age , BMI, CKD status ( yes or no ) , vitamin D , DM2 status ( Can I check how many non DM2 patients had documented HbA1C levels ? If this number is high , we could use HbA1C instead of DM2 status for adjustment ) , and BMD LSpine ( only for differences in TBS DM2 patients )

We thank the reviewer for this comment , we reduced the number of covariates in our multivariable analysis to the suggested covariates that are found to be significant in our univariate analysis. We have limited our variables analysed in the model to be age, BMI , race, presence of CKD , amputation and vitamin D status. 50% of our non-DM2 patients have HbA1C documented during the time of admission . As not all our non-DM2 patients had HbA1C recorded, we have opted to use DM2 status in our adjustments.

In table 2, for TBS it would also make an adjustment for Age, BMI and BMD lumbar spine, as we want to investigate the independence of TBS association between DM2 and non-DM2 from density.

- could we also do this for the supplementary 3 table for TBS fully adjusted multivariate model ( ie including the BMD L Spine in the model )

We thank the reviewer for the insightful comment and have added this into a third model in our adjustments to assess the independence of TBS association between DM2 and non DM2 patients. We found that TBS difference were not independent of LS BMD differences in both well controlled and poorly controlled DM2 elderly patients. We have included this findings into our results and discussions.

Results / discussion:

See some of my major comments above.

We thank the reviewer for this comment , we have inserted the TBS category for degradation - TBS value of � 1.35 is considered normal , � 1.20 to � 1.35 is considered to intermediate and �1.20 to be degraded. As we do not have a current normative data for our population, setting this in the context of a recent normative TBS and BMD study from a South East Asian population in Thailand ( Sritara et al Age-Adjusted Dual X-ray Absorptiometry-Derived Trabecular Bone Score Curve for the Lumbar Spine in Thai Females and Males JCD 2016 ; 19) , the value of TBS in this population is compared to the recorded value in the 70-80 y/o female TBS at 1.218. The TBS value in this population of severe osteoporotic elderly female non DM2 patient was recorded at 1.19 and elderly DM2 was 1.23. The same normative data in elderly (70-80 y/o) Thai male was documented in the study at TBS of 1.302, our study demonstrated TBS value of 1.31 in elderly non DM2 male and 1.32 in elderly DM2 male. We have added that future studies to better understand the role of TBS and BMD in our elderly population and normative data would be important to inform future studies in our population.

We thank the reviewer for the comment and have shortened this discussion and moved the figure to the supplementary section of the manuscript.

Your report significant relationship between TBS and insulin for women in suppl. Table 3. But it is not mentioned at all in the text. On purpose?

We thank the reviewer for the comment , as this was a cross sectional study we were not able to infer causality and simply stated the association as demonstrated in the multivariate analysis. Furthermore , we acknowledged the limitation of the small number of DM2 patients on insulin within this study ( n=29 ) as a limitation of the study. We cited previous studies on insulin and its effect in BMD .Exogenous insulin was found in various studies to increase, reduces or having a neutral effect on BMD. We added that it would be important for future longitudinal and larger study to better understand the relationship of exogenous insulin on BMD or TBS.

In your limitations, I would clearly states that there is no age matched controls.

We thank the reviewer for the comment , we have added this important limitation point into our discussion.

We thank the reviewer for the comment. We have amended our conclusion to state the pertinent findings in our re-analysis. We concluded that older DM2 patients presents with hip fractures at a higher TBS and BMD value compared to the non-DM2 patients. And these differences persist despite adjustment for covariates in DM2 women and poorly controlled DM2 men. We have further expounded on the differences in LS BMD, T score and Total hip T score in elderly men and women with DM2 and non DM2 in our discussions to clarify the corresponding values to the osteopenic or osteoporotic ranges .

We thank the reviewer for the comment. We agree that the results of this study is not applicable to all elderly patients and hence have limited our report to the findings of this study - which are the significantly higher TBS and BMD values in elderly patients with DM2 who present with hip fractures. However, women with poorly controlled DM2 had lower TBS values compared to those that were well controlled but this trend were reversed in DM2 men. It is interesting to note that when we looked at HbA1C as a continuous variable, differences in TBS and BMD did not exhibit significant associations. These may imply the need for future study of BMD and TBS differences in DM2 population to take into account the contribution of DM2 disease control and complications

We mentioned in our discussion caution to our readers to limit this finding to the elderly hip fracture population and suggested that further studies would be important in understanding these differences better in this population.

Corresponding Author

Linsey Gani

Attachment

Submitted filename: PLOSONEreply210720.docx

Click here for additional data file.^{(25.2KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0241616.r003

Decision Letter 1

Robert Daniel Blank

14 Aug 2020

PONE-D-20-14006R1

Bone Mineral Density and Trabecular Bone Score in Elderly Type 2 Diabetes Southeast Asian Patients with Severe Osteoporotic Hip Fractures

PLOS ONE

Dear Dr. Gani,

Please address the suggestions made by reviewer 3. Thank you for the substantial changes you have made to address the initial critique of your MS.

Please submit your revised manuscript by Sep 28 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Robert Daniel Blank, MD, PhD

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #1: Thank you for addressing my comments. I think the manuscript has improved. I have no further comments.

Reviewer #3: The paper has been significantly improved.

Please add on table 2 the age and BMI for each categories.

It can be misleading sometime in the text… when you are saying “elderly patients with DM2 and severe OP present with Hip fracture…. Compared to non DM2 patients”. It could almost applied that the non DM2 patients are not OP with hip fracture. So please throughout the manuscript, abstract and table make it clear….In patients with severe osteoporosis present with Hip fracture: DM2 vs non DM2

The fact that in poorly controlled women the TBS is lower than well controlled is now consistent with literature…. and clearly have clinical implication. This findings is already important and could be also highlighted in the abstract. Maybe by linking that with the results in men dilute the message. You may want to separate the sentence in two.

The fact that HbA1C as continuous variable does not work as good as category could be related to the non-gaussian distribution of the parameter (skewed)…. The log should then be used… (possible explanation….)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

PLoS One. 2020 Nov 19;15(11):e0241616. doi: 10.1371/journal.pone.0241616.r004

Author response to Decision Letter 1

9 Sep 2020

Dear Plos One Editor

Reviewer #3: The paper has been significantly improved.

Please add on table 2 the age and BMI for each categories.

Thank you for the comment , we have added age and BMI into table 2

Thank you for the comment , we have amended these accordingly to make it clear that these are DM2 and non DM2 patients with hip fracture .

Thank you for the comment, we have amended this in the abstract and hope that this clarifies the point better

Thank you for the comment and suggestion, we re-analyzed the multivariable regression using a log-transformation of the HbA1C and found no association of Log HbA1C with TBS. This is included in supplementary table 7. We have included this explanation in the body of the text to clarify this point.

Attachment

Submitted filename: Editorreply170820.docx

Click here for additional data file.^{(13.5KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0241616.r005

Decision Letter 2

Robert Daniel Blank

19 Oct 2020

Bone Mineral Density and Trabecular Bone Score in Elderly Type 2 Diabetes Southeast Asian Patients with Severe Osteoporotic Hip Fractures

PONE-D-20-14006R2

Dear Dr. Gani,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Robert Daniel Blank, MD, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #3: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #3: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Reviewer #1: Thank you for addressing reviewers' comments. The manuscript benefited from the revision. I have no further comments.

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

PLoS One. doi: 10.1371/journal.pone.0241616.r006

Acceptance letter

Robert Daniel Blank

22 Oct 2020

PONE-D-20-14006R2

Bone Mineral Density and Trabecular Bone Score in Elderly Type 2 Diabetes Southeast Asian Patients with Severe Osteoporotic Hip Fractures

Dear Dr. Gani:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr Robert Daniel Blank

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Fig. Relationship of trabecular bone score (TBS) and body mass index (BMI) in women and men with and without DM2.

(TIF)

Click here for additional data file.^{(333.2KB, tif)}

S1 Table. Duration (days) from date of admission to BMD analysis stratified by gender and DM2 status.

(DOCX)

Click here for additional data file.^{(21.4KB, docx)}

S2 Table. Correlation between TBS and BMD in DM2 and non DM2 patients stratified by gender.

(DOCX)

Click here for additional data file.^{(14KB, docx)}

S3 Table. Correlation between BMI with TBS and BMD in DM2 and nonDM2 patients stratified by gender.

(DOCX)

Click here for additional data file.^{(13.5KB, docx)}

S4 Table. Demographic and clinical variables of DM2 patients subdivided into gender and DM2 control.

(DOCX)

Click here for additional data file.^{(17.5KB, docx)}

S5 Table

(DOCX)

Click here for additional data file.^{(16.1KB, docx)}

S6 Table. Interaction term analysis between TBS and DM2 status with gender.

(DOCX)

Click here for additional data file.^{(12.9KB, docx)}

S7 Table. Relationship of trabecular bone score with diabetes mellitus (DM2) medications, complications and glycaemia control (β Coefficient).

(DOCX)

Click here for additional data file.^{(15KB, docx)}

Attachment

Submitted filename: PLOSONEreply210720.docx

Click here for additional data file.^{(25.2KB, docx)}

Attachment

Submitted filename: Editorreply170820.docx

Click here for additional data file.^{(13.5KB, docx)}

Data Availability Statement

[pone.0241616.ref001] 1.Sarwar N, Gao P, Seshasai SR, Gobin R, Kaptoge S, Di Angelantonio E, et al. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet (London, England). 2010;375(9733):2215–22. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref002] 2.Kanis JA, Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: Synopsis of a WHO report. Osteoporosis International. 1994;4(6):368–81. 10.1007/BF01622200 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref003] 3.Chan JC, Malik V, Jia W, Kadowaki T, Yajnik CS, Yoon KH, et al. Diabetes in Asia: epidemiology, risk factors, and pathophysiology. Jama. 2009;301(20):2129–40. 10.1001/jama.2009.726 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref004] 4.Phan TP, Alkema L, Tai ES, Tan KHX, Yang Q, Lim W-Y, et al. Forecasting the burden of type 2 diabetes in Singapore using a demographic epidemiological model of Singapore. BMJ Open Diabetes Research & Care. 2014;2(1):e000012 10.1136/bmjdrc-2013-000012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref005] 5.Cooper C, Campion G, Melton LJ 3rd. Hip fractures in the elderly: a world-wide projection. Osteoporos Int. 1992;2(6):285–9. 10.1007/BF01623184 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref006] 6.Janghorbani M, Van Dam RM, Willett WC, Hu FB. Systematic review of type 1 and type 2 diabetes mellitus and risk of fracture. Am J Epidemiol. 2007;166(5):495–505. 10.1093/aje/kwm106 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref007] 7.Janghorbani M, Feskanich D, Willett WC, Hu F. Prospective study of diabetes and risk of hip fracture: the Nurses’ Health Study. Diabetes Care. 2006;29(7):1573–8. 10.2337/dc06-0440 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref008] 8.Bonds DE, Larson JC, Schwartz AV, Strotmeyer ES, Robbins J, Rodriguez BL, et al. Risk of fracture in women with type 2 diabetes: the Women’s Health Initiative Observational Study. J Clin Endocrinol Metab. 2006;91(9):3404–10. 10.1210/jc.2006-0614 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref009] 9.de L II, van der Klift M, de Laet CE, van Daele PL, Hofman A, Pols HA. Bone mineral density and fracture risk in type-2 diabetes mellitus: the Rotterdam Study. Osteoporos Int. 2005;16(12):1713–20. 10.1007/s00198-005-1909-1 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref010] 10.Hanley D, Brown J, Tenenhouse A, Olszynski W, Ioannidis G, Berger C, et al. Associations Among Disease Conditions, Bone Mineral Density, and Prevalent Vertebral Deformities in Men and Women 50 Years of Age and Older: Cross-Sectional Results From the Canadian Multicentre Osteoporosis Study. Journal of Bone and Mineral Research. 2003;18(4):784–90. 10.1359/jbmr.2003.18.4.784 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref011] 11.Melton LJ 3rd, Riggs BL, Leibson CL, Achenbach SJ, Camp JJ, Bouxsein ML, et al. A bone structural basis for fracture risk in diabetes. J Clin Endocrinol Metab. 2008;93(12):4804–9. 10.1210/jc.2008-0639 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref012] 12.Chen H, Li X, Yue R, Ren X, Zhang X, Ni A. The effects of diabetes mellitus and diabetic nephropathy on bone and mineral metabolism in T2DM patients. Diabetes research and clinical practice. 2013;100(2):272–6. 10.1016/j.diabres.2013.03.007 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref013] 13.Krakauer JC, McKenna MJ, Buderer NF, Rao DS, Whitehouse FW, Parfitt AM. Bone loss and bone turnover in diabetes. Diabetes. 1995;44(7):775–82. 10.2337/diab.44.7.775 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref014] 14.Looker AC, Eberhardt MS, Saydah SH. Diabetes and fracture risk in older U.S. adults. Bone. 2016;82:9–15. 10.1016/j.bone.2014.12.008 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref015] 15.McCloskey EV, Oden A, Harvey NC, Leslie WD, Hans D, Johansson H, et al. A Meta-Analysis of Trabecular Bone Score in Fracture Risk Prediction and Its Relationship to FRAX. J Bone Miner Res. 2016;31(5):940–8. 10.1002/jbmr.2734 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref016] 16.Harvey NC, Gluer CC, Binkley N, McCloskey EV, Brandi ML, Cooper C, et al. Trabecular bone score (TBS) as a new complementary approach for osteoporosis evaluation in clinical practice. Bone. 2015;78:216–24. 10.1016/j.bone.2015.05.016 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref017] 17.Silva BC, Broy SB, Boutroy S, Schousboe JT, Shepherd JA, Leslie WD. Fracture Risk Prediction by Non-BMD DXA Measures: the 2015 ISCD Official Positions Part 2: Trabecular Bone Score. J Clin Densitom. 2015;18(3):309–30. 10.1016/j.jocd.2015.06.008 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref018] 18.Rianon N, Ambrose CG, Buni M, Watt G, Reyes-Ortiz C, Lee M, et al. Trabecular Bone Score Is a Valuable Addition to Bone Mineral Density for Bone Quality Assessment in Older Mexican American Women With Type 2 Diabetes. J Clin Densitom. 2018;21(3):355–9. 10.1016/j.jocd.2018.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref019] 19.Jain RK, Narang DK, Hans D, Vokes TJ. Ethnic Differences in Trabecular Bone Score. J Clin Densitom. 2017;20(2):172–9. 10.1016/j.jocd.2016.04.003 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref020] 20.Kim JH, Choi HJ, Ku EJ, Kim KM, Kim SW, Cho NH, et al. Trabecular bone score as an indicator for skeletal deterioration in diabetes. J Clin Endocrinol Metab. 2015;100(2):475–82. 10.1210/jc.2014-2047 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref021] 21.Ho-Pham LT, Tran B, Do AT, Nguyen TV. Association between pre-diabetes, type 2 diabetes and trabecular bone score: The Vietnam Osteoporosis Study. Diabetes research and clinical practice. 2019;155:107790 10.1016/j.diabres.2019.107790 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref022] 22.Iki M, Fujita Y, Kouda K, Yura A, Tachiki T, Tamaki J, et al. Hyperglycemia is associated with increased bone mineral density and decreased trabecular bone score in elderly Japanese men: The Fujiwara-kyo osteoporosis risk in men (FORMEN) study. Bone. 2017;105:18–25. 10.1016/j.bone.2017.08.007 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref023] 23.Martineau P, Leslie WD, Johansson H, Harvey NC, McCloskey EV, Hans D, et al. In which patients does lumbar spine trabecular bone score (TBS) have the largest effect? Bone. 2018;113:161–8. 10.1016/j.bone.2018.05.026 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref024] 24.Leslie WD, Krieg MA, Hans D. Clinical factors associated with trabecular bone score. Journal of clinical densitometry: the official journal of the International Society for Clinical Densitometry. 2013;16(3):374–9. 10.1016/j.jocd.2013.01.006 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref025] 25.Sritara C, Thakkinstian A, Ongphiphadhanakul B, Amnuaywattakorn S, Utamakul C, Akrawichien T, et al. Age-Adjusted Dual X-ray Absorptiometry-Derived Trabecular Bone Score Curve for the Lumbar Spine in Thai Females and Males. J Clin Densitom. 2016;19(4):494–501. 10.1016/j.jocd.2015.05.068 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref026] 26.Evans AL, Paggiosi MA, Eastell R, Walsh JS. Bone Density, Microstructure and Strength in Obese and Normal Weight Men and Women in Younger and Older Adulthood. Journal of Bone and Mineral Research. 2015;30(5):920–8. 10.1002/jbmr.2407 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref027] 27.Langsetmo L, Vo TN, Ensrud KE, Taylor BC, Cawthon PM, Schwartz AV, et al. The Association Between Trabecular Bone Score and Lumbar Spine Volumetric BMD Is Attenuated Among Older Men With High Body Mass Index. J Bone Miner Res. 2016;31(10):1820–6. 10.1002/jbmr.2867 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref028] 28.WHO. Classification of Diabetes Mellitus 2019 WHO. 2019.

[pone.0241616.ref029] 29.Tatangelo G, Watts J, Lim K, Connaughton C, Abimanyi-Ochom J, Borgström F, et al. The Cost of Osteoporosis, Osteopenia, and Associated Fractures in Australia in 2017. J Bone Miner Res. 2019;34(4):616–25. 10.1002/jbmr.3640 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref030] 30.Phan TP, Alkema L, Tai ES, Tan KHX, Yang Q, Lim W-Y, et al. Forecasting the burden of type 2 diabetes in Singapore using a demographic epidemiological model of Singapore. BMJ Open Diabetes Research & Care. 2014;2(1):e000012 10.1136/bmjdrc-2013-000012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref031] 31.Dufour R, Winzenrieth R, Heraud A, Hans D, Mehsen N. Generation and validation of a normative, age-specific reference curve for lumbar spine trabecular bone score (TBS) in French women. Osteoporosis International. 2013;24(11):2837–46. 10.1007/s00198-013-2384-8 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref032] 32.Leslie WD, Aubry-Rozier B, Lamy O, Hans D, Manitoba Bone Density P. TBS (trabecular bone score) and diabetes-related fracture risk. J Clin Endocrinol Metab. 2013;98(2):602–9. 10.1210/jc.2012-3118 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref033] 33.Thrailkill KM, Charles K. Lumpkin J, Bunn RC, Kemp SF, Fowlkes JL. Is insulin an anabolic agent in bone? Dissecting the diabetic bone for clues. American Journal of Physiology-Endocrinology and Metabolism. 2005;289(5):E735–E45. 10.1152/ajpendo.00159.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref034] 34.Ruppert K, Cauley J, Lian Y, Zgibor JC, Derby C, Solomon DH. The effect of insulin on bone mineral density among women with type 2 diabetes: a SWAN Pharmacoepidemiology study. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 2018;29(2):347–54. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref035] 35.de Waard EAC, Koster A, Melai T, van Geel TA, Henry RMA, Schram MT, et al. The association between glucose metabolism status, diabetes severity and a history of fractures and recent falls in participants of 50 years and older-the Maastricht Study. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 2016;27(11):3207–16. 10.1007/s00198-016-3645-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref036] 36.Starup-Linde J, Gregersen S, Vestergaard P. Associations with fracture in patients with diabetes: a nested case–control study. BMJ Open. 2016;6(2):e009686 10.1136/bmjopen-2015-009686 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref037] 37.Schacter GI, Leslie WD, Majumdar SR, Morin SN, Lix LM, Hans D. Clinical performance of an updated trabecular bone score (TBS) algorithm in men and women: the Manitoba BMD cohort. Osteoporos Int. 2017;28(11):3199–203. 10.1007/s00198-017-4166-1 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref038] 38.Kim SM, Moon YW, Lim SJ, Yoon BK, Min YK, Lee DY, et al. Prediction of survival, second fracture, and functional recovery following the first hip fracture surgery in elderly patients. Bone. 2012;50(6):1343–50. 10.1016/j.bone.2012.02.633 [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref039] 39.Lee WJ, Sobrin L, Lee MJ, Kang MH, Seong M, Cho H. The Relationship Between Diabetic Retinopathy and Diabetic Nephropathy in a Population-Based Study in Korea (KNHANES V-2, 3)Relationship Between DR and DN in Korea. Investigative Ophthalmology & Visual Science. 2014;55(10):6547–53. [DOI] [PubMed] [Google Scholar]

[pone.0241616.ref040] 40.Lee M-K, Han K-D, Lee J-H, Sohn S-Y, Hong O-K, Jeong J-S, et al. Normal-to-mildly increased albuminuria predicts the risk for diabetic retinopathy in patients with type 2 diabetes. Scientific Reports. 2017;7(1):11757 10.1038/s41598-017-11906-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0241616.ref041] 41.Compston J. Type 2 diabetes mellitus and bone. Journal of Internal Medicine. 2018;283(2):140–53. 10.1111/joim.12725 [DOI] [PubMed] [Google Scholar]

PERMALINK

Bone mineral density and trabecular bone score in elderly type 2 diabetes Southeast Asian patients with severe osteoporotic hip fractures

Linsey U Gani

Kundan R Saripalli

Karen Fernandes

Suet F Leong

Koh T Tsai

Pei T Tan

Le R Chong

Thomas F J King

Roles

Abstract

Introduction

Methods

Results

Conclusions

Introduction

Subjects and methods

Study population

Clinical evaluation

Biochemical evaluation

Bone mineral density and trabecular bone score measurements

Statistical analysis

Results

Cohort characteristics and demographics

Fig 1. Study subject inclusion and exclusion selection criteria.

Table 1. Baseline demographic and clinical variables, trabecular bone score (TBS), bone mineral density (BMD) in patients who are non-diabetic (ND) and those with diabetes mellitus type 2 (DM2).

TBS and BMD differences between women and men with well controlled (HbA1C < 7%) or poorly controlled DM2 (HbA1C ≥ 7%) compared to non DM2 patients with hip fractures

Table 2. TBS and BMD values in DM2 and non DM2 women and men stratified to well controlled (HbA1c < 7%) and poorly controlled (HbA1c ≥ 7%).

Table 3.

Association of TBS with LS BMD and BMI

Subgroup analysis in DM2 patients between well controlled vs poorly controlled DM2 patients with hip fractures

Discussion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Robert Daniel Blank

Roles

Author response to Decision Letter 0

Decision Letter 1

Robert Daniel Blank

Roles

Author response to Decision Letter 1

Decision Letter 2

Robert Daniel Blank

Roles

Acceptance letter

Robert Daniel Blank

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases