ORIGINAL E n d o c r i n e
ARTICLE C a r e
Factors Associated With Bone Mineral Density and Risk of Fall in Korean Adults With Type 2 Diabetes Mellitus Aged 50 Years and Older Kyoung Min Lee, Chin Youb Chung, Soon-Sun Kwon, Seung Yeol Lee, Tae Gyun Kim, Young Choi, and Moon Seok Park Department of Orthopaedic Surgery (K.M.L., C.Y.C., S.Y.L., T.G.K., Y.C., M.S.P.) and Biomedical Research Institute (S.-S.K.), Seoul National University Bundang Hospital, Kyungki 463-707, Korea
Context: Osteoporotic fractures in subjects with diabetes mellitus (DM) carry higher mortality and morbidity. Because bone strength and minor trauma, such as a falls, are considered to be significant factors contributing to osteoporotic fractures, it is important to elucidate the associated factors with these. Objective: This study was performed to investigate the factors associated with bone mineral density (BMD) and falls in noninstitutionalized subjects with DM aged 50 years or older. Design, Setting, and Patients: We used the database from the 2010 Fifth Korea National Health and Nutrition Examination Survey. Subjects with DM aged 50 years or older were selected and included in the data analyses. Associated factors with BMD of the femoral neck and lumbar spine and those with falls were analyzed using multiple linear regression and binary logistic regression analyses, respectively. Results: Three hundred sixty-two subjects [209 males; 153 females; average age, 66.0 y (SD 8.2 y)] were included. Among the male subjects, the total body muscle mass (P ⬍ .001), daily calcium intake (P ⫽ .001), ALP levels (P ⫽ .007), and body mass index (P ⫽ .027) were significantly associated with femoral neck BMD, whereas body mass index (P ⫽ .001) and ALP levels (P ⫽ .040) were associated with lumbar spine BMD. Among the female subjects, age (P ⬍ .001), daily calcium intake (P ⫽ .011) and total body muscle mass (P ⫽ .023) were found to be significantly associated factors with femoral neck BMD, whereas age (P ⬍ .001) and body mass index (P ⫽ .012) and daily calcium intake (P ⫽ .040) were those with lumbar spine BMD. Osteoarthritis (P ⫽ .024) and total body muscle mass (P ⫽ .028) were found to be significantly associated with the risk of falls. Conclusions: Total body muscle mass was the most prominent factor predicting femoral neck BMD and risk of falls in community-dwelling elderly subjects with DM. Further investigation is required to determine their role in preventing osteoporotic fractures in diabetic subjects. (J Clin Endocrinol Metab 99: 4206 – 4213, 2014)
ecently diabetes mellitus (DM) has been drawing clinical attention as an independent risk factor for fragility fracture, although several studies have shown that subjects with DM do not have lower bone mineral density (BMD) (1, 2). These authors suggested that the susceptibility to fracture in DM is influenced by bone quality and increased risk of fall and not by bone quantity (3– 6). How-
R
ever, bone quality measurement has not been generalized in current clinical practice. In addition, some previous studies have reported that subjects with DM who sustained incident fractures had decreased BMD compared with those subjects with DM without such fractures (7, 8). Therefore, bone quantity, as measured by BMD, is still an apparent risk factor for osteoporotic fractures, and in-
ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received February 10, 2014. Accepted August 5, 2014. First Published Online August 20, 2014
Abbreviations: ALP, alkaline phosphatase; BMD, bone mineral density; BMI, body mass index; BUN, blood urea nitrogen; DM, diabetes mellitus; DXA, dual-energy X-ray absorptiometry; FBS, fasting blood glucose; KNHNES V, Fifth Korea National Health and Nutrition Examination Survey; OA, osteoarthritis of hip or knee.
4206
jcem.endojournals.org
J Clin Endocrinol Metab, November 2014, 99(11):4206 – 4213
doi: 10.1210/jc.2014-1400
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-1400
jcem.endojournals.org
4207
creasing the BMD is the current clinical target to prevent such fractures (9 –11). Osteoporotic fractures in subjects with DM is considered to carry higher mortality, morbidity, and subsequent sociomedical costs than such fractures in those without DM because of various concomitant medical conditions, such as impaired cardiovascular function, renal function, and immune response (12, 13). As such, preventing osteoporotic fractures in subjects with DM is a crucial medical issue. Because bone strength and minor trauma, such as a falls, are considered to be significant factors contributing to osteoporotic fractures (14, 15), it is important to elucidate the variables predicting these factors in order to reduce and prevent fractures. However, these have not been sufficiently investigated in elderly subjects with DM, although osteoporotic fractures are primarily involved in elderly population and could be complicated by DM. Therefore, in the present study, we aimed to identify the factors associated with BMD and risk of falls in Korean adults with type 2 DM aged 50 years or older using data from the Fifth Korea National Health and Nutrition Examination Survey (KNHNES V).
Materials and Methods Subjects The KNHNES V was conducted by the Korean Centers for Disease Control and Prevention in 2010. In total, 12 722 subjects selected from stratified, multistage probability samples of Korean households were invited to participate in the survey, and 10 533 agreed to participate, corresponding to a response rate of 82.8%. A total of 504 subjects aged 50 years or older with DM were selected for inclusion in this study. Of these, subjects who were taking osteoporosis medication, oral contraceptives, or hormone replacement therapy; those with any malignancy; and those with incomplete data were excluded from the data analyses (Figure 1). The survey included a health interview, health behavior, health examination, and nutrition surveys; of these, the health interview, health behavior, and health examination components were used in the present study. Written informed consent was obtained from all participants by the Korean Centers for Disease Control and Prevention, and the study was exempt from approval by the Institutional Review Board of Seoul National University Bundang Hospital because it used a publicly accessible database and did not involve any potential violation of human rights.
Data collection The KNHNES V 2010 database contains information on demographics, including age; sex; body mass index (BMI); weight change in the past year (gain of ⬎ 10 kg, gain of 6 –10 kg, gain of 3– 6 kg, gain of ⬍ 3 kg, no change, loss of ⬍ 3 kg, loss of 3– 6 kg, loss of 6 –10 kg, and loss of ⬎ 10 kg); house income (below the 25th percentile, between the 25th and 50th percentiles, between the 50th and 75th percentiles, greater than the 75th per-
Figure 1. Flow diagram of the inclusion and exclusion criteria in the KNHANES V. A total of 362 adults with DM aged 50 years or older were finally included.
centile); education level (elementary school, middle school, high school, and college); amount of smoking (current and cumulative); amount of alcohol consumption in the past year; presence of other medical conditions [hypertension, DM, hyperlipidemia, coronary artery disease, cerebrovascular accident, osteoarthritis of hip or knee (OA), chronic renal failure]; and current use of osteoporosis medication. The presence of diabetic retinopathy also was assessed. Family history of osteoporosis was recorded. In addition, experience of falls in the past year was obtained using the questionnaire. The parameters of height and weight were measured using standardized instruments, and BMI was calculated using the height and weight measurements. Weight change in the past year and education level were obtained from the questionnaire. House income was based on the total house income reported by the subjects. The amount of cumulative smoking was calculated from the amount of daily smoking and the duration of smoking, which were included in the questionnaire. The amount of alcohol consumption was calculated using the data on the average frequency of drinking and the average amount of drinking per ses-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
4208
Lee et al
BMD and Falls in DM
sion (in glasses) obtained from the questionnaire, with the assumption that one glass contains 10 g of alcohol. The presence of other medical conditions was determined from the diagnoses obtained from the questionnaire. Activity level was evaluated by the short form of the International Physical Activity Questionnaire (16), and weekly hours of vigorous-intensity, moderateintensity, and walking activities were calculated. Blood tests included fasting blood glucose (FBS), glycosylated hemoglobin, insulin, total cholesterol, triglycerides, hemoglobin, blood urea nitrogen (BUN), creatinine, 25-hydroxyvitamin D, alkaline phosphatase (ALP), and PTH. Blood samples were obtained throughout the year after an 8-hour fast and were immediately processed, refrigerated, and transported in cold storage to the central testing institute (NeoDin Medical Institute, Seoul, South Korea), where they were analyzed within 24 hours. Fundus photography was performed in subjects with DM. Seven standard photographs from the Early Treatment for Diabetic Retinopathy Study were obtained for each eye after pharmacological pupil dilatation. Diabetic retinopathy was defined as the presence of one or more retinal microaneurysms or retinal blot hemorrhages with or without more severe lesions (hard exudates, soft exudates, intraretinal microvascular abnormalities, venous bleeding, new retinal vessels, and fibroproliferations). Daily calcium intake was calculated from the Food Composition Table (17) and 24-hour dietary recall questionnaire, which was evaluated by a dietitian. BMD (grams per square centimeter) was measured at the femoral neck and lumbar spine using dual-energy X-ray absorptiometry (DXA; DISCOVERY-W fan-beam densitometer; Hologic Inc) with a coefficients of variation of 1.9% and 2.5%, respectively. Body composition also was analyzed, and total body muscle mass and total body fat percentiles were obtained. Total body muscle mass was calculated as follows: total body skeletal muscle ⫽ (1.19 appendicular lean soft tissue) ⫺ 1.01 (appendicular lean soft tissue mass measured by DXA machine) (18)
Data analysis The factors associated with BMD were analyzed for male and female subjects separately, whereas the factors predicting the risk of falls were evaluated for a single group containing both male and female subjects. The candidate variables that were significant in univariate analysis were selected and included in multiple regression analysis to evaluate their association with BMD and included in binary logistic regression analysis to assess the significant predicting factors of the risk of falls, respectively.
Statistics A descriptive analysis including the average and SD or proportion was performed for all variables. Data normality was tested using the Kolmogorov-Smirnov test. Correlations between the variables and BMDs of the femoral neck and lumbar spine were evaluated using Pearson’s correlation coefficient or Spearman’s correlation coefficient, depending on the normality of the data. The comparison of BMD between the two groups was performed using the Student’s t test. Variables that significantly correlated with BMD or conditions that showed significantly different BMD between cases were selected and included in the multiple regression analysis. For male subjects, the total body muscle mass, daily calcium intake, ALP, BMI, age, current amount of smoking, and triglycerides were included in the multiple regression analysis to evaluate femoral neck BMD; further-
J Clin Endocrinol Metab, November 2014, 99(11):4206 – 4213
more, BMI, ALP, and total body muscle mass were included in the multiple regression analysis model to evaluate lumbar spine BMD. For female subjects, age, daily calcium intake, total body muscle mass, house income, education level, FBS, BUN, and cumulative amount of smoking were included in the multiple regression analysis model to evaluate femoral neck BMD, and age, BMI, daily calcium intake, and total body muscle mass were included in the multiple regression analysis model to evaluate lumbar spine BMD. The categorical and binary variables that were significantly associated with a history of falls in the 2 test or continuous variables that were significantly different according to a history of falls in Student’s t test were selected and included in the binary logistic regression analysis. These were OA and total body muscle mass. A stepwise selection of the variables was adopted in multiple regression analysis to identify the significant associated factors with BMD, in which the dependent variables were BMD values of the femoral neck and lumbar spine. Binary logistic regression analysis was utilized to examine the significant factors that predict the risk of falls. All statistical analyses were performed using SPSS version 18.0 software (IBM Corp), with statistical significance set at P ⬍ .05.
Results A total of 362 subjects (209 males, 153 females) were included in the study. The average age of the subjects was 66.0 years (SD 8.2 y), and the average time from diagnosis of DM was 8.9 years (SD 8.1 y). The average femoral neck BMD of the male and female subjects was 0.7543 g/cm2 (SD 0.1196 g/cm2) and 0.5937 g/cm2 (SD 0.1070 g/cm2), respectively, whereas the average lumbar spine BMD was 0.9892 g/cm2 (SD 0.1552 g/cm2) and 0.8049 g/cm2 (SD 0.1243 g/cm2), respectively. A total of 53 subjects had a history of falls in the previous year (Table 1). Male subjects had significantly higher total body muscle mass than female subjects, and the total body muscle mass decreases with increasing age in both genders (Figure 2). In male adults with DM aged ⱖ 50 years, total body muscle mass (P ⬍ .001), daily calcium intake (P ⫽ .001) ALP levels (P ⫽ .007), and BMI (P ⫽ .027) were found to be significant associated factors with femoral neck BMD, whereas BMI (P ⫽ .001) and ALP levels (P ⫽ .040) were significantly associated with lumbar spine BMD (Table 2). In female adults with DM ⱖ 50 years, age (P ⬍ .001), daily calcium intake (P ⫽.011), and total body muscle mass (P ⫽ .023) were found to be significant associated factors with femoral neck BMD, whereas age (P ⬍ .001) and BMI (P ⫽ .012), and daily calcium intake (P ⫽ .040) were significantly associated with lumbar spine BMD (Table 3). OA (P ⫽ .024) and total body muscle mass (P ⫽ .028) were found to be the only significant factors that predicted the risk of falls in adults with DM aged 50 years or older (Table 4).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-1400
Table 1.
jcem.endojournals.org
4209
Data Summary
n Age, y Time from diagnosis of DM, y BMI, kg/cm2 Weight change (⬎10 kg gain/6 –10 kg gain/3– 6 kg gain/0 –3 kg gain/no change/0 –3 kg loss/3– 6 kg loss/6 –10 kg loss/⬎10 kg loss) House income (0 –25/25–50/50 –75/75–100 percentiles Education level, elementary/middle/high/college Current amount of smoking, pieces Cumulative amount of smoking, pack Amount of alcohol consumption, g/d Family history of osteoporosis, yes/noa Medical diseases HT, yes/no HL, yes/no CAD, yes/no CVA, yes/no OA, yes/no CRF, yes/no Diabetic retinopathy, yes/no Activity Weekly duration of vigorous-intensity activity, min Weekly duration of moderate-intensity activity, min Weekly duration of walking, min Weekly total MET Blood test FBS, mg/dL HbAlc, % Insulin, IU/mL Total cholesterol, mg/dL Triglycerides, mg/dL Hb, g/dL BUN, mg/dL Cr, mg/dL 25(OH)D, ng/mL ALP, IU/L PTH, pg/mL Daily calcium intake, mg Total body muscle mass, kg Total body fat percentile, % BMD (femoral neck), g/cm2 BMD (lumbar spine), g/cm2 Falls in the past year, yes/no
Male Subjects
Female Subjects
P Value
209 64.6 (8.1) 8.8 (8.2) 24.8 (3.1) 4/9/33/146/14/2/1
153 68.0 (8.1) 9.1 (8.0) 25.8 (3.9) 2/10/30/100/10/1/0
NA ⬍.001 .716 ⬍.006 .660
80/53/42/34 74/52/54/29 4.9 (8.8) 5473.8 (8726.0) 14.2 (21.7) 20/180
75/39/20/19 123/13/14/3 0.7 (3.1) 170.6 (1007.3) 0.9 (5.3) 21/126
.134 ⬍.001 ⬍.001 ⬍.001 ⬍.001 .228
121/88 42/167 17/192 17/192 23/186 3/206 33/176
97/56 38/115 12/141 10/143 65/88 3/150 30/123
.277 .251 1.000 .678 ⬍.001 .700 .400
92.6 (226.0) 190.6 (410.8) 458.2 (572.9) 3015.7 (3566.9)
56.1 (201.5) 192.5 (504.4) 305.1 (569.4) 2226.0 (3612.2)
.107 .969 ⬍.001 .016
141.8 (57.4) 7.3 (1.5) 10.1 (4.9) 177.5 (38.4) 186.0 (132.3) 14.4 (1.4) 17.6 (5.9) 1.02 (0.322) 20.4 (7.4) 244.8 (85.5) 63.9 (24.8) 565.5 (416.6) 47.2 (6.6) 24.1 (5.2) 0.7543 (0.1196) 0.9892 (0.1552) 21/188
140.0 (43.2) 7.4 (1.3) 12.6 (7.5) 196.0 (37.4) 175.1 (94.2) 12.9 (1.1) 17.0 (6.2) 0.77 (0.178) 17.7 (7.1) 264.8 (83.7) 69.8 (26.2) 374.5 (261.4) 34.1 (4.5) 36.7 (5.2) 0.5937 (0.1070) 0.8049 (0.1243) 32/121
.776 .492 .004 ⬍.001 .429 ⬍.001 .348 ⬍.001 .001 .046 .026 ⬍.001 ⬍.001 ⬍.001 ⬍.001 ⬍.001 .008
Abbreviations: CAD, coronary artery disease; Cr, creatinine; CRF, chronic renal failure; CVA, cerebrovascular accident; Hb, hemoglobin; HbAlc, glycosylated hemoglobin; Hct, hematocrit; HDL, high-density lipoprotein; HL, hyperlipidemia; HT, hypertension; LDL, low-density lipoprotein; MET, metabolic equivalent; NA, not applicable; 25(OH)D, 25-hydroxyvitamin D. Data are presented as mean (SD). a
Fifteen subjects (nine males and six females) did not know the details about their family histories of osteoporosis or osteoporotic fractures.
Total body muscle mass was significantly correlated with weekly hours of vigorous-intensity activity (r ⫽ 0.144, P ⫽ .044) but not with moderate-intensity activity (r ⫽ 0.062, P ⫽ .384) and walking (r ⫽ 0.087, P ⫽ .224).
Discussion This study investigated the factors associated with BMD and the factors predicting the risk of falls in communitydwelling subjects with DM aged 50 years or older. Total
body muscle mass was found to be a significant factor associated with BMD of the femoral neck and a significant factor predicting the risk of falls, whereas BMI was a significantly associated factor with lumbar spine BMD in both sexes with DM. ALP levels in the male adults and age in the female adults were found to be significant factors associated with femoral neck and lumbar spine BMD. Recently DM has drawn attention as an independent risk factor for osteoporotic fractures. The relationship between DM and fragility fracture has been clarified and reportedly involves the impairment of bone mechanical
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
4210
Lee et al
BMD and Falls in DM
J Clin Endocrinol Metab, November 2014, 99(11):4206 – 4213
Figure 2. The distribution of total body muscle mass in male and female subjects with diabetes according to age. Male subjects show higher total body muscle mass than female subjects, and the total body muscle mass decreases with increasing age in both genders.
strength by the glycosylation of type I collagen (19 –21). Unlike type 1 DM with low BMD, type 2 DM has been reported to have increased fracture risk despite normal or increased BMD, which is paradoxical to the conventional view of osteoporosis (1). Impaired bone quality and increased risk of falls have been proposed to explain this paradox in subjects with type 2 DM (4). Although trials have been conducted to evaluate bone quality statically and dynamically using quantitative computed tomography and bone turnover markers (22–25), quantitative BMD is still an apparent risk factor for fracture and remains the mainstream target of clinical management in subjects with DM (9 –11). Additionally, previous studies have reported that subjects with DM who sustained fractures had lower BMD than those without fractures, suggesting the clinical role of quantitative bone mass in osteoporotic fractures among subjects with DM (7, 8). Therefore, this study focused on the two relevant factors for osteoporotic fractures in subjects with DM in current clinical practice: BMD and falls. Total body muscle mass was found to be a significant factor in both femoral neck BMD and the risk of falls.
Previous studies have shown the relationship between osteoporosis and sarcopenia as well as between the risk of falls and sarcopenia in the elderly population (26, 27). Furthermore, skeletal muscle is considered to be involved in the endocrine system by modulating inflammatory cytokines, which also play a role in the development of type 2 DM (28). Subjects with DM are reported to have impaired muscle quality and strength (29). Frail or atrophic elderly individuals are believed to be more susceptible to falls, whereas people who exercise frequently or have higher levels of physical activity would be expected to have greater muscle mass; therefore, they may have greater stability when walking, running or performing activities of daily living. Therefore, optimizing total body muscle mass in these subjects could result in increased BMD and decreased risk of falls, thus eventually preventing osteoporotic fractures and possibly improving diabetes control. A previous study has shown that resistance training increased muscle mass and insulin action and had a positive effect on multiple risk factors for osteoporotic fractures in sedentary postmenopausal women (30). We found a significant correlation between weekly hours of vigorousintensity physical activity and total body muscle mass, which is in agreement with the results of the previous study. However, the effect of resistance training on the risk of osteoporotic fractures in elderly subjects with DM needs to be investigated further using a prospective longitudinal study design. In addition, total body muscle mass may be a reflection of age, postmenopausal duration, and a combination of life-long and recent androgen and estrogen status. These factors should be investigated further in a well-designed study. In the present study, we noted that higher BMI was a protective factor for lumbar spine osteoporosis in elderly subjects with type 2 DM in both sexes, which is consistent with the findings of previous studies suggesting that increased BMD in subjects with type 2 DM is caused by increased BMI (31–33). Although these studies have
Table 2. Multiple Regression Analysis to Identify Significant Factors Associated With Femoral Neck and Lumbar Spine BMD in Male Adults With Diabetes Mellitus Aged 50 Years or Older Femoral Neck BMD
Total body muscle mass Daily calcium intake ALP BMI Age Smoking (current) Triglycerides Coefficient
Lumbar Spine BMD
Standardized 
t
P Value
0.333 0.216 ⫺0.170 0.210 ⫺0.114 0.029 0.119 0.112
3.613 3.548 ⫺2.743 2.238 ⫺1.631 0.456 1.933 1.206
⬍.001 .001 .007 .027 .105 .643 .055 .230
BMI ALP Muscle mass Coefficient
Standardized 
t
P Value
0.252 ⫺0.151 0.049 0.737
3.472 ⫺2.071 0.454 7.086
.001 .040 .650 ⬍.001
R2 ⫽ 0.417; R2 ⫽ 0.100.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-1400
jcem.endojournals.org
4211
Table 3. Multiple Regression Analysis to Identify Significant Factors Associated With Femoral Neck and Lumbar Spine BMD in Female Adults With Diabetes Mellitus Aged 50 Years or Older Femoral Neck BMD
Age Daily calcium intake Muscle mass Income Education FBS BUN Smoking (cumulative) Coefficient
Lumbar Spine BMD
Standardized 
t
P Value
⫺0.590 0.195 0.185 0.106 0.158 0.113 ⫺0.085 ⫺0.082 1.129
⫺7.727 2.581 2.299 1.329 1.897 1.462 ⫺1.088 ⫺1.041 16.076
⬍.001 .011 .023 .187 .060 .146 .279 .300 ⬍.001
Age BMI Daily Calcium intake Muscle mass Coefficient
Standardized 
t
P Value
⫺0.344 0.205 0.171 0.080 0.964
⫺4.184 2.541 2.079 0.685 8.471
⬍.001 .012 .040 .495 ⬍.001
R2 ⫽ 0.385; R2 ⫽ 0.227.
shown that BMI was positively correlated with BMD, other studies have shown that obesity is a risk factor for fractures (34 –36). Thus, increased BMI could be a protective factor for osteoporosis and a risk factor for osteoporotic fractures at the same time, which is somewhat paradoxical. Increased BMI could result in increased strain on the skeleton in daily activities improving bone mass, whereas metabolic derangement in obese subjects with type 2 DM has an impact on bone quality and strength, which cannot be measured by DXA. Another possible explanation of the increased number of fractures in obese subjects is that obesity is a risk factor for frequent falls and more stress is exerted on the bone at the time of injury, thus increasing the possibility of fractures. ALP is produced by several organs, including the liver, bile duct, kidneys, and bone, and is primarily known to be associated with osteoblastic activity in bone metabolism. ALP level was a significant factor associated with femoral neck and lumbar spine BMD in male subjects with type 2 DM in the present study, showing a negative correlation. This could be interpreted that the subject with low BMD is concurrent with compensatory increased osteoblastic activity in male diabetic subjects. These results are supported by previous studies reporting an increase in ALP activity in women with more severe osteoporosis (37, 38). The amount of daily calcium intake was found to be significantly associated with femoral neck BMD in male subjects and both femoral neck and lumbar spine BMD in
female subjects. A previous study reported the beneficial effect of calcium supplements on femoral BMD and vertebral fracture rate in vitamin D-replete elderly patients (39). The current cross-sectional study results show a positive correlation between calcium intake and BMD in elderly subjects with diabetes, but the causal relationship needs to be investigated further in a longitudinal study. The increased risk of falls in subjects with DM has been explained by poorer balance, which may be a result of peripheral neuropathy, reduced vision, nephropathy, and lower muscle strength (40 – 42). Although our study data did not include the presence of peripheral neuropathy, total body muscle mass was found to significantly predict the risk of falls, whereas diabetic retinopathy and nephropathy did not. Our study population comprised community-dwelling elderly subjects with DM who might be less severely affected compared with the subjects in previous studies. We believe that the different characteristics of the study cohorts could result in different predicting factors for the risk of falls. OA was also a significant factor predicting the risk of falls in our study. This was concurrent with previous studies that reported the association of falls and postural control with OA (43, 44). There are certain limitations to be addressed. First, this was a cross-sectional study and the causal relationship between BMD and its associated factors might not be certified from our data. Second, the study did not evaluate the effects of specific medications, including cyclooxygenase
Table 4. Binary Logistic Regression Analysis to Identify Significant Factors Predicting Risk of Falls in Adults With Diabetes Mellitus Aged 50 Years or Older Variables
Estimate
SE
Odds Ratio
95% CI
P Value
OA Muscle mass Coefficient
0.756 0.00004 ⫺0.218
0.334 0.000019 0.832
2.129 1.00004 0.804
1.106 to 4.096 1.000001 to 1.000008 NA
.024 .028 .793
Abbreviations: CI, confidence interval; NA, not applicable.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
4212
Lee et al
BMD and Falls in DM
II inhibitor, nonsteroidal antiinflammatory drugs, thiazide diuretics, thiazolidinedione, selective serotonin reuptake inhibitors, and corticosteroids, which have been reported to affect bone metabolism. Third, the final outcome variables in the analyses were BMD and experience of falls during the past year and not osteoporotic fracture itself. Adequate evaluation of fracture incidence requires longer-term follow-up, and such data were not available for this study. Fourth, detailed medical histories of the subjects, such as those regarding thyroid disease, parathyroid disease, Cushing’s syndrome, and diabetic control, were not included in the database of this study. Furthermore, the frequency and severity of hypoglycemia may affect the risk of falls and fractures, and this issue needs to be investigated further in additional research regarding diabetic control. This study is considered clinically important because it identified potentially modifiable clinical characteristics such as muscle mass that could be risk factors for low BMD and falls in community-dwelling elderly subjects with diabetes mellitus. The study results are useful for both clinical practice and future research. In conclusion, in this cross-sectional study, we noted that total body muscle mass was the most prominent factor in femoral neck BMD and risk of falls in communitydwelling elderly subjects with DM. BMI was a significant factor associated with lumbar spine BMD, but further investigation is required to determine their role in the risk of falls and osteoporotic fractures. Prevention of osteoporosis and osteoporotic fracture in subjects with DM requires clinical attention because of these subjects’ concomitant morbidity and mortality.
Acknowledgments The authors thank Mi Sun Ryu, BS, and Hyun Mi Kim, BS, for their technical support. Address all correspondence and requests for reprints to: Moon Seok Park, MD, Department of Orthopaedic Surgery, Seoul National University Bundang Hospital, 300 Gumi-Dong, Bundang-Gu, Sungnam, Kyungki 463-707, Korea. E-mail: [email protected]. This work was supported by the Ministry of Trade, Industry, and Energy of Korea (Grant 10045220). Disclosure Summary: The authors have nothing to declare.
References 1. de Liefde 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;16:1713– 1720.
J Clin Endocrinol Metab, November 2014, 99(11):4206 – 4213
2. Patel S, Hyer S, Tweed K, Kerry S, Allan K, Rodin A, Barron J. Risk factors for fractures and falls in older women with type 2 diabetes mellitus. Calcif Tissue Int. 2008;82:87–91. 3. Yamaguchi T, Sugimoto T. Bone metabolism and fracture risk in type 2 diabetes mellitus. Endocr J. 2011;58:613– 624. 4. Hofbauer LC, Brueck CC, Singh SK, Dobnig H. Osteoporosis in patients with diabetes mellitus. J Bone Miner Res. 2007;22:1317– 1328. 5. Roman de Mettelinge T, Cambier D, Calders P, Van Den Noortgate N, Delbaere K. Understanding the relationship between type 2 diabetes mellitus and falls in older adults: a prospective cohort study. PloS One. 2013;8:e67055. 6. Saito M, Marumo K. Bone quality in diabetes. Front Endocrinol. 2013;4:72. 7. Strotmeyer ES, Cauley JA, Schwartz AV, et al. Nontraumatic fracture risk with diabetes mellitus and impaired fasting glucose in older white and black adults: the health, aging, and body composition study. Arch Intern Med. 2005;165:1612–1617. 8. Patsch JM, Burghardt AJ, Yap SP, et al. Increased cortical porosity in type 2 diabetic postmenopausal women with fragility fractures. J Bone Miner Res. 2013;28:313–324. 9. Ikeda T, Manabe H, Iwata K. Clinical significance of alendronate in postmenopausal type 2 diabetes mellitus. Diabetes Metab. 2004;30: 355–358. 10. Keegan TH, Schwartz AV, Bauer DC, Sellmeyer DE, Kelsey JL. Effect of alendronate on bone mineral density and biochemical markers of bone turnover in type 2 diabetic women: the fracture intervention trial. Diabetes Care. 2004;27:1547–1553. 11. Leslie WD, Rubin MR, Schwartz AV, Kanis JA. Type 2 diabetes and bone. J Bone Miner Res. 2012;27:2231–2237. 12. Kong AP, Xu G, Brown N, So WY, Ma RC, Chan JC. Diabetes and its comorbidities—where East meets West. Nat Rev Endocrinol. 2013;9:537–547. 13. Knapp S. Diabetes and infection: is there a link? A mini-review. Gerontology. 2013;59:99 –104. 14. Hwang HF, Lee HD, Huang HH, Chen CY, Lin MR. Fall mechanisms, bone strength, and hip fractures in elderly men and women in Taiwan. Osteoporos Int. 2011;22:2385–2393. 15. van Geel TA, van den Bergh JP, Dinant GJ, Geusens PP. Individualizing fracture risk prediction. Maturitas. 2010;65:143–148. 16. Chun MY. Validity and reliability of the Korean version of the International Physical Activity Questionnaire Short Form in the elderly. Korean J Fam Med. 2012;33:144 –151. 17. National Rural Resources Development Institute. Food Composition Table, 7th ed. Suwon: Rural Development Administration; 2006. 18. Kim J, Wang Z, Heymsfield SB, Baumgartner RN, Gallagher D. Total-body skeletal muscle mass: estimation by a new dual-energy X-ray absorptiometry method. Am J Clin Nutr. 2002;76:378 –383. 19. Tang SY, Allen MR, Phipps R, Burr DB, Vashishth D. Changes in non-enzymatic glycation and its association with altered mechanical properties following 1-year treatment with risedronate or alendronate. Osteoporos Int. 2009;20:887– 894. 20. Paul RG, Bailey AJ. Glycation of collagen: the basis of its central role in the late complications of ageing and diabetes. Int J Biochem Cell Biol. 1996;28:1297–1310. 21. Vashishth D, Gibson GJ, Khoury JI, Schaffler MB, Kimura J, Fyhrie DP. Influence of nonenzymatic glycation on biomechanical properties of cortical bone. Bone. 2001;28:195–201. 22. Burghardt AJ, Issever AS, Schwartz AV, et al. High-resolution peripheral quantitative computed tomographic imaging of cortical and trabecular bone microarchitecture in patients with type 2 diabetes mellitus. J Clin Endocrinol Metab. 2010;95:5045–5055. 23. Dobnig H, Piswanger-Sölkner JC, Roth M, et al. Type 2 diabetes mellitus in nursing home patients: effects on bone turnover, bone mass, and fracture risk. J Clin Endocrinol Metab. 2006;91:3355– 3363. 24. Gerdhem P, Isaksson A, Akesson K, Obrant KJ. Increased bone
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-1400
25.
26.
27.
28. 29.
30.
31.
32.
33.
34.
density and decreased bone turnover, but no evident alteration of fracture susceptibility in elderly women with diabetes mellitus. Osteoporos Int. 2005;16:1506 –1512. Kanazawa I, Yamaguchi T, Yamamoto M, et al. Serum osteocalcin level is associated with glucose metabolism and atherosclerosis parameters in type 2 diabetes mellitus. J Clin Endocrinol Metab. 2009; 94:45– 49. Moseley KF, Dobrosielski DA, Stewart KJ, De Beur SM, Sellmeyer DE. Lean mass and fat mass predict bone mineral density in middleaged individuals with noninsulin-requiring type 2 diabetes mellitus. Clin Endocrinol (Oxf). 2011;74:565–571. Wickham C, Cooper C, Margetts BM, Barker DJ. Muscle strength, activity, housing and the risk of falls in elderly people. Age Ageing. 1989;18:47–51. Pratesi A, Tarantini F, Di Bari M. Skeletal muscle: an endocrine organ. Clin Cases Miner Bone Metab. 2013;10:11–14. Park SW, Goodpaster BH, Strotmeyer ES, et al. Decreased muscle strength and quality in older adults with type 2 diabetes: the health, aging, and body composition study. Diabetes. 2006;55:1813–1818. Nelson ME, Fiatarone MA, Morganti CM, Trice I, Greenberg RA, Evans WJ. Effects of high-intensity strength training on multiple risk factors for osteoporotic fractures. A randomized controlled trial. JAMA. 1994;272:1909 –1914. Bridges MJ, Moochhala SH, Barbour J, Kelly CA. Influence of diabetes on peripheral bone mineral density in men: a controlled study. Acta Diabetol. 2005;42:82– 86. Wakasugi M, Wakao R, Tawata M, Gan N, Koizumi K, Onaya T. Bone mineral density measured by dual energy x-ray absorptiometry in patients with non-insulin-dependent diabetes mellitus. Bone. 1993;14:29 –33. Méndez JP, Rojano-Mejía D, Pedraza J, et al. Bone mineral density in postmenopausal Mexican-Mestizo women with normal body mass index, overweight, or obesity. Menopause. 2013;20:568 –572. Compston JE, Watts NB, Chapurlat R, et al. Obesity is not protective against fracture in postmenopausal women: GLOW. Am J Med. 2011;124:1043–1050.
jcem.endojournals.org
4213
35. Nielson CM, Marshall LM, Adams AL, et al. Osteoporotic Fractures in Men Study Research Group. BMI and fracture risk in older men: the osteoporotic fractures in men study (MrOS). J Bone Miner Res. 2011;26:496 –502. 36. Tanaka S, Kuroda T, Saito M, Shiraki M. Overweight/obesity and underweight are both risk factors for osteoporotic fractures at different sites in Japanese postmenopausal women. Osteoporos Int. 2013;24:69 –76. 37. Hulth AG, Nilsson BE, Westlin NE, Wiklund PE. Alkaline phosphatase in women with osteoporosis. Acta Med Scand. 1979;206: 201–203. 38. Iki M, Akiba T, Matsumoto T, et al. Reference database of biochemical markers of bone turnover for the Japanese female population. Japanese Population-based Osteoporosis (JPOS) Study. Osteoporos Int. 2004;15:981–991. 39. Chevalley T1, Rizzoli R, Nydegger V, et al. Effects of calcium supplements on femoral bone mineral density and vertebral fracture rate in vitamin-D-replete elderly patients. Osteoporos Int. 1994;4:245– 252. 40. Schwartz AV, Hillier TA, Sellmeyer DE, et al. Older women with diabetes have a higher risk of falls: a prospective study. Diabetes Care. 2002;25:1749 –1754. 41. Schwartz AV, Vittinghoff E, Sellmeyer DE, et al. Health, Aging, and Body Composition Study. Diabetes-related complications, glycemic control, and falls in older adults. Diabetes Care. 2008;31:391–396. 42. Volpato S, Leveille SG, Blaum C, Fried LP, Guralnik JM. Risk factors for falls in older disabled women with diabetes: the women’s health and aging study. J Gerontol A Biol Sci Med Sci. 2005;60: 1539 –1545. 43. Jones G, Nguyen T, Sambrook PN, Lord SR, Kelly PJ, Eisman JA. Osteoarthritis, bone density, postural stability, and osteoporotic fractures: a population based study. J Rheumatol. 1995;22:921– 925. 44. Hoops ML, Rosenblatt NJ, Hurt CP, Crenshaw J, Grabiner MD. Does lower extremity osteoarthritis exacerbate risk factors for falls in older adults? Womens Health (Lond Engl). 2012;8:685– 696.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
ARTICLE C a r e
Factors Associated With Bone Mineral Density and Risk of Fall in Korean Adults With Type 2 Diabetes Mellitus Aged 50 Years and Older Kyoung Min Lee, Chin Youb Chung, Soon-Sun Kwon, Seung Yeol Lee, Tae Gyun Kim, Young Choi, and Moon Seok Park Department of Orthopaedic Surgery (K.M.L., C.Y.C., S.Y.L., T.G.K., Y.C., M.S.P.) and Biomedical Research Institute (S.-S.K.), Seoul National University Bundang Hospital, Kyungki 463-707, Korea
Context: Osteoporotic fractures in subjects with diabetes mellitus (DM) carry higher mortality and morbidity. Because bone strength and minor trauma, such as a falls, are considered to be significant factors contributing to osteoporotic fractures, it is important to elucidate the associated factors with these. Objective: This study was performed to investigate the factors associated with bone mineral density (BMD) and falls in noninstitutionalized subjects with DM aged 50 years or older. Design, Setting, and Patients: We used the database from the 2010 Fifth Korea National Health and Nutrition Examination Survey. Subjects with DM aged 50 years or older were selected and included in the data analyses. Associated factors with BMD of the femoral neck and lumbar spine and those with falls were analyzed using multiple linear regression and binary logistic regression analyses, respectively. Results: Three hundred sixty-two subjects [209 males; 153 females; average age, 66.0 y (SD 8.2 y)] were included. Among the male subjects, the total body muscle mass (P ⬍ .001), daily calcium intake (P ⫽ .001), ALP levels (P ⫽ .007), and body mass index (P ⫽ .027) were significantly associated with femoral neck BMD, whereas body mass index (P ⫽ .001) and ALP levels (P ⫽ .040) were associated with lumbar spine BMD. Among the female subjects, age (P ⬍ .001), daily calcium intake (P ⫽ .011) and total body muscle mass (P ⫽ .023) were found to be significantly associated factors with femoral neck BMD, whereas age (P ⬍ .001) and body mass index (P ⫽ .012) and daily calcium intake (P ⫽ .040) were those with lumbar spine BMD. Osteoarthritis (P ⫽ .024) and total body muscle mass (P ⫽ .028) were found to be significantly associated with the risk of falls. Conclusions: Total body muscle mass was the most prominent factor predicting femoral neck BMD and risk of falls in community-dwelling elderly subjects with DM. Further investigation is required to determine their role in preventing osteoporotic fractures in diabetic subjects. (J Clin Endocrinol Metab 99: 4206 – 4213, 2014)
ecently diabetes mellitus (DM) has been drawing clinical attention as an independent risk factor for fragility fracture, although several studies have shown that subjects with DM do not have lower bone mineral density (BMD) (1, 2). These authors suggested that the susceptibility to fracture in DM is influenced by bone quality and increased risk of fall and not by bone quantity (3– 6). How-
R
ever, bone quality measurement has not been generalized in current clinical practice. In addition, some previous studies have reported that subjects with DM who sustained incident fractures had decreased BMD compared with those subjects with DM without such fractures (7, 8). Therefore, bone quantity, as measured by BMD, is still an apparent risk factor for osteoporotic fractures, and in-
ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received February 10, 2014. Accepted August 5, 2014. First Published Online August 20, 2014
Abbreviations: ALP, alkaline phosphatase; BMD, bone mineral density; BMI, body mass index; BUN, blood urea nitrogen; DM, diabetes mellitus; DXA, dual-energy X-ray absorptiometry; FBS, fasting blood glucose; KNHNES V, Fifth Korea National Health and Nutrition Examination Survey; OA, osteoarthritis of hip or knee.
4206
jcem.endojournals.org
J Clin Endocrinol Metab, November 2014, 99(11):4206 – 4213
doi: 10.1210/jc.2014-1400
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-1400
jcem.endojournals.org
4207
creasing the BMD is the current clinical target to prevent such fractures (9 –11). Osteoporotic fractures in subjects with DM is considered to carry higher mortality, morbidity, and subsequent sociomedical costs than such fractures in those without DM because of various concomitant medical conditions, such as impaired cardiovascular function, renal function, and immune response (12, 13). As such, preventing osteoporotic fractures in subjects with DM is a crucial medical issue. Because bone strength and minor trauma, such as a falls, are considered to be significant factors contributing to osteoporotic fractures (14, 15), it is important to elucidate the variables predicting these factors in order to reduce and prevent fractures. However, these have not been sufficiently investigated in elderly subjects with DM, although osteoporotic fractures are primarily involved in elderly population and could be complicated by DM. Therefore, in the present study, we aimed to identify the factors associated with BMD and risk of falls in Korean adults with type 2 DM aged 50 years or older using data from the Fifth Korea National Health and Nutrition Examination Survey (KNHNES V).
Materials and Methods Subjects The KNHNES V was conducted by the Korean Centers for Disease Control and Prevention in 2010. In total, 12 722 subjects selected from stratified, multistage probability samples of Korean households were invited to participate in the survey, and 10 533 agreed to participate, corresponding to a response rate of 82.8%. A total of 504 subjects aged 50 years or older with DM were selected for inclusion in this study. Of these, subjects who were taking osteoporosis medication, oral contraceptives, or hormone replacement therapy; those with any malignancy; and those with incomplete data were excluded from the data analyses (Figure 1). The survey included a health interview, health behavior, health examination, and nutrition surveys; of these, the health interview, health behavior, and health examination components were used in the present study. Written informed consent was obtained from all participants by the Korean Centers for Disease Control and Prevention, and the study was exempt from approval by the Institutional Review Board of Seoul National University Bundang Hospital because it used a publicly accessible database and did not involve any potential violation of human rights.
Data collection The KNHNES V 2010 database contains information on demographics, including age; sex; body mass index (BMI); weight change in the past year (gain of ⬎ 10 kg, gain of 6 –10 kg, gain of 3– 6 kg, gain of ⬍ 3 kg, no change, loss of ⬍ 3 kg, loss of 3– 6 kg, loss of 6 –10 kg, and loss of ⬎ 10 kg); house income (below the 25th percentile, between the 25th and 50th percentiles, between the 50th and 75th percentiles, greater than the 75th per-
Figure 1. Flow diagram of the inclusion and exclusion criteria in the KNHANES V. A total of 362 adults with DM aged 50 years or older were finally included.
centile); education level (elementary school, middle school, high school, and college); amount of smoking (current and cumulative); amount of alcohol consumption in the past year; presence of other medical conditions [hypertension, DM, hyperlipidemia, coronary artery disease, cerebrovascular accident, osteoarthritis of hip or knee (OA), chronic renal failure]; and current use of osteoporosis medication. The presence of diabetic retinopathy also was assessed. Family history of osteoporosis was recorded. In addition, experience of falls in the past year was obtained using the questionnaire. The parameters of height and weight were measured using standardized instruments, and BMI was calculated using the height and weight measurements. Weight change in the past year and education level were obtained from the questionnaire. House income was based on the total house income reported by the subjects. The amount of cumulative smoking was calculated from the amount of daily smoking and the duration of smoking, which were included in the questionnaire. The amount of alcohol consumption was calculated using the data on the average frequency of drinking and the average amount of drinking per ses-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
4208
Lee et al
BMD and Falls in DM
sion (in glasses) obtained from the questionnaire, with the assumption that one glass contains 10 g of alcohol. The presence of other medical conditions was determined from the diagnoses obtained from the questionnaire. Activity level was evaluated by the short form of the International Physical Activity Questionnaire (16), and weekly hours of vigorous-intensity, moderateintensity, and walking activities were calculated. Blood tests included fasting blood glucose (FBS), glycosylated hemoglobin, insulin, total cholesterol, triglycerides, hemoglobin, blood urea nitrogen (BUN), creatinine, 25-hydroxyvitamin D, alkaline phosphatase (ALP), and PTH. Blood samples were obtained throughout the year after an 8-hour fast and were immediately processed, refrigerated, and transported in cold storage to the central testing institute (NeoDin Medical Institute, Seoul, South Korea), where they were analyzed within 24 hours. Fundus photography was performed in subjects with DM. Seven standard photographs from the Early Treatment for Diabetic Retinopathy Study were obtained for each eye after pharmacological pupil dilatation. Diabetic retinopathy was defined as the presence of one or more retinal microaneurysms or retinal blot hemorrhages with or without more severe lesions (hard exudates, soft exudates, intraretinal microvascular abnormalities, venous bleeding, new retinal vessels, and fibroproliferations). Daily calcium intake was calculated from the Food Composition Table (17) and 24-hour dietary recall questionnaire, which was evaluated by a dietitian. BMD (grams per square centimeter) was measured at the femoral neck and lumbar spine using dual-energy X-ray absorptiometry (DXA; DISCOVERY-W fan-beam densitometer; Hologic Inc) with a coefficients of variation of 1.9% and 2.5%, respectively. Body composition also was analyzed, and total body muscle mass and total body fat percentiles were obtained. Total body muscle mass was calculated as follows: total body skeletal muscle ⫽ (1.19 appendicular lean soft tissue) ⫺ 1.01 (appendicular lean soft tissue mass measured by DXA machine) (18)
Data analysis The factors associated with BMD were analyzed for male and female subjects separately, whereas the factors predicting the risk of falls were evaluated for a single group containing both male and female subjects. The candidate variables that were significant in univariate analysis were selected and included in multiple regression analysis to evaluate their association with BMD and included in binary logistic regression analysis to assess the significant predicting factors of the risk of falls, respectively.
Statistics A descriptive analysis including the average and SD or proportion was performed for all variables. Data normality was tested using the Kolmogorov-Smirnov test. Correlations between the variables and BMDs of the femoral neck and lumbar spine were evaluated using Pearson’s correlation coefficient or Spearman’s correlation coefficient, depending on the normality of the data. The comparison of BMD between the two groups was performed using the Student’s t test. Variables that significantly correlated with BMD or conditions that showed significantly different BMD between cases were selected and included in the multiple regression analysis. For male subjects, the total body muscle mass, daily calcium intake, ALP, BMI, age, current amount of smoking, and triglycerides were included in the multiple regression analysis to evaluate femoral neck BMD; further-
J Clin Endocrinol Metab, November 2014, 99(11):4206 – 4213
more, BMI, ALP, and total body muscle mass were included in the multiple regression analysis model to evaluate lumbar spine BMD. For female subjects, age, daily calcium intake, total body muscle mass, house income, education level, FBS, BUN, and cumulative amount of smoking were included in the multiple regression analysis model to evaluate femoral neck BMD, and age, BMI, daily calcium intake, and total body muscle mass were included in the multiple regression analysis model to evaluate lumbar spine BMD. The categorical and binary variables that were significantly associated with a history of falls in the 2 test or continuous variables that were significantly different according to a history of falls in Student’s t test were selected and included in the binary logistic regression analysis. These were OA and total body muscle mass. A stepwise selection of the variables was adopted in multiple regression analysis to identify the significant associated factors with BMD, in which the dependent variables were BMD values of the femoral neck and lumbar spine. Binary logistic regression analysis was utilized to examine the significant factors that predict the risk of falls. All statistical analyses were performed using SPSS version 18.0 software (IBM Corp), with statistical significance set at P ⬍ .05.
Results A total of 362 subjects (209 males, 153 females) were included in the study. The average age of the subjects was 66.0 years (SD 8.2 y), and the average time from diagnosis of DM was 8.9 years (SD 8.1 y). The average femoral neck BMD of the male and female subjects was 0.7543 g/cm2 (SD 0.1196 g/cm2) and 0.5937 g/cm2 (SD 0.1070 g/cm2), respectively, whereas the average lumbar spine BMD was 0.9892 g/cm2 (SD 0.1552 g/cm2) and 0.8049 g/cm2 (SD 0.1243 g/cm2), respectively. A total of 53 subjects had a history of falls in the previous year (Table 1). Male subjects had significantly higher total body muscle mass than female subjects, and the total body muscle mass decreases with increasing age in both genders (Figure 2). In male adults with DM aged ⱖ 50 years, total body muscle mass (P ⬍ .001), daily calcium intake (P ⫽ .001) ALP levels (P ⫽ .007), and BMI (P ⫽ .027) were found to be significant associated factors with femoral neck BMD, whereas BMI (P ⫽ .001) and ALP levels (P ⫽ .040) were significantly associated with lumbar spine BMD (Table 2). In female adults with DM ⱖ 50 years, age (P ⬍ .001), daily calcium intake (P ⫽.011), and total body muscle mass (P ⫽ .023) were found to be significant associated factors with femoral neck BMD, whereas age (P ⬍ .001) and BMI (P ⫽ .012), and daily calcium intake (P ⫽ .040) were significantly associated with lumbar spine BMD (Table 3). OA (P ⫽ .024) and total body muscle mass (P ⫽ .028) were found to be the only significant factors that predicted the risk of falls in adults with DM aged 50 years or older (Table 4).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-1400
Table 1.
jcem.endojournals.org
4209
Data Summary
n Age, y Time from diagnosis of DM, y BMI, kg/cm2 Weight change (⬎10 kg gain/6 –10 kg gain/3– 6 kg gain/0 –3 kg gain/no change/0 –3 kg loss/3– 6 kg loss/6 –10 kg loss/⬎10 kg loss) House income (0 –25/25–50/50 –75/75–100 percentiles Education level, elementary/middle/high/college Current amount of smoking, pieces Cumulative amount of smoking, pack Amount of alcohol consumption, g/d Family history of osteoporosis, yes/noa Medical diseases HT, yes/no HL, yes/no CAD, yes/no CVA, yes/no OA, yes/no CRF, yes/no Diabetic retinopathy, yes/no Activity Weekly duration of vigorous-intensity activity, min Weekly duration of moderate-intensity activity, min Weekly duration of walking, min Weekly total MET Blood test FBS, mg/dL HbAlc, % Insulin, IU/mL Total cholesterol, mg/dL Triglycerides, mg/dL Hb, g/dL BUN, mg/dL Cr, mg/dL 25(OH)D, ng/mL ALP, IU/L PTH, pg/mL Daily calcium intake, mg Total body muscle mass, kg Total body fat percentile, % BMD (femoral neck), g/cm2 BMD (lumbar spine), g/cm2 Falls in the past year, yes/no
Male Subjects
Female Subjects
P Value
209 64.6 (8.1) 8.8 (8.2) 24.8 (3.1) 4/9/33/146/14/2/1
153 68.0 (8.1) 9.1 (8.0) 25.8 (3.9) 2/10/30/100/10/1/0
NA ⬍.001 .716 ⬍.006 .660
80/53/42/34 74/52/54/29 4.9 (8.8) 5473.8 (8726.0) 14.2 (21.7) 20/180
75/39/20/19 123/13/14/3 0.7 (3.1) 170.6 (1007.3) 0.9 (5.3) 21/126
.134 ⬍.001 ⬍.001 ⬍.001 ⬍.001 .228
121/88 42/167 17/192 17/192 23/186 3/206 33/176
97/56 38/115 12/141 10/143 65/88 3/150 30/123
.277 .251 1.000 .678 ⬍.001 .700 .400
92.6 (226.0) 190.6 (410.8) 458.2 (572.9) 3015.7 (3566.9)
56.1 (201.5) 192.5 (504.4) 305.1 (569.4) 2226.0 (3612.2)
.107 .969 ⬍.001 .016
141.8 (57.4) 7.3 (1.5) 10.1 (4.9) 177.5 (38.4) 186.0 (132.3) 14.4 (1.4) 17.6 (5.9) 1.02 (0.322) 20.4 (7.4) 244.8 (85.5) 63.9 (24.8) 565.5 (416.6) 47.2 (6.6) 24.1 (5.2) 0.7543 (0.1196) 0.9892 (0.1552) 21/188
140.0 (43.2) 7.4 (1.3) 12.6 (7.5) 196.0 (37.4) 175.1 (94.2) 12.9 (1.1) 17.0 (6.2) 0.77 (0.178) 17.7 (7.1) 264.8 (83.7) 69.8 (26.2) 374.5 (261.4) 34.1 (4.5) 36.7 (5.2) 0.5937 (0.1070) 0.8049 (0.1243) 32/121
.776 .492 .004 ⬍.001 .429 ⬍.001 .348 ⬍.001 .001 .046 .026 ⬍.001 ⬍.001 ⬍.001 ⬍.001 ⬍.001 .008
Abbreviations: CAD, coronary artery disease; Cr, creatinine; CRF, chronic renal failure; CVA, cerebrovascular accident; Hb, hemoglobin; HbAlc, glycosylated hemoglobin; Hct, hematocrit; HDL, high-density lipoprotein; HL, hyperlipidemia; HT, hypertension; LDL, low-density lipoprotein; MET, metabolic equivalent; NA, not applicable; 25(OH)D, 25-hydroxyvitamin D. Data are presented as mean (SD). a
Fifteen subjects (nine males and six females) did not know the details about their family histories of osteoporosis or osteoporotic fractures.
Total body muscle mass was significantly correlated with weekly hours of vigorous-intensity activity (r ⫽ 0.144, P ⫽ .044) but not with moderate-intensity activity (r ⫽ 0.062, P ⫽ .384) and walking (r ⫽ 0.087, P ⫽ .224).
Discussion This study investigated the factors associated with BMD and the factors predicting the risk of falls in communitydwelling subjects with DM aged 50 years or older. Total
body muscle mass was found to be a significant factor associated with BMD of the femoral neck and a significant factor predicting the risk of falls, whereas BMI was a significantly associated factor with lumbar spine BMD in both sexes with DM. ALP levels in the male adults and age in the female adults were found to be significant factors associated with femoral neck and lumbar spine BMD. Recently DM has drawn attention as an independent risk factor for osteoporotic fractures. The relationship between DM and fragility fracture has been clarified and reportedly involves the impairment of bone mechanical
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
4210
Lee et al
BMD and Falls in DM
J Clin Endocrinol Metab, November 2014, 99(11):4206 – 4213
Figure 2. The distribution of total body muscle mass in male and female subjects with diabetes according to age. Male subjects show higher total body muscle mass than female subjects, and the total body muscle mass decreases with increasing age in both genders.
strength by the glycosylation of type I collagen (19 –21). Unlike type 1 DM with low BMD, type 2 DM has been reported to have increased fracture risk despite normal or increased BMD, which is paradoxical to the conventional view of osteoporosis (1). Impaired bone quality and increased risk of falls have been proposed to explain this paradox in subjects with type 2 DM (4). Although trials have been conducted to evaluate bone quality statically and dynamically using quantitative computed tomography and bone turnover markers (22–25), quantitative BMD is still an apparent risk factor for fracture and remains the mainstream target of clinical management in subjects with DM (9 –11). Additionally, previous studies have reported that subjects with DM who sustained fractures had lower BMD than those without fractures, suggesting the clinical role of quantitative bone mass in osteoporotic fractures among subjects with DM (7, 8). Therefore, this study focused on the two relevant factors for osteoporotic fractures in subjects with DM in current clinical practice: BMD and falls. Total body muscle mass was found to be a significant factor in both femoral neck BMD and the risk of falls.
Previous studies have shown the relationship between osteoporosis and sarcopenia as well as between the risk of falls and sarcopenia in the elderly population (26, 27). Furthermore, skeletal muscle is considered to be involved in the endocrine system by modulating inflammatory cytokines, which also play a role in the development of type 2 DM (28). Subjects with DM are reported to have impaired muscle quality and strength (29). Frail or atrophic elderly individuals are believed to be more susceptible to falls, whereas people who exercise frequently or have higher levels of physical activity would be expected to have greater muscle mass; therefore, they may have greater stability when walking, running or performing activities of daily living. Therefore, optimizing total body muscle mass in these subjects could result in increased BMD and decreased risk of falls, thus eventually preventing osteoporotic fractures and possibly improving diabetes control. A previous study has shown that resistance training increased muscle mass and insulin action and had a positive effect on multiple risk factors for osteoporotic fractures in sedentary postmenopausal women (30). We found a significant correlation between weekly hours of vigorousintensity physical activity and total body muscle mass, which is in agreement with the results of the previous study. However, the effect of resistance training on the risk of osteoporotic fractures in elderly subjects with DM needs to be investigated further using a prospective longitudinal study design. In addition, total body muscle mass may be a reflection of age, postmenopausal duration, and a combination of life-long and recent androgen and estrogen status. These factors should be investigated further in a well-designed study. In the present study, we noted that higher BMI was a protective factor for lumbar spine osteoporosis in elderly subjects with type 2 DM in both sexes, which is consistent with the findings of previous studies suggesting that increased BMD in subjects with type 2 DM is caused by increased BMI (31–33). Although these studies have
Table 2. Multiple Regression Analysis to Identify Significant Factors Associated With Femoral Neck and Lumbar Spine BMD in Male Adults With Diabetes Mellitus Aged 50 Years or Older Femoral Neck BMD
Total body muscle mass Daily calcium intake ALP BMI Age Smoking (current) Triglycerides Coefficient
Lumbar Spine BMD
Standardized 
t
P Value
0.333 0.216 ⫺0.170 0.210 ⫺0.114 0.029 0.119 0.112
3.613 3.548 ⫺2.743 2.238 ⫺1.631 0.456 1.933 1.206
⬍.001 .001 .007 .027 .105 .643 .055 .230
BMI ALP Muscle mass Coefficient
Standardized 
t
P Value
0.252 ⫺0.151 0.049 0.737
3.472 ⫺2.071 0.454 7.086
.001 .040 .650 ⬍.001
R2 ⫽ 0.417; R2 ⫽ 0.100.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-1400
jcem.endojournals.org
4211
Table 3. Multiple Regression Analysis to Identify Significant Factors Associated With Femoral Neck and Lumbar Spine BMD in Female Adults With Diabetes Mellitus Aged 50 Years or Older Femoral Neck BMD
Age Daily calcium intake Muscle mass Income Education FBS BUN Smoking (cumulative) Coefficient
Lumbar Spine BMD
Standardized 
t
P Value
⫺0.590 0.195 0.185 0.106 0.158 0.113 ⫺0.085 ⫺0.082 1.129
⫺7.727 2.581 2.299 1.329 1.897 1.462 ⫺1.088 ⫺1.041 16.076
⬍.001 .011 .023 .187 .060 .146 .279 .300 ⬍.001
Age BMI Daily Calcium intake Muscle mass Coefficient
Standardized 
t
P Value
⫺0.344 0.205 0.171 0.080 0.964
⫺4.184 2.541 2.079 0.685 8.471
⬍.001 .012 .040 .495 ⬍.001
R2 ⫽ 0.385; R2 ⫽ 0.227.
shown that BMI was positively correlated with BMD, other studies have shown that obesity is a risk factor for fractures (34 –36). Thus, increased BMI could be a protective factor for osteoporosis and a risk factor for osteoporotic fractures at the same time, which is somewhat paradoxical. Increased BMI could result in increased strain on the skeleton in daily activities improving bone mass, whereas metabolic derangement in obese subjects with type 2 DM has an impact on bone quality and strength, which cannot be measured by DXA. Another possible explanation of the increased number of fractures in obese subjects is that obesity is a risk factor for frequent falls and more stress is exerted on the bone at the time of injury, thus increasing the possibility of fractures. ALP is produced by several organs, including the liver, bile duct, kidneys, and bone, and is primarily known to be associated with osteoblastic activity in bone metabolism. ALP level was a significant factor associated with femoral neck and lumbar spine BMD in male subjects with type 2 DM in the present study, showing a negative correlation. This could be interpreted that the subject with low BMD is concurrent with compensatory increased osteoblastic activity in male diabetic subjects. These results are supported by previous studies reporting an increase in ALP activity in women with more severe osteoporosis (37, 38). The amount of daily calcium intake was found to be significantly associated with femoral neck BMD in male subjects and both femoral neck and lumbar spine BMD in
female subjects. A previous study reported the beneficial effect of calcium supplements on femoral BMD and vertebral fracture rate in vitamin D-replete elderly patients (39). The current cross-sectional study results show a positive correlation between calcium intake and BMD in elderly subjects with diabetes, but the causal relationship needs to be investigated further in a longitudinal study. The increased risk of falls in subjects with DM has been explained by poorer balance, which may be a result of peripheral neuropathy, reduced vision, nephropathy, and lower muscle strength (40 – 42). Although our study data did not include the presence of peripheral neuropathy, total body muscle mass was found to significantly predict the risk of falls, whereas diabetic retinopathy and nephropathy did not. Our study population comprised community-dwelling elderly subjects with DM who might be less severely affected compared with the subjects in previous studies. We believe that the different characteristics of the study cohorts could result in different predicting factors for the risk of falls. OA was also a significant factor predicting the risk of falls in our study. This was concurrent with previous studies that reported the association of falls and postural control with OA (43, 44). There are certain limitations to be addressed. First, this was a cross-sectional study and the causal relationship between BMD and its associated factors might not be certified from our data. Second, the study did not evaluate the effects of specific medications, including cyclooxygenase
Table 4. Binary Logistic Regression Analysis to Identify Significant Factors Predicting Risk of Falls in Adults With Diabetes Mellitus Aged 50 Years or Older Variables
Estimate
SE
Odds Ratio
95% CI
P Value
OA Muscle mass Coefficient
0.756 0.00004 ⫺0.218
0.334 0.000019 0.832
2.129 1.00004 0.804
1.106 to 4.096 1.000001 to 1.000008 NA
.024 .028 .793
Abbreviations: CI, confidence interval; NA, not applicable.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
4212
Lee et al
BMD and Falls in DM
II inhibitor, nonsteroidal antiinflammatory drugs, thiazide diuretics, thiazolidinedione, selective serotonin reuptake inhibitors, and corticosteroids, which have been reported to affect bone metabolism. Third, the final outcome variables in the analyses were BMD and experience of falls during the past year and not osteoporotic fracture itself. Adequate evaluation of fracture incidence requires longer-term follow-up, and such data were not available for this study. Fourth, detailed medical histories of the subjects, such as those regarding thyroid disease, parathyroid disease, Cushing’s syndrome, and diabetic control, were not included in the database of this study. Furthermore, the frequency and severity of hypoglycemia may affect the risk of falls and fractures, and this issue needs to be investigated further in additional research regarding diabetic control. This study is considered clinically important because it identified potentially modifiable clinical characteristics such as muscle mass that could be risk factors for low BMD and falls in community-dwelling elderly subjects with diabetes mellitus. The study results are useful for both clinical practice and future research. In conclusion, in this cross-sectional study, we noted that total body muscle mass was the most prominent factor in femoral neck BMD and risk of falls in communitydwelling elderly subjects with DM. BMI was a significant factor associated with lumbar spine BMD, but further investigation is required to determine their role in the risk of falls and osteoporotic fractures. Prevention of osteoporosis and osteoporotic fracture in subjects with DM requires clinical attention because of these subjects’ concomitant morbidity and mortality.
Acknowledgments The authors thank Mi Sun Ryu, BS, and Hyun Mi Kim, BS, for their technical support. Address all correspondence and requests for reprints to: Moon Seok Park, MD, Department of Orthopaedic Surgery, Seoul National University Bundang Hospital, 300 Gumi-Dong, Bundang-Gu, Sungnam, Kyungki 463-707, Korea. E-mail: [email protected]. This work was supported by the Ministry of Trade, Industry, and Energy of Korea (Grant 10045220). Disclosure Summary: The authors have nothing to declare.
References 1. de Liefde 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;16:1713– 1720.
J Clin Endocrinol Metab, November 2014, 99(11):4206 – 4213
2. Patel S, Hyer S, Tweed K, Kerry S, Allan K, Rodin A, Barron J. Risk factors for fractures and falls in older women with type 2 diabetes mellitus. Calcif Tissue Int. 2008;82:87–91. 3. Yamaguchi T, Sugimoto T. Bone metabolism and fracture risk in type 2 diabetes mellitus. Endocr J. 2011;58:613– 624. 4. Hofbauer LC, Brueck CC, Singh SK, Dobnig H. Osteoporosis in patients with diabetes mellitus. J Bone Miner Res. 2007;22:1317– 1328. 5. Roman de Mettelinge T, Cambier D, Calders P, Van Den Noortgate N, Delbaere K. Understanding the relationship between type 2 diabetes mellitus and falls in older adults: a prospective cohort study. PloS One. 2013;8:e67055. 6. Saito M, Marumo K. Bone quality in diabetes. Front Endocrinol. 2013;4:72. 7. Strotmeyer ES, Cauley JA, Schwartz AV, et al. Nontraumatic fracture risk with diabetes mellitus and impaired fasting glucose in older white and black adults: the health, aging, and body composition study. Arch Intern Med. 2005;165:1612–1617. 8. Patsch JM, Burghardt AJ, Yap SP, et al. Increased cortical porosity in type 2 diabetic postmenopausal women with fragility fractures. J Bone Miner Res. 2013;28:313–324. 9. Ikeda T, Manabe H, Iwata K. Clinical significance of alendronate in postmenopausal type 2 diabetes mellitus. Diabetes Metab. 2004;30: 355–358. 10. Keegan TH, Schwartz AV, Bauer DC, Sellmeyer DE, Kelsey JL. Effect of alendronate on bone mineral density and biochemical markers of bone turnover in type 2 diabetic women: the fracture intervention trial. Diabetes Care. 2004;27:1547–1553. 11. Leslie WD, Rubin MR, Schwartz AV, Kanis JA. Type 2 diabetes and bone. J Bone Miner Res. 2012;27:2231–2237. 12. Kong AP, Xu G, Brown N, So WY, Ma RC, Chan JC. Diabetes and its comorbidities—where East meets West. Nat Rev Endocrinol. 2013;9:537–547. 13. Knapp S. Diabetes and infection: is there a link? A mini-review. Gerontology. 2013;59:99 –104. 14. Hwang HF, Lee HD, Huang HH, Chen CY, Lin MR. Fall mechanisms, bone strength, and hip fractures in elderly men and women in Taiwan. Osteoporos Int. 2011;22:2385–2393. 15. van Geel TA, van den Bergh JP, Dinant GJ, Geusens PP. Individualizing fracture risk prediction. Maturitas. 2010;65:143–148. 16. Chun MY. Validity and reliability of the Korean version of the International Physical Activity Questionnaire Short Form in the elderly. Korean J Fam Med. 2012;33:144 –151. 17. National Rural Resources Development Institute. Food Composition Table, 7th ed. Suwon: Rural Development Administration; 2006. 18. Kim J, Wang Z, Heymsfield SB, Baumgartner RN, Gallagher D. Total-body skeletal muscle mass: estimation by a new dual-energy X-ray absorptiometry method. Am J Clin Nutr. 2002;76:378 –383. 19. Tang SY, Allen MR, Phipps R, Burr DB, Vashishth D. Changes in non-enzymatic glycation and its association with altered mechanical properties following 1-year treatment with risedronate or alendronate. Osteoporos Int. 2009;20:887– 894. 20. Paul RG, Bailey AJ. Glycation of collagen: the basis of its central role in the late complications of ageing and diabetes. Int J Biochem Cell Biol. 1996;28:1297–1310. 21. Vashishth D, Gibson GJ, Khoury JI, Schaffler MB, Kimura J, Fyhrie DP. Influence of nonenzymatic glycation on biomechanical properties of cortical bone. Bone. 2001;28:195–201. 22. Burghardt AJ, Issever AS, Schwartz AV, et al. High-resolution peripheral quantitative computed tomographic imaging of cortical and trabecular bone microarchitecture in patients with type 2 diabetes mellitus. J Clin Endocrinol Metab. 2010;95:5045–5055. 23. Dobnig H, Piswanger-Sölkner JC, Roth M, et al. Type 2 diabetes mellitus in nursing home patients: effects on bone turnover, bone mass, and fracture risk. J Clin Endocrinol Metab. 2006;91:3355– 3363. 24. Gerdhem P, Isaksson A, Akesson K, Obrant KJ. Increased bone
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-1400
25.
26.
27.
28. 29.
30.
31.
32.
33.
34.
density and decreased bone turnover, but no evident alteration of fracture susceptibility in elderly women with diabetes mellitus. Osteoporos Int. 2005;16:1506 –1512. Kanazawa I, Yamaguchi T, Yamamoto M, et al. Serum osteocalcin level is associated with glucose metabolism and atherosclerosis parameters in type 2 diabetes mellitus. J Clin Endocrinol Metab. 2009; 94:45– 49. Moseley KF, Dobrosielski DA, Stewart KJ, De Beur SM, Sellmeyer DE. Lean mass and fat mass predict bone mineral density in middleaged individuals with noninsulin-requiring type 2 diabetes mellitus. Clin Endocrinol (Oxf). 2011;74:565–571. Wickham C, Cooper C, Margetts BM, Barker DJ. Muscle strength, activity, housing and the risk of falls in elderly people. Age Ageing. 1989;18:47–51. Pratesi A, Tarantini F, Di Bari M. Skeletal muscle: an endocrine organ. Clin Cases Miner Bone Metab. 2013;10:11–14. Park SW, Goodpaster BH, Strotmeyer ES, et al. Decreased muscle strength and quality in older adults with type 2 diabetes: the health, aging, and body composition study. Diabetes. 2006;55:1813–1818. Nelson ME, Fiatarone MA, Morganti CM, Trice I, Greenberg RA, Evans WJ. Effects of high-intensity strength training on multiple risk factors for osteoporotic fractures. A randomized controlled trial. JAMA. 1994;272:1909 –1914. Bridges MJ, Moochhala SH, Barbour J, Kelly CA. Influence of diabetes on peripheral bone mineral density in men: a controlled study. Acta Diabetol. 2005;42:82– 86. Wakasugi M, Wakao R, Tawata M, Gan N, Koizumi K, Onaya T. Bone mineral density measured by dual energy x-ray absorptiometry in patients with non-insulin-dependent diabetes mellitus. Bone. 1993;14:29 –33. Méndez JP, Rojano-Mejía D, Pedraza J, et al. Bone mineral density in postmenopausal Mexican-Mestizo women with normal body mass index, overweight, or obesity. Menopause. 2013;20:568 –572. Compston JE, Watts NB, Chapurlat R, et al. Obesity is not protective against fracture in postmenopausal women: GLOW. Am J Med. 2011;124:1043–1050.
jcem.endojournals.org
4213
35. Nielson CM, Marshall LM, Adams AL, et al. Osteoporotic Fractures in Men Study Research Group. BMI and fracture risk in older men: the osteoporotic fractures in men study (MrOS). J Bone Miner Res. 2011;26:496 –502. 36. Tanaka S, Kuroda T, Saito M, Shiraki M. Overweight/obesity and underweight are both risk factors for osteoporotic fractures at different sites in Japanese postmenopausal women. Osteoporos Int. 2013;24:69 –76. 37. Hulth AG, Nilsson BE, Westlin NE, Wiklund PE. Alkaline phosphatase in women with osteoporosis. Acta Med Scand. 1979;206: 201–203. 38. Iki M, Akiba T, Matsumoto T, et al. Reference database of biochemical markers of bone turnover for the Japanese female population. Japanese Population-based Osteoporosis (JPOS) Study. Osteoporos Int. 2004;15:981–991. 39. Chevalley T1, Rizzoli R, Nydegger V, et al. Effects of calcium supplements on femoral bone mineral density and vertebral fracture rate in vitamin-D-replete elderly patients. Osteoporos Int. 1994;4:245– 252. 40. Schwartz AV, Hillier TA, Sellmeyer DE, et al. Older women with diabetes have a higher risk of falls: a prospective study. Diabetes Care. 2002;25:1749 –1754. 41. Schwartz AV, Vittinghoff E, Sellmeyer DE, et al. Health, Aging, and Body Composition Study. Diabetes-related complications, glycemic control, and falls in older adults. Diabetes Care. 2008;31:391–396. 42. Volpato S, Leveille SG, Blaum C, Fried LP, Guralnik JM. Risk factors for falls in older disabled women with diabetes: the women’s health and aging study. J Gerontol A Biol Sci Med Sci. 2005;60: 1539 –1545. 43. Jones G, Nguyen T, Sambrook PN, Lord SR, Kelly PJ, Eisman JA. Osteoarthritis, bone density, postural stability, and osteoporotic fractures: a population based study. J Rheumatol. 1995;22:921– 925. 44. Hoops ML, Rosenblatt NJ, Hurt CP, Crenshaw J, Grabiner MD. Does lower extremity osteoarthritis exacerbate risk factors for falls in older adults? Womens Health (Lond Engl). 2012;8:685– 696.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:21 For personal use only. No other uses without permission. . All rights reserved.
