SEARCH

SEARCH BY CITATION

Keywords:

  • obesity;
  • fractures;
  • bone;
  • osteoporosis;
  • postmenopause;
  • aged

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Low body mass index (BMI) is a recognized risk factor for fragility fracture, whereas obesity is widely believed to be protective. As part of a clinical audit of guidance from the National Institute of Health and Clinical Excellence (NICE), we have documented the prevalence of obesity and morbid obesity in postmenopausal women younger than 75 years of age presenting to our Fracture Liaison Service (FLS). Between January 2006 and December 2007, 1005 postmenopausal women aged less than 75 years with a low-trauma fracture were seen in the FLS. Of these women, 805 (80%) underwent assessment of bone mineral density (BMD) by dual energy X-ray absorptiometry (DXA), and values for BMI were available in 799. The prevalence of obesity (BMI 30 to 34.9 kg/m2) and morbid obesity (BMI ≥ 35 kg/m2) in this cohort was 19.3% and 8.4%, respectively. Normal BMD was reported in 59.1% of obese and 73.1% of morbidly obese women, and only 11.7% and 4.5%, respectively, had osteoporosis (p < .0001). Multiple regression analysis revealed significant negative associations between hip T-score and age (p < .0001) and significant positive associations with BMI (p < .0001) and previous fracture (p = .001). Our results demonstrate a surprisingly high prevalence of obesity in postmenopausal women presenting to the FLS with low-trauma fracture. Most of these women had normal BMD, as measured by DXA. Our findings have important public heath implications in view of the rapidly rising increase in obesity in many populations and emphasize the need for further studies to establish the pathogenesis of fractures in obese individuals and to determine appropriate preventive strategies. © 2010 American Society for Bone and Mineral Research


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Low-trauma fractures affect one in three postmenopausal women and are associated with significant morbidity, mortality, and economic cost. Low bone mineral density (BMD) is a well-documented risk factor for fracture, and falls also play an important role, particularly in the pathogenesis of nonvertebral fractures.1, 2 In addition, some clinical risk factors contribute to fracture risk independently of BMD; these include age, previous fracture, glucocorticoid therapy, family history of hip fracture, tobacco use, alcohol abuse, and rheumatoid arthritis.3, 4 Low body mass index (BMI) is also included in these risk factors, although its contribution is independent of BMD only for hip fracture, probably through an association with frailty and increased falls risk.5

Recent studies have demonstrated that over 50% of postmenopausal women with incident fractures have a BMD higher than the diagnostic threshold of osteoporosis proposed in the World Health Organization (WHO) classification, namely, a T-score of −2.5.6–10 This observation may be explained in part by the contribution in many women of independent clinical risk factors to fracture risk. In addition, BMD is a continuous variable, and its inverse relationship with fracture risk is seen across the full spectrum of values. Most studies have emphasized the high prevalence of osteopenia (BMD T-score between −1 and −2.5) rather than osteoporosis in postmenopausal women with fracture. However, a substantial proportion of women with low-trauma fractures also have normal BMD, defined as a BMD T-score higher than −1. In such women, the pathogenesis of fractures and the case for bone protective therapy are uncertain.

Guidance from the National Institute of Health and Clinical Excellence (NICE) recommends measurement of BMD by dual energy X-ray absorptiometry (DXA) in postmenopausal women aged less than 75 years who present with a low-trauma fracture, only women with a BMD T-score −2.5 or less being eligible for bone protective therapy.11 As part of an audit of adherence to NICE guidance in postmenopausal women aged less than 75 years presenting to our Fracture Liaison Service (FLS) with a fragility fracture, we observed that nearly 40% had a normal BMD (T-score higher than −1). Further analysis of this population revealed that the majority of such women are obese or morbidly obese, raising the possibility that obesity is a risk factor for fracture in postmenopausal women. The aim of this study was to document the prevalence of obesity and morbid obesity in postmenopausal women with low-trauma fractures presenting to our FLS and to examine the relationship between BMI and BMD in this cohort.

Subjects and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

The FLS at Addenbrooke's Hospital in Cambridge was set up in 2005. It is situated in the Fracture Clinic, where a specialist nurse (LP) sees all postmenopausal women attending the clinic with a recent low-trauma fracture. Patients are counseled about risk factors, and in keeping with current NICE guidance, those aged less than 75 years are referred for bone densitometry. This study was an anonymous audit of a clinical service, and informed consent therefore was not required. The audit was conducted to monitor adherence to NICE guidance.

Information recorded included fracture site, previous history of low-trauma fracture after age 45, tobacco and alcohol use, medications, age at menopause, and medical history. BMD in the lumbar spine (L1–4) and proximal hip was measured within 4 weeks of the fracture by DXA using a Hologic bone densitometer (Hologic QDR 4500A, Hologic, Inc., Bedford, MA, USA). The short term in vivo precision of measurement at these two sites is 1% and 1% to 2%, respectively. Total hip BMD T-scores were calculated using data from the National Health and Nutrition Examination Survey (NHANES) reference female population.

According to the WHO classification, osteoporosis was defined as a BMD T-score in the lumbar spine and/or total hip of −2.5 or less, osteopenia as a T-score between −1 and −2.5, and normal BMD as a T-score of −1 or more.12 If the T-score was −2.5 or less at one site, a diagnosis of osteoporosis was made regardless of the T-score at the other site. Osteopenia was diagnosed when at least one T-score was between −1 and −2.5 and neither T-score was −2.5 or less, and normal BMD was diagnosed only in cases where the T-score was higher than −1 at both sites. Because of the inclusion of hip but not spine BMD in the FRAX algorithm, we also examined the effect of using only the total hip BMD in the diagnosis of osteoporosis, osteopenia, and normal BMD.

Height and weight were measured at the time of bone density measurement, and the BMI was calculated as weight (kg)/[height (m)]2. Fractures were considered to be low-trauma fractures if they occurred spontaneously or from standing height or less. All incident fractures were verified radiologically.

Statistical analysis

Descriptive statistics were calculated to describe the overall patient population in the study. In addition, prevalences of osteoporosis and obesity were computed. Differences in age, fracture site, BMI, tobacco use, alcohol intake, and medications between women with or without DXA scans were tested by one-way ANOVA and chi-square. BMI was divided into quartiles (<25, 25–29, 30–34, ≥35 kg/m2). Fisher's exact test was used to evaluate differences in the prevalence of fracture sites in obese and nonobese women. Multiple regression analysis was used to determine associations between total hip and lumbar spine BMD T-score and age, previous fracture, BMI, smoking, and intake of more than 14 alcohol units per week. Differences were considered significant when the two-tailed p value was less than 0.05. The statistical analysis was performed using the SPSS statistics package (SPSS, Inc., Chicago, IL, USA) for Windows Version 16.0.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Between January 2006 and December 2007, 1641 postmenopausal women with a low-trauma fracture were seen in the FLS. A total of 1005 were aged 75 years or less, and bone densitometry was performed in 805 (80%) of these women. In the remaining 20%, the patient either declined to have bone densitometry or was unable to undergo the examination for physical or logistic reasons. Of the 805 women undergoing bone densitometry, BMI was available in 799. In these women, both lumbar spine and hip BMD were available in 785, only hip BMD was available in 789, and only lumbar spine BMD was available in 795.

Women who did not undergo BMD measurement did not differ significantly from those who did in terms of age, fracture site, BMI, tobacco use, or alcohol intake but were more likely to be taking glucocorticoids (3.6% versus 0.6%, p < .003) and bone-active medications (29.7% versus 7.2%, p < .0001). Demographic details of all 1005 women, including the fracture site, are shown in Table 1.

Table 1. Characteristics of Postmenopausal Women Aged 75 Years or Less Presenting to the Fracture Liaison Service Between January 2007 and December 2008
 Postmenopausal women
Clinical data 
 Age (years)62.4 ± 6.9
 BMI (kg/m2)27.4 ± 5.7
 Previous history of fracture257/1005 (25.6%)
 Current smoker63/1005 (6.3%)
 Intake of more than 14 units of alcohol per week21/1005 (2.1%)
 Current glucocorticoid use13/1005 (1.3%)
DXA data 
 Spine T-score−1.22 ± 1.35 (801)
 Total hip T-score−0.82 ± 1.04 (795)
Fracture sites 
 Wrist360 (35.9%)
 Lower leg204 (20.3%)
 Hand or foot176 (17.5%)
 Humerus119 (11.9%)
 Hip55 (5.5%)
 Other88 (8.9%)
 Elbow72 (7.2%)
 Clavicle9 (0.9%)
 Pelvis3 (0.3%)
 Ribs2 (0.2%)
 Spine2 (0.2%)
 Scapula1 (0.1%)
 Missing1 (0.1%)

Overall, when both spine and hip BMD were considered, osteoporosis was present in 19.3%, osteopenia in 41.2%, and normal BMD in 39.4% of the women. Of the 799 women in whom BMI and BMD were available, 19.3% [95% confidence interval (CI) 16.6–22.0] were obese (BMI 30 to 34.9 kg/m2), and 8.4% (95% CI 6.5–10.3) were morbidly obese (BMI ≥ 35 kg/m2; total 27.7%); the percentages of women ≤25 and 25 to 30 kg/m2 were 36.9% and 35.4%, respectively.

Table 2 and Figure 1 show the diagnosis based on lumbar spine and/or total hip BMD (normal, osteopenia, or osteoporosis) according to quartiles of BMI. Normal BMD was reported in 59.1% of obese and 73.1% of morbidly obese women, and only 11.7% and 4.5%, respectively, had osteoporosis (p < .0001). In contrast, most of the women with a BMI ≤ 30 kg/m2 had osteopenia or osteoporosis. Women with normal BMD were younger than women with osteopenia or osteoporosis (61.1 ± 6.9 versus 62.7 ± 6.7 and 63.8 ± 6.6 years, respectively, p < .0001).

Table 2. Age and BMI Quartile in Women With Normal BMD, Osteopenia, or Osteoporosis Based on Spine and/or Total Hip BMD (n = 799)
 NormalOsteopeniaOsteoporosis
Age (years)61.1 ± 6.962.7 ± 6.763.8 ± 6.6
BMI (kg/m2)   
 <2560 (20.3%)147 (49.8%)88 (29.8%)
 25–29114 (40.3%)121 (42.8%)48 (17.0%)
 30–3491 (59.1%)45 (29.2%)18 (11.7%)
 ≥3549 (73.1%)15 (22.4%)3 (4.5%)
thumbnail image

Figure 1. Percentage of women with normal BMD, osteopenia, or osteoporosis at the spine and/or hip (A) or at the hip only (B) according to BMI quartile.

Download figure to PowerPoint

When diagnosis was based only on total hip BMD, osteoporosis was present in 5.6% of women, osteopenia in 37.5%, and normal BMD in 56.9%. Table 3 shows the distribution of BMD diagnostic categories according to quartile of BMI. Thus 80.4% of obese and 89.4% of morbidly obese women had a T-score higher than −1, and only 1.3% and 1.5%, respectively, had osteoporosis. Women with normal BMD were younger than women with osteopenia or osteoporosis (p < .0001).

Table 3. Age and BMI Quartile in Women With Normal BMD, Osteopenia, or Osteoporosis Based on Total Hip BMD (n = 789)
 NormalOsteopeniaOsteoporosis
Age (years)61.0 ± 6.763.6 ± 6.665.3 ± 6.1
BMI (kg/m2)   
 <2599 (34.1%)164 (56.6%)27 (9.3%)
 25–29168 (60.0%)98 (35.0%)14 (5.0%)
 30-< 35123 (80.4%)28 (18.3%)2 (1.3%)
 ≥3559 (89.4%)6 (9.1%)1 (1.5%)

Lumbar spine and total hip T-scores by quartile of BMI are shown in Table 4. At both sites, BMD increased across quartiles of BMI. Table 5 compares the site of fracture in obese and nonobese women. Fractures of the wrist were significantly less common and hip fractures significantly more common in obese than in nonobese women. The prevalence of the other fracture sites was similar in the two groups.

Table 4. Age and BMD T-Scores According to Quartile of BMI
 BMI < 25 kg/m2BMI 25–29 kg/m2BMI 30–34 kg/m2BMI ≥ 35 kg/m2
Age (years)62.1 ± 6.662.9 ± 7.061.8 ± 6.861.6 ± 6.9
Spine T-score (n = 795)−1.72 ± 1.24−1.15 ± 1.30−0.75 ± 1.30−0.50 ± 1.34
Total hip T-score (n = 789)−1.34 ± 0.92−0.82 ± 0.94−0.28 ± 0.92+0.20 ± 1.01
Table 5. Site of Fracture According to BMI
 BMI < 30 kg/m2 (n = 578)BMI ≥ 30 kg/m2 (n = 221)p Value
Wrist224 (38.8%)67 (30.3%)0.027
Hip59 (10.2%)37 (16.7%)0.015
Lower leg33 (5.7%)6 (2.7%)0.098
Hand or foot108 (18.7%)48 (21.7%)0.369
Humerus99 (17.1%)46 (20.8%)0.259
Other55 (9.5%)17 (7.7%)0.578

Multiple regression analysis revealed significant negative associations between hip T-score and age (p < .0001) and significant positive associations with BMI (p < .0001) and previous fracture (p = .001) (Table 6). In the spine, there was a significant negative association between the T-score and age (p = .019) and a significant positive association between T-score and BMI (p < .0001).

Table 6. Final Multiple Linear Regression Model After Performing Backwards Model Selection to Investigate Factors Associated With Total Hip T-Score
 CoefficientStandard Errorp Value
  1. Variables that were included in the model: age, previous history of fracture, BMI, history of smoking, and history of use more than 14 alcohol units per week.

  2. Variables that remained in the model: age, previous history of fracture, and BMI.

  3. Dependent variable: total hip T-score. Adjusted R2 = 0.309 and residual standard deviation = 0.870.

Constant−1.8010.369<.0001
Age (years)−0.0320.005<.0001
Previous fracture0.2380.073.001
BMI (kg/m2)0.0930.006<.0001

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Consistent with previous reports, we found that the majority of postmenopausal women presenting with a low-trauma fracture did not have osteoporosis as defined by the WHO classification. However, a new finding from this audit was the high prevalence of obesity in these women, the majority of whom had a normal BMD as defined as a T-score higher than −1. Thus 28% of women presenting with a fracture were obese or morbidly obese, of whom 59% and 73%, respectively, had a normal BMD based on both spine and total hip measurements and 80% and 89%, respectively, based on total hip alone. After adjustment for a number of confounding variables in our cohort, BMI was shown to be an independent determinant of BMD in both the lumbar spine and the total hip. A significant negative association with age and, in the hip but not the spine, a significant positive association with previous history of fracture also were demonstrated in multiple regression analysis.

The prevalence of obesity in our cohort appears to be at least as great and possibly greater than expected. In the European Prospective Investigation into Cancer and Nutrition (EPIC) study, a prospective population-based cohort study of middle-aged men and women in Norfolk, United Kingdom, conducted between 1993 and 1997, the prevalence of obesity and morbid obesity in women was 14% and 5%, respectively.13 However, obesity rates have increased over the past 10 years, and overrepresentation of healthy women in the cohort may have reduced the apparent prevalence of obesity. Figures from the Department of Health suggest that the prevalence of obesity in women aged 55 to 74 years was 29% in 200314; however, there are large regional variations throughout the United Kingdom, and we do not have accurate figures for current rates of obesity in the area served by our hospital trust.

Changes in fat mass and its distribution affect both the accuracy and precision of BMD measurement by DXA, although the direction and magnitude of the effect are uncertain.15 The existing literature on this is sparse and limited to experiments in which blocks of lard have been used to simulate body fat. Data obtained using this approach indicate that the effect varies according to the type of DXA system (pin beam versus fan beam), the distribution of body fat, the manufacturer of the software versions, and scan mode. In a recent study by Evans and colleagues using the Hologic QDR 4500A (the same system as used in our study), it was found that adding lard resulted in a small decrease (1.6%) in spine BMD and a smaller increase (0.7%) in hip BMD.16 Thus, if anything, spine BMD would have been underestimated in our study and hip BMD overestimated, but the latter not sufficiently to raise the BMD value from osteopenic or osteoporotic ranges to normal. However, the use of lard to simulate body fat has obvious limitations, and further work is required to establish the effects of obesity on DXA-based measurements of BMD.

The inverse relationship between BMD and fracture risk in postmenopausal women is well documented and is seen across the full spectrum of BMD values in the population. The WHO classification of osteoporosis as a BMD T-score −2.5 or less was intended as an operational definition that identified women at highest risk of fracture, but few postmenopausal women have a BMD in this range, and numerically, most fractures occur in women with a BMD T-score higher than −2.5. Using bone density measurements in the heel, finger, or forearm in a large prospective observational study of postmenopausal women in primary care practices in the United States, Siris and colleagues6 reported that only 7.2% of women who developed a fracture within 1 year of BMD measurement had a T-score below −2.5, whereas 39.6% had a T-score between −1 and −2.5 and 53.2% had a T-score higher than −1. Subsequently, similar findings have been reported using DXA-based measurements of BMD in prospective population-based studies. Schuit and colleagues7 reported normal baseline femoral neck BMD in 12.6% of postmenopausal women with incident nonvertebral fractures during a mean follow-up period of 6.8 years, osteopenia in 43.3%, and osteoporosis in 44.1%. In the Study of Osteoporotic Fractures (SOF), baseline BMD measurement at the total hip showed osteoporosis in 46% of women who developed a hip fracture during 5 years of follow-up, whereas 48% were osteopenic and 6% had normal BMD.8 In a study from Australia, Pascoe and colleagues10 reported that only 26.9% of fractures over a median follow-up period of 5.6 years occurred in women with osteoporosis as defined by a baseline BMD T-score −2.5 or greater, whereas 56.5% occurred in women with osteopenia and 16.6% in women with a normal BMD.

As noted in previous studies, there was considerable discordance in the diagnosis of osteoporosis according to the site of measurement.17 As would be expected, the prevalence of osteoporosis was lower when only the total hip BMD was considered (as opposed to spine and/or total hip BMD), whereas the percentage of women with a normal BMD was higher. In addition, an even higher proportion of obese and morbidly obese women had a normal BMD when only the total hip was considered. In general, measurements of BMD at the hip are preferable when predicting fracture risk,18 but in some conditions the spine is preferentially affected by osteoporosis, and in such cases, spine BMD may be more appropriate than hip BMD as a basis for treatment decisions. In addition, hip measurements cannot be made in patients who have undergone bilateral hip replacement. However, with increasing age, spine BMD measurements become less accurate because of the presence of osteophytes, extravascular calcification, and scoliosis.19

The high prevalence of obesity in postmenopausal women attending our FLS was unexpected in view of the known association between low BMI and fracture risk. The risk gradient is steepest below BMI values of around 22 kg/m2 and is attributable mainly to low BMD. Conversely, obesity is associated with high BMD and generally is thought to be protective against fracture. However, the risk gradient between BMI and fracture risk is not constant and over BMI values of 25 kg/m2 is relatively small, arguing against a protective effect of obesity per se.5 Although BMD is higher in obese subjects, this can be seen as an adaptive response to the higher loading of the skeleton and therefore may not confer greater protection against fracture in those individuals.20–23 Furthermore, although greater soft tissue padding may reduce skeletal trauma following a fall, poorer protective responses to falling and the higher impact of the fall owing to high body weight may offset this potential benefit. In addition, the risk of falling may be increased in obese subjects as a result of reduced muscular strength and mobility.24

In clinical trials, the antifracture efficacy of bone protective treatments has mostly been established in postmenopausal women with low BMD, with or without prevalent fractures.25 The results of these studies provide a strong rationale for treating postmenopausal women who present with a fracture and have a low BMD; evidence from a number of sources supports the efficacy of these interventions in women with osteopenia as well as those with osteoporosis.26–31 In most clinical trials of osteoporosis treatment, the proportion of obese women has been small, and those that have been included generally have had a low BMD. However, in a recent randomized, controlled trial in which women were not selected on the basis of low BMD, a significant interaction between BMI and antifracture efficacy was shown in women treated with clodronate, fracture reduction being less in women with a BMI of 30 kg/m2; a similar although not statistically significant interaction remained when femoral neck BMD was included in calculation of fracture risk. The case for bone protective therapy in obese women with normal BMD as defined by the WHO thus requires further investigation.

Our study has several limitations. Vertebral fractures were underrepresented in this audit because the majority do not come to medical attention32 and those that do may be referred to a specialist metabolic bone clinic rather than a fracture clinic. Hip fractures also were underrepresented because they are not routinely followed up in the fracture clinic and most occur in women over the age of 75 years.33 We were only able to assess women who attended the fracture clinic and do not know the characteristics of those who failed to keep their appointment. Finally, since BMD measurement was only performed routinely in postmenopausal women aged 75 years or less, our results may not be applicable to older women with incident low-trauma fractures. Nevertheless, our results are representative of a large proportion of the fracture burden presenting to a fracture clinic with fractures other than those of the spine and hip, and women in this age group comprised 61% of all postmenopausal women seen in the FLS during the study period.

Our findings have major public health implications in view of the rapidly rising prevalence of obesity in many populations.14, 34, 35 The high proportion of obese and morbidly obese in our fracture clinic population indicates that obesity may not be protective against fracture and might even be a risk factor for fracture. Obesity also may be a contributory factor in the increased fracture risk reported in some studies of patients with type 2 diabetes mellitus.36–40 The morbidity and economic costs associated with fractures are likely to be higher in obese individuals because of a greater risk of nonunion41, 42 and, in those requiring surgery, a greater frequency of postoperative complications. National guidelines for osteoporosis increasingly use intervention thresholds based on 10-year fracture probability levels estimated by FRAX; this approach results in the recommendation for treatment in some women with osteopenia as well as those with osteoporosis.43–46 However, even in the presence of a previous fracture, 10-year fracture probabilities generated in obese women with normal BMD generally fall well below the intervention thresholds, and treatment therefore will not be recommended. Further studies are required to establish the prevalence and pathogenesis of fractures in obese individuals and to determine appropriate management strategies.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

This study was supported by NHS National Institute of Health Research. MOP is supported by the CAPES Foundation, Ministry of Education, Brazil.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  • 1
    Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ. 1996; 312: 12541259.
  • 2
    Kaptoge S, Benevolenskaya LI, Bhalla AK, et al. Low BMD is less predictive than reported falls for future limb fractures in women across Europe: results from the European Prospective Osteoporosis Study. Bone. 2005; 36: 387298.
  • 3
    Kanis JA. Diagnosis of osteoporosis and assessment of fracture risk. Lancet. 2002; 359: 19291936 .
  • 4
    Kanis JA, McCloskey EV, Johansson H, Strom O, Borgstrom F, Oden A. National Osteoporosis Guideline Group: case finding for the management of osteoporosis with FRAX-assessment and intervention thresholds for the UK. Osteoporos Int. 2008; 19: 13951408.
  • 5
    DeLaet C, Kanis JA, Oden A, et al. Body mass index as a predictor of fracture risk: a meta-analysis. Osteoporos Int. 2005; 16: 13301338.
  • 6
    Siris ES, Miller PD, Barrett-Connor E, et al. Identification and fracture outcomes of undiagnosed low bone mineral density in postmenopausal women: results from the National Osteoporosis Risk Assessment. JAMA. 2001; 286: 28152822.
  • 7
    Schuit SCE, van der Klift M, Weel AEAM, et al. Fracture incidence and association with bone mineral density in elderly men and women: the Rotterdam Study. Bone. 2004; 34: 195202.
  • 8
    Wainwright SA, Marshall LM, Ensrud KE, Cauley JA, Black DM. Hip fracture in women without osteoporosis. J Clin Endocrinol Metab 2005; 90: 27872893.
  • 9
    Sornay-Rendu E, Munoz F, Garnero P, DuBoeuf F, Delmas PD. Identification of osteopenic women at high risk of fracture: the OFELY Study. J Bone Miner Res 2005; 20: 18131819.
  • 10
    Pascoe JA, Seeman E, Henry MJ, Merriman EN, Nicholson GC, Kotowicz MA. The population burden of fractures originates in women with osteopenia, not osteoporosis. Osteoporos Int, 2006; 17: 14041409.
  • 11
    National Institute for Health and Clinical Excellence. Final Appraisal Determination: Alendronate, Etidronate, Risedronate, Raloxifene, Strontium Ranelate and Teriparatide for the Secondary Prevention of Osteoporotic Fragility Fractures in Postmenopausal Women. London, NICE. 2008.
  • 12
    World Health Organization Study Group. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Technical Report Series 1994; 843.
  • 13
    Canoy D, Boekholdt SM, Wareham N, et al. Body fat distribution and risk of coronary heart disease in men and women in the European Investigation into Cancer and Nutrition in Norfolk cohort: a population-based prospective study. Circulation. 2007; 116: 29332943.
  • 14
    Department of Health. Forecasting Obesity to 2010. London: Joint Health Surveys Unit, National Centre for Social Research and Department of Epidemiology and Public Health at the Royal Free and University College Medical School; 2006.
  • 15
    Tothill P, Laskey MA, Orphanidou CI, van Wijk M. Anomalies in dual energy X-ray absorptiometry measurements of total body bone mineral during weight change using Lunar, Hologic, and Norland instruments. Br J Radiol. 1999; 72: 661669.
  • 16
    Evans EM, Mojtahedi MC, Kessinger RB, Misic MM. Simulated change in body fatness affects Hologic QDR 4500A whole body and central DXA bone measures. J Clin Densitometry. 2006; 9: 315322.
  • 17
    O'Gradaigh D, Debiram I, Love S, Richards HK, Compston JE. A prospective study of discordance in diagnosis of osteoporosis using spine and proximal femur bone densitometry. Osteoporos Int. 2003; 14: 1318.
  • 18
    Kanis JA, Johnell O, Oden A, et al. The use of multiple sites for the diagnosis of osteoporosis. Osteoporos Int. 2006; 17: 527534.
  • 19
    Reid IR, Evans MC, Ames R, Wattie DJ. The influence of osteophytes and aortic calcification on spinal mineral density in postmenopausal women. J Clin Endocrinol Metab. 1991; 72: 13721374.
  • 20
    Seeman E, Melton LJ III, O'Fallon WM, Riggs BL. Risk factors for spinal osteoporosis in men. Am J Med. 1983; 75: 977983.
  • 21
    Albala C, Yáñez M, Devoto E, Sostin C, Zeballos L, Santos JL. Obesity as a protective factor for postmenopausal osteoporosis. Int J Obs Relat Metab Disord. 1996; 20: 10271032.
  • 22
    Pesonen J, Sirola J, Tuppurainen M, et al. High bone mineral density among perimenopausal women. Osteoporos Int. 2005; 16: 18991906 .
  • 23
    Henry MJ, Pasco JA, Sanders KM, Nicholson GC, Kotowicz MA. Fracture risk (FRISK) score: Geelong Osteoporosis Study. Radiology. 2006; 241: 190196.
  • 24
    Arden NK, Crozier S, Smith H, et al. Knee pain, knee osteoarthritis and the risk of fracture. Arthritis Rheum. 2006; 55: 610615.
  • 25
    Poole KE, Compston JE. Osteoporosis and its management. BMJ. 2006; 333: 12511256.
  • 26
    Trivedi DP, Doll R, Khaw KT. Effect of four monthly oral vitamin D3 (cholecalciferol) supplementation on fractures and mortality in men and women living in the community: randomized double blind controlled trial. BMJ. 2003; 326: 469.
  • 27
    Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative Randomised Controlled Trial. JAMA. 2002; 288: 321333.
  • 28
    Kanis JA, Johnell O, Black DM, et al. Effect of raloxifene on the risk of new vertebral fracture in postmenopausal women with osteopenia or osteoporosis: a reanalysis of the Multiple Outcomes of Raloxifene Evaluation Trial. Bone. 2003; 33: 293300.
  • 29
    Marcus R, Wang O, Satterwhite J, Mitlak B. The skeletal response to teriparatide is largely independent of age, initial bone mineral density and prevalent vertebral fractures in postmenopausal women with osteoporosis. J Bone Miner Res. 2003; 18: 1823.
  • 30
    Watts NB, Josse RG, Hamdy RC, et al. Risedronate prevents new vertebral fractures in postmenopausal women at high risk. J Clin Endocrinol Metab. 2003; 88: 542549.
  • 31
    McCloskey EV, Johannsson H, Oden A, et al. Ten-year fracture probability identifies women who will benefit from clodronate therapy: additional results from a double-blind, placebo-controlled randomised study. Osteoporos Int. 2009; 20: 811817.
  • 32
    Cooper C, Atkinson EJ, O'Fallon WM, Melton LJ. The incidence of clinically diagnosed vertebral fractures: a population based study in Rochester, Minnesota, 1985–1989. J Bone Miner Res 1992; 7: 221227.
  • 33
    van Staa TP, Dennison EM, Leufkens HG, Cooper C. Epidemiology of fractures in England and Wales. Bone. 2001; 29: 517522.
  • 34
    World Health Organization. Obesity: preventing and managing the global epidemic. Report of a WHO Consultation. WHO Tech Rep Ser. 2000; 894.
  • 35
    Zaninotto P, Head J, Stamatakis E, Wardle H, Mindell J. Trends in obesity among adults in England between 1993 and 2004 by age and social class and projections of prevalence to 2012. J Epidemiol Commun Health. 2009; 63: 140146.
  • 36
    Janghorbani M, Feskanich D, Willett WC, Hu F. Prospective study of diabetes mellitus and risk of hip fracture: the Nurses' Health Study. Diabetes Care. 2006; 29: 15731578.
  • 37
    Kelsey JL, Browner WS, Seeley DG, Nevitt MC, Cummings SR. Risk factors for fractures of the distal forearm and proximal humerus: the Study of Osteoporotic Fractures Research Group. Am J Epidemiol. 1992; 135: 477489.
  • 38
    Melton LJ III, Achenbach SJ, O'Fallon WM, Khosla S. Secondary osteoporosis and the risk of distal forearm fractures in men and women. Bone. 2002; 31: 119125.
  • 39
    Vogt MT, Cauley JA, Tomaino MM, Stone K, Williams JR, Herndon JH. Distal radius fractures in older women: a 10-year follow-up study of descriptive characteristics and risk factors. The Study of Osteoporotic Fractures. J Am Geriatr Soc. 2002; 50: 97103.
  • 40
    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: 495505.
  • 41
    Green E, Lubahn JD, Evans J. Risk factors, treatment, and outcomes associated with nonunion of the midshaft humerus fracture. J Surg Orthop Adv. 2005; 14: 6472.
  • 42
    King AR, Moran SL, Steinmann SP. Humeral nonunion. Hand Clin. 2007; 23: 449456.
  • 43
    Kanis JA on behalf of the World Health Organization Scientific Group. Assessment of osteoporosis at the primary health-care level. Technical report. WHO Collaborating Centre, University of Sheffield, UK; 2008.
  • 44
    Kanis JA, Burlet N, Cooper C, et al. European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int. 2008; 19: 399348.
  • 45
    National Osteoporosis Foundation. Clinician's guide to prevention and treatment of osteoporosis. 2008.
  • 46
    Compston J, Cooper A, Cooper C, et al. on behalf of the National Osteoporosis Guideline Group (NOGG). Guidelines for the diagnosis and management of osteoporosis in postmenopausal women and men from the age of 50 years in the UK. Maturitas. 2009; 62: 105108.