REVIEW URRENT C OPINION
Osteoporosis epidemiology 2013: implications for diagnosis, risk assessment, and treatment William D. Leslie a and Suzanne N. Morin b
Purpose of review To summarize the recently published studies that provide insights into the changing epidemiology of osteoporosis and fractures. Recent findings The main themes reviewed are fracture outcomes; trends in fractures rates; fracture risk assessment and monitoring; atypical femoral fractures; male osteoporosis; falls and physical activity; and sarcopenia, obesity, and metabolic syndrome. Summary Osteoporotic fractures were found to have long-term consequences on excess mortality (10 years) and economic costs (5 years). The large burden of nonhip nonvertebral fractures has been underestimated. Divergent (but mostly declining) trends in fracture rates were confirmed in several cohorts from around the world. This has significant implications for healthcare planners and clinicians responsible for the care of individuals with osteoporosis, and also impacts on the calibration of fracture prediction tools. Although fracture prediction tools differ in their complexity, performance characteristics are similar when applied to the general population. Large, high-quality comparative studies with different case mixes are needed. Fracture probability does not appear to be responsive enough to support goal-directed treatment at this time. A consensus on the diagnosis of osteoporosis in men has emerged, based upon the same absolute bone density cutoff for both men and women. Finally, a plethora of new data highlight the importance of falls, physical activity, and body composition as contributors to skeletal health. Keywords bone mineral density, clinical practice guidelines, epidemiology, fracture, osteoporosis
INTRODUCTION Osteoporosis is characterized by low bone mass and skeletal fragility resulting in increased susceptibility to low-trauma fractures, especially those affecting the vertebrae, proximal femur (hip), distal forearm, and proximal humerus which are collectively referred to as major osteoporotic fractures [1]. This update summarizes the recently published studies that provide insights into the changing epidemiology of osteoporosis and fractures.
FRACTURE OUTCOMES The consequences of osteoporotic fractures include re-fractures, excess mortality and morbidity, and economic costs for the healthcare system. Using a large U.S. administrative database, Tajeu et al. [2] documented that the year following hip fracture was associated with a two-fold increase in death, a twofold increase in destitution (entering into lowincome status), and a four-fold increase in debility (requiring long-term care). www.co-rheumatology.com
Three notable publications in 2013 were from the Dubbo Osteoporosis Epidemiology Study, a prospective cohort of community-dwelling participants aged 60 years and older who have been followed since 1989 [3,4 ,5 ]. Although adverse outcomes are most frequently reported in relation to hip and vertebral fractures, these and other recent data highlight the importance of nonhip nonvertebral (NHNV) fractures as they are collectively much more numerous and make a larger contribution to the burden of osteoporosis at the population level [6]. Of the 952 documented fractures in women and 343 &&
&
a
University of Manitoba, Winnipeg and bMcGill University, Montreal, Manitoba, Canada Correspondence to Dr William D. Leslie, MD, MSc, Department of Medicine (C5121), St Boniface General Hospital, 409 Tache Avenue, Winnipeg, MB R2H 2A6, Canada. Tel: +1 204 237 2311; fax: +1 204 237 2007; e-mail: [email protected] Curr Opin Rheumatol 2014, 26:440–446 DOI:10.1097/BOR.0000000000000064 Volume 26 Number 4 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Osteoporosis epidemiology 2013 Leslie and Morin
an initial hip fracture, age-specific and sex-specific survival relative to the general population was greatly reduced in women (83% at 1 years, 59% at 5 years, and 31% 10 years) and men (63% at 1 years, 48% at 5 years, and 36% 10 years) [5 ]. For every six women and for every three men with hip fracture, one extra death occurred above that expected in the general population. Excess mortality impacts on the long-term economic burden of fractures. A population-based Canadian study [7] (16 198 incident fractures cases and 48 594 matched nonfracture controls) reported high direct healthcare costs for the first year after fracture, greatest for hip fractures (median $25 306 in women and $21 396 in men). Incremental costs declined considerably after the first year, in part because of higher mortality among those with fractures who already were the heaviest users of healthcare (healthy survivor bias). From the healthcare payer perspective, total costs for all fractures combined eventually fell below the prefracture levels because of the excess mortality among fracture cases. However, among those surviving 5 years following fracture, healthcare costs remained above prefracture levels.
KEY POINTS Divergent (but mostly declining) trends in fracture rates were confirmed in several cohorts from around the world.
&
Although fracture prediction tools differ in their complexity, performance characteristics are similar when applied to the general population. Fracture probability does not appear to be responsive enough to support goal-directed treatment at this time. A consensus on the diagnosis of osteoporosis in men has emerged, based upon the same absolute bone density cutoff for both men and women.
in men, over half were NHNV fractures (486 in women and 173 in men) and these were associated with a two-fold relative risk (RR) of subsequent fracture in women with even greater risk in men [3]. More importantly, NHNV fractures were associated with 20% excess mortality in the first 5 years after fracture. Re-fractures contributed substantially to this excess mortality as demonstrated in a competing risk analysis [4 ]. Total mortality (following initial and subsequent fractures) was elevated for 10 years, and most of the later excess mortality (5–10 years) was associated with re-fractures (Fig. 1). Among 151 women and 55 men sustaining &&
TRENDS IN FRACTURE RATES Large international variations in osteoporotic fracture rates had been reported, with temporal trends
Women
(a)
Men
(b)
100
100
80
80
60
60
40
40
Cumulative incidences
Total deaths
20
Deaths following initial fracture Age-matched population deaths
20 Expected mortality in the general population
Expected mortality in the general population
0
0 0
Fracture population 952
5
10
15
20
0
5
10
15
20
487
230
90
20
343
151
68
33
7
Time, years
FIGURE 1. Stacked graph of cumulative incidences of mortality following initial osteoporotic fracture in black and following re-fracture in gray compared with an age-matched general population: (a) women and (b) men. The dark gray area represents the excess mortality (i.e. above population mortality) related to the initial fracture. The light gray area represents the excess mortality related to the re-fracture. The height of the jagged, black, and gray line represents the cumulative incidence of mortality in the general population, mortality following initial fracture, and total mortality at a given time. Reproduced with permission [4 ]. &&
1040-8711 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-rheumatology.com
441
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Metabolic bone disease
that differ between populations [8]. Data from the Rochester Epidemiology Project (Olmsted County, Minnesota, USA) were used to compare the fracture rates in 2009–2011 with those from 1989 to 1991. In women, hip fractures showed a 25% decline with a 26% reduction in distal forearm fractures. However, the overall age-adjusted fracture rate increased by 11%, largely attributed to incidentally diagnosed vertebral fractures. The basis for declining hip and nonhip fracture rates in industrialized countries remains a subject of uncertainty. Data from the National Health and Nutrition Examination Survey (NHANES) noted a 6% increase in femoral neck bone mineral density (BMD) among non-Hispanic white women between 1988–1994 and 2005–2008 adjusted for multiple determinants [9]. Similar data were reported from a large Canadian BMD registry that included baseline dual-energy X-ray absorptiometry (DXA) for 36 587 women aged 50 years and older [10 ]. Even after adjustment for increasing osteoporosis treatment and obesity, there was a significant linear increase in BMD at both the femoral neck (0.52% per year) and lumbar spine (0.32% per year). Moreover, when incident major osteoporotic fracture rates and hip fracture rates were studied in the same population, there was a significant linear decline from 1996 to 2006, but adjusting for BMD accounted for the observed reduction in fractures. These data would suggest that improvement in BMD, still unexplained and not attributable to greater rates of osteoporosis treatment or obesity, is a major factor contributing to reductions in osteoporotic fracture rates. It is interesting to note that a secular increase in BMD was also reported among Southern Chinese women between 1995–2000 and 2005–2010 after adjustment for other factors (lumbar spine 4.17%, femoral neck 9.02%, and total hip 9.34%) among women aged 50 years and older. If confirmed in other Asian populations, then this may herald the reversal in increasing fracture rates that have been noted in this part of the world. &&
FRACTURE RISK ASSESSMENT AND MONITORING The need for screening and fracture risk assessment tools that go beyond BMD is now widely accepted. Questions arise regarding the number of risk factors required, with some systems including a small number of risk factors (e.g. Garvan FRC, www.garvan. org.au/bone-fracture-risk/), whereas others contain a large number of risk factors (e.g. QFracture, www.qfracture.org/). There is a tradeoff between complexity (greater accuracy) and simplicity (adoption in clinical practice). Rubin et al. [11] compared 442
www.co-rheumatology.com
the WHO Fracture Risk Assessment (FRAX) tool without BMD (www.shef.ac.uk/FRAX/) and several simpler screening tools (OST, ORAI, OSIRIS, SCORE, and age alone) for predicting fractures in 3614 women aged 40–90 years followed through the Danish National Register over 3 years. All tools showed similar fracture discrimination (area under curve between 0.703 and 0.722). This may indicate that for populations with a low prevalence of risk factors, a simpler screening tool may be sufficient. The same group performed a systematic review of screening and risk assessment tools [12 ]. Of a total of 48 tools, only 6 had been tested more than once in a population-based setting with acceptable methodology. Once again, there was no consistent evidence that more complex tools (SCORE, FRAX, and QFracture) had better performance characteristics than simpler tools (OST, ORAI, and Garvan FRC). However, the paucity of head-to-head comparisons limits any definitive conclusions. Larger, high-quality studies with different case mixes are needed to address this important question and to determine these tools’ effectiveness in selecting patients for therapy. Changing fracture epidemiology has implications for fracture risk assessment using tools calibrated to estimate absolute fracture risk. For example, the WHO FRAX system estimates 10-year major osteoporotic fracture and 10-year hip fracture probability. Changes in a country’s fracture rates would necessitate FRAX recalibration. For many countries without high-quality nonhip fracture data, major osteoporotic fracture rates are estimated based upon an assumption that these show a similar nonhip-to-hip fracture ratio as historical Swedish data (1987–1996). Population-based data from Canada suggest that these ratios may underestimate the rate of major osteoporotic fractures, possibly because of declining hip fracture rates [13]. Although Swedish vertebral and hip fracture ratios were similar to Canada, ratios were significantly lower for other sites (men and women: 46 and 35% lower for forearm and hip ratios, 19 and 15% lower for humerus and hip ratios, and 19 and 23% lower for any major osteoporotic fractures and hip ratio, respectively). The need to develop more responsive methods for monitoring drug treatment in osteoporosis was highlighted in an initiative to develop goal-directed treatment: ‘treat-to-target’ guidelines for osteoporosis [14 ]. Despite the intuitive appeal of this approach, it is very challenging to identify suitably responsive indices and treatment targets. Fracture probability might appear to be a logical candidate. However, FRAX was not found to be sufficiently responsive to antiosteoporosis treatment for this &
&&
Volume 26 Number 4 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Osteoporosis epidemiology 2013 Leslie and Morin
purpose in an analysis of 11 049 previously untreated women aged 50 years and older from a large BMD registry with paired FRAX estimates (median interval 4 years) [15 ]. A total of 6534 women initiated treatment after the initial assessment (40% were highly adherent with medication possession ratio 80% or greater). BMD decreased in untreated women and showed expected gains in adherent women (5.6–7.8% for the lumbar spine and 2.8–3.0% for the femoral neck). Despite this, median FRAX probability still increased, predominantly because of increasing age. Only 2.2% of women had a clinically important decrease in major osteoporotic fracture probability (4% or greater) and only 1.2% had an important decrease in hip fracture probability (1% or greater). Baseline FRAX probability was strongly predictive of incident major osteoporotic fractures (hazard ratio 1.8 per SD) and hip fractures (hazard ratio 4.5 per SD), but change in FRAX score was not an independent predictor of fractures. At present, only BMD and bone turnover markers appear to be sufficiently responsive to pharmacologic therapy to serve as possible targets. The value of repeat BMD testing for prediction of incident fractures has been an area of controversy. Berry et al. [16 ] examined this in the populationbased Framingham osteoporosis study (310 men and 492 women with two BMD measurements within 4 years), with fracture outcomes assessed for 10 years. In untreated men and women aged 75 years and older, the second BMD measurement did not meaningfully improve fracture prediction (area under the curve 0.71 for baseline BMD, 0.72 after addition of BMD change). The net reclassification improvement was not statistically significant. &&
&
ATYPICAL FEMORAL FRACTURES Atypical subtrochanteric and diaphyseal femoral fractures figured prominently in 2013. The American Society of Bone and Mineral Research’s second Task Force report [17 ] highlighted changes in the case definition (see list below) and the accumulating evidence of the strong association between these fractures and potent antiresorptive treatments (bisphosphonates and denosumab), a risk that may decline with cessation of treatment or treatment with teriparatide. A nationwide study from Sweden (59 atypical and 218 usual femoral fractures) noted a unique stress-type pattern, a transverse fracture angle, and lateral cortical thickening callus reaction in those with atypical fractures. There was also a strong association with bisphosphonate use (specificity 93%). Adjudicated atypical femoral fractures (AFFs) in 2007–2011 from a large integrated &&
healthcare provider showed prior bisphosphonate exposure in 128 of 142 cases (90%). The age-adjusted incidence increased with longer duration of use (from 1.78/100 000 years after 0.1–1.9 years to 113.1/100 000 years after 8–9.9 years) but was much lower than the expected rate of typical hip fractures (384–544/100 000 years) [18]. 2010 American Society for Bone and Mineral Research (ASBMR) Task Force case definitions of AFFs are as follows (specifically excluded are fractures of the femoral neck, intertrochanteric fractures with spiral subtrochanteric extension, pathological fractures associated with primary or metastatic bone tumors, and periprosthetic fractures): (1) Major features (All major features are required to satisfy the case definition of AFF. None of the minor features is required but have been sometimes associated with these fractures.): (a) located anywhere along the femur from just distal to the lesser trochanter to just proximal to the supracondylar flare; (b) associated with no trauma or minimal trauma, as in a fall from a standing height or less; (c) transverse or short oblique configuration; (d) noncomminuted; (e) complete fractures extend through both cortices and may be associated with a medial spike; incomplete fractures involve only the lateral cortex; (2) Minor features: (a) localized periosteal reaction of the lateral cortex (often referred to in the literature as ‘beaking’ or ‘flaring’); (b) generalized increase in cortical thickness of the diaphysis; (c) prodromal symptoms such as dull or aching pain in the groin or thigh; (d) bilateral fractures and symptoms; (e) delayed healing; (f) comorbid conditions (e.g. vitamin D deficiency, rheumatoid arthritis, and hypophosphatasia); (g) use of pharmaceutical agents (e.g. bisphosphonates, glucocorticoids, and proton pump inhibitors). Methods for case finding prior to a complete fracture are needed. One intriguing option is to extend DXA scans to the lower femur at the time of BMD testing to look for localized periosteal reaction, an early sign of incomplete AFFs. One preliminary report in patients over age 50 years who had received bisphosphonate therapy for more than 5 years identified suggestive changes of atypical
1040-8711 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-rheumatology.com
443
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Metabolic bone disease
fractures in 19 of 2057 patients (7.4%), among whom 7 (2.7%) were confirmed to have AFFs [19].
MALE OSTEOPOROSIS In the absence of a defining fracture event, the diagnosis of osteoporosis is based on the measurement of BMD by DXA. Two decades ago, the WHO provided an operational definition of osteoporosis for postmenopausal women as a BMD that lies 2.5 standard deviations or more below the average mean value for young healthy women (T-score –2.5 SD) [1]. More recently, a standardized reference site, the femoral neck, and a standard reference range have been advocated for both men and women (the NHANES III data for women aged 20–29 years) [20] based upon the accumulating data that fracture rates for men and women are similar for a given absolute BMD [21 ]. Several of the studies cited to support this change were published in 2013 [22 ]. Implementation of these new recommendations will be facilitated by changes to DXA software. Analogous to the situation with men, the International Society for Clinical Densitometry now recommends the use of white female reference data for all ethnic groups subject to local requirements [21 ]. Z-scores are still adjusted for age and sex and (where possible) ethnicity, and can take advantage of local reference data where available. &&
&
&&
FALLS AND PHYSICAL ACTIVITY Large international variation in fracture risk appears to be partially related to falls. In a large crosssectional study [23] conducted in 10 998 men aged 65 years and older from three regions (Hong Kong, United States, and Sweden), self-reported falls and fractures in the preceding 12 months showed significant variation. The proportion of fallers was highest in the USA and lowest in Hong Kong. Of interest, there were no differences in the proportion of fallers in the U.S. men of different ethnicities. This suggests that social and environmental factors have a major influence on the geographic variation in falls and fractures. A systematic review of falls and fractures among older individuals found that fall rates were significantly higher for western countries compared with east Asian countries [23]. Among western cohorts, 4.1% of falls result in fracture, whereas 86–95% fractures are secondary to falls. In a longitudinal cohort study [24] involving 2731 men (mean age 79 years), physical activity showed a significant association with falling that was modified by age. Men younger than 80 with low energy expenditure had a lower risk of falling than those with high energy expenditure (RR 0.75), whereas 444
www.co-rheumatology.com
among men aged 80 years and older, low energy expenditure showed a higher risk of falling than for men with high energy expenditure (RR 1.43). After adjustment for health status, low active energy expenditure was associated with higher risk of fracture (hazard ratio 1.82, lowest quintile versus highest quintile), as it was less than 33 min per day of moderate activity (hazard ratio 1.70).
SARCOPENIA, OBESITY, AND METABOLIC SYNDROME Sarcopenia, which encompasses muscle mass, strength, and physical performance, is being recognized as an important determinant of bone health and osteoporotic fractures. In the Epidemiologie de l’Osteoporose cohort (975 women aged 75 years and older), skeletal muscle mass measured by DXA was correlated with functional decline over the ensuing 4 years [25]. Lower skeletal muscle mass, older age, one or more comorbidities, and impaired chair-stand test were associated with functional decline in at least one instrumental activity of daily living. Historically, obesity has been associated with higher BMD and lower fracture risk, though this simplistic view has been questioned based upon the emerging data that obese women and men are at higher fracture risk than expected [26 ]. A large meta-analysis examining the association between BMI and fracture risk in prospective cohorts from 25 countries (398 610 women average at 63 years, 2.2 million person-years follow-up) reported the prevalence of obesity as 22%, with 19% of major osteoporotic fractures and 13% of hip fractures occurring in obese women [27 ]. Obese women were at lower risk for major osteoporotic fractures when BMD was not considered (hazard ratio 0.87 for BMI 35 versus 25 kg/m2), but at higher risk when BMD was considered (hazard ratio 1.16). In this analysis, low BMI was a risk factor for hip and all osteoporotic fractures, and was associated with lower risk for lower leg fractures; conversely, high BMI was a risk factor for upper-arm fractures. Even after adjustment for BMD, low BMI was a risk factor for hip fracture, and high BMI remained a risk factor for upper-arm fracture and also all osteoporotic fractures. Despite this, there is evidence from the Study of Osteoporotic Fractures that FRAX is still of value in predicting hip and major osteoporotic fractures in obese postmenopausal women, particularly when BMD is used in the risk calculation [28]. The complexity of the BMI–fracture relationship was underscored in the Global Longitudinal study of Osteoporosis in Women (GLOW) [29] which found, in 52 939 women with 3628 (6.9%) incident clinical &
&&
Volume 26 Number 4 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Osteoporosis epidemiology 2013 Leslie and Morin
fractures, that relationships between fracture and weight, BMI, and height were site specific. Furthermore, use of antiosteoporosis medication was significantly lower in obese women who underwent a longer period of hospitalization for treatment and had poorer functional status and health-related quality of life [30]. The mechanism underlying the association between obesity and fracture risk may include metabolic factors in addition to biomechanical considerations. Metabolic syndrome has been associated with lower BMD and greater bone loss in three publications on Korean women and men [31–33]. Type 2 diabetes is now recognized as a risk factor for fracture. An interesting hypothesis evaluated in 2013 was that bone metabolism may actually affect risk for diabetes through the effect of undercarboxylated osteocalcin, which has been shown to promote insulin sensitivity and secretion in animal models [34]. To see whether this applied to humans, three randomized trials were pooled in a post hoc analysis to see whether incident diabetes was related to antiresorptive treatment (alendronate n ¼ 6151, zoledronic acid n ¼ 7113, or denosumab n ¼ 7076) [35 ]. By suppressing bone turnover, it was hypothesized that lower undercarboxylated osteocalcin might lead to an increase in diabetes. In fact, this was not the case and there was no suggestion that antiresorptive therapy had a clinically important effect on fasting glucose, weight, or diabetes risk. This would suggest that animal models of undercarboxylated osteocalcin as a regulator of energy homeostasis may not be applicable to humans. &&
CONCLUSION Osteoporotic fractures have long-term consequences with excess mortality up to 10 years and increased economic costs up to 5 years. The large burden of NHNV fractures has been underestimated. A plethora of new data highlight the importance of falls, physical activity, and body composition as contributors to skeletal health. Acknowledgements Sources of support: S.N.M. is a Research Scholar funded by Health Research in Quebec. Conflicts of interest Disclosures: W.D.L. (all fees paid to facility): Speaker bureau: Amgen, Eli Lilly, Novartis. Research grants: Amgen, Genzyme. S.N.M.: Consultant to: Amgen, Novartis, Eli Lilly, Merck. Speaker bureau: Amgen, Novartis. Research grant: Amgen.
REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest && of outstanding interest 1. Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. WHO Study Group. Osteoporos Int 1994; 4:368–381. 2. Tajeu GS, Delzell E, Smith W, et al. Death, debility, and destitution following hip fracture. J Gerontol A Biol Sci Med Sci 2014; 69:346–353. 3. Bliuc D,Nguyen TV, EismanJA, Center JR. The impact ofnonhip nonvertebralfractures in elderly women and men. J Clin Endocrinol Metab 2014; 99:415–423. 4. Bliuc D, Nguyen ND, Nguyen TV, et al. Compound risk of high mortality && following osteoporotic fracture and refracture in elderly women and men. J Bone Miner Res 2013; 28:2317–2324. Osteoporotic fractures are associated with higher short-term and long-term mortality. Most of the 5-year to 10-year excess mortality was associated with re-fracture and highlights the importance of optimizing care after the initial fracture event to interrupt the ‘fracture cascade’. 5. Frost SA, Nguyen ND, Center JR, et al. Excess mortality attributable to hip& fracture: a relative survival analysis. Bone 2013; 56:23–29. The Dubbo Osteoporosis Epidemiology Study is a prospective epidemiologic study of osteoporosis and fractures in more than 2000 men and women aged 60 and above, started in 1989 with 21 years of follow-up. This study is well positioned to examine long-term mortality after hip fracture and highlights progressive declines in relative survival rates at 1, 5, and 10 years. On average, posthip fracture women died 4 years earlier [median: 4.1, interquartile range (IQR) 1.7– 7.8] and men died 5 years earlier (median ¼ 4.8, IQR 2.4–7.0) than expected. 6. Ioannidis G, Flahive J, Pickard L, et al. Nonhip, nonspine fractures drive healthcare utilization following a fracture: the Global Longitudinal Study of Osteoporosis in Women (GLOW). Osteoporos Int 2013; 24:59–67. 7. Leslie WD, Lix LM, Finlayson GS, et al. Direct healthcare costs for 5 years postfracture in Canada: a long-term population-based assessment. Osteoporos Int 2013; 24:1697–1705. 8. Cooper C, Cole ZA, Holroyd CR, et al. Secular trends in the incidence of hip and other osteoporotic fractures. Osteoporos Int 2011; 22:1277–1288. 9. Looker AC, Melton LJ III, Borrud LG, Shepherd JA. Changes in femur neck bone density in US adults between 1988–1994 and 2005–2008: demographic patterns and possible determinants. Osteoporos Int 2012; 23:771–780. 10. Leslie WD, Lix LM, Yogendran MS, et al. Temporal trends in obesity, && osteoporosis treatment, bone mineral density and fracture rates: a population-based historical cohort study. J Bone Miner Res 2014; 29:952–959. The National Health and Nutrition Examination Survey (NHANES) III femoral neck BMD database in young non-Hispanic white women is the international reference standard for T-score calculation, diagnosis of osteoporosis and fracture probability estimation under FRAX. NHANES reported that femoral BMD increased by approximately 6% in non-Hispanic white women between 1988–1994 and 2005–2008. Two complementary population-based historical cohorts from the Province of Manitoba, Canada (1996–2006) confirmed declining major osteoporotic fractures rates and increasing BMD. Surprisingly, temporal increases in BMI, obesity, and osteoporosis treatment did not account for these changes. Improvements in BMD (still unexplained) were found to explain the decline in major osteoporotic fracture rates from 1996 to 2006. One corollary of this study is to support the continued validity and use of NHANES III (1988–2004) as the international reference standard for T-score calculation, diagnosis of osteoporosis, and fracture probability estimation under FRAX. 11. Rubin KH, Abrahamsen B, Friis-Holmberg T, et al. Comparison of different screening tools (FRAX(R), OST, ORAI, OSIRIS, SCORE and age alone) to identify women with increased risk of fracture: a population-based prospective study. Bone 2013; 56:16–22. 12. Rubin KH, Friis-Holmberg T, Hermann AP, et al. Risk assessment tools to & identify women with increased risk of osteoporotic fracture: complexity or simplicity? A systematic review. J Bone Miner Res 2013; 28:1701–1717. This important, systematic review of fracture risk assessment tools examined the PubMed, Embase, and Cochrane databases. Of 48 tools identified, only 6 had been externally validated and tested more than once in a population-based setting with acceptable methodological quality. None of the tools performed consistently better than the others; simple tools often did as well or better than more complex tools. The authors highlight the need for high-quality studies in population-based cohorts with different case mixes to determine the effectiveness of these tools in selecting patients for therapy. 13. Lam A, Leslie W, Lix L, et al. Major osteoporotic to hip fracture ratios in Canadian men and women with Swedish comparisons: a population based analysis. J Bone Miner Res 2013. [Epub ahead of print] 14. Cummings SR, Cosman F, Eastell R, et al. Goal-directed treatment of && osteoporosis. J Bone Miner Res 2013; 28:433–438. This study proposes a bold new way of looking at the initiation and monitoring of osteoporosis treatment. The idea is to follow the example of other conditions, such as hypertension, in which treatment is based on achieving a goal. Although there are many obstacles in setting treatment goals, the result could be more rational and effective use of the expanding array of treatments for osteoporosis.
1040-8711 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-rheumatology.com
445
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Metabolic bone disease 15. Leslie WD, Majumdar SR, Lix LM, et al. Can change in FRAX Score be used to ‘Treat-to-Target’? A population-based cohort study. J Bone Miner Res 2013. [Epub ahead of print] This study in 11 049 previously untreated women aged more than 50 years undergoing baseline and follow-up DXA examinations and FRAX probability calculations provided a clear answer to the question – ‘No’. FRAX scores increased over time and this increase was attenuated but not prevented by the treatment. Few women had meaningful reductions in FRAX scores. FRAX with BMD is not responsive enough to be used as a target for goal-directed treatment. 16. Berry SD, Samelson EJ, Pencina MJ, et al. Repeat bone mineral density & screening and prediction of hip and major osteoporotic fracture. JAMA 2013; 310:1256–1262. The role of BMD monitoring has been an area of unresolved – and often times heated – controversy. In untreated individuals, there have been conflicting data on whether BMD change can be used to refine fracture risk beyond the baseline measurement. This study found that a second BMD measurement after 4 years in untreated men and women of mean age 75 years did not meaningfully improve the prediction of hip or major osteoporotic fracture. 17. Shane E, Burr D, Abrahamsen B, et al. Atypical subtrochanteric and diaphy&& seal femoral fractures: second report of a task force of the American Society for Bone and Mineral Research. J Bone Miner Res 2014; 29:1–23. Atypical femur fractures (AFFs), initially reported in patients taking bisphosphonates and more recently in patients on denosumab, have generated fear from patients, healthcare providers, and regulators out of proportion to their infrequency. The original ASBMR Task Force report in 2010 has been updated to reflect emerging data on the epidemiology, pathogenesis, and medical management of AFFs. The case definition has also been modified: periosteal stress reaction at the fracture site was changed from a minor to a major feature. 18. Dell RM, Adams AL, Greene DF, et al. Incidence of atypical nontraumatic diaphyseal fractures of the femur. J Bone Miner Res 2012; 27:2544–2550. 19. McKenna MJ, van der Kamp S, Heffernan E, Hurson C. Incomplete atypical femoral fractures: assessing the diagnostic utility of DXA by extending femur length. J Clin Densitom 2013; 16:579–583. 20. Kanis JA, McCloskey EV, Johansson H, et al. A reference standard for the description of osteoporosis. Bone 2008; 42:467–475. 21. Schousboe JT, Shepherd JA, Bilezikian JP, Baim S. Executive summary of the && 2013 International Society for Clinical Densitometry Position Development Conference on bone densitometry. J Clin Densitom 2013; 16:455–466. The International Society for Clinical Densitometry (ISCD) convenes a Position Development Conference (PDC) every 2–3 years to make recommendations for guidelines and standards in the field of musculoskeletal measurement and assessment. This report summarizes the results of the 2013 ISCD PDC for vertebral fracture assessment/DXA and National Health and Nutrition Survey (NHANES) Reference Database Task Forces, and a separate article summarized the results of the Body Composition Analysis Task Forces. Of particular importance are the recommendations on Normative Databases for Bone Mineral Density T-Scores. 22. Watts NB, Leslie WD, Foldes AJ, Miller PD. 2013 International Society for & Clinical Densitometry Position Development Conference: Task Force on Normative Databases. J Clin Densitom 2013; 16:472–481. This report from the ISCD Task Force on Normative Databases provided the evidence base for the Position Development Conference (PDC) and new position statements: manufacturers should continue to use their own databases for the lumbar spine as the reference standard for T-scores; manufacturers should continue to use National Health and Nutrition Examination Survey III data as the reference standard for femoral neck and total hip T-scores; if local reference data are available, they should be used to calculate only Z-scores but not T-scores; and a uniform Caucasian (nonrace adjusted) female reference database should be used to calculate T-scores for men of all ethnic groups. &&
446
www.co-rheumatology.com
23. Karlsson MK, Ribom EL, Nilsson JA, et al. International and ethnic variability of falls in older men. Scand J Public Health 2014; 42:194–200. 24. Cauley JA, Harrison SL, Cawthon PM, et al. Objective measures of physical activity, fractures and falls: the osteoporotic fractures in men study. J Am Geriatr Soc 2013; 61:1080–1088. 25. Amigues I, Schott AM, Amine M, et al. Low skeletal muscle mass and risk of functional decline in elderly community-dwelling women: the prospective EPIDOS study. J Am Med Dir Assoc 2013; 14:352–357. 26. Compston J. Obesity and bone. Curr Osteoporos Rep 2013; 11:30–35. &
This review examines the recent studies showing that fractures in obese postmenopausal women and older men contribute significantly to the overall fracture burden. 27. Johansson H, Kanis JA, Oden A, et al. A meta-analysis of the association of && fracture risk and body mass index in women. J Bone Miner Res 2014; 29:223–233. This large-scale meta-analysis is the definitive word to date on the association between BMI (as an index of underweight and obesity) and its effect on fracture risk in postmenopausal women. At a population level, high BMI was a protective factor for most sites of fragility fracture, but the association is complex, differs across skeletal sites, and is modified by the interaction between BMI and BMD. Osteoporotic fracture risk was lower for a BMI of 35 kg/m2 compared with 25 kg/m2 [hazard ratio 0.87 with 95% confidence interval (CI) 0.85–0.90]. When adjusted for bone mineral density (BMD), however, the same comparison showed that the hazard ratio for osteoporotic fracture was increased (hazard ratio 1.16; 95% CI 1.09–1.23). Thus, higher BMD values seen in obesity may not exert the level of protection expected. 28. Premaor M, Parker RA, Cummings S, et al. Predictive value of FRAX for fracture in obese older women. J Bone Miner Res 2013; 28:188–195. 29. Compston JE, Flahive J, Hosmer DW, et al. Relationship of weight, height, and body mass index with fracture risk at different sites in postmenopausal women: the Global Longitudinal Study of Osteoporosis in Women (GLOW). J Bone Miner Res 2014; 29:487–493. 30. Compston JE, Flahive J, Hooven FH, et al. Obesity, health-care utilization, and health-related quality of life after fracture in postmenopausal women: Global Longitudinal Study of Osteoporosis in Women (GLOW). Calcif Tissue Int 2014; 94:223–231. 31. Kim BJ, Ahn SH, Bae SJ, et al. Association between metabolic syndrome and bone loss at various skeletal sites in postmenopausal women: a 3-year retrospective longitudinal study. Osteoporos Int 2013; 24:2243–2252. 32. Kim T, Park S, Pak YS, et al. Association between metabolic syndrome and bone mineral density in Korea: the Fourth Korea National Health and Nutrition Examination Survey (KNHANES IV), 2008. J Bone Miner Metab 2013; 31:652–662. 33. Kim YH, Cho KH, Choi YS, et al. Low bone mineral density is associated with metabolic syndrome in South Korean men but not in women: the 2008–2010 Korean National Health and Nutrition Examination Survey. Arch Osteoporos 2013; 8:142. 34. Lee NK, Sowa H, Hinoi E, et al. Endocrine regulation of energy metabolism by the skeleton. Cell 2007; 130:456–469. 35. Schwartz AV, Schafer AL, Grey A, et al. Effects of antiresorptive therapies on && glucose metabolism: results from the FIT, HORIZON-PFT, and FREEDOM trials. J Bone Miner Res 2013; 28:1348–1354. This is a good news story for osteoporosis. Evidence in animal models suggests that the skeleton controls energy homeostasis through the feedback action of (undercarboxylated) osteocalcin and adiponectin, an insulin-sensitizing adipokine. Contrary to the predictions from mouse models, pharmacological reductions in bone turnover do not appear to increase the risk of insulin resistance and diabetes in humans.
Volume 26 Number 4 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Osteoporosis epidemiology 2013: implications for diagnosis, risk assessment, and treatment William D. Leslie a and Suzanne N. Morin b
Purpose of review To summarize the recently published studies that provide insights into the changing epidemiology of osteoporosis and fractures. Recent findings The main themes reviewed are fracture outcomes; trends in fractures rates; fracture risk assessment and monitoring; atypical femoral fractures; male osteoporosis; falls and physical activity; and sarcopenia, obesity, and metabolic syndrome. Summary Osteoporotic fractures were found to have long-term consequences on excess mortality (10 years) and economic costs (5 years). The large burden of nonhip nonvertebral fractures has been underestimated. Divergent (but mostly declining) trends in fracture rates were confirmed in several cohorts from around the world. This has significant implications for healthcare planners and clinicians responsible for the care of individuals with osteoporosis, and also impacts on the calibration of fracture prediction tools. Although fracture prediction tools differ in their complexity, performance characteristics are similar when applied to the general population. Large, high-quality comparative studies with different case mixes are needed. Fracture probability does not appear to be responsive enough to support goal-directed treatment at this time. A consensus on the diagnosis of osteoporosis in men has emerged, based upon the same absolute bone density cutoff for both men and women. Finally, a plethora of new data highlight the importance of falls, physical activity, and body composition as contributors to skeletal health. Keywords bone mineral density, clinical practice guidelines, epidemiology, fracture, osteoporosis
INTRODUCTION Osteoporosis is characterized by low bone mass and skeletal fragility resulting in increased susceptibility to low-trauma fractures, especially those affecting the vertebrae, proximal femur (hip), distal forearm, and proximal humerus which are collectively referred to as major osteoporotic fractures [1]. This update summarizes the recently published studies that provide insights into the changing epidemiology of osteoporosis and fractures.
FRACTURE OUTCOMES The consequences of osteoporotic fractures include re-fractures, excess mortality and morbidity, and economic costs for the healthcare system. Using a large U.S. administrative database, Tajeu et al. [2] documented that the year following hip fracture was associated with a two-fold increase in death, a twofold increase in destitution (entering into lowincome status), and a four-fold increase in debility (requiring long-term care). www.co-rheumatology.com
Three notable publications in 2013 were from the Dubbo Osteoporosis Epidemiology Study, a prospective cohort of community-dwelling participants aged 60 years and older who have been followed since 1989 [3,4 ,5 ]. Although adverse outcomes are most frequently reported in relation to hip and vertebral fractures, these and other recent data highlight the importance of nonhip nonvertebral (NHNV) fractures as they are collectively much more numerous and make a larger contribution to the burden of osteoporosis at the population level [6]. Of the 952 documented fractures in women and 343 &&
&
a
University of Manitoba, Winnipeg and bMcGill University, Montreal, Manitoba, Canada Correspondence to Dr William D. Leslie, MD, MSc, Department of Medicine (C5121), St Boniface General Hospital, 409 Tache Avenue, Winnipeg, MB R2H 2A6, Canada. Tel: +1 204 237 2311; fax: +1 204 237 2007; e-mail: [email protected] Curr Opin Rheumatol 2014, 26:440–446 DOI:10.1097/BOR.0000000000000064 Volume 26 Number 4 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Osteoporosis epidemiology 2013 Leslie and Morin
an initial hip fracture, age-specific and sex-specific survival relative to the general population was greatly reduced in women (83% at 1 years, 59% at 5 years, and 31% 10 years) and men (63% at 1 years, 48% at 5 years, and 36% 10 years) [5 ]. For every six women and for every three men with hip fracture, one extra death occurred above that expected in the general population. Excess mortality impacts on the long-term economic burden of fractures. A population-based Canadian study [7] (16 198 incident fractures cases and 48 594 matched nonfracture controls) reported high direct healthcare costs for the first year after fracture, greatest for hip fractures (median $25 306 in women and $21 396 in men). Incremental costs declined considerably after the first year, in part because of higher mortality among those with fractures who already were the heaviest users of healthcare (healthy survivor bias). From the healthcare payer perspective, total costs for all fractures combined eventually fell below the prefracture levels because of the excess mortality among fracture cases. However, among those surviving 5 years following fracture, healthcare costs remained above prefracture levels.
KEY POINTS Divergent (but mostly declining) trends in fracture rates were confirmed in several cohorts from around the world.
&
Although fracture prediction tools differ in their complexity, performance characteristics are similar when applied to the general population. Fracture probability does not appear to be responsive enough to support goal-directed treatment at this time. A consensus on the diagnosis of osteoporosis in men has emerged, based upon the same absolute bone density cutoff for both men and women.
in men, over half were NHNV fractures (486 in women and 173 in men) and these were associated with a two-fold relative risk (RR) of subsequent fracture in women with even greater risk in men [3]. More importantly, NHNV fractures were associated with 20% excess mortality in the first 5 years after fracture. Re-fractures contributed substantially to this excess mortality as demonstrated in a competing risk analysis [4 ]. Total mortality (following initial and subsequent fractures) was elevated for 10 years, and most of the later excess mortality (5–10 years) was associated with re-fractures (Fig. 1). Among 151 women and 55 men sustaining &&
TRENDS IN FRACTURE RATES Large international variations in osteoporotic fracture rates had been reported, with temporal trends
Women
(a)
Men
(b)
100
100
80
80
60
60
40
40
Cumulative incidences
Total deaths
20
Deaths following initial fracture Age-matched population deaths
20 Expected mortality in the general population
Expected mortality in the general population
0
0 0
Fracture population 952
5
10
15
20
0
5
10
15
20
487
230
90
20
343
151
68
33
7
Time, years
FIGURE 1. Stacked graph of cumulative incidences of mortality following initial osteoporotic fracture in black and following re-fracture in gray compared with an age-matched general population: (a) women and (b) men. The dark gray area represents the excess mortality (i.e. above population mortality) related to the initial fracture. The light gray area represents the excess mortality related to the re-fracture. The height of the jagged, black, and gray line represents the cumulative incidence of mortality in the general population, mortality following initial fracture, and total mortality at a given time. Reproduced with permission [4 ]. &&
1040-8711 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-rheumatology.com
441
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Metabolic bone disease
that differ between populations [8]. Data from the Rochester Epidemiology Project (Olmsted County, Minnesota, USA) were used to compare the fracture rates in 2009–2011 with those from 1989 to 1991. In women, hip fractures showed a 25% decline with a 26% reduction in distal forearm fractures. However, the overall age-adjusted fracture rate increased by 11%, largely attributed to incidentally diagnosed vertebral fractures. The basis for declining hip and nonhip fracture rates in industrialized countries remains a subject of uncertainty. Data from the National Health and Nutrition Examination Survey (NHANES) noted a 6% increase in femoral neck bone mineral density (BMD) among non-Hispanic white women between 1988–1994 and 2005–2008 adjusted for multiple determinants [9]. Similar data were reported from a large Canadian BMD registry that included baseline dual-energy X-ray absorptiometry (DXA) for 36 587 women aged 50 years and older [10 ]. Even after adjustment for increasing osteoporosis treatment and obesity, there was a significant linear increase in BMD at both the femoral neck (0.52% per year) and lumbar spine (0.32% per year). Moreover, when incident major osteoporotic fracture rates and hip fracture rates were studied in the same population, there was a significant linear decline from 1996 to 2006, but adjusting for BMD accounted for the observed reduction in fractures. These data would suggest that improvement in BMD, still unexplained and not attributable to greater rates of osteoporosis treatment or obesity, is a major factor contributing to reductions in osteoporotic fracture rates. It is interesting to note that a secular increase in BMD was also reported among Southern Chinese women between 1995–2000 and 2005–2010 after adjustment for other factors (lumbar spine 4.17%, femoral neck 9.02%, and total hip 9.34%) among women aged 50 years and older. If confirmed in other Asian populations, then this may herald the reversal in increasing fracture rates that have been noted in this part of the world. &&
FRACTURE RISK ASSESSMENT AND MONITORING The need for screening and fracture risk assessment tools that go beyond BMD is now widely accepted. Questions arise regarding the number of risk factors required, with some systems including a small number of risk factors (e.g. Garvan FRC, www.garvan. org.au/bone-fracture-risk/), whereas others contain a large number of risk factors (e.g. QFracture, www.qfracture.org/). There is a tradeoff between complexity (greater accuracy) and simplicity (adoption in clinical practice). Rubin et al. [11] compared 442
www.co-rheumatology.com
the WHO Fracture Risk Assessment (FRAX) tool without BMD (www.shef.ac.uk/FRAX/) and several simpler screening tools (OST, ORAI, OSIRIS, SCORE, and age alone) for predicting fractures in 3614 women aged 40–90 years followed through the Danish National Register over 3 years. All tools showed similar fracture discrimination (area under curve between 0.703 and 0.722). This may indicate that for populations with a low prevalence of risk factors, a simpler screening tool may be sufficient. The same group performed a systematic review of screening and risk assessment tools [12 ]. Of a total of 48 tools, only 6 had been tested more than once in a population-based setting with acceptable methodology. Once again, there was no consistent evidence that more complex tools (SCORE, FRAX, and QFracture) had better performance characteristics than simpler tools (OST, ORAI, and Garvan FRC). However, the paucity of head-to-head comparisons limits any definitive conclusions. Larger, high-quality studies with different case mixes are needed to address this important question and to determine these tools’ effectiveness in selecting patients for therapy. Changing fracture epidemiology has implications for fracture risk assessment using tools calibrated to estimate absolute fracture risk. For example, the WHO FRAX system estimates 10-year major osteoporotic fracture and 10-year hip fracture probability. Changes in a country’s fracture rates would necessitate FRAX recalibration. For many countries without high-quality nonhip fracture data, major osteoporotic fracture rates are estimated based upon an assumption that these show a similar nonhip-to-hip fracture ratio as historical Swedish data (1987–1996). Population-based data from Canada suggest that these ratios may underestimate the rate of major osteoporotic fractures, possibly because of declining hip fracture rates [13]. Although Swedish vertebral and hip fracture ratios were similar to Canada, ratios were significantly lower for other sites (men and women: 46 and 35% lower for forearm and hip ratios, 19 and 15% lower for humerus and hip ratios, and 19 and 23% lower for any major osteoporotic fractures and hip ratio, respectively). The need to develop more responsive methods for monitoring drug treatment in osteoporosis was highlighted in an initiative to develop goal-directed treatment: ‘treat-to-target’ guidelines for osteoporosis [14 ]. Despite the intuitive appeal of this approach, it is very challenging to identify suitably responsive indices and treatment targets. Fracture probability might appear to be a logical candidate. However, FRAX was not found to be sufficiently responsive to antiosteoporosis treatment for this &
&&
Volume 26 Number 4 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Osteoporosis epidemiology 2013 Leslie and Morin
purpose in an analysis of 11 049 previously untreated women aged 50 years and older from a large BMD registry with paired FRAX estimates (median interval 4 years) [15 ]. A total of 6534 women initiated treatment after the initial assessment (40% were highly adherent with medication possession ratio 80% or greater). BMD decreased in untreated women and showed expected gains in adherent women (5.6–7.8% for the lumbar spine and 2.8–3.0% for the femoral neck). Despite this, median FRAX probability still increased, predominantly because of increasing age. Only 2.2% of women had a clinically important decrease in major osteoporotic fracture probability (4% or greater) and only 1.2% had an important decrease in hip fracture probability (1% or greater). Baseline FRAX probability was strongly predictive of incident major osteoporotic fractures (hazard ratio 1.8 per SD) and hip fractures (hazard ratio 4.5 per SD), but change in FRAX score was not an independent predictor of fractures. At present, only BMD and bone turnover markers appear to be sufficiently responsive to pharmacologic therapy to serve as possible targets. The value of repeat BMD testing for prediction of incident fractures has been an area of controversy. Berry et al. [16 ] examined this in the populationbased Framingham osteoporosis study (310 men and 492 women with two BMD measurements within 4 years), with fracture outcomes assessed for 10 years. In untreated men and women aged 75 years and older, the second BMD measurement did not meaningfully improve fracture prediction (area under the curve 0.71 for baseline BMD, 0.72 after addition of BMD change). The net reclassification improvement was not statistically significant. &&
&
ATYPICAL FEMORAL FRACTURES Atypical subtrochanteric and diaphyseal femoral fractures figured prominently in 2013. The American Society of Bone and Mineral Research’s second Task Force report [17 ] highlighted changes in the case definition (see list below) and the accumulating evidence of the strong association between these fractures and potent antiresorptive treatments (bisphosphonates and denosumab), a risk that may decline with cessation of treatment or treatment with teriparatide. A nationwide study from Sweden (59 atypical and 218 usual femoral fractures) noted a unique stress-type pattern, a transverse fracture angle, and lateral cortical thickening callus reaction in those with atypical fractures. There was also a strong association with bisphosphonate use (specificity 93%). Adjudicated atypical femoral fractures (AFFs) in 2007–2011 from a large integrated &&
healthcare provider showed prior bisphosphonate exposure in 128 of 142 cases (90%). The age-adjusted incidence increased with longer duration of use (from 1.78/100 000 years after 0.1–1.9 years to 113.1/100 000 years after 8–9.9 years) but was much lower than the expected rate of typical hip fractures (384–544/100 000 years) [18]. 2010 American Society for Bone and Mineral Research (ASBMR) Task Force case definitions of AFFs are as follows (specifically excluded are fractures of the femoral neck, intertrochanteric fractures with spiral subtrochanteric extension, pathological fractures associated with primary or metastatic bone tumors, and periprosthetic fractures): (1) Major features (All major features are required to satisfy the case definition of AFF. None of the minor features is required but have been sometimes associated with these fractures.): (a) located anywhere along the femur from just distal to the lesser trochanter to just proximal to the supracondylar flare; (b) associated with no trauma or minimal trauma, as in a fall from a standing height or less; (c) transverse or short oblique configuration; (d) noncomminuted; (e) complete fractures extend through both cortices and may be associated with a medial spike; incomplete fractures involve only the lateral cortex; (2) Minor features: (a) localized periosteal reaction of the lateral cortex (often referred to in the literature as ‘beaking’ or ‘flaring’); (b) generalized increase in cortical thickness of the diaphysis; (c) prodromal symptoms such as dull or aching pain in the groin or thigh; (d) bilateral fractures and symptoms; (e) delayed healing; (f) comorbid conditions (e.g. vitamin D deficiency, rheumatoid arthritis, and hypophosphatasia); (g) use of pharmaceutical agents (e.g. bisphosphonates, glucocorticoids, and proton pump inhibitors). Methods for case finding prior to a complete fracture are needed. One intriguing option is to extend DXA scans to the lower femur at the time of BMD testing to look for localized periosteal reaction, an early sign of incomplete AFFs. One preliminary report in patients over age 50 years who had received bisphosphonate therapy for more than 5 years identified suggestive changes of atypical
1040-8711 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-rheumatology.com
443
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Metabolic bone disease
fractures in 19 of 2057 patients (7.4%), among whom 7 (2.7%) were confirmed to have AFFs [19].
MALE OSTEOPOROSIS In the absence of a defining fracture event, the diagnosis of osteoporosis is based on the measurement of BMD by DXA. Two decades ago, the WHO provided an operational definition of osteoporosis for postmenopausal women as a BMD that lies 2.5 standard deviations or more below the average mean value for young healthy women (T-score –2.5 SD) [1]. More recently, a standardized reference site, the femoral neck, and a standard reference range have been advocated for both men and women (the NHANES III data for women aged 20–29 years) [20] based upon the accumulating data that fracture rates for men and women are similar for a given absolute BMD [21 ]. Several of the studies cited to support this change were published in 2013 [22 ]. Implementation of these new recommendations will be facilitated by changes to DXA software. Analogous to the situation with men, the International Society for Clinical Densitometry now recommends the use of white female reference data for all ethnic groups subject to local requirements [21 ]. Z-scores are still adjusted for age and sex and (where possible) ethnicity, and can take advantage of local reference data where available. &&
&
&&
FALLS AND PHYSICAL ACTIVITY Large international variation in fracture risk appears to be partially related to falls. In a large crosssectional study [23] conducted in 10 998 men aged 65 years and older from three regions (Hong Kong, United States, and Sweden), self-reported falls and fractures in the preceding 12 months showed significant variation. The proportion of fallers was highest in the USA and lowest in Hong Kong. Of interest, there were no differences in the proportion of fallers in the U.S. men of different ethnicities. This suggests that social and environmental factors have a major influence on the geographic variation in falls and fractures. A systematic review of falls and fractures among older individuals found that fall rates were significantly higher for western countries compared with east Asian countries [23]. Among western cohorts, 4.1% of falls result in fracture, whereas 86–95% fractures are secondary to falls. In a longitudinal cohort study [24] involving 2731 men (mean age 79 years), physical activity showed a significant association with falling that was modified by age. Men younger than 80 with low energy expenditure had a lower risk of falling than those with high energy expenditure (RR 0.75), whereas 444
www.co-rheumatology.com
among men aged 80 years and older, low energy expenditure showed a higher risk of falling than for men with high energy expenditure (RR 1.43). After adjustment for health status, low active energy expenditure was associated with higher risk of fracture (hazard ratio 1.82, lowest quintile versus highest quintile), as it was less than 33 min per day of moderate activity (hazard ratio 1.70).
SARCOPENIA, OBESITY, AND METABOLIC SYNDROME Sarcopenia, which encompasses muscle mass, strength, and physical performance, is being recognized as an important determinant of bone health and osteoporotic fractures. In the Epidemiologie de l’Osteoporose cohort (975 women aged 75 years and older), skeletal muscle mass measured by DXA was correlated with functional decline over the ensuing 4 years [25]. Lower skeletal muscle mass, older age, one or more comorbidities, and impaired chair-stand test were associated with functional decline in at least one instrumental activity of daily living. Historically, obesity has been associated with higher BMD and lower fracture risk, though this simplistic view has been questioned based upon the emerging data that obese women and men are at higher fracture risk than expected [26 ]. A large meta-analysis examining the association between BMI and fracture risk in prospective cohorts from 25 countries (398 610 women average at 63 years, 2.2 million person-years follow-up) reported the prevalence of obesity as 22%, with 19% of major osteoporotic fractures and 13% of hip fractures occurring in obese women [27 ]. Obese women were at lower risk for major osteoporotic fractures when BMD was not considered (hazard ratio 0.87 for BMI 35 versus 25 kg/m2), but at higher risk when BMD was considered (hazard ratio 1.16). In this analysis, low BMI was a risk factor for hip and all osteoporotic fractures, and was associated with lower risk for lower leg fractures; conversely, high BMI was a risk factor for upper-arm fractures. Even after adjustment for BMD, low BMI was a risk factor for hip fracture, and high BMI remained a risk factor for upper-arm fracture and also all osteoporotic fractures. Despite this, there is evidence from the Study of Osteoporotic Fractures that FRAX is still of value in predicting hip and major osteoporotic fractures in obese postmenopausal women, particularly when BMD is used in the risk calculation [28]. The complexity of the BMI–fracture relationship was underscored in the Global Longitudinal study of Osteoporosis in Women (GLOW) [29] which found, in 52 939 women with 3628 (6.9%) incident clinical &
&&
Volume 26 Number 4 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Osteoporosis epidemiology 2013 Leslie and Morin
fractures, that relationships between fracture and weight, BMI, and height were site specific. Furthermore, use of antiosteoporosis medication was significantly lower in obese women who underwent a longer period of hospitalization for treatment and had poorer functional status and health-related quality of life [30]. The mechanism underlying the association between obesity and fracture risk may include metabolic factors in addition to biomechanical considerations. Metabolic syndrome has been associated with lower BMD and greater bone loss in three publications on Korean women and men [31–33]. Type 2 diabetes is now recognized as a risk factor for fracture. An interesting hypothesis evaluated in 2013 was that bone metabolism may actually affect risk for diabetes through the effect of undercarboxylated osteocalcin, which has been shown to promote insulin sensitivity and secretion in animal models [34]. To see whether this applied to humans, three randomized trials were pooled in a post hoc analysis to see whether incident diabetes was related to antiresorptive treatment (alendronate n ¼ 6151, zoledronic acid n ¼ 7113, or denosumab n ¼ 7076) [35 ]. By suppressing bone turnover, it was hypothesized that lower undercarboxylated osteocalcin might lead to an increase in diabetes. In fact, this was not the case and there was no suggestion that antiresorptive therapy had a clinically important effect on fasting glucose, weight, or diabetes risk. This would suggest that animal models of undercarboxylated osteocalcin as a regulator of energy homeostasis may not be applicable to humans. &&
CONCLUSION Osteoporotic fractures have long-term consequences with excess mortality up to 10 years and increased economic costs up to 5 years. The large burden of NHNV fractures has been underestimated. A plethora of new data highlight the importance of falls, physical activity, and body composition as contributors to skeletal health. Acknowledgements Sources of support: S.N.M. is a Research Scholar funded by Health Research in Quebec. Conflicts of interest Disclosures: W.D.L. (all fees paid to facility): Speaker bureau: Amgen, Eli Lilly, Novartis. Research grants: Amgen, Genzyme. S.N.M.: Consultant to: Amgen, Novartis, Eli Lilly, Merck. Speaker bureau: Amgen, Novartis. Research grant: Amgen.
REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest && of outstanding interest 1. Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. WHO Study Group. Osteoporos Int 1994; 4:368–381. 2. Tajeu GS, Delzell E, Smith W, et al. Death, debility, and destitution following hip fracture. J Gerontol A Biol Sci Med Sci 2014; 69:346–353. 3. Bliuc D,Nguyen TV, EismanJA, Center JR. The impact ofnonhip nonvertebralfractures in elderly women and men. J Clin Endocrinol Metab 2014; 99:415–423. 4. Bliuc D, Nguyen ND, Nguyen TV, et al. Compound risk of high mortality && following osteoporotic fracture and refracture in elderly women and men. J Bone Miner Res 2013; 28:2317–2324. Osteoporotic fractures are associated with higher short-term and long-term mortality. Most of the 5-year to 10-year excess mortality was associated with re-fracture and highlights the importance of optimizing care after the initial fracture event to interrupt the ‘fracture cascade’. 5. Frost SA, Nguyen ND, Center JR, et al. Excess mortality attributable to hip& fracture: a relative survival analysis. Bone 2013; 56:23–29. The Dubbo Osteoporosis Epidemiology Study is a prospective epidemiologic study of osteoporosis and fractures in more than 2000 men and women aged 60 and above, started in 1989 with 21 years of follow-up. This study is well positioned to examine long-term mortality after hip fracture and highlights progressive declines in relative survival rates at 1, 5, and 10 years. On average, posthip fracture women died 4 years earlier [median: 4.1, interquartile range (IQR) 1.7– 7.8] and men died 5 years earlier (median ¼ 4.8, IQR 2.4–7.0) than expected. 6. Ioannidis G, Flahive J, Pickard L, et al. Nonhip, nonspine fractures drive healthcare utilization following a fracture: the Global Longitudinal Study of Osteoporosis in Women (GLOW). Osteoporos Int 2013; 24:59–67. 7. Leslie WD, Lix LM, Finlayson GS, et al. Direct healthcare costs for 5 years postfracture in Canada: a long-term population-based assessment. Osteoporos Int 2013; 24:1697–1705. 8. Cooper C, Cole ZA, Holroyd CR, et al. Secular trends in the incidence of hip and other osteoporotic fractures. Osteoporos Int 2011; 22:1277–1288. 9. Looker AC, Melton LJ III, Borrud LG, Shepherd JA. Changes in femur neck bone density in US adults between 1988–1994 and 2005–2008: demographic patterns and possible determinants. Osteoporos Int 2012; 23:771–780. 10. Leslie WD, Lix LM, Yogendran MS, et al. Temporal trends in obesity, && osteoporosis treatment, bone mineral density and fracture rates: a population-based historical cohort study. J Bone Miner Res 2014; 29:952–959. The National Health and Nutrition Examination Survey (NHANES) III femoral neck BMD database in young non-Hispanic white women is the international reference standard for T-score calculation, diagnosis of osteoporosis and fracture probability estimation under FRAX. NHANES reported that femoral BMD increased by approximately 6% in non-Hispanic white women between 1988–1994 and 2005–2008. Two complementary population-based historical cohorts from the Province of Manitoba, Canada (1996–2006) confirmed declining major osteoporotic fractures rates and increasing BMD. Surprisingly, temporal increases in BMI, obesity, and osteoporosis treatment did not account for these changes. Improvements in BMD (still unexplained) were found to explain the decline in major osteoporotic fracture rates from 1996 to 2006. One corollary of this study is to support the continued validity and use of NHANES III (1988–2004) as the international reference standard for T-score calculation, diagnosis of osteoporosis, and fracture probability estimation under FRAX. 11. Rubin KH, Abrahamsen B, Friis-Holmberg T, et al. Comparison of different screening tools (FRAX(R), OST, ORAI, OSIRIS, SCORE and age alone) to identify women with increased risk of fracture: a population-based prospective study. Bone 2013; 56:16–22. 12. Rubin KH, Friis-Holmberg T, Hermann AP, et al. Risk assessment tools to & identify women with increased risk of osteoporotic fracture: complexity or simplicity? A systematic review. J Bone Miner Res 2013; 28:1701–1717. This important, systematic review of fracture risk assessment tools examined the PubMed, Embase, and Cochrane databases. Of 48 tools identified, only 6 had been externally validated and tested more than once in a population-based setting with acceptable methodological quality. None of the tools performed consistently better than the others; simple tools often did as well or better than more complex tools. The authors highlight the need for high-quality studies in population-based cohorts with different case mixes to determine the effectiveness of these tools in selecting patients for therapy. 13. Lam A, Leslie W, Lix L, et al. Major osteoporotic to hip fracture ratios in Canadian men and women with Swedish comparisons: a population based analysis. J Bone Miner Res 2013. [Epub ahead of print] 14. Cummings SR, Cosman F, Eastell R, et al. Goal-directed treatment of && osteoporosis. J Bone Miner Res 2013; 28:433–438. This study proposes a bold new way of looking at the initiation and monitoring of osteoporosis treatment. The idea is to follow the example of other conditions, such as hypertension, in which treatment is based on achieving a goal. Although there are many obstacles in setting treatment goals, the result could be more rational and effective use of the expanding array of treatments for osteoporosis.
1040-8711 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-rheumatology.com
445
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Metabolic bone disease 15. Leslie WD, Majumdar SR, Lix LM, et al. Can change in FRAX Score be used to ‘Treat-to-Target’? A population-based cohort study. J Bone Miner Res 2013. [Epub ahead of print] This study in 11 049 previously untreated women aged more than 50 years undergoing baseline and follow-up DXA examinations and FRAX probability calculations provided a clear answer to the question – ‘No’. FRAX scores increased over time and this increase was attenuated but not prevented by the treatment. Few women had meaningful reductions in FRAX scores. FRAX with BMD is not responsive enough to be used as a target for goal-directed treatment. 16. Berry SD, Samelson EJ, Pencina MJ, et al. Repeat bone mineral density & screening and prediction of hip and major osteoporotic fracture. JAMA 2013; 310:1256–1262. The role of BMD monitoring has been an area of unresolved – and often times heated – controversy. In untreated individuals, there have been conflicting data on whether BMD change can be used to refine fracture risk beyond the baseline measurement. This study found that a second BMD measurement after 4 years in untreated men and women of mean age 75 years did not meaningfully improve the prediction of hip or major osteoporotic fracture. 17. Shane E, Burr D, Abrahamsen B, et al. Atypical subtrochanteric and diaphy&& seal femoral fractures: second report of a task force of the American Society for Bone and Mineral Research. J Bone Miner Res 2014; 29:1–23. Atypical femur fractures (AFFs), initially reported in patients taking bisphosphonates and more recently in patients on denosumab, have generated fear from patients, healthcare providers, and regulators out of proportion to their infrequency. The original ASBMR Task Force report in 2010 has been updated to reflect emerging data on the epidemiology, pathogenesis, and medical management of AFFs. The case definition has also been modified: periosteal stress reaction at the fracture site was changed from a minor to a major feature. 18. Dell RM, Adams AL, Greene DF, et al. Incidence of atypical nontraumatic diaphyseal fractures of the femur. J Bone Miner Res 2012; 27:2544–2550. 19. McKenna MJ, van der Kamp S, Heffernan E, Hurson C. Incomplete atypical femoral fractures: assessing the diagnostic utility of DXA by extending femur length. J Clin Densitom 2013; 16:579–583. 20. Kanis JA, McCloskey EV, Johansson H, et al. A reference standard for the description of osteoporosis. Bone 2008; 42:467–475. 21. Schousboe JT, Shepherd JA, Bilezikian JP, Baim S. Executive summary of the && 2013 International Society for Clinical Densitometry Position Development Conference on bone densitometry. J Clin Densitom 2013; 16:455–466. The International Society for Clinical Densitometry (ISCD) convenes a Position Development Conference (PDC) every 2–3 years to make recommendations for guidelines and standards in the field of musculoskeletal measurement and assessment. This report summarizes the results of the 2013 ISCD PDC for vertebral fracture assessment/DXA and National Health and Nutrition Survey (NHANES) Reference Database Task Forces, and a separate article summarized the results of the Body Composition Analysis Task Forces. Of particular importance are the recommendations on Normative Databases for Bone Mineral Density T-Scores. 22. Watts NB, Leslie WD, Foldes AJ, Miller PD. 2013 International Society for & Clinical Densitometry Position Development Conference: Task Force on Normative Databases. J Clin Densitom 2013; 16:472–481. This report from the ISCD Task Force on Normative Databases provided the evidence base for the Position Development Conference (PDC) and new position statements: manufacturers should continue to use their own databases for the lumbar spine as the reference standard for T-scores; manufacturers should continue to use National Health and Nutrition Examination Survey III data as the reference standard for femoral neck and total hip T-scores; if local reference data are available, they should be used to calculate only Z-scores but not T-scores; and a uniform Caucasian (nonrace adjusted) female reference database should be used to calculate T-scores for men of all ethnic groups. &&
446
www.co-rheumatology.com
23. Karlsson MK, Ribom EL, Nilsson JA, et al. International and ethnic variability of falls in older men. Scand J Public Health 2014; 42:194–200. 24. Cauley JA, Harrison SL, Cawthon PM, et al. Objective measures of physical activity, fractures and falls: the osteoporotic fractures in men study. J Am Geriatr Soc 2013; 61:1080–1088. 25. Amigues I, Schott AM, Amine M, et al. Low skeletal muscle mass and risk of functional decline in elderly community-dwelling women: the prospective EPIDOS study. J Am Med Dir Assoc 2013; 14:352–357. 26. Compston J. Obesity and bone. Curr Osteoporos Rep 2013; 11:30–35. &
This review examines the recent studies showing that fractures in obese postmenopausal women and older men contribute significantly to the overall fracture burden. 27. Johansson H, Kanis JA, Oden A, et al. A meta-analysis of the association of && fracture risk and body mass index in women. J Bone Miner Res 2014; 29:223–233. This large-scale meta-analysis is the definitive word to date on the association between BMI (as an index of underweight and obesity) and its effect on fracture risk in postmenopausal women. At a population level, high BMI was a protective factor for most sites of fragility fracture, but the association is complex, differs across skeletal sites, and is modified by the interaction between BMI and BMD. Osteoporotic fracture risk was lower for a BMI of 35 kg/m2 compared with 25 kg/m2 [hazard ratio 0.87 with 95% confidence interval (CI) 0.85–0.90]. When adjusted for bone mineral density (BMD), however, the same comparison showed that the hazard ratio for osteoporotic fracture was increased (hazard ratio 1.16; 95% CI 1.09–1.23). Thus, higher BMD values seen in obesity may not exert the level of protection expected. 28. Premaor M, Parker RA, Cummings S, et al. Predictive value of FRAX for fracture in obese older women. J Bone Miner Res 2013; 28:188–195. 29. Compston JE, Flahive J, Hosmer DW, et al. Relationship of weight, height, and body mass index with fracture risk at different sites in postmenopausal women: the Global Longitudinal Study of Osteoporosis in Women (GLOW). J Bone Miner Res 2014; 29:487–493. 30. Compston JE, Flahive J, Hooven FH, et al. Obesity, health-care utilization, and health-related quality of life after fracture in postmenopausal women: Global Longitudinal Study of Osteoporosis in Women (GLOW). Calcif Tissue Int 2014; 94:223–231. 31. Kim BJ, Ahn SH, Bae SJ, et al. Association between metabolic syndrome and bone loss at various skeletal sites in postmenopausal women: a 3-year retrospective longitudinal study. Osteoporos Int 2013; 24:2243–2252. 32. Kim T, Park S, Pak YS, et al. Association between metabolic syndrome and bone mineral density in Korea: the Fourth Korea National Health and Nutrition Examination Survey (KNHANES IV), 2008. J Bone Miner Metab 2013; 31:652–662. 33. Kim YH, Cho KH, Choi YS, et al. Low bone mineral density is associated with metabolic syndrome in South Korean men but not in women: the 2008–2010 Korean National Health and Nutrition Examination Survey. Arch Osteoporos 2013; 8:142. 34. Lee NK, Sowa H, Hinoi E, et al. Endocrine regulation of energy metabolism by the skeleton. Cell 2007; 130:456–469. 35. Schwartz AV, Schafer AL, Grey A, et al. Effects of antiresorptive therapies on && glucose metabolism: results from the FIT, HORIZON-PFT, and FREEDOM trials. J Bone Miner Res 2013; 28:1348–1354. This is a good news story for osteoporosis. Evidence in animal models suggests that the skeleton controls energy homeostasis through the feedback action of (undercarboxylated) osteocalcin and adiponectin, an insulin-sensitizing adipokine. Contrary to the predictions from mouse models, pharmacological reductions in bone turnover do not appear to increase the risk of insulin resistance and diabetes in humans.
Volume 26 Number 4 July 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
