Key Symposium. M. Lorentzon 1 & S. R. Cummings 2

Transcription

1 Key Symposium Click here for more articles from the symposium doi: /joim Watch Dr. Lorentzon and Dr. Cummings talk about the 11th Key Symposium: Osteoporosis: the evolution of a diagnosis here. Osteoporosis: the evolution of a diagnosis M. Lorentzon 1 & S. R. Cummings 2 From the 1 Geriatric Medicine, Institute of Medicine, Centre for Bone and Arthritis Research, Sahlgrenska Academy, M olndal, Sweden, and 2 University of California, the San Francisco Coordinating Center, California Pacific Medical Center Research Institute, San Francisco, CA, USA Abstract. Lorentzon M, Cummings SR (Institute of Medicine, M olndal, Sweden; and University of California, San Francisco, CA, USA). Osteoporosis: the evolution of a diagnosis (Key Symposium). J Intern Med 2015;277: The global trend towards increased longevity has resulted in ageing populations and a rise in diseases or conditions that primarily affect older persons. One such condition is osteoporosis (fragile or porous bones), which causes an increased fracture risk. Vertebral and hip fractures lead to increased morbidity and mortality and result in enormous healthcare costs. Here, we review the evolution of the diagnosis of osteoporosis. In an attempt to separate patients with normal bones from those with osteoporosis and to define the osteoporosis diagnosis, multiple factors and characteristics have been considered. These include pathology and histology of the disease, the endocrine regulation of bone metabolism, bone mineral density (BMD), fracture type or trauma severity, risk models for fracture prediction, and thresholds for pharmacological intervention. The femoral neck BMD 2.5 SDs cut-off for the diagnosis of osteoporosis is arbitrarily chosen, and there is no evidence to support the notion that fracture location (except vertebral fractures) or severity is useful to discriminate osteoporotic from normal bones. Fracture risk models (including factors unrelated to bone) dissociate bone strength from the diagnosis, and treatment thresholds are often based on health-economic considerations rather than bone properties. Vertebral fractures are a primary feature of osteoporosis, characterized by decreased bone mass, strength and quality, and a high risk of another such fracture that can be considerably reduced by treatment. We believe that the 2001 definition of osteoporosis by the National Institutes of Health Consensus Development Panel on Osteoporosis is still valid and useful: Osteoporosis is defined as a skeletal disorder characterized by compromised bone strength predisposing a person to an increased risk of fracture. Keywords: aetiology, fracture, osteoporosis. Introduction Improved healthcare, socio-economic and lifestyle changes have led to a dramatically increased life expectancy in modern societies during the past century. As a result, the incidence of disorders that primarily affect the elderly has increased considerably. The number of fractures due to fragile or porous bones (i.e. osteoporosis from the Greek terms for bone and pore ) has therefore increased in industrialized countries during the past 50 years, and a similar trend in developing countries is expected [1 4]. Fracture risk increases exponentially with age, and over 70% of all fractures affect women over 65 years old (Fig. 1). After the age of 50, almost one in two women and one in five men will sustain a fragility fracture during their remaining lifetime [3]. Vertebral and hip fractures lead to increased morbidity and mortality and result in enormous healthcare costs [5]. Therefore, the need to define a diagnosis for osteoporosis, a condition that increases the risk of fracture, has gradually emerged. In 2001, the National Institutes of Health Consensus Development Panel on Osteoporosis issued a consensus definition of osteoporosis: Osteoporosis is defined as a skeletal disorder characterized by compromised bone strength predisposing a person to an increased risk of fracture. Bone strength was considered to be primarily due to bone density and quality [6]. A deterioration of trabecular microstructure with loss of connectivity between trabeculae and cortical thinning is typical trait of the disease (Fig. 2). Here, we describe the evolution of the diagnosis of osteoporosis. Approaches to define the diagnosis of osteoporosis In an attempt to separate patients with normal bones from those with osteoporosis and to define 650 ª 2015 The Association for the Publication of the Journal of Internal Medicine

2 Rate per per year 400 Men Hip Radiographic vertebral Wrist Women The early history of osteoporosis as a clinical syndrome In 1822, Sir Astley Paston Cooper, a British surgeon and anatomist, commented on an observed association between abnormal bones and fractures. The French pathologist and surgeon Jean Lobstein was the first to use the term osteoporosis in 1835, although in the context of describing a condition with blue-grey sclera, which was probably osteogenesis imperfecta type I. In 1941, Fuller Albright reported the cases of women with vertebral fractures after loss of ovarian function. Oestrogen treatment improved the calcium balance in these women and arrested their height loss [7]. These findings were the basis for defining postmenopausal osteoporosis and established the link between osteoporosis and vertebral fractures. Several subsequent studies have demonstrated that vertebral fractures represent impaired bone quality and structural decay of the bone. Vertebral fractures reflect the severity of osteoporosis and are strong predictors of future fractures, thus serving as the hallmark of the disease [8, 9] >85 Age (years) Fig. 1 Fracture incidence by age and gender. From Ref. [3]. the osteoporosis diagnosis, multiple factors have been considered: The clinical investigation of patients; The endocrine regulation of bone metabolism and bone mass; Bone mineral density (BMD); The pathology and histology of the disease; The characteristics of fractures; Risk models for fracture prediction; and Thresholds for the initiation of pharmacological treatment. The clinical investigation of patients Osteoporosis is asymptomatic and patients usually first come to clinical attention after sustaining a fragility fracture, the primary manifestation of the disease [10]. Reduced bone strength is a disease characteristic that often leads to vertebral compression or other types of fragility fractures, usually as a result of low-energy trauma such as falling from standing height or less. In women with low BMD, clinical fragility fractures of the vertebra, humerus or forearm result in substantial disability [11]. Hip fractures lead to the most serious outcomes, and more than 50% of patients do not regain their physical function following this type of fracture [12, 13]. Excess mortality varies between studies, depending on the investigated population, ranging from 8% to 36% within the first year following hip fracture [14]. Thoracic kyphosis, back pain and limited physical function are common manifestations but not always present in patients with vertebral fractures [15]. When investigating patients with suspected osteoporosis, signs and symptoms due to any one of the many disorders that cause secondary osteoporosis should be sought. Examples of such conditions include malabsorption (e.g. due to coeliac or inflammatory bowel disease), hyperthyroidism, hyperparathyroidism, Cushing s disease, hypogonadism, ª 2015 The Association for the Publication of the Journal of Internal Medicine 651

3 Fig. 2 A compressed osteoporotic vertebra (right), with reduced trabecular number, connectivity and density. From Robbins Basic Pathology 9th ed. rheumatoid arthritis, alcoholism and chronic obstructive pulmonary disease [16]. The endocrine regulation of bone metabolism and bone mass In 1983, Riggs and Melton proposed that the osteoporosis diagnosis should be subdivided based on the proposed pathogenesis: Type 1 osteoporosis: due to decreased oestradiol levels and loss of trabecular bone (the major component of the vertebrae) following the first 3 5 years after the menopause, leading to increased risk of vertebral fractures; and Type 2 osteoporosis: a form of senile osteoporosis primarily due to advanced age with impaired calcium handling, resulting in reduced levels of circulating 1,25-OH-vitamin D, lower calcium absorption, and secondary hyperparathyroidism, coinciding with an increased risk of hip fracture [17]. Although this proposition was reasonable, based on the evidence present at that time, the results from later epidemiological and experimental studies did not support this division of the diagnosis for several reasons. First, the risk of vertebral fracture in women is not greatest in the first years following the menopause, but instead rises gradually with increasing age and is highest after the age of 75 years [18]. Secondly, low circulating levels of oestradiol were later shown to similarly predict both spine and hip fractures, whilst the levels of 25-OH-vitamin D and parathyroid hormone (PTH) were not associated with hip or spine fractures [19]. Finally, evidence has emerged from clinical studies in which bone characteristics have been measured using computed tomography (CT) to suggest that most accelerated trabecular bone loss begins in young adult life, when oestradiol levels are preserved, and that cortical bone loss accelerates following the menopause, indicating that oestradiol levels are primarily important for cortical and not trabecular bone loss [20]. Measuring BMD to define osteoporosis In the late 19th century, measurements of bone opacity using dental X-rays were first described in the journal Dental Cosmos [21]. Cameron and Sorensen significantly advanced the measurement of BMD by introducing the single-photon absorptiometry (SPA) technique in 1963 [22]. The equipment used was small, and the method precise and affordable, but SPA could only be used to measure BMD at peripheral skeletal sites. Dual-energy methods were later introduced so that BMD at axial sites, where the degree of soft tissue varies to a much greater extent, could be assessed. In 1976, Madsen et al. [23] in the USA developed dualphoton absorptiometry (DPA), a method using gamma rays of two different energies to measure BMD. The Swedish investigators Roos and Sk oldborn [24] described the use of DPA to measure BMD at the lumbar spine in With the introduction of dual-energy X-ray absorptiometry (DXA) in the late 1980s, acquisition time was dramatically shortened and the accuracy and precision of BMD measurement improved [25]. The DXA technique allowed rapid measurements, thus exposing the patient to very low radiation doses. Experimental research with human cadavers revealed that measuring areal BMD (g cm 2 ) 652 ª 2015 The Association for the Publication of the Journal of Internal Medicine

4 with DXA of the femur served as a good proxy for actual bone strength at this site, explaining about 70% of the variation [26], whereas BMD of the spine represented the bone strength of the vertebrae to a lesser extent (explaining 44% of the variation) [27]. The technique has now become the clinical gold standard to measure BMD, and a large number of studies have demonstrated the ability of DXA-derived BMD of the hip and spine to predict fractures. Bone mineral density measurement at the hip was found to predict hip fracture with an age-adjusted risk increase of 2.6 per SD decrease in BMD [28, 29]. The relationship between fracture risk and BMD is continuous with no apparent threshold for defining patients at high risk (Fig. 3). Thus, the relative risk of fracture increases gradually with decreasing BMD. Furthermore, BMD is a trait that displays a Gaussian distribution in the population, with no clear subgroup with extremely low values. As BMD decreases progressively with ageing, the proportion of patients with osteoporosis increases exponentially [30]. Therefore, it has been difficult to establish the level of BMD that should be considered normal. In an attempt to identify a suitable BMD cut-off level to define the osteoporosis diagnosis, a committee convened in Geneva in A consensus was reached; it was stated that osteoporosis was present if the patient s femoral neck BMD was 2.5 SDs below the mean of a young and healthy population (T-score), matched for gender and ethnic group. Using this definition, approximately 15% of white postmenopausal women had osteoporosis, which was similar to the 16% lifetime risk (in the USA) for this group to sustain a hip fracture, although these two facts were not related [31 33]. Low BMD, or osteopenia, was defined as a BMD value more than 1 SD but <2.5 SDs below that of a young healthy reference population (Fig. 4). A BMD T-score of at least 2.5 SDs or below in combination with a prevalent fracture was considered severe or established osteoporosis. The consensus meeting resulted in the 1994 World Health Organization (WHO) guidelines that today still define the osteoporosis diagnosis [34]. Applying these diagnostic criteria, using only hip bone density, resulted in labelling 13 18% and 37 50% of women over 50 years old in the USA as osteoporotic and osteopenic, respectively [35]. If a cut-off value of at least 2.5 SDs (using the same young healthy reference group) was used for the posterior anterior spine and for the bone mineral content of the mid-radius, in addition to the proximal femur, 30% of all postmenopausal women would be defined as osteoporotic [34]. Based on spine and hip BMD, the US National Osteoporosis Foundation (NOF) estimated that in 2014, there were nearly 10 million Americans with osteoporosis and over 43 million with osteopenia [36]. Some committees have proposed that BMD measured at other sites, such as the radius, should also be used to diagnose the disease, leading to even more patients being labelled osteoporotic. A diagnosis of osteoporosis can cause negative psychological effects. A majority of women (55%) remained worried about osteoporosis 12 months after a bone BMD (g cm 2 ) 1.24 Normal Osteopenia T above 1.0 Normal T 1.0 to 2.5 Osteopenia T 2.5 Osteoporosis X YAT-score 2 1 You have osteopenia You have osteoporosis X 4 Osteoporosis Age (years) Fig. 3 The relationship between bone density and fracture risk. Fig. 4 Definition of the osteoporosis diagnosis with examples from a bone densitometry measurement report. ª 2015 The Association for the Publication of the Journal of Internal Medicine 653

5 examination and 22% stopped an activity because of fear of falling, as a result of being diagnosed with the disease [37]. Furthermore, it has now become evident that including other risk factors and considering type of prevalent fracture (especially vertebral fractures) is in most cases necessary, in addition to diagnosing low BMD, to assess fracture risk and the severity of osteoporosis. It should be noted that defining osteoporosis using DXA has several limitations, including the confounding effects on the results of surrounding soft tissue, bone artefacts caused by osteoarthritis, degenerated discs, aortic calcification and vertebral compression fractures. In addition, DXA cannot be used to measure bone microstructure or quality, which are believed to influence fracture risk [25]. Because DXA is a two-dimensional technique, it also cannot be used to measure true volumetric BMD and bone size, or to separate trabecular from cortical bone. Thus, a large bone will yield a higher areal BMD than a small bone, although the volumetric BMD would be the same [38]. Can histopathology be used to define osteoporosis? Osteoporosis can be the result of both impaired development of peak bone mass early in life and later augmented bone loss [39, 40]. The structural integrity of the bone is maintained in adulthood by constant remodelling, a process in which osteoclasts resorb old or damaged bone which is replaced with new bone by osteoblasts [41]. With ageing, this process becomes imbalanced and the resorption exceeds the formation, leading to a net decrease in bone mass. In recent years, it has become clear that the receptor activator of NF-kB ligand (RANKL) pathway is the most important system for regulating bone resorption [42]. The 1994 WHO BMD-based definition of osteoporosis did not address bone microstructure or quality, which refers to bone microstructure, mineralization, turnover, and accumulation of damage, due to microfractures [6, 34]. With increasing age, cortical bone loss results in enlargement of the marrow cavity and cortical thinning as well as reduced volumetric BMD, at least partly due to increased cortical porosity [43]. Loss of trabecular bone results in an impaired trabecular structure, with reduced trabecular thickness and connectivity [44]. Altogether, these changes lead to reduced bone strength that can only partly be accounted for by BMD (as measured using DXA). Although large prospective studies are lacking, it has been reported that indices of trabecular structure and cortical traits, including thickness and volumetric BMD, are better than DXA-derived bone traits in separating men and women with and without fractures [45 48]. Experimental studies indicate that bone strength calculated using finite element analysis (FEA) based on high-resolution CT images is more strongly correlated with actual bone strength than DXA-derived bone traits [49]. In a large study of Icelandic women and men, Kopperdahl et al. [50] found that every SD reduction in FEA-calculated hip bone strength was associated with a more than fourfold increased risk of hip fracture. In a prospective study, Wang et al. investigated the ability of areal BMD measured by DXA and nonlinear FEA-calculated compressive bone strength (based on CT measurements) of the L1 vertebra to predict incident clinical vertebral fractures in elderly men. The investigators found that FEA-calculated bone strength was a significantly stronger predictor of fracture than BMD. Considerably higher hazard ratios (HRs) per SD change, also after adjusting for possible confounding variables, were observed for FEA-determined bone strength than for BMD measured by DXA [adjusted HR 7.2, 95% confidence interval (CI) vs. adjusted HR 3.2, 95% CI ] [51]. The technique of reference point indentation is a recently introduced invasive method to measure actual bone material strength (BMS) in humans [52]. Further development of the technology led to the advent of a hand-held device termed the Osteoprobe that measures the resistance of the mid-tibial bone to microfractures. The Osteoprobe utilizes a metal probe with a sharp conical tip applied with a standardized force of 10 N, and the patient is only exposed to minimal trauma and discomfort [53]. It has been reported that BMS assessed by this method is substantially lower in patients with prevalent hip fractures than in agematched controls [52, 54]. Future prospective studies are needed to determine whether BMS can improve fracture prediction in addition to DXA-derived BMD, CT-assessed bone structure, FEA of bone strength and clinical risk factors. These are very promising technologies for assessing the strength of bone and risk of fracture. Were they widely clinically available, they could describe the characteristics of bone on which a diagnosis of osteoporotic bone could be made. However, no cutoff values for strength of bone or risk of fracture have been proposed to define osteoporosis. 654 ª 2015 The Association for the Publication of the Journal of Internal Medicine

6 Table 1 Almost all fractures, not just major osteoporotic, are related to low BMD Relative risk/sd decrease in BMD of the Type Hip Radius Patella Major osteoporotic fractures Hip Clavicle Pelvis Humerus Spine Lower leg Wrist Hand Rib Ankle Fingers Toes Face ( NS ) BMD, bone mineral density. From Ref. [55]. Bold values indicates major osteoporotic fractures. Assuming that there is a continuous increase in the risk of fracture with a decrease in estimated strength, a diagnosis of osteoporosis based on a measure of bone structure would be affected by the same issues as the use of BMD to define a disease. Defining the osteoporosis diagnosis by fracture type Are there osteoporotic fractures? Historically, fractures of the hip, spine, humerus and forearm have been referred to as osteoporotic, but this terminology has been found to lack supporting evidence. The role of DXA-derived BMD in predicting different fracture types was investigated in the San Francisco Study of Osteoporotic Fractures (SOF), a 9- to 10-year prospective investigation of 9704 women aged 65 years, with 4172 fractures during follow-up. The investigators found that almost all fractures were related to low BMD [55]. Also, fracture types that are not traditionally considered osteoporotic, such as fractures of the patella, clavicle, pelvis and lower leg, were more strongly related than wrist fracture to femoral neck BMD (Table 1), contradicting the hypothesis that osteoporotic fractures are determined by fracture site. Interestingly, further analyses revealed that the proportion of fractures attributable to osteoporosis was rather low for all investigated fractures. Fractures of the clavicle (44%), spine (39%), patella (33%), humerus (31%), lower leg (30%) and hip (28%) were most attributable to osteoporosis measured at the spine or hip. Are only low-energy fractures due to osteoporosis? Another approach to define osteoporotic fractures has been to subdivide fractures by degree of trauma, assuming that low-trauma in contrast to high-trauma fractures (due to motor vehicle accidents or falls from above standing height) are primarily osteoporotic. This hypothesis was tested using data from the SOF cohort and the Osteoporotic Fractures in Men Study (MrOS) in the USA. The MrOS followed 5995 older men for 5.1 years. Nonspine fractures were recorded using radiographic reports and classified as high or low trauma. During follow-up in the two studies, 264 women and 94 men suffered high-energy fractures and 3211 and 346, respectively, sustained lowenergy fractures. As seen with low-trauma fractures, those caused by a high level of trauma were similarly associated with low BMD, indicating that trauma type does not discriminate osteoporotic from nonosteoporotic fractures (Table 2) [56]. Why vertebral fractures are unique Vertebral fractures are common, with a prevalence rate as high as 25 50% [5, 57] in women above 50 years of age, and lead to back pain and reduced physical function [15] as well as an increased risk of mortality [58, 59]. Most of these fractures occur after little or no trauma, and only about a third are clinically recognized [60]. The predictive value of a previous vertebral fracture is much stronger than that of low BMD or other prevalent fractures [8, 9]. It has been reported that spine BMD predicts longterm risk of incident vertebral fractures in women [61]. Vertebral fractures are associated with reduced bone strength [50] and deteriorated bone microstructure [62 64]. Thus, vertebral fractures are different from other types of fracture due to their substantial prognostic ability to predict additional fractures and their representativeness of deteriorated bone structure and bone strength. Therefore, they constitute a disease characteristic of osteoporosis. The response to pharmacological interventions to treat osteoporosis is considerably greater in preventing vertebral (50 70% relative ª 2015 The Association for the Publication of the Journal of Internal Medicine 655

7 Table 2 Relative hazard of high- and low-trauma fracture per 1 SD decrease in BMD at the total hip, femoral neck and lumbar spine for SOF women and MrOS men RH (95% CI) Women BMD location High-trauma fracture Low-trauma fracture High-trauma fracture Low-trauma fracture Total hip Age-adjusted a 1.47 ( ) 1.50 ( ) 1.58 ( ) 1.67 ( ) Multivariate-adjusted b 1.45 ( ) 1.49 ( ) 1.54 ( ) 1.69 ( ) Femoral neck Age-adjusted a 1.33 ( ) 1.49 ( ) 1.47 ( ) 1.65 ( ) Multivariate-adjusted b 1.23 ( ) 1.46 ( ) 1.41 ( ) 1.66 ( ) Lumbar spine Age-adjusted a 1.39 ( ) 1.38 ( ) 1.54 ( ) 1.47 ( ) Multivariate-adjusted b 1.30 ( ) 1.35 ( ) 1.47 ( ) 1.41 ( ) BMD, bone mineral density; CI, confidence interval; MrOS, Osteoporotic Fractures in Men Study; RH, relative hazard; SOF, Study of Osteoporotic Fractures. From Ref. [56]. a Per 5-year increase. b Multivariate-adjusted models include age, weight, height, walking as a form of exercise (yes/no), smoking history (ever versus never), fracture since age 50 years (yes/no), fall in the past year (yes/no), one or more medical conditions (yes/no), self-rated health (excellent/good versus fair/poor/very poor), ability to rise from chair five times without using one s arms (yes/no), current hormone therapy (yes/no, for SOF women only) and current bisphosphonate use (yes/no, for MrOS men only). Men Risk factors besides BMD (partial list) BMI Height at 25 Maternal hip fracture Smoking Needs arms to rise up from a chair Poor depth perception Digit symbol test (cognitive test) Parkinson s Hip Fx rate (per 1000 woman-years) Lowest third Heel BMD Middle third Highest third >=5 # Risk factors Fig. 5 Risk of hip fracture according to number of clinical risk factors and agespecific bone mineral density (BMD). BMI, body mass index. Adapted from Ref. [67]. risk reduction [65, 66]) than other fractures, which further emphasizes the unique role of vertebral fractures in osteoporosis and its treatment. Using fracture risk and treatment thresholds to define osteoporosis Several risk factors other than decreased bone strength, including older age, maternal hip fracture, oral glucocorticoid use, previous fracture, rheumatoid arthritis, Parkinson s disease, high at age 25 and smoking, predict fracture risk independently of BMD (Fig. 5) [67, 68]. The WHO fracture prediction tool FRAX ( was created using 12 prospective population-based cohorts with data regarding clinical risk factors, hip BMD (available 656 ª 2015 The Association for the Publication of the Journal of Internal Medicine

8 Table 3 Clinical risk factors for major osteoporotic (spine, hip, forearm or shoulder) and hip fractures using the FRAX calculator Risk factor Age Sex Height Weight Previous fracture Parental hip fracture Current smoking Oral glucocorticoid use Rheumatoid arthritis Secondary osteoporosis Alcohol use Femoral neck BMD Comment Only applicable for persons aged years Low-trauma fracture in adults Dose-dependent effect which is not taken into account by the calculator Ever use of corticosteroids for 3 months with an equivalent of 5 mg prednisolone daily. Dose-dependent effect which is not taken into account by the calculator A confirmed diagnosis of rheumatoid arthritis If the patient has type 1 diabetes mellitus, osteogenesis imperfecta, untreated long-term hyperthyroidism, chronic malnutrition, hypogonadism or premature menopause (<45 years), malabsorption and chronic liver disease 3 or more units per day. Dose-dependent effect which is not taken into account by the calculator BMD, bone mineral density. World Health Organization (WHO) scientific group on the assessment of osteoporosis at primary health care level. Paper presented at: Summary Meeting Report; May 5, 2007; Brussels, Belgium. in 75% of cohorts) and incident fractures. In total, more than 5000 fractures were observed in women and men during a total follow-up of person-years. Using the identified risk factors (Table 3), a model was created to allow prediction of the absolute risk of major osteoporotic (hip, spine, forearm and humerus) fracture, and the risk of hip fracture within 10 years [69]. The FRAX calculator incorporates several clinical risk factors, is freely available online and is country specific (taking into account the geographic variations in absolute fracture risk). It may be suggested that osteoporosis should be defined as a high fracture risk. However, a high risk may be due to many factors unrelated to bone. This would create a conundrum: a person with a certain bone density, strength and structure, but high risk due to age, residence in a high-risk region of the world or high risk of falling would be described as having osteoporosis, whilst a younger person with the same bone density, strength and structure in a low-risk region and at low risk of falls would not be diagnosed with the disease. Furthermore, as with BMD, the relationship between the estimates produced by models such as FRAX and risk of fracture is continuous with no inherent cutoff point for defining osteoporosis. Cut-off levels of risk for recommending drug treatment have been developed and might be considered to define osteoporosis. For example, the US NOF [36], the UK National Institute for Health and Care Excellence [70] and the Swedish National Board of Health and Welfare recommend the use of the FRAX tool in ascertaining who to test and who to treat for osteoporosis [71] and propose thresholds of risk above which patients should receive drug therapy to reduce fracture risk. Frequently in clinical practice, a diagnosis is accompanied by a treatment. Therefore, it could be argued that the osteoporosis diagnosis should be defined as a medical indication to initiate pharmacological treatment to prevent fractures. With a few exceptions, all randomized controlled trials to prevent fractures with medication for osteoporosis, such as alendronate [65, 72], zoledronic acid [66] or denosumab [73], have included women on the basis of a low BMD measured by DXA. Treatment with the bisphosphonate alendronate was reported to only reduce the risk of clinical fracture in postmenopausal women with osteoporosis and not in their osteopenic counterparts [74], indicating that antiresorptive agents are more potent if BMD is severely reduced. Similarly, subgroup analyses from the Freedom trial revealed ª 2015 The Association for the Publication of the Journal of Internal Medicine 657

9 that reduction in hip fracture risk with denosumab treatment was greater in women with osteoporosis at the hip than in those with osteopenia [75]. Thus, available evidence suggests that antiresorptive treatment is effective only if BMD is low, arguing that a low BMD is indicative of successful treatment and should therefore define the osteoporosis diagnosis. However, this argument is complicated by the low absolute fracture risk seen in early postmenopausal women with osteoporosis, but without other risk factors for fracture. For example, the number needed to treat with alendronate to prevent a vertebral fracture ranges from 20 in a high-risk population (with a 5% absolute risk reduction) to 200 in a low-risk population [65]. Treating large numbers of women despite a low observed fracture risk will likely induce many both common and uncommon side effects for each avoided fracture, leading to an unfavourable risk benefit balance, most importantly in terms of general health but also societal costs. Treatment thresholds are derived from cost-effectiveness analyses. For example, Tosteson et al. performed cost-effectiveness analyses for alendronate treatment in 2008, assuming 100% persistence to treatment, a 35% reduction in major fractures over 5 years and a willingness to spend $ per quality-adjusted life year saved. In this scenario, a 3% 10-year risk of hip fracture would be cost-effective [76]. Since then, the cost of generic alendronate has dropped from $600 per year to below $60 per year at present, which would substantially lower the 10-year probability for hip fracture treatment threshold to very low levels. If a diagnosis of osteoporosis was based on a treatment threshold, then the diagnosis would depend more on the cost of a drug than the properties of a bone or patient. The National Osteoporosis Guidance Group (NOGG) [77] in the UK recently presented treatment guidelines, in which treatment decisions are recommended based on previous fracture and fracture risk determined by FRAX, without BMD measurements. In this rational approach, the assessment of risk and treatment decisions are entirely dissociated from a diagnosis of osteoporosis or measurements of bone characteristics. diagnosis of osteoporosis should be based on the following: (i) occurrence of a vertebral fracture or low-trauma fracture of the proximal femur, humerus or pelvis; (ii) some types of wrist fractures in women with a T-score at any site between 1 and 2.5 (osteopenia); (iii) a T-score of less than 2.5 at the hip or spine; or (iv) a 10-year fracture risk at or above the treatment thresholds of 3% for hip fracture or 20% for major osteoporotic fractures determined by FRAX. None of the elements of this proposed definition except vertebral fracture is supported by the evidence reviewed above. There is evidence to support the following proposals that all fractures, not only low-trauma events, are related to bone strength and future risk, that fractures at other skeletal sites than those traditionally proposed are strongly related to BMD and future risk, that the T-score cut-off is arbitrary albeit firmly embedded in tradition and entry criteria for clinical trials, and that suggested (or any) treatment thresholds reflect health economics, rather than characteristics of the bone or patient, and would be specific to US practice guidelines. Conclusions There is no perfect definition of osteoporosis. However, if the term is used in clinical practice and research, it seems reasonable that it should describe properties of bone weakness that are exhibited by fractures resulting from decreased bone strength or measurements that reflect weak bones. Vertebral fractures remain a hallmark of osteoporosis reflecting decreased mass, strength and quality, and a very high risk of another such fracture that is substantially reduced by treatment. However, perhaps the best definition remains the consensus reached in 2001 by the NIH Consensus Development Panel on Osteoporosis. Conflict of interest statement SC has received consultation fees and honoraria from Lilly, Amgen and Merck. ML has received honoraria for lecturing from Lilly, Amgen, Novartis, Meda, Hologic and GE Lunar. The combination of these approaches A working group of the US National Bone Health Alliance [78] recently proposed that a clinical References 1 Gullberg B, Johnell O, Kanis JA. World-wide projections for hip fracture. Osteoporos Int 1997; 7: ª 2015 The Association for the Publication of the Journal of Internal Medicine

10 2 Ensrud KE. Epidemiology of fracture risk with advancing age. J Gerontol A Biol Sci Med Sci 2013; 68: Sambrook P, Cooper C. Osteoporosis. Lancet 2006; 367: Dodds MK, Codd MB, Looney A, Mulhall KJ. Incidence of hip fracture in the Republic of Ireland and future projections: a population-based study. Osteoporos Int 2009; 20: Cummings SR, Melton LJ. Epidemiology and outcomes of osteoporotic fractures. Lancet 2002; 359: NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA 2001; 285: Albright F, Burnett C, Cope O. Acute atrophy of bone (osteoporosis) simulating hyperparathyroidism. J Clin Endocrinol Metab 1941; 1: Ross PD, Davis JW, Epstein RS, Wasnich RD. Pre-existing fractures and bone mass predict vertebral fracture incidence in women. Ann Intern Med 1991; 114: Black DM, Arden NK, Palermo L, Pearson J, Cummings SR. Prevalent vertebral deformities predict hip fractures and new vertebral deformities but not wrist fractures. Study of Osteoporotic Fractures Research Group. J Bone Miner Res 1999; 14: Glaser DL, Kaplan FS. Osteoporosis. Definition and clinical presentation. Spine 1997; 22: 12S 6S. 11 Fink HA, Ensrud KE, Nelson DB et al. Disability after clinical fracture in postmenopausal women with low bone density: the fracture intervention trial (FIT). Osteoporos Int 2003; 14: Mossey JM, Mutran E, Knott K, Craik R. Determinants of recovery 12 months after hip fracture: the importance of psychosocial factors. Am J Public Health 1989; 79: Marottoli RA, Berkman LF, Cooney LM Jr. Decline in physical function following hip fracture. J Am Geriatr Soc 1992; 40: Sattui SE, Saag KG. Fracture mortality: associations with epidemiology and osteoporosis treatment. Nat Rev Endocrinol 2014; 10: Nevitt MC, Ettinger B, Black DM, et al. The association of radiographically detected vertebral fractures with back pain and function: a prospective study. Ann Intern Med 1998; 128: Hofbauer LC, Hamann C, Ebeling PR. Approach to the patient with secondary osteoporosis. Eur J Endocrinol 2010; 162: Riggs BL, Melton LJ 3rd. Evidence for two distinct syndromes of involutional osteoporosis. Am J Med 1983; 75: Felsenberg D, Silman AJ, Lunt M et al. Incidence of vertebral fracture in europe: results from the European Prospective Osteoporosis Study (EPOS). J Bone Miner Res 2002; 17: Cummings SR, Browner WS, Bauer D et al. Endogenous hormones and the risk of hip and vertebral fractures among older women. Study of Osteoporotic Fractures Research Group. N Engl J Med 1998; 339: Khosla S, Melton LJ 3rd, Riggs BL. The unitary model for estrogen deficiency and the pathogenesis of osteoporosis: is a revision needed? J Bone Miner Res 2011; 26: Dennis J. A new system of measurement in X-ray work. Dental Cosmos 1897; 39: Cameron JR, Sorenson J. Measurement of bone mineral in vivo: an improved method. Science 1963; 142: Madsen M, Peppler W, Mazess RB. Vertebral and total body bone mineral content by dual photon absorptiometry. Calcif Tissue Res 1976; 21(Suppl): Roos BO, Skoldborn H. Dual photon absorptiometry in lumbar vertebrae. I. Theory and method. Acta Radiol Ther Phys Biol 1974; 13: Bonnick SL. Bone Densitometry in Clinical Practice. New York, NY: Humana Press, a part of Springer Science and Business Media, Cheng XG, Lowet G, Boonen S et al. Assessment of the strength of proximal femur in vitro: relationship to femoral bone mineral density and femoral geometry. Bone 1997; 20: Cheng XG, Nicholson PH, Boonen S et al. Prediction of vertebral strength in vitro by spinal bone densitometry and calcaneal ultrasound. J Bone Miner Res 1997; 12: Cummings SR, Black DM, Nevitt MC et al. Bone density at various sites for prediction of hip fractures. The Study of Osteoporotic Fractures Research Group. Lancet 1993; 341: Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 1996; 312: Kanis JA, Johnell O, Oden A, Jonsson B, De Laet C, Dawson A. Risk of hip fracture according to the World Health Organization criteria for osteopenia and osteoporosis. Bone 2000; 27: Melton LJ 3rd, Atkinson EJ, O Fallon WM, Wahner HW, Riggs BL. Long-term fracture prediction by bone mineral assessed at different skeletal sites. J Bone Miner Res 1993; 8: Cummings SR, Black DM, Rubin SM. Lifetime risks of hip, Colles, or vertebral fracture and coronary heart disease among white postmenopausal women. Arch Intern Med 1989; 149: Kanis JA, Johnell O, De Laet C, Jonsson B, Oden A, Ogelsby AK. International variations in hip fracture probabilities: implications for risk assessment. J Bone Miner Res 2002; 17: Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Report of a WHO Study Group. World Health Organ Tech Rep Ser 1994; 843: Looker AC, Orwoll ES, Johnston CC Jr et al. Prevalence of low femoral bone density in older U.S. adults from NHANES III. J Bone Miner Res 1997; 12: National Osteoporosis Foundation. Clinician s Guide to Prevention and Treatment of Osteoporosis. Washinton, DC: National Osteoporosis Foundation, 2014; Rubin SM, Cummings SR. Results of bone densitometry affect women s decisions about taking measures to prevent fractures. Ann Intern Med 1992; 116: Carter D, Bouxsein M, Marcus R. New approaches for interpreting projected bone densitometry data. J Bone Miner Res 1992; 7: Seeman E. Pathogenesis of bone fragility in women and men. Lancet 2002; 359: Tabensky A, Duan Y, Edmonds J, Seeman E. The contribution of reduced peak accrual of bone and age-related bone loss to osteoporosis at the spine and hip: insights from the daughters of women with vertebral or hip fractures. J Bone Miner Res 2001; 16: Raisz LG. Pathogenesis of osteoporosis: concepts, conflicts, and prospects. J Clin Investig 2005; 115: ª 2015 The Association for the Publication of the Journal of Internal Medicine 659

11 42 Rachner TD, Khosla S, Hofbauer LC. Osteoporosis: now and the future. Lancet 2011; 377: Zebaze RM, Ghasem-Zadeh A, Bohte A et al. Intracortical remodelling and porosity in the distal radius and postmortem femurs of women: a cross-sectional study. Lancet 2010; 375: Brandi ML. Microarchitecture, the key to bone quality. Rheumatology (Oxford) 2009; 48(Suppl 4): iv Melton LJ 3rd, Christen D, Riggs BL et al. Assessing forearm fracture risk in postmenopausal women. Osteoporos Int 2010; 21: Szulc P, Boutroy S, Vilayphiou N, Chaitou A, Delmas PD, Chapurlat R. Cross-sectional analysis of the association between fragility fractures and bone microarchitecture in older men: the STRAMBO study. J Bone Miner Res 2011; 26: Rudang R, Darelid A, Nilsson M, Mellstrom D, Ohlsson C, Lorentzon M. X-ray-verified fractures are associated with finite element analysis-derived bone strength and trabecular microstructure in young adult men. J Bone Miner Res 2013; 28: Boutroy S, Van Rietbergen B, Sornay-Rendu E, Munoz F, Bouxsein ML, Delmas PD. Finite element analysis based on in vivo HR-pQCT images of the distal radius is associated with wrist fracture in postmenopausal women. J Bone Miner Res 2008; 23: Engelke K, Libanati C, Fuerst T, Zysset P, Genant HK. Advanced CT based in vivo methods for the assessment of bone density, structure, and strength. Curr Osteoporos Rep 2013; 11: Kopperdahl DL, Aspelund T, Hoffmann PF et al. Assessment of incident spine and hip fractures in women and men using finite element analysis of CT scans. J Bone Miner Res 2014; 29: Wang X, Sanyal A, Cawthon PM et al. Prediction of new clinical vertebral fractures in elderly men using finite element analysis of CT scans. J Bone Miner Res 2012; 27: Diez-Perez A, Guerri R, Nogues X et al. Microindentation for in vivo measurement of bone tissue mechanical properties in humans. J Bone Miner Res 2010; 25: Randall C, Bridges D, Guerri R et al. Applications of a new handheld reference point indentation instrument measuring bone material strength. J Med Devices 2013; 7: Guerri-Fernandez RC, Nogues X, Quesada Gomez JM et al. Microindentation for in vivo measurement of bone tissue material properties in atypical femoral fracture patients and controls. J Bone Miner Res 2013; 28: Stone KL, Seeley DG, Lui LY et al. BMD at multiple sites and risk of fracture of multiple types: long-term results from the Study of Osteoporotic Fractures. J Bone Miner Res 2003; 18: Mackey DC, Lui LY, Cawthon PM et al. High-trauma fractures and low bone mineral density in older women and men. JAMA 2007; 298: Cooper C, O Neill T, Silman A. The epidemiology of vertebral fractures. European Vertebral Osteoporosis Study Group. Bone 1993; 14(Suppl 1): S Hasserius R, Karlsson MK, Nilsson BE, Redlund-Johnell I, Johnell O. Prevalent vertebral deformities predict increased mortality and increased fracture rate in both men and women: a 10-year population-based study of 598 individuals from the Swedish cohort in the European Vertebral Osteoporosis Study. Osteoporos Int 2003; 14: Kado DM, Duong T, Stone KL et al. Incident vertebral fractures and mortality in older women: a prospective study. Osteoporos Int 2003; 14: Cooper C, Atkinson EJ, O Fallon WM, Melton LJ 3rd. Incidence of clinically diagnosed vertebral fractures: a population-based study in Rochester, Minnesota, J Bone Miner Res 1992; 7: Cauley JA, Hochberg MC, Lui LY et al. Long-term risk of incident vertebral fractures. JAMA 2007; 298: Melton LJ 3rd, Riggs BL, Keaveny TM et al. Structural determinants of vertebral fracture risk. J Bone Miner Res 2007; 22: Melton LJ 3rd, Riggs BL, Keaveny TM et al. Relation of vertebral deformities to bone density, structure, and strength. J Bone Miner Res 2010; 25: Ito M, Ikeda K, Nishiguchi M et al. Multi-detector row CT imaging of vertebral microstructure for evaluation of fracture risk. J Bone Miner Res 2005; 20: Wells GA, Cranney A, Peterson J et al. Alendronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database Syst Rev 2008: CD Black DM, Delmas PD, Eastell R et al. Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. N Engl J Med 2007; 356: Cummings SR, Nevitt MC, Browner WS et al. Risk factors for hip fracture in white women. Study of Osteoporotic Fractures Research Group. N Engl J Med 1995; 332: Kanis JA. Diagnosis of osteoporosis and assessment of fracture risk. Lancet 2002; 359: Kanis JA. Assessment of Osteoporosis at the Primary Health Care Level. Sheffield, UK: World Health Organization Collaborating Centre for Metabolic Disease, University of Sheffield, National Institute for Health and Care Excellence. Osteoporosis: Assessing the Risk of Fragility Fracture. NICE clinical guideline 146. Manchester, UK: National Institute for Health and Clinical Excellence, Socialstyrelsen. Nationella riktlinjer f or r orelseorganens sjukdomar Stockholm, Sweden, Black DM, Cummings SR, Karpf DB et al. Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Fracture Intervention Trial Research Group. Lancet 1996; 348: Cummings SR, San Martin J, McClung MR et al. Denosumab for prevention of fractures in postmenopausal women with osteoporosis. N Engl J Med 2009; 361: Cummings SR, Black DM, Thompson DE et al. Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. JAMA 1998; 280: Boonen S, Adachi JD, Man Z et al. Treatment with denosumab reduces the incidence of new vertebral and hip fractures in postmenopausal women at high risk. J Clin Endocrinol Metab 2011; 96: Tosteson AN, Melton LJ 3rd, Dawson-Hughes B et al. Costeffective osteoporosis treatment thresholds: the United States perspective. Osteoporos Int 2008; 19: Group NOG. Osteoporosis-Clinical guideline for prevention and treatment. National Osteoporosis Guideline Group on 660 ª 2015 The Association for the Publication of the Journal of Internal Medicine

12 behalf of the Bone Research Society, British Geriatrics Society, British Orthopaedic Association, British Orthopaedics Research Society, British Society of Rheumatology, National Osteoporosis Society, Osteoporosis 2000, Osteoporosis Dorset, Primary Care Rheumatology Society, Royal College of Physicians and Society for Endocrinology Siris ES, Adler R, Bilezikian J et al. The clinical diagnosis of osteoporosis: a position statement from the National Bone Health Alliance Working Group. Osteoporos Int 2014; 25: Correspondence: Mattias Lorentzon, MD, Geriatric Medicine, Institute of Medicine, Centre for Bone and Arthritis Research, Sahlgrenska Academy, Sahlgrenska University Hospital, Building K, 6th Floor, M olndal, Sweden. ( Mattias.Lorentzon@medic.gu.se). ª 2015 The Association for the Publication of the Journal of Internal Medicine 661