Osteoporosis has been identified by the US Surgeon General

New Guidelines for the Prevention and Treatment of Osteoporosis E. Michael Lewiecki, MD, and Nelson B. Watts, MD Abstract: The World Health Organization Fracture Risk Assessment Tool (FRAX ) and the National Osteoporosis Foundation s (NOF) Clinician s Guide to Prevention and Treatment of Osteoporosis are helpful clinical tools in the management of osteoporosis. Appropriate use of these tools requires a clear understanding of their limitations as well as their benefits. Good clinical judgment should be the ultimate determinant of treatment decisions. We anticipate that further refinements to FRAX and subsequent updates of the NOF guide will improve their clinical utility. Key Words: FRAX, guidelines, osteoporosis, treatment Osteoporosis has been identified by the US Surgeon General as a major public health concern. 1 It is a common skeletal disease characterized by low bone strength and increased risk of fracture. 2 Fractures are associated with adverse outcomes that include acute and chronic pain, diminished quality of life, disability, high risk of future fractures, increased mortality, and substantial healthcare expenses. Despite the widespread availability of devices to measure bone density (to identify patients at risk of fracture) and approved medications that safely and effectively reduce fracture risk, osteoporosis is underdiagnosed and undertreated. Even when treatment is started, it is commonly not taken correctly or long enough to reduce fracture risk. The National Osteoporosis Foundation (NOF), a not-forprofit US-based organization dedicated to improving patient access to high quality skeletal health care, released a longawaited update of its clinical practice guidelines on February 21, 2008. 3 On the same day, a companion online fracture risk assessment calculator (FRAX ) from the World Health Organization (WHO) was opened for public use. 4 Figures 1 and 2 show the FRAX home page and the risk calculator for US Caucasians, respectively. From the New Mexico Clinical Research & Osteoporosis Center, Albuquerque, NM; and University of Cincinnati Bone Health and Osteoporosis Center, Cincinnati, OH. Reprint requests to E. Michael Lewiecki, MD, New Mexico Clinical Research & Osteoporosis Center, 300 Oak Street NE, Albuquerque, NM, 87106. Email: LEWIECKI@aol.com Accepted July 18, 2008. Copyright 2009 by The Southern Medical Association 0038-4348/0 2000/10200-0175 New NOF Guidelines The NOF s Clinician s Guide to Prevention and Treatment of Osteoporosis is designed to serve as a reference for healthcare providers on the prevention, diagnosis and treatment of osteoporosis. The guide provides information on the epidemiology of osteoporosis, consequences of osteoporotic fractures, and evaluation of patients at risk for fracture. Advice is given on measures to improve lifestyle, optimize calcium and vitamin D intake, and reduce fall risk. Recommendations for pharmacologic treatment are based in part on the US adaptations of the WHO 10-year fracture probability model 5 and algorithms for determining treatment thresholds. 6 These recommendations are based on cost-effectiveness in populations of patients, and should be used together with other considerations in making treatment decisions for individual patients. Because the revised NOF guide represents a new paradigm for the management of osteoporosis, it deserves careful scrutiny to evaluate its clinical utility. The previous NOF guide, released in 2003, recommended pharmacologic therapy to reduce fracture risk in postmenopausal Caucasian women with a previous fracture of the hip or spine, when the T-score at the hip was below 2.0 regardless of risk factors, or when the T-score at the hip was less than 1.5 and at least one risk factor was present other than being postmenopausal and estrogen deficient. 3 While these recommendations were easy to grasp, they were limited in scope due to the population addressed (only postmenopausal Caucasian women not women of other ethnicities, Key Points The expression of fracture risk as 10-year probability of fracture is more clinically useful than relative risk of fracture. Bone mineral density (BMD) combined with clinical risk factors for fracture provides a better estimation of fracture risk than BMD or clinical risk factors alone. Fracture probability can be used to identify patients in whom pharmacological therapy is cost-effective. Factors in addition to cost-effectiveness should be considered in evaluating patients for treatment. Southern Medical Journal Volume 102, Number 2, February 2009 175

Lewiecki and Watts The Prevention and Treatment of Osteoporosis Fig. 1 The FRAX home page. Permission from http://www.shef.ac.uk/frax/index.htm. and not men); lack of weighting of risk factors (some, particularly age and previous fracture, being more robust predictors of fracture risk than others); and failure to quantify fracture risk in a way that was clinically useful. Fracture risk, often expressed as relative risk, does not provide a true assessment of the likelihood (probability) of fracture. For example, an early postmenopausal woman with a slightly low T-score may have a high risk of fracture compared with an age-matched population (where the fracture risk is low), while the actual probability of fracture over the next ten years might not be very high. As a result, there has been a bias toward treating early postmenopausal women with low probability of fracture and not treating older women with similar or even better T- scores who were at greater risk because of their older age. Hence, the concept evolved of estimating fracture probability using a combination of bone mineral density (BMD) and validated clinical risk factors, and expressing fracture risk as 10-year probability of fracture. The task of developing and validating methodologies for estimating fracture risk was undertaken by the WHO under the direction of Prof. John Kanis and the support of many individuals and organizations, including the NOF, International Society for Clinical Densitometry, American Society of Bone and Mineral Research, and the International Osteoporosis Foundation. The job of setting intervention thresholds was left to local or regional groups; in the United States, that was the NOF. Benefits of the New Guidelines The 2008 NOF recommendations for the treatment of osteoporosis are shown in Table 1, alongside the 2003 recommendations. Potential advantages of the new recommendations include better allocation of limited healthcare resources toward patients at higher risk for fracture and most likely to benefit from therapy. Risk factors are used systematically and are scientifically weighted. The expression of fracture risk as 10-year probability of fracture is an 176 2009 Southern Medical Association

Fig. 2 The risk calculator for US Caucasians. Permission from http://www.shef.ac.uk/frax/index.htm. improvement over relative risk; this should result in treatment being more likely to be offered to older patients who are at high risk of fracture but whose BMD is above the 2.5 threshold, and less likely to be offered to younger patients whose fracture risk is lower. The new guidelines 1 also address treatment in non-caucasians and in men. Table 1. Comparison of treatment thresholds in the National Osteoporosis Foundation guides of 2003 and 2008 Category 2003 2008 Applicable population Postmenopausal Caucasian women Postmenopausal women and men age 50 and older (ethnicity not specified) Previous fracture Hip or vertebral Hip or vertebral Regardless of clinical risk factors T-score less than 2.0 at the hip T-score 2.5 or less at the femoral neck, total hip, or spine Low bone mass and risk factors T-score between 1.5 and 2.0 at the hip and clinical risk factors. Major risk factors identified as family history of osteoporosis, personal history of low trauma fracture, cigarette smoking and low body weight (less than 127 lbs.), without weighting T-score between 1.0 and 2.5 at the femoral neck, total hip, or spine, and ten-year fracture risk as assessed by FRAX of 3% or more at the hip, 20% or more for major osteoporosis-related fracture (humerus, forearm, hip or clinical vertebral fracture) Southern Medical Journal Volume 102, Number 2, February 2009 177

Lewiecki and Watts The Prevention and Treatment of Osteoporosis Limitations of the New Guidelines There are potential downsides to the new recommendations. The new guidelines could have eliminated the use of T-scores and the category of osteopenia, but that was not achieved. However, the T-score threshold for treating postmenopausal women without additional risk factors has been lowered from 2.0 to 2.5; this number may be more appropriate, but could cause confusion as to whether patients previously treated should still be treated. Also, the calculation of fracture probability will identify patients previously not treated who should now be treated. While these are transitional issues that will diminish in importance as time passes, they should nevertheless be noted. It should also be pointed out that there is no evidence for reduction of fracture risk in treated patients with T-scores better than 1.5, some of whom would now be treated. Physicians, regulators, and payers of healthcare benefits must remain aware that recommendations are a guide for treatment and not intended to restrict treatment options. Other factors must also be considered in making treatment decisions for individual patients, including nonskeletal ones, such as the risk of falling. Limitations of FRAX Table 2 shows the risk factors considered in FRAX. The use of FRAX sometimes results in values for 10-year probability that are counterintuitive and inconsistent with some of the treatment recommendations. For example, a 50- year-old woman with no clinical risk factors, a body mass index (BMI) of 23.5 kg/m 2, and a T-score of 2.5 meets the T-score threshold for pharmacological therapy, yet has a calculated fracture probability (10-year risk of 2.5% for hip fracture and 11% for major osteoporotic fracture) that is below the treatment threshold. A 72-year-old woman with a T-score of 1.0 and a parent with a hip fracture has a 10-year Table 2. Clinical risk factors considered in the WHO fracture risk assessment tool Geographic region (China, France, Japan, Italy, Spain, Sweden, Turkey, United Kingdom, United States) Race (offered only in the USA Caucasian, black, Hispanic, Asian) Sex Weight in kg and height in cm for calculating body mass index (kg/m 2 ) with a conversion calculator from English units (pounds and inches) Previous spontaneous fracture or fragility as an adult Family history of osteoporosis (parent with hip fracture) Current smoking Glucocorticoid use (prednisolone 5 mg daily or more for 3 mo or longer) Rheumatoid arthritis (diagnosis confirmed by a health professional) Secondary osteoporosis (examples given include type 1 diabetes, osteogenesis imperfecta in adults, untreated long-standing hypothyroidism and hypogonadism or premature menopause) Alcohol more than 3 units daily (a unit of alcohol is equivalent to a glass of beer, an ounce of spirits or a medium-sized glass of wine) risk of 23% for major osteoporotic fracture a level of risk where treatment is recommended, yet no treatment has been shown to reduce fracture risk at this level of BMD. Not all risk factors are straightforward. Previous fractures are not clearly defined, and could include fracture sites not related to osteoporosis, such as fingers and toes. Additionally, secondary causes of osteoporosis are not explained well. Also, the fracture risk associated with corticosteroid use does not consider dose or duration of treatment. Finally, some patients with arthritis declare they have rheumatoid arthritis, although in fact it may be another form of arthritis. It must be recognized that not all risk factors are considered, so the calculated fracture risk may be less than the actual risk. The intervention threshold of 3% 10-year risk of hip fracture is a value that may be meaningful to health policy makers but possibly counterproductive for a patient considering treatment (ie, a 10-year risk of NOT having a hip fracture of 97% may seem quite acceptable). Also, the range of error for 10-year fracture risk is not clearly defined. It might be more appropriate to have an intervention threshold range (eg, above xx% fracture risk treatment should be strongly considered; between yy% and xx%, treatment should be considered based on individual patient circumstances). According to the FRAX model, it is possible for patients with normal T-scores ( 1.0 or better) to be identified as candidates for treatment even though drugs approved for treatment of osteoporosis have not been shown to reduce fracture risk in patients with T-scores better than 1.5. In fact, there is no strong evidence for effectiveness at reducing fracture risk in patients with T-scores better than 2.0. Moreover, within the same population, the calculated fracture risk may rise and fall abruptly with very minor changes in the variables. For example, for an American Caucasian woman with a T-score of 1.5, hip fracture risk is 0.8% at age 55, 0.1% at age 60, and 0.9% at age 65. And quite alarmingly, fracture risk varies widely across populations, often more than seems possible. An example of this can be seen in the 10-year risk of hip fracture for a 66-year-old woman, with a BMI of 23.4 kg/m 2, and a T-score of 2.5; her risk is calculated to be 5.1% in Sweden, 3.1% for a Caucasian in the United States, 2.5% in France, and 1.1% in Spain. Because fracture probability depends in part on life expectancy, 10-year risk decreases from age 80 to 85 and from 85 to 90; this could result in treatment not being offered to elderly patients for whom fractures are likely in the next 5 years. In the United States, it is not clear how to use FRAX in patients who are not Caucasian, black, Hispanic, or Asian. FRAX considers only femoral neck T-score, yet some patients may be at high risk of fracture with a low lumbar spine T-score despite a relatively good femoral neck T-score. Conclusion The WHO Fracture Risk Assessment Tool and the new NOF Clinician s Guide to Prevention and Treatment of 178 2009 Southern Medical Association

Osteoporosis represent advances in the care of osteoporosis, but more could be done to enhance their usefulness in managing patients. Healthcare providers, regulators, and payers must clearly understand the limitations as well as the benefits of these tools. They must also recognize that good clinical judgment should be the ultimate determinant of treatment decisions. We anticipate that further refinements to FRAX and subsequent updates of the NOF guide will improve their clinical utility. References 1. US Department of Health and Human Services. Bone Health and Osteoporosis: A Report of the Surgeon General. Rockville, MD, US Department of Health and Human Services, Office of the Surgeon General, 2004. 2. Klibanski A, Adams-Campbell L, Bassford T, et al. Osteoporosis prevention, diagnosis, and therapy. JAMA 2001;285:785 795. 3. National Osteoporosis Foundation. Clinician s Guide to Prevention and Treatment of Osteoporosis. Washington, DC, National Osteoporosis Foundation, 2008. Available at: http://www.nof.org/professionals/clinicians_ Guide.htm. Accessed July 10, 2008. 4. World Health Organization. WHO fracture risk assessment tool. Available at: http://www.shef.ac.uk/frax/. Accessed July 10, 2008. 5. Dawson-Hughes B, Tosteson AN, Melton LJ III, et al; National Osteoporosis Foundation Guide Committee. Implications of absolute fracture risk assessment for osteoporosis practice guidelines in the USA. Osteoporos Int 2008;19:449 458. 6. Tosteson AN, Melton LJ III, Dawson-Hughes B, et al; National Osteoporosis Foundation Guide Committee. Cost-effective osteoporosis treatment thresholds: the United States perspective. Osteoporos Int 2008;19: 437 447. I have learned that success is to be measured not so much by the position that one has reached in life as by the obstacles which he has overcome while trying to succeed. Booker T. Washington Southern Medical Journal Volume 102, Number 2, February 2009 179