Oestrogen derivatives and receptor agonists. Tamoxifen, raloxifene. Tibolone. Calcium. Calcitonins (salmon, eel, human, porcine). Bishosphonates. Etidronate.
Osteoporos Int (1997) 7:390–406 ß 1997 European Foundation for Osteoporosis and the National Osteoporosis Foundation
Osteoporosis International
Position Paper Guidelines for Diagnosis and Management of Osteoporosis J. A. Kanis, P. Delmas, P. Burckhardt, C. Cooper and D. Torgerson on behalf of the European Foundation for Osteoporosis and Bone Disease
Preamble. Significant developments have occurred in the field of osteoporosis over the past several years. There is now considerable information concerning its impact on general health and an international consensus concerning the definition of osteoporosis. Conceptually, this recognizes the multifactorial nature of the events which give rise to the fractures, but operational definitions have now been agreed and have gained a wide measure of acceptance. Accurate and precise diagnostic tools are also available. Finally, there is substantial evidence that the natural history of osteoporosis can be modulated by agents which in turn decrease the risk of fracture. Despite an increasing professional and public awareness of osteoporosis, the management of osteoporosis has been confined mainly to specialists. With the large number of affected individuals and the wider availability of diagnostic aids and safe treatments, there is a need for osteoporosis to be managed predominantly by the primary care physician. Against this background the European Foundation for Osteoporosis and Bone Disease through their Scientific Advisory Board has recognized a need to develop practice guidelines for primary care physicians which are summarized in this paper.
The clinical significance of osteoporosis lies in the fractures that arise. Common fractures include vertebral compression fractures, and fractures of the distal radius and the proximal femur (hip fracture). In addition, when the skeleton is osteoporotic, fractures occur more commonly at many other sites including the pelvis, proximal humerus, distal femur and ribs. Osteoporotic fractures occurring at the spine and the forearm are associated with significant morbidity, but the most serious consequences arise in patients with hip fracture, which is associated with a significant increase in mortality (15–20%), particularly in elderly men and women. Hip fractures account for more than 20% of orthopaedic bed occupancy in the UK, in Scandinavia and several other countries. After the age of 45 years, hip fractures account for a higher proportion of hospital bed occupancy than many other common disorders in women, including breast cancer and diabetes (Fig. 1).
Introduction The internationally agreed definition of osteoporosis is: ‘a progressive systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture’ [1]. ——————— Correspondence and offprint requests to: Professor John A. Kanis, WHO Collaborating Centre for Metabolic Bone Diseases, Department of Human Metabolism & Clinical Biochemistry, University of Sheffield Medical School, Beech Hill Road, Sheffield S10 2RX, UK. Tel: +44 (0)114 271 2649. Fax: +44 (0)114 273 9176.
Fig. 1. Hospital bed occupancy in women aged 45 years or more according to diagnostic category in the Trent Region of England (9% of the UK population). COAD, chronic obstructive airways disease; MI, myocardial infarction. (Source: Trent Health.)
Guidelines for Diagnosis and Management of Osteoporosis Table 1. Estimated remaining life-time fracture risk (%) with confidence intervals in women and men from Rochester, MN, USA at the age of 50 years Fracture site
Women
Men
Proximal femur Vertebrala Distal forearm Any of the above
17.5(16.8–18.2) 15.6(14.8–16.3) 16.0(15.7–16.7) 39.7(38.7–40.6)
6.0 (5.6–6.5) 5.0 (4.6–5.4) 2.5 (2.2–3.1) 13.1(12.4–13.7)
From Melton LJ, Atkinson EJ, O’Fallon WM, Wahner HW, Riggs BL. Long-term fracture risk prediction with bone mineral measurements made at various skeletal sites. J Bone Miner Res 1991;6 (Suppl 1): S136. a Clinically diagnosed fractures.
The likelihood that any individual will suffer an osteoporotic fracture is considerable. In many Western countries the remaining life-time risk of a hip fracture in white women at the age of menopause lies between 15% and 17%. The risk for other common types of osteoporotic fractures is nearly as high (Table 1), so that the combined fracture risk is 30–40%. Thus, more than one third of adult women will sustain one or more osteoporotic fractures in their life-time. This estimate is conservative since it does not include fractures at other sites and only takes into account those vertebral fractures which come to clinical attention, so that the true risk of fracture is higher. These indices of fracture risk compare with a life-time risk in women at the age of 50 years of 9–12% for breast cancer and 30–40% for cardiovascular disease. This indicates the widespread prevalence of osteoporosis in our society. In comparison, risks for men are about onethird of those in women, and are even lower for forearm fractures, but still represent a considerable burden. The increasing awareness of osteoporosis combined with the current availability and development of specific treatments is likely to increase the demand for management of patients with osteoporosis. In the past the management has been largely confined to specialists, but the increasing availability of diagnostic tools and wellproven treatments, and the increasing numbers of patients identified, indicate that the burden of managment will fall increasingly on the primary care physician. The aim of this document is to provide a framework for the cost-effective diagnosis and management of osteoporosis. These guidelines do not consider the identification of patients at risk from osteoporosis and strategies for prevention of the disease, though in many respects the approaches are similar.
Diagnosis of Osteoporosis The definition of osteoporosis captures the notion that low bone mass is an important component of the risk of fracture, but that other abnormalities occur in the skeleton, and that non-skeletal factors such as falls are also important. Nevertheless, it is only bone mass
391
measured as bone mineral that can presently be measured with precision and accuracy, and its measurement forms the basis for the diagnosis of osteoporosis.
Techniques to Measure Bone Mineral There are two widely utilized techniques to assess bone mass. They variously assess mineral content of regional sites, particularly those sites at risk of osteoporotic fracture such as the wrist, spine and hip, but also the whole skeleton. Single-Energy Absorptiometry. The technique measures bone mineral at peripheral (appendicular) sites such as the heel and the wrist. Single photon absorptiometry (SPA) utilizes a photon-emitting source such as iodine125, and the amount of bone mineral in the tissue traversed attenuates the photons from which the mineral content is calculated. Single-energy X-ray absorptiometry (SXA) has now supplanted SPA as the single-energy technique for scanning the wrist. It is more precise and avoids the need for isotopes. Dual-Energy Absorptiometry. Sites such as the spine and hip cannot be measured accurately by SPA or SXA. Dual-energy absorptiometry utilizing photons (DPA) or X-rays (DXA) permits bone mineral to be measured at these sites. The amount of bone mineral present at a specific site of a scan is termed the bone mineral content. When the bone mineral content is divided by the area or volume assessed (the region of interest), a value for bone mineral density is provided. With single- and dualenergy absorptiometry the bone mineral content is divided by the area assessed (because of the twodimensional scan) and is not, therefore, a true volumetric density but an areal density. A typical scan of the lumbar spine is shown in Fig. 2 and an analysis in Table 2. Several other techniques have been developed to measure bone mineral, but their use is less widespread and in some cases confined to clinical research. They include quantitative computed tomography, ultrasound evaluation of bone and several radiographic techniques including radiographic density measurements at the hand. Note that standard skeletal radiographs are a very inaccurate method of assessing the amount of bone mineral, but apparent osteoporosis on X-rays is a common reason for further assessment. The major limitations of computed tomography at the spine are the expense, low precision and comparatively high radiation dose. Computed tomography techniques to assess bone mineral density at appendicular sites are used in some countries with lower cost and radiation exposure and higher precision. Ultrasound attenuation and velocity measurements avoid exposure to radiation. They may be useful as a diagnostic aid, but their value in monitoring treatments is not yet established.
392
J. A. Kanis et al.
Fig. 2. Dual-energy X-ray absorptiometry scan of the lumbar spine in a healthy, premenopausal woman aged 33 years. The computer has selected the edge of the spine and the vertebrae have been distinguished by the operator to give the areas of interest to be computed.
Table 2. Measurements made from an anteroposterior scan using DXA at the lumbar spine in a 53-year-old woman at the menopause Region Area (cm2)
Mineral content (g)
Mineral density (g/cm2)
T-score SDa
(%)
SDa
(%)
adults (‘peak bone mass’) is approximately normal irrespective of the technique used. Because of this normal distribution, bone density values in individuals are often expressed in relation to a reference population in standard deviation units. This reduces the problems associated with differences in calibration between some instruments. When standard deviation units are used in relation to the young healthy population, this is referred to as the T-score. For diagnostic purposes two thresholds of bone mineral density have been proposed for Caucasian women based on the T-score [2]. The first defines the majority of individuals who will sustain a fracture in the future (osteoporosis) and the second a higher threshold that may identify those most likely to develop osteoporosis and those women who might benefit most by the prevention of bone loss at the time of the menopause (low bone mass or osteopenia). Osteoporosis denotes a value for bone mineral density or bone mineral content that is two and a half standard deviations or more below the young adult mean value (T-score 4–2.5). Osteopenia or low bone mass is a T-score that lies between –1 and –2.5. Severe or ‘established’ osteoporosis denotes osteoporosis as defined above in the presence of one or more documented fragility fractures, usually of the wrist, spine or hip. In the young healthy population, 15% of women will have a T-score of less than –1, and thus have osteopenia. Approximately half a per cent will already have osteoporosis (Fig. 3). Note that these thresholds apply to women only. Suitable diagnostic cut-off values for men are less secure. It has been suggested that a similar absolute value for bone mineral density as that used in women can be taken as a cut-off point for the diagnosis of
Z-score
L1 L2 L3 L4
12.24 13.03 14.51 15.08
10.29 13.04 15.00 15.81
0.841 1.000 1.034 1.048
–0.77 –0.25 –0.46 –0.62
91 97 95 94
+0.07 +0.68 +0.53 +0.39
101 108 106 104
Total
54.86
54.14
0.987
–0.55
94
+0.41
105
Note the increase in vertebral area, mineral content and density with the more caudal direction. The results are compared either with the premenopausal reference range (T-score) or with an age-matched population (Z-score). In this instance the bone mineral density is 6% below the average for peak bone density but 5% above average for age. a Standard deviation units.
Bone Mineral and Osteoporosis Skeletal mass is relatively constant once growth has ceased, until the age of 50 years or so. The distribution of bone mineral content or density in young healthy
Fig. 3. Distribution of bone mineral density in young healthy women aged 30–40 years. Because the distribution of bone density is normal, 15% of the population have a T-score of –1 or lower and less than 0.6% of the population have a T-score below –2.5.
Guidelines for Diagnosis and Management of Osteoporosis
osteoporosis; namely a value for bone mineral 2.5 standard deviations below the average for women.
Natural History of Bone Loss Bone mineral content and density are relatively constant in adult men and women under the age of 50 years. In women, loss of bone may occur before the age of menopause at some sites, but losses are small compared with those in later life. Enhanced bone loss occurs thereafter, and coincides with ovarian failure shortly before the menopause and continues throughout life. The
Fig. 4. Bone mineral density (BMD) in women at different ages and the prevalence of osteoporosis. Bone mineral density is normally distributed at all ages, but values decrease progressively with age. The proportion of patients with osteoporosis (with a bone mineral density of 2.5 standard deviation units or less than the young adult mean) increases exponentially with age. (From Kanis JA, et al. Osteoporos Int 1994;4:368–81.)
393
proportion of patients with osteoporosis increases exponentially with age because of the distribution of bone mineral density in the population (Fig. 4). Bone loss occurs with age in men, but the rate of bone loss is much slower so that the frequency of osteoporosis and of fractures is less. What Does This Tell us About Fracture Risk? The average life-time risk of the common osteoporotic fractures in white women and men is approximately 40% and 13% respectively at the age of 50 years (see Table 1). The risk of fracture approximately doubles for each standard deviation decrease in bone mineral density. The risk is more than doubled in individuals with low bone mass and nearly 4-fold greater in women with osteoporosis (50% life-time risk at the age of 50 years) compared with women with a normal bone mineral density (13% life-time risk at 50 years). The risks can be doubled again when individuals have had a fragility fracture beforehand (see below). Estimating fracture risk by bone mineral measurements is comparable to the assessment of the risk of stroke by blood pressure readings. Blood pressure values are continuously distributed in the population, as is bone mineral density. In the same way that a patient above a cut-off level for blood pressure is diagnosed as hypertensive, the diagnosis of osteoporosis is based on a value for bone mineral below a cut-off threshold. As is the case for blood pressure and stroke, there is no absolute threshold for bone mineral that discriminates absolutely who will or will not fracture. The performance of bone mineral density in predicting fracture is, however, at least as good as that of blood pressure in predicting stroke, and considerably better than the use of serum cholesterol to predict coronary artery disease. Nevertheless, it should be recognized that, just because bone mineral density is normal, this is no guarantee that fracture will not occur – only that the risk is decreased. If, however, bone mineral density is in the osteoporotic range, then fractures are likely. Which Site Should be Measured? The site for assessment depends upon the reason for undertaking the scan and also on the age of the patient. Osteoporosis is a systemic disease and loss of bone occurs at all sites. For this reason bone scans for diagnostic purposes should normally be undertaken at any one site. There is only a small advantage in the measurement of multiple sites. Because, however, the correlation between bone mass at different sites is less than perfect, osteoporosis at one site is not invariably associated with osteoporosis at other sites, and therefore assessment of the relevant biological site is preferable. For these reasons, measurements made at the wrist, spine or hip may be appropriate in younger individuals (to assess the risk of any fracture). In the elderly hip fractures are the major concern and carry the highest morbidity and mortality. If the clinician wishes to predict hip fracture risk with the greatest accuracy, then measurement at that site is more useful than elsewhere since measurements at the site of
394
biological relevance predict fractures at that site most accurately. In addition, osteoarthrosis is particularly common in the spine of men and in elderly women, and in such cases this site is less suitable for the measurement of bone density in these age groups. Changes in bone mineral density that occur in the immediate postmenopausal period or as a result of treatment are often more marked in the spine and can be detected earlier than at the hip or wrist.
Problems in the Assessment of Bone Mineral There are a number of problems and limitations in the assessment of bone mineral which should be recognized in the interpretation of bone scans (Table 3). Bone mineral density gives an estimate of bone mass based on its mineral content. The presence of osteomalacia, a disorder where mineralization of bone is defective that may occur in the elderly, will therefore, underestimate bone mass. Osteoarthrosis at both the spine and the hip are common in the elderly, and will contribute to the density measurement, partricularly at the spine, but not to skeletal strength. Heterogeneity of density due to osteoarthrosis or previous fracture can often be detected on the scan and in some cases be excluded from the analysis. In the case of the hip, smaller regions of interest can be selected to exclude the joint. Table 3. Problems in the interpretation of bone mineral measurements Osteomalacia Osteoarthritis (especially the spine) Vascular calcification (especially the spine) Overlying metal objects Contrast media (spine) Previous gold therapy Previous fracture (spine, hip and wrist) Severe scoliosis Small stature Vertebral deformities due to osteoarthrosis, Scheuermann’s disease Inadequate reference ranges
It is important that an appropriate normal reference range is used to interpret bone density values. The ranges supplied by manufacturers may variously over- or underestimate the proportion of a tested population categorized as having osteoporosis.
Identification of Patients with Osteoporosis At present there is no universally accepted policy for screening to identify patients with osteoporosis. With the increasing development of effective agents specifically affecting bone metabolism this view may change. In the absence of such policies patients are identified as having osteoporosis largely because of a fragility fracture or sometimes by the presence of strong risk factors. Bone mineral density measurements can be used to enhance
J. A. Kanis et al. Table 4. Clinical risk factors providing indications for the diagnostic use of bone densitometry 1. Presence of strong risk factors. Oestrogen deficiency Premature menopause (1 year) Primary hypogonadism Corticosteroid therapy (>7.5 mg/day for 1 year or more) Maternal family history of hip fracture Low body mass index (
