10.1.6 Bone disease in older people

Bone disease is a common problem in the elderly, and its clinical manifestation are a major preventable public health problem. The disorders of the skeleton have been classified in a variety of ways, an approach which tends to restrict understanding of the clinical problem in a particular patient. Frequently, several separate disorders coexist, each contributing to impairment of bone form or function, and each requiring a separate intervention.

The major categories of disorder are osteoporosis, too little bone within the bone, osteomalacia, impaired mineralization of bone matrix, and infiltration of bone with cancer cells. Each represents a distinct pathological processes that results in abnormal bone structure and function, which may present as bone pain and/or fracture.

Osteoporosis is defined as a condition characterized by micro-architectural deterioration of skeletal structure, predisposing to fragility fracture, its principal manifestation. The many causes of this condition are listed in Box 10.1.6.1. All share some clinical and anatomical aspects.

The clinical definition of osteoporosis emphasizes the failure of the mechanical function of the tissue, with consequent pain and deformity, and thus is a term with connotations similar to heart failure or kidney failure. Osteoporosis is the most common cause of bone disease in the elderly. Interest in osteoporosis as an age-related disorder is fuelled by the strong age dependence of fracture and the dramatic increase in life expectancy in most countries of the world.

The anatomical definition emphasizes the micro-architectural deterioration that affects both trabecular and cortical bone structures, and which forms the basis for the clinical presentation. The principle mechanism by which it occurs is via an abnormality of bone turnover. This is a process whereby osteoclast-mediated resorption of bone occurs on preformed surfaces, followed by osteoblast-mediated bone formation, which in early life restores the structure completely. However, in osteoporosis osteoclast-mediated bone resorption occurs without adequate compensatory osteoblast-mediated bone formation. The critical concept is that of an imbalance between the processes of resorption and formation. In trabecular bone the disease process leads to disconnection between the trabecular plates, rendering them mechanically incompetent. In cortical bone it leads to cortical thinning, primarily due to endocortical resorption and intracortical porosity, the latter being due to tunnelling of the osteoclast ‘cutting cone’ through the dense cortical bone. These deleterious effects are counteracted in part by primary bone apposition on periosteal surfaces, a process called modelling.

Bone formation requires the production of a protein matrix, called osteoid, by the osteoblast. Crystals of hydroxyapatite consisting of calcium phosphate and water then form on the osteoid to produce a tough, strong mechanical support structure. The definition of osteomalacia requires evidence of increased unmineralized osteoid due to a defect in mineralization. This appearance occurs as a result of deficient concentrations of calcium or phosphate for the formation of hydroxyapatite crystals on the osteoid formed by osteoblasts. The supply of calcium has to be reduced substantially to result in increased unmineralized osteoid, most frequently this is due to a combination of dietary calcium and vitamin D deficiency. However, it is important to understand that many elderly patients do indeed have substantial deficiencies of both factors. Phosphate deficiency causing osteomalacia is rare, and when developing in old age is usually due to oncogenic osteomalacia, although patients with familial hypophosphataemia do survive into old age.

The anatomical definition requires evidence of a delay in mineralization as evidenced by a prolonged mineralization lag time—the time between the production of the osteoid by the osteoblast and its full calcification. The gold standard for diagnosis requires a bone biopsy after giving the patient tetracycline to label newly calcifying osteoid, to demonstrate the calcification front. Two labels are given 2 weeks apart. In patients suffering from osteomalacia, two separate lines are not seen; rather, the label is not taken up or is smeared over the osteoid.

In the absence of a bone biopsy, serum markers of bone turnover such as bone-derived alkaline phosphatase levels are increased. This, together with biochemical evidence of calcium deficiency, low serum calcium, secondary hyperparathyroidism, and a low fasting urine calcium to creatinine ratio, is usually taken as evidence that osteomalacia may be present and the mineral deficit treated accordingly.

This term has been introduced recently and is synonymous with malignant bone disease. The bone defect may share some similarities with osteoporosis in that there may be dramatically increased bone resorption; hence, antiosteolytic pharmaceuticals may be effective in both conditions. In tumour induced osteolysis the resorptive process is driven either by malignant metastases to the bone, causing increased osteoclast activity by paracrine effects, or by solid tumours releasing endocrine factors such as parathyroid hormone related peptide (PTHRP) into the circulation.

Each of the three disorders outlined above can result in fracture. There are two main types of fracture: those due to a single application of force above the failure load of the bone, and those due to repeated applications of sub-failure load forces concentrated at a point because of the presence of a small localized fracture (stress fracture).

Classically it is considered that osteomalacia is the cause of stress fractures. However all three major categories of skeletal disorder outlined above can result in clinical stress fracture.

Accepted sites for the single event minimal trauma fracture include the hip, spine, and forearm. More recently it has been accepted that fractures at most sites in old age are osteoporotic in origin. Common sites for such fractures include the humerus, pelvis, and ribs. Certain skeletal sites are considered to be unlikely to be sites for osteoporotic fractures. Fractures of the face, skull, hands and feet do not increase with age and are often associated with a direct blow, suggesting that in the case of these fractures, skeletal fragility or osteoporosis is not a prominent feature. Furthermore, in a prospective study, fractures of the finger, ankle, and face were not associated with reduced bone mass. Fractures at these sites do not require bone-based pharmacological treatments.

Incidence of spine fracture rises after the age of 60, especially in women, and often occurs without any obviously excessive force. One difficulty in determining the epidemiology of vertebral fracture is the problem of specifying the criteria for diagnosis. The problem with using radiological criteria for fracture definition is that there is a wide range of variation in normal vertebral body morphology. This makes the differentiation of vertebral fracture from normality on X-ray difficult.

There are three types of vertebral fracture. Wedge fractures present with a reduction in anterior height measurements compared to posterior heights by 20% in the lumbar spine and 30% in the thoracic spine. Central fracture of the vertebral end plate, often associated with wedge fracture and crush fracture of the both posterior and anterior borders of the vertebra, is diagnosed with reference to the vertebra above and below. Clinically diagnosed vertebral fracture has an associated increased mortality with a five year relative survival of 0.81. Although up to 50% of patients with spine fracture diagnosed on radiological criteria alone often do not report a specific episode of back pain, such subjects do have an increased level of back symptoms compared to those without fracture.

In industrialized cultures, fracture risk after the age of 70 is associated with an increased propensity to fall. There are numerous clinical risk factors predisposing to hip fracture (1). The enormous impact of hip fracture on mortality has been recognized for many years with mortality rates increased by up to 20% during the 6–12 months following the fracture. The overall five year relative survival is 0.82. In addition there is an enormous reduction in functionality, as shown by performance on various activities of daily living. These disabilities result in increased use of community resources such as hostels and nursing homes in societies where these are available. Such care constitutes a major part of the cost of treating a fracture, which these communities have to carry.

In the same way that osteoporosis increases with age, so does the propensity to fall. Approximately 30% of self-caring subjects over the age of 65 years fall each year. Falls are a potent cause of osteoporotic fracture. In the elderly, 10–20% of all falls result in significant injury, including fracture. Falling, like bone fragility, has numerous causes, of which vitamin D deficiency is the one most closely related to bone disease. A review of the causes of falling is outside the scope of this chapter.

Clearly a propensity to fall will also increase the chances of minimal trauma fracture. The incidence of both osteoporosis and falls rise dramatically with age, accounting for the age dependence of osteoporotic fracture. At very low levels of bone mass, little force is required to fracture a bone. Typical examples of this are fractures associated with cancer induced osteoporosis. It should also be noted that force can be applied in ways other than falling; for example, lifting can cause vertebral fracture and squeezing can cause rib fracture. Thus, an osteoporotic fracture is one that occurs in the presence of force that the skeleton should normally be able to withstand.

In order to develop a fracture, a force that exceeds the mechanical strength of the bone must be applied. There is a linear relationship between measures of structure, e.g. bone mineral density (BMD), and the force required to fracture a bone. A large force will result in fracture irrespective of the bone architecture. This is the reason fractures following motor vehicle accidents are not considered to be associated with osteoporosis. A patient whose bones fracture after falling is more likely to have a reduced bone structure. However, in general only about half of these individuals actually have a bone mass at or below the threshold defined as indicative of osteoporosis using the gold standard for the diagnosis of osteoporosis: a dual energy x-ray absorptiometry (DEXA) BMD less than 2.5 SD below the young normal mean (T score < −2.5).

Age-related osteoporosis is an inclusive term used to specify low bone density in elderly individuals. Its aetiology varies from person to person and most suffer from more than one of the disorders listed in Box 10.1.6.1.

One way to integrate these disorders is to consider their activity on the only two cell types that can influence bone mass and structure: the osteoclast and osteoblast. Every adult continually regenerates their skeleton, such that on average no bone structure is older than 5–10 years. This occurs so that microfractures acquired as a result of forces applied to the skeleton are remodelled away. A feature of age related osteoporosis is the relative overactivity of the osteoclast compared to the osteoblast, resulting in bone loss and osteoporosis. Both problems are more important in causing fracture in individuals with low peak bone mass induced by genetic mechanisms.

The two most common causes of osteoclast over activity are firstly a deficiency of extracellular calcium and/or vitamin D, and secondly a deficiency of gonadal hormones acting directly to increase bone turnover.

This occurs as a result of loss of vitamin D action on calcium transport in the bowel, causing reduced calcium absorption. Reduced gonadal hormone levels also cause reduced intestinal calcium absorption and increased urine calcium loss (Fig 10.1.6.1). The extracellular calcium deficiency is sensed by the parathyroid glands, resulting in an increase in parathyroid hormone (PTH), so called secondary hyperparathyroidism. At the cellular level, PTH stimulates the formation of cytokines and secreted proteins such as RANKL, which stimulate osteoclasts and result in increased osteoclastic bone resorption.

Vitamin D produced in the skin under the influence of sunlight is the main physiological source of vitamin D in the body, where its production is tightly regulated. UVB radiation (290–320 nm) is required for the production of cholecalciferol in the skin. Factors that inhibit this photochemical reaction relate to factors preventing adequate incident sunlight falling on the skin. These include sun block, glass and low incident sunlight due to low zenith angle in high latitudes, especially in winter when little vitamin D is formed in the skin. The efficiency of formation of vitamin D in the skin also falls with increasing age. In the absence of adequate incident sunlight, dietary sources can replace the skin as the main supplier of vitamin D to the body. Because the supply of vitamin D in the diet is unregulated, vitamin D intoxication may occur at high vitamin D intakes.

The decline in gut calcium absorption is dependent on a reduction in the effectiveness of vitamin D on stimulating calcium absorption. This is due to a reduction in the concentration of the precursor of calcitriol (25-hydroxyvitamin D) made in the liver, due to lack of its precursor cholecalciferol and also of calcitriol itself due to renal impairment (Fig. 10.1.6.2).

Oestrogen deficiency has been reported to reduce intrinsic gut wall calcium transport and cause a rise in renal calcium excretion, due to a reduction in the reabsorption of calcium in the distal tubule (Fig 10.1.6.3). This is associated with the loss of oestrogen stimulation of plasma membrane Ca²⁺ ATPase-1 (PMCA1) previously called calcium ATPase. The net effect of these two processes is to increase bone resorption which can in part be corrected by increased calcium intake (2).

In addition to the indirect effects of gonadal hormone deficiency, there are direct effects on bone resulting in increased osteoclast activity. The exact mechanism remains uncertain but involves increased generation of active osteoclasts from monocyte precursors under the activity of cytokines generated in osteoblasts.

The increased dissolution of bone under the action of the osteoclast is further exacerbated by an age-related reduction in osteoblast activity, such that bone dissolved is not adequately replaced. The basis of this reduced activity has not been completely elucidated, but the concept of senescence of mesenchymal stem cells within the bone has been advanced. The basis for this is currently being explored, and in view of the fact that agents such as teriparatide (recombinant PTH) have been shown to stimulate increased osteoblast activity compared to the osteoclast, it is likely that this defect is correctable.

In addition to the disorders discussed above, the concepts that inherited disorders of bone biology can affect both peak bone mass and bone loss in old age are now accepted. The principal determinant of skeletal structure during bone growth in childhood and adolescence is genetic potential, which accounts for 60–80% of the population variance in peak bone density.

There is now a clear proof of principle that genetic effects can influence phenotype in the presentation of fracture in ageing. Epidemiological evidence indicates that a family history of hip fracture predicts occurrence of fracture in the patient. A complete family history of fracture should be obtained when evaluating a patient for osteoporosis risk. A polymorphism in an SP1 binding site of the collagen α1 gene promoter has been described. The phenotype is of low bone density and increased propensity to fracture in old age, without the more severe manifestations of classic osteogenesis imperfecta (3). A mutation in the aromatase gene, resulting in increased production of oestrogen, also reduces age related bone loss. Many other polymorphisms have been described in genes that regulate proteins involved in skeletal physiology. To date, few have held up as being of potential clinical use in order to evaluate future fracture risk. Nevertheless, discovery of polymorphisms that modify phenotype should allow greater understanding of the mechanisms of osteoporosis and direct new therapeutic interventions.

The role of reduced calorific nutritional intake in addition to a low calcium intake has been recognized as a potential cause of osteoporosis in the elderly. Patients with low body weight are more likely to have reduced bone density and fractures. In women, this may be due in part to low endogenous oestrogen production in diminished fat stores.

Reduction in physical activity reduces mechanically induced maintenance of bone microarchitecture in animal studies. This has been clearly demonstrated in the immobilization that follows fracture and in patients with spinal cord lesions. Equally, increasing stress-strain relationships in bone has been shown to increase bone mass in animal and human studies to a small extent.

Over their lifetimes, men can expect to sustain similar degrees of loss of trabecular bone as women. Factors that have been implicated include testosterone deficiency, inducing oestrogen deficiency, calcium deficiency, and deficiency in the vitamin D endocrine system. The incidence of true hypogonadism rises with age, although the precise level of testosterone at which treatment benefit can be expected is controversial. However, there is evidence that high rates of bone turnover can be prevented by testosterone therapy, with a consequent increase in bone density at the spine and the hip. The aetiology of increased bone turnover is related to decreased aromatization of oestrogen. There are also data available on the direct effects of nonaromatizable androgens on the bones and kidney.

Primary prevention is a major management aim that is attractive in osteoporosis, because of the large number of patients at risk, and because we have interventions that are effective in preventing bone loss. The primary prevention approach can be applied to the whole population without knowledge of the precise risks for fracture. The approach requires that the interventions recommended have few risks and can be implemented by health promotion methods (Box 10.1.6.2).

In operational clinical terms, a patient with osteoporosis is defined as an individual at increased risk of fragility fracture. Selecting the appropriate level of intervention depends on a clear understanding of the level of risk of fracture that the patient has. Although it is true that at some point in their life over 40% of women and 30% of men may sustain an osteoporotic fracture, in many this will occur in old age. In young healthy men and women the actual risk of sustaining a clinical osteoporotic fracture is less than 0.5% per year as opposed to 4% per year or higher in women over 80. Thus the principal determinant of fracture risk is age. In addition to age there are two clinically useful determinants of fracture risk: BMD measurement and history of previous fracture.

Bone densitometry is a generic term denoting the noninvasive measurement of bone structure in order to predict its strength. A variety of modalities are in current use. These include the use of electromagnetic radiation in quantitative computed tomography (QCT) and dual x-ray absorptiometry (DXA). High frequency sound waves are used in ultrasonographic measurement. An older technique that has been automated is radiographic morphometry using an X-ray, usually of the hand.

DXA scanning is currently the most widely used diagnostic modality for osteoporosis. DXA integrates the measurement of all the bone structures in the path of the scanning beam, including cortical and trabecular bone, into one value. Beams of radiation at two energy levels are generated from an X-ray source. and are scanned across the bone in a two dimensional fashion. The detector measures the attenuation of the two beams and by integrating the relative attenuation of the low and high-energy beams, is able to define the bone area and subtract attenuation due to soft tissue from the bone image. By dividing the bone mineral content (BMC) by the bone area, a third measurement, the areal BMD, is derived expressed in g/cm². It is important to understand that this is not a true density because it is only two-dimensional. Thus, variation in the third unmeasured dimension will affect the ‘density’ of the bone. In the absence of pathology, the true volumetric bone density of males and females are similar. However the male skeleton is on an average larger than the female in all dimensions. Thus the area bone density will appear to be larger in males than females because the unmeasured third dimension is larger. Combining true density and size in this way may have some advantages, as bone strength is dependent on both the amount of bone within the bone and the size of the bone.

Different DXA machines have been calibrated against standards differing in size and composition, so that comparisons of machine values in g/cm² are not valid. To overcome these problems, bone density values are commonly expressed in standard deviation units with reference to either the age matched range (Z-score) or the young normal range consisting of subjects at peak bone mass (T-score). These methods of expressing the bone density value are valid, because bone density values at a particular age have a normal distribution.

Osteoporosis may be localized to certain areas of the skeleton. Thus future fracture risk is better determined by bone density measurements at several sites. It is recommended that an appendicular and axial skeletal site be measured, usually the lumbar spine and hip. If there is significant degenerative joint disease, or fracture in the lumbar spine, the distal forearm provides a useful area of trabecular bone to sample.

The World Health Organization (WHO) has developed an operational definition of postmenopausal osteoporosis in women for purposes of clinical decision making as a DXA bone density value 2.5 standard deviations below the young normal mean. Individuals with values below this level are considered to be at increased risk of fracture. This level was chosen because it defined the lower limit of normal in relation to the young healthy skeleton, and also because it selected subjects at high risk of fracture. Therapy for osteoporosis should be considered in subjects with bone density below this level. This definition does not apply to other modalities of bone density measurement, especially ultrasound scans.

The WHO committee also introduced the concept of osteopenia, or low bone density, that applies to subjects with a T-score below −1.0. These individuals are at increased risk of fracture, but this is lower than for osteoporotic subjects. There is evidence that these classifications also apply to men with age-related osteoporosis.

Previous spine fractures are strong predictors of future fracture at the spine site (relative risk increased tenfold). Previous appendicular fracture also predicts spine fracture (relative risk increased twofold). Any previous fracture will increase the relative future risk of hip fracture twofold. Thus previous fracture not induced by excessive force such as accident predisposes to future fracture, and can be used in conjunction with bone density and age to select subjects at high risk of future fractures.

The lower the bone density T or Z scores the higher the risk of fracture. In age related and postmenopausal osteoporosis, the relative risk of fracture rises by about two times for each standard deviation below normal, irrespective of the age and gender (4). A patient with a Z score of −3 (that is, their bone density at that site is 3 standard deviations below the mean normal for age), is at an eight times (2 × 2 × 2) increased risk of fracture compared to a subject whose bone density is at the mean for their age.

This calculation has now been incorporated into multivariate models which also take account of age, gender, and previous fracture history to give a 5 or 10 year risk (4) (5). Patients who are at over 5 to 10% risk of fracture in five years should be considered for preventive pharmacological treatment, but only if they have a DXA BMD score of less than −2, as studies of individuals with higher bone density have not been undertaken. In general, most elderly people who have had an osteoporotic fracture and have a bone density T-score value less than or equal to −2.0 have a five-year risk of fracture over 5%.

Case finding implies diagnosis of osteoporosis when the patient may not have a symptomatic complaint such as minimal trauma fracture or the loss of height. The principle diagnostic approach that is recommended consists of two steps. The first is the recognition of people who may potentially have osteoporosis, followed by the second step, bone density estimation at two skeletal sites to assist in the evaluation of future risk of fracture. Patients with the conditions outlined in Box 10.1.6.1 are possible candidates for screening using DEXA. In the presence of persistent back pain or significant kyphosis a lateral X-ray of the lumbar and thoracic spine should be taken to diagnose vertebral fracture. If available, DXA BMD testing is useful to evaluate the extent of the condition and assist in follow-up.

This requires a detailed history and examination directed at the various causes of bone disease outlined in Box 10.1.6.1. The most common in men and women is age-related osteoporosis. Malabsorptive conditions can be difficult to diagnose. One should be aware of myeloma presenting as osteoporosis.

In all cases, a measurement of plasma calcium (preferably ionized), creatinine, and bone turnover is appropriate. Directed biochemical testing, e.g. thyroid-stimulating hormone (TSH) in suspected thyrotoxicosis, or an overnight dexamethasone suppression test in Cushing’s disease may be required. Although vitamin D deficiency is best diagnosed with a double tetracycline-labelled bone biopsy, a low 25-hydroxyvitamin D level with or without a raised PTH level is used as a surrogate diagnostic approach. The markers of bone turnover used should include a measure of bone formation, e.g. alkaline phosphatase, osteocalcin, or P1NP and one of bone resorption, e.g. the urine hydroxyproline/creatinine ratio or serum or urine type I collagen crosslinked C-telopeptide (CTX). Abnormalities in these markers may give diagnostic information as to the cause of the osteoporosis. High resorption may occur in calcium, vitamin D, or oestrogen deficiency. Low formation may occur in corticosteroid-associated osteoporosis. In age-related osteoporosis, bone formation and resorption markers predict future fracture independently of bone density, and may give early evidence of response to treatment before it is possible to detect changes in bone density.

The diagnosis of vitamin D deficiency requires a low level of 25-hydroxyvitamin D. Vitamin D deficiency is defined as a level below 30 nmol/l, whereas vitamin D insufficiency is diagnosed when the level is less than 60 nmol/l. Higher levels of 25-hydroxyvitamin D have been claimed to be beneficial, but as yet there is no clear evidence of clinical as opposed to biochemical benefit. A raised PTH level and high levels of markers of bone formation and resorption, plus low levels of calcium and phosphate, are supportive.

The single most useful marker of tumour-induced osteolysis is raised serum calcium in the presence of low parathyroid hormone, as this combination is almost completely specific for this disorder

Appropriate surgical techniques for immobilization and appropriate medical techniques for pain relief and management of complications must be considered. The rehabilitation of hip fracture patients is a complex area, which requires a proper team approach. Restoration of function after limb fracture requires careful physiotherapy and occupational therapy.

The control of pain following spinal fracture is often badly addressed. In addition to the use of paracetamol opiates may be required in the early stages. A significant proportion of subjects go on to develop a chronic pain syndrome. A correctable source of pain can arise from the facet joints in the area of the vertebral fracture. It is likely that the wedging of the vertebral body sets up significant anatomical strain in the facet joint that may respond to injection with local anaesthetics and long acting corticosteroids. A technique of injection of bone cement in to the fractured vertebral body, called vertebroplasty, has been shown to be effective for short term pain relief (6, 7).

The management of this condition is achieved with calcium at least 1000 mg per day and vitamin D therapy at least 1000 U per day. It may take 6 months to 1 year before the biochemical and clinical defects are corrected.

The management of these disorders is outside the scope of this chapter.

There are now numerous pharmacological, dietary and physical treatments available, which require skill and time to fit to the requirements of the patient. In all cases of osteoporosis, attention to the lifestyle factors outlined in Box 10.1.6.2 is appropriate. Physical activity increases muscle strength and reduces the risk of falling, in particular exercises that involve a component of balance such as Tai Chi. It is important to reduce psychotropic drug administration and to modify the home environment to reduce the risks of falling. In elderly institutionalized patients, the use of energy absorbent pads over the greater trochanter may reduce hip fracture rates, but compliance is low. The specific treatment approach depends on the cause, e.g. the osteoporosis of thyrotoxicosis is best managed by control of the thyrotoxicosis in the first instance.

A variety of effective interventions to prevent osteoporotic fracture exist. The principle mode of action is to reduce osteoclast activity relative to osteoblast activity. The aim is to repair bone surfaces and prevent further bone loss by inhibiting osteoclast-mediated bone resorption, allowing repair of the Howship’s lacunae by osteoblast-mediated bone formation. Agents effective in primary prevention of osteoporosis before the first fracture are also effective in secondary prevention of osteoporosis after the first fracture. This is because the pathophysiological processes are the same. The difference is that the skeletal structure is more severely damaged in secondary prevention, so that the risk of future fracture is much higher; the absolute size of the treatment effect is therefore increased.

There is strong evidence from controlled clinical trials that calcium supplementation, of about 1000 mg of elemental calcium, and vitamin D 1000 U per day reduces appendicular fractures. Calcium supplementation should be introduced in all patients with a diagnosis of osteoporosis. Vitamin D (or cholecalciferol) should be added, especially in subjects with evidence of vitamin D deficiency (8).

More recently, interest has developed in evidence that vitamin D deficiency results in an increased propensity to muscle weakness, and thereby an increased risk of falling (9).

These agents have revolutionized osteoporosis therapy. Bisphosphonates should be considered first-line pharmacological treatment in addition to calcium and vitamin D, if these are not considered sufficiently protective. The chemical structure consists of a backbone of P–C–P atoms with side chains attached to the carbon atom. This structure replicates the pyrophosphate molecule, consisting of P–O–P, which is highly bone seeking. After entering the circulation, 50% of the available bisphosphonates are retained at bone surfaces; the remainder is excreted in the kidney. This accounts for the specificity of these agents, which are retained by bone for long periods of time. Because these agents are less than 1% absorbed, they must be consumed on an empty stomach with water only for at least 30 minutes after consumption.

There are a large number of effective compounds of which etidronate was the first. Alendronate (10), risedronate (11), and ibandronate (12) given orally daily weekly, monthly, or three monthly have efficacy in preventing fractures, especially in subjects with a pre-existing history of vertebral fracture. An intravenous preparation of ibandronate is available. When administered by the oral route, oesophageal reflux is a contraindication to this medication, as severe oesophagitis and gastritis may occur if the mucosa is exposed for long periods of time. Recently, an infusion of zoledronate once a year has been shown to be effective in fracture prevention both before (13) and after hip fracture (14), as has intravenous ibandronate.

Other concerns after long term exposure from bisphosphonates are osteonecrosis of the jaw, perhaps better defined as osteomyelitis of the jaw, after tooth extraction and stress fracture due to a low bone remodelling rate, resulting in subtrochanteric femur fractures.

Recognition that oestrogen deficiency plays an important role in age-related osteoporosis, together with the evaluation of new selective oestrogen receptor modulators (SERMs), which do not stimulate the breast or endometrium, have raised the possibility of treating large numbers of elderly women at increased risk of fracture. Raloxifene, the first of these modified, oestrogen-like compounds, was shown to produce no endometrial stimulation or bleeding, and to reduce the risks of breast cancer while preventing bone loss and vertebral fracture (15). Unfortunately, there may be a small increase in cerebrovascular disease risk.

This is a monoclonal antibody administered by subcutaneous injection every six months, directed against circulating RANK ligand (RANKL). RANKL is a potent cytokine released by osteoblasts and directed at the RANK receptor on osteoclast precursors. It has a powerful effect on fracture reduction (16, 17) in short-term studies; its efficacy in suppressing bone turnover and increasing bone density exceeds that of some bisphosphonates.

Oestrogen has been used for years to treat postmenopausal osteoporosis. Its efficacy was strongly supported by the Women’s Health Initiative trial, which showed a 30% reduction in clinical fracture in patients unselected for osteoporosis (18). However, when used with progesterone to protect the endometrium there was a 50% increase in breast cancer risk. This translates to about a 1.5% risk over 5 years as opposed to a 1% risk, as well as a small increase in cardiovascular risk. Thus combined therapy is not recommended except for those individuals with unacceptable postmenopausal symptoms. In these cases tibolone should also be considered (19).

Oestrogen alone does not increase breast cancer risk, but an increase in cardiovascular risk still remains, especially in those with significant underlying vascular disease, a common problem in the elderly. Other significant deleterious effects include menstrual bleeding, breast stimulation, and deep vein thrombosis. Thus the only usual indication in the elderly is a combination of major postmenopausal symptoms and osteoporosis. Oestrogen therapy should be introduced at low doses (e.g. 0.31 mg conjugated equine oestrogen, 25 µg transdermal oestrogen). Calcium increases the effectiveness of oestrogen treatment.

Testosterone replacement is indicated in hypogonadal men with osteoporosis, especially if bone turnover is elevated. Primary hypogonadism is definite if luteinizing hormone levels are elevated. In the case of hypogonadotropic hypogonadism, if the testosterone is low and there is evidence of osteoporosis, testosterone replacement is indicated. There are various modes of administration including pills, patches, subcutaneous injection, and pellets. No randomized, controlled trials of this therapy have been performed, so the extent of potential adverse events such prostate cancer and cardiovascular disease is unknown. Nevertheless, a trial of therapy is indicated if in addition to osteoporosis there are psychological symptoms.

Nasal calcitonin may be effective in reducing fracture rates in patients with pre-existing spine fracture, but has been largely replaced by newer more effective agents.

This agent was shown to have similar therapeutic benefit to bisphosphonates in reducing fractures, although its mode of action remains unclear (20). It is administered once daily as a powder. It increases bone density in part by incorporation into hydroxyapatite crystals, where because of it greater atomic weight it increases bone mass by an effect independent of bone volume. Thus an increase in areal BMD does not necessarily indicate an increase in bone volume. Side effects include occasional gastrointestinal problems, and occasionally a potentially life-threatening allergic skin rash.

There is much interest in developing new methods of inducing primary bone formation, as occurs in fracture repair, to reform the connected trabecular bone structures important in skeletal strength.

Daily injections of both 1–34 (21) and 1–84 parathyroid hormone have been shown to stimulate osteoblastic bone formation and osteoclastic bone resorption. The balance is much in favour of bone formation, with large increases in bone mass and reduction in fractures recorded. This effect should not be confused with the increased bone resorption that occurs with continuous exposure to high PTH levels in hyperparathyroidism.

Compliance with the treatment regimen agreed upon with the patient is a major therapeutic problem. In general, less that 50% of patients are compliant with therapy. Monitoring after 6–12 weeks to assess acceptability and check for side- effects has been shown to improve compliance (Box 10.1.6.3). In selected cases at high risk of further bone loss, repetition of biochemical markers of bone turnover to assess whether this has been suppressed may give an early indication of treatment failure or success. Bone density measurement should be repeated at 1 year and thereafter at 1–2 yearly intervals. In the case of vertebral osteoporosis, the height should be checked at each review with repeat X-rays of the thoracic and lumbar spine at 1–2 yearly intervals. Review at yearly intervals is likely to improve compliance with treatment.

Continuing high bone turnover, or bone loss greater than three times the coefficient of variation of the method (usually 5%), should prompt a search for other causes of the osteoporotic process and reconsideration of the modality of treatment selected. It is important to explain to the patient who has already developed vertebral osteoporotic fractures that current treatment modalities only slow the rate of development of new fractures by about 50%, so that further fractures are likely to occur but at a reduced rate. Nevertheless, continuing fractures should prompt a review of the treatment regimen.