Hip and Vertebral Fractures – Demographic and Gender-Specific Factors, Social and Economic Factors, Falls
Osteoporosis is an asymptomatic condition, which exists only as a pathological or (disputably) radiological entity until and unless structural mechanical failure under loading leads to fracture of one or more bones. The epidemiology of male osteoporosis has been little studied until quite recently (Dennison and Cooper, 1996; Melton etal., 1992; Kannus etal., 1996), and there are only a small number of studies addressing the clinical and social consequences of fracture in men. In this post we will review the evidence available on the global epidemiology, clinical impact, and outcome of the two fracture types most clearly associated with skeletal fragility in men— vertebral compression fractures and femoral neck fractures.
II. Demographic Factors
The incidence of femoral neck fracture in men is rising in all countries for which data are available, and the rate of increase is itself accelerating in many countries (Obrant etal., 1989; Cooper etal., 1992a). A similar though less marked pattern is seen in the incidence of symptomatic vertebral fracture in most countries, although for reasons discussed later the trend appears to be leveling off in some developed nations. The most powerful force driving these increases is undoubtedly the extension of average lifespan which has occurred in all parts of the developed world and in many developing nations (Lau and Cooper, 1996; Kalache, 1996) during the last hundred years.
Not only has average life expectancy risen by decades in this time, but the age structure of populations has also undergone qualitative change. This effect is especially marked in countries which are in transition from “less developed” status to a social and economic structure more in line with the Western industrialized nations. These countries, of which Singapore is a good example, typically see a slowing of overall population growth accompanied by a rapid increase in the proportion of their society composed of older people (Kalache, 1996). It is often asserted that the “dependency ratio” of such a society is increasing (i.e., the ratio of economically dependent adults to economically productive adults), and this perception underlies recent efforts in many developed countries to control health care expenditure.
However, if one takes a broader view of the relevant population demographics by including children, it can be seen that the ratio of dependent citizens to productive adults is changing in a less straightforward way, and may even be falling in some countries. This effect is associated with a redistribution of the disease burden in a given society, from infectious diseases affecting predominantly children and younger adults to degenerative diseases of old age.
Insofar as the slow decrease in bone mineral density (BMD) from the fourth decade onward can be regarded as an age-related change rather than a pathological event, the rising absolute number of osteoporotic fractures in late life is inextricably though not exclusively linked to the absolute number of older people in a population. However, large changes in the relative prevalence of such conditions will occur only in the context of the change in age structure described previously, and it is this relative change which is of social and economic consequence to a society.
III. Gender-Specific Factors
The incidence and prevalence of femoral and vertebra! fractures in men differ appreciably from that in women in most populations studied. There is a particularly striking pattern in prevalence of vertebral deformity, with reversal of the sex ratio with increasing age (O’Neill et al., 1996). Gender-specific reasons for this disparity fall into three main categories: differences in skeletal growth in childhood, differences in the hormonal milieu during adult life, and differences in exposure to major trauma.
A. Skeletal Growth
The attainment of peak bone mass in both sexes is almost certainly controlled both directly by genetic factors and indirectly by genetic control of hormones and their receptors (Smith etal., 1994; Morishima et al., 1995; Kasperk et al., 1997). Skeletal size is clearly a heritable characteristic, but it is not clear whether the larger size of the average male skeleton is a result of specific habitus-determining genes (which would have to be on the Y chromosome) or whether men are constitutionally larger because of the effect of sex hormones on the agents of bone growth. The general observation that childhood growth is roughly comparable in boys and girls before the onset of puberty suggests the latter. Prevalent vertebral deformation is more common in young men than in young women, probably mainly as a result of trauma, but it may be that the axial skeleton of men in their early twenties is more sensitive to traumatic damage than women’s skeletons.
In eugonadal men, growth of the vertebral bodies continues for about five years longer than in women, and active haematopoietic (red) marrow is present in regions of trabecular bone formation, giving rise to a rich supply of bone cell precursors in proximity to active growth plates. There is fairly good evidence from trials of antiresorptive drugs in osteoporosis that bone turnover rates and the proportion of total bone volume participating in remodeling are risk factors for vertebral fracture independent of measured bone density (Ravn et al., 1997; Riggs et al., 1996), and it seems reasonable to speculate that resistance to trauma is lower in the spines of men than in women during a period of a few years in early adult life when injuries are fairly common. This might lead to a greater accumulation of minor vertebral deformations than in women of the same age whose epiphyses have closed, which are then detected in radiographic surveys many years later as prevalent deformities (O’Neill et al., 1996).
The effect on peak bone mass of defects at any point in the pathways by which sex hormones act is clear, with testosterone deficiency leading to late epiphyseal closure, increased height, and low bone mass (Horowitz et al., 1992) . Recently it has become apparent that “female” sex hormones are also necessary for normal skeletal growth in men (Smith et al., 1994; Selby et al., 1996) , a role mediated by stochastic conversion of testosterone to oestradiol by the cytochrome P-450 enzyme aromatase. Failure of this pathway, either because of aromatase inactivation or oestrogen receptor deficiency, leads to a physical and skeletal appearance similar to that seen in hypoandrogenic states, with delay of epiphyseal closure, low peak bone mass, and vertebral fractures (Morishima et al., 1995;Soule et al., 1995; Carani et al., 1997).
Risk factors other than bone mass and density are heritable; skeletal structure at the gross level is clearly under genetic control, as shown by variations in hip axis length according to race and family history (Nelson et al., 1995; Cummings et al., 1994a).
B. Hormonal Influences in Adult Life
Skeletal growth is complete by the middle of the third decade, and after a few years of balanced bone turnover men enter a state of net bone resorption which persists for the rest of their lives. This bone loss may be an inevitable and direct consequence of skeletal aging, or it may plausibly be a result of reducing physical activity with advancing years (Silman et al., 1997) . There is considerable debate over whether the rate of bone loss is more rapid in hormonally replete women than in men during this period, and there appears to be more variation between individuals than between the sexes. What is not in doubt is that once women enter the perimenopause they begin to lose bone mass—and also bone microstructure—at a rate which rapidly and permanently increases their fracture risk, both absolute and relative to men.
Gonadal failure in men is uncommon until late life and is not a normal consequence of aging even then. The absence of a universal “male menopause” is the dominant factor associated with the lower incidence of fragility fracture in men of all cultures and races studied. Nonetheless, gonadal failure for any reason is associated with a deterioration in bone quality (Stepan et al., 1989), and some degree of gonadal dysfunction is commonly detected in men with both hip (Jackson and Spiekerman, 1989; Stanley etai, 1991) and vertebral (Baillie etal., 1992) fractures. The degree of hypogonadism is often not otherwise clinically apparent, and appropriate endocrine assessment is a necessary part of the investigation of any man with vertebral fracture and is at least advisable in all but the most elderly men with hip fracture.
The prevalence of vertebral fracture is greatly influenced by the appreciably higher risk of traumatic fracture in men of most societies studied. Because the radiological techniques used in many epidemiological studies offer no insight into the causation of a vertebral abnormality, it has been difficult to distinguish between fragility-related and traumatic fracture with confidence. However, some much-needed light is being shed on this area by results from recent research such as the European vertebral osteoporosis study (EVOS) (O’Neill et al., 1996), although there are inevitably problems of recall bias, survivor effects, and all the usual hazards of retrospective analysis. It not only appears fairly clear that the risk attributable to trauma varies according to the fracture site studied, but it also seems much higher in studies of prevalent vertebral deformity than with clinical vertebral fracture or appendicular fractures at any site (Silman et al., 1997; Cooper et al., 1992b, 1993a; Melton et al., 1993). Societal influences on the risk of trauma are discussed later.
IV. Factors Affecting Case Definition
There is at present no universally agreed upon definition of a vertebral fracture, which might be expected to hinder any attempt to assess incidence or prevalence.
Prevalent vertebral deformity, assessed by spinal radiography, has been the main outcome measure of international epidemiological studies (O’Neill et al., 1996; Cooper et al., 1993a; Melton et al., 1993) and offers a useful comparative measurement for studies across populations, but only if exactly the same technique is used to interpret the radiographs in all the groups studied. Two techniques for quantitative assessment of vertebral deformity using standardized lateral thoracolumbar radiographs have been used in population studies including men, that of McCloskey et al. (1993) and Eas- tell et al. (1991). Differences in techniques for reading the radiographs mean that results obtained using the two techniques are not identical and that of Eastell generally gives higher prevalences. This may account for some part of the differences in prevalent vertebral deformity between European countries and the United States.
The task of accurately measuring the structural strength of a complex, irregularly shaped, partially hollow object of nonuniform composition which is exposed to loading in a variety of planes is one to give any materials scientist nightmares; in practical terms, it simply cannot be done. Fracture risk assessment is therefore based on indirect measures of bone strength, principally radiological measurement of bone density. In developed countries, the peak skeletal mass of men is 20-40% greater than in women, but “true” bone density is much the same when calculated by measurement of ash weight and bone volume in cadaveric specimens (Nielsen et al., 1980). This indicates that much of the difference between bone mineral content in men and women is associated with systematic differences in bone size. The relationship between bone size and bone density is neither simple nor linear, and fracture risk prediction in men cannot rely too heavily on a technique which has been validated mainly in women without taking this into account. There are particular problems in the interpretation of BMD results obtained by dual energy X-ray absorptiometry (DXA), which is vulnerable to systematic differences in bone size because results are expressed as “areal density” (mineral content per unit area), which does not adjust for bone size in the long axis of measurement.
An additional problem with DXA measurements in anteroposterior lumbar spine views is the presence of calcification outside the vertebral bodies but within the radiological “region of interest.” This “irrelevant” calcium is mainly associated with osteophyte formation at inter- vertebral joints with a small contribution from calcification in the aorta, both of which are more prevalent in men than in women (Liu et al., 1997). Both factors elevate measured BMD and thus interfere with attempts to generate risk ratios. Although authoritative working parties have proposed a classification of DXA results in men with the aim of allowing risk stratification in the same way as with women (Miller et al., 1996), the link with fracture risk is not nearly as well supported from research evidence. DXA measurement forms the basis of the World Health Organization risk classification for osteoporosis in women, but there is no equivalent statement in relation to men. DXA measurements at the hip can, however, validly be used in men to assess change in density over time. Of the other techniques available, only quantitative computed tomography (QCT) routinely estimates true volumetric density, but the extra information obtained on bone size has not generally been incorporated into fracture risk assessment because of lack of evidence regarding the nature of the relationship. In any case, the relatively large radiation dosage associated with QCT has proved a barrier to widespread acceptance of the technique in lower risk subjects.
It is estimated that only about 30% of incident vertebral deformations are diagnosed contemporaneously (Ross, 1997). This does not necessarily mean that the other 70% of events are asymptomatic; many of those afflicted probably resort to self-medication and rest and an unknown proportion are misdiagnosed by their medical practitioners as having a lesion of the spinal ligaments or intervertebral discs. The proportion of those with back pain who present to a doctor and the proportion whose lesion is correctly diagnosed at presentation are both highly likely to vary according to social factors and medical knowledge. The increasing awareness among the health care professions and their patients of osteoporosis as a condition relevant to men has very probably affected the reported incidence of clinical vertebral fractures, although the size of this effect is by definition impossible to judge. If a point is reached where most medical practitioners are aware of osteoporosis, there will be a leveling-off of reported rates regardless of the underlying incidence. Data on clinical vertebral fractures, particularly on secular trends in incidence, must therefore be viewed with considerable caution.
V. Social and Economic Factors
Many aspects of the structure of a society affect the risk of fracture, acting at various points on the pathway from risk of trauma to skeletal fragility. Most of these effects are at least partly gender-specific; occupation, diet, exercise and risk behaviors are all influenced by social attitudes to gender roles. Economic conditions prevailing during the working lifetime of a typical male citizen may have interacting influences on the risks of vertebral fracture; for example, in recessionary conditions, unemployment reduces the risk of exposure to trauma but reduces activity levels and increases the likelihood of poor dietary intake of important nutrients.
The aetiology of trauma of sufficient severity to cause vertebral deformation in men with “normal” bone mineral density and no other conditions detrimental to bone quality is very poorly understood. A review of the literature would suggest that most of these injuries are a result of playing hazardous sports, but this clearly reflects publication bias. Among the factors which are capable of accounting for the very large numbers of vertebral deformities detected in EVOS, the strongest contender is occupational injury, particularly among workers in the classic heavy industries of steelmaking, shipbuilding and mining, the construction industry, and agriculture. This kind of personal history will probably be familiar to many clinicians who see male patients with vertebral deformities, but the evidence is sparse to nonexistent. Road traffic accidents may also be a significant factor. There are therefore significant social and demographic influences on the prevalence of traumatic vertebral fracture between populations depending on the size of the labor force in “at risk” occupations within each country.
There is evidence in the results from the EVOS study (O’Neill et al., 1996) that European countries which have undergone a transition to an economy based on value-additive manufacture and service industries have a lower total prevalence and a different age distribution of male vertebral deformation when compared to countries whose economies still rely primarily on extractive industries and heavy engineering, but the association with occupation is not proven as yet. One unavoidable gap in the EVOS data is that it tells us relatively little about vertebral fracture prevalence in predominantly agrarian societies because none of the participating centers were based in countries with a substantial proportion of the workforce employed in agriculture— the agricultural workforce within countries studied ranging from less than 1% of the total up to a maximum of 22% in Greece (Anonymous, 1998).
In a given economy farm workers will benefit from sunlight exposure, regular physical exercise, and possibly better diet than an office or factory worker. This is partially offset by lower socioeconomic status for agricultural workers in most cultures and by a greater risk of trauma. Data on fracture prevalence and BMD collected from agrarian populations in less developed countries (Lau and Cooper, 1996; Solomon, 1979; Bloom and Pogrund, 1982; Aspray et al., 1996) cannot be compared directly either to nonwhite urban populations (Daniels et al., 1995; Cummings et al., 1994b) or to European or U.S. data because there are invariably large differences in ethnic composition or social structure of the populations studied. Direct comparisons could perhaps be made where ethnically concordant populations have been exposed to discordant social and environmental influences—one example would be Chinese citizens of Hong Kong island and of Guangdong province on the mainland—but there are too many other factors involved for this to be a straightforward undertaking.
The dietary components which are relevant to skeletal health—calcium, vitamin D, and protein—are obtained from a variety of sources. There are large variations among societies in the availability of these nutrients, and within a society there are often culturally determined gender-specific differences in their distribution, particularly in times of hardship. Although severe protein-energy malnutrition is now rare in developed countries, this has not always been the case during the lifespan of our older citizens. Whether the nutritional needs of males or females were given priority at such times may have affected skeletal health, especially in those who were peripubertal at the time. However, the effect of a given set of adverse circumstances may be unpredictable and even counter-intuitive, for example, the average British diet during the period of food rationing during and after the Second World War was appreciably better in nutritional terms than in the preceding years because of a more equitable distribution of supplies and the national implementation of nutritional guidelines. Serving military personnel (mostly young men) were favored in food distribution, as is often the case in times of conflict. However, women who contributed to the war effort by becoming agricultural workers (the “Land Army”) had opportunities to supplement their rations which were not available to those working in manufacture.
It is clear that cross-sectional dietary surveys must be limited by the complex interplay of cultural factors within and among societies during the lifespan of older adults.
Exercise is important for skeletal growth and maintenance, and in Western societies this century, men have had a greater average level of physical exercise than women. Physical labor as part of an occupation has been much more important than recreational activity until recently, but the number of men engaged in physically taxing employment has been falling for a generation, and the gender difference is narrowing (Cooper etal., 1990). However, this observation cannot be generalized to other cultures; in many societies, particularly those reliant on subsistence agriculture, women perform much more physical work than men (Anonymous, 1998). Continuing physical activity in late life is consistently shown to be negatively correlated with fracture risk, but causality is difficult to establish for this association because of the potential for confounding by poor general physical health (Grisso et al., 1997; Nguyen etal., 1996;Jaglal etal., 1993).
D. Leisure Activities
Smoking and excessive alcohol consumption are important risk factors for osteoporosis and fracture (Kiel et al., 1996; Diaz et al., 1997). Both activities are more prevalent in men in all societies where they are permitted.
Sporting and other outdoor leisure activities are also pursued more frequently by men and would generally be expected to have a favorable effect on bone health as a result of exercise and sunlight exposure. Unfortunately, this benefit is probably at least partially offset by an increased risk of fracture as a result of trauma (Silman et al., 1997).
The vast majority of osteoporosis-related fractures occur as a result of a fall. Factors which influence the likelihood of falling are therefore important contributors to overall fracture risk, especially in older patients and those who have already sustained a nonvertebral fracture. Separately, factors associated with protective responses and with force transmission from the im-pact site are highly relevant to femoral fracture risk (Table I). Studies have shown an increased risk in association with markers of frailty such as use of a walking aid and use of medications which impair neuromuscular coordination (Grisso etal., 1997; Nguyen etal., 1996; Dargent-Molina etal., 1996; Poor et al., 1995a; Cummings et al., 1995; Jacobsen et al., 1995; Ranstam et al., 1996). Force transmission is influenced by height and weight independently of BMD (Grisso et ai, 1997; Nguyen et al., 1996; Cummings et al., 1995) , and the geometry of bone affects the likelihood of structural failure in response to a given force, the best documented example being hip axis length (Karlsson et al., 1996). By comparison with vertebral fractures, the incidence of most types of nonvertebral fracture is more sensitive to falls risk and rather less affected by factors acting on BMD.
TABLE I Risk Factors for Hip Fracture in Men“
Adjusted odds ratio•
Femoral neck BMD
— 1 s.d.
+ + +
+ + +
- 1 s.d.
Impaired activities of
daily living (ADI.)
‘Risk factors for fracture are a combination of falls risk, general frailty and bone quality. Those factors which have been associated with fracture risk in men are shown, indicating the odds ratio and making a tentative allocation of mechanism to each risk. Data are mainly from Grisso et ill. (1997) and Nguyen at al. (1996) with supporting information from other studies (Dargent- Molina ct al., 1996; Poor ct al., 1995a; Cummings et al., 1995; Jacobsen et al., 1995;Ranstam etal., 1996).
‘’Unit of difference from the mean/median, or categorical label.
‘Crude O.R. adjusted for BMD where available.
‘Cognitive impairment; values shown are mid-range versus normal.
‘’Composite scores for multiple risk factors; values arc middle category and highest two categories versus lowest two categories.
VII. Consequences of Osteoporotic Vertebral and Femoral Fracture
Of all the consequences of osteoporosis, proximal femoral fracture is associated with the greatest morbidity, mortality, and economic consequences (Barrett-Connor, 1995). About 20-25% of proximal femoral fractures occur in men (Cooper et al., 1992a; McColl et al., 1998), and these fractures account for more than 85% of the total economic impact of osteoporosis in both sexes (Ray etal., 1997). Although some proximal femoral fractures are associated with major trauma, there is a consensus that the vast majority occur in the context of clinically important loss of skeletal integrity due to osteoporosis (Melton et al., 1997). The clinical consequences of proximal femoral fracture are early mortality and long-term morbidity. Mortality in the first six months after fracture is 18-34%, which is to 10 to 20 percentage points higher than in age-matched controls, and men appear to have a significantly higher mortality than age-matched women (Poor et al., 1995b,c; Sembo and Johnell, 1993).
However, in both personal and economic terms, the morbidity of femoral fracture is perhaps even more daunting than the mortality. Six months after the event, more than half of patients are still in pain and require assistance with walking (Sernbo and Johnell, 1993). Loss of independence is nearly universal because of both decreased mobility and loss of confidence (Jensen and Bagger, 1982), with one-third of patients moving into residential care or living permanently with relatives (Poor et al., 1995c; Jensen and Bagger, 1982). The economic impact of femoral fracture in men can be calculated from available data as approximately one-fifth of the total cost of osteoporosis, or about $2.75 billion annually in the United States alone (Ray et al., 1997; Dolan and Torgerson, 1998).
Vertebral fractures are associated with a much smaller proportion of the economic costs of osteoporosis, but morbidity is considerable (Ross, 1997; Kanis et al., 1992; Burger et al., 1997; Scane et al., 1994). In a study from the United Kingdom of men questioned at least six months after a symptomatic vertebral fracture (Scane et al., 1994), three-quarters reported sleep disturbance by pain and half were still using analgesics every day. No figures are available for loss of earnings. There are few deaths attributed directly to vertebral osteoporosis, but survival curves show a continuing excess of mortality during follow-up after incident fracture, which is probably due mainly to underlying disease to which the patient’s vertebral osteoporosis is secondary or incidental (Cooper et al., 1993b). Nonetheless, death from vertebral osteoporosis is not rare in male sufferers and is usually a result of a combination of elements such as respiratory compromise, toxicity from prescribed and other medications, and/or depression.
Specific pharmacological therapy for male osteoporosis is discussed in a later post, but in many patients with clinically important disease, drug therapy for osteoporosis is experimental, inappropriate, or ineffective. In almost all cases, antiresorptive therapy is reserved until the patient has been thoroughly investigated and other appropriate measures have been taken.
A. Pain Control
Analgesia is the cornerstone of the early management of osteoporotic fracture. Almost all patients require pain relief, and, in the initial stages after any fracture, strong opiates are usually indicated. The WHO “ladder of analgesia” is a reasonable guide to systemic treatment in the acute phase, and the administration of epidural, spinal, or regional anaesthesia can be of great benefit. In patients with chronic pain after vertebral fracture, the same initial approach is followed, but pain relief will be inadequate in a significant minority of patients. There is little evidence to support the use of different analgesics in combination or adjunctive agents such as tricyclic antidepressants, and polypharmacy may do more harm than good. Nonetheless, most practitioners who regularly see osteoporosis sufferers have their own empirical approach, often including combinations such as opiates and nonsteroidal anti-inflammatory drugs. The antiresorptive agent calcitonin appears to have analgesic properties when used for the acute treatment of vertebral fracture (Ljunghall et al., 1991), but evidence of its superiority over other analgesics is limited, and it is costly. The role of nonpharmacological therapy for pain relief is unclear and somewhat controversial, but a trial of transcutaneous electrical nerve stimulation (TENS) will be of benefit to some. Physiotherapy, particularly hydrotherapy, may be of considerable value where there is a large component of secondary muscle spasm. Acupuncture and complementary medicines have their advocates, but proof of efficacy is even more sparse than for “conventional” therapies. Osteopathic and chiropractic manipulation would seem likely on the face of it to be actively harmful, but in fact there is no convincing evidence either way.
Assessment of the risks and benefits of surgical intervention is appropriate for all nonvertebral fractures as soon as immediate analgesia has been given. In the case of proximal femoral fractures, the choice is stark: operative fixation of the fracture or nurse the patient in bed until their death, which is unlikely to be long delayed. The effective delivery of immediate care requires cooperation among specialties and among professions, with important roles for emergency room staff, physicians with acute elderly care experience, anaesthetists, orthopaedic surgeons, and nursing staff. The patient’s condition should be optimized for surgery as quickly as possible, and repair of the fracture should follow at the earliest opportunity. There is compelling evidence that the steps taken in the first 48 hours after a hip fracture have a significant effect on longer-term morbidity and resource usage (Millar and Hill, 1994; Parker and Pryor, 1992; Audit Commission, 1995). However, no effect on early mortality has been demonstrated, suggesting that there is a subgroup of patients whose other morbidities may be too great to overcome no matter how optimal their care. After surgery, if analgesia is carefully titrated to the patient’s needs, it should be possible to begin mobilization on the second postoperative day. Length of stay, after plunging dramatically in the 1970s, has continued to fall steadily in recent years with the appreciation that early repair and mobilization shortens recovery time in survivors (Millar and Hill, 1994).
C. Reduction of Falls Risk
Multiprofessional assessment of factors contributing to the risk of falls is an essential part of the management of patients who have sustained a low- trauma fracture (Tinetti et al., 1994; Clemson et al., 1996; Parker et al., 1996) . In the case of older patients, components of this would typically include physician review of medications, physiotherapy training to improve muscle tone and gait, occupational therapy assessment of safety in the home, and liaison with primary care workers in the community for monitoring of events after discharge from hospital. Where risk of further misadventure is high, it may be appropriate to consider moving the patient to a more protected environment such as residential care or a relative’s home; however, among the many problems with this course of action is obtaining consent from all involved parties. Older people place an extremely high value on their home environment (Sixsmith, 1986) and may lose the will to live if displaced—any change of address is associated with a significant early mortality.
D. Reduction of Forces Acting at the Impact Site
It is well recognized that factors affecting the impact force associated with a fall onto the hip have a major influence on fracture rates (Hayes etal., 1993; Maitland et al., 1993; Einhorn, 1992). Since the introduction of hip protectors in the late 1980s, there has been considerable interest in the use of strategies to minimize the likelihood of fracture in the event of a completed fall. Lauritzen et al. (1993) showed that in residents of nursing homes the wearing of hip protectors conferred almost complete immunity from proximal femoral fracture; unfortunately, there are practical difficulties with this approach, not the least of which is persuading at-risk older people to wear their hip protectors consistently enough to derive benefit (Villar etal., 1998; Ekman et al., 1997). Compliance with this intervention is in the range 30-50%, although it is fair to say that long-term compliance with most pharmaceutical therapies is no better.
IX. Future Trends
The enormous increase in the incidence of osteoporosis-related femoral and vertebral fractures in men is a problem which is extremely likely to get worse before it gets better. There are probably several distinct phenomena underlying the rising fracture rates seen in men in most developed countries. First, and most influentially, aging of the population brings more “low-risk” men into the osteoporotic range of bone density, which has a disproportionate effect on femoral fractures. Secondly, the rising age-specific rates reflect a composite of various factors influencing bone health and trauma risk whose importance we cannot easily judge; these include dietary, occupational, and social trends. Finally, increasing awareness of osteoporosis in men, coupled with universal access to health care in many developed countries (except perhaps the United States, where high-risk low-paid workers are most likely to be uninsured), is probably resulting in enhanced detection of vertebral fracture. The leveling-off of vertebral fracture incidence rates which has recently begun to emerge (Melton et al., 1996) may represent a true effect on osteoporosis (such as dietary improvement in the last 50 years), an effect on trauma caused by occupational change or the reaching of a “steady state” of diagnostic accuracy. Most probably it reflects all of these plus other factors which are as yet unknown.
Management of vertebral fracture depends on good analgesia and the prevention of further fracture by appropriate treatment. There is scope for great progress in this area, mainly because of our lack of existing knowledge as much as the current intensity of interest in male osteoporosis. Antiresorptive agents such as bisphosphonates are likely to become established, and new drugs acting by unrelated mechanisms will probably emerge.
Proximal femoral fracture will continue to kill a large number of elderly men, and in those in whom it occurs simply as a terminal event in a process of general physiological decompensation, we have little to offer except the basic Hippocratic values of care and comfort. However, for the larger proportion of patients whose physiology is salvageable, we can realistically aim for improved survival and reduced morbidity as a result of integrated multiprofessional care by the many health care workers involved.