Can Falls and Hip Fracture Be Prevented in Frail Older Adults?

The recent introduction of potent bone antiresorptive agents has focused strategies for the prevention of hip and osteoporotic fractures on the prevention and treatment of bone loss. Agents such as alendronate and raloxifene have been very effective in the prevention of osteoporotic fractures of the spine, with reductions of 40 to 48% being reported.1–3 However, reductions of nonvertebral fractures and hip fractures have not been significant, ranging from 3 to 20%. Vitamin D, on the other hand, which has modest effects on bone mineral density, seems to have disproportionate and significant effects on the reduction of nonvertebral fractures and hip fractures.4,5 Supplementing 3270 women in a French nursing home population with 800 units of vitamin D resulted in a 43% reduction in the incidence of hip fracture compared with placebo treatment.4 In men and women older than age 65 residing in Boston, 700 units of vitamin D supplementation resulted in a 55% reduction in nonvertebral fractures over 3 years.5 One has to be very cautious in comparing the French and the Boston studies with largely American studies because of differences in vitamin D status and the ages of the populations. Nevertheless, it is difficult to attribute all of the vitamin D effect on nonvertebral fracture incidence to an increase in bone strength (density). The changes in bone density at the hip achieved with vitamin D supplementation are less than those seen with bisphospho-nates. One clue to this paradox may be that in contrast to the spontaneous vertebral osteoporotic fractures, nonvertebral osteoporotic fractures are almost invariably preceded by a fall. Melton and colleagues suggest that after age 70, age-related factors other than bone density become the dominant determinant of hip fracture risk.6 An attractive candidate for these age-related factors is changes in the central nervous and neuromuscular system that predispose older adults to injurious falls. A slowing of the central processing of sensory information signaling a loss of balance and the generation of an appropriate and timely postural response seems to be a dominant determinant of injurious falls7 and a significant risk factor for hip fractures.8 With advancing age, especially after age 70, it is postulated that this slowing of central processing speed results in the failure to extend the forearm in time to break the fall. Thus, the full energy of the fall is more likely to be directed to the hip. The energy generated by a fall from a standing height is 2 to 3 times that necessary to fracture the average 65+ -year-old hip.9 Epidemiologic data supports this hypothesis.10 Whereas both osteoporotic fractures of the hip and wrist rise in parallel in women after age 50, after age 70, wrist fractures plateau or decline and hip fractures increase exponentially. Falls then become an increasingly important component of the hip fracture equation with the advancement of age. It is this component of the hip fracture equation on which vitamin D may be acting, especially in older people, who are at greatest risk of being vitamin D deficient. The report by Stein and colleagues in this issue of theJournalprovides evidence for an important link in the chain of logic that potentially connects vitamin D supplementation to the reduction of hip fracture. They have demonstrated that 25-hydroxyvitamin D (25-OHD) and parathyroid hormone levels (PTH) are associated with falls in frail older adults who are residents of either nursing homes or assisted living facilities. Residents who fell had lower levels of 25-OHD) (P < .02) and higher levels of PTH (P < .01). In a multiple logistic regression for falling, higher serum PTH remained independently associated with falling with an odds ratio for falling of 5.6 (CI, 1.7-18.5). As expected, independence in living and walking were significantly protective of falling. In their multiple regression, 25-OHD drops out, suggesting that PTH is an independent risk factor for falls. Indeed, there is evidence that PTH has direct effects on muscle and brain function as noted by the authors. An alternative explanation of PTH remaining in the multiple regression is that it may be a better biological marker than 25-OHD for vitamin D deficiency at the tissue level. Serum levels of 25-OHD do not reflect the tissue response to the sterol. This becomes particularly relevant when we consider the variable and increasing resistance to vitamin D with hormonal status and advancing age. In addition, multiple alleles in the vitamin D receptor protein gene also determine tissue responsivity. The expression of the 1,25-dihydroxyvitamin D receptor protein is dependent on estrogen.11 This dependence of the receptor protein on estrogen may be responsible for the increased requirement for vitamin D and calcium that occurs at the time of the menopause. To what extent then can we attribute the reduction in postural sway12 and falls13 observed with estrogen to the hormone's effect on increased expression of the 1,25-dihydroxyvitamin D receptor protein? An effect of estrogen on the tissue response to vitamin D raises the possibility of a synergistic interaction between these two hormones in the prevention of nonvertebral fractures. The answer to this intriguing possibility will soon be available with the report of the results of the STOPIT14 trial, which examines the effects of estrogen and vitamin D on nonvertebral fractures. The results of the MORE Trial' exploring the effects of raloxifene on osteoporotic fractures in more than 7000 women over 3 years are also of interest. In a population similar to the alendronate Fracture Intervention Trial (FIT),1 raloxifene reduced vertebral fractures by 40%, comparable to the reduction observed with alendronate. However, in contrast to alendronate, nonvertebral fractures were reduced by only 3%, and hip fractures increased by 14%.3 Although these changes were not significant, they do raise the concern that this agent, which causes hot flushes, may be acting as an antiestrogen on the CNS, thus increasing the risk of injurious falls and offsetting the gains in bone strength reflected in the reduction of vertebral fractures. Can we prevent hip fractures in frail older persons? Vitamin D deficiency (resistance) as discussed by Stein et al. seems to be an increasingly important determinant of hip fractures in adults older than age 70, who account for 90% of all hip fractures.15 Conversely, the age-related decline in bone strength (density) becomes increasingly less important in determining the risk of hip fracture with advancing age. Intervention strategies to prevent hip fracture that affect only the changes in bone strength (density) and not those factors resulting in falls may be of declining importance with advancing age and frailty. Secondly, agents that affect brain aging adversely, such as the selective estrogen receptor modulators, may actually increase falls and hip fractures. Thus, effective interventions to reduce hip fractures must, therefore, either reduce the liability of falling or, alternatively, reduce the energy of the impact of the fall on the hip. Vitamin D supplementation promises to achieve the former. The article by Stein et al. in this issue of the Journalis of importance because it provides the rationale for a safe and cost-effective intervention strategy for the prevention of nonvertebral fractures in frail older people. Investigators should now be encouraged to initiate prospective intervention trials to validate the associations proposed by Stein et al. and to determine the efficacy of vitamin D supplements in the prevention of falls in older people.