OSTEOPOROTIC FRACTURES HAVE GENERALLY BEEN defined as fractures that occur following relatively low trauma, such as a fall from standing height or less. Implicit in this definition is the assumption that this type of skeletal trauma would not result in fracture of a normal bone, since the strength of the bone should be able to withstand such force. A surrogate marker for bone strength now used and considered the current gold standard for diagnosing osteoporosis prior to the occurrence of a fracture is calculation of areal bone mineral density (BMD) by dual-energy x-ray absorptiometry. This technique measures the amount of mineral in bone (in grams), typically at the lumbar spine, proximal femur, or distal forearm, and divides this amount by the projected area of the bone, providing a density expressed in grams per centimeter squared. Low BMD is a robust predictor of osteoporotic fracture risk; conversely, low-trauma fractures are related to BMD. In contrast to low-trauma fractures, the conventional view has been that high-trauma fractures, such as those related to a motor vehicle crash or fall from greater than standing height, should not be considered “osteoporotic” fractures, since even normal bones might be expected to fail under these greater loads. While plausible, this concept has not been rigorously tested. Moreover, the definition of hightrauma fractures has been fairly broad, ranging from severe forces that might cause any bone to fracture, to falls from a chair height that might not uniformly cause fracture. The specific definition of high-trauma fractures is of considerably more than academic interest, because current recommendations are to evaluate patients with lowtrauma fractures at any age for an underlying metabolic bone disease, particularly osteoporosis, but not to pursue this assessment for patients sustaining high-trauma fractures. In addition, this definition has far-reaching consequences for the design and analysis of clinical trials involving prevention or treatment of osteoporosis. High-trauma fractures are generally excluded as end points in trials, a factor that has implications for determining the effects of a particular intervention on fracture risk. In this context, the study by Mackey and colleagues in this issue of JAMA is of particular interest and has potential clinical importance. The authors used data from 2 large prospective cohort studies in the United States, the Study of Osteoporotic Fractures (SOF) in women and the Osteoporotic Fractures in Men Study (MrOS) in men to test whether there was an association between BMD and high-trauma fractures and whether high-trauma fractures increased risk of subsequent fracture in older women and men. The somewhat surprising, and perhaps counterintuitive, finding was that among women each 1-SD reduction in BMD was associated with virtually identical increases in risk of high-trauma fractures (multivariable relative hazard [RH], 1.45; 95% confidence interval [CI], 1.23-1.72) and low-trauma fractures (RH, 1.49; 95% CI, 1.42-1.57). Moreover, among women, the risk of subsequent fracture following a high-trauma fracture was the same as that following a low-trauma fracture (RH, 1.31; 95% CI, 1.20-1.43). Thus, highand low-trauma fractures (at least based on how these are currently defined) increased risk of subsequent fracture to exactly the same extent. Similar trends were observed among men, although power was limited due to fewer fractures in men. The study by Mackey et al has several strengths. It involved large numbers of patients (8022 women and 5995 men) who were followed up for fairly long periods (9.1 and 5.1 years, respectively). The determination of fractures appeared complete, including data from questionnaires, personal interviews, and clinical records coupled with radiological confirmation. However, there also were several limitations. Spine fractures were not evaluated, so the findings of the study might not be applicable to lowvs hightrauma spine fractures. In addition, all participants were 65 years or older, so the results may not apply to younger individuals. A previous retrospective case-control study of Australian women older than 50 years with high-trauma fractures also found that these women had lower BMD at the hip, spine, forearm, and total body sites as compared with control women.
Read full abstract