Single BMD measurement, fracture history predict long-term fracture risk
A single bone mineral density (BMD) measurement and fracture history were able to predict the long-term risk of hip and nonvertebral fracture in postmenopausal women, according to a recent study.
“[The findings]... suggest that a single fracture risk assessment in a woman aged 65 and over may be useful in estimating her risk over 20 to 25 years. [U]nless a major change in health that impacts bone health occurs ... a single assessment of BMD may continue to be predictive of fracture risk for as long as 25 years,” said the researchers.
The study population comprised 7,959 women (mean age, 73.4 years) from the US-based Study of Osteoporotic Fractures (SOF) cohort. Researchers obtained DXA* scans of the right proximal femur and data on demographics, lifestyle, medication use, comorbidities, and fracture history. At baseline, 38.4 percent of women had experienced at least one nonvertebral fracture while 2.3 percent had an incidence of hip fracture since age 50 years.
Patients were followed up for 20 years for nonvertebral fractures and 25 years for hip fractures, where 43.7 and 15.9 percent of women experienced at least one incidence of nonvertebral or hip fracture, respectively (cumulative incidence of 46.2 and 17.9 percent, respectively).
Between the lowest and highest BMD quartiles, there was an almost fivefold risk of hip fracture (relative hazard [RH], 4.9; RH, 2.0 per standard deviation [SD] decrease in BMD), and a twofold risk of nonvertebral fractures (RH, 2.4; RH, 1.5 per SD decrease in BMD). [J Bone Miner Res 2017;doi:10.1002/jbmr.3194]
Women aged ≥80 years had a higher incidence of hip fracture at 25 years compared with those <70 years (22.6 percent vs 13.9 percent), while incidence of nonvertebral fracture at 20 years was 50 and 42.6 percent among women aged ≥80 years and <70 years, respectively.
History of hip fracture was a better predictor of hip fracture at 25 years compared with history of nonvertebral fracture (RH, 1.6 vs 1.4), and history of either similarly predicted 20-year risk of nonvertebral fracture (RH, 1.6).
However, these findings did not suggest an advantage of hip fracture over nonvertebral fracture history in predicting future fracture risk, said the researchers.
“[A] history of any fracture is much more common than a history of hip fracture, and therefore will be more useful on a population level as a predictor,” they said.
The predictive value of femoral neck BMD in predicting future hip and nonvertebral fractures remained statistically significant for up to 20–25 years, while fracture history predicted hip fracture risk for up to 15 years and nonvertebral fractures for up to 20 years.
The researchers acknowledged that as the study population consisted of Caucasian women, the findings may not be generalizable to men or other ethnicities. Furthermore, the impact of a more recent fracture on future fracture risk was not assessed.