Background: As a criterion for selecting an osteoporosis treatment target, this study investigates the effectiveness of clinically using the Fracture Risk Assessment Tool (FRAX®) developed by the World Health Organization (WHO) and evaluates changes in osteoporotic fracture risk prediction according to the bone mineral density (BMD) values of lumbar spine and hip.
Methods: Osteoporotic fracture occurred in 531 of 4556 patients who underwent hip and lumbar spine BMD measurement at our hospital from April 2003 until March 2013. Of the 531 patients, we excluded those who had no BMD measurements within 2 weeks of fracture, those who had no risk factor data needed to calculate the FRAX® value, and those who were already undergoing osteoporosis treatment. Finally, we analysed the FRAX® value in 445 patients by inputting the BMD values of lumbar spine and hip. In accordance with the standards of the National Osteoporosis Foundation (NOF), FRAX® values greater than 20% for major osteoporotic fracture or 3% for hip fracture were considered high risk.
Results: Among the 445 patients, the high risk group consisted of 330 patients (74%) when lumbar spine BMD was used to calculate the FRAX® value, 281 patients (63%) when hip BMD was used, and 258 patients (58%) when no BMD values were used. For the 84 patients with osteopenia, using the lumbar spine BMD in the model did not lead to a significant difference in the average FRAX® value compared with when hip BMD was used. However, the average FRAX® value was significantly higher when no BMD values were used in the model (p<0.001). For the osteopenia patients, the high risk group consisted of 39 patients (46%) when BMD was not used, 19 patients (23%) when hip BMD was used, and 14 patients (17%) when lumbar spine BMD was used. The highest osteoporotic fracture probabilities occurred when BMD was not used.
Conclusion: Clinicians using the FRAX® model to determine osteoporosis treatment for patients with osteopenia may be able to improve clinical efficacy by excluding BMD values from the model.
Disclosure: The authors declared no competing interests.