Accuracy of the Fracture Risk Assessment Tool for judging pharmacotherapy initiation for primary osteoporosis.

INTRODUCTION: This study aimed to determine whether the Fracture Risk Assessment Tool (FRAX®) is useful in assessing the criteria for the initiation of pharmacotherapy for primary osteoporosis based on the current diagnostic criteria in Japan.
MATERIALS AND METHODS: We enrolled 614 patients aged ≥ 40 years (average, 77.0 years) who were eligible for primary osteoporosis evaluation. Bone mineral density measurements of the lumbar spine, total hip, and femoral neck using ALPHYS LF (FUJIFILM, Tokyo, Japan) and imaging studies involving the lumbar spine were obtained and the FRAX® scores of each patient were calculated with and without the T-score of the femoral neck. The receiver operating characteristic curve analysis method was used to calculate the cut-off FRAX® scores with reference to the criteria for initiating pharmacotherapy for osteoporosis; the accuracies of both FRAX® scores were compared.
RESULTS: The FRAX® score calculated with the T-score was more accurate for hip fracture risk assessment [cut-off value 5.5%; the area under the curve (AUC) 0.946] than for major osteoporotic fracture risk assessment (cut-off value 17.0%; AUC 0.924) in judging the criteria (p = 0.001). Conversely, the FRAX® score calculated without the T-score was equally accurate for hip fracture risk assessment (AUC 0.796) and major osteoporotic fracture risk assessment (AUC 0.806) (p = 0.23).
CONCLUSION: The FRAX® score can accurately assess the criteria for initiating pharmacotherapy for primary osteoporosis based on the current Japanese diagnostic criteria, especially when the T-score is used.