Were you identified to be at high fracture risk by FRAX® before your osteoporotic fracture occurred?

With the increasing longevity and the growing size of the aging population, China is among the developing countries with the greatest burden of osteoporotic fractures (OFs). Identification of individuals at high fracture risk to provide early treatment in developing countries with limited resources remains a challenge. Our aim was to calculate the FRAX® values in a selected mainland Chinese population the day before the occurrence of an OF and to compare the results with the Osteoporosis Foundation (NOF) thresholds. To define a population at high fracture risk, we selected patients in the orthopedic emergency cohort from Shanghai Sixth People's Hospital, China. To calculate the FRAX® the day before the actual OF, the actual fracture was not considered as a previous fracture. Subjects with a FRAX® value lower than the NOF thresholds were not considered at high risk. One-hundred ninety-eight patients were included with a mean age of 69.9 ± 10.1 years old, and 78.8 % were female. Major fracture sites included hip (41.9 %), proximal humerus (22.2 %), forearm (27.8 %), spine (6.6 %), and pelvis (1.5 %). Overall, only two (1.0 %) patients had a 10-year major fracture probability ≥20 %, and 84 (42.4 %) patients had a 10-year hip fracture probability ≥3 %. More than half of the population was not identified to be at high risk by the FRAX® BMI (body mass index). For age groups 55-59, 60-69, 70-79, and 80-90 years old, respectively, the mean FRAX® BMI values (major OF/hip fracture) were 3.5/0.7, 5.7/1.8, 8.2/4.7, and 8.6/4.8 %. Dual-energy X-ray absorptiometry was available for 58 patients, and 41.2 % had a T-score ≤ -2.5 standard deviation (lumbar spine and/or femoral neck and/or total hip). Only six (17.6 %) patients with a normal or osteopenic BMD were classified at high risk by the FRAX® BMD (bone mineral density). The NOF thresholds could only identify fewer than half of the studied population to be at high risk the day before an actual OF, and the 10-year hip fracture probability seemed to perform better than the 10-year major fracture probability. Prospective population-based studies are needed to validate the prediction of Chinese FRAX® model and to establish ideal thresholds.