Judicious use of DXA-BMD in assessing fracture risk by using clinical risk factors in the Indian population

Summary FRAX scores were significantly higher in patients admitted with fragility fractures than controls and can be useful in choosing the right patients for bone density testing, thus using of an expensive test judiciously. Purpose This study was planned to compare the FRAX scores for the risk for major osteoporotic fracture (FRAX-MOF) and hip fracture (FRAX-HF) in patients with fragility fractures (cases) and those admitted for other indications (controls) in the orthopedic ward in our institute. Methods Historical and anthropometric data were prospectively recorded from 500 consecutive patients admitted in the orthopedic ward in in Medanta, the Medicity, Gurgaon, India. The receiver operating characteristic (ROC) curves were constructed for FRAX-MOF and FRAX-HF and the area under the curve (AUC) was calculated between cases and controls. Results The FRAX-MOF was significantly high in cases as compared to controls (7.34 ± 4.41 versus 5.64 ± 4.3; p = 0.001). The FRAX-HF was also significantly high in cases as compared to controls (2.95 ± 3.13 versus 1.67 ± 2.21; p < 0.001). The areas under the curves were 0.627 for FRAX-MOF and 0.654 for FRAX-HF. For FRAX-MOF, a cutoff of 2 has a 90% sensitivity, but only 15% specificity; whereas a cutoff of 10.5 had a specificity of 90% to differentiate those with and without fractures, but only 23% sensitivity. For FRAX-HF, a cutoff 0.3 had about 90% sensitivity and 20% specificity, whereas a cutoff of 3.5 had 90% specificity and 25% specificity to differentiate cases and controls. Conclusions This study compared the FRAX-MOF and FRAX-HF in patients with and without fragility fractures and derived cutoffs for practical clinical use of FRAX-MOF and FRAX-HF to optimize the use of DXA-BMD.