Performance of Sri Lankan FRAX algorithm without bone mineral density and with Quantitative Ultrasound data input

Abstract

Introduction

Fracture risk assessment algorithm (FRAX) is the most validated method available to predict fracture risk. Its use is restricted due to limited availability of Dual Energy X-ray Absorptiometry (DXA). FRAX has the option of assessing facture risk without BMD data.

Objectives

To assess the ability of Sri Lankan FRAX algorithm without BMD input in evaluating fracture risk. The possibility of replacing the BMD input with Quantitative Ultrasound (QUS) data of radius in calculating fracture risk also assessed.

Methods

Data of clinical risk factors associated with fractures were collected from community dwelling postmenopausal women (n=339). DXA scans were performed in all subjects and QUS scans (in radius) were performed in a randomly selected sample (n=207). Ten-year risks of major osteoporotic fracture (MOFR) and hip fracture (HFR) were calculated with BMD, without BMD (FRAX-FN0) and with US T score instead of BMD (FRAX-UST).

Results and conclusion

Nearly 35.7% had high risk of fractures. FRAX-FN0 had 79.2% sensitivity, 80.1% specificity, 68.8% positive predictive value (PPV) and 87.4% negative predictive value (NPV). FRAX-UST showed 78.4% sensitivity, 70% specificity, 59.8% PPV and 85% NPV. ROC AUCs were above 0.80 in both FRAX-FN0 and FRAX-UST. The standard errors of estimate (SEE) were less in FRAX-FN0 (3.96 and 2.76 for MOFR-FN0 and HFR-FN0 respectively) compared to FRAX-UST (6.13 and 4.83 for MOFR-UST and HFR-UST, respectively). In conclusion, Sri Lankan FRAX without BMD is an acceptable alternative in areas with restricted DXA facility. Radial QUS data cannot be used as a substitute to FN-BMD in Sri Lankan FRAX.