| Objective To statistically analyze the least significant change (LSC) in the measurement of bone mineral density (BMD) with dual-energy X-ray absorptiometry (DXA), and to explore its clinical value. Methods BMD of the lumbar vertebrae and the hip of 31 subjects were measured using the GE Lunar Prodigy bone densitometer by 4 technicians (A, B, C, and D). Each subject was measured for 2 consecutive times, and the precision error (PE) and LSC were calculated. Results (1) There were difference in BMD measurement of different positions between the technicians. The PE and LSC at each site were different measured by the same technician. The variation of PE and LSC in the lumbar region of interest was smaller than that in the bilateral femurs. In this experiment, there was no statistically significant difference in PE (CV) measured at L1-L4, left and right femur necks, and left and right total hips between 4 technicians (all P>0.05). (2) The PE of L1-L4 measured by the team was RMS-SD =0.011 g/cm2, RMS-CV=0.011, and the LSC was LSC-SD=0.031 g/cm2, LSC-CV=0.031.The PE of the left total hip was RMS-SD=0.013g/cm2, RMS-CV=0.014, and the LSC was LSC-SD=0.036g/cm2, LSC-CV =0.039.The PE of the right total hip was RMS-SD=0.010 g/cm2, RMS-CV=0.011, and the LSC was LSC-SD =0.026 g/cm2, LSC-CV=0.030. Conclusion PE is small and accurate when using DXA to measure BMD. The monitoring time interval (MTI) should be extended with the increase of PE in the area of interest and be shortened with the increase of the predicted annual change of BMD. |