Table 3.
Mean (SD) values and linear correlation coefficients (r) of the in vivo dual-energy X-ray absorptiometry (DXA) measurements with Prodigy and iDXA (n = 72). Bland and Altman analysis results with relative mean differences d%a as well as limits of agreement [d ± (1.96 × SD)]. Simple linear regression analysis of Prodigy (dependent) versus iDXA (independent) BMD data with standard error (SE) and standard errors of estimates (SEE). Systematic BMD differences were observed between Prodigy and iDXA. After linear regression correction equations were applied the differences were negligible.
| BMD ROI | Prodigy mean (SD) |
iDXA mean (SD) |
r | d%a | Intercept (SE)c | Slope (SE)c | SEE | SEE (%)d | Limits of agreement |
|---|---|---|---|---|---|---|---|---|---|
| Spine L2–L4 | 1.169 (0.145) | 1.186 (0.149)b | 0.990 | 1.5 | 0.985 (0.002) | 0.021 | 1.8 | −0.025 to 0.059 | |
| Femoral neck | 0.941 (0.129) | 0.946 (0.128)b | 0.986 | 0.5 | 0.995 (0.003) | 0.022 | 2.3 | −0.037 to 0.048 | |
| Ward's triangle | 0.761 (0.145) | 0.758 (0.148) | 0.985 | −0.4 | 1.002 (0.004) | 0.026 | 3.4 | −0.05 to 0.05 | |
| Trochanter | 0.800 (0.102) | 0.806 (0.108)b | 0.987 | 0.7 | 0.044 (0.015) | 0.939 (0.019) | 0.017 | 2.1 | −0.030 to 0.040 |
| Shaft | 1.162 (0.164) | 1.180 (0.168)b | 0.993 | 1.5 | 0.985 (0.002) | 0.020 | 1.7 | −0.022 to 0.058 | |
| Total hip | 0.993 (0.126) | 1.002 (0.129)b | 0.995 | 0.9 | 0.990 (0.002) | 0.013 | 1.3 | −0.016 to 0.034 |
aFormula for relative mean difference (d%): [(iDXA − Prodigy) × 100/Prodigy].
bSignificantly different (p < 0.05) mean difference (d = Prodigy − iDXA, paired t-test or nonparametric Wilcoxon signed-rank test in Ward' triangle BMD, BMC, or area ROIs).
cCorrection equation: Prodigy (BMD) = Slope × iDXA (BMD) + intercept.
dFormula for SEE (%): [SEE × 100/((Prodigy + iDXA)/2)].