Background. The aim of the study was to determine the relationship between dual energy x-ray absorptiometry (DXA) and quantitative ultrasonometry (QUS) of calcaneus and their correlation with axial bone mineral density. Methods and Results. 1284 subjects were tested for BMD (Bone Mineral Density) by DXA at the spine and hip (707 subjects by DPX-L, Lunar, and 577 subjects by QDR-4500 A, Hologic) and calcaneus (by PIXI, Lunar). The calcaneus was also measured using the QUS (Achilles Plus, Lunar), on the same day. The mean age of the patients was 56.5±11.6 years, mean height 166 cm, mean weight 70 kg. Three subjects were selected for precision error measurement with low, medium and high BMD of calcaneus (T-score of -2.2, -0.77 and 2.02, respectively) and scanned with re-positioning at the right heel (PIXI and Achilles Plus) 21 times on one day for short term precision error and over 21 consecutive days for long term precision error. The in vivo short term precision error of the heel measurement (BMD, SOS, BUA) in subjects with normal BMD was 0.67 %, 0.47 % and 1.87 %, respectively; the long term in vivo precision error was 1.14 %, 0.26 % and 2.95 %, respectively. No significant difference was found between BMD values on the right and left heel. A statistically significant correlation (p<0.001) was found between BUA and BMD (r=0.71), SOS and BMD (r=0.73), Stiffness and BMD (r=0.77). The heel BMD was also significantly correlated to BMD of the femoral neck (r=0.64) and BMD of total femur (r=0.70) and BMD of lumbar spine (r=0.59). Conclusions. The DXA of the heel underestimates the prevalence of osteoporosis. The results of the heel QUS (Stiffness) appear to be better correlated to femoral BMD than heel BMD. The observed correlation coefficient of 0.77 between QUS and DXA at the heel was statistically significant, but it explains only 60 % of variability of the QUS of the heel.

        Key words: calcaneus, dual energy X-ray absorptiometry, osteoporosis, quantitative ultrasonometry.