Water bath and contact methods in ultrasonic evaluation of bone.

Abstract

Ultrasonic devices for the measurement of speed of sound (SOS) and broadband ultrasonic attenuation (BUA) generally use either a contact or water bath method. The aim of this study was to compare these two methods while determining the influence of soft tissue, pathlength (heel width and bone width), and a fixed heel dimension on SOS (m/second) and BUA (dB/MHz). Ultrasonic measurements were made using a CUBA Research system utilizing a pair of 1 MHz unfocused transducers with mean precision CV = 0.7% and 6.0% for all SOS and BUA measurements, respectively. SOS and BUA were determined in 24 human cadaveric heels under three conditions: contact method (heel intact), water bath method (heel intact), water bath method (no soft tissue). Although there were significant differences between measurements using contact and water bath techniques (heel intact), their correlations were high (r = 0.858 for SOS and r = 0. 937 for BUA; P < 0.001). After removal of soft tissue, SOS significantly increased (78 m/second; P < 0.001) whereas there was no change in BUA (P > 0.05). Heel width correlated with SOS measurements (-0.224 < r < -0.347; P < 0.001) and bone width correlated with BUA measurements (0.198 < r < 0.276; P < 0.001). The practice of using a fixed heel dimension (Lunar Achilles) was investigated by comparing SOS calculated with measured heel thickness and a value of 4 cm (Lunar Achilles). SOS increased by 42 m/second (2.7%) using the fixed heel dimension compared with measured heel widths. This study demonstrates the similarity between contact and water bath-based methods, while showing that the presence of soft tissue reduces SOS but has no effect on BUA. The use of a fixed heel dimension for calculation of SOS overestimates values obtained when using measured heel dimensions, though the values correlate highly (r = 0.98, P < 0.001). In addition, an increase in heel width tends to cause an underestimation of SOS whereas an increase in bone width tends to overestimate BUA, although the effects are relatively small.