Rationale and objectives: The authors determined the effect of obesity on measurements of amplitude-dependent speed of bone ultrasound (Ad-SOS [m/sec]) and compared them to the total body bone mineral content (TBBMC/g).
Methods: A total of 25 women were studied (mean age 41.8 +/- 10.2 years). In all the subjects, body mass index (BMI) exceeded 30 kg/m2 (range, 31.12-47.47 kg/m2); mean body weight was 104 +/- 17 kg. Ad-SOS was measured at the proximal phalanges and TBBMC in whole body with dual-energy x-ray absorptiometry.
Results: Correlation study (Fisher's r to z) showed that Ad-SOS correlated negatively with weight (r = -0.85, P < 0.0001) and with TBBMC (r = -0.71, P < 0.0001). The correlation between TBBMC and weight was r = 0.76, P < 0.0001. Body fat percentage correlated partially with TBBMC (r = 0.40, P < 0.05) and negatively with Ad-SOS (r = -0.75, P < 0.0001). When the correlation test was adjusted for weight (partial correlation), the correlation between Ad-SOS and TBBMC was not significant (r = -0.21, P = NS), and the correlation between Ad-SOS and weight continued to be inversely significant (r = -0.67, P < 0.0001).
Conclusions: The results showed a clearly negative effect of weight on Ad-SOS measurements and indicated the limitations of this technique when employed in overweight and obese patients. Broad-band ultrasound attenuation and speed of sound, two commonly measured variables in bone ultrasound studies, may be differently affected by soft tissue.