Wiesler ER,Chloros GD,Cartwright MS,Smith BP,Rushing J,Walker FO.
The Journal of Hand Surgery. 2006;31-A(5):726-732.

44 wrists with CTS and 86 asymptomatic control wrists were ultrasounded with a 12/5 MHz linear array transducer and the cross-sectional area of the median nerve was measured at the distal wrist crease. All wrists also underwent NCS.
The average axial cross-sectional area at the wrist was 9mm2 (range 5-14 mm2) in the control group and 14 mm2 (range 5-32 mm2) in the CTS group. This difference in size was statistically significant (p<0.001). Higher values of cross-sectional areas were associated with abnormal NCS results (p=0.013, Pearson Correlation Coefficient). Sensitivity (91%) and specificity (84%) tests suggested that USS was a good diagnostic tool and adjunct to NCS.
The authors suggest that the ease of use, non-invasive nature and shorter time, with reduced costs makes USS an attractive modality.
This certainly would be a useful adjunct in the diagnosis of CTS as NCS are not always conclusive for patients who have CTS. It may prove to be a useful alternative diagnostic tool for an already burdened neurophysiology department. It could also allow the clinican to administer safe therapeutic injections thus avoiding any potential for median nerve injury.