ABSTRACT
Ultrasound criteria for carpal tunnel syndrome (CTS) may vary in different populations. To determine the ultrasonographic criteria for CTS in a Malaysian population and compare its usefulness with nerve conduction studies (NCS), we studied patients clinically diagnosed with CTS and normal controls by ultrasonography. All patients also underwent standard NCS. Median nerve Cross-Sectional Area (CSA) and Flattening Ratio (FR) at 3 different levels – proximal to tunnel inlet, at tunnel inlet and tunnel outlet were measured. Receiver operator characteristic (ROC) analyses were used to calculate the optimal discriminatory threshold values for CTS. Of 54 CTS hands, NCS was positive in 85.2%. Median nerve CSA at all 3 levels, were signifi cantly greater in CTS hands.FR was signifi cantly greater at tunnel inlet. A CSA threshold of 0.1 cm2 proximal to and at tunnel inlet had sensitivities of 70.4% and 63% and specifi cities of 85.2% and 88.5 % respectively. CSA at tunnel outlet had lower specifi city. If CSA of 2 levels (viz. proximal to or at tunnel inlet) were considered together, sensitivity and specifi city improved to 81.5% and 83.3%. Qualitative loss of fascicular discrimination of the nerve proximal to the inlet had sensitivity and specifi city of 77.8% and 96.3%. The most useful ultrasonographic parameter was median nerve CSA either proximal to or at tunnel inlet. However, the sensitivities were lower compared to NCS. Qualitative appearance of the median nerve is a useful adjunct to diagnosis. In conclusion, ultrasonography play an important complementary role to NCS in the diagnosis of CTS.