What is the most sensitive test for diagnosing carpal tunnel syndrome?

OBJECTIVE To compare sensitivities between 7 principal nerve conduction studies (NCS) for diagnosing carpal tunnel syndrome (CTS). METHOD In 104 CTS and 64 control hands, following "Standard" NCSs were examined simultaneously: (1) Median sensory NCS; (2) segmental wrist-palm sensory NCS; (3) 4th digit latency difference; (4) 1st digit latency difference and (5) palmar mixed nerve latency difference. As "Guideline" and "Option" NCSs, we also examined: (6) Median motor distal latency and (7) second lumbrical-interossei latency difference (2LILD). Forty-nine CTS hands were divided into a milder subgroup only if action potentials could be recorded using all tests applied; that is, those with any absent potentials were excluded from the subgroup. Sensitivities and specificities were compared to each other. RESULTS In all CTS hands, the sensitivity of test (1), (2), (3), (4), (5), (6) and (7) was 83, 87, 92, 90, 90, 70 and 92%, respectively. In the milder subgroup, it was 67, 78, 84, 82, 84, 43, and 84% in the same order. There was no statistical difference between Standard tests and 2LILD. Specificities of all tests were over 95%. CONCLUSIONS All "Standard" tests and 2LILD have high comparable sensitivities. Therefore, 2LILD should be recommended as "Standard" NCS detecting CTS.