OBJECTIVE: To demonstrate a conduction block of the median nerve at the flexor retinaculum (FR) in carpal tunnel syndrome (CTS), comparison of potentials obtained with stimulation of median nerve at the wrist and the palm may be required. METHOD: To determine the severity and incidence of conduction block in patients with CTS, seventy hands of neurologically healthy adults (mean age, 48 years) as control, and seventy hands of patients with CTS (mean age, 51 years) were tested. We performed median motor and middle finger recorded antidromic sensory conduction study with stimulation of the wrist and palm of a distance of 5 cm. The negative peak spike duration and baseline to peak amplitude of the compound muscle action potential (CMAP), and sensory nerve action potential (SNAP) with wrist and palm stimulations were measured. From these values, the wrist to palm duration ratio and amplitude ratio were obtained. RESULTS: The criteria of median motor nerve conduction block were a wrist to palm amplitude ratio of less than 0.7 and a wrist to palm duration ratio of less than 1.13. The criteria of median sensory conduction block were a wrist to palm amplitude ratio of less than 0.61 and a wrist to palm duration ratio of less than 1.33. In the patient group, 10 hands (14.3%) showed motor conduction block and 12 hands (17.1%) showed sensory conduction block and 3 hands (4.3%) showed both. The wrist to palm amplitude ratios of CMAP and SNAP in the patient showing conduction block were 0.6+/-0.1, and 0.4+/-0.2, respectively. There was no correlation between palm CMAP or SNAP amplitude and respective wrist to palm ratios. CONCLUSION: Comparison of the amplitude and duration of CMAP or SNAP obtained with stimulation of both wrist and palm may be able to differentiate between conduction block and axonal degeneration. These values may be useful in planning treatment and predicting outcome.