Low-frequency filter was 20 Hz, high-frequency filter was 2 kHz, duration of pulse was 0.2 ms, rate of pulse was 1/s, and the machine used was a Dantec Neuromatic 2,000 M electromyograph. Sensitivity and sweep speed were not specified.

This study was performed in the supine position, with the hand in its anatomical position.

Following the orthodromic method [1], sensory responses were recorded at the wrist (R) after the median nerve distal stimulation of the digital branches of digit I (Figs. 1 and 2), digit II (Figs. 3 and 4), digit III (Figs. 5 and 6), and digit IV (Fig. 7). Bipolar surface recording electrodes (silver/silver chloride electrodes covered by saline-soaked pads mounted in Perspex and fixed with a silicon rubber strap using minimal tension) were placed with a 2 cm interelectrode distance at the wrist, just proximal to the distal wrist crease (ideally between the tendons of the flexor carpi radialis (FCR) and the palmaris longus (PL) muscles). No fixed distances between the stimulating cathode and the recording electrodes were used. Digit I (thumb), digit II (index finger), digit III (middle finger), and digit IV (ring finger) recordings were made separately. The ground (G) electrode position was not mentioned in the text; the figures shows the ground electrode positioned on the palm.

The median nerve stimulations were applied distally to the digital branches of digit I, digit II, digit III, and digit IV. The stimulating electrodes (the authors used Dantec surface stimulating electrodes) were placed along each side (lateral radial and ulnar sides and medial side) of digit I, digit II, and digit III and on the lateral side (radial side) only of digit IV. A total of 7 digital branches were stimulated in turn, with the cathode (−) placed over the proximal phalanx and the anode (+) over the middle phalanx (for digit II, digit III, and digit IV stimulation) or terminal phalanx (digit I stimulation), with a 2 cm distance between the cathode and anode. Supramaximal stimulation was used, and 10 consecutive stimuli were used to average SNAPs.

Onset latency (ms) was measured from the onset of the stimulus to the onset of the negative deflection of the evoked sensory nerve action potential (SNAP); peak-to-peak amplitude (μV) was measured from the negative to the positive deflection of the wave. Sensory nerve conduction velocity (SNCV) was calculated using the conventional method and measured in meter per second (m/s). Sensory threshold (mA) to the stimulus was reported by the patient when the stimulus was first perceived. All results were divided into those for dominant and nondominant hands. Hand dominance was defined as the hand used for writing and other activities. All hands were warmed prior to the testing by seating the patient for 15 minutes in a room at 22–24 °C. The authors studied 23 median nerves (Table 1) from 12 women (12 dominant hands and 11 nondominant hands, age range 26–61 years, mean age 41 years) without clinical features of carpal tunnel syndrome (CTS) and 55 hands from 34 women (29 dominant hands and 26 nondominant hands, 22 bilateral CTS – 12 unilateral CTS, age range 29–67 years, mean age 44 years) with clinical features of CTS (Table 2).