Sensitivity was 20 μV/division, low-frequency filter was 20 Hz, high-frequency filter was 5 kHz, sweep speed was 1 ms/division, duration of pulse was 0.1 ms, and the machine used was a Medelec Synergy electromyograph. Rate of pulse was not specified.

This study was performed in the supine position, with the hands placed with the wrist in a neutral position and the digits relaxed.

Following the orthodromic method [1], the sensory nerve action potentials (SNAPs) were recorded using disk electrodes placed on the midpalm and the wrist. On the midpalm (R1), the active electrode (A) was placed 7 cm from the stimulating cathode (secured around the proximal phalanx of the index or middle finger in the case of digit II or digit III stimulation, respectively), while at the wrist (R2), the active electrode (A) was placed 14 cm from the stimulating cathode (secured around the proximal phalanx of the index or middle finger in the case of digit II or digit III stimulation, respectively). The reference (R) electrode was placed proximally (Figs. 1 and 2). The authors used simultaneous two-channel recording to avoid using different stimulus intensity. Following the antidromic method, the SNAPs were recorded using ring electrodes placed distally to digit II and to digit III. The active recording electrode was placed around the proximal phalanx of the index or middle finger in the case of the digit II or digit III recording, respectively. The reference (R) electrode was placed distally at about 4 cm from the active recording electrode (Figs. 3 and 4). The ground (G) electrode was placed between the stimulation and recording sites.

Following the orthodromic method, stimulation was applied to digit II and to digit III. The authors used ring electrodes; the stimulating cathode was secured around the proximal phalanx of the index or middle finger in the case of the digit II or digit III stimulation, respectively. The anode (−) was placed distally with an interelectrode separation always maintained at 4 cm. Following the antidromic method, stimulus was supramaximal at the midpalm and at the wrist, 7 cm (S1) and 14 cm (S2) proximal to the active recording electrode placed around the proximal phalanx of index and middle finger in the case of the digit II and the digit III recording, respectively. Stimuli of 0.1 ms duration and anodal rotation were used to reduce stimulus artifact.

Onset latency (ms) was measured from the stimulus onset to the onset of the negative peak of the sensory nerve action potentials (SNAPs). Peak latency (ms) was measured from the onset of the stimulus to the peak of the negative deflection of the SNAPs. Negative peak amplitude (μV) was determined from the onset to the peak of the negative peak of the SNAPs. The SNAPs were averaged ten times. Distances were measured with a caliper. Skin temperature was measured at the midpalm, and it was maintained at >33 °C using a hydrocollator pack. To compare the conduction times between the palm–digit and wrist–palm segments, the authors calculated the orthodromic and antidromic distoproximal ratio (DPLR) with Padua’s method [2], but using the onset and peak latencies (and not SNCV), as follows (digit III was reported, same for digit II):

Authors classified the distoproximal ratio (DPLR) data into three categories: DPLR = 1 (palm–wrist and palm–digit segments had the same conduction time), DPLR < 1 (palm–wrist segment conduction time was slower than that for the palm–digit segment), and DPLR > 1 (palm–wrist segment conduction time was faster than that of the palm–digit segment). In normal hands, they defined a normal DPLR as > 1 and a reversed ratio when DPLR was ≤1. Normal values (Table 1) were obtained from 60 control hands of 30 asymptomatic healthy volunteers (15 men and 15 women, age range 18–36 years, mean age 27.1 years, height range 155–181 cm, mean height 167.2 cm, mean men height 173.8 cm and women 160.5 cm).