With institutional approval and informed written consent, we studied healthy (ASA I) female Caucasian volunteers aged 18 to 40 years who had either natural bright red, black, or dark brown hair. We considered volunteers to be Caucasian if they reported being primarily of northern European descent. Volunteers were monetarily reimbursed for their time.
Other studies indicate that women report different pain experiences and more negative responses to pain than men,13-16
as well different responses to analgesics.17
We thus restricted our study population to women. A higher sensitivity to pain stimuli has been observed during the luteal phase of the menstrual cycle.18,19
We therefore also restricted studies to the first ten days of the participants' menstrual cycles unless they were using hormonal contraceptives.
Exclusion criteria included chemical hair treatment, any history of medical or psychiatric problems, history of current or past chronic pain conditions, skin abrasions or lesions on testing sites, pregnancy, body mass index >30 kg/m2
, recreational drug use, or use of medications other than oral contraceptives. Based on the means and standard deviations obtained from preliminary test results on the Neurometer CPT/C
device (Neurotron, Inc., Baltimore, MD),20
an a priori sample-size estimate suggested that eighteen subjects in each group would provide a 90% power for detecting a 40% reduction in pain tolerance thresholds at a two-tailed, unpaired alpha level of 0.05. We thus enrolled 30 volunteers with red hair and 30 with dark hair.
All studies were started at 8:00 am. Volunteers fasted and refrained from smoking for at least eight hours before the start of the protocol. Studies were conducted in a quiet room that was maintained at a comfortable ambient temperature. Each volunteer's height, weight, and age were recorded. Sensory and pain thresholds for each volunteer were tested with two devices: the Neurometer CPT/C (Neurotron, Inc., Baltimore, MD) and the TSA-II Neurosensory Analyzer (Medoc Ltd., Minneapolis, MN).
The Neurometer CPT/C
produces a biphasic sinusoid alternating current (AC) waveform stimulus at frequencies of 2000, 250, and 5 Hz.20,21
Sinewaves at 2000, 250, and 5 Hz correspond to depolarization periods of 0.25 msec, 2 msec, and 100 msec, respectively. The large diameter fibers can respond to the rapid 2000 Hz stimulus while the small unmyelinated fibers require several milliseconds of a continuous depolarization period to respond. However, large fibers repolarize faster than the slow increase of the 5-Hz stimulus can depolarize them and therefore do not achieve threshold potential with that stimulation. Together, these factors result in the 2000, 250 and 5 Hz sinewave depolarizations selectively evoking responses from the large myelinated A-beta, the small myelinated A-delta, and the small unmyelinated C fibers, respectively.22,23
automatically compensates for alterations in skin resistance by adjusting the voltage to maintain a constant current.21
was modified to deliver currents from zero to a maximum of 20 mA (milliampere).
For determination of the baseline sensory, pain perception, and pain tolerance thresholds, pairs of gold electrodes were positioned on glabrous skin of both the right and left ring fingers. For each specific threshold measurement, both sides (dominant and non-dominant hand) were tested sequentially, with the order determined by random assignment. The skin was prepared for testing with a gentle abrasive cleaning preparation. A pair of gold electrodes (1 cm diameter) separated by a 1.7 cm Mylar spreader was coated with a thin layer of chloride-free electroconductive gel and taped to the finger. Volunteers were instructed as to how to respond to each stimulus according to a standardized script and were blinded at all times to the threshold measurement results displayed on the Neurometer device.
The baseline current perception (sensory) thresholds were tested first at 2000, 250 and 5 Hz. The operator slowly increased the stimulus until the volunteer consistently reported detecting the stimulus at 0.02 mA value and not detecting the stimulus 0.2 mA units below this value. The manual intensity alignment was followed by auto test cycles controlled by the device. Auto test cycles with true or false testing confirmed that the volunteer responses were within ± 0.04 mA. After repeated and consistent crossovers occurred, the device determined the exact mA current value.
After completing the baseline current perception threshold measurements on both hands, baseline pain perception thresholds and then pain tolerance thresholds were measured on both hands. For pain perception and pain tolerance threshold measurements, the volunteer initiated the stimulation by pushing a button on the device. As long as the volunteer depressed the button, the amount of stimulation slowly increased automatically. We asked volunteers to release the button when they perceived the stimulus as painful (for pain perception thresholds) or it became intolerable (for pain tolerance thresholds). We allowed at least two minutes, or until sensation in the finger had returned to baseline, between successive stimuli. In summary, three different frequencies of stimulation (2000 Hz, 250 Hz, and 5 Hz) were used in order to obtain three different threshold measurements (current perception, pain perception, and maximum pain tolerance). These baseline sensory and pain thresholds in response to electrical stimulation were normally distributed.
In order to determine baseline thermal sensory, pain perception, and pain tolerance thresholds, contact heat stimuli were delivered using a computer-controlled thermal sensory analyzer (TSA-II Neurosensory Analyzer
). This device has been used extensively for quantitative assessment of thermal sensory and pain thresholds.18,24
The maximum delivered temperature was 50°C, and the minimum temperature was 0°C.
For temperature threshold testing, a 3 x 3 cm square thermode was positioned on the volar area of the volunteer's forearm. Both the dominant and non-dominant arms were tested sequentially, with the order determined randomly. From a baseline of 32°C, probe temperature was increased or decreased at a rate of 0.5°C/second until the volunteer responded. The slow rise presumably evokes mainly stimulation of C-nociceptive afferents.25
Volunteers were instructed as to how to respond to each stimulus according to a standardized script and were always blinded to results. Four trials of warm and cold sensory thresholds were given to each volunteer, followed by four trials of heat pain and cold pain (perception) thresholds and then four trials of heat pain and cold pain tolerance thresholds. Volunteers were instructed to press a button to interrupt the stimulus whenever they felt that the appropriate threshold had been reached. The four trials were averaged to determine the thresholds. A few volunteers reported mild tenderness at the testing site after maximum intensity stimulation; in these volunteers, the position of the thermode was altered slightly between trials in order to avoid either sensitization or habituation of cutaneous receptors. In addition, we allowed at least 60 seconds between successive stimuli. Subjects were blinded at all times to the temperature threshold values obtained with the TSA-II Neurosensory Analyzer
In addition to the baseline current and thermal threshold measurements, we also investigated how the pain tolerance thresholds were affected by local anesthetic. The methodology to evaluate the sensitivity to different local anesthetics using the Neurometer
has been described in previous studies.17,26
Briefly, the volar surface of the non-dominant forearm of the volunteer was divided into three areas, each measuring approximately 2 x 4 cm. One area functioned as control; the second area was covered with a 6-mm-thick layer of 4% liposomal lidocaine (ELA-Max, Ferndale laboratories, Ferndale, MI), which was wiped clean after 60 minutes. In a previous study, the average onset time for this formulation of lidocaine was 7 minutes,26
the application time of 60 minutes in this study should therefore be sufficient to provide cutaneous anesthesia. In the third area, we injected 2 mL of 1.0% lidocaine subcutaneously and allowed an onset time of at least 5 minutes.
Pain tolerance threshold values were determined at frequencies of 2000 Hz, 250 Hz, and 5 Hz as described above for all three areas. Subjects were blinded to the pain tolerance threshold values generated for each area. Pain tolerance thresholds of areas tested for local anesthetic sensitivity were found to have a censored outcome, i.e. many volunteers reached maximum pain tolerance thresholds and could go no higher.
We compared the demographics and other volunteer characteristics with unpaired, two-sided, t-tests for continuous variables and chi-square or Fisher's exact test for categorical variables. The analysis of perception and threshold outcomes and all continuous outcomes depended on the distribution of the data and whether any of the values were censored. Values were considered censored when the maximum possible pain threshold values were not reached, i.e., the maximum possible pain threshold values in these volunteers were a function of the limits of the Neurometer device (max. output is 20 mAmp) and not a physical limitation. In other words, the actual pain threshold values were not known because the values were truncated or “right censored” at 20 mAmps. Outcomes that were normally distributed were compared with unpaired, two-sided, t-tests. If the values were skewed, the Mann-Whitney rank-sum test was used for comparing the groups. If some of the data were censored, then we used Kaplan-Meier survival curves and log-rank tests. P < 0.05 was considered statistically significant.