Thermal detection thresholds of Aδ- and C-fibre afferents activated by brief CO2 laser pulses applied onto the human hairy skin

Abstract

Brief high-power laser pulses applied onto the hairy skin of the distal end of a limb generate a double sensation related to the activation of Aδ- and C-fibres, referred to as first and second pain. However, neurophysiological and behavioural responses related to the activation of C-fibres can be studied reliably only if the concomitant activation of Aδ-fibres is avoided. Here, using a novel CO(2) laser stimulator able to deliver constant-temperature heat pulses through a feedback regulation of laser power by an online measurement of skin temperature at target site, combined with an adaptive staircase algorithm using reaction-time to distinguish between responses triggered by Aδ- and C-fibre input, we show that it is possible to estimate robustly and independently the thermal detection thresholds of Aδ-fibres (46.9±1.7°C) and C-fibres (39.8±1.7°C). Furthermore, we show that both thresholds are dependent on the skin temperature preceding and/or surrounding the test stimulus, indicating that the Aδ- and C-fibre afferents triggering the behavioural responses to brief laser pulses behave, at least partially, as detectors of a change in skin temperature rather than as pure level detectors. Most importantly, our results show that the difference in threshold between Aδ- and C-fibre afferents activated by brief laser pulses can be exploited to activate C-fibres selectively and reliably, provided that the rise in skin temperature generated by the laser stimulator is well-controlled. Our approach could constitute a tool to explore, in humans, the physiological and pathophysiological mechanisms involved in processing C- and Aδ-fibre input, respectively.

Figure 1. Experimental paradigm.

A. Thermal stimuli were applied to the dorsum of the non-dominant hand using a CO₂ laser (beam diameter: 6 mm). The data shown corresponds to an actual measurement of skin temperature performed by the radiometer in three different trials (x-axis: time; y-axis: skin temperature). After measuring the baseline skin temperature for one second (T₀), the stimulated area was raised to the conditioning skin temperature using a 1-s heating ramp and was maintained at that temperature for 2 seconds (T₁). Three different conditioning skin temperatures were used: T₁ = unchanged (blue), T₁ = 34°C (green) and (c) T₁ = 38°C (red). Following this conditioning of the stimulated area, the test stimulus (T₂) was applied, consisting of a 10-ms heating ramp followed by a 40-ms plateau. B–D: The staircase procedure used to estimate Aδ- and C-fibre thresholds (one subject shown as example). Participants were asked to respond as quickly as possible by pressing a button held in the dominant hand when perceiving the test stimulus (T₂). Reaction times were used to discriminate between responses triggered by C-fibre input (reaction-time ≥650 ms) and responses triggered by Aδ-fibre input (reaction-time <650 ms). For all three conditions (Panel B: T₁ = unchanged, Panel C: T₁ = 34°C and Panel D: T₁ = 38°C), an adaptive staircase algorithm was used to estimate (1) the absolute detection threshold assumed to reflect the detection threshold of C-fibre input and (2) the detection threshold of responses with a reaction-time <650 ms assumed to reflect the detection threshold of Aδ-fibre input. The six staircases required for this procedure were presented in an interleaved fashion, such as to avoid order effects related to habituation and/or sensitization as well as a possible response bias due to anticipation.

Figure 2. Frequency distribution of the reaction-times to brief CO₂ laser pulses applied onto the dorsum of the non-dominant hand using target skin temperatures ranging from 35 to 51°C.

The upper panel shows the overall distribution of reaction-times regardless of the target skin temperature (T₂). The lower panel shows the distribution of reaction-times as a function of T₂. Note the bimodal distribution of reaction-times, related to the fact that lower target skin temperatures triggered late-latency detections compatible with the conduction velocity of unmyelinated C-fibres, whereas higher target skin temperatures triggered early-latency detections compatible with the conduction velocity of myelinated Aδ-fibres (at higher skin temperatures, the stimulus activated both Aδ- and C-fiber afferents, but detections were related to the first-arriving Aδ-fiber afferent volley). The bimodal nature of this distribution was confirmed by comparing the fitting of the data to a model describing a unimodal vs. a bimodal distribution of sensory-motor responses (see Results). The first function of the bimodal model characterized the distribution of the short-latency Aδ-fibre responses (a = 459 ms; b = 111 ms). The second function characterized the distribution of the late-latency C-fibre responses (a = 961 ms; b = 202 ms).

Figure 3. Thermal detection threshold of the behavioural responses triggered by Aδ- and C-fibre input obtained in the three experimental conditions (T₁ = unchanged, T₁ = 34°C and T₁ = 38°C).

The thin grey lines represent the individual thresholds obtained in each participant. The box plots represent the group-level median and interquartile range. Note that in all participants, the Aδ-fibre threshold is markedly greater than the C-fibre threshold. Also note that this difference is less marked in the condition T₁ = 38°C.

Figure 4. Thermal detection threshold of the behavioural responses triggered by Aδ- and C-fibre input obtained when the stimulated area is first heated to T₁ = 34°C or T₁ = 38°C during the two seconds preceding the test stimulus (T₂).

The thin grey lines represent the individual thresholds obtained in each participant. The box plots represent the group-level median and interquartile range. Note that the C-fibre threshold is increased for T₁ = 38°C vs. T₁ = 34°C, whereas the Aδ-fibre threshold appears unaffected by the conditioning skin temperature.

Figure 5. Correlation between baseline skin temperature (T0) measured immediately before trial onset and the thermal detection threshold of the behavioural responses triggered by Aδ- and C-fibre input, in each of the three experimental conditions (T₁ = unchanged, T₁ = 34°C and T₁ = 38°C) (Pearson's correlation coefficient, two-tailed p-values, adjusted for six comparisons using the Bonferroni correction).

Note the positive relationship between baseline skin temperature (T₀) and the Aδ-fibre threshold, in particular, when the stimulated area was brought for two seconds to T₁ = 34°C and T₁ = 38°C prior to applying the test stimulus.