"First pain" in humans: convergent and specific forebrain responses

Abstract

Background: Brief heat stimuli that excite nociceptors innervated by finely myelinated (Aδ) fibers evoke an initial, sharp, well-localized pain ("first pain") that is distinguishable from the delayed, less intense, more prolonged dull pain attributed to nociceptors innervated by unmyelinated (C) fibers ("second pain"). In the present study, we address the question of whether a brief, noxious heat stimulus that excites cutaneous Aδ fibers activates a distinct set of forebrain structures preferentially in addition to those with similar responses to converging input from C fibers. Heat stimuli at two temperatures were applied to the dorsum of the left hand of healthy volunteers in a functional brain imaging (fMRI) paradigm and responses analyzed in a set of volumes of interest (VOI).

Results: Brief 41°C stimuli were painless and evoked only C fiber responses, but 51°C stimuli were at pain threshold and preferentially evoked Aδ fiber responses. Most VOI responded to both intensities of stimulation. However, within volumes of interest, a contrast analysis and comparison of BOLD response latencies showed that the bilateral anterior insulae, the contralateral hippocampus, and the ipsilateral posterior insula were preferentially activated by painful heat stimulation that excited Aδ fibers.

Conclusions: These findings show that two sets of forebrain structures mediate the initial sharp pain evoked by brief cutaneous heat stimulation: those responding preferentially to the brief stimulation of Aδ heat nociceptors and those with similar responses to converging inputs from the painless stimulation of C fibers. Our results suggest a unique and specific physiological basis, at the forebrain level, for the "first pain" sensation that has long been attributed to Aδ fiber stimulation and support the concept that both specific and convergent mechanisms act concurrently to mediate pain.

Figure 1

Experimental protocol of fMRI experiment. The top row show the messages displayed to the subject in the scanner for each somatosensory stimulus. The stimulus was delivered after a 2-6 s delay ("Ready" message), followed by a 5 s "Wait" message before a 4 s "Rate stimulus" message was displayed, at which time the subject rated the stimulus on a five-button claw. The middle row illustrates how one stimulus (bars) is repeated 10 times, creating one run. The bottom row illustrates how each run is repeated with different stimulus modalities, making a total of 20 runs. Only stimulus intensities delivered to hairy skin (41°C and 51°C) is reported here. Intermingled were two heat stimulus intensities delivered to glabrous skin and one epidermal electrical stimulus. Results from these stimuli are reported elsewhere.

Figure 2

Vertex (Cz) potentials evoked by contact-heat pulses applied to the hairy skin of the left hand. Grand average responses were averaged across subjects after 10 stimuli delivered at 41°C (A) and at 51°C (B). Dotted lines mark the latencies of the major positive peaks (A and B).

Figure 3

Latency of major positive peak and pain intensity (NRS). A) Latency of the first major positivity in the heat-evoked potential at the vertex (electrode Cz). Latency of 51°C stimuli were significantly shorter than latency of 41°C stimuli (*p = 0.04). B) Average pain intensity ratings obtained during scanning. Ratings of 51°C stimuli were significantly more intense than ratings of 41°C stimuli (*p = 0.018). Values are mean ± standard deviation.

Figure 4

BOLD responses of Aδ predominant structures. Responses are from bilateral A Ins, contralateral hippocampus and ipsilateral P Ins. Each trace is the average BOLD response across subjects for 41°C (open circles) and 51°C (filled circles) ± standard error of the mean. The y-axis indicates % change from baseline (calculated as the average BOLD response over the preceding 10 s). Time resolution is 1.5 s (= TR). The 51°C BOLD peaks 1.5 - 3.0 s earlier than the 41°C BOLD in all these VOI, which is one of the three criteria established for Aδ predominant structures, as described in the text.

Figure 5

Brain locations of Aδ predominant activation. Activations in the bilateral A Ins, ipsilateral P Ins and contralateral hippocampus satisfied three criteria for being Aδ predominant: activation at 51°C, larger activation at 51°C than 41°C, and shorter 51°C BOLD latency than 41°C BOLD latency. Z-coordinates are in Talairach measurements.