The Journal of Neurophysiology Vol. 79 No. 5 May 1998, pp. 2557-2567
Copyright ©1998 by the American Physiological Society

Facilitation of a Nociceptive Flexion Reflex in Man by Nonnoxious Radiant Heat Produced by a Laser

Léon Plaghki¹, Dominique Bragard¹, Daniel Le Bars², Jean-Claude Willer³, and Jean-Marie Godfraind¹

¹ Faculté de Médecine, Université Catholique de Louvain, B-1200 Brussels, Belgium; ² Institut National de la Santé et de la Recherche Médicale U-161, 75014 Paris; and ³ Faculté de Médecine Pitié-Salpêtrière, Laboratoire de Neurophysiologie, 75634 Paris Cedex 13, France

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Plaghki, Léon, Dominique Bragard, Daniel Le Bars, Jean-Claude Willer, and Jean-Marie Godfraind. Facilitation of a nociceptive flexion reflex in man by nonnoxious radiant heat produced by a laser. J. Neurophysiol. 79: 2557-2567, 1998. Electromyographic recordings were made in healthy volunteers from the knee-flexor biceps femoris muscle of the nociceptive R_III reflex elicited by electrical stimulation of the cutaneous sural nerve. The stimulus intensity was adjusted to produce a moderate pricking-pain sensation. The test responses were conditioned by a nonnoxious thermal (40°C) stimulus applied to the receptive field of the sural nerve. This stimulus was delivered by a CO₂ laser stimulator and consisted of a 100-ms pulse of heat with a beam diameter of 20 mm. Its power was 22.7 ± 4.2 W (7.2 mJ/mm²), and it produced a sensation of warmth. The maximum surface temperature reached at the end of the period of stimulation was calculated to be 7°C above the actual reference temperature of the skin (32°C). The interval between the laser (conditioning) and electrical (test) stimuli was varied from 50 to 3,000 ms in steps of 50 ms. It was found that the nociceptive flexion reflex was facilitated by the thermal stimulus; this modulation occurred with particular conditioning-test intervals, which peaked at 500 and 1,100 ms with an additional late, long-lasting phase between 1,600 and 2,300 ms. It was calculated that the conduction velocities of the cutaneous afferent fibers responsible for facilitating the R_III reflex, fell into three ranges: one corresponding to A fibers (3.2 m/s) and two in the C fiber range (1.3 and 0.7 m/s). It is concluded that information emanating from warm receptors and nociceptors converges. In this respect, the present data show, for the first time, that in man, conditioning nonnociceptive warm thermoreceptive A and C fibers results in an interaction at the spinal level with a nociceptive reflex. This interaction may constitute a useful means whereby signals add together to trigger flexion reflexes in defensive reactions and other basic motor behaviors. It also may contribute to hyperalgesia in inflammatory processes. The methodology used in this study appears to be a useful noninvasive tool for exploring the thermoalgesic mechanisms in both experimental and clinical situations.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Direct investigation of neuronal convergence at the cellular level is difficult in animals and impossible in man. However, it is possible to use an indirect approach that involves testing interactions elicited by the activation of different pathways. In this respect, spatial and temporal interactions between afferent signals contributing to flexion reflexes have been studied in both animals and man (Lundberg 1979; Schomburg 1990). In all but one (Lundberg et al. 1977) of these studies, the nociceptive nature of the conditioning stimulation appeared to be the common denominator that allowed clear interactions between the afferent signals to be seen. More recently, Steffens and Schomburg (1993) showed that excitatory postsynaptic potentials evoked by cutaneous and joint afferents clearly were facilitated by conditioning radiant heat only if the stimulation temperatures were well above the pain threshold.

In animals, there are several lines of evidence that nociceptive flexor reflexes can be facilitated by homotopic noxious conditioning stimulation. Wall and Woolf (1984) reported that in the decerebrated spinal unanesthetized rat, a conditioning procedure of applying 20 electrical stimuli to the sural nerve at 1 Hz, resulted in a marked increase in the excitability of biceps femoris/semitendinosus motoneurons for 10 min. Only electrical conditioning stimuli at intensities that recruited afferent C fibers were effective in producing a prolonged facilitation of the reflex. This observation was confirmed repeatedly in the same preparation with nearly identical experimental protocols (Wiesenfeld-Hallin 1985; Wiesenfeld-Hallin et al. 1990, 1991; Woolf and Wall 1986; Woolf and Wiesenfeld-Hallin 1986). It also has been observed in spinal rabbits during recordings of reflex responses elicited in the nerves to the semitendinosus or gastrocnemius muscle by stimulation of the sural nerve. In this last case, the facilitation was seen after noxious stimulation of the heel or the application of a train of electrical stimuli to the sural nerve at an intensity that recruited A and C fibers (Catley et al. 1984; Clarke et al. 1992). Long-lasting postconditioning depolarizations have been observed during recordings from motoneurons in the neonatal rat spinal cord in vitro after homosynaptic or heterosynaptic activation by repetitive stimulation of high-threshold afferent fibers (Sivilotti et al. 1993; Thompson et al. 1993). Similarly, it has been reported that in the spinal unanesthetized rat, there is a decrease in the latency for hindpaw withdrawal from noxious heat after nociceptive electrical conditioning of the sciatic nerve (Cleland et al. 1994). A decrease in the tail flick latency also has been seen after noxious thermal conditioning of the tail in pentobarbital/chloral, lightly anesthetized intact rats (Cridland and Henry 1988; Yashpal et al. 1991, 1993).

Other studies of nociceptive reflexes have shown that noxious stimuli applied homotopically can be either facilitatory as mentioned earlier or inhibitory. Indeed, strong segmental inhibitory effects have been reported after high-intensity conditioning applied ipsilaterally on or near to the same dermatome (Catley et al. 1983; Chung et al. 1983; Clarke et al. 1988, 1989; Shin et al. 1986; Taylor et al. 1989-1991). Interestingly, it was found that the effects found in spinal animals, whether facilitatory or inhibitory, can be modified drastically in intact nontransected animals where descending inhibitory processes are fully operational (Fleischman and Urca 1988; Gozariu et al. 1995).

It is a common observation that conditioning nonnociceptive afferents do not reveal interactions with afferent signals produced by a nociceptive test stimulus; nevertheless it is still well accepted that fibers conveying nonnociceptive information contribute to the pool of "flexor reflex afferents" (FRA) (Eccles and Lundberg 1959). In view of the contribution of nonnociceptive afferents to FRA, the present study was aimed at reinvestigating the possibility that nonnociceptive thermal signals may interact with nociceptive signals in man by recording a well-defined nociceptive flexion reflex, the R_III reflex, elicited by applying test stimuli to a cutaneous nerve (the sural nerve); using a CO₂ laser stimulator to provide nonnoxious thermal conditioning stimulation of the receptive field of this nerve; and systematically varying the delay between the conditioning and test stimuli.

The R_III reflex is elicited mainly by activation of A- afferent fibers and represents an objective physiological index which is highly correlated with the sensation of pain produced by electrical stimulation of the sural nerve (Willer 1977, 1984; Willer and Albe-Fessard 1983).

Laser stimulation allows precise control of stimulation parameters, elicits a well-synchronized afferent volley of impulses and can produce natural and selective thermal stimulation of low-thresholds afferents. Depending on the amount of energy delivered, the CO₂ laser stimulator produces warm and/or painful sensations by direct activation of cutaneous thermoreceptors and/or nociceptors connected exclusively to A and C fibers (Bromm et al. 1984; Devor et al. 1982).

In this paper, it will be shown that conditioning nonnociceptive afferents does facilitate the R_III flexion reflex at various conditioning-test intervals that, by use of a mathematical model can be shown to correspond to the activation of peripheral A and C fibers.

	METHODS

Abstract Introduction Methods Results Discussion References

Subjects

Experiments were carried out on 15 healthy volunteers: 7 women and 8 men, mean age 26 ± 5.7 yr (range 20-45). The rules of the local ethics committee were followed. Subjects gave their informed consent after being carefully briefed on the experimental procedures and purposes. To ensure good muscle relaxation, they were installed comfortably in an armchair. They were familiarized with the perception of electrical and laser stimuli and instructed in all the usual manipulations performed during a session. They wore protective goggles in line with the guidelines for control of laser hazards. To avoid any auditory cues that may have been associated with the production of the stimuli, the subjects wore headphones (Ultramuffit, Racal Safety).

Procedure for the test R_III reflex

The method for eliciting and recording the nociceptive R_III flexion reflex from a knee-flexor muscle was based on that described previously (Willer 1984). Briefly, bipolar electrical stimuli were applied via a pair of surface electrodes placed 20 mm apart over the submalleolar pathway of the right sural nerve (Fig. 1A). The stimulus consisted of a volley of eight rectangular constant-current pulses, 0.5 ms in duration, delivered over 20 ms through an isolation device (Digitimer Model DS7) driven by a pulse-generator (Master-8, AMPI). The interstimulus interval varied from 5 to 11 s. Reflex activity evoked in the ipsilateral biceps femoris was recorded through a pair of surface Ag-AgCl electrodes (Meditrace model ECE1801) placed on the skin over the muscle, and connected to a Gould Universal Amplifier (Model 13 4615-58) with a band-pass of 10 Hz to 2 kHz and a gain of ×25,000. The upgoing edge of the electrical stimulus triggered data acquisition by a small computer (Cambridge Electronic Design Model 1401) for a period of 500 ms, at 4,000 cps. The threshold of the R_III reflex first was defined as the stimulus intensity eliciting liminal reflex responses with a probability of 80-90%. The current intensity was then adjusted to 25% above this threshold and was kept constant for the rest of the experiment. This intensity [mean: 14.4 ± 6.3 (SD) mA] produced supraliminal reflex responses with minimal fluctuations from one stimulus to the next (Fig. 1B) (Willer 1984) and a sensation of moderate pricking pain, which was well tolerated by the subjects at least for the duration of an experimental session (20 min).

View larger version (16K): [in this window] [in a new window]   FIG. 1. Experimental setup and recording examples. A: experimental set-up. Test RIII reflex was recorded from the ipsilateral biceps femoris after electrical stimulation delivered over the submalleolar path of the sural nerve at a constant intensity of 1.25 times threshold. This test electromyographic (EMG) response was or was not (controls) conditioned by a nonnoxious CO2 laser radiant heat pulse applied to the receptive field area of the sural nerve preceding the conditioned stimulation at random interval of between 50 and 3,000 ms. B: example of unconditioned and conditioned RIII reflex responses. As shown below the recordings, the responses were full-wave rectified. C: full-wave rectified responses in the period 80-160 ms after the electrical test stimulus onset were used to provide an initial representation of the effects of the conditioning stimulus. Median of the 60 unconditioned responses then was calculated, and the individual unconditioned response closest to that value was selected, as illustrated for 1 subject in Fig. 1C1 (). Variations of amplitude (in V) of the EMG responses were built (C2) by subtracting from each bin, the median rectified unconditioned EMG response ( in C1) from the individual conditioned EMG response ( in C1).

Conditioning procedure

The conditioning stimulus was delivered by a CO₂ laser to the cutaneous area situated at the dorsum of the right foot and corresponding to the receptive field of the sural nerve (Fig. 1A). It consisted of a 100-ms duration heat pulse with a beam diameter of 20 mm and power rating of 22.7 ± 4.2 W (7.2 mJ/mm²). With such parameters, the maximum temperature reached at the end of the stimulation period was 7°C (or 9°C with + SD of power) above the reference skin temperature, i.e., 32°C (for details of these computations, see Haimi-Cohen et al. 1983). In these conditions, the cutaneous temperature should not exceed 40°C in the superficial epidermis, and thus even the most superficial nociceptive endings should not be activated (Bromm and Treede 1983). Indeed, the stimuli were perceived as warm but never as painful. Before starting the experimental procedure, it was confirmed that the conditioning stimulus when delivered alone, did not elicit an electromyographic (EMG) response in the biceps femoris. The CO₂ laser device was an improved version of the machine conceived and built in the Department of Physics at the Université catholique de Louvain (Plaghki et al. 1989). The reproducibility of the laser stimuli was accurate at 1%. A He-Ne laser beam (Hughes Aircraft; wavelength 632.8 nm, power output 5 mW) was used to show visually the site of stimulation. The laser equipment was kept in an adjacent room to reduce noise and heat. The laser beam was deflected, through a hole in the wall, on to the foot of the subjects by means of a remotely controlled gold-coated flat mirror. The position of the laser beam was checked via a closed circuit television video; it was displaced slightly after each laser stimulus to avoid any overheating of the skin. All parameters (output power, stimulus duration, beam diameter and target site) were controlled from the experimental room.

General experimental procedure

During an experimental session, once the electrical stimulus intensity had been determined (as defined above), a total of 120 R_III reflex responses were studied. Sixty of these were unconditioned (control responses, i.e., those associated with a sham stimulation) and 60 were conditioned, i.e., preceded by a thermal stimulus. All these responses were elicited in a completely random order determined by a computer program. The conditioning-test interval (t) between the laser (conditioning) and electrical (test) stimuli was varied from 50 to 3,000 ms in steps of 50 ms.

Processing of data

The EMG records containing the reflex responses was full-wave rectified and integrated over a time window of 80 ms starting 80 ms after the test stimulus onset (Fig. 1B). The magnitudes of the responses (whether control or conditioned) measured in this fashion had a distribution that was skewed positively (Snedecor and Cochran 1989). In accord with the normality hypothesis being unfulfilled, medians rather than means were calculated.

The median of the 60 control responses was calculated for each of the 15 subjects. Then, the 60 conditioned responses, each corresponding to a given conditioning-test interval, and the 60 control responses were normalized to this median. This yielded for each conditioning-test interval, 15 normalized conditioned responses and 15 normalized control responses, the median of which was calculated separately and taken as the grand median of the control responses (i.e., the median of the 15 × 60 control responses). Based on these normalized data, time series vectors (control and conditioned) of equally spaced points in time (t = 50 ms; n = 60) were obtained and submitted to a resistant nonlinear smoothing procedure (3RSSH Tukey's notation; STATGRAPHICS, STSC) (see Figs. 2B and 3). This procedure of nonlinear smoothing was chosen because it is based on a "running" median and is resistant to "isolated outliers." Because this procedure has to find the median for three values surrounding each point, the two first and two last endpoints of the times series vectors were omitted in Fig. 3.

View larger version (28K): [in this window] [in a new window]   FIG. 2. Waterfall display of results obtained from 1 subject. A: typical example of results obtained from 1 subject (male, aged 29, stature 1.92 m) illustrating the influence of the laser thermal stimulus on the RIII reflex. Abscissa indicates time (in ms) with 0 corresponding to the occurrence of the test electrical stimulus; only the 80- to 160-ms time window after the stimulus is shown. Ordinate indicates the variation in amplitude (in µV) of the EMG responses by subtracting from each bin the median response of the 60 rectified control EMG responses from the individual conditioned EMG responses (see Fig. 1C). Sixty conditioned responses of the biceps femoris muscle are presented in a waterfall display. From bottom to top, each sweep corresponds to a response conditioned by a laser stimulus and elicited by an electrical test stimulus delivered with progressively increasing time intervals (t) in steps of 50 ms. B: graph of the ratio of the responses integrated and normalized to the median of the unconditioned responses as a function of the time interval t between conditioning and test stimuli. RIII reflex response was noticeably enhanced when intervals between the conditioning laser heat stimulus and the test electrical stimulus were ~500, 1,000, and >1,700 ms.

View larger version (39K): [in this window] [in a new window]   FIG. 3. Modulation of RIII reflex responses by conditioning nonnoxious heat stimulation. Influence of the conditioning laser thermal stimulus on the nociceptive reflex RIII shown for data pooled from all 15 subjects. Curve with filled circles represents the data of the smoothed normalized EMG evoked reflex activities (see METHODS) as a function of the time interval t between the conditioning laser stimulus and the test electrical stimulus. Curve with open circles represents the smoothed normalized data computed from the unconditioned responses, i.e., those associated with a sham stimulation. Two curves are clearly different. As compared with the controls, the curve obtained from the conditioned EMG responses shows 3 peaks at 500 ms (with the fastest conditioning effect noticed ~250 ms), 1,100 ms, and a late, large peak between 1,600 and 2,300 ms. Mean conduction velocities of any peripheral fibers excited by a conditioning stimulus when convergence and coincidence of the information conveyed by the afferent fibers activated by the test and the conditioning stimulus occurred, were computed as in Fig. 4 (top). t values corresponding to the 3 peaks matched the criteria defined by Eq. 2 and Fig. 4: 1 in the range of A fibers and 2 in the range of C fibers (see also Table 2).

To achieve a more comprehensible representation of data from a single subject (Fig. 2A), the variations in amplitude of the EMG responses were determined for the period 80-160 ms after the test stimulus; this was done by subtracting from each bin, the median value of the 60 rectified control responses from the individual conditioned response (see Fig. 1C).

Theoretical calculations

In this section, an attempt will be made to identify the type of peripheral nerve fibers that were activated when cutaneous sensory receptors were excited by the nonnociceptive thermal CO₂ laser stimuli. This will be done by calculating the fibers' conduction velocities by taking account of the spatio-temporal interactions, which occurred when changes in the excitability of the motoneuronal pool to the electrical test stimulus appeared after heat laser stimulation. As already shown in Fig. 3, such changes occurred only when the test response was timed properly after the conditioning stimulus.

Initial equation

During any period of facilitation of the R_III reflex, convergence and coincidence of information being conveyed by afferent fibers activated by the electrical test stimulus and the laser stimulus must have occurred in the CNS.

Thus for any timing of a facilitation, one can write the following equation

(1)

where

tpe =	propagation time for the electrical current through the skin
tae =	activation time of the nerve fibers by the electrical current
tde =	delay resulting from conduction time of peripheral fibers activated by the electrical current
tcde =	central delay time of circuits activated by the electrical current
t =	time interval between the conditioning and the test stimulus
Z =	uncertainty parameter that affects t
tpl =	propagation time of thermal energy through the superficial layers of the skin
tal =	activation time of sensory receptors by thermal energy
tdl =	delay resulting from conduction time of peripheral fibers activated by the thermal laser stimulus
tcdl =	central delay time of circuits activated by the thermal laser stimulus.

We will review all these parameters step by step. Because most of them could reasonably be estimated, we will propose a final Eq. 2 giving a relationship between the experimentally controlled parameter t and the conduction velocities (V_l) of peripheral fibers in the sural nerve, which we presume were activated by the laser heat stimulus.

Parameters related to peripheral processes elicited by electrical stimulation

It generally is accepted that electrical stimulation directly activates peripheral nerve fibers. The propagation time of the electrical current (t_pe) through the skin and body fluids and the activation time (t_ae) of the nerve fibers both may be regarded here as negligible values.

Based on the work of Willer and Albe-Fessard (1983) and the source book of Liveson and Ma (1992), the peripheral fibers of the sural nerve involved in the R_III reflex are known to conduct impulses at a velocity (V_e) ~35 m/s. Because the distance of the peripheral path (D_e) may be estimated easily by measuring the distance between the stimulation site and the spinal process of T12, the peripheral conduction time (t_de) can be calculated as: t_de = D_e/V_e.

Parameters related to peripheral processes elicited by laser stimulation

The thermal energy sent by the laser stimulator reaches sensory receptors after a propagation time (t_pl) through the superficial layers of the skin and activates the receptors after an activation time (t_al). Their respective values presented in Table 1 are derived from the work of Haimi-Cohen et al. (1983) and of Bromm and Treede (1983, 1984).

The peripheral path distance (D_l) corresponds to the distance between the stimulation site and the spinal process of T₁₂. Unlike the electrical test stimulus, the thermal laser stimulation used in the present study did not elicit a muscle reflex; therefore, neither the conduction velocity (V_l) nor the peripheral conduction time (t_dl) could be measured in the present noninvasive experiments. However the relation t_dl = V_l/D_l will be used in the calculations described later.

Parameters related to central delays

After the peripheral processes elicited by electrical stimulation, a central delay (t_cde) is required for related information to reach the motoneuronal pool and trigger activation of the biceps femoris muscle. After the peripheral processes elicited by laser stimulation, a central delay (t_cdl) also is required for related information to reach the motoneuronal pool and influence its excitability. If these central delays are equal, then these two parameters cancel each other in Eq. 1. If the values of these parameters are not equal, their difference t_d = |t_cde  t_cdl| should be taken into account. Such a difference could derive from the propagation of action potentials through different neuronal networks, one of these possibly including additional synapses. We decided to take into account a greatly overestimated possible difference of 20 synapses. In fact, taking into account a difference of 20 synapses, computational analysis based on Eq. 2 showed that this factor was of limited importance with regards to the other parameters.

Time elapsed between conditioning and test stimuli

The parameter that has been manipulated experimentally is t, i.e., the time elapsed between application of conditioning and test stimuli. More particularly, t was the time elapsed between the beginning of the 100-ms thermal stimulus and the beginning of the 20-ms train of electrical pulses. However, because the exact timing of the start of nerve fiber excitation may have varied during the stimulating period, the time elapsed between the actual effective conditioning and the test stimuli also may have varied. This uncertainty is represented by the parameter Z. According to the durations of both thermal and electrical stimuli, the (t + Z) value may vary from (t  100 ms) to (t + 20 ms).

Final equation and computations

By rearranging Eq. 1, and applying the simplifications mentioned above, it may be rewritten

(2)

Based on Eq. 2, the relation between V_l and t may be examined for the three sets of parameter values listed in Table 1. These values were chosen on the basis of the above-mentioned reports in the literature and taking into account under- and overestimations.

Results of the computational analysis are presented in Fig. 4. Three curves were calculated using underestimated, averaged, and overestimated parameters, respectively, as shown in Table 1. For t values >1 s, no significant differences were seen among the three curves (Fig. 4, front panel). By contrast, for t values <500 ms, and much more dramatically for t values <250 ms, differences between the curves were greatly increased at decreasingvalues of t (Fig. 4, inset). Thus Eq. 2 yields results that makesense from a biological point of view only for (t + Z + t_de t_pl  t_al + _td)  0. Indeed, the sum of the parameter values(Z + t_de  t_pl  t_al + _td) equal + 26, 100, and 180 ms, respectively, for the under-, mean, and overestimated parameters (Table 1). As a consequence, when using the mean and overestimated parameter values respectively, computing the conduction velocity for t lower than ~150 and 250 ms yields inconsistent figures.

View larger version (21K): [in this window] [in a new window]   FIG. 4. Computation of the conduction velocity of primary afferent fibers engaged in convergence. Theoretical relationship between the conditioning-test interval t and the conduction velocity Vl of any peripheral fibers excited by a conditioning stimulus when convergence and coincidence of the information conveyed by the afferent fibers activated by the test and the conditioning stimulus occurred. By computation of Eq. 2, Vl = Dl/(t + Z + tde  tpl  tal + td), 3 curves were obtained using the underestimated, averaged, and overestimated values of the parameters, respectively, as listed in Table 1. First part of the curves corresponding to the lower t values (0- to 0.75-s range), are shown at the back with a higher magnification of the abscissa. Calculated values of Vl were added as the top abscissa in Fig. 3.

For the t values corresponding to the peaks illustrated in Fig. 3, i.e., corresponding to a facilitation of the electrically induced nociceptive reflex, conduction velocities of the peripheral fibers excited by the thermal laser stimulus can be computed from the three curves shown in Fig. 4. The results are shown in Table 2. In addition, the conduction velocities of the peripheral fibers excited by the thermal laser stimulus, as estimated from the curve calculated with averaged parameters, are reported on the top abscissa of Fig. 3 for graphic examination. The facilitations elicited when t values were ~500 ms (with the fastest conditioning effect starting ~250 ms), correspond to laser-induced activities in small-diameter myelinated fibers of the A type. The facilitations observed for t values ~1,100 ms and those ~1,600-2,300 ms, result from laser-induced activities in thin nonmyelinated nerve fibers of the C type; in addition, two separate C-fiber groups could be defined: the first with conduction velocities of 1-1.5 m/s (C1), and the second with conduction velocities <1 m/s (C2).

	RESULTS

Abstract Introduction Methods Results Discussion References

Experimental observations

In the 15 healthy volunteers, we examined whether nonnoxious radiant heat produced by a laser could influence the R_III flexion reflex elicited by nociceptive electrical stimulation of the sural nerve (Fig. 1A). The thermal stimulus was perceived by the subjects as warm but never painful. When applied alone, it never elicited an EMG response in the biceps femoris.

As illustrated with an individual example (Fig. 1B), the application of the laser stimulus with a 1,950-ms conditioning-test interval resulted in a clear facilitation of the R_III response (Fig. 1B2) as compared with the control response (Fig. 1B1). Further processing of the data involved both responses being full-wave rectified (Fig. 1B, bottom) and considered within the period 80-160 ms after the test stimulus onset. The median of the 60 unconditioned responses then was calculated, and the individual unconditioned response closest to that value was selected, as illustrated for one subject in Fig. 1C1 (). Finally, the variations in amplitude of each conditioned EMG response were computed by subtracting from each bin, the value of the median rectified unconditioned response from the individual rectified conditioned response (Fig. 1C2).

Figure 2A shows a typical example of the result obtained in this way from one subject, illustrating the influence of the laser heat stimulus on the R_III reflex. This figure was constructed by sequentially displaying the increased reflex activity (ordinate,  amplitude) elicited by the conditioning thermal stimulus, with the time after the test stimulus running from left to right (abscissa) and the conditioning-test intervals (t) increasing from front to back (z axis). The facilitation of the 60 conditioned reflex responses are illustrated in perspective using a waterfall display presentation. The data, initially acquired with conditioning-test intervals applied at random in the 50-3,000 ms range, were rearranged in such a way that each sweep of the figure corresponds to a test response conditioned by a laser stimulus with progressively increasing conditioning-test intervals from bottom to top in 50-ms steps. Note the facilitation of the R_III reflex for some conditioning-test intervals, separated by periods of "silence" during which individual EMG responses were nearly identical to the control R_III reflex.

The results obtained from this individual subject are shown quantitatively in Fig. 2B. The results are expressed in terms of the ratio (ordinate) between each individual response and the median unconditioned response as a function of the conditioning-test interval (abscissa). Both response values were calculated during the 80-160 ms after the test stimulus. The curves were obtained after a resistant nonlinear smoothing treatment. Note that the R_III reflex clearly was enhanced when the conditioning-test intervals were ~500 and 1,000 ms and when they were >1,700 ms.

The profile of this curve based on a single subject (Fig. 2B) was similar to that for data pooled from the 15 subjects. Figure 3 shows two curves obtained after the resistant nonlinear smoothing treatment of the normalized conditioned and unconditioned responses (see Processing data). The curve with filled circles represents the data of the smoothed normalized EMG evoked activities as a function of the interval between the conditioning laser stimulus and the test electrical stimulus. The curve with open circles represents the smoothed normalized data computed from the unconditioned responses. The two curves are clearly different. The curve obtained from the conditioned EMG responses shows three peaks at conditioning-test intervals of 500 and 1,100 ms and between 1,600 and 2,300 ms. These peak values, together with other parameters were used to compute the conduction velocity of the cutaneous afferent fibers activated by the conditioning laser heat stimulus (see Theoretical calculations). In Fig. 3, the top abscissa shows the computed conduction velocity plotted against the corresponding EMG data. Based on the computed conduction velocity, three different groups of afferent fibers were recognized: one in the A-fiber range and two in the C-fiber range (see Table 2).

	DISCUSSION

Abstract Introduction Methods Results Discussion References

It was found that the nociceptive flexion reflex was facilitated by nonnoxious thermal stimulation applied to the peripheral receptive field of the sural nerve. Such a modulation occurred with definite time intervals between the conditioning and test stimuli, namely ~500, ~1,100 ms, and a late, long-lasting phase between 1,600 and 2,300 ms. The simplest interpretation of our observations is that nociceptive and nonnociceptive information converges and interacts on a common pool of interneurons. The occurrence of such interactions implies a temporal coincidence of information emanating from the two areas of stimulation. This was obvious only with adequate time intervals between the conditioning and test stimuli. Because the R_III reflex is present in patients with total sections of the spinal cord (Roby-Brami et al. 1987; Willer and Bussel 1980), one can assume that its basic pathway is mainly spinal. Thus the most parsimonious interpretation of our present data is that the convergence between noxious and nonnoxious signals occurred in the spinal cord with the various peaks of facilitation being determined mainly by the time required for the synchronized peripheral volleys elicited by the heat stimulus to reach the common final pool of interneurons. Therefore the different facilitatory waves can be explained by the triggering of activity in fibers with different peripheral conduction velocities.

In spite of the parsimony of our interpretation of the present results, three other alternative interpretations could be put forward: sensitization, variation in central delays, and oscillations of cord excitability. Sensitization (see references in Meyer et al. 1994) is unlikely to introduce a bias in the present results for two reasons. First, the position of the laser beam was moved slightly after each laser stimulus to avoid any skin damage due to repetitive stimulation. Second, the stimulus intervals were applied randomly, and thus the rank order of the applied stimulus did not appear to be a significant source of variability. Variations in central delays may result from additional synapses in segmental networksand/or in long loops. These variations appear to be of limited importance given that the introduction of an uncertainty factor as large as 20 ms in the mathematical model (see Z parameter in Theoretical calculations) had little impact on the determination of the conduction velocities. Finally, it could be questioned whether the three peaks do reflect oscillations of cord excitability triggered by the conditioning laser stimulus. Indeed mechanisms proposed to play a major role during spinal processing (convergence, temporal, and/or spatial summation and assembly coding) may depend on neuronal discharge synchronization. In this line, it has been reported that oscillations arise locally in the dorsal horn of the spinal cord and induce cooperative activity of spatially distributed neuronal subpopulations (Eblen-Zajjur and Sandkühler 1997). Dorsal root stimulation also has been shown to trigger oscillations in the motoneuron pool (Baranauskas and Nistri 1995) with a power spectrum peaking at 8 Hz. However, power spectrum of the present data peak <2 Hz. As no such low-frequency oscillator has been observed, our proposition seems more likely. Higher frequencies of oscillations, if present, should have been observed in our data given the t of sampling rate of 50 ms. Besides, radiant heat stimulation is known to activate different types of peripheral nerve fibers in the A and C range (references below).

The peak values in Fig. 3, together with the mathematical modeling, helped us to compute the conduction velocities of the cutaneous afferent fibers activated by the conditioning thermal stimulus. Three different groups of afferent fibers were identified: one in the A-fiber range (3.2 m/s) and two in the C-fiber range (1.3 and 0.7 m/s). This is consistent with the fact that the CO₂ laser stimulator activates exclusively A and C fibers (Bromm et al. 1984; Devor et al. 1982) and evokes sensations through direct activation of cutaneous thermoreceptors and/or nociceptors, depending on the amount of energy delivered.

The discussion will be organized into several sections as follows: facilitation of a flexion reflex by thermal stimuli; putative peripheral fibers involved in the facilitation; putative spinal neurons involved in the facilitation; and methodological considerations.

Facilitation of a flexion reflex by thermal nonnoxious or noxious stimuli

To the best of our knowledge, there have been only two studies that have used an experimental design close to that used in the present work and that merit consideration.

Andersen et al. (1994) reported that a reflex response from the tibialis anterior muscle was facilitated by a painful thermal conditioning stimulus applied to the sole of the ipsilateral foot with conditioning-test intervals from 600 to 3,000 ms. This effect was attributed to fibers with a broadband of conduction velocities within the C-fiber population being activated by the conditioning stimulus. However, because these authors used seven fixed conditioning-test intervals(0, 300, 600, 800, 1,000, 1,200, and 3,000 ms), it was quite impossible to differentiate with accuracy several peaks of facilitatory effect as we did here by considering 60 conditioning-test intervals between the laser and electrical stimuli. Besides these differences in the resolution of conditioning-test intervals, there were others between the present study and that work, notably because in the latter EMG reflex activity was elicited from the tibialis anterior muscle by electrical stimulation of the sole of the foot whereas we recorded a well-defined nociceptive flexion reflex from the biceps femoris muscle elicited by stimulation of A- fibers in a cutaneous nerve, the sural nerve; because the conditioning stimulation was provided by a Xenon lamp (visible light) and was noxious whereas in our study, a mild sensation of warmth was described by the subjects; and because the test stimulus was applied to the sole of the foot, whereas in our study it was applied on the dorsum of the foot, i.e., within the receptive field of the sural nerve. All these differences make comparison between the present study and that of Andersen et al. (1994) quite difficult.

In the second study (Willer and Albe-Fessard 1983), the R_III reflex was conditioned by painful electrical stimuli(30-mA intensity, 1-ms duration) applied to the sural nerve territory. Facilitations of the R_III reflex were induced by such conditioning stimuli and were attributed to the activation of afferent fibers with conduction velocities in the 0.8-25m/s range. In this case, however, the conditioning stimulus was also nociceptive in nature, and facilitations were in the 150-200% range. By contrast, the moderate facilitations (15-20%) observed in the present study were elicited by nonnoxious conditioning radiant heat. In any case, as far as we know, the facilitation of nociceptive reflexes by nonnoxious thermal stimuli has not been reported to date.

Putative peripheral fibers involved in the facilitation

It was possible to subdivide into three groups, the peripheral fibers which when activated facilitated the R_III reflex because of the time resolution resulting from the experimental design. However, as suggested in Fig. 3, the first peak of activation might in fact be composed of two components, both in the range of A-fiber conduction velocities. However, before a definite conclusion can be made about this, further experiments involving shorter time intervals (t) between the test and conditioning stimuli will be required.

It should be emphasized that the three groups of fibers that we are proposing were involved in the modulation of the R_III reflex, conveyed information subserving warm perception. This is supported not only by the verbal reports of the subjects but also by the computation of maximal temperatures reached in the center of the laser beam, which were always <40°C (see Conditioning procedures).

The responses of peripheral receptors and afferent fibers to thermal stimulation have been studied widely with microneurographic techniques (e.g., Beck et al. 1974; Vallbo et al. 1979; Van Hees and Gybels 1972). Several classes of cutaneous receptors that may respond to thermal stimuli are connected to slowly conducting thin fibers, either myelinated (A) or unmyelinated (C). The afferent fibers that selectively respond to thermal stimuli generally are classified either as cold receptors, connected to A fibers or warm receptors, connected to C fibers. This dichotomy is based mainly on the observation that warm sensibility is blocked earlier than cold sensibility after local anesthesia (Fruhstorfer et al. 1974; Sinclair and Hinshaw 1950). However, this separation between A cold and C warm receptors is not absolute (Spray 1986) and, as discussed later, several types of peripheral receptors might have contributed to the results reported herein.

In the myelinated afferent fiber group, the A cold receptors are known to operate optimally in the 15-30°C range. In the monkey, the mean conduction velocity of these afferent fibers is 9.0 m/s with a large range from 1.7 to 16.7 m/s (Dubner et al. 1975). These receptors present resting discharges at 11 Hz at the optimal temperature of 25°C and disappear beyond 38°C (Beitel et al. 1977; Darian-Smith et al. 1973; Sumino and Dubner 1981). Such a decreased activity of the A-cold receptors in the 25-38°C range makes the participation of this class of neurons unlikely in the facilitation of the R_III reflex.

Observations of A-fiber activity associated with warm perception have been reported but the proportion of such afferent fibers is low compared with cold receptors (Darian-Smith et al. 1973). In the monkey, facial warm receptors connected to A and C fibers with conduction velocities in the 0.9-6.7 m/s range have been reported (Sumino et al. 1973). Interestingly, the warm receptors described by Terashima and Liang (1993) in man, are linked to A fibers with conduction velocities in the 3.8-11.2 m/s range. These conduction velocities are reasonably close to those computed in the present study (3.1-8.1 m/s); such fibers may have contributed to the early facilitation of the R_III reflex reported herein.

A priori, two other types of A receptors also might have contributed to these effects. In the first subclass, the A-HTM (high-threshold mechanical) or type I A-MH (mechano-heat), the A receptors respond to strong mechanical stimulation and usually remain insensitive to noxious heat and chemical stimulation. They have high thermal thresholds (>49°C) and sometimes rapid conduction velocities (mean 31 m/s, range 5-53 m/s) (Campbell et al. 1979). In the second subclass, the mechano-thermal receptors (MTR) or type II A-MH, the A- receptors respond to strong mechanical stimuli, noxious heating, and algogenic chemicals. They present lower thermal thresholds (range: 37-47°C) and slower conduction velocities (mean: 15.2 ± 9.9 m/s) (Dubner et al. 1977). Although the former group is unlikely to have contributed to the early facilitation of the R_III reflex reported herein, the latter might have contributed to some extent.

Regarding the unmyelinated fibers, two classes of receptors are concerned with warm stimuli. In the first, theC-warm receptors are included in afferent fibers that respond optimally to temperatures ~40°C, i.e., temperatures that generally are perceived as gentle warmth. Their conduction velocities are reported to be between 0.6 and 2.4 m/s, with mean values between 0.7 and 1.2 m/s (Darian-Smith et al. 1979; Duclaux and Kenshalo 1980; Hensel and Iggo 1971; LaMotte and Campbell 1978); these are reasonably close to those computed in the present study (0.7-1.3 m/s). Such receptors were therefore probably involved in the late facilitations of the R_III reflex reported herein.

In the second class, the C-mechano-heat nociceptors(C-MHs), the unmyelinated afferent fibers respond to high-intensity mechanical and thermal stimuli. They were shown to sometimes respond to chemical stimuli and thus also are termed polymodal. Their thermal threshold ranges between 38 and 49°C with mean values ~44°C (Beitel and Dubner 1976; Robinson et al. 1983). They show no spontaneous activity below 38°C (Hallin et al. 1981; LaMotte and Campbell 1978). In humans, the conduction velocity for this group is reported to be in the 0.7-1.3 m/s range with mean values ~0.8 m/s (Gybels et al. 1979; Van Hees and Gybels 1972). Even though their participation cannot be ruled out completely, it is unlikely that such fibers would have contributed significantly to the thermally induced late facilitations of the R_III reflex.

Although the distribution of C-fiber conduction velocities is most commonly said to be unimodal, it was found to be bimodal in the present study. As far as we know, the existence of two groups of conduction velocities of C fibers was reported by Douglass et Ritchie (1959). However, these C fibers were not sensitive to nonnoxious warm stimulation. In a study of C-fiber warm receptors, Darian-Smith et al. (1979) provided a histogram of the distribution of their conduction velocities: it was clearly unimodal centered on 1.2 m/s. Two distinct classes of C-MH receptors were described in the monkey by Meyer and Campbell (1981) with either quickly or slowly adapting responses. However these authors do not rule out the possibility that this difference depended on the depth below the skin surface at which the receptor was located. Interestingly, latency distributions of C-fiber-evoked reflexes elicited by electrical stimulation of the sural nerve in the rat are bimodal, suggesting existence of two populations of C fibers with different conduction velocities (0.6 and 0.9 m/s) (see Falinower et al. 1994). Similar response latencies with a bimodal distribution also have been observed during recordings of rat dorsal horn convergent neurons after electrical stimulation of their receptive fields (e.g., see Villanueva et al. 1986; this study).

Putative spinal neurons involved in the facilitation

As mentioned already, the convergence of afferent signals evoked by the nonnoxious thermal conditioning and the nociceptive test stimuli most probably occurred in the spinal cord. For many reasons, such a convergence is not conceivable at the level of the motoneuronal pool responsible for the activation of the biceps femoris muscle. Radiant heat below the threshold for nociceptive afferent fibers did not elicit synaptic effects in  motoneurons nor facilitation on reflex pathways from Group Ib or II muscle afferents or mechanoreceptive cutaneous afferents (Behrends et al. 1983; Burke et al. 1971; Steffens and Schomburg 1993). Because all A and C fibers terminate in the dorsal horn of the spinal cord, it follows that the functional site of convergence was located on interneurons situated upstream of the motoneuronal pool.

One subpopulation of dorsal horn neurons fulfills some of the properties required for being the locus of convergence, and several pieces of evidence suggest that they may well belong to the reflex arc involved in the R_III reflex. Indeed, the R_III reflex recorded in humans and the convergent neurons recorded in animals are modulated by common factors: both are inhibited by systemic or intrathecal morphine (see ref. in Willer and Le Bars 1995); both are inhibited by transcutaneous electrical nerve stimulation (Handwerker et al. 1975; Rosenberg et al. 1996); both are inhibited by heterotopic noxious stimuli (see references in Willer et al. 1989) via supraspinal structures located in the lower medulla (Bouhassira et al. 1995; De Broucker et al. 1990). A subset of such neurons, the so-called "warming/noxious heat-convergent units" have thresholds mainly <42°C, increase their discharge over a limited range of temperatures and most commonly reach a plateau or even decline slightly when noxious temperatures are achieved (Burton 1975; Kanui 1988; Le Bars and Chitour 1983; Menétrey et al. 1979; Price and Browe 1975). This subset of neurons may well be a candidate as the locus of convergence. However, involvement of ventral horn interneurons cannot be excluded because temperature-sensitive neurons have been described within laminae VII and VIII (Burton 1975).

Methodological considerations

We believe that the present results provide and validate a potentially useful noninvasive method not only for further studies concerned with central interactions among information from various origins but also for those studies that will deal with the indirect study of peripheral fibers in man. It is clear that a certain degree of convergence is implied within the pathways underlying the R_III reflex. Admittedly, absence of facilitation by no means indicates absence of fiber activation by the conditioning stimulus; but occurrence of facilitation clearly indicates that a group of fibers actually have been activated by the conditioning stimulus. As quoted by Schomburg (1990) "spatial facilitation depends on a distinct subliminal fringe around the discharge zone in the interneuronal pool shared by the two pathways." Thus being aware of the possible pitfalls of false negative results and considering the difficulty of investigating, in man, peripheral fibers with slow conduction velocities, the experimental method used here may be a useful tool in both experimental and clinical situations. In this respect, it is important to note that the routine electrophysiological methods used in clinical departments investigate transmission only in afferent fibers with the largest diameter and cannot explore the thermoalgesic system (Noël and Desmedt 1975; Veilleux and Stevens 1987).

We will comment briefly on some possible functional implications of the present results. Under the term "erythralgia," Lewis (1942) described temperature-dependent pain elicited by experimentally induced local redness and tenderness of the skin. Using an in vitro testis-superior spermatic nerve preparation, Kumazawa et al. (1987) provided convincing evidence that such observations could be explained by a potentiation of the response of polymodal nociceptors by increases of temperature in the nonnoxious 34-43°C range (at least as far as the viscera are concerned). Because inflammation is associated with local increases in temperature, this may contribute to clinical hyperalgesia. However, as shown here for cutaneous stimuli, nonnoxious increases in temperature also could result in a central booster by means of convergence between information emanating from warm and nociceptive receptors. Together with the effects of the "inflammatory soup" (see reference in Dray 1994; Levine and Taiwo 1994), such factors potentially could contribute to the hyperalgesia seen during inflammatory processes.

In conclusion, the present data suggest indirectly that warm nonnociceptive information can be conveyed by both A- and C-afferent fibers and can modulate the R_III reflex in man. Although these effects were moderate, in the 15-20% range, they do suggest that such warm afferents may contribute to some extent to the FRA and add a useful signal to that of other information to trigger flexion reflexes in defensive reactions and other basic motor behaviors.

	ACKNOWLEDGEMENTS

  We thank Dr. S. W. Cadden for advice in the preparation of the manuscript. We also thank Prof. A. Fayt and P. Stouffs (physicist engineer) of the Molecular Spectroscopy Laboratory (Universit catholique de Louvain) for building the CO₂ laser.

  This work was supported by Fonds National de la Recherce Scientifique/Fondation pour la Recherche Scientifique Médicale Belgium and l'Institut National de la Santé et de la Recherche Médicale.

	FOOTNOTES

  Present address and address for reprint requests: L. Plaghki, Cliniques Universitaires St. Luc, 10 Ave. Hippocrate, B-1200 Brussels, Belgium.

  Received 23 July 1997; accepted in final form 21 January 1998.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ANDERSEN, O. K., JENSEN, L. M., BRENNUM, J., ARENDT-NIELSEN, L. Evidence for central summation of C and A nociceptive activity in man. Pain 59: 273-280, 1994.[Medline]
• BARANAUSKAS, G., NISTRI, A. Membrane potential oscillations of neonatal rat spinal motoneurons evoked by electrical stimulation of dorsal root fibres. Eur. J. Neurosci. 7: 2403-2408, 1995.[Medline]
• BECK, P. W., HANDWERKER, H. O., ZIMMERMANN, M. Nervous outflow from the cat's foot during noxious radiant heat stimulation. Brain Res. 67: 373-386, 1974.[Medline]
• BEHRENDS, T., SCHOMBURG, E. D., STEFFENS, H. Facilitatory interaction between cutaneous afferents from low threshold mechanoreceptors and noclceptors in segmental reflex pathways to alpha-motoneurons. Brain Res. 260: 131-134, 1983.[Medline]
• BEITEL, R. E., DUBNER, R. The response of unmyelinated (C) polymodal nociceptors to thermal stimuli applied to the monkey's face. J Neurophysiol. 39: 1160-1175, 1976.[Abstract/Free Full Text]
• BEITEL, R. E., DUBNER, R., HARRIS, R., SUMINO, R. Role of thermoreceptive afferents in behavioral reaction times to warming temperature shifts applied to the monkey's face. Brain Res. 138: 329-346, 1977.[Medline]
• BROMM, B., JAHNKE, M. T., TREEDE, R. D. Responses of human cutaneous afferents to CO₂ laser stimuli causing pain. Exp. Brain Res. 55: 158-166, 1984.[Medline]
• BROMM, B., TREEDE, R. D. CO₂ laser radiant heat pulses activate C nociceptors in man. Pflügers Arch. 399: 155-156, 1983.[Medline]
• BROMM, B., TREEDE, R. D. Nerve fibre discharges, cerebral potentials and sensations induced by CO₂-laser stimulation. Hum. Neurobiol. 3: 33-40, 1984.[Medline]
• BOUHASSIRA, D., CHITOUR, D., VILLANUEVA, L., LE BARS, D. The spinal transmission of nociceptive information: modulation by the caudal medulla. Neuroscience 69: 931-938, 1995.[Medline]
• BURKE, R. E., RUDOMIN, P., VYCKLICKY, L., ZAJAC, F. E. iii. Primary afferent depolarisation and flexion reflexes produced by radiant heat stimulation of the skin. J. Physiol. (Lond.) 207: 185-214, 1971.
• BURTON, H. Responses of spinal cord neurons to systematic changes in hindlimb skin temperatures in cats and primates. J. Neurophysiol. 38: 1060-1079, 1975.[Abstract/Free Full Text]
• CAMPBELL, J. M., MEYER, R. A., LA MOTTE, R. H. Sensitization of myelinated nociceptive afferents that innervate monkey hand. J. Neurophysiol. 49: 98-110, 1979.
• CATLEY, D. M., CLARKE, R. W., PASCOE, J. E. Naloxone enhancement of spinal reflexes in the rabbit. J. Physiol. (Lond.) 339: 61-73, 1983.[Abstract/Free Full Text]
• CATLEY, D. M, CLARKE, R. W., PASCOE, J. E. Post-tetanic depression of spinal reflexes in the rabbit and the possible involvement of opioid peptides. J. Physiol. (Lond.) 352: 483-493, 1984.[Abstract/Free Full Text]
• CHUNG, J. M., FANG, Z. R., CARGILL, C. L., WILLIS, W. D. Prolonged, naloxone-reversible inhibition of the flexion reflex in the cat. Pain 15: 35-53, 1983.
• CLARKE, R. W., FORD, T. W., HARRIS, J., TAYLOR, J. S. Thyrotropin releasing hormone, cholecystokinin and endogenous opioids in the modulation of spinal reflexes in the rabbit. Neuropharmacology 27: 1279-1284, 1988.[Medline]
• CLARKE, R. W., FORD, T. W., TAYLOR, J. S. Activation by high intensity peripheral nerve stimulation of adrenergic and opioidergic inhibition of a spinal reflex in the decerebrated rabbit. Brain Res. 505: 1-6, 1989.[Medline]
• CLARKE, R. W., HARRIS, J., FORD, T. W., TAYLOR, J. S. Prolonged potentiation of transmission through a withdrawal reflex pathway after noxious stimulation of the heel in the rabbit. Pain 49: 65-70, 1992.[Medline]
• CLELAND, C. L., FOONG-YEN, L., GEBHART, G. F. Pentobarbital prevents the development of C- fiber-induced hyperalgesia in the rat. Pain 57: 31-43, 1994.[Medline]
• CRIDLAND, R. A., HENRY, J. L. Facilitation of the tail-flick reflex by noxious cutaneous stimulation in the rat: antagonism by a substance P analogue. Brain Res. 462: 15-21, 1988.[Medline]
• DARIAN-SMITH, I., JOHNSON, K. O., DYKES, R. "Cold" fiber population innervating palmar and digital skin of the monkey: responses to cooling pulses. J. Neurophysiol. 36: 325-346, 1973.[Free Full Text]
• DARIAN-SMITH, I., JOHNSON, K. O., LAMOTTE, C., KENINS, P., SHIGENAGA, Y., MING, V. C. Coding of incremental changes in skin temperature by single warm fibers in the monkey. J. Neurophysiol. 42: 1316-1331, 1979.[Abstract/Free Full Text]
• DE BROUCKER, T., CESARO, P., WILLER, J. C., LE BARS, D. Diffuse noxious inhibitory controls (DNIC) in man: involvement of the spinoreticular tract. Brain 113: 1223-1234, 1990.[Abstract/Free Full Text]
• DEVOR, M., CARMON, A., FROSTIG, R. Primary afferent and spinal sensory neurones that respond to brief pulses of intense infra red laser radiation: a preliminary survey in rats. Exp. Neurol. 76: 483-494, 1982.[Medline]
• DOUGLAS, W. W., RITCHIE, J. M. The sensory functions of the non-myelinated afferent nerve fibres from the skin, In: Pain and Itch: Nervous Mechanisms, edited by G.E.W. Wolstenholme, and M. O'Connor London: Churchill, 1959, p. 26-40.
• DRAY, A. Tasting the inflammatory soup: the role of peripheral neurones. Pain Rev. 1: 153-171, 1994.
• DUBNER, R., PRICE, D. D., BEITEL, R. E., HU, J. W. Peripheral neural correlates of behavior in monkey and human related to sensory-discriminative aspects of pain. In: Pain in the Trigeminal Region, edited by D. J. Anderson, and B. Matthews Amsterdam: Elsevier, 1977, p. 57-66.
• DUBNER, R., SUMINO, R., WOOD, W. I. A peripheral "cold" fiber population responsive to innocuous and noxious thermal stimuli applied to monkey's face. J. Neurophysiol. 38: 1373-1389, 1975.[Abstract/Free Full Text]
• DUCLAUX, R., KENSHALO, D. R. Response characteristics of cutaneous warm receptors in the monkey. J. Neurophysiol. 43: 1-15, 1980.[Free Full Text]
• EBLEN-ZAJJUR, A. A., SANDKÜHLER, J. Synchronicity of nociceptive and non-nociceptive adjacent neurons in the spinal dorsal horn of the rat: stimulus-induced plasticity. Neuroscience 76: 39-54, 1997.[Medline]
• ECCLES, R. M., LUNDBERG, A. Synaptic actions in motoneurones by afferents which may evoke the flexion reflex. Arch. Ital. Biol. 97: 199-221, 1959.
• FALINOWER, S., WILLER, J. C., JUNIEN, J. L., LE BARS, D. A C-fiber reflex modulated by heterotopic noxious somatic stimuli in the rat. J. Neurophysiol. 72: 194-213, 1994.[Abstract/Free Full Text]
• FLEISHMAN, A., URCA, G. Different endogenous analgesia systems are activated by noxious stimulation of different body regions. Brain Res. 455: 49-57, 1988.[Medline]
• FRUHSTORFER, H., ZENZ, M., NOLTE, H., HENSEL, H. Dissociated loss of cold and warm sensibility during regional anaesthesia. Pflügers Arch. 349: 73-82, 1974.[Medline]
• GOZARIU, M., LE BARS, D., WILLER, J.-C. Temporal summations of C-fibre inputs: competition between facilitatory and inhibitory effects on a C-fibre reflex in the intact rat. Soc. Neurosci. Abstr. 21: 1408, 1995.
• GYBELS, J., HANDWERKER, H. O., VAN HEES, J. A comparison between the discharges of human nociceptive nerve fibres and the subject's ratings of his sensations. J. Physiol. (Lond.) 292: 193-206, 1979.[Abstract/Free Full Text]
• HAIMI-COHEN, R., COHEN, A., CARMON, A. A model for the temperature distribution in skin noxiously stimulated by a brief pulses of CO₂ laser radiation. J. Neurosci. Methods 8: 127-137, 1983.[Medline]
• HALLIN, R. G., TOREBJRK, H. E., WIESENFELD, Z. Nociceptors and warm receptors innervated by C fibres in human skin. J. Neurol. Neurosurg. Psychiatry 44: 313-319, 1981., 1981.
• HANDWERKER, H. O., IGGO, A., ZIMMERMANN, M. Segmental and supraspinal actions on dorsal horn neurons responding to noxious and non-noxious skin stimuli. Pain 1: 147-165, 1975.[Medline]
• HENSEL, H., IGGO, A. Analysis of cutaneous warm and cold fibres in primates. Pflügers Arch. 329: 1-8, 1971.[Medline]
• KANUI, T. I. Receptive field organisation and electrophysiological responses of spinal cord thermoreactive neurones in the rat. Exp. Brain Res. 71: 508-514, 1988.[Medline]
• KUMAZAWA, T., MIZUMURA, K., SATO, J. Thermally potentiated responses to algesic substances of visceral nociceptors. Pain 28: 255-264, 1987.[Medline]
• LE BARS, D., CHITOUR, D. Do convergent neurones in the spinal dorsal horn discriminate nociceptive from non-nociceptive information? Pain 17: 1-19, 1983.[Medline]
• LAMOTTE, R. H., CAMPBELL, J. N. Comparison of responses of warm and nociceptive C-fiber afferents in monkey with human judgments of thermal pain. J. Neurophysiol. 41: 509-528, 1978.[Abstract/Free Full Text]
• LEVINE, J., TAIWO, Y. Inflammatory pain. In: Textbook of Pain, edited by P. D. Wall, and R. Melzack Edinburgh: Churchill Livingstone, 1994, p. 45-56.
• LEWIS, T. In: Pain. New York: Macmillan, 1942.
• LIVESON, J. A., MA, D. M. In: Laboratory Reference for Clinical Neurpophysiology. Philadelphia, PA: F. A. Davis, 1992.
• LUNDBERG, A. Multisensory control of spinal reflex pathways. In: Progress in Brain Research. Reflex Control of Posture and Movement, edited by R. Granit, and D. Pomeiano Amsterdam: Elsevier, 1979, vol. 50, p. 11-28.
• LUNDBERG, A., MALMGREM, K., SCHOMBURG, E. D. Cutaneous facilitation of transmission in reflex pathways from Ib afferents to motoneurons. J. Physiol. (Lond.) 265: 763-780, 1977.[Abstract/Free Full Text]
• MENÉTREY, D., CHAOUCH, A., BESSON, J. M. Responses of spinal cord dorsal horn neurones to non-noxious and noxious cutaneous temperature changes in the spinal rat. Pain 6: 163-282, 1979.[Medline]
• MEYER, R. A., CAMPBELL, J. N. Evidence for two distinct classes of unmyelinated nociceptive afferents in monkey. Brain Res. 224: 149-152, 1981.[Medline]
• MEYER, R. A., CAMPBELL, J. N., RAJA, S. N. Peripheral neural mechanisms of nociception. In: Textbook of Pain, edited by P. D. Wall, and R. Melzack Edinburgh: Churchill Livingston, 1994, p. 13-44.
• NOËL, P., DESMEDT, J. E. Somatosensory cerebral evoked potentials after vascular lesions of the brainstem and diencephalon. Brain 98: 113-128, 1975.[Free Full Text]
• PLAGHKI, L., DELISLE, D., FAYT, A., GODFRAIND, J. M., STOUFFS, P. Algesimetry by brief pulses of CO₂ radiation (Abstract). Arch. Int. Physiol. Biochem. 97: 82, 1989.
• PRICE, D. D., BROWE, A. C. Responses of spinal cord neurons to graded noxious and non-noxious stimuli. Exp. Neurol. 48: 201-221, 1975.[Medline]
• ROBINSON, C. J., TOREBJRK, H. E., LA MOTTE, R. H. Psychophysical detection and pain ratings of incremental thermal stimuli: a comparison with nociceptor responses in humans. Brain Res. 274: 87-106, 1983.[Medline]
• ROBY-BRAMI, A., BUSSEL, B., WILLER, J.-C., LE BARS, D. An electrophysiological investigation into the pain relieving effects of heterotopic nociceptive stimuli: probable involvement of a supraspinal loop. Brain 110: 1497-1508, 1987.[Abstract/Free Full Text]
• ROSENBERG, S., DANZIGER, N., BOURGEOIS, P., CHARPENTIER, G., WILLER, J.-C. Inhibition of nociceptive reflexes by tens and piezo-electric currents. A neurophysiological investigation in normal human volunteers. In: Abstracts 8th World Congress on Pain. Seattle: IASP Press, 1996, p. 85.
• SCHOMBURG, E. D. Spinal sensorimotor systems and their supraspinal control. Neurosci. Res. 7: 265-340, 1990.[Medline]
• SHIN, H. K., KIM, J., CHUNG, J. N. Inhibition and excitation of the nociceptive flexion reflex by conditioning stimulation of a peripheral nerve in the cat. Exp. Neurol. 92: 335-348, 1986.[Medline]
• SINCLAIR, D. C., HINSHAW, J. R. A comparison of the sensory dissociation produced by procaine and by limb compression. Brain 73: 480-498, 1950.[Free Full Text]
• SIVILOTTI, L. G., THOMPSON, S.W.N., WOOLF, C. J. Rate of rise of the cumulative depolarization evoked by repetitive stimulation of small-caliber afferents is a predictor of action potential windup in rat spinal neurons in vitro. J. Neurophysiol. 69: 1621-1631, 1993.[Abstract/Free Full Text]
• SNEDECOR, G. W., COCHRAN, W. G. In: Statistical Methods. Iowa: Iowa State Univ. Press, 1989.
• SPRAY, D. C. Cutaneous temperature receptors. Annu. Rev. Physiol. 48: 625-638, 1986.[Medline], 1986.
• STEFFENS, H., SCHOMBURG, E. D. Convergence in segmental reflex pathways from nociceptive and non-nociceptive afferents to alpha-motoneurones in the cat. J. Physiol. (Lond.) 466: 191-211, 1993.[Abstract/Free Full Text]
• SUMINO, R., DUBNER, R. Response characteristics of specific thermoreceptive afferents innervating monkey facial skin and their relationship to human thermal sensitivity. Brain Res. Rev. 3: 105-122, 1981.
• SUMINO, R., DUBNER, R., STARKMAN, S. Responses of small myelinated "warm" fibers to noxious heat stimuli applied to the monkey's face. Brain Res. 62: 260-263, 1973.[Medline]
• TAYLOR, J. S., NEAL, R. I., CLARKE, R. W., FORD, T. W. Supression of a spinal reflex by intense stimulation of near and remote peripheral nerves in the rabbit. Proc. Int. Union Physiol. Sci. 17: 94, 1989.
• TAYLOR, J. S., NEAL, R. I., HARRIS, J., FORD, T. W., CLARKE, R. W. Prolonged inhibition of a spinal reflex after intense stimulation of distant peripheral nerves in the decerebrated rabbit. J. Physiol. (Lond.) 437: 71-83, 1991.[Abstract/Free Full Text]
• TAYLOR, J. S., PETTIT, J. S., HARRIS, J., FORD, T. W., CLARKE, R. W. Noxious stimulation of the toes evokes long-lasting, naloxone-reversible suppression of the sural-gastrocnemius reflex in the rabbit. Brain Res. 531: 263-268, 1990.[Medline]
• TERASHIMA, S. AND LIANG, Y.-F. Modality difference in the physiological properties and morphological characteristics of the trigeminal sensory neurons. Jpn. J. Physiol. 43, Suppl. 1: S267-S274, 1993.
• THOMPSON, S.W.N., WOOLF, C. J., SILVIOTTI, L. G. Small-caliber afferent inputs produce a heterosynaptic facilitation of the synaptic responses evoked by primary afferent A-fibers in the neonatal rat spinal cord in vitro. J. Neurophysiol. 69: 2116-2128, 1993.[Abstract/Free Full Text]
• VALLBO, A. B., HAGBARTH, K. E., TOREBJRK, H. E., WALLIN, B. G. Somatosensory, proprioceptive, and sympathetic activity in human peripheral nerves. Physiol. Rev. 59: 919-957, 1979.[Free Full Text]
• VAN HEES, J., GYBELS, J. Pain related to single afferent C fibers from human skin. Brain Res. 48: 397-400, 1972.[Medline]
• VEILLEUX, M., STEVENS, J. C. Syringomyelia: electrophysiological aspects. Muscle Nerve 10: 449-458, 1987.[Medline]
• VILLANUEVA, L., CHITOUR, D., LE BARS, D. Involvement of the dorsolateral funiculus in the descending spinal projections responsible for diffuse noxious inhibitory controls in the rat. J. Neurophysiol. 56: 1185-1195, 1986.[Abstract/Free Full Text]
• WALL, P. D., WOOLF, C. J. Muscle but not cutaneous C-afferent input produces prolonged increases in the excitability of the flexion reflex in the rat. J. Physiol. (Lond.) 356: 443-458, 1984.[Abstract/Free Full Text]
• WIESENFELD-HALLIN, Z. Intrathecal somatostatin modulates spinal sensory and reflex mechanisms: behavioral and electrophysiological studies in the rat. Neurosci. Lett. 62: 69-74, 1985.[Medline]
• WIESENFELD-HALLIN, Z., XU, X.-J., HAKANSON, R., FENG, D.-M., FOLKERS, K. The specific antagonistic effect of intrathecal spantide II on substance P- and C-fiber conditioning stimulation-induced facilitation of the nociceptive flexor reflex in the rat. Brain Res. 526: 234-290, 1990.
• WIESENFELD-HALLIN, Z., XU, X.-J., HAKANSON, R., FENG, D.-M., FOLKERS, K. Tachykinins mediate changes in spinal reflexes after activation of unmyelinated muscle afferents in the rat. Acta Physiol. Scan. 141: 57-61, 1991.[Medline]
• WILLER, J.-C. Comparative study of perceived pain and nociceptive flexion reflex in man. Pain 3: 69-80, 1977.[Medline]
• WILLER, J.-C. Nociceptive flexion reflex as a physiological correlate of pain sensation in humans. In: Pain Measurement in Man. Neurophysiological Correlates of Pain, edited by and B. Bromm Amsterdam: Elsevier, 1984, p. 87-110.
• WILLER, J.-C., ALBE-FESSARD, D. Further studies on the role of afferent input from relatively large diameter fibers in transmission of nociceptive messages in humans. Brain Res. 278: 318-321, 1983.[Medline]
• WILLER, J.-C., BUSSEL, B. Evidence for a direct spinal mechanism in morphine-induced inhibition of nociceptive reflexes in humans. Brain Res. 187: 212-215, 1980.[Medline]
• WILLER, J.-C., DE BROUCKER, T., LE BARS, D. Encoding of nociceptive thermal stimuli by diffuse noxious inhibitory controls in humans. J. Neurophysiol. 62: 1028-1038, 1989.[Abstract/Free Full Text]
• WILLER, J.-C., LE BARS, D. Electrophysiological studies on morphine analgesia in man. In: Advances in Pain Research and Therapy, edited by B. Bromm, and J. E. Desmedt New York: Raven Press, 1995, vol. 22, p. 541-557.
• WILLER, J.-C., ROBY, A., LE BARS, D. Psychophysical and electrophysiological approaches to the pain-relieving effects of heterotopic nociceptive stimuli. Brain 107: 1095-1112, 1984.[Abstract/Free Full Text]
• WOOLF, C. J. Long term alterations in the excitability of the flexion reflex produced by peripheral tissue injury in the chronic decerebrate rat. Pain 18: 325-343, 1984.[Medline]
• WOOLF, C. J., WALL, P. D. Relative effectiveness of C primary afferent fibres of different origins in evoking a prolonged facilitation of the flexor reflex in the rat. J. Neurosci. 6: 1433-1442, 1986.[Abstract]
• WOOLF, C. J., WIESENFELD-HALLIN, Z. Substance P and calcitonin gene-related peptide synergistically modulate the gain of the nociceptive flexor withdrawal reflex in the rat. Neurosci Lett. 66: 226-230, 1986.[Medline]
• YASHPAL, K., RADHAKRISHNAN, V., HENRY, J. L. NMDA receptor antagonist blocks the facilitation of the tail flick reflex in the rat induced by intrathecal administration of substance P and by noxious cutaneous stimulation. Neurosci. Lett. 128: 269-272, 1991.[Medline]
• YASHPAL, K., RADHAKRISHNAN, V., CODERRE, T. J., HENRY, J. L. CP-96,345, but not its stereoisomer, CP-96,344, blocks the nociceptive responses to intrathecally administered substance P and to noxious thermal and chemical stimuli in the rat. Neuroscience 52: 1039-1047, 1993.[Medline]

0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society