Evidence for central summation of C and Aδ nociceptive activity in man

273

Pain, 59 (1994) 273-280 0 1994 Elsevier Science B.V. All rights reserved 03043959/94/%07.00

PAIN 2596

Evidence for central summation of C and A6 nociceptive activity in man Ole Kzseler Andersen

a**,Lars Mohr Jensen a, Jannick Brennum b, Lars Arendt-Nielsen

a

’ Department

of Medicat Informatics, Laboratory for Experimental Pain Research, Aaiborg lJni~*ersity,DK-9220 Aalborg (Denmark) and b Department of Neurology, Laboratory for Sensory Physiology, Gentofte University Hospital, DK-2900 Hellerup (Denmark)

(Received 1.5November 1993; revision received 18 February 1994, accepted 15 March 1994)

Using two different stimulators, we have induced activity in A6 and C afferents in order to Summary investigate a possible summation of nociceptive activity from these two fiber types. We used nociceptive electrical st~ulation to evoke activity in AS fibers. High-intensi~ light from a xenon lamp, focused into a liquid light guide which was positioned on a spot painted black under the sole of the foot, resulted in a characteristic deiayed burning sensation, indicating selective C-fiber activation. By varying the delay between radiant heat and eIectrica1 stimuli (o-3000 msec), sensations evoked by these stimuli were brought to coincide. When we eIicited the electrical stimulation during on-going burning pain, corresponding to a delay of approximately 1 set between the stimulations, we found a significantly higher nociceptive withdrawal reflex in tibials anterior (P < 0.01) and a higher overall pain rating (P < 0.05). The existence of a summation mechanism at the spinal cord is the most likely explanation for our findings. Furthermore, the results demonstrate that the nociceptive reflex may be modulated by on-going C-fiber activity. Key words: Refiex; Summation; C fiber; A6 fiber; Nociception

Introduction The central integrative mechanism in the nociceptive system is assumed to be of importance for various pain syndromes. Integration of afferent activity can be both over time, temporal summation, and area, spatial summation. Spatial summation has been found for thermally induced pain (Price 1989; Douglass et al, 1992). The mechanism of convergence and spatial su~ation of afferent nociceptive activity probably involves wide dynamic range (WDR) neurons (Dubner 19911, also called multireceptive neurons, located in the dorsal horn. Temporal summation of afferent activity is more pronounced for activity in C-fibers than for AS-fiber activity (Price et al. 1977, 1978). Temporal summation is also thought to involve WDR neurons (Price et al.

* Comqxmding

author: Ole K. Andersen, Dept. of Medical Informatics, Laboratory for ~~~rnental Pain Research, Aalborg University, Frederik Bajers Vej 7D, DIG9220 Aalborg, Denmark. FAX: (45) 9815-4008.

SSDI 0304-3959(94)00072-M

19771, as noxious activity causes a proionged discharge from these neurons (Schouenborg and Dickenson 1988; Dubner 1991). Thus, repeated stimulation of C fibers at a frequency of at least l/3 Hz results in a progressive increase (wind-up) in the activity of WDR neurons (Price et al. 1978) and a psychophysical correlation has been found for similar frequencies (Price et al. 1977). Studies of summation are interesting as summation may be the first step in the creation of the central sensit~ation that follows prolonged C-fiber activity (Woolf and Thompson 1991). As most of the WDR neurons receive input from both AS and C fibers (Price 1988; Schouenborg and Dickenson 19&Q, it would be of interest to investigate the potency of summation of activity in these two fiber types. The nociceptive withdrawal reflex from the hind limb muscles in the cat (Price 1972) and in man (Arendt-Nielsen et al. 1993) has been shown to reflect central temporal integration of noxious afferent activity. Moreover, the amplitude (Shahani and Young 1971; Willer 19’77) and the duration (Shahani and Young 1971) of the nociceptive reflex have been shown to correlate with the intensity of the pain stimulus. Sha-

hani and Young (1971) also suggest the human nociceptive withdrawal reflex to be a quantitative measure of central integration in man. In post-surgical patients, Dahl et al. (1992) have found a decreased reflex threshold and increased reflex response, which they ascribed to tissue damage and the following central sensitization. A possible summation of A& and C-fiber activity could be an alternative explanation for the findings if C-fiber activity continue after the tissue damage. Investigations of a possible central integration of activity in A6 and C fibers require activation of nociceptors of both fiber types. Activation of nociceptors innervated by C fibers can be achieved by strong heat stimulation resulting in a delayed burning pain after the stimulation while A6 activity can be evoked by intense electrical stimulations. It remains, however, a problem to activate the two fiber populations selectively. The aim of the present study was to investigate a possible summation of nociceptive activity in A6 and C fibers evoked by two different stimulators. As the conduction velocity for the two fiber populations differs, it is necessary to make a delay between onset of the two stimuli to obtain synchronous arrival of activity in the two fiber populations at the level of the spinal cord.

Methods Subjects Twelve male volunteers (mean age: 24 years; range: 21-34 years) participated in the experiment. Informed consent was obtained from all subjects, and the Helsinki Declaration was respected. No medication had been used in the last 24 h prior to the test. The subjects were seated in a test chair. The thigh rested on the seat and the lower leg was free to swing so that stimufation of the sole of the foot was possible (Fig. 1).

Stimulation In the last decades, the sural nerve has been the usual stimulation site for eliciting withdrawa reflexes in the teg (Hugon 1973; Wilier 1977; Wall and Woolf 1984; DeBrouker et al. 1989). Natural painful stimuli seldom hit the lateral dorsal side of the foot but more likely the sole of the foot, e.g., during walking. Therefore, in this experiment, nociceptive electrical and light stimuli were delivered to the median arch of the foot. The electrical stimufi were delivered percutaneously by a constant-current rectangular pulse train, consisting of 5 pulses, each of 1 msec duration and delivered at 200 Hz. The stimulus intensity was 1.5 times the individual reflex threshold which was defined as the lowest stimulus intensity by which 3 successive reflexes could be evoked. A reflex was defined as EMG activity above 20 L;V for a period of at least 5 msec within a time window from 70 to 200 msec after stimulation. This reflex is normally denoted the RI11 reflex (Hugon 1973; Wifler 19771.The stimulation intensity of 1.5 times the reflex threshold was above the subjects’ pain threshold and ensured a stable nociceptive reflex in all subjects throughout the experiment. Secondly, heat stimulations were delivered by high-energy light pulses from a 1000 W xenon lamp (Cermax Xenon Illuminator, ILC Technology) focused into a liquid light guide (diameter: 3 mm) positioned

Fig. 1. Experimental set-up. The subjects were seated in a test chair and were asked to relax. EMG was recorded from the anterior tibia1 muscle. Stimulations were delivered to the plantar part of the foot from a constant current stimulator and/or a high-intensity xenon lamp. which was focused into a light guide positioned next to the electrode.

under the foot. The stimulation position was painted black to achieve maximum absorption of the light. With a fixed duration of 200 msec, the intensity of the Iight was adjusted until a characteristically painful burning sensation was evoked approximately 1 set after the onset of light stimulation. The end of the light guide was positioned next to the stimulation electrode. The interval between two stimulations was at least 8 sec. The light stimuli did not evoke a first pain component (i.e., pricking pain) (Price et al. 1977, 1978; Arendt-Nielsen 1990a) but only a second pain component (i.e., longer latency with burning or swelling quality) (Price et al. 1977, 1978). Usually, this was achieved at a position with thick epidermis. We marked an area of approximately 2 X 2 cm where this burning sensation could be evoked, making variations in the stimulation position possible in order to avoid sensitization/fatigue of the nociceptors. Electrical and light stimuli were applied either alone or in combination. In the latter configuration, the electrical stimulus was delayed with respect to the light onset. The delays were chosen as either 0,300, 600, 800, lOBI, 1200 or 3OfKlmsec. Stimulation series (5 stimulations) at each delay were given in a random sequence.

Electromyographic recordings The electromyogram (EMG) was recorded from the ipsilateral m. tibialis anterior (TA) using Ag-AgCl surface electrodes (diameter: 7 mm) placed parallel to the muscle fiber direction with an inter-electrode distance of appro~mateiy 30 mm. Prior to the electrode attachment, the skin was lightly abraded and cleaned with isopropyl alcohol. The EMG signals were filtered (20-500 Hz; 4. order), amplified (lO,OOO-50,000 times), sampled (2 kHz), monitored on a computer storage oscilloscope, and stored on disk for later analysis. EMG signals for 4 set were stored from 200 msec prior to light stimulation.

Pain sensation and reaction time The subjects (n = 10) were asked to rate the ouerull pain sensation after the actual stimulation configuration. The pain sensation was recorded using a visual analog scale (VAS) of light-emitting

275 diodes, covering a continuous scale from 0 to 10. Reaction time to the light stimuli was also measured, The subjects were isolated when recording the reaction time so that they obtained no visual or auditory cues of the stimulation. They were then asked to press a button when they felt any sensation following the light stimuli. An increased EMG response to an electrical nociceptive stimulation following C-fiber activation
Data analysis EMG recordings were rectified and averaged (5 stimulations). The onset latency and duration of the reflexes following the stimulations were measured from the averaged reflex by visual inspection. The criterion for determination of the onset and end of the reflex was presence of activity with an amplitude above the background recording noise (there was no background EMG activity). The rootmean-square amplitude (RMS) was measured in the SO-200 msec interval after electrical stimulation (or the light stimuli when these were given alone). Finally, the integrated energy (iE) of the reflex response was calculated in the window 50-300 msec. As a consequence of the very low background activity in the TA during relaxed rest, we saw no early components, so the iE solely represents the nociceptive reflex (RI11 reflex) (see Fig. 3). Pain sensations were tested statistically in the same way as the reflex parameters (against the sensations at no-delay and at 3000 msec delay). Light stimuli reaction times were averaged between the subjects, using 3 values for each subject. Motor neuron pool excitability is expressed as the H/M,,, relation, i.e., the relation between the peak-to-peak amplitude of the maximum recorded M-response and the actual H-reflex. The relation is an average of 5 measurements at the same delay.

Wilcoxon’s non-parametric rank sum test for paired data was for statistical analysis. A 2-tailed significance level of 5% was considered statistically significicant. used

0’

I Iys

I

I

I

I

el

0

300

600

I

I

I

,

800 1000 12003000

Delay fms) Fig. 2. Median RMS calcufations (n = 12) for the different conditions calculated in the 50-200 msec interval after the electrical stimulation. *Significant difference between this observation and the 0 msec delay condition. ‘Significant difference between this observation and 3000 msec delay condition.

msec (Table I) compared with the reflex duration when stimulating simultaneously (600 msec: P < 0.01; 800 msec: P < 0.03; 1000 msec: P < 0.01; 1200 msec: P < 0.01). The reflex duration of the 3000 msec condition decreased compared with the recording for the 1000 msec condition (3000 vs. 1000 msec: P < 0.03, TabIe I). The RMS value of the EMG activity in the 50-200 msec interval varied with the delay between the electrical and light stimuli as illustrated in Fig 2. The highest RMS values were obtained when the electricai stimulus was delayed 8~-12~ msec. Stimulation at these delays evoked significantly higher reflexes than reflexes evoked by simultaneous light and electrical stimuli (800 msec: P < 0.02; 1000 msec: P < 0.01; 1200 msec: P < 0.03). Furthermore, the reflex, evoked with a delay of

TABLE I DURATION AND PAIN RATINGS (median values) P values listed for significant results. p(O) means a paired test with the reflex activity at no delay and p(3000) when testing against the 3000 msec delay.

Results For the light stimuli there were no activity in the 50-200 msec interval after stimulation (the RMS value different from zero in Fig. 2 represents noise only). The onset latencies of the reflex evoked by electrical stimuli showed no significant variations for the different conditions having a global mean value of 78.4 f 10 msec (range: 65-130 msec). The duration of the electrically evoked reflex showed a significant increase when the electrical stimulation was delayed 600-1200

Delay Duration msec msec light el. 0 300 600 800 1000 1200 3000

55 55 58 75 78 95 78 60

p(O) duration

P P P P

< < < <

p(3000) duration

Pain ratings

1.5 2.9 2.9 3.3 0.01 3.5 0.03 3.4 0.01 P < 0.03 3.5 0.01 3.4 3

P(O) painrat.

P(3000) painrat.

P < 0.04 P P P < 0.04 P P < 0.04 P

< 0.01 < 0.03 < 0.01 < 0.01

3000msec had a lower RMS amplitude than the reflex evoked with a delay of 800- 1200 msec (P < 0.02).

I” < 0.02; 1000 msec: Y < 0.01; 1200 msec: Y < t1.05). Moreover, the iE value for the 3000 msec delay shows a significant decrease in energy compared with the 1000 msec recording (P < 0.05). Fig. 4 illustrates averaged curves (n = 12) for the iE calculations. The derivative of the iE represents the instantaneous power and, as Fig. 4 illustrates, the 1000 msec recordings show a higher power in the initial phase of the reflex which corresponds with the difference in the RMS calculations. The duration of the reflexes can be identified as the part of the curves where the energy increases, but temporal aspects of inter-person averages must be carefully considered as there are differences in the reflex onset. The pain ratings were highest in the 600-1200 msec

As the duration of the reflex varied, it is not ideal to calculate the RMS in a constant time window. Instead, an integrated parameter may be used with a time window that encapsuIates the reflex which is only possibIe when no background activity is present. As this is the case for the TA during relaxed conditions, we have accumulated the energy (iE) in the interval from SO to 300 msec and presented the value continuously as a function of time (see Fig. 3). The iE caIculations indicate the same facilitation as shown by the RMS calculations, i.e., the final values of the integrated energies (see Fig. 3) are significantly elevated in the intervals 800-1200 msec (800 msec:

A !ti-

c

ii

i'

I

.

-----100

150

Time

zoo

250

300

[ms]

Fig. 3. Averaged (n = 5) and rectified reflex recordings for 1 subject evoked at 3 different inter-stimulus delays. A: 0 msec; B: 1000 msec; C: 3000 msec. Integrated energy (iE) curves are also shown relative to the right scale.

217

(Torebj6rk and Ochoa 1980) arose about 1 set after onset of the light (mean reaction time: 961+ 413 msec). Furthermore, the most painful sensation was reported when the electrical stimulation was delivered during maximum burning pain, which was 800 and 1200 msec after light stimulation. Differences in the quality of the pain perception evoked by the two stimulus types might explain the relatively small differences in pain ratings found between the different conditions (Table I). Thus, most of the subjects were able to discriminate pain sensations following electrical and light stimulatiogs. However, when the sensations arose simultaneously, discrimination was difficult and the pain sensation was maximum. Several authors (Hugon 1973; Willer 1977; Chan and Tsang 1985) have reported on attempts to separate the EMG activity evoked by electrical stimulations into an early component evoked by activity in thick tactile fibers (RI1 reflex) having a latency between 40-60 msec and a later component evoked by AS fibers (the RI11 reflex) with latencies higher than 70 msec. The early component is not present when stimulating the skin innervated by the sural nerve (Willer 1977) only when stimulating the nerve directly. In this study, we also used cutaneous stimulations instead of direct nerve stimulations although it was not the receptive field of the sural nerve. Chan and Tsang (1985) only found RII-reflex components in 2 of 10 subjects when stimulating the plantar part of the foot (as in this study), but these RI1 components had onset latencies shorter than the latencies found in the present study. The EMG activity evoked by the electrical stimulations in this study is therefore likely a result of activity in A6 fibers (RI11 reflexes). The duration of the reflex varied with the delays

TABLE II RECORDINGS OF THE H-REFLEX (expressed as a percent of the maximum M-response for 3 different subjects) Delay msec

Subject 1 (% I$,,,)

Subject 2 (% M,,,)

Subject 3 (% M,,)

0 300 600 800 1000 1200 3000

24.6i6.3 26.2k3.8 32.0 + 7.0 29.3 f 7.6 25.7* 5.5 25.7 k 3.2 29.Ok6.0

35.0+ 9.0 29.2k 6.2 31.1+10 28.1 f 10 27.8+ 1.9 25.8+ 6.3 29.9f 7.4

23.5 f 2.0 23.1k2.0 26.3 f 2.2 25.1 f 2.2 24.6k 2.2 25.8*2.1 26.0+2:1

interval and significant differences were found (see Table I). The mean reaction time to light stimuli was 9612 413 msec. Table II shows the averaged H/M,, relation, and the variations between the different conditions for all 3 subjects show no tendency towards a facilitation of the motor neuron pool by the light stimulations.

Discussion The present study shows a significant increase in the nociceptive reflex size (duration and amplitude) at an inter-stimulus interval of approximately 1 set between heat and electrical stimulations. This result primarily indicates integration of afferent activity from two different nerve fiber populations. The afferent barrageevoked light stimulation is likely mediated by slow-conducting fibers, C fibers. Psychophysically, the pain perception reported by the subjects supports a main C-fiber as a burning and painful sensation activation,

10.

iE (nV’)

electrical

10.

iE (nV’)

0 Ins

10.

iE (nV’)

300

ms

10.

6.

6.

6.

6.

6.

4.

4.

4.

4.

2 0 50

10

iE (nV’)

600

l-no

a.

6.

6.

A’

fl 100

iE (nV’)

150

200

a00 ma

61

200

10

iE (nV’)

6

100 lima

(ma)

1000 I

ms

,-

150 Time

10

iE (nV’)

1200

ms

a

200 (me)

10

iE (d)

3000

250

300

250

300

Ins

aI

lime

(ms)

The

200 (mm)

Fig. 4. Integrated energy (iE) averaged across all subjects (n = 12) for all 8 conditions. The bottom line represents the energy and the top line the energy + SD. Note the differences in slope and delay from onset of the reflex until1 the asymptotic levels are reached, although differences in onset latency for the subjects must be considered. The large standard deviations, especially at loo0 msec delay, are partly due to the analysis method as the energy is a very sensitive parameter for inter-person variations.

278

between light and electrical stimuli. The longest reflex duration was found for the condition with 1000 msec delay. The recordings for this delay also had the highest iE. For the iE (Fig. 41, we found a difference in the final energy level between the recordings for electrical stimulations alone and for simultaneous electrical and light stimulations (0 msec delay). This might be an effect of a visual warning coming from the light instrument, as a flash appeared when stimulating. A warning is known to cause smaller responses with less variability in nociceptive-evoked potentials (Cruccu al. 1983; Svensson et al. 1992) but, so far, this effect has not been found in nociceptive reflexes. The interval in which an increase in the reflex size is found covers 600-3000 msec and might be a result of the large difference in the conduction velocity within the C-fiber population (0.5-1.5 m/set measured by Hallin and Torebjork 19731, resulting in a broad interval in which summation takes place. The reflex maximum might then correspond to the time when the maximum number of action potentials via C fibers reach the dorsal horn. The amplitude and duration of the EMG findings therefore support the suggested summation of the afferent activity as both parameters have been shown to increase at increasing stimulus intensities (Shahani and Young 1971). The central mechanism behind summation of afferent activity in different fiber types is unclear. Summation has primarily been shown for C-fiber activity (Price et al. 1977). This mechanism applies both to stimuli which activate the nociceptors over a time period ‘temporal summation’ (Price et al. 19781 and to activation of nociceptors distributed over a large area, ‘spatial summation’ (Price et al. 1989; Douglass et al. 1992). A recent study of spinal neurons in vitro has shown a wind-up effect following electrical stimulation of the dorsal root at A6 intensity, although wind-up following stimulation at C-fiber intensity was more pronounced (Sivilotti et al. 1993). Furthermore, activity in WDR neurons wind-up following repetitive stimulation at C-fiber intensity, while nociceptive-specific neurons (NS) are unaffected by repetitive stimulations at the same intensity (Schouenborg and Sjolund 19831. As activity in unmyelinated afferents evokes prolonged discharges in WDR neurons (Dubner 1991) and most of these neurons receive input from A6 and C fibers (Willis 1985; Price 1988>, it seems likely that these neurons are capable of performing this integration. Local integration of heat pain within the receptive field of a primary nociceptive neuron and spatial integration of output from various primary neurons within the dorsal horn may involve different types of neurons (Price et al. 1989). Electrophysiological and psychophysical findings also suggest that WDR neurons are capable of encoding both intensity and location of a painful stimulus (Coghill et al. 19931, while NS neu-

rons are involved in mechanical pain or phasic responses to thermal pain. Activity from WDR neurons alone is sufficient to cause pain (Coghill et al. 1993). As WDR neurons have a large receptive field compared with NS neurons (Price 1988; Dubner 19911, it is likely that they are involved in the integration, as a degree of spatial summation is needed because of the distance between the electrode and the end of the light guide. Diffuse noxious heat stimuli have previously been shown to depress the spinal nociceptive reflex (DNIC, diffuse noxious inhibitory control) (Willer et al. 1989). In DNIC, the noxious stimuli are applied to a heterotopic area (the contralateral hand when stimulating the sural nerve) so it is probably another central mechanism that is responsible for the depressive effect on the RI11 reflex and pain sensation than the mechanism responsible for our findings. Further, the duration of the noxious stimuli when the hand is immersed into a water bath is much longer than the transient heat pulse used in this study whereby other central plastic changes such as release of inhibitory mediators are more likely. Thus, the effect of DNIC outlasts the conditioning heat stimuli (Willer et al. 1989). Repeated heat stimulations to the same skin area are known to cause receptor sensitization of the heatsensitive nociceptors and the interval between consecutive stimulations influences the degree of sensitization (Bessou and Pearl 1969; see also Price et al. 1989). Based on reports from the subjects, we considered it likely that the high-energy heat stimulations used in this experiment have caused some degree of sensitization, despite the pause between consecutive stimulations. But as the stimulation position of the light and delay between light and electrical stimulations were varied randomly, the data were not biased. As previously mentioned, the pain perception reported by the subjects, the peak in reflex size, and the reaction time all indicate activation of C fibers by xenon lamp stimulations of the sole of the foot. Activation of C fibers for a short period of time (electrical stimulation at 1 Hz for 20 set> results in a prolonged facilitation of the reflex (Cook et al. 1986; Woolf and Wall 1986). This facilitation has been interpreted as a hyperalgesic state based on both central and peripheral alterations (Woolf 1983) in order to provide a better protection of any potential tissue damage (Clarke et al. 1992). Experiments have shown that hyperexcitability is neither a result of on-going activity in thin afferents from the inflamed area (Woolf 1983; Wall and Woolf 1984; Woolf and Wall 19861 nor of other peripheral receptor mechanisms (Dubner et al. 1989). The state of the primary neurons in the spinal cord and the motor pool excitability (Cook et al. 1986) reveal no sufficient explanation for the elevated excitability. Thus, the increased excitability has been regarded as a result of

279

changes in the inter-neurons, conveying information from the dorsal horn to the motor neurons. Release of neuropeptides, known to facilitate the depolarization through NMDA receptors by excitatory amino acids, is one theory behind the increased excitability (Dubner 1991; Woolf and Thompson 1991). The sensitization observed in this experiment could be an incipient experimentally induced h~erexcitabili~. However, no explanation for the peak in the reflex size at 1000 msec can be found in a sensitization that lasts several minutes. Considering the effect of C-fiber activity on the reflex shown by the present results, it is difficult to determine whether the increased reflex responses are a result of central hyperexcitability following tissue injury (Woolf and McMahon 1985; Hartell and Headley 1991; Dahl et al. 1992), or central summation of on-going C-fiber-mediated activity from the area of injury. Thus, Grdnross and Pertovaara (1993) have shown that ongoing activity from the periphe~ is necessary to maintain faciiitation of the nociceptive reflex in experimentally induced hyperaIgesia using capsaicin. The results of our control study bear no evidence of a change in the motor neuron pool excitability during the same time window as when the nociceptive reflexes were measured. However, Cook et al. (1986) found a major increase in the H-reflex following application of mustard oil, a selective C-fiber chemical irritant that lasted up to 10 min. An obvious difference in the temporal extent of the C-fiber volley between our study and that of Cook et al. (1986) could be one reason for the different findings. Thus, the results of the control study reveal no postsynaptic mechanism which could explain our reflex results. In the perspective of these alternative hypotheses, summation, presumably in second-order neurons, constitutes the most likely basis for our results since both ascending signals and motor neurons are affected. So far, selective activation of C fibers has been difficult. Chemical irritants (e.g., capsaicin) are a possibility but the intensity and duration of the pain is difficult to control. Use of the xenon lamp, as described in the present study, may therefore be usable in future studies. When carrying out electrophysiological studies of the nociceptive reflexes, electrical stimulation of the median arch of the foot is not used very often (Chan and Tsang 1985) as sural nerve stimulation has become the most commonly used technique in recent years. Yet, the original studies by Sherrington (1910) and Kugleberg et al. (1960) used cutaneous stimulations, e.g., the sole of the foot, presumably because stimulation of these areas represents more natural conditions and evokes a pure flexor reflex of the ankle, knee and hip, which is not normally seen when stimulating the sura1 nerve. Another advantage of cutaneous stimula-

tions compared to nerve bundle stimulations is a minimum of early tactile components mediated by A/3 fibers in the reflex (Willer 1977).

References Arendt”NieIsen, L., First pain event related potentials to argon laser stimuli: recording and quanti~cation, J. Neurot., Neurosurg. Psychiat., 53 (199Oa) 398-404. Arendt-Nielsen, L., Brennum, J., Sindrup, S., and Bak, P., Electrophysiological and psychophysical quantification of central temporal summation of the human nociceptive system, Eur. J. Appl. Physiol., in press. Bessou, P., and Perl, E.R., Response of cutaneous sensory units with unmyelinated fibers to noxious stimuli, J. Neurophysiol., 32 (1969) 1025-1043. Chan, C.W.Y., and Tsang, H.H.Y., A quantitative study of flexion reflex in man relevance to pain research. In: H.L. Fields et al. (Eds.), Advances in Pain Research and Therapy, Vol. 9, Raven Press, New York, 1985, pp. 361-370. Clarke, R.W., Harris, J., Ford, T.W., and Taylor, J.S., Prolonged potentiation of transmission through a withdrawal reflex pathway after noxious stimulation of the heel in the rabbit, Pain, 49 (1992) 65-70. Coghill, RX., Mayer, D.J., and Price, D., Wide dynamic range but not nociceptive-specific neurons encode multidimensional features of prolonged repetitive heat pain, J. Neurophysiol., 69 (1993) 703-716. Cook, A.J., Woolf, C.J., and Wall, P.D., Prolonged C-fibre mediated facilitation of the flexion reflex in the rat is not due to changes in afferent terminal or motorneurone excitability, Neurosci. Lett., 70 (1986) 91-96. Cruccu, C., Fornaretli, M., Inghillieri, M., and Manfredi, M., The limits of tooth pulp evoked potentials for pain quantitation, Physiol. Behav., 31 (1983) 339-342. Dahl, J.B., Erichsen, C.J., Fuglsang-Frederiksen, A., and Kehlet, H., Pain sensation and nociceptive reflex excitability in surgical patients and human volunteers, Br. J. Anaesth., 69 (1992) 117-121. DeBrouker, T., Wilier, J.C., and Bergeret S., The nociceptive reflex in humans: a specific and objective correlate of experimental pain. In: CR. Chapman, and J.D. Loeser (Eds.), Issues in Pain Measurement, Raven Press, New York, 1989, pp. 337-352. Dickenson, A.H., and Sullivan, A.F., NMDA receptors and central hyperalgesic states, Pain, 46 (1991) 344-345. Douglass, D.K., Carstens, E., and Watkins, L.R., Spatial summation in human thermal pain perception: comparison within and between dermatomes, Pain, 50 (1992) 197-202. Dubner, R., and Bennet, G.J., Spinal and trigeminal mechanisms of nociception, Annu. Rev. Neurosci., 6 (1983) 381-418. Dubner, R., Hylden, J.K.L., Nahin, R.L., and Traub, R.J., Neuronal plasticity in the superficial dorsal horn following peripheral tissue inflammation and nerve injury. In: F. Cervero, G.J. Bennet and P.M. Headley (Eds.), Processing of Sensory Information in the Superficial Dorsal Horn of the Spinal Cord, Plenum Press, New York, 1989, pp. 429.442. Dubner, R., Neuronal plasticity in the spinal and medullary dorsal horns: a possible role in central pain mechanisms. In: K.L. Casey (Ed.), Pain and Central Nervous Systems Disease: The Central Pain Syndromes, Raven Press, New York, 1991, pp. 143-155. GrGnross, M., and Pertovaara, A., Capsaicin-induced central facilitation of a nociceptive flexion reflex in humans, Neurosci, Lett., 159 (1993) 215-218. Hallin, R.G., and TorebjSrk, H.E., Electrically induced A and C

280 fibre responses in intact human skin nerves, Exp. Brain Res.. 16 (1973) 309-320. Hartell. N.A., and Headley. P.M., Preparative surgery enhances the direct spinal actions of three injectable anaesthetics in the anaesthetized rat, Pain, 46 (1991) 75-80. Hugon, M., Exteroceptive reflexes to stimulation of the sural nerve in human man. In: J.E. Desmedt (Ed.), New Developments in Electromyography and Clinical Neurophysiology. Vol. 3, Karger, Base]. 1973, pp. 713-729. Kugleberg, E., Eklund, K., and Lennart, G., An electromyographic study of the nociceptive reflex of the lower limb mechanism of the planter responses. Brain, 83 (1960) 394-410. Price, D.D., Characteristics of second pain and flexion reflexes indicative of prolonged central summation, Exp. Neural., 37 (1972) 371-387. Price, D.D., Psychological and Neural Mechanisms of Pain. Raven, New York. 1988. Price, D.D., Hu. J.W., Dubner, R., and Gracely, R.. Peripheral suppression of first pain and central summation of second pain evoked by noxious heat pulses, Pain, 3 (1977) 57-68. Price, D.D., Hayes, R., Ruda, M., and Dubner, R., Spatial and temporal transformation of input to spinothalamic tract neurons and their relations to somatic sensations, J. Neurophysiol., 41 (1978) 933-94s. Price, D.D., Mchaffie. J.G., Larson, M.A.. Spatial summation of heat-induced pain: influences of stimulus area and spatial separation of stimuli on perceived pain sensation, intensity and unpleasantness. J. Neurophysiol., 62 (1989) 1270-1279. Schouenborg, J., and Dickenson, A., Activity evoked by A- and C-afferent fibers in rat doral horn neurons and its relation to a flexion reflex, J. Neurophysiol.. 50 (1983) 1108-1121. Schouenhorg, J.. and Dickenson, A., Long-lasting neuronal activity in rat dorsal horn evoked by impulses in cutaneous C-fibres during noxious mechanical stimulation. Brain Res., 439 (1988) 56-63. Shahani. B., and Young, R., Human flexor reflexes, J. Neurol. Neurosurg. Psychiat.. 34 (1971) 616-627. Sivilotti, L.G., Thompson, S.W.N.. and Woolf, C.J., Rate of rise of the cumulative depolarisation evoked by repetitive stimulation of

small-caliber afferents is a predictor of potenttal windup in rot neurons in vitro, J. Neurophysiol., 69 (1993) 1621-1631. Svensson, P.. Arendt-Nielsen, L.. Kaaber, S., and Bjerring, P.. Vertex potentials evoked by nociceptive laser stimulation of oral mucosa: a comparison of four stimulation paradigms, Anesth. Pain Control Dent., 1 (1992) 222-228. Toft, E., and Sinkjzr, T., H-reflex changes during contractions of the ankle extensors in spastic patients, Acta Neurol. Stand., l (1993) l-7. Torebjiirk, H.E., and Ochoa, J.L., Specific sensations evoked by activity in single identified sensory units in man, Acta Physiol. Stand., 110 (1980) 445-447. Wall, P.D.. and Woolf C.J., Muscle but not cutaneous C-aiferent input produces prolonged increases in the excitability of the flexion reflex in the rat, J. Physiol., 356 (1984) 443-458. Wilier, J.C., Comparative study of perceived pain and nociceptive reflexes in man, Pain. 3 (1977) 69-80. Wilier, J.C., Boureau, F., and Albe-Fessard, D.. Role of large diameter cutaneous afferents in transmission of nociceptive messages: electrophysiological study in man. Brain Res., 152 (1978) 358-364. Wilier, J.C., De Broucker, T., and Le Bars, D., Encoding of nociceptive thermal stimuli by diffuse noxious inhibitory controls in humans. J. Neurophysiol., 62 (1989) 1028-1038. Willis, W.D., The Pain System: the Neural Basis of Nociceptive Transmission in the Mammalian Nervous System. Karger. New York. 1985. Woolf, C.J., Evidence for a central component of post-injury pain hypersensitivity, Nature, 306 (1983) 686-688. Woolf, C.J.. and McMahon, S.B., Injury-induced plasticity of the flexor reflex in chronic decerebrate rats, Neuroscience. 16 (198% 395-404. Woolf. C.J., and Thompson, S.W.N., the induction and maintenance of central sensitization is dependent on N-methyl-o-aspartic acid receptor activation: implications for the treatment of postinjury hypersensitivity states, Pain, 44 (1991) 293-299. Woolf, C.J., and Wall, P.D.. Relative effectiveness of C primary afferent fibers of different origins in evoking a prolonged facilitation of the flexor reflex in the rat, J. Neurosci., 6 (1986) 1433.-1442.