Hyperalgesia and allodynia are common clinical conditions characterized by perturbations in the relationship between a peripheral stimulus and the sensation of pain. A better understanding of the physiological and pharmacological mechanisms underlying these painful conditions might promote the development of new analgesic drugs. Many experimental studies have suggested that changes at both the periphery and secondary mechanisms in the central nervous system, such as central sensitization, are involved in the development of hyperalgesia/allodynia (1–3). Central sensitization corresponds to a state of hyperexcitability of central (notably spinal) nociceptive neurons induced by persistent or tonic nociceptive peripheral inputs (4,5).
The temporal summation of nociceptive inputs might play a significant role in the development of central sensitization and hyperalgesia. In animals, the repeated application of nociceptive stimuli at a fixed intensity and a relatively high frequency (i.e., >0.3 Hz) results in a progressive increase in the responses of spinal nociceptive neurons and reflexes (6–8). During the last few years, numerous experimental studies have been devoted to this “wind-up” phenomenon, as it may represent an elementary form of central sensitization. One of the main arguments to suggest such a relationship between these phenomena is that both wind-up (9,10) and central sensitization involve the activation of N-methyl-D-aspartate (NMDA) receptors (11,12).
Consistent with these findings, psychophysical experiments in humans have demonstrated that the temporal summation resulting from repetitive nociceptive stimuli results in a similar increase in the sensation of pain (13,14). This effect might be related to the wind-up of nociceptive responses described in animals because it is observed with similar frequencies of stimulation and is reduced after the administration of NMDA receptor antagonists such as ketamine or dextromethorphan (15–17). Thus, it is thought that NMDA-mediated mechanisms similar to those described in animals also function in humans and might lead to central sensitization under pathological conditions. Such a state of hyperexcitability of central nociceptive neurons might explain why, in some patients, non-nociceptive stimuli can induce pain (i.e., allodynia), whereas in others, normal nociceptive stimuli can induce abnormally large responses (i.e., hyperalgesia). In accordance with this hypothesis, several clinical studies have indicated that the administration of NMDA receptor antagonists is particularly effective in reducing evoked pain in patients suffering from neuropathic pain (18–21) or inflammatory pain from the region surrounding a surgical incision (22).
We sought to provide a useful psychophysiological index for quantifying the wind-up phenomenon and analyzing its mechanisms in humans. We compared the effects of small systemic doses of ketamine or a placebo, administered in a double-blinded, cross-over design, on the temporal summation of the nociceptive flexion (RIII) reflex and the sensation of pain in normal volunteers. The RIII reflex is a polysynaptic spinal reflex that can be easily elicited by electrical stimulation of a sensory nerve, such as the sural nerve, and recorded from a flexor muscle, such as the biceps femoris in the ipsilateral limb. This reflex is a useful tool for objectively studying the spinal mechanisms of nociception in humans because its threshold and amplitude are closely related to those of the concomitant pain evoked by electrical stimulation (23). In addition to investigating the selective action of ketamine on temporal summation, we also evaluated its effects on the recruitment (i.e., stimulus/response) curve for the RIII reflex up to the pain tolerance threshold.
Experiments were performed, with local ethics committee approval, on six healthy volunteers (men 25–40 yr old). The participants were carefully briefed on the experimental procedure, and they gave their informed, written consent.
The RIII reflex was evoked and recorded using a completely computerized system (Notocord Systems, Croissy, France), using techniques previously described and validated (23). In brief, the sural nerve was stimulated electrically using a pair of surface electrodes placed 2 cm apart on the degreased skin overlying the nerve within its retromalleolar path. Each electrical stimulus consisted of a train of five rectangular pulses, each of 1-ms duration, delivered over a 12-ms period, from a constant current stimulator. Electromyographic responses were recorded from the ipsilateral biceps femoris muscle via a pair of surface electrodes placed 2 cm apart on the skin over the muscle. The RIII reflex responses were identified as multiphasic waves appearing 90–180 ms after the onset of the stimuli. By restricting the study to responses within this time, it is possible to avoid the tactile (RII) reflex, which can occur between 50 and 70 ms, or artifacts produced by involuntary movements, which can be observed as early as 250–300 ms after the stimulus. Each reflex response was amplified, digitized, full wave-rectified, and integrated. The resulting integrals were used to quantify the RIII response.
The six volunteers underwent experiments on two occasions separated by an interval of 7 days. On each occasion, two experimental sequences were performed: one before (control period) and one 5 min after the IV administration (over a 1-min period) of 0.15 mg/kg ketamine or a placebo (saline). This dose was chosen based on the results of previous clinical studies (18, 19) and pilot experiments. The injections were administered in a double-blinded and randomized fashion. Each sequence consisted of 1) constructing the recruitment curve for the reflex as a function of stimulus intensity, which was increased progressively up to the tolerance threshold at a frequency of 0.17 Hz (i.e., 10 stimuli/min); 2) adjusting the stimulus intensity to 1.2 times the reflex threshold and checking the stability of the recordings and reflex responses over a 2-min period (with the stimuli at a frequency of 0.17 Hz); 3) applying a series of 15 such stimuli (i.e., 1.2 times threshold) at 1 Hz. The subjects were asked to use a 100-mm visual analog scale (VAS) graduated from 0 (no pain) to 100 (worst possible pain) to rate both the sensation evoked by the first stimulus in this series and the maximal pain produced by any of the stimuli. The stimulation frequency was chosen based on pilot experiments, which had shown that the magnitude of the wind-up of the reflex was much less at lower frequencies (e.g., 0.5 Hz), whereas higher frequencies (e.g., 1.5 or 2 Hz) resulted in fusion of successive responses, which hampered the analysis of each individual response. Similarly, with respect to the stimulus intensity, higher stimulus intensities (e.g., 1.4–1.5 times the threshold) induced intolerable increases in the sensations of pain and motor artifacts during the stimulation sequences.
Data are expressed as means ± SEM. The reflex threshold was defined as the intersection of the linear regression line for the RIII stimulus/response relationship with the abscissa. The tolerance threshold was the maximal intensity on the recruitment curve. To allow analysis of group data, each reflex response was expressed as a percentage of the maximal response observed during the control recruitment curve (i.e., before the injection). Recruitment curves were normalized between 0 and 30 mA such that when the tolerance threshold was <30 mA, the last value obtained for the reflex was assigned to all the higher intensities in the series. Temporal summation was analyzed by expressing each response during the sequence of 15 stimuli at 1 Hz as a percentage of the first response. Wilcoxon’s signed ranks test was used for comparison of paired data. The areas under the mean recruitment curves and the mean wind-up curves were calculated and used to compare the effects of the placebo and ketamine by using of repeated-measures analysis of variance. The influence of treatment order (placebo-ketamine or ketamine-placebo) and period (first or second stage) were also tested. Results were considered significant at P < 0.05.
The results indicate that the administration of a small systemic dose of ketamine, which did not significantly alter the recruitment curve for the RIII reflex, did selectively depress the progressive increase (i.e., wind-up) of both the reflex and the sensation of pain associated with the repeated application of nociceptive stimuli at a fixed intensity.
Effects of Ketamine on the Recruitment Curve of the Nociceptive Flexion Reflex
The recruitment curves of the reflex obtained with low-frequency electrical stimuli (i.e., 0.17 Hz) did not differ significantly between the two control periods and were not significantly altered after the injection of ketamine or placebo (Figure 1). The mean RIII reflex thresholds were 9.3 ± 1.4 and 9.1 ± 1.5 mA (not significant) in the first and second control periods, respectively, and were not significantly different after the administration of ketamine or placebo (9.5 ± 1.6 and 8.9 ± 1.3 mA). The mean tolerance thresholds (i.e., the maximal stimulus intensities on the recruitment curves) were 22.5 ± 4.1 and 25.3 ± 5.3 mA during the first and second control periods, respectively, and were not significantly different after the injection of ketamine or placebo (23.2 ± 2.6 and 24.6 ± 3.1 mA).
Effects of Ketamine on the Temporal Summation Affecting the Reflex and the Sensation of Pain
The application of a series of 15 stimuli at 1.2 times the reflex threshold and a frequency of 1 Hz induced a progressive increase in the reflex responses. An example is illustrated in Figure 2. The cumulative data presented in Figure 3A indicate that the application of stimuli at this frequency induced a strong facilitation of the reflex responses (up to 392% ± 94% of the first response), which peaked between the fifth and eighth stimuli in the series. The first response in the series was not significantly altered after the administration of ketamine or placebo. Thereafter, the progressive increase (i.e., wind-up) in the reflex was reduced significantly after the administration of ketamine (P < 0.01) but not placebo (Figures 2 and 3B).
In addition to the temporal facilitatory effects on the reflex responses, there was a concomitant increase in the sensations of pain. During repetitive stimulation, the subjects reported both an increase in pain intensity and an increased spread of the pain in the lower limb. This effect of temporal summation on pain was significantly reduced by ketamine but not placebo. The increase in pain during the series of 15 stimuli was similar in the two control periods (i.e., before the administration of ketamine or placebo). Baseline VAS scores (i.e., the VAS score for the first stimulus in the series) were 18 ± 9 and 18 ± 13 mm in the first and second control periods, respectively. The corresponding maximal VAS scores during the series were 43 ± 12 and 48 ± 12 mm. The baseline VAS scores were not significantly altered by the injection of ketamine or placebo (19 ± 11 and 20 ± 14 mm, respectively). By contrast, the maximal VAS scores were significantly reduced after the administration of ketamine (28 ± 13 mm;P < 0.05) but not placebo (55 ± 18 mm).
The sequence (placebo-ketamine or ketamine-placebo) and the period (first or second session) did not influence the results.
All the subjects reported one or several side effects after ketamine administration. These included sedation (one of six), illusions (two of six), paraesthesia in the limbs (one of six), increased sensitivity to light and sound (two of six), unpleasant dreams (one of six), and dizziness (one of six). The side effects were maximal 1–3 min after the injection.
The present data from humans indicate that the temporal summation of inputs from nociceptive stimuli results in a parallel increase in both the RIII reflex and the sensation of pain and that this wind-up phenomenon involves the activation of NMDA receptors, as it is selectively decreased after the administration of ketamine.
Our electrophysiological results are in keeping with those of Arendt-Nielsen et al. (17), who showed that the “temporal summation threshold” for the RIII reflex (i.e., the threshold of the reflex induced by the application of five electrical stimuli at 2 Hz) was significantly increased during an infusion of ketamine. However, these effects were not selective because the dose injected (a bolus of 0.5 mg/kg followed by an infusion of 9 μg · kg−1 · min−1) also induced a reduction in the responses to suprathreshold mechanical and electrical stimuli. By contrast, the smaller dose of ketamine we used (0.15 mg/kg) did not induce general analgesic effects. This was shown by the absence of a significant alteration in the stimulus/response curves for the reflex. One major problem associated with the use of ketamine is its numerous side effects. Although we used a very small systemic dose, all the subjects reported some side effects; consequently, the study was not completely blind. However, because of the selectivity of action of ketamine on the wind-up phenomenon, it is unlikely that the present results were due entirely to nonspecific effects.
Wind-up of nociceptive responses has been well characterized in animals. It has been demonstrated in spinal dorsal and ventral horn neurons, both in vivo and in vitro, after stimulation of thin myelinated Aδ or unmyelinated C fibers (6–10,24), as well as in nociceptive reflexes (25–27). Intracellular recordings have shown that, in contrast to what happens after activation of Aβ fibers, the activation of thin afferent fibers induces excitatory postsynaptic potentials in dorsal horn nociceptive neurons, which can last several hundreds of milliseconds (28). Thus, the application of repetitive stimuli at a relatively rapid frequency (i.e., >0.3 Hz) can induce an increasing and cumulative depolarization, resulting in an increase in the number of action potentials evoked by each stimulus. This slow depolarization is due to the co-release of excitatory amino acids (EAA), such as glutamate, and neuropeptides, such as substance P (29) by nociceptive primary afferents. The release of EAA in the spinal dorsal horn and the activation of NMDA receptors seem to play a major role in wind-up, because it is selectively reduced by NMDA receptor antagonists (9,10). However, the action of EAA is probably modulated by neuropeptides (29). In addition, recent in vitro experiments suggest that other mechanisms, notably those mediated by L-type Ca2+ currents, may also have a significant role in these processes (30).
The fact that wind-up is mediated through NMDA receptors has lead to the proposition that it may represent an intermediate step in the development of central sensitization. Central sensitization of dorsal horn nociceptive neurons is characterized by a reduction in their threshold for activation, an increase in their spontaneous activity and responses to noxious peripheral stimuli, and an expansion of their receptive fields. This phenomenon has been observed in models of tissue or nerve injury and is thought to represent a pathophysiological mechanism for various acute and chronic pain syndromes (1–5). The increased release of EAA in the spinal cord due to the tonic or persistent activation of nociceptive afferents under pathological conditions (e.g., inflammatory processes) could be sufficient to lead to an activation of NMDA receptors, which is capable of inducing long-term alterations in neuronal properties. Such modifications are mediated by the intracellular influx of Ca2+ and the activation of protein kinase C and NO synthase and could result in a state of hyperexcitability or central sensitization of nociceptive neurons. In accordance with this hypothesis, behavioral signs of hyperalgesia can be reduced by NMDA antagonists (3–5).
However, despite these apparent similarities, there are significant discrepancies between the electrophysiological results concerning wind-up and the behavioral indices suggestive of hyperalgesia in animals (31). One major difference is that, in contrast to hyperalgesia, which may last for minutes, hours, or days, wind-up subsides within several seconds. The present results confirm that hyperalgesia induced by wind-up does not outlast the period of high-frequency stimulation. In addition, it is likely that much more complex alterations are also induced by the temporal summation of nociceptive inputs, because this can also activate descending inhibitory controls that counteract the facilitatory processes in the spinal cord (26,27). Finally, it has been reported that wind-up does not induce all the characteristics of central sensitization of dorsal horn neurons (32). Although wind-up alone cannot trigger complete, long-lasting central sensitization, these phenomena have some common characteristics, such as the ability to induce hyperalgesia or to be reduced by NMDA antagonists. This reinforces the interest in wind-up as a useful model for physiological of pharmacological studies.
In keeping with previous psychophysical studies, we observed that the temporal summation of the nociceptive reflex was associated with a parallel increase in sensations of pain. Unfortunately, because of the relatively high frequency of stimulation, it was not possible to rate the sensations evoked by each stimulus in the series and, thus, to correlate precisely the increase in reflexes and sensations. Most of the subjects reported a progressive increase in the spread and duration of the painful sensations during the stimulation sequence, which also hampered analysis of the sensations elicited by each stimulus. A similar effect of temporal summation on pain has been observed with nociceptive thermal, mechanical, and electrical stimuli (13,14). It parallels the wind-up phenomenon observed in animals because it is frequency-dependent and is selectively reduced by NMDA receptor antagonists such as ketamine or dextromethorphan (15–17). Thus, it is likely that central mechanisms similar to those proposed in animals also exist in humans. Although central sensitization has not been directly demonstrated in humans, several lines of evidence suggest that the secondary hyperalgesia induced by experimental burn injury or the local application of chemical irritants, as well as brush-induced allodynia in patients with nerve injuries (33,34), are dependent on central mechanisms. In addition, ketamine can reduce the secondary hyperalgesia induced experimentally (35). Finally, several studies suggest that NMDA receptor antagonists are clinically effective and exert effects predominantly on evoked pains (i.e., allodynia and hyperalgesia) associated with nerve or tissue injuries (18–21). Unfortunately, the use of these drugs is hampered by their psychotomimetic side effects.
These clinical results encourage a search for new molecules with more favorable tolerance profiles. The present data suggest that the RIII reflex might be a useful tool for pharmacological studies, particularly for objectively evaluating the effects of such molecules in humans. Such a methodology could also be used to analyze the effects of temporal summation and the alterations in the spinal processing of pain under various pathological conditions.
We thank Dr. S. W. Cadden for advice in the preparation and correction of the manuscript.
1. Dubner R. Neuronal plasticity and pain following peripheral tissue inflammation or nerve injury. In: Bond MR, Charlton JE, Woolf CJ, eds. Proceedings of the VIth World Congress on Pain. Amsterdam: Elsevier, 1991: 264–76.
2. Dubner R, Ruda MA. Activity-dependent neuronal plasticity following tissue injury and inflammation. Trends Neurosci 1992; 15:96–103.
3. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993; 52:259–85.
4. Yaksh TL. The spinal pharmacology of facilitation of afferent processing evoked by high-threshold afferent input of the postinjury pain state. Curr Opin Neurol Neurosurg 1993; 6:250–6.
5. Price DD, Mao J, Mayer D. Central mechanism of normal and abnormal pain states. In: Fields HL, Liebeskind JC, eds. Progress in pain research and management. Vol.1. Seattle, WA: IASP, 1994: 61–84.
6. Mendell LM. Physiological properties of unmyelinated fiber projection to the spinal cord. Exp Neurol 1966; 16:616–22.
7. Wagman IH, Price DD. Responses of dorsal horn cells of M. mulatta to cutaneous and sural nerve A and C fiber stimuli. J Neurophysiol 1969; 32:803–17.
8. Price DD, Hayes RL, Ruda MA, Dubner R. Spatial and temporal transformations of input to spinothalamic tract neurons and their relation to somatic sensation. J Neurophysiol 1978; 41:933–47.
9. Davies SN, Lodge D. Evidence for involvement of N-methyl-D-aspartate receptors in “wind-up” of class 2 neurons in the dorsal horn of the rat. Brain Res 1987; 424:402–6.
10. Dickenson AH, Sullivan AF. Evidence for a role of the NMDA receptor in the frequency dependent potentiation of the deep rat dorsal horn nociceptive neurons following C fibre stimulation. Neuropharmacology 1987; 26:1235–8.
11. Ren K, Hylden JL, Williams GM, et al. The effects of a non-competitive receptor NMDA antagonist, MK-801, on behavioral hyperalgesia and dorsal horn neuronal activity in rats with unilateral inflammation. Pain 1992; 50:331–44.
12. Neugebauer V, Lucke T, Schaible HG. N-Methyl-D-Aspartate (NMDA) and non-NMDA receptor antagonists block the hyperexcitability of dorsal horn neurons during development of acute arthritis in the rat’s knee joint. J Neurophysiol 1993; 70:1365–77.
13. Price DD, Hu JW, Dubner R, Gracely RH. Peripheral suppression of first pain and central summation of second pain evoked by noxious heat pulses. Pain 1977; 3:57–68.
14. Arendt-Nielsen L, Brennum J, Sindrup S, Bak P. Electrophysiological and psychophysical quantification of temporal summation in the human nociceptive system. Eur J Appl Physiol 1994; 68:266–73.
15. Price DD, Mao J, Frenk H, Mayer DJ. The N -methyl-D-aspartate receptor antagonist dextromethorphan selectively reduces temporal summation of second pain in man. Pain 1994; 59:165–74.
16. Arendt-Nielsen L, Nielsen J, Petersen-Felix S, et al. Effect of racemic mixture and the (S+)-isomer of ketamine on temporal and spatial summation of pain. Br J Anaesth 1996; 77:625–31.
17. Arendt-Nielsen L, Petersen-Felix S, Fischer M, et al. The effects of N-methyl-D-aspartate antagonist (ketamine) on single and repeated nociceptive stimuli: a placebo-controlled experimental human study. Anesth Analg 1995; 81:63–8.
18. Backonja M, Arndt G, Gombar KA, et al. Responses of chronic neuropathic pain syndromes to ketamine: a preliminary study. Pain 1994; 56:51–7.
19. Eide PK, Jorum E, Stubhaug A, et al. Relief of post-herpetic neuralgia with the N-methyl-D-aspartic acid receptor antagonist ketamine: a double-blind, cross-over comparison with morphine and placebo. Pain 1994; 58:347–54.
20. Nikolajsen L, Hansen CL, Nielsen J, et al. The effects of ketamine on phantom pain: a central neuropathic disorder maintained by peripheral input. Pain 1996; 67:69–77.
21. Nikolajsen L, Hansen PO, Jensen TS. Oral ketamine therapy in the treatment of postamputation stump pain. Acta Anaesthesiol Scand 1997; 41:427–9.
22. Stubhaug A, Breivik H, Eide PK, et al. Mapping of punctate hyperalgesia around a surgical incision demonstrates that ketamine is a powerful suppressor of central sensitization to pain following surgery. Acta Anaesthesiol Scand 1997; 41:1124–32.
23. Willer JC. Comparative study of perceived pain and nociceptive flexion reflex in man. Pain 1977; 3:69–80.
24. Thompson SWN, King AE, Woolf CJ. Activity-dependent changes in rat ventral horn neurons in vitro: summation of prolonged afferent evoked postsynaptic depolarizations produces a D-2-amino-5-phoshonovaleric acid sensitive wind up. Eur J Neurosci 1990; 2:638–49.
25. Price DD. Characteristics of second pain and flexion reflexes indicative of prolonged central summation. Exp Neurol 1972; 37:371–87.
26. Gozariu M, Bouhassira D, Willer JC, et al. The influence of temporal summation on a C-fibre reflex in the rat: effects of lesions in the rostral ventromedial medulla (RVM). Brain Res 1998; 792:168–72.
27. Gozariu M, Bragard D, Le Bars D, et al. Temporal summation of C-fibre inputs: competition between facilitatory and inhibitory effects on a C-fibre reflex in the intact rat. J Neurophysiol 1997; 78:3165–79.
28. Price DD, Hull CD, Buchwald NA. Intracellular responses of dorsal horn cells to cutaneous and sural nerve A and C fiber stimuli. Exp Neurol 1971; 33:291–309.
29. Urban L, Thompson SW, Dray A. Modulation of spinal excitability: co-operation between neurokinin and excitatory amino acid transmitters. Trends Neurosci 1994; 17:432–8.
30. Russo RE, Nagy F, Hounsgaard J. Modulation of plateau properties in dorsal horn neurons in a slice preparation of the turtle spinal cord. J Physiol 1997; 499:459–74.
31. Woolf CJ. Wind up and central sensitization are not equivalent. Pain 1996; 66:105–8.
32. Li J, Simone DA, Larson AA. Windup leads to characteristics of central sensitization. Pain 1999; 79:75–82.
33. Treede RD, Meyer RA, Raja SN, et al. Peripheral and central mechanisms of cutaneous hyperalgesia. Prog Neurobiol 1992; 38:397–421.
34. Koltzenburg M, Torebjork HE, Wahren LK. Nociceptor modulated central sensitization causes mechanical hyperalgesia in acute chemogenic and neuropathic pain. Brain 1994; 117:579–91.
© 2000 International Anesthesia Research Society
35. Ilkjaer S, Petersen KL, Brennum J, et al. Effect of systemic N-methyl-D-aspartate receptor antagonist (ketamine) on primary and secondary hyperalgesia in humans. Br J Anaesth 1996; 76:829–34.