- Email: [email protected]
The effect of sympathetic activity on thermal hyperalgesia in capsaicin-treated skin during body cooling and warming
European Journal of Pain (2001) 5: 59–67 doi:10.1053/eujp.2000.0224, available online at http://www.idealibrary.com on
The effect of sympathetic activity on thermal hyperalgesia in capsaicin-treated skin during body cooling and warming Peter D. Drummond School of Psychology, Murdoch University, Western Australia
An adrenergic mechanism is thought to contribute to pain in conditions that sometimes develop during chronic inflammation and after nerve or tissue injury. There is some doubt, however, about whether adrenergic activity influences nociception in acute inflammation. To investigate this issue, the noncompetitive alpha-(α) adrenergic antagonist phenoxybenzamine was introduced by iontophoresis into the skin of 16 healthy volunteers either before or after the topical application of capsaicin. When applied before capsaicin, phenoxybenzamine increased thermal hyperalgesia at normal ambient temperatures and during body warming. These findings suggest that phenoxybenzamine blocked an analgesic mechanism when applied before the onset of inflammation. However, this effect disappeared during body cooling. When applied after capsaicin, phenoxybenzamine inhibited thermal hyperalgesia at normal ambient temperatures, and during body warming and cooling. Thus, phenoxybenzamine blocked a hyperalgesic mechanism when applied after the onset of inflammation. It was concluded that the presence of inflammation influences the nociceptive effect of α-adrenergic blockage, possibly by increasing access to excitatory adrenergic receptors on nociceptive afferents. An excitatory adrenergic influence on nociception may overcome an inhibitory adrenergic influence during inflammation. © 2001 European Federation of Chapters of the International Association for the Study of Pain KEYWORDS: thermal hyperalgesia, capsaicin, α-adrenoreceptors, inflammation.
INTRODUCTION An adrenergic mechanism contributes to chronic pain in conditions that sometimes develop after nerve and tissue injury (Torebjork et al., 1995; Choi and Rowbotham, 1997). Investigation of this mechanism in animal models of peripheral nerve injury has shown that sensitivity to electrical stimulation of the sympathetic trunk and to subcutaneous injection of noradrenaline develops in C-fibre polymodal nociceptors that survive partial nerve section (Sato & Perl, 1991). Conversely, pain behaviours and ectopic discharges in the injured Paper received 4 August 2000 and accepted for publication 11 December 2000. Correspondence to: Dr Peter Drummond, School of Psychology, Murdoch University, 6150, Western Australia. Fax: 61-8-93606492; E-mail: [email protected]
afferents of animals with neuropathic pain are inhibited by alpha-(α) adrenergic antagonists (Lee et al., 1999). Sensitivity to noradrenaline also develops in cutaneous C-fibre polymodal nociceptors in inflamed tissue (Hu & Zhu, 1989; Sato et al., 1993; Sato et al., 1994), suggesting that the inflammatory response to nerve and tissue injury might activate an adrenergic mechanism of pain. In humans, noradrenaline introduced into symptomatic skin sometimes increases pain and mechanical hyperalgesia in patients with complex regional pain syndrome (Davis et al., 1991; Torebjörk et al., 1995; Ali et al., 2000) and post-herpetic neuralgia (Choi and Rowbotham, 1997). Noradrenaline also increases thermal hyperalgesia in some patients with injured peripheral nerves (Choi and Rowbotham, 1997), and evokes a similar effect in the inflamed skin of healthy subjects. In particular, the intradermal administration of noradrenaline lowers the
1090-3801/01/010059 + 09 $35.00/0 © 2001 European Federation of Chapters of the International Association for the Study of Pain
60
heat–pain threshold in skin sensitized by capsaicin (Drummond, 1995; Drummond, 1998a; Drummond, 1998b; Drummond, 1999) and, to a lesser extent, in untreated skin (Drummond, 1996; Fuchs et al., 2001). Intravenous infusion of adrenaline also lowers the heat-pain threshold in capsaicin-treated and untreated skin (Janssen et al., 1998). Although the intradermal administration of noradrenaline clearly influences thermal hyperalgesia in inflamed skin, there is some debate about whether the normal release of endogenous stores of noradrenaline from sympathetic nerve endings increases nociceptor activity. For example, Pedersen et al. (1997) reported that lumbar sympathetic nerve block did not alter thermal or mechanical hyperalgesia in heat-injured skin in the leg. Kindgen-Milles and Holthusen (1997) found that neither noradrenaline nor the αadrenergic antagonist phentolamine influenced pain evoked by intravenous electrical stimulation or intravenous injection of hyperosmolar saline. Baron et al. (1999) reported that sensitivity to mechanical stimulation in skin sensitized by capsaicin was unaffected by body heating or cooling. In a direct test of the peripheral endogenous adrenergic influence on nociception, Elam et al. (1999) inserted microelectrodes into the peroneal nerve to measure nerve traffic in polymodal Cfibre afferents and sympathetic efferents. Activity increased in sympathetic efferents during the arousal provoked by mental stress (serial subtractions for 1 min) but did not change in nociceptor afferents, even after the afferents had been sensitized by mustard oil and brought close to threshold by repetitive electrical stimulation. In contrast to these negative findings, Kinnman et al. (1997) reported that subcutaneous injection of phentolamine inhibited ongoing pain and mechanical hyperalgesia around the site of intradermal capsaicin injection. In a similar study by Liu et al. (1996), intravenous injection of phentolamine inhibited the development of mechanical allodynia around the site of intradermal capsaicin injection. Finally, Drummond (1998a) reported that thermal hyperalgesia increased in capsaicintreated skin after the iontophoresis of tyramine, a substance that release endogenous stores of noradrenaline. Hyperalgesia to tyramine was European Journal of Pain (2001), 5
P.D. D RUMMOND
blocked by pretreatment with phenoxybenzamine, a noncompetitive α-adrenergic antagonist. Thus, it remains possible that the normal release of noradrenaline during sympathetic neural discharge influences the responsiveness of cutaneous nociceptors in some circumstances. The aim of the present study was to further explore the conditions in which sympathetic activity might influence nociceptor responsiveness. Since the pharmacological release of noradrenaline increases thermal hyperalgesia in the capsaicin-treated skin of healthy volunteers (Drummond, 1998a), it was hypothesized that an increase in sympathetic activity during body cooling would increase thermal hyperalgesia. To investigate the adrenergic component of nociception, phenoxybenzamine was introduced locally by iontophoresis into the capsaicin-treated skin. If endogenously released noradrenaline sensitizes nociceptors or increases their rate of discharge, thermal hyperalgesia should be lower at the site of α-adrenergic blockade than elsewhere in the capsaicin-treated skin.
METHOD Subjects The sample consisted of eight male and eight female university students aged between 17 and 40 years (mean age 21.6 years) who were in good health. Each subject provided informed consent for the procedures, which were approved by the Murdoch University Human Research Ethics committee. Procedures The experiments were run in a temperaturecontrolled laboratory, maintained at 24.1 ± 1.7˚C ( ± SD).
Application of capsaicin Capsaicin powder (Sigma, St Louis, MI, USA) was dissolved in 50% ethanol in distilled water at a concentration of 0.02 M (0.6%). The skin on the back of the hand or the volar aspect of the
α-A DRENERGIC
61
INFLUENCE ON THERMAL HYPERALGESIA
forearm near the elbow was cleaned with soap and water and an alcohol swab. If necessary, hair was shaved from the site with a disposable razor to facilitate the attachment of dressings and iontophoresis capsules to the skin. Care was taken not to touch the skin with the razor. The gauze pad of an elastic dressing (40 mm by 25 mm), containing 400 µl of the capsaicin solution, was then applied to the prepared skin. The dressing was covered by plastic tape to retard evaporation of the capsaicin solution. Thirty minutes later the dressing was removed and the treated skin was washed with soap and water.
Iontophoresis Phenoxybenzamine hydrochloride (ICN Pharmaceuticals, Costa Mesa, CA, USA) was prepared daily at 0.5 mM with distilled water from a 10 mM stock solution in distilled water. A capsule (internal diameter 0.8 cm) was attached to the site of application of capsaicin with an adhesive washer. The capsule was filled with the phenoxybenzamine solution, and a weak direct current (50 µA) was passed through the solution for 10 min to introduce the positively charged phenoxybenzamine ions into the skin. Iontophoresis was used in preference to intradermal injection to minimize systemic effects and to ensure that the phenoxybenzamine was distributed locally throughout the epidermis and upper layers of the dermis around the terminals of cutaneous nerves. The ground electrode was a silver plate measuring 3 cm by 5 cm, covered in electrode paste and attached to the volar aspect of the forearm near the wrist. In the first session, phenoxybenzamine was iontophoresed before the topical application of capsaicin (eight subjects), and shortly after the capsaicin had been washed from the skin (eight subjects). In the second session, phenoxybenzamine was iontophoresed before the topical application of capsaicin in five subjects, and shortly after the capsaicin had been washed from the skin in 11 subjects.
over light clothing to the subject’s chest and back. In addition, the subject’s face, trunk and legs were sprayed periodically with cold water while air was circulated around the subject with an electric fan. Care was taken to avoid directly cooling experimental sites on the forearm and hand.
Body warming (Second session.) Since the aim of this procedure was to minimize sympathetic vasoconstriction, body warming was always carried out in the second session when psychological responses to the laboratory setting were likely to be minimal. Approximately 1 h after the capsaicin had been washed from the skin, subjects were warmed gently with a convection heater placed behind their chair for 20 min. Experimental sites on the forearm and hand were not warmed directly.
Vascular responses The temperature of the fingertips was measured with a calibrated thermocouple before and after body cooling and warming. In addition, changes in blood flow in the capsaicin-treated skin were monitored with a wide surface area probe attached to a Moor MBF3D laser Doppler flowmeter (Moor Instruments, Axminster, UK). The probe consists of a central near-infrared laser beam emitter (wavelength 810 nm) surrounded by eight glass fibres, equally spaced on a 2-mm diameter circle, which collect the reflected light. The collected light was integrated by the MBF3D monitor to give an average flux over a surface area of approximately 7 mm2 and a depth of 1–2 mm. Blood flow was sampled once per second, and was later averaged using Moorsoft software for several minutes before and after cooling and warming. Because the flowmeter detected relative changes in blood flow, the change in flow after cooling or warming was expressed as the percent change from the level at baseline.
Body cooling (First session.) Approximately 1 h after the capsaicin had been washed from the skin, freezer packs (25 cm by 35 cm) were applied for 20 min
Heat–pain thresholds The radiant heat from a halogen globe was focused through a 6-mm diameter aperture European Journal of Pain (2001), 5
62
Statistical analyses Thermal hyperalgesia was defined as the difference between the heat–pain threshold in untreated skin before body cooling or warming and the heat–pain threshold at experimental sites in the capsaicin-treated skin. These scores were investigated in analyses of variance (SPSS for Windows, Version 8.1), with one betweensubjects factor of Timing of the Phenoxybenzamine Iontophoresis (phenoxybenzamine administered before versus after capsaicin) and repeated measures factors for site (the site pretreated with phenoxybenzamine versus the control sites) and time (before and after body cooling or warming). Results are presented as the mean ± SE.
RESULTS Body cooling Finger temperature decreased from 28.6 ± 1.0˚C to 22.3 ± 0.4 ˚C during body cooling [t(15) = 8.34, p < 0.001]. Blood flow decreased substantially through the capsaicin-treated skin of 14 of the 16 subjects, and remained close to baseline in the other two subjects [mean decrease 57 ± 8%, t(15) = 7.07, p < 0.001]. The heat–pain threshold in skin not treated with capsaicin averaged 44.0 ± 0.9˚C before cooling, and 43.6 ± 1.0˚C after European Journal of Pain (2001), 5
6
Phenoxybenzamine after capsaicin
*
4 2
*
0 −2 −4
Before cooling
After cooling
Control sites
Thermal hyperalgesia (°C)
Phenoxybenzamine before capsaicin Thermal hyperalgesia (°C)
placed just above the skin. Skin temperature was monitored with a thermocouple that touched the skin in the centre of the aperture. To assess the heat-pain threshold, skin temperature was brought to 32˚C for 10–15 s and then increased at 0.5˚C per second until the subject signalled pain by switching the lamp off or to a maximum of 49˚C. The heat–pain threshold was measured at the site of phenoxybenzamine iontophoresis and at three control sites in the capsaicin-treated skin, before and after body cooling and warming. The heat–pain threshold at each site was calculated as the average threshold from two or three temperature ramps. Sensitivity to heat–pain was also measured twice at two sites on the untreated forearm or hand, before and after body cooling and warming.
P.D. D RUMMOND
6 4 2 0 −2 −4
* Before cooling
After cooling
Phenoxybenzamine site
FIG 1. Effect of alpha-(α) adrenergic blockade on thermal hyperalgesia before and after 20 min of body cooling. Thermal hyperalgesia was defined as the difference between the heat–pain threshold in untreated skin before body cooling (representing the subject’s usual heat–pain threshold) and heat–pain thresholds in the capsaicin-treated skin. Phenoxy benzamine was administered before capsaicin in eight subjects, and after capsaicin in the other eight subjects. Difference between the phenoxybenzamine site (■) and the control sites (■ ■) statistically significant (* p < 0.05). Bars represent standard errors.
20 min of cooling (difference not significant). In contrast, the heat–pain threshold at control sites in the capsaicin-treated skin decreased from 42.6 ± 0.8˚C to 41.3 ± 0.9˚C after cooling [t(15) = 2.24, p < 0.05]. The effect of α-adrenergic blockade on thermal hyperalgesia depended on whether phenoxybenzamine was applied before or after the application of capsaicin (Fig. 1). The interaction between timing of the phenoxybenzamine iontophoresis, site and time was highly significant [F(1,14) = 47.8, p < 0.001], indicating that body cooling as well as the timing of the phenoxybenzamine iontophoresis influenced the effect of αadrenergic blockade on thermal hyperalgesia. When phenoxybenzamine was administered before capsaicin, thermal hyperalgesia was greater at the treated site than at the control sites before body cooling [t(7) = 2.68, p < 0.05]. However, hyperalgesia was greater at the control sites than at the treated site after body cooling [t(7) = 2.74, p < 0.05]. When phenoxybenzamine was administered after capsaicin, thermal hyperalgesia was greater at the control sites than at the treated site before body cooling [t(7) = 3.05, p < 0.05]. This persisted to some extent during body cooling (Fig. 1),
63
INFLUENCE ON THERMAL HYPERALGESIA
Body warming Finger temperature increased from 29.1 ± 1.0˚C before warming to 30.5 ± 0.7˚C after 20 min of warming (increase not significant). In the group as a whole, blood flow through the capsaicintreated skin did not change significantly during body warming (mean increase 8 ± 13%); however, blood flow increased 53 ± 18% in seven subjects and decreased 28 ± 6% in nine others. Heat–pain thresholds in skin not treated with capsaicin averaged 44.5 ± 0.5˚C before body warming and 44.3 ± 0.5˚C after 20 min of body warming (difference not significant). In contrast, heat–pain thresholds at control sites in the capsaicin-treated skin decreased from 41.0 ± 0.6˚C to 39.9 ± 0.7˚C during body warming [t(15) = 3.26, p < 0.01]. In the seven subjects with an increase in blood flow in the capsaicin-treated skin during body warming, the heat pain threshold decreased 1.9 ± 0.4˚C [t(6) = 5.42, p < 0.01]. In contrast, the heat pain threshold decreased only 0.5 ± 0.4˚C in the other nine subjects (mean decrease not significant). As in session 1, the effect of α-adrenergic blockade on thermal hyperalgesia depended on whether phenoxybenzamine was administered before or after capsaicin [interaction between timing of the phenoxybenzamine iontophoresis and site, F(1,14) = 31.0, p < 0.001] (Fig. 2). Body warming did not influence the effect of αadrenergic blockade on thermal hyperalgesia. When phenoxybenzamine was administered before capsaicin, thermal hyperalgesia was greater at the treated site than at control sites in the capsaicin-treated skin [mean heat–pain threshold 1.6 ± 0.5˚C lower at the phenoxybenzamine site than at the control sites, t(4) = 3.08, p < 0.05]. In contrast, when phenoxybenzamine was administered after capsaicin, thermal hyperalgesia was greater at control sites in the capsaicin-treated skin than at the phenoxybenzamine site [mean heat pain threshold 2.0 ± 0.4˚C higher at the phenoxybenzamine site than at the control sites, t(10) = 5.50, p < 0.001].
Phenoxybenzamine before capsaicin 8
*
6 4 2
Before warming
After warming
Control sites
Phenoxybenzamine after capsaicin Thermal hyperalgesia (°C)
although after 20 min of cooling the difference between the treated and control sites was no longer statistically significant.
Thermal hyperalgesia (°C)
α-A DRENERGIC
8 6 *
4 2
*
Before warming
After warming
Phenoxybenzamine site
FIG 2. Effect of alpha-(α) adrenergic blockade on sensitivity to heat–pain before and after 20 min of body warming. Thermal hyperalgesia was defined as the difference between the heat–pain threshold in untreated skin before body warming and heat–pain thresholds in the capsaicin-treated skin. Phenoxybenzamine was administered before capsaicin in five subjects, and after capsaicin in the other 11 subjects. Difference between the phenoxybenzamine site (■) and the control sites (■ ■) statistically significant (* p < 0.05). Bars represent standard errors.
DISCUSSION The main findings of the present study were as follows. Thermal hyperalgesia intensified at control sites in the capsaicin-treated skin during body cooling and heating. At normal ambient temperatures and during body warming, the effect of phenoxybenzamine depended on whether it was administered before or after the capsaicin treatment. When administered before the capsaicin treatment, phenoxybenzamine intensified thermal hyperalgesia at normal ambient temperatures and during body warming. However, when administered after the capsaicin treatment, phenoxybenzamine inhibited thermal hyperalgesia. Finally, when administered before the capsaicin treatment, phenoxybenzamine inhibited thermal hyperalgesia during body cooling. Hyperalgesic effect of phenoxybenzamine Capsaicin sensitizes polymodal nociceptors to heat (Winter et al., 1995), probably by acting on heat-sensitive ion channels (Cesare et al., 1999). Thermal hyperalgesia intensified in capsaicintreated skin during body warming, primarily in European Journal of Pain (2001), 5
64
association with increases in local blood flow. Thus, an increase in cutaneous temperature associated with an increase in blood flow may have contributed to hyperalgesia during body warming. Since release of vasoconstrictor tone by α-adrenergic blockade should also increase blood flow, a similar mechanism might have contributed to thermal hyperalgesia at sites of phenoxybenzamine administration. However, this does not fully account for the effect of phenoxybenzamine on thermal hyperalgesia, because hyperalgesia intensified only when phenoxybenzamine was administered before the capsaicin treatment. Furthermore, the hyperalgesic effect of phenoxybenzamine persisted during body warming, when sympathetic vasoconstrictor tone at control sites was likely to be minimal. Thus, additional mechanisms are required to account for the hyperalgesic effect of phenoxybenzamine. Phenoxybenzamine blocks prejunctional autoregulatory α2-adrenoceptors and inhibits the uptake of noradrenaline into adrenergic nerve terminals and non-neuronal tissue (Hoffman and Lefkowitz, 1996). Since noradrenaline increases thermal hyperalgesia in capsaicin-treated skin (Drummond, 1995), an increase in extracellular levels of noradrenaline at the site of phenoxybenzamine administration might have contributed to hyperalgesia during the neurogenic inflammation provoked subsequently by capsaicin. Alternatively, phenoxybenzamine may have blocked an inhibitory adrenergic influence on nociceptor discharge (Lindgren et al., 1987; Matran et al., 1989; Fuder and Selbach, 1993; Gentili et al., 1996). This inhibitory influence was investigated by Khasar et al. (1995), who reported that the α2-agonist clonidine inhibited hyperalgesia provoked by the intradermal injection of prostaglandin E2 into the rat paw. The inhibitory effect was reversed by the α2-antagonist rauwolscine. Since prostaglandin E2 produces hyperalgesia by acting directly on nociceptor afferents, Khasar et al. (1995) suggested that activation of α2-adrenoceptors directly inhibits nociceptor discharge. Less direct mechanisms involving α-adrenoceptors might also inhibit nociceptor discharge. For example, Nakamura European Journal of Pain (2001), 5
P.D. D RUMMOND
and Ferreira (1988) reported that the local analgesic effect of clonidine in inflamed rat paws was blocked by the opioid antagonist naloxone, implying that activation of α2-adrenoceptors leads to opioid release. Hypoalgesic influence of phenoxybenzamine When applied after the capsaicin treatment, phenoxybenzamine inhibited thermal hyperalgesia at normal ambient temperatures and during body cooling and warming, indicating that phenoxybenzamine blocked a nociceptive mechanism. In addition to blocking α-adrenoceptors, high concentrations of phenoxybenzamine block receptors for serotonin, acetylcholine and histamine (Hoffman and Lefkowitz, 1996). Since serotonin and histamine are nociceptive mediators, blockade of these receptors could account for the hypoalgesic effect of phenoxybenzamine. However, at the dose used in the present study, phenoxybenzamine blocks thermal hyperalgesia to tyramine (Drummond, 1998a), a substance that selectively releases noradrenaline from sympathetic nerve terminals (Hoffman & Lefkowitz, 1996). Thus, it seems likely that phenoxybenzamine blocked an adrenergic mechanism of thermal hyperalgesia. This mechanism may involve α-adrenoceptors on primary afferent nociceptors (Ouseph and Levine, 1995; Hong and Abbott, 1996) or on non-neural tissue. For example, α-adrenergic mechanisms influence the release of nerve growth factor, a potent nociceptive mediator, from vascular smooth muscle (Tuttle et al., 1993), and also influence the release of nitric oxide from the vascular endothelium (Hu et al., 1994). Another possibility is that loss of sympathetic vasoconstrictor tone due to α-adrenergic blockade facilitated the washout of inflammatory mediators. However, this explanation does not account for the increase in thermal hyperalgesia during body warming, particularly in association with increases in blood flow in the capsaicin-treated skin. The blockade of a nociceptive mechanism by phenoxybenzamine is consistent with the findings of Kinnman et al. (1997) and Liu et al.
α-A DRENERGIC
65
INFLUENCE ON THERMAL HYPERALGESIA
(1996), but is at odds with those of Pedersen et al. (1997) who reported that lumbar sympathetic block had no effect on thermal hyperalgesia in heat-injured skin. Although sympathetic ganglion blockade prevents the local release of noradrenaline, circulating catecholamines might still have exerted an adrenergic influence on nociceptors in Pedersen’s study. It is also worth noting that capsaicin rather than heat injury was used as a nociceptive stimulus in each of the studies that identified an α-adrenergic influence on nociception. Finger temperature and blood flow through the capsaicin-treated skin dropped substantially during body cooling, indicating that this was an effective stimulus for cutaneous vasoconstriction. Although heat–pain thresholds did not change in untreated skin during body cooling, thermal hyperalgesia intensified in capsaicin-treated skin. Since body cooling does not affect hyperalgesia to mechanical stimulation in capsaicin-treated skin (Baron et al., 1999), the effect of cooling may be specific to sensitized thermal nociceptors. Phenoxybenzamine administered after the capsaicin treatment did not fully inhibit thermal hyperalgesia during body cooling, suggesting that hyperalgesia was mediated in part by a nonadrenergic mechanism. One possibility is the release of nociceptive substances in addition to noradrenaline (e.g. neuropeptide Y, ATP or prostaglandins) from sympathetic vasoconstrictor nerve terminals (Levine et al., 1986; Gonzales et al., 1991; Burnstock & Wood, 1996; Tracey et al., 1995). Different effects of phenoxybenzamine administered before and after capsaicin Tight junctions between perineurial cells and additional tight junctions between endothelial cells in the vasa nervorum act as a barrier to the diffusion of substances from the extracellular fluid into nerve fascicles (Olsson, 1990; Kiernan, 1996). However, this barrier weakens during the inflammation provoked by capsaicin (Zochodne & Ho, 1993). This property of capsaicin was used previously to investigate whether the blood– nerve diffusion barrier modifies adrenergic influences on hyperalgesia (Drummond, 1998a). Both in the present and a previous study
(Drummond, 1998a), phenoxybenzamine administered after capsaicin inhibited hyperalgesia whereas phenoxybenzamine administered before capsaicin did not (at least at normal ambient temperatures). This implies that the blood–nerve diffusion barrier prevented phenoxybenzamine from blocking an adrenergic influence on nociception when administered before capsaicin. When administered before the capsaicin treatment, phenoxybenzamine intensified thermal hyperalgesia at normal ambient temperatures and during body warming. Paradoxically, however, the hyperalgesia effect of phenoxyb enzamine switched to an inhibitory influence during body cooling. That is, phenoxybenzamine apparently blocked a nociceptive mechanism or enhanced an analgesic mechanism that developed during cooling. Since phenoxybenzamine administered before capsaicin treatment did not inhibit the adrenergic hyperalgesia provoked by tyramine (Drummond, 1998a), this finding was unexpected and warrants further investigation. However, a hypothetical explanation of the effect is presented in Fig. 3.
Implications of the present findings The existence of an α-adrenergic influence on nociception in capsaicin-treated skin was questioned recently by Baron et al. (1999), who reported that sensitivity to mechanical stimulation in skin sensitized by capsaicin was unaffected by body heating or cooling. However, Baron et al. (1999) did not employ an antagonist to investigate α-adrenergic influences on nociception, did not distinguish between central and local influences on nociception and did not investigate thermal hyperalgesia. Elam et al. (1999) also raised doubts about the existence of an α-adrenergic influence on nociception in healthy subjects, because nociceptor discharge did not change during a brief period of sympathetic arousal (1 min). However, the present findings suggest that a longer period of sympathetic arousal may influence nociceptor discharge in inflamed skin, possibly by altering the balance between inhibitory and excitatory α-adrenergic influences on nociception (Fig. 3). European Journal of Pain (2001), 5
66
P.D. D RUMMOND
Phenoxybenzamine administered before capsaicin Phenoxybenzamine
After capsaicin and during body warming
pain
During body cooling
pain
Phenoxybenzamine administered after capsaicin After phenoxybenzamine and During body Capsaicin during body warming cooling
Rowbotham, 1997). The present findings suggest that an α-adrenergic mechanism that inhibits nociception may normally counteract direct adrenergic activation of primary afferent nociceptors. Disruption to this inhibitory influence after nerve or tissue injury might contribute to the adrenergic component of pain and hyperalgesia in some patients.
ACKNOWLEDGEMENTS
pain
pain
blood–nerve barrier intact;
blood-nerve barrier disrupted;
α2 receptor;
α1 receptor;
noradrenaline;
Phenoxybenzamine.
FIG. 3 Hypothesized distribution and effects of alpha-(α) adrenergic receptors on primary afferent nociceptors. It is postulated that nociceptive adrenoceptors are usually inactive because the blood–nerve barrier (represented by the solid box) limits the availability of agonists such as noradrenaline within nerve fascicles. When administered before capsaicin, the blood–nerve barrier also prevents phenoxybenzamine from blocking excitatory α1-adrenoceptors but not inhibitory α2-adrenoceptors at the exposed tips of nociceptive afferents (represented in the figure by a receptor spanning the blood–nerve barrier). Phenoxybenzamine might also increase extracellular levels of noradrenaline by blocking prejunctional autoregulatory α2-adrenoceptors and inhibiting the uptake of noradrenaline. Application of capsaicin disrupts the blood–nerve barrier (represented by the dotted boxes). Now, noradrenaline can increase thermal hyperalgesia via excitatory α1-adrenoceptors on nociceptive afferents; however, further release of noradrenaline during body cooling cancels out this effect via inhibitory α2-adrenoceptors. When administered after capsaicin, disruption of the blood–nerve barrier allows phenoxybenzamine to block inhibitory and excitatory α-adrenoceptors, resulting in a reduction of thermal hyperalgesia.
The existence of opposing α-adrenergic influences on nociception has implications for the mechanism of pain in conditions with a sympathetic component (Torebjörk et al., 1995; Choi and European Journal of Pain (2001), 5
This study was supported by the National Health and Medical Research Council of Australia, the Australian Research Council and Medtronic Australasia. I gratefully acknowledge the technical assistance of Ms Nadene Friday.
REFERENCES Ali Z, Raja SN, Wesselmann U, Fuchs PN, Meyer RA, Campbell JN. Intradermal injection of norepinephrine evokes pain in patients with sympathetically maintained pain. Pain 2000; 88: 161–168. Baron R, Wasner G, Borgstedt R, Hastedt E, Schulte H, Binder A, Kopper F, Rowbotham M, Levine JD, Fields HL. Effect of sympathetic activity on capsaicin-evoked pain, hyperalgesia, and vasodilatation. Neurology 1999; 52: 923–932. Burnstock G, Wood JN. Purinergic receptors: their role in nociception and primary afferent neurotransmission. Curr Opin Neurobiol 1996; 6: 526–532. Cesare P, Moriondo A, Vellani V, McNaughton PA. Ion channels gated by heat. Proc Natl Acad Sci USA 1999; 96: 7658–7663. Choi B, Rowbotham MC. Effect of adrenergic receptor activation on post-herpetic neuralgia and sensory disturbances. Pain 1997; 69: 55–63. Davis KD, Treede RD, Raja SN, Meyer RA, Campbell JN. Topical application of clonidine relieves hyperalgesia in patients with sympathetically maintained pain. Pain 1991; 47: 309–317. Drummond PD. Noradrenaline increases hyperalgesia to heat in skin sensitized by capsaicin. Pain 1995; 60: 311–315. Drummond PD. Independent effects of ischaemia and noradrenaline on thermal hyperalgesia in capsaicin-treated skin. Pain 1996; 67: 129–133. Drummond PD. Enhancement of thermal hyperalgesia by α-adrenoceptors in capsaicin-treated skin. J Auton Nerv Syst 1998a; 69: 96–102. Drummond PD. The effect of noradrenaline, angiotensin II and vasopressin on blood flow and sensitivity to heat in capsaicin-treated skin. Clin Auton Res 1998b; 8: 87–93. Drummond PD. Nitroprusside inhibits thermal hyperalgesia induced by noradrenaline in capsaicin-treated skin. Pain 1999; 80: 405–412.
α-A DRENERGIC
INFLUENCE ON THERMAL HYPERALGESIA
Elam M, Olausson B, Skarphedinsson JO, Wallin BG. Does sympathetic nerve discharge affect the firing of polymodal C-fibre afferents in humans? Brain 1999; 122: 2237–2244. Fuchs PN, Meyer RA, Raja SN. Heat, but not mechanical hyperalgesia, following adrenergic injections in normal human skin. Pain. 2001; 90: 15–23. Fuder H, Selbach M. Characterization of sensory neurotransmission and its inhibition via α2B-adrenoceptors and via non-α2-receptors in rabbit iris. NaunynSchmiedeberg’s Arch Pharmacol 1993; 347: 394–401. Gentili M, Juhel A, Bonnet F. Peripheral analgesic effect of intra-articular clonidine. Pain 1996; 64: 593–596. Gonzales R, Sherbourne CD, Goldyne ME, Levine JD. Noradrenaline-induced prostaglandin production by sympathetic postganglionic neurons is mediated by α2adrenergic receptors. J Neurochem 1991; 57: 1145–1150. Hoffman BB, Lefkowitz RJ. Catecholamines, sympathomimetic drugs, and adrenergic receptor antagonists. In: Hardman JG, Limbird LE, Molinoff PB, Ruddon RW, Gilman AG, editors. Goodman & Gilman’s The Pharmacological Basis of Therapeutics, 9th Edition. New York: McGraw-Hill, 1996; 199–248. Hong Y, Abbott FV. Contribution of peripheral α1Aadrenoceptors to pain induced by formalin or α-methyl5-hydroxytryptamine plus noradrenaline. Eur J Pharmacol 1996; 301: 41–48. Hu ZW, Miller JW, Hoffman BB. Induction of enhanced release of endothelium-derived relaxing factor after prolonged exposure to α-adrenergic agonists: role in desensitization of smooth muscle contraction. J Cardiovasc Pharmacol 1994; 23: 337–343. Hu S, Zhu J. Sympathetic facilitation of sustained discharges of polymodal nociceptors. Pain 1989; 38: 85–90. Janssen SA, Arntz A, Bouts S. Anxiety and pain: epinephrine-induced hyperalgesia and attentional influences. Pain 1998; 76: 309–316. Khasar SG, Green PG, Chou B, Levine JD. Peripheral nociceptive effects of a2-adrenergic receptor agonists in the rat. Neurosci 1995; 66: 427–432. Kiernan JA. Vascular permeability in the peripheral autonomic and somatic nervous systems: controversial aspects and comparisons with the blood-brain barrier. Microsci Res Tech 1996; 35: 122–136. Kindgen-Milles D, Holthusen H. Nociception from blood vessels is independent of the sympathetic nervous system under physiological conditions in humans. Eur J Pharmacol 1997; 328: 41–44. Kinnman E, Nygårds EB, Hansson P. Peripheral α-adrenoreceptors are involved in the development of capsaicin induced ongoing and stimulus evoked pain in humans. Pain 1997; 69: 79–85. Lee DH, Liu X, Kim HT, Chung K, Chung JM. Receptor subtype mediating the adrenergic sensitivity of pain behavior and ectopic discharges in neuropathic Lewis rats. J Neurophysiol 1999; 81: 2226–2233.
67
Levine JD, Taiwo YO, Collins SD, Tam JK. Noradrenaline hyperalgesia is mediated through interaction with sympathetic postganglionic neurone terminals rather than activation of primary afferent nociceptors. Nature 1986; 323: 158–160. Lindgren BR, Anderson CD, Frödin T, Andersson RGG. Inhibitory effects of clonidine on allergic reactions in guinea pig skin. Eur J Pharmacol 1987; 134: 339–343. Liu M, Max MB, Parada S, Rowan JS, Bennett GJ. The sympathetic nervous system contributes to capsaicinevoked mechanical allodynia but not pinprick hyperalgesia in humans. J Neurosci 1996; 16: 7331–7335. Matran R, Martling CR, Lundberg JM. Inhibition of cholinergic and non-adrenergic, non-cholinergic bronchoconstriction in the guinea pig mediated by neuropeptide Y and α2-adrenoceptors and opiate receptors. Eur J Pharmacol 1989; 163: 15–23. Nakamura M, Ferreira SH. Peripheral analgesic action of clonidine: mediation by release of endogenous enkephalin-like substances. Eur J Pharmacol 1988; 146: 223–228. Olsson Y. Microenvironment of the peripheral nervous system under normal and pathological conditions. Crit Rev Neurobiol 1990; 5: 265–311. Ouseph AK, Levine JD. α1-adrenoceptor-mediated sympathetically dependent mechanical hyperalgesia in the rat. Eur J Pharmacol 1995; 273: 107–112. Pedersen JL, Rung GW, Kehlet H. Effect of sympathetic nerve block on acute inflammatory pain and hyperalgesia. Anesthesiol 1997; 86: 293–301. Sato J, Perl ER. Adrenergic excitation of cutaneous pain receptors induced by peripheral nerve injury. Science 1991; 251: 1608–1610. Sato J, Suzuki S, Iseki T, Kumazawa T. Adrenergic excitation of cutaneous nociceptors in chronically inflamed rats. Neurosci Lett 1993; 164: 225–228. Sato J, Suzuki S, Tamura R, Kumazawa T. Norepinephrine excitation of cutaneous nociceptors in adjuvant-induced inflamed rats does not depend on sympathetic neurons. Neurosci Lett 1994; 177: 135–138. Torebjörk E, Wahren LK, Wallin G, Hallin R, Koltzenburg M. Noradrenaline-evoked pain in neuralgia. Pain 1995; 63: 11–20. Tracey DJ, Romm MA, Yao NNL. Peripheral hyperalgesia in experimental neuropathy: exacerbation by neuropeptide Y. Brain Res 1995; 669: 245–254. Tuttle JB, Etheridge R, Creedon DJ. Receptor-mediated stimulation and inhibition of nerve growth factor secretion by vascular smooth muscle. Exp Cell Res 1993; 208: 350–361. Winter J, Bevan S, Campbell EA. Capsaicin and pain mechanisms. Br J Anaesth 1995; 75: 157–168. Zochodne DW, Ho LT. Evidence that capsaicin hyperaemia of rat sciatic vasa nervorum is local, opiate sensitive and involves mast cells. J Physiol 1993; 468: 325–333.
European Journal of Pain (2001), 5
The effect of sympathetic activity on thermal hyperalgesia in capsaicin-treated skin during body cooling and warming Peter D. Drummond School of Psychology, Murdoch University, Western Australia
An adrenergic mechanism is thought to contribute to pain in conditions that sometimes develop during chronic inflammation and after nerve or tissue injury. There is some doubt, however, about whether adrenergic activity influences nociception in acute inflammation. To investigate this issue, the noncompetitive alpha-(α) adrenergic antagonist phenoxybenzamine was introduced by iontophoresis into the skin of 16 healthy volunteers either before or after the topical application of capsaicin. When applied before capsaicin, phenoxybenzamine increased thermal hyperalgesia at normal ambient temperatures and during body warming. These findings suggest that phenoxybenzamine blocked an analgesic mechanism when applied before the onset of inflammation. However, this effect disappeared during body cooling. When applied after capsaicin, phenoxybenzamine inhibited thermal hyperalgesia at normal ambient temperatures, and during body warming and cooling. Thus, phenoxybenzamine blocked a hyperalgesic mechanism when applied after the onset of inflammation. It was concluded that the presence of inflammation influences the nociceptive effect of α-adrenergic blockage, possibly by increasing access to excitatory adrenergic receptors on nociceptive afferents. An excitatory adrenergic influence on nociception may overcome an inhibitory adrenergic influence during inflammation. © 2001 European Federation of Chapters of the International Association for the Study of Pain KEYWORDS: thermal hyperalgesia, capsaicin, α-adrenoreceptors, inflammation.
INTRODUCTION An adrenergic mechanism contributes to chronic pain in conditions that sometimes develop after nerve and tissue injury (Torebjork et al., 1995; Choi and Rowbotham, 1997). Investigation of this mechanism in animal models of peripheral nerve injury has shown that sensitivity to electrical stimulation of the sympathetic trunk and to subcutaneous injection of noradrenaline develops in C-fibre polymodal nociceptors that survive partial nerve section (Sato & Perl, 1991). Conversely, pain behaviours and ectopic discharges in the injured Paper received 4 August 2000 and accepted for publication 11 December 2000. Correspondence to: Dr Peter Drummond, School of Psychology, Murdoch University, 6150, Western Australia. Fax: 61-8-93606492; E-mail: [email protected]
afferents of animals with neuropathic pain are inhibited by alpha-(α) adrenergic antagonists (Lee et al., 1999). Sensitivity to noradrenaline also develops in cutaneous C-fibre polymodal nociceptors in inflamed tissue (Hu & Zhu, 1989; Sato et al., 1993; Sato et al., 1994), suggesting that the inflammatory response to nerve and tissue injury might activate an adrenergic mechanism of pain. In humans, noradrenaline introduced into symptomatic skin sometimes increases pain and mechanical hyperalgesia in patients with complex regional pain syndrome (Davis et al., 1991; Torebjörk et al., 1995; Ali et al., 2000) and post-herpetic neuralgia (Choi and Rowbotham, 1997). Noradrenaline also increases thermal hyperalgesia in some patients with injured peripheral nerves (Choi and Rowbotham, 1997), and evokes a similar effect in the inflamed skin of healthy subjects. In particular, the intradermal administration of noradrenaline lowers the
1090-3801/01/010059 + 09 $35.00/0 © 2001 European Federation of Chapters of the International Association for the Study of Pain
60
heat–pain threshold in skin sensitized by capsaicin (Drummond, 1995; Drummond, 1998a; Drummond, 1998b; Drummond, 1999) and, to a lesser extent, in untreated skin (Drummond, 1996; Fuchs et al., 2001). Intravenous infusion of adrenaline also lowers the heat-pain threshold in capsaicin-treated and untreated skin (Janssen et al., 1998). Although the intradermal administration of noradrenaline clearly influences thermal hyperalgesia in inflamed skin, there is some debate about whether the normal release of endogenous stores of noradrenaline from sympathetic nerve endings increases nociceptor activity. For example, Pedersen et al. (1997) reported that lumbar sympathetic nerve block did not alter thermal or mechanical hyperalgesia in heat-injured skin in the leg. Kindgen-Milles and Holthusen (1997) found that neither noradrenaline nor the αadrenergic antagonist phentolamine influenced pain evoked by intravenous electrical stimulation or intravenous injection of hyperosmolar saline. Baron et al. (1999) reported that sensitivity to mechanical stimulation in skin sensitized by capsaicin was unaffected by body heating or cooling. In a direct test of the peripheral endogenous adrenergic influence on nociception, Elam et al. (1999) inserted microelectrodes into the peroneal nerve to measure nerve traffic in polymodal Cfibre afferents and sympathetic efferents. Activity increased in sympathetic efferents during the arousal provoked by mental stress (serial subtractions for 1 min) but did not change in nociceptor afferents, even after the afferents had been sensitized by mustard oil and brought close to threshold by repetitive electrical stimulation. In contrast to these negative findings, Kinnman et al. (1997) reported that subcutaneous injection of phentolamine inhibited ongoing pain and mechanical hyperalgesia around the site of intradermal capsaicin injection. In a similar study by Liu et al. (1996), intravenous injection of phentolamine inhibited the development of mechanical allodynia around the site of intradermal capsaicin injection. Finally, Drummond (1998a) reported that thermal hyperalgesia increased in capsaicintreated skin after the iontophoresis of tyramine, a substance that release endogenous stores of noradrenaline. Hyperalgesia to tyramine was European Journal of Pain (2001), 5
P.D. D RUMMOND
blocked by pretreatment with phenoxybenzamine, a noncompetitive α-adrenergic antagonist. Thus, it remains possible that the normal release of noradrenaline during sympathetic neural discharge influences the responsiveness of cutaneous nociceptors in some circumstances. The aim of the present study was to further explore the conditions in which sympathetic activity might influence nociceptor responsiveness. Since the pharmacological release of noradrenaline increases thermal hyperalgesia in the capsaicin-treated skin of healthy volunteers (Drummond, 1998a), it was hypothesized that an increase in sympathetic activity during body cooling would increase thermal hyperalgesia. To investigate the adrenergic component of nociception, phenoxybenzamine was introduced locally by iontophoresis into the capsaicin-treated skin. If endogenously released noradrenaline sensitizes nociceptors or increases their rate of discharge, thermal hyperalgesia should be lower at the site of α-adrenergic blockade than elsewhere in the capsaicin-treated skin.
METHOD Subjects The sample consisted of eight male and eight female university students aged between 17 and 40 years (mean age 21.6 years) who were in good health. Each subject provided informed consent for the procedures, which were approved by the Murdoch University Human Research Ethics committee. Procedures The experiments were run in a temperaturecontrolled laboratory, maintained at 24.1 ± 1.7˚C ( ± SD).
Application of capsaicin Capsaicin powder (Sigma, St Louis, MI, USA) was dissolved in 50% ethanol in distilled water at a concentration of 0.02 M (0.6%). The skin on the back of the hand or the volar aspect of the
α-A DRENERGIC
61
INFLUENCE ON THERMAL HYPERALGESIA
forearm near the elbow was cleaned with soap and water and an alcohol swab. If necessary, hair was shaved from the site with a disposable razor to facilitate the attachment of dressings and iontophoresis capsules to the skin. Care was taken not to touch the skin with the razor. The gauze pad of an elastic dressing (40 mm by 25 mm), containing 400 µl of the capsaicin solution, was then applied to the prepared skin. The dressing was covered by plastic tape to retard evaporation of the capsaicin solution. Thirty minutes later the dressing was removed and the treated skin was washed with soap and water.
Iontophoresis Phenoxybenzamine hydrochloride (ICN Pharmaceuticals, Costa Mesa, CA, USA) was prepared daily at 0.5 mM with distilled water from a 10 mM stock solution in distilled water. A capsule (internal diameter 0.8 cm) was attached to the site of application of capsaicin with an adhesive washer. The capsule was filled with the phenoxybenzamine solution, and a weak direct current (50 µA) was passed through the solution for 10 min to introduce the positively charged phenoxybenzamine ions into the skin. Iontophoresis was used in preference to intradermal injection to minimize systemic effects and to ensure that the phenoxybenzamine was distributed locally throughout the epidermis and upper layers of the dermis around the terminals of cutaneous nerves. The ground electrode was a silver plate measuring 3 cm by 5 cm, covered in electrode paste and attached to the volar aspect of the forearm near the wrist. In the first session, phenoxybenzamine was iontophoresed before the topical application of capsaicin (eight subjects), and shortly after the capsaicin had been washed from the skin (eight subjects). In the second session, phenoxybenzamine was iontophoresed before the topical application of capsaicin in five subjects, and shortly after the capsaicin had been washed from the skin in 11 subjects.
over light clothing to the subject’s chest and back. In addition, the subject’s face, trunk and legs were sprayed periodically with cold water while air was circulated around the subject with an electric fan. Care was taken to avoid directly cooling experimental sites on the forearm and hand.
Body warming (Second session.) Since the aim of this procedure was to minimize sympathetic vasoconstriction, body warming was always carried out in the second session when psychological responses to the laboratory setting were likely to be minimal. Approximately 1 h after the capsaicin had been washed from the skin, subjects were warmed gently with a convection heater placed behind their chair for 20 min. Experimental sites on the forearm and hand were not warmed directly.
Vascular responses The temperature of the fingertips was measured with a calibrated thermocouple before and after body cooling and warming. In addition, changes in blood flow in the capsaicin-treated skin were monitored with a wide surface area probe attached to a Moor MBF3D laser Doppler flowmeter (Moor Instruments, Axminster, UK). The probe consists of a central near-infrared laser beam emitter (wavelength 810 nm) surrounded by eight glass fibres, equally spaced on a 2-mm diameter circle, which collect the reflected light. The collected light was integrated by the MBF3D monitor to give an average flux over a surface area of approximately 7 mm2 and a depth of 1–2 mm. Blood flow was sampled once per second, and was later averaged using Moorsoft software for several minutes before and after cooling and warming. Because the flowmeter detected relative changes in blood flow, the change in flow after cooling or warming was expressed as the percent change from the level at baseline.
Body cooling (First session.) Approximately 1 h after the capsaicin had been washed from the skin, freezer packs (25 cm by 35 cm) were applied for 20 min
Heat–pain thresholds The radiant heat from a halogen globe was focused through a 6-mm diameter aperture European Journal of Pain (2001), 5
62
Statistical analyses Thermal hyperalgesia was defined as the difference between the heat–pain threshold in untreated skin before body cooling or warming and the heat–pain threshold at experimental sites in the capsaicin-treated skin. These scores were investigated in analyses of variance (SPSS for Windows, Version 8.1), with one betweensubjects factor of Timing of the Phenoxybenzamine Iontophoresis (phenoxybenzamine administered before versus after capsaicin) and repeated measures factors for site (the site pretreated with phenoxybenzamine versus the control sites) and time (before and after body cooling or warming). Results are presented as the mean ± SE.
RESULTS Body cooling Finger temperature decreased from 28.6 ± 1.0˚C to 22.3 ± 0.4 ˚C during body cooling [t(15) = 8.34, p < 0.001]. Blood flow decreased substantially through the capsaicin-treated skin of 14 of the 16 subjects, and remained close to baseline in the other two subjects [mean decrease 57 ± 8%, t(15) = 7.07, p < 0.001]. The heat–pain threshold in skin not treated with capsaicin averaged 44.0 ± 0.9˚C before cooling, and 43.6 ± 1.0˚C after European Journal of Pain (2001), 5
6
Phenoxybenzamine after capsaicin
*
4 2
*
0 −2 −4
Before cooling
After cooling
Control sites
Thermal hyperalgesia (°C)
Phenoxybenzamine before capsaicin Thermal hyperalgesia (°C)
placed just above the skin. Skin temperature was monitored with a thermocouple that touched the skin in the centre of the aperture. To assess the heat-pain threshold, skin temperature was brought to 32˚C for 10–15 s and then increased at 0.5˚C per second until the subject signalled pain by switching the lamp off or to a maximum of 49˚C. The heat–pain threshold was measured at the site of phenoxybenzamine iontophoresis and at three control sites in the capsaicin-treated skin, before and after body cooling and warming. The heat–pain threshold at each site was calculated as the average threshold from two or three temperature ramps. Sensitivity to heat–pain was also measured twice at two sites on the untreated forearm or hand, before and after body cooling and warming.
P.D. D RUMMOND
6 4 2 0 −2 −4
* Before cooling
After cooling
Phenoxybenzamine site
FIG 1. Effect of alpha-(α) adrenergic blockade on thermal hyperalgesia before and after 20 min of body cooling. Thermal hyperalgesia was defined as the difference between the heat–pain threshold in untreated skin before body cooling (representing the subject’s usual heat–pain threshold) and heat–pain thresholds in the capsaicin-treated skin. Phenoxy benzamine was administered before capsaicin in eight subjects, and after capsaicin in the other eight subjects. Difference between the phenoxybenzamine site (■) and the control sites (■ ■) statistically significant (* p < 0.05). Bars represent standard errors.
20 min of cooling (difference not significant). In contrast, the heat–pain threshold at control sites in the capsaicin-treated skin decreased from 42.6 ± 0.8˚C to 41.3 ± 0.9˚C after cooling [t(15) = 2.24, p < 0.05]. The effect of α-adrenergic blockade on thermal hyperalgesia depended on whether phenoxybenzamine was applied before or after the application of capsaicin (Fig. 1). The interaction between timing of the phenoxybenzamine iontophoresis, site and time was highly significant [F(1,14) = 47.8, p < 0.001], indicating that body cooling as well as the timing of the phenoxybenzamine iontophoresis influenced the effect of αadrenergic blockade on thermal hyperalgesia. When phenoxybenzamine was administered before capsaicin, thermal hyperalgesia was greater at the treated site than at the control sites before body cooling [t(7) = 2.68, p < 0.05]. However, hyperalgesia was greater at the control sites than at the treated site after body cooling [t(7) = 2.74, p < 0.05]. When phenoxybenzamine was administered after capsaicin, thermal hyperalgesia was greater at the control sites than at the treated site before body cooling [t(7) = 3.05, p < 0.05]. This persisted to some extent during body cooling (Fig. 1),
63
INFLUENCE ON THERMAL HYPERALGESIA
Body warming Finger temperature increased from 29.1 ± 1.0˚C before warming to 30.5 ± 0.7˚C after 20 min of warming (increase not significant). In the group as a whole, blood flow through the capsaicintreated skin did not change significantly during body warming (mean increase 8 ± 13%); however, blood flow increased 53 ± 18% in seven subjects and decreased 28 ± 6% in nine others. Heat–pain thresholds in skin not treated with capsaicin averaged 44.5 ± 0.5˚C before body warming and 44.3 ± 0.5˚C after 20 min of body warming (difference not significant). In contrast, heat–pain thresholds at control sites in the capsaicin-treated skin decreased from 41.0 ± 0.6˚C to 39.9 ± 0.7˚C during body warming [t(15) = 3.26, p < 0.01]. In the seven subjects with an increase in blood flow in the capsaicin-treated skin during body warming, the heat pain threshold decreased 1.9 ± 0.4˚C [t(6) = 5.42, p < 0.01]. In contrast, the heat pain threshold decreased only 0.5 ± 0.4˚C in the other nine subjects (mean decrease not significant). As in session 1, the effect of α-adrenergic blockade on thermal hyperalgesia depended on whether phenoxybenzamine was administered before or after capsaicin [interaction between timing of the phenoxybenzamine iontophoresis and site, F(1,14) = 31.0, p < 0.001] (Fig. 2). Body warming did not influence the effect of αadrenergic blockade on thermal hyperalgesia. When phenoxybenzamine was administered before capsaicin, thermal hyperalgesia was greater at the treated site than at control sites in the capsaicin-treated skin [mean heat–pain threshold 1.6 ± 0.5˚C lower at the phenoxybenzamine site than at the control sites, t(4) = 3.08, p < 0.05]. In contrast, when phenoxybenzamine was administered after capsaicin, thermal hyperalgesia was greater at control sites in the capsaicin-treated skin than at the phenoxybenzamine site [mean heat pain threshold 2.0 ± 0.4˚C higher at the phenoxybenzamine site than at the control sites, t(10) = 5.50, p < 0.001].
Phenoxybenzamine before capsaicin 8
*
6 4 2
Before warming
After warming
Control sites
Phenoxybenzamine after capsaicin Thermal hyperalgesia (°C)
although after 20 min of cooling the difference between the treated and control sites was no longer statistically significant.
Thermal hyperalgesia (°C)
α-A DRENERGIC
8 6 *
4 2
*
Before warming
After warming
Phenoxybenzamine site
FIG 2. Effect of alpha-(α) adrenergic blockade on sensitivity to heat–pain before and after 20 min of body warming. Thermal hyperalgesia was defined as the difference between the heat–pain threshold in untreated skin before body warming and heat–pain thresholds in the capsaicin-treated skin. Phenoxybenzamine was administered before capsaicin in five subjects, and after capsaicin in the other 11 subjects. Difference between the phenoxybenzamine site (■) and the control sites (■ ■) statistically significant (* p < 0.05). Bars represent standard errors.
DISCUSSION The main findings of the present study were as follows. Thermal hyperalgesia intensified at control sites in the capsaicin-treated skin during body cooling and heating. At normal ambient temperatures and during body warming, the effect of phenoxybenzamine depended on whether it was administered before or after the capsaicin treatment. When administered before the capsaicin treatment, phenoxybenzamine intensified thermal hyperalgesia at normal ambient temperatures and during body warming. However, when administered after the capsaicin treatment, phenoxybenzamine inhibited thermal hyperalgesia. Finally, when administered before the capsaicin treatment, phenoxybenzamine inhibited thermal hyperalgesia during body cooling. Hyperalgesic effect of phenoxybenzamine Capsaicin sensitizes polymodal nociceptors to heat (Winter et al., 1995), probably by acting on heat-sensitive ion channels (Cesare et al., 1999). Thermal hyperalgesia intensified in capsaicintreated skin during body warming, primarily in European Journal of Pain (2001), 5
64
association with increases in local blood flow. Thus, an increase in cutaneous temperature associated with an increase in blood flow may have contributed to hyperalgesia during body warming. Since release of vasoconstrictor tone by α-adrenergic blockade should also increase blood flow, a similar mechanism might have contributed to thermal hyperalgesia at sites of phenoxybenzamine administration. However, this does not fully account for the effect of phenoxybenzamine on thermal hyperalgesia, because hyperalgesia intensified only when phenoxybenzamine was administered before the capsaicin treatment. Furthermore, the hyperalgesic effect of phenoxybenzamine persisted during body warming, when sympathetic vasoconstrictor tone at control sites was likely to be minimal. Thus, additional mechanisms are required to account for the hyperalgesic effect of phenoxybenzamine. Phenoxybenzamine blocks prejunctional autoregulatory α2-adrenoceptors and inhibits the uptake of noradrenaline into adrenergic nerve terminals and non-neuronal tissue (Hoffman and Lefkowitz, 1996). Since noradrenaline increases thermal hyperalgesia in capsaicin-treated skin (Drummond, 1995), an increase in extracellular levels of noradrenaline at the site of phenoxybenzamine administration might have contributed to hyperalgesia during the neurogenic inflammation provoked subsequently by capsaicin. Alternatively, phenoxybenzamine may have blocked an inhibitory adrenergic influence on nociceptor discharge (Lindgren et al., 1987; Matran et al., 1989; Fuder and Selbach, 1993; Gentili et al., 1996). This inhibitory influence was investigated by Khasar et al. (1995), who reported that the α2-agonist clonidine inhibited hyperalgesia provoked by the intradermal injection of prostaglandin E2 into the rat paw. The inhibitory effect was reversed by the α2-antagonist rauwolscine. Since prostaglandin E2 produces hyperalgesia by acting directly on nociceptor afferents, Khasar et al. (1995) suggested that activation of α2-adrenoceptors directly inhibits nociceptor discharge. Less direct mechanisms involving α-adrenoceptors might also inhibit nociceptor discharge. For example, Nakamura European Journal of Pain (2001), 5
P.D. D RUMMOND
and Ferreira (1988) reported that the local analgesic effect of clonidine in inflamed rat paws was blocked by the opioid antagonist naloxone, implying that activation of α2-adrenoceptors leads to opioid release. Hypoalgesic influence of phenoxybenzamine When applied after the capsaicin treatment, phenoxybenzamine inhibited thermal hyperalgesia at normal ambient temperatures and during body cooling and warming, indicating that phenoxybenzamine blocked a nociceptive mechanism. In addition to blocking α-adrenoceptors, high concentrations of phenoxybenzamine block receptors for serotonin, acetylcholine and histamine (Hoffman and Lefkowitz, 1996). Since serotonin and histamine are nociceptive mediators, blockade of these receptors could account for the hypoalgesic effect of phenoxybenzamine. However, at the dose used in the present study, phenoxybenzamine blocks thermal hyperalgesia to tyramine (Drummond, 1998a), a substance that selectively releases noradrenaline from sympathetic nerve terminals (Hoffman & Lefkowitz, 1996). Thus, it seems likely that phenoxybenzamine blocked an adrenergic mechanism of thermal hyperalgesia. This mechanism may involve α-adrenoceptors on primary afferent nociceptors (Ouseph and Levine, 1995; Hong and Abbott, 1996) or on non-neural tissue. For example, α-adrenergic mechanisms influence the release of nerve growth factor, a potent nociceptive mediator, from vascular smooth muscle (Tuttle et al., 1993), and also influence the release of nitric oxide from the vascular endothelium (Hu et al., 1994). Another possibility is that loss of sympathetic vasoconstrictor tone due to α-adrenergic blockade facilitated the washout of inflammatory mediators. However, this explanation does not account for the increase in thermal hyperalgesia during body warming, particularly in association with increases in blood flow in the capsaicin-treated skin. The blockade of a nociceptive mechanism by phenoxybenzamine is consistent with the findings of Kinnman et al. (1997) and Liu et al.
α-A DRENERGIC
65
INFLUENCE ON THERMAL HYPERALGESIA
(1996), but is at odds with those of Pedersen et al. (1997) who reported that lumbar sympathetic block had no effect on thermal hyperalgesia in heat-injured skin. Although sympathetic ganglion blockade prevents the local release of noradrenaline, circulating catecholamines might still have exerted an adrenergic influence on nociceptors in Pedersen’s study. It is also worth noting that capsaicin rather than heat injury was used as a nociceptive stimulus in each of the studies that identified an α-adrenergic influence on nociception. Finger temperature and blood flow through the capsaicin-treated skin dropped substantially during body cooling, indicating that this was an effective stimulus for cutaneous vasoconstriction. Although heat–pain thresholds did not change in untreated skin during body cooling, thermal hyperalgesia intensified in capsaicin-treated skin. Since body cooling does not affect hyperalgesia to mechanical stimulation in capsaicin-treated skin (Baron et al., 1999), the effect of cooling may be specific to sensitized thermal nociceptors. Phenoxybenzamine administered after the capsaicin treatment did not fully inhibit thermal hyperalgesia during body cooling, suggesting that hyperalgesia was mediated in part by a nonadrenergic mechanism. One possibility is the release of nociceptive substances in addition to noradrenaline (e.g. neuropeptide Y, ATP or prostaglandins) from sympathetic vasoconstrictor nerve terminals (Levine et al., 1986; Gonzales et al., 1991; Burnstock & Wood, 1996; Tracey et al., 1995). Different effects of phenoxybenzamine administered before and after capsaicin Tight junctions between perineurial cells and additional tight junctions between endothelial cells in the vasa nervorum act as a barrier to the diffusion of substances from the extracellular fluid into nerve fascicles (Olsson, 1990; Kiernan, 1996). However, this barrier weakens during the inflammation provoked by capsaicin (Zochodne & Ho, 1993). This property of capsaicin was used previously to investigate whether the blood– nerve diffusion barrier modifies adrenergic influences on hyperalgesia (Drummond, 1998a). Both in the present and a previous study
(Drummond, 1998a), phenoxybenzamine administered after capsaicin inhibited hyperalgesia whereas phenoxybenzamine administered before capsaicin did not (at least at normal ambient temperatures). This implies that the blood–nerve diffusion barrier prevented phenoxybenzamine from blocking an adrenergic influence on nociception when administered before capsaicin. When administered before the capsaicin treatment, phenoxybenzamine intensified thermal hyperalgesia at normal ambient temperatures and during body warming. Paradoxically, however, the hyperalgesia effect of phenoxyb enzamine switched to an inhibitory influence during body cooling. That is, phenoxybenzamine apparently blocked a nociceptive mechanism or enhanced an analgesic mechanism that developed during cooling. Since phenoxybenzamine administered before capsaicin treatment did not inhibit the adrenergic hyperalgesia provoked by tyramine (Drummond, 1998a), this finding was unexpected and warrants further investigation. However, a hypothetical explanation of the effect is presented in Fig. 3.
Implications of the present findings The existence of an α-adrenergic influence on nociception in capsaicin-treated skin was questioned recently by Baron et al. (1999), who reported that sensitivity to mechanical stimulation in skin sensitized by capsaicin was unaffected by body heating or cooling. However, Baron et al. (1999) did not employ an antagonist to investigate α-adrenergic influences on nociception, did not distinguish between central and local influences on nociception and did not investigate thermal hyperalgesia. Elam et al. (1999) also raised doubts about the existence of an α-adrenergic influence on nociception in healthy subjects, because nociceptor discharge did not change during a brief period of sympathetic arousal (1 min). However, the present findings suggest that a longer period of sympathetic arousal may influence nociceptor discharge in inflamed skin, possibly by altering the balance between inhibitory and excitatory α-adrenergic influences on nociception (Fig. 3). European Journal of Pain (2001), 5
66
P.D. D RUMMOND
Phenoxybenzamine administered before capsaicin Phenoxybenzamine
After capsaicin and during body warming
pain
During body cooling
pain
Phenoxybenzamine administered after capsaicin After phenoxybenzamine and During body Capsaicin during body warming cooling
Rowbotham, 1997). The present findings suggest that an α-adrenergic mechanism that inhibits nociception may normally counteract direct adrenergic activation of primary afferent nociceptors. Disruption to this inhibitory influence after nerve or tissue injury might contribute to the adrenergic component of pain and hyperalgesia in some patients.
ACKNOWLEDGEMENTS
pain
pain
blood–nerve barrier intact;
blood-nerve barrier disrupted;
α2 receptor;
α1 receptor;
noradrenaline;
Phenoxybenzamine.
FIG. 3 Hypothesized distribution and effects of alpha-(α) adrenergic receptors on primary afferent nociceptors. It is postulated that nociceptive adrenoceptors are usually inactive because the blood–nerve barrier (represented by the solid box) limits the availability of agonists such as noradrenaline within nerve fascicles. When administered before capsaicin, the blood–nerve barrier also prevents phenoxybenzamine from blocking excitatory α1-adrenoceptors but not inhibitory α2-adrenoceptors at the exposed tips of nociceptive afferents (represented in the figure by a receptor spanning the blood–nerve barrier). Phenoxybenzamine might also increase extracellular levels of noradrenaline by blocking prejunctional autoregulatory α2-adrenoceptors and inhibiting the uptake of noradrenaline. Application of capsaicin disrupts the blood–nerve barrier (represented by the dotted boxes). Now, noradrenaline can increase thermal hyperalgesia via excitatory α1-adrenoceptors on nociceptive afferents; however, further release of noradrenaline during body cooling cancels out this effect via inhibitory α2-adrenoceptors. When administered after capsaicin, disruption of the blood–nerve barrier allows phenoxybenzamine to block inhibitory and excitatory α-adrenoceptors, resulting in a reduction of thermal hyperalgesia.
The existence of opposing α-adrenergic influences on nociception has implications for the mechanism of pain in conditions with a sympathetic component (Torebjörk et al., 1995; Choi and European Journal of Pain (2001), 5
This study was supported by the National Health and Medical Research Council of Australia, the Australian Research Council and Medtronic Australasia. I gratefully acknowledge the technical assistance of Ms Nadene Friday.
REFERENCES Ali Z, Raja SN, Wesselmann U, Fuchs PN, Meyer RA, Campbell JN. Intradermal injection of norepinephrine evokes pain in patients with sympathetically maintained pain. Pain 2000; 88: 161–168. Baron R, Wasner G, Borgstedt R, Hastedt E, Schulte H, Binder A, Kopper F, Rowbotham M, Levine JD, Fields HL. Effect of sympathetic activity on capsaicin-evoked pain, hyperalgesia, and vasodilatation. Neurology 1999; 52: 923–932. Burnstock G, Wood JN. Purinergic receptors: their role in nociception and primary afferent neurotransmission. Curr Opin Neurobiol 1996; 6: 526–532. Cesare P, Moriondo A, Vellani V, McNaughton PA. Ion channels gated by heat. Proc Natl Acad Sci USA 1999; 96: 7658–7663. Choi B, Rowbotham MC. Effect of adrenergic receptor activation on post-herpetic neuralgia and sensory disturbances. Pain 1997; 69: 55–63. Davis KD, Treede RD, Raja SN, Meyer RA, Campbell JN. Topical application of clonidine relieves hyperalgesia in patients with sympathetically maintained pain. Pain 1991; 47: 309–317. Drummond PD. Noradrenaline increases hyperalgesia to heat in skin sensitized by capsaicin. Pain 1995; 60: 311–315. Drummond PD. Independent effects of ischaemia and noradrenaline on thermal hyperalgesia in capsaicin-treated skin. Pain 1996; 67: 129–133. Drummond PD. Enhancement of thermal hyperalgesia by α-adrenoceptors in capsaicin-treated skin. J Auton Nerv Syst 1998a; 69: 96–102. Drummond PD. The effect of noradrenaline, angiotensin II and vasopressin on blood flow and sensitivity to heat in capsaicin-treated skin. Clin Auton Res 1998b; 8: 87–93. Drummond PD. Nitroprusside inhibits thermal hyperalgesia induced by noradrenaline in capsaicin-treated skin. Pain 1999; 80: 405–412.
α-A DRENERGIC
INFLUENCE ON THERMAL HYPERALGESIA
Elam M, Olausson B, Skarphedinsson JO, Wallin BG. Does sympathetic nerve discharge affect the firing of polymodal C-fibre afferents in humans? Brain 1999; 122: 2237–2244. Fuchs PN, Meyer RA, Raja SN. Heat, but not mechanical hyperalgesia, following adrenergic injections in normal human skin. Pain. 2001; 90: 15–23. Fuder H, Selbach M. Characterization of sensory neurotransmission and its inhibition via α2B-adrenoceptors and via non-α2-receptors in rabbit iris. NaunynSchmiedeberg’s Arch Pharmacol 1993; 347: 394–401. Gentili M, Juhel A, Bonnet F. Peripheral analgesic effect of intra-articular clonidine. Pain 1996; 64: 593–596. Gonzales R, Sherbourne CD, Goldyne ME, Levine JD. Noradrenaline-induced prostaglandin production by sympathetic postganglionic neurons is mediated by α2adrenergic receptors. J Neurochem 1991; 57: 1145–1150. Hoffman BB, Lefkowitz RJ. Catecholamines, sympathomimetic drugs, and adrenergic receptor antagonists. In: Hardman JG, Limbird LE, Molinoff PB, Ruddon RW, Gilman AG, editors. Goodman & Gilman’s The Pharmacological Basis of Therapeutics, 9th Edition. New York: McGraw-Hill, 1996; 199–248. Hong Y, Abbott FV. Contribution of peripheral α1Aadrenoceptors to pain induced by formalin or α-methyl5-hydroxytryptamine plus noradrenaline. Eur J Pharmacol 1996; 301: 41–48. Hu ZW, Miller JW, Hoffman BB. Induction of enhanced release of endothelium-derived relaxing factor after prolonged exposure to α-adrenergic agonists: role in desensitization of smooth muscle contraction. J Cardiovasc Pharmacol 1994; 23: 337–343. Hu S, Zhu J. Sympathetic facilitation of sustained discharges of polymodal nociceptors. Pain 1989; 38: 85–90. Janssen SA, Arntz A, Bouts S. Anxiety and pain: epinephrine-induced hyperalgesia and attentional influences. Pain 1998; 76: 309–316. Khasar SG, Green PG, Chou B, Levine JD. Peripheral nociceptive effects of a2-adrenergic receptor agonists in the rat. Neurosci 1995; 66: 427–432. Kiernan JA. Vascular permeability in the peripheral autonomic and somatic nervous systems: controversial aspects and comparisons with the blood-brain barrier. Microsci Res Tech 1996; 35: 122–136. Kindgen-Milles D, Holthusen H. Nociception from blood vessels is independent of the sympathetic nervous system under physiological conditions in humans. Eur J Pharmacol 1997; 328: 41–44. Kinnman E, Nygårds EB, Hansson P. Peripheral α-adrenoreceptors are involved in the development of capsaicin induced ongoing and stimulus evoked pain in humans. Pain 1997; 69: 79–85. Lee DH, Liu X, Kim HT, Chung K, Chung JM. Receptor subtype mediating the adrenergic sensitivity of pain behavior and ectopic discharges in neuropathic Lewis rats. J Neurophysiol 1999; 81: 2226–2233.
67
Levine JD, Taiwo YO, Collins SD, Tam JK. Noradrenaline hyperalgesia is mediated through interaction with sympathetic postganglionic neurone terminals rather than activation of primary afferent nociceptors. Nature 1986; 323: 158–160. Lindgren BR, Anderson CD, Frödin T, Andersson RGG. Inhibitory effects of clonidine on allergic reactions in guinea pig skin. Eur J Pharmacol 1987; 134: 339–343. Liu M, Max MB, Parada S, Rowan JS, Bennett GJ. The sympathetic nervous system contributes to capsaicinevoked mechanical allodynia but not pinprick hyperalgesia in humans. J Neurosci 1996; 16: 7331–7335. Matran R, Martling CR, Lundberg JM. Inhibition of cholinergic and non-adrenergic, non-cholinergic bronchoconstriction in the guinea pig mediated by neuropeptide Y and α2-adrenoceptors and opiate receptors. Eur J Pharmacol 1989; 163: 15–23. Nakamura M, Ferreira SH. Peripheral analgesic action of clonidine: mediation by release of endogenous enkephalin-like substances. Eur J Pharmacol 1988; 146: 223–228. Olsson Y. Microenvironment of the peripheral nervous system under normal and pathological conditions. Crit Rev Neurobiol 1990; 5: 265–311. Ouseph AK, Levine JD. α1-adrenoceptor-mediated sympathetically dependent mechanical hyperalgesia in the rat. Eur J Pharmacol 1995; 273: 107–112. Pedersen JL, Rung GW, Kehlet H. Effect of sympathetic nerve block on acute inflammatory pain and hyperalgesia. Anesthesiol 1997; 86: 293–301. Sato J, Perl ER. Adrenergic excitation of cutaneous pain receptors induced by peripheral nerve injury. Science 1991; 251: 1608–1610. Sato J, Suzuki S, Iseki T, Kumazawa T. Adrenergic excitation of cutaneous nociceptors in chronically inflamed rats. Neurosci Lett 1993; 164: 225–228. Sato J, Suzuki S, Tamura R, Kumazawa T. Norepinephrine excitation of cutaneous nociceptors in adjuvant-induced inflamed rats does not depend on sympathetic neurons. Neurosci Lett 1994; 177: 135–138. Torebjörk E, Wahren LK, Wallin G, Hallin R, Koltzenburg M. Noradrenaline-evoked pain in neuralgia. Pain 1995; 63: 11–20. Tracey DJ, Romm MA, Yao NNL. Peripheral hyperalgesia in experimental neuropathy: exacerbation by neuropeptide Y. Brain Res 1995; 669: 245–254. Tuttle JB, Etheridge R, Creedon DJ. Receptor-mediated stimulation and inhibition of nerve growth factor secretion by vascular smooth muscle. Exp Cell Res 1993; 208: 350–361. Winter J, Bevan S, Campbell EA. Capsaicin and pain mechanisms. Br J Anaesth 1995; 75: 157–168. Zochodne DW, Ho LT. Evidence that capsaicin hyperaemia of rat sciatic vasa nervorum is local, opiate sensitive and involves mast cells. J Physiol 1993; 468: 325–333.
European Journal of Pain (2001), 5