Altered heat pain thresholds and cerebral event-related potentials following painful CO2 laser stimulation in subjects with fibromyalgia syndrome

Pain, 58 (1994) 185-193

185

0 1994 Elsevier Science B.V. All rights reserved 0304-3959/94/$07.00

PAIN 2538

Altered heat pain thresholds and cerebral event-related potentials following painful CO, laser stimulation in subjects with fibromyalgia syndrome S.J. Gibson a National

a,*, G.O. Littlejohn

b, M.M. Gorman

a, R.D. Helme

a and G. Granges

b

Research Institute of Gerontology and Geriatric Medicine, North West Hospital, Mount Royal Campus, Parkville, Kctoria 3052 (Australia) and b Department of Medicine, Monash Medical Centre, Clayton, Victoria 3168 (Australia)

(Received 7 October 1992, revision received 4 June 1993, accepted 22 December 1994)

Summary A decrease in mechanical pressure pain thresholds, particularly over pre-designated tender points, is one of the defining characteristics of fibromyalgia syndrome (FS); however, changes in thermal pain sensitivity have not been investigated. The present study examined heat pain thresholds and cerebral event-related potentials following CO, laser stimulation in 10 subjects with FS and 10 age-matched control volunteers. The results indicate that patients with FS exhibit a significant reduction in heat pain threshold when tested on the dorsal surface of the hand. In accordance with previous research, we also found a decrease in mechanical pain threshold over pre-designated tender points and at control sites as well as a significantly larger mechanically induced neurogenic flare response. These measures were highly correlated with thermal pain threshold even though different anatomical sites were stimulated. Hence, it seems likely that FS patients display a multimodal change in pain sensitivity which is generalized rather than anatomically restricted. Patients with FS also displayed a significant increase in the peak-to-peak amplitude of the cerebral potential evoked by CO, laser stimulation at pain threshold intensity and 1.5 times pain threshold intensity. These findings suggest a greater activation of central nervous system (CNS) pathways following noxious input. Putative explanations for the increased CNS response are discussed, including mechanisms of peripheral nociceptor sensitization, altered CNS function and the role of psychological factors.

Key words: Fibromyalgia

syndrome; Pain threshold; Cerebral event-related

Introduction

Primary fibromyalgia syndrome (FS) is a chronic musculoskeletal pain condition with defined diagnostic criteria and a consistent pattern of symptoms and signs (Yunus et al. 1989; Wolfe et al. 1990; Granges and Littlejohn 1993). However, the aetiology and pathophysiology of this condition remains poorly understood (Boissevain and McCain 1991). Characteristic symptoms include diffuse aching and soreness, a non-re-

* Corresponding author; S.J. Gibson, National Research Institute of

Gerontology and Geriatric Medicine, North West Hospital, Mount Royal Campus, Poplar Road, Parkville, Victoria 3052, Australia. SSDI 0304-3959(94)00014-6

potential; CO, laser; Chronic pain

storative sleep pattern, morning stiffness and fatigue, non-dermatomal paraesthesia and subjective feelings of swelling (Quimby et al. 1988; Simms and Goldenberg 1988; Wolfe et al. 1989; Yunus et al. 1989). Affective disturbance, particularly high levels of anxiety, depression, hypochondriasis and hysteria have also been noted amongst some patients (Scudds et al. 1987; Goldenberg 1989). However, perhaps the most salient feature of this syndrome is a predictable anatomical pattern of tender points elicited by manual palpation (Yunus et al. 1989; Wolfe et al. 1990). In fact, the existence and number of such tender point sites now constitutes one of the major defining characteristics of FS and helps differentiate this syndrome from other closely related musculoskeletal pain conditions, such as

myofascial pain syndromes and articular dysfunction (see Simons 1988; McCain and Scudds 1988 for full discussion). The increase in mechanical pain sensitivity manifested clinically as tender point sites in FS patients has been examined recently using more quantitative methods. One study of tender point sites using a springloaded pressure algometer revealed a significantly lower pressure pain threshold in patients with FS (Wolfe et al. 1990). It has also been reported that, when compared to controls, FS patients have a lowered mechanical pain threshold at non-tender point sites (Scudds et al. 1987; Tunks et al. 19881, a lower tolerance for mechanically induced pain (Scudds 1987; Quimby et al. 1988) and a non-significant trend for reduced electrical pain thresholds (Scudds et al. 1987). Primary afferent C and AS fibres with polymodal nociceptors play a major role in the transmission of nociceptive stimuli to the CNS. Stimulation of these fibres also induces neurogenic inflammation, that is flare and plasma extravasation in skin. This flare response is thought to occur via an axon reflex mechanism involving the efferent release of substance P and CGRP from primary afferent terminals (White and Helme 1985; Magi and Meli 1988). We have previously shown an increased neurogenic flare response following noxious mechanical and chemical stimulation in FS patients as well as a reduced chemical threshold for flare induction (Littlejohn et al. 1987). The size of neurogenic flare was found to correlate with the number of tender points and, given the evidence of a decrease in mechanical pain threshold, it was suggested that many of the symptoms of FS, including pain, may result from a heightened activation of primary afferent fibres with polymodal nociceptors (Littlejohn et al. 1987). According to this hypothesis, any alteration in peripheral nociceptive mechanisms involving polymodal nociceptor function would be expected to affect the response to chemical, mechanical and thermal modalities of noxious stimulation. Although there is clear evidence of alterations in mechanical and irritant chemical pain thresholds, to the best of our knowledge no study has examined thermal pain thresholds in these patients. In addition to the alterations in peripheral nociceptive function, recent evidence indicates that CNS mechanisms might also play an important role in this syndrome. Although somewhat equivocal (see Biossevain and McCain 1991) abnormalities in the CNS metabolism of serotonin, tryptophan and catecholamine have been documented (Moldofsky and Warsh 1978; Russell et al. 1986). Increased concentrations of substance P in the CSF (Vaeroy et al. 1988) but not in plasma (Reynolds et al. 1988) have been reported and FS patients may exhibit abnormal interactions between substance P and CGRP (Vaeroy et al.

1989). As these neuropeptides are known to be involved in pain transmission these findings have been taken as support for the suggestion that FS might represent a disorder of pain perception or pain modulation (Smythe 1976; Vaeroy et al. 1989). It is also of interest that hyperacousis, decreased sound pain thresholds and vestibular hyper-reactivity have been described in some patients; this data supports a generalised disturbance in the CNS processing of sensory information (Gerster and Hadj-djilani 1984; Hadjdjilani and Gerster 1984). However, to date there has been no direct investigation of CNS processing of nociceptive input in patients with FS. One method which offers the potential to examine pain-related CNS processing involves the measurement of cerebral event-related potentials recorded in response to painful stimulation. Cerebral event-related potentials are stimulus-induced changes in the EEG and are thought to provide an index of CNS activation following an event. These electrical waveforms can be divided into early and late components which are denoted by their polarity (P = positive; N = negative) and latency after stimulus onset (Bromm and Treede 1991). The appearance of late component-evoked responses (i.e., 100-800 msec), such as the well-known P300, require experiments dealing with attentional, perceptual and cognitive processes (Prichard 1981). In contrast, the nociceptive-evoked response (NER), which is thought to reflect secondary processing of noxious input (i.e., magnitude and stimulus quality), can be elicited without the need for complex cognitive paradigms, presumably because the sensation of pain is compelling and requires immediate attention. Previous research has shown a strong relationship between the amplitude of the positive peak (P280-380), the intensity of noxious stimulation and subjective pain rating (Carmen et al. 1978, 1980; Gibson et al. 1991a). In addition, the NER has been shown to be a sensitive measure of analgesic efficacy (Hill et al. 1990). However, this waveform also appears to be influenced by factors other than supraspinal nociceptive input (Downman 1991). In common with measures of subjective appraisal, peak amplitude of the NER is known to vary as a function of the level of attention (Miltner et al. 1989), arousal (Bromm and Treede 1991) and anxiety (Gibson et al. 1991b). Hence, several investigators have suggested that the NER represents a physiological correlate of the global or integrated CNS processing which underlies the perception of pain (Carmen et al. 1980; Bromm and Treede 1991; Gibson et al. 1991a). The aim of the present study was to examine heat pain thresholds and the NER following noxious CO, laser thermal stimulation in patients suffering from FS. The severity of FS pain, the number of tender points, the neurogenic flare response and mechanical pain

187

sensitivity were also monitored and compared heat pain threshold and NER responses.

to the

Methods and materials Patients Eleven consecutive female outpatients attending a private rheumatology practice and fulfilling the criteria for primary fibromyalgia (Wolfe et al. 1990) were entered into the study. All patients had been attending the practice for at least 12 months. Only 2 patients were taking regular medication (simple analgesics and hypnotics) and these subjects were instructed to cease all medication at least 24 h prior to testing. Eleven age- and sex-matched control volunteers were also examined. All control volunteers were pain free at the time testing and had no previous history of musculoskeletal pain.

Methods On the day of testing all subjects underwent a clinical examination which included an assessment of the 18 pre-designated tender point sites used in the American College of Rheumatology (ACR) classification for FS (Wolfe et al. 1990). The assessment of tender point sites was made using a pressure algometer (Pain Diagnostics Thermography) calibrated in kg/cm’ (Fischer 1987). A point was considered tender if the subject indicated the presence of a painful sensation at pressures of less than 4 kg/cm*. A total myalgic score (TMS) was computed based on the sum of the pressure values to induce pain at the 18 tender points. Four points not designated as ACR tender point sites were used as control points, these were the right and left thumb nails and right and left mid-deltoid regions. The sum of the pressure pain thresholds at these sites was used to index mechanical pain sensitivity in control regions (CMS). A structured interview was conducted to elicit information on the characteristic symptoms of FS. Pain and fatigue were quantified on separate IO-cm visual analogue scales (VAS) in which the left extreme was labelled “no pain” or “not tired” and the right extreme was labelled with “the worst pain ever” or “extremely tired”. The constancy of pain was assessed on an ordinal scale ranging from “always present” to “less than once per month”. The number of painful regions (i.e., cervical, thoracic, lumbar, arms, legs) was also noted. Non-restorative sleep, subjective feelings of swelling and paraesthesia were coded as present or absent. Duration and severity of morning stiffness was coded on a 4-point ordinal scale. Reduced grip strength as assessed by sphygmomometer values of less than 300 mm Hg and skinfold tenderness on the mid-back dorsum, upper and lower limbs were also noted. The scores from all of these measures were then combined to provide a pseudo-interval scaling of symptom severity (see Granges and Littlejohn 1993). Finally, a mechanically induced neurogenie flare response was elicited by the application of firm pressure with a swab stick. The swab stick was positioned perpendicular to the skin, 3 cm lateral to the upper thoracic spine and then dragged over the skin in a downward direction for a distance of 10 cm and at a velocity of approximately 10 cm/set. The pressure applied to the swab stick was sufficient to induce a just noticeable scratch on the skin surface. This procedure was undertaken on both the right and left sides of the upper thoracic spine. Neurogenic flare size was quantified as the maximum width (mm) of induced skin flare 2 min after mechanical stimulation. The assessment of neurogenic flare was undertaken by a second investigator who was blinded to patient diagnosis and presenting clinical features. After obtaining informed consent subjects were seated in a comfortable chair in a temperature-controlled room shielded against electrical and acoustical interference. The subject remained alone in the test room during threshold and NER measurement procedures.

A closed-circuit television provided an uninterrupted view of the subject and a 2-way intercom system allowed for verbal communication throughout the session. A CO, laser (10.6 pm wavelength) situated in an adjoining room was used to deliver radiant heat pulses, 5 mm in diameter and of 33 msec duration. Previous research has shown that skin reflectance is less than 2% for radiation wavelengths of greater than 4 pm, irrespective of skin pigmentation, and the incident energy is almost completely absorbed within 50 pm skin depth (Biehl 1984). This temporal-spatial profile ensures that the laser will raise skin temperature to painful levels almost instantaneously and microneurographic studies have shown that high intensity CO, stimulation selectively activates A6 and C nociceptive fibres (but not AP fibres) in animals (Devor et al. 1982) and in man (Bromm and Treede 1991). It has also been argued that the contribution of non-nociceptive thermoreceptors to laser-evoked sensation is negligible (Bromm and Treede 1991), although it should be noted that at low-intensity laser stimulation tactile and warm sensations are sometimes reported (Pertovaara et al. 1984). This finding might suggest a concurrent activation of C warm and possibly AS mechanosensitive non-nociceptive primary afferent fibres during laser stimulation. The computer-controlled laser pulse was directed through a hollow tube into the subject test room and onto the dorsal surface of either the right or left hand. It should be noted that the hand dorsum is not one of the tender point sites used in the ACR classification of FS. All subjects were familiarized with the threshold determination procedures and exposed to the thermal stimulus at several intensities prior to formal testing. Detection and pain threshold were determined using a double random staircase procedure, which combines the up-down trials of the method of limits with the efficiency of the method of adjustment (see Gracely 1988 for a full description). Briefly, two independent staircase series are used: one for detection threshold and one for pain threshold. On each trial one of the two staircases is randomly chosen and the appropriate stimulus intensity for that staircase is presented. The response from the subject determines the next stimulus presented by that staircase the next time it is randomly selected, lowering the intensity for a positive response (i.e., “painful” or “detected”) and increasing it for a negative response (i.e., “not painful” or “not detected”). The procedure is repeated for a selected number of trials resulting in two ascending-descending staircase series which titrate the intensity required to elicit a report of just noticeable sensation or just noticeable pain. In the present study the starting intensity of stimulation for each staircase was determined by a short ascending series which was terminated when the subject first reported the presence of just noticeable sensation or just noticeable pain. Using the staircase procedure, detection and pain threshold were defined as the mean value of 4 consecutive 1 W up-down alternations in the intensity of laser stimulation. Subjects were always tested blind to stimulus onset and intensity. Following threshold determination, a block of 36 stimuli were presented with a randomised interstimulus interval of between 20 and 40 set and at psuedo-random intensities, either pain threshold or 1.5 pain threshold. Approximately 3 set after each laser stimulus, subjects were instructed to move their hand a few millimeters in order to minimize the effects of habituation and receptor fatigue. They were also asked to rate each stimulus using an 8-item word descriptor scale varying from “just noticeable” through to “excruciating”. A list of word descriptors developed by Bromm et al. (1984) was used to describe the quality of sensation. This list could be divided into words representing general tactile sensation (i.e., tingling, touching, warm) and noxious sensation (i.e., pricking stinging, burning). At all other times throughout the session, subjects were asked to move as little as possible. Following each CO, laser stimulus, EEG segments of 1000 msec duration were recorded from Ag/AgCI electrodes attached to the vertex (Cz) versus linked ears (AlA2) with forehead as ground (Fpz). lnfra- and supra-orbital

188 electrodes were used to monitor gross eye movements (EOGJ. Interelectrode impedance was reduced to less than 5 KR by abrading the skin surface. EEG samples were amplified (100 k), filtered with a OS-70 Hz bandpass, digitized with a sampling rate of 600 Hz and stored on computer for off-line analysis. Single trial NERs were averaged for each intensity of laser stimulation after eliminating sweeps contaminated with eye movement artifact. Measures of peak latency and peak-to-peak amplitude were calculated for each major waveform component within the averaged NER. A computer program automatically selected the major positive and negative peaks within the 100-600 msec poststimulus segment of the waveform. As the computer-derived calculation of latency and amplitude were occasionally inaccurate, for instance in cases of multiple peaks or lack of clear peak definition within the 100-600 msec window, all measures were checked visually and then adjusted if necessary. This method has been shown to have very good test-retest reliability (Gibson et al. 1991a). In common with the measurement of flare size, the determination of peak amplitude and latency was performed in a single blind manner.

Results

Eleven FS patients and 11 controls were originally recruited into the study. In 1 control and 1 FS subject the intensity of stimulation required for threshold report of pain exceeded our safety guidelines (i.e., > 40 W) for the use of CO, laser stimulation on exposed skin. As a result these subjects were excluded from all further testing and analysis. Descriptive information on the remaining 10 subjects from each group is presented in Table I. Analysis of this data using univariate t test revealed that the mean age and general mood state of FS patients was not significantly different from control volunteers. The mean duration of clinical pain was 74 + 14.5 months, with a VAS pain score of 3.8 and a corresponding rating of “mild” to “moderate” pain on the word descriptor scale. As expected the mean clinical symptom score CT,, = 7.68, P < 0.0000, the mean number of tender points CT,, = 5.42, P < 0.0001) and total myalgic score CT,, = 5.05, P < 0.0001) of FS patients was significantly different than controls. The pressure pain threshold score from control points was TABLE I DESCRIPTIVE AND CLINICAL INFORMATION FOR SUBJECTS WITH FS (n = 10) AND PAIN-FREE CONTROL VOLUNTEERS (n = 10) VALUES REFER TO MEAN (+ SEM)

Age VkS mood Pain duration Symptom score Number of TEPs Total myalgic score Control myalgic score Pain descriptor VAS pain

Fibromyalgia patients

Control subjects

P

28.3 (2.6) 3.6 (0.8) 74.4 (15.4) 10.8 (1.4) 13.4 (1.5) 58.3 (5.1) 19.8 (2.3) 4.3 (0.5) 3.8 (0.8)

26.6 (2.1) 3.8 (0.6) 0.0 (0.01 0.2 (0.1) 3.6 (1.0) 103.2 (6.7) 32.1 (2.9) 0.0 (0.0) 0.0 (0.0)

0.5601 0.7806 0.0001 0.0001 0.0001 0.0037 -

DETECTION

PAIN

THRESHOLD

THRESHOLD

Fig. 1. Mean (i SEM) detection and pain thresholds to thermal CO, laser stimulation in patients with FS (n = 20) and pain-free control volunteers (n = 20). * Differences between groups (P < 0.05).

also found to be significantly lower in FS patients (T,, = 3.19, P < 0.005). It should be noted that on the day of testing 2 FS patients had less than 11 tender points on our defined pressure algometry criteria (subject R.M., 8 TEPs; subject L.C., 6 TEPs). This contrasts with the minimum number of tender point sites required to satisfy a diagnosis of FS according to the criteria of Wolfe et al. (1990) when using manual palpation. However, these subjects were not rejected from the FS sample for the following reasons. The subjects had been attending a rheumatology practice for at least 12 months and had fulfilled the TEPs criteria on a number of previous occasions. On the day of testing both subjects were still suffering from diffuse musculoskeletal pain and detailed a number of characteristic symptoms (i.e., nonrestorative sleep, fatigue) and signs (i.e., skinfold tenderness, dermatographia). Finally the determination of TEPs was made using a pressure algometer rather than manual palpation as in the Wolfe (1990) criteria, In this respect it is of interest that both subjects exhibited an additional 5 TEPs sites in which the pressure pain threshold was between 4.0 and 4.5 kg/cm2. In order to examine the main effects for group (FS vs. control) and side of testing (left vs. right), the measures of neurogenic flare, threshold and NER were analysed cojointly using MANOVA. As a significant overall difference was observed between FS patients and controls
189 TABLE II MEAN (* SEM) NOCICEPTIVE-EVOKED RESPONSE AMPLITUDE, LATENCY, SUBJECTIVE RATING AND THE PERCENTAGE OF STIMULI CLASSIFIED AS NOXIOUS FOLLOWING PAIN THRESHOLD INTENSITY AND 1.5 TIMES PAIN THRESHOLD INTENSITY CO, LASER STIMULATION IN PATIENTS WITH FS (n = 20) AND CONTROL VOLUNTEERS (II = 20)

Pain threshold intensity Latency N270 (msec) Latency P370 (msec) NER amplitude (VI Subjective rating Stimuli rated as ‘pricking’ or ‘stinging’ 1.5 times pain threshold intensity Latency N270 (msec) Latency P370 (msec) NER amplitude (V) Subjective rating Stimuli rated as ‘pricking’ or ‘stinging’

Fibromyalgia patients

Control subjects

269.4 (6.2) 371.2 (6.7) 34.2 (2.6) 2.9 (0.22) 85.7% (5.7)

277.5 (6.5) 376.5 (6.1) 25.4 (1.6) * 2.8 (0.23) 89.9% (3.5)

267.2 (5.4) 364.9 (6.6) 50.5 (3.1) 4.9 (0.14) 89.5% (3.9)

272.5 (6.7) 368.7 (5.8) 31.4 (1.8) * 4.6 (0.14) 92.1% (2.8)

* Difference between groups (P < 0.0001).

0.0001). The mean (+ S.E.M) detection threshold and pain threshold values in response to CO, laser stimulation are presented in Fig. 1. Although detection threshold intensity was similar in both groups (F(,, 38)= 0.9, P < 0.6781, FS subjects exhibited a significant reduction in CO, laser pain threshold intensity when compared to controls (F(,, 38)= 5.617, P < 0.022). Hence, when

CONTROL eO

wit&

recording the NER following pain threshold and 1.5 times pain threshold laser stimulation, the patients with FS actually received a lower intensity of stimulation (pain threshold X= 12.5 W; 1.5 x threshold X; = 18.8 W> when compared to control volunteers (pain threshold (X = 18.7 W; 1.5 x threshold X = 28.0 WI. The grand mean NER in subjects with FS and in

CONTROL

SIJSJECTS

SUBJECTS

tr0

F~SROMYALGIA

FIBROMYALGIA SUBJECTS

w 8o

Rnplittic(uU)

i 30

Fig. 2. a: grand mean nociceptive-evoked response in patients with FS (n = 20) and control volunteers (n = 20) following pain threshold CO, laser stimulation (left panel). b: grand mean nociceptive-evoked response in patients with FS (n = 20) and control volunteers (n = 20) following 1.5 times pain threshold CO, laser stimulation (right panel).

190 TABLE

III

PEARSON CORRELATION THRESHOLD MEASURES SURES (n = 40)

Pain threshold NER amplitude Subjective rating Flare VAS pain VAS mood Symptoms

COEFFICIENTS IN FIBROMYALGIA

BETWEEN CLINICAL CHARACTERISTICS, AND CONTROL SUBJECTS COMBINING

VAS pain

VAS mood

Symptoms

TMS

- 0.25 0.41 ** 0.32 0.53 **

-0.15 0.29 0.26 0.19

- 0.33 0.64 ** 0.36 0.76 **

0.46 -0.60 -0.48 -0.86

_

-0.12 _ _

0.78 ** 0.09 _

NER, NEUROGENIC FLARE THE RIGHT AND LEFT SIDE

CMS

Flare

* ** * **

0.38 * ~ 0.43 * - 0.48 * - 0.75 **

- 0.46 * 0.55 ** 0.39 * _

-0.67 ** 0.17 -0.84 **

-- 0.43 * - 0.20 -0.63 **

0.53 ** 0.19 0.76 **

AND MEA-

* P
pain-free control volunteers following 1.5 times pain threshold CO, laser stimulation is presented in Fig. 2a,b, respectively. Regardless of intensity the NER was characterised by a small negative peak at approximately 270 msec (N270) post-stimulus and a high-amplitude positive peak at 370 msec (P370). The latency and peak-to-peak amplitude of the major waveform components following laser stimulation at pain threshold and 1.5 times pain threshold intensity are summarized in Table II. Also included in Table II are the percentages of CO, laser stimuli classified as stinging or pricking as well as the subjective rating of stimulus magnitude. Univariate analysis of the main effects for the NER data was undertaken using a 2-way ANOVA (Group x Intensity). Although measures of peak latency remained unchanged, the peak-to-peak amplitude of the NER was significantly increased in patients with FS (F,r, 3X) = 18.04, P < 0.0001) and, as expected, higher intensity stimulation (i.e., pain threshold vs. 1.5 times threshold) resulted in an increased peak-to-peak amplitude of the NER (F,, 3xj = 37.16, P < 0.0001). A significant Group x Intensity interaction effect was also observed (F,,, s8)= 8.12, P = 0.007), suggesting that whilst both groups displayed an increase in NER amplitude between pain threshold and 1.5 times pain threshold stimulation, the magnitude of this increase was greater in patients with FS (see Table II). With respect to the subjective appraisal of CO, laser stimulation, a univariate 2-way ANOVA revealed that both groups rated 1.5 times pain threshold stimulation as being more intense than pain threshold stimulation (F(, 3x)= 310.9, P < 0.0001) (see Table II). However, despite a very weak trend, there was no significant difference in the subjective estimate of stimulation magnitude between FS patients and control volunteers = 0.91, P = 0.347), and no interaction effect was W.38, observed (F,,, 3xj = 1.54, P = 0.102). This latter finding was not unexpected given that the intensity of CO, laser stimulation used for NER recording was adjusted

to accord with the pain threshold intensity of each individual. The choice of qualitative word descriptor was similar in both groups and across both intensities of stimulation with approximately 90% of laser stimuli being rated as pricking or stinging. The remaining stimuli were generally rated as touching or tingling. A bivariate Pearson correlation coefficient (combining data from both intensities) revealed a significant positive association between the subjective estimate of stimulus intensity and the peak-to-peak amplitude of the NER in both FS patients (r = 0.7178, P < 0.0001) and control volunteers (r = 0.6659, P < 0.0001). Pearson’s correlation coefficients were also used to explore the relationship between the clinical characteristics, flare and threshold measures as well as the subjective rating of stimulus intensity and NER amplitude following laser stimulation at 1.5 times pain threshold intensity. The results of this analysis are presented in Table III. As can be seen multiple significant correlations were observed. To summarize these relationships it appears that the severity of FS as indexed by the symptom score, VAS pain ratings and particularly the mechanical pain sensitivity (i.e., TMS and CMS), was highly correlated with the size of neurogenic flare response, the amplitude of the NER, and to a less extent, the subjective rating of stimulus magnitude. An increased flare size was also significantly associated with higher amplitude NERs and subjective estimate of stimulus magnitude as well as lower thermal pain thresholds. Finally, as might be expected, the intensity of clinical pain, symptom score and the myalgic score at tender points as well as control sites were significantly inter-related (see Table III>.

Discussion

The findings from the present study indicate that patients with FS exhibit a significant reduction in CO,

191

laser heat pain thresholds when compared to agematched pain-free control volunteers. These results complement previous findings of lowered mechanical pain threshold (Scudds et al. 1987; Tunks et al. 1988) and tolerance in FS patients (Scudds et al. 1987; Quimby et al. 1988) and suggests an increased pain sensitivity to thermal as well as mechanical stimulation. The change in thermal pain sensitivity was noted following stimulation of the dorsal surface of the hand, rather than over sites designated as “tender points”. Similarly, an overall lowering of pressure pain threshold was observed, with a significant reduction in mylagic score at both pre-designated sites and at control sites which included the right and left thumb nails and mid-deltoid regions. On the basis of these results it seems likely that FS patients display a generalised and multimodal change in cutaneous pain sensitivity, although given the recent findings of Wolfe et al. (19901, it remains likely that this change is more prominent at certain anatomical locations, where the degree of change in pressure pain threshold allows for the clinically useful sign of the “tender point” sites. The intensity of clinical pain and the Symptom Score Index, comprising self-reported levels of fatigue, stiffness, paraesthesia, sleep quality, grip strength and pain intensity, constancy and location(s), were highly correlated with the mechanical pressure pain threshold at the predesignated tender point sites. Mechanical pain threshold at control sites was also strongly related to the level of clinical pain and symptoms. In contrast, although there was some degree of association between thermal pain threshold and the clinical condition of the patient, this trend failed to reach statistical significance. Hence, despite the multimodal nature of change in pain sensitivity exhibited by FS patients, it appears that factors which contribute to the degree of mechanical pain sensitivity assume a greater degree of importance in the expression of the clinical pain condition. In a previous study we demonstrated an increased neurogenic flare response following noxious mechanical and chemical stimulation (Littlejohn et al. 1987). It was suggested that the exaggerated flare in FS patients may reflect an increased responsiveness of polymodal nociceptors on primary afferent C fibres. The present results add support to this view. Firstly, FS patients were again shown to exhibit a significant enhancement in the size of mechanically induced flare. Second, there is now clear evidence of a reduction in thermal and mechanical pain thresholds, thereby emphasising the polymodal nature of the change in cutaneous pain sensitivity. Finally, we observed a highly significant negative correlation between the size of neurogenic flare and the degree of mechanical and thermal pain sensitivity. In other words, the patients who exhibited the largest flare response also displayed the greatest reduction in CO, laser heat pain threshold and in the

pressure pain threshold over predesignated tender points and control sites. The apparent strength of this association strongly suggests some common underlying mechanism, such as an increased responsivity in polymodal nociceptor function. However, it should be noted that although the flare response can be initiated without the need for CNS involvement (Holzer 1988; Magi and Meli 1988>, alterations in sympathetic vasoconstrictor tone can lead to a relative change in the size of neurogenic flare (Hornyak et al. 19901. In this regard there have been reports of altered activity in the sympathetic nervous system in some patients with FS (Vaeroy et al. 1989) as well as reports of no difference (Elam et al. 1992; Vaeroy et al. 1989 cited in Vaeroy et al. 1989). More direct methods of evaluating primary afferent function, such as microneurographic recording techniques, should be undertaken in order to resolve this issue. The other major finding from the present study was the significant increase in the peak-to-peak amplitude of the cerebral NER in patients with FS. This increase was observed in response to noxious CO, laser stimuli delivered at pain threshold as well as stimuli delivered at 1.5 times pain threshold intensity. Consistent with previous studies (Carmon et al. 1978, 1980; Gibson et al. 1991b), higher-intensity laser stimulation was strongly associated with higher subjective estimates of pain magnitude and a corresponding increase in the amplitude of the NER, and this occurred in both4FS patients and in pain-free control volunteers. However, a significant Intensity x Group interaction was also found, suggesting that whilst FS patients exhibited elevated NER responses following pain threshold stimulation, there was an even greater increase relative to controls at suprathreshold intensity laser stimulation. It is of interest to note that at an equivalent intensity of CO, laser stimulation, which fortuitously corresponds to the pain threshold intensity of controls (18.7 W) and 1.5 times pain threshold intensity in FS patients (18.8 W), the amplitude of the NER was approximately double that of the control volunteers and FS patients rated the stimulus as “moderate” pain compared to a mean rating of “weak” pain in the control group. As the NER is thought to represent the integrated CNS processing which underlies the perception of pain (Carmen et al. 1980; Bromm and Treede 1991; Gibson et al. 1991a;) it seems likely that FS patients exhibit a marked increase in the CNS response to incoming nociceptive input. Given the suggestion of a heightened responsivity of primary afferent fibres with polymodal nociceptors, it is conceivable that the observed change in CNS processing may be entirely due to mechanisms of peripheral sensitization. The significant correlation between the amplitude of the NER and the size of neurogenic flare provides some support for this hypothesis. How-

ever, in recording the NER, care was taken to adjust stimulus intensity as a function of each individual’s pain threshold. Hence, for FS patients a lower level of stimulus intensity was used to generate the NER. As the NER was still increased in amplitude despite this adjustment it appears that there may be some change in CNS function which is over and above the changes due to the sensitization of peripheral primary afferent pathways. It has been suggested that FS might represent a disorder of pain modulation involving endogenous opioid systems (Smythe 1976) and the recent findings of an alteration in the CNS concentration of substance P and CGRP have been taken as support for this idea (Vaeroy et al. 1988, 1989). Any change in the function of inhibitory central mechanisms at the level of the spinal cord or mid-brain-brain stem could also result in an increased NER in FS patients. Another possible explanation for the increase in CNS response involves more generalised psychological influences. The level of arousal (Bromm and Treede 1991), attention (Miltner et al. 1989) and anxiety (Gibson et al. 1991b) is known to effect the amplitude of the NER. Increased levels of psychological disturbance, particularly with respect to anxiety, depression and hysteria have been previously noted in FS patients (Goldenberg 1989; Biossevain and McCain 1991). One could also argue that an alteration in thermal and mechanical pain threshold which is not site specific is more consistent with a change in CNS function rather than localized pathophysiology. In addition, it has been suggested that a decrease in pain threshold, such as seen in patients with FS, might reflect a state of hypervigilance to somatic signals resulting in an exaggerated focus on incoming sensations, including noxious input (Chapman 1978). In this respect, the demonstration of hyperacousis, decreased sound pain threshold and vestibular hyper-reactivity in FS patients (Gerster and Hadj-Djilani 1984; Hadj-Djilani and Gerster 1984) is of particular interest. When assessing the importance of psychological factors, it should be remembered that general mood state, as measured by a VAS, was not different between our controls and FS subjects, and did not correlate with the amplitude of the NER, the subjective rating of thermal stimuli or the level of mechanical and thermal pain sensitivity. The NER was recorded using standardised instructions, a familiarization protocol, a standard test paradigm and test environment in order to precisely control the level of arousal and attention throughout the session. Moreover, the detection threshold intensity of CO, laser stimulation was almost identical in control and FS subjects, yet variations in the level of arousal or attention might be expected to equally affect these responses. Thus, it would seem unlikely that these psychological influences play any major role in

altering the CNS response. It remains possible that more specific psychological states, such as hypervigilance with an exaggerated focus on painful somatic sensation, might contribute to the increased amplitude of the NER. Nonetheless, irrespective of whether peripheral and/or central factors contribute to the increase in NER amplitude, it is apparent that FS subjects display a much greater CNS activation in response to noxious CO, laser stimulation. Overall, the present study demonstrates that patients with FS exhibit an increased neurogenic flare and cerebral NER with an accompanying increase in mechanical and thermal pain sensitivity. These changes are clearly related to the severity of the clinical condition, as evidenced by the significant correlations between pressure pain thresholds, the size of neurogenic flare, the amplitude of the NER and the intensity of clinical pain and the number of characteristic symptoms. It is tempting to speculate that a change in the responsivity of peripheral primary afferent mechanisms as well as an alteration in CNS processing may play an important role in the pathophysiology of FS. That is, an increased afferent response and CNS activation following low intensity chemical, mechanical or thermal stimulation could be directly responsible for the pain and regional hyperalgesia experienced by these patients. However, in the absence of demonstrable tissue damage or injury, the cause of such changes in nociceptive function remains unknown and whether these changes are a primary factor in the etiology of FS or merely secondary consequences of this chronic pain disorder awaits further research.

References Biehl, R., Treede, R.D. and Bromm, B., Pain ratings and short radiant heat pulses. In: B. Bromm (Ed.), Pain Measurement in Man. Neurophysiological Correlates of Pain, Elsevier, Amsterdam, 1984, pp. 397-408. Boissevain, M.D. and McCain, G.A., Toward an integrated understanding of fibromyalgia syndrome. I. Medical and pathophysiological aspects, Pain, 45 (1991) 227-23X. Bromm, B. and Treede. R.D., Nerve fibre discharges, cerebral potentials and sensations induced by CO, laser stimulation, Hum. Neurobiol.. 3 (1984) 33-40. Bromm, B. and Treede, R.D.. Laser-evoked cerebral potentials in the assessment of cutaneous pain sensitivity in normal subjects and patients, Rev. Neurol. (Paris), 147 (1991) 625-643. Carmon, A., Dotan, Y. and Same, Y.. Correlations of subjective pain experience with cerebral evoked responses to noxious thermal stimulations, Exp. Brain Res., 33 (1978) 445-453. Carmon, A., Friedman, Y., Coger, R. and Kenton, B.. Single trial analysis of evoked potentials to noxious thermal stimulation in man, Pain, 8 (1980) 21-32. Chapman, C.R., The perception of noxious events. In: R.A. Sternbath (Ed.), The Psychology of Pain, Raven Press. New York, 1978, pp. 423-433. Devor. M., C’armon, A. and Frostig, R., Primary afferents and spinal sensory neurons that repond to brief pulses of intense infrared

193 laser radiation: a preliminary study in rats, Exp. Neurol., 76 (1982) 483-494. Downman, R., Spinal and supraspinal correlates of nociception in man, Pain, 45 (1991) 269-281. Elam, M., Johansson, G. and Wallin, B.G., Do patients with primary fibromyalgia have an altered muscle sympathetic nerve activity, Pain, 48 (1992) 371-375. Fischer, A.A., Pressure algometry over normal muscles, standard values, validity and reproducibility of pressure threshold, Pain, 30 (1987) 115-126. Gerster, J.C. and Hadj-djilani, A., Hearing and vestibular abnormalities in primary fibrositis syndrome, J. Rheumatol., 11 (1984) 678-680. Gibson, S.J., Gorman, M.M. and Helme, R.D., Assessment of pain in the elderly using event-related cerebral potentials. In: M.R. Bond, J.E. Charlton and C.J. Woolf (Eds.), Proc. VIth World Congress on Pain, Elsevier, Amsterdam, 1991a, pp. 523-529. Gibson, S.J., Le Vasseur, S.A. and Helme, R.D., Cerebral event-realted responses induced by CO, laser stimulation in subjects suffering from cetvico-brachial pain syndrome, Pain, 47 (1991b) 173182. Goldenberg, D.L., An overview of psychologic studies in fibromyalgia, J. Rheumatol., 16 (1989) 12-14. Gracely, R.H, Lota, L., Walter, D.J. and Dubner, R., A multiple random staircase method of psychophysical assessment, Pain, 32 (1988) 5.5-63. Granges, G. and Littlejohn, G.O., A comparative study of clinical signs in fibromyalgia/fibrositis syndrome, normal and exercising subjects, J. Rheumatol., 20 (1993) 344-351. Hadj-djilani, A. and Gerster, J.C., Meniere’s disease and fibrositis syndrome (psychogenic rheumatism): relationship in audiometric and nystagmographic results, Acta Octolaryngol. (Stockh.), 406 (1984) 67-71. Hill, H.F., Chapman. C.R., Saeger, L. et al., Steady-state infusions of opioids in human. II. Concentration-effect relationships and therapuetic margins, Pain, 43 (1990) 69-80. Holzer, P., Local effector functions of capsaicin-sensitive sensory nerve endings: invovlement of tachykinnins, CGRP and other neuropeptides, Neuroscience, 24 (1988) 739-768. Hornyak, M.E., Naver, H.K., Rydenhag, B. and Wallin, B.G., Sympathetic activity influences the vascular axon reflex in the skin, Acta Physiol. Scand., 139 (1990) 77-84. Littlejohn, G.O., Weinstein, C. and Helme, R.D., Increased neurogenie inflammation in fibrositis syndrome, J. Rheumatol., 14 (1987) 1022-1025. Magi, C.A. and Meli, A., The sensory-efferent function of capsaicinsensitive sensory neurons, Gen. Pharmacol., 19 (1988) l-43. McCain, G.A. and Scudds, R.A., The concept of primary fibromyalgia (fibrositis): clinical value, relation and significance to other chronic musculoskeletal pain syndromes, Pain, 33 (1988) 273-288. Miltner, W., Johnson, R., Braun, C. and Larbig, W., Somatosensory event related potentials to painful and non-painful stimuli: effects of attention, Pain, 38 (1989) 303-312.

Moldofsky, H. and Warsh, J.J., Plasma tryptophan and musculoskeletal pain in non-articular rheumatism, Pain, 5 (1978) 65-71. Pertovaara, A., Reinikainen, K. and Hari, R., The activation of unmyelinated or myelinated afferent fibres by brief infrared laser pulses varies with skin type, Brain Res., 307 (1984) 341-343. Prichard, W.S., Psychophysiology of P300, Psychol. Bull., 89 (1981) 506-540. Quimby, L.G., Block, S.R. and Gratwick, G.M., Fibromyalgia: generalized pain intolerance and manifold symptom reporting, J. Rheumatol., 15 (1988) 1264-1270. Reynolds, W.J., Chiu, B. and Inman, R.D., Plasma substance P levels in fibrositis, J. Rheumatol., 15 (1988) 1802-1803. Russell, I.J., Vipraio, G.A., Morgan, W.M. and Bowden, C.L., Is there a metabolic basis for the fibrositis syndrome?, Am. J. Med., 81 (1986) 50-54. Scudds, R.A., Rollman, G.B., Harth, M. and McCain, G.A., Pain perception and personality measures as discriminators in the classification of fibrositis, J. Rheumatol., 14 (1987) 563-569. Simms, R.W. and Goldenberg, D.L., Symptoms micking neurologic disorders in fibromyalgia syndrome, J. Rheumatol., I5 (1988) 1271-1273. Simons, D.G., Myofascial pain syndromes of the head, neck and low back. In: R. Dubner, G.F. Gebhart and M.R. Bond (Eds.1, Proc. Vth World Congress on Pain, Elsevier, Amsterdam, 1988, pp. 186-200. Smythe, H.A., Fibrositis as a disorder of pain modulation, Clin. Rheumat. Dis., 5 (1976) 823-832. Tunks, E., Crook, J., Norman, G. and Kalaher, S., Tender points in fibromyalgia, Pain, 34 (1988) 11-19. Vaeroy, H., Helle, R., Forre, O., Kass, E. and Terenius, L., Elevated CSF levels of substance P and high incidence of Reynaud phenomenon in patients with fibromyalgia: new features for diagnosis, Pain, 32 (19881 21-26. Vaeroy, H., Sakurada, T., Forre, O., Kass, E. and Terenius, L., Modulation of pain in fibromyalgia: cerebrospinal fluid (CSF) investigation of pain related neuropeptides with special reference to calcitonin gene related peptide (CGRP), J. Rheumatol., 16 (1989) 94-97. Vaeroy, H., Qiao, Z., Morkrid, L. and Forre, O., Altered sympathetic nervous system response in patients with fibroyalgia (fibrositis syndrome), J. Rheumatol., 16 (1989) 1460-1465. White, D.M. and Helme, R.D., Release of substance P from peripheral nerve terminals following stimulation of the sciatic nerve, Brain Res., 336 (1985) 15-27. Wolfe. F., Fibromyalgia: the clinical syndrome, Rheum. Dis. Clin. North Am., 15 (1989) I-18. Wolfe, F., Smythe, H.A., Yunus, M.B. et al., The American College of Rheumatology 1990 criteria for the classification of fibromyalgia, Arthr. Rheum., 33 (1990) 160-172. Yunus, M., Masi, A.T. and Aldag, J.C., A controlled study of Primary fibromyalgia syndrome: clinical features and association with other functional syndromes, J. Rheumatol., 16 (1989) 62-71.