Increased pain sensitivity in fibromyalgia: effects of... : PAIN (original) (raw)
1 Introduction
Fibromyalgia (FM) is defined on the basis of chronic widespread pain and generalized tenderness (Wolfe et al., 1990). In clinical practice, tenderness is typically assessed by performing a manual tender point examination. The American College of Rheumatology (ACR) criteria for FM requires a minimum of 11 (pain at pressure >4 kg) out of the 18 tender points. In research settings, tenderness is also assessed by a variety of mechanical dolorimeters (i.e. pressure algometer). These methods show that tenderness in FM is not limited to defined tender points, but instead extends throughout the entire body (Scudds et al., 1987; Quimby et al., 1988; Mikkelsson et al., 1992; Wolfe et al., 1995; Sorensen et al., 1998; Petzke et al., 2001).
The enhanced pain sensitivity observed in FM is not only confined to blunt pressure, but also includes increased sensitivity to electrical stimulation, radiant and cutaneous heat stimulation (Arroyo and Cohen, 1993; Gibson et al., 1994; Lautenbacher et al., 1994; Kosek et al., 1995; Dohrenbusch, 2001), and cold stimulation, which evokes a notable dysaesthetic component (Bengtsson et al., 1986; Kosek et al., 1996; Berglund et al., 2002). Although enhanced sensitivity to several stimulus modalities is consistent with a central neurobiological mechanism, this consistency could also be due to the manner in which pain sensitivity is assessed. Previous studies have usually evaluated pain sensitivity by ascending methods, which are particularly vulnerable to response biases (Gracely et al., 1979; Gracely and Kwilosz, 1988; Gracely and Naliboff, 1996; Petzke et al., 2003). These biases may result in reports of increased pain sensitivity due to anticipation or expectancy of a painful stimulation or as a result of generalized psychological hypervigilance (McDermid et al., 1996). Alternatively, the predictable nature of an ascending series may reduce anxiety and instill a sense of control, decreasing reported pain sensitivity.
There is evidence that the use of psychophysical direct scaling methods which control common biases also show increased pain sensitivity in FM patients, either to pressure applied to tender muscle (Bendtsen et al., 1997) or to heat applied to multiple cutaneous regions (Kosek et al., 1996). In addition, the multiple random staircase (MRS) method (Cornsweet, 1962; Gracely et al., 1988) and its variants have demonstrated increased pain sensitivity in FM (Gibson et al., 1995; Lautenbacher and Rollman, 1997).
This study assessed pain sensitivity to blunt pressure and contact heat in FM patients and age- and gender-matched controls. We hypothesized that both stimulus modalities would elicit increased pain sensitivity in the patient group, supporting a central mechanism of altered sensory processing in FM.
For both modalities, this study also compared the results from predictable ascending threshold and random suprathreshold paradigms to determine both the influence of ‘expectancy’ or ‘psychological hypervigilance’ and whether these effects would vary between the patient and control groups. The effects of sensitization were evaluated in the two groups by repeating the random testing of pressure stimuli after a defined interval and by assessing the time course of heat pain sensitivity in the MRS procedure.
2 Methods
2.1 Subjects
Patients under treatment at the Georgetown University Medical Center for an established diagnosis of FM were invited to participate in this study. Patients with concurrent inflammatory rheumatic conditions and severe other medical conditions were excluded.
Healthy controls (HC) were recruited through flyers and newspaper advertisements, compensated for their participation and matched by age and gender to the patient population. All participants read and signed an informed consent form. The informed consent and protocol were approved by the Georgetown University Institutional Review Board.
All female participants were screened for menstrual status on the day of pain testing. Subjects were asked about menstrual history, average cycle length, last menstrual period, and use of medications (birth control pill, hormone replacement therapy). An approximate cycle stage was determined if possible and grouped into the following phases: menstruation, follicular (including midcycle), luteal (including premenstrual), and postmenopausal (including perimenopausal women and women with hysterectomy).
2.2 Psychophysical testing—general procedure
Patients participating in pain testing were allowed to continue their regular medication; however, they were advised to avoid taking any opioid or non-opioid pain medication for 24 h prior to the testing session. HC were also advised to avoid taking any analgesic medication 24 h prior to the pain testing.
On arrival to the research center, all subjects were familiarized with the testing environment. During this time, the consent form was signed, a menstrual history was obtained, and subjects were asked about current medications and their intake over the previous 24 h. After establishing historical presence of chronic widespread pain, a manual tender point count was performed to confirm a current diagnosis of FM. Only patients who met ACR criteria were included in the analysis. HC with a history of any current or chronic pain for more than 1 week were excluded.
A brief demonstration of the pain testing equipment, including application of several pressure and thermal stimuli, served to familiarize subjects with the procedure. For each pain testing session, subjects received instructions from a standardized script. Pain intensity ratings were recorded on a 21-box combined numerical analog descriptor scale using standardized instructions (Sternberg et al., 1998; Eliav and Gracely, 1998). The sequence of the testing is shown in Table 1 and was identical for all subjects.
Sequence and selection of tests of pain sensitivity and respective outcome measures
2.2.1 Pressure pain testing
All subjects received a dolorimeter examination at 18 defined tender points and four control points (Wolfe et al., 1990). Pressure was increased at a rate of 1 kg/s using a 3.14 cm2 footplate size and subjects were instructed to indicate when they first perceived pain. Pressure was increased up to 12 kg if necessary. If this did not elicit a pain response, 12 kg was recorded as pain threshold.
Manual tender point examinations and dolorimetry can induce bias as some FM patients might expect to be tender in a ‘tender point’. These areas were avoided for more advanced psychophysical testing. In both random and ascending paradigms, discrete pressure stimuli were applied using a remote stimulation device to eliminate any direct examiner/subject interaction. This device employed a hydraulic system to apply pressure to the right or left thumbnail via a 1 cm2 hard rubber circular probe. The administrator activated the hydraulic system by placing calibrated weights on a moveable table and adjusting valves to control stimulus timing. The probe was encased in plastic housing, positioned over the subject's thumbnail, and lowered to apply pressure consistent with the weight on the moveable table. The combination of valves and calibrated weights produced controlled, repeatable stimulation that approached a rectangular waveform.
2.2.1.1 Discrete ascending series (ASC-P)
An ascending series of discrete stimuli, each of 5 s duration, was applied to the right thumbnail. Initial stimulation pressure was 0.45 kg; thereafter, stimulation pressures increased in 0.45 kg increments until either a subject's level of pain tolerance or a maximum of 9.1 kg was reached. Upon rating a stimulus with a pain intensity of 10 (mild to moderate pain), subjects were asked permission to continue after each subsequent stimulus. Inter-stimulus interval was 30 s.
2.2.1.2 Discrete random series (RAN-P)
Subjects were informed that they would receive a different series of stimuli within the range of the previous ascending series. Seven stimuli (0.45, 0.91, 1.36, 1.82, 2.73, 3.64, 4.54 kg) were presented twice, each in random order, to the left thumb and pain intensity ratings were recorded. Preliminary psychophysical testing in patients and controls indicated that this weight distribution should result in at least three values that would fall between pain threshold and tolerance in most subjects.
2.2.1.3 Repeat discrete random series (REPEAT-P)
After a 15-min pause in pressure stimulation, the discrete random series was repeated on the left thumb using a different stimulation sequence. For both random series, inter-stimulus interval was 30 s.
2.2.2 Heat pain testing
Heat pain testing was performed with a 1-cm diameter contact thermode system (Gracely, 1988). A low-mass electrical heater on a water-perfused cold sink with feedback circuitry delivered precise stimulus waveforms at rates exceeding 10 °C/s. Heat stimuli, each 3 s in duration, were applied to the left forearm at 20-s intervals, again avoiding tender points to reduce response bias. The volar surface of the forearm was divided into a grid of three columns by seven rows. Stimulation was started in the first row on the lateral column near the elbow. When all 21 locations were stimulated, testing restarted at the initial position, thus allowing 7 min between repeated stimulation at the same location. The probe was placed on the respective site about 5 s prior to stimulus application to adapt skin temperature to the baseline temperature.
2.2.2.1 Discrete ascending series (ASC-H)
A discrete ascending series of stimuli was applied to the left forearm starting from a baseline temperature of 32 °C. Initial stimulus intensity was 37 °C, and stimulus intensity was increased in 1 °C increments until either the subject's level of pain tolerance or a maximum of 51 °C was reached. Similar to the ascending pressure series, subjects were asked permission to continue with each subsequent stimulus after reaching a pain intensity rating of 10.
2.2.2.2 Discrete random series (MRS-H)
In this paradigm, the stimulus temperature was determined interactively by the MRS, whereby, a computer program continuously adjusted the stimulus temperatures in each staircase to produce the same response distribution in each subject (Gracely et al., 1988). Three staircase pairs were used to titrate stimulus temperatures to pain responses of 0.5 (none to faint pain), 9.5 (mild to moderate pain), and 13.5 (moderate to strong pain). The starting points for each pair were chosen from the discrete ascending series. The program switched between staircases in pseudorandomized order, with each of the six staircases delivering 12 stimuli (72 total). Baseline temperature was 32 °C, and the stimuli were delivered in a range between 35 and 51 °C.
2.3 Questionnaires
Clinical pain was assessed by the short form of the McGill pain questionnaire (SFMG), using the total score, as well as subscores for sensory and affective dimensions (Melzack, 1987). The present pain intensity (PPI) was recorded on a 6-point verbal rating scale. In addition, a visual analogue scale for current pain intensity (VAS) and the regional pain score (RPS) were used. The RPS, developed for patients with FM, is a 6-point (0–5) Likert-scale rating pain at 21 predefined sites (Lautenschlager et al., 1993; Finckh et al., 1998).
2.4 Statistical analysis
The major outcome measures of pain sensitivity are listed in Table 1. Analyses examined differences between groups, stimulus modalities and modes of stimulus presentation, and the interaction of these effects.
2.4.1 Threshold measures
Pain thresholds were determined specifically by dolorimetry and the MRS for heat (MRS-H: scale value 0.5). Two derived measures of pain threshold were calculated from the ascending and random data for pressure (ASC-P, RAN-P) and from the ascending data for heat stimuli (ASC-H): (1) a cumulative frequency of pain reports for increasing stimulus intensity and (2) the proportion of stimuli judged as painful over a predefined range (pressure 0.45–4.5 kg, heat 37–45 °C).
2.4.2 Suprathreshold measures
Two comparisons were performed for suprathreshold pressure pain sensitivity. First, the area under the curve (AUC) for the stimulus range (0.45–4.5 kg) was calculated for each individual subject in both groups for each of the three discrete stimulation paradigms (ASC-P, RAN-P, REPEAT-P) and then used for group comparisons. Since the curve was defined as a set of pairs of stimulation pressures (xi) and respective pain intensity ratings (yi), the AUC was calculated as follows: AUC=sum((y(i+1)+y(i))/2× x(i+1)−x(i)). The sum included all available data points in the given stimulus range. In subjects who could not tolerate the whole testing range, the highest pain rating was substituted for the respective missing values. A second analysis compared stimulus response functions in a common overlapping region of 2.7, 3.6, and 4.5 kg, which evoked pain in both groups. However, six FM patients could not tolerate these stimuli and were thus excluded from this analysis, leading to a more conservative estimate of pain sensitivity in these patients.
The methodological differences between the discrete ascending series (ASC-H) and the MRS-H used for the heat pain testing required transformation of the original data for direct comparison. Linear regression curves were fitted to each individual discrete ascending series, and stimulation temperature values for pain scale response levels used in the MRS-H (0.5, 9.5, and 13.5) were estimated. These values were also compared with similarly computed values from linear regressions for each of the six individual staircase results of the MRS paradigm.
2.4.3 Expectancy effects
Expectancy, for the purpose of this study, was defined as a greater pain sensitivity (i.e. greater pain report or lower pain threshold) in ascending than random paradigms (ASC>RAN). The opposite of expectancy, termed ‘control’, was defined as a lower pain sensitivity (i.e. smaller subjective pain report or a higher pain threshold) in ascending than random paradigms (ASC<RAN). Subjects were categorized accordingly for threshold and suprathreshold measures for pressure (percentage of painful responses and AUC) and heat stimuli (linear regression values for 0.5 and 9.5).
2.4.4 Sensitization
Comparing the difference in AUC between RAN-P and REPEAT-P assessed sensitization to painful pressure stimulation, which was defined as an increase in AUC with REPEAT-P. In contrast, desensitization was defined as a decrease in AUC. For painful heat, the time course of the applied heat stimuli over the MRS session was used to describe sensitization (decrease of temperatures to evoke the respective pain intensities [0.5, 9.5, 13.5]) or desensitization (increase of stimulus temperatures to evoke the respective pain intensities [0.5, 9.5, 13.5]).
2.4.5 Statistical procedures
Comparisons between group means were analyzed using Students _t_-test. Two-way ANOVA was used to evaluate between- and within-group comparisons. Stimulus response curves were compared using appropriate repeated measure ANOVA procedures for both between- and within-group comparisons. Pearson's correlation coefficients were used to evaluate relationships between variables. All data are shown as mean ±1 SEM unless stated otherwise. SPSS 9.0 and MS Excel were used for data analysis.
3 Results
3.1 Subjects
Fifty-three of 165 clinic patients agreed to participate in the study and were examined with the evoked-pain testing protocol. Forty-three patients fulfilled the ACR criteria for FM on the day of testing. Only their data are included in the analysis. A comparison of the selected and non-selected patients revealed significant differences: patients who did not meet ACR criteria reported less subjective pain and less tenderness (Table 2).
Average subject questionnaire responses, tender point counts, and dolorimetry pain threshold in FM patients meeting or not meeting the ACR criteria on the day of testing
The 43 selected FM patients were age- and gender-matched with the 28 HC (Table 3). The slight difference in age was not statistically significant (_p_=0.26). Distribution of menstrual stages in the two groups is shown in Table 3 and the difference between groups was not statistically significant.
Age in years, sex distribution in percentage, and menstrual status in percentage of female subjects in the FM and HC groups, respectively
3.2 Dolorimetry at tender points vs. thumbnails
Data on dolorimetry thresholds at tender points and both thumbnails are summarized in Table 4. As expected, all thresholds were significantly higher in HC. Within-group measures at the thumbs and tender points were highly inter-correlated, more so in the patient group, possibly due to range restriction in the controls.
Dolorimetry pain thresholds at tender points and both thumbnails: differences between the two groups and correlations within groups
3.3 Pain threshold
3.3.1 Pressure
FM patients had lower pressure pain thresholds and increased pain report in both the ascending and the random paradigms (Fig. 1A, B). Both groups reported more painful trials, and thus had a lower pain threshold in the random paradigm than the ascending paradigm (Fig. 1A, B). A 2×2 ANOVA of the percentage of painful trials showed a significant main effect for the ascending vs. random paradigms, but no interaction with the subject group (Fig. 1B).
Pressure pain thresholds: (A) cumulative pain report for ascending and random paradigms up to 4.54 kg/cm2 stimulation in FM patients and HC; (B) percentage of painful trials. A 2×2 ANOVA showed a significant main effect for paradigms used (ASC-P vs. RAN-P, _F_=17.8, p<0.0001) as well as for subject group (HC vs. FM, _F_=35.8, p<0.0001), but no interaction (_F_=1.21, _p_=0.27).
3.3.2 Heat
Heat pain thresholds derived from both the ascending (Fig. 2A, B) and MRS (Fig. 2C) paradigms were significantly lower in FM patients. To directly compare the results of the ascending vs. MRS heat pain thresholds, thresholds in the ascending paradigm were estimated with individual linear regression equations and computed for a pain scale value of 0.5. The respective heat pain thresholds in the ascending paradigm were 41.8±0.32 °C for HC and 40.4±0.23 °C for FM patients. Comparison with the MRS values (41.6±0.36 °C for HC and 39.7±0.27 °C for FM patients) using a 2×2 ANOVA revealed significant main effects for group (HC vs. FM, _F_=17.3, p<0.0001) and method (ASC vs. RAN, _F_=8.1, p<0.006), but no significant interaction.
Heat pain thresholds: (A) cumulative pain report for the ascending paradigm (ASC-H) up to 45 °C stimulation in FM patients and HC; (B) percentage of painful trials in the ascending paradigm (ASC-H) with FM patients reporting pain more often (p<0.006); (C) pain threshold in MRS-H paradigm. FM patients have a lower pain threshold (**p<0.0001).
3.4 Suprathreshold
3.4.1 Pressure
The results for the AUC are shown in Fig. 3A, demonstrating increased pressure pain sensitivity in FM patients in both ascending and random paradigms compared with HC. Again, the random paradigm led to greater mean ratings in both groups. Comparison of the ascending and random paradigms with a 2×2 ANOVA revealed no interaction, indicating a similar difference in response between the two groups. This lack of interaction is depicted clearly by the stimulus response graph shown in Fig. 3B. There is a striking parallel ‘shift to the left’ or ‘higher pain sensitivity’ in FM patients for both paradigms.
Pressure pain—suprathreshold: (A) AUC for both paradigms and groups. A 2×2 ANOVA revealed significant main effects for method (ASC-P vs. RAN-P, _F_=61.5, p<0.0001) and group (HC vs. FM, _F_=27.1, p<0.0001), but no interaction (_F_=0.84, _p_=0.36). (B) Stimulus response functions. FM subjects display a parallel shift to the left in both paradigms. A repeat measure ANOVA shows significant main effects for method (ASC-P vs. RAN-P, _F_=49.8, p<0.0001), stimulus intensity (_F_=127.8, p<0.0001), and group (FM vs. HC, _F_=16.2, p<0.0001), but no significant interactions.
3.4.2 Heat
Fig. 4 shows that the effects found for suprathreshold pressure pain sensitivity are also found for suprathreshold painful contact heat. FM patients, in comparison with HC, show a parallel shift to the left, indicating greater pain sensitivity to heat. Both groups produced significantly greater ratings in the random than ascending paradigms.
Heat pain—suprathreshold and stimulus response functions. Data were computed from individual linear regression functions; temperatures for pain scale values 0.5, 9.5, and 13.5 are shown. FM subjects again display a parallel shift to the left in both paradigms. A repeat measure ANOVA shows significant main effects for method (ASC-H vs. MRS-H, _F_=17.9, p<0.0001), stimulus intensity (_F_=261.7, p<0.0001), and group (FM vs. HC, _F_=15.0, p<0.0001), but no significant interactions (_p_=n.s.).
3.5 Expectancy effects
Table 5 shows the distribution of expectancy effects across the two groups for both heat and pressure modalities and for the threshold and suprathreshold measures. There is no significant difference for any of the measures, indicating that the increased sensitivity in FM is not due to an increased proportion of FM patients classified as expectant.
Expectancy effects (ASC>RAN) for both pressure and heat vs. control effects (ASC<RAN) for threshold and suprathreshold measures
3.6 Sensitization
3.6.1 Pressure
Fig. 5A shows a similar degree of sensitization for both groups, indicated by an increase in AUC in the retest paradigm (REPEAT-P). A 2×2 ANOVA revealed significant main effects for sensitization (RAN-P vs. REPEAT-P, _F_=13.0, p<0.001) and group (HC vs. FM, _F_=18.9, p<0.0001), but importantly, no interaction for group and sensitization (_F_=0.46, _p_=0.5). Fig. 5B demonstrates that although the sensitization effect is small in both groups (HC: 27.8±14.2%, vs. FM: 22.7±8.5%, _p_=0.74), it occurs throughout the entire range of the suprathreshold stimulus response curve.
Sensitization of pressure pain: (A) AUC for the random and retest paradigms and the two groups; (B) stimulus response functions for RAN-P and REPEAT-P paradigms. FM subjects and controls display a parallel shift to the left in the REPEAT-P paradigm.
3.6.2 Heat
In FM patients, the stimulus temperatures required to produce a fixed subjective response decreased over time for the lower staircase level that was titrated at pain threshold (PI 0.5). In contrast, the temperatures required to evoke suprathreshold sensations increased over the course of the session, indicating desensitization that may be due to factors such as adaptation, habituation, or receptor suppression. Despite the increased heat pain sensitivity at pain threshold and decreased suprathreshold sensitivity, there is no evident difference between groups in the time course of both threshold (PI 0.5) and suprathreshold (PI 9.5 and 13.5) stimulation temperatures (Fig. 6A–C).
Sensitization and desensitization of heat pain. Curves for FM patients generally reach the same subjective endpoint at lower temperatures; however, time course runs almost parallel for all three pain levels. (A) Time course for pain scale value 0.5. FM subjects have a lower pain threshold than HC (p<0.0001). (B) Time course for pain scale value 9.5. FM subjects have a lower endpoint than HC (p<0.003). (C) Time course for pain scale value 13.5. FM subjects have a lower endpoint than HC (p<0.003). The temporal response at pain threshold does not predict the suprathreshold time course; the pain threshold temperatures increase, indicating sensitization, while the suprathreshold temperatures decrease, indicating desensitization.
3.7 Correlations between pressure and heat sensitivities
Table 6 shows the results of inter-correlations for the different methods, modalities, and groups. Pressure pain sensitivity is expressed by the AUC for the ascending and random methods. Heat pain sensitivity is expressed by the AUC for the ascending method and by the mean of all staircase endpoints for the MRS paradigm. Except for the random pressure and ascending heat paradigms in HC, all correlations within the respective groups were significant, while the absolute mean of the correlation coefficients was similar for HC (0.58) and FM patients (0.53).
Correlation between pressure and heat pain measurements (AUC for random (RAN-P) and ascending pressures (ASC-P) and ascending heat (ASC-H) and mean MRS values) for patients with FM and HC
4 Discussion
Statistically, the FM patients appeared ‘psychophysically’ identical to the HC, except for a left-shift in their responsiveness to pressure and thermal stimuli. This finding supports growing evidence that the clinical features of FM result from a central augmentation of pain processing (Lorenz et al., 1996; Kosek and Hansson, 1997; Mountz et al., 1998; Staud et al., 2001; Gracely et al., 2002; Price et al., 2002). This augmentation is not likely due to psychological factors such as expectancy that might result in only semantic re-labeling of an otherwise unaltered pain experience.
We hypothesized, and found, that centrally mediated increases in pressure pain sensitivity in FM patients would be supported by parallel increases in heat pain sensitivity and that these effects would be observed with both ascending and random presentations of stimuli. Absence of increased thermal pain sensitivity would suggest that mechanical tenderness might be related to peripheral factors (i.e. affecting certain types of nociceptors or sensory nerves) and not to central augmentation of pain sensitivity. Similarly, an increase in pain sensitivity to any stimulus modality that is found only in the ascending paradigms would provide support for a cognitive and/or affective mechanism in which subjects respond relatively earlier due to response bias (e.g. expectancy) or hypervigilance (lower than expected pain threshold on ascending paradigms compared with controls) (Gracely and Naliboff, 1996; McDermid et al., 1996).
The results, however, did not completely support our original hypotheses. We anticipated that both groups of subjects would rate pain intensity lower in random than ascending paradigms. We found the opposite; the majority of individuals in each group rated pain higher on random than ascending paradigms. This suggests that factors that might increase sensitivity to experimental pain in random paradigms are more influential than factors that might decrease sensitivity (e.g. predictability or perceived increase in control). Despite this increased sensitivity, there is evidence that random measures are influenced less by psychological factors, such as an individual's level of distress. In a different study of HC, we showed that the same random paradigm was relatively immune to individuals' level of distress, while the number of positive tender points and, to a lesser degree, dolorimetry were clearly associated with measures of distress (Petzke et al., 2003). Previous population-based studies have similarly demonstrated that the number of positive tender points (an ascending measure with no control on the part of the subject) is more strongly associated with measures of distress than dolorimetry (performed at the same points, but ‘controllable’ because the stimulus is terminated when the subject states that their pain threshold has been reached) (Wolfe, 1997).
It is well documented that aversive stimuli that are less predictable are perceived as being more threatening. In comparison with predictable stimulation, unpredictable aversive stimulation increases ratings of pain unpleasantness and pain reactivity, interferes with learned and feeding behavior, and increases physiological responses to stress evaluated by measures such as blood pressure, immune responsiveness, dopamine secretion, and gastric ulceration (Lawler et al., 1993; Orsini et al., 2002; Price et al., 1980; Rollnik et al., 2001). Thus, it is possible that the random presentation of stimulus intensities increased perceived pain report in our study, while providing relative immunity to external bias from psychological factors. Alternatively, in the absence of a strong expectancy effect, the task demands of the ascending method might cause lowered responses. Subjects may simply choose a starting response and then increase their responses to subsequent stimuli without attending to each evoked sensation. In determining which result should be regarded as the ‘gold standard’, we propose that the random methods are preferable, due to the low association with psychological distress.
This study used psychophysical methodology to suggest that expectancy, response bias, or hypervigilance are not playing a prominent role in FM patients. In contrast, studies of irritable bowel syndrome (IBS) have observed a different effect. Naliboff et al. (1997) delivered rectal distension stimuli to IBS patients and controls using both an ascending series of discrete stimuli and a random threshold tracking task. In comparison with controls, discomfort thresholds in the patients were lower in the ascending series, but not in the tracking task. This result suggests that hypervigilance may play a prominent role in the perception and rating of noxious stimuli in this group of patients. In a second study, Chang et al. (2000) used the same methods to assess somatic perception in female IBS patients with and without concurrent FM. A series of continuously ascending stimuli (Method of Limits) and a randomized series of fixed stimuli were manually applied to both tender and control points. Both patients with IBS alone and IBS with FM were characterized as hypervigilant.
Thus, an individual's performance on ascending vs. random paradigms may depend on multiple factors. In the present study, for example, individuals could have been reacting to predictability of stimulus onset and intensity, perceived control over the stimulus (subjects could easily withdraw their thumb), and the lack of real danger in the testing paradigm (excessive stimulation of the thumb is not expected to lead to permanent injury). With rectal balloon testing, sensations are less predictable, the application of the stimulus is less controllable, and the procedure is perceived as possibly causing harm. Additionally, temporal characteristics of the pressure and rectal balloon stimuli used in the two studies were markedly different (phasic vs. tonic). Integration of these characteristics by the central nervous system may be another reason for the observed differences between FM and IBS patients.
Several issues with respect to methodology and interpretation of data deserve mention. First, we chose to apply pressure and thermal stimuli to regions of the body not considered to be ‘tender points’. One reason for this was pragmatism: we found it difficult to construct a device that would deliver consistent, discrete pressure to any of the typical tender points reliably. In contrast, the thumb could be easily positioned within a cylindrical object to ensure that a rubber probe would rest precisely over the thumbnail and that pressure would be applied reliably. We also avoided testing tender points to avoid bias associated with differences in group attitudes toward these locations. Tender point regions would likely have little meaning to control subjects but might have considerable psychological significance to the patients who have undergone repeated tender point testing. Because FM patients have been ‘taught’ that tender points are more tender, this knowledge could influence subjective reports of sensations evoked by stimuli applied to these points.
The validity of using the thumb for pressure testing is well supported by existing data. Studies have shown that tenderness in FM extends throughout the entire body and is similar in tender points and ‘control points’ (i.e. forehead and thumbnail) (Granges and Littlejohn, 1993; Lautenbacher and Rollman, 1997). Our group and others have shown that dolorimetry values at the thumbnail (or any 2–4 points throughout the body) are highly correlated with the average obtained from all 18 points suggesting that this is an appropriate measure of overall pressure pain sensitivity (Granges and Littlejohn, 1993; Petzke et al., 2001). The results of this study do not support continued use of the term ‘control points’, as the same group difference in pressure pain sensitivity commonly observed at tender points could be demonstrated at the thumbnail. For these reasons, Wolfe (1997) suggested using the term ‘high threshold tender points’ instead of ‘control points’.
For all subjects, evaluation of suprathreshold sensitivity to pressure and heat involved similar tasks. Subjects received stimuli with unpredictable magnitude at regular intervals and used the same scale to report subjective pain intensity. These methods differed in execution and analysis, and thus introduced a possible confound between the random scaling method and stimulus modality. Since the use of discrete rectangular stimuli has been applied to clinical pressure pain evaluation only recently, we chose to use the direct scaling method for suprathreshold pain assessment to provide stimulus/response functions that could serve as normative data in future studies. Since such normative data are readily available for painful contact heat, we chose to use the more recent MRS method for suprathreshold heat assessment. We assumed that these methods, employing similar tasks, would reveal similar effects. Our results confirmed similar findings for both measures, increasing confidence that the difference between these suprathreshold scaling measures and the ascending paradigm was related to robust, common features of the random tasks (random, unpredictable stimulus intensity, predictable onset, use of combined numerical/category scale) and not to peculiarities of the task execution.
It is also important to point out that although the FM patients behaved similarly to the controls with respect to psychophysical characteristics, both groups exhibited a large range of responses. Further analyses of the within-group differences might identify subsets of patients with different psychological contributions to their report of evoked pain, and that these subsets (e.g. identification of ‘expectant’ patients) might have clinical relevance. Although our data suggest that psychological factors (i.e. expectancy and hypervigilance) are not playing a major role in the report of evoked pain in FM, there are numerous other psychological factors that contribute to pain report.
Acknowledgements
This work was supported by Department of Army grant DAMD 17–002–0018 and F.P. was supported by a grant from the Deutsche Forschungsgemeinschaft (Pe 713/1–1).
References
Arroyo JF, Cohen ML. Abnormal responses to electrocutaneous stimulation in fibromyalgia. J Rheumatol. 1993;20:1925-1931.
Bendtsen L, Norregaard J, Jensen R, Olesen J. Evidence of qualitatively altered nociception in patients with fibromyalgia. Arthritis Rheum. 1997;40:98-102.
Bengtsson A, Henriksson KG, Jorfeldt L, Kagedal B, Lennmarken C, Lindstrom F. Primary fibromyalgia. A clinical and laboratory study of 55 patients. Scand J Rheumatol. 1986;15:340-347.
Berglund B, Harju EL, Kosek E, Lindblom U. Quantitative and qualitative perceptual analysis of cold dysesthesia and hyperalgesia in fibromyalgia. Pain. 2002;96:177-187.
Chang L, Mayer EA, Johnson T, FitzGerald LZ, Naliboff B. Differences in somatic perception in female patients with irritable bowel syndrome with and without fibromyalgia. Pain. 2000;84:297-307.
Cornsweet TN. The staircase methods in psycophysics. Am J Psychol. 1962;5:413-417.
Dohrenbusch R. Are patients with fibromyalgia ‘hypervigilant’? Schmerz. 2001;15:38-47.
Eliav E, Gracely RH. Sensory changes in the territory of the lingual and inferior alveolar nerves following lower third molar extraction. Pain. 1998;77:191-199.
Finckh A, Morabia A, Deluze C, Vischer T. Validation of questionnaire-based response criteria of treatment efficacy in the fibromyalgia syndrome. Arthritis Care Res. 1998;11:116-123.
Gibson SJ, Granges G, Littlejohn GO, Helme RD., 1995. Increased thermal pain sensitivity in patients with fibromyalgia syndrome. In: Bromm B, Desmedt JE, editors., Pain and the brain. Raven Press, New York, NY, pp. 401-411.
Gibson SJ, Littlejohn GO, Gorman MM, Helme RD, Granges G. Altered heat pain thresholds and cerebral event-related potentials following painful CO2 laser stimulation in subjects with fibromyalgia syndrome. Pain. 1994;58:185-193.
Gracely RH, Dubner R, McGrath PJ. Narcotic analgesic: fentanyl reduces the intensity but the unpleasantness of painful tooth pulp sensations. Science. 1979;203:1261-1263.
Gracely RH. Multiple random staircase assessment of thermal pain perception. In: Dubner R, Bond M, Gebhart G, editors. Proceedings of the Fifth World Congress on Pain. Amsterdam: Elsevier; 1988. p. 391–4.
Gracely RH, Lota L, Walther DJ, Dubner R. A multiple random staircase method of psychophysical pain assessment. Pain. 1988;32:55-63.
Gracely R, Kwilosz DM. The descriptor differential scale: applying psychophysical principles to clinical pain assessment. Pain. 1988;35:279-288.
Gracely RH, Naliboff BD., 1996. Measurment of pain sensation. In: Kruger L, editor., Handbook of perception and cognition: somatosensory systems. Raven Press, New York, NY, pp. 243-313.
Gracely RH, Petzke F, Wolf JM, Clauw DJ. Functional magnetic resonance imaging evidence of augmented pain processing in fibromyalgia. Arthritis Rheum. 2002;46:1333-1343.
Granges G, Littlejohn G. Pressure pain threshold in pain-free subjects, in patients with chronic regional pain syndromes, and in patients with fibromyalgia syndrome. Arthritis Rheum. 1993;36:642-646.
Kosek E, Ekholm J, Hansson P. Increased pressure pain sensibility in fibromyalgia patients is located deep to the skin but not restricted to muscle tissue. Pain. 1995;63:335-339.
Kosek E, Ekholm J, Hansson P. Sensory dysfunction in fibromyalgia patients with implications for pathogenic mechanisms. Pain. 1996;68:375-383.
Kosek E, Hansson P. Modulatory influence on somatosensory perception from vibration and heterotopic noxious conditioning stimulation (HNCS) in fibromyalgia patients and healthy subjects. Pain. 1997;70:41-51.
Lautenbacher S, Rollman GB. Possible deficiencies of pain modulation in fibromyalgia. Clin J Pain. 1997;13:189-196.
Lautenbacher S, Rollman GB, McCain GA. Multi-method assessment of experimental and clinical pain in patients with fibromyalgia. Pain. 1994;59:45-53.
Lawler JE, Naylor SK, Abel MM. Predictability of foot shock differentially affects the phasic blood pressure of SHR, BHR, and WKY rats. Physiol Behav. 1993;54:369-374.
Lautenschlager J, Seglias J, Bruckle W, Muller W. Comparisons of spontaneous pain and tenderness in patients with primary fibromyalgia. Clin Rheumatol. 1993;10:168-173.
Lorenz J, Grasedyck K, Bromm B. Middle and long latency somatosensory evoked potentials after painful laser stimulation in patients with fibromyalgia syndrome. Electroenceph clin Neurophysiol. 1996;100:165-168.
McDermid AJ, Rollman GB, McCain GA. Generalized hypervigilance in fibromyalgia: evidence of perceptual amplification. Pain. 1996;66:133-144.
Melzack R. The short-form McGill pain questionnaire. Pain. 1987;30:191-197.
Mikkelsson M, Latikka P, Kautiainen H, Isomeri R, Isomaki H. Muscle and bone pressure pain threshold and pain tolerance in fibromyalgia patients and controls. Arch Phys Med Rehabil. 1992;73:814-818.
Mountz JM, Bradley LA, Alarcon GS. Abnormal functional activity of the central nervous system in fibromyalgia syndrome. Am J Med Sci. 1998;315:385-396.
Naliboff BD, Munakata J, Fullerton S, Gracely RH, Kodner A, Harraf F, Mayer EA. Evidence for two distinct perceptual alterations in irritable bowel syndrome. Gut. 1997;41:505-512.
Orsini C, Ventura R, Lucchese F, Puglisi-Allegra S, Cabib S. Predictable stress promotes place preference and low mesoaccumbens dopamine response. Physiol Behav. 2002;75:135-141.
Petzke F, Gracely RH, Park KM, Ambrose K, Clauw DJ. What do tender points measure? Influence of distress on four measures of tenderness. J Rheumatol. 2003;30:567-574.
Petzke F, Khine A, Williams D, Groner K, Clauw DJ, Gracely RH. Dolorimetry performed at 3 paired tender points highly predicts overall tenderness. J Rheumatol. 2001;28:2568-2569.
Price DD, Barrell JJ, Gracely RH. A psychophysical analysis of experiential factors that selectively influence the affective dimension of pain. Pain. 1980;8:137-149.
Price DD, Staud R, Robinson ME, Mauderli AP, Cannon R, Vierck CJ. Enhanced temporal summation of second pain and its central modulation in fibromyalgia patients. Pain. 2002;99:49-59.
Quimby LG, Block SR, Gratwick GM. Fibromyalgia: generalized pain intolerance and manifold symptom reporting. J Rheumatol. 1988;15:1264-1270.
Rollnik JD, Schmitz N, Kugler J. Cardiovascular reactions induced by unpredictable, predictable, and controllable painful stimuli during sphygmomanometry. Int J Psychophysiol. 2001;40:161-165.
Scudds RA, Rollman GB, Harth M, McCain GA. Pain perception and the personality measures as discriminators in the classification of fibrosis. J Rheumatol. 1987;14:563-569.
Sorensen J, Graven-Nielsen T, Henriksson KG, Bengtsson M, Arendt-Nielsen L. Hyperexcitability in fibromyalgia. J Rheumatol. 1998;25:152-155.
Staud R, Vierck CJ, Cannon RL, Mauderli AP, Price DD. Abnormal sensitization and temporal summation of second pain (wind-up) in patients with fibromyalgia syndrome. Pain. 2001;91:165-175.
Sternberg WF, Bailin D, Grant M, Gracely RH. Competition alters perception of noxious heat and cold in male and female athletes. Pain. 1998;76:231-238.
Wolfe F. The relation between tender points and fibromyalgia symptom variables: evidence that fibromyalgia is not a discrete disorder in the clinic. Ann Rheum Dis. 1997;56:268-271.
Wolfe F, Ross K, Anderson J, Russell IJ. Aspects of fibromyalgia in the general population: sex, pain threshold, and fibromyalgia symptoms. J Rheumatol. 1995;22:151-156.
Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C, Goldenberg DL, Tugwell P, Campbell SM, Abeles M, Clark P. The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum. 1990;33:160-172.
Keywords:
Fibromyalgia; Pain; Pressure; Thermal
© 2003 Lippincott Williams & Wilkins, Inc.