PAIN 152 (2011) 123130

www.elsevier.com/locate/pain

Sad mood increases pain sensitivity upon thermal grill illusion stimulation: Implications for central pain processing
Michael Karl Boettger a,,1, Christiane Schwier b,1, Karl-Jrgen Br b
a b

Institute of Physiology I, University Hospital, Jena, Germany Department of Psychiatry and Psychotherapy, University Hospital, Jena, Germany

a r t i c l e

i n f o

a b s t r a c t
In different elds of neuroscience research, illusions have successfully been used to unravel underlying mechanisms of stimulus processing. One such illusion existing for the eld of pain research is the socalled thermal grill illusion. Here, painful sensations are elicited by interlacing warm and cold bars, with stimulus intensities (temperatures) of these bars being below the respective heat pain or cold pain thresholds. To date, the underlying mechanisms of this phenomenon are not completely understood. There is some agreement, however, that the sensation evoked by this stimulation is generated by central nervous interactions. Therefore, we followed two approaches in this study: rstly, we aimed at developing and validating a water-driven device which might be used in fMRI scanners in future studies subject to minor adaptations. Secondly, we aimed to interfere with this illusion by induction of a sad mood state, a procedure which is suggested to inuence central nervous structures that are also involved in pain processing. The newly developed device induced thermal grill sensations similar to those reported in the literature. Induction of sad, but not neutral mood states, resulted in higher pain and unpleasantness ratings of the painful illusion. These ndings might be of importance for the understanding of pain processing in healthy volunteers, but putatively even more so in patients with major depressive disorder. Moreover, our results might indicate that central nervous structures involved in the affective domain or cognitive domain of pain processing might be involved in the perception of the illusion. 2010 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

Article history: Received 1 March 2010 Received in revised form 18 September 2010 Accepted 4 October 2010

Keywords: Thermal grill illusion Pain Mood induction Central pain processing Cold pain thresholds Heat pain thresholds

1. Introduction The phenomenon of a moderately painful and awkward sensation evoked by spatially alternating warm and cold stimuli below the respective pain thresholds has rst been described in 1896 [45] and was later called illusion of pain or painful grill illusion. Only recently electrophysiological and psychophysiological methods have been applied to examine putative underlying mechanisms in greater detail. In their landmark paper, Craig and Bushnell [13] could show that nearly all participants perceived the combination of temperatures, that caused warmth or cold feelings when presented on their own, to be painful. In the same study, recordings from the cat spinal cord revealed a dramatically reduced response of thermoreceptive-specic COLD cells in lamina I upon simultaneous warm and cold stimulus presentation, while the response of

Corresponding author. Address: Universittsklinikum Jena, Institut fr Physiologie I/Neurophysiologie, Teichgraben 8, 07743 Jena, Germany. Tel.: +49 (0) 3641 938864; fax: +49 (0) 3641 938812. E-mail address: mkboettger@gmx.de (M.K. Boettger). 1 These authors contributed equally to this work.

heat, pinch and cold (HPC) cells was unaltered. However, it remains speculation at which level of sensory processing the painful sensation is actually generated. Besides convergence or inhibitory mechanisms at the spinal level [13,16,21,22], the hypothalamus and various regions of the cortex have been discussed as candidate regions [13,14]. In particular, the anterior cingulate cortex (ACC) has been shown to be activated by grill stimuli in imaging experiments, while the isolated warm and cold stimuli of the grill were not capable of evoking similar activation in the same study [14]. The ACC further appears to be crucially involved in central processing of emotional and motivational aspects of pain such as unpleasantness [14,43]. Due to the nature of the evoked sensation, the thermal grill illusion has in addition been proposed as an experimental model for central pain states [15]. Taken together, a close examination of this phenomenon promises to reveal important insights into central pain processing. Like in other elds of neuroscience, e.g., in the visual system, the application of an illusion bears an enormous potential to uncover underlying physiological and pathophysiological mechanisms [18]. Since a Peltier-element driven generation of different temperatures [11,14] makes an application in functional magnetic resonance imaging (fMRI) scanners impossible, we aimed at

0304-3959/$36.00 2010 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2010.10.003

124

M.K. Boettger et al. / PAIN 152 (2011) 123130

Fig. 1. Experimental setup for eliciting the thermal grill illusion. (A) Water is cooled and heated, respectively, in two thermostats (1, 2) and pumped through every other bar in the grill device (3 and B). At the measuring site where the palm is being placed (3), the temperature is being measured by a digital thermometer (4).

validating a water-driven thermal grill device that can be used in such experiments in an adapted design in future studies. Thus, we obtained basic data and assessed temperature differences at which the alternating stimuli were perceived painful including respective pain ratings and compared these to those described in the literature. Considering the body of evidence pointing towards a central nervous generation of this pain illusion [14,15,25], we further aimed to interfere with pain processing on the cortical level. We have previously shown that sad mood induction leads to a sensitization to heat pain [46,47]. Therefore, participants were subjected to a modied Velten mood-induction procedure (MIP) to compare sad and neutral emotional states. In both conditions, the painful sensation elicited by the thermal grill was quantied regarding pain intensity and unpleasantness of the sensation applying visual analogue scales (VAS). 2. Materials and methods 2.1. Thermal grill A thermal grill device was designed according to previous literature [11,13] consisting of 15 tubes with 16 cm length and 10 mm diameter and a distance between tubes of 3 mm (number 3 in Fig. 1A and B). Heating and cooling of these tubes were accomplished by perfusion with either warm or cold water which was regulated to the respective temperatures using thermostats (Huber, Germany; Julabo, Germany, 1 and 2 in Figure 1A). Effective temperatures at the tubes were controlled using a digital thermometer (4 in Figure 1A). 2.2. Subjects For validation of the water-driven thermal grill (protocol 1, also see Fig. 2A), 26 healthy participants were recruited (for demographic data, see Table 1). For the investigation of emotional state on the thermal grill perception (protocol 2, Fig. 2B), 32 additional healthy female volunteers between 20 and 28 years of age participated in this study (see Table 1). They were recruited by yer advertisement and email requests. All participants were right-handed, according to the German version of the Edinburgh Handedness Inventory [34]. Prior to the experiment, participants were asked to complete the Beck Depression Inventory (BDI, [7]), the state-trait-anxietyinventory (STAI, [40]) and a questionnaire about current clinical symptoms and past medical history. Potential subjects who presented with one of the following conditions were excluded: any organic origin of pain complaints,

chronic pain, any neurological or psychiatric signs or symptoms as assessed by a standardized interview and a clinical examination, current use of analgesic or antidepressant medication, alcohol or substance dependency and use of alcohol within 12 h before the experiment. For inclusion, all subjects were required to give written informed consent to a protocol approved by the Ethics Committee of the Medical Faculty of the Friedrich-Schiller-University, Jena. 2.3. Experimental design 2.3.1. Protocol 1 (validation of the device, Fig. 2A) At rst, cold and heat pain thresholds (CPT and HPT) were determined consecutively by an ascending method of limits as described previously [4,39]. In brief, a 9 cm2 contact thermode (TSA2001; Medoc, Israel) was placed on the palm of both hands one at a time, i.e., the same area at which grill testing was performed later, and temperature was decreased or increased at a rate of 0.5 C/s (baseline temperature: 32.0 C; minimal temperature: 0 C, maximal temperature: 53.0 C). To determine thermal pain thresholds, participants were asked to follow the written instruction: When

Fig. 2. Protocols for validation (A, protocol 1) and the mood-induction procedure (MIP, B, protocol 2). (A) For validation, cold pain thresholds (CPT) and heat pain thresholds (HPT) were obtained in random order (X). From these thresholds, the temperatures applied to the different bars were calculated. In protocol 1, 7 temperature combinations were used (T1T7). In the rst test (T1), cold bars were set 15 C above CPT, and warm bars to 15 C below HPT (represented by the numbers displayed). Then, T2T7 were performed using differences from the respective thresholds of 13, 11, 9, 7, 5 and 3 C. (B) In order to assess the inuence sad mood induction on the thermal grill perception, CPT and HPT were obtained corresponding to protocol 1. Then, pain and unpleasantness ratings were assessed for the grill illusion using a temperature 5 C above CPT for the cold bars and 5 C below HPT for the warm bars (according to T6 in protocol 1). Following this, either sad or neutral mood was induced on one of the two testing days and the grill perception was repeated. The order of mood induction (sad or neutral) was randomized across subjects (X).

M.K. Boettger et al. / PAIN 152 (2011) 123130 Table 1 Demographic data of participants. Protocol 1 Number Age [years] Male/female BDI STAI-T STAI-S 26 24.1 3.2 12/14 3.8 3.1 34.9 9.1 38.7 8.2 Protocol 2 32 23.3 3.6 0/32 3.7 3.7 35.2 8.4 37.5 7.6

125

Table 2 Pain thresholds and temperature combinations for the different grill illusion stimuli. Protocol 1 Pain thresholds Left side HPT CPT GPT 46.3 0.5 14.4 1.3 15.4 1.9 Right side 46.0 0.5 13.2 1.4 16.6 1.9

BDI, beck depression inventory; STAI-T, state-trait-anxiety-inventory: trait; STAI-S, state-trait-anxiety-inventory: state.

Temperature combinations (grill) Test Left side Twarm T1 T2 T3 T4 T5 T6 T7 31.3 33.3 35.3 37.3 39.3 41.3 43.3 Tcold 29.4 27.4 25.4 23.4 21.4 19.4 17.4 Difference 1.9 5.9 9.9 13.9 17.9 21.9 25.9 Right side Twarm 31.0 33.0 35.0 37.0 39.0 41.0 43.0 Tcold 28.2 26.2 24.2 22.2 20.2 18.2 16.2 Difference 2.8 6.8 10.8 14.8 18.8 22.8 26.8

thermal perception becomes painful, press the stop button immediately. The investigation started with three learning trials and continued with ve consecutive tests. The mean of the last ve trials was calculated and used as the CPT or HPT for further experiments. Subjects were further asked to rate cold and heat stimuli in two categories: pain intensity and amount of unpleasantness, for thresholds and all grill experiments (see below). For this assessment, visual analogue scales (VAS) of 10 cm length were used on which participants indicated pain/unpleasantness intensity. Here, the very left end of VAS indicated no pain/no unpleasant feeling (0 mm), while the very right end represented worst imaginable pain/very unpleasant feeling (100 mm). Furthermore, participants were asked to indicate the quality of the stimulus as hot or cold or other, with the latter to be specied. Multiple answers were allowed. Temperatures for the warm and cold bars in order to elicit a thermal grill illusion were adapted to the individual heat and cold pain thresholds. In the rst temperature combination with the smallest difference between warm and cold stimuli (T1), the cold bar was set 15 C above the CPT of the respective subject, and the warm bar was heated to 15 C below HPT. The same test was performed repeatedly with temperatures gradually being decreased for the cold and increased for the warm bars, respectively. Using this protocol, a total of 7 test conditions were applied (T1T7) using 15 C (T1), 13 C (T2), 11 C (T3), 9 C (T4), 7 C (T5), 5 C (T6) and 3 C (T7) as differences from the respective CPT and HPT (also see Table 2). After reaching these temperatures as controlled by the attached digital thermometer, participants were asked to place their right or left hand in randomized order on the device for 30 s. After removing the hand participants had to indicate pain and unpleasantness on the VAS. Intervals between measurements were at least 3 min. 2.3.2. Protocol 2 (mood induction, Fig. 2B) After determining thermal thresholds as described above, the parameters dened for T6 in protocol 1 (Temperature of cold bars: 5 C above CPT; temperature of warm bars: 5 C below HPT) were chosen to elicit a grill sensation in protocol 2 (also see Table 2). All 32 subjects underwent two experimental conditions (withinsubject-design). In the rst condition, subjects were asked to rate their perceived pain intensity, the unpleasant feeling and indicate the quality of the perceived stimuli induced by a thermal grill before and after sad mood induction. In a second condition, sad mood induction was replaced by the induction of neutral mood. The order of these two conditions was randomized between subjects. The conditions took place at two different time points with a between-tests interval of at least 2 days, but not more then 7 days. 2.4. Mood induction procedure For the investigation of sad and neutral emotional states, we used the modied Velten mood-induction procedure (MIP) as described previously [44,47]. During MIP, subjects were exposed to

Protocol 2 Pain thresholds Neutral mood HPT CPT 47.7 0.3 11.0 1.5 Sad mood 47.1 0.5 10.0 1.5

Temperature combination (grill) Test Neutral mood Twarm T6 42.7 Tcold 16.0 Difference 26.7 Sad mood Twarm 42.1 Tcold 15.0 Difference 27.1

HPT, heat pain threshold; CPT, cold pain threshold; GPT, grill pain threshold. Twarm, temperature of warm bars; Tcold, temperature of cold bars. Difference: Difference between temperatures of cold and warm bars presented simultaneously for grill stimulation. In protocol 2, only one temperature combination was presented, according to T6 in protocol 1. Data are presented as temperatures in C. For HPT, CPT and GPT, mean SEM are presented, while for temperature combinations, only the respective mean values for all participants are depicted. There were no significant side differences in protocol 1 and no signicant temperature differences before mood induction as presented for protocol 2.

a series of 21 self-referent sad-mood statements (e.g. Life is a heavy burden) or neutral statements (e.g. An orange is a citrus fruit), which were presented twice. Participants had to read the statements aloud. While reading the statements, participants were asked to attempt to experience the mood suggested by the statements. To further facilitate mood induction, participants were exposed to sad music from Damien Rice in the sad mood condition and to Mozarts Piano concerto No. 21 in C Major in the neutral mood condition [42]. The whole mood-induction procedure lasted approximately 12 min. For the assessment of affective changes during MIP, participants were asked to rate their mood valence and arousal before, immediately after negative and neutral MIP and at the end of the experimental procedure. For that purpose, an affective rating system (Self-Assessment Manikin, SAM) was used [28]. The ratings of valence are indicated by 5 graphical representations of facial expressions ranging from a severe frown (most sad = 4) to a broad smile (most positive = +4). For arousal, the manikin reached from a state of low to high agitation (9 represents a high rating and 1 represents a low rating). 2.5. Statistical analyses For statistical analyses, SPSS for Windows (version 17.0) was used. Normal distribution of data was controlled for using KolmogorovSmirnov tests.

126

M.K. Boettger et al. / PAIN 152 (2011) 123130

For protocol 1, VAS ratings of pain and unpleasantness were compared between the rst grill illusion (thresholds 15 C, T1) and all other conditions (T2T7) using paired t-tests applying BonferroniHolm correction for multiple comparisons. For protocol 2, repeated measures analyses of variance (ANOVA) applying the factors Time (before vs. after mood induction) and Treatment (sad vs. neutral mood condition) were calculated to uncover differences in the mood dimensions as well as differences in the ratings of thermal grill stimuli. For further examination, planned post hoc t-tests were performed. Interactions between Time Treatment were of special interest. Re-test reliability was tested using Pearsons chi-square-tests in the participants of protocol 2, since thresholds and ratings for pain thresholds and stimulus perception were obtained twice. Here, only values before MIP were considered. In order to investigate whether those patients showing the strongest changes in mood are those displaying the greatest differences in pain and unpleasantness ratings, differences for valence, arousal, pain intensity and unpleasantness between the time points before mood induction and after the experiment were calculated for the sad mood condition. These parameters were then correlated using Pearsons correlation coefcient. Statistical signicance was accepted for p < 0.05.

3. Results 3.1. Validation of the device (Protocol 1) Cold, heat and grill pain thresholds (CPT, HPT, GPT) for the left and right hand are presented in Table 2. Pain intensity and unpleasantness ratings obtained after every single threshold testing (cold, heat, grill) were similar across stimulations (Figure 3A and C). Both ratings increased with larger grill temperature differences, becoming signicantly different from the rst applied grill stimulus (T1) starting at temperatures 7 C above CPT for the cold bars and 7 C below HPT for the warm bars (T5) for pain ratings and 9 C (T4), respectively, for unpleasantness ratings (Figure 3B and D). No signicant gender or age differences were apparent for any of the parameters obtained. Of the participants in protocol 1, two did not experience any pain irrespective of the temperature difference presented, while showing normal cold and heat pain thresholds. Data of these participants were included in the analyses. The quality of the thermal grill sensation upon the combined stimulus of 5 C below HPT for the warm bars and 5 C above CPT for the cold bars (T6) was indicated as hot by 49 out of 58 participants, as cold by 28 participants and as tingling by 4 participants. Other attributes given by the participants were neutral (n = 2), burning (n = 4), insect-like (n = 1) and sharp (n = 1). 3.2. Changes in mood quality (valence) and intensity (arousal) induced by MIP (Protocol 2) A reliable sad mood induction was indicated by a highly significant interaction effect of Time Treatment for valence [F(2, 61) = 15.695, p < 0.001, Fig. 4A] and arousal [F(2, 61) = 4.768, p = 0.012, Fig. 4B]. Thus, sad mood induction resulted in a signicantly decreased mood quality after MIP in contrast to neutral MIP. This was observed shortly after mood induction and lasted until the end of the experiment. In the neutral condition subjects did not report a signicant change in the valence dimension after mood induction (p = 0.48; Fig. 4A). Furthermore, after sad mood induction subjects tended to rate their mood intensity lower than immediately before mood induction (p = 0.091), which became signicant after the experiment (Fig. 4B). Neutral mood induction induced no signicant change in arousal. The order of paradigms (sad mood condition rst or neutral condition rst) did not inuence the amount of mood change. 3.3. Changes in pain thresholds and pain intensity rating upon MIP (Protocol 2) Cold and heat pain thresholds were not signicantly different before either sad MIP (Table 2) was performed in the crossover design as described above. Likewise, ratings of pain and unpleasantness did not signicantly differ between conditions before MIP (Fig. 5A and B). The mean differences between warm and cold bars used for eliciting the thermal grill illusion were 26.7 1.7 C and

Fig. 3. Characteristics of the thermal grill illusion regarding perceived pain and unpleasantness of the sensation. (A) Pain intensity ratings as obtained from visual analogue scales (VAS) at the cold pain threshold (CPT), heat pain threshold (HPT) and the grill pain threshold (GPT), i.e., the temperature difference at which participants rst perceived the sensation painful. (B) Pain intensity ratings upon presentation of different temperatures for warm and cold bars which were CPT + X and HPTX C (X = 15, 13, 11, 9, 7, 5, 3; according to temperature combination 17, i.e., T1T7). Pain intensities increased with the temperature difference between bars. There were no differences between left and right hands. (C) Ratings of unpleasantness at the respective thresholds. (D) Unpleasantness ratings at the different presented grill temperature combinations. Again, no side differences were obvious and unpleasantness increased with temperature difference. Data are presented as mean SEM.

M.K. Boettger et al. / PAIN 152 (2011) 123130

127

Fig. 4. Mood changes upon neutral and sad mood induction. (A) Valence of mood as assessed by the self-assessment manikin (SAM) was unchanged upon neutral mood induction, but signicantly decreased after sad mood induction and remained signicantly decreased until after the experiment. (B) Arousal showed a different time course with a slight decrease immediately after sad mood induction, but a signicant decrease at the time when the experiment ended. No signicant change in arousal was observed for neutral mood induction. Data are presented as mean SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

27.1 1.9 C for neutral and sad mood induction, respectively (p = 0.86, also see Table 2). A total of 4 participants did not experience any pain upon thermal grill stimulation, irrespective of the temperature combination provided and of the mood induced. Data of these participants were included in the analyses. Signicant Time Treatment interactions were identied for ratings of the thermal grill illusion, both for pain intensity [F(1, 62) = 20.517, p < 0.001] and unpleasantness [F(1, 62) = 11.059, p = 0.002]. In particular, after sad mood induction thermal grill stimuli were rated signicantly more painful (Fig. 5A) and more unpleasant (Fig. 5B). No signicant change in pain or unpleasantness rating was revealed after neutral mood induction (p = 0.42 and p = 0.11, respectively). Prior to mood induction, neither pain intensity nor unpleasantness of the thermal grill was different (p = 0.89 and p = 0.95, respectively). 3.4. Reliability of pain perception and ratings between testing days Pearsons chi-square-test identied signicant re-test correlations between rst and second assessment of cold and hot pain thresholds at the two different time points (before neutral mood induction and before sad mod induction). The same was true for pain ratings (see Table 3). Likewise, we found a signicant correlation between rst and second measurement of pre-MIP thermal grill stimuli perception (Table 3).

Fig. 5. Changes in grill perception upon mood induction. Before either sad or neutral mood-induction procedure (MIP), there were no differences in pain ratings (A) and unpleasantness ratings (B) for cold pain thresholds (CPT), heat pain thresholds (HPT) and the dened grill stimulus. Both pain intensity (A) and unpleasantness (B) were rated signicantly higher in the sad mood condition as compared to the neutral mood condition, and pain intensity was even signicantly increased as compared to values before MIP. Data are presented as mean SEM. *p < 0.05; **p < 0.01.

3.5. Correlations between MIP-induced changes in pain and mood Weak correlations were observed between changes in mood quality (valence) and changes in grill sensation pain intensity induced by sad mood induction (r = 0.270; p = 0.031) as well as for
Table 3 Re-test reliability of pain testing. Correlation between rst and second measurement Cold stimuli Pain thresholds (ascending method of limits) Pain rating (VAS) Unpleasant feeling rating (VAS) Heat stimuli Pain thresholds (ascending method of limits) Pain rating (VAS) Unpleasant feeling rating (VAS) Thermal grill stimuli Pain rating (VAS) Unpleasant feeling rating (VAS) VAS, visual analogue scale.

v2
0.86 0.76 0.69 0.80 0.60 0.56 0.58 0.54

p-value <0.001 <0.001 <0.001 <0.001 <0.001 =0.001 <0.001 =0.002

128

M.K. Boettger et al. / PAIN 152 (2011) 123130

valence and arousal with unpleasantness (r = 0.258; p = 0.040 and r = 0.301; p = 0.023, respectively), indicating an association between subjective ratings of both pain perception and mood.

4. Discussion In this study, we aimed to amend the understanding of the thermal grill illusion which is elicited by interlacing warm and cold bars, yet without reaching subjective pain thresholds for a uniform temperature stimulus as obtained in the same experiment. In particular, we found that sad mood induction in contrast to neutral mood signicantly increased the perceived pain sensation as well as the perceived unpleasantness during thermal grill stimulation. 4.1. Validation of the water-driven thermal grill Recent studies using thermal grill devices employed either Peltier elements or water for generating the desired temperatures [11,13,14,16,31]. Since no thermal grill testing systems are commercially available at present, we designed a self-made device using water for future application in fMRI scanners. Introducing this device, our rst aim was to compare the obtained data to those described previously and apply a study protocol similar to the one used by Bouhassira et al. [11]. In our study, taking into account the cold and heat pain thresholds of participants, different combinations of temperatures were presented, dened by the differences to the respective thresholds. The pain and unpleasantness scores evoked by these temperature differences were then scored on visual analogue scales. In our analysis, pain ratings were comparable to those obtained in several reference studies [11,13,25,26]. In particular, it is evident from all studies that larger differences between temperatures lead to more painful and more unpleasant sensations, while hardly any participants feel a painful sensation at all when the presented temperatures only differ marginally. In our study, values of pain and unpleasantness ratings at the respective thresholds were rather constant across stimulations, i.e., cold, heat and grill (also see Fig. 3). While the spatial resolution of the thermal grill has been shown to be of minor importance [16], our device still represents similar characteristics to those mentioned above. 4.2. Effects of mood induction on thermal grill perception Mood induction, which was reliably induced in all participants according to our method [44,47], led to a signicantly decreased mood quality in the sad mood condition outlasting the duration of the thermal grill perception testing, while the neutral condition did not signicantly alter quality or intensity of mood. Thus, in the sad mood condition, participants rated the perceived pain and unpleasantness of the grill sensation signicantly higher than in the neutral condition, i.e., this phenomenon was perceived more painful. The calculated difference in pain and unpleasantness ratings before and after MIP showed a correlation with mood quality changes, i.e., those patients with the strongest mood-induction effects were those who scored the evoked sensations most painful or unpleasant, respectively. Baseline CPT and HPT as well as pain ratings for the grill stimulation before mood induction did not differ between conditions, therefore, making confounding inuences by factors other than mood induction less likely. Furthermore, it has been demonstrated that sad mood is capable of reducing pain thresholds in depressed patients and controls [44,47] by 0.51.0 C. Considering the temperature difference to the respective thresholds of 5 C in our study, the inuence of this mood-induced threshold change is unlikely to explain the increased pain rating.

Previous studies also found increased pain ratings upon noxious stimuli during negative mood induction as compared to neutral or positive mood induction [24,38]. However, one has to bear in mind that the painful sensation induced using a thermal grill does not involve supra-threshold noxious stimulation, but reects a phenomenon that has been discussed to evolve from inhibitory and disinhibitory mechanisms on the spinal [13,16,21,22] or the cerebral level [13,14]. Concerning the latter, a model has been proposed in which the lateral lamina I spino-thalamo-cortical terminating in the insula and the medial lamina I pathway activating the anterior cingulate cortex (ACC) interact, particularly since the ACC appeared to be the only region which was activated by the grill, but not by isolated cold and warm stimuli [14]. States of negative emotion as induced in this study have further shown an enhanced pain-evoked activity both in the ACC and the insular cortex (IC) [8,36], which are strongly involved both in central processing of noxious stimuli [2], but also, as described above, in those mediated by the grill illusion [14]. Adding to this, correlation analyses in patients with bromyalgia revealed a strong association between depressive symptoms and activation not only in the IC, but also in the amygdala, i.e., areas that are thought to be involved in affective pain processing [20]. Besides such increased brain responses to noxious stimuli, sad mood induction has further been shown to lead to signicantly less deactivation, for instance in the angular gyrus or the posterior cingulate [8]. Therefore, the combination of sad mood induction leading to central nervous effects on pain perception and quantication of pain intensity with a centrally generated pain-like phenomenon, the thermal grill illusion, might in fact represent an effective tool to further identify the central mechanisms that are important for the internal evaluation of a perceived stimulus that is encoded to be painful. 4.3. Putative clinical implications of the ndings Besides a gain in the understanding of central pain processing and the rating of stimuli as painful in healthy subjects, the results described herein might be of potential importance for the understanding of chronic pain states, particularly those in which the underlying cause, e.g., tissue damage, is no longer present and in which pain outlasts the underlying pathology, but also in central neuropathic pain states as suggested previously [15]. Furthermore, there might be an implication for the study of pain perception in patients with major depressive disorder (MDD). These patients have been described to report more pain complaints than respective reference populations [12], although these results need to be considered with some care [19]. On the other hand, experimental pain thresholds for noxious stimuli applied to the skin have repeatedly been shown to be increased in patients with MDD and related diseases [1,35,9,10,17,29,30,44]. Since the sensory-discriminative component of stimulus perception including pain appears to be intact as assessed by quantitative sensory testing [27], higher central nervous nuclei mediating the affective, autonomic or cognitive components of pain perception might be involved in the sense of a maladaptive activation of the neural network [41]. For instance, enhanced activity in the lateral prefrontal cortex has been shown in depressed patients upon noxious stimulation [6]. Such an increase in prefrontal activity has been suggested in the attentional modication of pain perception [35] and appears to suppress the unpleasantness of a perceived stimulus [33], thereby possibly inuencing both the affective and the cognitive components. Since the prefrontal cortex is further closely linked to the ACC [2], it will be interesting to examine patients with major depression with the thermal grill illusion in future studies, since here, no noxious thresholds are being crossed, therefore providing insights in central stimulus processing of depressed patients.

M.K. Boettger et al. / PAIN 152 (2011) 123130

129

4.4. Limitations Cold and heat pain thresholds were obtained using a 9 cm2 square thermode, while the stimulation area using the grill device was larger (approximately 26 cm2). Such increase in stimulus area has been shown to putatively alter heat pain perception, with a doubling of stimulus size corresponding to an increase of 1.3 C [37], thereby putatively limiting our results. However, since particularly in the mood-induction procedure, stimulus intensities 5 C below respective thresholds were applied, a major confound caused by different stimulation areas appears rather unlikely in our study. Furthermore, a recent study employing a thermal grill revealed that the perception of pain in this illusion is irrespective of the number of bars used, and thereby the area that is being stimulated [32]. In protocol 2, only female participants were examined. This has mainly two reasons: rstly, we aimed at excluding gender effects as a putative confounding factor, not because of the induction of the thermal grill illusion which we could show to be similar across sexes, but rather since it has been shown that women are more prone to react to sad mood induction [23]. Secondly, more women suffer from major depressive disorder than men. For these reasons, the results presented herein might not be transferable to male participants. Furthermore, we aimed to include a rather homogenous study population of young aged women in order to exclude putative confounders like neuropathy or chronic pain complaints. This, however, might further limit the applicability of the results obtained on other age groups. 4.5. Conclusions We found that sad mood induction was capable of enhancing the painful sensation elicited by the non-noxious temperatures presented, i.e., a central amplication of a centrally generated pain-like sensation. Therefore, this combination of methods provides a promising tool for studying central pain processing, particularly when an application together with functional brain imaging becomes available, following minor adaptations of the device. This approach might further be valuable for dissecting the described alterations of pain perception in major depression. Acknowledgements The authors would particularly like to thank Gnter Ditze and Gnther Sammler, central manufacturing unit at the University Hospital Jena (Zentralwerkstatt), for their help with building the thermal grill device. Furthermore, we would like to thank Janneke Terhaar, Anna Kliem and Caroline Gerdes, Department of Psychiatry, for recruiting and helping with participant assessment as well as Elisabeth Lehmkuhl and David Grossmann, Institute of Physiology I, for their assistance in obtaining the data for the validation of the device. In addition, we are grateful to Hans-Georg Schaible, Institute of Physiology I, for helpful suggestions and discussions. There are no actual or potential conicts of interest for any of the authors. This study was supported by Deutsche Forschungsgemeinschaft (DFG) BO3235/3-1. References
[1] Adler G, Gattaz WF. Pain perception threshold in major depression. Biol Psychiatry 1993;34:6879. [2] Apkarian AV, Bushnell MC, Treede RD, Zubieta JK. Human brain mechanisms of pain perception and regulation in health and disease. Eur J Pain 2005;9:46384. [3] Br KJ, Greiner W, Letsch A, Kbele R, Sauer H. Inuence of gender and hemispheric lateralization on heat pain perception in major depression. J Psychiatr Res 2003;37:34553.

[4] Br KJ, Brehm S, Boettger MK, Boettger S, Wagner G, Sauer H. Pain perception in major depression depends on pain modality. Pain 2005;117:97103. [5] Br KJ, Brehm S, Boettger MK, Wagner G, Boettger S, Sauer H. Decreased sensitivity to experimental pain in adjustment disorder. Eur J Pain 2006;10:46771. [6] Br KJ, Wagner G, Koschke M, Boettger S, Boettger MK, Schlsser R, Sauer H. Increased prefrontal activation during pain perception in major depression. Biol Psychiatry 2007;62:12817. [7] Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:56171. [8] Berna C, Leknes S, Holmes EA, Edwards RR, Goodwin GM, Tracey I. Induction of depressed mood disrupts emotion regulation neurocircuitry and enhances pain unpleasantness. Biol Psychiatry 2010;67:108390. [9] Boettger MK, Greiner W, Rachow T, Brhl C, Br KJ. Sympathetic skin response following painful electrical stimulation is increased in major depression. Pain 2010;149:1304. [10] Boettger MK, Br KJ. Perception for ischemic pain shows similarities in adjustment disorder and major depression. Eur J Pain 2007;11:81922. [11] Bouhassira D, Kern D, Rouaud J, Pelle-Lancien E, Morain F. Investigation of the paradoxical painful sensation (illusion of pain) produced by a thermal grill. Pain 2005;114:1607. [12] Corruble E, Guel JD. Pain complaints in depressed inpatients. Psychopathology 2000;33:3079. [13] Craig AD, Bushnell MC. The thermal grill illusion: unmasking the burn of cold pain. Science 1994;265:2525. [14] Craig AD, Reiman EM, Evans A, Bushnell MC. Functional imaging of an illusion of pain. Nature 1996;384:25860. [15] Craig AD. Can the basis for central neuropathic pain be identied by using a thermal grill? Pain 2008;135:2156. [16] Defrin R, Benstein-Sheraizin A, Bezalel A, Mantzur O, Arendt-Nielsen L. The spatial characteristics of the painful thermal grill illusion. Pain 2008;138:57786. [17] Dickens C, McGowan L, Dale S. Impact of depression on experimental pain perception: a systematic review of the literature with meta-analysis. Psychosom Med 2003;65:36975. [18] Eagleman DM. Visual illusions and neurobiology. Nat Rev Neurosci 2001;2:9206. [19] Garcia-Cebrian A, Gandhi P, Demyttenaere K, Peveler R. The association of depression and painful physical symptoms a review of the European literature. Eur Psychiatry 2006;21:37988. [20] Giesecke T, Gracely RH, Williams DA, Geisser ME, Petzke FW, Clauw DJ. The relationship between depression, clinical pain, and experimental pain in a chronic pain cohort. Arthritis Rheum 2005;52:157784. [21] Green BG. Synthetic heat at mild temperatures. Somatosens Mot Res 2002;19:1308. [22] Green BG, Pope JV. Innocuous cooling can produce nociceptive sensations that are inhibited during dynamic mechanical contact. Exp Brain Res 2003;148:2909. [23] Kenntner-Mabiala R, Gorges S, Alpers GW, Lehmann AC, Pauli P. Musically induced arousal affects pain perception in females but not in males: a psychophysiological examination. Biol Psychol 2007;75:1923. [24] Kenntner-Mabiala R, Andreatta M, Wieser MJ, Muhlberger A, Pauli P. Distinct effects of attention and affect on pain perception and somatosensory evoked potentials. Biol Psychol 2008;78:11422. [25] Kern D, Pelle-Lancien E, Luce V, Bouhassira D. Pharmacological dissection of the paradoxical pain induced by a thermal grill. Pain 2008;135:2919. [26] Kern D, Plantevin F, Bouhassira D. Effects of morphine on the experimental illusion of pain produced by a thermal grill. Pain 2008;139:6539. [27] Klauenberg S, Maier C, Assion HJ, Hoffmann A, Krumova EK, Magerl W, Scherens A, Treede RD, Juckel G. Depression and changed pain perception: hints for a central disinhibition mechanism. Pain 2008;140:33243. [28] Lang PJ. Behavioral treatment and bio-behavioral assessment. In: Sidowski JB, Johnson JH, Williams TA, editors. Technology in mental health care delivery systems. Norwood, NJ: Ablex; 1980. p. 11937. [29] Lautenbacher S, Krieg JC. Pain perception in psychiatric disorders: a review of the literature. J Psychiatr Res 1994;28:10922. [30] Lautenbacher S, Spernal J, Schreiber W, Krieg JC. Relationship between clinical pain complaints and pain sensitivity in patients with depression and panic disorder. Psychosom Med 1999;61:8227. [31] Leung AY, Wallace MS, Schulteis G, Yaksh TL. Qualitative and quantitative characterization of the thermal grill. Pain 2005;116:2632. [32] Li X, Petrini L, Wang L, Defrin R, Arendt-Nielsen A. The importance of stimulus parameters for the experience of the thermal grill illusion. Neurophysiol Clin 2009;39:27582. [33] Lorenz J, Minoshima S, Casey KL. Keeping pain out of mind: the role of the dorsolateral prefrontal cortex in pain modulation. Brain 2003;126:107991. [34] Oldeld RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97113. [35] Petrovic P, Petersson KM, Ghatan PH, Stone-Elander S, Ingvar M. Pain-related cerebral activation is altered by a distracting cognitive task. Pain 2000;85:1930. [36] Phillips ML, Gregory LJ, Cullen S, Coen S, Ng V, Andrew C, Giampietro V, Bullmore E, Zelaya F, Amaro E, Thompson DG, Hobson AR, Williams SC, Brammer M, Aziz Q. The effect of negative emotional context on

130

M.K. Boettger et al. / PAIN 152 (2011) 123130 neural and behavioural responses to oesophageal stimulation. Brain 2003;126:66984. Price DD, McHafe JG, Larson MA. Spatial summation of heat-induced pain: inuence of stimulation area and spatial separation on perceived pain sensation intensity and unpleasantness. J Neurophysiol 1989;62:12709. Rainville P, Bao QV, Chretien P. Pain-related emotions modulate experimental pain perception and autonomic responses. Pain 2005;118:30618. Schwier C, Kliem A, Boettger MK, Br KJ. Increased cold-pain thresholds in major depression. J Pain 2010;11:28790. Spielberger CD, Gorsuch RL, Lushene RE. Manual for the state-train anxiety inventory. New York: Consulting Psychologist Press; 1979. Strigo IA, Simmons AN, Matthews SC, Craig AD, Paulus MP. Association of major depressive disorder with altered functional brain response during anticipation and processing of heat pain. Arch Gen Psychiatry 2008;65:127584. [42] Sutherland G, Newman B, Rachman S. Experimental investigations of the relations between mood and intrusive unwanted cognitions. Br J Med Psychol 1982;55:12738. [43] Talbot JD, Marrett S, Evans AC, Meyer E, Bushnell MC, Duncan GH. Multiple representations of pain in human cerebral cortex. Science 1991;251:13558. [44] Terhaar J, Boettger MK, Schwier C, Wagner G, Israel AK, Br KJ. Increased sensitivity to heat pain after sad mood induction in female patients with major depression. Eur J Pain 2010;14:55963. [45] Thunberg T. Frnimmelserna vid till samma stlle lokasirerad, samtidigtb pagaende kld och vrmeretning. Uppsala Lkkfren Fhr 1896:48995. [46] Velten Jr E. A laboratory task for induction of mood states. Behav Res Ther 1968;6:47382. [47] Wagner G, Koschke M, Leuf T, Schlsser R, Br KJ. Reduced heat pain thresholds after sad-mood induction are associated with changes in thalamic activity. Neuropsychologia 2009;47:9807.

[37]

[38] [39] [40] [41]