- Email: [email protected]
Functional brain imaging of trigeminal neuralgia
European Journal of Pain 15 (2011) 124–131
Contents lists available at ScienceDirect
European Journal of Pain journal homepage: www.EuropeanJournalPain.com
Functional brain imaging of trigeminal neuralgia Xavier Moisset a, Nicolas Villain b, Denis Ducreux c, Alain Serrie d, Gérard Cunin d, Dominique Valade e, Bernard Calvino f, Didier Bouhassira a,* a
INSERM U-987, Physiopathologie et Pharmacologie Clinique de la Douleur, Hôpital Ambroise Paré, Boulogne-Billancourt F-92100, France INSERM-EPHE-Université de Caen/Basse-Normandie, Unité U923, GIP Cyceron, CHU Côte de Nacre, Caen F-14074, France c Service de Neuroradiologie et CIERM, AP-HP Hôpital de Bicêtre, Le Kremlin Bicêtre F-94275, France d Fédération d’Evaluation et de Traitement de la douleur, Hôpital Lariboisière, Paris F-75475, France e Service Urgences Céphalées, Hôpital Lariboisière, Paris F-75475, France f Laboratoire de Neurobiologie, UMR ESPCI-CNRS 7637, Paris F-75005, France b
a r t i c l e
i n f o
Article history: Received 7 January 2010 Received in revised form 17 May 2010 Accepted 6 June 2010 Available online 6 July 2010 Keywords: Neuropathic pain Functional imaging Trigeminal neuralgia Allodynia
a b s t r a c t We used functional magnetic resonance imaging (fMRI) to analyze changes in brain activity associated with stimulation of the cutaneous trigger zone in patients with classic trigeminal neuralgia (CTN). Fifteen consecutive patients with CTN in the second or third division of the nerve, were included in this study. The fMRI paradigm consisted of light tactile stimuli of the trigger zone and the homologous contralateral area. Stimulation of the affected side induced pain in seven patients, but was not painful in eight patients on the day of the experiment. Painful stimuli were associated with significantly increased activity in the spinal trigeminal nucleus (SpV), thalamus, primary and secondary somatosensory cortices (S1, S2), anterior cingulate cortex (ACC), insula, premotor/motor cortex, prefrontal areas, putamen, hippocampus and brainstem. Nonpainful stimulation of the trigger zone activated all but three of these structures (SpV, brainstem and ACC). After a successful surgical treatment, activation induced by stimulation of the operated side was confined to S1 and S2. Our data demonstrate the pathological hyperexcitability of the trigeminal nociceptive system, including the second order trigeminal sensory neurons during evoked attacks of CTN. Such sensitization may depend on pain modulatory systems involving both the brainstem (i.e. periaqueductal gray and adjacent structures) and interconnected cortical structures (i.e. ACC). The fact that large portions of the classical ‘pain neuromatrix’ were also activated during nonpainful stimulation of the trigger zone, could reflect a state of maintained sensitization of the trigeminal nociceptive systems in CTN. Ó 2010 European Federation of International Association for the Study of Pain Chapters. Published by Elsevier Ltd. All rights reserved.
1. Introduction Idiopathic or classic trigeminal neuralgia (CTN) is severe unilateral paroxysmal pain in one or more branches of the fifth cranial nerve, typically the maxillary or mandibular branches. CTN is characterized by brief (a few seconds) paroxysmal pain, most frequently described as a ‘stabbing’- or ‘electric-shock’-like pain. Between attacks the patient is pain-free and the standard clinical examination is normal. The paroxysms may be spontaneous or provoked by various precipitating factors (yawning, chewing, etc.) and/or light tactile stimulation of a specific cutaneous area, the ‘‘trigger zone” (Zakrzewska, 2002). Thus, CTN is a unique neuro-
* Corresponding author. Address: INSERM U-987, Centre d’Evaluation et de Traitement de la Douleur, Hôpital Ambroise Paré, 9 Avenue Charles de Gaulle, 92100 Boulogne-Billancourt, France. Tel.: +33 1 49 09 45 56; fax: +33 1 49 09 44 35. E-mail address: [email protected] (D. Bouhassira).
pathic pain syndrome combining paroxysmal pain and evoked pain which might be regarded as a special form of dynamic mechanical allodynia (i.e. pain induced by normally nonpainful tactile stimuli). Pain evoked by tactile stimuli is a major feature of both peripheral and central neuropathic pain syndromes (Attal et al., 2008). Several studies have investigated changes in brain activity associated with this symptom to determine whether this aberrant pain results from an abnormal activation of the physiological ‘‘pain matrix” (Kupers and Kehlet, 2006; Moisset and Bouhassira, 2007). However, the results obtained have been very variable, probably reflecting the heterogeneity of patients in terms of etiology, lesion topography and symptomatology. In particular, variations in the activation of somatosensory systems may have been due to differences in the magnitude of sensory deafferentation, independent of pain. In addition, the fact that a majority of patients presented also with spontaneous continuous pain may have masked the effects of evoked pain.
1090-3801/$36.00 Ó 2010 European Federation of International Association for the Study of Pain Chapters. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ejpain.2010.06.006
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
We used functional magnetic resonance imaging (fMRI) in a group of patients with CTN since this ‘‘pathological model” appears particularly suited to investigating the brain correlates of pain evoked by tactile stimuli. Another unique feature of CTN is that, in contrast with other neuropathic pain syndromes, it can be readily cured by surgery (Lopez et al., 2004). Thus, we also tested our patients after a successful percutaneous radiofrequency thermocoagulation of the Gasserian ganglion to determine whether the changes in brain activity were reversed. 2. Materials and methods This study was approved by the Ambroise Paré Local Ethical Committee. All patients and healthy volunteers gave written informed consent. 2.1. Patients Patients referred to the Lariboisière or Ambroise Paré Hospitals for refractory trigeminal neuralgia and possible indication for surgical treatment were eligible for this study. Inclusion criteria included classical (idiopathic) trigeminal neuralgia (CTN), as defined by IHS criteria (Headache Classification Subcommittee, 2004), located within the maxillary and/or mandibular branch (V2–V3) of the trigeminal nerve, with a trigger zone sensitive to light tactile stimuli within the same territory. For obvious ethical reasons and to limit the risk of facial spasms (i.e. tic douloureux) or other movements during fMRI, stimulation of the trigger zone should evoke a short (i.e. less than 25 s) attack of pain with a moderate bearable intensity (i.e. 4–6 on a visual analog scale). Exclusion criteria were: symptomatic or atypical neuralgia (e.g. presence of spontaneous ongoing pain and/or sensory deficits); severe neuralgia requiring emergency surgery; neuralgia with frequent spontaneous attacks (i.e. >30/day) which could disrupt the scanning session; previous surgical treatment for the neuralgia; presence of other pain; major psychiatric disease. At inclusion (3–7 days before fMRI), the patients had a complete neurological examination including a comparative sensory examination between the normal and affected sides of the face. Thermal sensitivity was assessed with two thermo-rollers (Somedic, Sweden) at constant temperatures of 40 °C (warm) and 25 °C (cold). Mechanical sensitivity (detection and pain thresholds) was assessed with calibrated Von Frey hairs. Dynamic tactile allodynia
125
was assessed (intensity and quality) by stimulation of the trigger zone with a cotton swab. The mean daily number of attacks and their intensity and quality were also assessed with the Neuropathic Pain Symptom Inventory (NPSI) (Bouhassira et al., 2004). This assessment was systematically performed before the first fMRI in all patients and before the second fMRI in patients seen after a percutaneous radiofrequency thermocoagulation of the Gasserian ganglion. 2.2. Surgical procedure A standardized controlled retrogasserian percutaneous radiofrequency thermocoagulation was performed in pharmaco-resistant patients by the same trained investigators (GC and AS) using the classical technique (Sweet and Wepsic, 1974; Siegfried, 1977). The needle was introduced into the foramen ovale under short-term general anesthesia. Its position was controlled with imaging intensifier and modified according to the electrophysiological test carried out after the patient was woken up. The temperature of the tip of the needle was maintained between 65 and 70 °C and the duration of the lesion was adapted to each patient to produce a pinprick hypoalgesia limited to the territory of the affected branch of the trigeminal nerve. 2.3. Experimental design Patients on analgesic medication were asked to discontinue their medication 12 h before their scheduled scanning session. Patients were systematically tested before entering the scanner to check for allodynia and to assess its intensity and duration. The fMRI session consisted of one anatomical and three functional scanning sessions. Each functional scanning session consisted of four acquisition epochs without simulation (27 s long; rest condition) alternating with three epochs (27 s long) with light tactile stimulation of one of the three sites: trigger zone, contralateral homologous side or dorsum of the right hand (as a control of activation of the somatosensory system). The stimuli were applied by the same trained investigator using a cotton swab (attached to a 0.7 m long wood stick to avoid magnetic inhomogeneity) with a frequency of 1 per second (Fig. 1). After each scanning session, the patients rated their mean pain intensity over the three stimulation epochs on a 100 mm visual analog scale (VAS; from 0: ‘‘no pain” to 100: ‘‘worst possible pain”).
Fig. 1. Schematic illustration of the experimental paradigm. Each patient participated in three functional scanning sessions. Each session consisted of four acquisition epochs without simulation (27 s long; rest condition) alternating with three epochs (27 s long) with light tactile stimulation of one of the three sites: (A) trigger zone, (B) contralateral homologous side or (C) dorsum of the right hand. After each scanning session, the patients rated their mean pain intensity over the three stimulation epochs on a 100 mm visual analog scale (VAS; from 0: ‘‘no pain” to 100: ‘‘worst possible pain”).
126
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
2.4. fMRI sequences MRI was carried out using a 1.5 T Philips high-speed scanner with a multi-channel head coil. Anatomical scans were collected using a high-resolution gradient echo T1-weighted 3D Fast SPGR anatomical protocol (TE = 5 ms, TR = 140, matrix 256 256, FOV 240 240 mm, 124 slices with 1.5 mm thickness) after a sagittal T1-weighted fast scoot view. The functional scans were collected using a blood oxygen level-dependent contrast (BOLD) protocol with a T2-weighted gradient echo planar imaging (EPI) sequence (TR = 3000, TE = 60 ms, Flip angle = 90°). Each functional scan consisted of a 42 volumes interleaved acquisition (30 axial slices, 5 mm thick, 240 240 mm FOV, 128 128 matrix, 1.88 1.88 mm in-plane resolution). The scanning planes were oriented parallel to the anterior commissure – posterior commissure line and covered the top of the cortex down to the medulla. Subjects were placed in a supine position on the MRI table and made comfortable. They were instructed to close their eyes, relax and stay still for the duration of the procedure. A Velcro band held the subject’s head inside the antenna and limited macro-motion artefacts. Image acquisition time did not exceed 40 min. 2.5. Image processing and analyses All data were processed using statistical parametric mapping 5 (SPM, Welcome Department of Cognitive Neurology, London, UK). Anatomical images were transformed stereotactically to Talairach coordinates using the standard template of the Montreal Neurological Institute. Since the Talairach system is not specific for brainstem structures, the location of activated areas was based on the landmark-based topographical approach described by DaSilva et al. (2002). For each subject, EPI volumes were corrected for slice timing and then realigned on the first volume of the first functional scanning session. Individual T1-to-EPI rigid coregistration was then performed. The T1 image was then segmented/normalised using the SPM5 ‘Segment’ procedure with the ICBM/MNI priors and the resulting normalisation parameters were applied to the coregistered EPI images. In order to pool ipsi- and contra-lateral data within (trigger zone vs. contralateral homologous area stimulation sessions) and across (patients with right- vs. left-sided allodynia) patients, a flip procedure (including the flip itself and the application of a non-flipped to flipped MNI template deformation) was then performed so that all data correspond to left-sided stimulations. Finally, the whole fMRI data set was smoothed with a Gaussian kernel of 8 mm full-width-at-half-maximum. Statistical analyses were conducted on functional images using the general linear model approach on a voxel-by-voxel basis employing a random effects model implemented with a two level procedure. For each functional scanning session corresponding to the three sites (i.e. trigger zone, contralateral side, right hand), stimulation and rest conditions were modelled as boxcar functions at each stimulus onset and length according to stimulation and rest epoch (i.e. six experimental conditions: trigger zone stimulation, trigger zone rest, contralateral stimulation, contralateral rest, right hand stimulation and right hand rest). The ensuing hæmodynamic response was modelled by convolving these boxcar functions with a canonical hæmodynamic response function. Four contrasts were computed for each patient: ‘‘Trigger Zone Stimulation vs. Trigger Zone Rest”, ‘‘Contralateral Stimulation vs. Contralateral Rest”, ‘‘Right Hand Stimulation vs. Right Hand Rest”, the direct contrast ‘‘Trigger Zone Stimulation vs. Trigger Zone Rest > Contralateral Stimulation vs. Contralateral Rest”, and all resulting contrast images were then entered into second-level random effect analyses. First, independent one-sample t-tests were performed for each group of patients (with or without evoked pain) before and after surgery. Second, so as to directly and statistically assess the group
differences before surgery, two-sample t-tests were performed. For all these analyses, only regions surviving an uncorrected p < 0.001 threshold were considered as significant. 3. Results Fifteen consecutive patients (six women, 67.2 ± 9.8 years old) participated in this study. All patients presented with CTN (mean duration: 8.3 ± 6.9 years) without clinically detectable sensory (thermal or tactile) deficits. They suffered both spontaneous attacks (mean = 11 ± 6 per day) and attacks provoked by classical precipitating factors (speaking, swallowing, shaving, yawning, etc.). Additionally, all patients reported an ipsilateral cutaneous trigger zone, within the territory of the 2nd or 3rd division of the Vth nerve (ala nasi (n = 6), upper lip (n = 3) chin (n = 4), cheek (n = 2)). The majority of patients (8 of 15) reported that the ‘efficacy’ of their trigger zone was variable and that tactile stimulation did not systematically elicit a painful sensation. They all received a drug-based treatment including, one anticonvulsant (carbamazepine: 10 patients, oxcarbazepine: 3 patients) or a combination of anticonvulsants (carbamazepine with pregabalin or gabapentin: 2 patients). Upon examination at inclusion, eight patients were scheduled for a radiofrequency thermocoagulation and seven patients had their drug treatment re-adjusted. 3.1. Functional imaging study before surgery All but one patients could discontinue their usual treatment 12 h before fMRI. One patient could not stop his treatment (carbamazepine 1800 mg/day), but halved the dosage. The sensory examination before the scanning session confirmed that stroking within the trigger zone induced a clear (although tolerable) pain sensation (i.e. allodynia) in seven patients. However, in eight patients, stimulation of their usual trigger zone did not elicit any pain or unpleasant sensation (i.e. dysesthesia); but a normal tactile sensation. Both groups of patients (with or without pain) completed the same fMRI protocol. No patient suffered a spontaneous attack during the fMRI session. 3.1.1. Changes in brain activity in patients with allodynia In the seven patients with evoked pain, light tactile stimulation of the trigger zone induced pain with an intensity of 50.8 ± 5.8, described as electric shocks (n = 5), stabbing (n = 1) or shooting (n = 1). This evoked pain was associated with bilateral activation in the primary (S1) and secondary (S2) somatosensory cortices, anterior cingulate cortex (ACC), prefrontal cortex; contralateral activation in the anterior insula, thalamus; ipsilateral activation in the medium cingulate cortex (MCC) and caudal medulla in the spinal trigeminal nucleus (SpV); activation in the medial brainstem area including the periaqueductal grey (PAG) (Fig. 2A, Table 1). Significant activation was also observed in the hippocampal/parahippocampal areas, precentral cortex, supplementary motor area (SMA), putamen and cerebellum (Table 1). Stimulation on the contralateral homologous nonpainful side induced bilateral activation in S1, S2 and precentral area and contralateral activation in SMA (Table 1). The comparative analysis of affected and unaffected sides confirmed that activation induced by stimulation of the affected side was stronger in all the structures mentioned above (Table 2). Activation induced by stimulation of the right hand was confined to S1 contralaterally and S2 bilaterally (results not shown). 3.1.2. Changes in brain activity in patients without allodynia In eight patients without pain on the day of fMRI, light tactile stimulation of their usual trigger zone was associated with
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
127
Fig. 2. Brain areas activated (in yellow) during painful (A: patients with evoked pain) or nonpainful (B: patients without pain) stimulation of the cutaneous trigger zone (i.e., contrast: ‘‘Trigger Zone Stimulation vs. Trigger Zone Rest”) in patients with CTN. Abbreviations: S1: primary somatosensory cortex; S2: secondary somatosensory cortex; ant. insula: anterior insular cortex; post. insula: posterior insular cortex; ACC: anterior cingulate area; MCC: medial cingulate cortex; thal: thalamus; SMA: supplementary motor area; SpV: Spinal trigeminal nucleus. Only regions surviving an uncorrected p < 0.001 threshold were considered as significant.
activation in bilateral S1 and S2 and precentral cortex, contralateral SMA, thalamus, anterior and posterior insula, prefrontal cortex, putamen, ipsilateral MCC, hippocampus/parahippocampus area and cerebellum (Table 1 and Fig. 2B). Stimulation of the contralateral nonaffected side activated S1, S2 and precentral cortex bilaterally, SMA and inferior frontal cortex controlaterally (Table 1). The comparative analysis of affected and unaffected sides confirmed that activation induced by stimulation of the affected side was stronger in all the structures mentioned above (Table 2). Stimulation of the right hand induced activation of S1 contralaterally and S2 bilaterally (results not shown). 3.1.3. Comparison of brain activity between patients with or without allodynia The results of the direct comparison of changes in brain activity associated with stimulation of the affected side between patient
with or without allodynia are summarized in Table 2. This comparison confirmed that stimulation of the affected side in patients with allodynia induced stronger activation in several areas including: contralateral S1, SMA, ACC, prefrontal cortex, brainstem and ipsilateral medulla.
3.2. Functional imaging study after surgery Six of the seven patients who experienced allodynia during the first fMRI session were treated with retrogasserian radiofrequency thermocoagulation of the Gasserian ganglion. Five patients were seen 46.1 ± 12.9 days after surgery for a second fMRI. One patient was lost to follow-up. These five patients reported complete pain relief. Four patients had discontinued their drug-based treatment and one patient was taking a reduced dose of carbamazepine (600 mg/day).
128
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
Table 1 Average z scores and activation during stimulation of the affected (‘‘Trigger Zone Stimulation vs. Trigger Zone Rest”) or unaffected side (‘‘Contralateral Stimulation vs. Contralateral Rest”) in patients with CTN, with evoked during stimulation of the trigger zone or without pain during stimulation of the trigger zone. The coordinates are expressed in millimeters according to the reference lines defined by Talairach and Tournoux (1988): X = medial–lateral: 0 midline, + right; Y = anterior–posterior: 0 anterior commissure, + anterior; Z = superior–inferior: 0 commissural line,+ superior. Abbreviations: ant: anterior; post: posterior; sup: superior; BA = Brodmann area; ipsi: ipsilateral; contra: contralateral; SMA: supplementary motor area; S1: primary motor cortex; S2: secondary somatosensory cortex; ACC: anterior cingulate cortex; M1: primary motor cortex; MCC: medial cingulate cortex; hipp/parahipp: hippocampal/parahippocampal area. Patients with evoked pain Nonaffected side coordinates
Affected side coordinates
Nonaffected side coordinates
z
z
z
BA
Side
z
S1
1, 2, 3
S2
– –
Ipsi Contra Ipsi Contra Ipsi Contra Contra Contra Contra Contra Ipsi Contra Ipsi Ipsi Contra
4.82 5.79 4.70 5.41 5.89 4.17 4.68 4.99 4.11
52 56 58 60 56 48 6 8 36
12 18 20 26 18 6 6 10 10
40 38 38 28 46 44 66 6 5
5.72 4.57 4.37 5.43 5.90 3.99
8 4 8 46 46 48
22 22 22 31 32 44
31 24 48 26 22 2
M1/precentral SMA Thalamus Ant. insula Post. insula ACC MCC Frontal
Temporal sup. Hipp/parahipp Cerebelum Putamen Brainstem Medulla
6 – – – 24
45, 46
47 20 – – – – –
Contra Ipsi Ipsi Contra Ipsi
Patients without pain
Affected side coordinates X
4.27 4.31 5.58 4.62 3.32
Y
26 10 22 2 6
Z
12 74 18 26 40
16 20 6 10 48
X
3.92 5.42 4.56 4.83 3.56 3.58 4.43
Y 56 54 60 58 36 46 4
3.98
Z 20 10 32 19 8 8 2
10
40 40 20 13 60 30 56
64
22
X
Y
Z
4.97 5.78 4.16 5.40 5.30 4.09 4.22 5.34 4.05 4.17
53 59 60 52 42 54 2 14 38 45
13 19 26 14 10 2 4 20 24 11
45 38 16 28 58 40 56 6 4 11
3.77
13
15
57
4.28
36
40
22
4.03 3.44 4.02 4.59
48 34 8 26
16 20 44 6
18 16 12 8
X
Y
Z
5.26 4.28 4.13 5.89 4.29 4.01
55 46 62 56 42 54
17 6 24 22 4 6
43 50 28 22 58 24
4.08
40
44
8
Table 2 Average z scores and comparative activation between stimulation of affected and unaffected sides (i.e., contrast: ‘‘Trigger Zone Stimulation vs. Trigger Zone Rest > Contralateral Stimulation vs. Contralateral Rest”) in patients with (first column) or without (second column) evoked pain. The third column shows the comparative activation between the two groups of patients (i.e. two-sample t-tests were performed on images resulting from the contrast described above). Abbreviations: ant: anterior; post: posterior; inf: inferior; sup: superior; med: medial; SMA: supplementary motor area; S1: primary somatosensory cortex; S2: secondary somatosensory cortex; ACC: anterior cingulate area, M1: primary motor cortex; MCC: medial cingulare cortex; hipp/parahipp: hippocampal/parahippocampal area. Affected side vs unaffected side
Patients with evoked pain vs patients without pain
Patients with evoked pain
Patients without pain
Coordinates z
Coordinates
X
Y
Z
z
S1 S2
4.32
48
8
38
Parietal inf. M1/precentral SMA Thalamus Ant. insula Post. insula ACC MCC Frontal
4.83 3.75 4.74 3.32 3.93
40 38 2 16 44
46 0 6 20 8
62 36 62 10 6
3.77 4.81 5.31 4.53 3.78 4.27 4.54 3.73 3.59 3.57 3.92
6 14 50 46 40 20 30 36 28 2 4
20 22 20 30 20 34 18 52 2 26 38
26 46 2 22 22 6 8 36 6 12 48
Temporal sup. Hipp/parahipp Cerebelum Putamen Brainstem Medulla
Coordinates
X
Y
Z
z
4.51 4.71 3.21
40 52 50
28 28 24
58 14 16
3.35 4.06 3.50 4.83 4.27
42 14 22 46 44
10 12 16 9 12
58 6 0 3 12
4.25 4.13 3.62 3.89 3.45 3.86 3.74 3.80
2 40 56 52 34 34 10 34
4 42 8 6 26 20 54 0
36 16 10 10 8 16 6 0
All patients presented with a thermal (warm and cold) hypoesthesia and lower tactile sensitivity in the V2–V3 territory on the operated side. However, they could still clearly feel the tactile stimulation during the fMRI, but with a lower intensity on the operated side than on the nonoperated side. Stimulation of the operated and nonoperated sides induced similar changes in brain activity including bilateral activations in
X
Y
Z
3.28
62
32
48
3.20
6
22
62
4.16
0
18
26
3.99 3.17
50 42
14 54
38 16
3.24
20
24
22
3.30
14
90
32
3.25 3.24
0 8
32 24
4 18
S1 and S2 and contralateral activation in the precentral cortex (Fig. 3). 4. Discussion Pain induced by light tactile stimulation of the trigger zone in patients with CTN was associated with a pathological activation
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
129
Fig. 3. Brain areas activated (in yellow) during stimulation of the former trigger zone on the operated (A) and stimulation of the homologous area on the nonoperated side (B), after successful radiofrequency thermocoagulation of the Gasserian ganglion. Abbreviations: S1: primary somatosensory cortex; S2: secondary somatosensory cortex; precentral: precentral cortex. Only regions surviving an uncorrected p < 0.001 threshold were considered as significant.
(i.e. sensitization) of the whole trigeminal nociceptive systems including the spinal trigeminal nucleus (SpV) and structures involved in pain modulation. Such an abnormal activation, not observed after contralateral or extra-trigeminal (right hand) stimulation, was confined to the ipsilateral trigeminal system and was no longer observed after a successful surgical treatment of CTN. Many functional neuroimaging studies in healthy volunteers have led to the characterization of a network of brain areas forming a ‘‘pain matrix”, including the primary and secondary somatosensory cortices (S1, S2), the insular cortex (IC), the anterior cingulate cortex (ACC), the thalamus and the prefrontal cortex (PFC), which are involved in the different dimensions of pain perception (Treede et al., 1999; Peyron et al., 2000; Davis, 2000; Apkarian et al., 2005; Tracey and Mantyh, 2007). Although probably overly simplistic (Mouraux and Iannetti, 2009), this conceptual model has largely prevailed in the debates about the brain correlates of pain over the last 15 years. However, it is unclear whether pathological pain conditions, notably neuropathic pain, involve the same brain network (Kupers and Kehlet, 2006; Moisset and Bouhassira, 2007). Indeed, some studies in patients with peripheral or central neuropathic pain have reported that brush-evoked allodynia was not associated with activation of major components of the pain matrix such as the ACC or insula (Peyron et al., 1998, 2004; Ducreux et al., 2006; Witting et al., 2006), whereas these structures were activated in other studies (Schweinhardt et al., 2006; Becerra et al., 2006). In addition to methodological differences, these discrepancies may be explained by the high level of heterogeneity of patients included in these studies. In particular, the presence of various types and varying severity of sensory deafferentation may have complicated the interpretation of the results (Moisset
and Bouhassira, 2007). Another confounding factor was that in most of patients, provoked pain was associated with spontaneous continuous pain of variable intensity, potentially masking detection of activation related to allodynia. The different, but also overlapping, patterns of brain activation due to spontaneous and evoked pains has been recently investigated in patients with neuropathic pain (Geha et al., 2008). Our study of patients with ‘‘pure” evoked pain (i.e. without spontaneous pain nor sensory deficits), clearly demonstrates that pathological pain provoked by tactile stimuli may be associated with activation of all the major brain areas included in the ‘‘pain neuromatrix”. One limitation of our study, however, was that we did not investigate the changes in brain activity associated with ‘‘normal pain” in our patients, which could have revealed changes in the ‘‘pain matrix” associated with CTN. Other limitations include possible movement artefacts and the fact that the stimuli were applied manually whithout a strict control of the pressure by an investigator creating a possible magnetic inhomogeneity. Also the relatively small number of patients did not allow corrections for multiple comparisons. However, due to the strong consistency of our data with previous pain neuroimaging studies, it is highly unlikely that our results were only due to artefacts or nonspecific effects. Activation of the trigeminal nociceptive systems has previously been investigated in healthy volunteers (DaSilva et al., 2002; Borsook et al., 2003; Mainero et al., 2007) and in patients with trigeminal postraumatic nerve injury (Becerra et al., 2006). However, brain correlates of attacks of CTN have only been reported in one previous study in one single patient (Borsook et al., 2007). Here, ipsilateral activation of SpV was only observed during painful attacks of CTN, not during tactile stimulation of the contralateral nonaffected side, or stimulation of the affected side in patients
130
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
without pain. Thus, SpV activation was specifically related to evoked pain and may well reflect the hyperexcitability (i.e. sensitization) of nociceptive trigeminal neurons. Animal studies have long suggested a major role for central sensitization of spinal or trigeminal neurons in allodynia associated with nerve lesions (Woolf and Mannion, 1999). The development of central sensitization involves a number of molecular and cellular changes induced by peripheral nerve injury (Woolf and Salter, 2000; Baron, 2006). Whether such mechanisms also apply to CTN remains unclear. In many patients, CTN may be related to the chronic compression of the trigeminal roots in the brainstem by adjacent vascular loops (Jannetta, 1967; Fromm et al., 1984). This chronic ‘‘irritation” of the nerve root may result in focal demyelination and the generation of aberrant activities (i.e. ectopic discharges) and pathological cross-activation between large and small afferent fibers in the Gasserian ganglion or trigeminal nerve roots (Rappaport and Devor, 1994; Devor et al., 2002). These peripheral abnormalities may induce secondary central changes resulting in hyperexcitability (sensitization) of nociceptive neurons in the trigeminal nucleus and/or higher brain structures (Fromm et al., 1984; Dubner et al., 1987). Consistent with this notion, in our patients, ipsilateral sensitization of SpV was no longer observed following successful thermocoagulation of the Gasserian ganglion, confirming that it depended and/ or was maintained by peripheral alterations. CTN attacks were also associated with significant activation of brain areas involved in pain modulation (Basbaum and Fields, 1984; Millan, 2002), notably in the brainstem (PAG and adjacent structures). Our data suggest a preferential activation of the dorsal part of PAG. However, this should be confirmed with higher resolution imaging, which would also allow observation of more specific changes in activity in PAG and adjacent structures. Early reports highlighted the role of PAG in descending inhibitory controls of spinal or trigeminal nociceptive neurons activity (Basbaum and Fields, 1984); however, the modulatory role of the PAG is probably more complex. More recent studies in both animals and humans, have shown that brainstem structures including the PAG – notably through its connection to the rostral ventromedial medulla – may also be involved in the development and/or maintenance of persistant pain and central sensitization (Vanegas and Schaible, 2004; Zambreanu et al., 2005; Bingel and Tracey, 2008). In particular, brainstem activation was specifically associated with the maintenance of mechanical allodynia in the area of secondary hyperalgesia induced by capsaicine administration, an experimental model of central sensitization (Lee et al., 2008). Thus, activation of the PAG and adjacent structures in CTN may correspond to compensatory mechanisms and reflect abnormal (i.e. ineffective) overactivation of inhibitory processes and/or correspond to pain facilitatory processes. Brainstem areas are also connected to numerous brain regions including the ACC, insula and prefrontal cortex and also participate in higher level top-down pain modulation (Casey, 1999; Lorenz et al., 2003; Bingel and Tracey, 2008). The concomitant activation of brainstem, ACC and prefrontal cortex, which were only activated during painful stimulation in our patients, emphasizes the role of pain modulatory systems in evoked attacks of CTN, possibly in the development and/or maintenance of sensitization of trigeminal sensory neurons. Consistent with the results of previous studies, a number of other cortical and subcortical areas were also activated in our patients (Apkarian et al., 2005; Kupers and Kehlet, 2006; Tracey and Mantyh, 2007). These structures included, most notably, the motor/premotor cortex, the supplementary motor area, the caudate and putamen, the hippocampal/parahippocampal area and the cerebellum. Although these structures may participate directly in pain perception, their activation may also be related to various contextual features of the experimental setting (Tracey and Mantyh, 2007). Interestingly, most of these structures (S1,
S2, insula, M1, SMA, thalamus, mid-cingulum, frontal areas, hippocampus, putamen, cerebellum) were also activated in CTN patients without evoked pain on the day of the experiment. One cannot exclude that the absence of pain and the apparent different changes in brain activity were related, in some of these patients, to the effect of the treatment which was stopped only 12 h before the experiment. In any case, changes in activity in these structures might reflect a maintained sensitization of parts of the trigeminal somatosensory systems. Such a chronic sensitization has been suggested by electrophysiological studies in patients with trigeminal neuropathic pain (Obermann et al., 2007), although mostly in patients with atypical trigeminal neuralgia (i.e. with spontaneous pain). By contrast, in a recent study in patients with CTN (Blatow et al., 2009), no significant differences were observed between activation in S1, S2 and thalamus induced by stimulation of the affected or unaffected sides. However, it is unclear whether stimuli were applied in a trigger zone. Alternatively, the pattern of brain activation observed in our patients without evoked pain may depend also on psychological and/ or cognitive factors related to the specific salience of stimulation of a potential painful area (i.e. the trigger zone). Several studies have consistently shown, mostly in healthy volunteers, that various cognitive and psychological factors (e.g. anxiety, attention, expectation, anticipation of pain) can modulate brain activation associated with pain (Porro, 2003; Kupers et al., 2005; Rainvil, 2002; Wiech et al., 2008). Interestingly, in our patients, ‘‘ineffective” stimulation of the trigger zone was associated with significant changes in S1, anterior insula, mid-cingulate but not in the ACC, a pattern of activation which has been associated with uncertain expectation of pain (Ploghaus et al., 2003). In this respect it would be of interest to address further the role of pain anticipation/expectation in future studies in CTN by using an event related paradigm allowing to compare directly in a single patient the activation associated with painful or nonpainful stimuli.
Acknowledgment Xavier Moisset and Nicolas Villain were supported by the INSERM MD–PhD program. The authors have no conflict of interest related to this study.
References Apkarian AV, Bushnell MC, Treede RD, Zubieta JK. Human brain mechanisms of pain perception and regulation in health and disease. Eur J Pain 2005;9:463–84. Attal N, Fermanian C, Fermanian J, Lanteri-Minet M, Alchaar H, Bouhassira D. Neuropathic pain: are there distinct subtypes depending on the aetiology or anatomical lesion? Pain 2008;138:343–53. Baron R. Mechanisms of disease: neuropathic pain – a clinical perspective. Nat Clin Pract Neurol 2006;2:95–106. Basbaum AI, Fields HL. Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu Rev Neurosci 1984;7:309–38. Becerra L, Morris S, Bazes S, Gostic R, Sherman S, Gostic J, et al. Trigeminal neuropathic pain alters responses in CNS circuits to mechanical (brush) and thermal (cold and heat) stimuli. J Neurosci 2006;26:10646–57. Bingel U, Tracey I. Imaging CNS modulation of pain in humans. Physiology (Bethesda) 2008;23:371–80. Blatow M, Nennig E, Sarpaczki E, Reinhardt J, Schlieter M, Herweh C, et al. Altered somatosensory processing in trigeminal neuralgia. Hum Brain Mapp 2009;30:3495–508. Borsook D, DaSilva AF, Ploghaus A, Becerra L. Specific and somatotopic functional magnetic resonance imaging activation in the trigeminal ganglion by brush and noxious heat. J Neurosci 2003;23:7897–903. Borsook D, Moulton EA, Pendse G, Morris S, Cole SH, Aiello-Lammens M, et al. Comparison of evoked vs. spontaneous tics in a patient with trigeminal neuralgia (tic doloureux). Mol Pain 2007;6:3–34. Bouhassira D, Attal N, Fermanian J, Alchaar H, Gautron M, Masquelier E, et al. Development and validation of the Neuropathic Pain Symptom Inventory. Pain 2004;108:248–57. Casey KL. Forebrain mechanisms of nociception and pain: analysis through imaging. Proc Natl Acad Sci USA 1999;96:7668–74.
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131 DaSilva AF, Becerra L, Makris N, Strassman AM, Gonzalez RG, Geatrakis N, et al. Somatotopic activation in the human trigeminal pain pathway. J Neurosci 2002;22:8183–92. Davis KD. The neural circuitry of pain as explored with functional MRI. Neurol Res 2000;22:313–7. Devor M, Amir R, Rappaport ZH. Pathophysiology of trigeminal neuralgia: the ignition hypothesis. Clin J Pain 2002;18:4–13. Dubner R, Sharav Y, Gracely RH, Price DD. Idiopathic trigeminal neuralgia: sensory features and pain mechanisms. Pain 1987;31:23–33. Ducreux D, Attal N, Parker F, Bouhassira D. Mechanisms of central neuropathic pain: a combined psychophysical and fMRI study in syringomyelia. Brain 2006;129:963–76. Fromm GH, Terrence CF, Maroon JC. Trigeminal neuralgia: current concepts regarding etiology and pathogenesis. Arch Neurol 1984;41:1204–7. Geha PY, Baliki MN, Wang X, Harden RN, Paice JA, Apkarian AV. Brain dynamics for perception of tactile allodynia (touch-induced pain) in postherpetic neuralgia. Pain 2008;138:641–56. Headache Classification Subcommittee of the International Headache Society. The international classification of headache disorders. 2nd ed.. Cephalalgia 2004;24(Suppl. 1):9–160. Jannetta PJ. Arterial compression of the trigeminal nerve at the pons in patients with trigeminal neuralgia. J Neurosurg 1967;26(Suppl.):159–62. Kupers R, Faymonville ME, Laureys S. The cognitive modulation of pain: hypnosisand placebo-induced analgesia. Prog Brain Res 2005;150:251–69. Kupers R, Kehlet H. Brain imaging of clinical pain states: a critical review and strategies for future studies. Lancet Neurol 2006;5:1033–44. Lee MC, Zambreanu L, Menon DK, Tracey I. Identifying brain activity specifically related to the maintenance and perceptual consequence of central sensitization in humans. J Neurosci 2008;28:11642–9. Lopez BC, Hamlyn PJ, Zakrzewska JM. Systematic review of ablative neurosurgical techniques for the treatment of trigeminal neuralgia. Neurosurgery 2004;54:973–82. Lorenz J, Minoshima S, Casey KL. Keeping pain out of mind: the role of the dorsolateral prefrontal cortex in pain modulation. Brain 2003;126:1079–91. Mainero C, Zhang WT, Kumar A, Rosen BR, Sorensen AG. Mapping the spinal and supraspinal pathways of dynamic mechanical allodynia in the human trigeminal system using cardiac-gated fMRI. Neuroimage 2007;35:1201–10. Millan MJ. Descending control of pain. Prog Neurobiol 2002;66:355–474. Moisset X, Bouhassira D. Brain imaging of neuropathic pain. Neuroimage 2007;37(Suppl. 1):S80–8. Mouraux A, Iannetti GD. Nociceptive laser-evoked brain potentials do not reflect nociceptive-specific neural activity. J Neurophysiol 2009;101:3258–69. Obermann M, Yoon MS, Ese D, Maschke M, Kaube H, Diener HC, et al. Impaired trigeminal nociceptive processing in patients with trigeminal neuralgia. Neurology 2007;69:835–41. Peyron R, Garcia-Larrea L, Gregoire MC, Convers P, Lavenne F, Veyre L, et al. Allodynia after lateral-medullary (Wallenberg) infarct. A PET study. Brain 1998;121:345–56.
131
Peyron R, Laurent B, Garcia-Larrea L. Functional imaging of brain responses to pain. A review and meta-analysis. Neurophysiol Clin 2000;30:263–88. Peyron R, Schneider F, Faillenot I, Convers P, Barral FG, Garcia-Larrea L, et al. An fMRI study of cortical representation of mechanical allodynia in patients with neuropathic pain. Neurology 2004;63:1838–46. Ploghaus A, Becerra L, Borras C, Borsook D. Neural circuitry underlying pain modulation: expectation, hypnosis, placebo. Trends Cogn Sci 2003;7: 197–200. Porro CA. Functional imaging and pain: behavior, perception, and modulation. Neuroscientist 2003;9:354–69. Rainville P. Brain mechanisms of pain affect and pain modulation. Curr Opin Neurobiol 2002;12:195–204. Rappaport ZH, Devor M. Trigeminal neuralgia: the role of self-sustaining discharge in the trigeminal ganglion. Pain 1994;56:127–38. Schweinhardt P, Glynn C, Brooks J, McQuay H, Jack T, Chessell I, et al. An fMRI study of cerebral processing of brush-evoked allodynia in neuropathic pain patients. Neuroimage 2006;32:256–65. Siegfried J. 500 percutaneous thermocoagulations of the Gasserian ganglion for trigeminal pain. Surg Neurol 1977;8:126–31. Sweet WH, Wepsic JG. Controlled thermocoagualtion of trigeminal ganglion and rootlets for differential destruction of pain fibers. 1. Trigeminal neuralgia. J Neurosurg 1974;21:278–310. Talairach J, Tournoux P. Co-planar Stereotaxic atlas of the human brain: 3dimensional proportional system: an approach to cerebral imaging. New York, NY: Thieme Medical Publishers, Inc; 1998. Tracey I, Mantyh PW. The cerebral signature for pain perception and its modulation. Neuron 2007;55:377–91. Treede RD, Kenshalo DR, Gracely RH, Jones AK. The cortical representation of pain. Pain 1999;79:105–11. Vanegas H, Schaible HG. Descending control of persistent pain: inhibitory or facilitatory? Brain Res Brain Res Rev 2004;46:295–309. Wiech K, Ploner M, Tracey I. Neurocognitive aspects of pain perception. Trends Cogn Sci 2008;12:306–13. Witting N, Kupers RC, Svensson P, Jensen TS. A PET activation study of brushevoked allodynia in patients with nerve injury pain. Pain 2006;120: 145–54. Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000;288:1765–9. Woolf CJ, Mannion RJ. Neuropathic pain: aetiology, symptoms, mechanisms, and management. Lancet 1999;353:1959–64. Zakrzewska JM. Diagnosis and differential diagnosis of trigeminal neuralgia. Clin J Pain 2002;18:14–21. Zambreanu L, Wise RG, Brooks JC, Iannetti GD, Tracey I. A role for the brainstem in central sensitisation in humans. Evidence from functional magnetic resonance imaging. Pain 2005;114:397–407.
Contents lists available at ScienceDirect
European Journal of Pain journal homepage: www.EuropeanJournalPain.com
Functional brain imaging of trigeminal neuralgia Xavier Moisset a, Nicolas Villain b, Denis Ducreux c, Alain Serrie d, Gérard Cunin d, Dominique Valade e, Bernard Calvino f, Didier Bouhassira a,* a
INSERM U-987, Physiopathologie et Pharmacologie Clinique de la Douleur, Hôpital Ambroise Paré, Boulogne-Billancourt F-92100, France INSERM-EPHE-Université de Caen/Basse-Normandie, Unité U923, GIP Cyceron, CHU Côte de Nacre, Caen F-14074, France c Service de Neuroradiologie et CIERM, AP-HP Hôpital de Bicêtre, Le Kremlin Bicêtre F-94275, France d Fédération d’Evaluation et de Traitement de la douleur, Hôpital Lariboisière, Paris F-75475, France e Service Urgences Céphalées, Hôpital Lariboisière, Paris F-75475, France f Laboratoire de Neurobiologie, UMR ESPCI-CNRS 7637, Paris F-75005, France b
a r t i c l e
i n f o
Article history: Received 7 January 2010 Received in revised form 17 May 2010 Accepted 6 June 2010 Available online 6 July 2010 Keywords: Neuropathic pain Functional imaging Trigeminal neuralgia Allodynia
a b s t r a c t We used functional magnetic resonance imaging (fMRI) to analyze changes in brain activity associated with stimulation of the cutaneous trigger zone in patients with classic trigeminal neuralgia (CTN). Fifteen consecutive patients with CTN in the second or third division of the nerve, were included in this study. The fMRI paradigm consisted of light tactile stimuli of the trigger zone and the homologous contralateral area. Stimulation of the affected side induced pain in seven patients, but was not painful in eight patients on the day of the experiment. Painful stimuli were associated with significantly increased activity in the spinal trigeminal nucleus (SpV), thalamus, primary and secondary somatosensory cortices (S1, S2), anterior cingulate cortex (ACC), insula, premotor/motor cortex, prefrontal areas, putamen, hippocampus and brainstem. Nonpainful stimulation of the trigger zone activated all but three of these structures (SpV, brainstem and ACC). After a successful surgical treatment, activation induced by stimulation of the operated side was confined to S1 and S2. Our data demonstrate the pathological hyperexcitability of the trigeminal nociceptive system, including the second order trigeminal sensory neurons during evoked attacks of CTN. Such sensitization may depend on pain modulatory systems involving both the brainstem (i.e. periaqueductal gray and adjacent structures) and interconnected cortical structures (i.e. ACC). The fact that large portions of the classical ‘pain neuromatrix’ were also activated during nonpainful stimulation of the trigger zone, could reflect a state of maintained sensitization of the trigeminal nociceptive systems in CTN. Ó 2010 European Federation of International Association for the Study of Pain Chapters. Published by Elsevier Ltd. All rights reserved.
1. Introduction Idiopathic or classic trigeminal neuralgia (CTN) is severe unilateral paroxysmal pain in one or more branches of the fifth cranial nerve, typically the maxillary or mandibular branches. CTN is characterized by brief (a few seconds) paroxysmal pain, most frequently described as a ‘stabbing’- or ‘electric-shock’-like pain. Between attacks the patient is pain-free and the standard clinical examination is normal. The paroxysms may be spontaneous or provoked by various precipitating factors (yawning, chewing, etc.) and/or light tactile stimulation of a specific cutaneous area, the ‘‘trigger zone” (Zakrzewska, 2002). Thus, CTN is a unique neuro-
* Corresponding author. Address: INSERM U-987, Centre d’Evaluation et de Traitement de la Douleur, Hôpital Ambroise Paré, 9 Avenue Charles de Gaulle, 92100 Boulogne-Billancourt, France. Tel.: +33 1 49 09 45 56; fax: +33 1 49 09 44 35. E-mail address: [email protected] (D. Bouhassira).
pathic pain syndrome combining paroxysmal pain and evoked pain which might be regarded as a special form of dynamic mechanical allodynia (i.e. pain induced by normally nonpainful tactile stimuli). Pain evoked by tactile stimuli is a major feature of both peripheral and central neuropathic pain syndromes (Attal et al., 2008). Several studies have investigated changes in brain activity associated with this symptom to determine whether this aberrant pain results from an abnormal activation of the physiological ‘‘pain matrix” (Kupers and Kehlet, 2006; Moisset and Bouhassira, 2007). However, the results obtained have been very variable, probably reflecting the heterogeneity of patients in terms of etiology, lesion topography and symptomatology. In particular, variations in the activation of somatosensory systems may have been due to differences in the magnitude of sensory deafferentation, independent of pain. In addition, the fact that a majority of patients presented also with spontaneous continuous pain may have masked the effects of evoked pain.
1090-3801/$36.00 Ó 2010 European Federation of International Association for the Study of Pain Chapters. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ejpain.2010.06.006
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
We used functional magnetic resonance imaging (fMRI) in a group of patients with CTN since this ‘‘pathological model” appears particularly suited to investigating the brain correlates of pain evoked by tactile stimuli. Another unique feature of CTN is that, in contrast with other neuropathic pain syndromes, it can be readily cured by surgery (Lopez et al., 2004). Thus, we also tested our patients after a successful percutaneous radiofrequency thermocoagulation of the Gasserian ganglion to determine whether the changes in brain activity were reversed. 2. Materials and methods This study was approved by the Ambroise Paré Local Ethical Committee. All patients and healthy volunteers gave written informed consent. 2.1. Patients Patients referred to the Lariboisière or Ambroise Paré Hospitals for refractory trigeminal neuralgia and possible indication for surgical treatment were eligible for this study. Inclusion criteria included classical (idiopathic) trigeminal neuralgia (CTN), as defined by IHS criteria (Headache Classification Subcommittee, 2004), located within the maxillary and/or mandibular branch (V2–V3) of the trigeminal nerve, with a trigger zone sensitive to light tactile stimuli within the same territory. For obvious ethical reasons and to limit the risk of facial spasms (i.e. tic douloureux) or other movements during fMRI, stimulation of the trigger zone should evoke a short (i.e. less than 25 s) attack of pain with a moderate bearable intensity (i.e. 4–6 on a visual analog scale). Exclusion criteria were: symptomatic or atypical neuralgia (e.g. presence of spontaneous ongoing pain and/or sensory deficits); severe neuralgia requiring emergency surgery; neuralgia with frequent spontaneous attacks (i.e. >30/day) which could disrupt the scanning session; previous surgical treatment for the neuralgia; presence of other pain; major psychiatric disease. At inclusion (3–7 days before fMRI), the patients had a complete neurological examination including a comparative sensory examination between the normal and affected sides of the face. Thermal sensitivity was assessed with two thermo-rollers (Somedic, Sweden) at constant temperatures of 40 °C (warm) and 25 °C (cold). Mechanical sensitivity (detection and pain thresholds) was assessed with calibrated Von Frey hairs. Dynamic tactile allodynia
125
was assessed (intensity and quality) by stimulation of the trigger zone with a cotton swab. The mean daily number of attacks and their intensity and quality were also assessed with the Neuropathic Pain Symptom Inventory (NPSI) (Bouhassira et al., 2004). This assessment was systematically performed before the first fMRI in all patients and before the second fMRI in patients seen after a percutaneous radiofrequency thermocoagulation of the Gasserian ganglion. 2.2. Surgical procedure A standardized controlled retrogasserian percutaneous radiofrequency thermocoagulation was performed in pharmaco-resistant patients by the same trained investigators (GC and AS) using the classical technique (Sweet and Wepsic, 1974; Siegfried, 1977). The needle was introduced into the foramen ovale under short-term general anesthesia. Its position was controlled with imaging intensifier and modified according to the electrophysiological test carried out after the patient was woken up. The temperature of the tip of the needle was maintained between 65 and 70 °C and the duration of the lesion was adapted to each patient to produce a pinprick hypoalgesia limited to the territory of the affected branch of the trigeminal nerve. 2.3. Experimental design Patients on analgesic medication were asked to discontinue their medication 12 h before their scheduled scanning session. Patients were systematically tested before entering the scanner to check for allodynia and to assess its intensity and duration. The fMRI session consisted of one anatomical and three functional scanning sessions. Each functional scanning session consisted of four acquisition epochs without simulation (27 s long; rest condition) alternating with three epochs (27 s long) with light tactile stimulation of one of the three sites: trigger zone, contralateral homologous side or dorsum of the right hand (as a control of activation of the somatosensory system). The stimuli were applied by the same trained investigator using a cotton swab (attached to a 0.7 m long wood stick to avoid magnetic inhomogeneity) with a frequency of 1 per second (Fig. 1). After each scanning session, the patients rated their mean pain intensity over the three stimulation epochs on a 100 mm visual analog scale (VAS; from 0: ‘‘no pain” to 100: ‘‘worst possible pain”).
Fig. 1. Schematic illustration of the experimental paradigm. Each patient participated in three functional scanning sessions. Each session consisted of four acquisition epochs without simulation (27 s long; rest condition) alternating with three epochs (27 s long) with light tactile stimulation of one of the three sites: (A) trigger zone, (B) contralateral homologous side or (C) dorsum of the right hand. After each scanning session, the patients rated their mean pain intensity over the three stimulation epochs on a 100 mm visual analog scale (VAS; from 0: ‘‘no pain” to 100: ‘‘worst possible pain”).
126
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
2.4. fMRI sequences MRI was carried out using a 1.5 T Philips high-speed scanner with a multi-channel head coil. Anatomical scans were collected using a high-resolution gradient echo T1-weighted 3D Fast SPGR anatomical protocol (TE = 5 ms, TR = 140, matrix 256 256, FOV 240 240 mm, 124 slices with 1.5 mm thickness) after a sagittal T1-weighted fast scoot view. The functional scans were collected using a blood oxygen level-dependent contrast (BOLD) protocol with a T2-weighted gradient echo planar imaging (EPI) sequence (TR = 3000, TE = 60 ms, Flip angle = 90°). Each functional scan consisted of a 42 volumes interleaved acquisition (30 axial slices, 5 mm thick, 240 240 mm FOV, 128 128 matrix, 1.88 1.88 mm in-plane resolution). The scanning planes were oriented parallel to the anterior commissure – posterior commissure line and covered the top of the cortex down to the medulla. Subjects were placed in a supine position on the MRI table and made comfortable. They were instructed to close their eyes, relax and stay still for the duration of the procedure. A Velcro band held the subject’s head inside the antenna and limited macro-motion artefacts. Image acquisition time did not exceed 40 min. 2.5. Image processing and analyses All data were processed using statistical parametric mapping 5 (SPM, Welcome Department of Cognitive Neurology, London, UK). Anatomical images were transformed stereotactically to Talairach coordinates using the standard template of the Montreal Neurological Institute. Since the Talairach system is not specific for brainstem structures, the location of activated areas was based on the landmark-based topographical approach described by DaSilva et al. (2002). For each subject, EPI volumes were corrected for slice timing and then realigned on the first volume of the first functional scanning session. Individual T1-to-EPI rigid coregistration was then performed. The T1 image was then segmented/normalised using the SPM5 ‘Segment’ procedure with the ICBM/MNI priors and the resulting normalisation parameters were applied to the coregistered EPI images. In order to pool ipsi- and contra-lateral data within (trigger zone vs. contralateral homologous area stimulation sessions) and across (patients with right- vs. left-sided allodynia) patients, a flip procedure (including the flip itself and the application of a non-flipped to flipped MNI template deformation) was then performed so that all data correspond to left-sided stimulations. Finally, the whole fMRI data set was smoothed with a Gaussian kernel of 8 mm full-width-at-half-maximum. Statistical analyses were conducted on functional images using the general linear model approach on a voxel-by-voxel basis employing a random effects model implemented with a two level procedure. For each functional scanning session corresponding to the three sites (i.e. trigger zone, contralateral side, right hand), stimulation and rest conditions were modelled as boxcar functions at each stimulus onset and length according to stimulation and rest epoch (i.e. six experimental conditions: trigger zone stimulation, trigger zone rest, contralateral stimulation, contralateral rest, right hand stimulation and right hand rest). The ensuing hæmodynamic response was modelled by convolving these boxcar functions with a canonical hæmodynamic response function. Four contrasts were computed for each patient: ‘‘Trigger Zone Stimulation vs. Trigger Zone Rest”, ‘‘Contralateral Stimulation vs. Contralateral Rest”, ‘‘Right Hand Stimulation vs. Right Hand Rest”, the direct contrast ‘‘Trigger Zone Stimulation vs. Trigger Zone Rest > Contralateral Stimulation vs. Contralateral Rest”, and all resulting contrast images were then entered into second-level random effect analyses. First, independent one-sample t-tests were performed for each group of patients (with or without evoked pain) before and after surgery. Second, so as to directly and statistically assess the group
differences before surgery, two-sample t-tests were performed. For all these analyses, only regions surviving an uncorrected p < 0.001 threshold were considered as significant. 3. Results Fifteen consecutive patients (six women, 67.2 ± 9.8 years old) participated in this study. All patients presented with CTN (mean duration: 8.3 ± 6.9 years) without clinically detectable sensory (thermal or tactile) deficits. They suffered both spontaneous attacks (mean = 11 ± 6 per day) and attacks provoked by classical precipitating factors (speaking, swallowing, shaving, yawning, etc.). Additionally, all patients reported an ipsilateral cutaneous trigger zone, within the territory of the 2nd or 3rd division of the Vth nerve (ala nasi (n = 6), upper lip (n = 3) chin (n = 4), cheek (n = 2)). The majority of patients (8 of 15) reported that the ‘efficacy’ of their trigger zone was variable and that tactile stimulation did not systematically elicit a painful sensation. They all received a drug-based treatment including, one anticonvulsant (carbamazepine: 10 patients, oxcarbazepine: 3 patients) or a combination of anticonvulsants (carbamazepine with pregabalin or gabapentin: 2 patients). Upon examination at inclusion, eight patients were scheduled for a radiofrequency thermocoagulation and seven patients had their drug treatment re-adjusted. 3.1. Functional imaging study before surgery All but one patients could discontinue their usual treatment 12 h before fMRI. One patient could not stop his treatment (carbamazepine 1800 mg/day), but halved the dosage. The sensory examination before the scanning session confirmed that stroking within the trigger zone induced a clear (although tolerable) pain sensation (i.e. allodynia) in seven patients. However, in eight patients, stimulation of their usual trigger zone did not elicit any pain or unpleasant sensation (i.e. dysesthesia); but a normal tactile sensation. Both groups of patients (with or without pain) completed the same fMRI protocol. No patient suffered a spontaneous attack during the fMRI session. 3.1.1. Changes in brain activity in patients with allodynia In the seven patients with evoked pain, light tactile stimulation of the trigger zone induced pain with an intensity of 50.8 ± 5.8, described as electric shocks (n = 5), stabbing (n = 1) or shooting (n = 1). This evoked pain was associated with bilateral activation in the primary (S1) and secondary (S2) somatosensory cortices, anterior cingulate cortex (ACC), prefrontal cortex; contralateral activation in the anterior insula, thalamus; ipsilateral activation in the medium cingulate cortex (MCC) and caudal medulla in the spinal trigeminal nucleus (SpV); activation in the medial brainstem area including the periaqueductal grey (PAG) (Fig. 2A, Table 1). Significant activation was also observed in the hippocampal/parahippocampal areas, precentral cortex, supplementary motor area (SMA), putamen and cerebellum (Table 1). Stimulation on the contralateral homologous nonpainful side induced bilateral activation in S1, S2 and precentral area and contralateral activation in SMA (Table 1). The comparative analysis of affected and unaffected sides confirmed that activation induced by stimulation of the affected side was stronger in all the structures mentioned above (Table 2). Activation induced by stimulation of the right hand was confined to S1 contralaterally and S2 bilaterally (results not shown). 3.1.2. Changes in brain activity in patients without allodynia In eight patients without pain on the day of fMRI, light tactile stimulation of their usual trigger zone was associated with
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
127
Fig. 2. Brain areas activated (in yellow) during painful (A: patients with evoked pain) or nonpainful (B: patients without pain) stimulation of the cutaneous trigger zone (i.e., contrast: ‘‘Trigger Zone Stimulation vs. Trigger Zone Rest”) in patients with CTN. Abbreviations: S1: primary somatosensory cortex; S2: secondary somatosensory cortex; ant. insula: anterior insular cortex; post. insula: posterior insular cortex; ACC: anterior cingulate area; MCC: medial cingulate cortex; thal: thalamus; SMA: supplementary motor area; SpV: Spinal trigeminal nucleus. Only regions surviving an uncorrected p < 0.001 threshold were considered as significant.
activation in bilateral S1 and S2 and precentral cortex, contralateral SMA, thalamus, anterior and posterior insula, prefrontal cortex, putamen, ipsilateral MCC, hippocampus/parahippocampus area and cerebellum (Table 1 and Fig. 2B). Stimulation of the contralateral nonaffected side activated S1, S2 and precentral cortex bilaterally, SMA and inferior frontal cortex controlaterally (Table 1). The comparative analysis of affected and unaffected sides confirmed that activation induced by stimulation of the affected side was stronger in all the structures mentioned above (Table 2). Stimulation of the right hand induced activation of S1 contralaterally and S2 bilaterally (results not shown). 3.1.3. Comparison of brain activity between patients with or without allodynia The results of the direct comparison of changes in brain activity associated with stimulation of the affected side between patient
with or without allodynia are summarized in Table 2. This comparison confirmed that stimulation of the affected side in patients with allodynia induced stronger activation in several areas including: contralateral S1, SMA, ACC, prefrontal cortex, brainstem and ipsilateral medulla.
3.2. Functional imaging study after surgery Six of the seven patients who experienced allodynia during the first fMRI session were treated with retrogasserian radiofrequency thermocoagulation of the Gasserian ganglion. Five patients were seen 46.1 ± 12.9 days after surgery for a second fMRI. One patient was lost to follow-up. These five patients reported complete pain relief. Four patients had discontinued their drug-based treatment and one patient was taking a reduced dose of carbamazepine (600 mg/day).
128
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
Table 1 Average z scores and activation during stimulation of the affected (‘‘Trigger Zone Stimulation vs. Trigger Zone Rest”) or unaffected side (‘‘Contralateral Stimulation vs. Contralateral Rest”) in patients with CTN, with evoked during stimulation of the trigger zone or without pain during stimulation of the trigger zone. The coordinates are expressed in millimeters according to the reference lines defined by Talairach and Tournoux (1988): X = medial–lateral: 0 midline, + right; Y = anterior–posterior: 0 anterior commissure, + anterior; Z = superior–inferior: 0 commissural line,+ superior. Abbreviations: ant: anterior; post: posterior; sup: superior; BA = Brodmann area; ipsi: ipsilateral; contra: contralateral; SMA: supplementary motor area; S1: primary motor cortex; S2: secondary somatosensory cortex; ACC: anterior cingulate cortex; M1: primary motor cortex; MCC: medial cingulate cortex; hipp/parahipp: hippocampal/parahippocampal area. Patients with evoked pain Nonaffected side coordinates
Affected side coordinates
Nonaffected side coordinates
z
z
z
BA
Side
z
S1
1, 2, 3
S2
– –
Ipsi Contra Ipsi Contra Ipsi Contra Contra Contra Contra Contra Ipsi Contra Ipsi Ipsi Contra
4.82 5.79 4.70 5.41 5.89 4.17 4.68 4.99 4.11
52 56 58 60 56 48 6 8 36
12 18 20 26 18 6 6 10 10
40 38 38 28 46 44 66 6 5
5.72 4.57 4.37 5.43 5.90 3.99
8 4 8 46 46 48
22 22 22 31 32 44
31 24 48 26 22 2
M1/precentral SMA Thalamus Ant. insula Post. insula ACC MCC Frontal
Temporal sup. Hipp/parahipp Cerebelum Putamen Brainstem Medulla
6 – – – 24
45, 46
47 20 – – – – –
Contra Ipsi Ipsi Contra Ipsi
Patients without pain
Affected side coordinates X
4.27 4.31 5.58 4.62 3.32
Y
26 10 22 2 6
Z
12 74 18 26 40
16 20 6 10 48
X
3.92 5.42 4.56 4.83 3.56 3.58 4.43
Y 56 54 60 58 36 46 4
3.98
Z 20 10 32 19 8 8 2
10
40 40 20 13 60 30 56
64
22
X
Y
Z
4.97 5.78 4.16 5.40 5.30 4.09 4.22 5.34 4.05 4.17
53 59 60 52 42 54 2 14 38 45
13 19 26 14 10 2 4 20 24 11
45 38 16 28 58 40 56 6 4 11
3.77
13
15
57
4.28
36
40
22
4.03 3.44 4.02 4.59
48 34 8 26
16 20 44 6
18 16 12 8
X
Y
Z
5.26 4.28 4.13 5.89 4.29 4.01
55 46 62 56 42 54
17 6 24 22 4 6
43 50 28 22 58 24
4.08
40
44
8
Table 2 Average z scores and comparative activation between stimulation of affected and unaffected sides (i.e., contrast: ‘‘Trigger Zone Stimulation vs. Trigger Zone Rest > Contralateral Stimulation vs. Contralateral Rest”) in patients with (first column) or without (second column) evoked pain. The third column shows the comparative activation between the two groups of patients (i.e. two-sample t-tests were performed on images resulting from the contrast described above). Abbreviations: ant: anterior; post: posterior; inf: inferior; sup: superior; med: medial; SMA: supplementary motor area; S1: primary somatosensory cortex; S2: secondary somatosensory cortex; ACC: anterior cingulate area, M1: primary motor cortex; MCC: medial cingulare cortex; hipp/parahipp: hippocampal/parahippocampal area. Affected side vs unaffected side
Patients with evoked pain vs patients without pain
Patients with evoked pain
Patients without pain
Coordinates z
Coordinates
X
Y
Z
z
S1 S2
4.32
48
8
38
Parietal inf. M1/precentral SMA Thalamus Ant. insula Post. insula ACC MCC Frontal
4.83 3.75 4.74 3.32 3.93
40 38 2 16 44
46 0 6 20 8
62 36 62 10 6
3.77 4.81 5.31 4.53 3.78 4.27 4.54 3.73 3.59 3.57 3.92
6 14 50 46 40 20 30 36 28 2 4
20 22 20 30 20 34 18 52 2 26 38
26 46 2 22 22 6 8 36 6 12 48
Temporal sup. Hipp/parahipp Cerebelum Putamen Brainstem Medulla
Coordinates
X
Y
Z
z
4.51 4.71 3.21
40 52 50
28 28 24
58 14 16
3.35 4.06 3.50 4.83 4.27
42 14 22 46 44
10 12 16 9 12
58 6 0 3 12
4.25 4.13 3.62 3.89 3.45 3.86 3.74 3.80
2 40 56 52 34 34 10 34
4 42 8 6 26 20 54 0
36 16 10 10 8 16 6 0
All patients presented with a thermal (warm and cold) hypoesthesia and lower tactile sensitivity in the V2–V3 territory on the operated side. However, they could still clearly feel the tactile stimulation during the fMRI, but with a lower intensity on the operated side than on the nonoperated side. Stimulation of the operated and nonoperated sides induced similar changes in brain activity including bilateral activations in
X
Y
Z
3.28
62
32
48
3.20
6
22
62
4.16
0
18
26
3.99 3.17
50 42
14 54
38 16
3.24
20
24
22
3.30
14
90
32
3.25 3.24
0 8
32 24
4 18
S1 and S2 and contralateral activation in the precentral cortex (Fig. 3). 4. Discussion Pain induced by light tactile stimulation of the trigger zone in patients with CTN was associated with a pathological activation
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
129
Fig. 3. Brain areas activated (in yellow) during stimulation of the former trigger zone on the operated (A) and stimulation of the homologous area on the nonoperated side (B), after successful radiofrequency thermocoagulation of the Gasserian ganglion. Abbreviations: S1: primary somatosensory cortex; S2: secondary somatosensory cortex; precentral: precentral cortex. Only regions surviving an uncorrected p < 0.001 threshold were considered as significant.
(i.e. sensitization) of the whole trigeminal nociceptive systems including the spinal trigeminal nucleus (SpV) and structures involved in pain modulation. Such an abnormal activation, not observed after contralateral or extra-trigeminal (right hand) stimulation, was confined to the ipsilateral trigeminal system and was no longer observed after a successful surgical treatment of CTN. Many functional neuroimaging studies in healthy volunteers have led to the characterization of a network of brain areas forming a ‘‘pain matrix”, including the primary and secondary somatosensory cortices (S1, S2), the insular cortex (IC), the anterior cingulate cortex (ACC), the thalamus and the prefrontal cortex (PFC), which are involved in the different dimensions of pain perception (Treede et al., 1999; Peyron et al., 2000; Davis, 2000; Apkarian et al., 2005; Tracey and Mantyh, 2007). Although probably overly simplistic (Mouraux and Iannetti, 2009), this conceptual model has largely prevailed in the debates about the brain correlates of pain over the last 15 years. However, it is unclear whether pathological pain conditions, notably neuropathic pain, involve the same brain network (Kupers and Kehlet, 2006; Moisset and Bouhassira, 2007). Indeed, some studies in patients with peripheral or central neuropathic pain have reported that brush-evoked allodynia was not associated with activation of major components of the pain matrix such as the ACC or insula (Peyron et al., 1998, 2004; Ducreux et al., 2006; Witting et al., 2006), whereas these structures were activated in other studies (Schweinhardt et al., 2006; Becerra et al., 2006). In addition to methodological differences, these discrepancies may be explained by the high level of heterogeneity of patients included in these studies. In particular, the presence of various types and varying severity of sensory deafferentation may have complicated the interpretation of the results (Moisset
and Bouhassira, 2007). Another confounding factor was that in most of patients, provoked pain was associated with spontaneous continuous pain of variable intensity, potentially masking detection of activation related to allodynia. The different, but also overlapping, patterns of brain activation due to spontaneous and evoked pains has been recently investigated in patients with neuropathic pain (Geha et al., 2008). Our study of patients with ‘‘pure” evoked pain (i.e. without spontaneous pain nor sensory deficits), clearly demonstrates that pathological pain provoked by tactile stimuli may be associated with activation of all the major brain areas included in the ‘‘pain neuromatrix”. One limitation of our study, however, was that we did not investigate the changes in brain activity associated with ‘‘normal pain” in our patients, which could have revealed changes in the ‘‘pain matrix” associated with CTN. Other limitations include possible movement artefacts and the fact that the stimuli were applied manually whithout a strict control of the pressure by an investigator creating a possible magnetic inhomogeneity. Also the relatively small number of patients did not allow corrections for multiple comparisons. However, due to the strong consistency of our data with previous pain neuroimaging studies, it is highly unlikely that our results were only due to artefacts or nonspecific effects. Activation of the trigeminal nociceptive systems has previously been investigated in healthy volunteers (DaSilva et al., 2002; Borsook et al., 2003; Mainero et al., 2007) and in patients with trigeminal postraumatic nerve injury (Becerra et al., 2006). However, brain correlates of attacks of CTN have only been reported in one previous study in one single patient (Borsook et al., 2007). Here, ipsilateral activation of SpV was only observed during painful attacks of CTN, not during tactile stimulation of the contralateral nonaffected side, or stimulation of the affected side in patients
130
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131
without pain. Thus, SpV activation was specifically related to evoked pain and may well reflect the hyperexcitability (i.e. sensitization) of nociceptive trigeminal neurons. Animal studies have long suggested a major role for central sensitization of spinal or trigeminal neurons in allodynia associated with nerve lesions (Woolf and Mannion, 1999). The development of central sensitization involves a number of molecular and cellular changes induced by peripheral nerve injury (Woolf and Salter, 2000; Baron, 2006). Whether such mechanisms also apply to CTN remains unclear. In many patients, CTN may be related to the chronic compression of the trigeminal roots in the brainstem by adjacent vascular loops (Jannetta, 1967; Fromm et al., 1984). This chronic ‘‘irritation” of the nerve root may result in focal demyelination and the generation of aberrant activities (i.e. ectopic discharges) and pathological cross-activation between large and small afferent fibers in the Gasserian ganglion or trigeminal nerve roots (Rappaport and Devor, 1994; Devor et al., 2002). These peripheral abnormalities may induce secondary central changes resulting in hyperexcitability (sensitization) of nociceptive neurons in the trigeminal nucleus and/or higher brain structures (Fromm et al., 1984; Dubner et al., 1987). Consistent with this notion, in our patients, ipsilateral sensitization of SpV was no longer observed following successful thermocoagulation of the Gasserian ganglion, confirming that it depended and/ or was maintained by peripheral alterations. CTN attacks were also associated with significant activation of brain areas involved in pain modulation (Basbaum and Fields, 1984; Millan, 2002), notably in the brainstem (PAG and adjacent structures). Our data suggest a preferential activation of the dorsal part of PAG. However, this should be confirmed with higher resolution imaging, which would also allow observation of more specific changes in activity in PAG and adjacent structures. Early reports highlighted the role of PAG in descending inhibitory controls of spinal or trigeminal nociceptive neurons activity (Basbaum and Fields, 1984); however, the modulatory role of the PAG is probably more complex. More recent studies in both animals and humans, have shown that brainstem structures including the PAG – notably through its connection to the rostral ventromedial medulla – may also be involved in the development and/or maintenance of persistant pain and central sensitization (Vanegas and Schaible, 2004; Zambreanu et al., 2005; Bingel and Tracey, 2008). In particular, brainstem activation was specifically associated with the maintenance of mechanical allodynia in the area of secondary hyperalgesia induced by capsaicine administration, an experimental model of central sensitization (Lee et al., 2008). Thus, activation of the PAG and adjacent structures in CTN may correspond to compensatory mechanisms and reflect abnormal (i.e. ineffective) overactivation of inhibitory processes and/or correspond to pain facilitatory processes. Brainstem areas are also connected to numerous brain regions including the ACC, insula and prefrontal cortex and also participate in higher level top-down pain modulation (Casey, 1999; Lorenz et al., 2003; Bingel and Tracey, 2008). The concomitant activation of brainstem, ACC and prefrontal cortex, which were only activated during painful stimulation in our patients, emphasizes the role of pain modulatory systems in evoked attacks of CTN, possibly in the development and/or maintenance of sensitization of trigeminal sensory neurons. Consistent with the results of previous studies, a number of other cortical and subcortical areas were also activated in our patients (Apkarian et al., 2005; Kupers and Kehlet, 2006; Tracey and Mantyh, 2007). These structures included, most notably, the motor/premotor cortex, the supplementary motor area, the caudate and putamen, the hippocampal/parahippocampal area and the cerebellum. Although these structures may participate directly in pain perception, their activation may also be related to various contextual features of the experimental setting (Tracey and Mantyh, 2007). Interestingly, most of these structures (S1,
S2, insula, M1, SMA, thalamus, mid-cingulum, frontal areas, hippocampus, putamen, cerebellum) were also activated in CTN patients without evoked pain on the day of the experiment. One cannot exclude that the absence of pain and the apparent different changes in brain activity were related, in some of these patients, to the effect of the treatment which was stopped only 12 h before the experiment. In any case, changes in activity in these structures might reflect a maintained sensitization of parts of the trigeminal somatosensory systems. Such a chronic sensitization has been suggested by electrophysiological studies in patients with trigeminal neuropathic pain (Obermann et al., 2007), although mostly in patients with atypical trigeminal neuralgia (i.e. with spontaneous pain). By contrast, in a recent study in patients with CTN (Blatow et al., 2009), no significant differences were observed between activation in S1, S2 and thalamus induced by stimulation of the affected or unaffected sides. However, it is unclear whether stimuli were applied in a trigger zone. Alternatively, the pattern of brain activation observed in our patients without evoked pain may depend also on psychological and/ or cognitive factors related to the specific salience of stimulation of a potential painful area (i.e. the trigger zone). Several studies have consistently shown, mostly in healthy volunteers, that various cognitive and psychological factors (e.g. anxiety, attention, expectation, anticipation of pain) can modulate brain activation associated with pain (Porro, 2003; Kupers et al., 2005; Rainvil, 2002; Wiech et al., 2008). Interestingly, in our patients, ‘‘ineffective” stimulation of the trigger zone was associated with significant changes in S1, anterior insula, mid-cingulate but not in the ACC, a pattern of activation which has been associated with uncertain expectation of pain (Ploghaus et al., 2003). In this respect it would be of interest to address further the role of pain anticipation/expectation in future studies in CTN by using an event related paradigm allowing to compare directly in a single patient the activation associated with painful or nonpainful stimuli.
Acknowledgment Xavier Moisset and Nicolas Villain were supported by the INSERM MD–PhD program. The authors have no conflict of interest related to this study.
References Apkarian AV, Bushnell MC, Treede RD, Zubieta JK. Human brain mechanisms of pain perception and regulation in health and disease. Eur J Pain 2005;9:463–84. Attal N, Fermanian C, Fermanian J, Lanteri-Minet M, Alchaar H, Bouhassira D. Neuropathic pain: are there distinct subtypes depending on the aetiology or anatomical lesion? Pain 2008;138:343–53. Baron R. Mechanisms of disease: neuropathic pain – a clinical perspective. Nat Clin Pract Neurol 2006;2:95–106. Basbaum AI, Fields HL. Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu Rev Neurosci 1984;7:309–38. Becerra L, Morris S, Bazes S, Gostic R, Sherman S, Gostic J, et al. Trigeminal neuropathic pain alters responses in CNS circuits to mechanical (brush) and thermal (cold and heat) stimuli. J Neurosci 2006;26:10646–57. Bingel U, Tracey I. Imaging CNS modulation of pain in humans. Physiology (Bethesda) 2008;23:371–80. Blatow M, Nennig E, Sarpaczki E, Reinhardt J, Schlieter M, Herweh C, et al. Altered somatosensory processing in trigeminal neuralgia. Hum Brain Mapp 2009;30:3495–508. Borsook D, DaSilva AF, Ploghaus A, Becerra L. Specific and somatotopic functional magnetic resonance imaging activation in the trigeminal ganglion by brush and noxious heat. J Neurosci 2003;23:7897–903. Borsook D, Moulton EA, Pendse G, Morris S, Cole SH, Aiello-Lammens M, et al. Comparison of evoked vs. spontaneous tics in a patient with trigeminal neuralgia (tic doloureux). Mol Pain 2007;6:3–34. Bouhassira D, Attal N, Fermanian J, Alchaar H, Gautron M, Masquelier E, et al. Development and validation of the Neuropathic Pain Symptom Inventory. Pain 2004;108:248–57. Casey KL. Forebrain mechanisms of nociception and pain: analysis through imaging. Proc Natl Acad Sci USA 1999;96:7668–74.
X. Moisset et al. / European Journal of Pain 15 (2011) 124–131 DaSilva AF, Becerra L, Makris N, Strassman AM, Gonzalez RG, Geatrakis N, et al. Somatotopic activation in the human trigeminal pain pathway. J Neurosci 2002;22:8183–92. Davis KD. The neural circuitry of pain as explored with functional MRI. Neurol Res 2000;22:313–7. Devor M, Amir R, Rappaport ZH. Pathophysiology of trigeminal neuralgia: the ignition hypothesis. Clin J Pain 2002;18:4–13. Dubner R, Sharav Y, Gracely RH, Price DD. Idiopathic trigeminal neuralgia: sensory features and pain mechanisms. Pain 1987;31:23–33. Ducreux D, Attal N, Parker F, Bouhassira D. Mechanisms of central neuropathic pain: a combined psychophysical and fMRI study in syringomyelia. Brain 2006;129:963–76. Fromm GH, Terrence CF, Maroon JC. Trigeminal neuralgia: current concepts regarding etiology and pathogenesis. Arch Neurol 1984;41:1204–7. Geha PY, Baliki MN, Wang X, Harden RN, Paice JA, Apkarian AV. Brain dynamics for perception of tactile allodynia (touch-induced pain) in postherpetic neuralgia. Pain 2008;138:641–56. Headache Classification Subcommittee of the International Headache Society. The international classification of headache disorders. 2nd ed.. Cephalalgia 2004;24(Suppl. 1):9–160. Jannetta PJ. Arterial compression of the trigeminal nerve at the pons in patients with trigeminal neuralgia. J Neurosurg 1967;26(Suppl.):159–62. Kupers R, Faymonville ME, Laureys S. The cognitive modulation of pain: hypnosisand placebo-induced analgesia. Prog Brain Res 2005;150:251–69. Kupers R, Kehlet H. Brain imaging of clinical pain states: a critical review and strategies for future studies. Lancet Neurol 2006;5:1033–44. Lee MC, Zambreanu L, Menon DK, Tracey I. Identifying brain activity specifically related to the maintenance and perceptual consequence of central sensitization in humans. J Neurosci 2008;28:11642–9. Lopez BC, Hamlyn PJ, Zakrzewska JM. Systematic review of ablative neurosurgical techniques for the treatment of trigeminal neuralgia. Neurosurgery 2004;54:973–82. Lorenz J, Minoshima S, Casey KL. Keeping pain out of mind: the role of the dorsolateral prefrontal cortex in pain modulation. Brain 2003;126:1079–91. Mainero C, Zhang WT, Kumar A, Rosen BR, Sorensen AG. Mapping the spinal and supraspinal pathways of dynamic mechanical allodynia in the human trigeminal system using cardiac-gated fMRI. Neuroimage 2007;35:1201–10. Millan MJ. Descending control of pain. Prog Neurobiol 2002;66:355–474. Moisset X, Bouhassira D. Brain imaging of neuropathic pain. Neuroimage 2007;37(Suppl. 1):S80–8. Mouraux A, Iannetti GD. Nociceptive laser-evoked brain potentials do not reflect nociceptive-specific neural activity. J Neurophysiol 2009;101:3258–69. Obermann M, Yoon MS, Ese D, Maschke M, Kaube H, Diener HC, et al. Impaired trigeminal nociceptive processing in patients with trigeminal neuralgia. Neurology 2007;69:835–41. Peyron R, Garcia-Larrea L, Gregoire MC, Convers P, Lavenne F, Veyre L, et al. Allodynia after lateral-medullary (Wallenberg) infarct. A PET study. Brain 1998;121:345–56.
131
Peyron R, Laurent B, Garcia-Larrea L. Functional imaging of brain responses to pain. A review and meta-analysis. Neurophysiol Clin 2000;30:263–88. Peyron R, Schneider F, Faillenot I, Convers P, Barral FG, Garcia-Larrea L, et al. An fMRI study of cortical representation of mechanical allodynia in patients with neuropathic pain. Neurology 2004;63:1838–46. Ploghaus A, Becerra L, Borras C, Borsook D. Neural circuitry underlying pain modulation: expectation, hypnosis, placebo. Trends Cogn Sci 2003;7: 197–200. Porro CA. Functional imaging and pain: behavior, perception, and modulation. Neuroscientist 2003;9:354–69. Rainville P. Brain mechanisms of pain affect and pain modulation. Curr Opin Neurobiol 2002;12:195–204. Rappaport ZH, Devor M. Trigeminal neuralgia: the role of self-sustaining discharge in the trigeminal ganglion. Pain 1994;56:127–38. Schweinhardt P, Glynn C, Brooks J, McQuay H, Jack T, Chessell I, et al. An fMRI study of cerebral processing of brush-evoked allodynia in neuropathic pain patients. Neuroimage 2006;32:256–65. Siegfried J. 500 percutaneous thermocoagulations of the Gasserian ganglion for trigeminal pain. Surg Neurol 1977;8:126–31. Sweet WH, Wepsic JG. Controlled thermocoagualtion of trigeminal ganglion and rootlets for differential destruction of pain fibers. 1. Trigeminal neuralgia. J Neurosurg 1974;21:278–310. Talairach J, Tournoux P. Co-planar Stereotaxic atlas of the human brain: 3dimensional proportional system: an approach to cerebral imaging. New York, NY: Thieme Medical Publishers, Inc; 1998. Tracey I, Mantyh PW. The cerebral signature for pain perception and its modulation. Neuron 2007;55:377–91. Treede RD, Kenshalo DR, Gracely RH, Jones AK. The cortical representation of pain. Pain 1999;79:105–11. Vanegas H, Schaible HG. Descending control of persistent pain: inhibitory or facilitatory? Brain Res Brain Res Rev 2004;46:295–309. Wiech K, Ploner M, Tracey I. Neurocognitive aspects of pain perception. Trends Cogn Sci 2008;12:306–13. Witting N, Kupers RC, Svensson P, Jensen TS. A PET activation study of brushevoked allodynia in patients with nerve injury pain. Pain 2006;120: 145–54. Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000;288:1765–9. Woolf CJ, Mannion RJ. Neuropathic pain: aetiology, symptoms, mechanisms, and management. Lancet 1999;353:1959–64. Zakrzewska JM. Diagnosis and differential diagnosis of trigeminal neuralgia. Clin J Pain 2002;18:14–21. Zambreanu L, Wise RG, Brooks JC, Iannetti GD, Tracey I. A role for the brainstem in central sensitisation in humans. Evidence from functional magnetic resonance imaging. Pain 2005;114:397–407.