Human cerebellar responses to brush and heat stimuli ... - Springer Link

1 downloads 0 Views 4MB Size Report
1PAIN Group, Brain Imaging Center, McLean Hospital, Harvard Medical School, ... Unit, Department of Neurology, Massachusetts General Hospital and Harvard ...
The Cerebellum 2008, 252–272 ORIGINAL ARTICLE

Human cerebellar responses to brush and heat stimuli in healthy and neuropathic pain subjects

D. BORSOOK1, E. A. MOULTON1, S. TULLY1, J. D. SCHMAHMANN2 & L. BECERRA1 1

PAIN Group, Brain Imaging Center, McLean Hospital, Harvard Medical School, and 2Cerebellar Research Group, Ataxia Unit, Department of Neurology, Massachusetts General Hospital and Harvard Medical School, Massachusetts, USA

Abstract Though human pain imaging studies almost always demonstrate activation in the cerebellum, the role of the cerebellum in pain function is not well understood. Here we present results from two studies on the effects of noxious thermal heat and brush applied to the right side of the face in a group of healthy subjects (Group I) and a group of patients with neuropathic pain (Group II) who are more sensitive to both thermal and mechanical stimuli. Statistically significant activations and volumes of activations were defined in the cerebellum. Activated cerebellar structures were identified by colocalization of fMRI activation with the ‘MRI Atlas of the Human Cerebellum’. Functional data (obtained using a 3T magnet) were defined in terms of maximum voxels and volume of activation in the cerebellum. Volume maps were then mapped onto two millimeter serial slices taken through the cerebellum in order to identify activation within regions defined by the activation volume. The data indicate that different regions of the cerebellum are involved in acute and chronic pain processing. Heat produces greater contralateral activation compared with brush, while brush resulted in more ipsilateral/bilateral cerebellar activation. Further, innocuous brush stimuli in healthy subjects produced decreased cerebellar activation in lobules concerned with somatosensory processing. The data also suggest a dichotomy of innocuous stimuli/sensorimotor cerebellum activation versus noxious experience/cognitive/limbic cerebellum activation. These results lead us to propose that the cerebellum may modulate the emotional and cognitive experience that distinguishes the perception of pain from the appreciation of innocuous sensory stimulation.

Key words: trigemino-cerebellar pathways, chronic pain, allodynia, cerebellar muclei, pain, fMRI, BOLD

Introduction The cerebellum may have a role in a number of integrative functions including memory, associative learning, motor control (1–3), and more recently in sensory processing including nociception (4). The structure is traditionally divided into a medial zone involved in somatic and autonomic reflexes as well as complex movements; an intermediate zone that is involved in voluntary movements; and a lateral zone involved in higher order functions such as memory and cognitive functions in association with the cortex including language (5–7). In addition, connections between the cerebellum and the hypothalamus suggest a possible role in autonomic function, as well as a link to limbic structures involved with emotion (8–10). Thus, the cerebellum appears to be involved in integrating motor, sensory, autonomic, and cognitive responses to environmental stimuli including acute and chronic pain. Most fMRI studies of pain show activation in the cerebellum (See Table I) mostly described as either

midline (vermis) or in the cerebellar hemispheres (11,12). One fMRI report specifically evaluated the effects of pain in the cerebellum (13). In the latter, a parametric analysis of the responses to four subjectapplied temperatures, ranging in intensity from innocuous to painful, suggests that stimulus-intensity could be encoded in the vermis and ipsilateral hemispheric lobule VI. In addition, a number of cerebellar areas activated non-discriminately to different temperatures, including the anterior vermis (lobules III-V), contralateral lobule VIII, and bilaterally in hemispheric lobule III-VI. Here we have taken separate studies of mechanical and thermal heat stimuli applied to the region of the right maxillary division of the trigeminal nerve in a group of healthy subjects (Moulton et al., submitted) and in a group of patients with chronic neuropathic pain (14), and mapped cerebellar activations evoked by thermal (heat) and mechanical (brush) stimuli. Cerebellar structures were identified using the atlas of the cerebellum (15). Although the studies were separate (14, Moulton et al., submitted), the

Correspondence: David Borsook, MD, PhD, PAIN Group, Brain Imaging Center, McLean Hospital, 115 Mill Street, Belmont, MA, USA. E-mail: [email protected] ISSN 1473-4222 print/ISSN 1473-4230 online # 2008 Springer Science + Business Media, LLC DOI: 10.1007/s12311-008-0011-6 Online first: 3 April 2008

Pain and the cerebellum

253

Table I. Functional imaging of cerebellar activation in acute and neuropathic pain studies. (A) Experimental pain (evoked stimuli). Paper

Method

Stimulus

Site

Hsieh 1996 Casey 1996

PET PET

Svensson 1997

PET

Xu 1997 Derbyshire 1998 Iadarola 1998 May 1998 Paulson 1998

PET PET PET PET PET

Ethanol injection Thermode Cold water bath Laser Electric Laser Hot water bath Capsaicin injection Capsaicin injection Thermode

R arm L arm L hand L arm L arm L hand R hand L arm R forehead L arm

Becerra 1999 Coghill 1999 Peyron 1999 Becerra 2001 Casey 2001 Coghill 2001 Bingel 2002 Derbyshire 2002 Helmchen 2003 Koyama 2003 Strigo 2003

fMRI PET PET fMRI PET PET fMRI PET fMRI fMRI fMRI

Thermode Thermode Thermode Thermode Thermode Thermode Laser Thermode Thermode Thermode Thermode

Helmchen 2004 Ibinson 2004 Wager 2004

fMRI fMRI fMRI

Wiech 2005 Albuquerque 2006 Choi 2006 Kong 2006 Seminowicz 2006 Staud 2006

fMRI fMRI fMRI fMRI fMRI fMRI

Thermode Electric Electric Thermode Thermode Thermode Hot water bath Thermode Electric Thermode

L hand R arm R/L hand L hand L arm R/L arm R/L hand R hand R hand R calf Esophagus Upper chest R hand R arm R wrist L arm L arm R masseter L finger R arm L arm R foot

Dominant activation I (R) B B I (L) I (L) I (L) I (R) B B I (L – males) B (females) B B B B B I B B B B B B I (R) I (R) I (R) I (L) C (R) I (R) C (R) B C (R) I (R)

Ref (105) (106) (107) (108) (109) (110) (111) (112) (113) (114) (115) (116) (117) (118) (119) (120) (121) (122) (123) (13) (124) (125) (126) (127) (33) (128) (129) (62)

B, bilateral; I, ipsilateral; C, contralateral; (L), left; (R), right.

(B) Studies on cognitive processes affecting pain-related activation. Paper

Method

Cognitive process

Stimulus

Site

Ploghaus 1999 Bantick 2002 Brooks 2002

fMRI fMRI fMRI

Smith 2002 Gracely 2004 Singer 2004 Wager 2004 Jackson 2005 Wiech 2005 Keltner 2006 Moriguchi 2006 Ogino 2006 Seminowicz 2006

fMRI fMRI fMRI fMRI fMRI fMRI fMRI fMRI fMRI fMRI

Anticipation Distraction Attention Distraction Attention Distraction Anticipation Catastrophizing Empathy Placebo Empathy Distraction Expectation Empathy Imagination Distraction

Before heat pain During heat pain During heat pain During heat pain During heat pain During heat pain Before heat pain During pressure pain Partner in pain During heat pain Pain-related pictures During heat pain During heat pain Pain-related pictures Pain-related pictures During electric pain

L hand L hand R hand R hand L hand L hand L hand L finger (Visual) L arm (Visual) L arm L hand (Visual) (Visual) L arm

Effect I (L) + B2 B+ I (R) + I (L) + none B+ I (L) + B+ I2 B+ B2 I (L) + R+ L+ C (R) 2

Ref (96) (130) (131)

(132) (83) (18) (125) (17) (126) (85) (133) (84) (129)

B, bilateral; I, ipsilateral; (L), left; (R), right; +, increased activation; 2, decreased activation.

experimental imaging paradigm was similar across the groups of healthy subjects and patients with a sensitized pain state (see Methods). Both data from human and animal studies strongly implicate the cerebellum in the modulation of pain. In

human studies, for example, transcranial magnetic stimulation of the cerebellum produces alteration of sensory thresholds and attenuation of cold pain sensation (16); functional imaging studies show cerebellar activation to pain stimuli (see above);

254

D. Borsook et al.

(C) Neuropathic pain (Evoked stimuli).

Paper

Method

Pathology

Stimulation

Derbyshire 2002 Albuquerque 2006 Becerra 2006

PET fMRI fMRI

Back pain Burning mouth Neuropathy

Ducreux 2006 Schweinhardt 2006 Witting 2006 Geha 2007

fMRI fMRI PET fMRI

Syringomyelia Neuropathy Neuropathy Post-herpetic

Heat pain/R hand Heat hyperalgesia/R face Cold allodynia/R face Brush allodynia/R face Heat hyperalgesia/R face Cold allodynia/affected hand Brush allodynia/Varied Brush allodynia/Varied Spontaneous pain/Varied Neuralgia

Dominant activation B B B B B I B I (L)

Ref (120) (127) (14)

(134) (135) (136) (137)

B, bilateral; I, ipsilateral, (L), left.

empathetic pain also produces similar cerebellar activation (17,18); and studies of patients with cerebellar damage impairs detection of somatosensory input changes (19). Animal studies indicate that cerebellar activation may be modulated by peripheral afferent sensory inputs (20,21) including pain (22). The trigeminal system has specific direct or collateral inputs into the cerebellum (23–26). For example, in rabbits, the paramedian lobule and the uvula receive independent trigeminal sensory information from neurons located in separate regions of the trigeminal sensory nucleus (23; see Discussion – Trigeminocerebellar connections). Deafferentation of the infraobital branch of the trigeminal nerve has been reported to result in reorganization of regions of the cerebellum including Crus IIa (27). One report suggests that C-fibers (that are activated by a number of noxious stimuli including heat), may reach the cerebellar Purkinje cells through climbing and mossy fibers (28). Taken together, such clinical and preclinical data suggest that there are direct and indirect pathways from sensory systems, including the trigeminal system, to the cerebellum. The salience of trigeminal sensory input from the face is not known but may be interpreted as being important in orienting the face to protective motor planning (for example, the blink reflex, nocifensive response to pain, facial movements). Such processing of sensory signals may be important in motor planning and execution (see 29) in which the cerebellum may play an integrative role. Given that the cerebellum may be important in modulating sensory stimuli and a motor response to such stimuli (30), the evaluation of specific cerebellar functional activity in different anatomical regions in response to noxious and nonnoxious stimuli in humans may provide insights into how this structure may integrate or modulate this type of information. Methods Data was obtained from two experiments using a trigeminal model of pain (for a review, see 31). The

overall approach is shown in Figure 1. In our model, pain was applied to the right side of the face in the territory of the second division of the trigeminal nerve (V2). The two experiments examined: (Group I) acute noxious thermal pain in healthy subjects (32), and (Group II) stimuli applied to the same region of the face in patients with neuropathic pain (14) affecting the second division of the trigeminal nerve. All studies were thus similar or identical in the following aspects – stimulation sites (second division of the trigeminal nerve, V2); use of the same 3T scanner; identical equipment for stimulation (1.661.6 mm thermal contact probe/Peltier; brush); and all experiments were performed by the same research team at the same location. The research protocols were approved by the institutional IRB and all pain paradigms were conducted in accordance with the Declaration of Helsinki.

Subjects All subjects underwent a clinical review of systems and urine screens for recreational drugs including phencyclidine, barbiturates, tetra-hydrocannabinol, opiates, benzodiazepines, amphetamines, and cocaine. (Cortez Diagnostics 7 Drug RapidDip INsta-scan). Subjects who were healthy and tested negative for the drug screens were recruited to the study. Two groups of subjects were recruited to two separate studies. Group I: 12 healthy volunteers (9 men and 3 women), with an average age of 27¡10 years, were recruited for a study on facial cortical somatotopy through advertisements circulated in the Boston metropolitan area. Subjects were imaged once (Moulton et al., in preparation). Group II: For details see (14). Only brush and heat stimuli to the affected side are reported here. Briefly, 6 subjects with right-sided neuropathic pain (1 man and 5 women), with an average age of 49¡8 years, were recruited to the study (see Table A). Subjects were recruited through advertisements placed in the local newspaper or at physicians’ offices (details are

Pain and the cerebellum

255

Figure 1. Top: Summary table of the two groups. Bottom: Summary of data acquisition and anatomical mapping.

provided in a published report, 14). Pain was rated on a visual analogue scale (VAS; see methods). The average spontaneous pain rating was 7.7¡0.6 (mean¡SEM), and the reported prior history of pain evoked by stimuli (e.g., touching the area, clothing etc) was 7.2¡0.42 years. Subjects were imaged twice, with the second session coming 2–6 months after their first imaging session. Subjects attended a screening session where a full medical examination, medical history, and compliance with inclusion/exclusion criteria were performed before enrollment into the study. Medications were discontinued for one dosing interval. Note that most of the subjects were women, and all premenopausal women were scanned during their mid-follicular phase (days 5–7) of their menstrual cycle, since studies suggest that the variation in pain sensitivity

across the menstrual phase is lowest at this time (33) and more consistent with responses in men. Experimental paradigm The experimental paradigm for data acquisition was similar in each group. In addition, the tools used to apply stimuli to the face were identical, and the timing of the application of the brush and thermal stimuli in both groups was similar. Temperatures applied were different for Group I (46˚C) and Group II (threshold-based; see below). Group I: The experimental paradigm consisted of one MR scanning visit. Anatomical scans were collected, followed by separate functional scans for heat (46˚C) and brush applied to the right V2 area of the face.

Table A. Clinical data on patients with trigeminal neuropathic pain.* Subject

Age

Gender

1

54

F

2

39

3

Affected side

Origin of pain

Diagnosis

Right V2

Following Antibiotic Rx

F

Right V2

Car accident

57

F

Right Vi/V2/V3

Herpes Zoster

4

48

F

Right V1/V2

Ski accident

Idiopathic facial Neuropathy Post traumatic Facial neuropathy Post-herpetic Neuralgia Post traumatic Facial neuropathy

5 6

54 41

M F

Right V2 Right V2/V3

Herpes Zoster Car accident

*Adapted from our previous study (14).

Post-herpetic Post traumatic

Pain medications None Tylenol Neurontin Buproprion Setraline Tylenol #3 None Vicodin Paroxetine Clonazepam

256

D. Borsook et al.

Group II: The experimental paradigm comprised of two MR scanning visits, with the second visit occurring 2–3 months after the first. For each visit, anatomical scans were collected, followed by functional scans of sensory stimulation to the face. The sensory stimulation included heat and brush stimuli, presented in separate blocks. During the functional scans, sensory stimulation was applied to either the maxillary division of the trigeminal nerve of the painaffected facial location (V2A), or the mirror location on the unaffected contralateral side (V2U).

Ratings in the scanner. For online ratings of the VAS, subjects used a dial that could be easily rotated between index finger and thumb. An onscreen bar was presented which had ‘no pain’ at one end and ‘max pain’ at the other. By rotating the dial, a cursor could be moved along the bar. During a stimulus, subjects used their left hand to rate their pain. Thus, subjects rated pain using their hand opposite to the side of the face that was stimulated, so as to potentially segregate activity due to sensory vs. motor events. Imaging parameters

Stimuli. Mechanical stimulation. Brush stimuli, applied by a Velcro-topped (soft side) stick, were administered at 1–2 Hz (1–2 strokes per second) within the V2 area. Brush stimuli were applied to the face using a lever system attached to a wooden frame overlying the MRI headcoil. Plastic/nylon materials allowed us to brush the face while standing at the feet of the subject. Thus, there was no manual reaching into the head coil, and all movement occurred inferior to the location of the brain. In Group I, stimuli were applied 3 times, each for a period of 16 sec separated by 30 sec of no stimulus. In Group II, stimuli were also applied 3 times, each for 25 sec and 30 sec of rest. Thermal stimulation. An FDA approved Thermal Sensory Analyzer (TSA, MEDOC, Haifa, Israel) was used to deliver heat stimuli through a probe that has been adapted to rest on the face. The peltier probe is 1.661.6 cm, or about half the size of the thumb. Temperatures applied to the V2 area were different for Group I (46˚C) and II (V2A pain threshold +1˚C). For both groups, the thermode was heated at a rate of 4˚C/sec to the target temperature, which was maintained for 15 sec. The temperature returned to the baseline at a rate of 4˚C/sec to end the stimulus event. The inter-stimulus interval was 30 sec, with 3 total stimulus events. For each group, thermal stimuli were applied using the same timing as described for the brush. The approach for thermal stimulation of the face in the magnet has been used by our group (14,31,32,34) and also recently by others (35). Stimulation of the face in the scanner. Prior to stimuli administration, the right V2 areas of the face to be stimulated (Group I applied to normal skin; Group II applied to painful/neuropathic skin) were marked with a water-soluble pen. A specially designed module that allowed for placement of thermal probes and the ability to apply brush stimuli to the specific regions was used. While in the scanner, the subjects were told to rate the intensity of stimulusevoked pain using an online visual analog scale (VAS) from ‘No pain’ to ‘Max pain’, a method we have previously described (36).

Imaging was carried out in a 3.0 T Siemens Trio scanner. For anatomical localization, an MPRAGE was used (16161.6 mm resolution). For Group I, each functional scan consisted of 33 slices (3.5 mm isotropic resolution) coronally oriented to match the brainstem axis and covering the middle region of the cerebrum. Functional images were acquired in Group I using Gradient Echo (GE) echo planar imaging (EPI) sequence with TR/TE52.5s/30 ms, for a total of 74 volumes per functional scan. For Group II, functional scans were acquired using a GE EPI sequence with isotropic resolution of 3.5 mm, 41 slices (no-gaps) were prescribed obliquely perpendicular to the AC-PC axis, a TR/TE52.5 s/30 ms was used and 128 volumes were acquired per functional scan. For both groups field map images were acquired with the same prescription as for the functional images. These images were then used to correct for susceptibility-induced distortions in functional scans. Note that in the case of healthy subjects the whole cerebellum was scanned, but for the patients approximately 2/3 of the cerebellum was scanned, with the posterior component not included (Figure 2). Due to a limitation of the phase array coil to acquire more slices (i.e., 41 instead of 33) within the same TR, approximately 2/3 of the patients (Group II) cerebellum was scanned wit the posterior component not included (14). Group I had enough coverage with 41 slices to scan the whole cerebellum. Data analysis The image analysis package fsl 3.2 (FMRIB, University of Oxford, UK; www.fmrib.ox.ac.uk/fsl) and in-house programs implemented via MATLAB (Release 7.2, Mathworks Inc., Natick, MA, USA) were used for data processing. For both groups, head motion correction, spatial smoothing (Group I: 6 mm Gaussian kernal; Group II: 5 mm), and prewhitening was performed using fsl tools. For Group I, a high-pass filter of 75 sec was also used. The general linear model (GLM) was utilized (fsl) for individual subject statistical analysis. For Groups I and II a contrast explanatory variable was constructed comparing periods of stimulation with resting periods based on the temperature or brush

Pain and the cerebellum

257

Figure 2. Masks defined from functional scans used for cerebellar activation for Group I – Healthy Controls (Top) and Group II – Patients (Bottom). MNI coordinates were adjusted to match the cerebellar atlas (15) – see text.

stroking temporal profiles (14, Moulton et al., in preparation). For Group I, a mixed-effect analysis was performed using fsl after transformation of individual results to a standard brain (MNI Standard Brain). For Group II, we combined all subjects and both visits to report aggregate results using a fixed effects approach. For both groups, thresholds were determined from a partition of the statistical map distribution utilizing a gaussian mixture model approach (37). This approach determines statistical thresholds from a partition of the statistical map distribution by deconstructing the

overall statistical map into several optimized Gaussian distributions. The group activation map (zstat image) obtained from standard statistical analysis is further classified using a generalized mixture model approach (37). The classification scheme produces a series of classes (active, deactive, belonging to the null distribution). Each voxel has a probability to be in a particular class. Each class (or the others) has a corresponding map in which each voxel depicts the probability that it belongs to its class. Thresholding the activation class map with pw0.5 (standard in classification schemes) allow us

258

D. Borsook et al.

to determine a mask for which voxels can be considered active. The mask is then applied to the original zstat image. In order to align fMRI images with the atlas, fMRI images were transformed. The transformation consisted of flipping the x-coordinate (atlas is in neurological convention while fMRI images are in radiological convention). Furthermore, fMRI data was rendered over a 26262 mm3 template while the atlas was on a 16161 mm3 atlas. The most superior and inferior axial MRI images of the cerebellum were matched with the atlas and labeled accordingly in Figures 2–6 with the actual coordinates in MNI space. Anatomical evaluation of cerebellar activations To measure regions activated in the cerebellum, we first evaluated zmax and the volume of activations. These measures were mapped using an automated approach to identify regions activated (38). Once

zmax and volumes of activation were recorded, coordinates of zmax were transformed onto the cerebellum atlas (15). This was necessary because the MNI standard does not correspond precisely to the Talairach-based cerebellum atlas. Cerebellar activations were then determined using the following approach: activations were grouped into the following domains: Vermis, Hemisphere, and Cerebellar nuclei (dentate, eboliform, globose, and fastigial) as defined in the Atlas (15). Vermal and hemisphere activations were defined in the horizontal plane by examination of the identical slice with the Atlas, which displays specific coordinates in mm (Figure 3). To identify activations of the nuclei, significant activations were evaluated relative to the Atlas in the horizontal, coronal, and sagittal planes (Figure A). In this report, we use ‘cluster’ to refer to a group of activated voxels that by statistical and spatial extension can be considered as a single focus of activation. A cluster is recorded in terms of peak activation coordinates and its volume. If a cluster’s

Figure 3. Cerebellar vermal and hemisphere activations. Top: Serial sections with number indicating z axis as taken from the Atlas of Schmahmann et al. Bottom: Mapping activations using MRI Atlas of Human Cerebellum (15). In this case the figure shows horizontal slice 251 from the atlas and the funtional activations mapped onto the same horizontal anatomical section acquired in the study. Note that the numbers for the functional maps represent n-1 to match the cerebellar atlas cordinates. See text.

Pain and the cerebellum

259

Figure 4. Activations for heat and brush – Group I. Circles mark activation within nuclei. MNI coordinates were adjusted to match the cerebellar atlas (15) – see text.

volume of activation encompassed any of the regions within the cerebellum (i.e., lobule, vermis, or nucleus), that region was considered active within that cluster. We also include cerebellar structures encompassed within the extension of each cluster in a specific region (lobule, vermis or nucleus). Common activations for healthy vs. neuropathic The main focus of this report is to localize specific activation in the cerebellum to thermal and mechanical

stimuli in these two groups, performed in separate studies. Although not an optimal comparison due to confounding issues (viz., un-matched controls for age, gender, pain level, disease state, and medications) we compared thermal and mechanical stimuli for the ‘neuropathic state’ vs. the ‘healthy state’. To identify common activations for brush and for heat for Group I and Group II, overlapping maps were created. Since subjects rated pain intensity using the hand opposite to the side of the face that was stimulated, activity in the cerebellum may also be observed due

260

D. Borsook et al.

Figure 5. Activations for heat – Group II. MNI coordinates were adjusted to match the cerebellar atlas (15) – see text.

to the motor input. The left hand was used to rate both brush and heat. In order to evaluate the potential contribution of the motor task, we evaluated overlapping areas of activation in the cerebellum in each group separately. Results The results are presented as activations for heat and brush in healthy subjects and in patients with neuropathic pain (sensitized state; see 14).

Group I: Healthy subjects Psychophysical ratings. During the functional scan, heat to V2 produced significantly greater pain intensity ratings than the near zero values recorded for brush (paired-t-test t(11)53.72, pv0.01). The mean (¡SE) pain intensity ratings for brush were 0.04¡0.04, while the ratings for heat were 2.59¡0.69. Only one subject reported pain with brushing, with his average rating being 0.5.

Pain and the cerebellum

261

Figure 6. Commonly activated areas for heat and brush for healthy subjects and neuropathic patients. MNI coordinates were adjusted to match the cerebellar atlas (15) – see text.

Activation by heat and brush. Figure 4 shows serial sections (2 mm thick) through the cerebellum matching to the horizontal (z) axis of the cerebellar atlas (15) showing activation following heat (Figure 4 – top) or brush (Figure 4 – bottom) stimulation. Based on the data, tables of activation clusters (for max voxel, volume, coordinates for each max volume) were created. Table IIA shows activations following noxious heat. A total of 6 clusters met the criteria for significance (see Methods). For increased signal, 2 clusters

contralateral to the stimulus and 2 clusters ipsilateral were significant. For the larger volumes or activated (cerebellum_6_R545.92 cm3 more than one cerebellum_6_L51.30 cm3) cerebellar region was observed within each cluster. Two clusters showed significantly decreased activation (Table IIB). For both increased and decreased signal, activation was found in the cerebellar hemispheres (n58) or dentate nucleus (n52). A total of 5 right-sided (5 increased) activations and 5 left-sided activations (3

262

D. Borsook et al.

Figure A. Method for evaluating cerebellar nuclei activations. Top: The location of the cerebellar nuclei within the x, y and z planes (coordinates with atlas limits) defined by the Atlas (Schmahmann et al., 2004). D, dentate nucleus, E, emboliform nucleus; F, fastigial nucleus; and G, globose nucleus. Bottom: Two examples of evaluating activation in nuclei by examination of activation in the x,y and z planes. In both examples the activation maps are mapped onto the MRI and anatomical sections from the Atlas of Schmahmann et al. (15).

Pain and the cerebellum

263

Table II. Healthy subjects (heat). (A) Increased signal change. Coordinates zmax Cluster Cerebellum_6_R

Cerebellum_7b_R Cerebellum_6_L Cerebellum_Crus2_L

Cerebellar regions

Vox Vol cluster

zstat max cluster

VIIB_R CrII_R Dentate_R VI_R VIIB_R CRII_L Dentate_L CRII_L

5.92

4.2898

0.26 1.30

x

y

z

18

254

228

3.3082 3.7647

6 222

274 262

244 234

0.20

3.3097

230

272

240

Cerebellar regions

Vox Vol cluster

zstat max cluster

x

y

z

IV_L VI_L

0.92 0.37

3.1881 2.8163

222 228

238 244

218 228

(B) Decreased signal change. Coordinates zmax Cluster Cerebellum_IV_L Cerebellum_6_L

increased, 2 decreased) were observed in specific cerebellar regions. Following brush applied to the right side of the face, two clusters were observed for both increased (Table IIIA) and decreased (Table IIIB) activation. One cluster for decreased signal change (Cerebelum_4_5_L) was large (volume 37.4 cm3) and encompassed 6 cerebellar regions. For brush, activation was predominantly bilateral in cerebellar regions as noted in Table IIIB. Figure 4 (bottom) shows increased and decreased signal activation following brush through 2 mm serial slices in the cerebellum. A total of 12 activations were observed in the cerebellar hemisphere regions and 2 in the cerebellar nuclei (the latter only for decreased signal).

On closer inspection of Figure 4, heat produces more activation in vermal lobules V and VI compared with brush (region IX). Activations in the cerebellar hemispheres were widely distributed. Of note, decreased activations were observed for brush in region IV and for heat in lobule V, in both cases in the right hemisphere. Some differences in the lamina activated included VIIIA where noxious heat activated seemed to produce greater activation. Group II: Neuropathic pain patients Psychophysical ratings. As reported in Becerra et al. (14), the average spontaneous pain rating of the affected area was 7.7¡0.6 (mean¡SEM), and the reported prior history of pain evoked by stimuli (e.g.,

Table III. Healthy subjects (brush). (A) Increased signal change. Coordinates zmax Cluster

Cerebellar regions

Volume (cm3)

zstat max cluster

x

y

z

Cerebellum_8_R Cerebellum_6_L

VIIIB_R VI_L

0.24 0.38

3.0868 3.2978

18 222

274 268

256 228

Volume (cm3)

zstat max cluster

x

y

Z

37.4

4.75

210

242

216

0

266

236

(B) Decreased signal change – brush. Coordinates zmax Cluster Cerebellum_4_5_L

Vermis_8

Cerebellar regions VIIIB_R/L IX_R/L Dentate_R/L VI_R/L V_R/L IV_R/L VIIIA

0.22

2.623

264

D. Borsook et al.

touching the area, clothing etc) was 7.2¡0.42. Subjects heat pain threshold (measured prior to scanning was 37.7¡0.6 (scan 1) and 39.2¡0.5˚C (scan 2) – indicating that this was below the normal range of around 43˚C. During the scanning the subjects online VAS reporting of pain for brush and heat were 4.0¡0.77 and 5.8¡0.79 respectively. Activation by heat and brush. Figure 5 shows increased (red) and decreased (blue) activation through sequential 2 mm horizontal sections of the cerebellum following heat (Figure 5 – top) and for brush (Figure 5 – bottom) applied to the right V2 division of the face. Table IVA shows increased signal change for clusters in the cerebellum following heat applied to the affected areas on the right side of the face in patients with neuropathic pain. Only one area showed decreased signal (Table IVB). For all activations, regions that met levels of significance for increased activation were mostly contralateral (n511 vs. ipsilateral n55) to the stimulus. Activation was predominantly in the cerebellar hemispheres. The dentate nucleus was activated bilaterally. Brush applied to the affected region of the face in the neuropathic patients produced the most activations (21 clusters; 36 regions) for increased and decreased signal change (Tables VA and VB). Many of the clusters had voxel volumes of w100 (see Table VA). Most of these were ipsilateral or in the vermis (n527 regions) which contrasted to the heat stimulus in these patients.

Comparing Group I and Group II The comparison really reflects differences between mild pain (Group I) and more significant pain (Group II) as a result of allodynia. Figure 6 shows activations in serial 2 mm sections through the cerebellum, common to heat (Figure 6 – top) or brush (Figure 6 – bottom) for healthy and neuropathic patients. As can be seen from the figure, there are only 2–3 activations that overlap. There was overlap in activation in number of cerebellar regions following heat including lobule VI (L[eft]/R[ight]); lobule V (L/R); lobule Crus I (CrI) (L). In the case of brush, common activations were observed in lobule VI (L/R), a small activation in lobule IX, and lobule CrI (L/R). Thus there was a common activation pattern for both stimuli in lobule VI (L/ R) and lobule CrI (L). These common activations may represent motor activity, as subjects were told to rate their pain using their left hand (see Methods).

Discussion In this study we compared activations in the cerebellum following thermal and brush stimuli applied to the same region of the face (V2 or maxillary division of the trigeminal nerve) in two groups of subjects – healthy controls and neuropathic pain patients derived from two separate studies. In the latter, all patients had thermal and dynamic mechanical allodynia (i.e., reported pain to

Table IV. Neuropathic subjects (heat). (A) Increased signal change. Coordinates zmax Cluster Cerebellum_6_L

Cerebellum_Crus1_R No ROI identified Cerebellum_6_L Cerebellum_Crus1_L Cerebellum_Crus1_L Cerebellum_Crus1_L Cerebellum_Crus1_L

Cerebellar regions

Volume (cm3)

zstat max cluster

x

X_L VIIB_R VI_R/L VIIB_L CrI_R Dentate_R/L V_R/L IV_L III CrI_R VIIIB_R VI_L CrI_L CrI_L CrI_L CrI_L

18.53

4.9975

232

236

238

1.02 0.38 0.82 0.26 0.47 0.53 0.20

3.4409 3.3187 3.4799 2.6322 3.5677 3.7231 3.115

42 12 232 238 246 246 254

250 262 266 248 262 258 262

234 262 220 236 226 230 226

y

z

(B) Decreased signal change. Coordinates zmax Cluster Cerebellum_8_L

Cerebellar regions VIIA_L

Volume (cm3)

zstat max cluster

x

y

z

0.78

3.6578

236

260

254

265

Pain and the cerebellum Table V. Neuropathic subjects (brush). (A) Increased signal change. Coordinates zmax Cluster Cerebellum_6_R

Cerebellum_6_R Cerebellum_8_R Cerebellum_4_5_R Vermis_7 Vermis_9 Cerebellum_9_R Vermis_6 Vermis_8 Vermis_4_5 Vermis_4_5 Vermis_8 Cerebellum_4_5_L No ROI identified No ROI identified No ROI identified Cerebellum_8_L No ROI identified Cerebellum_6_L

Cerebellar Regions

Volume (cm3)

zstat max Cluster

VIIB_R CrII_R CrI_R VI_R VI_R VIIIA_R V_R VIIB IX IX_R VI VIIIA IV/V III VIIIA V_L Dentate_L White matter_L Dentate_L Dentate_L VIIIA_L CrII_L CrI_L VI_L

9.67

5.8835

0.28 0.59 0.22 0.65 0.37 0.30 0.74 0.38 0.53 0.42 1.01 1.46 1.77 0.61

x

y

z

32

262

230

3.6028 3.801 3.3338 3.6009 3.6141 3.6149 4.2462 3.9221 4.0999 4.1865 3.9685 4.3947 4.8849 4.4094

38 16 8 6 4 4 2 2 0 0 22 24 216 222

242 266 256 268 260 256 268 262 254 246 264 260 256 254

232 256 218 226 242 248 212 236 212 26 234 216 228 240

1.63 0.89 1.35

5.1305 4.2204 4.3673

222 224 230

254 240 252

232 246 240

0.22

3.0393

234

246

232

(B) Decreased signal change. Coordinates zmax Cluster No ROI identified No ROI identified No ROI identified

Cerebellar Regions

Volume (cm3)

zstat max Cluster

x

y

z

V_R IV_R VIIIB_R IX_R/L VIIIA_R/L Dentate_R/L CrI_R/L IX VI_R/L V_R/L IV_R/L III

0.48 0.26 41.30

0.93465 22.4171 0.97867

26 12 212

232 238 232

234 232 228

normally non-noxious thermal and mechanical (brush) stimuli). Both stimuli produced more pain in the patients than heat and brush in healthy subjects. Overall, heat and brush produced relatively few activations in healthy subjects while a larger number of activations were observed in patients. In the latter group, with brush-induced pain (allodynia) produced more activation than heat. In patients, heat and brush had mirrored asymmetrical activation patterns: noxious heat produced a predominantly contralateral activation in the patient (and healthy) group, while brush produced a greater number of activations in the ipsilateral cerebellar hemispheres (more so in patients than in healthy subjects). A number of studies have reported cerebellar activation following noxious stimuli and

these are summarized in Table I with a summary of reports of cerebellar activation in acute experimental pain (Table IA), cognitive influences on cerebellar activation (Table IB) and in neuropathic pain (Table IC). Note that most of the activations from noxious stimuli were bilateral and or ipsilateral. Our results show that cerebellar activation is observed in distinct regions in response to brush and heat in two different groups as discussed below. Heat and brush activation in healthy subjects: Cognitive vs. sensory processing Our data show that heat produced activation in areas thought to be involved in cognitive processing (39) (i.e., lobules CrII and VIIB) and in sensory-cognitive

266

D. Borsook et al.

processing (VI). However, in healthy subjects brush stimuli produced predominantly decreased signal in areas involved in sensory-motor integration (lobules IV, V and VI), and in secondary sensory processing (lobule VIIIB). Thus, there is a differentiation of increased cognitive cerebellar regional activity for a low heat stimulus vs. a predominant decrease in the sensory processing component for brush. We infer from these results that there is a distinct difference in cerebellar mechanisms integrating information for mild thermal heat vs. brush, the latter producing activation almost completely restricted to cerebellar sensory processing. Heat and brush activation in patients: Cognitive vs. sensory processing In patients, both heat and brush produces pain (see Figure 1) when applied to the area of the face affected by nerve damage resulting in the clinical condition of neuropathic pain. Heat pain produced activation in regions involved in sensory-motor processing (anterior lobe, lobules III–V, and lobules VI and VIIIA), but also included areas involved in cognitive processing (lobule CRI). The activation pattern was not very different from heat in healthy subjects. For brush-induced allodynia, activation in the cerebellum was observed in sensorimotor regions in lobules III to V, putative secondary somatosensory regions (lobule VIII; (40)), vestibular regions (lobule IX), cognitive regions (lobules VIIB, CRI, CRII), and prominent dentate nucleus activation. The distribution of sensory and cognitive activation following brush may result from the nature of the sensation: pain, dysesthesia, differential input by the same fiber type (Aß fibers) that normally conveys the sensation of mechanical stimuli such as brush. Thus, there might be additive inputs from the trigeminal nuclei that connect with various brain regions including the cerebellum in response to mechanical stimuli as well as painful stimuli. In the current study, there is nothing disrupting cerebellar function (i.e., no injury), rather, cognitive cerebellar regions are active in the cognitive state. Thus, the cerebellum appears to be active in its sensorimotor as well as its cognitive and limbic regions for situations in which a subject’s experience of a stimulus is painful. Trigemino-cerebellar pathways Possible pathways mediating noxious (thermal and mechanical) and innocuous (mechanical) information to the cerebellum have been described using electorphysiological and anatomical tracing techniques in mammals and birds (21,25,41–46). Relatively few trigeminal afferents project to the cerebellar nuclei (47) and most project to the cerebellar cortex. Trigeminal afferents are reported to arise in the the nucleus interpolaris,nucleus oralis

and principal nucleus, and project ipsilaterally to the three cerebellar cortical regions – the lobulus simplex and part of lobule V, rostral folia of the paramedian lobule with surrounding parts of crus I and II, and lobule IX (48). Some projections are observed from the mesencepahlic nucleus and nucleus caudalis directed to vermal regions. In a study of trigeminocerebellar projections in sheep, secondary trigeminocerebellar connections have been described in some detail using tract tracing techniques (49). The results indicate most of the cerebellar cortex receives bilateral (but mostly ipsilateral) fibers from the trigeminal nuclei (except flocculus, ventral paraflocculus and lobules I–IV) with some topographic organization (mesencephalic nucleus to the anterior lobe, lobules VI, VIII and dorsal paraflocculus; principal nucleus to all lobules in vermis and hemispheres). These projections have not been described in man. Our results may be interpreted in the light of these studies. For noxious heat, the primary inputs are to the trigeminal nucleus caudalis and interpolaris (50–52), and based on the rat data, projections would be to vermal regions. Non-noxious brush stimuli are transmitted via the main or principal sensory nucleus and have collateral projections to the more caudal nuclei (53– 57). Inputs to the trigeminal nucleus are complex (see (31) for a review) and given patterns we observe (e.g., heat pain activating a disproportionate number of Crus I and Crus II regions and painful brush activating a large number of vermal regions), it is difficult to extrapolate from the animal studies. In addition, many indirect pathways, for example, the trigemino-olivary-cerebellar projection (46) may also contribute. MRI tract-tracing studies may help delineate these pathways in humans (58,59). Pain and the cerebellum Although there are numerous pain studies that report activation in the cerebellum, only one group has reported activation in specific cerebellar regions (13,60). In these two prior studies, the authors employed a similar method for producing heat, but applied heat to the hand in healthy subjects. Activations were reported in the anterior vermis (lobules II–V), and bilaterally in the hemispheres (lobules III–VI). These activations were observed predominantly ipsilateral to the stimulus. The studies were performed in healthy volunteers. Other studies have shown activation in the cerebellum to temporal summation of C-fiber evoked pain, and suggest cerebellar activation may correlate with premotor activity (61,62). In our studies, we observed more widespread activations within the vermis and hemispheres for the patient group and in the nuclei. Based on our current understanding of input-output systems of the cerebellum the activations in the cerebellar cortex are a result of afferent

Pain and the cerebellum sensory inputs (63), including pain (64,65), while activation in the nuclei relate to outputs (66). Decreased activations were also observed in the present study (Tables IB–IVB). If decreased activation represents inhibitory processing (67,68), we interpret these changes as such and relate to inhibitory systems in cerebellar cortical regions. The cerebellum has been called ‘the great modulator of neurologic function’ (69). It has outputs to numerous limbic structures (including the hippocampus, amygdala, intralaminar thalamic nuclei, the hypothalamus, the periaqaueductal gray (70–75) and thus may influence emotion, cognition, and sensory function – all dimensions involved in the response to pain or pain modulation. Having said this, the role of the cerebellum in emotional processing remains controversial (see 76). The role of the cerebellum in nociception has been reviewed by Saab and Willis (4). In preclinical studies, cerebellar stimulation modulates thalamic neural responses to painful stimuli (77) and also modulates the intensity of a visceral nociceptive reflex in rats (78). Conversely, nociceptive stimuli modulate the activity of cerebellar Purkinje cells (22,64,65). Pharmacological manipulation of the cerebellum also produces analgesia. For example, microinjection of morphine into the anterior cerebellum of rats produced analgesia that was reversible by systemic naloxone (79). In addition, the same authors reported that brief electrical stimulation of the same area resulted in analgesia after the stimulation ended. Lesions of the cerebellar vermis in rats results in a larger increase in the threshold to the reaction to electrical shock in rats compared with lesions of the cerebellar cortex (80). Human studies have not been able to produce controlled lesions or stimulation for pain, although cerebellar stimulation has been used for motor disorders (81). Pain imaging studies across the spectrum from visceral (82) to somatic, acute to chronic (83) all show cerebellar activation. Few other structures are activated in such a consistent manner. In addition to responding to direct pain stimulation, activation is reported across a number of pain experiments including imagination of pain (84), modulation of pain (85) and perception of pain in others (17), empathy (18), and acupuncture stimulation (86). Indeed many aspects of sensory discrimination described above pertain to anticipation, error prediction, cognitive responses, attention, and integration of the brain’s response to sensation (87). Little is known about the specific role of the cerebellum in nociceptive processing. Are these systems generic to sensory processing, or is there specific modulation relating to pain and integrating the CNS response including autonomic, cognitive, emotional and sensory dimensions? Although the cerebellum has been considered to be involved in cognitive behaviors, no clinical or preclinical studies

267

have yet shown a role in acute or chronic pain. Recent work has considered many pain responses as learned (88,89) and the role of the cerebellum is still unclear. The possible role of the cerebellum in pain processing may be considered in the light of several theoretical formulations: (i)

The cerebellum may optimize performance by modulating behavior automatically according to context (39) such as the need to integrate appropriate motor function. This modulation may be applied to learned behaviors in pain (sensory) processing, as the cerebellum has been considered to be involved in monitoring and adjusting sensory acquisition (90). The cerebellum is considered to be involved in controlling error signals, and may be a comparator for errors in somatosensory processing (64,65,91,92). Such error signals may play a role, for example, in wrongly executed movement (4). (ii) Cerebellar processing may play an important role in response to expected vs. unexpected events. The cerebellum may play a bigger role in processing unexpected events than expected events. In a recent imaging study of spontaneous trigeminal neuralgia (Borsook et al., under review), no activation was observed in the cerebellum following evoked (expected) tics but activation was observed following spontaneous (unexpected tics). This type of information may infer a role in processing anticipatory sensory input with a high level of temporal accuracy, and optimize temporal responses in the sensory and integrative systems (93,94). For example, measures of cerebellar activity using MEG in response to electric shocks of the median nerve suggests that the signals are probably elicited by the first afferent sensory volley from peripheral nerve endings and mediated by spinocerebellar (cuneocerebellar) tracts (95). The results imply strong coherent activation of cerebellar neuronal populations after purely sensory stimulation based on observation of changes notes measured in millisecond scale temporal resolution. Whether the cerebellum is directly involved in anticipation of pain (96) or plays an integrative role is unknown. (iii) The cerebellum may play an integrative role with respect to sensory information flow to and from the somatosensory cortex. Alterations in somatosensory cortex have been reported in patients with cerebellar lesions (97), and altered sensory processing has also been reported in such patients (19). In the latter study, the authors report that the cerebellum may be involved in pre-attentive detection of incoming somatosensory inputs.

268

D. Borsook et al.

Methodological limitations A number of methodological caveats apply to this report: (i)

(ii)

(iii)

Separate studies. The data for heat and brush were from two different studies and although many processes were the same (3T magnet, number of stimuli, exactly the same method used for applying the stimuli etc), there are a number of differences. Differences in the scanning were present for stimulation sites and number of times subjects were scanned (twice in Group II). However as reported in our paper (14) we did not observe significant differences to pain thresholds prior to scanning, or differences in activation for the two scans. Patients were scanned twice to enhance the power or the study. Un-matched studies. The main data reported relate to separate activation patterns to heat and brush in two groups – healthy and neuropathic pain. As noted data reported here were derived from two prior studies (Moulton et al., in preparation; (14)) that were not specifically designed to compare activation between healthy and neuropathic pain subjects. As a consequence, the groups were not matched for age or gender, or ongoing background pain. While some issues can be addressed in a proper comparison of such data, there are a few that are always problematic when using a patient group. These include the disease state for which there are no good surrogate models (except perhaps capsaicin hyperalgesia, but this does not reflect the underlying changes in neuropathic pain including altered chemical, anatomical and functional processes observed in neuropathic pain (98–100). In addition, patients are usually on or have been on medication for long periods of time, and thus a true comparison is very difficult. What we show in the comparison are essentially areas of activation that are common or different between these groups. One way to interpret this may be the effects of a low level input to heat and brush in Group I and a high level input to heat and brush in Group II. Such a group would still have some inherent inequities e.g., analgesics. Motor function. Another confound may relate to the motor function by left hand movements for rating pain that could have produced activation within the cerebellum, since the cerebellum is well known to be involved in motor integration (101). The common and similar activation sites for both heat and brush stimuli suggest that motor activation is present but does not encompass all of the regions demonstrated in the two cohorts. Further, we

(iv)

(v)

do not see common activation in the dentate nuclei. In an fMRI study, activation in cerebellar output nuclei (dentate) showed increases only when subjects experienced cutaneous stimulation alone but showed little activation in nuclei with combined sensory stimulation and coordinated movements of the hand (see (102)). This and other imaging studies (103,104) suggest that the cerebellum has an important role in sensory discrimination. Cerebellar acquisition. The cerebellum was not fully imaged in the patient cohort, potentially missing activation in the most posterior aspect of the cerebellum. However, healthy subjects, that had full coverage of the cerebellum, did not display activation beyond y5274, the posterior extent for the patient group, and hence our common activation results are likely valid. Differences in pain responses. In addition, heat applied in the control subjects produced mild pain whereas the pain produced by heat in the neuropathic pain subjects was more severe. An increase in the intensity of pain would probably enhance cerebellar activation.

Conclusions Cerebellar sensorimotor regions are activated by touch, whereas cerebellar lobules thought to be involved in cognitive and emotional processing are activated during noxious stimulation both in healthy subjects and in those with neuropathic pain. These results led us to propose that the cerebellum modulates the emotional and cognitive experience that distinguishes the perception of pain from the appreciation of innocuous sensory stimulation. These results add to the weight of preclinical and clinical imaging studies in healthy subjects and patients that demonstrate cerebellar activation by painful stimulation, but a clinical condition in which there is altered pain processing as a result of cerebellar lesions in humans has not yet been described.

Acknowledgements This work was supported by a grant from NINDS (NS 042721).

References 1. Ito M. Cerebellar circuitry as a neuronal machine. Prog Neurobiol. 2006;78:272–303. 2. Schmahmann JD. An emerging concept. The cerebellar contribution to higher function. Arch Neurol. 1991;48: 1178–87.

Pain and the cerebellum 3. Schmahmann JD. Rediscovery of an early concept. Int Rev Neurobiol. 1997;41:3–27. 4. Saab CY, Willis WD. The cerebellum: organization, functions and its role in nociception. Brain Res Brain Res Rev. 2003;42:85–95. 5. Chambers WW, Sprague JM. Functional localization in the cerebellum. II. Somatotopic organization in cortex and nuclei. AMA Arch Neurol Psychiatry. 1955;74:653–80. 6. Leiner HC, Leiner AL, Dow RS. Does the cerebellum contribute to mental skills? Behav Neurosci. 1986;100: 443–54. 7. Schmahmann JD, Sherman JC. Cerebellar cognitive affective syndrome. Int Rev Neurobiol. 1997;41:433–40. 8. Haines DE, Dietrichs E. An HRP study of hypothalamocerebellar and cerebello-hypothalamic connections in squirrel monkey (Saimiri sciureus). J Comp Neurol. 1984;229: 559–75. 9. Levisohn L, Cronin-Golomb A, Schmahmann JD. Neuropsychological consequences of cerebellar tumour resection in children: cerebellar cognitive affective syndrome in a paediatric population. Brain. 2000;123(Pt 5):1041–50. 10. Schutter DJ, van Honk J. The cerebellum on the rise in human emotion. Cerebellum. 2005;4:290–4. 11. 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. 12. Peyron R, Laurent B, Garcia-Larrea L. Functional imaging of brain responses to pain. A review and meta-analysis (2000). Neurophysiol Clin. 2000;30:263–88. 13. Helmchen C, Mohr C, Erdmann C, Binkofski F. Cerebellar neural responses related to actively and passively applied noxious thermal stimulation in human subjects: a parametric fMRI study. Neurosci Lett. 2004;361:237–40. 14. 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. 15. Schmahmann JD, Doyon J, Toga AW, Petrides M, Evans AC. MRI atlas of the human cerebellum. San Diego: Academic Press, 2000. 16. Landgrebe M, Langguth B, Barta W, Hajak G, Eichhammer P. [Modulation of cold-/warm-sensation by cerebellar repetitive transcranial magnetic stimulation (rTMS).]. Psychiatr Prax. 2007;34(1 Suppl.):10–12. 17. Jackson PL, Meltzoff AN, Decety J. How do we perceive the pain of others? A window into the neural processes involved in empathy. Neuroimage. 2005;24:771–9. 18. Singer T, Seymour B, O’Doherty J, Kaube H, Dolan RJ, Frith CD. Empathy for pain involves the affective but not sensory components of pain. Science. 2004;303(5661): 1157–62. 19. Restuccia D, Marca GD, Valeriani M, Leggio MG, Molinari M. Cerebellar damage impairs detection of somatosensory input changes. A somatosensory mismatch-negativity study. Brain. 2007;130(Pt 1):276–87. 20. Hartmann MJ, Bower JM. Tactile responses in the granule cell layer of cerebellar folium crus IIa of freely behaving rats. J Neurosci. 2001;21:3549–63. 21. Holtzman T, Rajapaksa T, Mostofi A, Edgley SA. Different responses of rat cerebellar Purkinje cells and Golgi cells evoked by widespread convergent sensory inputs. J Physiol. 2006;574(Pt 2):491–507. 22. Saab CY, Willis WD. Nociceptive visceral stimulation modulates the activity of cerebellar Purkinje cells. Exp Brain Res. 2001;140:122–6. 23. Bukowska D, Mierzejewska-Krzyzowska B, Zguczynski L. Topography and axonal collaterals of trigeminocerebellar projection to the paramedian lobule and uvula in the rabbit cerebellum. Acta Neurobiol Exp (Wars). 2006;66: 145–51.

269

24. Patrick GW, Robinson MA. Collateral projections from trigeminal sensory nuclei to ventrobasal thalamus and cerebellar cortex in rats. J Morphol. 1987;192:229–36. 25. Somana R, Kotchabhakdi N, Walberg F. Cerebellar afferents from the trigeminal sensory nuclei in the cat. Exp Brain Res. 1980;38:57–64. 26. Steindler DA. Trigeminocerebellar, trigeminotectal, and trigeminothalamic projections: a double retrograde axonal tracing study in the mouse. J Comp Neurol. 1985;237: 155–75. 27. Shumway CA, Morissette J, Gruen P, Bower JM. Plasticity in cerebellar tactile maps in the adult rat. J Comp Neurol. 1999;413:583–92. 28. Jie W, Pei-Xi C. Discharge response of cerebellar Purkinje cells to stimulation of C-fiber in cat saphenous nerve. Brain Res. 1992;581:269–72. 29. Heekeren HR, Marrett S, Ruff DA, Bandettini PA, Ungerleider LG. Involvement of human left dorsolateral prefrontal cortex in perceptual decision making is independent of response modality. Proc Natl Acad Sci USA. 2006;103:10023–8. 30. Molinari M, Leggio MG, Thaut MH. The cerebellum and neural networks for rhythmic sensorimotor synchronization in the human brain. Cerebellum. 2007;6:18–23. 31. Borsook D, Burstein R, Becerra L. Functional imaging of the human trigeminal system: opportunities for new insights into pain processing in health and disease. J Neurobiol. 2004;61:107–25. 32. Moulton E, Pendse G, Morris S, Strassman A, AielloLammens M, Becerra L, et al. Capsaicin-induced thermal hyperalgesia and sensitization in the human trigeminal nociceptive pathway: an fMRI study. NeuroImage. 2007; 35(4):1586–600. 33. Choi JC, Park SK, Kim YH, Shin YW, Kwon JS, Kim JS, et al. Different brain activation patterns to pain and painrelated unpleasantness during the menstrual cycle. Anesthesiology. 2006;105:120–7. 34. 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. 35. de Leeuw R, Davis CE, Albuquerque R, Carlson CR, Andersen AH. Brain activity during stimulation of the trigeminal nerve with noxious heat. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2006;102:750–7. 36. Borras MC, Becerra L, Ploghaus A, Gostic JM, DaSilva A, Gonzalez RG, et al. fMRI measurement of CNS responses to naloxone infusion and subsequent mild noxious thermal stimuli in healthy volunteers. J Neurophysiol. 2004;91: 2723–33. 37. Pendse G, Borsook D, Becerra L. A generalized mixture modeling approach applied to the problem of thresholding fMRI statistical maps. Society for Neuroscience. Atlanta, GA; 2006. 38. Maldjian JA, Laurienti PJ, Kraft RA, Burdette JH. An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets. Neuroimage. 2003;19:1233–9. 39. Schmahmann JD. Disorders of the cerebellum: ataxia, dysmetria of thought, and the cerebellar cognitive affective syndrome. J Neuropsychiatry Clin Neurosci. 2004;16: 367–78. 40. Woolsey T. Summary of papers on the cerebellum. In: Philip B, editor. Patterns of organization in the central nervous system, 1952. pp 334–6. 41. Arends JJ, Zeigler HP. Cerebellar connections of the trigeminal system in the pigeon (Columba livia). Brain Res. 1989;487:69–78. 42. Bukowska D. Trigeminocerebellar projection to the paramedian lobule with emphasis on the climbing fibre zones: a

270

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

D. Borsook et al. retrograde tracing study in the rabbit. J Hirnforsch. 1996;37:159–72. Elias SA, Taylor A, Somjen G. Direct and relayed projection of periodontal receptor afferents to the cerebellum in the ferret. Proc R Soc Lond B Biol Sci. 1987;231(1263): 199–216. Falls WM. Direct connections of primary trigeminal afferent axons with trigeminocerebellar projection neurons in the border zone of rat trigeminal nucleus oralis. Neurosci Lett. 1987;83:247–52. Hayashi H, Sumino R, Sessle BJ. Functional organization of trigeminal subnucleus interpolaris: nociceptive and innocuous afferent inputs, projections to thalamus, cerebellum, and spinal cord, and descending modulation from periaqueductal gray. J Neurophysiol. 1984;51:890–905. Yatim N, Billig I, Compoint C, Buisseret P, BuisseretDelmas C. Trigeminocerebellar and trigemino-olivary projections in rats. Neurosci Res. 1996;25:267–83. Dietrichs E, Walberg F. Cerebellar nuclear afferents – where do they originate? A re-evaluation of the projections from some lower brain stem nuclei. Anat Embryol (Berl). 1987;177:165–72. Jacquin MF, Semba K, Rhoades RW, Egger MD. Trigeminal primary afferents project bilaterally to dorsal horn and ipsilaterally to cerebellum, reticular formation, and cuneate, solitary, supratrigeminal and vagal nuclei. Brain Res. 1982;246:285–91. Saigal RP, Karamanlidis AN, Voogd J, Mangana O, Michaloudi H. Secondary trigeminocerebellar projections in sheep studied with the horseradish peroxidase tracing method. J Comp Neurol. 1980;189:537–53. Amano N, Hu JW, Sessle BJ. Responses of neurons in feline trigeminal subnucleus caudalis (medullary dorsal horn) to cutaneous, intraoral, and muscle afferent stimuli. J Neurophysiol. 1986;55:227–43. Sessle BJ, Greenwood LF. Inputs to trigeminal brain stem neurones from facial, oral, tooth pulp and pharyngolaryngeal tissues: I. Responses to innocuous and noxious stimuli. Brain Res. 1976;117:211–26. Young RF, Perryman KM. Pathways for orofacial pain sensation in the trigeminal brain-stem nuclear complex of the Macaque monkey. J Neurosurg. 1984;61: 563–8. Jacquin MF, Chiaia NL, Klein BG, Rhoades RW. Structurefunction relationships in the rat brainstem subnucleus interpolaris: VI. Cervical convergence in cells deafferented at birth and a potential primary afferent substrate. J Comp Neurol. 1989;283:513–25. Jacquin MF, Mooney RD, Rhoades RW. Morphology, response properties, and collateral projections of trigeminothalamic neurons in brainstem subnucleus interpolaris of rat. Exp Brain Res. 1986;61:457–68. Jacquin MF, Renehan WE, Mooney RD, Rhoades RW. Structure-function relationships in rat medullary and cervical dorsal horns. I. Trigeminal primary afferents. J Neurophysiol. 1986;55:1153–86. Jacquin MF, Rhoades RW. Cell structure and response properties in the trigeminal subnucleus oralis. Somatosens Mot Res. 1990;7:265–88. Renehan WE, Jacquin MF, Mooney RD, Rhoades RW. Structure-function relationships in rat medullary and cervical dorsal horns. II. Medullary dorsal horn cells. J Neurophysiol. 1986;55:1187–201. Mori S, Zhang J. Principles of diffusion tensor imaging and its applications to basic neuroscience research. Neuron. 2006;51:527–39. Yamada K, Nagakane Y, Yoshikawa K, Kizu O, Ito H, Kubota T, et al. Somatotopic organization of thalamocortical projection fibers as assessed with MR tractography. Radiology. 2007;242:840–5.

60. Helmchen C, Rambold H, Sprenger A, Erdmann C, Binkofski F. Cerebellar activation in opsoclonus: an fMRI study. Neurology. 2003;61:412–5. 61. Christmann C, Koeppe C, Braus DF, Ruf M, Flor H. A simultaneous EEG-fMRI study of painful electric stimulation. Neuroimage. 2007;34:1428–37. 62. Staud R, Craggs JG, Robinson ME, Perlstein WM, Price DD. Brain activity related to temporal summation of C-fiber evoked pain. Pain. 2007;129(1–2):130–42. 63. Jorntell H, Ekerot CF. Properties of somatosensory synaptic integration in cerebellar granule cells in vivo. J Neurosci. 2006;26:11786–97. 64. Ekerot CF, Gustavsson P, Oscarsson O, Schouenborg J. Climbing fibres projecting to cat cerebellar anterior lobe activated by cutaneous A and C fibres. J Physiol. 1987;386: 529–38. 65. Ekerot CF, Oscarsson O, Schouenborg J. Stimulation of cat cutaneous nociceptive C fibres causing tonic and synchronous activity in climbing fibres. J Physiol. 1987;386:539–46. 66. Delgado-Garcia JM. [Structure and function of the cerebellum]. Rev Neurol. 2001;33:635–42. 67. Harel N, Lee SP, Nagaoka T, Kim DS, Kim SG. Origin of negative blood oxygenation level-dependent fMRI signals. J Cereb Blood Flow Metab. 2002;22:908–17. 68. Stefanovic B, Warnking JM, Pike GB. Hemodynamic and metabolic responses to neuronal inhibition. Neuroimage. 2004;22:771–8. 69. Snider RS. Recent contributions to the anatomy and physiology of the cerebellum. Arch Neurol Psych, 1950. pp 196–219. 70. Allen G, McColl R, Barnard H, Ringe WK, Fleckenstein J, Cullum CM. Magnetic resonance imaging of cerebellarprefrontal and cerebellar-parietal functional connectivity. Neuroimage. 2005;28:39–48. 71. Braak H, Braak E, Yilmazer D, Bohl J. Functional anatomy of human hippocampal formation and related structures. J Child Neurol. 1996;11:265–75. 72. Middleton FA, Strick PL. Cerebellar output channels. Int Rev Neurobiol. 1997;41:61–82. 73. Onat F, Cavdar S. Cerebellar connections: hypothalamus. Cerebellum. 2003;2:263–9. 74. Schmahmann JD, Pandya DN. The cerebrocerebellar system. Int Rev Neurobiol. 1997;41:31–60. 75. Snider RS, Maiti A. Cerebellar contributions to the Papez circuit. J Neurosci Res. 1976;2:133–46. 76. Manto MU. On the cerebello-cerebral interactions. Cerebellum. 2006;5:286–8. 77. Liu FY, Qiao JT, Dafny N. Cerebellar stimulation modulates thalamic noxious-evoked responses. Brain Res Bull. 1993;30:529–34. 78. Saab CY, Willis WD. Cerebellar stimulation modulates the intensity of a visceral nociceptive reflex in the rat. Exp Brain Res. 2002;146:117–21. 79. Dey PK, Ray AK. Anterior cerebellum as a site for morphine analgesia and post-stimulation analgesia. Indian J Physiol Pharmacol. 1982;26:3–12. 80. Peters M, Bleek C, Monjan AA. Reaction to electrical shock after cerebellar lesions in the rat. Physiol Behav. 1973;10:429–33. 81. Iwata NK, Ugawa Y. The effects of cerebellar stimulation on the motor cortical excitability in neurological disorders: a review. Cerebellum. 2005;4:218–23. 82. Lu CL, Wu YT, Yeh TC, Chen LF, Chang FY, Lee SD, et al. Neuronal correlates of gastric pain induced by fundus distension: a 3T-fMRI study. Neurogastroenterol Motil. 2004;16:575–87. 83. Gracely RH, Geisser ME, Giesecke T, Grant MA, Petzke F, Williams DA, et al. Pain catastrophizing and neural responses to pain among persons with fibromyalgia. Brain. 2004;127(Pt 4):835–43.

Pain and the cerebellum 84. Ogino Y, Nemoto H, Inui K, Saito S, Kakigi R, Goto F. Inner experience of pain: imagination of pain while viewing images showing painful events forms subjective pain representation in human brain. Cereb Cortex. 2007;17(5):1139–46. 85. Keltner JR, Furst A, Fan C, Redfern R, Inglis B, Fields HL. Isolating the modulatory effect of expectation on pain transmission: a functional magnetic resonance imaging study. J Neurosci. 2006;26:4437–43. 86. Yoo SS, Teh EK, Blinder RA, Jolesz FA. Modulation of cerebellar activities by acupuncture stimulation: evidence from fMRI study. Neuroimage. 2004;22:932–40. 87. Ito M. Bases and implications of learning in the cerebellum – adaptive control and internal model mechanism. Prog Brain Res. 2005;148:95–109. 88. Johansen JP, Fields HL. Glutamatergic activation of anterior cingulate cortex produces an aversive teaching signal. Nat Neurosci. 2004;7:398–403. 89. Seymour B, O’Doherty JP, Koltzenburg M, Wiech K, Frackowiak R, Friston K, et al. Opponent appetitiveaversive neural processes underlie predictive learning of pain relief. Nat Neurosci. 2005;8:1234–40. 90. Bower JM. Control of sensory data acquisition. Int Rev Neurobiol. 1997;41:489–513. 91. Apps R, Garwicz M. Anatomical and physiological foundations of cerebellar information processing. Nat Rev Neurosci. 2005;6:297–311. 92. Eccles J, Ito M, Szentagothai J. The cerebellum as a neuronal machine. New York: Springer-Verlag, 1967. 93. Ohyama T, Nores WL, Murphy M, Mauk MD. What the cerebellum computes. Trends Neurosci. 2003;26:222–7. 94. Tesche CD, Karhu JJ. Anticipatory cerebellar responses during somatosensory omission in man. Hum Brain Mapp. 2000;9:119–42. 95. Tesche CD, Karhu J. Somatosensory evoked magnetic fields arising from sources in the human cerebellum. Brain Res. 1997;744:23–31. 96. Ploghaus A, Tracey I, Gati JS, Clare S, Menon RS, Matthews PM, et al. Dissociating pain from its anticipation in the human brain. Science. 1999;284(5422):1979–81. 97. Restuccia D, Valeriani M, Barba C, Le Pera D, Capecci M, Filippini V, et al. Functional changes of the primary somatosensory cortex in patients with unilateral cerebellar lesions. Brain. 2001;124(Pt 4):757–68. 98. Basbaum AI. Distinct neurochemical features of acute and persistent pain. Proc Natl Acad Sci USA. 1999;96:7739–43. 99. Woolf CJ, Mannion RJ. Neuropathic pain: aetiology, symptoms, mechanisms, and management. Lancet. 1999; 353(9168):1959–64. 100. Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science. 2000;288(5472):1765–9. 101. Thach WT, Bastian AJ. Role of the cerebellum in the control and adaptation of gait in health and disease. Prog Brain Res. 2004;143:353–66. 102. Bower JM. Is the cerebellum sensory for motor’s sake, or motor for sensory’s sake: the view from the whiskers of a rat? Prog Brain Res. 1997;114:463–96. 103. Fox PT, Raichle ME, Thach WT. Functional mapping of the human cerebellum with positron emission tomography. Proc Natl Acad Sci USA. 1985;82(21):7462–6. 104. Kim SG, Ugurbil K, Strick PL. Activation of a cerebellar output nucleus during cognitive processing. Science. 1994;265(5174):949–51. 105. Hsieh JC, Stahle-Backdahl M, Hagermark O, StoneElander S, Rosenquist G, Ingvar M. Traumatic nociceptive pain activates the hypothalamus and the periaqueductal gray: a positron emission tomography study. Pain. 1996;64: 303–14. 106. Casey KL, Minoshima S, Morrow TJ, Koeppe RA. Comparison of human cerebral activation pattern during

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

271

cutaneous warmth, heat pain, and deep cold pain. J Neurophysiol. 1996;76:571–81. Svensson P, Minoshima S, Beydoun A, Morrow TJ, Casey KL. Cerebral processing of acute skin and muscle pain in humans. J Neurophysiol. 1997;78:450–60. Xu X, Fukuyama H, Yazawa S, Mima T, Hanakawa T, Magata Y, et al. Functional localization of pain perception in the human brain studied by PET. Neuroreport. 1997;8: 555–9. Derbyshire SW, Jones AK. Cerebral responses to a continual tonic pain stimulus measured using positron emission tomography. Pain. 1998;76:127–35. Iadarola MJ, Berman KF, Zeffiro TA, Byas-Smith MG, Gracely RH, Max MB, et al. Neural activation during acute capsaicin-evoked pain and allodynia assessed with PET. Brain. 1998;121(Pt 5):931–47. May A, Kaube H, Buchel C, Eichten C, Rijntjes M, Juptner M, et al. Experimental cranial pain elicited by capsaicin: a PET study. Pain. 1998;74:61–6. Paulson PE, Minoshima S, Morrow TJ, Casey KL. Gender differences in pain perception and patterns of cerebral activation during noxious heat stimulation in humans. Pain. 1998;76:223–9. Becerra LR, Breiter HC, Stojanovic M, Fishman S, Edwards A, Comite AR, et al. Human brain activation under controlled thermal stimulation and habituation to noxious heat: an fMRI study. Magn Reson Med. 1999;41:1044–57. Coghill RC, Sang CN, Maisog JM, Iadarola MJ. Pain intensity processing within the human brain: a bilateral, distributed mechanism. J Neurophysiol. 1999;82:1934–43. Peyron R, Garcia-Larrea L, Gregoire MC, Costes N, Convers P, Lavenne F, et al. Haemodynamic brain responses to acute pain in humans: sensory and attentional networks. Brain. 1999;122(Pt 9):1765–80. Becerra L, Breiter HC, Wise R, Gonzalez RG, Borsook D. Reward circuitry activation by noxious thermal stimuli. Neuron. 2001;32:927–46. Casey KL, Morrow TJ, Lorenz J, Minoshima S. Temporal and spatial dynamics of human forebrain activity during heat pain: analysis by positron emission tomography. J Neurophysiol. 2001;85:951–9. Coghill RC, Gilron I, Iadarola MJ. Hemispheric lateralization of somatosensory processing. J Neurophysiol. 2001;85:2602–12. Bingel U, Quante M, Knab R, Bromm B, Weiller C, Buchel C. Subcortical structures involved in pain processing: evidence from single-trial fMRI. Pain. 2002;99:313–21. Derbyshire SW, Jones AK, Creed F, Starz T, Meltzer CC, Townsend DW, et al. Cerebral responses to noxious thermal stimulation in chronic low back pain patients and normal controls. Neuroimage. 2002;16:158–68. Helmchen C, Mohr C, Erdmann C, Petersen D, Nitschke MF. Differential cerebellar activation related to perceived pain intensity during noxious thermal stimulation in humans: a functional magnetic resonance imaging study. Neurosci Lett. 2003;335:202–6. Koyama T, McHaffie JG, Laurienti PJ, Coghill RC. The single-epoch fMRI design: validation of a simplified paradigm for the collection of subjective ratings. Neuroimage. 2003;19:976–87. Strigo IA, Duncan GH, Boivin M, Bushnell MC. Differentiation of visceral and cutaneous pain in the human brain. J Neurophysiol. 2003;89:3294–303. Ibinson JW, Small RH, Algaze A, Roberts CJ, Clark DL, Schmalbrock P. Functional magnetic resonance imaging studies of pain: an investigation of signal decay during and across sessions. Anesthesiology. 2004;101:960–9. Wager TD, Rilling JK, Smith EE, Sokolik A, Casey KL, Davidson RJ, et al. Placebo-induced changes in FMRI in the

272

126.

127.

128.

129.

130.

131.

D. Borsook et al. anticipation and experience of pain. Science. 2004; 303(5661):1162–7. Wiech K, Seymour B, Kalisch R, Stephan KE, Koltzenburg M, Driver J, et al. Modulation of pain processing in hyperalgesia by cognitive demand. Neuroimage. 2005;27:59–69. Albuquerque RJ, de Leeuw R, Carlson CR, Okeson JP, Miller CS, Andersen AH. Cerebral activation during thermal stimulation of patients who have burning mouth disorder: an fMRI study. Pain. 2006;122:223–34. Kong J, White NS, Kwong KK, Vangel MG, Rosman IS, Gracely RH, et al. Using fMRI to dissociate sensory encoding from cognitive evaluation of heat pain intensity. Hum Brain Mapp. 2006;27:715–21. Seminowicz DA, Davis KD. Interactions of pain intensity and cognitive load: the brain stays on task. Cereb Cortex. 2007;17(6):1412–22. Bantick SJ, Wise RG, Ploghaus A, Clare S, Smith SM, Tracey I. Imaging how attention modulates pain in humans using functional MRI. Brain. 2002;125(Pt 2):310–9. Brooks JC, Nurmikko TJ, Bimson WE, Singh KD, Roberts N. fMRI of thermal pain: effects of stimulus laterality and attention. Neuroimage. 2002;15:293–301.

132. Smith KA, Ploghaus A, Cowen PJ, McCleery JM, Goodwin GM, Smith S, et al. Cerebellar responses during anticipation of noxious stimuli in subjects recovered from depression. Functional magnetic resonance imaging study. Br J Psychiatry. 2002;181:411–5. 133. Moriguchi Y, Decety J, Ohnishi T, Maeda M, Mori T, Nemoto K, et al. Empathy and judging other’s pain: an fMRI study of alexithymia. Cereb Cortex. 2006 Dec 5; [Epub ahead of print]. 134. Ducreux D, Attal N, Parker F, Bouhassira D. Mechanisms of central neuropathic pain: a combined psychophysical and fMRI study in syringomyelia. Brain. 2006;129(Pt 4):963–76. 135. 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. 136. Witting N, Kupers RC, Svensson P, Jensen TS. A PET activation study of brush-evoked allodynia in patients with nerve injury pain. Pain. 2006;120:145–54. 137. Geha PY, Baliki MN, Chialvo DR, Harden RN, Paice JA, Apkarian AV. Brain activity for spontaneous pain of postherpetic neuralgia and its modulation by lidocaine patch therapy. Pain. 2007;128:88–100.